diff --git a/.gitattributes b/.gitattributes index a6344aac8c09253b3b630fb776ae94478aa0275b..5f2d567169f2cfa9f214b4ac7658b44d9b1cc378 100644 --- a/.gitattributes +++ b/.gitattributes @@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text *.zip filter=lfs diff=lfs merge=lfs -text *.zst filter=lfs diff=lfs merge=lfs -text *tfevents* filter=lfs diff=lfs merge=lfs -text +*.ttf filter=lfs diff=lfs merge=lfs -text diff --git a/app.py b/app.py new file mode 100644 index 0000000000000000000000000000000000000000..c2da852da8361b3732166878065f1ddab3f124d5 --- /dev/null +++ b/app.py @@ -0,0 +1,401 @@ +''' +AnyText: Multilingual Visual Text Generation And Editing +Paper: https://arxiv.org/abs/2311.03054 +Code: https://github.com/tyxsspa/AnyText +Copyright (c) Alibaba, Inc. and its affiliates. +''' +import os +from modelscope.pipelines import pipeline +import cv2 +import gradio as gr +import numpy as np +import re +from gradio.components import Component +from util import check_channels, resize_image, save_images +import json + +BBOX_MAX_NUM = 8 +img_save_folder = 'SaveImages' +load_model = True +if load_model: + inference = pipeline('my-anytext-task', model='damo/cv_anytext_text_generation_editing', model_revision='v1.1.0') + + +def count_lines(prompt): + prompt = prompt.replace('“', '"') + prompt = prompt.replace('”', '"') + p = '"(.*?)"' + strs = re.findall(p, prompt) + if len(strs) == 0: + strs = [' '] + return len(strs) + + +def generate_rectangles(w, h, n, max_trys=200): + img = np.zeros((h, w, 1), dtype=np.uint8) + rectangles = [] + attempts = 0 + n_pass = 0 + low_edge = int(max(w, h)*0.3 if n <= 3 else max(w, h)*0.2) # ~150, ~100 + while attempts < max_trys: + rect_w = min(np.random.randint(max((w*0.5)//n, low_edge), w), int(w*0.8)) + ratio = np.random.uniform(4, 10) + rect_h = max(low_edge, int(rect_w/ratio)) + rect_h = min(rect_h, int(h*0.8)) + # gen rotate angle + rotation_angle = 0 + rand_value = np.random.rand() + if rand_value < 0.7: + pass + elif rand_value < 0.8: + rotation_angle = np.random.randint(0, 40) + elif rand_value < 0.9: + rotation_angle = np.random.randint(140, 180) + else: + rotation_angle = np.random.randint(85, 95) + # rand position + x = np.random.randint(0, w - rect_w) + y = np.random.randint(0, h - rect_h) + # get vertex + rect_pts = cv2.boxPoints(((rect_w/2, rect_h/2), (rect_w, rect_h), rotation_angle)) + rect_pts = np.int32(rect_pts) + # move + rect_pts += (x, y) + # check boarder + if np.any(rect_pts < 0) or np.any(rect_pts[:, 0] >= w) or np.any(rect_pts[:, 1] >= h): + attempts += 1 + continue + # check overlap + if any(check_overlap_polygon(rect_pts, rp) for rp in rectangles): + attempts += 1 + continue + n_pass += 1 + cv2.fillPoly(img, [rect_pts], 255) + rectangles.append(rect_pts) + if n_pass == n: + break + print("attempts:", attempts) + if len(rectangles) != n: + raise gr.Error(f'Failed in auto generate positions after {attempts} attempts, try again!') + return img + + +def check_overlap_polygon(rect_pts1, rect_pts2): + poly1 = cv2.convexHull(rect_pts1) + poly2 = cv2.convexHull(rect_pts2) + rect1 = cv2.boundingRect(poly1) + rect2 = cv2.boundingRect(poly2) + if rect1[0] + rect1[2] >= rect2[0] and rect2[0] + rect2[2] >= rect1[0] and rect1[1] + rect1[3] >= rect2[1] and rect2[1] + rect2[3] >= rect1[1]: + return True + return False + + +def draw_rects(width, height, rects): + img = np.zeros((height, width, 1), dtype=np.uint8) + for rect in rects: + x1 = int(rect[0] * width) + y1 = int(rect[1] * height) + w = int(rect[2] * width) + h = int(rect[3] * height) + x2 = x1 + w + y2 = y1 + h + cv2.rectangle(img, (x1, y1), (x2, y2), 255, -1) + return img + + +def process(mode, prompt, pos_radio, sort_radio, revise_pos, show_debug, draw_img, rect_img, ref_img, ori_img, img_count, ddim_steps, w, h, strength, cfg_scale, seed, eta, a_prompt, n_prompt, *rect_list): + n_lines = count_lines(prompt) + # Text Generation + if mode == 'gen': + # create pos_imgs + if pos_radio == 'Manual-draw(手绘)': + if draw_img is not None: + pos_imgs = 255 - draw_img['image'] + if 'mask' in draw_img: + pos_imgs = pos_imgs.astype(np.float32) + draw_img['mask'][..., 0:3].astype(np.float32) + pos_imgs = pos_imgs.clip(0, 255).astype(np.uint8) + else: + pos_imgs = np.zeros((w, h, 1)) + elif pos_radio == 'Manual-rect(拖框)': + rect_check = rect_list[:BBOX_MAX_NUM] + rect_xywh = rect_list[BBOX_MAX_NUM:] + checked_rects = [] + for idx, c in enumerate(rect_check): + if c: + _xywh = rect_xywh[4*idx:4*(idx+1)] + checked_rects += [_xywh] + pos_imgs = draw_rects(w, h, checked_rects) + elif pos_radio == 'Auto-rand(随机)': + pos_imgs = generate_rectangles(w, h, n_lines, max_trys=500) + # Text Editing + elif mode == 'edit': + revise_pos = False # disable pos revise in edit mode + if ref_img is None or ori_img is None: + raise gr.Error('No reference image, please upload one for edit!') + edit_image = ori_img.clip(1, 255) # for mask reason + edit_image = check_channels(edit_image) + edit_image = resize_image(edit_image, max_length=768) + h, w = edit_image.shape[:2] + if isinstance(ref_img, dict) and 'mask' in ref_img and ref_img['mask'].mean() > 0: + pos_imgs = 255 - edit_image + edit_mask = cv2.resize(ref_img['mask'][..., 0:3], (w, h)) + pos_imgs = pos_imgs.astype(np.float32) + edit_mask.astype(np.float32) + pos_imgs = pos_imgs.clip(0, 255).astype(np.uint8) + else: + if isinstance(ref_img, dict) and 'image' in ref_img: + ref_img = ref_img['image'] + pos_imgs = 255 - ref_img # example input ref_img is used as pos + cv2.imwrite('pos_imgs.png', 255-pos_imgs[..., ::-1]) + params = { + "sort_priority": sort_radio, + "show_debug": show_debug, + "revise_pos": revise_pos, + "image_count": img_count, + "ddim_steps": ddim_steps, + "image_width": w, + "image_height": h, + "strength": strength, + "cfg_scale": cfg_scale, + "eta": eta, + "a_prompt": a_prompt, + "n_prompt": n_prompt + } + input_data = { + "prompt": prompt, + "seed": seed, + "draw_pos": pos_imgs, + "ori_image": ori_img, + } + results, rtn_code, rtn_warning, debug_info = inference(input_data, mode=mode, **params) + if rtn_code >= 0: + # save_images(results, img_save_folder) + # print(f'Done, result images are saved in: {img_save_folder}') + if rtn_warning: + gr.Warning(rtn_warning) + else: + raise gr.Error(rtn_warning) + return results, gr.Markdown(debug_info, visible=show_debug) + + +def create_canvas(w=512, h=512, c=3, line=5): + image = np.full((h, w, c), 200, dtype=np.uint8) + for i in range(h): + if i % (w//line) == 0: + image[i, :, :] = 150 + for j in range(w): + if j % (w//line) == 0: + image[:, j, :] = 150 + image[h//2-8:h//2+8, w//2-8:w//2+8, :] = [200, 0, 0] + return image + + +def resize_w(w, img1, img2): + if isinstance(img2, dict): + img2 = img2['image'] + return [cv2.resize(img1, (w, img1.shape[0])), cv2.resize(img2, (w, img2.shape[0]))] + + +def resize_h(h, img1, img2): + if isinstance(img2, dict): + img2 = img2['image'] + return [cv2.resize(img1, (img1.shape[1], h)), cv2.resize(img2, (img2.shape[1], h))] + + +is_t2i = 'true' +block = gr.Blocks(css='style.css', theme=gr.themes.Soft()).queue() + +with open('javascript/bboxHint.js', 'r') as file: + value = file.read() +escaped_value = json.dumps(value) + +with block: + block.load(fn=None, + _js=f"""() => {{ + const script = document.createElement("script"); + const text = document.createTextNode({escaped_value}); + script.appendChild(text); + document.head.appendChild(script); + }}""") + gr.HTML('
\ +
\ + [arXiv] \ + [Code] \ + [ModelScope]\ + version: 1.1.0
') + with gr.Row(variant='compact'): + with gr.Column(): + with gr.Accordion('🕹Instructions(说明)', open=False,): + with gr.Tabs(): + with gr.Tab("English"): + gr.Markdown('Run Examples') + gr.Markdown('AnyText has two modes: Text Generation and Text Editing, and we provides a variety of examples. Select one, click on [Run!] button to run.') + gr.Markdown('Please note, before running examples, ensure the manual draw area is empty, otherwise may get wrong results. Additionally, different examples use \ + different parameters (such as resolution, seed, etc.). When generate your own, please pay attention to the parameter changes, or refresh the page to restore the default parameters.') + gr.Markdown('Text Generation') + gr.Markdown('Enter the textual description (in Chinese or English) of the image you want to generate in [Prompt]. Each text line that needs to be generated should be \ + enclosed in double quotes. Then, manually draw the specified position for each text line to generate the image.\ + The drawing of text positions is crucial to the quality of the resulting image, \ + please do not draw too casually or too small. The number of positions should match the number of text lines, and the size of each position should be matched \ + as closely as possible to the length or width of the corresponding text line. If [Manual-draw] is inconvenient, you can try dragging rectangles [Manual-rect] or random positions [Auto-rand].') + gr.Markdown('When generating multiple lines, each position is matched with the text line according to a certain rule. The [Sort Position] option is used to \ + determine whether to prioritize sorting from top to bottom or from left to right. You can open the [Show Debug] option in the parameter settings to observe the text position and glyph image \ + in the result. You can also select the [Revise Position] which uses the bounding box of the rendered text as the revised position. However, it is occasionally found that the creativity of the \ + generated text is slightly lower using this method.') + gr.Markdown('Text Editing') + gr.Markdown('Please upload an image in [Ref] as a reference image, then adjust the brush size, and mark the area(s) to be edited. Input the textual description and \ + the new text to be modified in [Prompt], then generate the image.') + gr.Markdown('The reference image can be of any resolution, but it will be internally processed with a limit that the longer side cannot exceed 768 pixels, and the \ + width and height will both be scaled to multiples of 64.') + with gr.Tab("简体中文"): + gr.Markdown('运行示例') + gr.Markdown('AnyText有两种运行模式:文字生成和文字编辑,每种模式下提供了丰富的示例,选择一个,点击[Run!]即可。') + gr.Markdown('请注意,运行示例前确保手绘位置区域是空的,防止影响示例结果,另外不同示例使用不同的参数(如分辨率,种子数等),如果要自行生成时,请留意参数变化,或刷新页面恢复到默认参数。') + gr.Markdown('文字生成') + gr.Markdown('在Prompt中输入描述提示词(支持中英文),需要生成的每一行文字用双引号包裹,然后依次手绘指定每行文字的位置,生成图片。\ + 文字位置的绘制对成图质量很关键, \ + 请不要画的太随意或太小,位置的数量要与文字行数量一致,每个位置的尺寸要与对应的文字行的长短或宽高尽量匹配。如果手绘(Manual-draw)不方便,\ + 可以尝试拖框矩形(Manual-rect)或随机生成(Auto-rand)。') + gr.Markdown('多行生成时,每个位置按照一定规则排序后与文字行做对应,Sort Position选项用于确定排序时优先从上到下还是从左到右。\ + 可以在参数设置中打开Show Debug选项,在结果图像中观察文字位置和字形图。也可以勾选Revise Position选项,这样会用渲染文字的外接矩形作为修正后的位置,不过偶尔发现这样生成的文字创造性略低。') + gr.Markdown('文字编辑') + gr.Markdown('请上传一张待编辑的图片作为参考图(Ref),然后调整笔触大小后,在参考图上涂抹要编辑的位置,在Prompt中输入描述提示词和要修改的文字内容,生成图片。') + gr.Markdown('参考图可以为任意分辨率,但内部处理时会限制长边不能超过768,并且宽高都被缩放为64的整数倍。') + with gr.Accordion('🛠Parameters(参数)', open=False): + with gr.Row(variant='compact'): + img_count = gr.Slider(label="Image Count(图片数)", minimum=1, maximum=12, value=4, step=1) + ddim_steps = gr.Slider(label="Steps(步数)", minimum=1, maximum=100, value=20, step=1) + with gr.Row(variant='compact'): + image_width = gr.Slider(label="Image Width(宽度)", minimum=256, maximum=768, value=512, step=64) + image_height = gr.Slider(label="Image Height(高度)", minimum=256, maximum=768, value=512, step=64) + with gr.Row(variant='compact'): + strength = gr.Slider(label="Strength(控制力度)", minimum=0.0, maximum=2.0, value=1.0, step=0.01) + cfg_scale = gr.Slider(label="CFG-Scale(CFG强度)", minimum=0.1, maximum=30.0, value=9.0, step=0.1) + with gr.Row(variant='compact'): + seed = gr.Slider(label="Seed(种子数)", minimum=-1, maximum=99999999, step=1, randomize=False, value=-1) + eta = gr.Number(label="eta (DDIM)", value=0.0) + with gr.Row(variant='compact'): + show_debug = gr.Checkbox(label='Show Debug(调试信息)', value=False) + gr.Markdown('whether show glyph image and debug information in the result(是否在结果中显示glyph图以及调试信息)') + a_prompt = gr.Textbox(label="Added Prompt(附加提示词)", value='best quality, extremely detailed,4k, HD, supper legible text, clear text edges, clear strokes, neat writing, no watermarks') + n_prompt = gr.Textbox(label="Negative Prompt(负向提示词)", value='low-res, bad anatomy, extra digit, fewer digits, cropped, worst quality, low quality, watermark, unreadable text, messy words, distorted text, disorganized writing, advertising picture') + prompt = gr.Textbox(label="Prompt(提示词)") + with gr.Tabs() as tab_modes: + with gr.Tab("🖼Text Generation(文字生成)", elem_id='MD-tab-t2i') as mode_gen: + pos_radio = gr.Radio(["Manual-draw(手绘)", "Manual-rect(拖框)", "Auto-rand(随机)"], value='Manual-draw(手绘)', label="Pos-Method(位置方式)", info="choose a method to specify text positions(选择方法用于指定文字位置).") + with gr.Row(): + sort_radio = gr.Radio(["↕", "↔"], value='↕', label="Sort Position(位置排序)", info="position sorting priority(位置排序时的优先级)") + revise_pos = gr.Checkbox(label='Revise Position(修正位置)', value=False) + # gr.Markdown('try to revise according to text\'s bounding rectangle(尝试通过渲染后的文字行的外接矩形框修正位置)') + with gr.Row(variant='compact'): + rect_cb_list: list[Component] = [] + rect_xywh_list: list[Component] = [] + for i in range(BBOX_MAX_NUM): + e = gr.Checkbox(label=f'{i}', value=False, visible=False, min_width='10') + x = gr.Slider(label='x', value=0.4, minimum=0.0, maximum=1.0, step=0.0001, elem_id=f'MD-t2i-{i}-x', visible=False) + y = gr.Slider(label='y', value=0.4, minimum=0.0, maximum=1.0, step=0.0001, elem_id=f'MD-t2i-{i}-y', visible=False) + w = gr.Slider(label='w', value=0.2, minimum=0.0, maximum=1.0, step=0.0001, elem_id=f'MD-t2i-{i}-w', visible=False) + h = gr.Slider(label='h', value=0.2, minimum=0.0, maximum=1.0, step=0.0001, elem_id=f'MD-t2i-{i}-h', visible=False) + x.change(fn=None, inputs=x, outputs=x, _js=f'v => onBoxChange({is_t2i}, {i}, "x", v)', show_progress=False, queue=False) + y.change(fn=None, inputs=y, outputs=y, _js=f'v => onBoxChange({is_t2i}, {i}, "y", v)', show_progress=False, queue=False) + w.change(fn=None, inputs=w, outputs=w, _js=f'v => onBoxChange({is_t2i}, {i}, "w", v)', show_progress=False, queue=False) + h.change(fn=None, inputs=h, outputs=h, _js=f'v => onBoxChange({is_t2i}, {i}, "h", v)', show_progress=False, queue=False) + + e.change(fn=None, inputs=e, outputs=e, _js=f'e => onBoxEnableClick({is_t2i}, {i}, e)', queue=False) + rect_cb_list.extend([e]) + rect_xywh_list.extend([x, y, w, h]) + + rect_img = gr.Image(value=create_canvas(), label="Rext Position(方框位置)", elem_id="MD-bbox-rect-t2i", show_label=False, visible=False) + draw_img = gr.Image(value=create_canvas(), label="Draw Position(绘制位置)", visible=True, tool='sketch', show_label=False, brush_radius=60) + + def re_draw(): + return [gr.Image(value=create_canvas(), tool='sketch'), gr.Slider(value=512), gr.Slider(value=512)] + draw_img.clear(re_draw, None, [draw_img, image_width, image_height]) + image_width.release(resize_w, [image_width, rect_img, draw_img], [rect_img, draw_img]) + image_height.release(resize_h, [image_height, rect_img, draw_img], [rect_img, draw_img]) + + def change_options(selected_option): + return [gr.Checkbox(visible=selected_option == 'Manual-rect(拖框)')] * BBOX_MAX_NUM + \ + [gr.Image(visible=selected_option == 'Manual-rect(拖框)'), + gr.Image(visible=selected_option == 'Manual-draw(手绘)'), + gr.Radio(visible=selected_option != 'Auto-rand(随机)'), + gr.Checkbox(value=selected_option == 'Auto-rand(随机)')] + pos_radio.change(change_options, pos_radio, rect_cb_list + [rect_img, draw_img, sort_radio, revise_pos], show_progress=False, queue=False) + with gr.Row(): + gr.Markdown("") + run_gen = gr.Button(value="Run(运行)!", scale=0.3, elem_classes='run') + gr.Markdown("") + + def exp_gen_click(): + return [gr.Slider(value=512), gr.Slider(value=512)] # all examples are 512x512, refresh draw_img + exp_gen = gr.Examples( + [ + ['一只浣熊站在黑板前,上面写着"深度学习"', "example_images/gen1.png", "Manual-draw(手绘)", "↕", False, 4, 81808278], + ['一个儿童蜡笔画,森林里有一个可爱的蘑菇形状的房子,标题是"森林小屋"', "example_images/gen16.png", "Manual-draw(手绘)", "↕", False, 4, 40173333], + ['一个精美设计的logo,画的是一个黑白风格的厨师,带着厨师帽,logo下方写着“深夜食堂”', "example_images/gen14.png", "Manual-draw(手绘)", "↕", False, 4, 6970544], + ['photo of caramel macchiato coffee on the table, top-down perspective, with "Any" "Text" written on it using cream', "example_images/gen9.png", "Manual-draw(手绘)", "↕", False, 4, 66273235], + ['一张户外雪地靴的电商广告,上面写着 “双12大促!”,“立减50”,“加绒加厚”,“穿脱方便”,“温暖24小时送达”, “包邮”,高级设计感,精美构图', "example_images/gen15.png", "Manual-draw(手绘)", "↕", False, 4, 66980376], + ['Sign on the clean building that reads "科学" and "과학" and "ステップ" and "SCIENCE"', "example_images/gen6.png", "Manual-draw(手绘)", "↕", True, 4, 13246309], + ['一个精致的马克杯,上面雕刻着一首中国古诗,内容是 "花落知多少" "夜来风雨声" "处处闻啼鸟" "春眠不觉晓"', "example_images/gen3.png", "Manual-draw(手绘)", "↔", False, 4, 60358279], + ['A delicate square cake, cream and fruit, with "CHEERS" "to the" and "GRADUATE" written in chocolate', "example_images/gen8.png", "Manual-draw(手绘)", "↕", False, 4, 93424638], + ['一件精美的毛衣,上面有针织的文字:"通义丹青"', "example_images/gen4.png", "Manual-draw(手绘)", "↕", False, 4, 48769450], + ['一个双肩包的特写照,上面用针织文字写着”为了无法“ ”计算的价值“', "example_images/gen12.png", "Manual-draw(手绘)", "↕", False, 4, 35552323], + ['A nice drawing in pencil of Michael Jackson, with the words "Micheal" and "Jackson" written on it', "example_images/gen7.png", "Manual-draw(手绘)", "↕", False, 4, 83866922], + ['一个漂亮的蜡笔画,有行星,宇航员,还有宇宙飞船,上面写的是"去火星旅行", "王小明", "11月1日"', "example_images/gen5.png", "Manual-draw(手绘)", "↕", False, 4, 42328250], + ['一个装饰华丽的蛋糕,上面用奶油写着“阿里云”和"APSARA"', "example_images/gen13.png", "Manual-draw(手绘)", "↕", False, 4, 62357019], + ['一张关于墙上的彩色涂鸦艺术的摄影作品,上面写着“人工智能" 和 "神经网络"', "example_images/gen10.png", "Manual-draw(手绘)", "↕", False, 4, 64722007], + ['一枚中国古代铜钱, 上面的文字是 "康" "寶" "通" "熙"', "example_images/gen2.png", "Manual-draw(手绘)", "↕", False, 4, 24375031], + ['a well crafted ice sculpture that made with "Happy" and "Holidays". Dslr photo, perfect illumination', "example_images/gen11.png", "Manual-draw(手绘)", "↕", True, 4, 64901362], + ], + [prompt, draw_img, pos_radio, sort_radio, revise_pos, img_count, seed], + examples_per_page=5, + ) + exp_gen.dataset.click(exp_gen_click, None, [image_width, image_height]) + + with gr.Tab("🎨Text Editing(文字编辑)") as mode_edit: + with gr.Row(variant='compact'): + ref_img = gr.Image(label='Ref(参考图)', source='upload') + ori_img = gr.Image(label='Ori(原图)') + + def upload_ref(x): + return [gr.Image(type="numpy", brush_radius=60, tool='sketch'), + gr.Image(value=x)] + + def clear_ref(x): + return gr.Image(source='upload', tool=None) + ref_img.upload(upload_ref, ref_img, [ref_img, ori_img]) + ref_img.clear(clear_ref, ref_img, ref_img) + with gr.Row(): + gr.Markdown("") + run_edit = gr.Button(value="Run(运行)!", scale=0.3, elem_classes='run') + gr.Markdown("") + gr.Examples( + [ + ['精美的书法作品,上面写着“志” “存” “高” ”远“', "example_images/ref10.jpg", "example_images/edit10.png", 4, 98053044], + ['一个表情包,小猪说 "下班"', "example_images/ref2.jpg", "example_images/edit2.png", 2, 43304008], + ['Characters written in chalk on the blackboard that says "DADDY"', "example_images/ref8.jpg", "example_images/edit8.png", 4, 73556391], + ['一个中国古代铜钱,上面写着"乾" "隆"', "example_images/ref12.png", "example_images/edit12.png", 4, 89159482], + ['黑板上写着"Here"', "example_images/ref11.jpg", "example_images/edit11.png", 2, 15353513], + ['A letter picture that says "THER"', "example_images/ref6.jpg", "example_images/edit6.png", 4, 72321415], + ['一堆水果, 中间写着“UIT”', "example_images/ref13.jpg", "example_images/edit13.png", 4, 54263567], + ['一个漫画,上面写着" "', "example_images/ref14.png", "example_images/edit14.png", 4, 94081527], + ['一个黄色标志牌,上边写着"不要" 和 "大意"', "example_images/ref3.jpg", "example_images/edit3.png", 2, 64010349], + ['A cake with colorful characters that reads "EVERYDAY"', "example_images/ref7.jpg", "example_images/edit7.png", 4, 8943410], + ['一个青铜鼎,上面写着" "和" "', "example_images/ref4.jpg", "example_images/edit4.png", 4, 71139289], + ['一个建筑物前面的字母标牌, 上面写着 " "', "example_images/ref5.jpg", "example_images/edit5.png", 4, 50416289], + ], + [prompt, ori_img, ref_img, img_count, seed], + examples_per_page=5, + ) + with gr.Column(): + result_gallery = gr.Gallery(label='Result(结果)', show_label=True, preview=True, columns=2, allow_preview=True, height=600) + result_info = gr.Markdown('', visible=False) + ips = [prompt, pos_radio, sort_radio, revise_pos, show_debug, draw_img, rect_img, ref_img, ori_img, img_count, ddim_steps, image_width, image_height, strength, cfg_scale, seed, eta, a_prompt, n_prompt, *(rect_cb_list+rect_xywh_list)] + run_gen.click(fn=process, inputs=[gr.State('gen')] + ips, outputs=[result_gallery, result_info]) + run_edit.click(fn=process, inputs=[gr.State('edit')] + ips, outputs=[result_gallery, result_info]) + +block.launch( + server_name='0.0.0.0' if os.getenv('GRADIO_LISTEN', '') != '' else "127.0.0.1", + share=False, + root_path=f"/{os.getenv('GRADIO_PROXY_PATH')}" if os.getenv('GRADIO_PROXY_PATH') else "" +) +# block.launch(server_name='0.0.0.0') diff --git a/bert_tokenizer.py b/bert_tokenizer.py new file mode 100644 index 0000000000000000000000000000000000000000..1f7315134f6e8182c50c87d30dd976e144a2cd89 --- /dev/null +++ b/bert_tokenizer.py @@ -0,0 +1,421 @@ +# Copyright 2018 The Google AI Language Team Authors. +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +"""Tokenization classes.""" + +from __future__ import absolute_import, division, print_function +import collections +import re +import unicodedata + +import six + + +def validate_case_matches_checkpoint(do_lower_case, init_checkpoint): + """Checks whether the casing config is consistent with the checkpoint name.""" + + # The casing has to be passed in by the user and there is no explicit check + # as to whether it matches the checkpoint. The casing information probably + # should have been stored in the bert_config.json file, but it's not, so + # we have to heuristically detect it to validate. + + if not init_checkpoint: + return + + m = re.match('^.*?([A-Za-z0-9_-]+)/bert_model.ckpt', init_checkpoint) + if m is None: + return + + model_name = m.group(1) + + lower_models = [ + 'uncased_L-24_H-1024_A-16', 'uncased_L-12_H-768_A-12', + 'multilingual_L-12_H-768_A-12', 'chinese_L-12_H-768_A-12' + ] + + cased_models = [ + 'cased_L-12_H-768_A-12', 'cased_L-24_H-1024_A-16', + 'multi_cased_L-12_H-768_A-12' + ] + + is_bad_config = False + if model_name in lower_models and not do_lower_case: + is_bad_config = True + actual_flag = 'False' + case_name = 'lowercased' + opposite_flag = 'True' + + if model_name in cased_models and do_lower_case: + is_bad_config = True + actual_flag = 'True' + case_name = 'cased' + opposite_flag = 'False' + + if is_bad_config: + raise ValueError( + 'You passed in `--do_lower_case=%s` with `--init_checkpoint=%s`. ' + 'However, `%s` seems to be a %s model, so you ' + 'should pass in `--do_lower_case=%s` so that the fine-tuning matches ' + 'how the model was pre-training. If this error is wrong, please ' + 'just comment out this check.' % + (actual_flag, init_checkpoint, model_name, case_name, + opposite_flag)) + + +def convert_to_unicode(text): + """Converts `text` to Unicode (if it's not already), assuming utf-8 input.""" + if six.PY3: + if isinstance(text, str): + return text + elif isinstance(text, bytes): + return text.decode('utf-8', 'ignore') + else: + raise ValueError('Unsupported string type: %s' % (type(text))) + elif six.PY2: + if isinstance(text, str): + return text.decode('utf-8', 'ignore') + elif isinstance(text, unicode): + return text + else: + raise ValueError('Unsupported string type: %s' % (type(text))) + else: + raise ValueError('Not running on Python2 or Python 3?') + + +def printable_text(text): + """Returns text encoded in a way suitable for print or `tf.logging`.""" + + # These functions want `str` for both Python2 and Python3, but in one case + # it's a Unicode string and in the other it's a byte string. + if six.PY3: + if isinstance(text, str): + return text + elif isinstance(text, bytes): + return text.decode('utf-8', 'ignore') + else: + raise ValueError('Unsupported string type: %s' % (type(text))) + elif six.PY2: + if isinstance(text, str): + return text + elif isinstance(text, unicode): + return text.encode('utf-8') + else: + raise ValueError('Unsupported string type: %s' % (type(text))) + else: + raise ValueError('Not running on Python2 or Python 3?') + + +def load_vocab(vocab_file): + """Loads a vocabulary file into a dictionary.""" + vocab = collections.OrderedDict() + index = 0 + with open(vocab_file, 'r', encoding='utf-8') as reader: + while True: + token = convert_to_unicode(reader.readline()) + if not token: + break + token = token.strip() + vocab[token] = index + index += 1 + return vocab + + +def convert_by_vocab(vocab, items): + """Converts a sequence of [tokens|ids] using the vocab.""" + output = [] + for item in items: + output.append(vocab[item]) + return output + + +def convert_tokens_to_ids(vocab, tokens): + return convert_by_vocab(vocab, tokens) + + +def convert_ids_to_tokens(inv_vocab, ids): + return convert_by_vocab(inv_vocab, ids) + + +def whitespace_tokenize(text): + """Runs basic whitespace cleaning and splitting on a piece of text.""" + text = text.strip() + if not text: + return [] + tokens = text.split() + return tokens + + +class FullTokenizer(object): + """Runs end-to-end tokenziation.""" + + def __init__(self, vocab_file, do_lower_case=True): + self.vocab = load_vocab(vocab_file) + self.inv_vocab = {v: k for k, v in self.vocab.items()} + self.basic_tokenizer = BasicTokenizer(do_lower_case=do_lower_case) + self.wordpiece_tokenizer = WordpieceTokenizer(vocab=self.vocab) + + def tokenize(self, text): + split_tokens = [] + for token in self.basic_tokenizer.tokenize(text): + for sub_token in self.wordpiece_tokenizer.tokenize(token): + split_tokens.append(sub_token) + + return split_tokens + + def convert_tokens_to_ids(self, tokens): + return convert_by_vocab(self.vocab, tokens) + + def convert_ids_to_tokens(self, ids): + return convert_by_vocab(self.inv_vocab, ids) + + @staticmethod + def convert_tokens_to_string(tokens, clean_up_tokenization_spaces=True): + """ Converts a sequence of tokens (string) in a single string. """ + + def clean_up_tokenization(out_string): + """ Clean up a list of simple English tokenization artifacts + like spaces before punctuations and abreviated forms. + """ + out_string = ( + out_string.replace(' .', '.').replace(' ?', '?').replace( + ' !', '!').replace(' ,', ',').replace(" ' ", "'").replace( + " n't", "n't").replace(" 'm", "'m").replace( + " 's", "'s").replace(" 've", + "'ve").replace(" 're", "'re")) + return out_string + + text = ' '.join(tokens).replace(' ##', '').strip() + if clean_up_tokenization_spaces: + clean_text = clean_up_tokenization(text) + return clean_text + else: + return text + + def vocab_size(self): + return len(self.vocab) + + +class BasicTokenizer(object): + """Runs basic tokenization (punctuation splitting, lower casing, etc.).""" + + def __init__(self, do_lower_case=True): + """Constructs a BasicTokenizer. + + Args: + do_lower_case: Whether to lower case the input. + """ + self.do_lower_case = do_lower_case + + def tokenize(self, text): + """Tokenizes a piece of text.""" + text = convert_to_unicode(text) + text = self._clean_text(text) + + # This was added on November 1st, 2018 for the multilingual and Chinese + # models. This is also applied to the English models now, but it doesn't + # matter since the English models were not trained on any Chinese data + # and generally don't have any Chinese data in them (there are Chinese + # characters in the vocabulary because Wikipedia does have some Chinese + # words in the English Wikipedia.). + text = self._tokenize_chinese_chars(text) + + orig_tokens = whitespace_tokenize(text) + split_tokens = [] + for token in orig_tokens: + if self.do_lower_case: + token = token.lower() + token = self._run_strip_accents(token) + split_tokens.extend(self._run_split_on_punc(token)) + + output_tokens = whitespace_tokenize(' '.join(split_tokens)) + return output_tokens + + def _run_strip_accents(self, text): + """Strips accents from a piece of text.""" + text = unicodedata.normalize('NFD', text) + output = [] + for char in text: + cat = unicodedata.category(char) + if cat == 'Mn': + continue + output.append(char) + return ''.join(output) + + def _run_split_on_punc(self, text): + """Splits punctuation on a piece of text.""" + chars = list(text) + i = 0 + start_new_word = True + output = [] + while i < len(chars): + char = chars[i] + if _is_punctuation(char): + output.append([char]) + start_new_word = True + else: + if start_new_word: + output.append([]) + start_new_word = False + output[-1].append(char) + i += 1 + + return [''.join(x) for x in output] + + def _tokenize_chinese_chars(self, text): + """Adds whitespace around any CJK character.""" + output = [] + for char in text: + cp = ord(char) + if self._is_chinese_char(cp): + output.append(' ') + output.append(char) + output.append(' ') + else: + output.append(char) + return ''.join(output) + + def _is_chinese_char(self, cp): + """Checks whether CP is the codepoint of a CJK character.""" + # This defines a "chinese character" as anything in the CJK Unicode block: + # https://en.wikipedia.org/wiki/CJK_Unified_Ideographs_(Unicode_block) + # + # Note that the CJK Unicode block is NOT all Japanese and Korean characters, + # despite its name. The modern Korean Hangul alphabet is a different block, + # as is Japanese Hiragana and Katakana. Those alphabets are used to write + # space-separated words, so they are not treated specially and handled + # like the all of the other languages. + if ((cp >= 0x4E00 and cp <= 0x9FFF) or (cp >= 0x3400 and cp <= 0x4DBF) + or (cp >= 0x20000 and cp <= 0x2A6DF) + or (cp >= 0x2A700 and cp <= 0x2B73F) + or (cp >= 0x2B740 and cp <= 0x2B81F) + or (cp >= 0x2B820 and cp <= 0x2CEAF) + or (cp >= 0xF900 and cp <= 0xFAFF) + or (cp >= 0x2F800 and cp <= 0x2FA1F)): + return True + + return False + + def _clean_text(self, text): + """Performs invalid character removal and whitespace cleanup on text.""" + output = [] + for char in text: + cp = ord(char) + if cp == 0 or cp == 0xfffd or _is_control(char): + continue + if _is_whitespace(char): + output.append(' ') + else: + output.append(char) + return ''.join(output) + + +class WordpieceTokenizer(object): + """Runs WordPiece tokenziation.""" + + def __init__(self, vocab, unk_token='[UNK]', max_input_chars_per_word=200): + self.vocab = vocab + self.unk_token = unk_token + self.max_input_chars_per_word = max_input_chars_per_word + + def tokenize(self, text): + """Tokenizes a piece of text into its word pieces. + + This uses a greedy longest-match-first algorithm to perform tokenization + using the given vocabulary. + + For example: + input = "unaffable" + output = ["un", "##aff", "##able"] + + Args: + text: A single token or whitespace separated tokens. This should have + already been passed through `BasicTokenizer. + + Returns: + A list of wordpiece tokens. + """ + + text = convert_to_unicode(text) + + output_tokens = [] + for token in whitespace_tokenize(text): + chars = list(token) + if len(chars) > self.max_input_chars_per_word: + output_tokens.append(self.unk_token) + continue + + is_bad = False + start = 0 + sub_tokens = [] + while start < len(chars): + end = len(chars) + cur_substr = None + while start < end: + substr = ''.join(chars[start:end]) + if start > 0: + substr = '##' + substr + if substr in self.vocab: + cur_substr = substr + break + end -= 1 + if cur_substr is None: + is_bad = True + break + sub_tokens.append(cur_substr) + start = end + + if is_bad: + output_tokens.append(self.unk_token) + else: + output_tokens.extend(sub_tokens) + return output_tokens + + +def _is_whitespace(char): + """Checks whether `chars` is a whitespace character.""" + # \t, \n, and \r are technically contorl characters but we treat them + # as whitespace since they are generally considered as such. + if char == ' ' or char == '\t' or char == '\n' or char == '\r': + return True + cat = unicodedata.category(char) + if cat == 'Zs': + return True + return False + + +def _is_control(char): + """Checks whether `chars` is a control character.""" + # These are technically control characters but we count them as whitespace + # characters. + if char == '\t' or char == '\n' or char == '\r': + return False + cat = unicodedata.category(char) + if cat in ('Cc', 'Cf'): + return True + return False + + +def _is_punctuation(char): + """Checks whether `chars` is a punctuation character.""" + cp = ord(char) + # We treat all non-letter/number ASCII as punctuation. + # Characters such as "^", "$", and "`" are not in the Unicode + # Punctuation class but we treat them as punctuation anyways, for + # consistency. + if ((cp >= 33 and cp <= 47) or (cp >= 58 and cp <= 64) + or (cp >= 91 and cp <= 96) or (cp >= 123 and cp <= 126)): + return True + cat = unicodedata.category(char) + if cat.startswith('P'): + return True + return False diff --git a/cldm/cldm.py b/cldm/cldm.py new file mode 100644 index 0000000000000000000000000000000000000000..978f5b1d3fc0b7f4e68ef0f48d74d4e7c026bce8 --- /dev/null +++ b/cldm/cldm.py @@ -0,0 +1,617 @@ +import einops +import torch +import torch as th +import torch.nn as nn +import copy +from easydict import EasyDict as edict + +from ldm.modules.diffusionmodules.util import ( + conv_nd, + linear, + zero_module, + timestep_embedding, +) + +from einops import rearrange, repeat +from torchvision.utils import make_grid +from ldm.modules.attention import SpatialTransformer +from ldm.modules.diffusionmodules.openaimodel import UNetModel, TimestepEmbedSequential, ResBlock, Downsample, AttentionBlock +from ldm.models.diffusion.ddpm import LatentDiffusion +from ldm.util import log_txt_as_img, exists, instantiate_from_config +from ldm.models.diffusion.ddim import DDIMSampler +from ldm.modules.distributions.distributions import DiagonalGaussianDistribution +from .recognizer import TextRecognizer, create_predictor + + +def count_parameters(model): + return sum(p.numel() for p in model.parameters() if p.requires_grad) + + +class ControlledUnetModel(UNetModel): + def forward(self, x, timesteps=None, context=None, control=None, only_mid_control=False, **kwargs): + hs = [] + with torch.no_grad(): + t_emb = timestep_embedding(timesteps, self.model_channels, repeat_only=False) + emb = self.time_embed(t_emb) + h = x.type(self.dtype) + for module in self.input_blocks: + h = module(h, emb, context) + hs.append(h) + h = self.middle_block(h, emb, context) + + if control is not None: + h += control.pop() + + for i, module in enumerate(self.output_blocks): + if only_mid_control or control is None: + h = torch.cat([h, hs.pop()], dim=1) + else: + h = torch.cat([h, hs.pop() + control.pop()], dim=1) + h = module(h, emb, context) + + h = h.type(x.dtype) + return self.out(h) + + +class ControlNet(nn.Module): + def __init__( + self, + image_size, + in_channels, + model_channels, + glyph_channels, + position_channels, + num_res_blocks, + attention_resolutions, + dropout=0, + channel_mult=(1, 2, 4, 8), + conv_resample=True, + dims=2, + use_checkpoint=False, + use_fp16=False, + num_heads=-1, + num_head_channels=-1, + num_heads_upsample=-1, + use_scale_shift_norm=False, + resblock_updown=False, + use_new_attention_order=False, + use_spatial_transformer=False, # custom transformer support + transformer_depth=1, # custom transformer support + context_dim=None, # custom transformer support + n_embed=None, # custom support for prediction of discrete ids into codebook of first stage vq model + legacy=True, + disable_self_attentions=None, + num_attention_blocks=None, + disable_middle_self_attn=False, + use_linear_in_transformer=False, + ): + super().__init__() + if use_spatial_transformer: + assert context_dim is not None, 'Fool!! You forgot to include the dimension of your cross-attention conditioning...' + + if context_dim is not None: + assert use_spatial_transformer, 'Fool!! You forgot to use the spatial transformer for your cross-attention conditioning...' + from omegaconf.listconfig import ListConfig + if type(context_dim) == ListConfig: + context_dim = list(context_dim) + + if num_heads_upsample == -1: + num_heads_upsample = num_heads + + if num_heads == -1: + assert num_head_channels != -1, 'Either num_heads or num_head_channels has to be set' + + if num_head_channels == -1: + assert num_heads != -1, 'Either num_heads or num_head_channels has to be set' + self.dims = dims + self.image_size = image_size + self.in_channels = in_channels + self.model_channels = model_channels + if isinstance(num_res_blocks, int): + self.num_res_blocks = len(channel_mult) * [num_res_blocks] + else: + if len(num_res_blocks) != len(channel_mult): + raise ValueError("provide num_res_blocks either as an int (globally constant) or " + "as a list/tuple (per-level) with the same length as channel_mult") + self.num_res_blocks = num_res_blocks + if disable_self_attentions is not None: + # should be a list of booleans, indicating whether to disable self-attention in TransformerBlocks or not + assert len(disable_self_attentions) == len(channel_mult) + if num_attention_blocks is not None: + assert len(num_attention_blocks) == len(self.num_res_blocks) + assert all(map(lambda i: self.num_res_blocks[i] >= num_attention_blocks[i], range(len(num_attention_blocks)))) + print(f"Constructor of UNetModel received num_attention_blocks={num_attention_blocks}. " + f"This option has LESS priority than attention_resolutions {attention_resolutions}, " + f"i.e., in cases where num_attention_blocks[i] > 0 but 2**i not in attention_resolutions, " + f"attention will still not be set.") + + self.attention_resolutions = attention_resolutions + self.dropout = dropout + self.channel_mult = channel_mult + self.conv_resample = conv_resample + self.use_checkpoint = use_checkpoint + self.dtype = th.float16 if use_fp16 else th.float32 + self.num_heads = num_heads + self.num_head_channels = num_head_channels + self.num_heads_upsample = num_heads_upsample + self.predict_codebook_ids = n_embed is not None + + time_embed_dim = model_channels * 4 + self.time_embed = nn.Sequential( + linear(model_channels, time_embed_dim), + nn.SiLU(), + linear(time_embed_dim, time_embed_dim), + ) + + self.input_blocks = nn.ModuleList( + [ + TimestepEmbedSequential( + conv_nd(dims, in_channels, model_channels, 3, padding=1) + ) + ] + ) + self.zero_convs = nn.ModuleList([self.make_zero_conv(model_channels)]) + + self.glyph_block = TimestepEmbedSequential( + conv_nd(dims, glyph_channels, 8, 3, padding=1), + nn.SiLU(), + conv_nd(dims, 8, 8, 3, padding=1), + nn.SiLU(), + conv_nd(dims, 8, 16, 3, padding=1, stride=2), + nn.SiLU(), + conv_nd(dims, 16, 16, 3, padding=1), + nn.SiLU(), + conv_nd(dims, 16, 32, 3, padding=1, stride=2), + nn.SiLU(), + conv_nd(dims, 32, 32, 3, padding=1), + nn.SiLU(), + conv_nd(dims, 32, 96, 3, padding=1, stride=2), + nn.SiLU(), + conv_nd(dims, 96, 96, 3, padding=1), + nn.SiLU(), + conv_nd(dims, 96, 256, 3, padding=1, stride=2), + nn.SiLU(), + ) + + self.position_block = TimestepEmbedSequential( + conv_nd(dims, position_channels, 8, 3, padding=1), + nn.SiLU(), + conv_nd(dims, 8, 8, 3, padding=1), + nn.SiLU(), + conv_nd(dims, 8, 16, 3, padding=1, stride=2), + nn.SiLU(), + conv_nd(dims, 16, 16, 3, padding=1), + nn.SiLU(), + conv_nd(dims, 16, 32, 3, padding=1, stride=2), + nn.SiLU(), + conv_nd(dims, 32, 32, 3, padding=1), + nn.SiLU(), + conv_nd(dims, 32, 64, 3, padding=1, stride=2), + nn.SiLU(), + ) + + self.fuse_block = zero_module(conv_nd(dims, 256+64+4, model_channels, 3, padding=1)) + + self._feature_size = model_channels + input_block_chans = [model_channels] + ch = model_channels + ds = 1 + for level, mult in enumerate(channel_mult): + for nr in range(self.num_res_blocks[level]): + layers = [ + ResBlock( + ch, + time_embed_dim, + dropout, + out_channels=mult * model_channels, + dims=dims, + use_checkpoint=use_checkpoint, + use_scale_shift_norm=use_scale_shift_norm, + ) + ] + ch = mult * model_channels + if ds in attention_resolutions: + if num_head_channels == -1: + dim_head = ch // num_heads + else: + num_heads = ch // num_head_channels + dim_head = num_head_channels + if legacy: + # num_heads = 1 + dim_head = ch // num_heads if use_spatial_transformer else num_head_channels + if exists(disable_self_attentions): + disabled_sa = disable_self_attentions[level] + else: + disabled_sa = False + + if not exists(num_attention_blocks) or nr < num_attention_blocks[level]: + layers.append( + AttentionBlock( + ch, + use_checkpoint=use_checkpoint, + num_heads=num_heads, + num_head_channels=dim_head, + use_new_attention_order=use_new_attention_order, + ) if not use_spatial_transformer else SpatialTransformer( + ch, num_heads, dim_head, depth=transformer_depth, context_dim=context_dim, + disable_self_attn=disabled_sa, use_linear=use_linear_in_transformer, + use_checkpoint=use_checkpoint + ) + ) + self.input_blocks.append(TimestepEmbedSequential(*layers)) + self.zero_convs.append(self.make_zero_conv(ch)) + self._feature_size += ch + input_block_chans.append(ch) + if level != len(channel_mult) - 1: + out_ch = ch + self.input_blocks.append( + TimestepEmbedSequential( + ResBlock( + ch, + time_embed_dim, + dropout, + out_channels=out_ch, + dims=dims, + use_checkpoint=use_checkpoint, + use_scale_shift_norm=use_scale_shift_norm, + down=True, + ) + if resblock_updown + else Downsample( + ch, conv_resample, dims=dims, out_channels=out_ch + ) + ) + ) + ch = out_ch + input_block_chans.append(ch) + self.zero_convs.append(self.make_zero_conv(ch)) + ds *= 2 + self._feature_size += ch + + if num_head_channels == -1: + dim_head = ch // num_heads + else: + num_heads = ch // num_head_channels + dim_head = num_head_channels + if legacy: + # num_heads = 1 + dim_head = ch // num_heads if use_spatial_transformer else num_head_channels + self.middle_block = TimestepEmbedSequential( + ResBlock( + ch, + time_embed_dim, + dropout, + dims=dims, + use_checkpoint=use_checkpoint, + use_scale_shift_norm=use_scale_shift_norm, + ), + AttentionBlock( + ch, + use_checkpoint=use_checkpoint, + num_heads=num_heads, + num_head_channels=dim_head, + use_new_attention_order=use_new_attention_order, + ) if not use_spatial_transformer else SpatialTransformer( # always uses a self-attn + ch, num_heads, dim_head, depth=transformer_depth, context_dim=context_dim, + disable_self_attn=disable_middle_self_attn, use_linear=use_linear_in_transformer, + use_checkpoint=use_checkpoint + ), + ResBlock( + ch, + time_embed_dim, + dropout, + dims=dims, + use_checkpoint=use_checkpoint, + use_scale_shift_norm=use_scale_shift_norm, + ), + ) + self.middle_block_out = self.make_zero_conv(ch) + self._feature_size += ch + + def make_zero_conv(self, channels): + return TimestepEmbedSequential(zero_module(conv_nd(self.dims, channels, channels, 1, padding=0))) + + def forward(self, x, hint, text_info, timesteps, context, **kwargs): + t_emb = timestep_embedding(timesteps, self.model_channels, repeat_only=False) + emb = self.time_embed(t_emb) + + # guided_hint from text_info + B, C, H, W = x.shape + glyphs = torch.cat(text_info['glyphs'], dim=1).sum(dim=1, keepdim=True) + positions = torch.cat(text_info['positions'], dim=1).sum(dim=1, keepdim=True) + enc_glyph = self.glyph_block(glyphs, emb, context) + enc_pos = self.position_block(positions, emb, context) + guided_hint = self.fuse_block(torch.cat([enc_glyph, enc_pos, text_info['masked_x']], dim=1)) + + outs = [] + + h = x.type(self.dtype) + for module, zero_conv in zip(self.input_blocks, self.zero_convs): + if guided_hint is not None: + h = module(h, emb, context) + h += guided_hint + guided_hint = None + else: + h = module(h, emb, context) + outs.append(zero_conv(h, emb, context)) + + h = self.middle_block(h, emb, context) + outs.append(self.middle_block_out(h, emb, context)) + + return outs + + +class ControlLDM(LatentDiffusion): + + def __init__(self, control_stage_config, control_key, glyph_key, position_key, only_mid_control, loss_alpha=0, loss_beta=0, with_step_weight=False, use_vae_upsample=False, latin_weight=1.0, embedding_manager_config=None, *args, **kwargs): + super().__init__(*args, **kwargs) + self.control_model = instantiate_from_config(control_stage_config) + self.control_key = control_key + self.glyph_key = glyph_key + self.position_key = position_key + self.only_mid_control = only_mid_control + self.control_scales = [1.0] * 13 + self.loss_alpha = loss_alpha + self.loss_beta = loss_beta + self.with_step_weight = with_step_weight + self.use_vae_upsample = use_vae_upsample + self.latin_weight = latin_weight + if embedding_manager_config is not None and embedding_manager_config.params.valid: + self.embedding_manager = self.instantiate_embedding_manager(embedding_manager_config, self.cond_stage_model) + for param in self.embedding_manager.embedding_parameters(): + param.requires_grad = True + else: + self.embedding_manager = None + if self.loss_alpha > 0 or self.loss_beta > 0 or self.embedding_manager: + if embedding_manager_config.params.emb_type == 'ocr': + self.text_predictor = create_predictor().eval() + args = edict() + args.rec_image_shape = "3, 48, 320" + args.rec_batch_num = 6 + args.rec_char_dict_path = './ocr_recog/ppocr_keys_v1.txt' + self.cn_recognizer = TextRecognizer(args, self.text_predictor) + for param in self.text_predictor.parameters(): + param.requires_grad = False + if self.embedding_manager: + self.embedding_manager.recog = self.cn_recognizer + + @torch.no_grad() + def get_input(self, batch, k, bs=None, *args, **kwargs): + if self.embedding_manager is None: # fill in full caption + self.fill_caption(batch) + x, c, mx = super().get_input(batch, self.first_stage_key, mask_k='masked_img', *args, **kwargs) + control = batch[self.control_key] # for log_images and loss_alpha, not real control + if bs is not None: + control = control[:bs] + control = control.to(self.device) + control = einops.rearrange(control, 'b h w c -> b c h w') + control = control.to(memory_format=torch.contiguous_format).float() + + inv_mask = batch['inv_mask'] + if bs is not None: + inv_mask = inv_mask[:bs] + inv_mask = inv_mask.to(self.device) + inv_mask = einops.rearrange(inv_mask, 'b h w c -> b c h w') + inv_mask = inv_mask.to(memory_format=torch.contiguous_format).float() + + glyphs = batch[self.glyph_key] + gly_line = batch['gly_line'] + positions = batch[self.position_key] + n_lines = batch['n_lines'] + language = batch['language'] + texts = batch['texts'] + assert len(glyphs) == len(positions) + for i in range(len(glyphs)): + if bs is not None: + glyphs[i] = glyphs[i][:bs] + gly_line[i] = gly_line[i][:bs] + positions[i] = positions[i][:bs] + n_lines = n_lines[:bs] + glyphs[i] = glyphs[i].to(self.device) + gly_line[i] = gly_line[i].to(self.device) + positions[i] = positions[i].to(self.device) + glyphs[i] = einops.rearrange(glyphs[i], 'b h w c -> b c h w') + gly_line[i] = einops.rearrange(gly_line[i], 'b h w c -> b c h w') + positions[i] = einops.rearrange(positions[i], 'b h w c -> b c h w') + glyphs[i] = glyphs[i].to(memory_format=torch.contiguous_format).float() + gly_line[i] = gly_line[i].to(memory_format=torch.contiguous_format).float() + positions[i] = positions[i].to(memory_format=torch.contiguous_format).float() + info = {} + info['glyphs'] = glyphs + info['positions'] = positions + info['n_lines'] = n_lines + info['language'] = language + info['texts'] = texts + info['img'] = batch['img'] # nhwc, (-1,1) + info['masked_x'] = mx + info['gly_line'] = gly_line + info['inv_mask'] = inv_mask + return x, dict(c_crossattn=[c], c_concat=[control], text_info=info) + + def apply_model(self, x_noisy, t, cond, *args, **kwargs): + assert isinstance(cond, dict) + diffusion_model = self.model.diffusion_model + _cond = torch.cat(cond['c_crossattn'], 1) + _hint = torch.cat(cond['c_concat'], 1) + control = self.control_model(x=x_noisy, timesteps=t, context=_cond, hint=_hint, text_info=cond['text_info']) + control = [c * scale for c, scale in zip(control, self.control_scales)] + eps = diffusion_model(x=x_noisy, timesteps=t, context=_cond, control=control, only_mid_control=self.only_mid_control) + + return eps + + def instantiate_embedding_manager(self, config, embedder): + model = instantiate_from_config(config, embedder=embedder) + return model + + @torch.no_grad() + def get_unconditional_conditioning(self, N): + return self.get_learned_conditioning(dict(c_crossattn=[[""] * N], text_info=None)) + + def get_learned_conditioning(self, c): + if self.cond_stage_forward is None: + if hasattr(self.cond_stage_model, 'encode') and callable(self.cond_stage_model.encode): + if self.embedding_manager is not None and c['text_info'] is not None: + self.embedding_manager.encode_text(c['text_info']) + if isinstance(c, dict): + cond_txt = c['c_crossattn'][0] + else: + cond_txt = c + if self.embedding_manager is not None: + cond_txt = self.cond_stage_model.encode(cond_txt, embedding_manager=self.embedding_manager) + else: + cond_txt = self.cond_stage_model.encode(cond_txt) + if isinstance(c, dict): + c['c_crossattn'][0] = cond_txt + else: + c = cond_txt + if isinstance(c, DiagonalGaussianDistribution): + c = c.mode() + else: + c = self.cond_stage_model(c) + else: + assert hasattr(self.cond_stage_model, self.cond_stage_forward) + c = getattr(self.cond_stage_model, self.cond_stage_forward)(c) + return c + + def fill_caption(self, batch, place_holder='*'): + bs = len(batch['n_lines']) + cond_list = copy.deepcopy(batch[self.cond_stage_key]) + for i in range(bs): + n_lines = batch['n_lines'][i] + if n_lines == 0: + continue + cur_cap = cond_list[i] + for j in range(n_lines): + r_txt = batch['texts'][j][i] + cur_cap = cur_cap.replace(place_holder, f'"{r_txt}"', 1) + cond_list[i] = cur_cap + batch[self.cond_stage_key] = cond_list + + @torch.no_grad() + def log_images(self, batch, N=4, n_row=2, sample=False, ddim_steps=50, ddim_eta=0.0, return_keys=None, + quantize_denoised=True, inpaint=True, plot_denoise_rows=False, plot_progressive_rows=True, + plot_diffusion_rows=False, unconditional_guidance_scale=9.0, unconditional_guidance_label=None, + use_ema_scope=True, + **kwargs): + use_ddim = ddim_steps is not None + + log = dict() + z, c = self.get_input(batch, self.first_stage_key, bs=N) + if self.cond_stage_trainable: + with torch.no_grad(): + c = self.get_learned_conditioning(c) + c_crossattn = c["c_crossattn"][0][:N] + c_cat = c["c_concat"][0][:N] + text_info = c["text_info"] + text_info['glyphs'] = [i[:N] for i in text_info['glyphs']] + text_info['gly_line'] = [i[:N] for i in text_info['gly_line']] + text_info['positions'] = [i[:N] for i in text_info['positions']] + text_info['n_lines'] = text_info['n_lines'][:N] + text_info['masked_x'] = text_info['masked_x'][:N] + text_info['img'] = text_info['img'][:N] + + N = min(z.shape[0], N) + n_row = min(z.shape[0], n_row) + log["reconstruction"] = self.decode_first_stage(z) + log["masked_image"] = self.decode_first_stage(text_info['masked_x']) + log["control"] = c_cat * 2.0 - 1.0 + log["img"] = text_info['img'].permute(0, 3, 1, 2) # log source image if needed + # get glyph + glyph_bs = torch.stack(text_info['glyphs']) + glyph_bs = torch.sum(glyph_bs, dim=0) * 2.0 - 1.0 + log["glyph"] = torch.nn.functional.interpolate(glyph_bs, size=(512, 512), mode='bilinear', align_corners=True,) + # fill caption + if not self.embedding_manager: + self.fill_caption(batch) + captions = batch[self.cond_stage_key] + log["conditioning"] = log_txt_as_img((512, 512), captions, size=16) + + if plot_diffusion_rows: + # get diffusion row + diffusion_row = list() + z_start = z[:n_row] + for t in range(self.num_timesteps): + if t % self.log_every_t == 0 or t == self.num_timesteps - 1: + t = repeat(torch.tensor([t]), '1 -> b', b=n_row) + t = t.to(self.device).long() + noise = torch.randn_like(z_start) + z_noisy = self.q_sample(x_start=z_start, t=t, noise=noise) + diffusion_row.append(self.decode_first_stage(z_noisy)) + + diffusion_row = torch.stack(diffusion_row) # n_log_step, n_row, C, H, W + diffusion_grid = rearrange(diffusion_row, 'n b c h w -> b n c h w') + diffusion_grid = rearrange(diffusion_grid, 'b n c h w -> (b n) c h w') + diffusion_grid = make_grid(diffusion_grid, nrow=diffusion_row.shape[0]) + log["diffusion_row"] = diffusion_grid + + if sample: + # get denoise row + samples, z_denoise_row = self.sample_log(cond={"c_concat": [c_cat], "c_crossattn": [c], "text_info": text_info}, + batch_size=N, ddim=use_ddim, + ddim_steps=ddim_steps, eta=ddim_eta) + x_samples = self.decode_first_stage(samples) + log["samples"] = x_samples + if plot_denoise_rows: + denoise_grid = self._get_denoise_row_from_list(z_denoise_row) + log["denoise_row"] = denoise_grid + + if unconditional_guidance_scale > 1.0: + uc_cross = self.get_unconditional_conditioning(N) + uc_cat = c_cat # torch.zeros_like(c_cat) + uc_full = {"c_concat": [uc_cat], "c_crossattn": [uc_cross['c_crossattn'][0]], "text_info": text_info} + samples_cfg, tmps = self.sample_log(cond={"c_concat": [c_cat], "c_crossattn": [c_crossattn], "text_info": text_info}, + batch_size=N, ddim=use_ddim, + ddim_steps=ddim_steps, eta=ddim_eta, + unconditional_guidance_scale=unconditional_guidance_scale, + unconditional_conditioning=uc_full, + ) + x_samples_cfg = self.decode_first_stage(samples_cfg) + log[f"samples_cfg_scale_{unconditional_guidance_scale:.2f}"] = x_samples_cfg + pred_x0 = False # wether log pred_x0 + if pred_x0: + for idx in range(len(tmps['pred_x0'])): + pred_x0 = self.decode_first_stage(tmps['pred_x0'][idx]) + log[f"pred_x0_{tmps['index'][idx]}"] = pred_x0 + + return log + + @torch.no_grad() + def sample_log(self, cond, batch_size, ddim, ddim_steps, **kwargs): + ddim_sampler = DDIMSampler(self) + b, c, h, w = cond["c_concat"][0].shape + shape = (self.channels, h // 8, w // 8) + samples, intermediates = ddim_sampler.sample(ddim_steps, batch_size, shape, cond, verbose=False, log_every_t=5, **kwargs) + return samples, intermediates + + def configure_optimizers(self): + lr = self.learning_rate + params = list(self.control_model.parameters()) + if self.embedding_manager: + params += list(self.embedding_manager.embedding_parameters()) + if not self.sd_locked: + # params += list(self.model.diffusion_model.input_blocks.parameters()) + # params += list(self.model.diffusion_model.middle_block.parameters()) + params += list(self.model.diffusion_model.output_blocks.parameters()) + params += list(self.model.diffusion_model.out.parameters()) + if self.unlockKV: + nCount = 0 + for name, param in self.model.diffusion_model.named_parameters(): + if 'attn2.to_k' in name or 'attn2.to_v' in name: + params += [param] + nCount += 1 + print(f'Cross attention is unlocked, and {nCount} Wk or Wv are added to potimizers!!!') + + opt = torch.optim.AdamW(params, lr=lr) + return opt + + def low_vram_shift(self, is_diffusing): + if is_diffusing: + self.model = self.model.cuda() + self.control_model = self.control_model.cuda() + self.first_stage_model = self.first_stage_model.cpu() + self.cond_stage_model = self.cond_stage_model.cpu() + else: + self.model = self.model.cpu() + self.control_model = self.control_model.cpu() + self.first_stage_model = self.first_stage_model.cuda() + self.cond_stage_model = self.cond_stage_model.cuda() diff --git a/cldm/ddim_hacked.py b/cldm/ddim_hacked.py new file mode 100644 index 0000000000000000000000000000000000000000..25b1bc947272ad14d7f7e5e4d1809005253b63d0 --- /dev/null +++ b/cldm/ddim_hacked.py @@ -0,0 +1,317 @@ +"""SAMPLING ONLY.""" + +import torch +import numpy as np +from tqdm import tqdm + +from ldm.modules.diffusionmodules.util import make_ddim_sampling_parameters, make_ddim_timesteps, noise_like, extract_into_tensor + + +class DDIMSampler(object): + def __init__(self, model, schedule="linear", **kwargs): + super().__init__() + self.model = model + self.ddpm_num_timesteps = model.num_timesteps + self.schedule = schedule + + def register_buffer(self, name, attr): + if type(attr) == torch.Tensor: + if attr.device != torch.device("cuda"): + attr = attr.to(torch.device("cuda")) + setattr(self, name, attr) + + def make_schedule(self, ddim_num_steps, ddim_discretize="uniform", ddim_eta=0., verbose=True): + self.ddim_timesteps = make_ddim_timesteps(ddim_discr_method=ddim_discretize, num_ddim_timesteps=ddim_num_steps, + num_ddpm_timesteps=self.ddpm_num_timesteps,verbose=verbose) + alphas_cumprod = self.model.alphas_cumprod + assert alphas_cumprod.shape[0] == self.ddpm_num_timesteps, 'alphas have to be defined for each timestep' + to_torch = lambda x: x.clone().detach().to(torch.float32).to(self.model.device) + + self.register_buffer('betas', to_torch(self.model.betas)) + self.register_buffer('alphas_cumprod', to_torch(alphas_cumprod)) + self.register_buffer('alphas_cumprod_prev', to_torch(self.model.alphas_cumprod_prev)) + + # calculations for diffusion q(x_t | x_{t-1}) and others + self.register_buffer('sqrt_alphas_cumprod', to_torch(np.sqrt(alphas_cumprod.cpu()))) + self.register_buffer('sqrt_one_minus_alphas_cumprod', to_torch(np.sqrt(1. - alphas_cumprod.cpu()))) + self.register_buffer('log_one_minus_alphas_cumprod', to_torch(np.log(1. - alphas_cumprod.cpu()))) + self.register_buffer('sqrt_recip_alphas_cumprod', to_torch(np.sqrt(1. / alphas_cumprod.cpu()))) + self.register_buffer('sqrt_recipm1_alphas_cumprod', to_torch(np.sqrt(1. / alphas_cumprod.cpu() - 1))) + + # ddim sampling parameters + ddim_sigmas, ddim_alphas, ddim_alphas_prev = make_ddim_sampling_parameters(alphacums=alphas_cumprod.cpu(), + ddim_timesteps=self.ddim_timesteps, + eta=ddim_eta,verbose=verbose) + self.register_buffer('ddim_sigmas', ddim_sigmas) + self.register_buffer('ddim_alphas', ddim_alphas) + self.register_buffer('ddim_alphas_prev', ddim_alphas_prev) + self.register_buffer('ddim_sqrt_one_minus_alphas', np.sqrt(1. - ddim_alphas)) + sigmas_for_original_sampling_steps = ddim_eta * torch.sqrt( + (1 - self.alphas_cumprod_prev) / (1 - self.alphas_cumprod) * ( + 1 - self.alphas_cumprod / self.alphas_cumprod_prev)) + self.register_buffer('ddim_sigmas_for_original_num_steps', sigmas_for_original_sampling_steps) + + @torch.no_grad() + def sample(self, + S, + batch_size, + shape, + conditioning=None, + callback=None, + normals_sequence=None, + img_callback=None, + quantize_x0=False, + eta=0., + mask=None, + x0=None, + temperature=1., + noise_dropout=0., + score_corrector=None, + corrector_kwargs=None, + verbose=True, + x_T=None, + log_every_t=100, + unconditional_guidance_scale=1., + unconditional_conditioning=None, # this has to come in the same format as the conditioning, # e.g. as encoded tokens, ... + dynamic_threshold=None, + ucg_schedule=None, + **kwargs + ): + if conditioning is not None: + if isinstance(conditioning, dict): + ctmp = conditioning[list(conditioning.keys())[0]] + while isinstance(ctmp, list): ctmp = ctmp[0] + cbs = ctmp.shape[0] + if cbs != batch_size: + print(f"Warning: Got {cbs} conditionings but batch-size is {batch_size}") + + elif isinstance(conditioning, list): + for ctmp in conditioning: + if ctmp.shape[0] != batch_size: + print(f"Warning: Got {cbs} conditionings but batch-size is {batch_size}") + + else: + if conditioning.shape[0] != batch_size: + print(f"Warning: Got {conditioning.shape[0]} conditionings but batch-size is {batch_size}") + + self.make_schedule(ddim_num_steps=S, ddim_eta=eta, verbose=verbose) + # sampling + C, H, W = shape + size = (batch_size, C, H, W) + print(f'Data shape for DDIM sampling is {size}, eta {eta}') + + samples, intermediates = self.ddim_sampling(conditioning, size, + callback=callback, + img_callback=img_callback, + quantize_denoised=quantize_x0, + mask=mask, x0=x0, + ddim_use_original_steps=False, + noise_dropout=noise_dropout, + temperature=temperature, + score_corrector=score_corrector, + corrector_kwargs=corrector_kwargs, + x_T=x_T, + log_every_t=log_every_t, + unconditional_guidance_scale=unconditional_guidance_scale, + unconditional_conditioning=unconditional_conditioning, + dynamic_threshold=dynamic_threshold, + ucg_schedule=ucg_schedule + ) + return samples, intermediates + + @torch.no_grad() + def ddim_sampling(self, cond, shape, + x_T=None, ddim_use_original_steps=False, + callback=None, timesteps=None, quantize_denoised=False, + mask=None, x0=None, img_callback=None, log_every_t=100, + temperature=1., noise_dropout=0., score_corrector=None, corrector_kwargs=None, + unconditional_guidance_scale=1., unconditional_conditioning=None, dynamic_threshold=None, + ucg_schedule=None): + device = self.model.betas.device + b = shape[0] + if x_T is None: + img = torch.randn(shape, device=device) + else: + img = x_T + + if timesteps is None: + timesteps = self.ddpm_num_timesteps if ddim_use_original_steps else self.ddim_timesteps + elif timesteps is not None and not ddim_use_original_steps: + subset_end = int(min(timesteps / self.ddim_timesteps.shape[0], 1) * self.ddim_timesteps.shape[0]) - 1 + timesteps = self.ddim_timesteps[:subset_end] + + intermediates = {'x_inter': [img], 'pred_x0': [img]} + time_range = reversed(range(0,timesteps)) if ddim_use_original_steps else np.flip(timesteps) + total_steps = timesteps if ddim_use_original_steps else timesteps.shape[0] + print(f"Running DDIM Sampling with {total_steps} timesteps") + + iterator = tqdm(time_range, desc='DDIM Sampler', total=total_steps) + + for i, step in enumerate(iterator): + index = total_steps - i - 1 + ts = torch.full((b,), step, device=device, dtype=torch.long) + + if mask is not None: + assert x0 is not None + img_orig = self.model.q_sample(x0, ts) # TODO: deterministic forward pass? + img = img_orig * mask + (1. - mask) * img + + if ucg_schedule is not None: + assert len(ucg_schedule) == len(time_range) + unconditional_guidance_scale = ucg_schedule[i] + + outs = self.p_sample_ddim(img, cond, ts, index=index, use_original_steps=ddim_use_original_steps, + quantize_denoised=quantize_denoised, temperature=temperature, + noise_dropout=noise_dropout, score_corrector=score_corrector, + corrector_kwargs=corrector_kwargs, + unconditional_guidance_scale=unconditional_guidance_scale, + unconditional_conditioning=unconditional_conditioning, + dynamic_threshold=dynamic_threshold) + img, pred_x0 = outs + if callback: callback(i) + if img_callback: img_callback(pred_x0, i) + + if index % log_every_t == 0 or index == total_steps - 1: + intermediates['x_inter'].append(img) + intermediates['pred_x0'].append(pred_x0) + + return img, intermediates + + @torch.no_grad() + def p_sample_ddim(self, x, c, t, index, repeat_noise=False, use_original_steps=False, quantize_denoised=False, + temperature=1., noise_dropout=0., score_corrector=None, corrector_kwargs=None, + unconditional_guidance_scale=1., unconditional_conditioning=None, + dynamic_threshold=None): + b, *_, device = *x.shape, x.device + + if unconditional_conditioning is None or unconditional_guidance_scale == 1.: + model_output = self.model.apply_model(x, t, c) + else: + model_t = self.model.apply_model(x, t, c) + model_uncond = self.model.apply_model(x, t, unconditional_conditioning) + model_output = model_uncond + unconditional_guidance_scale * (model_t - model_uncond) + + if self.model.parameterization == "v": + e_t = self.model.predict_eps_from_z_and_v(x, t, model_output) + else: + e_t = model_output + + if score_corrector is not None: + assert self.model.parameterization == "eps", 'not implemented' + e_t = score_corrector.modify_score(self.model, e_t, x, t, c, **corrector_kwargs) + + alphas = self.model.alphas_cumprod if use_original_steps else self.ddim_alphas + alphas_prev = self.model.alphas_cumprod_prev if use_original_steps else self.ddim_alphas_prev + sqrt_one_minus_alphas = self.model.sqrt_one_minus_alphas_cumprod if use_original_steps else self.ddim_sqrt_one_minus_alphas + sigmas = self.model.ddim_sigmas_for_original_num_steps if use_original_steps else self.ddim_sigmas + # select parameters corresponding to the currently considered timestep + a_t = torch.full((b, 1, 1, 1), alphas[index], device=device) + a_prev = torch.full((b, 1, 1, 1), alphas_prev[index], device=device) + sigma_t = torch.full((b, 1, 1, 1), sigmas[index], device=device) + sqrt_one_minus_at = torch.full((b, 1, 1, 1), sqrt_one_minus_alphas[index],device=device) + + # current prediction for x_0 + if self.model.parameterization != "v": + pred_x0 = (x - sqrt_one_minus_at * e_t) / a_t.sqrt() + else: + pred_x0 = self.model.predict_start_from_z_and_v(x, t, model_output) + + if quantize_denoised: + pred_x0, _, *_ = self.model.first_stage_model.quantize(pred_x0) + + if dynamic_threshold is not None: + raise NotImplementedError() + + # direction pointing to x_t + dir_xt = (1. - a_prev - sigma_t**2).sqrt() * e_t + noise = sigma_t * noise_like(x.shape, device, repeat_noise) * temperature + if noise_dropout > 0.: + noise = torch.nn.functional.dropout(noise, p=noise_dropout) + x_prev = a_prev.sqrt() * pred_x0 + dir_xt + noise + return x_prev, pred_x0 + + @torch.no_grad() + def encode(self, x0, c, t_enc, use_original_steps=False, return_intermediates=None, + unconditional_guidance_scale=1.0, unconditional_conditioning=None, callback=None): + timesteps = np.arange(self.ddpm_num_timesteps) if use_original_steps else self.ddim_timesteps + num_reference_steps = timesteps.shape[0] + + assert t_enc <= num_reference_steps + num_steps = t_enc + + if use_original_steps: + alphas_next = self.alphas_cumprod[:num_steps] + alphas = self.alphas_cumprod_prev[:num_steps] + else: + alphas_next = self.ddim_alphas[:num_steps] + alphas = torch.tensor(self.ddim_alphas_prev[:num_steps]) + + x_next = x0 + intermediates = [] + inter_steps = [] + for i in tqdm(range(num_steps), desc='Encoding Image'): + t = torch.full((x0.shape[0],), timesteps[i], device=self.model.device, dtype=torch.long) + if unconditional_guidance_scale == 1.: + noise_pred = self.model.apply_model(x_next, t, c) + else: + assert unconditional_conditioning is not None + e_t_uncond, noise_pred = torch.chunk( + self.model.apply_model(torch.cat((x_next, x_next)), torch.cat((t, t)), + torch.cat((unconditional_conditioning, c))), 2) + noise_pred = e_t_uncond + unconditional_guidance_scale * (noise_pred - e_t_uncond) + + xt_weighted = (alphas_next[i] / alphas[i]).sqrt() * x_next + weighted_noise_pred = alphas_next[i].sqrt() * ( + (1 / alphas_next[i] - 1).sqrt() - (1 / alphas[i] - 1).sqrt()) * noise_pred + x_next = xt_weighted + weighted_noise_pred + if return_intermediates and i % ( + num_steps // return_intermediates) == 0 and i < num_steps - 1: + intermediates.append(x_next) + inter_steps.append(i) + elif return_intermediates and i >= num_steps - 2: + intermediates.append(x_next) + inter_steps.append(i) + if callback: callback(i) + + out = {'x_encoded': x_next, 'intermediate_steps': inter_steps} + if return_intermediates: + out.update({'intermediates': intermediates}) + return x_next, out + + @torch.no_grad() + def stochastic_encode(self, x0, t, use_original_steps=False, noise=None): + # fast, but does not allow for exact reconstruction + # t serves as an index to gather the correct alphas + if use_original_steps: + sqrt_alphas_cumprod = self.sqrt_alphas_cumprod + sqrt_one_minus_alphas_cumprod = self.sqrt_one_minus_alphas_cumprod + else: + sqrt_alphas_cumprod = torch.sqrt(self.ddim_alphas) + sqrt_one_minus_alphas_cumprod = self.ddim_sqrt_one_minus_alphas + + if noise is None: + noise = torch.randn_like(x0) + return (extract_into_tensor(sqrt_alphas_cumprod, t, x0.shape) * x0 + + extract_into_tensor(sqrt_one_minus_alphas_cumprod, t, x0.shape) * noise) + + @torch.no_grad() + def decode(self, x_latent, cond, t_start, unconditional_guidance_scale=1.0, unconditional_conditioning=None, + use_original_steps=False, callback=None): + + timesteps = np.arange(self.ddpm_num_timesteps) if use_original_steps else self.ddim_timesteps + timesteps = timesteps[:t_start] + + time_range = np.flip(timesteps) + total_steps = timesteps.shape[0] + print(f"Running DDIM Sampling with {total_steps} timesteps") + + iterator = tqdm(time_range, desc='Decoding image', total=total_steps) + x_dec = x_latent + for i, step in enumerate(iterator): + index = total_steps - i - 1 + ts = torch.full((x_latent.shape[0],), step, device=x_latent.device, dtype=torch.long) + x_dec, _ = self.p_sample_ddim(x_dec, cond, ts, index=index, use_original_steps=use_original_steps, + unconditional_guidance_scale=unconditional_guidance_scale, + unconditional_conditioning=unconditional_conditioning) + if callback: callback(i) + return x_dec diff --git a/cldm/embedding_manager.py b/cldm/embedding_manager.py new file mode 100644 index 0000000000000000000000000000000000000000..be397d23d1db1d97d536d1bfdf0a2301807ad77e --- /dev/null +++ b/cldm/embedding_manager.py @@ -0,0 +1,165 @@ +''' +Copyright (c) Alibaba, Inc. and its affiliates. +''' +import torch +import torch.nn as nn +import torch.nn.functional as F +from functools import partial +from ldm.modules.diffusionmodules.util import conv_nd, linear + + +def get_clip_token_for_string(tokenizer, string): + batch_encoding = tokenizer(string, truncation=True, max_length=77, return_length=True, + return_overflowing_tokens=False, padding="max_length", return_tensors="pt") + tokens = batch_encoding["input_ids"] + assert torch.count_nonzero(tokens - 49407) == 2, f"String '{string}' maps to more than a single token. Please use another string" + return tokens[0, 1] + + +def get_bert_token_for_string(tokenizer, string): + token = tokenizer(string) + assert torch.count_nonzero(token) == 3, f"String '{string}' maps to more than a single token. Please use another string" + token = token[0, 1] + return token + + +def get_clip_vision_emb(encoder, processor, img): + _img = img.repeat(1, 3, 1, 1)*255 + inputs = processor(images=_img, return_tensors="pt") + inputs['pixel_values'] = inputs['pixel_values'].to(img.device) + outputs = encoder(**inputs) + emb = outputs.image_embeds + return emb + + +def get_recog_emb(encoder, img_list): + _img_list = [(img.repeat(1, 3, 1, 1)*255)[0] for img in img_list] + encoder.predictor.eval() + _, preds_neck = encoder.pred_imglist(_img_list, show_debug=False) + return preds_neck + + +def pad_H(x): + _, _, H, W = x.shape + p_top = (W - H) // 2 + p_bot = W - H - p_top + return F.pad(x, (0, 0, p_top, p_bot)) + + +class EncodeNet(nn.Module): + def __init__(self, in_channels, out_channels): + super(EncodeNet, self).__init__() + chan = 16 + n_layer = 4 # downsample + + self.conv1 = conv_nd(2, in_channels, chan, 3, padding=1) + self.conv_list = nn.ModuleList([]) + _c = chan + for i in range(n_layer): + self.conv_list.append(conv_nd(2, _c, _c*2, 3, padding=1, stride=2)) + _c *= 2 + self.conv2 = conv_nd(2, _c, out_channels, 3, padding=1) + self.avgpool = nn.AdaptiveAvgPool2d(1) + self.act = nn.SiLU() + + def forward(self, x): + x = self.act(self.conv1(x)) + for layer in self.conv_list: + x = self.act(layer(x)) + x = self.act(self.conv2(x)) + x = self.avgpool(x) + x = x.view(x.size(0), -1) + return x + + +class EmbeddingManager(nn.Module): + def __init__( + self, + embedder, + valid=True, + glyph_channels=20, + position_channels=1, + placeholder_string='*', + add_pos=False, + emb_type='ocr', + **kwargs + ): + super().