inference/t2i_demo.py

import os
import yaml
import torch
import sys
sys.path.append(os.path.abspath('./'))
from inference.utils import *
from train import WurstCoreB
from gdf import DDPMSampler
from train import WurstCore_t2i as WurstCoreC
from core.utils import load_or_fail
import numpy as np
import random
import argparse
import gradio as gr


def parse_args():
    parser = argparse.ArgumentParser()
    parser.add_argument( '--height', type=int, default=2560, help='image height')
    parser.add_argument('--width', type=int, default=5120, help='image width')
    parser.add_argument('--seed', type=int, default=123, help='random seed')
    parser.add_argument('--dtype', type=str, default='bf16', help=' if bf16 does not work, change it to float32 ')
    parser.add_argument('--config_c', type=str, 
    default='configs/training/t2i.yaml' ,help='config file for stage c, latent generation')
    parser.add_argument('--config_b', type=str, 
    default='configs/inference/stage_b_1b.yaml' ,help='config file for stage b, latent decoding')
    parser.add_argument( '--prompt', type=str,
     default='A photo-realistic image of a west