__init__() + if hasattr(embedder, 'tokenizer'): # using Stable Diffusion's CLIP encoder + get_token_for_string = partial(get_clip_token_for_string, embedder.tokenizer) + token_dim = 768 + if hasattr(embedder, 'vit'): + assert emb_type == 'vit' + self.get_vision_emb = partial(get_clip_vision_emb, embedder.vit, embedder.processor) + self.get_recog_emb = None + else: # using LDM's BERT encoder + get_token_for_string = partial(get_bert_token_for_string, embedder.tknz_fn) + token_dim = 1280 + self.token_dim = token_dim + self.emb_type = emb_type + + self.add_pos = add_pos + if add_pos: + self.position_encoder = EncodeNet(position_channels, token_dim) + if emb_type == 'ocr': + self.proj = linear(40*64, token_dim) + if emb_type == 'conv': + self.glyph_encoder = EncodeNet(glyph_channels, token_dim) + + self.placeholder_token = get_token_for_string(placeholder_string) + + def encode_text(self, text_info): + if self.get_recog_emb is None and self.emb_type == 'ocr': + self.get_recog_emb = partial(get_recog_emb, self.recog) + + gline_list = [] + pos_list = [] + for i in range(len(text_info['n_lines'])): # sample index in a batch + n_lines = text_info['n_lines'][i] + for j in range(n_lines): # line + gline_list += [text_info['gly_line'][j][i:i+1]] + if self.add_pos: + pos_list += [text_info['positions'][j][i:i+1]] + + if len(gline_list) > 0: + if self.emb_type == 'ocr': + recog_emb = self.get_recog_emb(gline_list) + enc_glyph = self.proj(recog_emb.reshape(recog_emb.shape[0], -1)) + elif self.emb_type == 'vit': + enc_glyph = self.get_vision_emb(pad_H(torch.cat(gline_list, dim=0))) + elif self.emb_type == 'conv': + enc_glyph = self.glyph_encoder(pad_H(torch.cat(gline_list, dim=0))) + if self.add_pos: + enc_pos = self.position_encoder(torch.cat(gline_list, dim=0)) + enc_glyph = enc_glyph+enc_pos + + self.text_embs_all = [] + n_idx = 0 + for i in range(len(text_info['n_lines'])): # sample index in a batch + n_lines = text_info['n_lines'][i] + text_embs = [] + for j in range(n_lines): # line + text_embs += [enc_glyph[n_idx:n_idx+1]] + n_idx += 1 + self.text_embs_all += [text_embs] + + def forward( + self, + tokenized_text, + embedded_text, + ): + b, device = tokenized_text.shape[0], tokenized_text.device + for i in range(b): + idx = tokenized_text[i] == self.placeholder_token.to(device) + if sum(idx) > 0: + if i >= len(self.text_embs_all): + print('truncation for log images...') + break + text_emb = torch.cat(self.text_embs_all[i], dim=0) + if sum(idx) != len(text_emb): + print('truncation for long caption...') + embedded_text[i][idx] = text_emb[:sum(idx)] + return embedded_text + + def embedding_parameters(self): + return self.parameters() diff --git a/cldm/hack.py b/cldm/hack.py new file mode 100644 index 0000000000000000000000000000000000000000..454361e9d036cd1a6a79122c2fd16b489e4767b1 --- /dev/null +++ b/cldm/hack.py @@ -0,0 +1,111 @@ +import torch +import einops + +import ldm.modules.encoders.modules +import ldm.modules.attention + +from transformers import logging +from ldm.modules.attention import default + + +def disable_verbosity(): + logging.set_verbosity_error() + print('logging improved.') + return + + +def enable_sliced_attention(): + ldm.modules.attention.CrossAttention.forward = _hacked_sliced_attentin_forward + print('Enabled sliced_attention.') + return + + +def hack_everything(clip_skip=0): + disable_verbosity() + ldm.modules.encoders.modules.FrozenCLIPEmbedder.forward = _hacked_clip_forward + ldm.modules.encoders.modules.FrozenCLIPEmbedder.clip_skip = clip_skip + print('Enabled clip hacks.') + return + + +# Written by Lvmin +def _hacked_clip_forward(self, text): + PAD = self.tokenizer.pad_token_id + EOS = self.tokenizer.eos_token_id + BOS = self.tokenizer.bos_token_id + + def tokenize(t): + return self.tokenizer(t, truncation=False, add_special_tokens=False)["input_ids"] + + def transformer_encode(t): + if self.clip_skip > 1: + rt = self.transformer(input_ids=t, output_hidden_states=True) + return self.transformer.text_model.final_layer_norm(rt.hidden_states[-self.clip_skip]) + else: + return self.transformer(input_ids=t, output_hidden_states=False).last_hidden_state + + def split(x): + return x[75 * 0: 75 * 1], x[75 * 1: 75 * 2], x[75 * 2: 75 * 3] + + def pad(x, p, i): + return x[:i] if len(x) >= i else x + [p] * (i - len(x)) + + raw_tokens_list = tokenize(text) + tokens_list = [] + + for raw_tokens in raw_tokens_list: + raw_tokens_123 = split(raw_tokens) + raw_tokens_123 = [[BOS] + raw_tokens_i + [EOS] for raw_tokens_i in raw_tokens_123] + raw_tokens_123 = [pad(raw_tokens_i, PAD, 77) for raw_tokens_i in raw_tokens_123] + tokens_list.append(raw_tokens_123) + + tokens_list = torch.IntTensor(tokens_list).to(self.device) + + feed = einops.rearrange(tokens_list, 'b f i -> (b f) i') + y = transformer_encode(feed) + z = einops.rearrange(y, '(b f) i c -> b (f i) c', f=3) + + return z + + +# Stolen from https://github.com/basujindal/stable-diffusion/blob/main/optimizedSD/splitAttention.py +def _hacked_sliced_attentin_forward(self, x, context=None, mask=None): + h = self.heads + + q = self.to_q(x) + context = default(context, x) + k = self.to_k(context) + v = self.to_v(context) + del context, x + + q, k, v = map(lambda t: einops.rearrange(t, 'b n (h d) -> (b h) n d', h=h), (q, k, v)) + + limit = k.shape[0] + att_step = 1 + q_chunks = list(torch.tensor_split(q, limit // att_step, dim=0)) + k_chunks = list(torch.tensor_split(k, limit // att_step, dim=0)) + v_chunks = list(torch.tensor_split(v, limit // att_step, dim=0)) + + q_chunks.reverse() + k_chunks.reverse() + v_chunks.reverse() + sim = torch.zeros(q.shape[0], q.shape[1], v.shape[2], device=q.device) + del k, q, v + for i in range(0, limit, att_step): + q_buffer = q_chunks.pop() + k_buffer = k_chunks.pop() + v_buffer = v_chunks.pop() + sim_buffer = torch.einsum('b i d, b j d -> b i j', q_buffer, k_buffer) * self.scale + + del k_buffer, q_buffer + # attention, what we cannot get enough of, by chunks + + sim_buffer = sim_buffer.softmax(dim=-1) + + sim_buffer = torch.einsum('b i j, b j d -> b i d', sim_buffer, v_buffer) + del v_buffer + sim[i:i + att_step, :, :] = sim_buffer + + del sim_buffer + sim = einops.rearrange(sim, '(b h) n d -> b n (h d)', h=h) + return self.to_out(sim) diff --git a/cldm/logger.py b/cldm/logger.py new file mode 100644 index 0000000000000000000000000000000000000000..6a8803846f2a8979f87f3cf9ea5b12869439e62f --- /dev/null +++ b/cldm/logger.py @@ -0,0 +1,76 @@ +import os + +import numpy as np +import torch +import torchvision +from PIL import Image +from pytorch_lightning.callbacks import Callback +from pytorch_lightning.utilities.distributed import rank_zero_only + + +class ImageLogger(Callback): + def __init__(self, batch_frequency=2000, max_images=4, clamp=True, increase_log_steps=True, + rescale=True, disabled=False, log_on_batch_idx=False, log_first_step=False, + log_images_kwargs=None): + super().__init__() + self.rescale = rescale + self.batch_freq = batch_frequency + self.max_images = max_images + if not increase_log_steps: + self.log_steps = [self.batch_freq] + self.clamp = clamp + self.disabled = disabled + self.log_on_batch_idx = log_on_batch_idx + self.log_images_kwargs = log_images_kwargs if log_images_kwargs else {} + self.log_first_step = log_first_step + + @rank_zero_only + def log_local(self, save_dir, split, images, global_step, current_epoch, batch_idx): + root = os.path.join(save_dir, "image_log", split) + for k in images: + grid = torchvision.utils.make_grid(images[k], nrow=4) + if self.rescale: + grid = (grid + 1.0) / 2.0 # -1,1 -> 0,1; c,h,w + grid = grid.transpose(0, 1).transpose(1, 2).squeeze(-1) + grid = grid.numpy() + grid = (grid * 255).astype(np.uint8) + filename = "{}_gs-{:06}_e-{:06}_b-{:06}.png".format(k, global_step, current_epoch, batch_idx) + path = os.path.join(root, filename) + os.makedirs(os.path.split(path)[0], exist_ok=True) + Image.fromarray(grid).save(path) + + def log_img(self, pl_module, batch, batch_idx, split="train"): + check_idx = batch_idx # if self.log_on_batch_idx else pl_module.global_step + if (self.check_frequency(check_idx) and # batch_idx % self.batch_freq == 0 + hasattr(pl_module, "log_images") and + callable(pl_module.log_images) and + self.max_images > 0): + logger = type(pl_module.logger) + + is_train = pl_module.training + if is_train: + pl_module.eval() + + with torch.no_grad(): + images = pl_module.log_images(batch, split=split, **self.log_images_kwargs) + + for k in images: + N = min(images[k].shape[0], self.max_images) + images[k] = images[k][:N] + if isinstance(images[k], torch.Tensor): + images[k] = images[k].detach().cpu() + if self.clamp: + images[k] = torch.clamp(images[k], -1., 1.) + + self.log_local(pl_module.logger.save_dir, split, images, + pl_module.global_step, pl_module.current_epoch, batch_idx) + + if is_train: + pl_module.train() + + def check_frequency(self, check_idx): + return check_idx % self.batch_freq == 0 + + def on_train_batch_end(self, trainer, pl_module, outputs, batch, batch_idx, dataloader_idx): + if not self.disabled: + self.log_img(pl_module, batch, batch_idx, split="train") diff --git a/cldm/model.py b/cldm/model.py new file mode 100644 index 0000000000000000000000000000000000000000..b4404622322e05fd8fea4ba0a15912b88c3b2e2a --- /dev/null +++ b/cldm/model.py @@ -0,0 +1,30 @@ +import os +import torch + +from omegaconf import OmegaConf +from ldm.util import instantiate_from_config + + +def get_state_dict(d): + return d.get('state_dict', d) + + +def load_state_dict(ckpt_path, location='cpu'): + _, extension = os.path.splitext(ckpt_path) + if extension.lower() == ".safetensors": + import safetensors.torch + state_dict = safetensors.torch.load_file(ckpt_path, device=location) + else: + state_dict = get_state_dict(torch.load(ckpt_path, map_location=torch.device(location))) + state_dict = get_state_dict(state_dict) + print(f'Loaded state_dict from [{ckpt_path}]') + return state_dict + + +def create_model(config_path, cond_stage_path=None): + config = OmegaConf.load(config_path) + if cond_stage_path: + config.model.params.cond_stage_config.params.version = cond_stage_path # use pre-downloaded ckpts, in case blocked + model = instantiate_from_config(config.model).cpu() + print(f'Loaded model config from [{config_path}]') + return model diff --git a/cldm/recognizer.py b/cldm/recognizer.py new file mode 100755 index 0000000000000000000000000000000000000000..2defa9117b73bd741ef2bdaecb320a33f58e6933 --- /dev/null +++ b/cldm/recognizer.py @@ -0,0 +1,303 @@ +''' +Copyright (c) Alibaba, Inc. and its affiliates. +''' +import os +import sys +sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__), '..'))) +import cv2 +import numpy as np +import math +import traceback +from easydict import EasyDict as edict +import time +from ocr_recog.RecModel import RecModel +import torch +import torch.nn.functional as F +from skimage.transform._geometric import _umeyama as get_sym_mat + + +def min_bounding_rect(img): + ret, thresh = cv2.threshold(img, 127, 255, 0) + contours, hierarchy = cv2.findContours(thresh, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE) + if len(contours) == 0: + print('Bad contours, using fake bbox...') + return np.array([[0, 0], [100, 0], [100, 100], [0, 100]]) + max_contour = max(contours, key=cv2.contourArea) + rect = cv2.minAreaRect(max_contour) + box = cv2.boxPoints(rect) + box = np.int0(box) + # sort + x_sorted = sorted(box, key=lambda x: x[0]) + left = x_sorted[:2] + right = x_sorted[2:] + left = sorted(left, key=lambda x: x[1]) + (tl, bl) = left + right = sorted(right, key=lambda x: x[1]) + (tr, br) = right + if tl[1] > bl[1]: + (tl, bl) = (bl, tl) + if tr[1] > br[1]: + (tr, br) = (br, tr) + return np.array([tl, tr, br, bl]) + + +def adjust_image(box, img): + pts1 = np.float32([box[0], box[1], box[2], box[3]]) + width = max(np.linalg.norm(pts1[0]-pts1[1]), np.linalg.norm(pts1[2]-pts1[3])) + height = max(np.linalg.norm(pts1[0]-pts1[3]), np.linalg.norm(pts1[1]-pts1[2])) + pts2 = np.float32([[0, 0], [width, 0], [width, height], [0, height]]) + # get transform matrix + M = get_sym_mat(pts1, pts2, estimate_scale=True) + C, H, W = img.shape + T = np.array([[2 / W, 0, -1], [0, 2 / H, -1], [0, 0, 1]]) + theta = np.linalg.inv(T @ M @ np.linalg.inv(T)) + theta = torch.from_numpy(theta[:2, :]).unsqueeze(0).type(torch.float32).to(img.device) + grid = F.affine_grid(theta, torch.Size([1, C, H, W]), align_corners=True) + result = F.grid_sample(img.unsqueeze(0), grid, align_corners=True) + result = torch.clamp(result.squeeze(0), 0, 255) + # crop + result = result[:, :int(height), :int(width)] + return result + + +''' +mask: numpy.ndarray, mask of textual, HWC +src_img: torch.Tensor, source image, CHW +''' +def crop_image(src_img, mask): + box = min_bounding_rect(mask) + result = adjust_image(box, src_img) + if len(result.shape) == 2: + result = torch.stack([result]*3, axis=-1) + return result + + +def create_predictor(model_dir=None, model_lang='ch', is_onnx=False): + model_file_path = model_dir + if model_file_path is not None and not os.path.exists(model_file_path): + raise ValueError("not find model file path {}".format(model_file_path)) + + if is_onnx: + import onnxruntime as ort + sess = ort.InferenceSession(model_file_path, providers=['CPUExecutionProvider']) # 'TensorrtExecutionProvider', 'CUDAExecutionProvider', 'CPUExecutionProvider' + return sess + else: + if model_lang == 'ch': + n_class = 6625 + elif model_lang == 'en': + n_class = 97 + else: + raise ValueError(f"Unsupported OCR recog model_lang: {model_lang}") + rec_config = edict( + in_channels=3, + backbone=edict(type='MobileNetV1Enhance', scale=0.5, last_conv_stride=[1, 2], last_pool_type='avg'), + neck=edict(type='SequenceEncoder', encoder_type="svtr", dims=64, depth=2, hidden_dims=120, use_guide=True), + head=edict(type='CTCHead', fc_decay=0.00001, out_channels=n_class, return_feats=True) + ) + + rec_model = RecModel(rec_config) + if model_file_path is not None: + rec_model.load_state_dict(torch.load(model_file_path, map_location="cpu")) + rec_model.eval() + return rec_model.eval() + + +def _check_image_file(path): + img_end = {'jpg', 'bmp', 'png', 'jpeg', 'rgb', 'tif', 'tiff'} + return any([path.lower().endswith(e) for e in img_end]) + + +def get_image_file_list(img_file): + imgs_lists = [] + if img_file is None or not os.path.exists(img_file): + raise Exception("not found any img file in {}".format(img_file)) + if os.path.isfile(img_file) and _check_image_file(img_file): + imgs_lists.append(img_file) + elif os.path.isdir(img_file): + for single_file in os.listdir(img_file): + file_path = os.path.join(img_file, single_file) + if os.path.isfile(file_path) and _check_image_file(file_path): + imgs_lists.append(file_path) + if len(imgs_lists) == 0: + raise Exception("not found any img file in {}".format(img_file)) + imgs_lists = sorted(imgs_lists) + return imgs_lists + + +class TextRecognizer(object): + def __init__(self, args, predictor): + self.rec_image_shape = [int(v) for v in args.rec_image_shape.split(",")] + self.rec_batch_num = args.rec_batch_num + self.predictor = predictor + self.chars = self.get_char_dict(args.rec_char_dict_path) + self.char2id = {x: i for i, x in enumerate(self.chars)} + self.is_onnx = not isinstance(self.predictor, torch.nn.Module) + + # img: CHW + def resize_norm_img(self, img, max_wh_ratio): + imgC, imgH, imgW = self.rec_image_shape + assert imgC == img.shape[0] + imgW = int((imgH * max_wh_ratio)) + + h, w = img.shape[1:] + ratio = w / float(h) + if math.ceil(imgH * ratio) > imgW: + resized_w = imgW + else: + resized_w = int(math.ceil(imgH * ratio)) + resized_image = torch.nn.functional.interpolate( + img.unsqueeze(0), + size=(imgH, resized_w), + mode='bilinear', + align_corners=True, + ) + resized_image /= 255.0 + resized_image -= 0.5 + resized_image /= 0.5 + padding_im = torch.zeros((imgC, imgH, imgW), dtype=torch.float32).to(img.device) + padding_im[:, :, 0:resized_w] = resized_image[0] + return padding_im + + # img_list: list of tensors with shape chw 0-255 + def pred_imglist(self, img_list, show_debug=False, is_ori=False): + img_num = len(img_list) + assert img_num > 0 + # Calculate the aspect ratio of all text bars + width_list = [] + for img in img_list: + width_list.append(img.shape[2] / float(img.shape[1])) + # Sorting can speed up the recognition process + indices = torch.from_numpy(np.argsort(np.array(width_list))) + batch_num = self.rec_batch_num + preds_all = [None] * img_num + preds_neck_all = [None] * img_num + for beg_img_no in range(0, img_num, batch_num): + end_img_no = min(img_num, beg_img_no + batch_num) + norm_img_batch = [] + + imgC, imgH, imgW = self.rec_image_shape[:3] + max_wh_ratio = imgW / imgH + for ino in range(beg_img_no, end_img_no): + h, w = img_list[indices[ino]].shape[1:] + if h > w * 1.2: + img = img_list[indices[ino]] + img = torch.transpose(img, 1, 2).flip(dims=[1]) + img_list[indices[ino]] = img + h, w = img.shape[1:] + # wh_ratio = w * 1.0 / h + # max_wh_ratio = max(max_wh_ratio, wh_ratio) # comment to not use different ratio + for ino in range(beg_img_no, end_img_no): + norm_img = self.resize_norm_img(img_list[indices[ino]], max_wh_ratio) + norm_img = norm_img.unsqueeze(0) + norm_img_batch.append(norm_img) + norm_img_batch = torch.cat(norm_img_batch, dim=0) + if show_debug: + for i in range(len(norm_img_batch)): + _img = norm_img_batch[i].permute(1, 2, 0).detach().cpu().numpy() + _img = (_img + 0.5)*255 + _img = _img[:, :, ::-1] + file_name = f'{indices[beg_img_no + i]}' + file_name = file_name + '_ori' if is_ori else file_name + cv2.imwrite(file_name + '.jpg', _img) + if self.is_onnx: + input_dict = {} + input_dict[self.predictor.get_inputs()[0].name] = norm_img_batch.detach().cpu().numpy() + outputs = self.predictor.run(None, input_dict) + preds = {} + preds['ctc'] = torch.from_numpy(outputs[0]) + preds['ctc_neck'] = [torch.zeros(1)] * img_num + else: + preds = self.predictor(norm_img_batch) + for rno in range(preds['ctc'].shape[0]): + preds_all[indices[beg_img_no + rno]] = preds['ctc'][rno] + preds_neck_all[indices[beg_img_no + rno]] = preds['ctc_neck'][rno] + + return torch.stack(preds_all, dim=0), torch.stack(preds_neck_all, dim=0) + + def get_char_dict(self, character_dict_path): + character_str = [] + with open(character_dict_path, "rb") as fin: + lines = fin.readlines() + for line in lines: + line = line.decode('utf-8').strip("\n").strip("\r\n") + character_str.append(line) + dict_character = list(character_str) + dict_character = ['sos'] + dict_character + [' '] # eos is space + return dict_character + + def get_text(self, order): + char_list = [self.chars[text_id] for text_id in order] + return ''.join(char_list) + + def decode(self, mat): + text_index = mat.detach().cpu().numpy().argmax(axis=1) + ignored_tokens = [0] + selection = np.ones(len(text_index), dtype=bool) + selection[1:] = text_index[1:] != text_index[:-1] + for ignored_token in ignored_tokens: + selection &= text_index != ignored_token + return text_index[selection], np.where(selection)[0] + + def get_ctcloss(self, preds, gt_text, weight): + if not isinstance(weight, torch.Tensor): + weight = torch.tensor(weight).to(preds.device) + ctc_loss = torch.nn.CTCLoss(reduction='none') + log_probs = preds.log_softmax(dim=2).permute(1, 0, 2) # NTC-->TNC + targets = [] + target_lengths = [] + for t in gt_text: + targets += [self.char2id.get(i, len(self.chars)-1) for i in t] + target_lengths += [len(t)] + targets = torch.tensor(targets).to(preds.device) + target_lengths = torch.tensor(target_lengths).to(preds.device) + input_lengths = torch.tensor([log_probs.shape[0]]*(log_probs.shape[1])).to(preds.device) + loss = ctc_loss(log_probs, targets, input_lengths, target_lengths) + loss = loss / input_lengths * weight + return loss + + +def main(): + rec_model_dir = "./ocr_weights/ppv3_rec.pth" + predictor = create_predictor(rec_model_dir) + args = edict() + args.rec_image_shape = "3, 48, 320" + args.rec_char_dict_path = './ocr_weights/ppocr_keys_v1.txt' + args.rec_batch_num = 6 + text_recognizer = TextRecognizer(args, predictor) + image_dir = './test_imgs_cn' + gt_text = ['韩国小馆']*14 + + image_file_list = get_image_file_list(image_dir) + valid_image_file_list = [] + img_list = [] + + for image_file in image_file_list: + img = cv2.imread(image_file) + if img is None: + print("error in loading image:{}".format(image_file)) + continue + valid_image_file_list.append(image_file) + img_list.append(torch.from_numpy(img).permute(2, 0, 1).float()) + try: + tic = time.time() + times = [] + for i in range(10): + preds, _ = text_recognizer.pred_imglist(img_list) # get text + preds_all = preds.softmax(dim=2) + times += [(time.time()-tic)*1000.] + tic = time.time() + print(times) + print(np.mean(times[1:]) / len(preds_all)) + weight = np.ones(len(gt_text)) + loss = text_recognizer.get_ctcloss(preds, gt_text, weight) + for i in range(len(valid_image_file_list)): + pred = preds_all[i] + order, idx = text_recognizer.decode(pred) + text = text_recognizer.get_text(order) + print(f'{valid_image_file_list[i]}: pred/gt="{text}"/"{gt_text[i]}", loss={loss[i]:.2f}') + except Exception as E: + print(traceback.format_exc(), E) + + +if __name__ == "__main__": + main() diff --git a/dataset_util.py b/dataset_util.py new file mode 100644 index 0000000000000000000000000000000000000000..f1e44640e88d6bf99ffbaf81051a0677bf9c11ba --- /dev/null +++ b/dataset_util.py @@ -0,0 +1,77 @@ + +import json +import pathlib + +__all__ = ['load', 'save', 'show_bbox_on_image'] + + +def load(file_path: str): + file_path = pathlib.Path(file_path) + func_dict = {'.txt': load_txt, '.json': load_json, '.list': load_txt} + assert file_path.suffix in func_dict + return func_dict[file_path.suffix](file_path) + + +def load_txt(file_path: str): + with open(file_path, 'r', encoding='utf8') as f: + content = [x.strip().strip('\ufeff').strip('\xef\xbb\xbf') for x in f.readlines()] + return content + + +def load_json(file_path: str): + with open(file_path, 'r', encoding='utf8') as f: + content = json.load(f) + return content + + +def save(data, file_path): + file_path = pathlib.Path(file_path) + func_dict = {'.txt': save_txt, '.json': save_json} + assert file_path.suffix in func_dict + return func_dict[file_path.suffix](data, file_path) + + +def save_txt(data, file_path): + if not isinstance(data, list): + data = [data] + with open(file_path, mode='w', encoding='utf8') as f: + f.write('\n'.join(data)) + + +def save_json(data, file_path): + with open(file_path, 'w', encoding='utf-8') as json_file: + json.dump(data, json_file, ensure_ascii=False, indent=4) + + +def show_bbox_on_image(image, polygons=None, txt=None, color=None, font_path='./font/Arial_Unicode.ttf'): + from PIL import ImageDraw, ImageFont + image = image.convert('RGB') + draw = ImageDraw.Draw(image) + if len(txt) == 0: + txt = None + if color is None: + color = (255, 0, 0) + if txt is not None: + font = ImageFont.truetype(font_path, 20) + for i, box in enumerate(polygons): + box = box[0] + if txt is not None: + draw.text((int(box[0][0]) + 20, int(box[0][1]) - 20), str(txt[i]), fill='red', font=font) + for j in range(len(box) - 1): + draw.line((box[j][0], box[j][1], box[j + 1][0], box[j + 1][1]), fill=color, width=2) + draw.line((box[-1][0], box[-1][1], box[0][0], box[0][1]), fill=color, width=2) + return image + + +def show_glyphs(glyphs, name): + import numpy as np + import cv2 + size = 64 + gap = 5 + n_char = 20 + canvas = np.ones((size, size*n_char + gap*(n_char-1), 1))*0.5 + x = 0 + for i in range(glyphs.shape[-1]): + canvas[:, x:x + size, :] = glyphs[..., i:i+1] + x += size+gap + cv2.imwrite(name, canvas*255) diff --git a/example_images/banner.png b/example_images/banner.png new file mode 100644 index 0000000000000000000000000000000000000000..4dc7f1ac0828b1b97ac49598c694bfe94d0bd0c6 Binary files /dev/null and b/example_images/banner.png differ diff --git a/example_images/edit1.png b/example_images/edit1.png new file mode 100644 index 0000000000000000000000000000000000000000..e9eb8bd3fe5c124b73850d8f559d1bfad559f78e Binary files /dev/null and b/example_images/edit1.png differ diff --git a/example_images/edit10.png b/example_images/edit10.png new file mode 100644 index 0000000000000000000000000000000000000000..2b953032224d6b466c4225c14b91e48cc8cb92dc Binary files /dev/null and b/example_images/edit10.png differ diff --git a/example_images/edit11.png b/example_images/edit11.png new file mode 100644 index 0000000000000000000000000000000000000000..b0dc5d3c961e80eb865c82456bc21070198aef0b Binary files /dev/null and b/example_images/edit11.png differ diff --git a/example_images/edit12.png b/example_images/edit12.png new file mode 100644 index 0000000000000000000000000000000000000000..2e2b417c9cad122e7f12f39655e1e4c5443849f2 Binary files /dev/null and b/example_images/edit12.png differ diff --git a/example_images/edit13.png b/example_images/edit13.png new file mode 100644 index 0000000000000000000000000000000000000000..77840c4cf7801c7b231ad743bece6ff7f9eaeb8b Binary files /dev/null and b/example_images/edit13.png differ diff --git a/example_images/edit14.png b/example_images/edit14.png new file mode 100644 index 0000000000000000000000000000000000000000..c823c41380a9c28330ffe555bdab48111749e387 Binary files /dev/null and b/example_images/edit14.png differ diff --git a/example_images/edit2.png b/example_images/edit2.png new file mode 100644 index 0000000000000000000000000000000000000000..19bc46f6db4f98c650db038009717702872c2cb3 Binary files /dev/null and b/example_images/edit2.png differ diff --git a/example_images/edit3.png b/example_images/edit3.png new file mode 100644 index 0000000000000000000000000000000000000000..343c0dbfff29da58ceab148f907bf30e5fe49e63 Binary files /dev/null and b/example_images/edit3.png differ diff --git a/example_images/edit4.png b/example_images/edit4.png new file mode 100644 index 0000000000000000000000000000000000000000..f6cff25d2383f08eb548927d74529a64bcfc221c Binary files /dev/null and b/example_images/edit4.png differ diff --git a/example_images/edit5.png b/example_images/edit5.png new file mode 100644 index 0000000000000000000000000000000000000000..8d7b41d87a04b391d69d8dbcdc66f8554c46f551 Binary files /dev/null and b/example_images/edit5.png differ diff --git a/example_images/edit6.png b/example_images/edit6.png new file mode 100644 index 0000000000000000000000000000000000000000..6eedce647acf5fdfcf1bff80d4c91d95a003fded Binary files /dev/null and b/example_images/edit6.png differ diff --git a/example_images/edit7.png b/example_images/edit7.png new file mode 100644 index 0000000000000000000000000000000000000000..5390d6fd193f47937ba58df0b9bb41344e4df321 Binary files /dev/null and b/example_images/edit7.png differ diff --git a/example_images/edit8.png b/example_images/edit8.png new file mode 100644 index 0000000000000000000000000000000000000000..036c1d27f326fac76beb64044591a4143b845e22 Binary files /dev/null and b/example_images/edit8.png differ diff --git a/example_images/edit9.png b/example_images/edit9.png new file mode 100644 index 0000000000000000000000000000000000000000..7010574e488c3dea76d1d4e82fbee1d5fbbafd06 Binary files /dev/null and b/example_images/edit9.png differ diff --git a/example_images/gen1.png b/example_images/gen1.png new file mode 100644 index 0000000000000000000000000000000000000000..e0ca934937e6785f5ac19667a6cee5b82ace3dea Binary files /dev/null and b/example_images/gen1.png differ diff --git a/example_images/gen10.png b/example_images/gen10.png new file mode 100644 index 0000000000000000000000000000000000000000..13a440cc173de1289f195c98037e20bce1ded951 Binary files /dev/null and b/example_images/gen10.png differ diff --git a/example_images/gen11.png b/example_images/gen11.png new file mode 100644 index 0000000000000000000000000000000000000000..fbb56162d4b2862d50c8401883ddc7965f9d0b9c Binary files /dev/null and b/example_images/gen11.png differ diff --git a/example_images/gen12.png b/example_images/gen12.png new file mode 100644 index 0000000000000000000000000000000000000000..b224a4638c16f6316f40bcbe7f2e59c3167ed5c9 Binary files /dev/null and b/example_images/gen12.png differ diff --git a/example_images/gen13.png b/example_images/gen13.png new file mode 100644 index 0000000000000000000000000000000000000000..b99f87a2fbc368097d66440a52cda01792778396 Binary files /dev/null and b/example_images/gen13.png differ diff --git a/example_images/gen14.png b/example_images/gen14.png new file mode 100644 index 0000000000000000000000000000000000000000..50695245ce9f9e55646b018f45ee29a19cc99b20 Binary files /dev/null and b/example_images/gen14.png differ diff --git a/example_images/gen15.png b/example_images/gen15.png new file mode 100644 index 0000000000000000000000000000000000000000..cf096e1c30ce81b783b0827f2706611b95be34d3 Binary files /dev/null and b/example_images/gen15.png differ diff --git a/example_images/gen16.png b/example_images/gen16.png new file mode 100644 index 0000000000000000000000000000000000000000..77f06a6eab3cef917160164b87d229fe49ec5073 Binary files /dev/null and b/example_images/gen16.png differ diff --git a/example_images/gen2.png b/example_images/gen2.png new file mode 100644 index 0000000000000000000000000000000000000000..cc3285d63614e962c8d5702b8302ed25b8aed4ee Binary files /dev/null and b/example_images/gen2.png differ diff --git a/example_images/gen3.png b/example_images/gen3.png new file mode 100644 index 0000000000000000000000000000000000000000..fbdf379f40d4687221b20b98e8ba612877fd00a9 Binary files /dev/null and b/example_images/gen3.png differ diff --git a/example_images/gen4.png b/example_images/gen4.png new file mode 100644 index 0000000000000000000000000000000000000000..ffb94dfdc2c3b307d559a2c933e1f89a70f80b02 Binary files /dev/null and b/example_images/gen4.png differ diff --git a/example_images/gen5.png b/example_images/gen5.png new file mode 100644 index 0000000000000000000000000000000000000000..335f9bd2fe0a493d3e343d50d5765f4033a968de Binary files /dev/null and b/example_images/gen5.png differ diff --git a/example_images/gen6.png b/example_images/gen6.png new file mode 100644 index 0000000000000000000000000000000000000000..b6f03092edf35715294978973ea4d9b9b14cdf8e Binary files /dev/null and b/example_images/gen6.png differ diff --git a/example_images/gen7.png b/example_images/gen7.png new file mode 100644 index 0000000000000000000000000000000000000000..385d1dbe4243e66be718359f2616b99037a62376 Binary files /dev/null and b/example_images/gen7.png differ diff --git a/example_images/gen8.png b/example_images/gen8.png new file mode 100644 index 0000000000000000000000000000000000000000..4b7dcf6e4f586f860f45b25bdc4d1fb7ad1be465 Binary files /dev/null and b/example_images/gen8.png differ diff --git a/example_images/gen9.png b/example_images/gen9.png new file mode 100644 index 0000000000000000000000000000000000000000..a6c68dadc3d686d359cdb69a35a81aeecd10ccc1 Binary files /dev/null and b/example_images/gen9.png differ diff --git a/example_images/ref1.jpg b/example_images/ref1.jpg new file mode 100644 index 0000000000000000000000000000000000000000..e051af3cd9bcb77805d3e95ab9f9e88f65952773 Binary files /dev/null and b/example_images/ref1.jpg differ diff --git a/example_images/ref10.jpg b/example_images/ref10.jpg new file mode 100644 index 0000000000000000000000000000000000000000..5515ac2cc041a257cf7ae2944182eadecc162ec7 Binary files /dev/null and b/example_images/ref10.jpg differ diff --git a/example_images/ref11.jpg b/example_images/ref11.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f44a8fe449fd532b84d519a09734f5f76c9709e9 Binary files /dev/null and b/example_images/ref11.jpg differ diff --git a/example_images/ref12.png b/example_images/ref12.png new file mode 100644 index 0000000000000000000000000000000000000000..2182c327c8582eba88dea839211769105c154f19 Binary files /dev/null and b/example_images/ref12.png differ diff --git a/example_images/ref13.jpg b/example_images/ref13.jpg new file mode 100644 index 0000000000000000000000000000000000000000..227fc9b2b8a21d0671cb66468c7ef9a1113736c6 Binary files /dev/null and b/example_images/ref13.jpg differ diff --git a/example_images/ref14.png b/example_images/ref14.png new file mode 100644 index 0000000000000000000000000000000000000000..78a2b6af97727e743e3873b5076f65827d1bb0ff Binary files /dev/null and b/example_images/ref14.png differ diff --git a/example_images/ref2.jpg b/example_images/ref2.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f4878c58a7b09c540a8deb1a7685b8c017cc1dbf Binary files /dev/null and b/example_images/ref2.jpg differ diff --git a/example_images/ref3.jpg b/example_images/ref3.jpg new file mode 100644 index 0000000000000000000000000000000000000000..67aac42212aac484d6fed6b032dc3aca43189620 Binary files /dev/null and b/example_images/ref3.jpg differ diff --git a/example_images/ref4.jpg b/example_images/ref4.jpg new file mode 100644 index 0000000000000000000000000000000000000000..36897cc34399c3c3e849e70b161090a34920c309 Binary files /dev/null and b/example_images/ref4.jpg differ diff --git a/example_images/ref5.jpg b/example_images/ref5.jpg new file mode 100644 index 0000000000000000000000000000000000000000..33cdfe4c95a124b06f829d1f774e847fecefd6ad Binary files /dev/null and b/example_images/ref5.jpg differ diff --git a/example_images/ref6.jpg b/example_images/ref6.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d375a3f1bb966114237ba2eb36ef6f8b506835df Binary files /dev/null and b/example_images/ref6.jpg differ diff --git a/example_images/ref7.jpg b/example_images/ref7.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8eb7ef4901eac5b727d775fef83a6cace00eebab Binary files /dev/null and b/example_images/ref7.jpg differ diff --git a/example_images/ref8.jpg b/example_images/ref8.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c36b3c5b2e340b665e6547eac85044a6af72392a Binary files /dev/null and b/example_images/ref8.jpg differ diff --git a/example_images/ref9.jpg b/example_images/ref9.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8c23e56cf899f4874eb7d0e4b9adf056aa0ea5ef Binary files /dev/null and b/example_images/ref9.jpg differ diff --git a/font/Arial_Unicode.ttf b/font/Arial_Unicode.ttf new file mode 100644 index 0000000000000000000000000000000000000000..2bc3e08a7fbd6d2c40fad21b056a1ed865c1a2b9 --- /dev/null +++ b/font/Arial_Unicode.ttf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:876af2cd4854644e7f3e7feb2f688997fdb3343c6df6693611209c9dfb47ccec +size 23278008 diff --git a/javascript/bboxHint.js b/javascript/bboxHint.js new file mode 100644 index 0000000000000000000000000000000000000000..a17a233c8472cbcd960f2b1f949cf9a5a8021d05 --- /dev/null +++ b/javascript/bboxHint.js @@ -0,0 +1,554 @@ +/* +Part of the implementation is borrowed and modified from multidiffusion-upscaler-for-automatic1111, +publicly available at https://github.com/pkuliyi2015/multidiffusion-upscaler-for-automatic1111 +*/ + +const BBOX_MAX_NUM = 16; +const BBOX_WARNING_SIZE = 1280; +const DEFAULT_X = 0.4; +const DEFAULT_Y = 0.4; +const DEFAULT_H = 0.2; +const DEFAULT_W = 0.2; + +// ref: https://html-color.codes/ +const COLOR_MAP = [ + ['#ff0000', 'rgba(255, 0, 0, 0.3)'], // red + ['#ff9900', 'rgba(255, 153, 0, 0.3)'], // orange + ['#ffff00', 'rgba(255, 255, 0, 0.3)'], // yellow + ['#33cc33', 'rgba(51, 204, 51, 0.3)'], // green + ['#33cccc', 'rgba(51, 204, 204, 0.3)'], // indigo + ['#0066ff', 'rgba(0, 102, 255, 0.3)'], // blue + ['#6600ff', 'rgba(102, 0, 255, 0.3)'], // purple + ['#cc00cc', 'rgba(204, 0, 204, 0.3)'], // dark pink + ['#ff6666', 'rgba(255, 102, 102, 0.3)'], // light red + ['#ffcc66', 'rgba(255, 204, 102, 0.3)'], // light orange + ['#99cc00', 'rgba(153, 204, 0, 0.3)'], // lime green + ['#00cc99', 'rgba(0, 204, 153, 0.3)'], // teal + ['#0099cc', 'rgba(0, 153, 204, 0.3)'], // steel blue + ['#9933cc', 'rgba(153, 51, 204, 0.3)'], // lavender + ['#ff3399', 'rgba(255, 51, 153, 0.3)'], // hot pink + ['#996633', 'rgba(153, 102, 51, 0.3)'], // brown +]; + +const RESIZE_BORDER = 5; +const MOVE_BORDER = 5; + +const t2i_bboxes = new Array(BBOX_MAX_NUM).fill(null); +const i2i_bboxes = new Array(BBOX_MAX_NUM).fill(null); + +function gradioApp() { + const elems = document.getElementsByTagName('gradio-app') + const gradioShadowRoot = elems.length == 0 ? null : elems[0].shadowRoot + return !!gradioShadowRoot ? gradioShadowRoot : document; +} + +// ↓↓↓ called from gradio ↓↓↓ + +function onCreateT2IRefClick(overwrite) { + let width, height; + if (overwrite) { + const overwriteInputs = gradioApp().querySelectorAll('#MD-overwrite-width-t2i input, #MD-overwrite-height-t2i input'); + width = parseInt(overwriteInputs[0].value); + height = parseInt(overwriteInputs[2].value); + } else { + const sizeInputs = gradioApp().querySelectorAll('#txt2img_width input, #txt2img_height input'); + width = parseInt(sizeInputs[0].value); + height = parseInt(sizeInputs[2].value); + } + + if (isNaN(width)) width = 512; + if (isNaN(height)) height = 512; + + // Concat it to string to bypass the gradio bug + // 向黑恶势力低头 + return width.toString() + 'x' + height.toString(); +} + +function onCreateI2IRefClick() { + const canvas = gradioApp().querySelector('#img2img_image img'); + return canvas.src; +} + +function onBoxEnableClick(is_t2i, idx, enable) { + let canvas = null; + let bboxes = null; + let locator = null; + if (is_t2i) { + // locator = () => gradioApp().querySelector('#MD-bbox-ref-t2i'); + locator = () => gradioApp().querySelector('#MD-bbox-rect-t2i'); + bboxes = t2i_bboxes; + + } else { + locator = () => gradioApp().querySelector('#MD-bbox-ref-i2i'); + bboxes = i2i_bboxes; + } + ref_div = locator(); + canvas = ref_div.querySelector('img'); + if (!canvas) { return false; } + + if (enable) { + // Check if the bounding box already exists + if (!bboxes[idx]) { + // Initialize bounding box + const bbox = [DEFAULT_X, DEFAULT_Y, DEFAULT_W, DEFAULT_H]; + const colorMap = COLOR_MAP[idx % COLOR_MAP.length]; + const div = document.createElement('div'); + div.id = 'MD-bbox-' + (is_t2i ? 't2i-' : 'i2i-') + idx; + div.style.left = '0px'; + div.style.top = '0px'; + div.style.width = '0px'; + div.style.height = '0px'; + div.style.position = 'absolute'; + div.style.border = '2px solid ' + colorMap[0]; + div.style.background = colorMap[1]; + div.style.zIndex = '900'; + div.style.display = 'none'; + // A text tip to warn the user if bbox is too large + const tip = document.createElement('span'); + tip.id = 'MD-tip-' + (is_t2i ? 't2i-' : 'i2i-') + idx; + tip.style.left = '50%'; + tip.style.top = '50%'; + tip.style.position = 'absolute'; + tip.style.transform = 'translate(-50%, -50%)'; + tip.style.fontSize = '12px'; + tip.style.fontWeight = 'bold'; + tip.style.textAlign = 'center'; + tip.style.color = colorMap[0]; + tip.style.zIndex = '901'; + tip.style.display = 'none'; + tip.innerHTML = 'Warning: Region very large!
Take care of VRAM usage!'; + div.appendChild(tip); + div.addEventListener('mousedown', function (e) { + if (e.button === 0) { onBoxMouseDown(e, is_t2i, idx); } + }); + div.addEventListener('mousemove', function (e) { + updateCursorStyle(e, is_t2i, idx); + }); + + const shower = function() { // insert to DOM if necessary + if (!gradioApp().querySelector('#' + div.id)) { + locator().appendChild(div); + } + } + bboxes[idx] = [div, bbox, shower]; + } + + // Show the bounding box + displayBox(canvas, is_t2i, bboxes[idx]); + return true; + } else { + if (!bboxes[idx]) { return false; } + const [div, bbox, shower] = bboxes[idx]; + div.style.display = 'none'; + } + return false; +} + +function onBoxChange(is_t2i, idx, what, v) { + // This function handles all the changes of the bounding box + // Including the rendering and python slider update + let bboxes = null; + let canvas = null; + if (is_t2i) { + bboxes = t2i_bboxes; + canvas = gradioApp().querySelector('#MD-bbox-rect-t2i img'); + } else { + bboxes = i2i_bboxes; + canvas = gradioApp().querySelector('#MD-bbox-ref-i2i img'); + } + if (!bboxes[idx] || !canvas) { + switch (what) { + case 'x': return DEFAULT_X; + case 'y': return DEFAULT_Y; + case 'w': return DEFAULT_W; + case 'h': return DEFAULT_H; + } + } + const [div, bbox, shower] = bboxes[idx]; + if (div.style.display === 'none') { return v; } + + // parse trigger + switch (what) { + case 'x': bbox[0] = v; break; + case 'y': bbox[1] = v; break; + case 'w': bbox[2] = v; break; + case 'h': bbox[3] = v; break; + } + displayBox(canvas, is_t2i, bboxes[idx]); + return v; +} + +// ↓↓↓ called from js ↓↓↓ + +function getSeedInfo(is_t2i, id, current_seed) { + const info_id = is_t2i ? '#html_info_txt2img' : '#html_info_img2img'; + const info_div = gradioApp().querySelector(info_id); + try{ + current_seed = parseInt(current_seed); + } catch(e) { + current_seed = -1; + } + if (!info_div) return current_seed; + let info = info_div.innerHTML; + if (!info) return current_seed; + // remove all html tags + info = info.replace(/<[^>]*>/g, ''); + // Find a json string 'region control:' in the info + // get its index + idx = info.indexOf('Region control'); + if (idx == -1) return current_seed; + // get the json string (detect the bracket) + // find the first '{' + let start_idx = info.indexOf('{', idx); + let bracket = 1; + let end_idx = start_idx + 1; + while (bracket > 0 && end_idx < info.length) { + if (info[end_idx] == '{') bracket++; + if (info[end_idx] == '}') bracket--; + end_idx++; + } + if (bracket > 0) { + return current_seed; + } + // get the json string + let json_str = info.substring(start_idx, end_idx); + // replace the single quote to double quote + json_str = json_str.replace(/'/g, '"'); + // replace python True to javascript true, False to false + json_str = json_str.replace(/True/g, 'true'); + // parse the json string + let json = JSON.parse(json_str); + // get the seed if the region id is in the json + const region_id = 'Region ' + id.toString(); + if (!(region_id in json)) return current_seed; + const region = json[region_id]; + if (!('seed' in region)) return current_seed; + let seed = region['seed']; + try{ + seed = parseInt(seed); + } catch(e) { + return current_seed; + } + return seed; +} + +function displayBox(canvas, is_t2i, bbox_info) { + // check null input + const [div, bbox, shower] = bbox_info; + const [x, y, w, h] = bbox; + if (!canvas || !div || x == null || y == null || w == null || h == null) { return; } + + // client: canvas widget display size + // natural: content image real size + let vpScale = Math.min(canvas.clientWidth / canvas.naturalWidth, canvas.clientHeight / canvas.naturalHeight); + let canvasCenterX = canvas.clientWidth / 2; + let canvasCenterY = canvas.clientHeight / 2; + let scaledX = canvas.naturalWidth * vpScale; + let scaledY = canvas.naturalHeight * vpScale; + let viewRectLeft = canvasCenterX - scaledX / 2; + let viewRectRight = canvasCenterX + scaledX / 2; + let viewRectTop = canvasCenterY - scaledY / 2; + let viewRectDown = canvasCenterY + scaledY / 2; + + let xDiv = viewRectLeft + scaledX * x; + let yDiv = viewRectTop + scaledY * y; + let wDiv = Math.min(scaledX * w, viewRectRight - xDiv); + let hDiv = Math.min(scaledY * h, viewRectDown - yDiv); + + // Calculate warning bbox size + let upscalerFactor = 1.0; + if (!is_t2i) { + const upscalerInput = parseFloat(gradioApp().querySelector('#MD-i2i-upscaler-factor input').value); + if (!isNaN(upscalerInput)) upscalerFactor = upscalerInput; + } + let maxSize = BBOX_WARNING_SIZE / upscalerFactor * vpScale; + let maxW = maxSize / scaledX; + let maxH = maxSize / scaledY; + if (w > maxW || h > maxH) { + div.querySelector('span').style.display = 'block'; + } else { + div.querySelector('span').style.display = 'none'; + } + + // update
when not equal + div.style.left = xDiv + 'px'; + div.style.top = yDiv + 'px'; + div.style.width = wDiv + 'px'; + div.style.height = hDiv + 'px'; + div.style.display = 'block'; + + // insert it to DOM if not appear + shower(); +} + +function onBoxMouseDown(e, is_t2i, idx) { + let bboxes = null; + let canvas = null; + if (is_t2i) { + bboxes = t2i_bboxes; + canvas = gradioApp().querySelector('#MD-bbox-rect-t2i img'); + } else { + bboxes = i2i_bboxes; + canvas = gradioApp().querySelector('#MD-bbox-ref-i2i img'); + } + // Get the bounding box + if (!canvas || !bboxes[idx]) { return; } + const [div, bbox, shower] = bboxes[idx]; + + // Check if the click is inside the bounding box + const boxRect = div.getBoundingClientRect(); + let mouseX = e.clientX; + let mouseY = e.clientY; + + const resizeLeft = mouseX >= boxRect.left && mouseX <= boxRect.left + RESIZE_BORDER; + const resizeRight = mouseX >= boxRect.right - RESIZE_BORDER && mouseX <= boxRect.right; + const resizeTop = mouseY >= boxRect.top && mouseY <= boxRect.top + RESIZE_BORDER; + const resizeBottom = mouseY >= boxRect.bottom - RESIZE_BORDER && mouseY <= boxRect.bottom; + + const moveHorizontal = mouseX >= boxRect.left + MOVE_BORDER && mouseX <= boxRect.right - MOVE_BORDER; + const moveVertical = mouseY >= boxRect.top + MOVE_BORDER && mouseY <= boxRect.bottom - MOVE_BORDER; + + if (!resizeLeft && !resizeRight && !resizeTop && !resizeBottom && !moveHorizontal && !moveVertical) { return; } + + const horizontalPivot = resizeLeft ? bbox[0] + bbox[2] : bbox[0]; + const verticalPivot = resizeTop ? bbox[1] + bbox[3] : bbox[1]; + + // Canvas can be regarded as invariant during the drag operation + // Calculate in advance to reduce overhead + + // Calculate viewport scale based on the current canvas size and the natural image size + let vpScale = Math.min(canvas.clientWidth / canvas.naturalWidth, canvas.clientHeight / canvas.naturalHeight); + let vpOffset = canvas.getBoundingClientRect(); + + // Calculate scaled dimensions of the canvas + let scaledX = canvas.naturalWidth * vpScale; + let scaledY = canvas.naturalHeight * vpScale; + + // Calculate the canvas center and view rectangle coordinates + let canvasCenterX = (vpOffset.left + window.scrollX) + canvas.clientWidth / 2; + let canvasCenterY = (vpOffset.top + window.scrollY) + canvas.clientHeight / 2; + let viewRectLeft = canvasCenterX - scaledX / 2 - window.scrollX; + let viewRectRight = canvasCenterX + scaledX / 2 - window.scrollX; + let viewRectTop = canvasCenterY - scaledY / 2 - window.scrollY; + let viewRectDown = canvasCenterY + scaledY / 2 - window.scrollY; + + mouseX = Math.min(Math.max(mouseX, viewRectLeft), viewRectRight); + mouseY = Math.min(Math.max(mouseY, viewRectTop), viewRectDown); + + //const accordion = gradioApp().querySelector(`#MD-accordion-${is_t2i ? 't2i' : 'i2i'}-${idx}`); + const accordion = gradioApp().querySelector('#MD-tab-t2i'); + + // Move or resize the bounding box on mousemove + function onMouseMove(e) { + // Prevent selecting anything irrelevant + e.preventDefault(); + + // Get the new mouse position + let newMouseX = e.clientX; + let newMouseY = e.clientY; + + // clamp the mouse position to the view rectangle + newMouseX = Math.min(Math.max(newMouseX, viewRectLeft), viewRectRight); + newMouseY = Math.min(Math.max(newMouseY, viewRectTop), viewRectDown); + + // Calculate the mouse movement delta + const dx = (newMouseX - mouseX) / scaledX; + const dy = (newMouseY - mouseY) / scaledY; + + // Update the mouse position + mouseX = newMouseX; + mouseY = newMouseY; + + // if no move just return + if (dx === 0 && dy === 0) { return; } + + // Update the mouse position + let [x, y, w, h] = bbox; + if (moveHorizontal && moveVertical) { + // If moving the bounding box + x = Math.min(Math.max(x + dx, 0), 1 - w); + y = Math.min(Math.max(y + dy, 0), 1 - h); + } else { + // If resizing the bounding box + if (resizeLeft || resizeRight) { + if (x < horizontalPivot) { + if (dx <= w) { + // If still within the left side of the pivot + x = x + dx; + w = w - dx; + } else { + // If crossing the pivot + w = dx - w; + x = horizontalPivot; + } + } else { + if (w + dx < 0) { + // If still within the right side of the pivot + x = horizontalPivot + w + dx; + w = - dx - w; + } else { + // If crossing the pivot + x = horizontalPivot; + w = w + dx; + } + } + + // Clamp the bounding box to the image + if (x < 0) { + w = w + x; + x = 0; + } else if (x + w > 1) { + w = 1 - x; + } + } + // Same as above, but for the vertical axis + if (resizeTop || resizeBottom) { + if (y < verticalPivot) { + if (dy <= h) { + y = y + dy; + h = h - dy; + } else { + h = dy - h; + y = verticalPivot; + } + } else { + if (h + dy < 0) { + y = verticalPivot + h + dy; + h = - dy - h; + } else { + y = verticalPivot; + h = h + dy; + } + } + if (y < 0) { + h = h + y; + y = 0; + } else if (y + h > 1) { + h = 1 - y; + } + } + } + const [div, old_bbox, _] = bboxes[idx]; + + // If all the values are the same, just return + if (old_bbox[0] === x && old_bbox[1] === y && old_bbox[2] === w && old_bbox[3] === h) { return; } + // else update the bbox + const event = new Event('input'); + const coords = [x, y, w, h]; + // The querySelector is not very efficient, so we query it once and reuse it + // caching will result gradio bugs that stucks bbox and cannot move & drag + const sliderIds = ['x', 'y', 'w', 'h']; + // We try to select the input sliders + const sliderSelectors = sliderIds.map(id => `#MD-${is_t2i ? 't2i' : 'i2i'}-${idx}-${id} input`).join(', '); + let sliderInputs = accordion.querySelectorAll(sliderSelectors); + // alert(sliderInputs.length) + if (sliderInputs.length == 0) { + // If we failed, the accordion is probably closed and sliders are removed in the dom, so we open it + accordion.querySelector('.label-wrap').click(); + // and try again + sliderInputs = accordion.querySelectorAll(sliderSelectors); + // If we still failed, we just return + if (sliderInputs.length == 0) { return; } + } + for (let i = 0; i < 4; i++) { + if (old_bbox[i] !== coords[i]) { + sliderInputs[2*i].value = coords[i]; + sliderInputs[2*i].dispatchEvent(event); + } + } + } + + // Remove the mousemove and mouseup event listeners + function onMouseUp() { + document.removeEventListener('mousemove', onMouseMove); + document.removeEventListener('mouseup', onMouseUp); + } + + // Add the event listeners + document.addEventListener('mousemove', onMouseMove); + document.addEventListener('mouseup', onMouseUp); +} + +function updateCursorStyle(e, is_t2i, idx) { + // This function changes the cursor style when hovering over the bounding box + const bboxes = is_t2i ? t2i_bboxes : i2i_bboxes; + if (!bboxes[idx]) return; + + const div = bboxes[idx][0]; + const boxRect = div.getBoundingClientRect(); + const mouseX = e.clientX; + const mouseY = e.clientY; + + const resizeLeft = mouseX >= boxRect.left && mouseX <= boxRect.left + RESIZE_BORDER; + const resizeRight = mouseX >= boxRect.right - RESIZE_BORDER && mouseX <= boxRect.right; + const resizeTop = mouseY >= boxRect.top && mouseY <= boxRect.top + RESIZE_BORDER; + const resizeBottom = mouseY >= boxRect.bottom - RESIZE_BORDER && mouseY <= boxRect.bottom; + + if ((resizeLeft && resizeTop) || (resizeRight && resizeBottom)) { + div.style.cursor = 'nwse-resize'; + } else if ((resizeLeft && resizeBottom) || (resizeRight && resizeTop)) { + div.style.cursor = 'nesw-resize'; + } else if (resizeLeft || resizeRight) { + div.style.cursor = 'ew-resize'; + } else if (resizeTop || resizeBottom) { + div.style.cursor = 'ns-resize'; + } else { + div.style.cursor = 'move'; + } +} + +// ↓↓↓ auto called event listeners ↓↓↓ + +function updateBoxes(is_t2i) { + // This function redraw all bounding boxes + let bboxes = null; + let canvas = null; + if (is_t2i) { + bboxes = t2i_bboxes; + canvas = gradioApp().querySelector('#MD-bbox-rect-t2i img'); + } else { + bboxes = i2i_bboxes; + canvas = gradioApp().querySelector('#MD-bbox-ref-i2i img'); + } + if (!canvas) return; + + for (let idx = 0; idx < bboxes.length; idx++) { + if (!bboxes[idx]) continue; + const [div, bbox, shower] = bboxes[idx]; + if (div.style.display === 'none') { return; } + + displayBox(canvas, is_t2i, bboxes[idx]); + } +} + +window.addEventListener('resize', _ => { + updateBoxes(true); + updateBoxes(false); +}); + +// ======== Gradio Bug Fix ======== +// For Gradio versions > 3.16.0 and < 3.29.0, the accordion DOM will be deleted when it is closed. +// We need to judge the versions and listen to the accordion open event, rerender the bbox at that time. +// This silly bug fix is only for compatibility, we recommend to update the gradio version to 3.29.0 or higher. +try { + const GRADIO_VERSIONS = window.gradio_config["version"].split("."); + const gradio_major_version = parseInt(GRADIO_VERSIONS[0]); + const gradio_minor_version = parseInt(GRADIO_VERSIONS[1]); + if (gradio_major_version == 3 && gradio_minor_version > 16 && gradio_minor_version < 29) { + let listener = e => { + if (!e) { return; } + if (!e.target) { return; } + if (!e.target.classList) { return; } + if (!e.target.classList.contains('label-wrap')) { return; } + for (let tab of ['t2i', 'i2i']) { + const div = gradioApp().querySelector('#MD-bbox-control-' + tab +' div.label-wrap'); + if (!div) { continue; } + updateBoxes(tab === 't2i'); + } + }; + window.addEventListener('DOMNodeInserted', listener); + } +} catch (ignored) { + // If the above code failed, the gradio version shouldn't be in the range of 3.16.0 to 3.29.0, so we just return. +} +// ======== Gradio Bug Fix ======== diff --git a/ldm/data/__init__.py b/ldm/data/__init__.py new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/ldm/data/util.py b/ldm/data/util.py new file mode 100644 index 0000000000000000000000000000000000000000..5b60ceb2349e3bd7900ff325740e2022d2903b1c --- /dev/null +++ b/ldm/data/util.py @@ -0,0 +1,24 @@ +import torch + +from ldm.modules.midas.api import load_midas_transform + + +class AddMiDaS(object): + def __init__(self, model_type): + super().__init__() + self.transform = load_midas_transform(model_type) + + def pt2np(self, x): + x = ((x + 1.0) * .5).detach().cpu().numpy() + return x + + def np2pt(self, x): + x = torch.from_numpy(x) * 2 - 1. + return x + + def __call__(self, sample): + # sample['jpg'] is tensor hwc in [-1, 1] at this point + x = self.pt2np(sample['jpg']) + x = self.transform({"image": x})["image"] + sample['midas_in'] = x + return sample \ No newline at end of file diff --git a/ldm/models/autoencoder.py b/ldm/models/autoencoder.py new file mode 100644 index 0000000000000000000000000000000000000000..d122549995ce2cd64092c81a58419ed4a15a02fd --- /dev/null +++ b/ldm/models/autoencoder.py @@ -0,0 +1,219 @@ +import torch +import pytorch_lightning as pl +import torch.nn.functional as F +from contextlib import contextmanager + +from ldm.modules.diffusionmodules.model import Encoder, Decoder +from ldm.modules.distributions.distributions import DiagonalGaussianDistribution + +from ldm.util import instantiate_from_config +from ldm.modules.ema import LitEma + + +class AutoencoderKL(pl.LightningModule): + def __init__(self, + ddconfig, + lossconfig, + embed_dim, + ckpt_path=None, + ignore_keys=[], + image_key="image", + colorize_nlabels=None, + monitor=None, + ema_decay=None, + learn_logvar=False + ): + super().__init__() + self.learn_logvar = learn_logvar + self.image_key = image_key + self.encoder = Encoder(**ddconfig) + self.decoder = Decoder(**ddconfig) + self.loss = instantiate_from_config(lossconfig) + assert ddconfig["double_z"] + self.quant_conv = torch.nn.Conv2d(2*ddconfig["z_channels"], 2*embed_dim, 1) + self.post_quant_conv = torch.nn.Conv2d(embed_dim, ddconfig["z_channels"], 1) + self.embed_dim = embed_dim + if colorize_nlabels is not None: + assert type(colorize_nlabels)==int + self.register_buffer("colorize", torch.randn(3, colorize_nlabels, 1, 1)) + if monitor is not None: + self.monitor = monitor + + self.use_ema = ema_decay is not None + if self.use_ema: + self.ema_decay = ema_decay + assert 0. < ema_decay < 1. + self.model_ema = LitEma(self, decay=ema_decay) + print(f"Keeping EMAs of {len(list(self.model_ema.buffers()))}.") + + if ckpt_path is not None: + self.init_from_ckpt(ckpt_path, ignore_keys=ignore_keys) + + def init_from_ckpt(self, path, ignore_keys=list()): + sd = torch.load(path, map_location="cpu")["state_dict"] + keys = list(sd.keys()) + for k in keys: + for ik in ignore_keys: + if k.startswith(ik): + print("Deleting key {} from state_dict.".format(k)) + del sd[k] + self.load_state_dict(sd, strict=False) + print(f"Restored from {path}") + + @contextmanager + def ema_scope(self, context=None): + if self.use_ema: + self.model_ema.store(self.parameters()) + self.model_ema.copy_to(self) + if context is not None: + print(f"{context}: Switched to EMA weights") + try: + yield None + finally: + if self.use_ema: + self.model_ema.restore(self.parameters()) + if context is not None: + print(f"{context}: Restored training weights") + + def on_train_batch_end(self, *args, **kwargs): + if self.use_ema: + self.model_ema(self) + + def encode(self, x): + h = self.encoder(x) + moments = self.quant_conv(h) + posterior = DiagonalGaussianDistribution(moments) + return posterior + + def decode(self, z): + z = self.post_quant_conv(z) + dec = self.decoder(z) + return dec + + def forward(self, input, sample_posterior=True): + posterior = self.encode(input) + if sample_posterior: + z = posterior.sample() + else: + z = posterior.mode() + dec = self.decode(z) + return dec, posterior + + def get_input(self, batch, k): + x = batch[k] + if len(x.shape) == 3: + x = x[..., None] + x = x.permute(0, 3, 1, 2).to(memory_format=torch.contiguous_format).float() + return x + + def training_step(self, batch, batch_idx, optimizer_idx): + inputs = self.get_input(batch, self.image_key) + reconstructions, posterior = self(inputs) + + if optimizer_idx == 0: + # train encoder+decoder+logvar + aeloss, log_dict_ae = self.loss(inputs, reconstructions, posterior, optimizer_idx, self.global_step, + last_layer=self.get_last_layer(), split="train") + self.log("aeloss", aeloss, prog_bar=True, logger=True, on_step=True, on_epoch=True) + self.log_dict(log_dict_ae, prog_bar=False, logger=True, on_step=True, on_epoch=False) + return aeloss + + if optimizer_idx == 1: + # train the discriminator + discloss, log_dict_disc = self.loss(inputs, reconstructions, posterior, optimizer_idx, self.global_step, + last_layer=self.get_last_layer(), split="train") + + self.log("discloss", discloss, prog_bar=True, logger=True, on_step=True, on_epoch=True) + self.log_dict(log_dict_disc, prog_bar=False, logger=True, on_step=True, on_epoch=False) + return discloss + + def validation_step(self, batch, batch_idx): + log_dict = self._validation_step(batch, batch_idx) + with self.ema_scope(): + log_dict_ema = self._validation_step(batch, batch_idx, postfix="_ema") + return log_dict + + def _validation_step(self, batch, batch_idx, postfix=""): + inputs = self.get_input(batch, self.image_key) + reconstructions, posterior = self(inputs) + aeloss, log_dict_ae = self.loss(inputs, reconstructions, posterior, 0, self.global_step, + last_layer=self.get_last_layer(), split="val"+postfix) + + discloss, log_dict_disc = self.loss(inputs, reconstructions, posterior, 1, self.global_step, + last_layer=self.get_last_layer(), split="val"+postfix) + + self.log(f"val{postfix}/rec_loss", log_dict_ae[f"val{postfix}/rec_loss"]) + self.log_dict(log_dict_ae) + self.log_dict(log_dict_disc) + return self.log_dict + + def configure_optimizers(self): + lr = self.learning_rate + ae_params_list = list(self.encoder.parameters()) + list(self.decoder.parameters()) + list( + self.quant_conv.parameters()) + list(self.post_quant_conv.parameters()) + if self.learn_logvar: + print(f"{self.__class__.__name__}: Learning logvar") + ae_params_list.append(self.loss.logvar) + opt_ae = torch.optim.Adam(ae_params_list, + lr=lr, betas=(0.5, 0.9)) + opt_disc = torch.optim.Adam(self.loss.discriminator.parameters(), + lr=lr, betas=(0.5, 0.9)) + return [opt_ae, opt_disc], [] + + def get_last_layer(self): + return self.decoder.conv_out.weight + + @torch.no_grad() + def log_images(self, batch, only_inputs=False, log_ema=False, **kwargs): + log = dict() + x = self.get_input(batch, self.image_key) + x = x.to(self.device) + if not only_inputs: + xrec, posterior = self(x) + if x.shape[1] > 3: + # colorize with random projection + assert xrec.shape[1] > 3 + x = self.to_rgb(x) + xrec = self.to_rgb(xrec) + log["samples"] = self.decode(torch.randn_like(posterior.sample())) + log["reconstructions"] = xrec + if log_ema or self.use_ema: + with self.ema_scope(): + xrec_ema, posterior_ema = self(x) + if x.shape[1] > 3: + # colorize with random projection + assert xrec_ema.shape[1] > 3 + xrec_ema = self.to_rgb(xrec_ema) + log["samples_ema"] = self.decode(torch.randn_like(posterior_ema.sample())) + log["reconstructions_ema"] = xrec_ema + log["inputs"] = x + return log + + def to_rgb(self, x): + assert self.image_key == "segmentation" + if not hasattr(self, "colorize"): + self.register_buffer("colorize", torch.randn(3, x.shape[1], 1, 1).to(x)) + x = F.conv2d(x, weight=self.colorize) + x = 2.*(x-x.min())/(x.max()-x.min()) - 1. + return x + + +class IdentityFirstStage(torch.nn.Module): + def __init__(self, *args, vq_interface=False, **kwargs): + self.vq_interface = vq_interface + super().__init__() + + def encode(self, x, *args, **kwargs): + return x + + def decode(self, x, *args, **kwargs): + return x + + def quantize(self, x, *args, **kwargs): + if self.vq_interface: + return x, None, [None, None, None] + return x + + def forward(self, x, *args, **kwargs): + return x + diff --git a/ldm/models/diffusion/__init__.py b/ldm/models/diffusion/__init__.py new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/ldm/models/diffusion/ddim.py b/ldm/models/diffusion/ddim.py new file mode 100644 index 0000000000000000000000000000000000000000..676813b7ecaa3f82338d4e99d6549251872d31a1 --- /dev/null +++ b/ldm/models/diffusion/ddim.py @@ -0,0 +1,354 @@ +"""SAMPLING ONLY.""" + +import torch +import numpy as np +from tqdm import tqdm + +from ldm.modules.diffusionmodules.util import make_ddim_sampling_parameters, make_ddim_timesteps, noise_like, extract_into_tensor + + +class DDIMSampler(object): + def __init__(self, model, schedule="linear", **kwargs): + super().__init__() + self.model = model + self.ddpm_num_timesteps = model.num_timesteps + self.schedule = schedule + + def register_buffer(self, name, attr): + if type(attr) == torch.Tensor: + if attr.device != torch.device("cuda"): + attr = attr.to(torch.device("cuda")) + setattr(self, name, attr) + + def make_schedule(self, ddim_num_steps, ddim_discretize="uniform", ddim_eta=0., verbose=True): + self.ddim_timesteps = make_ddim_timesteps(ddim_discr_method=ddim_discretize, num_ddim_timesteps=ddim_num_steps, + num_ddpm_timesteps=self.ddpm_num_timesteps,verbose=verbose) + alphas_cumprod = self.model.alphas_cumprod + assert alphas_cumprod.shape[0] == self.ddpm_num_timesteps, 'alphas have to be defined for each timestep' + to_torch = lambda x: x.clone().detach().to(torch.float32).to(self.model.device) + + self.register_buffer('betas', to_torch(self.model.betas)) + self.register_buffer('alphas_cumprod', to_torch(alphas_cumprod)) + self.register_buffer('alphas_cumprod_prev', to_torch(self.model.alphas_cumprod_prev)) + + # calculations for diffusion q(x_t | x_{t-1}) and others + self.register_buffer('sqrt_alphas_cumprod', to_torch(np.sqrt(alphas_cumprod.cpu()))) + self.register_buffer('sqrt_one_minus_alphas_cumprod', to_torch(np.sqrt(1. - alphas_cumprod.cpu()))) + self.register_buffer('log_one_minus_alphas_cumprod', to_torch(np.log(1. - alphas_cumprod.cpu()))) + self.register_buffer('sqrt_recip_alphas_cumprod', to_torch(np.sqrt(1. / alphas_cumprod.cpu()))) + self.register_buffer('sqrt_recipm1_alphas_cumprod', to_torch(np.sqrt(1. / alphas_cumprod.cpu() - 1))) + + # ddim sampling parameters + ddim_sigmas, ddim_alphas, ddim_alphas_prev = make_ddim_sampling_parameters(alphacums=alphas_cumprod.cpu(), + ddim_timesteps=self.ddim_timesteps, + eta=ddim_eta,verbose=verbose) + self.register_buffer('ddim_sigmas', ddim_sigmas) + self.register_buffer('ddim_alphas', ddim_alphas) + self.register_buffer('ddim_alphas_prev', ddim_alphas_prev) + self.register_buffer('ddim_sqrt_one_minus_alphas', np.sqrt(1. - ddim_alphas)) + sigmas_for_original_sampling_steps = ddim_eta * torch.sqrt( + (1 - self.alphas_cumprod_prev) / (1 - self.alphas_cumprod) * ( + 1 - self.alphas_cumprod / self.alphas_cumprod_prev)) + self.register_buffer('ddim_sigmas_for_original_num_steps', sigmas_for_original_sampling_steps) + + @torch.no_grad() + def sample(self, + S, + batch_size, + shape, + conditioning=None, + callback=None, + normals_sequence=None, + img_callback=None, + quantize_x0=False, + eta=0., + mask=None, + x0=None, + temperature=1., + noise_dropout=0., + score_corrector=None, + corrector_kwargs=None, + verbose=True, + x_T=None, + log_every_t=100, + unconditional_guidance_scale=1., + unconditional_conditioning=None, # this has to come in the same format as the conditioning, # e.g. as encoded tokens, ... + dynamic_threshold=None, + ucg_schedule=None, + **kwargs + ): + if conditioning is not None: + if isinstance(conditioning, dict): + ctmp = conditioning[list(conditioning.keys())[0]] + while isinstance(ctmp, list): ctmp = ctmp[0] + cbs = ctmp.shape[0] + # cbs = len(ctmp[0]) + if cbs != batch_size: + print(f"Warning: Got {cbs} conditionings but batch-size is {batch_size}") + + elif isinstance(conditioning, list): + for ctmp in conditioning: + if ctmp.shape[0] != batch_size: + print(f"Warning: Got {cbs} conditionings but batch-size is {batch_size}") + + else: + if conditioning.shape[0] != batch_size: + print(f"Warning: Got {conditioning.shape[0]} conditionings but batch-size is {batch_size}") + + self.make_schedule(ddim_num_steps=S, ddim_eta=eta, verbose=verbose) + # sampling + C, H, W = shape + size = (batch_size, C, H, W) + print(f'Data shape for DDIM sampling is {size}, eta {eta}') + + samples, intermediates = self.ddim_sampling(conditioning, size, + callback=callback, + img_callback=img_callback, + quantize_denoised=quantize_x0, + mask=mask, x0=x0, + ddim_use_original_steps=False, + noise_dropout=noise_dropout, + temperature=temperature, + score_corrector=score_corrector, + corrector_kwargs=corrector_kwargs, + x_T=x_T, + log_every_t=log_every_t, + unconditional_guidance_scale=unconditional_guidance_scale, + unconditional_conditioning=unconditional_conditioning, + dynamic_threshold=dynamic_threshold, + ucg_schedule=ucg_schedule + ) + return samples, intermediates + + @torch.no_grad() + def ddim_sampling(self, cond, shape, + x_T=None, ddim_use_original_steps=False, + callback=None, timesteps=None, quantize_denoised=False, + mask=None, x0=None, img_callback=None, log_every_t=100, + temperature=1., noise_dropout=0., score_corrector=None, corrector_kwargs=None, + unconditional_guidance_scale=1., unconditional_conditioning=None, dynamic_threshold=None, + ucg_schedule=None): + device = self.model.betas.device + b = shape[0] + if x_T is None: + img = torch.randn(shape, device=device) + else: + img = x_T + + if timesteps is None: + timesteps = self.ddpm_num_timesteps if ddim_use_original_steps else self.ddim_timesteps + elif timesteps is not None and not ddim_use_original_steps: + subset_end = int(min(timesteps / self.ddim_timesteps.shape[0], 1) * self.ddim_timesteps.shape[0]) - 1 + timesteps = self.ddim_timesteps[:subset_end] + + intermediates = {'x_inter': [img], 'pred_x0': [img], "index": [10000]} + time_range = reversed(range(0, timesteps)) if ddim_use_original_steps else np.flip(timesteps) + total_steps = timesteps if ddim_use_original_steps else timesteps.shape[0] + print(f"Running DDIM Sampling with {total_steps} timesteps") + + iterator = tqdm(time_range, desc='DDIM Sampler', total=total_steps) + + for i, step in enumerate(iterator): + index = total_steps - i - 1 + ts = torch.full((b,), step, device=device, dtype=torch.long) + + if mask is not None: + assert x0 is not None + img_orig = self.model.q_sample(x0, ts) # TODO: deterministic forward pass? + img = img_orig * mask + (1. - mask) * img + + if ucg_schedule is not None: + assert len(ucg_schedule) == len(time_range) + unconditional_guidance_scale = ucg_schedule[i] + + outs = self.p_sample_ddim(img, cond, ts, index=index, use_original_steps=ddim_use_original_steps, + quantize_denoised=quantize_denoised, temperature=temperature, + noise_dropout=noise_dropout, score_corrector=score_corrector, + corrector_kwargs=corrector_kwargs, + unconditional_guidance_scale=unconditional_guidance_scale, + unconditional_conditioning=unconditional_conditioning, + dynamic_threshold=dynamic_threshold) + img, pred_x0 = outs + if callback: + callback(i) + if img_callback: + img_callback(pred_x0, i) + + if index % log_every_t == 0 or index == total_steps - 1: + intermediates['x_inter'].append(img) + intermediates['pred_x0'].append(pred_x0) + intermediates['index'].append(index) + + return img, intermediates + + @torch.no_grad() + def p_sample_ddim(self, x, c, t, index, repeat_noise=False, use_original_steps=False, quantize_denoised=False, + temperature=1., noise_dropout=0., score_corrector=None, corrector_kwargs=None, + unconditional_guidance_scale=1., unconditional_conditioning=None, + dynamic_threshold=None): + b, *_, device = *x.shape, x.device + + if unconditional_conditioning is None or unconditional_guidance_scale == 1.: + model_output = self.model.apply_model(x, t, c) + else: + x_in = torch.cat([x] * 2) + t_in = torch.cat([t] * 2) + if isinstance(c, dict): + assert isinstance(unconditional_conditioning, dict) + c_in = dict() + for k in c: + if isinstance(c[k], list): + c_in[k] = [torch.cat([ + unconditional_conditioning[k][i], + c[k][i]]) for i in range(len(c[k]))] + elif isinstance(c[k], dict): + c_in[k] = dict() + for key in c[k]: + if isinstance(c[k][key], list): + if not isinstance(c[k][key][0], torch.Tensor): + continue + c_in[k][key] = [torch.cat([ + unconditional_conditioning[k][key][i], + c[k][key][i]]) for i in range(len(c[k][key]))] + else: + c_in[k][key] = torch.cat([ + unconditional_conditioning[k][key], + c[k][key]]) + + else: + c_in[k] = torch.cat([ + unconditional_conditioning[k], + c[k]]) + elif isinstance(c, list): + c_in = list() + assert isinstance(unconditional_conditioning, list) + for i in range(len(c)): + c_in.append(torch.cat([unconditional_conditioning[i], c[i]])) + else: + c_in = torch.cat([unconditional_conditioning, c]) + model_uncond, model_t = self.model.apply_model(x_in, t_in, c_in).chunk(2) + model_output = model_uncond + unconditional_guidance_scale * (model_t - model_uncond) + + if self.model.parameterization == "v": + e_t = self.model.predict_eps_from_z_and_v(x, t, model_output) + else: + e_t = model_output + + if score_corrector is not None: + assert self.model.parameterization == "eps", 'not implemented' + e_t = score_corrector.modify_score(self.model, e_t, x, t, c, **corrector_kwargs) + + alphas = self.model.alphas_cumprod if use_original_steps else self.ddim_alphas + alphas_prev = self.model.alphas_cumprod_prev if use_original_steps else self.ddim_alphas_prev + sqrt_one_minus_alphas = self.model.sqrt_one_minus_alphas_cumprod if use_original_steps else self.ddim_sqrt_one_minus_alphas + sigmas = self.model.ddim_sigmas_for_original_num_steps if use_original_steps else self.ddim_sigmas + # select parameters corresponding to the currently considered timestep + a_t = torch.full((b, 1, 1, 1), alphas[index], device=device) + a_prev = torch.full((b, 1, 1, 1), alphas_prev[index], device=device) + sigma_t = torch.full((b, 1, 1, 1), sigmas[index], device=device) + sqrt_one_minus_at = torch.full((b, 1, 1, 1), sqrt_one_minus_alphas[index],device=device) + + # current prediction for x_0 + if self.model.parameterization != "v": + pred_x0 = (x - sqrt_one_minus_at * e_t) / a_t.sqrt() + else: + pred_x0 = self.model.predict_start_from_z_and_v(x, t, model_output) + + if quantize_denoised: + pred_x0, _, *_ = self.model.first_stage_model.quantize(pred_x0) + + if dynamic_threshold is not None: + raise NotImplementedError() + + # direction pointing to x_t + dir_xt = (1. - a_prev - sigma_t**2).sqrt() * e_t + noise = sigma_t * noise_like(x.shape, device, repeat_noise) * temperature + if noise_dropout > 0.: + noise = torch.nn.functional.dropout(noise, p=noise_dropout) + x_prev = a_prev.sqrt() * pred_x0 + dir_xt + noise + return x_prev, pred_x0 + + @torch.no_grad() + def encode(self, x0, c, t_enc, use_original_steps=False, return_intermediates=None, + unconditional_guidance_scale=1.0, unconditional_conditioning=None, callback=None): + num_reference_steps = self.ddpm_num_timesteps if use_original_steps else self.ddim_timesteps.shape[0] + + assert t_enc <= num_reference_steps + num_steps = t_enc + + if use_original_steps: + alphas_next = self.alphas_cumprod[:num_steps] + alphas = self.alphas_cumprod_prev[:num_steps] + else: + alphas_next = self.ddim_alphas[:num_steps] + alphas = torch.tensor(self.ddim_alphas_prev[:num_steps]) + + x_next = x0 + intermediates = [] + inter_steps = [] + for i in tqdm(range(num_steps), desc='Encoding Image'): + t = torch.full((x0.shape[0],), i, device=self.model.device, dtype=torch.long) + if unconditional_guidance_scale == 1.: + noise_pred = self.model.apply_model(x_next, t, c) + else: + assert unconditional_conditioning is not None + e_t_uncond, noise_pred = torch.chunk( + self.model.apply_model(torch.cat((x_next, x_next)), torch.cat((t, t)), + torch.cat((unconditional_conditioning, c))), 2) + noise_pred = e_t_uncond + unconditional_guidance_scale * (noise_pred - e_t_uncond) + + xt_weighted = (alphas_next[i] / alphas[i]).sqrt() * x_next + weighted_noise_pred = alphas_next[i].sqrt() * ( + (1 / alphas_next[i] - 1).sqrt() - (1 / alphas[i] - 1).sqrt()) * noise_pred + x_next = xt_weighted + weighted_noise_pred + if return_intermediates and i % ( + num_steps // return_intermediates) == 0 and i < num_steps - 1: + intermediates.append(x_next) + inter_steps.append(i) + elif return_intermediates and i >= num_steps - 2: + intermediates.append(x_next) + inter_steps.append(i) + if callback: callback(i) + + out = {'x_encoded': x_next, 'intermediate_steps': inter_steps} + if return_intermediates: + out.update({'intermediates': intermediates}) + return x_next, out + + @torch.no_grad() + def stochastic_encode(self, x0, t, use_original_steps=False, noise=None): + # fast, but does not allow for exact reconstruction + # t serves as an index to gather the correct alphas + if use_original_steps: + sqrt_alphas_cumprod = self.sqrt_alphas_cumprod + sqrt_one_minus_alphas_cumprod = self.sqrt_one_minus_alphas_cumprod + else: + sqrt_alphas_cumprod = torch.sqrt(self.ddim_alphas) + sqrt_one_minus_alphas_cumprod = self.ddim_sqrt_one_minus_alphas + + if noise is None: + noise = torch.randn_like(x0) + return (extract_into_tensor(sqrt_alphas_cumprod, t, x0.shape) * x0 + + extract_into_tensor(sqrt_one_minus_alphas_cumprod, t, x0.shape) * noise) + + @torch.no_grad() + def decode(self, x_latent, cond, t_start, unconditional_guidance_scale=1.0, unconditional_conditioning=None, + use_original_steps=False, callback=None): + + timesteps = np.arange(self.ddpm_num_timesteps) if use_original_steps else self.ddim_timesteps + timesteps = timesteps[:t_start] + + time_range = np.flip(timesteps) + total_steps = timesteps.shape[0] + print(f"Running DDIM Sampling with {total_steps} timesteps") + + iterator = tqdm(time_range, desc='Decoding image', total=total_steps) + x_dec = x_latent + for i, step in enumerate(iterator): + index = total_steps - i - 1 + ts = torch.full((x_latent.shape[0],), step, device=x_latent.device, dtype=torch.long) + x_dec, _ = self.p_sample_ddim(x_dec, cond, ts, index=index, use_original_steps=use_original_steps, + unconditional_guidance_scale=unconditional_guidance_scale, + unconditional_conditioning=unconditional_conditioning) + if callback: callback(i) + return x_dec \ No newline at end of file diff --git a/ldm/models/diffusion/ddpm.py b/ldm/models/diffusion/ddpm.py new file mode 100644 index 0000000000000000000000000000000000000000..43a3d590a0e5621afde6ed9e3899094fe2322555 --- /dev/null +++ b/ldm/models/diffusion/ddpm.py @@ -0,0 +1,1799 @@ +""" +Part of the implementation is borrowed and modified from ControlNet, publicly available at https://github.com/lllyasviel/ControlNet/blob/main/ldm/models/diffusion/ddpm.py +""" + +import torch +import torch.nn as nn +import numpy as np +import pytorch_lightning as pl +from torch.optim.lr_scheduler import LambdaLR +from einops import rearrange, repeat +from contextlib import contextmanager, nullcontext +from functools import partial +import itertools +from tqdm import tqdm +from torchvision.utils import make_grid +from pytorch_lightning.utilities.distributed import rank_zero_only +from omegaconf import ListConfig + +from ldm.util import log_txt_as_img, exists, default, ismap, isimage, mean_flat, count_params, instantiate_from_config +from ldm.modules.ema import LitEma +from ldm.modules.distributions.distributions import normal_kl, DiagonalGaussianDistribution +from ldm.models.autoencoder import IdentityFirstStage, AutoencoderKL +from ldm.modules.diffusionmodules.util import make_beta_schedule, extract_into_tensor, noise_like +from ldm.models.diffusion.ddim import DDIMSampler +from cldm.recognizer import crop_image +import cv2 + + +__conditioning_keys__ = {'concat': 'c_concat', + 'crossattn': 'c_crossattn', + 'adm': 'y'} + +PRINT_DEBUG = False + + +def print_grad(grad): + # print('Gradient:', grad) + # print(grad.shape) + a = grad.max() + b = grad.min() + # print(f'mean={grad.mean():.4f}, max={a:.4f}, min={b:.4f}') + s = 255./(a-b) + c = 255*(-b/(a-b)) + grad = grad * s + c + # print(f'mean={grad.mean():.4f}, max={grad.max():.4f}, min={grad.min():.4f}') + img = grad[0].permute(1, 2, 0).detach().cpu().numpy() + if img.shape[0] == 512: + cv2.imwrite('grad-img.jpg', img) + elif img.shape[0] == 64: + cv2.imwrite('grad-latent.jpg', img) + + +def disabled_train(self, mode=True): + """Overwrite model.train with this function to make sure train/eval mode + does not change anymore.""" + return self + + +def uniform_on_device(r1, r2, shape, device): + return (r1 - r2) * torch.rand(*shape, device=device) + r2 + + +class DDPM(pl.LightningModule): + # classic DDPM with Gaussian diffusion, in image space + def __init__(self, + unet_config, + timesteps=1000, + beta_schedule="linear", + loss_type="l2", + ckpt_path=None, + ignore_keys=[], + load_only_unet=False, + monitor="val/loss", + use_ema=True, + first_stage_key="image", + image_size=256, + channels=3, + log_every_t=100, + clip_denoised=True, + linear_start=1e-4, + linear_end=2e-2, + cosine_s=8e-3, + given_betas=None, + original_elbo_weight=0., + v_posterior=0., # weight for choosing posterior variance as sigma = (1-v) * beta_tilde + v * beta + l_simple_weight=1., + conditioning_key=None, + parameterization="eps", # all assuming fixed variance schedules + scheduler_config=None, + use_positional_encodings=False, + learn_logvar=False, + logvar_init=0., + make_it_fit=False, + ucg_training=None, + reset_ema=False, + reset_num_ema_updates=False, + ): + super().__init__() + assert parameterization in ["eps", "x0", "v"], 'currently only supporting "eps" and "x0" and "v"' + self.parameterization = parameterization + print(f"{self.__class__.__name__}: Running in {self.parameterization}-prediction mode") + self.cond_stage_model = None + self.clip_denoised = clip_denoised + self.log_every_t = log_every_t + self.first_stage_key = first_stage_key + self.image_size = image_size # try conv? + self.channels = channels + self.use_positional_encodings = use_positional_encodings + self.model = DiffusionWrapper(unet_config, conditioning_key) + count_params(self.model, verbose=True) + self.use_ema = use_ema + if self.use_ema: + self.model_ema = LitEma(self.model) + print(f"Keeping EMAs of {len(list(self.model_ema.buffers()))}.") + + self.use_scheduler = scheduler_config is not None + if self.use_scheduler: + self.scheduler_config = scheduler_config + + self.v_posterior = v_posterior + self.original_elbo_weight = original_elbo_weight + self.l_simple_weight = l_simple_weight + + if monitor is not None: + self.monitor = monitor + self.make_it_fit = make_it_fit + if reset_ema: assert exists(ckpt_path) + if ckpt_path is not None: + self.init_from_ckpt(ckpt_path, ignore_keys=ignore_keys, only_model=load_only_unet) + if reset_ema: + assert self.use_ema + print(f"Resetting ema to pure model weights. This is useful when restoring from an ema-only checkpoint.") + self.model_ema = LitEma(self.model) + if reset_num_ema_updates: + print(" +++++++++++ WARNING: RESETTING NUM_EMA UPDATES TO ZERO +++++++++++ ") + assert self.use_ema + self.model_ema.reset_num_updates() + + self.register_schedule(given_betas=given_betas, beta_schedule=beta_schedule, timesteps=timesteps, + linear_start=linear_start, linear_end=linear_end, cosine_s=cosine_s) + + self.loss_type = loss_type + + self.learn_logvar = learn_logvar + logvar = torch.full(fill_value=logvar_init, size=(self.num_timesteps,)) + if self.learn_logvar: + self.logvar = nn.Parameter(self.logvar, requires_grad=True) + else: + self.register_buffer('logvar', logvar) + + self.ucg_training = ucg_training or dict() + if self.ucg_training: + self.ucg_prng = np.random.RandomState() + + def register_schedule(self, given_betas=None, beta_schedule="linear", timesteps=1000, + linear_start=1e-4, linear_end=2e-2, cosine_s=8e-3): + if exists(given_betas): + betas = given_betas + else: + betas = make_beta_schedule(beta_schedule, timesteps, linear_start=linear_start, linear_end=linear_end, + cosine_s=cosine_s) + alphas = 1. - betas + alphas_cumprod = np.cumprod(alphas, axis=0) + # np.save('1.npy', alphas_cumprod) + alphas_cumprod_prev = np.append(1., alphas_cumprod[:-1]) + + timesteps, = betas.shape + self.num_timesteps = int(timesteps) + self.linear_start = linear_start + self.linear_end = linear_end + assert alphas_cumprod.shape[0] == self.num_timesteps, 'alphas have to be defined for each timestep' + + to_torch = partial(torch.tensor, dtype=torch.float32) + + self.register_buffer('betas', to_torch(betas)) + self.register_buffer('alphas_cumprod', to_torch(alphas_cumprod)) + self.register_buffer('alphas_cumprod_prev', to_torch(alphas_cumprod_prev)) + + # calculations for diffusion q(x_t | x_{t-1}) and others + self.register_buffer('sqrt_alphas_cumprod', to_torch(np.sqrt(alphas_cumprod))) + self.register_buffer('sqrt_one_minus_alphas_cumprod', to_torch(np.sqrt(1. - alphas_cumprod))) + self.register_buffer('log_one_minus_alphas_cumprod', to_torch(np.log(1. - alphas_cumprod))) + self.register_buffer('sqrt_recip_alphas_cumprod', to_torch(np.sqrt(1. / alphas_cumprod))) + self.register_buffer('sqrt_recipm1_alphas_cumprod', to_torch(np.sqrt(1. / alphas_cumprod - 1))) + + # calculations for posterior q(x_{t-1} | x_t, x_0) + posterior_variance = (1 - self.v_posterior) * betas * (1. - alphas_cumprod_prev) / ( + 1. - alphas_cumprod) + self.v_posterior * betas + # above: equal to 1. / (1. / (1. - alpha_cumprod_tm1) + alpha_t / beta_t) + self.register_buffer('posterior_variance', to_torch(posterior_variance)) + # below: log calculation clipped because the posterior variance is 0 at the beginning of the diffusion chain + self.register_buffer('posterior_log_variance_clipped', to_torch(np.log(np.maximum(posterior_variance, 1e-20)))) + self.register_buffer('posterior_mean_coef1', to_torch( + betas * np.sqrt(alphas_cumprod_prev) / (1. - alphas_cumprod))) + self.register_buffer('posterior_mean_coef2', to_torch( + (1. - alphas_cumprod_prev) * np.sqrt(alphas) / (1. - alphas_cumprod))) + + if self.parameterization == "eps": + lvlb_weights = self.betas ** 2 / ( + 2 * self.posterior_variance * to_torch(alphas) * (1 - self.alphas_cumprod)) + elif self.parameterization == "x0": + lvlb_weights = 0.5 * np.sqrt(torch.Tensor(alphas_cumprod)) / (2. * 1 - torch.Tensor(alphas_cumprod)) + elif self.parameterization == "v": + lvlb_weights = torch.ones_like(self.betas ** 2 / ( + 2 * self.posterior_variance * to_torch(alphas) * (1 - self.alphas_cumprod))) + else: + raise NotImplementedError("mu not supported") + lvlb_weights[0] = lvlb_weights[1] + self.register_buffer('lvlb_weights', lvlb_weights, persistent=False) + assert not torch.isnan(self.lvlb_weights).all() + + @contextmanager + def ema_scope(self, context=None): + if self.use_ema: + self.model_ema.store(self.model.parameters()) + self.model_ema.copy_to(self.model) + if context is not None: + print(f"{context}: Switched to EMA weights") + try: + yield None + finally: + if self.use_ema: + self.model_ema.restore(self.model.parameters()) + if context is not None: + print(f"{context}: Restored training weights") + + @torch.no_grad() + def init_from_ckpt(self, path, ignore_keys=list(), only_model=False): + sd = torch.load(path, map_location="cpu") + if "state_dict" in list(sd.keys()): + sd = sd["state_dict"] + keys = list(sd.keys()) + for k in keys: + for ik in ignore_keys: + if k.startswith(ik): + print("Deleting key {} from state_dict.".format(k)) + del sd[k] + if self.make_it_fit: + n_params = len([name for name, _ in + itertools.chain(self.named_parameters(), + self.named_buffers())]) + for name, param in tqdm( + itertools.chain(self.named_parameters(), + self.named_buffers()), + desc="Fitting old weights to new weights", + total=n_params + ): + if not name in sd: + continue + old_shape = sd[name].shape + new_shape = param.shape + assert len(old_shape) == len(new_shape) + if len(new_shape) > 2: + # we only modify first two axes + assert new_shape[2:] == old_shape[2:] + # assumes first axis corresponds to output dim + if not new_shape == old_shape: + new_param = param.clone() + old_param = sd[name] + if len(new_shape) == 1: + for i in range(new_param.shape[0]): + new_param[i] = old_param[i % old_shape[0]] + elif len(new_shape) >= 2: + for i in range(new_param.shape[0]): + for j in range(new_param.shape[1]): + new_param[i, j] = old_param[i % old_shape[0], j % old_shape[1]] + + n_used_old = torch.ones(old_shape[1]) + for j in range(new_param.shape[1]): + n_used_old[j % old_shape[1]] += 1 + n_used_new = torch.zeros(new_shape[1]) + for j in range(new_param.shape[1]): + n_used_new[j] = n_used_old[j % old_shape[1]] + + n_used_new = n_used_new[None, :] + while len(n_used_new.shape) < len(new_shape): + n_used_new = n_used_new.unsqueeze(-1) + new_param /= n_used_new + + sd[name] = new_param + + missing, unexpected = self.load_state_dict(sd, strict=False) if not only_model else self.model.load_state_dict( + sd, strict=False) + print(f"Restored from {path} with {len(missing)} missing and {len(unexpected)} unexpected keys") + if len(missing) > 0: + print(f"Missing Keys:\n {missing}") + if len(unexpected) > 0: + print(f"\nUnexpected Keys:\n {unexpected}") + + def q_mean_variance(self, x_start, t): + """ + Get the distribution q(x_t | x_0). + :param x_start: the [N x C x ...] tensor of noiseless inputs. + :param t: the number of diffusion steps (minus 1). Here, 0 means one step. + :return: A tuple (mean, variance, log_variance), all of x_start's shape. + """ + mean = (extract_into_tensor(self.sqrt_alphas_cumprod, t, x_start.shape) * x_start) + variance = extract_into_tensor(1.0 - self.alphas_cumprod, t, x_start.shape) + log_variance = extract_into_tensor(self.log_one_minus_alphas_cumprod, t, x_start.shape) + return mean, variance, log_variance + + def predict_start_from_noise(self, x_t, t, noise): + return ( + extract_into_tensor(self.sqrt_recip_alphas_cumprod, t, x_t.shape) * x_t - + extract_into_tensor(self.sqrt_recipm1_alphas_cumprod, t, x_t.shape) * noise + ) + + def predict_start_from_z_and_v(self, x_t, t, v): + # self.register_buffer('sqrt_alphas_cumprod', to_torch(np.sqrt(alphas_cumprod))) + # self.register_buffer('sqrt_one_minus_alphas_cumprod', to_torch(np.sqrt(1. - alphas_cumprod))) + return ( + extract_into_tensor(self.sqrt_alphas_cumprod, t, x_t.shape) * x_t - + extract_into_tensor(self.sqrt_one_minus_alphas_cumprod, t, x_t.shape) * v + ) + + def predict_eps_from_z_and_v(self, x_t, t, v): + return ( + extract_into_tensor(self.sqrt_alphas_cumprod, t, x_t.shape) * v + + extract_into_tensor(self.sqrt_one_minus_alphas_cumprod, t, x_t.shape) * x_t + ) + + def q_posterior(self, x_start, x_t, t): + posterior_mean = ( + extract_into_tensor(self.posterior_mean_coef1, t, x_t.shape) * x_start + + extract_into_tensor(self.posterior_mean_coef2, t, x_t.shape) * x_t + ) + posterior_variance = extract_into_tensor(self.posterior_variance, t, x_t.shape) + posterior_log_variance_clipped = extract_into_tensor(self.posterior_log_variance_clipped, t, x_t.shape) + return posterior_mean, posterior_variance, posterior_log_variance_clipped + + def p_mean_variance(self, x, t, clip_denoised: bool): + model_out = self.model(x, t) + if self.parameterization == "eps": + x_recon = self.predict_start_from_noise(x, t=t, noise=model_out) + elif self.parameterization == "x0": + x_recon = model_out + if clip_denoised: + x_recon.clamp_(-1., 1.) + + model_mean, posterior_variance, posterior_log_variance = self.q_posterior(x_start=x_recon, x_t=x, t=t) + return model_mean, posterior_variance, posterior_log_variance + + @torch.no_grad() + def p_sample(self, x, t, clip_denoised=True, repeat_noise=False): + b, *_, device = *x.shape, x.device + model_mean, _, model_log_variance = self.p_mean_variance(x=x, t=t, clip_denoised=clip_denoised) + noise = noise_like(x.shape, device, repeat_noise) + # no noise when t == 0 + nonzero_mask = (1 - (t == 0).float()).reshape(b, *((1,) * (len(x.shape) - 1))) + return model_mean + nonzero_mask * (0.5 * model_log_variance).exp() * noise + + @torch.no_grad() + def p_sample_loop(self, shape, return_intermediates=False): + device = self.betas.device + b = shape[0] + img = torch.randn(shape, device=device) + intermediates = [img] + for i in tqdm(reversed(range(0, self.num_timesteps)), desc='Sampling t', total=self.num_timesteps): + img = self.p_sample(img, torch.full((b,), i, device=device, dtype=torch.long), + clip_denoised=self.clip_denoised) + if i % self.log_every_t == 0 or i == self.num_timesteps - 1: + intermediates.append(img) + if return_intermediates: + return img, intermediates + return img + + @torch.no_grad() + def sample(self, batch_size=16, return_intermediates=False): + image_size = self.image_size + channels = self.channels + return self.p_sample_loop((batch_size, channels, image_size, image_size), + return_intermediates=return_intermediates) + + def q_sample(self, x_start, t, noise=None): + noise = default(noise, lambda: torch.randn_like(x_start)) + return (extract_into_tensor(self.sqrt_alphas_cumprod, t, x_start.shape) * x_start + + extract_into_tensor(self.sqrt_one_minus_alphas_cumprod, t, x_start.shape) * noise) + + def get_v(self, x, noise, t): + return ( + extract_into_tensor(self.sqrt_alphas_cumprod, t, x.shape) * noise - + extract_into_tensor(self.sqrt_one_minus_alphas_cumprod, t, x.shape) * x + ) + + def get_loss(self, pred, target, mean=True): + if self.loss_type == 'l1': + loss = (target - pred).abs() + if mean: + loss = loss.mean() + elif self.loss_type == 'l2': + if mean: + loss = torch.nn.functional.mse_loss(target, pred) + else: + loss = torch.nn.functional.mse_loss(target, pred, reduction='none') + else: + raise NotImplementedError("unknown loss type '{loss_type}'") + + return loss + + def p_losses(self, x_start, t, noise=None): + noise = default(noise, lambda: torch.randn_like(x_start)) + x_noisy = self.q_sample(x_start=x_start, t=t, noise=noise) + model_out = self.model(x_noisy, t) + + loss_dict = {} + if self.parameterization == "eps": + target = noise + elif self.parameterization == "x0": + target = x_start + elif self.parameterization == "v": + target = self.get_v(x_start, noise, t) + else: + raise NotImplementedError(f"Parameterization {self.parameterization} not yet supported") + + loss = self.get_loss(model_out, target, mean=False).mean(dim=[1, 2, 3]) + + log_prefix = 'train' if self.training else 'val' + + loss_dict.update({f'{log_prefix}/loss_simple': loss.mean()}) + loss_simple = loss.mean() * self.l_simple_weight + + loss_vlb = (self.lvlb_weights[t] * loss).mean() + loss_dict.update({f'{log_prefix}/loss_vlb': loss_vlb}) + + loss = loss_simple + self.original_elbo_weight * loss_vlb + + loss_dict.update({f'{log_prefix}/loss': loss}) + + return loss, loss_dict + + def forward(self, x, *args, **kwargs): + # b, c, h, w, device, img_size, = *x.shape, x.device, self.image_size + # assert h == img_size and w == img_size, f'height and width of image must be {img_size}' + t = torch.randint(0, self.num_timesteps, (x.shape[0],), device=self.device).long() + return self.p_losses(x, t, *args, **kwargs) + + def get_input(self, batch, k): + x = batch[k] + if len(x.shape) == 3: + x = x[..., None] + x = rearrange(x, 'b h w c -> b c h w') + x = x.to(memory_format=torch.contiguous_format).float() + return x + + def shared_step(self, batch): + x = self.get_input(batch, self.first_stage_key) + loss, loss_dict = self(x) + return loss, loss_dict + + def training_step(self, batch, batch_idx): + for k in self.ucg_training: + p = self.ucg_training[k]["p"] + val = self.ucg_training[k]["val"] + if val is None: + val = "" + for i in range(len(batch[k])): + if self.ucg_prng.choice(2, p=[1 - p, p]): + batch[k][i] = val + + loss, loss_dict = self.shared_step(batch) + + self.log_dict(loss_dict, prog_bar=True, + logger=True, on_step=True, on_epoch=True) + + self.log("global_step", self.global_step, + prog_bar=True, logger=True, on_step=True, on_epoch=False) + + if self.use_scheduler: + lr = self.optimizers().param_groups[0]['lr'] + self.log('lr_abs', lr, prog_bar=True, logger=True, on_step=True, on_epoch=False) + + return loss + + @torch.no_grad() + def validation_step(self, batch, batch_idx): + _, loss_dict_no_ema = self.shared_step(batch) + with self.ema_scope(): + _, loss_dict_ema = self.shared_step(batch) + loss_dict_ema = {key + '_ema': loss_dict_ema[key] for key in loss_dict_ema} + self.log_dict(loss_dict_no_ema, prog_bar=False, logger=True, on_step=False, on_epoch=True) + self.log_dict(loss_dict_ema, prog_bar=False, logger=True, on_step=False, on_epoch=True) + + def on_train_batch_end(self, *args, **kwargs): + if self.use_ema: + self.model_ema(self.model) + + def _get_rows_from_list(self, samples): + n_imgs_per_row = len(samples) + denoise_grid = rearrange(samples, 'n b c h w -> b n c h w') + denoise_grid = rearrange(denoise_grid, 'b n c h w -> (b n) c h w') + denoise_grid = make_grid(denoise_grid, nrow=n_imgs_per_row) + return denoise_grid + + @torch.no_grad() + def log_images(self, batch, N=8, n_row=2, sample=True, return_keys=None, **kwargs): + log = dict() + x = self.get_input(batch, self.first_stage_key) + N = min(x.shape[0], N) + n_row = min(x.shape[0], n_row) + x = x.to(self.device)[:N] + log["inputs"] = x + + # get diffusion row + diffusion_row = list() + x_start = x[:n_row] + + for t in range(self.num_timesteps): + if t % self.log_every_t == 0 or t == self.num_timesteps - 1: + t = repeat(torch.tensor([t]), '1 -> b', b=n_row) + t = t.to(self.device).long() + noise = torch.randn_like(x_start) + x_noisy = self.q_sample(x_start=x_start, t=t, noise=noise) + diffusion_row.append(x_noisy) + + log["diffusion_row"] = self._get_rows_from_list(diffusion_row) + + if sample: + # get denoise row + with self.ema_scope("Plotting"): + samples, denoise_row = self.sample(batch_size=N, return_intermediates=True) + + log["samples"] = samples + log["denoise_row"] = self._get_rows_from_list(denoise_row) + + if return_keys: + if np.intersect1d(list(log.keys()), return_keys).shape[0] == 0: + return log + else: + return {key: log[key] for key in return_keys} + return log + + def configure_optimizers(self): + lr = self.learning_rate + params = list(self.model.parameters()) + if self.learn_logvar: + params = params + [self.logvar] + opt = torch.optim.AdamW(params, lr=lr) + return opt + + +class LatentDiffusion(DDPM): + """main class""" + + def __init__(self, + first_stage_config, + cond_stage_config, + num_timesteps_cond=None, + cond_stage_key="image", + cond_stage_trainable=False, + concat_mode=True, + cond_stage_forward=None, + conditioning_key=None, + scale_factor=1.0, + scale_by_std=False, + force_null_conditioning=False, + *args, **kwargs): + self.force_null_conditioning = force_null_conditioning + self.num_timesteps_cond = default(num_timesteps_cond, 1) + self.scale_by_std = scale_by_std + assert self.num_timesteps_cond <= kwargs['timesteps'] + # for backwards compatibility after implementation of DiffusionWrapper + if conditioning_key is None: + conditioning_key = 'concat' if concat_mode else 'crossattn' + if cond_stage_config == '__is_unconditional__' and not self.force_null_conditioning: + conditioning_key = None + ckpt_path = kwargs.pop("ckpt_path", None) + reset_ema = kwargs.pop("reset_ema", False) + reset_num_ema_updates = kwargs.pop("reset_num_ema_updates", False) + ignore_keys = kwargs.pop("ignore_keys", []) + super().__init__(conditioning_key=conditioning_key, *args, **kwargs) + self.concat_mode = concat_mode + self.cond_stage_trainable = cond_stage_trainable + self.cond_stage_key = cond_stage_key + try: + self.num_downs = len(first_stage_config.params.ddconfig.ch_mult) - 1 + except: + self.num_downs = 0 + if not scale_by_std: + self.scale_factor = scale_factor + else: + self.register_buffer('scale_factor', torch.tensor(scale_factor)) + self.instantiate_first_stage(first_stage_config) + self.instantiate_cond_stage(cond_stage_config) + self.cond_stage_forward = cond_stage_forward + self.clip_denoised = False + self.bbox_tokenizer = None + + self.restarted_from_ckpt = False + if ckpt_path is not None: + self.init_from_ckpt(ckpt_path, ignore_keys) + self.restarted_from_ckpt = True + if reset_ema: + assert self.use_ema + print( + f"Resetting ema to pure model weights. This is useful when restoring from an ema-only checkpoint.") + self.model_ema = LitEma(self.model) + if reset_num_ema_updates: + print(" +++++++++++ WARNING: RESETTING NUM_EMA UPDATES TO ZERO +++++++++++ ") + assert self.use_ema + self.model_ema.reset_num_updates() + + def make_cond_schedule(self, ): + self.cond_ids = torch.full(size=(self.num_timesteps,), fill_value=self.num_timesteps - 1, dtype=torch.long) + ids = torch.round(torch.linspace(0, self.num_timesteps - 1, self.num_timesteps_cond)).long() + self.cond_ids[:self.num_timesteps_cond] = ids + + @rank_zero_only + @torch.no_grad() + def on_train_batch_start(self, batch, batch_idx, dataloader_idx): + # only for very first batch + if self.scale_by_std and self.current_epoch == 0 and self.global_step == 0 and batch_idx == 0 and not self.restarted_from_ckpt: + assert self.scale_factor == 1., 'rather not use custom rescaling and std-rescaling simultaneously' + # set rescale weight to 1./std of encodings + print("### USING STD-RESCALING ###") + x = super().get_input(batch, self.first_stage_key) + x = x.to(self.device) + encoder_posterior = self.encode_first_stage(x) + z = self.get_first_stage_encoding(encoder_posterior).detach() + del self.scale_factor + self.register_buffer('scale_factor', 1. / z.flatten().std()) + print(f"setting self.scale_factor to {self.scale_factor}") + print("### USING STD-RESCALING ###") + + def register_schedule(self, + given_betas=None, beta_schedule="linear", timesteps=1000, + linear_start=1e-4, linear_end=2e-2, cosine_s=8e-3): + super().register_schedule(given_betas, beta_schedule, timesteps, linear_start, linear_end, cosine_s) + + self.shorten_cond_schedule = self.num_timesteps_cond > 1 + if self.shorten_cond_schedule: + self.make_cond_schedule() + + def instantiate_first_stage(self, config): + model = instantiate_from_config(config) + self.first_stage_model = model.eval() + self.first_stage_model.train = disabled_train + for param in self.first_stage_model.parameters(): + param.requires_grad = False + + def instantiate_cond_stage(self, config): + if not self.cond_stage_trainable: + if config == "__is_first_stage__": + print("Using first stage also as cond stage.") + self.cond_stage_model = self.first_stage_model + elif config == "__is_unconditional__": + print(f"Training {self.__class__.__name__} as an unconditional model.") + self.cond_stage_model = None + # self.be_unconditional = True + else: + model = instantiate_from_config(config) + self.cond_stage_model = model.eval() + self.cond_stage_model.train = disabled_train + for param in self.cond_stage_model.parameters(): + param.requires_grad = False + else: + assert config != '__is_first_stage__' + assert config != '__is_unconditional__' + model = instantiate_from_config(config) + self.cond_stage_model = model + + def _get_denoise_row_from_list(self, samples, desc='', force_no_decoder_quantization=False): + denoise_row = [] + for zd in tqdm(samples, desc=desc): + denoise_row.append(self.decode_first_stage(zd.to(self.device), + force_not_quantize=force_no_decoder_quantization)) + n_imgs_per_row = len(denoise_row) + denoise_row = torch.stack(denoise_row) # n_log_step, n_row, C, H, W + denoise_grid = rearrange(denoise_row, 'n b c h w -> b n c h w') + denoise_grid = rearrange(denoise_grid, 'b n c h w -> (b n) c h w') + denoise_grid = make_grid(denoise_grid, nrow=n_imgs_per_row) + return denoise_grid + + def get_first_stage_encoding(self, encoder_posterior): + if isinstance(encoder_posterior, DiagonalGaussianDistribution): + z = encoder_posterior.sample() + elif isinstance(encoder_posterior, torch.Tensor): + z = encoder_posterior + else: + raise NotImplementedError(f"encoder_posterior of type '{type(encoder_posterior)}' not yet implemented") + return self.scale_factor * z + + def get_learned_conditioning(self, c): + if self.cond_stage_forward is None: + if hasattr(self.cond_stage_model, 'encode') and callable(self.cond_stage_model.encode): + c = self.cond_stage_model.encode(c) + if isinstance(c, DiagonalGaussianDistribution): + c = c.mode() + else: + c = self.cond_stage_model(c) + else: + assert hasattr(self.cond_stage_model, self.cond_stage_forward) + c = getattr(self.cond_stage_model, self.cond_stage_forward)(c) + return c + + def meshgrid(self, h, w): + y = torch.arange(0, h).view(h, 1, 1).repeat(1, w, 1) + x = torch.arange(0, w).view(1, w, 1).repeat(h, 1, 1) + + arr = torch.cat([y, x], dim=-1) + return arr + + def delta_border(self, h, w): + """ + :param h: height + :param w: width + :return: normalized distance to image border, + wtith min distance = 0 at border and max dist = 0.5 at image center + """ + lower_right_corner = torch.tensor([h - 1, w - 1]).view(1, 1, 2) + arr = self.meshgrid(h, w) / lower_right_corner + dist_left_up = torch.min(arr, dim=-1, keepdims=True)[0] + dist_right_down = torch.min(1 - arr, dim=-1, keepdims=True)[0] + edge_dist = torch.min(torch.cat([dist_left_up, dist_right_down], dim=-1), dim=-1)[0] + return edge_dist + + def get_weighting(self, h, w, Ly, Lx, device): + weighting = self.delta_border(h, w) + weighting = torch.clip(weighting, self.split_input_params["clip_min_weight"], + self.split_input_params["clip_max_weight"], ) + weighting = weighting.view(1, h * w, 1).repeat(1, 1, Ly * Lx).to(device) + + if self.split_input_params["tie_braker"]: + L_weighting = self.delta_border(Ly, Lx) + L_weighting = torch.clip(L_weighting, + self.split_input_params["clip_min_tie_weight"], + self.split_input_params["clip_max_tie_weight"]) + + L_weighting = L_weighting.view(1, 1, Ly * Lx).to(device) + weighting = weighting * L_weighting + return weighting + + def get_fold_unfold(self, x, kernel_size, stride, uf=1, df=1): # todo load once not every time, shorten code + """ + :param x: img of size (bs, c, h, w) + :return: n img crops of size (n, bs, c, kernel_size[0], kernel_size[1]) + """ + bs, nc, h, w = x.shape + + # number of crops in image + Ly = (h - kernel_size[0]) // stride[0] + 1 + Lx = (w - kernel_size[1]) // stride[1] + 1 + + if uf == 1 and df == 1: + fold_params = dict(kernel_size=kernel_size, dilation=1, padding=0, stride=stride) + unfold = torch.nn.Unfold(**fold_params) + + fold = torch.nn.Fold(output_size=x.shape[2:], **fold_params) + + weighting = self.get_weighting(kernel_size[0], kernel_size[1], Ly, Lx, x.device).to(x.dtype) + normalization = fold(weighting).view(1, 1, h, w) # normalizes the overlap + weighting = weighting.view((1, 1, kernel_size[0], kernel_size[1], Ly * Lx)) + + elif uf > 1 and df == 1: + fold_params = dict(kernel_size=kernel_size, dilation=1, padding=0, stride=stride) + unfold = torch.nn.Unfold(**fold_params) + + fold_params2 = dict(kernel_size=(kernel_size[0] * uf, kernel_size[0] * uf), + dilation=1, padding=0, + stride=(stride[0] * uf, stride[1] * uf)) + fold = torch.nn.Fold(output_size=(x.shape[2] * uf, x.shape[3] * uf), **fold_params2) + + weighting = self.get_weighting(kernel_size[0] * uf, kernel_size[1] * uf, Ly, Lx, x.device).to(x.dtype) + normalization = fold(weighting).view(1, 1, h * uf, w * uf) # normalizes the overlap + weighting = weighting.view((1, 1, kernel_size[0] * uf, kernel_size[1] * uf, Ly * Lx)) + + elif df > 1 and uf == 1: + fold_params = dict(kernel_size=kernel_size, dilation=1, padding=0, stride=stride) + unfold = torch.nn.Unfold(**fold_params) + + fold_params2 = dict(kernel_size=(kernel_size[0] // df, kernel_size[0] // df), + dilation=1, padding=0, + stride=(stride[0] // df, stride[1] // df)) + fold = torch.nn.Fold(output_size=(x.shape[2] // df, x.shape[3] // df), **fold_params2) + + weighting = self.get_weighting(kernel_size[0] // df, kernel_size[1] // df, Ly, Lx, x.device).to(x.dtype) + normalization = fold(weighting).view(1, 1, h // df, w // df) # normalizes the overlap + weighting = weighting.view((1, 1, kernel_size[0] // df, kernel_size[1] // df, Ly * Lx)) + + else: + raise NotImplementedError + + return fold, unfold, normalization, weighting + + @torch.no_grad() + def get_input(self, batch, k, return_first_stage_outputs=False, force_c_encode=False, + cond_key=None, return_original_cond=False, bs=None, return_x=False, mask_k=None): + x = super().get_input(batch, k) + if bs is not None: + x = x[:bs] + x = x.to(self.device) + encoder_posterior = self.encode_first_stage(x) + z = self.get_first_stage_encoding(encoder_posterior).detach() + + if mask_k is not None: + mx = super().get_input(batch, mask_k) + if bs is not None: + mx = mx[:bs] + mx = mx.to(self.device) + encoder_posterior = self.encode_first_stage(mx) + mx = self.get_first_stage_encoding(encoder_posterior).detach() + + if self.model.conditioning_key is not None and not self.force_null_conditioning: + if cond_key is None: + cond_key = self.cond_stage_key + if cond_key != self.first_stage_key: + if cond_key in ['caption', 'coordinates_bbox', "txt"]: + xc = batch[cond_key] + elif cond_key in ['class_label', 'cls']: + xc = batch + else: + xc = super().get_input(batch, cond_key).to(self.device) + else: + xc = x + if not self.cond_stage_trainable or force_c_encode: + if isinstance(xc, dict) or isinstance(xc, list): + c = self.get_learned_conditioning(xc) + else: + c = self.get_learned_conditioning(xc.to(self.device)) + else: + c = xc + if bs is not None: + c = c[:bs] + + if self.use_positional_encodings: + pos_x, pos_y = self.compute_latent_shifts(batch) + ckey = __conditioning_keys__[self.model.conditioning_key] + c = {ckey: c, 'pos_x': pos_x, 'pos_y': pos_y} + + else: + c = None + xc = None + if self.use_positional_encodings: + pos_x, pos_y = self.compute_latent_shifts(batch) + c = {'pos_x': pos_x, 'pos_y': pos_y} + out = [z, c] + if return_first_stage_outputs: + xrec = self.decode_first_stage(z) + out.extend([x, xrec]) + if return_x: + out.extend([x]) + if return_original_cond: + out.append(xc) + if mask_k: + out.append(mx) + return out + + @torch.no_grad() + def decode_first_stage(self, z, predict_cids=False, force_not_quantize=False): + if predict_cids: + if z.dim() == 4: + z = torch.argmax(z.exp(), dim=1).long() + z = self.first_stage_model.quantize.get_codebook_entry(z, shape=None) + z = rearrange(z, 'b h w c -> b c h w').contiguous() + + z = 1. / self.scale_factor * z + return self.first_stage_model.decode(z) + + def decode_first_stage_grad(self, z, predict_cids=False, force_not_quantize=False): + if predict_cids: + if z.dim() == 4: + z = torch.argmax(z.exp(), dim=1).long() + z = self.first_stage_model.quantize.get_codebook_entry(z, shape=None) + z = rearrange(z, 'b h w c -> b c h w').contiguous() + + z = 1. / self.scale_factor * z + return self.first_stage_model.decode(z) + + @torch.no_grad() + def encode_first_stage(self, x): + return self.first_stage_model.encode(x) + + def shared_step(self, batch, **kwargs): + x, c = self.get_input(batch, self.first_stage_key) + loss = self(x, c) + return loss + + def forward(self, x, c, *args, **kwargs): + t = torch.randint(0, self.num_timesteps, (x.shape[0],), device=self.device).long() + # t = torch.randint(500, 501, (x.shape[0],), device=self.device).long() + if self.model.conditioning_key is not None: + assert c is not None + if self.cond_stage_trainable: + c = self.get_learned_conditioning(c) + if self.shorten_cond_schedule: # TODO: drop this option + tc = self.cond_ids[t].to(self.device) + c = self.q_sample(x_start=c, t=tc, noise=torch.randn_like(c.float())) + return self.p_losses(x, c, t, *args, **kwargs) + + def apply_model(self, x_noisy, t, cond, return_ids=False): + if isinstance(cond, dict): + # hybrid case, cond is expected to be a dict + pass + else: + if not isinstance(cond, list): + cond = [cond] + key = 'c_concat' if self.model.conditioning_key == 'concat' else 'c_crossattn' + cond = {key: cond} + + x_recon = self.model(x_noisy, t, **cond) + + if isinstance(x_recon, tuple) and not return_ids: + return x_recon[0] + else: + return x_recon + + def _predict_eps_from_xstart(self, x_t, t, pred_xstart): + return (extract_into_tensor(self.sqrt_recip_alphas_cumprod, t, x_t.shape) * x_t - pred_xstart) / \ + extract_into_tensor(self.sqrt_recipm1_alphas_cumprod, t, x_t.shape) + + def _prior_bpd(self, x_start): + """ + Get the prior KL term for the variational lower-bound, measured in + bits-per-dim. + This term can't be optimized, as it only depends on the encoder. + :param x_start: the [N x C x ...] tensor of inputs. + :return: a batch of [N] KL values (in bits), one per batch element. + """ + batch_size = x_start.shape[0] + t = torch.tensor([self.num_timesteps - 1] * batch_size, device=x_start.device) + qt_mean, _, qt_log_variance = self.q_mean_variance(x_start, t) + kl_prior = normal_kl(mean1=qt_mean, logvar1=qt_log_variance, mean2=0.0, logvar2=0.0) + return mean_flat(kl_prior) / np.log(2.0) + + def p_mean_variance(self, x, c, t, clip_denoised: bool, return_codebook_ids=False, quantize_denoised=False, + return_x0=False, score_corrector=None, corrector_kwargs=None): + t_in = t + model_out = self.apply_model(x, t_in, c, return_ids=return_codebook_ids) + + if score_corrector is not None: + assert self.parameterization == "eps" + model_out = score_corrector.modify_score(self, model_out, x, t, c, **corrector_kwargs) + + if return_codebook_ids: + model_out, logits = model_out + + if self.parameterization == "eps": + x_recon = self.predict_start_from_noise(x, t=t, noise=model_out) + elif self.parameterization == "x0": + x_recon = model_out + else: + raise NotImplementedError() + + if clip_denoised: + x_recon.clamp_(-1., 1.) + if quantize_denoised: + x_recon, _, [_, _, indices] = self.first_stage_model.quantize(x_recon) + model_mean, posterior_variance, posterior_log_variance = self.q_posterior(x_start=x_recon, x_t=x, t=t) + if return_codebook_ids: + return model_mean, posterior_variance, posterior_log_variance, logits + elif return_x0: + return model_mean, posterior_variance, posterior_log_variance, x_recon + else: + return model_mean, posterior_variance, posterior_log_variance + + @torch.no_grad() + def p_sample(self, x, c, t, clip_denoised=False, repeat_noise=False, + return_codebook_ids=False, quantize_denoised=False, return_x0=False, + temperature=1., noise_dropout=0., score_corrector=None, corrector_kwargs=None): + b, *_, device = *x.shape, x.device + outputs = self.p_mean_variance(x=x, c=c, t=t, clip_denoised=clip_denoised, + return_codebook_ids=return_codebook_ids, + quantize_denoised=quantize_denoised, + return_x0=return_x0, + score_corrector=score_corrector, corrector_kwargs=corrector_kwargs) + if return_codebook_ids: + raise DeprecationWarning("Support dropped.") + model_mean, _, model_log_variance, logits = outputs + elif return_x0: + model_mean, _, model_log_variance, x0 = outputs + else: + model_mean, _, model_log_variance = outputs + + noise = noise_like(x.shape, device, repeat_noise) * temperature + if noise_dropout > 0.: + noise = torch.nn.functional.dropout(noise, p=noise_dropout) + # no noise when t == 0 + nonzero_mask = (1 - (t == 0).float()).reshape(b, *((1,) * (len(x.shape) - 1))) + + if return_codebook_ids: + return model_mean + nonzero_mask * (0.5 * model_log_variance).exp() * noise, logits.argmax(dim=1) + if return_x0: + return model_mean + nonzero_mask * (0.5 * model_log_variance).exp() * noise, x0 + else: + return model_mean + nonzero_mask * (0.5 * model_log_variance).exp() * noise + + @torch.no_grad() + def progressive_denoising(self, cond, shape, verbose=True, callback=None, quantize_denoised=False, + img_callback=None, mask=None, x0=None, temperature=1., noise_dropout=0., + score_corrector=None, corrector_kwargs=None, batch_size=None, x_T=None, start_T=None, + log_every_t=None): + if not log_every_t: + log_every_t = self.log_every_t + timesteps = self.num_timesteps + if batch_size is not None: + b = batch_size if batch_size is not None else shape[0] + shape = [batch_size] + list(shape) + else: + b = batch_size = shape[0] + if x_T is None: + img = torch.randn(shape, device=self.device) + else: + img = x_T + intermediates = [] + if cond is not None: + if isinstance(cond, dict): + cond = {key: cond[key][:batch_size] if not isinstance(cond[key], list) else + list(map(lambda x: x[:batch_size], cond[key])) for key in cond} + else: + cond = [c[:batch_size] for c in cond] if isinstance(cond, list) else cond[:batch_size] + + if start_T is not None: + timesteps = min(timesteps, start_T) + iterator = tqdm(reversed(range(0, timesteps)), desc='Progressive Generation', + total=timesteps) if verbose else reversed( + range(0, timesteps)) + if type(temperature) == float: + temperature = [temperature] * timesteps + + for i in iterator: + ts = torch.full((b,), i, device=self.device, dtype=torch.long) + if self.shorten_cond_schedule: + assert self.model.conditioning_key != 'hybrid' + tc = self.cond_ids[ts].to(cond.device) + cond = self.q_sample(x_start=cond, t=tc, noise=torch.randn_like(cond)) + + img, x0_partial = self.p_sample(img, cond, ts, + clip_denoised=self.clip_denoised, + quantize_denoised=quantize_denoised, return_x0=True, + temperature=temperature[i], noise_dropout=noise_dropout, + score_corrector=score_corrector, corrector_kwargs=corrector_kwargs) + if mask is not None: + assert x0 is not None + img_orig = self.q_sample(x0, ts) + img = img_orig * mask + (1. - mask) * img + + if i % log_every_t == 0 or i == timesteps - 1: + intermediates.append(x0_partial) + if callback: callback(i) + if img_callback: img_callback(img, i) + return img, intermediates + + @torch.no_grad() + def p_sample_loop(self, cond, shape, return_intermediates=False, + x_T=None, verbose=True, callback=None, timesteps=None, quantize_denoised=False, + mask=None, x0=None, img_callback=None, start_T=None, + log_every_t=None): + + if not log_every_t: + log_every_t = self.log_every_t + device = self.betas.device + b = shape[0] + if x_T is None: + img = torch.randn(shape, device=device) + else: + img = x_T + + intermediates = [img] + if timesteps is None: + timesteps = self.num_timesteps + + if start_T is not None: + timesteps = min(timesteps, start_T) + iterator = tqdm(reversed(range(0, timesteps)), desc='Sampling t', total=timesteps) if verbose else reversed( + range(0, timesteps)) + + if mask is not None: + assert x0 is not None + assert x0.shape[2:3] == mask.shape[2:3] # spatial size has to match + + for i in iterator: + ts = torch.full((b,), i, device=device, dtype=torch.long) + if self.shorten_cond_schedule: + assert self.model.conditioning_key != 'hybrid' + tc = self.cond_ids[ts].to(cond.device) + cond = self.q_sample(x_start=cond, t=tc, noise=torch.randn_like(cond)) + + img = self.p_sample(img, cond, ts, + clip_denoised=self.clip_denoised, + quantize_denoised=quantize_denoised) + if mask is not None: + img_orig = self.q_sample(x0, ts) + img = img_orig * mask + (1. - mask) * img + + if i % log_every_t == 0 or i == timesteps - 1: + intermediates.append(img) + if callback: callback(i) + if img_callback: img_callback(img, i) + + if return_intermediates: + return img, intermediates + return img + + @torch.no_grad() + def sample(self, cond, batch_size=16, return_intermediates=False, x_T=None, + verbose=True, timesteps=None, quantize_denoised=False, + mask=None, x0=None, shape=None, **kwargs): + if shape is None: + shape = (batch_size, self.channels, self.image_size, self.image_size) + if cond is not None: + if isinstance(cond, dict): + cond = {key: cond[key][:batch_size] if not isinstance(cond[key], list) else + list(map(lambda x: x[:batch_size], cond[key])) for key in cond} + else: + cond = [c[:batch_size] for c in cond] if isinstance(cond, list) else cond[:batch_size] + return self.p_sample_loop(cond, + shape, + return_intermediates=return_intermediates, x_T=x_T, + verbose=verbose, timesteps=timesteps, quantize_denoised=quantize_denoised, + mask=mask, x0=x0) + + @torch.no_grad() + def sample_log(self, cond, batch_size, ddim, ddim_steps, **kwargs): + if ddim: + ddim_sampler = DDIMSampler(self) + shape = (self.channels, self.image_size, self.image_size) + samples, intermediates = ddim_sampler.sample(ddim_steps, batch_size, + shape, cond, verbose=False, **kwargs) + + else: + samples, intermediates = self.sample(cond=cond, batch_size=batch_size, + return_intermediates=True, **kwargs) + + return samples, intermediates + + @torch.no_grad() + def get_unconditional_conditioning(self, batch_size, null_label=None): + if null_label is not None: + xc = null_label + if isinstance(xc, ListConfig): + xc = list(xc) + if isinstance(xc, dict) or isinstance(xc, list): + c = self.get_learned_conditioning(xc) + else: + if hasattr(xc, "to"): + xc = xc.to(self.device) + c = self.get_learned_conditioning(xc) + else: + if self.cond_stage_key in ["class_label", "cls"]: + xc = self.cond_stage_model.get_unconditional_conditioning(batch_size, device=self.device) + return self.get_learned_conditioning(xc) + else: + raise NotImplementedError("todo") + if isinstance(c, list): # in case the encoder gives us a list + for i in range(len(c)): + c[i] = repeat(c[i], '1 ... -> b ...', b=batch_size).to(self.device) + else: + c = repeat(c, '1 ... -> b ...', b=batch_size).to(self.device) + return c + + @torch.no_grad() + def log_images(self, batch, N=8, n_row=4, sample=True, ddim_steps=50, ddim_eta=0., return_keys=None, + quantize_denoised=True, inpaint=True, plot_denoise_rows=False, plot_progressive_rows=True, + plot_diffusion_rows=True, unconditional_guidance_scale=1., unconditional_guidance_label=None, + use_ema_scope=True, + **kwargs): + ema_scope = self.ema_scope if use_ema_scope else nullcontext + use_ddim = ddim_steps is not None + + log = dict() + z, c, x, xrec, xc = self.get_input(batch, self.first_stage_key, + return_first_stage_outputs=True, + force_c_encode=True, + return_original_cond=True, + bs=N) + N = min(x.shape[0], N) + n_row = min(x.shape[0], n_row) + log["inputs"] = x + log["reconstruction"] = xrec + if self.model.conditioning_key is not None: + if hasattr(self.cond_stage_model, "decode"): + xc = self.cond_stage_model.decode(c) + log["conditioning"] = xc + elif self.cond_stage_key in ["caption", "txt"]: + xc = log_txt_as_img((x.shape[2], x.shape[3]), batch[self.cond_stage_key], size=x.shape[2] // 25) + log["conditioning"] = xc + elif self.cond_stage_key in ['class_label', "cls"]: + try: + xc = log_txt_as_img((x.shape[2], x.shape[3]), batch["human_label"], size=x.shape[2] // 25) + log['conditioning'] = xc + except KeyError: + # probably no "human_label" in batch + pass + elif isimage(xc): + log["conditioning"] = xc + if ismap(xc): + log["original_conditioning"] = self.to_rgb(xc) + + if plot_diffusion_rows: + # get diffusion row + diffusion_row = list() + z_start = z[:n_row] + for t in range(self.num_timesteps): + if t % self.log_every_t == 0 or t == self.num_timesteps - 1: + t = repeat(torch.tensor([t]), '1 -> b', b=n_row) + t = t.to(self.device).long() + noise = torch.randn_like(z_start) + z_noisy = self.q_sample(x_start=z_start, t=t, noise=noise) + diffusion_row.append(self.decode_first_stage(z_noisy)) + + diffusion_row = torch.stack(diffusion_row) # n_log_step, n_row, C, H, W + diffusion_grid = rearrange(diffusion_row, 'n b c h w -> b n c h w') + diffusion_grid = rearrange(diffusion_grid, 'b n c h w -> (b n) c h w') + diffusion_grid = make_grid(diffusion_grid, nrow=diffusion_row.shape[0]) + log["diffusion_row"] = diffusion_grid + + if sample: + # get denoise row + with ema_scope("Sampling"): + samples, z_denoise_row = self.sample_log(cond=c, batch_size=N, ddim=use_ddim, + ddim_steps=ddim_steps, eta=ddim_eta) + # samples, z_denoise_row = self.sample(cond=c, batch_size=N, return_intermediates=True) + x_samples = self.decode_first_stage(samples) + log["samples"] = x_samples + if plot_denoise_rows: + denoise_grid = self._get_denoise_row_from_list(z_denoise_row) + log["denoise_row"] = denoise_grid + + if quantize_denoised and not isinstance(self.first_stage_model, AutoencoderKL) and not isinstance( + self.first_stage_model, IdentityFirstStage): + # also display when quantizing x0 while sampling + with ema_scope("Plotting Quantized Denoised"): + samples, z_denoise_row = self.sample_log(cond=c, batch_size=N, ddim=use_ddim, + ddim_steps=ddim_steps, eta=ddim_eta, + quantize_denoised=True) + # samples, z_denoise_row = self.sample(cond=c, batch_size=N, return_intermediates=True, + # quantize_denoised=True) + x_samples = self.decode_first_stage(samples.to(self.device)) + log["samples_x0_quantized"] = x_samples + + if unconditional_guidance_scale > 1.0: + uc = self.get_unconditional_conditioning(N, unconditional_guidance_label) + if self.model.conditioning_key == "crossattn-adm": + uc = {"c_crossattn": [uc], "c_adm": c["c_adm"]} + with ema_scope("Sampling with classifier-free guidance"): + samples_cfg, _ = self.sample_log(cond=c, batch_size=N, ddim=use_ddim, + ddim_steps=ddim_steps, eta=ddim_eta, + unconditional_guidance_scale=unconditional_guidance_scale, + unconditional_conditioning=uc, + ) + x_samples_cfg = self.decode_first_stage(samples_cfg) + log[f"samples_cfg_scale_{unconditional_guidance_scale:.2f}"] = x_samples_cfg + + if inpaint: + # make a simple center square + b, h, w = z.shape[0], z.shape[2], z.shape[3] + mask = torch.ones(N, h, w).to(self.device) + # zeros will be filled in + mask[:, h // 4:3 * h // 4, w // 4:3 * w // 4] = 0. + mask = mask[:, None, ...] + with ema_scope("Plotting Inpaint"): + samples, _ = self.sample_log(cond=c, batch_size=N, ddim=use_ddim, eta=ddim_eta, + ddim_steps=ddim_steps, x0=z[:N], mask=mask) + x_samples = self.decode_first_stage(samples.to(self.device)) + log["samples_inpainting"] = x_samples + log["mask"] = mask + + # outpaint + mask = 1. - mask + with ema_scope("Plotting Outpaint"): + samples, _ = self.sample_log(cond=c, batch_size=N, ddim=use_ddim, eta=ddim_eta, + ddim_steps=ddim_steps, x0=z[:N], mask=mask) + x_samples = self.decode_first_stage(samples.to(self.device)) + log["samples_outpainting"] = x_samples + + if plot_progressive_rows: + with ema_scope("Plotting Progressives"): + img, progressives = self.progressive_denoising(c, + shape=(self.channels, self.image_size, self.image_size), + batch_size=N) + prog_row = self._get_denoise_row_from_list(progressives, desc="Progressive Generation") + log["progressive_row"] = prog_row + + if return_keys: + if np.intersect1d(list(log.keys()), return_keys).shape[0] == 0: + return log + else: + return {key: log[key] for key in return_keys} + return log + + def configure_optimizers(self): + lr = self.learning_rate + params = list(self.model.parameters()) + if self.cond_stage_trainable: + print(f"{self.__class__.__name__}: Also optimizing conditioner params!") + params = params + list(self.cond_stage_model.parameters()) + if self.learn_logvar: + print('Diffusion model optimizing logvar') + params.append(self.logvar) + opt = torch.optim.AdamW(params, lr=lr) + if self.use_scheduler: + assert 'target' in self.scheduler_config + scheduler = instantiate_from_config(self.scheduler_config) + + print("Setting up LambdaLR scheduler...") + scheduler = [ + { + 'scheduler': LambdaLR(opt, lr_lambda=scheduler.schedule), + 'interval': 'step', + 'frequency': 1 + }] + return [opt], scheduler + return opt + + @torch.no_grad() + def to_rgb(self, x): + x = x.float() + if not hasattr(self, "colorize"): + self.colorize = torch.randn(3, x.shape[1], 1, 1).to(x) + x = nn.functional.conv2d(x, weight=self.colorize) + x = 2. * (x - x.min()) / (x.max() - x.min()) - 1. + return x + + +class DiffusionWrapper(pl.LightningModule): + def __init__(self, diff_model_config, conditioning_key): + super().__init__() + self.sequential_cross_attn = diff_model_config.pop("sequential_crossattn", False) + self.diffusion_model = instantiate_from_config(diff_model_config) + self.conditioning_key = conditioning_key + assert self.conditioning_key in [None, 'concat', 'crossattn', 'hybrid', 'adm', 'hybrid-adm', 'crossattn-adm'] + + def forward(self, x, t, c_concat: list = None, c_crossattn: list = None, c_adm=None): + if self.conditioning_key is None: + out = self.diffusion_model(x, t) + elif self.conditioning_key == 'concat': + xc = torch.cat([x] + c_concat, dim=1) + out = self.diffusion_model(xc, t) + elif self.conditioning_key == 'crossattn': + if not self.sequential_cross_attn: + cc = torch.cat(c_crossattn, 1) + else: + cc = c_crossattn + out = self.diffusion_model(x, t, context=cc) + elif self.conditioning_key == 'hybrid': + xc = torch.cat([x] + c_concat, dim=1) + cc = torch.cat(c_crossattn, 1) + out = self.diffusion_model(xc, t, context=cc) + elif self.conditioning_key == 'hybrid-adm': + assert c_adm is not None + xc = torch.cat([x] + c_concat, dim=1) + cc = torch.cat(c_crossattn, 1) + out = self.diffusion_model(xc, t, context=cc, y=c_adm) + elif self.conditioning_key == 'crossattn-adm': + assert c_adm is not None + cc = torch.cat(c_crossattn, 1) + out = self.diffusion_model(x, t, context=cc, y=c_adm) + elif self.conditioning_key == 'adm': + cc = c_crossattn[0] + out = self.diffusion_model(x, t, y=cc) + else: + raise NotImplementedError() + + return out + + +class LatentUpscaleDiffusion(LatentDiffusion): + def __init__(self, *args, low_scale_config, low_scale_key="LR", noise_level_key=None, **kwargs): + super().__init__(*args, **kwargs) + # assumes that neither the cond_stage nor the low_scale_model contain trainable params + assert not self.cond_stage_trainable + self.instantiate_low_stage(low_scale_config) + self.low_scale_key = low_scale_key + self.noise_level_key = noise_level_key + + def instantiate_low_stage(self, config): + model = instantiate_from_config(config) + self.low_scale_model = model.eval() + self.low_scale_model.train = disabled_train + for param in self.low_scale_model.parameters(): + param.requires_grad = False + + @torch.no_grad() + def get_input(self, batch, k, cond_key=None, bs=None, log_mode=False): + if not log_mode: + z, c = super().get_input(batch, k, force_c_encode=True, bs=bs) + else: + z, c, x, xrec, xc = super().get_input(batch, self.first_stage_key, return_first_stage_outputs=True, + force_c_encode=True, return_original_cond=True, bs=bs) + x_low = batch[self.low_scale_key][:bs] + x_low = rearrange(x_low, 'b h w c -> b c h w') + x_low = x_low.to(memory_format=torch.contiguous_format).float() + zx, noise_level = self.low_scale_model(x_low) + if self.noise_level_key is not None: + # get noise level from batch instead, e.g. when extracting a custom noise level for bsr + raise NotImplementedError('TODO') + + all_conds = {"c_concat": [zx], "c_crossattn": [c], "c_adm": noise_level} + if log_mode: + # TODO: maybe disable if too expensive + x_low_rec = self.low_scale_model.decode(zx) + return z, all_conds, x, xrec, xc, x_low, x_low_rec, noise_level + return z, all_conds + + @torch.no_grad() + def log_images(self, batch, N=8, n_row=4, sample=True, ddim_steps=200, ddim_eta=1., return_keys=None, + plot_denoise_rows=False, plot_progressive_rows=True, plot_diffusion_rows=True, + unconditional_guidance_scale=1., unconditional_guidance_label=None, use_ema_scope=True, + **kwargs): + ema_scope = self.ema_scope if use_ema_scope else nullcontext + use_ddim = ddim_steps is not None + + log = dict() + z, c, x, xrec, xc, x_low, x_low_rec, noise_level = self.get_input(batch, self.first_stage_key, bs=N, + log_mode=True) + N = min(x.shape[0], N) + n_row = min(x.shape[0], n_row) + log["inputs"] = x + log["reconstruction"] = xrec + log["x_lr"] = x_low + log[f"x_lr_rec_@noise_levels{'-'.join(map(lambda x: str(x), list(noise_level.cpu().numpy())))}"] = x_low_rec + if self.model.conditioning_key is not None: + if hasattr(self.cond_stage_model, "decode"): + xc = self.cond_stage_model.decode(c) + log["conditioning"] = xc + elif self.cond_stage_key in ["caption", "txt"]: + xc = log_txt_as_img((x.shape[2], x.shape[3]), batch[self.cond_stage_key], size=x.shape[2] // 25) + log["conditioning"] = xc + elif self.cond_stage_key in ['class_label', 'cls']: + xc = log_txt_as_img((x.shape[2], x.shape[3]), batch["human_label"], size=x.shape[2] // 25) + log['conditioning'] = xc + elif isimage(xc): + log["conditioning"] = xc + if ismap(xc): + log["original_conditioning"] = self.to_rgb(xc) + + if plot_diffusion_rows: + # get diffusion row + diffusion_row = list() + z_start = z[:n_row] + for t in range(self.num_timesteps): + if t % self.log_every_t == 0 or t == self.num_timesteps - 1: + t = repeat(torch.tensor([t]), '1 -> b', b=n_row) + t = t.to(self.device).long() + noise = torch.randn_like(z_start) + z_noisy = self.q_sample(x_start=z_start, t=t, noise=noise) + diffusion_row.append(self.decode_first_stage(z_noisy)) + + diffusion_row = torch.stack(diffusion_row) # n_log_step, n_row, C, H, W + diffusion_grid = rearrange(diffusion_row, 'n b c h w -> b n c h w') + diffusion_grid = rearrange(diffusion_grid, 'b n c h w -> (b n) c h w') + diffusion_grid = make_grid(diffusion_grid, nrow=diffusion_row.shape[0]) + log["diffusion_row"] = diffusion_grid + + if sample: + # get denoise row + with ema_scope("Sampling"): + samples, z_denoise_row = self.sample_log(cond=c, batch_size=N, ddim=use_ddim, + ddim_steps=ddim_steps, eta=ddim_eta) + # samples, z_denoise_row = self.sample(cond=c, batch_size=N, return_intermediates=True) + x_samples = self.decode_first_stage(samples) + log["samples"] = x_samples + if plot_denoise_rows: + denoise_grid = self._get_denoise_row_from_list(z_denoise_row) + log["denoise_row"] = denoise_grid + + if unconditional_guidance_scale > 1.0: + uc_tmp = self.get_unconditional_conditioning(N, unconditional_guidance_label) + # TODO explore better "unconditional" choices for the other keys + # maybe guide away from empty text label and highest noise level and maximally degraded zx? + uc = dict() + for k in c: + if k == "c_crossattn": + assert isinstance(c[k], list) and len(c[k]) == 1 + uc[k] = [uc_tmp] + elif k == "c_adm": # todo: only run with text-based guidance? + assert isinstance(c[k], torch.Tensor) + #uc[k] = torch.ones_like(c[k]) * self.low_scale_model.max_noise_level + uc[k] = c[k] + elif isinstance(c[k], list): + uc[k] = [c[k][i] for i in range(len(c[k]))] + else: + uc[k] = c[k] + + with ema_scope("Sampling with classifier-free guidance"): + samples_cfg, _ = self.sample_log(cond=c, batch_size=N, ddim=use_ddim, + ddim_steps=ddim_steps, eta=ddim_eta, + unconditional_guidance_scale=unconditional_guidance_scale, + unconditional_conditioning=uc, + ) + x_samples_cfg = self.decode_first_stage(samples_cfg) + log[f"samples_cfg_scale_{unconditional_guidance_scale:.2f}"] = x_samples_cfg + + if plot_progressive_rows: + with ema_scope("Plotting Progressives"): + img, progressives = self.progressive_denoising(c, + shape=(self.channels, self.image_size, self.image_size), + batch_size=N) + prog_row = self._get_denoise_row_from_list(progressives, desc="Progressive Generation") + log["progressive_row"] = prog_row + + return log + + +class LatentFinetuneDiffusion(LatentDiffusion): + """ + Basis for different finetunas, such as inpainting or depth2image + To disable finetuning mode, set finetune_keys to None + """ + + def __init__(self, + concat_keys: tuple, + finetune_keys=("model.diffusion_model.input_blocks.0.0.weight", + "model_ema.diffusion_modelinput_blocks00weight" + ), + keep_finetune_dims=4, + # if model was trained without concat mode before and we would like to keep these channels + c_concat_log_start=None, # to log reconstruction of c_concat codes + c_concat_log_end=None, + *args, **kwargs + ): + ckpt_path = kwargs.pop("ckpt_path", None) + ignore_keys = kwargs.pop("ignore_keys", list()) + super().__init__(*args, **kwargs) + self.finetune_keys = finetune_keys + self.concat_keys = concat_keys + self.keep_dims = keep_finetune_dims + self.c_concat_log_start = c_concat_log_start + self.c_concat_log_end = c_concat_log_end + if exists(self.finetune_keys): assert exists(ckpt_path), 'can only finetune from a given checkpoint' + if exists(ckpt_path): + self.init_from_ckpt(ckpt_path, ignore_keys) + + def init_from_ckpt(self, path, ignore_keys=list(), only_model=False): + sd = torch.load(path, map_location="cpu") + if "state_dict" in list(sd.keys()): + sd = sd["state_dict"] + keys = list(sd.keys()) + for k in keys: + for ik in ignore_keys: + if k.startswith(ik): + print("Deleting key {} from state_dict.".format(k)) + del sd[k] + + # make it explicit, finetune by including extra input channels + if exists(self.finetune_keys) and k in self.finetune_keys: + new_entry = None + for name, param in self.named_parameters(): + if name in self.finetune_keys: + print( + f"modifying key '{name}' and keeping its original {self.keep_dims} (channels) dimensions only") + new_entry = torch.zeros_like(param) # zero init + assert exists(new_entry), 'did not find matching parameter to modify' + new_entry[:, :self.keep_dims, ...] = sd[k] + sd[k] = new_entry + + missing, unexpected = self.load_state_dict(sd, strict=False) if not only_model else self.model.load_state_dict( + sd, strict=False) + print(f"Restored from {path} with {len(missing)} missing and {len(unexpected)} unexpected keys") + if len(missing) > 0: + print(f"Missing Keys: {missing}") + if len(unexpected) > 0: + print(f"Unexpected Keys: {unexpected}") + + @torch.no_grad() + def log_images(self, batch, N=8, n_row=4, sample=True, ddim_steps=200, ddim_eta=1., return_keys=None, + quantize_denoised=True, inpaint=True, plot_denoise_rows=False, plot_progressive_rows=True, + plot_diffusion_rows=True, unconditional_guidance_scale=1., unconditional_guidance_label=None, + use_ema_scope=True, + **kwargs): + ema_scope = self.ema_scope if use_ema_scope else nullcontext + use_ddim = ddim_steps is not None + + log = dict() + z, c, x, xrec, xc = self.get_input(batch, self.first_stage_key, bs=N, return_first_stage_outputs=True) + c_cat, c = c["c_concat"][0], c["c_crossattn"][0] + N = min(x.shape[0], N) + n_row = min(x.shape[0], n_row) + log["inputs"] = x + log["reconstruction"] = xrec + if self.model.conditioning_key is not None: + if hasattr(self.cond_stage_model, "decode"): + xc = self.cond_stage_model.decode(c) + log["conditioning"] = xc + elif self.cond_stage_key in ["caption", "txt"]: + xc = log_txt_as_img((x.shape[2], x.shape[3]), batch[self.cond_stage_key], size=x.shape[2] // 25) + log["conditioning"] = xc + elif self.cond_stage_key in ['class_label', 'cls']: + xc = log_txt_as_img((x.shape[2], x.shape[3]), batch["human_label"], size=x.shape[2] // 25) + log['conditioning'] = xc + elif isimage(xc): + log["conditioning"] = xc + if ismap(xc): + log["original_conditioning"] = self.to_rgb(xc) + + if not (self.c_concat_log_start is None and self.c_concat_log_end is None): + log["c_concat_decoded"] = self.decode_first_stage(c_cat[:, self.c_concat_log_start:self.c_concat_log_end]) + + if plot_diffusion_rows: + # get diffusion row + diffusion_row = list() + z_start = z[:n_row] + for t in range(self.num_timesteps): + if t % self.log_every_t == 0 or t == self.num_timesteps - 1: + t = repeat(torch.tensor([t]), '1 -> b', b=n_row) + t = t.to(self.device).long() + noise = torch.randn_like(z_start) + z_noisy = self.q_sample(x_start=z_start, t=t, noise=noise) + diffusion_row.append(self.decode_first_stage(z_noisy)) + + diffusion_row = torch.stack(diffusion_row) # n_log_step, n_row, C, H, W + diffusion_grid = rearrange(diffusion_row, 'n b c h w -> b n c h w') + diffusion_grid = rearrange(diffusion_grid, 'b n c h w -> (b n) c h w') + diffusion_grid = make_grid(diffusion_grid, nrow=diffusion_row.shape[0]) + log["diffusion_row"] = diffusion_grid + + if sample: + # get denoise row + with ema_scope("Sampling"): + samples, z_denoise_row = self.sample_log(cond={"c_concat": [c_cat], "c_crossattn": [c]}, + batch_size=N, ddim=use_ddim, + ddim_steps=ddim_steps, eta=ddim_eta) + # samples, z_denoise_row = self.sample(cond=c, batch_size=N, return_intermediates=True) + x_samples = self.decode_first_stage(samples) + log["samples"] = x_samples + if plot_denoise_rows: + denoise_grid = self._get_denoise_row_from_list(z_denoise_row) + log["denoise_row"] = denoise_grid + + if unconditional_guidance_scale > 1.0: + uc_cross = self.get_unconditional_conditioning(N, unconditional_guidance_label) + uc_cat = c_cat + uc_full = {"c_concat": [uc_cat], "c_crossattn": [uc_cross]} + with ema_scope("Sampling with classifier-free guidance"): + samples_cfg, _ = self.sample_log(cond={"c_concat": [c_cat], "c_crossattn": [c]}, + batch_size=N, ddim=use_ddim, + ddim_steps=ddim_steps, eta=ddim_eta, + unconditional_guidance_scale=unconditional_guidance_scale, + unconditional_conditioning=uc_full, + ) + x_samples_cfg = self.decode_first_stage(samples_cfg) + log[f"samples_cfg_scale_{unconditional_guidance_scale:.2f}"] = x_samples_cfg + + return log + + +class LatentInpaintDiffusion(LatentFinetuneDiffusion): + """ + can either run as pure inpainting model (only concat mode) or with mixed conditionings, + e.g. mask as concat and text via cross-attn. + To disable finetuning mode, set finetune_keys to None + """ + + def __init__(self, + concat_keys=("mask", "masked_image"), + masked_image_key="masked_image", + *args, **kwargs + ): + super().__init__(concat_keys, *args, **kwargs) + self.masked_image_key = masked_image_key + assert self.masked_image_key in concat_keys + + @torch.no_grad() + def get_input(self, batch, k, cond_key=None, bs=None, return_first_stage_outputs=False): + # note: restricted to non-trainable encoders currently + assert not self.cond_stage_trainable, 'trainable cond stages not yet supported for inpainting' + z, c, x, xrec, xc = super().get_input(batch, self.first_stage_key, return_first_stage_outputs=True, + force_c_encode=True, return_original_cond=True, bs=bs) + + assert exists(self.concat_keys) + c_cat = list() + for ck in self.concat_keys: + cc = rearrange(batch[ck], 'b h w c -> b c h w').to(memory_format=torch.contiguous_format).float() + if bs is not None: + cc = cc[:bs] + cc = cc.to(self.device) + bchw = z.shape + if ck != self.masked_image_key: + cc = torch.nn.functional.interpolate(cc, size=bchw[-2:]) + else: + cc = self.get_first_stage_encoding(self.encode_first_stage(cc)) + c_cat.append(cc) + c_cat = torch.cat(c_cat, dim=1) + all_conds = {"c_concat": [c_cat], "c_crossattn": [c]} + if return_first_stage_outputs: + return z, all_conds, x, xrec, xc + return z, all_conds + + @torch.no_grad() + def log_images(self, *args, **kwargs): + log = super(LatentInpaintDiffusion, self).log_images(*args, **kwargs) + log["masked_image"] = rearrange(args[0]["masked_image"], + 'b h w c -> b c h w').to(memory_format=torch.contiguous_format).float() + return log + + +class LatentDepth2ImageDiffusion(LatentFinetuneDiffusion): + """ + condition on monocular depth estimation + """ + + def __init__(self, depth_stage_config, concat_keys=("midas_in",), *args, **kwargs): + super().__init__(concat_keys=concat_keys, *args, **kwargs) + self.depth_model = instantiate_from_config(depth_stage_config) + self.depth_stage_key = concat_keys[0] + + @torch.no_grad() + def get_input(self, batch, k, cond_key=None, bs=None, return_first_stage_outputs=False): + # note: restricted to non-trainable encoders currently + assert not self.cond_stage_trainable, 'trainable cond stages not yet supported for depth2img' + z, c, x, xrec, xc = super().get_input(batch, self.first_stage_key, return_first_stage_outputs=True, + force_c_encode=True, return_original_cond=True, bs=bs) + + assert exists(self.concat_keys) + assert len(self.concat_keys) == 1 + c_cat = list() + for ck in self.concat_keys: + cc = batch[ck] + if bs is not None: + cc = cc[:bs] + cc = cc.to(self.device) + cc = self.depth_model(cc) + cc = torch.nn.functional.interpolate( + cc, + size=z.shape[2:], + mode="bicubic", + align_corners=False, + ) + + depth_min, depth_max = torch.amin(cc, dim=[1, 2, 3], keepdim=True), torch.amax(cc, dim=[1, 2, 3], + keepdim=True) + cc = 2. * (cc - depth_min) / (depth_max - depth_min + 0.001) - 1. + c_cat.append(cc) + c_cat = torch.cat(c_cat, dim=1) + all_conds = {"c_concat": [c_cat], "c_crossattn": [c]} + if return_first_stage_outputs: + return z, all_conds, x, xrec, xc + return z, all_conds + + @torch.no_grad() + def log_images(self, *args, **kwargs): + log = super().log_images(*args, **kwargs) + depth = self.depth_model(args[0][self.depth_stage_key]) + depth_min, depth_max = torch.amin(depth, dim=[1, 2, 3], keepdim=True), \ + torch.amax(depth, dim=[1, 2, 3], keepdim=True) + log["depth"] = 2. * (depth - depth_min) / (depth_max - depth_min) - 1. + return log + + +class LatentUpscaleFinetuneDiffusion(LatentFinetuneDiffusion): + """ + condition on low-res image (and optionally on some spatial noise augmentation) + """ + def __init__(self, concat_keys=("lr",), reshuffle_patch_size=None, + low_scale_config=None, low_scale_key=None, *args, **kwargs): + super().__init__(concat_keys=concat_keys, *args, **kwargs) + self.reshuffle_patch_size = reshuffle_patch_size + self.low_scale_model = None + if low_scale_config is not None: + print("Initializing a low-scale model") + assert exists(low_scale_key) + self.instantiate_low_stage(low_scale_config) + self.low_scale_key = low_scale_key + + def instantiate_low_stage(self, config): + model = instantiate_from_config(config) + self.low_scale_model = model.eval() + self.low_scale_model.train = disabled_train + for param in self.low_scale_model.parameters(): + param.requires_grad = False + + @torch.no_grad() + def get_input(self, batch, k, cond_key=None, bs=None, return_first_stage_outputs=False): + # note: restricted to non-trainable encoders currently + assert not self.cond_stage_trainable, 'trainable cond stages not yet supported for upscaling-ft' + z, c, x, xrec, xc = super().get_input(batch, self.first_stage_key, return_first_stage_outputs=True, + force_c_encode=True, return_original_cond=True, bs=bs) + + assert exists(self.concat_keys) + assert len(self.concat_keys) == 1 + # optionally make spatial noise_level here + c_cat = list() + noise_level = None + for ck in self.concat_keys: + cc = batch[ck] + cc = rearrange(cc, 'b h w c -> b c h w') + if exists(self.reshuffle_patch_size): + assert isinstance(self.reshuffle_patch_size, int) + cc = rearrange(cc, 'b c (p1 h) (p2 w) -> b (p1 p2 c) h w', + p1=self.reshuffle_patch_size, p2=self.reshuffle_patch_size) + if bs is not None: + cc = cc[:bs] + cc = cc.to(self.device) + if exists(self.low_scale_model) and ck == self.low_scale_key: + cc, noise_level = self.low_scale_model(cc) + c_cat.append(cc) + c_cat = torch.cat(c_cat, dim=1) + if exists(noise_level): + all_conds = {"c_concat": [c_cat], "c_crossattn": [c], "c_adm": noise_level} + else: + all_conds = {"c_concat": [c_cat], "c_crossattn": [c]} + if return_first_stage_outputs: + return z, all_conds, x, xrec, xc + return z, all_conds + + @torch.no_grad() + def log_images(self, *args, **kwargs): + log = super().log_images(*args, **kwargs) + log["lr"] = rearrange(args[0]["lr"], 'b h w c -> b c h w') + return log diff --git a/ldm/models/diffusion/dpm_solver/__init__.py b/ldm/models/diffusion/dpm_solver/__init__.py new file mode 100644 index 0000000000000000000000000000000000000000..7427f38c07530afbab79154ea8aaf88c4bf70a08 --- /dev/null +++ b/ldm/models/diffusion/dpm_solver/__init__.py @@ -0,0 +1 @@ +from .sampler import DPMSolverSampler \ No newline at end of file diff --git a/ldm/models/diffusion/dpm_solver/dpm_solver.py b/ldm/models/diffusion/dpm_solver/dpm_solver.py new file mode 100644 index 0000000000000000000000000000000000000000..095e5ba3ce0b1aa7f4b3f1e2e5d8fff7cfe6dc8c --- /dev/null +++ b/ldm/models/diffusion/dpm_solver/dpm_solver.py @@ -0,0 +1,1154 @@ +import torch +import torch.nn.functional as F +import math +from tqdm import tqdm + + +class NoiseScheduleVP: + def __init__( + self, + schedule='discrete', + betas=None, + alphas_cumprod=None, + continuous_beta_0=0.1, + continuous_beta_1=20., + ): + """Create a wrapper class for the forward SDE (VP type). + *** + Update: We support discrete-time diffusion models by implementing a picewise linear interpolation for log_alpha_t. + We recommend to use schedule='discrete' for the discrete-time diffusion models, especially for high-resolution images. + *** + The forward SDE ensures that the condition distribution q_{t|0}(x_t | x_0) = N ( alpha_t * x_0, sigma_t^2 * I ). + We further define lambda_t = log(alpha_t) - log(sigma_t), which is the half-logSNR (described in the DPM-Solver paper). + Therefore, we implement the functions for computing alpha_t, sigma_t and lambda_t. For t in [0, T], we have: + log_alpha_t = self.marginal_log_mean_coeff(t) + sigma_t = self.marginal_std(t) + lambda_t = self.marginal_lambda(t) + Moreover, as lambda(t) is an invertible function, we also support its inverse function: + t = self.inverse_lambda(lambda_t) + =============================================================== + We support both discrete-time DPMs (trained on n = 0, 1, ..., N-1) and continuous-time DPMs (trained on t in [t_0, T]). + 1. For discrete-time DPMs: + For discrete-time DPMs trained on n = 0, 1, ..., N-1, we convert the discrete steps to continuous time steps by: + t_i = (i + 1) / N + e.g. for N = 1000, we have t_0 = 1e-3 and T = t_{N-1} = 1. + We solve the corresponding diffusion ODE from time T = 1 to time t_0 = 1e-3. + Args: + betas: A `torch.Tensor`. The beta array for the discrete-time DPM. (See the original DDPM paper for details) + alphas_cumprod: A `torch.Tensor`. The cumprod alphas for the discrete-time DPM. (See the original DDPM paper for details) + Note that we always have alphas_cumprod = cumprod(betas). Therefore, we only need to set one of `betas` and `alphas_cumprod`. + **Important**: Please pay special attention for the args for `alphas_cumprod`: + The `alphas_cumprod` is the \hat{alpha_n} arrays in the notations of DDPM. Specifically, DDPMs assume that + q_{t_n | 0}(x_{t_n} | x_0) = N ( \sqrt{\hat{alpha_n}} * x_0, (1 - \hat{alpha_n}) * I ). + Therefore, the notation \hat{alpha_n} is different from the notation alpha_t in DPM-Solver. In fact, we have + alpha_{t_n} = \sqrt{\hat{alpha_n}}, + and + log(alpha_{t_n}) = 0.5 * log(\hat{alpha_n}). + 2. For continuous-time DPMs: + We support two types of VPSDEs: linear (DDPM) and cosine (improved-DDPM). The hyperparameters for the noise + schedule are the default settings in DDPM and improved-DDPM: + Args: + beta_min: A `float` number. The smallest beta for the linear schedule. + beta_max: A `float` number. The largest beta for the linear schedule. + cosine_s: A `float` number. The hyperparameter in the cosine schedule. + cosine_beta_max: A `float` number. The hyperparameter in the cosine schedule. + T: A `float` number. The ending time of the forward process. + =============================================================== + Args: + schedule: A `str`. The noise schedule of the forward SDE. 'discrete' for discrete-time DPMs, + 'linear' or 'cosine' for continuous-time DPMs. + Returns: + A wrapper object of the forward SDE (VP type). + + =============================================================== + Example: + # For discrete-time DPMs, given betas (the beta array for n = 0, 1, ..., N - 1): + >>> ns = NoiseScheduleVP('discrete', betas=betas) + # For discrete-time DPMs, given alphas_cumprod (the \hat{alpha_n} array for n = 0, 1, ..., N - 1): + >>> ns = NoiseScheduleVP('discrete', alphas_cumprod=alphas_cumprod) + # For continuous-time DPMs (VPSDE), linear schedule: + >>> ns = NoiseScheduleVP('linear', continuous_beta_0=0.1, continuous_beta_1=20.) + """ + + if schedule not in ['discrete', 'linear', 'cosine']: + raise ValueError( + "Unsupported noise schedule {}. The schedule needs to be 'discrete' or 'linear' or 'cosine'".format( + schedule)) + + self.schedule = schedule + if schedule == 'discrete': + if betas is not None: + log_alphas = 0.5 * torch.log(1 - betas).cumsum(dim=0) + else: + assert alphas_cumprod is not None + log_alphas = 0.5 * torch.log(alphas_cumprod) + self.total_N = len(log_alphas) + self.T = 1. + self.t_array = torch.linspace(0., 1., self.total_N + 1)[1:].reshape((1, -1)) + self.log_alpha_array = log_alphas.reshape((1, -1,)) + else: + self.total_N = 1000 + self.beta_0 = continuous_beta_0 + self.beta_1 = continuous_beta_1 + self.cosine_s = 0.008 + self.cosine_beta_max = 999. + self.cosine_t_max = math.atan(self.cosine_beta_max * (1. + self.cosine_s) / math.pi) * 2. * ( + 1. + self.cosine_s) / math.pi - self.cosine_s + self.cosine_log_alpha_0 = math.log(math.cos(self.cosine_s / (1. + self.cosine_s) * math.pi / 2.)) + self.schedule = schedule + if schedule == 'cosine': + # For the cosine schedule, T = 1 will have numerical issues. So we manually set the ending time T. + # Note that T = 0.9946 may be not the optimal setting. However, we find it works well. + self.T = 0.9946 + else: + self.T = 1. + + def marginal_log_mean_coeff(self, t): + """ + Compute log(alpha_t) of a given continuous-time label t in [0, T]. + """ + if self.schedule == 'discrete': + return interpolate_fn(t.reshape((-1, 1)), self.t_array.to(t.device), + self.log_alpha_array.to(t.device)).reshape((-1)) + elif self.schedule == 'linear': + return -0.25 * t ** 2 * (self.beta_1 - self.beta_0) - 0.5 * t * self.beta_0 + elif self.schedule == 'cosine': + log_alpha_fn = lambda s: torch.log(torch.cos((s + self.cosine_s) / (1. + self.cosine_s) * math.pi / 2.)) + log_alpha_t = log_alpha_fn(t) - self.cosine_log_alpha_0 + return log_alpha_t + + def marginal_alpha(self, t): + """ + Compute alpha_t of a given continuous-time label t in [0, T]. + """ + return torch.exp(self.marginal_log_mean_coeff(t)) + + def marginal_std(self, t): + """ + Compute sigma_t of a given continuous-time label t in [0, T]. + """ + return torch.sqrt(1. - torch.exp(2. * self.marginal_log_mean_coeff(t))) + + def marginal_lambda(self, t): + """ + Compute lambda_t = log(alpha_t) - log(sigma_t) of a given continuous-time label t in [0, T]. + """ + log_mean_coeff = self.marginal_log_mean_coeff(t) + log_std = 0.5 * torch.log(1. - torch.exp(2. * log_mean_coeff)) + return log_mean_coeff - log_std + + def inverse_lambda(self, lamb): + """ + Compute the continuous-time label t in [0, T] of a given half-logSNR lambda_t. + """ + if self.schedule == 'linear': + tmp = 2. * (self.beta_1 - self.beta_0) * torch.logaddexp(-2. * lamb, torch.zeros((1,)).to(lamb)) + Delta = self.beta_0 ** 2 + tmp + return tmp / (torch.sqrt(Delta) + self.beta_0) / (self.beta_1 - self.beta_0) + elif self.schedule == 'discrete': + log_alpha = -0.5 * torch.logaddexp(torch.zeros((1,)).to(lamb.device), -2. * lamb) + t = interpolate_fn(log_alpha.reshape((-1, 1)), torch.flip(self.log_alpha_array.to(lamb.device), [1]), + torch.flip(self.t_array.to(lamb.device), [1])) + return t.reshape((-1,)) + else: + log_alpha = -0.5 * torch.logaddexp(-2. * lamb, torch.zeros((1,)).to(lamb)) + t_fn = lambda log_alpha_t: torch.arccos(torch.exp(log_alpha_t + self.cosine_log_alpha_0)) * 2. * ( + 1. + self.cosine_s) / math.pi - self.cosine_s + t = t_fn(log_alpha) + return t + + +def model_wrapper( + model, + noise_schedule, + model_type="noise", + model_kwargs={}, + guidance_type="uncond", + condition=None, + unconditional_condition=None, + guidance_scale=1., + classifier_fn=None, + classifier_kwargs={}, +): + """Create a wrapper function for the noise prediction model. + DPM-Solver needs to solve the continuous-time diffusion ODEs. For DPMs trained on discrete-time labels, we need to + firstly wrap the model function to a noise prediction model that accepts the continuous time as the input. + We support four types of the diffusion model by setting `model_type`: + 1. "noise": noise prediction model. (Trained by predicting noise). + 2. "x_start": data prediction model. (Trained by predicting the data x_0 at time 0). + 3. "v": velocity prediction model. (Trained by predicting the velocity). + The "v" prediction is derivation detailed in Appendix D of [1], and is used in Imagen-Video [2]. + [1] Salimans, Tim, and Jonathan Ho. "Progressive distillation for fast sampling of diffusion models." + arXiv preprint arXiv:2202.00512 (2022). + [2] Ho, Jonathan, et al. "Imagen Video: High Definition Video Generation with Diffusion Models." + arXiv preprint arXiv:2210.02303 (2022). + + 4. "score": marginal score function. (Trained by denoising score matching). + Note that the score function and the noise prediction model follows a simple relationship: + ``` + noise(x_t, t) = -sigma_t * score(x_t, t) + ``` + We support three types of guided sampling by DPMs by setting `guidance_type`: + 1. "uncond": unconditional sampling by DPMs. + The input `model` has the following format: + `` + model(x, t_input, **model_kwargs) -> noise | x_start | v | score + `` + 2. "classifier": classifier guidance sampling [3] by DPMs and another classifier. + The input `model` has the following format: + `` + model(x, t_input, **model_kwargs) -> noise | x_start | v | score + `` + The input `classifier_fn` has the following format: + `` + classifier_fn(x, t_input, cond, **classifier_kwargs) -> logits(x, t_input, cond) + `` + [3] P. Dhariwal and A. Q. Nichol, "Diffusion models beat GANs on image synthesis," + in Advances in Neural Information Processing Systems, vol. 34, 2021, pp. 8780-8794. + 3. "classifier-free": classifier-free guidance sampling by conditional DPMs. + The input `model` has the following format: + `` + model(x, t_input, cond, **model_kwargs) -> noise | x_start | v | score + `` + And if cond == `unconditional_condition`, the model output is the unconditional DPM output. + [4] Ho, Jonathan, and Tim Salimans. "Classifier-free diffusion guidance." + arXiv preprint arXiv:2207.12598 (2022). + + The `t_input` is the time label of the model, which may be discrete-time labels (i.e. 0 to 999) + or continuous-time labels (i.e. epsilon to T). + We wrap the model function to accept only `x` and `t_continuous` as inputs, and outputs the predicted noise: + `` + def model_fn(x, t_continuous) -> noise: + t_input = get_model_input_time(t_continuous) + return noise_pred(model, x, t_input, **model_kwargs) + `` + where `t_continuous` is the continuous time labels (i.e. epsilon to T). And we use `model_fn` for DPM-Solver. + =============================================================== + Args: + model: A diffusion model with the corresponding format described above. + noise_schedule: A noise schedule object, such as NoiseScheduleVP. + model_type: A `str`. The parameterization type of the diffusion model. + "noise" or "x_start" or "v" or "score". + model_kwargs: A `dict`. A dict for the other inputs of the model function. + guidance_type: A `str`. The type of the guidance for sampling. + "uncond" or "classifier" or "classifier-free". + condition: A pytorch tensor. The condition for the guided sampling. + Only used for "classifier" or "classifier-free" guidance type. + unconditional_condition: A pytorch tensor. The condition for the unconditional sampling. + Only used for "classifier-free" guidance type. + guidance_scale: A `float`. The scale for the guided sampling. + classifier_fn: A classifier function. Only used for the classifier guidance. + classifier_kwargs: A `dict`. A dict for the other inputs of the classifier function. + Returns: + A noise prediction model that accepts the noised data and the continuous time as the inputs. + """ + + def get_model_input_time(t_continuous): + """ + Convert the continuous-time `t_continuous` (in [epsilon, T]) to the model input time. + For discrete-time DPMs, we convert `t_continuous` in [1 / N, 1] to `t_input` in [0, 1000 * (N - 1) / N]. + For continuous-time DPMs, we just use `t_continuous`. + """ + if noise_schedule.schedule == 'discrete': + return (t_continuous - 1. / noise_schedule.total_N) * 1000. + else: + return t_continuous + + def noise_pred_fn(x, t_continuous, cond=None): + if t_continuous.reshape((-1,)).shape[0] == 1: + t_continuous = t_continuous.expand((x.shape[0])) + t_input = get_model_input_time(t_continuous) + if cond is None: + output = model(x, t_input, **model_kwargs) + else: + output = model(x, t_input, cond, **model_kwargs) + if model_type == "noise": + return output + elif model_type == "x_start": + alpha_t, sigma_t = noise_schedule.marginal_alpha(t_continuous), noise_schedule.marginal_std(t_continuous) + dims = x.dim() + return (x - expand_dims(alpha_t, dims) * output) / expand_dims(sigma_t, dims) + elif model_type == "v": + alpha_t, sigma_t = noise_schedule.marginal_alpha(t_continuous), noise_schedule.marginal_std(t_continuous) + dims = x.dim() + return expand_dims(alpha_t, dims) * output + expand_dims(sigma_t, dims) * x + elif model_type == "score": + sigma_t = noise_schedule.marginal_std(t_continuous) + dims = x.dim() + return -expand_dims(sigma_t, dims) * output + + def cond_grad_fn(x, t_input): + """ + Compute the gradient of the classifier, i.e. nabla_{x} log p_t(cond | x_t). + """ + with torch.enable_grad(): + x_in = x.detach().requires_grad_(True) + log_prob = classifier_fn(x_in, t_input, condition, **classifier_kwargs) + return torch.autograd.grad(log_prob.sum(), x_in)[0] + + def model_fn(x, t_continuous): + """ + The noise predicition model function that is used for DPM-Solver. + """ + if t_continuous.reshape((-1,)).shape[0] == 1: + t_continuous = t_continuous.expand((x.shape[0])) + if guidance_type == "uncond": + return noise_pred_fn(x, t_continuous) + elif guidance_type == "classifier": + assert classifier_fn is not None + t_input = get_model_input_time(t_continuous) + cond_grad = cond_grad_fn(x, t_input) + sigma_t = noise_schedule.marginal_std(t_continuous) + noise = noise_pred_fn(x, t_continuous) + return noise - guidance_scale * expand_dims(sigma_t, dims=cond_grad.dim()) * cond_grad + elif guidance_type == "classifier-free": + if guidance_scale == 1. or unconditional_condition is None: + return noise_pred_fn(x, t_continuous, cond=condition) + else: + x_in = torch.cat([x] * 2) + t_in = torch.cat([t_continuous] * 2) + c_in = torch.cat([unconditional_condition, condition]) + noise_uncond, noise = noise_pred_fn(x_in, t_in, cond=c_in).chunk(2) + return noise_uncond + guidance_scale * (noise - noise_uncond) + + assert model_type in ["noise", "x_start", "v"] + assert guidance_type in ["uncond", "classifier", "classifier-free"] + return model_fn + + +class DPM_Solver: + def __init__(self, model_fn, noise_schedule, predict_x0=False, thresholding=False, max_val=1.): + """Construct a DPM-Solver. + We support both the noise prediction model ("predicting epsilon") and the data prediction model ("predicting x0"). + If `predict_x0` is False, we use the solver for the noise prediction model (DPM-Solver). + If `predict_x0` is True, we use the solver for the data prediction model (DPM-Solver++). + In such case, we further support the "dynamic thresholding" in [1] when `thresholding` is True. + The "dynamic thresholding" can greatly improve the sample quality for pixel-space DPMs with large guidance scales. + Args: + model_fn: A noise prediction model function which accepts the continuous-time input (t in [epsilon, T]): + `` + def model_fn(x, t_continuous): + return noise + `` + noise_schedule: A noise schedule object, such as NoiseScheduleVP. + predict_x0: A `bool`. If true, use the data prediction model; else, use the noise prediction model. + thresholding: A `bool`. Valid when `predict_x0` is True. Whether to use the "dynamic thresholding" in [1]. + max_val: A `float`. Valid when both `predict_x0` and `thresholding` are True. The max value for thresholding. + + [1] Chitwan Saharia, William Chan, Saurabh Saxena, Lala Li, Jay Whang, Emily Denton, Seyed Kamyar Seyed Ghasemipour, Burcu Karagol Ayan, S Sara Mahdavi, Rapha Gontijo Lopes, et al. Photorealistic text-to-image diffusion models with deep language understanding. arXiv preprint arXiv:2205.11487, 2022b. + """ + self.model = model_fn + self.noise_schedule = noise_schedule + self.predict_x0 = predict_x0 + self.thresholding = thresholding + self.max_val = max_val + + def noise_prediction_fn(self, x, t): + """ + Return the noise prediction model. + """ + return self.model(x, t) + + def data_prediction_fn(self, x, t): + """ + Return the data prediction model (with thresholding). + """ + noise = self.noise_prediction_fn(x, t) + dims = x.dim() + alpha_t, sigma_t = self.noise_schedule.marginal_alpha(t), self.noise_schedule.marginal_std(t) + x0 = (x - expand_dims(sigma_t, dims) * noise) / expand_dims(alpha_t, dims) + if self.thresholding: + p = 0.995 # A hyperparameter in the paper of "Imagen" [1]. + s = torch.quantile(torch.abs(x0).reshape((x0.shape[0], -1)), p, dim=1) + s = expand_dims(torch.maximum(s, self.max_val * torch.ones_like(s).to(s.device)), dims) + x0 = torch.clamp(x0, -s, s) / s + return x0 + + def model_fn(self, x, t): + """ + Convert the model to the noise prediction model or the data prediction model. + """ + if self.predict_x0: + return self.data_prediction_fn(x, t) + else: + return self.noise_prediction_fn(x, t) + + def get_time_steps(self, skip_type, t_T, t_0, N, device): + """Compute the intermediate time steps for sampling. + Args: + skip_type: A `str`. The type for the spacing of the time steps. We support three types: + - 'logSNR': uniform logSNR for the time steps. + - 'time_uniform': uniform time for the time steps. (**Recommended for high-resolutional data**.) + - 'time_quadratic': quadratic time for the time steps. (Used in DDIM for low-resolutional data.) + t_T: A `float`. The starting time of the sampling (default is T). + t_0: A `float`. The ending time of the sampling (default is epsilon). + N: A `int`. The total number of the spacing of the time steps. + device: A torch device. + Returns: + A pytorch tensor of the time steps, with the shape (N + 1,). + """ + if skip_type == 'logSNR': + lambda_T = self.noise_schedule.marginal_lambda(torch.tensor(t_T).to(device)) + lambda_0 = self.noise_schedule.marginal_lambda(torch.tensor(t_0).to(device)) + logSNR_steps = torch.linspace(lambda_T.cpu().item(), lambda_0.cpu().item(), N + 1).to(device) + return self.noise_schedule.inverse_lambda(logSNR_steps) + elif skip_type == 'time_uniform': + return torch.linspace(t_T, t_0, N + 1).to(device) + elif skip_type == 'time_quadratic': + t_order = 2 + t = torch.linspace(t_T ** (1. / t_order), t_0 ** (1. / t_order), N + 1).pow(t_order).to(device) + return t + else: + raise ValueError( + "Unsupported skip_type {}, need to be 'logSNR' or 'time_uniform' or 'time_quadratic'".format(skip_type)) + + def get_orders_and_timesteps_for_singlestep_solver(self, steps, order, skip_type, t_T, t_0, device): + """ + Get the order of each step for sampling by the singlestep DPM-Solver. + We combine both DPM-Solver-1,2,3 to use all the function evaluations, which is named as "DPM-Solver-fast". + Given a fixed number of function evaluations by `steps`, the sampling procedure by DPM-Solver-fast is: + - If order == 1: + We take `steps` of DPM-Solver-1 (i.e. DDIM). + - If order == 2: + - Denote K = (steps // 2). We take K or (K + 1) intermediate time steps for sampling. + - If steps % 2 == 0, we use K steps of DPM-Solver-2. + - If steps % 2 == 1, we use K steps of DPM-Solver-2 and 1 step of DPM-Solver-1. + - If order == 3: + - Denote K = (steps // 3 + 1). We take K intermediate time steps for sampling. + - If steps % 3 == 0, we use (K - 2) steps of DPM-Solver-3, and 1 step of DPM-Solver-2 and 1 step of DPM-Solver-1. + - If steps % 3 == 1, we use (K - 1) steps of DPM-Solver-3 and 1 step of DPM-Solver-1. + - If steps % 3 == 2, we use (K - 1) steps of DPM-Solver-3 and 1 step of DPM-Solver-2. + ============================================ + Args: + order: A `int`. The max order for the solver (2 or 3). + steps: A `int`. The total number of function evaluations (NFE). + skip_type: A `str`. The type for the spacing of the time steps. We support three types: + - 'logSNR': uniform logSNR for the time steps. + - 'time_uniform': uniform time for the time steps. (**Recommended for high-resolutional data**.) + - 'time_quadratic': quadratic time for the time steps. (Used in DDIM for low-resolutional data.) + t_T: A `float`. The starting time of the sampling (default is T). + t_0: A `float`. The ending time of the sampling (default is epsilon). + device: A torch device. + Returns: + orders: A list of the solver order of each step. + """ + if order == 3: + K = steps // 3 + 1 + if steps % 3 == 0: + orders = [3, ] * (K - 2) + [2, 1] + elif steps % 3 == 1: + orders = [3, ] * (K - 1) + [1] + else: + orders = [3, ] * (K - 1) + [2] + elif order == 2: + if steps % 2 == 0: + K = steps // 2 + orders = [2, ] * K + else: + K = steps // 2 + 1 + orders = [2, ] * (K - 1) + [1] + elif order == 1: + K = 1 + orders = [1, ] * steps + else: + raise ValueError("'order' must be '1' or '2' or '3'.") + if skip_type == 'logSNR': + # To reproduce the results in DPM-Solver paper + timesteps_outer = self.get_time_steps(skip_type, t_T, t_0, K, device) + else: + timesteps_outer = self.get_time_steps(skip_type, t_T, t_0, steps, device)[ + torch.cumsum(torch.tensor([0, ] + orders)).to(device)] + return timesteps_outer, orders + + def denoise_to_zero_fn(self, x, s): + """ + Denoise at the final step, which is equivalent to solve the ODE from lambda_s to infty by first-order discretization. + """ + return self.data_prediction_fn(x, s) + + def dpm_solver_first_update(self, x, s, t, model_s=None, return_intermediate=False): + """ + DPM-Solver-1 (equivalent to DDIM) from time `s` to time `t`. + Args: + x: A pytorch tensor. The initial value at time `s`. + s: A pytorch tensor. The starting time, with the shape (x.shape[0],). + t: A pytorch tensor. The ending time, with the shape (x.shape[0],). + model_s: A pytorch tensor. The model function evaluated at time `s`. + If `model_s` is None, we evaluate the model by `x` and `s`; otherwise we directly use it. + return_intermediate: A `bool`. If true, also return the model value at time `s`. + Returns: + x_t: A pytorch tensor. The approximated solution at time `t`. + """ + ns = self.noise_schedule + dims = x.dim() + lambda_s, lambda_t = ns.marginal_lambda(s), ns.marginal_lambda(t) + h = lambda_t - lambda_s + log_alpha_s, log_alpha_t = ns.marginal_log_mean_coeff(s), ns.marginal_log_mean_coeff(t) + sigma_s, sigma_t = ns.marginal_std(s), ns.marginal_std(t) + alpha_t = torch.exp(log_alpha_t) + + if self.predict_x0: + phi_1 = torch.expm1(-h) + if model_s is None: + model_s = self.model_fn(x, s) + x_t = ( + expand_dims(sigma_t / sigma_s, dims) * x + - expand_dims(alpha_t * phi_1, dims) * model_s + ) + if return_intermediate: + return x_t, {'model_s': model_s} + else: + return x_t + else: + phi_1 = torch.expm1(h) + if model_s is None: + model_s = self.model_fn(x, s) + x_t = ( + expand_dims(torch.exp(log_alpha_t - log_alpha_s), dims) * x + - expand_dims(sigma_t * phi_1, dims) * model_s + ) + if return_intermediate: + return x_t, {'model_s': model_s} + else: + return x_t + + def singlestep_dpm_solver_second_update(self, x, s, t, r1=0.5, model_s=None, return_intermediate=False, + solver_type='dpm_solver'): + """ + Singlestep solver DPM-Solver-2 from time `s` to time `t`. + Args: + x: A pytorch tensor. The initial value at time `s`. + s: A pytorch tensor. The starting time, with the shape (x.shape[0],). + t: A pytorch tensor. The ending time, with the shape (x.shape[0],). + r1: A `float`. The hyperparameter of the second-order solver. + model_s: A pytorch tensor. The model function evaluated at time `s`. + If `model_s` is None, we evaluate the model by `x` and `s`; otherwise we directly use it. + return_intermediate: A `bool`. If true, also return the model value at time `s` and `s1` (the intermediate time). + solver_type: either 'dpm_solver' or 'taylor'. The type for the high-order solvers. + The type slightly impacts the performance. We recommend to use 'dpm_solver' type. + Returns: + x_t: A pytorch tensor. The approximated solution at time `t`. + """ + if solver_type not in ['dpm_solver', 'taylor']: + raise ValueError("'solver_type' must be either 'dpm_solver' or 'taylor', got {}".format(solver_type)) + if r1 is None: + r1 = 0.5 + ns = self.noise_schedule + dims = x.dim() + lambda_s, lambda_t = ns.marginal_lambda(s), ns.marginal_lambda(t) + h = lambda_t - lambda_s + lambda_s1 = lambda_s + r1 * h + s1 = ns.inverse_lambda(lambda_s1) + log_alpha_s, log_alpha_s1, log_alpha_t = ns.marginal_log_mean_coeff(s), ns.marginal_log_mean_coeff( + s1), ns.marginal_log_mean_coeff(t) + sigma_s, sigma_s1, sigma_t = ns.marginal_std(s), ns.marginal_std(s1), ns.marginal_std(t) + alpha_s1, alpha_t = torch.exp(log_alpha_s1), torch.exp(log_alpha_t) + + if self.predict_x0: + phi_11 = torch.expm1(-r1 * h) + phi_1 = torch.expm1(-h) + + if model_s is None: + model_s = self.model_fn(x, s) + x_s1 = ( + expand_dims(sigma_s1 / sigma_s, dims) * x + - expand_dims(alpha_s1 * phi_11, dims) * model_s + ) + model_s1 = self.model_fn(x_s1, s1) + if solver_type == 'dpm_solver': + x_t = ( + expand_dims(sigma_t / sigma_s, dims) * x + - expand_dims(alpha_t * phi_1, dims) * model_s + - (0.5 / r1) * expand_dims(alpha_t * phi_1, dims) * (model_s1 - model_s) + ) + elif solver_type == 'taylor': + x_t = ( + expand_dims(sigma_t / sigma_s, dims) * x + - expand_dims(alpha_t * phi_1, dims) * model_s + + (1. / r1) * expand_dims(alpha_t * ((torch.exp(-h) - 1.) / h + 1.), dims) * ( + model_s1 - model_s) + ) + else: + phi_11 = torch.expm1(r1 * h) + phi_1 = torch.expm1(h) + + if model_s is None: + model_s = self.model_fn(x, s) + x_s1 = ( + expand_dims(torch.exp(log_alpha_s1 - log_alpha_s), dims) * x + - expand_dims(sigma_s1 * phi_11, dims) * model_s + ) + model_s1 = self.model_fn(x_s1, s1) + if solver_type == 'dpm_solver': + x_t = ( + expand_dims(torch.exp(log_alpha_t - log_alpha_s), dims) * x + - expand_dims(sigma_t * phi_1, dims) * model_s + - (0.5 / r1) * expand_dims(sigma_t * phi_1, dims) * (model_s1 - model_s) + ) + elif solver_type == 'taylor': + x_t = ( + expand_dims(torch.exp(log_alpha_t - log_alpha_s), dims) * x + - expand_dims(sigma_t * phi_1, dims) * model_s + - (1. / r1) * expand_dims(sigma_t * ((torch.exp(h) - 1.) / h - 1.), dims) * (model_s1 - model_s) + ) + if return_intermediate: + return x_t, {'model_s': model_s, 'model_s1': model_s1} + else: + return x_t + + def singlestep_dpm_solver_third_update(self, x, s, t, r1=1. / 3., r2=2. / 3., model_s=None, model_s1=None, + return_intermediate=False, solver_type='dpm_solver'): + """ + Singlestep solver DPM-Solver-3 from time `s` to time `t`. + Args: + x: A pytorch tensor. The initial value at time `s`. + s: A pytorch tensor. The starting time, with the shape (x.shape[0],). + t: A pytorch tensor. The ending time, with the shape (x.shape[0],). + r1: A `float`. The hyperparameter of the third-order solver. + r2: A `float`. The hyperparameter of the third-order solver. + model_s: A pytorch tensor. The model function evaluated at time `s`. + If `model_s` is None, we evaluate the model by `x` and `s`; otherwise we directly use it. + model_s1: A pytorch tensor. The model function evaluated at time `s1` (the intermediate time given by `r1`). + If `model_s1` is None, we evaluate the model at `s1`; otherwise we directly use it. + return_intermediate: A `bool`. If true, also return the model value at time `s`, `s1` and `s2` (the intermediate times). + solver_type: either 'dpm_solver' or 'taylor'. The type for the high-order solvers. + The type slightly impacts the performance. We recommend to use 'dpm_solver' type. + Returns: + x_t: A pytorch tensor. The approximated solution at time `t`. + """ + if solver_type not in ['dpm_solver', 'taylor']: + raise ValueError("'solver_type' must be either 'dpm_solver' or 'taylor', got {}".format(solver_type)) + if r1 is None: + r1 = 1. / 3. + if r2 is None: + r2 = 2. / 3. + ns = self.noise_schedule + dims = x.dim() + lambda_s, lambda_t = ns.marginal_lambda(s), ns.marginal_lambda(t) + h = lambda_t - lambda_s + lambda_s1 = lambda_s + r1 * h + lambda_s2 = lambda_s + r2 * h + s1 = ns.inverse_lambda(lambda_s1) + s2 = ns.inverse_lambda(lambda_s2) + log_alpha_s, log_alpha_s1, log_alpha_s2, log_alpha_t = ns.marginal_log_mean_coeff( + s), ns.marginal_log_mean_coeff(s1), ns.marginal_log_mean_coeff(s2), ns.marginal_log_mean_coeff(t) + sigma_s, sigma_s1, sigma_s2, sigma_t = ns.marginal_std(s), ns.marginal_std(s1), ns.marginal_std( + s2), ns.marginal_std(t) + alpha_s1, alpha_s2, alpha_t = torch.exp(log_alpha_s1), torch.exp(log_alpha_s2), torch.exp(log_alpha_t) + + if self.predict_x0: + phi_11 = torch.expm1(-r1 * h) + phi_12 = torch.expm1(-r2 * h) + phi_1 = torch.expm1(-h) + phi_22 = torch.expm1(-r2 * h) / (r2 * h) + 1. + phi_2 = phi_1 / h + 1. + phi_3 = phi_2 / h - 0.5 + + if model_s is None: + model_s = self.model_fn(x, s) + if model_s1 is None: + x_s1 = ( + expand_dims(sigma_s1 / sigma_s, dims) * x + - expand_dims(alpha_s1 * phi_11, dims) * model_s + ) + model_s1 = self.model_fn(x_s1, s1) + x_s2 = ( + expand_dims(sigma_s2 / sigma_s, dims) * x + - expand_dims(alpha_s2 * phi_12, dims) * model_s + + r2 / r1 * expand_dims(alpha_s2 * phi_22, dims) * (model_s1 - model_s) + ) + model_s2 = self.model_fn(x_s2, s2) + if solver_type == 'dpm_solver': + x_t = ( + expand_dims(sigma_t / sigma_s, dims) * x + - expand_dims(alpha_t * phi_1, dims) * model_s + + (1. / r2) * expand_dims(alpha_t * phi_2, dims) * (model_s2 - model_s) + ) + elif solver_type == 'taylor': + D1_0 = (1. / r1) * (model_s1 - model_s) + D1_1 = (1. / r2) * (model_s2 - model_s) + D1 = (r2 * D1_0 - r1 * D1_1) / (r2 - r1) + D2 = 2. * (D1_1 - D1_0) / (r2 - r1) + x_t = ( + expand_dims(sigma_t / sigma_s, dims) * x + - expand_dims(alpha_t * phi_1, dims) * model_s + + expand_dims(alpha_t * phi_2, dims) * D1 + - expand_dims(alpha_t * phi_3, dims) * D2 + ) + else: + phi_11 = torch.expm1(r1 * h) + phi_12 = torch.expm1(r2 * h) + phi_1 = torch.expm1(h) + phi_22 = torch.expm1(r2 * h) / (r2 * h) - 1. + phi_2 = phi_1 / h - 1. + phi_3 = phi_2 / h - 0.5 + + if model_s is None: + model_s = self.model_fn(x, s) + if model_s1 is None: + x_s1 = ( + expand_dims(torch.exp(log_alpha_s1 - log_alpha_s), dims) * x + - expand_dims(sigma_s1 * phi_11, dims) * model_s + ) + model_s1 = self.model_fn(x_s1, s1) + x_s2 = ( + expand_dims(torch.exp(log_alpha_s2 - log_alpha_s), dims) * x + - expand_dims(sigma_s2 * phi_12, dims) * model_s + - r2 / r1 * expand_dims(sigma_s2 * phi_22, dims) * (model_s1 - model_s) + ) + model_s2 = self.model_fn(x_s2, s2) + if solver_type == 'dpm_solver': + x_t = ( + expand_dims(torch.exp(log_alpha_t - log_alpha_s), dims) * x + - expand_dims(sigma_t * phi_1, dims) * model_s + - (1. / r2) * expand_dims(sigma_t * phi_2, dims) * (model_s2 - model_s) + ) + elif solver_type == 'taylor': + D1_0 = (1. / r1) * (model_s1 - model_s) + D1_1 = (1. / r2) * (model_s2 - model_s) + D1 = (r2 * D1_0 - r1 * D1_1) / (r2 - r1) + D2 = 2. * (D1_1 - D1_0) / (r2 - r1) + x_t = ( + expand_dims(torch.exp(log_alpha_t - log_alpha_s), dims) * x + - expand_dims(sigma_t * phi_1, dims) * model_s + - expand_dims(sigma_t * phi_2, dims) * D1 + - expand_dims(sigma_t * phi_3, dims) * D2 + ) + + if return_intermediate: + return x_t, {'model_s': model_s, 'model_s1': model_s1, 'model_s2': model_s2} + else: + return x_t + + def multistep_dpm_solver_second_update(self, x, model_prev_list, t_prev_list, t, solver_type="dpm_solver"): + """ + Multistep solver DPM-Solver-2 from time `t_prev_list[-1]` to time `t`. + Args: + x: A pytorch tensor. The initial value at time `s`. + model_prev_list: A list of pytorch tensor. The previous computed model values. + t_prev_list: A list of pytorch tensor. The previous times, each time has the shape (x.shape[0],) + t: A pytorch tensor. The ending time, with the shape (x.shape[0],). + solver_type: either 'dpm_solver' or 'taylor'. The type for the high-order solvers. + The type slightly impacts the performance. We recommend to use 'dpm_solver' type. + Returns: + x_t: A pytorch tensor. The approximated solution at time `t`. + """ + if solver_type not in ['dpm_solver', 'taylor']: + raise ValueError("'solver_type' must be either 'dpm_solver' or 'taylor', got {}".format(solver_type)) + ns = self.noise_schedule + dims = x.dim() + model_prev_1, model_prev_0 = model_prev_list + t_prev_1, t_prev_0 = t_prev_list + lambda_prev_1, lambda_prev_0, lambda_t = ns.marginal_lambda(t_prev_1), ns.marginal_lambda( + t_prev_0), ns.marginal_lambda(t) + log_alpha_prev_0, log_alpha_t = ns.marginal_log_mean_coeff(t_prev_0), ns.marginal_log_mean_coeff(t) + sigma_prev_0, sigma_t = ns.marginal_std(t_prev_0), ns.marginal_std(t) + alpha_t = torch.exp(log_alpha_t) + + h_0 = lambda_prev_0 - lambda_prev_1 + h = lambda_t - lambda_prev_0 + r0 = h_0 / h + D1_0 = expand_dims(1. / r0, dims) * (model_prev_0 - model_prev_1) + if self.predict_x0: + if solver_type == 'dpm_solver': + x_t = ( + expand_dims(sigma_t / sigma_prev_0, dims) * x + - expand_dims(alpha_t * (torch.exp(-h) - 1.), dims) * model_prev_0 + - 0.5 * expand_dims(alpha_t * (torch.exp(-h) - 1.), dims) * D1_0 + ) + elif solver_type == 'taylor': + x_t = ( + expand_dims(sigma_t / sigma_prev_0, dims) * x + - expand_dims(alpha_t * (torch.exp(-h) - 1.), dims) * model_prev_0 + + expand_dims(alpha_t * ((torch.exp(-h) - 1.) / h + 1.), dims) * D1_0 + ) + else: + if solver_type == 'dpm_solver': + x_t = ( + expand_dims(torch.exp(log_alpha_t - log_alpha_prev_0), dims) * x + - expand_dims(sigma_t * (torch.exp(h) - 1.), dims) * model_prev_0 + - 0.5 * expand_dims(sigma_t * (torch.exp(h) - 1.), dims) * D1_0 + ) + elif solver_type == 'taylor': + x_t = ( + expand_dims(torch.exp(log_alpha_t - log_alpha_prev_0), dims) * x + - expand_dims(sigma_t * (torch.exp(h) - 1.), dims) * model_prev_0 + - expand_dims(sigma_t * ((torch.exp(h) - 1.) / h - 1.), dims) * D1_0 + ) + return x_t + + def multistep_dpm_solver_third_update(self, x, model_prev_list, t_prev_list, t, solver_type='dpm_solver'): + """ + Multistep solver DPM-Solver-3 from time `t_prev_list[-1]` to time `t`. + Args: + x: A pytorch tensor. The initial value at time `s`. + model_prev_list: A list of pytorch tensor. The previous computed model values. + t_prev_list: A list of pytorch tensor. The previous times, each time has the shape (x.shape[0],) + t: A pytorch tensor. The ending time, with the shape (x.shape[0],). + solver_type: either 'dpm_solver' or 'taylor'. The type for the high-order solvers. + The type slightly impacts the performance. We recommend to use 'dpm_solver' type. + Returns: + x_t: A pytorch tensor. The approximated solution at time `t`. + """ + ns = self.noise_schedule + dims = x.dim() + model_prev_2, model_prev_1, model_prev_0 = model_prev_list + t_prev_2, t_prev_1, t_prev_0 = t_prev_list + lambda_prev_2, lambda_prev_1, lambda_prev_0, lambda_t = ns.marginal_lambda(t_prev_2), ns.marginal_lambda( + t_prev_1), ns.marginal_lambda(t_prev_0), ns.marginal_lambda(t) + log_alpha_prev_0, log_alpha_t = ns.marginal_log_mean_coeff(t_prev_0), ns.marginal_log_mean_coeff(t) + sigma_prev_0, sigma_t = ns.marginal_std(t_prev_0), ns.marginal_std(t) + alpha_t = torch.exp(log_alpha_t) + + h_1 = lambda_prev_1 - lambda_prev_2 + h_0 = lambda_prev_0 - lambda_prev_1 + h = lambda_t - lambda_prev_0 + r0, r1 = h_0 / h, h_1 / h + D1_0 = expand_dims(1. / r0, dims) * (model_prev_0 - model_prev_1) + D1_1 = expand_dims(1. / r1, dims) * (model_prev_1 - model_prev_2) + D1 = D1_0 + expand_dims(r0 / (r0 + r1), dims) * (D1_0 - D1_1) + D2 = expand_dims(1. / (r0 + r1), dims) * (D1_0 - D1_1) + if self.predict_x0: + x_t = ( + expand_dims(sigma_t / sigma_prev_0, dims) * x + - expand_dims(alpha_t * (torch.exp(-h) - 1.), dims) * model_prev_0 + + expand_dims(alpha_t * ((torch.exp(-h) - 1.) / h + 1.), dims) * D1 + - expand_dims(alpha_t * ((torch.exp(-h) - 1. + h) / h ** 2 - 0.5), dims) * D2 + ) + else: + x_t = ( + expand_dims(torch.exp(log_alpha_t - log_alpha_prev_0), dims) * x + - expand_dims(sigma_t * (torch.exp(h) - 1.), dims) * model_prev_0 + - expand_dims(sigma_t * ((torch.exp(h) - 1.) / h - 1.), dims) * D1 + - expand_dims(sigma_t * ((torch.exp(h) - 1. - h) / h ** 2 - 0.5), dims) * D2 + ) + return x_t + + def singlestep_dpm_solver_update(self, x, s, t, order, return_intermediate=False, solver_type='dpm_solver', r1=None, + r2=None): + """ + Singlestep DPM-Solver with the order `order` from time `s` to time `t`. + Args: + x: A pytorch tensor. The initial value at time `s`. + s: A pytorch tensor. The starting time, with the shape (x.shape[0],). + t: A pytorch tensor. The ending time, with the shape (x.shape[0],). + order: A `int`. The order of DPM-Solver. We only support order == 1 or 2 or 3. + return_intermediate: A `bool`. If true, also return the model value at time `s`, `s1` and `s2` (the intermediate times). + solver_type: either 'dpm_solver' or 'taylor'. The type for the high-order solvers. + The type slightly impacts the performance. We recommend to use 'dpm_solver' type. + r1: A `float`. The hyperparameter of the second-order or third-order solver. + r2: A `float`. The hyperparameter of the third-order solver. + Returns: + x_t: A pytorch tensor. The approximated solution at time `t`. + """ + if order == 1: + return self.dpm_solver_first_update(x, s, t, return_intermediate=return_intermediate) + elif order == 2: + return self.singlestep_dpm_solver_second_update(x, s, t, return_intermediate=return_intermediate, + solver_type=solver_type, r1=r1) + elif order == 3: + return self.singlestep_dpm_solver_third_update(x, s, t, return_intermediate=return_intermediate, + solver_type=solver_type, r1=r1, r2=r2) + else: + raise ValueError("Solver order must be 1 or 2 or 3, got {}".format(order)) + + def multistep_dpm_solver_update(self, x, model_prev_list, t_prev_list, t, order, solver_type='dpm_solver'): + """ + Multistep DPM-Solver with the order `order` from time `t_prev_list[-1]` to time `t`. + Args: + x: A pytorch tensor. The initial value at time `s`. + model_prev_list: A list of pytorch tensor. The previous computed model values. + t_prev_list: A list of pytorch tensor. The previous times, each time has the shape (x.shape[0],) + t: A pytorch tensor. The ending time, with the shape (x.shape[0],). + order: A `int`. The order of DPM-Solver. We only support order == 1 or 2 or 3. + solver_type: either 'dpm_solver' or 'taylor'. The type for the high-order solvers. + The type slightly impacts the performance. We recommend to use 'dpm_solver' type. + Returns: + x_t: A pytorch tensor. The approximated solution at time `t`. + """ + if order == 1: + return self.dpm_solver_first_update(x, t_prev_list[-1], t, model_s=model_prev_list[-1]) + elif order == 2: + return self.multistep_dpm_solver_second_update(x, model_prev_list, t_prev_list, t, solver_type=solver_type) + elif order == 3: + return self.multistep_dpm_solver_third_update(x, model_prev_list, t_prev_list, t, solver_type=solver_type) + else: + raise ValueError("Solver order must be 1 or 2 or 3, got {}".format(order)) + + def dpm_solver_adaptive(self, x, order, t_T, t_0, h_init=0.05, atol=0.0078, rtol=0.05, theta=0.9, t_err=1e-5, + solver_type='dpm_solver'): + """ + The adaptive step size solver based on singlestep DPM-Solver. + Args: + x: A pytorch tensor. The initial value at time `t_T`. + order: A `int`. The (higher) order of the solver. We only support order == 2 or 3. + t_T: A `float`. The starting time of the sampling (default is T). + t_0: A `float`. The ending time of the sampling (default is epsilon). + h_init: A `float`. The initial step size (for logSNR). + atol: A `float`. The absolute tolerance of the solver. For image data, the default setting is 0.0078, followed [1]. + rtol: A `float`. The relative tolerance of the solver. The default setting is 0.05. + theta: A `float`. The safety hyperparameter for adapting the step size. The default setting is 0.9, followed [1]. + t_err: A `float`. The tolerance for the time. We solve the diffusion ODE until the absolute error between the + current time and `t_0` is less than `t_err`. The default setting is 1e-5. + solver_type: either 'dpm_solver' or 'taylor'. The type for the high-order solvers. + The type slightly impacts the performance. We recommend to use 'dpm_solver' type. + Returns: + x_0: A pytorch tensor. The approximated solution at time `t_0`. + [1] A. Jolicoeur-Martineau, K. Li, R. Piché-Taillefer, T. Kachman, and I. Mitliagkas, "Gotta go fast when generating data with score-based models," arXiv preprint arXiv:2105.14080, 2021. + """ + ns = self.noise_schedule + s = t_T * torch.ones((x.shape[0],)).to(x) + lambda_s = ns.marginal_lambda(s) + lambda_0 = ns.marginal_lambda(t_0 * torch.ones_like(s).to(x)) + h = h_init * torch.ones_like(s).to(x) + x_prev = x + nfe = 0 + if order == 2: + r1 = 0.5 + lower_update = lambda x, s, t: self.dpm_solver_first_update(x, s, t, return_intermediate=True) + higher_update = lambda x, s, t, **kwargs: self.singlestep_dpm_solver_second_update(x, s, t, r1=r1, + solver_type=solver_type, + **kwargs) + elif order == 3: + r1, r2 = 1. / 3., 2. / 3. + lower_update = lambda x, s, t: self.singlestep_dpm_solver_second_update(x, s, t, r1=r1, + return_intermediate=True, + solver_type=solver_type) + higher_update = lambda x, s, t, **kwargs: self.singlestep_dpm_solver_third_update(x, s, t, r1=r1, r2=r2, + solver_type=solver_type, + **kwargs) + else: + raise ValueError("For adaptive step size solver, order must be 2 or 3, got {}".format(order)) + while torch.abs((s - t_0)).mean() > t_err: + t = ns.inverse_lambda(lambda_s + h) + x_lower, lower_noise_kwargs = lower_update(x, s, t) + x_higher = higher_update(x, s, t, **lower_noise_kwargs) + delta = torch.max(torch.ones_like(x).to(x) * atol, rtol * torch.max(torch.abs(x_lower), torch.abs(x_prev))) + norm_fn = lambda v: torch.sqrt(torch.square(v.reshape((v.shape[0], -1))).mean(dim=-1, keepdim=True)) + E = norm_fn((x_higher - x_lower) / delta).max() + if torch.all(E <= 1.): + x = x_higher + s = t + x_prev = x_lower + lambda_s = ns.marginal_lambda(s) + h = torch.min(theta * h * torch.float_power(E, -1. / order).float(), lambda_0 - lambda_s) + nfe += order + print('adaptive solver nfe', nfe) + return x + + def sample(self, x, steps=20, t_start=None, t_end=None, order=3, skip_type='time_uniform', + method='singlestep', lower_order_final=True, denoise_to_zero=False, solver_type='dpm_solver', + atol=0.0078, rtol=0.05, + ): + """ + Compute the sample at time `t_end` by DPM-Solver, given the initial `x` at time `t_start`. + ===================================================== + We support the following algorithms for both noise prediction model and data prediction model: + - 'singlestep': + Singlestep DPM-Solver (i.e. "DPM-Solver-fast" in the paper), which combines different orders of singlestep DPM-Solver. + We combine all the singlestep solvers with order <= `order` to use up all the function evaluations (steps). + The total number of function evaluations (NFE) == `steps`. + Given a fixed NFE == `steps`, the sampling procedure is: + - If `order` == 1: + - Denote K = steps. We use K steps of DPM-Solver-1 (i.e. DDIM). + - If `order` == 2: + - Denote K = (steps // 2) + (steps % 2). We take K intermediate time steps for sampling. + - If steps % 2 == 0, we use K steps of singlestep DPM-Solver-2. + - If steps % 2 == 1, we use (K - 1) steps of singlestep DPM-Solver-2 and 1 step of DPM-Solver-1. + - If `order` == 3: + - Denote K = (steps // 3 + 1). We take K intermediate time steps for sampling. + - If steps % 3 == 0, we use (K - 2) steps of singlestep DPM-Solver-3, and 1 step of singlestep DPM-Solver-2 and 1 step of DPM-Solver-1. + - If steps % 3 == 1, we use (K - 1) steps of singlestep DPM-Solver-3 and 1 step of DPM-Solver-1. + - If steps % 3 == 2, we use (K - 1) steps of singlestep DPM-Solver-3 and 1 step of singlestep DPM-Solver-2. + - 'multistep': + Multistep DPM-Solver with the order of `order`. The total number of function evaluations (NFE) == `steps`. + We initialize the first `order` values by lower order multistep solvers. + Given a fixed NFE == `steps`, the sampling procedure is: + Denote K = steps. + - If `order` == 1: + - We use K steps of DPM-Solver-1 (i.e. DDIM). + - If `order` == 2: + - We firstly use 1 step of DPM-Solver-1, then use (K - 1) step of multistep DPM-Solver-2. + - If `order` == 3: + - We firstly use 1 step of DPM-Solver-1, then 1 step of multistep DPM-Solver-2, then (K - 2) step of multistep DPM-Solver-3. + - 'singlestep_fixed': + Fixed order singlestep DPM-Solver (i.e. DPM-Solver-1 or singlestep DPM-Solver-2 or singlestep DPM-Solver-3). + We use singlestep DPM-Solver-`order` for `order`=1 or 2 or 3, with total [`steps` // `order`] * `order` NFE. + - 'adaptive': + Adaptive step size DPM-Solver (i.e. "DPM-Solver-12" and "DPM-Solver-23" in the paper). + We ignore `steps` and use adaptive step size DPM-Solver with a higher order of `order`. + You can adjust the absolute tolerance `atol` and the relative tolerance `rtol` to balance the computatation costs + (NFE) and the sample quality. + - If `order` == 2, we use DPM-Solver-12 which combines DPM-Solver-1 and singlestep DPM-Solver-2. + - If `order` == 3, we use DPM-Solver-23 which combines singlestep DPM-Solver-2 and singlestep DPM-Solver-3. + ===================================================== + Some advices for choosing the algorithm: + - For **unconditional sampling** or **guided sampling with small guidance scale** by DPMs: + Use singlestep DPM-Solver ("DPM-Solver-fast" in the paper) with `order = 3`. + e.g. + >>> dpm_solver = DPM_Solver(model_fn, noise_schedule, predict_x0=False) + >>> x_sample = dpm_solver.sample(x, steps=steps, t_start=t_start, t_end=t_end, order=3, + skip_type='time_uniform', method='singlestep') + - For **guided sampling with large guidance scale** by DPMs: + Use multistep DPM-Solver with `predict_x0 = True` and `order = 2`. + e.g. + >>> dpm_solver = DPM_Solver(model_fn, noise_schedule, predict_x0=True) + >>> x_sample = dpm_solver.sample(x, steps=steps, t_start=t_start, t_end=t_end, order=2, + skip_type='time_uniform', method='multistep') + We support three types of `skip_type`: + - 'logSNR': uniform logSNR for the time steps. **Recommended for low-resolutional images** + - 'time_uniform': uniform time for the time steps. **Recommended for high-resolutional images**. + - 'time_quadratic': quadratic time for the time steps. + ===================================================== + Args: + x: A pytorch tensor. The initial value at time `t_start` + e.g. if `t_start` == T, then `x` is a sample from the standard normal distribution. + steps: A `int`. The total number of function evaluations (NFE). + t_start: A `float`. The starting time of the sampling. + If `T` is None, we use self.noise_schedule.T (default is 1.0). + t_end: A `float`. The ending time of the sampling. + If `t_end` is None, we use 1. / self.noise_schedule.total_N. + e.g. if total_N == 1000, we have `t_end` == 1e-3. + For discrete-time DPMs: + - We recommend `t_end` == 1. / self.noise_schedule.total_N. + For continuous-time DPMs: + - We recommend `t_end` == 1e-3 when `steps` <= 15; and `t_end` == 1e-4 when `steps` > 15. + order: A `int`. The order of DPM-Solver. + skip_type: A `str`. The type for the spacing of the time steps. 'time_uniform' or 'logSNR' or 'time_quadratic'. + method: A `str`. The method for sampling. 'singlestep' or 'multistep' or 'singlestep_fixed' or 'adaptive'. + denoise_to_zero: A `bool`. Whether to denoise to time 0 at the final step. + Default is `False`. If `denoise_to_zero` is `True`, the total NFE is (`steps` + 1). + This trick is firstly proposed by DDPM (https://arxiv.org/abs/2006.11239) and + score_sde (https://arxiv.org/abs/2011.13456). Such trick can improve the FID + for diffusion models sampling by diffusion SDEs for low-resolutional images + (such as CIFAR-10). However, we observed that such trick does not matter for + high-resolutional images. As it needs an additional NFE, we do not recommend + it for high-resolutional images. + lower_order_final: A `bool`. Whether to use lower order solvers at the final steps. + Only valid for `method=multistep` and `steps < 15`. We empirically find that + this trick is a key to stabilizing the sampling by DPM-Solver with very few steps + (especially for steps <= 10). So we recommend to set it to be `True`. + solver_type: A `str`. The taylor expansion type for the solver. `dpm_solver` or `taylor`. We recommend `dpm_solver`. + atol: A `float`. The absolute tolerance of the adaptive step size solver. Valid when `method` == 'adaptive'. + rtol: A `float`. The relative tolerance of the adaptive step size solver. Valid when `method` == 'adaptive'. + Returns: + x_end: A pytorch tensor. The approximated solution at time `t_end`. + """ + t_0 = 1. / self.noise_schedule.total_N if t_end is None else t_end + t_T = self.noise_schedule.T if t_start is None else t_start + device = x.device + if method == 'adaptive': + with torch.no_grad(): + x = self.dpm_solver_adaptive(x, order=order, t_T=t_T, t_0=t_0, atol=atol, rtol=rtol, + solver_type=solver_type) + elif method == 'multistep': + assert steps >= order + timesteps = self.get_time_steps(skip_type=skip_type, t_T=t_T, t_0=t_0, N=steps, device=device) + assert timesteps.shape[0] - 1 == steps + with torch.no_grad(): + vec_t = timesteps[0].expand((x.shape[0])) + model_prev_list = [self.model_fn(x, vec_t)] + t_prev_list = [vec_t] + # Init the first `order` values by lower order multistep DPM-Solver. + for init_order in tqdm(range(1, order), desc="DPM init order"): + vec_t = timesteps[init_order].expand(x.shape[0]) + x = self.multistep_dpm_solver_update(x, model_prev_list, t_prev_list, vec_t, init_order, + solver_type=solver_type) + model_prev_list.append(self.model_fn(x, vec_t)) + t_prev_list.append(vec_t) + # Compute the remaining values by `order`-th order multistep DPM-Solver. + for step in tqdm(range(order, steps + 1), desc="DPM multistep"): + vec_t = timesteps[step].expand(x.shape[0]) + if lower_order_final and steps < 15: + step_order = min(order, steps + 1 - step) + else: + step_order = order + x = self.multistep_dpm_solver_update(x, model_prev_list, t_prev_list, vec_t, step_order, + solver_type=solver_type) + for i in range(order - 1): + t_prev_list[i] = t_prev_list[i + 1] + model_prev_list[i] = model_prev_list[i + 1] + t_prev_list[-1] = vec_t + # We do not need to evaluate the final model value. + if step < steps: + model_prev_list[-1] = self.model_fn(x, vec_t) + elif method in ['singlestep', 'singlestep_fixed']: + if method == 'singlestep': + timesteps_outer, orders = self.get_orders_and_timesteps_for_singlestep_solver(steps=steps, order=order, + skip_type=skip_type, + t_T=t_T, t_0=t_0, + device=device) + elif method == 'singlestep_fixed': + K = steps // order + orders = [order, ] * K + timesteps_outer = self.get_time_steps(skip_type=skip_type, t_T=t_T, t_0=t_0, N=K, device=device) + for i, order in enumerate(orders): + t_T_inner, t_0_inner = timesteps_outer[i], timesteps_outer[i + 1] + timesteps_inner = self.get_time_steps(skip_type=skip_type, t_T=t_T_inner.item(), t_0=t_0_inner.item(), + N=order, device=device) + lambda_inner = self.noise_schedule.marginal_lambda(timesteps_inner) + vec_s, vec_t = t_T_inner.tile(x.shape[0]), t_0_inner.tile(x.shape[0]) + h = lambda_inner[-1] - lambda_inner[0] + r1 = None if order <= 1 else (lambda_inner[1] - lambda_inner[0]) / h + r2 = None if order <= 2 else (lambda_inner[2] - lambda_inner[0]) / h + x = self.singlestep_dpm_solver_update(x, vec_s, vec_t, order, solver_type=solver_type, r1=r1, r2=r2) + if denoise_to_zero: + x = self.denoise_to_zero_fn(x, torch.ones((x.shape[0],)).to(device) * t_0) + return x + + +############################################################# +# other utility functions +############################################################# + +def interpolate_fn(x, xp, yp): + """ + A piecewise linear function y = f(x), using xp and yp as keypoints. + We implement f(x) in a differentiable way (i.e. applicable for autograd). + The function f(x) is well-defined for all x-axis. (For x beyond the bounds of xp, we use the outmost points of xp to define the linear function.) + Args: + x: PyTorch tensor with shape [N, C], where N is the batch size, C is the number of channels (we use C = 1 for DPM-Solver). + xp: PyTorch tensor with shape [C, K], where K is the number of keypoints. + yp: PyTorch tensor with shape [C, K]. + Returns: + The function values f(x), with shape [N, C]. + """ + N, K = x.shape[0], xp.shape[1] + all_x = torch.cat([x.unsqueeze(2), xp.unsqueeze(0).repeat((N, 1, 1))], dim=2) + sorted_all_x, x_indices = torch.sort(all_x, dim=2) + x_idx = torch.argmin(x_indices, dim=2) + cand_start_idx = x_idx - 1 + start_idx = torch.where( + torch.eq(x_idx, 0), + torch.tensor(1, device=x.device), + torch.where( + torch.eq(x_idx, K), torch.tensor(K - 2, device=x.device), cand_start_idx, + ), + ) + end_idx = torch.where(torch.eq(start_idx, cand_start_idx), start_idx + 2, start_idx + 1) + start_x = torch.gather(sorted_all_x, dim=2, index=start_idx.unsqueeze(2)).squeeze(2) + end_x = torch.gather(sorted_all_x, dim=2, index=end_idx.unsqueeze(2)).squeeze(2) + start_idx2 = torch.where( + torch.eq(x_idx, 0), + torch.tensor(0, device=x.device), + torch.where( + torch.eq(x_idx, K), torch.tensor(K - 2, device=x.device), cand_start_idx, + ), + ) + y_positions_expanded = yp.unsqueeze(0).expand(N, -1, -1) + start_y = torch.gather(y_positions_expanded, dim=2, index=start_idx2.unsqueeze(2)).squeeze(2) + end_y = torch.gather(y_positions_expanded, dim=2, index=(start_idx2 + 1).unsqueeze(2)).squeeze(2) + cand = start_y + (x - start_x) * (end_y - start_y) / (end_x - start_x) + return cand + + +def expand_dims(v, dims): + """ + Expand the tensor `v` to the dim `dims`. + Args: + `v`: a PyTorch tensor with shape [N]. + `dim`: a `int`. + Returns: + a PyTorch tensor with shape [N, 1, 1, ..., 1] and the total dimension is `dims`. + """ + return v[(...,) + (None,) * (dims - 1)] \ No newline at end of file diff --git a/ldm/models/diffusion/dpm_solver/sampler.py b/ldm/models/diffusion/dpm_solver/sampler.py new file mode 100644 index 0000000000000000000000000000000000000000..7d137b8cf36718c1c58faa09f9dd919e5fb2977b --- /dev/null +++ b/ldm/models/diffusion/dpm_solver/sampler.py @@ -0,0 +1,87 @@ +"""SAMPLING ONLY.""" +import torch + +from .dpm_solver import NoiseScheduleVP, model_wrapper, DPM_Solver + + +MODEL_TYPES = { + "eps": "noise", + "v": "v" +} + + +class DPMSolverSampler(object): + def __init__(self, model, **kwargs): + super().__init__() + self.model = model + to_torch = lambda x: x.clone().detach().to(torch.float32).to(model.device) + self.register_buffer('alphas_cumprod', to_torch(model.alphas_cumprod)) + + def register_buffer(self, name, attr): + if type(attr) == torch.Tensor: + if attr.device != torch.device("cuda"): + attr = attr.to(torch.device("cuda")) + setattr(self, name, attr) + + @torch.no_grad() + def sample(self, + S, + batch_size, + shape, + conditioning=None, + callback=None, + normals_sequence=None, + img_callback=None, + quantize_x0=False, + eta=0., + mask=None, + x0=None, + temperature=1., + noise_dropout=0., + score_corrector=None, + corrector_kwargs=None, + verbose=True, + x_T=None, + log_every_t=100, + unconditional_guidance_scale=1., + unconditional_conditioning=None, + # this has to come in the same format as the conditioning, # e.g. as encoded tokens, ... + **kwargs + ): + if conditioning is not None: + if isinstance(conditioning, dict): + cbs = conditioning[list(conditioning.keys())[0]].shape[0] + if cbs != batch_size: + print(f"Warning: Got {cbs} conditionings but batch-size is {batch_size}") + else: + if conditioning.shape[0] != batch_size: + print(f"Warning: Got {conditioning.shape[0]} conditionings but batch-size is {batch_size}") + + # sampling + C, H, W = shape + size = (batch_size, C, H, W) + + print(f'Data shape for DPM-Solver sampling is {size}, sampling steps {S}') + + device = self.model.betas.device + if x_T is None: + img = torch.randn(size, device=device) + else: + img = x_T + + ns = NoiseScheduleVP('discrete', alphas_cumprod=self.alphas_cumprod) + + model_fn = model_wrapper( + lambda x, t, c: self.model.apply_model(x, t, c), + ns, + model_type=MODEL_TYPES[self.model.parameterization], + guidance_type="classifier-free", + condition=conditioning, + unconditional_condition=unconditional_conditioning, + guidance_scale=unconditional_guidance_scale, + ) + + dpm_solver = DPM_Solver(model_fn, ns, predict_x0=True, thresholding=False) + x = dpm_solver.sample(img, steps=S, skip_type="time_uniform", method="multistep", order=2, lower_order_final=True) + + return x.to(device), None \ No newline at end of file diff --git a/ldm/models/diffusion/plms.py b/ldm/models/diffusion/plms.py new file mode 100644 index 0000000000000000000000000000000000000000..7002a365d27168ced0a04e9a4d83e088f8284eae --- /dev/null +++ b/ldm/models/diffusion/plms.py @@ -0,0 +1,244 @@ +"""SAMPLING ONLY.""" + +import torch +import numpy as np +from tqdm import tqdm +from functools import partial + +from ldm.modules.diffusionmodules.util import make_ddim_sampling_parameters, make_ddim_timesteps, noise_like +from ldm.models.diffusion.sampling_util import norm_thresholding + + +class PLMSSampler(object): + def __init__(self, model, schedule="linear", **kwargs): + super().__init__() + self.model = model + self.ddpm_num_timesteps = model.num_timesteps + self.schedule = schedule + + def register_buffer(self, name, attr): + if type(attr) == torch.Tensor: + if attr.device != torch.device("cuda"): + attr = attr.to(torch.device("cuda")) + setattr(self, name, attr) + + def make_schedule(self, ddim_num_steps, ddim_discretize="uniform", ddim_eta=0., verbose=True): + if ddim_eta != 0: + raise ValueError('ddim_eta must be 0 for PLMS') + self.ddim_timesteps = make_ddim_timesteps(ddim_discr_method=ddim_discretize, num_ddim_timesteps=ddim_num_steps, + num_ddpm_timesteps=self.ddpm_num_timesteps,verbose=verbose) + alphas_cumprod = self.model.alphas_cumprod + assert alphas_cumprod.shape[0] == self.ddpm_num_timesteps, 'alphas have to be defined for each timestep' + to_torch = lambda x: x.clone().detach().to(torch.float32).to(self.model.device) + + self.register_buffer('betas', to_torch(self.model.betas)) + self.register_buffer('alphas_cumprod', to_torch(alphas_cumprod)) + self.register_buffer('alphas_cumprod_prev', to_torch(self.model.alphas_cumprod_prev)) + + # calculations for diffusion q(x_t | x_{t-1}) and others + self.register_buffer('sqrt_alphas_cumprod', to_torch(np.sqrt(alphas_cumprod.cpu()))) + self.register_buffer('sqrt_one_minus_alphas_cumprod', to_torch(np.sqrt(1. - alphas_cumprod.cpu()))) + self.register_buffer('log_one_minus_alphas_cumprod', to_torch(np.log(1. - alphas_cumprod.cpu()))) + self.register_buffer('sqrt_recip_alphas_cumprod', to_torch(np.sqrt(1. / alphas_cumprod.cpu()))) + self.register_buffer('sqrt_recipm1_alphas_cumprod', to_torch(np.sqrt(1. / alphas_cumprod.cpu() - 1))) + + # ddim sampling parameters + ddim_sigmas, ddim_alphas, ddim_alphas_prev = make_ddim_sampling_parameters(alphacums=alphas_cumprod.cpu(), + ddim_timesteps=self.ddim_timesteps, + eta=ddim_eta,verbose=verbose) + self.register_buffer('ddim_sigmas', ddim_sigmas) + self.register_buffer('ddim_alphas', ddim_alphas) + self.register_buffer('ddim_alphas_prev', ddim_alphas_prev) + self.register_buffer('ddim_sqrt_one_minus_alphas', np.sqrt(1. - ddim_alphas)) + sigmas_for_original_sampling_steps = ddim_eta * torch.sqrt( + (1 - self.alphas_cumprod_prev) / (1 - self.alphas_cumprod) * ( + 1 - self.alphas_cumprod / self.alphas_cumprod_prev)) + self.register_buffer('ddim_sigmas_for_original_num_steps', sigmas_for_original_sampling_steps) + + @torch.no_grad() + def sample(self, + S, + batch_size, + shape, + conditioning=None, + callback=None, + normals_sequence=None, + img_callback=None, + quantize_x0=False, + eta=0., + mask=None, + x0=None, + temperature=1., + noise_dropout=0., + score_corrector=None, + corrector_kwargs=None, + verbose=True, + x_T=None, + log_every_t=100, + unconditional_guidance_scale=1., + unconditional_conditioning=None, + # this has to come in the same format as the conditioning, # e.g. as encoded tokens, ... + dynamic_threshold=None, + **kwargs + ): + if conditioning is not None: + if isinstance(conditioning, dict): + cbs = conditioning[list(conditioning.keys())[0]].shape[0] + if cbs != batch_size: + print(f"Warning: Got {cbs} conditionings but batch-size is {batch_size}") + else: + if conditioning.shape[0] != batch_size: + print(f"Warning: Got {conditioning.shape[0]} conditionings but batch-size is {batch_size}") + + self.make_schedule(ddim_num_steps=S, ddim_eta=eta, verbose=verbose) + # sampling + C, H, W = shape + size = (batch_size, C, H, W) + print(f'Data shape for PLMS sampling is {size}') + + samples, intermediates = self.plms_sampling(conditioning, size, + callback=callback, + img_callback=img_callback, + quantize_denoised=quantize_x0, + mask=mask, x0=x0, + ddim_use_original_steps=False, + noise_dropout=noise_dropout, + temperature=temperature, + score_corrector=score_corrector, + corrector_kwargs=corrector_kwargs, + x_T=x_T, + log_every_t=log_every_t, + unconditional_guidance_scale=unconditional_guidance_scale, + unconditional_conditioning=unconditional_conditioning, + dynamic_threshold=dynamic_threshold, + ) + return samples, intermediates + + @torch.no_grad() + def plms_sampling(self, cond, shape, + x_T=None, ddim_use_original_steps=False, + callback=None, timesteps=None, quantize_denoised=False, + mask=None, x0=None, img_callback=None, log_every_t=100, + temperature=1., noise_dropout=0., score_corrector=None, corrector_kwargs=None, + unconditional_guidance_scale=1., unconditional_conditioning=None, + dynamic_threshold=None): + device = self.model.betas.device + b = shape[0] + if x_T is None: + img = torch.randn(shape, device=device) + else: + img = x_T + + if timesteps is None: + timesteps = self.ddpm_num_timesteps if ddim_use_original_steps else self.ddim_timesteps + elif timesteps is not None and not ddim_use_original_steps: + subset_end = int(min(timesteps / self.ddim_timesteps.shape[0], 1) * self.ddim_timesteps.shape[0]) - 1 + timesteps = self.ddim_timesteps[:subset_end] + + intermediates = {'x_inter': [img], 'pred_x0': [img]} + time_range = list(reversed(range(0,timesteps))) if ddim_use_original_steps else np.flip(timesteps) + total_steps = timesteps if ddim_use_original_steps else timesteps.shape[0] + print(f"Running PLMS Sampling with {total_steps} timesteps") + + iterator = tqdm(time_range, desc='PLMS Sampler', total=total_steps) + old_eps = [] + + for i, step in enumerate(iterator): + index = total_steps - i - 1 + ts = torch.full((b,), step, device=device, dtype=torch.long) + ts_next = torch.full((b,), time_range[min(i + 1, len(time_range) - 1)], device=device, dtype=torch.long) + + if mask is not None: + assert x0 is not None + img_orig = self.model.q_sample(x0, ts) # TODO: deterministic forward pass? + img = img_orig * mask + (1. - mask) * img + + outs = self.p_sample_plms(img, cond, ts, index=index, use_original_steps=ddim_use_original_steps, + quantize_denoised=quantize_denoised, temperature=temperature, + noise_dropout=noise_dropout, score_corrector=score_corrector, + corrector_kwargs=corrector_kwargs, + unconditional_guidance_scale=unconditional_guidance_scale, + unconditional_conditioning=unconditional_conditioning, + old_eps=old_eps, t_next=ts_next, + dynamic_threshold=dynamic_threshold) + img, pred_x0, e_t = outs + old_eps.append(e_t) + if len(old_eps) >= 4: + old_eps.pop(0) + if callback: callback(i) + if img_callback: img_callback(pred_x0, i) + + if index % log_every_t == 0 or index == total_steps - 1: + intermediates['x_inter'].append(img) + intermediates['pred_x0'].append(pred_x0) + + return img, intermediates + + @torch.no_grad() + def p_sample_plms(self, x, c, t, index, repeat_noise=False, use_original_steps=False, quantize_denoised=False, + temperature=1., noise_dropout=0., score_corrector=None, corrector_kwargs=None, + unconditional_guidance_scale=1., unconditional_conditioning=None, old_eps=None, t_next=None, + dynamic_threshold=None): + b, *_, device = *x.shape, x.device + + def get_model_output(x, t): + if unconditional_conditioning is None or unconditional_guidance_scale == 1.: + e_t = self.model.apply_model(x, t, c) + else: + x_in = torch.cat([x] * 2) + t_in = torch.cat([t] * 2) + c_in = torch.cat([unconditional_conditioning, c]) + e_t_uncond, e_t = self.model.apply_model(x_in, t_in, c_in).chunk(2) + e_t = e_t_uncond + unconditional_guidance_scale * (e_t - e_t_uncond) + + if score_corrector is not None: + assert self.model.parameterization == "eps" + e_t = score_corrector.modify_score(self.model, e_t, x, t, c, **corrector_kwargs) + + return e_t + + alphas = self.model.alphas_cumprod if use_original_steps else self.ddim_alphas + alphas_prev = self.model.alphas_cumprod_prev if use_original_steps else self.ddim_alphas_prev + sqrt_one_minus_alphas = self.model.sqrt_one_minus_alphas_cumprod if use_original_steps else self.ddim_sqrt_one_minus_alphas + sigmas = self.model.ddim_sigmas_for_original_num_steps if use_original_steps else self.ddim_sigmas + + def get_x_prev_and_pred_x0(e_t, index): + # select parameters corresponding to the currently considered timestep + a_t = torch.full((b, 1, 1, 1), alphas[index], device=device) + a_prev = torch.full((b, 1, 1, 1), alphas_prev[index], device=device) + sigma_t = torch.full((b, 1, 1, 1), sigmas[index], device=device) + sqrt_one_minus_at = torch.full((b, 1, 1, 1), sqrt_one_minus_alphas[index],device=device) + + # current prediction for x_0 + pred_x0 = (x - sqrt_one_minus_at * e_t) / a_t.sqrt() + if quantize_denoised: + pred_x0, _, *_ = self.model.first_stage_model.quantize(pred_x0) + if dynamic_threshold is not None: + pred_x0 = norm_thresholding(pred_x0, dynamic_threshold) + # direction pointing to x_t + dir_xt = (1. - a_prev - sigma_t**2).sqrt() * e_t + noise = sigma_t * noise_like(x.shape, device, repeat_noise) * temperature + if noise_dropout > 0.: + noise = torch.nn.functional.dropout(noise, p=noise_dropout) + x_prev = a_prev.sqrt() * pred_x0 + dir_xt + noise + return x_prev, pred_x0 + + e_t = get_model_output(x, t) + if len(old_eps) == 0: + # Pseudo Improved Euler (2nd order) + x_prev, pred_x0 = get_x_prev_and_pred_x0(e_t, index) + e_t_next = get_model_output(x_prev, t_next) + e_t_prime = (e_t + e_t_next) / 2 + elif len(old_eps) == 1: + # 2nd order Pseudo Linear Multistep (Adams-Bashforth) + e_t_prime = (3 * e_t - old_eps[-1]) / 2 + elif len(old_eps) == 2: + # 3nd order Pseudo Linear Multistep (Adams-Bashforth) + e_t_prime = (23 * e_t - 16 * old_eps[-1] + 5 * old_eps[-2]) / 12 + elif len(old_eps) >= 3: + # 4nd order Pseudo Linear Multistep (Adams-Bashforth) + e_t_prime = (55 * e_t - 59 * old_eps[-1] + 37 * old_eps[-2] - 9 * old_eps[-3]) / 24 + + x_prev, pred_x0 = get_x_prev_and_pred_x0(e_t_prime, index) + + return x_prev, pred_x0, e_t diff --git a/ldm/models/diffusion/sampling_util.py b/ldm/models/diffusion/sampling_util.py new file mode 100644 index 0000000000000000000000000000000000000000..7eff02be6d7c54d43ee6680636ac0698dd3b3f33 --- /dev/null +++ b/ldm/models/diffusion/sampling_util.py @@ -0,0 +1,22 @@ +import torch +import numpy as np + + +def append_dims(x, target_dims): + """Appends dimensions to the end of a tensor until it has target_dims dimensions. + From https://github.com/crowsonkb/k-diffusion/blob/master/k_diffusion/utils.py""" + dims_to_append = target_dims - x.ndim + if dims_to_append < 0: + raise ValueError(f'input has {x.ndim} dims but target_dims is {target_dims}, which is less') + return x[(...,) + (None,) * dims_to_append] + + +def norm_thresholding(x0, value): + s = append_dims(x0.pow(2).flatten(1).mean(1).sqrt().clamp(min=value), x0.ndim) + return x0 * (value / s) + + +def spatial_norm_thresholding(x0, value): + # b c h w + s = x0.pow(2).mean(1, keepdim=True).sqrt().clamp(min=value) + return x0 * (value / s) \ No newline at end of file diff --git a/ldm/modules/attention.py b/ldm/modules/attention.py new file mode 100644 index 0000000000000000000000000000000000000000..509cd873768f0dd75a75ab3fcdd652822b12b59f --- /dev/null +++ b/ldm/modules/attention.py @@ -0,0 +1,341 @@ +from inspect import isfunction +import math +import torch +import torch.nn.functional as F +from torch import nn, einsum +from einops import rearrange, repeat +from typing import Optional, Any + +from ldm.modules.diffusionmodules.util import checkpoint + + +try: + import xformers + import xformers.ops + XFORMERS_IS_AVAILBLE = True +except: + XFORMERS_IS_AVAILBLE = False + +# CrossAttn precision handling +import os +_ATTN_PRECISION = os.environ.get("ATTN_PRECISION", "fp32") + +def exists(val): + return val is not None + + +def uniq(arr): + return{el: True for el in arr}.keys() + + +def default(val, d): + if exists(val): + return val + return d() if isfunction(d) else d + + +def max_neg_value(t): + return -torch.finfo(t.dtype).max + + +def init_(tensor): + dim = tensor.shape[-1] + std = 1 / math.sqrt(dim) + tensor.uniform_(-std, std) + return tensor + + +# feedforward +class GEGLU(nn.Module): + def __init__(self, dim_in, dim_out): + super().__init__() + self.proj = nn.Linear(dim_in, dim_out * 2) + + def forward(self, x): + x, gate = self.proj(x).chunk(2, dim=-1) + return x * F.gelu(gate) + + +class FeedForward(nn.Module): + def __init__(self, dim, dim_out=None, mult=4, glu=False, dropout=0.): + super().__init__() + inner_dim = int(dim * mult) + dim_out = default(dim_out, dim) + project_in = nn.Sequential( + nn.Linear(dim, inner_dim), + nn.GELU() + ) if not glu else GEGLU(dim, inner_dim) + + self.net = nn.Sequential( + project_in, + nn.Dropout(dropout), + nn.Linear(inner_dim, dim_out) + ) + + def forward(self, x): + return self.net(x) + + +def zero_module(module): + """ + Zero out the parameters of a module and return it. + """ + for p in module.parameters(): + p.detach().zero_() + return module + + +def Normalize(in_channels): + return torch.nn.GroupNorm(num_groups=32, num_channels=in_channels, eps=1e-6, affine=True) + + +class SpatialSelfAttention(nn.Module): + def __init__(self, in_channels): + super().__init__() + self.in_channels = in_channels + + self.norm = Normalize(in_channels) + self.q = torch.nn.Conv2d(in_channels, + in_channels, + kernel_size=1, + stride=1, + padding=0) + self.k = torch.nn.Conv2d(in_channels, + in_channels, + kernel_size=1, + stride=1, + padding=0) + self.v = torch.nn.Conv2d(in_channels, + in_channels, + kernel_size=1, + stride=1, + padding=0) + self.proj_out = torch.nn.Conv2d(in_channels, + in_channels, + kernel_size=1, + stride=1, + padding=0) + + def forward(self, x): + h_ = x + h_ = self.norm(h_) + q = self.q(h_) + k = self.k(h_) + v = self.v(h_) + + # compute attention + b,c,h,w = q.shape + q = rearrange(q, 'b c h w -> b (h w) c') + k = rearrange(k, 'b c h w -> b c (h w)') + w_ = torch.einsum('bij,bjk->bik', q, k) + + w_ = w_ * (int(c)**(-0.5)) + w_ = torch.nn.functional.softmax(w_, dim=2) + + # attend to values + v = rearrange(v, 'b c h w -> b c (h w)') + w_ = rearrange(w_, 'b i j -> b j i') + h_ = torch.einsum('bij,bjk->bik', v, w_) + h_ = rearrange(h_, 'b c (h w) -> b c h w', h=h) + h_ = self.proj_out(h_) + + return x+h_ + + +class CrossAttention(nn.Module): + def __init__(self, query_dim, context_dim=None, heads=8, dim_head=64, dropout=0.): + super().__init__() + inner_dim = dim_head * heads + context_dim = default(context_dim, query_dim) + + self.scale = dim_head ** -0.5 + self.heads = heads + + self.to_q = nn.Linear(query_dim, inner_dim, bias=False) + self.to_k = nn.Linear(context_dim, inner_dim, bias=False) + self.to_v = nn.Linear(context_dim, inner_dim, bias=False) + + self.to_out = nn.Sequential( + nn.Linear(inner_dim, query_dim), + nn.Dropout(dropout) + ) + + def forward(self, x, context=None, mask=None): + h = self.heads + + q = self.to_q(x) + context = default(context, x) + k = self.to_k(context) + v = self.to_v(context) + + q, k, v = map(lambda t: rearrange(t, 'b n (h d) -> (b h) n d', h=h), (q, k, v)) + + # force cast to fp32 to avoid overflowing + if _ATTN_PRECISION =="fp32": + with torch.autocast(enabled=False, device_type = 'cuda'): + q, k = q.float(), k.float() + sim = einsum('b i d, b j d -> b i j', q, k) * self.scale + else: + sim = einsum('b i d, b j d -> b i j', q, k) * self.scale + + del q, k + + if exists(mask): + mask = rearrange(mask, 'b ... -> b (...)') + max_neg_value = -torch.finfo(sim.dtype).max + mask = repeat(mask, 'b j -> (b h) () j', h=h) + sim.masked_fill_(~mask, max_neg_value) + + # attention, what we cannot get enough of + sim = sim.softmax(dim=-1) + + out = einsum('b i j, b j d -> b i d', sim, v) + out = rearrange(out, '(b h) n d -> b n (h d)', h=h) + return self.to_out(out) + + +class MemoryEfficientCrossAttention(nn.Module): + # https://github.com/MatthieuTPHR/diffusers/blob/d80b531ff8060ec1ea982b65a1b8df70f73aa67c/src/diffusers/models/attention.py#L223 + def __init__(self, query_dim, context_dim=None, heads=8, dim_head=64, dropout=0.0): + super().__init__() + print(f"Setting up {self.__class__.__name__}. Query dim is {query_dim}, context_dim is {context_dim} and using " + f"{heads} heads.") + inner_dim = dim_head * heads + context_dim = default(context_dim, query_dim) + + self.heads = heads + self.dim_head = dim_head + + self.to_q = nn.Linear(query_dim, inner_dim, bias=False) + self.to_k = nn.Linear(context_dim, inner_dim, bias=False) + self.to_v = nn.Linear(context_dim, inner_dim, bias=False) + + self.to_out = nn.Sequential(nn.Linear(inner_dim, query_dim), nn.Dropout(dropout)) + self.attention_op: Optional[Any] = None + + def forward(self, x, context=None, mask=None): + q = self.to_q(x) + context = default(context, x) + k = self.to_k(context) + v = self.to_v(context) + + b, _, _ = q.shape + q, k, v = map( + lambda t: t.unsqueeze(3) + .reshape(b, t.shape[1], self.heads, self.dim_head) + .permute(0, 2, 1, 3) + .reshape(b * self.heads, t.shape[1], self.dim_head) + .contiguous(), + (q, k, v), + ) + + # actually compute the attention, what we cannot get enough of + out = xformers.ops.memory_efficient_attention(q, k, v, attn_bias=None, op=self.attention_op) + + if exists(mask): + raise NotImplementedError + out = ( + out.unsqueeze(0) + .reshape(b, self.heads, out.shape[1], self.dim_head) + .permute(0, 2, 1, 3) + .reshape(b, out.shape[1], self.heads * self.dim_head) + ) + return self.to_out(out) + + +class BasicTransformerBlock(nn.Module): + ATTENTION_MODES = { + "softmax": CrossAttention, # vanilla attention + "softmax-xformers": MemoryEfficientCrossAttention + } + def __init__(self, dim, n_heads, d_head, dropout=0., context_dim=None, gated_ff=True, checkpoint=True, + disable_self_attn=False): + super().__init__() + attn_mode = "softmax-xformers" if XFORMERS_IS_AVAILBLE else "softmax" + assert attn_mode in self.ATTENTION_MODES + attn_cls = self.ATTENTION_MODES[attn_mode] + self.disable_self_attn = disable_self_attn + self.attn1 = attn_cls(query_dim=dim, heads=n_heads, dim_head=d_head, dropout=dropout, + context_dim=context_dim if self.disable_self_attn else None) # is a self-attention if not self.disable_self_attn + self.ff = FeedForward(dim, dropout=dropout, glu=gated_ff) + self.attn2 = attn_cls(query_dim=dim, context_dim=context_dim, + heads=n_heads, dim_head=d_head, dropout=dropout) # is self-attn if context is none + self.norm1 = nn.LayerNorm(dim) + self.norm2 = nn.LayerNorm(dim) + self.norm3 = nn.LayerNorm(dim) + self.checkpoint = checkpoint + + def forward(self, x, context=None): + return checkpoint(self._forward, (x, context), self.parameters(), self.checkpoint) + + def _forward(self, x, context=None): + x = self.attn1(self.norm1(x), context=context if self.disable_self_attn else None) + x + x = self.attn2(self.norm2(x), context=context) + x + x = self.ff(self.norm3(x)) + x + return x + + +class SpatialTransformer(nn.Module): + """ + Transformer block for image-like data. + First, project the input (aka embedding) + and reshape to b, t, d. + Then apply standard transformer action. + Finally, reshape to image + NEW: use_linear for more efficiency instead of the 1x1 convs + """ + def __init__(self, in_channels, n_heads, d_head, + depth=1, dropout=0., context_dim=None, + disable_self_attn=False, use_linear=False, + use_checkpoint=True): + super().__init__() + if exists(context_dim) and not isinstance(context_dim, list): + context_dim = [context_dim] + self.in_channels = in_channels + inner_dim = n_heads * d_head + self.norm = Normalize(in_channels) + if not use_linear: + self.proj_in = nn.Conv2d(in_channels, + inner_dim, + kernel_size=1, + stride=1, + padding=0) + else: + self.proj_in = nn.Linear(in_channels, inner_dim) + + self.transformer_blocks = nn.ModuleList( + [BasicTransformerBlock(inner_dim, n_heads, d_head, dropout=dropout, context_dim=context_dim[d], + disable_self_attn=disable_self_attn, checkpoint=use_checkpoint) + for d in range(depth)] + ) + if not use_linear: + self.proj_out = zero_module(nn.Conv2d(inner_dim, + in_channels, + kernel_size=1, + stride=1, + padding=0)) + else: + self.proj_out = zero_module(nn.Linear(in_channels, inner_dim)) + self.use_linear = use_linear + + def forward(self, x, context=None): + # note: if no context is given, cross-attention defaults to self-attention + if not isinstance(context, list): + context = [context] + b, c, h, w = x.shape + x_in = x + x = self.norm(x) + if not self.use_linear: + x = self.proj_in(x) + x = rearrange(x, 'b c h w -> b (h w) c').contiguous() + if self.use_linear: + x = self.proj_in(x) + for i, block in enumerate(self.transformer_blocks): + x = block(x, context=context[i]) + if self.use_linear: + x = self.proj_out(x) + x = rearrange(x, 'b (h w) c -> b c h w', h=h, w=w).contiguous() + if not self.use_linear: + x = self.proj_out(x) + return x + x_in + diff --git a/ldm/modules/diffusionmodules/__init__.py b/ldm/modules/diffusionmodules/__init__.py new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/ldm/modules/diffusionmodules/model.py b/ldm/modules/diffusionmodules/model.py new file mode 100644 index 0000000000000000000000000000000000000000..b089eebbe1676d8249005bb9def002ff5180715b --- /dev/null +++ b/ldm/modules/diffusionmodules/model.py @@ -0,0 +1,852 @@ +# pytorch_diffusion + derived encoder decoder +import math +import torch +import torch.nn as nn +import numpy as np +from einops import rearrange +from typing import Optional, Any + +from ldm.modules.attention import MemoryEfficientCrossAttention + +try: + import xformers + import xformers.ops + XFORMERS_IS_AVAILBLE = True +except: + XFORMERS_IS_AVAILBLE = False + print("No module 'xformers'. Proceeding without it.") + + +def get_timestep_embedding(timesteps, embedding_dim): + """ + This matches the implementation in Denoising Diffusion Probabilistic Models: + From Fairseq. + Build sinusoidal embeddings. + This matches the implementation in tensor2tensor, but differs slightly + from the description in Section 3.5 of "Attention Is All You Need". + """ + assert len(timesteps.shape) == 1 + + half_dim = embedding_dim // 2 + emb = math.log(10000) / (half_dim - 1) + emb = torch.exp(torch.arange(half_dim, dtype=torch.float32) * -emb) + emb = emb.to(device=timesteps.device) + emb = timesteps.float()[:, None] * emb[None, :] + emb = torch.cat([torch.sin(emb), torch.cos(emb)], dim=1) + if embedding_dim % 2 == 1: # zero pad + emb = torch.nn.functional.pad(emb, (0,1,0,0)) + return emb + + +def nonlinearity(x): + # swish + return x*torch.sigmoid(x) + + +def Normalize(in_channels, num_groups=32): + return torch.nn.GroupNorm(num_groups=num_groups, num_channels=in_channels, eps=1e-6, affine=True) + + +class Upsample(nn.Module): + def __init__(self, in_channels, with_conv): + super().__init__() + self.with_conv = with_conv + if self.with_conv: + self.conv = torch.nn.Conv2d(in_channels, + in_channels, + kernel_size=3, + stride=1, + padding=1) + + def forward(self, x): + x = torch.nn.functional.interpolate(x, scale_factor=2.0, mode="nearest") + if self.with_conv: + x = self.conv(x) + return x + + +class Downsample(nn.Module): + def __init__(self, in_channels, with_conv): + super().__init__() + self.with_conv = with_conv + if self.with_conv: + # no asymmetric padding in torch conv, must do it ourselves + self.conv = torch.nn.Conv2d(in_channels, + in_channels, + kernel_size=3, + stride=2, + padding=0) + + def forward(self, x): + if self.with_conv: + pad = (0,1,0,1) + x = torch.nn.functional.pad(x, pad, mode="constant", value=0) + x = self.conv(x) + else: + x = torch.nn.functional.avg_pool2d(x, kernel_size=2, stride=2) + return x + + +class ResnetBlock(nn.Module): + def __init__(self, *, in_channels, out_channels=None, conv_shortcut=False, + dropout, temb_channels=512): + super().__init__() + self.in_channels = in_channels + out_channels = in_channels if out_channels is None else out_channels + self.out_channels = out_channels + self.use_conv_shortcut = conv_shortcut + + self.norm1 = Normalize(in_channels) + self.conv1 = torch.nn.Conv2d(in_channels, + out_channels, + kernel_size=3, + stride=1, + padding=1) + if temb_channels > 0: + self.temb_proj = torch.nn.Linear(temb_channels, + out_channels) + self.norm2 = Normalize(out_channels) + self.dropout = torch.nn.Dropout(dropout) + self.conv2 = torch.nn.Conv2d(out_channels, + out_channels, + kernel_size=3, + stride=1, + padding=1) + if self.in_channels != self.out_channels: + if self.use_conv_shortcut: + self.conv_shortcut = torch.nn.Conv2d(in_channels, + out_channels, + kernel_size=3, + stride=1, + padding=1) + else: + self.nin_shortcut = torch.nn.Conv2d(in_channels, + out_channels, + kernel_size=1, + stride=1, + padding=0) + + def forward(self, x, temb): + h = x + h = self.norm1(h) + h = nonlinearity(h) + h = self.conv1(h) + + if temb is not None: + h = h + self.temb_proj(nonlinearity(temb))[:,:,None,None] + + h = self.norm2(h) + h = nonlinearity(h) + h = self.dropout(h) + h = self.conv2(h) + + if self.in_channels != self.out_channels: + if self.use_conv_shortcut: + x = self.conv_shortcut(x) + else: + x = self.nin_shortcut(x) + + return x+h + + +class AttnBlock(nn.Module): + def __init__(self, in_channels): + super().__init__() + self.in_channels = in_channels + + self.norm = Normalize(in_channels) + self.q = torch.nn.Conv2d(in_channels, + in_channels, + kernel_size=1, + stride=1, + padding=0) + self.k = torch.nn.Conv2d(in_channels, + in_channels, + kernel_size=1, + stride=1, + padding=0) + self.v = torch.nn.Conv2d(in_channels, + in_channels, + kernel_size=1, + stride=1, + padding=0) + self.proj_out = torch.nn.Conv2d(in_channels, + in_channels, + kernel_size=1, + stride=1, + padding=0) + + def forward(self, x): + h_ = x + h_ = self.norm(h_) + q = self.q(h_) + k = self.k(h_) + v = self.v(h_) + + # compute attention + b,c,h,w = q.shape + q = q.reshape(b,c,h*w) + q = q.permute(0,2,1) # b,hw,c + k = k.reshape(b,c,h*w) # b,c,hw + w_ = torch.bmm(q,k) # b,hw,hw w[b,i,j]=sum_c q[b,i,c]k[b,c,j] + w_ = w_ * (int(c)**(-0.5)) + w_ = torch.nn.functional.softmax(w_, dim=2) + + # attend to values + v = v.reshape(b,c,h*w) + w_ = w_.permute(0,2,1) # b,hw,hw (first hw of k, second of q) + h_ = torch.bmm(v,w_) # b, c,hw (hw of q) h_[b,c,j] = sum_i v[b,c,i] w_[b,i,j] + h_ = h_.reshape(b,c,h,w) + + h_ = self.proj_out(h_) + + return x+h_ + +class MemoryEfficientAttnBlock(nn.Module): + """ + Uses xformers efficient implementation, + see https://github.com/MatthieuTPHR/diffusers/blob/d80b531ff8060ec1ea982b65a1b8df70f73aa67c/src/diffusers/models/attention.py#L223 + Note: this is a single-head self-attention operation + """ + # + def __init__(self, in_channels): + super().__init__() + self.in_channels = in_channels + + self.norm = Normalize(in_channels) + self.q = torch.nn.Conv2d(in_channels, + in_channels, + kernel_size=1, + stride=1, + padding=0) + self.k = torch.nn.Conv2d(in_channels, + in_channels, + kernel_size=1, + stride=1, + padding=0) + self.v = torch.nn.Conv2d(in_channels, + in_channels, + kernel_size=1, + stride=1, + padding=0) + self.proj_out = torch.nn.Conv2d(in_channels, + in_channels, + kernel_size=1, + stride=1, + padding=0) + self.attention_op: Optional[Any] = None + + def forward(self, x): + h_ = x + h_ = self.norm(h_) + q = self.q(h_) + k = self.k(h_) + v = self.v(h_) + + # compute attention + B, C, H, W = q.shape + q, k, v = map(lambda x: rearrange(x, 'b c h w -> b (h w) c'), (q, k, v)) + + q, k, v = map( + lambda t: t.unsqueeze(3) + .reshape(B, t.shape[1], 1, C) + .permute(0, 2, 1, 3) + .reshape(B * 1, t.shape[1], C) + .contiguous(), + (q, k, v), + ) + out = xformers.ops.memory_efficient_attention(q, k, v, attn_bias=None, op=self.attention_op) + + out = ( + out.unsqueeze(0) + .reshape(B, 1, out.shape[1], C) + .permute(0, 2, 1, 3) + .reshape(B, out.shape[1], C) + ) + out = rearrange(out, 'b (h w) c -> b c h w', b=B, h=H, w=W, c=C) + out = self.proj_out(out) + return x+out + + +class MemoryEfficientCrossAttentionWrapper(MemoryEfficientCrossAttention): + def forward(self, x, context=None, mask=None): + b, c, h, w = x.shape + x = rearrange(x, 'b c h w -> b (h w) c') + out = super().forward(x, context=context, mask=mask) + out = rearrange(out, 'b (h w) c -> b c h w', h=h, w=w, c=c) + return x + out + + +def make_attn(in_channels, attn_type="vanilla", attn_kwargs=None): + assert attn_type in ["vanilla", "vanilla-xformers", "memory-efficient-cross-attn", "linear", "none"], f'attn_type {attn_type} unknown' + if XFORMERS_IS_AVAILBLE and attn_type == "vanilla": + attn_type = "vanilla-xformers" + print(f"making attention of type '{attn_type}' with {in_channels} in_channels") + if attn_type == "vanilla": + assert attn_kwargs is None + return AttnBlock(in_channels) + elif attn_type == "vanilla-xformers": + print(f"building MemoryEfficientAttnBlock with {in_channels} in_channels...") + return MemoryEfficientAttnBlock(in_channels) + elif type == "memory-efficient-cross-attn": + attn_kwargs["query_dim"] = in_channels + return MemoryEfficientCrossAttentionWrapper(**attn_kwargs) + elif attn_type == "none": + return nn.Identity(in_channels) + else: + raise NotImplementedError() + + +class Model(nn.Module): + def __init__(self, *, ch, out_ch, ch_mult=(1,2,4,8), num_res_blocks, + attn_resolutions, dropout=0.0, resamp_with_conv=True, in_channels, + resolution, use_timestep=True, use_linear_attn=False, attn_type="vanilla"): + super().__init__() + if use_linear_attn: attn_type = "linear" + self.ch = ch + self.temb_ch = self.ch*4 + self.num_resolutions = len(ch_mult) + self.num_res_blocks = num_res_blocks + self.resolution = resolution + self.in_channels = in_channels + + self.use_timestep = use_timestep + if self.use_timestep: + # timestep embedding + self.temb = nn.Module() + self.temb.dense = nn.ModuleList([ + torch.nn.Linear(self.ch, + self.temb_ch), + torch.nn.Linear(self.temb_ch, + self.temb_ch), + ]) + + # downsampling + self.conv_in = torch.nn.Conv2d(in_channels, + self.ch, + kernel_size=3, + stride=1, + padding=1) + + curr_res = resolution + in_ch_mult = (1,)+tuple(ch_mult) + self.down = nn.ModuleList() + for i_level in range(self.num_resolutions): + block = nn.ModuleList() + attn = nn.ModuleList() + block_in = ch*in_ch_mult[i_level] + block_out = ch*ch_mult[i_level] + for i_block in range(self.num_res_blocks): + block.append(ResnetBlock(in_channels=block_in, + out_channels=block_out, + temb_channels=self.temb_ch, + dropout=dropout)) + block_in = block_out + if curr_res in attn_resolutions: + attn.append(make_attn(block_in, attn_type=attn_type)) + down = nn.Module() + down.block = block + down.attn = attn + if i_level != self.num_resolutions-1: + down.downsample = Downsample(block_in, resamp_with_conv) + curr_res = curr_res // 2 + self.down.append(down) + + # middle + self.mid = nn.Module() + self.mid.block_1 = ResnetBlock(in_channels=block_in, + out_channels=block_in, + temb_channels=self.temb_ch, + dropout=dropout) + self.mid.attn_1 = make_attn(block_in, attn_type=attn_type) + self.mid.block_2 = ResnetBlock(in_channels=block_in, + out_channels=block_in, + temb_channels=self.temb_ch, + dropout=dropout) + + # upsampling + self.up = nn.ModuleList() + for i_level in reversed(range(self.num_resolutions)): + block = nn.ModuleList() + attn = nn.ModuleList() + block_out = ch*ch_mult[i_level] + skip_in = ch*ch_mult[i_level] + for i_block in range(self.num_res_blocks+1): + if i_block == self.num_res_blocks: + skip_in = ch*in_ch_mult[i_level] + block.append(ResnetBlock(in_channels=block_in+skip_in, + out_channels=block_out, + temb_channels=self.temb_ch, + dropout=dropout)) + block_in = block_out + if curr_res in attn_resolutions: + attn.append(make_attn(block_in, attn_type=attn_type)) + up = nn.Module() + up.block = block + up.attn = attn + if i_level != 0: + up.upsample = Upsample(block_in, resamp_with_conv) + curr_res = curr_res * 2 + self.up.insert(0, up) # prepend to get consistent order + + # end + self.norm_out = Normalize(block_in) + self.conv_out = torch.nn.Conv2d(block_in, + out_ch, + kernel_size=3, + stride=1, + padding=1) + + def forward(self, x, t=None, context=None): + #assert x.shape[2] == x.shape[3] == self.resolution + if context is not None: + # assume aligned context, cat along channel axis + x = torch.cat((x, context), dim=1) + if self.use_timestep: + # timestep embedding + assert t is not None + temb = get_timestep_embedding(t, self.ch) + temb = self.temb.dense[0](temb) + temb = nonlinearity(temb) + temb = self.temb.dense[1](temb) + else: + temb = None + + # downsampling + hs = [self.conv_in(x)] + for i_level in range(self.num_resolutions): + for i_block in range(self.num_res_blocks): + h = self.down[i_level].block[i_block](hs[-1], temb) + if len(self.down[i_level].attn) > 0: + h = self.down[i_level].attn[i_block](h) + hs.append(h) + if i_level != self.num_resolutions-1: + hs.append(self.down[i_level].downsample(hs[-1])) + + # middle + h = hs[-1] + h = self.mid.block_1(h, temb) + h = self.mid.attn_1(h) + h = self.mid.block_2(h, temb) + + # upsampling + for i_level in reversed(range(self.num_resolutions)): + for i_block in range(self.num_res_blocks+1): + h = self.up[i_level].block[i_block]( + torch.cat([h, hs.pop()], dim=1), temb) + if len(self.up[i_level].attn) > 0: + h = self.up[i_level].attn[i_block](h) + if i_level != 0: + h = self.up[i_level].upsample(h) + + # end + h = self.norm_out(h) + h = nonlinearity(h) + h = self.conv_out(h) + return h + + def get_last_layer(self): + return self.conv_out.weight + + +class Encoder(nn.Module): + def __init__(self, *, ch, out_ch, ch_mult=(1,2,4,8), num_res_blocks, + attn_resolutions, dropout=0.0, resamp_with_conv=True, in_channels, + resolution, z_channels, double_z=True, use_linear_attn=False, attn_type="vanilla", + **ignore_kwargs): + super().__init__() + if use_linear_attn: attn_type = "linear" + self.ch = ch + self.temb_ch = 0 + self.num_resolutions = len(ch_mult) + self.num_res_blocks = num_res_blocks + self.resolution = resolution + self.in_channels = in_channels + + # downsampling + self.conv_in = torch.nn.Conv2d(in_channels, + self.ch, + kernel_size=3, + stride=1, + padding=1) + + curr_res = resolution + in_ch_mult = (1,)+tuple(ch_mult) + self.in_ch_mult = in_ch_mult + self.down = nn.ModuleList() + for i_level in range(self.num_resolutions): + block = nn.ModuleList() + attn = nn.ModuleList() + block_in = ch*in_ch_mult[i_level] + block_out = ch*ch_mult[i_level] + for i_block in range(self.num_res_blocks): + block.append(ResnetBlock(in_channels=block_in, + out_channels=block_out, + temb_channels=self.temb_ch, + dropout=dropout)) + block_in = block_out + if curr_res in attn_resolutions: + attn.append(make_attn(block_in, attn_type=attn_type)) + down = nn.Module() + down.block = block + down.attn = attn + if i_level != self.num_resolutions-1: + down.downsample = Downsample(block_in, resamp_with_conv) + curr_res = curr_res // 2 + self.down.append(down) + + # middle + self.mid = nn.Module() + self.mid.block_1 = ResnetBlock(in_channels=block_in, + out_channels=block_in, + temb_channels=self.temb_ch, + dropout=dropout) + self.mid.attn_1 = make_attn(block_in, attn_type=attn_type) + self.mid.block_2 = ResnetBlock(in_channels=block_in, + out_channels=block_in, + temb_channels=self.temb_ch, + dropout=dropout) + + # end + self.norm_out = Normalize(block_in) + self.conv_out = torch.nn.Conv2d(block_in, + 2*z_channels if double_z else z_channels, + kernel_size=3, + stride=1, + padding=1) + + def forward(self, x): + # timestep embedding + temb = None + + # downsampling + hs = [self.conv_in(x)] + for i_level in range(self.num_resolutions): + for i_block in range(self.num_res_blocks): + h = self.down[i_level].block[i_block](hs[-1], temb) + if len(self.down[i_level].attn) > 0: + h = self.down[i_level].attn[i_block](h) + hs.append(h) + if i_level != self.num_resolutions-1: + hs.append(self.down[i_level].downsample(hs[-1])) + + # middle + h = hs[-1] + h = self.mid.block_1(h, temb) + h = self.mid.attn_1(h) + h = self.mid.block_2(h, temb) + + # end + h = self.norm_out(h) + h = nonlinearity(h) + h = self.conv_out(h) + return h + + +class Decoder(nn.Module): + def __init__(self, *, ch, out_ch, ch_mult=(1,2,4,8), num_res_blocks, + attn_resolutions, dropout=0.0, resamp_with_conv=True, in_channels, + resolution, z_channels, give_pre_end=False, tanh_out=False, use_linear_attn=False, + attn_type="vanilla", **ignorekwargs): + super().__init__() + if use_linear_attn: attn_type = "linear" + self.ch = ch + self.temb_ch = 0 + self.num_resolutions = len(ch_mult) + self.num_res_blocks = num_res_blocks + self.resolution = resolution + self.in_channels = in_channels + self.give_pre_end = give_pre_end + self.tanh_out = tanh_out + + # compute in_ch_mult, block_in and curr_res at lowest res + in_ch_mult = (1,)+tuple(ch_mult) + block_in = ch*ch_mult[self.num_resolutions-1] + curr_res = resolution // 2**(self.num_resolutions-1) + self.z_shape = (1,z_channels,curr_res,curr_res) + print("Working with z of shape {} = {} dimensions.".format( + self.z_shape, np.prod(self.z_shape))) + + # z to block_in + self.conv_in = torch.nn.Conv2d(z_channels, + block_in, + kernel_size=3, + stride=1, + padding=1) + + # middle + self.mid = nn.Module() + self.mid.block_1 = ResnetBlock(in_channels=block_in, + out_channels=block_in, + temb_channels=self.temb_ch, + dropout=dropout) + self.mid.attn_1 = make_attn(block_in, attn_type=attn_type) + self.mid.block_2 = ResnetBlock(in_channels=block_in, + out_channels=block_in, + temb_channels=self.temb_ch, + dropout=dropout) + + # upsampling + self.up = nn.ModuleList() + for i_level in reversed(range(self.num_resolutions)): + block = nn.ModuleList() + attn = nn.ModuleList() + block_out = ch*ch_mult[i_level] + for i_block in range(self.num_res_blocks+1): + block.append(ResnetBlock(in_channels=block_in, + out_channels=block_out, + temb_channels=self.temb_ch, + dropout=dropout)) + block_in = block_out + if curr_res in attn_resolutions: + attn.append(make_attn(block_in, attn_type=attn_type)) + up = nn.Module() + up.block = block + up.attn = attn + if i_level != 0: + up.upsample = Upsample(block_in, resamp_with_conv) + curr_res = curr_res * 2 + self.up.insert(0, up) # prepend to get consistent order + + # end + self.norm_out = Normalize(block_in) + self.conv_out = torch.nn.Conv2d(block_in, + out_ch, + kernel_size=3, + stride=1, + padding=1) + + def forward(self, z): + #assert z.shape[1:] == self.z_shape[1:] + self.last_z_shape = z.shape + + # timestep embedding + temb = None + + # z to block_in + h = self.conv_in(z) + + # middle + h = self.mid.block_1(h, temb) + h = self.mid.attn_1(h) + h = self.mid.block_2(h, temb) + + # upsampling + for i_level in reversed(range(self.num_resolutions)): + for i_block in range(self.num_res_blocks+1): + h = self.up[i_level].block[i_block](h, temb) + if len(self.up[i_level].attn) > 0: + h = self.up[i_level].attn[i_block](h) + if i_level != 0: + h = self.up[i_level].upsample(h) + + # end + if self.give_pre_end: + return h + + h = self.norm_out(h) + h = nonlinearity(h) + h = self.conv_out(h) + if self.tanh_out: + h = torch.tanh(h) + return h + + +class SimpleDecoder(nn.Module): + def __init__(self, in_channels, out_channels, *args, **kwargs): + super().__init__() + self.model = nn.ModuleList([nn.Conv2d(in_channels, in_channels, 1), + ResnetBlock(in_channels=in_channels, + out_channels=2 * in_channels, + temb_channels=0, dropout=0.0), + ResnetBlock(in_channels=2 * in_channels, + out_channels=4 * in_channels, + temb_channels=0, dropout=0.0), + ResnetBlock(in_channels=4 * in_channels, + out_channels=2 * in_channels, + temb_channels=0, dropout=0.0), + nn.Conv2d(2*in_channels, in_channels, 1), + Upsample(in_channels, with_conv=True)]) + # end + self.norm_out = Normalize(in_channels) + self.conv_out = torch.nn.Conv2d(in_channels, + out_channels, + kernel_size=3, + stride=1, + padding=1) + + def forward(self, x): + for i, layer in enumerate(self.model): + if i in [1,2,3]: + x = layer(x, None) + else: + x = layer(x) + + h = self.norm_out(x) + h = nonlinearity(h) + x = self.conv_out(h) + return x + + +class UpsampleDecoder(nn.Module): + def __init__(self, in_channels, out_channels, ch, num_res_blocks, resolution, + ch_mult=(2,2), dropout=0.0): + super().__init__() + # upsampling + self.temb_ch = 0 + self.num_resolutions = len(ch_mult) + self.num_res_blocks = num_res_blocks + block_in = in_channels + curr_res = resolution // 2 ** (self.num_resolutions - 1) + self.res_blocks = nn.ModuleList() + self.upsample_blocks = nn.ModuleList() + for i_level in range(self.num_resolutions): + res_block = [] + block_out = ch * ch_mult[i_level] + for i_block in range(self.num_res_blocks + 1): + res_block.append(ResnetBlock(in_channels=block_in, + out_channels=block_out, + temb_channels=self.temb_ch, + dropout=dropout)) + block_in = block_out + self.res_blocks.append(nn.ModuleList(res_block)) + if i_level != self.num_resolutions - 1: + self.upsample_blocks.append(Upsample(block_in, True)) + curr_res = curr_res * 2 + + # end + self.norm_out = Normalize(block_in) + self.conv_out = torch.nn.Conv2d(block_in, + out_channels, + kernel_size=3, + stride=1, + padding=1) + + def forward(self, x): + # upsampling + h = x + for k, i_level in enumerate(range(self.num_resolutions)): + for i_block in range(self.num_res_blocks + 1): + h = self.res_blocks[i_level][i_block](h, None) + if i_level != self.num_resolutions - 1: + h = self.upsample_blocks[k](h) + h = self.norm_out(h) + h = nonlinearity(h) + h = self.conv_out(h) + return h + + +class LatentRescaler(nn.Module): + def __init__(self, factor, in_channels, mid_channels, out_channels, depth=2): + super().__init__() + # residual block, interpolate, residual block + self.factor = factor + self.conv_in = nn.Conv2d(in_channels, + mid_channels, + kernel_size=3, + stride=1, + padding=1) + self.res_block1 = nn.ModuleList([ResnetBlock(in_channels=mid_channels, + out_channels=mid_channels, + temb_channels=0, + dropout=0.0) for _ in range(depth)]) + self.attn = AttnBlock(mid_channels) + self.res_block2 = nn.ModuleList([ResnetBlock(in_channels=mid_channels, + out_channels=mid_channels, + temb_channels=0, + dropout=0.0) for _ in range(depth)]) + + self.conv_out = nn.Conv2d(mid_channels, + out_channels, + kernel_size=1, + ) + + def forward(self, x): + x = self.conv_in(x) + for block in self.res_block1: + x = block(x, None) + x = torch.nn.functional.interpolate(x, size=(int(round(x.shape[2]*self.factor)), int(round(x.shape[3]*self.factor)))) + x = self.attn(x) + for block in self.res_block2: + x = block(x, None) + x = self.conv_out(x) + return x + + +class MergedRescaleEncoder(nn.Module): + def __init__(self, in_channels, ch, resolution, out_ch, num_res_blocks, + attn_resolutions, dropout=0.0, resamp_with_conv=True, + ch_mult=(1,2,4,8), rescale_factor=1.0, rescale_module_depth=1): + super().__init__() + intermediate_chn = ch * ch_mult[-1] + self.encoder = Encoder(in_channels=in_channels, num_res_blocks=num_res_blocks, ch=ch, ch_mult=ch_mult, + z_channels=intermediate_chn, double_z=False, resolution=resolution, + attn_resolutions=attn_resolutions, dropout=dropout, resamp_with_conv=resamp_with_conv, + out_ch=None) + self.rescaler = LatentRescaler(factor=rescale_factor, in_channels=intermediate_chn, + mid_channels=intermediate_chn, out_channels=out_ch, depth=rescale_module_depth) + + def forward(self, x): + x = self.encoder(x) + x = self.rescaler(x) + return x + + +class MergedRescaleDecoder(nn.Module): + def __init__(self, z_channels, out_ch, resolution, num_res_blocks, attn_resolutions, ch, ch_mult=(1,2,4,8), + dropout=0.0, resamp_with_conv=True, rescale_factor=1.0, rescale_module_depth=1): + super().__init__() + tmp_chn = z_channels*ch_mult[-1] + self.decoder = Decoder(out_ch=out_ch, z_channels=tmp_chn, attn_resolutions=attn_resolutions, dropout=dropout, + resamp_with_conv=resamp_with_conv, in_channels=None, num_res_blocks=num_res_blocks, + ch_mult=ch_mult, resolution=resolution, ch=ch) + self.rescaler = LatentRescaler(factor=rescale_factor, in_channels=z_channels, mid_channels=tmp_chn, + out_channels=tmp_chn, depth=rescale_module_depth) + + def forward(self, x): + x = self.rescaler(x) + x = self.decoder(x) + return x + + +class Upsampler(nn.Module): + def __init__(self, in_size, out_size, in_channels, out_channels, ch_mult=2): + super().__init__() + assert out_size >= in_size + num_blocks = int(np.log2(out_size//in_size))+1 + factor_up = 1.+ (out_size % in_size) + print(f"Building {self.__class__.__name__} with in_size: {in_size} --> out_size {out_size} and factor {factor_up}") + self.rescaler = LatentRescaler(factor=factor_up, in_channels=in_channels, mid_channels=2*in_channels, + out_channels=in_channels) + self.decoder = Decoder(out_ch=out_channels, resolution=out_size, z_channels=in_channels, num_res_blocks=2, + attn_resolutions=[], in_channels=None, ch=in_channels, + ch_mult=[ch_mult for _ in range(num_blocks)]) + + def forward(self, x): + x = self.rescaler(x) + x = self.decoder(x) + return x + + +class Resize(nn.Module): + def __init__(self, in_channels=None, learned=False, mode="bilinear"): + super().__init__() + self.with_conv = learned + self.mode = mode + if self.with_conv: + print(f"Note: {self.__class__.__name} uses learned downsampling and will ignore the fixed {mode} mode") + raise NotImplementedError() + assert in_channels is not None + # no asymmetric padding in torch conv, must do it ourselves + self.conv = torch.nn.Conv2d(in_channels, + in_channels, + kernel_size=4, + stride=2, + padding=1) + + def forward(self, x, scale_factor=1.0): + if scale_factor==1.0: + return x + else: + x = torch.nn.functional.interpolate(x, mode=self.mode, align_corners=False, scale_factor=scale_factor) + return x diff --git a/ldm/modules/diffusionmodules/openaimodel.py b/ldm/modules/diffusionmodules/openaimodel.py new file mode 100644 index 0000000000000000000000000000000000000000..7df6b5abfe8eff07f0c8e8703ba8aee90d45984b --- /dev/null +++ b/ldm/modules/diffusionmodules/openaimodel.py @@ -0,0 +1,786 @@ +from abc import abstractmethod +import math + +import numpy as np +import torch as th +import torch.nn as nn +import torch.nn.functional as F + +from ldm.modules.diffusionmodules.util import ( + checkpoint, + conv_nd, + linear, + avg_pool_nd, + zero_module, + normalization, + timestep_embedding, +) +from ldm.modules.attention import SpatialTransformer +from ldm.util import exists + + +# dummy replace +def convert_module_to_f16(x): + pass + +def convert_module_to_f32(x): + pass + + +## go +class AttentionPool2d(nn.Module): + """ + Adapted from CLIP: https://github.com/openai/CLIP/blob/main/clip/model.py + """ + + def __init__( + self, + spacial_dim: int, + embed_dim: int, + num_heads_channels: int, + output_dim: int = None, + ): + super().__init__() + self.positional_embedding = nn.Parameter(th.randn(embed_dim, spacial_dim ** 2 + 1) / embed_dim ** 0.5) + self.qkv_proj = conv_nd(1, embed_dim, 3 * embed_dim, 1) + self.c_proj = conv_nd(1, embed_dim, output_dim or embed_dim, 1) + self.num_heads = embed_dim // num_heads_channels + self.attention = QKVAttention(self.num_heads) + + def forward(self, x): + b, c, *_spatial = x.shape + x = x.reshape(b, c, -1) # NC(HW) + x = th.cat([x.mean(dim=-1, keepdim=True), x], dim=-1) # NC(HW+1) + x = x + self.positional_embedding[None, :, :].to(x.dtype) # NC(HW+1) + x = self.qkv_proj(x) + x = self.attention(x) + x = self.c_proj(x) + return x[:, :, 0] + + +class TimestepBlock(nn.Module): + """ + Any module where forward() takes timestep embeddings as a second argument. + """ + + @abstractmethod + def forward(self, x, emb): + """ + Apply the module to `x` given `emb` timestep embeddings. + """ + + +class TimestepEmbedSequential(nn.Sequential, TimestepBlock): + """ + A sequential module that passes timestep embeddings to the children that + support it as an extra input. + """ + + def forward(self, x, emb, context=None): + for layer in self: + if isinstance(layer, TimestepBlock): + x = layer(x, emb) + elif isinstance(layer, SpatialTransformer): + x = layer(x, context) + else: + x = layer(x) + return x + + +class Upsample(nn.Module): + """ + An upsampling layer with an optional convolution. + :param channels: channels in the inputs and outputs. + :param use_conv: a bool determining if a convolution is applied. + :param dims: determines if the signal is 1D, 2D, or 3D. If 3D, then + upsampling occurs in the inner-two dimensions. + """ + + def __init__(self, channels, use_conv, dims=2, out_channels=None, padding=1): + super().__init__() + self.channels = channels + self.out_channels = out_channels or channels + self.use_conv = use_conv + self.dims = dims + if use_conv: + self.conv = conv_nd(dims, self.channels, self.out_channels, 3, padding=padding) + + def forward(self, x): + assert x.shape[1] == self.channels + if self.dims == 3: + x = F.interpolate( + x, (x.shape[2], x.shape[3] * 2, x.shape[4] * 2), mode="nearest" + ) + else: + x = F.interpolate(x, scale_factor=2, mode="nearest") + if self.use_conv: + x = self.conv(x) + return x + +class TransposedUpsample(nn.Module): + 'Learned 2x upsampling without padding' + def __init__(self, channels, out_channels=None, ks=5): + super().__init__() + self.channels = channels + self.out_channels = out_channels or channels + + self.up = nn.ConvTranspose2d(self.channels,self.out_channels,kernel_size=ks,stride=2) + + def forward(self,x): + return self.up(x) + + +class Downsample(nn.Module): + """ + A downsampling layer with an optional convolution. + :param channels: channels in the inputs and outputs. + :param use_conv: a bool determining if a convolution is applied. + :param dims: determines if the signal is 1D, 2D, or 3D. If 3D, then + downsampling occurs in the inner-two dimensions. + """ + + def __init__(self, channels, use_conv, dims=2, out_channels=None,padding=1): + super().__init__() + self.channels = channels + self.out_channels = out_channels or channels + self.use_conv = use_conv + self.dims = dims + stride = 2 if dims != 3 else (1, 2, 2) + if use_conv: + self.op = conv_nd( + dims, self.channels, self.out_channels, 3, stride=stride, padding=padding + ) + else: + assert self.channels == self.out_channels + self.op = avg_pool_nd(dims, kernel_size=stride, stride=stride) + + def forward(self, x): + assert x.shape[1] == self.channels + return self.op(x) + + +class ResBlock(TimestepBlock): + """ + A residual block that can optionally change the number of channels. + :param channels: the number of input channels. + :param emb_channels: the number of timestep embedding channels. + :param dropout: the rate of dropout. + :param out_channels: if specified, the number of out channels. + :param use_conv: if True and out_channels is specified, use a spatial + convolution instead of a smaller 1x1 convolution to change the + channels in the skip connection. + :param dims: determines if the signal is 1D, 2D, or 3D. + :param use_checkpoint: if True, use gradient checkpointing on this module. + :param up: if True, use this block for upsampling. + :param down: if True, use this block for downsampling. + """ + + def __init__( + self, + channels, + emb_channels, + dropout, + out_channels=None, + use_conv=False, + use_scale_shift_norm=False, + dims=2, + use_checkpoint=False, + up=False, + down=False, + ): + super().__init__() + self.channels = channels + self.emb_channels = emb_channels + self.dropout = dropout + self.out_channels = out_channels or channels + self.use_conv = use_conv + self.use_checkpoint = use_checkpoint + self.use_scale_shift_norm = use_scale_shift_norm + + self.in_layers = nn.Sequential( + normalization(channels), + nn.SiLU(), + conv_nd(dims, channels, self.out_channels, 3, padding=1), + ) + + self.updown = up or down + + if up: + self.h_upd = Upsample(channels, False, dims) + self.x_upd = Upsample(channels, False, dims) + elif down: + self.h_upd = Downsample(channels, False, dims) + self.x_upd = Downsample(channels, False, dims) + else: + self.h_upd = self.x_upd = nn.Identity() + + self.emb_layers = nn.Sequential( + nn.SiLU(), + linear( + emb_channels, + 2 * self.out_channels if use_scale_shift_norm else self.out_channels, + ), + ) + self.out_layers = nn.Sequential( + normalization(self.out_channels), + nn.SiLU(), + nn.Dropout(p=dropout), + zero_module( + conv_nd(dims, self.out_channels, self.out_channels, 3, padding=1) + ), + ) + + if self.out_channels == channels: + self.skip_connection = nn.Identity() + elif use_conv: + self.skip_connection = conv_nd( + dims, channels, self.out_channels, 3, padding=1 + ) + else: + self.skip_connection = conv_nd(dims, channels, self.out_channels, 1) + + def forward(self, x, emb): + """ + Apply the block to a Tensor, conditioned on a timestep embedding. + :param x: an [N x C x ...] Tensor of features. + :param emb: an [N x emb_channels] Tensor of timestep embeddings. + :return: an [N x C x ...] Tensor of outputs. + """ + return checkpoint( + self._forward, (x, emb), self.parameters(), self.use_checkpoint + ) + + + def _forward(self, x, emb): + if self.updown: + in_rest, in_conv = self.in_layers[:-1], self.in_layers[-1] + h = in_rest(x) + h = self.h_upd(h) + x = self.x_upd(x) + h = in_conv(h) + else: + h = self.in_layers(x) + emb_out = self.emb_layers(emb).type(h.dtype) + while len(emb_out.shape) < len(h.shape): + emb_out = emb_out[..., None] + if self.use_scale_shift_norm: + out_norm, out_rest = self.out_layers[0], self.out_layers[1:] + scale, shift = th.chunk(emb_out, 2, dim=1) + h = out_norm(h) * (1 + scale) + shift + h = out_rest(h) + else: + h = h + emb_out + h = self.out_layers(h) + return self.skip_connection(x) + h + + +class AttentionBlock(nn.Module): + """ + An attention block that allows spatial positions to attend to each other. + Originally ported from here, but adapted to the N-d case. + https://github.com/hojonathanho/diffusion/blob/1e0dceb3b3495bbe19116a5e1b3596cd0706c543/diffusion_tf/models/unet.py#L66. + """ + + def __init__( + self, + channels, + num_heads=1, + num_head_channels=-1, + use_checkpoint=False, + use_new_attention_order=False, + ): + super().__init__() + self.channels = channels + if num_head_channels == -1: + self.num_heads = num_heads + else: + assert ( + channels % num_head_channels == 0 + ), f"q,k,v channels {channels} is not divisible by num_head_channels {num_head_channels}" + self.num_heads = channels // num_head_channels + self.use_checkpoint = use_checkpoint + self.norm = normalization(channels) + self.qkv = conv_nd(1, channels, channels * 3, 1) + if use_new_attention_order: + # split qkv before split heads + self.attention = QKVAttention(self.num_heads) + else: + # split heads before split qkv + self.attention = QKVAttentionLegacy(self.num_heads) + + self.proj_out = zero_module(conv_nd(1, channels, channels, 1)) + + def forward(self, x): + return checkpoint(self._forward, (x,), self.parameters(), True) # TODO: check checkpoint usage, is True # TODO: fix the .half call!!! + #return pt_checkpoint(self._forward, x) # pytorch + + def _forward(self, x): + b, c, *spatial = x.shape + x = x.reshape(b, c, -1) + qkv = self.qkv(self.norm(x)) + h = self.attention(qkv) + h = self.proj_out(h) + return (x + h).reshape(b, c, *spatial) + + +def count_flops_attn(model, _x, y): + """ + A counter for the `thop` package to count the operations in an + attention operation. + Meant to be used like: + macs, params = thop.profile( + model, + inputs=(inputs, timestamps), + custom_ops={QKVAttention: QKVAttention.count_flops}, + ) + """ + b, c, *spatial = y[0].shape + num_spatial = int(np.prod(spatial)) + # We perform two matmuls with the same number of ops. + # The first computes the weight matrix, the second computes + # the combination of the value vectors. + matmul_ops = 2 * b * (num_spatial ** 2) * c + model.total_ops += th.DoubleTensor([matmul_ops]) + + +class QKVAttentionLegacy(nn.Module): + """ + A module which performs QKV attention. Matches legacy QKVAttention + input/ouput heads shaping + """ + + def __init__(self, n_heads): + super().__init__() + self.n_heads = n_heads + + def forward(self, qkv): + """ + Apply QKV attention. + :param qkv: an [N x (H * 3 * C) x T] tensor of Qs, Ks, and Vs. + :return: an [N x (H * C) x T] tensor after attention. + """ + bs, width, length = qkv.shape + assert width % (3 * self.n_heads) == 0 + ch = width // (3 * self.n_heads) + q, k, v = qkv.reshape(bs * self.n_heads, ch * 3, length).split(ch, dim=1) + scale = 1 / math.sqrt(math.sqrt(ch)) + weight = th.einsum( + "bct,bcs->bts", q * scale, k * scale + ) # More stable with f16 than dividing afterwards + weight = th.softmax(weight.float(), dim=-1).type(weight.dtype) + a = th.einsum("bts,bcs->bct", weight, v) + return a.reshape(bs, -1, length) + + @staticmethod + def count_flops(model, _x, y): + return count_flops_attn(model, _x, y) + + +class QKVAttention(nn.Module): + """ + A module which performs QKV attention and splits in a different order. + """ + + def __init__(self, n_heads): + super().__init__() + self.n_heads = n_heads + + def forward(self, qkv): + """ + Apply QKV attention. + :param qkv: an [N x (3 * H * C) x T] tensor of Qs, Ks, and Vs. + :return: an [N x (H * C) x T] tensor after attention. + """ + bs, width, length = qkv.shape + assert width % (3 * self.n_heads) == 0 + ch = width // (3 * self.n_heads) + q, k, v = qkv.chunk(3, dim=1) + scale = 1 / math.sqrt(math.sqrt(ch)) + weight = th.einsum( + "bct,bcs->bts", + (q * scale).view(bs * self.n_heads, ch, length), + (k * scale).view(bs * self.n_heads, ch, length), + ) # More stable with f16 than dividing afterwards + weight = th.softmax(weight.float(), dim=-1).type(weight.dtype) + a = th.einsum("bts,bcs->bct", weight, v.reshape(bs * self.n_heads, ch, length)) + return a.reshape(bs, -1, length) + + @staticmethod + def count_flops(model, _x, y): + return count_flops_attn(model, _x, y) + + +class UNetModel(nn.Module): + """ + The full UNet model with attention and timestep embedding. + :param in_channels: channels in the input Tensor. + :param model_channels: base channel count for the model. + :param out_channels: channels in the output Tensor. + :param num_res_blocks: number of residual blocks per downsample. + :param attention_resolutions: a collection of downsample rates at which + attention will take place. May be a set, list, or tuple. + For example, if this contains 4, then at 4x downsampling, attention + will be used. + :param dropout: the dropout probability. + :param channel_mult: channel multiplier for each level of the UNet. + :param conv_resample: if True, use learned convolutions for upsampling and + downsampling. + :param dims: determines if the signal is 1D, 2D, or 3D. + :param num_classes: if specified (as an int), then this model will be + class-conditional with `num_classes` classes. + :param use_checkpoint: use gradient checkpointing to reduce memory usage. + :param num_heads: the number of attention heads in each attention layer. + :param num_heads_channels: if specified, ignore num_heads and instead use + a fixed channel width per attention head. + :param num_heads_upsample: works with num_heads to set a different number + of heads for upsampling. Deprecated. + :param use_scale_shift_norm: use a FiLM-like conditioning mechanism. + :param resblock_updown: use residual blocks for up/downsampling. + :param use_new_attention_order: use a different attention pattern for potentially + increased efficiency. + """ + + def __init__( + self, + image_size, + in_channels, + model_channels, + out_channels, + num_res_blocks, + attention_resolutions, + dropout=0, + channel_mult=(1, 2, 4, 8), + conv_resample=True, + dims=2, + num_classes=None, + use_checkpoint=False, + use_fp16=False, + num_heads=-1, + num_head_channels=-1, + num_heads_upsample=-1, + use_scale_shift_norm=False, + resblock_updown=False, + use_new_attention_order=False, + use_spatial_transformer=False, # custom transformer support + transformer_depth=1, # custom transformer support + context_dim=None, # custom transformer support + n_embed=None, # custom support for prediction of discrete ids into codebook of first stage vq model + legacy=True, + disable_self_attentions=None, + num_attention_blocks=None, + disable_middle_self_attn=False, + use_linear_in_transformer=False, + ): + super().__init__() + if use_spatial_transformer: + assert context_dim is not None, 'Fool!! You forgot to include the dimension of your cross-attention conditioning...' + + if context_dim is not None: + assert use_spatial_transformer, 'Fool!! You forgot to use the spatial transformer for your cross-attention conditioning...' + from omegaconf.listconfig import ListConfig + if type(context_dim) == ListConfig: + context_dim = list(context_dim) + + if num_heads_upsample == -1: + num_heads_upsample = num_heads + + if num_heads == -1: + assert num_head_channels != -1, 'Either num_heads or num_head_channels has to be set' + + if num_head_channels == -1: + assert num_heads != -1, 'Either num_heads or num_head_channels has to be set' + + self.image_size = image_size + self.in_channels = in_channels + self.model_channels = model_channels + self.out_channels = out_channels + if isinstance(num_res_blocks, int): + self.num_res_blocks = len(channel_mult) * [num_res_blocks] + else: + if len(num_res_blocks) != len(channel_mult): + raise ValueError("provide num_res_blocks either as an int (globally constant) or " + "as a list/tuple (per-level) with the same length as channel_mult") + self.num_res_blocks = num_res_blocks + if disable_self_attentions is not None: + # should be a list of booleans, indicating whether to disable self-attention in TransformerBlocks or not + assert len(disable_self_attentions) == len(channel_mult) + if num_attention_blocks is not None: + assert len(num_attention_blocks) == len(self.num_res_blocks) + assert all(map(lambda i: self.num_res_blocks[i] >= num_attention_blocks[i], range(len(num_attention_blocks)))) + print(f"Constructor of UNetModel received num_attention_blocks={num_attention_blocks}. " + f"This option has LESS priority than attention_resolutions {attention_resolutions}, " + f"i.e., in cases where num_attention_blocks[i] > 0 but 2**i not in attention_resolutions, " + f"attention will still not be set.") + + self.attention_resolutions = attention_resolutions + self.dropout = dropout + self.channel_mult = channel_mult + self.conv_resample = conv_resample + self.num_classes = num_classes + self.use_checkpoint = use_checkpoint + self.dtype = th.float16 if use_fp16 else th.float32 + self.num_heads = num_heads + self.num_head_channels = num_head_channels + self.num_heads_upsample = num_heads_upsample + self.predict_codebook_ids = n_embed is not None + + time_embed_dim = model_channels * 4 + self.time_embed = nn.Sequential( + linear(model_channels, time_embed_dim), + nn.SiLU(), + linear(time_embed_dim, time_embed_dim), + ) + + if self.num_classes is not None: + if isinstance(self.num_classes, int): + self.label_emb = nn.Embedding(num_classes, time_embed_dim) + elif self.num_classes == "continuous": + print("setting up linear c_adm embedding layer") + self.label_emb = nn.Linear(1, time_embed_dim) + else: + raise ValueError() + + self.input_blocks = nn.ModuleList( + [ + TimestepEmbedSequential( + conv_nd(dims, in_channels, model_channels, 3, padding=1) + ) + ] + ) + self._feature_size = model_channels + input_block_chans = [model_channels] + ch = model_channels + ds = 1 + for level, mult in enumerate(channel_mult): + for nr in range(self.num_res_blocks[level]): + layers = [ + ResBlock( + ch, + time_embed_dim, + dropout, + out_channels=mult * model_channels, + dims=dims, + use_checkpoint=use_checkpoint, + use_scale_shift_norm=use_scale_shift_norm, + ) + ] + ch = mult * model_channels + if ds in attention_resolutions: + if num_head_channels == -1: + dim_head = ch // num_heads + else: + num_heads = ch // num_head_channels + dim_head = num_head_channels + if legacy: + #num_heads = 1 + dim_head = ch // num_heads if use_spatial_transformer else num_head_channels + if exists(disable_self_attentions): + disabled_sa = disable_self_attentions[level] + else: + disabled_sa = False + + if not exists(num_attention_blocks) or nr < num_attention_blocks[level]: + layers.append( + AttentionBlock( + ch, + use_checkpoint=use_checkpoint, + num_heads=num_heads, + num_head_channels=dim_head, + use_new_attention_order=use_new_attention_order, + ) if not use_spatial_transformer else SpatialTransformer( + ch, num_heads, dim_head, depth=transformer_depth, context_dim=context_dim, + disable_self_attn=disabled_sa, use_linear=use_linear_in_transformer, + use_checkpoint=use_checkpoint + ) + ) + self.input_blocks.append(TimestepEmbedSequential(*layers)) + self._feature_size += ch + input_block_chans.append(ch) + if level != len(channel_mult) - 1: + out_ch = ch + self.input_blocks.append( + TimestepEmbedSequential( + ResBlock( + ch, + time_embed_dim, + dropout, + out_channels=out_ch, + dims=dims, + use_checkpoint=use_checkpoint, + use_scale_shift_norm=use_scale_shift_norm, + down=True, + ) + if resblock_updown + else Downsample( + ch, conv_resample, dims=dims, out_channels=out_ch + ) + ) + ) + ch = out_ch + input_block_chans.append(ch) + ds *= 2 + self._feature_size += ch + + if num_head_channels == -1: + dim_head = ch // num_heads + else: + num_heads = ch // num_head_channels + dim_head = num_head_channels + if legacy: + #num_heads = 1 + dim_head = ch // num_heads if use_spatial_transformer else num_head_channels + self.middle_block = TimestepEmbedSequential( + ResBlock( + ch, + time_embed_dim, + dropout, + dims=dims, + use_checkpoint=use_checkpoint, + use_scale_shift_norm=use_scale_shift_norm, + ), + AttentionBlock( + ch, + use_checkpoint=use_checkpoint, + num_heads=num_heads, + num_head_channels=dim_head, + use_new_attention_order=use_new_attention_order, + ) if not use_spatial_transformer else SpatialTransformer( # always uses a self-attn + ch, num_heads, dim_head, depth=transformer_depth, context_dim=context_dim, + disable_self_attn=disable_middle_self_attn, use_linear=use_linear_in_transformer, + use_checkpoint=use_checkpoint + ), + ResBlock( + ch, + time_embed_dim, + dropout, + dims=dims, + use_checkpoint=use_checkpoint, + use_scale_shift_norm=use_scale_shift_norm, + ), + ) + self._feature_size += ch + + self.output_blocks = nn.ModuleList([]) + for level, mult in list(enumerate(channel_mult))[::-1]: + for i in range(self.num_res_blocks[level] + 1): + ich = input_block_chans.pop() + layers = [ + ResBlock( + ch + ich, + time_embed_dim, + dropout, + out_channels=model_channels * mult, + dims=dims, + use_checkpoint=use_checkpoint, + use_scale_shift_norm=use_scale_shift_norm, + ) + ] + ch = model_channels * mult + if ds in attention_resolutions: + if num_head_channels == -1: + dim_head = ch // num_heads + else: + num_heads = ch // num_head_channels + dim_head = num_head_channels + if legacy: + #num_heads = 1 + dim_head = ch // num_heads if use_spatial_transformer else num_head_channels + if exists(disable_self_attentions): + disabled_sa = disable_self_attentions[level] + else: + disabled_sa = False + + if not exists(num_attention_blocks) or i < num_attention_blocks[level]: + layers.append( + AttentionBlock( + ch, + use_checkpoint=use_checkpoint, + num_heads=num_heads_upsample, + num_head_channels=dim_head, + use_new_attention_order=use_new_attention_order, + ) if not use_spatial_transformer else SpatialTransformer( + ch, num_heads, dim_head, depth=transformer_depth, context_dim=context_dim, + disable_self_attn=disabled_sa, use_linear=use_linear_in_transformer, + use_checkpoint=use_checkpoint + ) + ) + if level and i == self.num_res_blocks[level]: + out_ch = ch + layers.append( + ResBlock( + ch, + time_embed_dim, + dropout, + out_channels=out_ch, + dims=dims, + use_checkpoint=use_checkpoint, + use_scale_shift_norm=use_scale_shift_norm, + up=True, + ) + if resblock_updown + else Upsample(ch, conv_resample, dims=dims, out_channels=out_ch) + ) + ds //= 2 + self.output_blocks.append(TimestepEmbedSequential(*layers)) + self._feature_size += ch + + self.out = nn.Sequential( + normalization(ch), + nn.SiLU(), + zero_module(conv_nd(dims, model_channels, out_channels, 3, padding=1)), + ) + if self.predict_codebook_ids: + self.id_predictor = nn.Sequential( + normalization(ch), + conv_nd(dims, model_channels, n_embed, 1), + #nn.LogSoftmax(dim=1) # change to cross_entropy and produce non-normalized logits + ) + + def convert_to_fp16(self): + """ + Convert the torso of the model to float16. + """ + self.input_blocks.apply(convert_module_to_f16) + self.middle_block.apply(convert_module_to_f16) + self.output_blocks.apply(convert_module_to_f16) + + def convert_to_fp32(self): + """ + Convert the torso of the model to float32. + """ + self.input_blocks.apply(convert_module_to_f32) + self.middle_block.apply(convert_module_to_f32) + self.output_blocks.apply(convert_module_to_f32) + + def forward(self, x, timesteps=None, context=None, y=None,**kwargs): + """ + Apply the model to an input batch. + :param x: an [N x C x ...] Tensor of inputs. + :param timesteps: a 1-D batch of timesteps. + :param context: conditioning plugged in via crossattn + :param y: an [N] Tensor of labels, if class-conditional. + :return: an [N x C x ...] Tensor of outputs. + """ + assert (y is not None) == ( + self.num_classes is not None + ), "must specify y if and only if the model is class-conditional" + hs = [] + t_emb = timestep_embedding(timesteps, self.model_channels, repeat_only=False) + emb = self.time_embed(t_emb) + + if self.num_classes is not None: + assert y.shape[0] == x.shape[0] + emb = emb + self.label_emb(y) + + h = x.type(self.dtype) + for module in self.input_blocks: + h = module(h, emb, context) + hs.append(h) + h = self.middle_block(h, emb, context) + for module in self.output_blocks: + h = th.cat([h, hs.pop()], dim=1) + h = module(h, emb, context) + h = h.type(x.dtype) + if self.predict_codebook_ids: + return self.id_predictor(h) + else: + return self.out(h) diff --git a/ldm/modules/diffusionmodules/upscaling.py b/ldm/modules/diffusionmodules/upscaling.py new file mode 100644 index 0000000000000000000000000000000000000000..03816662098ce1ffac79bd939b892e867ab91988 --- /dev/null +++ b/ldm/modules/diffusionmodules/upscaling.py @@ -0,0 +1,81 @@ +import torch +import torch.nn as nn +import numpy as np +from functools import partial + +from ldm.modules.diffusionmodules.util import extract_into_tensor, make_beta_schedule +from ldm.util import default + + +class AbstractLowScaleModel(nn.Module): + # for concatenating a downsampled image to the latent representation + def __init__(self, noise_schedule_config=None): + super(AbstractLowScaleModel, self).__init__() + if noise_schedule_config is not None: + self.register_schedule(**noise_schedule_config) + + def register_schedule(self, beta_schedule="linear", timesteps=1000, + linear_start=1e-4, linear_end=2e-2, cosine_s=8e-3): + betas = make_beta_schedule(beta_schedule, timesteps, linear_start=linear_start, linear_end=linear_end, + cosine_s=cosine_s) + alphas = 1. - betas + alphas_cumprod = np.cumprod(alphas, axis=0) + alphas_cumprod_prev = np.append(1., alphas_cumprod[:-1]) + + timesteps, = betas.shape + self.num_timesteps = int(timesteps) + self.linear_start = linear_start + self.linear_end = linear_end + assert alphas_cumprod.shape[0] == self.num_timesteps, 'alphas have to be defined for each timestep' + + to_torch = partial(torch.tensor, dtype=torch.float32) + + self.register_buffer('betas', to_torch(betas)) + self.register_buffer('alphas_cumprod', to_torch(alphas_cumprod)) + self.register_buffer('alphas_cumprod_prev', to_torch(alphas_cumprod_prev)) + + # calculations for diffusion q(x_t | x_{t-1}) and others + self.register_buffer('sqrt_alphas_cumprod', to_torch(np.sqrt(alphas_cumprod))) + self.register_buffer('sqrt_one_minus_alphas_cumprod', to_torch(np.sqrt(1. - alphas_cumprod))) + self.register_buffer('log_one_minus_alphas_cumprod', to_torch(np.log(1. - alphas_cumprod))) + self.register_buffer('sqrt_recip_alphas_cumprod', to_torch(np.sqrt(1. / alphas_cumprod))) + self.register_buffer('sqrt_recipm1_alphas_cumprod', to_torch(np.sqrt(1. / alphas_cumprod - 1))) + + def q_sample(self, x_start, t, noise=None): + noise = default(noise, lambda: torch.randn_like(x_start)) + return (extract_into_tensor(self.sqrt_alphas_cumprod, t, x_start.shape) * x_start + + extract_into_tensor(self.sqrt_one_minus_alphas_cumprod, t, x_start.shape) * noise) + + def forward(self, x): + return x, None + + def decode(self, x): + return x + + +class SimpleImageConcat(AbstractLowScaleModel): + # no noise level conditioning + def __init__(self): + super(SimpleImageConcat, self).__init__(noise_schedule_config=None) + self.max_noise_level = 0 + + def forward(self, x): + # fix to constant noise level + return x, torch.zeros(x.shape[0], device=x.device).long() + + +class ImageConcatWithNoiseAugmentation(AbstractLowScaleModel): + def __init__(self, noise_schedule_config, max_noise_level=1000, to_cuda=False): + super().__init__(noise_schedule_config=noise_schedule_config) + self.max_noise_level = max_noise_level + + def forward(self, x, noise_level=None): + if noise_level is None: + noise_level = torch.randint(0, self.max_noise_level, (x.shape[0],), device=x.device).long() + else: + assert isinstance(noise_level, torch.Tensor) + z = self.q_sample(x, noise_level) + return z, noise_level + + + diff --git a/ldm/modules/diffusionmodules/util.py b/ldm/modules/diffusionmodules/util.py new file mode 100644 index 0000000000000000000000000000000000000000..637363dfe34799e70cfdbcd11445212df9d9ca1f --- /dev/null +++ b/ldm/modules/diffusionmodules/util.py @@ -0,0 +1,270 @@ +# adopted from +# https://github.com/openai/improved-diffusion/blob/main/improved_diffusion/gaussian_diffusion.py +# and +# https://github.com/lucidrains/denoising-diffusion-pytorch/blob/7706bdfc6f527f58d33f84b7b522e61e6e3164b3/denoising_diffusion_pytorch/denoising_diffusion_pytorch.py +# and +# https://github.com/openai/guided-diffusion/blob/0ba878e517b276c45d1195eb29f6f5f72659a05b/guided_diffusion/nn.py +# +# thanks! + + +import os +import math +import torch +import torch.nn as nn +import numpy as np +from einops import repeat + +from ldm.util import instantiate_from_config + + +def make_beta_schedule(schedule, n_timestep, linear_start=1e-4, linear_end=2e-2, cosine_s=8e-3): + if schedule == "linear": + betas = ( + torch.linspace(linear_start ** 0.5, linear_end ** 0.5, n_timestep, dtype=torch.float64) ** 2 + ) + + elif schedule == "cosine": + timesteps = ( + torch.arange(n_timestep + 1, dtype=torch.float64) / n_timestep + cosine_s + ) + alphas = timesteps / (1 + cosine_s) * np.pi / 2 + alphas = torch.cos(alphas).pow(2) + alphas = alphas / alphas[0] + betas = 1 - alphas[1:] / alphas[:-1] + betas = np.clip(betas, a_min=0, a_max=0.999) + + elif schedule == "sqrt_linear": + betas = torch.linspace(linear_start, linear_end, n_timestep, dtype=torch.float64) + elif schedule == "sqrt": + betas = torch.linspace(linear_start, linear_end, n_timestep, dtype=torch.float64) ** 0.5 + else: + raise ValueError(f"schedule '{schedule}' unknown.") + return betas.numpy() + + +def make_ddim_timesteps(ddim_discr_method, num_ddim_timesteps, num_ddpm_timesteps, verbose=True): + if ddim_discr_method == 'uniform': + c = num_ddpm_timesteps // num_ddim_timesteps + ddim_timesteps = np.asarray(list(range(0, num_ddpm_timesteps, c))) + elif ddim_discr_method == 'quad': + ddim_timesteps = ((np.linspace(0, np.sqrt(num_ddpm_timesteps * .8), num_ddim_timesteps)) ** 2).astype(int) + else: + raise NotImplementedError(f'There is no ddim discretization method called "{ddim_discr_method}"') + + # assert ddim_timesteps.shape[0] == num_ddim_timesteps + # add one to get the final alpha values right (the ones from first scale to data during sampling) + steps_out = ddim_timesteps + 1 + if verbose: + print(f'Selected timesteps for ddim sampler: {steps_out}') + return steps_out + + +def make_ddim_sampling_parameters(alphacums, ddim_timesteps, eta, verbose=True): + # select alphas for computing the variance schedule + alphas = alphacums[ddim_timesteps] + alphas_prev = np.asarray([alphacums[0]] + alphacums[ddim_timesteps[:-1]].tolist()) + + # according the the formula provided in https://arxiv.org/abs/2010.02502 + sigmas = eta * np.sqrt((1 - alphas_prev) / (1 - alphas) * (1 - alphas / alphas_prev)) + if verbose: + print(f'Selected alphas for ddim sampler: a_t: {alphas}; a_(t-1): {alphas_prev}') + print(f'For the chosen value of eta, which is {eta}, ' + f'this results in the following sigma_t schedule for ddim sampler {sigmas}') + return sigmas, alphas, alphas_prev + + +def betas_for_alpha_bar(num_diffusion_timesteps, alpha_bar, max_beta=0.999): + """ + Create a beta schedule that discretizes the given alpha_t_bar function, + which defines the cumulative product of (1-beta) over time from t = [0,1]. + :param num_diffusion_timesteps: the number of betas to produce. + :param alpha_bar: a lambda that takes an argument t from 0 to 1 and + produces the cumulative product of (1-beta) up to that + part of the diffusion process. + :param max_beta: the maximum beta to use; use values lower than 1 to + prevent singularities. + """ + betas = [] + for i in range(num_diffusion_timesteps): + t1 = i / num_diffusion_timesteps + t2 = (i + 1) / num_diffusion_timesteps + betas.append(min(1 - alpha_bar(t2) / alpha_bar(t1), max_beta)) + return np.array(betas) + + +def extract_into_tensor(a, t, x_shape): + b, *_ = t.shape + out = a.gather(-1, t) + return out.reshape(b, *((1,) * (len(x_shape) - 1))) + + +def checkpoint(func, inputs, params, flag): + """ + Evaluate a function without caching intermediate activations, allowing for + reduced memory at the expense of extra compute in the backward pass. + :param func: the function to evaluate. + :param inputs: the argument sequence to pass to `func`. + :param params: a sequence of parameters `func` depends on but does not + explicitly take as arguments. + :param flag: if False, disable gradient checkpointing. + """ + if flag: + args = tuple(inputs) + tuple(params) + return CheckpointFunction.apply(func, len(inputs), *args) + else: + return func(*inputs) + + +class CheckpointFunction(torch.autograd.Function): + @staticmethod + def forward(ctx, run_function, length, *args): + ctx.run_function = run_function + ctx.input_tensors = list(args[:length]) + ctx.input_params = list(args[length:]) + ctx.gpu_autocast_kwargs = {"enabled": torch.is_autocast_enabled(), + "dtype": torch.get_autocast_gpu_dtype(), + "cache_enabled": torch.is_autocast_cache_enabled()} + with torch.no_grad(): + output_tensors = ctx.run_function(*ctx.input_tensors) + return output_tensors + + @staticmethod + def backward(ctx, *output_grads): + ctx.input_tensors = [x.detach().requires_grad_(True) for x in ctx.input_tensors] + with torch.enable_grad(), \ + torch.cuda.amp.autocast(**ctx.gpu_autocast_kwargs): + # Fixes a bug where the first op in run_function modifies the + # Tensor storage in place, which is not allowed for detach()'d + # Tensors. + shallow_copies = [x.view_as(x) for x in ctx.input_tensors] + output_tensors = ctx.run_function(*shallow_copies) + input_grads = torch.autograd.grad( + output_tensors, + ctx.input_tensors + ctx.input_params, + output_grads, + allow_unused=True, + ) + del ctx.input_tensors + del ctx.input_params + del output_tensors + return (None, None) + input_grads + + +def timestep_embedding(timesteps, dim, max_period=10000, repeat_only=False): + """ + Create sinusoidal timestep embeddings. + :param timesteps: a 1-D Tensor of N indices, one per batch element. + These may be fractional. + :param dim: the dimension of the output. + :param max_period: controls the minimum frequency of the embeddings. + :return: an [N x dim] Tensor of positional embeddings. + """ + if not repeat_only: + half = dim // 2 + freqs = torch.exp( + -math.log(max_period) * torch.arange(start=0, end=half, dtype=torch.float32) / half + ).to(device=timesteps.device) + args = timesteps[:, None].float() * freqs[None] + embedding = torch.cat([torch.cos(args), torch.sin(args)], dim=-1) + if dim % 2: + embedding = torch.cat([embedding, torch.zeros_like(embedding[:, :1])], dim=-1) + else: + embedding = repeat(timesteps, 'b -> b d', d=dim) + return embedding + + +def zero_module(module): + """ + Zero out the parameters of a module and return it. + """ + for p in module.parameters(): + p.detach().zero_() + return module + + +def scale_module(module, scale): + """ + Scale the parameters of a module and return it. + """ + for p in module.parameters(): + p.detach().mul_(scale) + return module + + +def mean_flat(tensor): + """ + Take the mean over all non-batch dimensions. + """ + return tensor.mean(dim=list(range(1, len(tensor.shape)))) + + +def normalization(channels): + """ + Make a standard normalization layer. + :param channels: number of input channels. + :return: an nn.Module for normalization. + """ + return GroupNorm32(32, channels) + + +# PyTorch 1.7 has SiLU, but we support PyTorch 1.5. +class SiLU(nn.Module): + def forward(self, x): + return x * torch.sigmoid(x) + + +class GroupNorm32(nn.GroupNorm): + def forward(self, x): + return super().forward(x.float()).type(x.dtype) + +def conv_nd(dims, *args, **kwargs): + """ + Create a 1D, 2D, or 3D convolution module. + """ + if dims == 1: + return nn.Conv1d(*args, **kwargs) + elif dims == 2: + return nn.Conv2d(*args, **kwargs) + elif dims == 3: + return nn.Conv3d(*args, **kwargs) + raise ValueError(f"unsupported dimensions: {dims}") + + +def linear(*args, **kwargs): + """ + Create a linear module. + """ + return nn.Linear(*args, **kwargs) + + +def avg_pool_nd(dims, *args, **kwargs): + """ + Create a 1D, 2D, or 3D average pooling module. + """ + if dims == 1: + return nn.AvgPool1d(*args, **kwargs) + elif dims == 2: + return nn.AvgPool2d(*args, **kwargs) + elif dims == 3: + return nn.AvgPool3d(*args, **kwargs) + raise ValueError(f"unsupported dimensions: {dims}") + + +class HybridConditioner(nn.Module): + + def __init__(self, c_concat_config, c_crossattn_config): + super().__init__() + self.concat_conditioner = instantiate_from_config(c_concat_config) + self.crossattn_conditioner = instantiate_from_config(c_crossattn_config) + + def forward(self, c_concat, c_crossattn): + c_concat = self.concat_conditioner(c_concat) + c_crossattn = self.crossattn_conditioner(c_crossattn) + return {'c_concat': [c_concat], 'c_crossattn': [c_crossattn]} + + +def noise_like(shape, device, repeat=False): + repeat_noise = lambda: torch.randn((1, *shape[1:]), device=device).repeat(shape[0], *((1,) * (len(shape) - 1))) + noise = lambda: torch.randn(shape, device=device) + return repeat_noise() if repeat else noise() \ No newline at end of file diff --git a/ldm/modules/distributions/__init__.py b/ldm/modules/distributions/__init__.py new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/ldm/modules/distributions/distributions.py b/ldm/modules/distributions/distributions.py new file mode 100644 index 0000000000000000000000000000000000000000..f2b8ef901130efc171aa69742ca0244d94d3f2e9 --- /dev/null +++ b/ldm/modules/distributions/distributions.py @@ -0,0 +1,92 @@ +import torch +import numpy as np + + +class AbstractDistribution: + def sample(self): + raise NotImplementedError() + + def mode(self): + raise NotImplementedError() + + +class DiracDistribution(AbstractDistribution): + def __init__(self, value): + self.value = value + + def sample(self): + return self.value + + def mode(self): + return self.value + + +class DiagonalGaussianDistribution(object): + def __init__(self, parameters, deterministic=False): + self.parameters = parameters + self.mean, self.logvar = torch.chunk(parameters, 2, dim=1) + self.logvar = torch.clamp(self.logvar, -30.0, 20.0) + self.deterministic = deterministic + self.std = torch.exp(0.5 * self.logvar) + self.var = torch.exp(self.logvar) + if self.deterministic: + self.var = self.std = torch.zeros_like(self.mean).to(device=self.parameters.device) + + def sample(self): + x = self.mean + self.std * torch.randn(self.mean.shape).to(device=self.parameters.device) + return x + + def kl(self, other=None): + if self.deterministic: + return torch.Tensor([0.]) + else: + if other is None: + return 0.5 * torch.sum(torch.pow(self.mean, 2) + + self.var - 1.0 - self.logvar, + dim=[1, 2, 3]) + else: + return 0.5 * torch.sum( + torch.pow(self.mean - other.mean, 2) / other.var + + self.var / other.var - 1.0 - self.logvar + other.logvar, + dim=[1, 2, 3]) + + def nll(self, sample, dims=[1,2,3]): + if self.deterministic: + return torch.Tensor([0.]) + logtwopi = np.log(2.0 * np.pi) + return 0.5 * torch.sum( + logtwopi + self.logvar + torch.pow(sample - self.mean, 2) / self.var, + dim=dims) + + def mode(self): + return self.mean + + +def normal_kl(mean1, logvar1, mean2, logvar2): + """ + source: https://github.com/openai/guided-diffusion/blob/27c20a8fab9cb472df5d6bdd6c8d11c8f430b924/guided_diffusion/losses.py#L12 + Compute the KL divergence between two gaussians. + Shapes are automatically broadcasted, so batches can be compared to + scalars, among other use cases. + """ + tensor = None + for obj in (mean1, logvar1, mean2, logvar2): + if isinstance(obj, torch.Tensor): + tensor = obj + break + assert tensor is not None, "at least one argument must be a Tensor" + + # Force variances to be Tensors. Broadcasting helps convert scalars to + # Tensors, but it does not work for torch.exp(). + logvar1, logvar2 = [ + x if isinstance(x, torch.Tensor) else torch.tensor(x).to(tensor) + for x in (logvar1, logvar2) + ] + + return 0.5 * ( + -1.0 + + logvar2 + - logvar1 + + torch.exp(logvar1 - logvar2) + + ((mean1 - mean2) ** 2) * torch.exp(-logvar2) + ) diff --git a/ldm/modules/ema.py b/ldm/modules/ema.py new file mode 100644 index 0000000000000000000000000000000000000000..bded25019b9bcbcd0260f0b8185f8c7859ca58c4 --- /dev/null +++ b/ldm/modules/ema.py @@ -0,0 +1,80 @@ +import torch +from torch import nn + + +class LitEma(nn.Module): + def __init__(self, model, decay=0.9999, use_num_upates=True): + super().__init__() + if decay < 0.0 or decay > 1.0: + raise ValueError('Decay must be between 0 and 1') + + self.m_name2s_name = {} + self.register_buffer('decay', torch.tensor(decay, dtype=torch.float32)) + self.register_buffer('num_updates', torch.tensor(0, dtype=torch.int) if use_num_upates + else torch.tensor(-1, dtype=torch.int)) + + for name, p in model.named_parameters(): + if p.requires_grad: + # remove as '.'-character is not allowed in buffers + s_name = name.replace('.', '') + self.m_name2s_name.update({name: s_name}) + self.register_buffer(s_name, p.clone().detach().data) + + self.collected_params = [] + + def reset_num_updates(self): + del self.num_updates + self.register_buffer('num_updates', torch.tensor(0, dtype=torch.int)) + + def forward(self, model): + decay = self.decay + + if self.num_updates >= 0: + self.num_updates += 1 + decay = min(self.decay, (1 + self.num_updates) / (10 + self.num_updates)) + + one_minus_decay = 1.0 - decay + + with torch.no_grad(): + m_param = dict(model.named_parameters()) + shadow_params = dict(self.named_buffers()) + + for key in m_param: + if m_param[key].requires_grad: + sname = self.m_name2s_name[key] + shadow_params[sname] = shadow_params[sname].type_as(m_param[key]) + shadow_params[sname].sub_(one_minus_decay * (shadow_params[sname] - m_param[key])) + else: + assert not key in self.m_name2s_name + + def copy_to(self, model): + m_param = dict(model.named_parameters()) + shadow_params = dict(self.named_buffers()) + for key in m_param: + if m_param[key].requires_grad: + m_param[key].data.copy_(shadow_params[self.m_name2s_name[key]].data) + else: + assert not key in self.m_name2s_name + + def store(self, parameters): + """ + Save the current parameters for restoring later. + Args: + parameters: Iterable of `torch.nn.Parameter`; the parameters to be + temporarily stored. + """ + self.collected_params = [param.clone() for param in parameters] + + def restore(self, parameters): + """ + Restore the parameters stored with the `store` method. + Useful to validate the model with EMA parameters without affecting the + original optimization process. Store the parameters before the + `copy_to` method. After validation (or model saving), use this to + restore the former parameters. + Args: + parameters: Iterable of `torch.nn.Parameter`; the parameters to be + updated with the stored parameters. + """ + for c_param, param in zip(self.collected_params, parameters): + param.data.copy_(c_param.data) diff --git a/ldm/modules/encoders/__init__.py b/ldm/modules/encoders/__init__.py new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/ldm/modules/encoders/modules.py b/ldm/modules/encoders/modules.py new file mode 100644 index 0000000000000000000000000000000000000000..0f3638a8ec4062d6aa4cd7046f34be502f785a88 --- /dev/null +++ b/ldm/modules/encoders/modules.py @@ -0,0 +1,385 @@ +import torch +import torch.nn as nn +from torch.utils.checkpoint import checkpoint + +from transformers import T5Tokenizer, T5EncoderModel, CLIPTokenizer, CLIPTextModel, AutoProcessor, CLIPVisionModelWithProjection + +import open_clip +from ldm.util import count_params + + +def _expand_mask(mask, dtype, tgt_len=None): + """ + Expands attention_mask from `[bsz, seq_len]` to `[bsz, 1, tgt_seq_len, src_seq_len]`. + """ + bsz, src_len = mask.size() + tgt_len = tgt_len if tgt_len is not None else src_len + + expanded_mask = mask[:, None, None, :].expand(bsz, 1, tgt_len, src_len).to(dtype) + + inverted_mask = 1.0 - expanded_mask + + return inverted_mask.masked_fill(inverted_mask.to(torch.bool), torch.finfo(dtype).min) + + +def _build_causal_attention_mask(bsz, seq_len, dtype): + # lazily create causal attention mask, with full attention between the vision tokens + # pytorch uses additive attention mask; fill with -inf + mask = torch.empty(bsz, seq_len, seq_len, dtype=dtype) + mask.fill_(torch.tensor(torch.finfo(dtype).min)) + mask.triu_(1) # zero out the lower diagonal + mask = mask.unsqueeze(1) # expand mask + return mask + +class AbstractEncoder(nn.Module): + def __init__(self): + super().__init__() + + def encode(self, *args, **kwargs): + raise NotImplementedError + + +class IdentityEncoder(AbstractEncoder): + + def encode(self, x): + return x + + +class ClassEmbedder(nn.Module): + def __init__(self, embed_dim, n_classes=1000, key='class', ucg_rate=0.1): + super().__init__() + self.key = key + self.embedding = nn.Embedding(n_classes, embed_dim) + self.n_classes = n_classes + self.ucg_rate = ucg_rate + + def forward(self, batch, key=None, disable_dropout=False): + if key is None: + key = self.key + # this is for use in crossattn + c = batch[key][:, None] + if self.ucg_rate > 0. and not disable_dropout: + mask = 1. - torch.bernoulli(torch.ones_like(c) * self.ucg_rate) + c = mask * c + (1-mask) * torch.ones_like(c)*(self.n_classes-1) + c = c.long() + c = self.embedding(c) + return c + + def get_unconditional_conditioning(self, bs, device="cuda"): + uc_class = self.n_classes - 1 # 1000 classes --> 0 ... 999, one extra class for ucg (class 1000) + uc = torch.ones((bs,), device=device) * uc_class + uc = {self.key: uc} + return uc + + +def disabled_train(self, mode=True): + """Overwrite model.train with this function to make sure train/eval mode + does not change anymore.""" + return self + + +class FrozenT5Embedder(AbstractEncoder): + """Uses the T5 transformer encoder for text""" + def __init__(self, version="google/t5-v1_1-large", device="cuda", max_length=77, freeze=True): # others are google/t5-v1_1-xl and google/t5-v1_1-xxl + super().__init__() + self.tokenizer = T5Tokenizer.from_pretrained(version) + self.transformer = T5EncoderModel.from_pretrained(version) + self.device = device + self.max_length = max_length # TODO: typical value? + if freeze: + self.freeze() + + def freeze(self): + self.transformer = self.transformer.eval() + #self.train = disabled_train + for param in self.parameters(): + param.requires_grad = False + + def forward(self, text): + batch_encoding = self.tokenizer(text, truncation=True, max_length=self.max_length, return_length=True, + return_overflowing_tokens=False, padding="max_length", return_tensors="pt") + tokens = batch_encoding["input_ids"].to(self.device) + outputs = self.transformer(input_ids=tokens) + + z = outputs.last_hidden_state + return z + + def encode(self, text): + return self(text) + + +class FrozenCLIPEmbedder(AbstractEncoder): + """Uses the CLIP transformer encoder for text (from huggingface)""" + LAYERS = [ + "last", + "pooled", + "hidden" + ] + def __init__(self, version="openai/clip-vit-large-patch14", device="cuda", max_length=77, + freeze=True, layer="last", layer_idx=None): # clip-vit-base-patch32 + super().__init__() + assert layer in self.LAYERS + self.tokenizer = CLIPTokenizer.from_pretrained(version) + self.transformer = CLIPTextModel.from_pretrained(version) + self.device = device + self.max_length = max_length + if freeze: + self.freeze() + self.layer = layer + self.layer_idx = layer_idx + if layer == "hidden": + assert layer_idx is not None + assert 0 <= abs(layer_idx) <= 12 + + def freeze(self): + self.transformer = self.transformer.eval() + # self.train = disabled_train + for param in self.parameters(): + param.requires_grad = False + + def forward(self, text): + batch_encoding = self.tokenizer(text, truncation=True, max_length=self.max_length, return_length=True, + return_overflowing_tokens=False, padding="max_length", return_tensors="pt") + tokens = batch_encoding["input_ids"].to(self.device) + outputs = self.transformer(input_ids=tokens, output_hidden_states=self.layer=="hidden") + if self.layer == "last": + z = outputs.last_hidden_state + elif self.layer == "pooled": + z = outputs.pooler_output[:, None, :] + else: + z = outputs.hidden_states[self.layer_idx] + return z + + def encode(self, text): + return self(text) + + +class FrozenOpenCLIPEmbedder(AbstractEncoder): + """ + Uses the OpenCLIP transformer encoder for text + """ + LAYERS = [ + # "pooled", + "last", + "penultimate" + ] + + def __init__(self, arch="ViT-H-14", version="laion2b_s32b_b79k", device="cuda", max_length=77, + freeze=True, layer="last"): + super().__init__() + assert layer in self.LAYERS + model, _, _ = open_clip.create_model_and_transforms(arch, device=torch.device('cpu'), pretrained=version) + del model.visual + self.model = model + + self.device = device + self.max_length = max_length + if freeze: + self.freeze() + self.layer = layer + if self.layer == "last": + self.layer_idx = 0 + elif self.layer == "penultimate": + self.layer_idx = 1 + else: + raise NotImplementedError() + + def freeze(self): + self.model = self.model.eval() + for param in self.parameters(): + param.requires_grad = False + + def forward(self, text): + tokens = open_clip.tokenize(text) + z = self.encode_with_transformer(tokens.to(self.device)) + return z + + def encode_with_transformer(self, text): + x = self.model.token_embedding(text) # [batch_size, n_ctx, d_model] + x = x + self.model.positional_embedding + x = x.permute(1, 0, 2) # NLD -> LND + x = self.text_transformer_forward(x, attn_mask=self.model.attn_mask) + x = x.permute(1, 0, 2) # LND -> NLD + x = self.model.ln_final(x) + return x + + def text_transformer_forward(self, x: torch.Tensor, attn_mask=None): + for i, r in enumerate(self.model.transformer.resblocks): + if i == len(self.model.transformer.resblocks) - self.layer_idx: + break + if self.model.transformer.grad_checkpointing and not torch.jit.is_scripting(): + x = checkpoint(r, x, attn_mask) + else: + x = r(x, attn_mask=attn_mask) + return x + + def encode(self, text): + return self(text) + + +class FrozenCLIPT5Encoder(AbstractEncoder): + def __init__(self, clip_version="openai/clip-vit-large-patch14", t5_version="google/t5-v1_1-xl", device="cuda", + clip_max_length=77, t5_max_length=77): + super().__init__() + self.clip_encoder = FrozenCLIPEmbedder(clip_version, device, max_length=clip_max_length) + self.t5_encoder = FrozenT5Embedder(t5_version, device, max_length=t5_max_length) + print(f"{self.clip_encoder.__class__.__name__} has {count_params(self.clip_encoder)*1.e-6:.2f} M parameters, " + f"{self.t5_encoder.__class__.__name__} comes with {count_params(self.t5_encoder)*1.e-6:.2f} M params.") + + def encode(self, text): + return self(text) + + def forward(self, text): + clip_z = self.clip_encoder.encode(text) + t5_z = self.t5_encoder.encode(text) + return [clip_z, t5_z] + + +class FrozenCLIPEmbedderT3(AbstractEncoder): + """Uses the CLIP transformer encoder for text (from Hugging Face)""" + def __init__(self, version="openai/clip-vit-large-patch14", device="cuda", max_length=77, freeze=True, use_vision=False): + super().__init__() + self.tokenizer = CLIPTokenizer.from_pretrained(version) + self.transformer = CLIPTextModel.from_pretrained(version) + if use_vision: + self.vit = CLIPVisionModelWithProjection.from_pretrained(version) + self.processor = AutoProcessor.from_pretrained(version) + self.device = device + self.max_length = max_length + if freeze: + self.freeze() + + def embedding_forward( + self, + input_ids=None, + position_ids=None, + inputs_embeds=None, + embedding_manager=None, + ): + seq_length = input_ids.shape[-1] if input_ids is not None else inputs_embeds.shape[-2] + if position_ids is None: + position_ids = self.position_ids[:, :seq_length] + if inputs_embeds is None: + inputs_embeds = self.token_embedding(input_ids) + if embedding_manager is not None: + inputs_embeds = embedding_manager(input_ids, inputs_embeds) + position_embeddings = self.position_embedding(position_ids) + embeddings = inputs_embeds + position_embeddings + return embeddings + + self.transformer.text_model.embeddings.forward = embedding_forward.__get__(self.transformer.text_model.embeddings) + + def encoder_forward( + self, + inputs_embeds, + attention_mask=None, + causal_attention_mask=None, + output_attentions=None, + output_hidden_states=None, + return_dict=None, + ): + output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions + output_hidden_states = ( + output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states + ) + return_dict = return_dict if return_dict is not None else self.config.use_return_dict + encoder_states = () if output_hidden_states else None + all_attentions = () if output_attentions else None + hidden_states = inputs_embeds + for idx, encoder_layer in enumerate(self.layers): + if output_hidden_states: + encoder_states = encoder_states + (hidden_states,) + layer_outputs = encoder_layer( + hidden_states, + attention_mask, + causal_attention_mask, + output_attentions=output_attentions, + ) + hidden_states = layer_outputs[0] + if output_attentions: + all_attentions = all_attentions + (layer_outputs[1],) + if output_hidden_states: + encoder_states = encoder_states + (hidden_states,) + return hidden_states + + self.transformer.text_model.encoder.forward = encoder_forward.__get__(self.transformer.text_model.encoder) + + def text_encoder_forward( + self, + input_ids=None, + attention_mask=None, + position_ids=None, + output_attentions=None, + output_hidden_states=None, + return_dict=None, + embedding_manager=None, + ): + output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions + output_hidden_states = ( + output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states + ) + return_dict = return_dict if return_dict is not None else self.config.use_return_dict + if input_ids is None: + raise ValueError("You have to specify either input_ids") + input_shape = input_ids.size() + input_ids = input_ids.view(-1, input_shape[-1]) + hidden_states = self.embeddings(input_ids=input_ids, position_ids=position_ids, embedding_manager=embedding_manager) + bsz, seq_len = input_shape + # CLIP's text model uses causal mask, prepare it here. + # https://github.com/openai/CLIP/blob/cfcffb90e69f37bf2ff1e988237a0fbe41f33c04/clip/model.py#L324 + causal_attention_mask = _build_causal_attention_mask(bsz, seq_len, hidden_states.dtype).to( + hidden_states.device + ) + # expand attention_mask + if attention_mask is not None: + # [bsz, seq_len] -> [bsz, 1, tgt_seq_len, src_seq_len] + attention_mask = _expand_mask(attention_mask, hidden_states.dtype) + last_hidden_state = self.encoder( + inputs_embeds=hidden_states, + attention_mask=attention_mask, + causal_attention_mask=causal_attention_mask, + output_attentions=output_attentions, + output_hidden_states=output_hidden_states, + return_dict=return_dict, + ) + last_hidden_state = self.final_layer_norm(last_hidden_state) + return last_hidden_state + + self.transformer.text_model.forward = text_encoder_forward.__get__(self.transformer.text_model) + + def transformer_forward( + self, + input_ids=None, + attention_mask=None, + position_ids=None, + output_attentions=None, + output_hidden_states=None, + return_dict=None, + embedding_manager=None, + ): + return self.text_model( + input_ids=input_ids, + attention_mask=attention_mask, + position_ids=position_ids, + output_attentions=output_attentions, + output_hidden_states=output_hidden_states, + return_dict=return_dict, + embedding_manager=embedding_manager + ) + + self.transformer.forward = transformer_forward.__get__(self.transformer) + + def freeze(self): + self.transformer = self.transformer.eval() + for param in self.parameters(): + param.requires_grad = False + + def forward(self, text, **kwargs): + batch_encoding = self.tokenizer(text, truncation=True, max_length=self.max_length, return_length=True, + return_overflowing_tokens=False, padding="max_length", return_tensors="pt") + tokens = batch_encoding["input_ids"].to(self.device) + z = self.transformer(input_ids=tokens, **kwargs) + return z + + def encode(self, text, **kwargs): + return self(text, **kwargs) diff --git a/ldm/modules/image_degradation/__init__.py b/ldm/modules/image_degradation/__init__.py new file mode 100644 index 0000000000000000000000000000000000000000..7836cada81f90ded99c58d5942eea4c3477f58fc --- /dev/null +++ b/ldm/modules/image_degradation/__init__.py @@ -0,0 +1,2 @@ +from ldm.modules.image_degradation.bsrgan import degradation_bsrgan_variant as degradation_fn_bsr +from ldm.modules.image_degradation.bsrgan_light import degradation_bsrgan_variant as degradation_fn_bsr_light diff --git a/ldm/modules/image_degradation/bsrgan.py b/ldm/modules/image_degradation/bsrgan.py new file mode 100644 index 0000000000000000000000000000000000000000..32ef56169978e550090261cddbcf5eb611a6173b --- /dev/null +++ b/ldm/modules/image_degradation/bsrgan.py @@ -0,0 +1,730 @@ +# -*- coding: utf-8 -*- +""" +# -------------------------------------------- +# Super-Resolution +# -------------------------------------------- +# +# Kai Zhang (cskaizhang@gmail.com) +# https://github.com/cszn +# From 2019/03--2021/08 +# -------------------------------------------- +""" + +import numpy as np +import cv2 +import torch + +from functools import partial +import random +from scipy import ndimage +import scipy +import scipy.stats as ss +from scipy.interpolate import interp2d +from scipy.linalg import orth +import albumentations + +import ldm.modules.image_degradation.utils_image as util + + +def modcrop_np(img, sf): + ''' + Args: + img: numpy image, WxH or WxHxC + sf: scale factor + Return: + cropped image + ''' + w, h = img.shape[:2] + im = np.copy(img) + return im[:w - w % sf, :h - h % sf, ...] + + +""" +# -------------------------------------------- +# anisotropic Gaussian kernels +# -------------------------------------------- +""" + + +def analytic_kernel(k): + """Calculate the X4 kernel from the X2 kernel (for proof see appendix in paper)""" + k_size = k.shape[0] + # Calculate the big kernels size + big_k = np.zeros((3 * k_size - 2, 3 * k_size - 2)) + # Loop over the small kernel to fill the big one + for r in range(k_size): + for c in range(k_size): + big_k[2 * r:2 * r + k_size, 2 * c:2 * c + k_size] += k[r, c] * k + # Crop the edges of the big kernel to ignore very small values and increase run time of SR + crop = k_size // 2 + cropped_big_k = big_k[crop:-crop, crop:-crop] + # Normalize to 1 + return cropped_big_k / cropped_big_k.sum() + + +def anisotropic_Gaussian(ksize=15, theta=np.pi, l1=6, l2=6): + """ generate an anisotropic Gaussian kernel + Args: + ksize : e.g., 15, kernel size + theta : [0, pi], rotation angle range + l1 : [0.1,50], scaling of eigenvalues + l2 : [0.1,l1], scaling of eigenvalues + If l1 = l2, will get an isotropic Gaussian kernel. + Returns: + k : kernel + """ + + v = np.dot(np.array([[np.cos(theta), -np.sin(theta)], [np.sin(theta), np.cos(theta)]]), np.array([1., 0.])) + V = np.array([[v[0], v[1]], [v[1], -v[0]]]) + D = np.array([[l1, 0], [0, l2]]) + Sigma = np.dot(np.dot(V, D), np.linalg.inv(V)) + k = gm_blur_kernel(mean=[0, 0], cov=Sigma, size=ksize) + + return k + + +def gm_blur_kernel(mean, cov, size=15): + center = size / 2.0 + 0.5 + k = np.zeros([size, size]) + for y in range(size): + for x in range(size): + cy = y - center + 1 + cx = x - center + 1 + k[y, x] = ss.multivariate_normal.pdf([cx, cy], mean=mean, cov=cov) + + k = k / np.sum(k) + return k + + +def shift_pixel(x, sf, upper_left=True): + """shift pixel for super-resolution with different scale factors + Args: + x: WxHxC or WxH + sf: scale factor + upper_left: shift direction + """ + h, w = x.shape[:2] + shift = (sf - 1) * 0.5 + xv, yv = np.arange(0, w, 1.0), np.arange(0, h, 1.0) + if upper_left: + x1 = xv + shift + y1 = yv + shift + else: + x1 = xv - shift + y1 = yv - shift + + x1 = np.clip(x1, 0, w - 1) + y1 = np.clip(y1, 0, h - 1) + + if x.ndim == 2: + x = interp2d(xv, yv, x)(x1, y1) + if x.ndim == 3: + for i in range(x.shape[-1]): + x[:, :, i] = interp2d(xv, yv, x[:, :, i])(x1, y1) + + return x + + +def blur(x, k): + ''' + x: image, NxcxHxW + k: kernel, Nx1xhxw + ''' + n, c = x.shape[:2] + p1, p2 = (k.shape[-2] - 1) // 2, (k.shape[-1] - 1) // 2 + x = torch.nn.functional.pad(x, pad=(p1, p2, p1, p2), mode='replicate') + k = k.repeat(1, c, 1, 1) + k = k.view(-1, 1, k.shape[2], k.shape[3]) + x = x.view(1, -1, x.shape[2], x.shape[3]) + x = torch.nn.functional.conv2d(x, k, bias=None, stride=1, padding=0, groups=n * c) + x = x.view(n, c, x.shape[2], x.shape[3]) + + return x + + +def gen_kernel(k_size=np.array([15, 15]), scale_factor=np.array([4, 4]), min_var=0.6, max_var=10., noise_level=0): + """" + # modified version of https://github.com/assafshocher/BlindSR_dataset_generator + # Kai Zhang + # min_var = 0.175 * sf # variance of the gaussian kernel will be sampled between min_var and max_var + # max_var = 2.5 * sf + """ + # Set random eigen-vals (lambdas) and angle (theta) for COV matrix + lambda_1 = min_var + np.random.rand() * (max_var - min_var) + lambda_2 = min_var + np.random.rand() * (max_var - min_var) + theta = np.random.rand() * np.pi # random theta + noise = -noise_level + np.random.rand(*k_size) * noise_level * 2 + + # Set COV matrix using Lambdas and Theta + LAMBDA = np.diag([lambda_1, lambda_2]) + Q = np.array([[np.cos(theta), -np.sin(theta)], + [np.sin(theta), np.cos(theta)]]) + SIGMA = Q @ LAMBDA @ Q.T + INV_SIGMA = np.linalg.inv(SIGMA)[None, None, :, :] + + # Set expectation position (shifting kernel for aligned image) + MU = k_size // 2 - 0.5 * (scale_factor - 1) # - 0.5 * (scale_factor - k_size % 2) + MU = MU[None, None, :, None] + + # Create meshgrid for Gaussian + [X, Y] = np.meshgrid(range(k_size[0]), range(k_size[1])) + Z = np.stack([X, Y], 2)[:, :, :, None] + + # Calcualte Gaussian for every pixel of the kernel + ZZ = Z - MU + ZZ_t = ZZ.transpose(0, 1, 3, 2) + raw_kernel = np.exp(-0.5 * np.squeeze(ZZ_t @ INV_SIGMA @ ZZ)) * (1 + noise) + + # shift the kernel so it will be centered + # raw_kernel_centered = kernel_shift(raw_kernel, scale_factor) + + # Normalize the kernel and return + # kernel = raw_kernel_centered / np.sum(raw_kernel_centered) + kernel = raw_kernel / np.sum(raw_kernel) + return kernel + + +def fspecial_gaussian(hsize, sigma): + hsize = [hsize, hsize] + siz = [(hsize[0] - 1.0) / 2.0, (hsize[1] - 1.0) / 2.0] + std = sigma + [x, y] = np.meshgrid(np.arange(-siz[1], siz[1] + 1), np.arange(-siz[0], siz[0] + 1)) + arg = -(x * x + y * y) / (2 * std * std) + h = np.exp(arg) + h[h < scipy.finfo(float).eps * h.max()] = 0 + sumh = h.sum() + if sumh != 0: + h = h / sumh + return h + + +def fspecial_laplacian(alpha): + alpha = max([0, min([alpha, 1])]) + h1 = alpha / (alpha + 1) + h2 = (1 - alpha) / (alpha + 1) + h = [[h1, h2, h1], [h2, -4 / (alpha + 1), h2], [h1, h2, h1]] + h = np.array(h) + return h + + +def fspecial(filter_type, *args, **kwargs): + ''' + python code from: + https://github.com/ronaldosena/imagens-medicas-2/blob/40171a6c259edec7827a6693a93955de2bd39e76/Aulas/aula_2_-_uniform_filter/matlab_fspecial.py + ''' + if filter_type == 'gaussian': + return fspecial_gaussian(*args, **kwargs) + if filter_type == 'laplacian': + return fspecial_laplacian(*args, **kwargs) + + +""" +# -------------------------------------------- +# degradation models +# -------------------------------------------- +""" + + +def bicubic_degradation(x, sf=3): + ''' + Args: + x: HxWxC image, [0, 1] + sf: down-scale factor + Return: + bicubicly downsampled LR image + ''' + x = util.imresize_np(x, scale=1 / sf) + return x + + +def srmd_degradation(x, k, sf=3): + ''' blur + bicubic downsampling + Args: + x: HxWxC image, [0, 1] + k: hxw, double + sf: down-scale factor + Return: + downsampled LR image + Reference: + @inproceedings{zhang2018learning, + title={Learning a single convolutional super-resolution network for multiple degradations}, + author={Zhang, Kai and Zuo, Wangmeng and Zhang, Lei}, + booktitle={IEEE Conference on Computer Vision and Pattern Recognition}, + pages={3262--3271}, + year={2018} + } + ''' + x = ndimage.filters.convolve(x, np.expand_dims(k, axis=2), mode='wrap') # 'nearest' | 'mirror' + x = bicubic_degradation(x, sf=sf) + return x + + +def dpsr_degradation(x, k, sf=3): + ''' bicubic downsampling + blur + Args: + x: HxWxC image, [0, 1] + k: hxw, double + sf: down-scale factor + Return: + downsampled LR image + Reference: + @inproceedings{zhang2019deep, + title={Deep Plug-and-Play Super-Resolution for Arbitrary Blur Kernels}, + author={Zhang, Kai and Zuo, Wangmeng and Zhang, Lei}, + booktitle={IEEE Conference on Computer Vision and Pattern Recognition}, + pages={1671--1681}, + year={2019} + } + ''' + x = bicubic_degradation(x, sf=sf) + x = ndimage.filters.convolve(x, np.expand_dims(k, axis=2), mode='wrap') + return x + + +def classical_degradation(x, k, sf=3): + ''' blur + downsampling + Args: + x: HxWxC image, [0, 1]/[0, 255] + k: hxw, double + sf: down-scale factor + Return: + downsampled LR image + ''' + x = ndimage.filters.convolve(x, np.expand_dims(k, axis=2), mode='wrap') + # x = filters.correlate(x, np.expand_dims(np.flip(k), axis=2)) + st = 0 + return x[st::sf, st::sf, ...] + + +def add_sharpening(img, weight=0.5, radius=50, threshold=10): + """USM sharpening. borrowed from real-ESRGAN + Input image: I; Blurry image: B. + 1. K = I + weight * (I - B) + 2. Mask = 1 if abs(I - B) > threshold, else: 0 + 3. Blur mask: + 4. Out = Mask * K + (1 - Mask) * I + Args: + img (Numpy array): Input image, HWC, BGR; float32, [0, 1]. + weight (float): Sharp weight. Default: 1. + radius (float): Kernel size of Gaussian blur. Default: 50. + threshold (int): + """ + if radius % 2 == 0: + radius += 1 + blur = cv2.GaussianBlur(img, (radius, radius), 0) + residual = img - blur + mask = np.abs(residual) * 255 > threshold + mask = mask.astype('float32') + soft_mask = cv2.GaussianBlur(mask, (radius, radius), 0) + + K = img + weight * residual + K = np.clip(K, 0, 1) + return soft_mask * K + (1 - soft_mask) * img + + +def add_blur(img, sf=4): + wd2 = 4.0 + sf + wd = 2.0 + 0.2 * sf + if random.random() < 0.5: + l1 = wd2 * random.random() + l2 = wd2 * random.random() + k = anisotropic_Gaussian(ksize=2 * random.randint(2, 11) + 3, theta=random.random() * np.pi, l1=l1, l2=l2) + else: + k = fspecial('gaussian', 2 * random.randint(2, 11) + 3, wd * random.random()) + img = ndimage.filters.convolve(img, np.expand_dims(k, axis=2), mode='mirror') + + return img + + +def add_resize(img, sf=4): + rnum = np.random.rand() + if rnum > 0.8: # up + sf1 = random.uniform(1, 2) + elif rnum < 0.7: # down + sf1 = random.uniform(0.5 / sf, 1) + else: + sf1 = 1.0 + img = cv2.resize(img, (int(sf1 * img.shape[1]), int(sf1 * img.shape[0])), interpolation=random.choice([1, 2, 3])) + img = np.clip(img, 0.0, 1.0) + + return img + + +# def add_Gaussian_noise(img, noise_level1=2, noise_level2=25): +# noise_level = random.randint(noise_level1, noise_level2) +# rnum = np.random.rand() +# if rnum > 0.6: # add color Gaussian noise +# img += np.random.normal(0, noise_level / 255.0, img.shape).astype(np.float32) +# elif rnum < 0.4: # add grayscale Gaussian noise +# img += np.random.normal(0, noise_level / 255.0, (*img.shape[:2], 1)).astype(np.float32) +# else: # add noise +# L = noise_level2 / 255. +# D = np.diag(np.random.rand(3)) +# U = orth(np.random.rand(3, 3)) +# conv = np.dot(np.dot(np.transpose(U), D), U) +# img += np.random.multivariate_normal([0, 0, 0], np.abs(L ** 2 * conv), img.shape[:2]).astype(np.float32) +# img = np.clip(img, 0.0, 1.0) +# return img + +def add_Gaussian_noise(img, noise_level1=2, noise_level2=25): + noise_level = random.randint(noise_level1, noise_level2) + rnum = np.random.rand() + if rnum > 0.6: # add color Gaussian noise + img = img + np.random.normal(0, noise_level / 255.0, img.shape).astype(np.float32) + elif rnum < 0.4: # add grayscale Gaussian noise + img = img + np.random.normal(0, noise_level / 255.0, (*img.shape[:2], 1)).astype(np.float32) + else: # add noise + L = noise_level2 / 255. + D = np.diag(np.random.rand(3)) + U = orth(np.random.rand(3, 3)) + conv = np.dot(np.dot(np.transpose(U), D), U) + img = img + np.random.multivariate_normal([0, 0, 0], np.abs(L ** 2 * conv), img.shape[:2]).astype(np.float32) + img = np.clip(img, 0.0, 1.0) + return img + + +def add_speckle_noise(img, noise_level1=2, noise_level2=25): + noise_level = random.randint(noise_level1, noise_level2) + img = np.clip(img, 0.0, 1.0) + rnum = random.random() + if rnum > 0.6: + img += img * np.random.normal(0, noise_level / 255.0, img.shape).astype(np.float32) + elif rnum < 0.4: + img += img * np.random.normal(0, noise_level / 255.0, (*img.shape[:2], 1)).astype(np.float32) + else: + L = noise_level2 / 255. + D = np.diag(np.random.rand(3)) + U = orth(np.random.rand(3, 3)) + conv = np.dot(np.dot(np.transpose(U), D), U) + img += img * np.random.multivariate_normal([0, 0, 0], np.abs(L ** 2 * conv), img.shape[:2]).astype(np.float32) + img = np.clip(img, 0.0, 1.0) + return img + + +def add_Poisson_noise(img): + img = np.clip((img * 255.0).round(), 0, 255) / 255. + vals = 10 ** (2 * random.random() + 2.0) # [2, 4] + if random.random() < 0.5: + img = np.random.poisson(img * vals).astype(np.float32) / vals + else: + img_gray = np.dot(img[..., :3], [0.299, 0.587, 0.114]) + img_gray = np.clip((img_gray * 255.0).round(), 0, 255) / 255. + noise_gray = np.random.poisson(img_gray * vals).astype(np.float32) / vals - img_gray + img += noise_gray[:, :, np.newaxis] + img = np.clip(img, 0.0, 1.0) + return img + + +def add_JPEG_noise(img): + quality_factor = random.randint(30, 95) + img = cv2.cvtColor(util.single2uint(img), cv2.COLOR_RGB2BGR) + result, encimg = cv2.imencode('.jpg', img, [int(cv2.IMWRITE_JPEG_QUALITY), quality_factor]) + img = cv2.imdecode(encimg, 1) + img = cv2.cvtColor(util.uint2single(img), cv2.COLOR_BGR2RGB) + return img + + +def random_crop(lq, hq, sf=4, lq_patchsize=64): + h, w = lq.shape[:2] + rnd_h = random.randint(0, h - lq_patchsize) + rnd_w = random.randint(0, w - lq_patchsize) + lq = lq[rnd_h:rnd_h + lq_patchsize, rnd_w:rnd_w + lq_patchsize, :] + + rnd_h_H, rnd_w_H = int(rnd_h * sf), int(rnd_w * sf) + hq = hq[rnd_h_H:rnd_h_H + lq_patchsize * sf, rnd_w_H:rnd_w_H + lq_patchsize * sf, :] + return lq, hq + + +def degradation_bsrgan(img, sf=4, lq_patchsize=72, isp_model=None): + """ + This is the degradation model of BSRGAN from the paper + "Designing a Practical Degradation Model for Deep Blind Image Super-Resolution" + ---------- + img: HXWXC, [0, 1], its size should be large than (lq_patchsizexsf)x(lq_patchsizexsf) + sf: scale factor + isp_model: camera ISP model + Returns + ------- + img: low-quality patch, size: lq_patchsizeXlq_patchsizeXC, range: [0, 1] + hq: corresponding high-quality patch, size: (lq_patchsizexsf)X(lq_patchsizexsf)XC, range: [0, 1] + """ + isp_prob, jpeg_prob, scale2_prob = 0.25, 0.9, 0.25 + sf_ori = sf + + h1, w1 = img.shape[:2] + img = img.copy()[:w1 - w1 % sf, :h1 - h1 % sf, ...] # mod crop + h, w = img.shape[:2] + + if h < lq_patchsize * sf or w < lq_patchsize * sf: + raise ValueError(f'img size ({h1}X{w1}) is too small!') + + hq = img.copy() + + if sf == 4 and random.random() < scale2_prob: # downsample1 + if np.random.rand() < 0.5: + img = cv2.resize(img, (int(1 / 2 * img.shape[1]), int(1 / 2 * img.shape[0])), + interpolation=random.choice([1, 2, 3])) + else: + img = util.imresize_np(img, 1 / 2, True) + img = np.clip(img, 0.0, 1.0) + sf = 2 + + shuffle_order = random.sample(range(7), 7) + idx1, idx2 = shuffle_order.index(2), shuffle_order.index(3) + if idx1 > idx2: # keep downsample3 last + shuffle_order[idx1], shuffle_order[idx2] = shuffle_order[idx2], shuffle_order[idx1] + + for i in shuffle_order: + + if i == 0: + img = add_blur(img, sf=sf) + + elif i == 1: + img = add_blur(img, sf=sf) + + elif i == 2: + a, b = img.shape[1], img.shape[0] + # downsample2 + if random.random() < 0.75: + sf1 = random.uniform(1, 2 * sf) + img = cv2.resize(img, (int(1 / sf1 * img.shape[1]), int(1 / sf1 * img.shape[0])), + interpolation=random.choice([1, 2, 3])) + else: + k = fspecial('gaussian', 25, random.uniform(0.1, 0.6 * sf)) + k_shifted = shift_pixel(k, sf) + k_shifted = k_shifted / k_shifted.sum() # blur with shifted kernel + img = ndimage.filters.convolve(img, np.expand_dims(k_shifted, axis=2), mode='mirror') + img = img[0::sf, 0::sf, ...] # nearest downsampling + img = np.clip(img, 0.0, 1.0) + + elif i == 3: + # downsample3 + img = cv2.resize(img, (int(1 / sf * a), int(1 / sf * b)), interpolation=random.choice([1, 2, 3])) + img = np.clip(img, 0.0, 1.0) + + elif i == 4: + # add Gaussian noise + img = add_Gaussian_noise(img, noise_level1=2, noise_level2=25) + + elif i == 5: + # add JPEG noise + if random.random() < jpeg_prob: + img = add_JPEG_noise(img) + + elif i == 6: + # add processed camera sensor noise + if random.random() < isp_prob and isp_model is not None: + with torch.no_grad(): + img, hq = isp_model.forward(img.copy(), hq) + + # add final JPEG compression noise + img = add_JPEG_noise(img) + + # random crop + img, hq = random_crop(img, hq, sf_ori, lq_patchsize) + + return img, hq + + +# todo no isp_model? +def degradation_bsrgan_variant(image, sf=4, isp_model=None): + """ + This is the degradation model of BSRGAN from the paper + "Designing a Practical Degradation Model for Deep Blind Image Super-Resolution" + ---------- + sf: scale factor + isp_model: camera ISP model + Returns + ------- + img: low-quality patch, size: lq_patchsizeXlq_patchsizeXC, range: [0, 1] + hq: corresponding high-quality patch, size: (lq_patchsizexsf)X(lq_patchsizexsf)XC, range: [0, 1] + """ + image = util.uint2single(image) + isp_prob, jpeg_prob, scale2_prob = 0.25, 0.9, 0.25 + sf_ori = sf + + h1, w1 = image.shape[:2] + image = image.copy()[:w1 - w1 % sf, :h1 - h1 % sf, ...] # mod crop + h, w = image.shape[:2] + + hq = image.copy() + + if sf == 4 and random.random() < scale2_prob: # downsample1 + if np.random.rand() < 0.5: + image = cv2.resize(image, (int(1 / 2 * image.shape[1]), int(1 / 2 * image.shape[0])), + interpolation=random.choice([1, 2, 3])) + else: + image = util.imresize_np(image, 1 / 2, True) + image = np.clip(image, 0.0, 1.0) + sf = 2 + + shuffle_order = random.sample(range(7), 7) + idx1, idx2 = shuffle_order.index(2), shuffle_order.index(3) + if idx1 > idx2: # keep downsample3 last + shuffle_order[idx1], shuffle_order[idx2] = shuffle_order[idx2], shuffle_order[idx1] + + for i in shuffle_order: + + if i == 0: + image = add_blur(image, sf=sf) + + elif i == 1: + image = add_blur(image, sf=sf) + + elif i == 2: + a, b = image.shape[1], image.shape[0] + # downsample2 + if random.random() < 0.75: + sf1 = random.uniform(1, 2 * sf) + image = cv2.resize(image, (int(1 / sf1 * image.shape[1]), int(1 / sf1 * image.shape[0])), + interpolation=random.choice([1, 2, 3])) + else: + k = fspecial('gaussian', 25, random.uniform(0.1, 0.6 * sf)) + k_shifted = shift_pixel(k, sf) + k_shifted = k_shifted / k_shifted.sum() # blur with shifted kernel + image = ndimage.filters.convolve(image, np.expand_dims(k_shifted, axis=2), mode='mirror') + image = image[0::sf, 0::sf, ...] # nearest downsampling + image = np.clip(image, 0.0, 1.0) + + elif i == 3: + # downsample3 + image = cv2.resize(image, (int(1 / sf * a), int(1 / sf * b)), interpolation=random.choice([1, 2, 3])) + image = np.clip(image, 0.0, 1.0) + + elif i == 4: + # add Gaussian noise + image = add_Gaussian_noise(image, noise_level1=2, noise_level2=25) + + elif i == 5: + # add JPEG noise + if random.random() < jpeg_prob: + image = add_JPEG_noise(image) + + # elif i == 6: + # # add processed camera sensor noise + # if random.random() < isp_prob and isp_model is not None: + # with torch.no_grad(): + # img, hq = isp_model.forward(img.copy(), hq) + + # add final JPEG compression noise + image = add_JPEG_noise(image) + image = util.single2uint(image) + example = {"image":image} + return example + + +# TODO incase there is a pickle error one needs to replace a += x with a = a + x in add_speckle_noise etc... +def degradation_bsrgan_plus(img, sf=4, shuffle_prob=0.5, use_sharp=True, lq_patchsize=64, isp_model=None): + """ + This is an extended degradation model by combining + the degradation models of BSRGAN and Real-ESRGAN + ---------- + img: HXWXC, [0, 1], its size should be large than (lq_patchsizexsf)x(lq_patchsizexsf) + sf: scale factor + use_shuffle: the degradation shuffle + use_sharp: sharpening the img + Returns + ------- + img: low-quality patch, size: lq_patchsizeXlq_patchsizeXC, range: [0, 1] + hq: corresponding high-quality patch, size: (lq_patchsizexsf)X(lq_patchsizexsf)XC, range: [0, 1] + """ + + h1, w1 = img.shape[:2] + img = img.copy()[:w1 - w1 % sf, :h1 - h1 % sf, ...] # mod crop + h, w = img.shape[:2] + + if h < lq_patchsize * sf or w < lq_patchsize * sf: + raise ValueError(f'img size ({h1}X{w1}) is too small!') + + if use_sharp: + img = add_sharpening(img) + hq = img.copy() + + if random.random() < shuffle_prob: + shuffle_order = random.sample(range(13), 13) + else: + shuffle_order = list(range(13)) + # local shuffle for noise, JPEG is always the last one + shuffle_order[2:6] = random.sample(shuffle_order[2:6], len(range(2, 6))) + shuffle_order[9:13] = random.sample(shuffle_order[9:13], len(range(9, 13))) + + poisson_prob, speckle_prob, isp_prob = 0.1, 0.1, 0.1 + + for i in shuffle_order: + if i == 0: + img = add_blur(img, sf=sf) + elif i == 1: + img = add_resize(img, sf=sf) + elif i == 2: + img = add_Gaussian_noise(img, noise_level1=2, noise_level2=25) + elif i == 3: + if random.random() < poisson_prob: + img = add_Poisson_noise(img) + elif i == 4: + if random.random() < speckle_prob: + img = add_speckle_noise(img) + elif i == 5: + if random.random() < isp_prob and isp_model is not None: + with torch.no_grad(): + img, hq = isp_model.forward(img.copy(), hq) + elif i == 6: + img = add_JPEG_noise(img) + elif i == 7: + img = add_blur(img, sf=sf) + elif i == 8: + img = add_resize(img, sf=sf) + elif i == 9: + img = add_Gaussian_noise(img, noise_level1=2, noise_level2=25) + elif i == 10: + if random.random() < poisson_prob: + img = add_Poisson_noise(img) + elif i == 11: + if random.random() < speckle_prob: + img = add_speckle_noise(img) + elif i == 12: + if random.random() < isp_prob and isp_model is not None: + with torch.no_grad(): + img, hq = isp_model.forward(img.copy(), hq) + else: + print('check the shuffle!') + + # resize to desired size + img = cv2.resize(img, (int(1 / sf * hq.shape[1]), int(1 / sf * hq.shape[0])), + interpolation=random.choice([1, 2, 3])) + + # add final JPEG compression noise + img = add_JPEG_noise(img) + + # random crop + img, hq = random_crop(img, hq, sf, lq_patchsize) + + return img, hq + + +if __name__ == '__main__': + print("hey") + img = util.imread_uint('utils/test.png', 3) + print(img) + img = util.uint2single(img) + print(img) + img = img[:448, :448] + h = img.shape[0] // 4 + print("resizing to", h) + sf = 4 + deg_fn = partial(degradation_bsrgan_variant, sf=sf) + for i in range(20): + print(i) + img_lq = deg_fn(img) + print(img_lq) + img_lq_bicubic = albumentations.SmallestMaxSize(max_size=h, interpolation=cv2.INTER_CUBIC)(image=img)["image"] + print(img_lq.shape) + print("bicubic", img_lq_bicubic.shape) + print(img_hq.shape) + lq_nearest = cv2.resize(util.single2uint(img_lq), (int(sf * img_lq.shape[1]), int(sf * img_lq.shape[0])), + interpolation=0) + lq_bicubic_nearest = cv2.resize(util.single2uint(img_lq_bicubic), (int(sf * img_lq.shape[1]), int(sf * img_lq.shape[0])), + interpolation=0) + img_concat = np.concatenate([lq_bicubic_nearest, lq_nearest, util.single2uint(img_hq)], axis=1) + util.imsave(img_concat, str(i) + '.png') + + diff --git a/ldm/modules/image_degradation/bsrgan_light.py b/ldm/modules/image_degradation/bsrgan_light.py new file mode 100644 index 0000000000000000000000000000000000000000..808c7f882cb75e2ba2340d5b55881d11927351f0 --- /dev/null +++ b/ldm/modules/image_degradation/bsrgan_light.py @@ -0,0 +1,651 @@ +# -*- coding: utf-8 -*- +import numpy as np +import cv2 +import torch + +from functools import partial +import random +from scipy import ndimage +import scipy +import scipy.stats as ss +from scipy.interpolate import interp2d +from scipy.linalg import orth +import albumentations + +import ldm.modules.image_degradation.utils_image as util + +""" +# -------------------------------------------- +# Super-Resolution +# -------------------------------------------- +# +# Kai Zhang (cskaizhang@gmail.com) +# https://github.com/cszn +# From 2019/03--2021/08 +# -------------------------------------------- +""" + +def modcrop_np(img, sf): + ''' + Args: + img: numpy image, WxH or WxHxC + sf: scale factor + Return: + cropped image + ''' + w, h = img.shape[:2] + im = np.copy(img) + return im[:w - w % sf, :h - h % sf, ...] + + +""" +# -------------------------------------------- +# anisotropic Gaussian kernels +# -------------------------------------------- +""" + + +def analytic_kernel(k): + """Calculate the X4 kernel from the X2 kernel (for proof see appendix in paper)""" + k_size = k.shape[0] + # Calculate the big kernels size + big_k = np.zeros((3 * k_size - 2, 3 * k_size - 2)) + # Loop over the small kernel to fill the big one + for r in range(k_size): + for c in range(k_size): + big_k[2 * r:2 * r + k_size, 2 * c:2 * c + k_size] += k[r, c] * k + # Crop the edges of the big kernel to ignore very small values and increase run time of SR + crop = k_size // 2 + cropped_big_k = big_k[crop:-crop, crop:-crop] + # Normalize to 1 + return cropped_big_k / cropped_big_k.sum() + + +def anisotropic_Gaussian(ksize=15, theta=np.pi, l1=6, l2=6): + """ generate an anisotropic Gaussian kernel + Args: + ksize : e.g., 15, kernel size + theta : [0, pi], rotation angle range + l1 : [0.1,50], scaling of eigenvalues + l2 : [0.1,l1], scaling of eigenvalues + If l1 = l2, will get an isotropic Gaussian kernel. + Returns: + k : kernel + """ + + v = np.dot(np.array([[np.cos(theta), -np.sin(theta)], [np.sin(theta), np.cos(theta)]]), np.array([1., 0.])) + V = np.array([[v[0], v[1]], [v[1], -v[0]]]) + D = np.array([[l1, 0], [0, l2]]) + Sigma = np.dot(np.dot(V, D), np.linalg.inv(V)) + k = gm_blur_kernel(mean=[0, 0], cov=Sigma, size=ksize) + + return k + + +def gm_blur_kernel(mean, cov, size=15): + center = size / 2.0 + 0.5 + k = np.zeros([size, size]) + for y in range(size): + for x in range(size): + cy = y - center + 1 + cx = x - center + 1 + k[y, x] = ss.multivariate_normal.pdf([cx, cy], mean=mean, cov=cov) + + k = k / np.sum(k) + return k + + +def shift_pixel(x, sf, upper_left=True): + """shift pixel for super-resolution with different scale factors + Args: + x: WxHxC or WxH + sf: scale factor + upper_left: shift direction + """ + h, w = x.shape[:2] + shift = (sf - 1) * 0.5 + xv, yv = np.arange(0, w, 1.0), np.arange(0, h, 1.0) + if upper_left: + x1 = xv + shift + y1 = yv + shift + else: + x1 = xv - shift + y1 = yv - shift + + x1 = np.clip(x1, 0, w - 1) + y1 = np.clip(y1, 0, h - 1) + + if x.ndim == 2: + x = interp2d(xv, yv, x)(x1, y1) + if x.ndim == 3: + for i in range(x.shape[-1]): + x[:, :, i] = interp2d(xv, yv, x[:, :, i])(x1, y1) + + return x + + +def blur(x, k): + ''' + x: image, NxcxHxW + k: kernel, Nx1xhxw + ''' + n, c = x.shape[:2] + p1, p2 = (k.shape[-2] - 1) // 2, (k.shape[-1] - 1) // 2 + x = torch.nn.functional.pad(x, pad=(p1, p2, p1, p2), mode='replicate') + k = k.repeat(1, c, 1, 1) + k = k.view(-1, 1, k.shape[2], k.shape[3]) + x = x.view(1, -1, x.shape[2], x.shape[3]) + x = torch.nn.functional.conv2d(x, k, bias=None, stride=1, padding=0, groups=n * c) + x = x.view(n, c, x.shape[2], x.shape[3]) + + return x + + +def gen_kernel(k_size=np.array([15, 15]), scale_factor=np.array([4, 4]), min_var=0.6, max_var=10., noise_level=0): + """" + # modified version of https://github.com/assafshocher/BlindSR_dataset_generator + # Kai Zhang + # min_var = 0.175 * sf # variance of the gaussian kernel will be sampled between min_var and max_var + # max_var = 2.5 * sf + """ + # Set random eigen-vals (lambdas) and angle (theta) for COV matrix + lambda_1 = min_var + np.random.rand() * (max_var - min_var) + lambda_2 = min_var + np.random.rand() * (max_var - min_var) + theta = np.random.rand() * np.pi # random theta + noise = -noise_level + np.random.rand(*k_size) * noise_level * 2 + + # Set COV matrix using Lambdas and Theta + LAMBDA = np.diag([lambda_1, lambda_2]) + Q = np.array([[np.cos(theta), -np.sin(theta)], + [np.sin(theta), np.cos(theta)]]) + SIGMA = Q @ LAMBDA @ Q.T + INV_SIGMA = np.linalg.inv(SIGMA)[None, None, :, :] + + # Set expectation position (shifting kernel for aligned image) + MU = k_size // 2 - 0.5 * (scale_factor - 1) # - 0.5 * (scale_factor - k_size % 2) + MU = MU[None, None, :, None] + + # Create meshgrid for Gaussian + [X, Y] = np.meshgrid(range(k_size[0]), range(k_size[1])) + Z = np.stack([X, Y], 2)[:, :, :, None] + + # Calcualte Gaussian for every pixel of the kernel + ZZ = Z - MU + ZZ_t = ZZ.transpose(0, 1, 3, 2) + raw_kernel = np.exp(-0.5 * np.squeeze(ZZ_t @ INV_SIGMA @ ZZ)) * (1 + noise) + + # shift the kernel so it will be centered + # raw_kernel_centered = kernel_shift(raw_kernel, scale_factor) + + # Normalize the kernel and return + # kernel = raw_kernel_centered / np.sum(raw_kernel_centered) + kernel = raw_kernel / np.sum(raw_kernel) + return kernel + + +def fspecial_gaussian(hsize, sigma): + hsize = [hsize, hsize] + siz = [(hsize[0] - 1.0) / 2.0, (hsize[1] - 1.0) / 2.0] + std = sigma + [x, y] = np.meshgrid(np.arange(-siz[1], siz[1] + 1), np.arange(-siz[0], siz[0] + 1)) + arg = -(x * x + y * y) / (2 * std * std) + h = np.exp(arg) + h[h < scipy.finfo(float).eps * h.max()] = 0 + sumh = h.sum() + if sumh != 0: + h = h / sumh + return h + + +def fspecial_laplacian(alpha): + alpha = max([0, min([alpha, 1])]) + h1 = alpha / (alpha + 1) + h2 = (1 - alpha) / (alpha + 1) + h = [[h1, h2, h1], [h2, -4 / (alpha + 1), h2], [h1, h2, h1]] + h = np.array(h) + return h + + +def fspecial(filter_type, *args, **kwargs): + ''' + python code from: + https://github.com/ronaldosena/imagens-medicas-2/blob/40171a6c259edec7827a6693a93955de2bd39e76/Aulas/aula_2_-_uniform_filter/matlab_fspecial.py + ''' + if filter_type == 'gaussian': + return fspecial_gaussian(*args, **kwargs) + if filter_type == 'laplacian': + return fspecial_laplacian(*args, **kwargs) + + +""" +# -------------------------------------------- +# degradation models +# -------------------------------------------- +""" + + +def bicubic_degradation(x, sf=3): + ''' + Args: + x: HxWxC image, [0, 1] + sf: down-scale factor + Return: + bicubicly downsampled LR image + ''' + x = util.imresize_np(x, scale=1 / sf) + return x + + +def srmd_degradation(x, k, sf=3): + ''' blur + bicubic downsampling + Args: + x: HxWxC image, [0, 1] + k: hxw, double + sf: down-scale factor + Return: + downsampled LR image + Reference: + @inproceedings{zhang2018learning, + title={Learning a single convolutional super-resolution network for multiple degradations}, + author={Zhang, Kai and Zuo, Wangmeng and Zhang, Lei}, + booktitle={IEEE Conference on Computer Vision and Pattern Recognition}, + pages={3262--3271}, + year={2018} + } + ''' + x = ndimage.convolve(x, np.expand_dims(k, axis=2), mode='wrap') # 'nearest' | 'mirror' + x = bicubic_degradation(x, sf=sf) + return x + + +def dpsr_degradation(x, k, sf=3): + ''' bicubic downsampling + blur + Args: + x: HxWxC image, [0, 1] + k: hxw, double + sf: down-scale factor + Return: + downsampled LR image + Reference: + @inproceedings{zhang2019deep, + title={Deep Plug-and-Play Super-Resolution for Arbitrary Blur Kernels}, + author={Zhang, Kai and Zuo, Wangmeng and Zhang, Lei}, + booktitle={IEEE Conference on Computer Vision and Pattern Recognition}, + pages={1671--1681}, + year={2019} + } + ''' + x = bicubic_degradation(x, sf=sf) + x = ndimage.convolve(x, np.expand_dims(k, axis=2), mode='wrap') + return x + + +def classical_degradation(x, k, sf=3): + ''' blur + downsampling + Args: + x: HxWxC image, [0, 1]/[0, 255] + k: hxw, double + sf: down-scale factor + Return: + downsampled LR image + ''' + x = ndimage.convolve(x, np.expand_dims(k, axis=2), mode='wrap') + # x = filters.correlate(x, np.expand_dims(np.flip(k), axis=2)) + st = 0 + return x[st::sf, st::sf, ...] + + +def add_sharpening(img, weight=0.5, radius=50, threshold=10): + """USM sharpening. borrowed from real-ESRGAN + Input image: I; Blurry image: B. + 1. K = I + weight * (I - B) + 2. Mask = 1 if abs(I - B) > threshold, else: 0 + 3. Blur mask: + 4. Out = Mask * K + (1 - Mask) * I + Args: + img (Numpy array): Input image, HWC, BGR; float32, [0, 1]. + weight (float): Sharp weight. Default: 1. + radius (float): Kernel size of Gaussian blur. Default: 50. + threshold (int): + """ + if radius % 2 == 0: + radius += 1 + blur = cv2.GaussianBlur(img, (radius, radius), 0) + residual = img - blur + mask = np.abs(residual) * 255 > threshold + mask = mask.astype('float32') + soft_mask = cv2.GaussianBlur(mask, (radius, radius), 0) + + K = img + weight * residual + K = np.clip(K, 0, 1) + return soft_mask * K + (1 - soft_mask) * img + + +def add_blur(img, sf=4): + wd2 = 4.0 + sf + wd = 2.0 + 0.2 * sf + + wd2 = wd2/4 + wd = wd/4 + + if random.random() < 0.5: + l1 = wd2 * random.random() + l2 = wd2 * random.random() + k = anisotropic_Gaussian(ksize=random.randint(2, 11) + 3, theta=random.random() * np.pi, l1=l1, l2=l2) + else: + k = fspecial('gaussian', random.randint(2, 4) + 3, wd * random.random()) + img = ndimage.convolve(img, np.expand_dims(k, axis=2), mode='mirror') + + return img + + +def add_resize(img, sf=4): + rnum = np.random.rand() + if rnum > 0.8: # up + sf1 = random.uniform(1, 2) + elif rnum < 0.7: # down + sf1 = random.uniform(0.5 / sf, 1) + else: + sf1 = 1.0 + img = cv2.resize(img, (int(sf1 * img.shape[1]), int(sf1 * img.shape[0])), interpolation=random.choice([1, 2, 3])) + img = np.clip(img, 0.0, 1.0) + + return img + + +# def add_Gaussian_noise(img, noise_level1=2, noise_level2=25): +# noise_level = random.randint(noise_level1, noise_level2) +# rnum = np.random.rand() +# if rnum > 0.6: # add color Gaussian noise +# img += np.random.normal(0, noise_level / 255.0, img.shape).astype(np.float32) +# elif rnum < 0.4: # add grayscale Gaussian noise +# img += np.random.normal(0, noise_level / 255.0, (*img.shape[:2], 1)).astype(np.float32) +# else: # add noise +# L = noise_level2 / 255. +# D = np.diag(np.random.rand(3)) +# U = orth(np.random.rand(3, 3)) +# conv = np.dot(np.dot(np.transpose(U), D), U) +# img += np.random.multivariate_normal([0, 0, 0], np.abs(L ** 2 * conv), img.shape[:2]).astype(np.float32) +# img = np.clip(img, 0.0, 1.0) +# return img + +def add_Gaussian_noise(img, noise_level1=2, noise_level2=25): + noise_level = random.randint(noise_level1, noise_level2) + rnum = np.random.rand() + if rnum > 0.6: # add color Gaussian noise + img = img + np.random.normal(0, noise_level / 255.0, img.shape).astype(np.float32) + elif rnum < 0.4: # add grayscale Gaussian noise + img = img + np.random.normal(0, noise_level / 255.0, (*img.shape[:2], 1)).astype(np.float32) + else: # add noise + L = noise_level2 / 255. + D = np.diag(np.random.rand(3)) + U = orth(np.random.rand(3, 3)) + conv = np.dot(np.dot(np.transpose(U), D), U) + img = img + np.random.multivariate_normal([0, 0, 0], np.abs(L ** 2 * conv), img.shape[:2]).astype(np.float32) + img = np.clip(img, 0.0, 1.0) + return img + + +def add_speckle_noise(img, noise_level1=2, noise_level2=25): + noise_level = random.randint(noise_level1, noise_level2) + img = np.clip(img, 0.0, 1.0) + rnum = random.random() + if rnum > 0.6: + img += img * np.random.normal(0, noise_level / 255.0, img.shape).astype(np.float32) + elif rnum < 0.4: + img += img * np.random.normal(0, noise_level / 255.0, (*img.shape[:2], 1)).astype(np.float32) + else: + L = noise_level2 / 255. + D = np.diag(np.random.rand(3)) + U = orth(np.random.rand(3, 3)) + conv = np.dot(np.dot(np.transpose(U), D), U) + img += img * np.random.multivariate_normal([0, 0, 0], np.abs(L ** 2 * conv), img.shape[:2]).astype(np.float32) + img = np.clip(img, 0.0, 1.0) + return img + + +def add_Poisson_noise(img): + img = np.clip((img * 255.0).round(), 0, 255) / 255. + vals = 10 ** (2 * random.random() + 2.0) # [2, 4] + if random.random() < 0.5: + img = np.random.poisson(img * vals).astype(np.float32) / vals + else: + img_gray = np.dot(img[..., :3], [0.299, 0.587, 0.114]) + img_gray = np.clip((img_gray * 255.0).round(), 0, 255) / 255. + noise_gray = np.random.poisson(img_gray * vals).astype(np.float32) / vals - img_gray + img += noise_gray[:, :, np.newaxis] + img = np.clip(img, 0.0, 1.0) + return img + + +def add_JPEG_noise(img): + quality_factor = random.randint(80, 95) + img = cv2.cvtColor(util.single2uint(img), cv2.COLOR_RGB2BGR) + result, encimg = cv2.imencode('.jpg', img, [int(cv2.IMWRITE_JPEG_QUALITY), quality_factor]) + img = cv2.imdecode(encimg, 1) + img = cv2.cvtColor(util.uint2single(img), cv2.COLOR_BGR2RGB) + return img + + +def random_crop(lq, hq, sf=4, lq_patchsize=64): + h, w = lq.shape[:2] + rnd_h = random.randint(0, h - lq_patchsize) + rnd_w = random.randint(0, w - lq_patchsize) + lq = lq[rnd_h:rnd_h + lq_patchsize, rnd_w:rnd_w + lq_patchsize, :] + + rnd_h_H, rnd_w_H = int(rnd_h * sf), int(rnd_w * sf) + hq = hq[rnd_h_H:rnd_h_H + lq_patchsize * sf, rnd_w_H:rnd_w_H + lq_patchsize * sf, :] + return lq, hq + + +def degradation_bsrgan(img, sf=4, lq_patchsize=72, isp_model=None): + """ + This is the degradation model of BSRGAN from the paper + "Designing a Practical Degradation Model for Deep Blind Image Super-Resolution" + ---------- + img: HXWXC, [0, 1], its size should be large than (lq_patchsizexsf)x(lq_patchsizexsf) + sf: scale factor + isp_model: camera ISP model + Returns + ------- + img: low-quality patch, size: lq_patchsizeXlq_patchsizeXC, range: [0, 1] + hq: corresponding high-quality patch, size: (lq_patchsizexsf)X(lq_patchsizexsf)XC, range: [0, 1] + """ + isp_prob, jpeg_prob, scale2_prob = 0.25, 0.9, 0.25 + sf_ori = sf + + h1, w1 = img.shape[:2] + img = img.copy()[:w1 - w1 % sf, :h1 - h1 % sf, ...] # mod crop + h, w = img.shape[:2] + + if h < lq_patchsize * sf or w < lq_patchsize * sf: + raise ValueError(f'img size ({h1}X{w1}) is too small!') + + hq = img.copy() + + if sf == 4 and random.random() < scale2_prob: # downsample1 + if np.random.rand() < 0.5: + img = cv2.resize(img, (int(1 / 2 * img.shape[1]), int(1 / 2 * img.shape[0])), + interpolation=random.choice([1, 2, 3])) + else: + img = util.imresize_np(img, 1 / 2, True) + img = np.clip(img, 0.0, 1.0) + sf = 2 + + shuffle_order = random.sample(range(7), 7) + idx1, idx2 = shuffle_order.index(2), shuffle_order.index(3) + if idx1 > idx2: # keep downsample3 last + shuffle_order[idx1], shuffle_order[idx2] = shuffle_order[idx2], shuffle_order[idx1] + + for i in shuffle_order: + + if i == 0: + img = add_blur(img, sf=sf) + + elif i == 1: + img = add_blur(img, sf=sf) + + elif i == 2: + a, b = img.shape[1], img.shape[0] + # downsample2 + if random.random() < 0.75: + sf1 = random.uniform(1, 2 * sf) + img = cv2.resize(img, (int(1 / sf1 * img.shape[1]), int(1 / sf1 * img.shape[0])), + interpolation=random.choice([1, 2, 3])) + else: + k = fspecial('gaussian', 25, random.uniform(0.1, 0.6 * sf)) + k_shifted = shift_pixel(k, sf) + k_shifted = k_shifted / k_shifted.sum() # blur with shifted kernel + img = ndimage.convolve(img, np.expand_dims(k_shifted, axis=2), mode='mirror') + img = img[0::sf, 0::sf, ...] # nearest downsampling + img = np.clip(img, 0.0, 1.0) + + elif i == 3: + # downsample3 + img = cv2.resize(img, (int(1 / sf * a), int(1 / sf * b)), interpolation=random.choice([1, 2, 3])) + img = np.clip(img, 0.0, 1.0) + + elif i == 4: + # add Gaussian noise + img = add_Gaussian_noise(img, noise_level1=2, noise_level2=8) + + elif i == 5: + # add JPEG noise + if random.random() < jpeg_prob: + img = add_JPEG_noise(img) + + elif i == 6: + # add processed camera sensor noise + if random.random() < isp_prob and isp_model is not None: + with torch.no_grad(): + img, hq = isp_model.forward(img.copy(), hq) + + # add final JPEG compression noise + img = add_JPEG_noise(img) + + # random crop + img, hq = random_crop(img, hq, sf_ori, lq_patchsize) + + return img, hq + + +# todo no isp_model? +def degradation_bsrgan_variant(image, sf=4, isp_model=None, up=False): + """ + This is the degradation model of BSRGAN from the paper + "Designing a Practical Degradation Model for Deep Blind Image Super-Resolution" + ---------- + sf: scale factor + isp_model: camera ISP model + Returns + ------- + img: low-quality patch, size: lq_patchsizeXlq_patchsizeXC, range: [0, 1] + hq: corresponding high-quality patch, size: (lq_patchsizexsf)X(lq_patchsizexsf)XC, range: [0, 1] + """ + image = util.uint2single(image) + isp_prob, jpeg_prob, scale2_prob = 0.25, 0.9, 0.25 + sf_ori = sf + + h1, w1 = image.shape[:2] + image = image.copy()[:w1 - w1 % sf, :h1 - h1 % sf, ...] # mod crop + h, w = image.shape[:2] + + hq = image.copy() + + if sf == 4 and random.random() < scale2_prob: # downsample1 + if np.random.rand() < 0.5: + image = cv2.resize(image, (int(1 / 2 * image.shape[1]), int(1 / 2 * image.shape[0])), + interpolation=random.choice([1, 2, 3])) + else: + image = util.imresize_np(image, 1 / 2, True) + image = np.clip(image, 0.0, 1.0) + sf = 2 + + shuffle_order = random.sample(range(7), 7) + idx1, idx2 = shuffle_order.index(2), shuffle_order.index(3) + if idx1 > idx2: # keep downsample3 last + shuffle_order[idx1], shuffle_order[idx2] = shuffle_order[idx2], shuffle_order[idx1] + + for i in shuffle_order: + + if i == 0: + image = add_blur(image, sf=sf) + + # elif i == 1: + # image = add_blur(image, sf=sf) + + if i == 0: + pass + + elif i == 2: + a, b = image.shape[1], image.shape[0] + # downsample2 + if random.random() < 0.8: + sf1 = random.uniform(1, 2 * sf) + image = cv2.resize(image, (int(1 / sf1 * image.shape[1]), int(1 / sf1 * image.shape[0])), + interpolation=random.choice([1, 2, 3])) + else: + k = fspecial('gaussian', 25, random.uniform(0.1, 0.6 * sf)) + k_shifted = shift_pixel(k, sf) + k_shifted = k_shifted / k_shifted.sum() # blur with shifted kernel + image = ndimage.convolve(image, np.expand_dims(k_shifted, axis=2), mode='mirror') + image = image[0::sf, 0::sf, ...] # nearest downsampling + + image = np.clip(image, 0.0, 1.0) + + elif i == 3: + # downsample3 + image = cv2.resize(image, (int(1 / sf * a), int(1 / sf * b)), interpolation=random.choice([1, 2, 3])) + image = np.clip(image, 0.0, 1.0) + + elif i == 4: + # add Gaussian noise + image = add_Gaussian_noise(image, noise_level1=1, noise_level2=2) + + elif i == 5: + # add JPEG noise + if random.random() < jpeg_prob: + image = add_JPEG_noise(image) + # + # elif i == 6: + # # add processed camera sensor noise + # if random.random() < isp_prob and isp_model is not None: + # with torch.no_grad(): + # img, hq = isp_model.forward(img.copy(), hq) + + # add final JPEG compression noise + image = add_JPEG_noise(image) + image = util.single2uint(image) + if up: + image = cv2.resize(image, (w1, h1), interpolation=cv2.INTER_CUBIC) # todo: random, as above? want to condition on it then + example = {"image": image} + return example + + + + +if __name__ == '__main__': + print("hey") + img = util.imread_uint('utils/test.png', 3) + img = img[:448, :448] + h = img.shape[0] // 4 + print("resizing to", h) + sf = 4 + deg_fn = partial(degradation_bsrgan_variant, sf=sf) + for i in range(20): + print(i) + img_hq = img + img_lq = deg_fn(img)["image"] + img_hq, img_lq = util.uint2single(img_hq), util.uint2single(img_lq) + print(img_lq) + img_lq_bicubic = albumentations.SmallestMaxSize(max_size=h, interpolation=cv2.INTER_CUBIC)(image=img_hq)["image"] + print(img_lq.shape) + print("bicubic", img_lq_bicubic.shape) + print(img_hq.shape) + lq_nearest = cv2.resize(util.single2uint(img_lq), (int(sf * img_lq.shape[1]), int(sf * img_lq.shape[0])), + interpolation=0) + lq_bicubic_nearest = cv2.resize(util.single2uint(img_lq_bicubic), + (int(sf * img_lq.shape[1]), int(sf * img_lq.shape[0])), + interpolation=0) + img_concat = np.concatenate([lq_bicubic_nearest, lq_nearest, util.single2uint(img_hq)], axis=1) + util.imsave(img_concat, str(i) + '.png') diff --git a/ldm/modules/image_degradation/utils/test.png b/ldm/modules/image_degradation/utils/test.png new file mode 100644 index 0000000000000000000000000000000000000000..4249b43de0f22707758d13c240268a401642f6e6 Binary files /dev/null and b/ldm/modules/image_degradation/utils/test.png differ diff --git a/ldm/modules/image_degradation/utils_image.py b/ldm/modules/image_degradation/utils_image.py new file mode 100644 index 0000000000000000000000000000000000000000..0175f155ad900ae33c3c46ed87f49b352e3faf98 --- /dev/null +++ b/ldm/modules/image_degradation/utils_image.py @@ -0,0 +1,916 @@ +import os +import math +import random +import numpy as np +import torch +import cv2 +from torchvision.utils import make_grid +from datetime import datetime +#import matplotlib.pyplot as plt # TODO: check with Dominik, also bsrgan.py vs bsrgan_light.py + + +os.environ["KMP_DUPLICATE_LIB_OK"]="TRUE" + + +''' +# -------------------------------------------- +# Kai Zhang (github: https://github.com/cszn) +# 03/Mar/2019 +# -------------------------------------------- +# https://github.com/twhui/SRGAN-pyTorch +# https://github.com/xinntao/BasicSR +# -------------------------------------------- +''' + + +IMG_EXTENSIONS = ['.jpg', '.JPG', '.jpeg', '.JPEG', '.png', '.PNG', '.ppm', '.PPM', '.bmp', '.BMP', '.tif'] + + +def is_image_file(filename): + return any(filename.endswith(extension) for extension in IMG_EXTENSIONS) + + +def get_timestamp(): + return datetime.now().strftime('%y%m%d-%H%M%S') + + +def imshow(x, title=None, cbar=False, figsize=None): + plt.figure(figsize=figsize) + plt.imshow(np.squeeze(x), interpolation='nearest', cmap='gray') + if title: + plt.title(title) + if cbar: + plt.colorbar() + plt.show() + + +def surf(Z, cmap='rainbow', figsize=None): + plt.figure(figsize=figsize) + ax3 = plt.axes(projection='3d') + + w, h = Z.shape[:2] + xx = np.arange(0,w,1) + yy = np.arange(0,h,1) + X, Y = np.meshgrid(xx, yy) + ax3.plot_surface(X,Y,Z,cmap=cmap) + #ax3.contour(X,Y,Z, zdim='z',offset=-2,cmap=cmap) + plt.show() + + +''' +# -------------------------------------------- +# get image pathes +# -------------------------------------------- +''' + + +def get_image_paths(dataroot): + paths = None # return None if dataroot is None + if dataroot is not None: + paths = sorted(_get_paths_from_images(dataroot)) + return paths + + +def _get_paths_from_images(path): + assert os.path.isdir(path), '{:s} is not a valid directory'.format(path) + images = [] + for dirpath, _, fnames in sorted(os.walk(path)): + for fname in sorted(fnames): + if is_image_file(fname): + img_path = os.path.join(dirpath, fname) + images.append(img_path) + assert images, '{:s} has no valid image file'.format(path) + return images + + +''' +# -------------------------------------------- +# split large images into small images +# -------------------------------------------- +''' + + +def patches_from_image(img, p_size=512, p_overlap=64, p_max=800): + w, h = img.shape[:2] + patches = [] + if w > p_max and h > p_max: + w1 = list(np.arange(0, w-p_size, p_size-p_overlap, dtype=np.int)) + h1 = list(np.arange(0, h-p_size, p_size-p_overlap, dtype=np.int)) + w1.append(w-p_size) + h1.append(h-p_size) +# print(w1) +# print(h1) + for i in w1: + for j in h1: + patches.append(img[i:i+p_size, j:j+p_size,:]) + else: + patches.append(img) + + return patches + + +def imssave(imgs, img_path): + """ + imgs: list, N images of size WxHxC + """ + img_name, ext = os.path.splitext(os.path.basename(img_path)) + + for i, img in enumerate(imgs): + if img.ndim == 3: + img = img[:, :, [2, 1, 0]] + new_path = os.path.join(os.path.dirname(img_path), img_name+str('_s{:04d}'.format(i))+'.png') + cv2.imwrite(new_path, img) + + +def split_imageset(original_dataroot, taget_dataroot, n_channels=3, p_size=800, p_overlap=96, p_max=1000): + """ + split the large images from original_dataroot into small overlapped images with size (p_size)x(p_size), + and save them into taget_dataroot; only the images with larger size than (p_max)x(p_max) + will be splitted. + Args: + original_dataroot: + taget_dataroot: + p_size: size of small images + p_overlap: patch size in training is a good choice + p_max: images with smaller size than (p_max)x(p_max) keep unchanged. + """ + paths = get_image_paths(original_dataroot) + for img_path in paths: + # img_name, ext = os.path.splitext(os.path.basename(img_path)) + img = imread_uint(img_path, n_channels=n_channels) + patches = patches_from_image(img, p_size, p_overlap, p_max) + imssave(patches, os.path.join(taget_dataroot,os.path.basename(img_path))) + #if original_dataroot == taget_dataroot: + #del img_path + +''' +# -------------------------------------------- +# makedir +# -------------------------------------------- +''' + + +def mkdir(path): + if not os.path.exists(path): + os.makedirs(path) + + +def mkdirs(paths): + if isinstance(paths, str): + mkdir(paths) + else: + for path in paths: + mkdir(path) + + +def mkdir_and_rename(path): + if os.path.exists(path): + new_name = path + '_archived_' + get_timestamp() + print('Path already exists. Rename it to [{:s}]'.format(new_name)) + os.rename(path, new_name) + os.makedirs(path) + + +''' +# -------------------------------------------- +# read image from path +# opencv is fast, but read BGR numpy image +# -------------------------------------------- +''' + + +# -------------------------------------------- +# get uint8 image of size HxWxn_channles (RGB) +# -------------------------------------------- +def imread_uint(path, n_channels=3): + # input: path + # output: HxWx3(RGB or GGG), or HxWx1 (G) + if n_channels == 1: + img = cv2.imread(path, 0) # cv2.IMREAD_GRAYSCALE + img = np.expand_dims(img, axis=2) # HxWx1 + elif n_channels == 3: + img = cv2.imread(path, cv2.IMREAD_UNCHANGED) # BGR or G + if img.ndim == 2: + img = cv2.cvtColor(img, cv2.COLOR_GRAY2RGB) # GGG + else: + img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB) # RGB + return img + + +# -------------------------------------------- +# matlab's imwrite +# -------------------------------------------- +def imsave(img, img_path): + img = np.squeeze(img) + if img.ndim == 3: + img = img[:, :, [2, 1, 0]] + cv2.imwrite(img_path, img) + +def imwrite(img, img_path): + img = np.squeeze(img) + if img.ndim == 3: + img = img[:, :, [2, 1, 0]] + cv2.imwrite(img_path, img) + + + +# -------------------------------------------- +# get single image of size HxWxn_channles (BGR) +# -------------------------------------------- +def read_img(path): + # read image by cv2 + # return: Numpy float32, HWC, BGR, [0,1] + img = cv2.imread(path, cv2.IMREAD_UNCHANGED) # cv2.IMREAD_GRAYSCALE + img = img.astype(np.float32) / 255. + if img.ndim == 2: + img = np.expand_dims(img, axis=2) + # some images have 4 channels + if img.shape[2] > 3: + img = img[:, :, :3] + return img + + +''' +# -------------------------------------------- +# image format conversion +# -------------------------------------------- +# numpy(single) <---> numpy(unit) +# numpy(single) <---> tensor +# numpy(unit) <---> tensor +# -------------------------------------------- +''' + + +# -------------------------------------------- +# numpy(single) [0, 1] <---> numpy(unit) +# -------------------------------------------- + + +def uint2single(img): + + return np.float32(img/255.) + + +def single2uint(img): + + return np.uint8((img.clip(0, 1)*255.).round()) + + +def uint162single(img): + + return np.float32(img/65535.) + + +def single2uint16(img): + + return np.uint16((img.clip(0, 1)*65535.).round()) + + +# -------------------------------------------- +# numpy(unit) (HxWxC or HxW) <---> tensor +# -------------------------------------------- + + +# convert uint to 4-dimensional torch tensor +def uint2tensor4(img): + if img.ndim == 2: + img = np.expand_dims(img, axis=2) + return torch.from_numpy(np.ascontiguousarray(img)).permute(2, 0, 1).float().div(255.).unsqueeze(0) + + +# convert uint to 3-dimensional torch tensor +def uint2tensor3(img): + if img.ndim == 2: + img = np.expand_dims(img, axis=2) + return torch.from_numpy(np.ascontiguousarray(img)).permute(2, 0, 1).float().div(255.) + + +# convert 2/3/4-dimensional torch tensor to uint +def tensor2uint(img): + img = img.data.squeeze().float().clamp_(0, 1).cpu().numpy() + if img.ndim == 3: + img = np.transpose(img, (1, 2, 0)) + return np.uint8((img*255.0).round()) + + +# -------------------------------------------- +# numpy(single) (HxWxC) <---> tensor +# -------------------------------------------- + + +# convert single (HxWxC) to 3-dimensional torch tensor +def single2tensor3(img): + return torch.from_numpy(np.ascontiguousarray(img)).permute(2, 0, 1).float() + + +# convert single (HxWxC) to 4-dimensional torch tensor +def single2tensor4(img): + return torch.from_numpy(np.ascontiguousarray(img)).permute(2, 0, 1).float().unsqueeze(0) + + +# convert torch tensor to single +def tensor2single(img): + img = img.data.squeeze().float().cpu().numpy() + if img.ndim == 3: + img = np.transpose(img, (1, 2, 0)) + + return img + +# convert torch tensor to single +def tensor2single3(img): + img = img.data.squeeze().float().cpu().numpy() + if img.ndim == 3: + img = np.transpose(img, (1, 2, 0)) + elif img.ndim == 2: + img = np.expand_dims(img, axis=2) + return img + + +def single2tensor5(img): + return torch.from_numpy(np.ascontiguousarray(img)).permute(2, 0, 1, 3).float().unsqueeze(0) + + +def single32tensor5(img): + return torch.from_numpy(np.ascontiguousarray(img)).float().unsqueeze(0).unsqueeze(0) + + +def single42tensor4(img): + return torch.from_numpy(np.ascontiguousarray(img)).permute(2, 0, 1, 3).float() + + +# from skimage.io import imread, imsave +def tensor2img(tensor, out_type=np.uint8, min_max=(0, 1)): + ''' + Converts a torch Tensor into an image Numpy array of BGR channel order + Input: 4D(B,(3/1),H,W), 3D(C,H,W), or 2D(H,W), any range, RGB channel order + Output: 3D(H,W,C) or 2D(H,W), [0,255], np.uint8 (default) + ''' + tensor = tensor.squeeze().float().cpu().clamp_(*min_max) # squeeze first, then clamp + tensor = (tensor - min_max[0]) / (min_max[1] - min_max[0]) # to range [0,1] + n_dim = tensor.dim() + if n_dim == 4: + n_img = len(tensor) + img_np = make_grid(tensor, nrow=int(math.sqrt(n_img)), normalize=False).numpy() + img_np = np.transpose(img_np[[2, 1, 0], :, :], (1, 2, 0)) # HWC, BGR + elif n_dim == 3: + img_np = tensor.numpy() + img_np = np.transpose(img_np[[2, 1, 0], :, :], (1, 2, 0)) # HWC, BGR + elif n_dim == 2: + img_np = tensor.numpy() + else: + raise TypeError( + 'Only support 4D, 3D and 2D tensor. But received with dimension: {:d}'.format(n_dim)) + if out_type == np.uint8: + img_np = (img_np * 255.0).round() + # Important. Unlike matlab, numpy.unit8() WILL NOT round by default. + return img_np.astype(out_type) + + +''' +# -------------------------------------------- +# Augmentation, flipe and/or rotate +# -------------------------------------------- +# The following two are enough. +# (1) augmet_img: numpy image of WxHxC or WxH +# (2) augment_img_tensor4: tensor image 1xCxWxH +# -------------------------------------------- +''' + + +def augment_img(img, mode=0): + '''Kai Zhang (github: https://github.com/cszn) + ''' + if mode == 0: + return img + elif mode == 1: + return np.flipud(np.rot90(img)) + elif mode == 2: + return np.flipud(img) + elif mode == 3: + return np.rot90(img, k=3) + elif mode == 4: + return np.flipud(np.rot90(img, k=2)) + elif mode == 5: + return np.rot90(img) + elif mode == 6: + return np.rot90(img, k=2) + elif mode == 7: + return np.flipud(np.rot90(img, k=3)) + + +def augment_img_tensor4(img, mode=0): + '''Kai Zhang (github: https://github.com/cszn) + ''' + if mode == 0: + return img + elif mode == 1: + return img.rot90(1, [2, 3]).flip([2]) + elif mode == 2: + return img.flip([2]) + elif mode == 3: + return img.rot90(3, [2, 3]) + elif mode == 4: + return img.rot90(2, [2, 3]).flip([2]) + elif mode == 5: + return img.rot90(1, [2, 3]) + elif mode == 6: + return img.rot90(2, [2, 3]) + elif mode == 7: + return img.rot90(3, [2, 3]).flip([2]) + + +def augment_img_tensor(img, mode=0): + '''Kai Zhang (github: https://github.com/cszn) + ''' + img_size = img.size() + img_np = img.data.cpu().numpy() + if len(img_size) == 3: + img_np = np.transpose(img_np, (1, 2, 0)) + elif len(img_size) == 4: + img_np = np.transpose(img_np, (2, 3, 1, 0)) + img_np = augment_img(img_np, mode=mode) + img_tensor = torch.from_numpy(np.ascontiguousarray(img_np)) + if len(img_size) == 3: + img_tensor = img_tensor.permute(2, 0, 1) + elif len(img_size) == 4: + img_tensor = img_tensor.permute(3, 2, 0, 1) + + return img_tensor.type_as(img) + + +def augment_img_np3(img, mode=0): + if mode == 0: + return img + elif mode == 1: + return img.transpose(1, 0, 2) + elif mode == 2: + return img[::-1, :, :] + elif mode == 3: + img = img[::-1, :, :] + img = img.transpose(1, 0, 2) + return img + elif mode == 4: + return img[:, ::-1, :] + elif mode == 5: + img = img[:, ::-1, :] + img = img.transpose(1, 0, 2) + return img + elif mode == 6: + img = img[:, ::-1, :] + img = img[::-1, :, :] + return img + elif mode == 7: + img = img[:, ::-1, :] + img = img[::-1, :, :] + img = img.transpose(1, 0, 2) + return img + + +def augment_imgs(img_list, hflip=True, rot=True): + # horizontal flip OR rotate + hflip = hflip and random.random() < 0.5 + vflip = rot and random.random() < 0.5 + rot90 = rot and random.random() < 0.5 + + def _augment(img): + if hflip: + img = img[:, ::-1, :] + if vflip: + img = img[::-1, :, :] + if rot90: + img = img.transpose(1, 0, 2) + return img + + return [_augment(img) for img in img_list] + + +''' +# -------------------------------------------- +# modcrop and shave +# -------------------------------------------- +''' + + +def modcrop(img_in, scale): + # img_in: Numpy, HWC or HW + img = np.copy(img_in) + if img.ndim == 2: + H, W = img.shape + H_r, W_r = H % scale, W % scale + img = img[:H - H_r, :W - W_r] + elif img.ndim == 3: + H, W, C = img.shape + H_r, W_r = H % scale, W % scale + img = img[:H - H_r, :W - W_r, :] + else: + raise ValueError('Wrong img ndim: [{:d}].'.format(img.ndim)) + return img + + +def shave(img_in, border=0): + # img_in: Numpy, HWC or HW + img = np.copy(img_in) + h, w = img.shape[:2] + img = img[border:h-border, border:w-border] + return img + + +''' +# -------------------------------------------- +# image processing process on numpy image +# channel_convert(in_c, tar_type, img_list): +# rgb2ycbcr(img, only_y=True): +# bgr2ycbcr(img, only_y=True): +# ycbcr2rgb(img): +# -------------------------------------------- +''' + + +def rgb2ycbcr(img, only_y=True): + '''same as matlab rgb2ycbcr + only_y: only return Y channel + Input: + uint8, [0, 255] + float, [0, 1] + ''' + in_img_type = img.dtype + img.astype(np.float32) + if in_img_type != np.uint8: + img *= 255. + # convert + if only_y: + rlt = np.dot(img, [65.481, 128.553, 24.966]) / 255.0 + 16.0 + else: + rlt = np.matmul(img, [[65.481, -37.797, 112.0], [128.553, -74.203, -93.786], + [24.966, 112.0, -18.214]]) / 255.0 + [16, 128, 128] + if in_img_type == np.uint8: + rlt = rlt.round() + else: + rlt /= 255. + return rlt.astype(in_img_type) + + +def ycbcr2rgb(img): + '''same as matlab ycbcr2rgb + Input: + uint8, [0, 255] + float, [0, 1] + ''' + in_img_type = img.dtype + img.astype(np.float32) + if in_img_type != np.uint8: + img *= 255. + # convert + rlt = np.matmul(img, [[0.00456621, 0.00456621, 0.00456621], [0, -0.00153632, 0.00791071], + [0.00625893, -0.00318811, 0]]) * 255.0 + [-222.921, 135.576, -276.836] + if in_img_type == np.uint8: + rlt = rlt.round() + else: + rlt /= 255. + return rlt.astype(in_img_type) + + +def bgr2ycbcr(img, only_y=True): + '''bgr version of rgb2ycbcr + only_y: only return Y channel + Input: + uint8, [0, 255] + float, [0, 1] + ''' + in_img_type = img.dtype + img.astype(np.float32) + if in_img_type != np.uint8: + img *= 255. + # convert + if only_y: + rlt = np.dot(img, [24.966, 128.553, 65.481]) / 255.0 + 16.0 + else: + rlt = np.matmul(img, [[24.966, 112.0, -18.214], [128.553, -74.203, -93.786], + [65.481, -37.797, 112.0]]) / 255.0 + [16, 128, 128] + if in_img_type == np.uint8: + rlt = rlt.round() + else: + rlt /= 255. + return rlt.astype(in_img_type) + + +def channel_convert(in_c, tar_type, img_list): + # conversion among BGR, gray and y + if in_c == 3 and tar_type == 'gray': # BGR to gray + gray_list = [cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) for img in img_list] + return [np.expand_dims(img, axis=2) for img in gray_list] + elif in_c == 3 and tar_type == 'y': # BGR to y + y_list = [bgr2ycbcr(img, only_y=True) for img in img_list] + return [np.expand_dims(img, axis=2) for img in y_list] + elif in_c == 1 and tar_type == 'RGB': # gray/y to BGR + return [cv2.cvtColor(img, cv2.COLOR_GRAY2BGR) for img in img_list] + else: + return img_list + + +''' +# -------------------------------------------- +# metric, PSNR and SSIM +# -------------------------------------------- +''' + + +# -------------------------------------------- +# PSNR +# -------------------------------------------- +def calculate_psnr(img1, img2, border=0): + # img1 and img2 have range [0, 255] + #img1 = img1.squeeze() + #img2 = img2.squeeze() + if not img1.shape == img2.shape: + raise ValueError('Input images must have the same dimensions.') + h, w = img1.shape[:2] + img1 = img1[border:h-border, border:w-border] + img2 = img2[border:h-border, border:w-border] + + img1 = img1.astype(np.float64) + img2 = img2.astype(np.float64) + mse = np.mean((img1 - img2)**2) + if mse == 0: + return float('inf') + return 20 * math.log10(255.0 / math.sqrt(mse)) + + +# -------------------------------------------- +# SSIM +# -------------------------------------------- +def calculate_ssim(img1, img2, border=0): + '''calculate SSIM + the same outputs as MATLAB's + img1, img2: [0, 255] + ''' + #img1 = img1.squeeze() + #img2 = img2.squeeze() + if not img1.shape == img2.shape: + raise ValueError('Input images must have the same dimensions.') + h, w = img1.shape[:2] + img1 = img1[border:h-border, border:w-border] + img2 = img2[border:h-border, border:w-border] + + if img1.ndim == 2: + return ssim(img1, img2) + elif img1.ndim == 3: + if img1.shape[2] == 3: + ssims = [] + for i in range(3): + ssims.append(ssim(img1[:,:,i], img2[:,:,i])) + return np.array(ssims).mean() + elif img1.shape[2] == 1: + return ssim(np.squeeze(img1), np.squeeze(img2)) + else: + raise ValueError('Wrong input image dimensions.') + + +def ssim(img1, img2): + C1 = (0.01 * 255)**2 + C2 = (0.03 * 255)**2 + + img1 = img1.astype(np.float64) + img2 = img2.astype(np.float64) + kernel = cv2.getGaussianKernel(11, 1.5) + window = np.outer(kernel, kernel.transpose()) + + mu1 = cv2.filter2D(img1, -1, window)[5:-5, 5:-5] # valid + mu2 = cv2.filter2D(img2, -1, window)[5:-5, 5:-5] + mu1_sq = mu1**2 + mu2_sq = mu2**2 + mu1_mu2 = mu1 * mu2 + sigma1_sq = cv2.filter2D(img1**2, -1, window)[5:-5, 5:-5] - mu1_sq + sigma2_sq = cv2.filter2D(img2**2, -1, window)[5:-5, 5:-5] - mu2_sq + sigma12 = cv2.filter2D(img1 * img2, -1, window)[5:-5, 5:-5] - mu1_mu2 + + ssim_map = ((2 * mu1_mu2 + C1) * (2 * sigma12 + C2)) / ((mu1_sq + mu2_sq + C1) * + (sigma1_sq + sigma2_sq + C2)) + return ssim_map.mean() + + +''' +# -------------------------------------------- +# matlab's bicubic imresize (numpy and torch) [0, 1] +# -------------------------------------------- +''' + + +# matlab 'imresize' function, now only support 'bicubic' +def cubic(x): + absx = torch.abs(x) + absx2 = absx**2 + absx3 = absx**3 + return (1.5*absx3 - 2.5*absx2 + 1) * ((absx <= 1).type_as(absx)) + \ + (-0.5*absx3 + 2.5*absx2 - 4*absx + 2) * (((absx > 1)*(absx <= 2)).type_as(absx)) + + +def calculate_weights_indices(in_length, out_length, scale, kernel, kernel_width, antialiasing): + if (scale < 1) and (antialiasing): + # Use a modified kernel to simultaneously interpolate and antialias- larger kernel width + kernel_width = kernel_width / scale + + # Output-space coordinates + x = torch.linspace(1, out_length, out_length) + + # Input-space coordinates. Calculate the inverse mapping such that 0.5 + # in output space maps to 0.5 in input space, and 0.5+scale in output + # space maps to 1.5 in input space. + u = x / scale + 0.5 * (1 - 1 / scale) + + # What is the left-most pixel that can be involved in the computation? + left = torch.floor(u - kernel_width / 2) + + # What is the maximum number of pixels that can be involved in the + # computation? Note: it's OK to use an extra pixel here; if the + # corresponding weights are all zero, it will be eliminated at the end + # of this function. + P = math.ceil(kernel_width) + 2 + + # The indices of the input pixels involved in computing the k-th output + # pixel are in row k of the indices matrix. + indices = left.view(out_length, 1).expand(out_length, P) + torch.linspace(0, P - 1, P).view( + 1, P).expand(out_length, P) + + # The weights used to compute the k-th output pixel are in row k of the + # weights matrix. + distance_to_center = u.view(out_length, 1).expand(out_length, P) - indices + # apply cubic kernel + if (scale < 1) and (antialiasing): + weights = scale * cubic(distance_to_center * scale) + else: + weights = cubic(distance_to_center) + # Normalize the weights matrix so that each row sums to 1. + weights_sum = torch.sum(weights, 1).view(out_length, 1) + weights = weights / weights_sum.expand(out_length, P) + + # If a column in weights is all zero, get rid of it. only consider the first and last column. + weights_zero_tmp = torch.sum((weights == 0), 0) + if not math.isclose(weights_zero_tmp[0], 0, rel_tol=1e-6): + indices = indices.narrow(1, 1, P - 2) + weights = weights.narrow(1, 1, P - 2) + if not math.isclose(weights_zero_tmp[-1], 0, rel_tol=1e-6): + indices = indices.narrow(1, 0, P - 2) + weights = weights.narrow(1, 0, P - 2) + weights = weights.contiguous() + indices = indices.contiguous() + sym_len_s = -indices.min() + 1 + sym_len_e = indices.max() - in_length + indices = indices + sym_len_s - 1 + return weights, indices, int(sym_len_s), int(sym_len_e) + + +# -------------------------------------------- +# imresize for tensor image [0, 1] +# -------------------------------------------- +def imresize(img, scale, antialiasing=True): + # Now the scale should be the same for H and W + # input: img: pytorch tensor, CHW or HW [0,1] + # output: CHW or HW [0,1] w/o round + need_squeeze = True if img.dim() == 2 else False + if need_squeeze: + img.unsqueeze_(0) + in_C, in_H, in_W = img.size() + out_C, out_H, out_W = in_C, math.ceil(in_H * scale), math.ceil(in_W * scale) + kernel_width = 4 + kernel = 'cubic' + + # Return the desired dimension order for performing the resize. The + # strategy is to perform the resize first along the dimension with the + # smallest scale factor. + # Now we do not support this. + + # get weights and indices + weights_H, indices_H, sym_len_Hs, sym_len_He = calculate_weights_indices( + in_H, out_H, scale, kernel, kernel_width, antialiasing) + weights_W, indices_W, sym_len_Ws, sym_len_We = calculate_weights_indices( + in_W, out_W, scale, kernel, kernel_width, antialiasing) + # process H dimension + # symmetric copying + img_aug = torch.FloatTensor(in_C, in_H + sym_len_Hs + sym_len_He, in_W) + img_aug.narrow(1, sym_len_Hs, in_H).copy_(img) + + sym_patch = img[:, :sym_len_Hs, :] + inv_idx = torch.arange(sym_patch.size(1) - 1, -1, -1).long() + sym_patch_inv = sym_patch.index_select(1, inv_idx) + img_aug.narrow(1, 0, sym_len_Hs).copy_(sym_patch_inv) + + sym_patch = img[:, -sym_len_He:, :] + inv_idx = torch.arange(sym_patch.size(1) - 1, -1, -1).long() + sym_patch_inv = sym_patch.index_select(1, inv_idx) + img_aug.narrow(1, sym_len_Hs + in_H, sym_len_He).copy_(sym_patch_inv) + + out_1 = torch.FloatTensor(in_C, out_H, in_W) + kernel_width = weights_H.size(1) + for i in range(out_H): + idx = int(indices_H[i][0]) + for j in range(out_C): + out_1[j, i, :] = img_aug[j, idx:idx + kernel_width, :].transpose(0, 1).mv(weights_H[i]) + + # process W dimension + # symmetric copying + out_1_aug = torch.FloatTensor(in_C, out_H, in_W + sym_len_Ws + sym_len_We) + out_1_aug.narrow(2, sym_len_Ws, in_W).copy_(out_1) + + sym_patch = out_1[:, :, :sym_len_Ws] + inv_idx = torch.arange(sym_patch.size(2) - 1, -1, -1).long() + sym_patch_inv = sym_patch.index_select(2, inv_idx) + out_1_aug.narrow(2, 0, sym_len_Ws).copy_(sym_patch_inv) + + sym_patch = out_1[:, :, -sym_len_We:] + inv_idx = torch.arange(sym_patch.size(2) - 1, -1, -1).long() + sym_patch_inv = sym_patch.index_select(2, inv_idx) + out_1_aug.narrow(2, sym_len_Ws + in_W, sym_len_We).copy_(sym_patch_inv) + + out_2 = torch.FloatTensor(in_C, out_H, out_W) + kernel_width = weights_W.size(1) + for i in range(out_W): + idx = int(indices_W[i][0]) + for j in range(out_C): + out_2[j, :, i] = out_1_aug[j, :, idx:idx + kernel_width].mv(weights_W[i]) + if need_squeeze: + out_2.squeeze_() + return out_2 + + +# -------------------------------------------- +# imresize for numpy image [0, 1] +# -------------------------------------------- +def imresize_np(img, scale, antialiasing=True): + # Now the scale should be the same for H and W + # input: img: Numpy, HWC or HW [0,1] + # output: HWC or HW [0,1] w/o round + img = torch.from_numpy(img) + need_squeeze = True if img.dim() == 2 else False + if need_squeeze: + img.unsqueeze_(2) + + in_H, in_W, in_C = img.size() + out_C, out_H, out_W = in_C, math.ceil(in_H * scale), math.ceil(in_W * scale) + kernel_width = 4 + kernel = 'cubic' + + # Return the desired dimension order for performing the resize. The + # strategy is to perform the resize first along the dimension with the + # smallest scale factor. + # Now we do not support this. + + # get weights and indices + weights_H, indices_H, sym_len_Hs, sym_len_He = calculate_weights_indices( + in_H, out_H, scale, kernel, kernel_width, antialiasing) + weights_W, indices_W, sym_len_Ws, sym_len_We = calculate_weights_indices( + in_W, out_W, scale, kernel, kernel_width, antialiasing) + # process H dimension + # symmetric copying + img_aug = torch.FloatTensor(in_H + sym_len_Hs + sym_len_He, in_W, in_C) + img_aug.narrow(0, sym_len_Hs, in_H).copy_(img) + + sym_patch = img[:sym_len_Hs, :, :] + inv_idx = torch.arange(sym_patch.size(0) - 1, -1, -1).long() + sym_patch_inv = sym_patch.index_select(0, inv_idx) + img_aug.narrow(0, 0, sym_len_Hs).copy_(sym_patch_inv) + + sym_patch = img[-sym_len_He:, :, :] + inv_idx = torch.arange(sym_patch.size(0) - 1, -1, -1).long() + sym_patch_inv = sym_patch.index_select(0, inv_idx) + img_aug.narrow(0, sym_len_Hs + in_H, sym_len_He).copy_(sym_patch_inv) + + out_1 = torch.FloatTensor(out_H, in_W, in_C) + kernel_width = weights_H.size(1) + for i in range(out_H): + idx = int(indices_H[i][0]) + for j in range(out_C): + out_1[i, :, j] = img_aug[idx:idx + kernel_width, :, j].transpose(0, 1).mv(weights_H[i]) + + # process W dimension + # symmetric copying + out_1_aug = torch.FloatTensor(out_H, in_W + sym_len_Ws + sym_len_We, in_C) + out_1_aug.narrow(1, sym_len_Ws, in_W).copy_(out_1) + + sym_patch = out_1[:, :sym_len_Ws, :] + inv_idx = torch.arange(sym_patch.size(1) - 1, -1, -1).long() + sym_patch_inv = sym_patch.index_select(1, inv_idx) + out_1_aug.narrow(1, 0, sym_len_Ws).copy_(sym_patch_inv) + + sym_patch = out_1[:, -sym_len_We:, :] + inv_idx = torch.arange(sym_patch.size(1) - 1, -1, -1).long() + sym_patch_inv = sym_patch.index_select(1, inv_idx) + out_1_aug.narrow(1, sym_len_Ws + in_W, sym_len_We).copy_(sym_patch_inv) + + out_2 = torch.FloatTensor(out_H, out_W, in_C) + kernel_width = weights_W.size(1) + for i in range(out_W): + idx = int(indices_W[i][0]) + for j in range(out_C): + out_2[:, i, j] = out_1_aug[:, idx:idx + kernel_width, j].mv(weights_W[i]) + if need_squeeze: + out_2.squeeze_() + + return out_2.numpy() + + +if __name__ == '__main__': + print('---') +# img = imread_uint('test.bmp', 3) +# img = uint2single(img) +# img_bicubic = imresize_np(img, 1/4) \ No newline at end of file diff --git a/ldm/modules/midas/__init__.py b/ldm/modules/midas/__init__.py new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/ldm/modules/midas/api.py b/ldm/modules/midas/api.py new file mode 100644 index 0000000000000000000000000000000000000000..b58ebbffd942a2fc22264f0ab47e400c26b9f41c --- /dev/null +++ b/ldm/modules/midas/api.py @@ -0,0 +1,170 @@ +# based on https://github.com/isl-org/MiDaS + +import cv2 +import torch +import torch.nn as nn +from torchvision.transforms import Compose + +from ldm.modules.midas.midas.dpt_depth import DPTDepthModel +from ldm.modules.midas.midas.midas_net import MidasNet +from ldm.modules.midas.midas.midas_net_custom import MidasNet_small +from ldm.modules.midas.midas.transforms import Resize, NormalizeImage, PrepareForNet + + +ISL_PATHS = { + "dpt_large": "midas_models/dpt_large-midas-2f21e586.pt", + "dpt_hybrid": "midas_models/dpt_hybrid-midas-501f0c75.pt", + "midas_v21": "", + "midas_v21_small": "", +} + + +def disabled_train(self, mode=True): + """Overwrite model.train with this function to make sure train/eval mode + does not change anymore.""" + return self + + +def load_midas_transform(model_type): + # https://github.com/isl-org/MiDaS/blob/master/run.py + # load transform only + if model_type == "dpt_large": # DPT-Large + net_w, net_h = 384, 384 + resize_mode = "minimal" + normalization = NormalizeImage(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5]) + + elif model_type == "dpt_hybrid": # DPT-Hybrid + net_w, net_h = 384, 384 + resize_mode = "minimal" + normalization = NormalizeImage(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5]) + + elif model_type == "midas_v21": + net_w, net_h = 384, 384 + resize_mode = "upper_bound" + normalization = NormalizeImage(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) + + elif model_type == "midas_v21_small": + net_w, net_h = 256, 256 + resize_mode = "upper_bound" + normalization = NormalizeImage(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) + + else: + assert False, f"model_type '{model_type}' not implemented, use: --model_type large" + + transform = Compose( + [ + Resize( + net_w, + net_h, + resize_target=None, + keep_aspect_ratio=True, + ensure_multiple_of=32, + resize_method=resize_mode, + image_interpolation_method=cv2.INTER_CUBIC, + ), + normalization, + PrepareForNet(), + ] + ) + + return transform + + +def load_model(model_type): + # https://github.com/isl-org/MiDaS/blob/master/run.py + # load network + model_path = ISL_PATHS[model_type] + if model_type == "dpt_large": # DPT-Large + model = DPTDepthModel( + path=model_path, + backbone="vitl16_384", + non_negative=True, + ) + net_w, net_h = 384, 384 + resize_mode = "minimal" + normalization = NormalizeImage(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5]) + + elif model_type == "dpt_hybrid": # DPT-Hybrid + model = DPTDepthModel( + path=model_path, + backbone="vitb_rn50_384", + non_negative=True, + ) + net_w, net_h = 384, 384 + resize_mode = "minimal" + normalization = NormalizeImage(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5]) + + elif model_type == "midas_v21": + model = MidasNet(model_path, non_negative=True) + net_w, net_h = 384, 384 + resize_mode = "upper_bound" + normalization = NormalizeImage( + mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225] + ) + + elif model_type == "midas_v21_small": + model = MidasNet_small(model_path, features=64, backbone="efficientnet_lite3", exportable=True, + non_negative=True, blocks={'expand': True}) + net_w, net_h = 256, 256 + resize_mode = "upper_bound" + normalization = NormalizeImage( + mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225] + ) + + else: + print(f"model_type '{model_type}' not implemented, use: --model_type large") + assert False + + transform = Compose( + [ + Resize( + net_w, + net_h, + resize_target=None, + keep_aspect_ratio=True, + ensure_multiple_of=32, + resize_method=resize_mode, + image_interpolation_method=cv2.INTER_CUBIC, + ), + normalization, + PrepareForNet(), + ] + ) + + return model.eval(), transform + + +class MiDaSInference(nn.Module): + MODEL_TYPES_TORCH_HUB = [ + "DPT_Large", + "DPT_Hybrid", + "MiDaS_small" + ] + MODEL_TYPES_ISL = [ + "dpt_large", + "dpt_hybrid", + "midas_v21", + "midas_v21_small", + ] + + def __init__(self, model_type): + super().__init__() + assert (model_type in self.MODEL_TYPES_ISL) + model, _ = load_model(model_type) + self.model = model + self.model.train = disabled_train + + def forward(self, x): + # x in 0..1 as produced by calling self.transform on a 0..1 float64 numpy array + # NOTE: we expect that the correct transform has been called during dataloading. + with torch.no_grad(): + prediction = self.model(x) + prediction = torch.nn.functional.interpolate( + prediction.unsqueeze(1), + size=x.shape[2:], + mode="bicubic", + align_corners=False, + ) + assert prediction.shape == (x.shape[0], 1, x.shape[2], x.shape[3]) + return prediction + diff --git a/ldm/modules/midas/midas/__init__.py b/ldm/modules/midas/midas/__init__.py new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/ldm/modules/midas/midas/base_model.py b/ldm/modules/midas/midas/base_model.py new file mode 100644 index 0000000000000000000000000000000000000000..5cf430239b47ec5ec07531263f26f5c24a2311cd --- /dev/null +++ b/ldm/modules/midas/midas/base_model.py @@ -0,0 +1,16 @@ +import torch + + +class BaseModel(torch.nn.Module): + def load(self, path): + """Load model from file. + + Args: + path (str): file path + """ + parameters = torch.load(path, map_location=torch.device('cpu')) + + if "optimizer" in parameters: + parameters = parameters["model"] + + self.load_state_dict(parameters) diff --git a/ldm/modules/midas/midas/blocks.py b/ldm/modules/midas/midas/blocks.py new file mode 100644 index 0000000000000000000000000000000000000000..2145d18fa98060a618536d9a64fe6589e9be4f78 --- /dev/null +++ b/ldm/modules/midas/midas/blocks.py @@ -0,0 +1,342 @@ +import torch +import torch.nn as nn + +from .vit import ( + _make_pretrained_vitb_rn50_384, + _make_pretrained_vitl16_384, + _make_pretrained_vitb16_384, + forward_vit, +) + +def _make_encoder(backbone, features, use_pretrained, groups=1, expand=False, exportable=True, hooks=None, use_vit_only=False, use_readout="ignore",): + if backbone == "vitl16_384": + pretrained = _make_pretrained_vitl16_384( + use_pretrained, hooks=hooks, use_readout=use_readout + ) + scratch = _make_scratch( + [256, 512, 1024, 1024], features, groups=groups, expand=expand + ) # ViT-L/16 - 85.0% Top1 (backbone) + elif backbone == "vitb_rn50_384": + pretrained = _make_pretrained_vitb_rn50_384( + use_pretrained, + hooks=hooks, + use_vit_only=use_vit_only, + use_readout=use_readout, + ) + scratch = _make_scratch( + [256, 512, 768, 768], features, groups=groups, expand=expand + ) # ViT-H/16 - 85.0% Top1 (backbone) + elif backbone == "vitb16_384": + pretrained = _make_pretrained_vitb16_384( + use_pretrained, hooks=hooks, use_readout=use_readout + ) + scratch = _make_scratch( + [96, 192, 384, 768], features, groups=groups, expand=expand + ) # ViT-B/16 - 84.6% Top1 (backbone) + elif backbone == "resnext101_wsl": + pretrained = _make_pretrained_resnext101_wsl(use_pretrained) + scratch = _make_scratch([256, 512, 1024, 2048], features, groups=groups, expand=expand) # efficientnet_lite3 + elif backbone == "efficientnet_lite3": + pretrained = _make_pretrained_efficientnet_lite3(use_pretrained, exportable=exportable) + scratch = _make_scratch([32, 48, 136, 384], features, groups=groups, expand=expand) # efficientnet_lite3 + else: + print(f"Backbone '{backbone}' not implemented") + assert False + + return pretrained, scratch + + +def _make_scratch(in_shape, out_shape, groups=1, expand=False): + scratch = nn.Module() + + out_shape1 = out_shape + out_shape2 = out_shape + out_shape3 = out_shape + out_shape4 = out_shape + if expand==True: + out_shape1 = out_shape + out_shape2 = out_shape*2 + out_shape3 = out_shape*4 + out_shape4 = out_shape*8 + + scratch.layer1_rn = nn.Conv2d( + in_shape[0], out_shape1, kernel_size=3, stride=1, padding=1, bias=False, groups=groups + ) + scratch.layer2_rn = nn.Conv2d( + in_shape[1], out_shape2, kernel_size=3, stride=1, padding=1, bias=False, groups=groups + ) + scratch.layer3_rn = nn.Conv2d( + in_shape[2], out_shape3, kernel_size=3, stride=1, padding=1, bias=False, groups=groups + ) + scratch.layer4_rn = nn.Conv2d( + in_shape[3], out_shape4, kernel_size=3, stride=1, padding=1, bias=False, groups=groups + ) + + return scratch + + +def _make_pretrained_efficientnet_lite3(use_pretrained, exportable=False): + efficientnet = torch.hub.load( + "rwightman/gen-efficientnet-pytorch", + "tf_efficientnet_lite3", + pretrained=use_pretrained, + exportable=exportable + ) + return _make_efficientnet_backbone(efficientnet) + + +def _make_efficientnet_backbone(effnet): + pretrained = nn.Module() + + pretrained.layer1 = nn.Sequential( + effnet.conv_stem, effnet.bn1, effnet.act1, *effnet.blocks[0:2] + ) + pretrained.layer2 = nn.Sequential(*effnet.blocks[2:3]) + pretrained.layer3 = nn.Sequential(*effnet.blocks[3:5]) + pretrained.layer4 = nn.Sequential(*effnet.blocks[5:9]) + + return pretrained + + +def _make_resnet_backbone(resnet): + pretrained = nn.Module() + pretrained.layer1 = nn.Sequential( + resnet.conv1, resnet.bn1, resnet.relu, resnet.maxpool, resnet.layer1 + ) + + pretrained.layer2 = resnet.layer2 + pretrained.layer3 = resnet.layer3 + pretrained.layer4 = resnet.layer4 + + return pretrained + + +def _make_pretrained_resnext101_wsl(use_pretrained): + resnet = torch.hub.load("facebookresearch/WSL-Images", "resnext101_32x8d_wsl") + return _make_resnet_backbone(resnet) + + + +class Interpolate(nn.Module): + """Interpolation module. + """ + + def __init__(self, scale_factor, mode, align_corners=False): + """Init. + + Args: + scale_factor (float): scaling + mode (str): interpolation mode + """ + super(Interpolate, self).__init__() + + self.interp = nn.functional.interpolate + self.scale_factor = scale_factor + self.mode = mode + self.align_corners = align_corners + + def forward(self, x): + """Forward pass. + + Args: + x (tensor): input + + Returns: + tensor: interpolated data + """ + + x = self.interp( + x, scale_factor=self.scale_factor, mode=self.mode, align_corners=self.align_corners + ) + + return x + + +class ResidualConvUnit(nn.Module): + """Residual convolution module. + """ + + def __init__(self, features): + """Init. + + Args: + features (int): number of features + """ + super().__init__() + + self.conv1 = nn.Conv2d( + features, features, kernel_size=3, stride=1, padding=1, bias=True + ) + + self.conv2 = nn.Conv2d( + features, features, kernel_size=3, stride=1, padding=1, bias=True + ) + + self.relu = nn.ReLU(inplace=True) + + def forward(self, x): + """Forward pass. + + Args: + x (tensor): input + + Returns: + tensor: output + """ + out = self.relu(x) + out = self.conv1(out) + out = self.relu(out) + out = self.conv2(out) + + return out + x + + +class FeatureFusionBlock(nn.Module): + """Feature fusion block. + """ + + def __init__(self, features): + """Init. + + Args: + features (int): number of features + """ + super(FeatureFusionBlock, self).__init__() + + self.resConfUnit1 = ResidualConvUnit(features) + self.resConfUnit2 = ResidualConvUnit(features) + + def forward(self, *xs): + """Forward pass. + + Returns: + tensor: output + """ + output = xs[0] + + if len(xs) == 2: + output += self.resConfUnit1(xs[1]) + + output = self.resConfUnit2(output) + + output = nn.functional.interpolate( + output, scale_factor=2, mode="bilinear", align_corners=True + ) + + return output + + + + +class ResidualConvUnit_custom(nn.Module): + """Residual convolution module. + """ + + def __init__(self, features, activation, bn): + """Init. + + Args: + features (int): number of features + """ + super().__init__() + + self.bn = bn + + self.groups=1 + + self.conv1 = nn.Conv2d( + features, features, kernel_size=3, stride=1, padding=1, bias=True, groups=self.groups + ) + + self.conv2 = nn.Conv2d( + features, features, kernel_size=3, stride=1, padding=1, bias=True, groups=self.groups + ) + + if self.bn==True: + self.bn1 = nn.BatchNorm2d(features) + self.bn2 = nn.BatchNorm2d(features) + + self.activation = activation + + self.skip_add = nn.quantized.FloatFunctional() + + def forward(self, x): + """Forward pass. + + Args: + x (tensor): input + + Returns: + tensor: output + """ + + out = self.activation(x) + out = self.conv1(out) + if self.bn==True: + out = self.bn1(out) + + out = self.activation(out) + out = self.conv2(out) + if self.bn==True: + out = self.bn2(out) + + if self.groups > 1: + out = self.conv_merge(out) + + return self.skip_add.add(out, x) + + # return out + x + + +class FeatureFusionBlock_custom(nn.Module): + """Feature fusion block. + """ + + def __init__(self, features, activation, deconv=False, bn=False, expand=False, align_corners=True): + """Init. + + Args: + features (int): number of features + """ + super(FeatureFusionBlock_custom, self).__init__() + + self.deconv = deconv + self.align_corners = align_corners + + self.groups=1 + + self.expand = expand + out_features = features + if self.expand==True: + out_features = features//2 + + self.out_conv = nn.Conv2d(features, out_features, kernel_size=1, stride=1, padding=0, bias=True, groups=1) + + self.resConfUnit1 = ResidualConvUnit_custom(features, activation, bn) + self.resConfUnit2 = ResidualConvUnit_custom(features, activation, bn) + + self.skip_add = nn.quantized.FloatFunctional() + + def forward(self, *xs): + """Forward pass. + + Returns: + tensor: output + """ + output = xs[0] + + if len(xs) == 2: + res = self.resConfUnit1(xs[1]) + output = self.skip_add.add(output, res) + # output += res + + output = self.resConfUnit2(output) + + output = nn.functional.interpolate( + output, scale_factor=2, mode="bilinear", align_corners=self.align_corners + ) + + output = self.out_conv(output) + + return output + diff --git a/ldm/modules/midas/midas/dpt_depth.py b/ldm/modules/midas/midas/dpt_depth.py new file mode 100644 index 0000000000000000000000000000000000000000..4e9aab5d2767dffea39da5b3f30e2798688216f1 --- /dev/null +++ b/ldm/modules/midas/midas/dpt_depth.py @@ -0,0 +1,109 @@ +import torch +import torch.nn as nn +import torch.nn.functional as F + +from .base_model import BaseModel +from .blocks import ( + FeatureFusionBlock, + FeatureFusionBlock_custom, + Interpolate, + _make_encoder, + forward_vit, +) + + +def _make_fusion_block(features, use_bn): + return FeatureFusionBlock_custom( + features, + nn.ReLU(False), + deconv=False, + bn=use_bn, + expand=False, + align_corners=True, + ) + + +class DPT(BaseModel): + def __init__( + self, + head, + features=256, + backbone="vitb_rn50_384", + readout="project", + channels_last=False, + use_bn=False, + ): + + super(DPT, self).__init__() + + self.channels_last = channels_last + + hooks = { + "vitb_rn50_384": [0, 1, 8, 11], + "vitb16_384": [2, 5, 8, 11], + "vitl16_384": [5, 11, 17, 23], + } + + # Instantiate backbone and reassemble blocks + self.pretrained, self.scratch = _make_encoder( + backbone, + features, + False, # Set to true of you want to train from scratch, uses ImageNet weights + groups=1, + expand=False, + exportable=False, + hooks=hooks[backbone], + use_readout=readout, + ) + + self.scratch.refinenet1 = _make_fusion_block(features, use_bn) + self.scratch.refinenet2 = _make_fusion_block(features, use_bn) + self.scratch.refinenet3 = _make_fusion_block(features, use_bn) + self.scratch.refinenet4 = _make_fusion_block(features, use_bn) + + self.scratch.output_conv = head + + + def forward(self, x): + if self.channels_last == True: + x.contiguous(memory_format=torch.channels_last) + + layer_1, layer_2, layer_3, layer_4 = forward_vit(self.pretrained, x) + + layer_1_rn = self.scratch.layer1_rn(layer_1) + layer_2_rn = self.scratch.layer2_rn(layer_2) + layer_3_rn = self.scratch.layer3_rn(layer_3) + layer_4_rn = self.scratch.layer4_rn(layer_4) + + path_4 = self.scratch.refinenet4(layer_4_rn) + path_3 = self.scratch.refinenet3(path_4, layer_3_rn) + path_2 = self.scratch.refinenet2(path_3, layer_2_rn) + path_1 = self.scratch.refinenet1(path_2, layer_1_rn) + + out = self.scratch.output_conv(path_1) + + return out + + +class DPTDepthModel(DPT): + def __init__(self, path=None, non_negative=True, **kwargs): + features = kwargs["features"] if "features" in kwargs else 256 + + head = nn.Sequential( + nn.Conv2d(features, features // 2, kernel_size=3, stride=1, padding=1), + Interpolate(scale_factor=2, mode="bilinear", align_corners=True), + nn.Conv2d(features // 2, 32, kernel_size=3, stride=1, padding=1), + nn.ReLU(True), + nn.Conv2d(32, 1, kernel_size=1, stride=1, padding=0), + nn.ReLU(True) if non_negative else nn.Identity(), + nn.Identity(), + ) + + super().__init__(head, **kwargs) + + if path is not None: + self.load(path) + + def forward(self, x): + return super().forward(x).squeeze(dim=1) + diff --git a/ldm/modules/midas/midas/midas_net.py b/ldm/modules/midas/midas/midas_net.py new file mode 100644 index 0000000000000000000000000000000000000000..8a954977800b0a0f48807e80fa63041910e33c1f --- /dev/null +++ b/ldm/modules/midas/midas/midas_net.py @@ -0,0 +1,76 @@ +"""MidashNet: Network for monocular depth estimation trained by mixing several datasets. +This file contains code that is adapted from +https://github.com/thomasjpfan/pytorch_refinenet/blob/master/pytorch_refinenet/refinenet/refinenet_4cascade.py +""" +import torch +import torch.nn as nn + +from .base_model import BaseModel +from .blocks import FeatureFusionBlock, Interpolate, _make_encoder + + +class MidasNet(BaseModel): + """Network for monocular depth estimation. + """ + + def __init__(self, path=None, features=256, non_negative=True): + """Init. + + Args: + path (str, optional): Path to saved model. Defaults to None. + features (int, optional): Number of features. Defaults to 256. + backbone (str, optional): Backbone network for encoder. Defaults to resnet50 + """ + print("Loading weights: ", path) + + super(MidasNet, self).__init__() + + use_pretrained = False if path is None else True + + self.pretrained, self.scratch = _make_encoder(backbone="resnext101_wsl", features=features, use_pretrained=use_pretrained) + + self.scratch.refinenet4 = FeatureFusionBlock(features) + self.scratch.refinenet3 = FeatureFusionBlock(features) + self.scratch.refinenet2 = FeatureFusionBlock(features) + self.scratch.refinenet1 = FeatureFusionBlock(features) + + self.scratch.output_conv = nn.Sequential( + nn.Conv2d(features, 128, kernel_size=3, stride=1, padding=1), + Interpolate(scale_factor=2, mode="bilinear"), + nn.Conv2d(128, 32, kernel_size=3, stride=1, padding=1), + nn.ReLU(True), + nn.Conv2d(32, 1, kernel_size=1, stride=1, padding=0), + nn.ReLU(True) if non_negative else nn.Identity(), + ) + + if path: + self.load(path) + + def forward(self, x): + """Forward pass. + + Args: + x (tensor): input data (image) + + Returns: + tensor: depth + """ + + layer_1 = self.pretrained.layer1(x) + layer_2 = self.pretrained.layer2(layer_1) + layer_3 = self.pretrained.layer3(layer_2) + layer_4 = self.pretrained.layer4(layer_3) + + layer_1_rn = self.scratch.layer1_rn(layer_1) + layer_2_rn = self.scratch.layer2_rn(layer_2) + layer_3_rn = self.scratch.layer3_rn(layer_3) + layer_4_rn = self.scratch.layer4_rn(layer_4) + + path_4 = self.scratch.refinenet4(layer_4_rn) + path_3 = self.scratch.refinenet3(path_4, layer_3_rn) + path_2 = self.scratch.refinenet2(path_3, layer_2_rn) + path_1 = self.scratch.refinenet1(path_2, layer_1_rn) + + out = self.scratch.output_conv(path_1) + + return torch.squeeze(out, dim=1) diff --git a/ldm/modules/midas/midas/midas_net_custom.py b/ldm/modules/midas/midas/midas_net_custom.py new file mode 100644 index 0000000000000000000000000000000000000000..50e4acb5e53d5fabefe3dde16ab49c33c2b7797c --- /dev/null +++ b/ldm/modules/midas/midas/midas_net_custom.py @@ -0,0 +1,128 @@ +"""MidashNet: Network for monocular depth estimation trained by mixing several datasets. +This file contains code that is adapted from +https://github.com/thomasjpfan/pytorch_refinenet/blob/master/pytorch_refinenet/refinenet/refinenet_4cascade.py +""" +import torch +import torch.nn as nn + +from .base_model import BaseModel +from .blocks import FeatureFusionBlock, FeatureFusionBlock_custom, Interpolate, _make_encoder + + +class MidasNet_small(BaseModel): + """Network for monocular depth estimation. + """ + + def __init__(self, path=None, features=64, backbone="efficientnet_lite3", non_negative=True, exportable=True, channels_last=False, align_corners=True, + blocks={'expand': True}): + """Init. + + Args: + path (str, optional): Path to saved model. Defaults to None. + features (int, optional): Number of features. Defaults to 256. + backbone (str, optional): Backbone network for encoder. Defaults to resnet50 + """ + print("Loading weights: ", path) + + super(MidasNet_small, self).__init__() + + use_pretrained = False if path else True + + self.channels_last = channels_last + self.blocks = blocks + self.backbone = backbone + + self.groups = 1 + + features1=features + features2=features + features3=features + features4=features + self.expand = False + if "expand" in self.blocks and self.blocks['expand'] == True: + self.expand = True + features1=features + features2=features*2 + features3=features*4 + features4=features*8 + + self.pretrained, self.scratch = _make_encoder(self.backbone, features, use_pretrained, groups=self.groups, expand=self.expand, exportable=exportable) + + self.scratch.activation = nn.ReLU(False) + + self.scratch.refinenet4 = FeatureFusionBlock_custom(features4, self.scratch.activation, deconv=False, bn=False, expand=self.expand, align_corners=align_corners) + self.scratch.refinenet3 = FeatureFusionBlock_custom(features3, self.scratch.activation, deconv=False, bn=False, expand=self.expand, align_corners=align_corners) + self.scratch.refinenet2 = FeatureFusionBlock_custom(features2, self.scratch.activation, deconv=False, bn=False, expand=self.expand, align_corners=align_corners) + self.scratch.refinenet1 = FeatureFusionBlock_custom(features1, self.scratch.activation, deconv=False, bn=False, align_corners=align_corners) + + + self.scratch.output_conv = nn.Sequential( + nn.Conv2d(features, features//2, kernel_size=3, stride=1, padding=1, groups=self.groups), + Interpolate(scale_factor=2, mode="bilinear"), + nn.Conv2d(features//2, 32, kernel_size=3, stride=1, padding=1), + self.scratch.activation, + nn.Conv2d(32, 1, kernel_size=1, stride=1, padding=0), + nn.ReLU(True) if non_negative else nn.Identity(), + nn.Identity(), + ) + + if path: + self.load(path) + + + def forward(self, x): + """Forward pass. + + Args: + x (tensor): input data (image) + + Returns: + tensor: depth + """ + if self.channels_last==True: + print("self.channels_last = ", self.channels_last) + x.contiguous(memory_format=torch.channels_last) + + + layer_1 = self.pretrained.layer1(x) + layer_2 = self.pretrained.layer2(layer_1) + layer_3 = self.pretrained.layer3(layer_2) + layer_4 = self.pretrained.layer4(layer_3) + + layer_1_rn = self.scratch.layer1_rn(layer_1) + layer_2_rn = self.scratch.layer2_rn(layer_2) + layer_3_rn = self.scratch.layer3_rn(layer_3) + layer_4_rn = self.scratch.layer4_rn(layer_4) + + + path_4 = self.scratch.refinenet4(layer_4_rn) + path_3 = self.scratch.refinenet3(path_4, layer_3_rn) + path_2 = self.scratch.refinenet2(path_3, layer_2_rn) + path_1 = self.scratch.refinenet1(path_2, layer_1_rn) + + out = self.scratch.output_conv(path_1) + + return torch.squeeze(out, dim=1) + + + +def fuse_model(m): + prev_previous_type = nn.Identity() + prev_previous_name = '' + previous_type = nn.Identity() + previous_name = '' + for name, module in m.named_modules(): + if prev_previous_type == nn.Conv2d and previous_type == nn.BatchNorm2d and type(module) == nn.ReLU: + # print("FUSED ", prev_previous_name, previous_name, name) + torch.quantization.fuse_modules(m, [prev_previous_name, previous_name, name], inplace=True) + elif prev_previous_type == nn.Conv2d and previous_type == nn.BatchNorm2d: + # print("FUSED ", prev_previous_name, previous_name) + torch.quantization.fuse_modules(m, [prev_previous_name, previous_name], inplace=True) + # elif previous_type == nn.Conv2d and type(module) == nn.ReLU: + # print("FUSED ", previous_name, name) + # torch.quantization.fuse_modules(m, [previous_name, name], inplace=True) + + prev_previous_type = previous_type + prev_previous_name = previous_name + previous_type = type(module) + previous_name = name \ No newline at end of file diff --git a/ldm/modules/midas/midas/transforms.py b/ldm/modules/midas/midas/transforms.py new file mode 100644 index 0000000000000000000000000000000000000000..350cbc11662633ad7f8968eb10be2e7de6e384e9 --- /dev/null +++ b/ldm/modules/midas/midas/transforms.py @@ -0,0 +1,234 @@ +import numpy as np +import cv2 +import math + + +def apply_min_size(sample, size, image_interpolation_method=cv2.INTER_AREA): + """Rezise the sample to ensure the given size. Keeps aspect ratio. + + Args: + sample (dict): sample + size (tuple): image size + + Returns: + tuple: new size + """ + shape = list(sample["disparity"].shape) + + if shape[0] >= size[0] and shape[1] >= size[1]: + return sample + + scale = [0, 0] + scale[0] = size[0] / shape[0] + scale[1] = size[1] / shape[1] + + scale = max(scale) + + shape[0] = math.ceil(scale * shape[0]) + shape[1] = math.ceil(scale * shape[1]) + + # resize + sample["image"] = cv2.resize( + sample["image"], tuple(shape[::-1]), interpolation=image_interpolation_method + ) + + sample["disparity"] = cv2.resize( + sample["disparity"], tuple(shape[::-1]), interpolation=cv2.INTER_NEAREST + ) + sample["mask"] = cv2.resize( + sample["mask"].astype(np.float32), + tuple(shape[::-1]), + interpolation=cv2.INTER_NEAREST, + ) + sample["mask"] = sample["mask"].astype(bool) + + return tuple(shape) + + +class Resize(object): + """Resize sample to given size (width, height). + """ + + def __init__( + self, + width, + height, + resize_target=True, + keep_aspect_ratio=False, + ensure_multiple_of=1, + resize_method="lower_bound", + image_interpolation_method=cv2.INTER_AREA, + ): + """Init. + + Args: + width (int): desired output width + height (int): desired output height + resize_target (bool, optional): + True: Resize the full sample (image, mask, target). + False: Resize image only. + Defaults to True. + keep_aspect_ratio (bool, optional): + True: Keep the aspect ratio of the input sample. + Output sample might not have the given width and height, and + resize behaviour depends on the parameter 'resize_method'. + Defaults to False. + ensure_multiple_of (int, optional): + Output width and height is constrained to be multiple of this parameter. + Defaults to 1. + resize_method (str, optional): + "lower_bound": Output will be at least as large as the given size. + "upper_bound": Output will be at max as large as the given size. (Output size might be smaller than given size.) + "minimal": Scale as least as possible. (Output size might be smaller than given size.) + Defaults to "lower_bound". + """ + self.__width = width + self.__height = height + + self.__resize_target = resize_target + self.__keep_aspect_ratio = keep_aspect_ratio + self.__multiple_of = ensure_multiple_of + self.__resize_method = resize_method + self.__image_interpolation_method = image_interpolation_method + + def constrain_to_multiple_of(self, x, min_val=0, max_val=None): + y = (np.round(x / self.__multiple_of) * self.__multiple_of).astype(int) + + if max_val is not None and y > max_val: + y = (np.floor(x / self.__multiple_of) * self.__multiple_of).astype(int) + + if y < min_val: + y = (np.ceil(x / self.__multiple_of) * self.__multiple_of).astype(int) + + return y + + def get_size(self, width, height): + # determine new height and width + scale_height = self.__height / height + scale_width = self.__width / width + + if self.__keep_aspect_ratio: + if self.__resize_method == "lower_bound": + # scale such that output size is lower bound + if scale_width > scale_height: + # fit width + scale_height = scale_width + else: + # fit height + scale_width = scale_height + elif self.__resize_method == "upper_bound": + # scale such that output size is upper bound + if scale_width < scale_height: + # fit width + scale_height = scale_width + else: + # fit height + scale_width = scale_height + elif self.__resize_method == "minimal": + # scale as least as possbile + if abs(1 - scale_width) < abs(1 - scale_height): + # fit width + scale_height = scale_width + else: + # fit height + scale_width = scale_height + else: + raise ValueError( + f"resize_method {self.__resize_method} not implemented" + ) + + if self.__resize_method == "lower_bound": + new_height = self.constrain_to_multiple_of( + scale_height * height, min_val=self.__height + ) + new_width = self.constrain_to_multiple_of( + scale_width * width, min_val=self.__width + ) + elif self.__resize_method == "upper_bound": + new_height = self.constrain_to_multiple_of( + scale_height * height, max_val=self.__height + ) + new_width = self.constrain_to_multiple_of( + scale_width * width, max_val=self.__width + ) + elif self.__resize_method == "minimal": + new_height = self.constrain_to_multiple_of(scale_height * height) + new_width = self.constrain_to_multiple_of(scale_width * width) + else: + raise ValueError(f"resize_method {self.__resize_method} not implemented") + + return (new_width, new_height) + + def __call__(self, sample): + width, height = self.get_size( + sample["image"].shape[1], sample["image"].shape[0] + ) + + # resize sample + sample["image"] = cv2.resize( + sample["image"], + (width, height), + interpolation=self.__image_interpolation_method, + ) + + if self.__resize_target: + if "disparity" in sample: + sample["disparity"] = cv2.resize( + sample["disparity"], + (width, height), + interpolation=cv2.INTER_NEAREST, + ) + + if "depth" in sample: + sample["depth"] = cv2.resize( + sample["depth"], (width, height), interpolation=cv2.INTER_NEAREST + ) + + sample["mask"] = cv2.resize( + sample["mask"].astype(np.float32), + (width, height), + interpolation=cv2.INTER_NEAREST, + ) + sample["mask"] = sample["mask"].astype(bool) + + return sample + + +class NormalizeImage(object): + """Normlize image by given mean and std. + """ + + def __init__(self, mean, std): + self.__mean = mean + self.__std = std + + def __call__(self, sample): + sample["image"] = (sample["image"] - self.__mean) / self.__std + + return sample + + +class PrepareForNet(object): + """Prepare sample for usage as network input. + """ + + def __init__(self): + pass + + def __call__(self, sample): + image = np.transpose(sample["image"], (2, 0, 1)) + sample["image"] = np.ascontiguousarray(image).astype(np.float32) + + if "mask" in sample: + sample["mask"] = sample["mask"].astype(np.float32) + sample["mask"] = np.ascontiguousarray(sample["mask"]) + + if "disparity" in sample: + disparity = sample["disparity"].astype(np.float32) + sample["disparity"] = np.ascontiguousarray(disparity) + + if "depth" in sample: + depth = sample["depth"].astype(np.float32) + sample["depth"] = np.ascontiguousarray(depth) + + return sample diff --git a/ldm/modules/midas/midas/vit.py b/ldm/modules/midas/midas/vit.py new file mode 100644 index 0000000000000000000000000000000000000000..ea46b1be88b261b0dec04f3da0256f5f66f88a74 --- /dev/null +++ b/ldm/modules/midas/midas/vit.py @@ -0,0 +1,491 @@ +import torch +import torch.nn as nn +import timm +import types +import math +import torch.nn.functional as F + + +class Slice(nn.Module): + def __init__(self, start_index=1): + super(Slice, self).__init__() + self.start_index = start_index + + def forward(self, x): + return x[:, self.start_index :] + + +class AddReadout(nn.Module): + def __init__(self, start_index=1): + super(AddReadout, self).__init__() + self.start_index = start_index + + def forward(self, x): + if self.start_index == 2: + readout = (x[:, 0] + x[:, 1]) / 2 + else: + readout = x[:, 0] + return x[:, self.start_index :] + readout.unsqueeze(1) + + +class ProjectReadout(nn.Module): + def __init__(self, in_features, start_index=1): + super(ProjectReadout, self).__init__() + self.start_index = start_index + + self.project = nn.Sequential(nn.Linear(2 * in_features, in_features), nn.GELU()) + + def forward(self, x): + readout = x[:, 0].unsqueeze(1).expand_as(x[:, self.start_index :]) + features = torch.cat((x[:, self.start_index :], readout), -1) + + return self.project(features) + + +class Transpose(nn.Module): + def __init__(self, dim0, dim1): + super(Transpose, self).__init__() + self.dim0 = dim0 + self.dim1 = dim1 + + def forward(self, x): + x = x.transpose(self.dim0, self.dim1) + return x + + +def forward_vit(pretrained, x): + b, c, h, w = x.shape + + glob = pretrained.model.forward_flex(x) + + layer_1 = pretrained.activations["1"] + layer_2 = pretrained.activations["2"] + layer_3 = pretrained.activations["3"] + layer_4 = pretrained.activations["4"] + + layer_1 = pretrained.act_postprocess1[0:2](layer_1) + layer_2 = pretrained.act_postprocess2[0:2](layer_2) + layer_3 = pretrained.act_postprocess3[0:2](layer_3) + layer_4 = pretrained.act_postprocess4[0:2](layer_4) + + unflatten = nn.Sequential( + nn.Unflatten( + 2, + torch.Size( + [ + h // pretrained.model.patch_size[1], + w // pretrained.model.patch_size[0], + ] + ), + ) + ) + + if layer_1.ndim == 3: + layer_1 = unflatten(layer_1) + if layer_2.ndim == 3: + layer_2 = unflatten(layer_2) + if layer_3.ndim == 3: + layer_3 = unflatten(layer_3) + if layer_4.ndim == 3: + layer_4 = unflatten(layer_4) + + layer_1 = pretrained.act_postprocess1[3 : len(pretrained.act_postprocess1)](layer_1) + layer_2 = pretrained.act_postprocess2[3 : len(pretrained.act_postprocess2)](layer_2) + layer_3 = pretrained.act_postprocess3[3 : len(pretrained.act_postprocess3)](layer_3) + layer_4 = pretrained.act_postprocess4[3 : len(pretrained.act_postprocess4)](layer_4) + + return layer_1, layer_2, layer_3, layer_4 + + +def _resize_pos_embed(self, posemb, gs_h, gs_w): + posemb_tok, posemb_grid = ( + posemb[:, : self.start_index], + posemb[0, self.start_index :], + ) + + gs_old = int(math.sqrt(len(posemb_grid))) + + posemb_grid = posemb_grid.reshape(1, gs_old, gs_old, -1).permute(0, 3, 1, 2) + posemb_grid = F.interpolate(posemb_grid, size=(gs_h, gs_w), mode="bilinear") + posemb_grid = posemb_grid.permute(0, 2, 3, 1).reshape(1, gs_h * gs_w, -1) + + posemb = torch.cat([posemb_tok, posemb_grid], dim=1) + + return posemb + + +def forward_flex(self, x): + b, c, h, w = x.shape + + pos_embed = self._resize_pos_embed( + self.pos_embed, h // self.patch_size[1], w // self.patch_size[0] + ) + + B = x.shape[0] + + if hasattr(self.patch_embed, "backbone"): + x = self.patch_embed.backbone(x) + if isinstance(x, (list, tuple)): + x = x[-1] # last feature if backbone outputs list/tuple of features + + x = self.patch_embed.proj(x).flatten(2).transpose(1, 2) + + if getattr(self, "dist_token", None) is not None: + cls_tokens = self.cls_token.expand( + B, -1, -1 + ) # stole cls_tokens impl from Phil Wang, thanks + dist_token = self.dist_token.expand(B, -1, -1) + x = torch.cat((cls_tokens, dist_token, x), dim=1) + else: + cls_tokens = self.cls_token.expand( + B, -1, -1 + ) # stole cls_tokens impl from Phil Wang, thanks + x = torch.cat((cls_tokens, x), dim=1) + + x = x + pos_embed + x = self.pos_drop(x) + + for blk in self.blocks: + x = blk(x) + + x = self.norm(x) + + return x + + +activations = {} + + +def get_activation(name): + def hook(model, input, output): + activations[name] = output + + return hook + + +def get_readout_oper(vit_features, features, use_readout, start_index=1): + if use_readout == "ignore": + readout_oper = [Slice(start_index)] * len(features) + elif use_readout == "add": + readout_oper = [AddReadout(start_index)] * len(features) + elif use_readout == "project": + readout_oper = [ + ProjectReadout(vit_features, start_index) for out_feat in features + ] + else: + assert ( + False + ), "wrong operation for readout token, use_readout can be 'ignore', 'add', or 'project'" + + return readout_oper + + +def _make_vit_b16_backbone( + model, + features=[96, 192, 384, 768], + size=[384, 384], + hooks=[2, 5, 8, 11], + vit_features=768, + use_readout="ignore", + start_index=1, +): + pretrained = nn.Module() + + pretrained.model = model + pretrained.model.blocks[hooks[0]].register_forward_hook(get_activation("1")) + pretrained.model.blocks[hooks[1]].register_forward_hook(get_activation("2")) + pretrained.model.blocks[hooks[2]].register_forward_hook(get_activation("3")) + pretrained.model.blocks[hooks[3]].register_forward_hook(get_activation("4")) + + pretrained.activations = activations + + readout_oper = get_readout_oper(vit_features, features, use_readout, start_index) + + # 32, 48, 136, 384 + pretrained.act_postprocess1 = nn.Sequential( + readout_oper[0], + Transpose(1, 2), + nn.Unflatten(2, torch.Size([size[0] // 16, size[1] // 16])), + nn.Conv2d( + in_channels=vit_features, + out_channels=features[0], + kernel_size=1, + stride=1, + padding=0, + ), + nn.ConvTranspose2d( + in_channels=features[0], + out_channels=features[0], + kernel_size=4, + stride=4, + padding=0, + bias=True, + dilation=1, + groups=1, + ), + ) + + pretrained.act_postprocess2 = nn.Sequential( + readout_oper[1], + Transpose(1, 2), + nn.Unflatten(2, torch.Size([size[0] // 16, size[1] // 16])), + nn.Conv2d( + in_channels=vit_features, + out_channels=features[1], + kernel_size=1, + stride=1, + padding=0, + ), + nn.ConvTranspose2d( + in_channels=features[1], + out_channels=features[1], + kernel_size=2, + stride=2, + padding=0, + bias=True, + dilation=1, + groups=1, + ), + ) + + pretrained.act_postprocess3 = nn.Sequential( + readout_oper[2], + Transpose(1, 2), + nn.Unflatten(2, torch.Size([size[0] // 16, size[1] // 16])), + nn.Conv2d( + in_channels=vit_features, + out_channels=features[2], + kernel_size=1, + stride=1, + padding=0, + ), + ) + + pretrained.act_postprocess4 = nn.Sequential( + readout_oper[3], + Transpose(1, 2), + nn.Unflatten(2, torch.Size([size[0] // 16, size[1] // 16])), + nn.Conv2d( + in_channels=vit_features, + out_channels=features[3], + kernel_size=1, + stride=1, + padding=0, + ), + nn.Conv2d( + in_channels=features[3], + out_channels=features[3], + kernel_size=3, + stride=2, + padding=1, + ), + ) + + pretrained.model.start_index = start_index + pretrained.model.patch_size = [16, 16] + + # We inject this function into the VisionTransformer instances so that + # we can use it with interpolated position embeddings without modifying the library source. + pretrained.model.forward_flex = types.MethodType(forward_flex, pretrained.model) + pretrained.model._resize_pos_embed = types.MethodType( + _resize_pos_embed, pretrained.model + ) + + return pretrained + + +def _make_pretrained_vitl16_384(pretrained, use_readout="ignore", hooks=None): + model = timm.create_model("vit_large_patch16_384", pretrained=pretrained) + + hooks = [5, 11, 17, 23] if hooks == None else hooks + return _make_vit_b16_backbone( + model, + features=[256, 512, 1024, 1024], + hooks=hooks, + vit_features=1024, + use_readout=use_readout, + ) + + +def _make_pretrained_vitb16_384(pretrained, use_readout="ignore", hooks=None): + model = timm.create_model("vit_base_patch16_384", pretrained=pretrained) + + hooks = [2, 5, 8, 11] if hooks == None else hooks + return _make_vit_b16_backbone( + model, features=[96, 192, 384, 768], hooks=hooks, use_readout=use_readout + ) + + +def _make_pretrained_deitb16_384(pretrained, use_readout="ignore", hooks=None): + model = timm.create_model("vit_deit_base_patch16_384", pretrained=pretrained) + + hooks = [2, 5, 8, 11] if hooks == None else hooks + return _make_vit_b16_backbone( + model, features=[96, 192, 384, 768], hooks=hooks, use_readout=use_readout + ) + + +def _make_pretrained_deitb16_distil_384(pretrained, use_readout="ignore", hooks=None): + model = timm.create_model( + "vit_deit_base_distilled_patch16_384", pretrained=pretrained + ) + + hooks = [2, 5, 8, 11] if hooks == None else hooks + return _make_vit_b16_backbone( + model, + features=[96, 192, 384, 768], + hooks=hooks, + use_readout=use_readout, + start_index=2, + ) + + +def _make_vit_b_rn50_backbone( + model, + features=[256, 512, 768, 768], + size=[384, 384], + hooks=[0, 1, 8, 11], + vit_features=768, + use_vit_only=False, + use_readout="ignore", + start_index=1, +): + pretrained = nn.Module() + + pretrained.model = model + + if use_vit_only == True: + pretrained.model.blocks[hooks[0]].register_forward_hook(get_activation("1")) + pretrained.model.blocks[hooks[1]].register_forward_hook(get_activation("2")) + else: + pretrained.model.patch_embed.backbone.stages[0].register_forward_hook( + get_activation("1") + ) + pretrained.model.patch_embed.backbone.stages[1].register_forward_hook( + get_activation("2") + ) + + pretrained.model.blocks[hooks[2]].register_forward_hook(get_activation("3")) + pretrained.model.blocks[hooks[3]].register_forward_hook(get_activation("4")) + + pretrained.activations = activations + + readout_oper = get_readout_oper(vit_features, features, use_readout, start_index) + + if use_vit_only == True: + pretrained.act_postprocess1 = nn.Sequential( + readout_oper[0], + Transpose(1, 2), + nn.Unflatten(2, torch.Size([size[0] // 16, size[1] // 16])), + nn.Conv2d( + in_channels=vit_features, + out_channels=features[0], + kernel_size=1, + stride=1, + padding=0, + ), + nn.ConvTranspose2d( + in_channels=features[0], + out_channels=features[0], + kernel_size=4, + stride=4, + padding=0, + bias=True, + dilation=1, + groups=1, + ), + ) + + pretrained.act_postprocess2 = nn.Sequential( + readout_oper[1], + Transpose(1, 2), + nn.Unflatten(2, torch.Size([size[0] // 16, size[1] // 16])), + nn.Conv2d( + in_channels=vit_features, + out_channels=features[1], + kernel_size=1, + stride=1, + padding=0, + ), + nn.ConvTranspose2d( + in_channels=features[1], + out_channels=features[1], + kernel_size=2, + stride=2, + padding=0, + bias=True, + dilation=1, + groups=1, + ), + ) + else: + pretrained.act_postprocess1 = nn.Sequential( + nn.Identity(), nn.Identity(), nn.Identity() + ) + pretrained.act_postprocess2 = nn.Sequential( + nn.Identity(), nn.Identity(), nn.Identity() + ) + + pretrained.act_postprocess3 = nn.Sequential( + readout_oper[2], + Transpose(1, 2), + nn.Unflatten(2, torch.Size([size[0] // 16, size[1] // 16])), + nn.Conv2d( + in_channels=vit_features, + out_channels=features[2], + kernel_size=1, + stride=1, + padding=0, + ), + ) + + pretrained.act_postprocess4 = nn.Sequential( + readout_oper[3], + Transpose(1, 2), + nn.Unflatten(2, torch.Size([size[0] // 16, size[1] // 16])), + nn.Conv2d( + in_channels=vit_features, + out_channels=features[3], + kernel_size=1, + stride=1, + padding=0, + ), + nn.Conv2d( + in_channels=features[3], + out_channels=features[3], + kernel_size=3, + stride=2, + padding=1, + ), + ) + + pretrained.model.start_index = start_index + pretrained.model.patch_size = [16, 16] + + # We inject this function into the VisionTransformer instances so that + # we can use it with interpolated position embeddings without modifying the library source. + pretrained.model.forward_flex = types.MethodType(forward_flex, pretrained.model) + + # We inject this function into the VisionTransformer instances so that + # we can use it with interpolated position embeddings without modifying the library source. + pretrained.model._resize_pos_embed = types.MethodType( + _resize_pos_embed, pretrained.model + ) + + return pretrained + + +def _make_pretrained_vitb_rn50_384( + pretrained, use_readout="ignore", hooks=None, use_vit_only=False +): + model = timm.create_model("vit_base_resnet50_384", pretrained=pretrained) + + hooks = [0, 1, 8, 11] if hooks == None else hooks + return _make_vit_b_rn50_backbone( + model, + features=[256, 512, 768, 768], + size=[384, 384], + hooks=hooks, + use_vit_only=use_vit_only, + use_readout=use_readout, + ) diff --git a/ldm/modules/midas/utils.py b/ldm/modules/midas/utils.py new file mode 100644 index 0000000000000000000000000000000000000000..9a9d3b5b66370fa98da9e067ba53ead848ea9a59 --- /dev/null +++ b/ldm/modules/midas/utils.py @@ -0,0 +1,189 @@ +"""Utils for monoDepth.""" +import sys +import re +import numpy as np +import cv2 +import torch + + +def read_pfm(path): + """Read pfm file. + + Args: + path (str): path to file + + Returns: + tuple: (data, scale) + """ + with open(path, "rb") as file: + + color = None + width = None + height = None + scale = None + endian = None + + header = file.readline().rstrip() + if header.decode("ascii") == "PF": + color = True + elif header.decode("ascii") == "Pf": + color = False + else: + raise Exception("Not a PFM file: " + path) + + dim_match = re.match(r"^(\d+)\s(\d+)\s$", file.readline().decode("ascii")) + if dim_match: + width, height = list(map(int, dim_match.groups())) + else: + raise Exception("Malformed PFM header.") + + scale = float(file.readline().decode("ascii").rstrip()) + if scale < 0: + # little-endian + endian = "<" + scale = -scale + else: + # big-endian + endian = ">" + + data = np.fromfile(file, endian + "f") + shape = (height, width, 3) if color else (height, width) + + data = np.reshape(data, shape) + data = np.flipud(data) + + return data, scale + + +def write_pfm(path, image, scale=1): + """Write pfm file. + + Args: + path (str): pathto file + image (array): data + scale (int, optional): Scale. Defaults to 1. + """ + + with open(path, "wb") as file: + color = None + + if image.dtype.name != "float32": + raise Exception("Image dtype must be float32.") + + image = np.flipud(image) + + if len(image.shape) == 3 and image.shape[2] == 3: # color image + color = True + elif ( + len(image.shape) == 2 or len(image.shape) == 3 and image.shape[2] == 1 + ): # greyscale + color = False + else: + raise Exception("Image must have H x W x 3, H x W x 1 or H x W dimensions.") + + file.write("PF\n" if color else "Pf\n".encode()) + file.write("%d %d\n".encode() % (image.shape[1], image.shape[0])) + + endian = image.dtype.byteorder + + if endian == "<" or endian == "=" and sys.byteorder == "little": + scale = -scale + + file.write("%f\n".encode() % scale) + + image.tofile(file) + + +def read_image(path): + """Read image and output RGB image (0-1). + + Args: + path (str): path to file + + Returns: + array: RGB image (0-1) + """ + img = cv2.imread(path) + + if img.ndim == 2: + img = cv2.cvtColor(img, cv2.COLOR_GRAY2BGR) + + img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB) / 255.0 + + return img + + +def resize_image(img): + """Resize image and make it fit for network. + + Args: + img (array): image + + Returns: + tensor: data ready for network + """ + height_orig = img.shape[0] + width_orig = img.shape[1] + + if width_orig > height_orig: + scale = width_orig / 384 + else: + scale = height_orig / 384 + + height = (np.ceil(height_orig / scale / 32) * 32).astype(int) + width = (np.ceil(width_orig / scale / 32) * 32).astype(int) + + img_resized = cv2.resize(img, (width, height), interpolation=cv2.INTER_AREA) + + img_resized = ( + torch.from_numpy(np.transpose(img_resized, (2, 0, 1))).contiguous().float() + ) + img_resized = img_resized.unsqueeze(0) + + return img_resized + + +def resize_depth(depth, width, height): + """Resize depth map and bring to CPU (numpy). + + Args: + depth (tensor): depth + width (int): image width + height (int): image height + + Returns: + array: processed depth + """ + depth = torch.squeeze(depth[0, :, :, :]).to("cpu") + + depth_resized = cv2.resize( + depth.numpy(), (width, height), interpolation=cv2.INTER_CUBIC + ) + + return depth_resized + +def write_depth(path, depth, bits=1): + """Write depth map to pfm and png file. + + Args: + path (str): filepath without extension + depth (array): depth + """ + write_pfm(path + ".pfm", depth.astype(np.float32)) + + depth_min = depth.min() + depth_max = depth.max() + + max_val = (2**(8*bits))-1 + + if depth_max - depth_min > np.finfo("float").eps: + out = max_val * (depth - depth_min) / (depth_max - depth_min) + else: + out = np.zeros(depth.shape, dtype=depth.type) + + if bits == 1: + cv2.imwrite(path + ".png", out.astype("uint8")) + elif bits == 2: + cv2.imwrite(path + ".png", out.astype("uint16")) + + return diff --git a/ldm/util.py b/ldm/util.py new file mode 100644 index 0000000000000000000000000000000000000000..d456a86a60968788abf9f5235a41fa826ba578dc --- /dev/null +++ b/ldm/util.py @@ -0,0 +1,197 @@ +import importlib + +import torch +from torch import optim +import numpy as np + +from inspect import isfunction +from PIL import Image, ImageDraw, ImageFont + + +def log_txt_as_img(wh, xc, size=10): + # wh a tuple of (width, height) + # xc a list of captions to plot + b = len(xc) + txts = list() + for bi in range(b): + txt = Image.new("RGB", wh, color="white") + draw = ImageDraw.Draw(txt) + font = ImageFont.truetype('font/Arial_Unicode.ttf', size=size) + nc = int(32 * (wh[0] / 256)) + lines = "\n".join(xc[bi][start:start + nc] for start in range(0, len(xc[bi]), nc)) + + try: + draw.text((0, 0), lines, fill="black", font=font) + except UnicodeEncodeError: + print("Cant encode string for logging. Skipping.") + + txt = np.array(txt).transpose(2, 0, 1) / 127.5 - 1.0 + txts.append(txt) + txts = np.stack(txts) + txts = torch.tensor(txts) + return txts + + +def ismap(x): + if not isinstance(x, torch.Tensor): + return False + return (len(x.shape) == 4) and (x.shape[1] > 3) + + +def isimage(x): + if not isinstance(x,torch.Tensor): + return False + return (len(x.shape) == 4) and (x.shape[1] == 3 or x.shape[1] == 1) + + +def exists(x): + return x is not None + + +def default(val, d): + if exists(val): + return val + return d() if isfunction(d) else d + + +def mean_flat(tensor): + """ + https://github.com/openai/guided-diffusion/blob/27c20a8fab9cb472df5d6bdd6c8d11c8f430b924/guided_diffusion/nn.py#L86 + Take the mean over all non-batch dimensions. + """ + return tensor.mean(dim=list(range(1, len(tensor.shape)))) + + +def count_params(model, verbose=False): + total_params = sum(p.numel() for p in model.parameters()) + if verbose: + print(f"{model.__class__.__name__} has {total_params*1.e-6:.2f} M params.") + return total_params + + +def instantiate_from_config(config, **kwargs): + if "target" not in config: + if config == '__is_first_stage__': + return None + elif config == "__is_unconditional__": + return None + raise KeyError("Expected key `target` to instantiate.") + return get_obj_from_str(config["target"])(**config.get("params", dict()), **kwargs) + + +def get_obj_from_str(string, reload=False): + module, cls = string.rsplit(".", 1) + if reload: + module_imp = importlib.import_module(module) + importlib.reload(module_imp) + return getattr(importlib.import_module(module, package=None), cls) + + +class AdamWwithEMAandWings(optim.Optimizer): + # credit to https://gist.github.com/crowsonkb/65f7265353f403714fce3b2595e0b298 + def __init__(self, params, lr=1.e-3, betas=(0.9, 0.999), eps=1.e-8, # TODO: check hyperparameters before using + weight_decay=1.e-2, amsgrad=False, ema_decay=0.9999, # ema decay to match previous code + ema_power=1., param_names=()): + """AdamW that saves EMA versions of the parameters.""" + if not 0.0 <= lr: + raise ValueError("Invalid learning rate: {}".format(lr)) + if not 0.0 <= eps: + raise ValueError("Invalid epsilon value: {}".format(eps)) + if not 0.0 <= betas[0] < 1.0: + raise ValueError("Invalid beta parameter at index 0: {}".format(betas[0])) + if not 0.0 <= betas[1] < 1.0: + raise ValueError("Invalid beta parameter at index 1: {}".format(betas[1])) + if not 0.0 <= weight_decay: + raise ValueError("Invalid weight_decay value: {}".format(weight_decay)) + if not 0.0 <= ema_decay <= 1.0: + raise ValueError("Invalid ema_decay value: {}".format(ema_decay)) + defaults = dict(lr=lr, betas=betas, eps=eps, + weight_decay=weight_decay, amsgrad=amsgrad, ema_decay=ema_decay, + ema_power=ema_power, param_names=param_names) + super().__init__(params, defaults) + + def __setstate__(self, state): + super().__setstate__(state) + for group in self.param_groups: + group.setdefault('amsgrad', False) + + @torch.no_grad() + def step(self, closure=None): + """Performs a single optimization step. + Args: + closure (callable, optional): A closure that reevaluates the model + and returns the loss. + """ + loss = None + if closure is not None: + with torch.enable_grad(): + loss = closure() + + for group in self.param_groups: + params_with_grad = [] + grads = [] + exp_avgs = [] + exp_avg_sqs = [] + ema_params_with_grad = [] + state_sums = [] + max_exp_avg_sqs = [] + state_steps = [] + amsgrad = group['amsgrad'] + beta1, beta2 = group['betas'] + ema_decay = group['ema_decay'] + ema_power = group['ema_power'] + + for p in group['params']: + if p.grad is None: + continue + params_with_grad.append(p) + if p.grad.is_sparse: + raise RuntimeError('AdamW does not support sparse gradients') + grads.append(p.grad) + + state = self.state[p] + + # State initialization + if len(state) == 0: + state['step'] = 0 + # Exponential moving average of gradient values + state['exp_avg'] = torch.zeros_like(p, memory_format=torch.preserve_format) + # Exponential moving average of squared gradient values + state['exp_avg_sq'] = torch.zeros_like(p, memory_format=torch.preserve_format) + if amsgrad: + # Maintains max of all exp. moving avg. of sq. grad. values + state['max_exp_avg_sq'] = torch.zeros_like(p, memory_format=torch.preserve_format) + # Exponential moving average of parameter values + state['param_exp_avg'] = p.detach().float().clone() + + exp_avgs.append(state['exp_avg']) + exp_avg_sqs.append(state['exp_avg_sq']) + ema_params_with_grad.append(state['param_exp_avg']) + + if amsgrad: + max_exp_avg_sqs.append(state['max_exp_avg_sq']) + + # update the steps for each param group update + state['step'] += 1 + # record the step after step update + state_steps.append(state['step']) + + optim._functional.adamw(params_with_grad, + grads, + exp_avgs, + exp_avg_sqs, + max_exp_avg_sqs, + state_steps, + amsgrad=amsgrad, + beta1=beta1, + beta2=beta2, + lr=group['lr'], + weight_decay=group['weight_decay'], + eps=group['eps'], + maximize=False) + + cur_ema_decay = min(ema_decay, 1 - state['step'] ** -ema_power) + for param, ema_param in zip(params_with_grad, ema_params_with_grad): + ema_param.mul_(cur_ema_decay).add_(param.float(), alpha=1 - cur_ema_decay) + + return loss \ No newline at end of file diff --git a/models_yaml/anytext_sd15.yaml b/models_yaml/anytext_sd15.yaml new file mode 100644 index 0000000000000000000000000000000000000000..3509940b54e87a7c564bf31f638e2df3f68301f9 --- /dev/null +++ b/models_yaml/anytext_sd15.yaml @@ -0,0 +1,99 @@ +model: + target: cldm.cldm.ControlLDM + params: + linear_start: 0.00085 + linear_end: 0.0120 + num_timesteps_cond: 1 + log_every_t: 200 + timesteps: 1000 + first_stage_key: "img" + cond_stage_key: "caption" + control_key: "hint" + glyph_key: "glyphs" + position_key: "positions" + image_size: 64 + channels: 4 + cond_stage_trainable: true # need be true when embedding_manager is valid + conditioning_key: crossattn + monitor: val/loss_simple_ema + scale_factor: 0.18215 + use_ema: False + only_mid_control: False + loss_alpha: 0 # perceptual loss, 0.003 + loss_beta: 0 # ctc loss + latin_weight: 1.0 # latin text line may need smaller weigth + with_step_weight: true + use_vae_upsample: true + embedding_manager_config: + target: cldm.embedding_manager.EmbeddingManager + params: + valid: true # v6 + emb_type: ocr # ocr, vit, conv + glyph_channels: 1 + position_channels: 1 + add_pos: false + placeholder_string: '*' + + control_stage_config: + target: cldm.cldm.ControlNet + params: + image_size: 32 # unused + in_channels: 4 + model_channels: 320 + glyph_channels: 1 + position_channels: 1 + attention_resolutions: [ 4, 2, 1 ] + num_res_blocks: 2 + channel_mult: [ 1, 2, 4, 4 ] + num_heads: 8 + use_spatial_transformer: True + transformer_depth: 1 + context_dim: 768 + use_checkpoint: True + legacy: False + + unet_config: + target: cldm.cldm.ControlledUnetModel + params: + image_size: 32 # unused + in_channels: 4 + out_channels: 4 + model_channels: 320 + attention_resolutions: [ 4, 2, 1 ] + num_res_blocks: 2 + channel_mult: [ 1, 2, 4, 4 ] + num_heads: 8 + use_spatial_transformer: True + transformer_depth: 1 + context_dim: 768 + use_checkpoint: True + legacy: False + + first_stage_config: + target: ldm.models.autoencoder.AutoencoderKL + params: + embed_dim: 4 + monitor: val/rec_loss + ddconfig: + double_z: true + z_channels: 4 + resolution: 256 + in_channels: 3 + out_ch: 3 + ch: 128 + ch_mult: + - 1 + - 2 + - 4 + - 4 + num_res_blocks: 2 + attn_resolutions: [] + dropout: 0.0 + lossconfig: + target: torch.nn.Identity + + cond_stage_config: + target: ldm.modules.encoders.modules.FrozenCLIPEmbedderT3 + params: + version: ./models/clip-vit-large-patch14 + use_vision: false # v6 diff --git a/models_yaml/anytext_sd15_conv.yaml b/models_yaml/anytext_sd15_conv.yaml new file mode 100644 index 0000000000000000000000000000000000000000..025158f3a0a864f4f390fe6d6bccb88bdbb13e3d --- /dev/null +++ b/models_yaml/anytext_sd15_conv.yaml @@ -0,0 +1,99 @@ +model: + target: cldm.cldm.ControlLDM + params: + linear_start: 0.00085 + linear_end: 0.0120 + num_timesteps_cond: 1 + log_every_t: 200 + timesteps: 1000 + first_stage_key: "img" + cond_stage_key: "caption" + control_key: "hint" + glyph_key: "glyphs" + position_key: "positions" + image_size: 64 + channels: 4 + cond_stage_trainable: true # need be true when embedding_manager is valid + conditioning_key: crossattn + monitor: val/loss_simple_ema + scale_factor: 0.18215 + use_ema: False + only_mid_control: False + loss_alpha: 0 # perceptual loss, 0.003 + loss_beta: 0 # ctc loss + latin_weight: 1.0 # latin text line may need smaller weigth + with_step_weight: true + use_vae_upsample: true + embedding_manager_config: + target: cldm.embedding_manager.EmbeddingManager + params: + valid: true # v6 + emb_type: conv # ocr, vit, conv + glyph_channels: 1 + position_channels: 1 + add_pos: false + placeholder_string: '*' + + control_stage_config: + target: cldm.cldm.ControlNet + params: + image_size: 32 # unused + in_channels: 4 + model_channels: 320 + glyph_channels: 1 + position_channels: 1 + attention_resolutions: [ 4, 2, 1 ] + num_res_blocks: 2 + channel_mult: [ 1, 2, 4, 4 ] + num_heads: 8 + use_spatial_transformer: True + transformer_depth: 1 + context_dim: 768 + use_checkpoint: True + legacy: False + + unet_config: + target: cldm.cldm.ControlledUnetModel + params: + image_size: 32 # unused + in_channels: 4 + out_channels: 4 + model_channels: 320 + attention_resolutions: [ 4, 2, 1 ] + num_res_blocks: 2 + channel_mult: [ 1, 2, 4, 4 ] + num_heads: 8 + use_spatial_transformer: True + transformer_depth: 1 + context_dim: 768 + use_checkpoint: True + legacy: False + + first_stage_config: + target: ldm.models.autoencoder.AutoencoderKL + params: + embed_dim: 4 + monitor: val/rec_loss + ddconfig: + double_z: true + z_channels: 4 + resolution: 256 + in_channels: 3 + out_ch: 3 + ch: 128 + ch_mult: + - 1 + - 2 + - 4 + - 4 + num_res_blocks: 2 + attn_resolutions: [] + dropout: 0.0 + lossconfig: + target: torch.nn.Identity + + cond_stage_config: + target: ldm.modules.encoders.modules.FrozenCLIPEmbedderT3 + params: + version: ./models/clip-vit-large-patch14 + use_vision: false # v6 diff --git a/models_yaml/anytext_sd15_perloss.yaml b/models_yaml/anytext_sd15_perloss.yaml new file mode 100644 index 0000000000000000000000000000000000000000..7b9a49d0b212aedb4f5d51e2abdcc019fa221491 --- /dev/null +++ b/models_yaml/anytext_sd15_perloss.yaml @@ -0,0 +1,99 @@ +model: + target: cldm.cldm.ControlLDM + params: + linear_start: 0.00085 + linear_end: 0.0120 + num_timesteps_cond: 1 + log_every_t: 200 + timesteps: 1000 + first_stage_key: "img" + cond_stage_key: "caption" + control_key: "hint" + glyph_key: "glyphs" + position_key: "positions" + image_size: 64 + channels: 4 + cond_stage_trainable: true # need be true when embedding_manager is valid + conditioning_key: crossattn + monitor: val/loss_simple_ema + scale_factor: 0.18215 + use_ema: False + only_mid_control: False + loss_alpha: 0.003 # perceptual loss, 0.003 + loss_beta: 0 # ctc loss + latin_weight: 1.0 # latin text line may need smaller weigth + with_step_weight: true + use_vae_upsample: true + embedding_manager_config: + target: cldm.embedding_manager.EmbeddingManager + params: + valid: true # v6 + emb_type: ocr # ocr, vit, conv + glyph_channels: 1 + position_channels: 1 + add_pos: false + placeholder_string: '*' + + control_stage_config: + target: cldm.cldm.ControlNet + params: + image_size: 32 # unused + in_channels: 4 + model_channels: 320 + glyph_channels: 1 + position_channels: 1 + attention_resolutions: [ 4, 2, 1 ] + num_res_blocks: 2 + channel_mult: [ 1, 2, 4, 4 ] + num_heads: 8 + use_spatial_transformer: True + transformer_depth: 1 + context_dim: 768 + use_checkpoint: True + legacy: False + + unet_config: + target: cldm.cldm.ControlledUnetModel + params: + image_size: 32 # unused + in_channels: 4 + out_channels: 4 + model_channels: 320 + attention_resolutions: [ 4, 2, 1 ] + num_res_blocks: 2 + channel_mult: [ 1, 2, 4, 4 ] + num_heads: 8 + use_spatial_transformer: True + transformer_depth: 1 + context_dim: 768 + use_checkpoint: True + legacy: False + + first_stage_config: + target: ldm.models.autoencoder.AutoencoderKL + params: + embed_dim: 4 + monitor: val/rec_loss + ddconfig: + double_z: true + z_channels: 4 + resolution: 256 + in_channels: 3 + out_ch: 3 + ch: 128 + ch_mult: + - 1 + - 2 + - 4 + - 4 + num_res_blocks: 2 + attn_resolutions: [] + dropout: 0.0 + lossconfig: + target: torch.nn.Identity + + cond_stage_config: + target: ldm.modules.encoders.modules.FrozenCLIPEmbedderT3 + params: + version: ./models/clip-vit-large-patch14 + use_vision: false # v6 diff --git a/models_yaml/anytext_sd15_vit.yaml b/models_yaml/anytext_sd15_vit.yaml new file mode 100644 index 0000000000000000000000000000000000000000..831d3227ccb6182b4916ce5921abf4c2ae97d576 --- /dev/null +++ b/models_yaml/anytext_sd15_vit.yaml @@ -0,0 +1,99 @@ +model: + target: cldm.cldm.ControlLDM + params: + linear_start: 0.00085 + linear_end: 0.0120 + num_timesteps_cond: 1 + log_every_t: 200 + timesteps: 1000 + first_stage_key: "img" + cond_stage_key: "caption" + control_key: "hint" + glyph_key: "glyphs" + position_key: "positions" + image_size: 64 + channels: 4 + cond_stage_trainable: true # need be true when embedding_manager is valid + conditioning_key: crossattn + monitor: val/loss_simple_ema + scale_factor: 0.18215 + use_ema: False + only_mid_control: False + loss_alpha: 0 # perceptual loss, 0.003 + loss_beta: 0 # ctc loss + latin_weight: 1.0 # latin text line may need smaller weigth + with_step_weight: true + use_vae_upsample: true + embedding_manager_config: + target: cldm.embedding_manager.EmbeddingManager + params: + valid: true # v6 + emb_type: vit # ocr, vit, conv + glyph_channels: 1 + position_channels: 1 + add_pos: false + placeholder_string: '*' + + control_stage_config: + target: cldm.cldm.ControlNet + params: + image_size: 32 # unused + in_channels: 4 + model_channels: 320 + glyph_channels: 1 + position_channels: 1 + attention_resolutions: [ 4, 2, 1 ] + num_res_blocks: 2 + channel_mult: [ 1, 2, 4, 4 ] + num_heads: 8 + use_spatial_transformer: True + transformer_depth: 1 + context_dim: 768 + use_checkpoint: True + legacy: False + + unet_config: + target: cldm.cldm.ControlledUnetModel + params: + image_size: 32 # unused + in_channels: 4 + out_channels: 4 + model_channels: 320 + attention_resolutions: [ 4, 2, 1 ] + num_res_blocks: 2 + channel_mult: [ 1, 2, 4, 4 ] + num_heads: 8 + use_spatial_transformer: True + transformer_depth: 1 + context_dim: 768 + use_checkpoint: True + legacy: False + + first_stage_config: + target: ldm.models.autoencoder.AutoencoderKL + params: + embed_dim: 4 + monitor: val/rec_loss + ddconfig: + double_z: true + z_channels: 4 + resolution: 256 + in_channels: 3 + out_ch: 3 + ch: 128 + ch_mult: + - 1 + - 2 + - 4 + - 4 + num_res_blocks: 2 + attn_resolutions: [] + dropout: 0.0 + lossconfig: + target: torch.nn.Identity + + cond_stage_config: + target: ldm.modules.encoders.modules.FrozenCLIPEmbedderT3 + params: + version: ./models/clip-vit-large-patch14 + use_vision: true # v6 diff --git a/ocr_recog/RNN.py b/ocr_recog/RNN.py new file mode 100755 index 0000000000000000000000000000000000000000..cf16855b37112c34a722a9ae7d21578a82d8c6d8 --- /dev/null +++ b/ocr_recog/RNN.py @@ -0,0 +1,210 @@ +from torch import nn +import torch +from .RecSVTR import Block + +class Swish(nn.Module): + def __int__(self): + super(Swish, self).__int__() + + def forward(self,x): + return x*torch.sigmoid(x) + +class Im2Im(nn.Module): + def __init__(self, in_channels, **kwargs): + super().__init__() + self.out_channels = in_channels + + def forward(self, x): + return x + +class Im2Seq(nn.Module): + def __init__(self, in_channels, **kwargs): + super().__init__() + self.out_channels = in_channels + + def forward(self, x): + B, C, H, W = x.shape + # assert H == 1 + x = x.reshape(B, C, H * W) + x = x.permute((0, 2, 1)) + return x + +class EncoderWithRNN(nn.Module): + def __init__(self, in_channels,**kwargs): + super(EncoderWithRNN, self).__init__() + hidden_size = kwargs.get('hidden_size', 256) + self.out_channels = hidden_size * 2 + self.lstm = nn.LSTM(in_channels, hidden_size, bidirectional=True, num_layers=2,batch_first=True) + + def forward(self, x): + self.lstm.flatten_parameters() + x, _ = self.lstm(x) + return x + +class SequenceEncoder(nn.Module): + def __init__(self, in_channels, encoder_type='rnn', **kwargs): + super(SequenceEncoder, self).__init__() + self.encoder_reshape = Im2Seq(in_channels) + self.out_channels = self.encoder_reshape.out_channels + self.encoder_type = encoder_type + if encoder_type == 'reshape': + self.only_reshape = True + else: + support_encoder_dict = { + 'reshape': Im2Seq, + 'rnn': EncoderWithRNN, + 'svtr': EncoderWithSVTR + } + assert encoder_type in support_encoder_dict, '{} must in {}'.format( + encoder_type, support_encoder_dict.keys()) + + self.encoder = support_encoder_dict[encoder_type]( + self.encoder_reshape.out_channels,**kwargs) + self.out_channels = self.encoder.out_channels + self.only_reshape = False + + def forward(self, x): + if self.encoder_type != 'svtr': + x = self.encoder_reshape(x) + if not self.only_reshape: + x = self.encoder(x) + return x + else: + x = self.encoder(x) + x = self.encoder_reshape(x) + return x + +class ConvBNLayer(nn.Module): + def __init__(self, + in_channels, + out_channels, + kernel_size=3, + stride=1, + padding=0, + bias_attr=False, + groups=1, + act=nn.GELU): + super().__init__() + self.conv = nn.Conv2d( + in_channels=in_channels, + out_channels=out_channels, + kernel_size=kernel_size, + stride=stride, + padding=padding, + groups=groups, + # weight_attr=paddle.ParamAttr(initializer=nn.initializer.KaimingUniform()), + bias=bias_attr) + self.norm = nn.BatchNorm2d(out_channels) + self.act = Swish() + + def forward(self, inputs): + out = self.conv(inputs) + out = self.norm(out) + out = self.act(out) + return out + + +class EncoderWithSVTR(nn.Module): + def __init__( + self, + in_channels, + dims=64, # XS + depth=2, + hidden_dims=120, + use_guide=False, + num_heads=8, + qkv_bias=True, + mlp_ratio=2.0, + drop_rate=0.1, + attn_drop_rate=0.1, + drop_path=0., + qk_scale=None): + super(EncoderWithSVTR, self).__init__() + self.depth = depth + self.use_guide = use_guide + self.conv1 = ConvBNLayer( + in_channels, in_channels // 8, padding=1, act='swish') + self.conv2 = ConvBNLayer( + in_channels // 8, hidden_dims, kernel_size=1, act='swish') + + self.svtr_block = nn.ModuleList([ + Block( + dim=hidden_dims, + num_heads=num_heads, + mixer='Global', + HW=None, + mlp_ratio=mlp_ratio, + qkv_bias=qkv_bias, + qk_scale=qk_scale, + drop=drop_rate, + act_layer='swish', + attn_drop=attn_drop_rate, + drop_path=drop_path, + norm_layer='nn.LayerNorm', + epsilon=1e-05, + prenorm=False) for i in range(depth) + ]) + self.norm = nn.LayerNorm(hidden_dims, eps=1e-6) + self.conv3 = ConvBNLayer( + hidden_dims, in_channels, kernel_size=1, act='swish') + # last conv-nxn, the input is concat of input tensor and conv3 output tensor + self.conv4 = ConvBNLayer( + 2 * in_channels, in_channels // 8, padding=1, act='swish') + + self.conv1x1 = ConvBNLayer( + in_channels // 8, dims, kernel_size=1, act='swish') + self.out_channels = dims + self.apply(self._init_weights) + + def _init_weights(self, m): + # weight initialization + if isinstance(m, nn.Conv2d): + nn.init.kaiming_normal_(m.weight, mode='fan_out') + if m.bias is not None: + nn.init.zeros_(m.bias) + elif isinstance(m, nn.BatchNorm2d): + nn.init.ones_(m.weight) + nn.init.zeros_(m.bias) + elif isinstance(m, nn.Linear): + nn.init.normal_(m.weight, 0, 0.01) + if m.bias is not None: + nn.init.zeros_(m.bias) + elif isinstance(m, nn.ConvTranspose2d): + nn.init.kaiming_normal_(m.weight, mode='fan_out') + if m.bias is not None: + nn.init.zeros_(m.bias) + elif isinstance(m, nn.LayerNorm): + nn.init.ones_(m.weight) + nn.init.zeros_(m.bias) + + def forward(self, x): + # for use guide + if self.use_guide: + z = x.clone() + z.stop_gradient = True + else: + z = x + # for short cut + h = z + # reduce dim + z = self.conv1(z) + z = self.conv2(z) + # SVTR global block + B, C, H, W = z.shape + z = z.flatten(2).permute(0, 2, 1) + + for blk in self.svtr_block: + z = blk(z) + + z = self.norm(z) + # last stage + z = z.reshape([-1, H, W, C]).permute(0, 3, 1, 2) + z = self.conv3(z) + z = torch.cat((h, z), dim=1) + z = self.conv1x1(self.conv4(z)) + + return z + +if __name__=="__main__": + svtrRNN = EncoderWithSVTR(56) + print(svtrRNN) \ No newline at end of file diff --git a/ocr_recog/RecCTCHead.py b/ocr_recog/RecCTCHead.py new file mode 100755 index 0000000000000000000000000000000000000000..867ede9916b10d49dc18e8633ae1cb3c4c87ad9d --- /dev/null +++ b/ocr_recog/RecCTCHead.py @@ -0,0 +1,48 @@ +from torch import nn + + +class CTCHead(nn.Module): + def __init__(self, + in_channels, + out_channels=6625, + fc_decay=0.0004, + mid_channels=None, + return_feats=False, + **kwargs): + super(CTCHead, self).__init__() + if mid_channels is None: + self.fc = nn.Linear( + in_channels, + out_channels, + bias=True,) + else: + self.fc1 = nn.Linear( + in_channels, + mid_channels, + bias=True, + ) + self.fc2 = nn.Linear( + mid_channels, + out_channels, + bias=True, + ) + + self.out_channels = out_channels + self.mid_channels = mid_channels + self.return_feats = return_feats + + def forward(self, x, labels=None): + if self.mid_channels is None: + predicts = self.fc(x) + else: + x = self.fc1(x) + predicts = self.fc2(x) + + if self.return_feats: + result = dict() + result['ctc'] = predicts + result['ctc_neck'] = x + else: + result = predicts + + return result diff --git a/ocr_recog/RecModel.py b/ocr_recog/RecModel.py new file mode 100755 index 0000000000000000000000000000000000000000..c2313bf02c952d7c5351175ccf36482cddc76cac --- /dev/null +++ b/ocr_recog/RecModel.py @@ -0,0 +1,45 @@ +from torch import nn +from .RNN import SequenceEncoder, Im2Seq, Im2Im +from .RecMv1_enhance import MobileNetV1Enhance + +from .RecCTCHead import CTCHead + +backbone_dict = {"MobileNetV1Enhance":MobileNetV1Enhance} +neck_dict = {'SequenceEncoder': SequenceEncoder, 'Im2Seq': Im2Seq,'None':Im2Im} +head_dict = {'CTCHead':CTCHead} + + +class RecModel(nn.Module): + def __init__(self, config): + super().__init__() + assert 'in_channels' in config, 'in_channels must in model config' + backbone_type = config.backbone.pop('type') + assert backbone_type in backbone_dict, f'backbone.type must in {backbone_dict}' + self.backbone = backbone_dict[backbone_type](config.in_channels, **config.backbone) + + neck_type = config.neck.pop('type') + assert neck_type in neck_dict, f'neck.type must in {neck_dict}' + self.neck = neck_dict[neck_type](self.backbone.out_channels, **config.neck) + + head_type = config.head.pop('type') + assert head_type in head_dict, f'head.type must in {head_dict}' + self.head = head_dict[head_type](self.neck.out_channels, **config.head) + + self.name = f'RecModel_{backbone_type}_{neck_type}_{head_type}' + + def load_3rd_state_dict(self, _3rd_name, _state): + self.backbone.load_3rd_state_dict(_3rd_name, _state) + self.neck.load_3rd_state_dict(_3rd_name, _state) + self.head.load_3rd_state_dict(_3rd_name, _state) + + def forward(self, x): + x = self.backbone(x) + x = self.neck(x) + x = self.head(x) + return x + + def encode(self, x): + x = self.backbone(x) + x = self.neck(x) + x = self.head.ctc_encoder(x) + return x diff --git a/ocr_recog/RecMv1_enhance.py b/ocr_recog/RecMv1_enhance.py new file mode 100644 index 0000000000000000000000000000000000000000..d5c848533dd35108bc963be4e6f0ea8c76564006 --- /dev/null +++ b/ocr_recog/RecMv1_enhance.py @@ -0,0 +1,233 @@ +import os, sys +import torch +import torch.nn as nn +import torch.nn.functional as F +from .common import Activation + + +class ConvBNLayer(nn.Module): + def __init__(self, + num_channels, + filter_size, + num_filters, + stride, + padding, + channels=None, + num_groups=1, + act='hard_swish'): + super(ConvBNLayer, self).__init__() + self.act = act + self._conv = nn.Conv2d( + in_channels=num_channels, + out_channels=num_filters, + kernel_size=filter_size, + stride=stride, + padding=padding, + groups=num_groups, + bias=False) + + self._batch_norm = nn.BatchNorm2d( + num_filters, + ) + if self.act is not None: + self._act = Activation(act_type=act, inplace=True) + + def forward(self, inputs): + y = self._conv(inputs) + y = self._batch_norm(y) + if self.act is not None: + y = self._act(y) + return y + + +class DepthwiseSeparable(nn.Module): + def __init__(self, + num_channels, + num_filters1, + num_filters2, + num_groups, + stride, + scale, + dw_size=3, + padding=1, + use_se=False): + super(DepthwiseSeparable, self).__init__() + self.use_se = use_se + self._depthwise_conv = ConvBNLayer( + num_channels=num_channels, + num_filters=int(num_filters1 * scale), + filter_size=dw_size, + stride=stride, + padding=padding, + num_groups=int(num_groups * scale)) + if use_se: + self._se = SEModule(int(num_filters1 * scale)) + self._pointwise_conv = ConvBNLayer( + num_channels=int(num_filters1 * scale), + filter_size=1, + num_filters=int(num_filters2 * scale), + stride=1, + padding=0) + + def forward(self, inputs): + y = self._depthwise_conv(inputs) + if self.use_se: + y = self._se(y) + y = self._pointwise_conv(y) + return y + + +class MobileNetV1Enhance(nn.Module): + def __init__(self, + in_channels=3, + scale=0.5, + last_conv_stride=1, + last_pool_type='max', + **kwargs): + super().__init__() + self.scale = scale + self.block_list = [] + + self.conv1 = ConvBNLayer( + num_channels=in_channels, + filter_size=3, + channels=3, + num_filters=int(32 * scale), + stride=2, + padding=1) + + conv2_1 = DepthwiseSeparable( + num_channels=int(32 * scale), + num_filters1=32, + num_filters2=64, + num_groups=32, + stride=1, + scale=scale) + self.block_list.append(conv2_1) + + conv2_2 = DepthwiseSeparable( + num_channels=int(64 * scale), + num_filters1=64, + num_filters2=128, + num_groups=64, + stride=1, + scale=scale) + self.block_list.append(conv2_2) + + conv3_1 = DepthwiseSeparable( + num_channels=int(128 * scale), + num_filters1=128, + num_filters2=128, + num_groups=128, + stride=1, + scale=scale) + self.block_list.append(conv3_1) + + conv3_2 = DepthwiseSeparable( + num_channels=int(128 * scale), + num_filters1=128, + num_filters2=256, + num_groups=128, + stride=(2, 1), + scale=scale) + self.block_list.append(conv3_2) + + conv4_1 = DepthwiseSeparable( + num_channels=int(256 * scale), + num_filters1=256, + num_filters2=256, + num_groups=256, + stride=1, + scale=scale) + self.block_list.append(conv4_1) + + conv4_2 = DepthwiseSeparable( + num_channels=int(256 * scale), + num_filters1=256, + num_filters2=512, + num_groups=256, + stride=(2, 1), + scale=scale) + self.block_list.append(conv4_2) + + for _ in range(5): + conv5 = DepthwiseSeparable( + num_channels=int(512 * scale), + num_filters1=512, + num_filters2=512, + num_groups=512, + stride=1, + dw_size=5, + padding=2, + scale=scale, + use_se=False) + self.block_list.append(conv5) + + conv5_6 = DepthwiseSeparable( + num_channels=int(512 * scale), + num_filters1=512, + num_filters2=1024, + num_groups=512, + stride=(2, 1), + dw_size=5, + padding=2, + scale=scale, + use_se=True) + self.block_list.append(conv5_6) + + conv6 = DepthwiseSeparable( + num_channels=int(1024 * scale), + num_filters1=1024, + num_filters2=1024, + num_groups=1024, + stride=last_conv_stride, + dw_size=5, + padding=2, + use_se=True, + scale=scale) + self.block_list.append(conv6) + + self.block_list = nn.Sequential(*self.block_list) + if last_pool_type == 'avg': + self.pool = nn.AvgPool2d(kernel_size=2, stride=2, padding=0) + else: + self.pool = nn.MaxPool2d(kernel_size=2, stride=2, padding=0) + self.out_channels = int(1024 * scale) + + def forward(self, inputs): + y = self.conv1(inputs) + y = self.block_list(y) + y = self.pool(y) + return y + +def hardsigmoid(x): + return F.relu6(x + 3., inplace=True) / 6. + +class SEModule(nn.Module): + def __init__(self, channel, reduction=4): + super(SEModule, self).__init__() + self.avg_pool = nn.AdaptiveAvgPool2d(1) + self.conv1 = nn.Conv2d( + in_channels=channel, + out_channels=channel // reduction, + kernel_size=1, + stride=1, + padding=0, + bias=True) + self.conv2 = nn.Conv2d( + in_channels=channel // reduction, + out_channels=channel, + kernel_size=1, + stride=1, + padding=0, + bias=True) + + def forward(self, inputs): + outputs = self.avg_pool(inputs) + outputs = self.conv1(outputs) + outputs = F.relu(outputs) + outputs = self.conv2(outputs) + outputs = hardsigmoid(outputs) + x = torch.mul(inputs, outputs) + + return x diff --git a/ocr_recog/RecSVTR.py b/ocr_recog/RecSVTR.py new file mode 100644 index 0000000000000000000000000000000000000000..484b3df991255590616f552d9f942d5a6d6973c9 --- /dev/null +++ b/ocr_recog/RecSVTR.py @@ -0,0 +1,591 @@ +import torch +import torch.nn as nn +import numpy as np +from torch.nn.init import trunc_normal_, zeros_, ones_ +from torch.nn import functional + + +def drop_path(x, drop_prob=0., training=False): + """Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks). + the original name is misleading as 'Drop Connect' is a different form of dropout in a separate paper... + See discussion: https://github.com/tensorflow/tpu/issues/494#issuecomment-532968956 ... + """ + if drop_prob == 0. or not training: + return x + keep_prob = torch.tensor(1 - drop_prob) + shape = (x.size()[0], ) + (1, ) * (x.ndim - 1) + random_tensor = keep_prob + torch.rand(shape, dtype=x.dtype) + random_tensor = torch.floor(random_tensor) # binarize + output = x.divide(keep_prob) * random_tensor + return output + + +class Swish(nn.Module): + def __int__(self): + super(Swish, self).__int__() + + def forward(self,x): + return x*torch.sigmoid(x) + + +class ConvBNLayer(nn.Module): + def __init__(self, + in_channels, + out_channels, + kernel_size=3, + stride=1, + padding=0, + bias_attr=False, + groups=1, + act=nn.GELU): + super().__init__() + self.conv = nn.Conv2d( + in_channels=in_channels, + out_channels=out_channels, + kernel_size=kernel_size, + stride=stride, + padding=padding, + groups=groups, + # weight_attr=paddle.ParamAttr(initializer=nn.initializer.KaimingUniform()), + bias=bias_attr) + self.norm = nn.BatchNorm2d(out_channels) + self.act = act() + + def forward(self, inputs): + out = self.conv(inputs) + out = self.norm(out) + out = self.act(out) + return out + + +class DropPath(nn.Module): + """Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks). + """ + + def __init__(self, drop_prob=None): + super(DropPath, self).__init__() + self.drop_prob = drop_prob + + def forward(self, x): + return drop_path(x, self.drop_prob, self.training) + + +class Identity(nn.Module): + def __init__(self): + super(Identity, self).__init__() + + def forward(self, input): + return input + + +class Mlp(nn.Module): + def __init__(self, + in_features, + hidden_features=None, + out_features=None, + act_layer=nn.GELU, + drop=0.): + super().__init__() + out_features = out_features or in_features + hidden_features = hidden_features or in_features + self.fc1 = nn.Linear(in_features, hidden_features) + if isinstance(act_layer, str): + self.act = Swish() + else: + self.act = act_layer() + self.fc2 = nn.Linear(hidden_features, out_features) + self.drop = nn.Dropout(drop) + + def forward(self, x): + x = self.fc1(x) + x = self.act(x) + x = self.drop(x) + x = self.fc2(x) + x = self.drop(x) + return x + + +class ConvMixer(nn.Module): + def __init__( + self, + dim, + num_heads=8, + HW=(8, 25), + local_k=(3, 3), ): + super().__init__() + self.HW = HW + self.dim = dim + self.local_mixer = nn.Conv2d( + dim, + dim, + local_k, + 1, (local_k[0] // 2, local_k[1] // 2), + groups=num_heads, + # weight_attr=ParamAttr(initializer=KaimingNormal()) + ) + + def forward(self, x): + h = self.HW[0] + w = self.HW[1] + x = x.transpose([0, 2, 1]).reshape([0, self.dim, h, w]) + x = self.local_mixer(x) + x = x.flatten(2).transpose([0, 2, 1]) + return x + + +class Attention(nn.Module): + def __init__(self, + dim, + num_heads=8, + mixer='Global', + HW=(8, 25), + local_k=(7, 11), + qkv_bias=False, + qk_scale=None, + attn_drop=0., + proj_drop=0.): + super().__init__() + self.num_heads = num_heads + head_dim = dim // num_heads + self.scale = qk_scale or head_dim**-0.5 + + self.qkv = nn.Linear(dim, dim * 3, bias=qkv_bias) + self.attn_drop = nn.Dropout(attn_drop) + self.proj = nn.Linear(dim, dim) + self.proj_drop = nn.Dropout(proj_drop) + self.HW = HW + if HW is not None: + H = HW[0] + W = HW[1] + self.N = H * W + self.C = dim + if mixer == 'Local' and HW is not None: + hk = local_k[0] + wk = local_k[1] + mask = torch.ones([H * W, H + hk - 1, W + wk - 1]) + for h in range(0, H): + for w in range(0, W): + mask[h * W + w, h:h + hk, w:w + wk] = 0. + mask_paddle = mask[:, hk // 2:H + hk // 2, wk // 2:W + wk // + 2].flatten(1) + mask_inf = torch.full([H * W, H * W],fill_value=float('-inf')) + mask = torch.where(mask_paddle < 1, mask_paddle, mask_inf) + self.mask = mask[None,None,:] + # self.mask = mask.unsqueeze([0, 1]) + self.mixer = mixer + + def forward(self, x): + if self.HW is not None: + N = self.N + C = self.C + else: + _, N, C = x.shape + qkv = self.qkv(x).reshape((-1, N, 3, self.num_heads, C //self.num_heads)).permute((2, 0, 3, 1, 4)) + q, k, v = qkv[0] * self.scale, qkv[1], qkv[2] + + attn = (q.matmul(k.permute((0, 1, 3, 2)))) + if self.mixer == 'Local': + attn += self.mask + attn = functional.softmax(attn, dim=-1) + attn = self.attn_drop(attn) + + x = (attn.matmul(v)).permute((0, 2, 1, 3)).reshape((-1, N, C)) + x = self.proj(x) + x = self.proj_drop(x) + return x + + +class Block(nn.Module): + def __init__(self, + dim, + num_heads, + mixer='Global', + local_mixer=(7, 11), + HW=(8, 25), + mlp_ratio=4., + qkv_bias=False, + qk_scale=None, + drop=0., + attn_drop=0., + drop_path=0., + act_layer=nn.GELU, + norm_layer='nn.LayerNorm', + epsilon=1e-6, + prenorm=True): + super().__init__() + if isinstance(norm_layer, str): + self.norm1 = eval(norm_layer)(dim, eps=epsilon) + else: + self.norm1 = norm_layer(dim) + if mixer == 'Global' or mixer == 'Local': + + self.mixer = Attention( + dim, + num_heads=num_heads, + mixer=mixer, + HW=HW, + local_k=local_mixer, + qkv_bias=qkv_bias, + qk_scale=qk_scale, + attn_drop=attn_drop, + proj_drop=drop) + elif mixer == 'Conv': + self.mixer = ConvMixer( + dim, num_heads=num_heads, HW=HW, local_k=local_mixer) + else: + raise TypeError("The mixer must be one of [Global, Local, Conv]") + + self.drop_path = DropPath(drop_path) if drop_path > 0. else Identity() + if isinstance(norm_layer, str): + self.norm2 = eval(norm_layer)(dim, eps=epsilon) + else: + self.norm2 = norm_layer(dim) + mlp_hidden_dim = int(dim * mlp_ratio) + self.mlp_ratio = mlp_ratio + self.mlp = Mlp(in_features=dim, + hidden_features=mlp_hidden_dim, + act_layer=act_layer, + drop=drop) + self.prenorm = prenorm + + def forward(self, x): + if self.prenorm: + x = self.norm1(x + self.drop_path(self.mixer(x))) + x = self.norm2(x + self.drop_path(self.mlp(x))) + else: + x = x + self.drop_path(self.mixer(self.norm1(x))) + x = x + self.drop_path(self.mlp(self.norm2(x))) + return x + + +class PatchEmbed(nn.Module): + """ Image to Patch Embedding + """ + + def __init__(self, + img_size=(32, 100), + in_channels=3, + embed_dim=768, + sub_num=2): + super().__init__() + num_patches = (img_size[1] // (2 ** sub_num)) * \ + (img_size[0] // (2 ** sub_num)) + self.img_size = img_size + self.num_patches = num_patches + self.embed_dim = embed_dim + self.norm = None + if sub_num == 2: + self.proj = nn.Sequential( + ConvBNLayer( + in_channels=in_channels, + out_channels=embed_dim // 2, + kernel_size=3, + stride=2, + padding=1, + act=nn.GELU, + bias_attr=False), + ConvBNLayer( + in_channels=embed_dim // 2, + out_channels=embed_dim, + kernel_size=3, + stride=2, + padding=1, + act=nn.GELU, + bias_attr=False)) + if sub_num == 3: + self.proj = nn.Sequential( + ConvBNLayer( + in_channels=in_channels, + out_channels=embed_dim // 4, + kernel_size=3, + stride=2, + padding=1, + act=nn.GELU, + bias_attr=False), + ConvBNLayer( + in_channels=embed_dim // 4, + out_channels=embed_dim // 2, + kernel_size=3, + stride=2, + padding=1, + act=nn.GELU, + bias_attr=False), + ConvBNLayer( + in_channels=embed_dim // 2, + out_channels=embed_dim, + kernel_size=3, + stride=2, + padding=1, + act=nn.GELU, + bias_attr=False)) + + def forward(self, x): + B, C, H, W = x.shape + assert H == self.img_size[0] and W == self.img_size[1], \ + f"Input image size ({H}*{W}) doesn't match model ({self.img_size[0]}*{self.img_size[1]})." + x = self.proj(x).flatten(2).permute(0, 2, 1) + return x + + +class SubSample(nn.Module): + def __init__(self, + in_channels, + out_channels, + types='Pool', + stride=(2, 1), + sub_norm='nn.LayerNorm', + act=None): + super().__init__() + self.types = types + if types == 'Pool': + self.avgpool = nn.AvgPool2d( + kernel_size=(3, 5), stride=stride, padding=(1, 2)) + self.maxpool = nn.MaxPool2d( + kernel_size=(3, 5), stride=stride, padding=(1, 2)) + self.proj = nn.Linear(in_channels, out_channels) + else: + self.conv = nn.Conv2d( + in_channels, + out_channels, + kernel_size=3, + stride=stride, + padding=1, + # weight_attr=ParamAttr(initializer=KaimingNormal()) + ) + self.norm = eval(sub_norm)(out_channels) + if act is not None: + self.act = act() + else: + self.act = None + + def forward(self, x): + + if self.types == 'Pool': + x1 = self.avgpool(x) + x2 = self.maxpool(x) + x = (x1 + x2) * 0.5 + out = self.proj(x.flatten(2).permute((0, 2, 1))) + else: + x = self.conv(x) + out = x.flatten(2).permute((0, 2, 1)) + out = self.norm(out) + if self.act is not None: + out = self.act(out) + + return out + + +class SVTRNet(nn.Module): + def __init__( + self, + img_size=[48, 100], + in_channels=3, + embed_dim=[64, 128, 256], + depth=[3, 6, 3], + num_heads=[2, 4, 8], + mixer=['Local'] * 6 + ['Global'] * + 6, # Local atten, Global atten, Conv + local_mixer=[[7, 11], [7, 11], [7, 11]], + patch_merging='Conv', # Conv, Pool, None + mlp_ratio=4, + qkv_bias=True, + qk_scale=None, + drop_rate=0., + last_drop=0.1, + attn_drop_rate=0., + drop_path_rate=0.1, + norm_layer='nn.LayerNorm', + sub_norm='nn.LayerNorm', + epsilon=1e-6, + out_channels=192, + out_char_num=25, + block_unit='Block', + act='nn.GELU', + last_stage=True, + sub_num=2, + prenorm=True, + use_lenhead=False, + **kwargs): + super().__init__() + self.img_size = img_size + self.embed_dim = embed_dim + self.out_channels = out_channels + self.prenorm = prenorm + patch_merging = None if patch_merging != 'Conv' and patch_merging != 'Pool' else patch_merging + self.patch_embed = PatchEmbed( + img_size=img_size, + in_channels=in_channels, + embed_dim=embed_dim[0], + sub_num=sub_num) + num_patches = self.patch_embed.num_patches + self.HW = [img_size[0] // (2**sub_num), img_size[1] // (2**sub_num)] + self.pos_embed = nn.Parameter(torch.zeros(1, num_patches, embed_dim[0])) + # self.pos_embed = self.create_parameter( + # shape=[1, num_patches, embed_dim[0]], default_initializer=zeros_) + + # self.add_parameter("pos_embed", self.pos_embed) + + self.pos_drop = nn.Dropout(p=drop_rate) + Block_unit = eval(block_unit) + + dpr = np.linspace(0, drop_path_rate, sum(depth)) + self.blocks1 = nn.ModuleList( + [ + Block_unit( + dim=embed_dim[0], + num_heads=num_heads[0], + mixer=mixer[0:depth[0]][i], + HW=self.HW, + local_mixer=local_mixer[0], + mlp_ratio=mlp_ratio, + qkv_bias=qkv_bias, + qk_scale=qk_scale, + drop=drop_rate, + act_layer=eval(act), + attn_drop=attn_drop_rate, + drop_path=dpr[0:depth[0]][i], + norm_layer=norm_layer, + epsilon=epsilon, + prenorm=prenorm) for i in range(depth[0]) + ] + ) + if patch_merging is not None: + self.sub_sample1 = SubSample( + embed_dim[0], + embed_dim[1], + sub_norm=sub_norm, + stride=[2, 1], + types=patch_merging) + HW = [self.HW[0] // 2, self.HW[1]] + else: + HW = self.HW + self.patch_merging = patch_merging + self.blocks2 = nn.ModuleList([ + Block_unit( + dim=embed_dim[1], + num_heads=num_heads[1], + mixer=mixer[depth[0]:depth[0] + depth[1]][i], + HW=HW, + local_mixer=local_mixer[1], + mlp_ratio=mlp_ratio, + qkv_bias=qkv_bias, + qk_scale=qk_scale, + drop=drop_rate, + act_layer=eval(act), + attn_drop=attn_drop_rate, + drop_path=dpr[depth[0]:depth[0] + depth[1]][i], + norm_layer=norm_layer, + epsilon=epsilon, + prenorm=prenorm) for i in range(depth[1]) + ]) + if patch_merging is not None: + self.sub_sample2 = SubSample( + embed_dim[1], + embed_dim[2], + sub_norm=sub_norm, + stride=[2, 1], + types=patch_merging) + HW = [self.HW[0] // 4, self.HW[1]] + else: + HW = self.HW + self.blocks3 = nn.ModuleList([ + Block_unit( + dim=embed_dim[2], + num_heads=num_heads[2], + mixer=mixer[depth[0] + depth[1]:][i], + HW=HW, + local_mixer=local_mixer[2], + mlp_ratio=mlp_ratio, + qkv_bias=qkv_bias, + qk_scale=qk_scale, + drop=drop_rate, + act_layer=eval(act), + attn_drop=attn_drop_rate, + drop_path=dpr[depth[0] + depth[1]:][i], + norm_layer=norm_layer, + epsilon=epsilon, + prenorm=prenorm) for i in range(depth[2]) + ]) + self.last_stage = last_stage + if last_stage: + self.avg_pool = nn.AdaptiveAvgPool2d((1, out_char_num)) + self.last_conv = nn.Conv2d( + in_channels=embed_dim[2], + out_channels=self.out_channels, + kernel_size=1, + stride=1, + padding=0, + bias=False) + self.hardswish = nn.Hardswish() + self.dropout = nn.Dropout(p=last_drop) + if not prenorm: + self.norm = eval(norm_layer)(embed_dim[-1], epsilon=epsilon) + self.use_lenhead = use_lenhead + if use_lenhead: + self.len_conv = nn.Linear(embed_dim[2], self.out_channels) + self.hardswish_len = nn.Hardswish() + self.dropout_len = nn.Dropout( + p=last_drop) + + trunc_normal_(self.pos_embed,std=.02) + self.apply(self._init_weights) + + def _init_weights(self, m): + if isinstance(m, nn.Linear): + trunc_normal_(m.weight,std=.02) + if isinstance(m, nn.Linear) and m.bias is not None: + zeros_(m.bias) + elif isinstance(m, nn.LayerNorm): + zeros_(m.bias) + ones_(m.weight) + + def forward_features(self, x): + x = self.patch_embed(x) + x = x + self.pos_embed + x = self.pos_drop(x) + for blk in self.blocks1: + x = blk(x) + if self.patch_merging is not None: + x = self.sub_sample1( + x.permute([0, 2, 1]).reshape( + [-1, self.embed_dim[0], self.HW[0], self.HW[1]])) + for blk in self.blocks2: + x = blk(x) + if self.patch_merging is not None: + x = self.sub_sample2( + x.permute([0, 2, 1]).reshape( + [-1, self.embed_dim[1], self.HW[0] // 2, self.HW[1]])) + for blk in self.blocks3: + x = blk(x) + if not self.prenorm: + x = self.norm(x) + return x + + def forward(self, x): + x = self.forward_features(x) + if self.use_lenhead: + len_x = self.len_conv(x.mean(1)) + len_x = self.dropout_len(self.hardswish_len(len_x)) + if self.last_stage: + if self.patch_merging is not None: + h = self.HW[0] // 4 + else: + h = self.HW[0] + x = self.avg_pool( + x.permute([0, 2, 1]).reshape( + [-1, self.embed_dim[2], h, self.HW[1]])) + x = self.last_conv(x) + x = self.hardswish(x) + x = self.dropout(x) + if self.use_lenhead: + return x, len_x + return x + + +if __name__=="__main__": + a = torch.rand(1,3,48,100) + svtr = SVTRNet() + + out = svtr(a) + print(svtr) + print(out.size()) \ No newline at end of file diff --git a/ocr_recog/common.py b/ocr_recog/common.py new file mode 100644 index 0000000000000000000000000000000000000000..a328bb034a37934b7437893b5c2e42cd3504c17f --- /dev/null +++ b/ocr_recog/common.py @@ -0,0 +1,74 @@ + + +import torch +import torch.nn as nn +import torch.nn.functional as F + + +class Hswish(nn.Module): + def __init__(self, inplace=True): + super(Hswish, self).__init__() + self.inplace = inplace + + def forward(self, x): + return x * F.relu6(x + 3., inplace=self.inplace) / 6. + +# out = max(0, min(1, slop*x+offset)) +# paddle.fluid.layers.hard_sigmoid(x, slope=0.2, offset=0.5, name=None) +class Hsigmoid(nn.Module): + def __init__(self, inplace=True): + super(Hsigmoid, self).__init__() + self.inplace = inplace + + def forward(self, x): + # torch: F.relu6(x + 3., inplace=self.inplace) / 6. + # paddle: F.relu6(1.2 * x + 3., inplace=self.inplace) / 6. + return F.relu6(1.2 * x + 3., inplace=self.inplace) / 6. + +class GELU(nn.Module): + def __init__(self, inplace=True): + super(GELU, self).__init__() + self.inplace = inplace + + def forward(self, x): + return torch.nn.functional.gelu(x) + + +class Swish(nn.Module): + def __init__(self, inplace=True): + super(Swish, self).__init__() + self.inplace = inplace + + def forward(self, x): + if self.inplace: + x.mul_(torch.sigmoid(x)) + return x + else: + return x*torch.sigmoid(x) + + +class Activation(nn.Module): + def __init__(self, act_type, inplace=True): + super(Activation, self).__init__() + act_type = act_type.lower() + if act_type == 'relu': + self.act = nn.ReLU(inplace=inplace) + elif act_type == 'relu6': + self.act = nn.ReLU6(inplace=inplace) + elif act_type == 'sigmoid': + raise NotImplementedError + elif act_type == 'hard_sigmoid': + self.act = Hsigmoid(inplace) + elif act_type == 'hard_swish': + self.act = Hswish(inplace=inplace) + elif act_type == 'leakyrelu': + self.act = nn.LeakyReLU(inplace=inplace) + elif act_type == 'gelu': + self.act = GELU(inplace=inplace) + elif act_type == 'swish': + self.act = Swish(inplace=inplace) + else: + raise NotImplementedError + + def forward(self, inputs): + return self.act(inputs) \ No newline at end of file diff --git a/ocr_recog/en_dict.txt b/ocr_recog/en_dict.txt new file mode 100644 index 0000000000000000000000000000000000000000..7677d31b9d3f08eef2823c2cf051beeab1f0470b --- /dev/null +++ b/ocr_recog/en_dict.txt @@ -0,0 +1,95 @@ +0 +1 +2 +3 +4 +5 +6 +7 +8 +9 +: +; +< += +> +? +@ +A +B +C +D +E +F +G +H +I +J +K +L +M +N +O +P +Q +R +S +T +U +V +W +X +Y +Z +[ +\ +] +^ +_ +` +a +b +c +d +e +f +g +h +i +j +k +l +m +n +o +p +q +r +s +t +u +v +w +x +y +z +{ +| +} +~ +! +" +# +$ +% +& +' +( +) +* ++ +, +- +. +/ + diff --git a/ocr_recog/ppocr_keys_v1.txt b/ocr_recog/ppocr_keys_v1.txt new file mode 100644 index 0000000000000000000000000000000000000000..84b885d8352226e49b1d5d791b8f43a663e246aa --- /dev/null +++ b/ocr_recog/ppocr_keys_v1.txt @@ -0,0 +1,6623 @@ +' +疗 +绚 +诚 +娇 +溜 +题 +贿 +者 +廖 +更 +纳 +加 +奉 +公 +一 +就 +汴 +计 +与 +路 +房 +原 +妇 +2 +0 +8 +- +7 +其 +> +: +] +, +, +骑 +刈 +全 +消 +昏 +傈 +安 +久 +钟 +嗅 +不 +影 +处 +驽 +蜿 +资 +关 +椤 +地 +瘸 +专 +问 +忖 +票 +嫉 +炎 +韵 +要 +月 +田 +节 +陂 +鄙 +捌 +备 +拳 +伺 +眼 +网 +盎 +大 +傍 +心 +东 +愉 +汇 +蹿 +科 +每 +业 +里 +航 +晏 +字 +平 +录 +先 +1 +3 +彤 +鲶 +产 +稍 +督 +腴 +有 +象 +岳 +注 +绍 +在 +泺 +文 +定 +核 +名 +水 +过 +理 +让 +偷 +率 +等 +这 +发 +” +为 +含 +肥 +酉 +相 +鄱 +七 +编 +猥 +锛 +日 +镀 +蒂 +掰 +倒 +辆 +栾 +栗 +综 +涩 +州 +雌 +滑 +馀 +了 +机 +块 +司 +宰 +甙 +兴 +矽 +抚 +保 +用 +沧 +秩 +如 +收 +息 +滥 +页 +疑 +埠 +! +! +姥 +异 +橹 +钇 +向 +下 +跄 +的 +椴 +沫 +国 +绥 +獠 +报 +开 +民 +蜇 +何 +分 +凇 +长 +讥 +藏 +掏 +施 +羽 +中 +讲 +派 +嘟 +人 +提 +浼 +间 +世 +而 +古 +多 +倪 +唇 +饯 +控 +庚 +首 +赛 +蜓 +味 +断 +制 +觉 +技 +替 +艰 +溢 +潮 +夕 +钺 +外 +摘 +枋 +动 +双 +单 +啮 +户 +枇 +确 +锦 +曜 +杜 +或 +能 +效 +霜 +盒 +然 +侗 +电 +晁 +放 +步 +鹃 +新 +杖 +蜂 +吒 +濂 +瞬 +评 +总 +隍 +对 +独 +合 +也 +是 +府 +青 +天 +诲 +墙 +组 +滴 +级 +邀 +帘 +示 +已 +时 +骸 +仄 +泅 +和 +遨 +店 +雇 +疫 +持 +巍 +踮 +境 +只 +亨 +目 +鉴 +崤 +闲 +体 +泄 +杂 +作 +般 +轰 +化 +解 +迂 +诿 +蛭 +璀 +腾 +告 +版 +服 +省 +师 +小 +规 +程 +线 +海 +办 +引 +二 +桧 +牌 +砺 +洄 +裴 +修 +图 +痫 +胡 +许 +犊 +事 +郛 +基 +柴 +呼 +食 +研 +奶 +律 +蛋 +因 +葆 +察 +戏 +褒 +戒 +再 +李 +骁 +工 +貂 +油 +鹅 +章 +啄 +休 +场 +给 +睡 +纷 +豆 +器 +捎 +说 +敏 +学 +会 +浒 +设 +诊 +格 +廓 +查 +来 +霓 +室 +溆 +¢ +诡 +寥 +焕 +舜 +柒 +狐 +回 +戟 +砾 +厄 +实 +翩 +尿 +五 +入 +径 +惭 +喹 +股 +宇 +篝 +| +; +美 +期 +云 +九 +祺 +扮 +靠 +锝 +槌 +系 +企 +酰 +阊 +暂 +蚕 +忻 +豁 +本 +羹 +执 +条 +钦 +H +獒 +限 +进 +季 +楦 +于 +芘 +玖 +铋 +茯 +未 +答 +粘 +括 +样 +精 +欠 +矢 +甥 +帷 +嵩 +扣 +令 +仔 +风 +皈 +行 +支 +部 +蓉 +刮 +站 +蜡 +救 +钊 +汗 +松 +嫌 +成 +可 +. +鹤 +院 +从 +交 +政 +怕 +活 +调 +球 +局 +验 +髌 +第 +韫 +谗 +串 +到 +圆 +年 +米 +/ +* +友 +忿 +检 +区 +看 +自 +敢 +刃 +个 +兹 +弄 +流 +留 +同 +没 +齿 +星 +聆 +轼 +湖 +什 +三 +建 +蛔 +儿 +椋 +汕 +震 +颧 +鲤 +跟 +力 +情 +璺 +铨 +陪 +务 +指 +族 +训 +滦 +鄣 +濮 +扒 +商 +箱 +十 +召 +慷 +辗 +所 +莞 +管 +护 +臭 +横 +硒 +嗓 +接 +侦 +六 +露 +党 +馋 +驾 +剖 +高 +侬 +妪 +幂 +猗 +绺 +骐 +央 +酐 +孝 +筝 +课 +徇 +缰 +门 +男 +西 +项 +句 +谙 +瞒 +秃 +篇 +教 +碲 +罚 +声 +呐 +景 +前 +富 +嘴 +鳌 +稀 +免 +朋 +啬 +睐 +去 +赈 +鱼 +住 +肩 +愕 +速 +旁 +波 +厅 +健 +茼 +厥 +鲟 +谅 +投 +攸 +炔 +数 +方 +击 +呋 +谈 +绩 +别 +愫 +僚 +躬 +鹧 +胪 +炳 +招 +喇 +膨 +泵 +蹦 +毛 +结 +5 +4 +谱 +识 +陕 +粽 +婚 +拟 +构 +且 +搜 +任 +潘 +比 +郢 +妨 +醪 +陀 +桔 +碘 +扎 +选 +哈 +骷 +楷 +亿 +明 +缆 +脯 +监 +睫 +逻 +婵 +共 +赴 +淝 +凡 +惦 +及 +达 +揖 +谩 +澹 +减 +焰 +蛹 +番 +祁 +柏 +员 +禄 +怡 +峤 +龙 +白 +叽 +生 +闯 +起 +细 +装 +谕 +竟 +聚 +钙 +上 +导 +渊 +按 +艾 +辘 +挡 +耒 +盹 +饪 +臀 +记 +邮 +蕙 +受 +各 +医 +搂 +普 +滇 +朗 +茸 +带 +翻 +酚 +( +光 +堤 +墟 +蔷 +万 +幻 +〓 +瑙 +辈 +昧 +盏 +亘 +蛀 +吉 +铰 +请 +子 +假 +闻 +税 +井 +诩 +哨 +嫂 +好 +面 +琐 +校 +馊 +鬣 +缂 +营 +访 +炖 +占 +农 +缀 +否 +经 +钚 +棵 +趟 +张 +亟 +吏 +茶 +谨 +捻 +论 +迸 +堂 +玉 +信 +吧 +瞠 +乡 +姬 +寺 +咬 +溏 +苄 +皿 +意 +赉 +宝 +尔 +钰 +艺 +特 +唳 +踉 +都 +荣 +倚 +登 +荐 +丧 +奇 +涵 +批 +炭 +近 +符 +傩 +感 +道 +着 +菊 +虹 +仲 +众 +懈 +濯 +颞 +眺 +南 +释 +北 +缝 +标 +既 +茗 +整 +撼 +迤 +贲 +挎 +耱 +拒 +某 +妍 +卫 +哇 +英 +矶 +藩 +治 +他 +元 +领 +膜 +遮 +穗 +蛾 +飞 +荒 +棺 +劫 +么 +市 +火 +温 +拈 +棚 +洼 +转 +果 +奕 +卸 +迪 +伸 +泳 +斗 +邡 +侄 +涨 +屯 +萋 +胭 +氡 +崮 +枞 +惧 +冒 +彩 +斜 +手 +豚 +随 +旭 +淑 +妞 +形 +菌 +吲 +沱 +争 +驯 +歹 +挟 +兆 +柱 +传 +至 +包 +内 +响 +临 +红 +功 +弩 +衡 +寂 +禁 +老 +棍 +耆 +渍 +织 +害 +氵 +渑 +布 +载 +靥 +嗬 +虽 +苹 +咨 +娄 +库 +雉 +榜 +帜 +嘲 +套 +瑚 +亲 +簸 +欧 +边 +6 +腿 +旮 +抛 +吹 +瞳 +得 +镓 +梗 +厨 +继 +漾 +愣 +憨 +士 +策 +窑 +抑 +躯 +襟 +脏 +参 +贸 +言 +干 +绸 +鳄 +穷 +藜 +音 +折 +详 +) +举 +悍 +甸 +癌 +黎 +谴 +死 +罩 +迁 +寒 +驷 +袖 +媒 +蒋 +掘 +模 +纠 +恣 +观 +祖 +蛆 +碍 +位 +稿 +主 +澧 +跌 +筏 +京 +锏 +帝 +贴 +证 +糠 +才 +黄 +鲸 +略 +炯 +饱 +四 +出 +园 +犀 +牧 +容 +汉 +杆 +浈 +汰 +瑷 +造 +虫 +瘩 +怪 +驴 +济 +应 +花 +沣 +谔 +夙 +旅 +价 +矿 +以 +考 +s +u +呦 +晒 +巡 +茅 +准 +肟 +瓴 +詹 +仟 +褂 +译 +桌 +混 +宁 +怦 +郑 +抿 +些 +余 +鄂 +饴 +攒 +珑 +群 +阖 +岔 +琨 +藓 +预 +环 +洮 +岌 +宀 +杲 +瀵 +最 +常 +囡 +周 +踊 +女 +鼓 +袭 +喉 +简 +范 +薯 +遐 +疏 +粱 +黜 +禧 +法 +箔 +斤 +遥 +汝 +奥 +直 +贞 +撑 +置 +绱 +集 +她 +馅 +逗 +钧 +橱 +魉 +[ +恙 +躁 +唤 +9 +旺 +膘 +待 +脾 +惫 +购 +吗 +依 +盲 +度 +瘿 +蠖 +俾 +之 +镗 +拇 +鲵 +厝 +簧 +续 +款 +展 +啃 +表 +剔 +品 +钻 +腭 +损 +清 +锶 +统 +涌 +寸 +滨 +贪 +链 +吠 +冈 +伎 +迥 +咏 +吁 +览 +防 +迅 +失 +汾 +阔 +逵 +绀 +蔑 +列 +川 +凭 +努 +熨 +揪 +利 +俱 +绉 +抢 +鸨 +我 +即 +责 +膦 +易 +毓 +鹊 +刹 +玷 +岿 +空 +嘞 +绊 +排 +术 +估 +锷 +违 +们 +苟 +铜 +播 +肘 +件 +烫 +审 +鲂 +广 +像 +铌 +惰 +铟 +巳 +胍 +鲍 +康 +憧 +色 +恢 +想 +拷 +尤 +疳 +知 +S +Y +F +D +A +峄 +裕 +帮 +握 +搔 +氐 +氘 +难 +墒 +沮 +雨 +叁 +缥 +悴 +藐 +湫 +娟 +苑 +稠 +颛 +簇 +后 +阕 +闭 +蕤 +缚 +怎 +佞 +码 +嘤 +蔡 +痊 +舱 +螯 +帕 +赫 +昵 +升 +烬 +岫 +、 +疵 +蜻 +髁 +蕨 +隶 +烛 +械 +丑 +盂 +梁 +强 +鲛 +由 +拘 +揉 +劭 +龟 +撤 +钩 +呕 +孛 +费 +妻 +漂 +求 +阑 +崖 +秤 +甘 +通 +深 +补 +赃 +坎 +床 +啪 +承 +吼 +量 +暇 +钼 +烨 +阂 +擎 +脱 +逮 +称 +P +神 +属 +矗 +华 +届 +狍 +葑 +汹 +育 +患 +窒 +蛰 +佼 +静 +槎 +运 +鳗 +庆 +逝 +曼 +疱 +克 +代 +官 +此 +麸 +耧 +蚌 +晟 +例 +础 +榛 +副 +测 +唰 +缢 +迹 +灬 +霁 +身 +岁 +赭 +扛 +又 +菡 +乜 +雾 +板 +读 +陷 +徉 +贯 +郁 +虑 +变 +钓 +菜 +圾 +现 +琢 +式 +乐 +维 +渔 +浜 +左 +吾 +脑 +钡 +警 +T +啵 +拴 +偌 +漱 +湿 +硕 +止 +骼 +魄 +积 +燥 +联 +踢 +玛 +则 +窿 +见 +振 +畿 +送 +班 +钽 +您 +赵 +刨 +印 +讨 +踝 +籍 +谡 +舌 +崧 +汽 +蔽 +沪 +酥 +绒 +怖 +财 +帖 +肱 +私 +莎 +勋 +羔 +霸 +励 +哼 +帐 +将 +帅 +渠 +纪 +婴 +娩 +岭 +厘 +滕 +吻 +伤 +坝 +冠 +戊 +隆 +瘁 +介 +涧 +物 +黍 +并 +姗 +奢 +蹑 +掣 +垸 +锴 +命 +箍 +捉 +病 +辖 +琰 +眭 +迩 +艘 +绌 +繁 +寅 +若 +毋 +思 +诉 +类 +诈 +燮 +轲 +酮 +狂 +重 +反 +职 +筱 +县 +委 +磕 +绣 +奖 +晋 +濉 +志 +徽 +肠 +呈 +獐 +坻 +口 +片 +碰 +几 +村 +柿 +劳 +料 +获 +亩 +惕 +晕 +厌 +号 +罢 +池 +正 +鏖 +煨 +家 +棕 +复 +尝 +懋 +蜥 +锅 +岛 +扰 +队 +坠 +瘾 +钬 +@ +卧 +疣 +镇 +譬 +冰 +彷 +频 +黯 +据 +垄 +采 +八 +缪 +瘫 +型 +熹 +砰 +楠 +襁 +箐 +但 +嘶 +绳 +啤 +拍 +盥 +穆 +傲 +洗 +盯 +塘 +怔 +筛 +丿 +台 +恒 +喂 +葛 +永 +¥ +烟 +酒 +桦 +书 +砂 +蚝 +缉 +态 +瀚 +袄 +圳 +轻 +蛛 +超 +榧 +遛 +姒 +奘 +铮 +右 +荽 +望 +偻 +卡 +丶 +氰 +附 +做 +革 +索 +戚 +坨 +桷 +唁 +垅 +榻 +岐 +偎 +坛 +莨 +山 +殊 +微 +骇 +陈 +爨 +推 +嗝 +驹 +澡 +藁 +呤 +卤 +嘻 +糅 +逛 +侵 +郓 +酌 +德 +摇 +※ +鬃 +被 +慨 +殡 +羸 +昌 +泡 +戛 +鞋 +河 +宪 +沿 +玲 +鲨 +翅 +哽 +源 +铅 +语 +照 +邯 +址 +荃 +佬 +顺 +鸳 +町 +霭 +睾 +瓢 +夸 +椁 +晓 +酿 +痈 +咔 +侏 +券 +噎 +湍 +签 +嚷 +离 +午 +尚 +社 +锤 +背 +孟 +使 +浪 +缦 +潍 +鞅 +军 +姹 +驶 +笑 +鳟 +鲁 +》 +孽 +钜 +绿 +洱 +礴 +焯 +椰 +颖 +囔 +乌 +孔 +巴 +互 +性 +椽 +哞 +聘 +昨 +早 +暮 +胶 +炀 +隧 +低 +彗 +昝 +铁 +呓 +氽 +藉 +喔 +癖 +瑗 +姨 +权 +胱 +韦 +堑 +蜜 +酋 +楝 +砝 +毁 +靓 +歙 +锲 +究 +屋 +喳 +骨 +辨 +碑 +武 +鸠 +宫 +辜 +烊 +适 +坡 +殃 +培 +佩 +供 +走 +蜈 +迟 +翼 +况 +姣 +凛 +浔 +吃 +飘 +债 +犟 +金 +促 +苛 +崇 +坂 +莳 +畔 +绂 +兵 +蠕 +斋 +根 +砍 +亢 +欢 +恬 +崔 +剁 +餐 +榫 +快 +扶 +‖ +濒 +缠 +鳜 +当 +彭 +驭 +浦 +篮 +昀 +锆 +秸 +钳 +弋 +娣 +瞑 +夷 +龛 +苫 +拱 +致 +% +嵊 +障 +隐 +弑 +初 +娓 +抉 +汩 +累 +蓖 +" +唬 +助 +苓 +昙 +押 +毙 +破 +城 +郧 +逢 +嚏 +獭 +瞻 +溱 +婿 +赊 +跨 +恼 +璧 +萃 +姻 +貉 +灵 +炉 +密 +氛 +陶 +砸 +谬 +衔 +点 +琛 +沛 +枳 +层 +岱 +诺 +脍 +榈 +埂 +征 +冷 +裁 +打 +蹴 +素 +瘘 +逞 +蛐 +聊 +激 +腱 +萘 +踵 +飒 +蓟 +吆 +取 +咙 +簋 +涓 +矩 +曝 +挺 +揣 +座 +你 +史 +舵 +焱 +尘 +苏 +笈 +脚 +溉 +榨 +诵 +樊 +邓 +焊 +义 +庶 +儋 +蟋 +蒲 +赦 +呷 +杞 +诠 +豪 +还 +试 +颓 +茉 +太 +除 +紫 +逃 +痴 +草 +充 +鳕 +珉 +祗 +墨 +渭 +烩 +蘸 +慕 +璇 +镶 +穴 +嵘 +恶 +骂 +险 +绋 +幕 +碉 +肺 +戳 +刘 +潞 +秣 +纾 +潜 +銮 +洛 +须 +罘 +销 +瘪 +汞 +兮 +屉 +r +林 +厕 +质 +探 +划 +狸 +殚 +善 +煊 +烹 +〒 +锈 +逯 +宸 +辍 +泱 +柚 +袍 +远 +蹋 +嶙 +绝 +峥 +娥 +缍 +雀 +徵 +认 +镱 +谷 += +贩 +勉 +撩 +鄯 +斐 +洋 +非 +祚 +泾 +诒 +饿 +撬 +威 +晷 +搭 +芍 +锥 +笺 +蓦 +候 +琊 +档 +礁 +沼 +卵 +荠 +忑 +朝 +凹 +瑞 +头 +仪 +弧 +孵 +畏 +铆 +突 +衲 +车 +浩 +气 +茂 +悖 +厢 +枕 +酝 +戴 +湾 +邹 +飚 +攘 +锂 +写 +宵 +翁 +岷 +无 +喜 +丈 +挑 +嗟 +绛 +殉 +议 +槽 +具 +醇 +淞 +笃 +郴 +阅 +饼 +底 +壕 +砚 +弈 +询 +缕 +庹 +翟 +零 +筷 +暨 +舟 +闺 +甯 +撞 +麂 +茌 +蔼 +很 +珲 +捕 +棠 +角 +阉 +媛 +娲 +诽 +剿 +尉 +爵 +睬 +韩 +诰 +匣 +危 +糍 +镯 +立 +浏 +阳 +少 +盆 +舔 +擘 +匪 +申 +尬 +铣 +旯 +抖 +赘 +瓯 +居 +ˇ +哮 +游 +锭 +茏 +歌 +坏 +甚 +秒 +舞 +沙 +仗 +劲 +潺 +阿 +燧 +郭 +嗖 +霏 +忠 +材 +奂 +耐 +跺 +砀 +输 +岖 +媳 +氟 +极 +摆 +灿 +今 +扔 +腻 +枝 +奎 +药 +熄 +吨 +话 +q +额 +慑 +嘌 +协 +喀 +壳 +埭 +视 +著 +於 +愧 +陲 +翌 +峁 +颅 +佛 +腹 +聋 +侯 +咎 +叟 +秀 +颇 +存 +较 +罪 +哄 +岗 +扫 +栏 +钾 +羌 +己 +璨 +枭 +霉 +煌 +涸 +衿 +键 +镝 +益 +岢 +奏 +连 +夯 +睿 +冥 +均 +糖 +狞 +蹊 +稻 +爸 +刿 +胥 +煜 +丽 +肿 +璃 +掸 +跚 +灾 +垂 +樾 +濑 +乎 +莲 +窄 +犹 +撮 +战 +馄 +软 +络 +显 +鸢 +胸 +宾 +妲 +恕 +埔 +蝌 +份 +遇 +巧 +瞟 +粒 +恰 +剥 +桡 +博 +讯 +凯 +堇 +阶 +滤 +卖 +斌 +骚 +彬 +兑 +磺 +樱 +舷 +两 +娱 +福 +仃 +差 +找 +桁 +÷ +净 +把 +阴 +污 +戬 +雷 +碓 +蕲 +楚 +罡 +焖 +抽 +妫 +咒 +仑 +闱 +尽 +邑 +菁 +爱 +贷 +沥 +鞑 +牡 +嗉 +崴 +骤 +塌 +嗦 +订 +拮 +滓 +捡 +锻 +次 +坪 +杩 +臃 +箬 +融 +珂 +鹗 +宗 +枚 +降 +鸬 +妯 +阄 +堰 +盐 +毅 +必 +杨 +崃 +俺 +甬 +状 +莘 +货 +耸 +菱 +腼 +铸 +唏 +痤 +孚 +澳 +懒 +溅 +翘 +疙 +杷 +淼 +缙 +骰 +喊 +悉 +砻 +坷 +艇 +赁 +界 +谤 +纣 +宴 +晃 +茹 +归 +饭 +梢 +铡 +街 +抄 +肼 +鬟 +苯 +颂 +撷 +戈 +炒 +咆 +茭 +瘙 +负 +仰 +客 +琉 +铢 +封 +卑 +珥 +椿 +镧 +窨 +鬲 +寿 +御 +袤 +铃 +萎 +砖 +餮 +脒 +裳 +肪 +孕 +嫣 +馗 +嵇 +恳 +氯 +江 +石 +褶 +冢 +祸 +阻 +狈 +羞 +银 +靳 +透 +咳 +叼 +敷 +芷 +啥 +它 +瓤 +兰 +痘 +懊 +逑 +肌 +往 +捺 +坊 +甩 +呻 +〃 +沦 +忘 +膻 +祟 +菅 +剧 +崆 +智 +坯 +臧 +霍 +墅 +攻 +眯 +倘 +拢 +骠 +铐 +庭 +岙 +瓠 +′ +缺 +泥 +迢 +捶 +? +? +郏 +喙 +掷 +沌 +纯 +秘 +种 +听 +绘 +固 +螨 +团 +香 +盗 +妒 +埚 +蓝 +拖 +旱 +荞 +铀 +血 +遏 +汲 +辰 +叩 +拽 +幅 +硬 +惶 +桀 +漠 +措 +泼 +唑 +齐 +肾 +念 +酱 +虚 +屁 +耶 +旗 +砦 +闵 +婉 +馆 +拭 +绅 +韧 +忏 +窝 +醋 +葺 +顾 +辞 +倜 +堆 +辋 +逆 +玟 +贱 +疾 +董 +惘 +倌 +锕 +淘 +嘀 +莽 +俭 +笏 +绑 +鲷 +杈 +择 +蟀 +粥 +嗯 +驰 +逾 +案 +谪 +褓 +胫 +哩 +昕 +颚 +鲢 +绠 +躺 +鹄 +崂 +儒 +俨 +丝 +尕 +泌 +啊 +萸 +彰 +幺 +吟 +骄 +苣 +弦 +脊 +瑰 +〈 +诛 +镁 +析 +闪 +剪 +侧 +哟 +框 +螃 +守 +嬗 +燕 +狭 +铈 +缮 +概 +迳 +痧 +鲲 +俯 +售 +笼 +痣 +扉 +挖 +满 +咋 +援 +邱 +扇 +歪 +便 +玑 +绦 +峡 +蛇 +叨 +〖 +泽 +胃 +斓 +喋 +怂 +坟 +猪 +该 +蚬 +炕 +弥 +赞 +棣 +晔 +娠 +挲 +狡 +创 +疖 +铕 +镭 +稷 +挫 +弭 +啾 +翔 +粉 +履 +苘 +哦 +楼 +秕 +铂 +土 +锣 +瘟 +挣 +栉 +习 +享 +桢 +袅 +磨 +桂 +谦 +延 +坚 +蔚 +噗 +署 +谟 +猬 +钎 +恐 +嬉 +雒 +倦 +衅 +亏 +璩 +睹 +刻 +殿 +王 +算 +雕 +麻 +丘 +柯 +骆 +丸 +塍 +谚 +添 +鲈 +垓 +桎 +蚯 +芥 +予 +飕 +镦 +谌 +窗 +醚 +菀 +亮 +搪 +莺 +蒿 +羁 +足 +J +真 +轶 +悬 +衷 +靛 +翊 +掩 +哒 +炅 +掐 +冼 +妮 +l +谐 +稚 +荆 +擒 +犯 +陵 +虏 +浓 +崽 +刍 +陌 +傻 +孜 +千 +靖 +演 +矜 +钕 +煽 +杰 +酗 +渗 +伞 +栋 +俗 +泫 +戍 +罕 +沾 +疽 +灏 +煦 +芬 +磴 +叱 +阱 +榉 +湃 +蜀 +叉 +醒 +彪 +租 +郡 +篷 +屎 +良 +垢 +隗 +弱 +陨 +峪 +砷 +掴 +颁 +胎 +雯 +绵 +贬 +沐 +撵 +隘 +篙 +暖 +曹 +陡 +栓 +填 +臼 +彦 +瓶 +琪 +潼 +哪 +鸡 +摩 +啦 +俟 +锋 +域 +耻 +蔫 +疯 +纹 +撇 +毒 +绶 +痛 +酯 +忍 +爪 +赳 +歆 +嘹 +辕 +烈 +册 +朴 +钱 +吮 +毯 +癜 +娃 +谀 +邵 +厮 +炽 +璞 +邃 +丐 +追 +词 +瓒 +忆 +轧 +芫 +谯 +喷 +弟 +半 +冕 +裙 +掖 +墉 +绮 +寝 +苔 +势 +顷 +褥 +切 +衮 +君 +佳 +嫒 +蚩 +霞 +佚 +洙 +逊 +镖 +暹 +唛 +& +殒 +顶 +碗 +獗 +轭 +铺 +蛊 +废 +恹 +汨 +崩 +珍 +那 +杵 +曲 +纺 +夏 +薰 +傀 +闳 +淬 +姘 +舀 +拧 +卷 +楂 +恍 +讪 +厩 +寮 +篪 +赓 +乘 +灭 +盅 +鞣 +沟 +慎 +挂 +饺 +鼾 +杳 +树 +缨 +丛 +絮 +娌 +臻 +嗳 +篡 +侩 +述 +衰 +矛 +圈 +蚜 +匕 +筹 +匿 +濞 +晨 +叶 +骋 +郝 +挚 +蚴 +滞 +增 +侍 +描 +瓣 +吖 +嫦 +蟒 +匾 +圣 +赌 +毡 +癞 +恺 +百 +曳 +需 +篓 +肮 +庖 +帏 +卿 +驿 +遗 +蹬 +鬓 +骡 +歉 +芎 +胳 +屐 +禽 +烦 +晌 +寄 +媾 +狄 +翡 +苒 +船 +廉 +终 +痞 +殇 +々 +畦 +饶 +改 +拆 +悻 +萄 +£ +瓿 +乃 +訾 +桅 +匮 +溧 +拥 +纱 +铍 +骗 +蕃 +龋 +缬 +父 +佐 +疚 +栎 +醍 +掳 +蓄 +x +惆 +颜 +鲆 +榆 +〔 +猎 +敌 +暴 +谥 +鲫 +贾 +罗 +玻 +缄 +扦 +芪 +癣 +落 +徒 +臾 +恿 +猩 +托 +邴 +肄 +牵 +春 +陛 +耀 +刊 +拓 +蓓 +邳 +堕 +寇 +枉 +淌 +啡 +湄 +兽 +酷 +萼 +碚 +濠 +萤 +夹 +旬 +戮 +梭 +琥 +椭 +昔 +勺 +蜊 +绐 +晚 +孺 +僵 +宣 +摄 +冽 +旨 +萌 +忙 +蚤 +眉 +噼 +蟑 +付 +契 +瓜 +悼 +颡 +壁 +曾 +窕 +颢 +澎 +仿 +俑 +浑 +嵌 +浣 +乍 +碌 +褪 +乱 +蔟 +隙 +玩 +剐 +葫 +箫 +纲 +围 +伐 +决 +伙 +漩 +瑟 +刑 +肓 +镳 +缓 +蹭 +氨 +皓 +典 +畲 +坍 +铑 +檐 +塑 +洞 +倬 +储 +胴 +淳 +戾 +吐 +灼 +惺 +妙 +毕 +珐 +缈 +虱 +盖 +羰 +鸿 +磅 +谓 +髅 +娴 +苴 +唷 +蚣 +霹 +抨 +贤 +唠 +犬 +誓 +逍 +庠 +逼 +麓 +籼 +釉 +呜 +碧 +秧 +氩 +摔 +霄 +穸 +纨 +辟 +妈 +映 +完 +牛 +缴 +嗷 +炊 +恩 +荔 +茆 +掉 +紊 +慌 +莓 +羟 +阙 +萁 +磐 +另 +蕹 +辱 +鳐 +湮 +吡 +吩 +唐 +睦 +垠 +舒 +圜 +冗 +瞿 +溺 +芾 +囱 +匠 +僳 +汐 +菩 +饬 +漓 +黑 +霰 +浸 +濡 +窥 +毂 +蒡 +兢 +驻 +鹉 +芮 +诙 +迫 +雳 +厂 +忐 +臆 +猴 +鸣 +蚪 +栈 +箕 +羡 +渐 +莆 +捍 +眈 +哓 +趴 +蹼 +埕 +嚣 +骛 +宏 +淄 +斑 +噜 +严 +瑛 +垃 +椎 +诱 +压 +庾 +绞 +焘 +廿 +抡 +迄 +棘 +夫 +纬 +锹 +眨 +瞌 +侠 +脐 +竞 +瀑 +孳 +骧 +遁 +姜 +颦 +荪 +滚 +萦 +伪 +逸 +粳 +爬 +锁 +矣 +役 +趣 +洒 +颔 +诏 +逐 +奸 +甭 +惠 +攀 +蹄 +泛 +尼 +拼 +阮 +鹰 +亚 +颈 +惑 +勒 +〉 +际 +肛 +爷 +刚 +钨 +丰 +养 +冶 +鲽 +辉 +蔻 +画 +覆 +皴 +妊 +麦 +返 +醉 +皂 +擀 +〗 +酶 +凑 +粹 +悟 +诀 +硖 +港 +卜 +z +杀 +涕 +± +舍 +铠 +抵 +弛 +段 +敝 +镐 +奠 +拂 +轴 +跛 +袱 +e +t +沉 +菇 +俎 +薪 +峦 +秭 +蟹 +历 +盟 +菠 +寡 +液 +肢 +喻 +染 +裱 +悱 +抱 +氙 +赤 +捅 +猛 +跑 +氮 +谣 +仁 +尺 +辊 +窍 +烙 +衍 +架 +擦 +倏 +璐 +瑁 +币 +楞 +胖 +夔 +趸 +邛 +惴 +饕 +虔 +蝎 +§ +哉 +贝 +宽 +辫 +炮 +扩 +饲 +籽 +魏 +菟 +锰 +伍 +猝 +末 +琳 +哚 +蛎 +邂 +呀 +姿 +鄞 +却 +歧 +仙 +恸 +椐 +森 +牒 +寤 +袒 +婆 +虢 +雅 +钉 +朵 +贼 +欲 +苞 +寰 +故 +龚 +坭 +嘘 +咫 +礼 +硷 +兀 +睢 +汶 +’ +铲 +烧 +绕 +诃 +浃 +钿 +哺 +柜 +讼 +颊 +璁 +腔 +洽 +咐 +脲 +簌 +筠 +镣 +玮 +鞠 +谁 +兼 +姆 +挥 +梯 +蝴 +谘 +漕 +刷 +躏 +宦 +弼 +b +垌 +劈 +麟 +莉 +揭 +笙 +渎 +仕 +嗤 +仓 +配 +怏 +抬 +错 +泯 +镊 +孰 +猿 +邪 +仍 +秋 +鼬 +壹 +歇 +吵 +炼 +< +尧 +射 +柬 +廷 +胧 +霾 +凳 +隋 +肚 +浮 +梦 +祥 +株 +堵 +退 +L +鹫 +跎 +凶 +毽 +荟 +炫 +栩 +玳 +甜 +沂 +鹿 +顽 +伯 +爹 +赔 +蛴 +徐 +匡 +欣 +狰 +缸 +雹 +蟆 +疤 +默 +沤 +啜 +痂 +衣 +禅 +w +i +h +辽 +葳 +黝 +钗 +停 +沽 +棒 +馨 +颌 +肉 +吴 +硫 +悯 +劾 +娈 +马 +啧 +吊 +悌 +镑 +峭 +帆 +瀣 +涉 +咸 +疸 +滋 +泣 +翦 +拙 +癸 +钥 +蜒 ++ +尾 +庄 +凝 +泉 +婢 +渴 +谊 +乞 +陆 +锉 +糊 +鸦 +淮 +I +B +N +晦 +弗 +乔 +庥 +葡 +尻 +席 +橡 +傣 +渣 +拿 +惩 +麋 +斛 +缃 +矮 +蛏 +岘 +鸽 +姐 +膏 +催 +奔 +镒 +喱 +蠡 +摧 +钯 +胤 +柠 +拐 +璋 +鸥 +卢 +荡 +倾 +^ +_ +珀 +逄 +萧 +塾 +掇 +贮 +笆 +聂 +圃 +冲 +嵬 +M +滔 +笕 +值 +炙 +偶 +蜱 +搐 +梆 +汪 +蔬 +腑 +鸯 +蹇 +敞 +绯 +仨 +祯 +谆 +梧 +糗 +鑫 +啸 +豺 +囹 +猾 +巢 +柄 +瀛 +筑 +踌 +沭 +暗 +苁 +鱿 +蹉 +脂 +蘖 +牢 +热 +木 +吸 +溃 +宠 +序 +泞 +偿 +拜 +檩 +厚 +朐 +毗 +螳 +吞 +媚 +朽 +担 +蝗 +橘 +畴 +祈 +糟 +盱 +隼 +郜 +惜 +珠 +裨 +铵 +焙 +琚 +唯 +咚 +噪 +骊 +丫 +滢 +勤 +棉 +呸 +咣 +淀 +隔 +蕾 +窈 +饨 +挨 +煅 +短 +匙 +粕 +镜 +赣 +撕 +墩 +酬 +馁 +豌 +颐 +抗 +酣 +氓 +佑 +搁 +哭 +递 +耷 +涡 +桃 +贻 +碣 +截 +瘦 +昭 +镌 +蔓 +氚 +甲 +猕 +蕴 +蓬 +散 +拾 +纛 +狼 +猷 +铎 +埋 +旖 +矾 +讳 +囊 +糜 +迈 +粟 +蚂 +紧 +鲳 +瘢 +栽 +稼 +羊 +锄 +斟 +睁 +桥 +瓮 +蹙 +祉 +醺 +鼻 +昱 +剃 +跳 +篱 +跷 +蒜 +翎 +宅 +晖 +嗑 +壑 +峻 +癫 +屏 +狠 +陋 +袜 +途 +憎 +祀 +莹 +滟 +佶 +溥 +臣 +约 +盛 +峰 +磁 +慵 +婪 +拦 +莅 +朕 +鹦 +粲 +裤 +哎 +疡 +嫖 +琵 +窟 +堪 +谛 +嘉 +儡 +鳝 +斩 +郾 +驸 +酊 +妄 +胜 +贺 +徙 +傅 +噌 +钢 +栅 +庇 +恋 +匝 +巯 +邈 +尸 +锚 +粗 +佟 +蛟 +薹 +纵 +蚊 +郅 +绢 +锐 +苗 +俞 +篆 +淆 +膀 +鲜 +煎 +诶 +秽 +寻 +涮 +刺 +怀 +噶 +巨 +褰 +魅 +灶 +灌 +桉 +藕 +谜 +舸 +薄 +搀 +恽 +借 +牯 +痉 +渥 +愿 +亓 +耘 +杠 +柩 +锔 +蚶 +钣 +珈 +喘 +蹒 +幽 +赐 +稗 +晤 +莱 +泔 +扯 +肯 +菪 +裆 +腩 +豉 +疆 +骜 +腐 +倭 +珏 +唔 +粮 +亡 +润 +慰 +伽 +橄 +玄 +誉 +醐 +胆 +龊 +粼 +塬 +陇 +彼 +削 +嗣 +绾 +芽 +妗 +垭 +瘴 +爽 +薏 +寨 +龈 +泠 +弹 +赢 +漪 +猫 +嘧 +涂 +恤 +圭 +茧 +烽 +屑 +痕 +巾 +赖 +荸 +凰 +腮 +畈 +亵 +蹲 +偃 +苇 +澜 +艮 +换 +骺 +烘 +苕 +梓 +颉 +肇 +哗 +悄 +氤 +涠 +葬 +屠 +鹭 +植 +竺 +佯 +诣 +鲇 +瘀 +鲅 +邦 +移 +滁 +冯 +耕 +癔 +戌 +茬 +沁 +巩 +悠 +湘 +洪 +痹 +锟 +循 +谋 +腕 +鳃 +钠 +捞 +焉 +迎 +碱 +伫 +急 +榷 +奈 +邝 +卯 +辄 +皲 +卟 +醛 +畹 +忧 +稳 +雄 +昼 +缩 +阈 +睑 +扌 +耗 +曦 +涅 +捏 +瞧 +邕 +淖 +漉 +铝 +耦 +禹 +湛 +喽 +莼 +琅 +诸 +苎 +纂 +硅 +始 +嗨 +傥 +燃 +臂 +赅 +嘈 +呆 +贵 +屹 +壮 +肋 +亍 +蚀 +卅 +豹 +腆 +邬 +迭 +浊 +} +童 +螂 +捐 +圩 +勐 +触 +寞 +汊 +壤 +荫 +膺 +渌 +芳 +懿 +遴 +螈 +泰 +蓼 +蛤 +茜 +舅 +枫 +朔 +膝 +眙 +避 +梅 +判 +鹜 +璜 +牍 +缅 +垫 +藻 +黔 +侥 +惚 +懂 +踩 +腰 +腈 +札 +丞 +唾 +慈 +顿 +摹 +荻 +琬 +~ +斧 +沈 +滂 +胁 +胀 +幄 +莜 +Z +匀 +鄄 +掌 +绰 +茎 +焚 +赋 +萱 +谑 +汁 +铒 +瞎 +夺 +蜗 +野 +娆 +冀 +弯 +篁 +懵 +灞 +隽 +芡 +脘 +俐 +辩 +芯 +掺 +喏 +膈 +蝈 +觐 +悚 +踹 +蔗 +熠 +鼠 +呵 +抓 +橼 +峨 +畜 +缔 +禾 +崭 +弃 +熊 +摒 +凸 +拗 +穹 +蒙 +抒 +祛 +劝 +闫 +扳 +阵 +醌 +踪 +喵 +侣 +搬 +仅 +荧 +赎 +蝾 +琦 +买 +婧 +瞄 +寓 +皎 +冻 +赝 +箩 +莫 +瞰 +郊 +笫 +姝 +筒 +枪 +遣 +煸 +袋 +舆 +痱 +涛 +母 +〇 +启 +践 +耙 +绲 +盘 +遂 +昊 +搞 +槿 +诬 +纰 +泓 +惨 +檬 +亻 +越 +C +o +憩 +熵 +祷 +钒 +暧 +塔 +阗 +胰 +咄 +娶 +魔 +琶 +钞 +邻 +扬 +杉 +殴 +咽 +弓 +〆 +髻 +】 +吭 +揽 +霆 +拄 +殖 +脆 +彻 +岩 +芝 +勃 +辣 +剌 +钝 +嘎 +甄 +佘 +皖 +伦 +授 +徕 +憔 +挪 +皇 +庞 +稔 +芜 +踏 +溴 +兖 +卒 +擢 +饥 +鳞 +煲 +‰ +账 +颗 +叻 +斯 +捧 +鳍 +琮 +讹 +蛙 +纽 +谭 +酸 +兔 +莒 +睇 +伟 +觑 +羲 +嗜 +宜 +褐 +旎 +辛 +卦 +诘 +筋 +鎏 +溪 +挛 +熔 +阜 +晰 +鳅 +丢 +奚 +灸 +呱 +献 +陉 +黛 +鸪 +甾 +萨 +疮 +拯 +洲 +疹 +辑 +叙 +恻 +谒 +允 +柔 +烂 +氏 +逅 +漆 +拎 +惋 +扈 +湟 +纭 +啕 +掬 +擞 +哥 +忽 +涤 +鸵 +靡 +郗 +瓷 +扁 +廊 +怨 +雏 +钮 +敦 +E +懦 +憋 +汀 +拚 +啉 +腌 +岸 +f +痼 +瞅 +尊 +咀 +眩 +飙 +忌 +仝 +迦 +熬 +毫 +胯 +篑 +茄 +腺 +凄 +舛 +碴 +锵 +诧 +羯 +後 +漏 +汤 +宓 +仞 +蚁 +壶 +谰 +皑 +铄 +棰 +罔 +辅 +晶 +苦 +牟 +闽 +\ +烃 +饮 +聿 +丙 +蛳 +朱 +煤 +涔 +鳖 +犁 +罐 +荼 +砒 +淦 +妤 +黏 +戎 +孑 +婕 +瑾 +戢 +钵 +枣 +捋 +砥 +衩 +狙 +桠 +稣 +阎 +肃 +梏 +诫 +孪 +昶 +婊 +衫 +嗔 +侃 +塞 +蜃 +樵 +峒 +貌 +屿 +欺 +缫 +阐 +栖 +诟 +珞 +荭 +吝 +萍 +嗽 +恂 +啻 +蜴 +磬 +峋 +俸 +豫 +谎 +徊 +镍 +韬 +魇 +晴 +U +囟 +猜 +蛮 +坐 +囿 +伴 +亭 +肝 +佗 +蝠 +妃 +胞 +滩 +榴 +氖 +垩 +苋 +砣 +扪 +馏 +姓 +轩 +厉 +夥 +侈 +禀 +垒 +岑 +赏 +钛 +辐 +痔 +披 +纸 +碳 +“ +坞 +蠓 +挤 +荥 +沅 +悔 +铧 +帼 +蒌 +蝇 +a +p +y +n +g +哀 +浆 +瑶 +凿 +桶 +馈 +皮 +奴 +苜 +佤 +伶 +晗 +铱 +炬 +优 +弊 +氢 +恃 +甫 +攥 +端 +锌 +灰 +稹 +炝 +曙 +邋 +亥 +眶 +碾 +拉 +萝 +绔 +捷 +浍 +腋 +姑 +菖 +凌 +涞 +麽 +锢 +桨 +潢 +绎 +镰 +殆 +锑 +渝 +铬 +困 +绽 +觎 +匈 +糙 +暑 +裹 +鸟 +盔 +肽 +迷 +綦 +『 +亳 +佝 +俘 +钴 +觇 +骥 +仆 +疝 +跪 +婶 +郯 +瀹 +唉 +脖 +踞 +针 +晾 +忒 +扼 +瞩 +叛 +椒 +疟 +嗡 +邗 +肆 +跆 +玫 +忡 +捣 +咧 +唆 +艄 +蘑 +潦 +笛 +阚 +沸 +泻 +掊 +菽 +贫 +斥 +髂 +孢 +镂 +赂 +麝 +鸾 +屡 +衬 +苷 +恪 +叠 +希 +粤 +爻 +喝 +茫 +惬 +郸 +绻 +庸 +撅 +碟 +宄 +妹 +膛 +叮 +饵 +崛 +嗲 +椅 +冤 +搅 +咕 +敛 +尹 +垦 +闷 +蝉 +霎 +勰 +败 +蓑 +泸 +肤 +鹌 +幌 +焦 +浠 +鞍 +刁 +舰 +乙 +竿 +裔 +。 +茵 +函 +伊 +兄 +丨 +娜 +匍 +謇 +莪 +宥 +似 +蝽 +翳 +酪 +翠 +粑 +薇 +祢 +骏 +赠 +叫 +Q +噤 +噻 +竖 +芗 +莠 +潭 +俊 +羿 +耜 +O +郫 +趁 +嗪 +囚 +蹶 +芒 +洁 +笋 +鹑 +敲 +硝 +啶 +堡 +渲 +揩 +』 +携 +宿 +遒 +颍 +扭 +棱 +割 +萜 +蔸 +葵 +琴 +捂 +饰 +衙 +耿 +掠 +募 +岂 +窖 +涟 +蔺 +瘤 +柞 +瞪 +怜 +匹 +距 +楔 +炜 +哆 +秦 +缎 +幼 +茁 +绪 +痨 +恨 +楸 +娅 +瓦 +桩 +雪 +嬴 +伏 +榔 +妥 +铿 +拌 +眠 +雍 +缇 +‘ +卓 +搓 +哌 +觞 +噩 +屈 +哧 +髓 +咦 +巅 +娑 +侑 +淫 +膳 +祝 +勾 +姊 +莴 +胄 +疃 +薛 +蜷 +胛 +巷 +芙 +芋 +熙 +闰 +勿 +窃 +狱 +剩 +钏 +幢 +陟 +铛 +慧 +靴 +耍 +k +浙 +浇 +飨 +惟 +绗 +祜 +澈 +啼 +咪 +磷 +摞 +诅 +郦 +抹 +跃 +壬 +吕 +肖 +琏 +颤 +尴 +剡 +抠 +凋 +赚 +泊 +津 +宕 +殷 +倔 +氲 +漫 +邺 +涎 +怠 +$ +垮 +荬 +遵 +俏 +叹 +噢 +饽 +蜘 +孙 +筵 +疼 +鞭 +羧 +牦 +箭 +潴 +c +眸 +祭 +髯 +啖 +坳 +愁 +芩 +驮 +倡 +巽 +穰 +沃 +胚 +怒 +凤 +槛 +剂 +趵 +嫁 +v +邢 +灯 +鄢 +桐 +睽 +檗 +锯 +槟 +婷 +嵋 +圻 +诗 +蕈 +颠 +遭 +痢 +芸 +怯 +馥 +竭 +锗 +徜 +恭 +遍 +籁 +剑 +嘱 +苡 +龄 +僧 +桑 +潸 +弘 +澶 +楹 +悲 +讫 +愤 +腥 +悸 +谍 +椹 +呢 +桓 +葭 +攫 +阀 +翰 +躲 +敖 +柑 +郎 +笨 +橇 +呃 +魁 +燎 +脓 +葩 +磋 +垛 +玺 +狮 +沓 +砜 +蕊 +锺 +罹 +蕉 +翱 +虐 +闾 +巫 +旦 +茱 +嬷 +枯 +鹏 +贡 +芹 +汛 +矫 +绁 +拣 +禺 +佃 +讣 +舫 +惯 +乳 +趋 +疲 +挽 +岚 +虾 +衾 +蠹 +蹂 +飓 +氦 +铖 +孩 +稞 +瑜 +壅 +掀 +勘 +妓 +畅 +髋 +W +庐 +牲 +蓿 +榕 +练 +垣 +唱 +邸 +菲 +昆 +婺 +穿 +绡 +麒 +蚱 +掂 +愚 +泷 +涪 +漳 +妩 +娉 +榄 +讷 +觅 +旧 +藤 +煮 +呛 +柳 +腓 +叭 +庵 +烷 +阡 +罂 +蜕 +擂 +猖 +咿 +媲 +脉 +【 +沏 +貅 +黠 +熏 +哲 +烁 +坦 +酵 +兜 +× +潇 +撒 +剽 +珩 +圹 +乾 +摸 +樟 +帽 +嗒 +襄 +魂 +轿 +憬 +锡 +〕 +喃 +皆 +咖 +隅 +脸 +残 +泮 +袂 +鹂 +珊 +囤 +捆 +咤 +误 +徨 +闹 +淙 +芊 +淋 +怆 +囗 +拨 +梳 +渤 +R +G +绨 +蚓 +婀 +幡 +狩 +麾 +谢 +唢 +裸 +旌 +伉 +纶 +裂 +驳 +砼 +咛 +澄 +樨 +蹈 +宙 +澍 +倍 +貔 +操 +勇 +蟠 +摈 +砧 +虬 +够 +缁 +悦 +藿 +撸 +艹 +摁 +淹 +豇 +虎 +榭 +ˉ +吱 +d +° +喧 +荀 +踱 +侮 +奋 +偕 +饷 +犍 +惮 +坑 +璎 +徘 +宛 +妆 +袈 +倩 +窦 +昂 +荏 +乖 +K +怅 +撰 +鳙 +牙 +袁 +酞 +X +痿 +琼 +闸 +雁 +趾 +荚 +虻 +涝 +《 +杏 +韭 +偈 +烤 +绫 +鞘 +卉 +症 +遢 +蓥 +诋 +杭 +荨 +匆 +竣 +簪 +辙 +敕 +虞 +丹 +缭 +咩 +黟 +m +淤 +瑕 +咂 +铉 +硼 +茨 +嶂 +痒 +畸 +敬 +涿 +粪 +窘 +熟 +叔 +嫔 +盾 +忱 +裘 +憾 +梵 +赡 +珙 +咯 +娘 +庙 +溯 +胺 +葱 +痪 +摊 +荷 +卞 +乒 +髦 +寐 +铭 +坩 +胗 +枷 +爆 +溟 +嚼 +羚 +砬 +轨 +惊 +挠 +罄 +竽 +菏 +氧 +浅 +楣 +盼 +枢 +炸 +阆 +杯 +谏 +噬 +淇 +渺 +俪 +秆 +墓 +泪 +跻 +砌 +痰 +垡 +渡 +耽 +釜 +讶 +鳎 +煞 +呗 +韶 +舶 +绷 +鹳 +缜 +旷 +铊 +皱 +龌 +檀 +霖 +奄 +槐 +艳 +蝶 +旋 +哝 +赶 +骞 +蚧 +腊 +盈 +丁 +` +蜚 +矸 +蝙 +睨 +嚓 +僻 +鬼 +醴 +夜 +彝 +磊 +笔 +拔 +栀 +糕 +厦 +邰 +纫 +逭 +纤 +眦 +膊 +馍 +躇 +烯 +蘼 +冬 +诤 +暄 +骶 +哑 +瘠 +」 +臊 +丕 +愈 +咱 +螺 +擅 +跋 +搏 +硪 +谄 +笠 +淡 +嘿 +骅 +谧 +鼎 +皋 +姚 +歼 +蠢 +驼 +耳 +胬 +挝 +涯 +狗 +蒽 +孓 +犷 +凉 +芦 +箴 +铤 +孤 +嘛 +坤 +V +茴 +朦 +挞 +尖 +橙 +诞 +搴 +碇 +洵 +浚 +帚 +蜍 +漯 +柘 +嚎 +讽 +芭 +荤 +咻 +祠 +秉 +跖 +埃 +吓 +糯 +眷 +馒 +惹 +娼 +鲑 +嫩 +讴 +轮 +瞥 +靶 +褚 +乏 +缤 +宋 +帧 +删 +驱 +碎 +扑 +俩 +俄 +偏 +涣 +竹 +噱 +皙 +佰 +渚 +唧 +斡 +# +镉 +刀 +崎 +筐 +佣 +夭 +贰 +肴 +峙 +哔 +艿 +匐 +牺 +镛 +缘 +仡 +嫡 +劣 +枸 +堀 +梨 +簿 +鸭 +蒸 +亦 +稽 +浴 +{ +衢 +束 +槲 +j +阁 +揍 +疥 +棋 +潋 +聪 +窜 +乓 +睛 +插 +冉 +阪 +苍 +搽 +「 +蟾 +螟 +幸 +仇 +樽 +撂 +慢 +跤 +幔 +俚 +淅 +覃 +觊 +溶 +妖 +帛 +侨 +曰 +妾 +泗 +· +: +瀘 +風 +Ë +( +) +∶ +紅 +紗 +瑭 +雲 +頭 +鶏 +財 +許 +• +¥ +樂 +焗 +麗 +— +; +滙 +東 +榮 +繪 +興 +… +門 +業 +π +楊 +國 +顧 +é +盤 +寳 +Λ +龍 +鳳 +島 +誌 +緣 +結 +銭 +萬 +勝 +祎 +璟 +優 +歡 +臨 +時 +購 += +★ +藍 +昇 +鐵 +觀 +勅 +農 +聲 +畫 +兿 +術 +發 +劉 +記 +專 +耑 +園 +書 +壴 +種 +Ο +● +褀 +號 +銀 +匯 +敟 +锘 +葉 +橪 +廣 +進 +蒄 +鑽 +阝 +祙 +貢 +鍋 +豊 +夬 +喆 +團 +閣 +開 +燁 +賓 +館 +酡 +沔 +順 ++ +硚 +劵 +饸 +陽 +車 +湓 +復 +萊 +氣 +軒 +華 +堃 +迮 +纟 +戶 +馬 +學 +裡 +電 +嶽 +獨 +マ +シ +サ +ジ +燘 +袪 +環 +❤ +臺 +灣 +専 +賣 +孖 +聖 +攝 +線 +▪ +α +傢 +俬 +夢 +達 +莊 +喬 +貝 +薩 +劍 +羅 +壓 +棛 +饦 +尃 +璈 +囍 +醫 +G +I +A +# +N +鷄 +髙 +嬰 +啓 +約 +隹 +潔 +賴 +藝 +~ +寶 +籣 +麺 +  +嶺 +√ +義 +網 +峩 +長 +∧ +魚 +機 +構 +② +鳯 +偉 +L +B +㙟 +畵 +鴿 +' +詩 +溝 +嚞 +屌 +藔 +佧 +玥 +蘭 +織 +1 +3 +9 +0 +7 +點 +砭 +鴨 +鋪 +銘 +廳 +弍 +‧ +創 +湯 +坶 +℃ +卩 +骝 +& +烜 +荘 +當 +潤 +扞 +係 +懷 +碶 +钅 +蚨 +讠 +☆ +叢 +爲 +埗 +涫 +塗 +→ +楽 +現 +鯨 +愛 +瑪 +鈺 +忄 +悶 +藥 +飾 +樓 +視 +孬 +ㆍ +燚 +苪 +師 +① +丼 +锽 +│ +韓 +標 +è +兒 +閏 +匋 +張 +漢 +Ü +髪 +會 +閑 +檔 +習 +裝 +の +峯 +菘 +輝 +И +雞 +釣 +億 +浐 +K +O +R +8 +H +E +P +T +W +D +S +C +M +F +姌 +饹 +» +晞 +廰 +ä +嵯 +鷹 +負 +飲 +絲 +冚 +楗 +澤 +綫 +區 +❋ +← +質 +靑 +揚 +③ +滬 +統 +産 +協 +﹑ +乸 +畐 +經 +運 +際 +洺 +岽 +為 +粵 +諾 +崋 +豐 +碁 +ɔ +V +2 +6 +齋 +誠 +訂 +´ +勑 +雙 +陳 +無 +í +泩 +媄 +夌 +刂 +i +c +t +o +r +a +嘢 +耄 +燴 +暃 +壽 +媽 +靈 +抻 +體 +唻 +É +冮 +甹 +鎮 +錦 +ʌ +蜛 +蠄 +尓 +駕 +戀 +飬 +逹 +倫 +貴 +極 +Я +Й +寬 +磚 +嶪 +郎 +職 +| +間 +n +d +剎 +伈 +課 +飛 +橋 +瘊 +№ +譜 +骓 +圗 +滘 +縣 +粿 +咅 +養 +濤 +彳 +® +% +Ⅱ +啰 +㴪 +見 +矞 +薬 +糁 +邨 +鲮 +顔 +罱 +З +選 +話 +贏 +氪 +俵 +競 +瑩 +繡 +枱 +β +綉 +á +獅 +爾 +™ +麵 +戋 +淩 +徳 +個 +劇 +場 +務 +簡 +寵 +h +實 +膠 +轱 +圖 +築 +嘣 +樹 +㸃 +營 +耵 +孫 +饃 +鄺 +飯 +麯 +遠 +輸 +坫 +孃 +乚 +閃 +鏢 +㎡ +題 +廠 +關 +↑ +爺 +將 +軍 +連 +篦 +覌 +參 +箸 +- +窠 +棽 +寕 +夀 +爰 +歐 +呙 +閥 +頡 +熱 +雎 +垟 +裟 +凬 +勁 +帑 +馕 +夆 +疌 +枼 +馮 +貨 +蒤 +樸 +彧 +旸 +靜 +龢 +暢 +㐱 +鳥 +珺 +鏡 +灡 +爭 +堷 +廚 +Ó +騰 +診 +┅ +蘇 +褔 +凱 +頂 +豕 +亞 +帥 +嘬 +⊥ +仺 +桖 +複 +饣 +絡 +穂 +顏 +棟 +納 +▏ +濟 +親 +設 +計 +攵 +埌 +烺 +ò +頤 +燦 +蓮 +撻 +節 +講 +濱 +濃 +娽 +洳 +朿 +燈 +鈴 +護 +膚 +铔 +過 +補 +Z +U +5 +4 +坋 +闿 +䖝 +餘 +缐 +铞 +貿 +铪 +桼 +趙 +鍊 +[ +㐂 +垚 +菓 +揸 +捲 +鐘 +滏 +𣇉 +爍 +輪 +燜 +鴻 +鮮 +動 +鹞 +鷗 +丄 +慶 +鉌 +翥 +飮 +腸 +⇋ +漁 +覺 +來 +熘 +昴 +翏 +鲱 +圧 +鄉 +萭 +頔 +爐 +嫚 +г +貭 +類 +聯 +幛 +輕 +訓 +鑒 +夋 +锨 +芃 +珣 +䝉 +扙 +嵐 +銷 +處 +ㄱ +語 +誘 +苝 +歸 +儀 +燒 +楿 +內 +粢 +葒 +奧 +麥 +礻 +滿 +蠔 +穵 +瞭 +態 +鱬 +榞 +硂 +鄭 +黃 +煙 +祐 +奓 +逺 +* +瑄 +獲 +聞 +薦 +讀 +這 +樣 +決 +問 +啟 +們 +執 +説 +轉 +單 +隨 +唘 +帶 +倉 +庫 +還 +贈 +尙 +皺 +■ +餅 +產 +○ +∈ +報 +狀 +楓 +賠 +琯 +嗮 +禮 +` +傳 +> +≤ +嗞 +Φ +≥ +換 +咭 +∣ +↓ +曬 +ε +応 +寫 +″ +終 +様 +純 +費 +療 +聨 +凍 +壐 +郵 +ü +黒 +∫ +製 +塊 +調 +軽 +確 +撃 +級 +馴 +Ⅲ +涇 +繹 +數 +碼 +證 +狒 +処 +劑 +< +晧 +賀 +衆 +] +櫥 +兩 +陰 +絶 +對 +鯉 +憶 +◎ +p +e +Y +蕒 +煖 +頓 +測 +試 +鼽 +僑 +碩 +妝 +帯 +≈ +鐡 +舖 +權 +喫 +倆 +ˋ +該 +悅 +ā +俫 +. +f +s +b +m +k +g +u +j +貼 +淨 +濕 +針 +適 +備 +l +/ +給 +謢 +強 +觸 +衛 +與 +⊙ +$ +緯 +變 +⑴ +⑵ +⑶ +㎏ +殺 +∩ +幚 +─ +價 +▲ +離 +ú +ó +飄 +烏 +関 +閟 +﹝ +﹞ +邏 +輯 +鍵 +驗 +訣 +導 +歷 +屆 +層 +▼ +儱 +錄 +熳 +ē +艦 +吋 +錶 +辧 +飼 +顯 +④ +禦 +販 +気 +対 +枰 +閩 +紀 +幹 +瞓 +貊 +淚 +△ +眞 +墊 +Ω +獻 +褲 +縫 +緑 +亜 +鉅 +餠 +{ +} +◆ +蘆 +薈 +█ +◇ +溫 +彈 +晳 +粧 +犸 +穩 +訊 +崬 +凖 +熥 +П +舊 +條 +紋 +圍 +Ⅳ +筆 +尷 +難 +雜 +錯 +綁 +識 +頰 +鎖 +艶 +□ +殁 +殼 +⑧ +├ +▕ +鵬 +ǐ +ō +ǒ +糝 +綱 +▎ +μ +盜 +饅 +醬 +籤 +蓋 +釀 +鹽 +據 +à +ɡ +辦 +◥ +彐 +┌ +婦 +獸 +鲩 +伱 +ī +蒟 +蒻 +齊 +袆 +腦 +寧 +凈 +妳 +煥 +詢 +偽 +謹 +啫 +鯽 +騷 +鱸 +損 +傷 +鎻 +髮 +買 +冏 +儥 +両 +﹢ +∞ +載 +喰 +z +羙 +悵 +燙 +曉 +員 +組 +徹 +艷 +痠 +鋼 +鼙 +縮 +細 +嚒 +爯 +≠ +維 +" +鱻 +壇 +厍 +帰 +浥 +犇 +薡 +軎 +² +應 +醜 +刪 +緻 +鶴 +賜 +噁 +軌 +尨 +镔 +鷺 +槗 +彌 +葚 +濛 +請 +溇 +緹 +賢 +訪 +獴 +瑅 +資 +縤 +陣 +蕟 +栢 +韻 +祼 +恁 +伢 +謝 +劃 +涑 +總 +衖 +踺 +砋 +凉 +籃 +駿 +苼 +瘋 +昽 +紡 +驊 +腎 +﹗ +響 +杋 +剛 +嚴 +禪 +歓 +槍 +傘 +檸 +檫 +炣 +勢 +鏜 +鎢 +銑 +尐 +減 +奪 +惡 +θ +僮 +婭 +臘 +ū +ì +殻 +鉄 +∑ +蛲 +焼 +緖 +續 +紹 +懮 \ No newline at end of file diff --git a/requirements.txt b/requirements.txt new file mode 100644 index 0000000000000000000000000000000000000000..2f78d9ab775346d2688a980a4184859d5c94e5b6 --- /dev/null +++ b/requirements.txt @@ -0,0 +1,3 @@ +Pillow==9.5.0 +gradio==3.50.0 +xformers==0.0.20 \ No newline at end of file diff --git a/style.css b/style.css new file mode 100644 index 0000000000000000000000000000000000000000..b86d2b64169ed66088f12dbf14d030af0125c600 --- /dev/null +++ b/style.css @@ -0,0 +1,32 @@ +#banner { + max-width: 400px; + margin: auto; + box-shadow: 0 2px 20px rgba(0, 0, 0, 0.5) !important; + border-radius: 20px; +} + +.run { + background-color: #624AFF !important; + color: #FFFFFF !important; + border-radius: 2px !important; + box-shadow: 0 3px 5px rgba(0, 0, 0, 0.5) !important; +} +.run:active { + background-color: #d96565 !important; +} +.run:hover { + background-color: #a079f5 !important; +} +/* tab button style */ +button.svelte-kqij2n { + margin-bottom: -1px; + border: 1px solid transparent; + border-color: transparent; + border-bottom: none; + color: #9CA3AF !important; + font-size: 16px; +} +button.selected.svelte-kqij2n { + background: #ddd8f9 !important; + color: rgb(62, 7, 240) !important; +} \ No newline at end of file diff --git a/t3_dataset.py b/t3_dataset.py new file mode 100644 index 0000000000000000000000000000000000000000..669b4598baf12654f8badc9bbbd5556ba33b8239 --- /dev/null +++ b/t3_dataset.py @@ -0,0 +1,447 @@ +import os +import numpy as np +import cv2 +import random +import math +from PIL import Image, ImageDraw, ImageFont +from torch.utils.data import Dataset, DataLoader +from dataset_util import load, show_bbox_on_image + + +phrase_list = [ + ', content and position of the texts are ', + ', textual material depicted in the image are ', + ', texts that says ', + ', captions shown in the snapshot are ', + ', with the words of ', + ', that reads ', + ', the written materials on the picture: ', + ', these texts are written on it: ', + ', captions are ', + ', content of the text in the graphic is ' +] + + +def insert_spaces(string, nSpace): + if nSpace == 0: + return string + new_string = "" + for char in string: + new_string += char + " " * nSpace + return new_string[:-nSpace] + + +def draw_glyph(font, text): + g_size = 50 + W, H = (512, 80) + new_font = font.font_variant(size=g_size) + img = Image.new(mode='1', size=(W, H), color=0) + draw = ImageDraw.Draw(img) + left, top, right, bottom = new_font.getbbox(text) + text_width = max(right-left, 5) + text_height = max(bottom - top, 5) + ratio = min(W*0.9/text_width, H*0.9/text_height) + new_font = font.font_variant(size=int(g_size*ratio)) + + text_width, text_height = new_font.getsize(text) + offset_x, offset_y = new_font.getoffset(text) + x = (img.width - text_width) // 2 + y = (img.height - text_height) // 2 - offset_y//2 + draw.text((x, y), text, font=new_font, fill='white') + img = np.expand_dims(np.array(img), axis=2).astype(np.float64) + return img + + +def draw_glyph2(font, text, polygon, vertAng=10, scale=1, width=512, height=512, add_space=True): + enlarge_polygon = polygon*scale + rect = cv2.minAreaRect(enlarge_polygon) + box = cv2.boxPoints(rect) + box = np.int0(box) + w, h = rect[1] + angle = rect[2] + if angle < -45: + angle += 90 + angle = -angle + if w < h: + angle += 90 + + vert = False + if (abs(angle) % 90 < vertAng or abs(90-abs(angle) % 90) % 90 < vertAng): + _w = max(box[:, 0]) - min(box[:, 0]) + _h = max(box[:, 1]) - min(box[:, 1]) + if _h >= _w: + vert = True + angle = 0 + + img = np.zeros((height*scale, width*scale, 3), np.uint8) + img = Image.fromarray(img) + + # infer font size + image4ratio = Image.new("RGB", img.size, "white") + draw = ImageDraw.Draw(image4ratio) + _, _, _tw, _th = draw.textbbox(xy=(0, 0), text=text, font=font) + text_w = min(w, h) * (_tw / _th) + if text_w <= max(w, h): + # add space + if len(text) > 1 and not vert and add_space: + for i in range(1, 100): + text_space = insert_spaces(text, i) + _, _, _tw2, _th2 = draw.textbbox(xy=(0, 0), text=text_space, font=font) + if min(w, h) * (_tw2 / _th2) > max(w, h): + break + text = insert_spaces(text, i-1) + font_size = min(w, h)*0.80 + else: + shrink = 0.75 if vert else 0.85 + font_size = min(w, h) / (text_w/max(w, h)) * shrink + new_font = font.font_variant(size=int(font_size)) + + left, top, right, bottom = new_font.getbbox(text) + text_width = right-left + text_height = bottom - top + + layer = Image.new('RGBA', img.size, (0, 0, 0, 0)) + draw = ImageDraw.Draw(layer) + if not vert: + draw.text((rect[0][0]-text_width//2, rect[0][1]-text_height//2-top), text, font=new_font, fill=(255, 255, 255, 255)) + else: + x_s = min(box[:, 0]) + _w//2 - text_height//2 + y_s = min(box[:, 1]) + for c in text: + draw.text((x_s, y_s), c, font=new_font, fill=(255, 255, 255, 255)) + _, _t, _, _b = new_font.getbbox(c) + y_s += _b + + rotated_layer = layer.rotate(angle, expand=1, center=(rect[0][0], rect[0][1])) + + x_offset = int((img.width - rotated_layer.width) / 2) + y_offset = int((img.height - rotated_layer.height) / 2) + img.paste(rotated_layer, (x_offset, y_offset), rotated_layer) + img = np.expand_dims(np.array(img.convert('1')), axis=2).astype(np.float64) + return img + + +def get_caption_pos(ori_caption, pos_idxs, prob=1.0, place_holder='*'): + idx2pos = { + 0: " top left", + 1: " top", + 2: " top right", + 3: " left", + 4: random.choice([" middle", " center"]), + 5: " right", + 6: " bottom left", + 7: " bottom", + 8: " bottom right" + } + new_caption = ori_caption + random.choice(phrase_list) + pos = '' + for i in range(len(pos_idxs)): + if random.random() < prob and pos_idxs[i] > 0: + pos += place_holder + random.choice([' located', ' placed', ' positioned', '']) + random.choice([' at', ' in', ' on']) + idx2pos[pos_idxs[i]] + ', ' + else: + pos += place_holder + ' , ' + pos = pos[:-2] + '.' + new_caption += pos + return new_caption + + +def generate_random_rectangles(w, h, box_num): + rectangles = [] + for i in range(box_num): + x = random.randint(0, w) + y = random.randint(0, h) + w = random.randint(16, 256) + h = random.randint(16, 96) + angle = random.randint(-45, 45) + p1 = (x, y) + p2 = (x + w, y) + p3 = (x + w, y + h) + p4 = (x, y + h) + center = ((x + x + w) / 2, (y + y + h) / 2) + p1 = rotate_point(p1, center, angle) + p2 = rotate_point(p2, center, angle) + p3 = rotate_point(p3, center, angle) + p4 = rotate_point(p4, center, angle) + rectangles.append((p1, p2, p3, p4)) + return rectangles + + +def rotate_point(point, center, angle): + # rotation + angle = math.radians(angle) + x = point[0] - center[0] + y = point[1] - center[1] + x1 = x * math.cos(angle) - y * math.sin(angle) + y1 = x * math.sin(angle) + y * math.cos(angle) + x1 += center[0] + y1 += center[1] + return int(x1), int(y1) + + +class T3DataSet(Dataset): + def __init__( + self, + json_path, + max_lines=5, + max_chars=20, + place_holder='*', + font_path='./font/Arial_Unicode.ttf', + caption_pos_prob=1.0, + mask_pos_prob=1.0, + mask_img_prob=0.5, + for_show=False, + using_dlc=False, + glyph_scale=1, + percent=1.0, + debug=False, + wm_thresh=1.0, + ): + assert isinstance(json_path, (str, list)) + if isinstance(json_path, str): + json_path = [json_path] + data_list = [] + self.using_dlc = using_dlc + self.max_lines = max_lines + self.max_chars = max_chars + self.place_holder = place_holder + self.font = ImageFont.truetype(font_path, size=60) + self.caption_pos_porb = caption_pos_prob + self.mask_pos_prob = mask_pos_prob + self.mask_img_prob = mask_img_prob + self.for_show = for_show + self.glyph_scale = glyph_scale + self.wm_thresh = wm_thresh + for jp in json_path: + data_list += self.load_data(jp, percent) + self.data_list = data_list + print(f'All dataset loaded, imgs={len(self.data_list)}') + self.debug = debug + if self.debug: + self.tmp_items = [i for i in range(100)] + + def load_data(self, json_path, percent): + content = load(json_path) + d = [] + count = 0 + wm_skip = 0 + max_img = len(content['data_list']) * percent + for gt in content['data_list']: + if len(d) > max_img: + break + if 'wm_score' in gt and gt['wm_score'] > self.wm_thresh: # wm_score > thresh will be skiped as an img with watermark + wm_skip += 1 + continue + data_root = content['data_root'] + if self.using_dlc: + data_root = data_root.replace('/data/vdb', '/mnt/data', 1) + img_path = os.path.join(data_root, gt['img_name']) + info = {} + info['img_path'] = img_path + info['caption'] = gt['caption'] if 'caption' in gt else '' + if self.place_holder in info['caption']: + count += 1 + info['caption'] = info['caption'].replace(self.place_holder, " ") + if 'annotations' in gt: + polygons = [] + invalid_polygons = [] + texts = [] + languages = [] + pos = [] + for annotation in gt['annotations']: + if len(annotation['polygon']) == 0: + continue + if 'valid' in annotation and annotation['valid'] is False: + invalid_polygons.append(annotation['polygon']) + continue + polygons.append(annotation['polygon']) + texts.append(annotation['text']) + languages.append(annotation['language']) + if 'pos' in annotation: + pos.append(annotation['pos']) + info['polygons'] = [np.array(i) for i in polygons] + info['invalid_polygons'] = [np.array(i) for i in invalid_polygons] + info['texts'] = texts + info['language'] = languages + info['pos'] = pos + d.append(info) + print(f'{json_path} loaded, imgs={len(d)}, wm_skip={wm_skip}') + if count > 0: + print(f"Found {count} image's caption contain placeholder: {self.place_holder}, change to ' '...") + return d + + def __getitem__(self, item): + item_dict = {} + if self.debug: # sample fixed items + item = self.tmp_items.pop() + print(f'item = {item}') + cur_item = self.data_list[item] + # img + target = np.array(Image.open(cur_item['img_path']).convert('RGB')) + if target.shape[0] != 512 or target.shape[1] != 512: + target = cv2.resize(target, (512, 512)) + target = (target.astype(np.float32) / 127.5) - 1.0 + item_dict['img'] = target + # caption + item_dict['caption'] = cur_item['caption'] + item_dict['glyphs'] = [] + item_dict['gly_line'] = [] + item_dict['positions'] = [] + item_dict['texts'] = [] + item_dict['language'] = [] + item_dict['inv_mask'] = [] + texts = cur_item.get('texts', []) + if len(texts) > 0: + idxs = [i for i in range(len(texts))] + if len(texts) > self.max_lines: + sel_idxs = random.sample(idxs, self.max_lines) + unsel_idxs = [i for i in idxs if i not in sel_idxs] + else: + sel_idxs = idxs + unsel_idxs = [] + if len(cur_item['pos']) > 0: + pos_idxs = [cur_item['pos'][i] for i in sel_idxs] + else: + pos_idxs = [-1 for i in sel_idxs] + item_dict['caption'] = get_caption_pos(item_dict['caption'], pos_idxs, self.caption_pos_porb, self.place_holder) + item_dict['polygons'] = [cur_item['polygons'][i] for i in sel_idxs] + item_dict['texts'] = [cur_item['texts'][i][:self.max_chars] for i in sel_idxs] + item_dict['language'] = [cur_item['language'][i] for i in sel_idxs] + # glyphs + for idx, text in enumerate(item_dict['texts']): + gly_line = draw_glyph(self.font, text) + glyphs = draw_glyph2(self.font, text, item_dict['polygons'][idx], scale=self.glyph_scale) + item_dict['glyphs'] += [glyphs] + item_dict['gly_line'] += [gly_line] + # mask_pos + for polygon in item_dict['polygons']: + item_dict['positions'] += [self.draw_pos(polygon, self.mask_pos_prob)] + # inv_mask + invalid_polygons = cur_item['invalid_polygons'] if 'invalid_polygons' in cur_item else [] + if len(texts) > 0: + invalid_polygons += [cur_item['polygons'][i] for i in unsel_idxs] + item_dict['inv_mask'] = self.draw_inv_mask(invalid_polygons) + item_dict['hint'] = self.get_hint(item_dict['positions']) + if random.random() < self.mask_img_prob: + # randomly generate 0~3 masks + box_num = random.randint(0, 3) + boxes = generate_random_rectangles(512, 512, box_num) + boxes = np.array(boxes) + pos_list = item_dict['positions'].copy() + for i in range(box_num): + pos_list += [self.draw_pos(boxes[i], self.mask_pos_prob)] + mask = self.get_hint(pos_list) + masked_img = target*(1-mask) + else: + masked_img = np.zeros_like(target) + item_dict['masked_img'] = masked_img + + if self.for_show: + item_dict['img_name'] = os.path.split(cur_item['img_path'])[-1] + return item_dict + if len(texts) > 0: + del item_dict['polygons'] + # padding + n_lines = min(len(texts), self.max_lines) + item_dict['n_lines'] = n_lines + n_pad = self.max_lines - n_lines + if n_pad > 0: + item_dict['glyphs'] += [np.zeros((512*self.glyph_scale, 512*self.glyph_scale, 1))] * n_pad + item_dict['gly_line'] += [np.zeros((80, 512, 1))] * n_pad + item_dict['positions'] += [np.zeros((512, 512, 1))] * n_pad + item_dict['texts'] += [' '] * n_pad + item_dict['language'] += [' '] * n_pad + + return item_dict + + def __len__(self): + return len(self.data_list) + + def draw_inv_mask(self, polygons): + img = np.zeros((512, 512)) + for p in polygons: + pts = p.reshape((-1, 1, 2)) + cv2.fillPoly(img, [pts], color=255) + img = img[..., None] + return img/255. + + def draw_pos(self, ploygon, prob=1.0): + img = np.zeros((512, 512)) + rect = cv2.minAreaRect(ploygon) + w, h = rect[1] + small = False + if w < 20 or h < 20: + small = True + if random.random() < prob: + pts = ploygon.reshape((-1, 1, 2)) + cv2.fillPoly(img, [pts], color=255) + # 10% dilate / 10% erode / 5% dilatex2 5% erodex2 + random_value = random.random() + kernel = np.ones((3, 3), dtype=np.uint8) + if random_value < 0.7: + pass + elif random_value < 0.8: + img = cv2.dilate(img.astype(np.uint8), kernel, iterations=1) + elif random_value < 0.9 and not small: + img = cv2.erode(img.astype(np.uint8), kernel, iterations=1) + elif random_value < 0.95: + img = cv2.dilate(img.astype(np.uint8), kernel, iterations=2) + elif random_value < 1.0 and not small: + img = cv2.erode(img.astype(np.uint8), kernel, iterations=2) + img = img[..., None] + return img/255. + + def get_hint(self, positions): + if len(positions) == 0: + return np.zeros((512, 512, 1)) + return np.sum(positions, axis=0).clip(0, 1) + + +if __name__ == '__main__': + ''' + Run this script to show details of your dataset, such as ocr annotations, glyphs, prompts, etc. + ''' + from tqdm import tqdm + from matplotlib import pyplot as plt + import shutil + + show_imgs_dir = 'show_results' + show_count = 50 + if os.path.exists(show_imgs_dir): + shutil.rmtree(show_imgs_dir) + os.makedirs(show_imgs_dir) + plt.rcParams['axes.unicode_minus'] = False + json_paths = [ + '/path/of/your/dataset/data1.json', + '/path/of/your/dataset/data2.json', + # ... + ] + + dataset = T3DataSet(json_paths, for_show=True, max_lines=20, glyph_scale=2, mask_img_prob=1.0, caption_pos_prob=0.0) + train_loader = DataLoader(dataset=dataset, batch_size=1, shuffle=False, num_workers=0) + pbar = tqdm(total=show_count) + for i, data in enumerate(train_loader): + if i == show_count: + break + img = ((data['img'][0].numpy() + 1.0) / 2.0 * 255).astype(np.uint8) + masked_img = ((data['masked_img'][0].numpy() + 1.0) / 2.0 * 255)[..., ::-1].astype(np.uint8) + cv2.imwrite(os.path.join(show_imgs_dir, f'plots_{i}_masked.jpg'), masked_img) + if 'texts' in data and len(data['texts']) > 0: + texts = [x[0] for x in data['texts']] + img = show_bbox_on_image(Image.fromarray(img), data['polygons'], texts) + cv2.imwrite(os.path.join(show_imgs_dir, f'plots_{i}.jpg'), np.array(img)[..., ::-1]) + with open(os.path.join(show_imgs_dir, f'plots_{i}.txt'), 'w') as fin: + fin.writelines([data['caption'][0]]) + all_glyphs = [] + for k, glyphs in enumerate(data['glyphs']): + cv2.imwrite(os.path.join(show_imgs_dir, f'plots_{i}_glyph_{k}.jpg'), glyphs[0].numpy().astype(np.int32)*255) + all_glyphs += [glyphs[0].numpy().astype(np.int32)*255] + cv2.imwrite(os.path.join(show_imgs_dir, f'plots_{i}_allglyphs.jpg'), np.sum(all_glyphs, axis=0)) + for k, gly_line in enumerate(data['gly_line']): + cv2.imwrite(os.path.join(show_imgs_dir, f'plots_{i}_gly_line_{k}.jpg'), gly_line[0].numpy().astype(np.int32)*255) + for k, position in enumerate(data['positions']): + if position is not None: + cv2.imwrite(os.path.join(show_imgs_dir, f'plots_{i}_pos_{k}.jpg'), position[0].numpy().astype(np.int32)*255) + cv2.imwrite(os.path.join(show_imgs_dir, f'plots_{i}_hint.jpg'), data['hint'][0].numpy().astype(np.int32)*255) + cv2.imwrite(os.path.join(show_imgs_dir, f'plots_{i}_inv_mask.jpg'), np.array(img)[..., ::-1]*(1-data['inv_mask'][0].numpy().astype(np.int32))) + pbar.update(1) + pbar.close() diff --git a/util.py b/util.py new file mode 100644 index 0000000000000000000000000000000000000000..9c8b953f0d18a29f846763346150d7af086207d7 --- /dev/null +++ b/util.py @@ -0,0 +1,43 @@ +import datetime +import os +import cv2 + + +def save_images(img_list, folder): + if not os.path.exists(folder): + os.makedirs(folder) + now = datetime.datetime.now() + date_str = now.strftime("%Y-%m-%d") + folder_path = os.path.join(folder, date_str) + if not os.path.exists(folder_path): + os.makedirs(folder_path) + time_str = now.strftime("%H_%M_%S") + for idx, img in enumerate(img_list): + image_number = idx + 1 + filename = f"{time_str}_{image_number}.jpg" + save_path = os.path.join(folder_path, filename) + cv2.imwrite(save_path, img[..., ::-1]) + + +def check_channels(image): + channels = image.shape[2] if len(image.shape) == 3 else 1 + if channels == 1: + image = cv2.cvtColor(image, cv2.COLOR_GRAY2BGR) + elif channels > 3: + image = image[:, :, :3] + return image + + +def resize_image(img, max_length=768): + height, width = img.shape[:2] + max_dimension = max(height, width) + + if max_dimension > max_length: + scale_factor = max_length / max_dimension + new_width = int(round(width * scale_factor)) + new_height = int(round(height * scale_factor)) + new_size = (new_width, new_height) + img = cv2.resize(img, new_size) + height, width = img.shape[:2] + img = cv2.resize(img, (width-(width % 64), height-(height % 64))) + return img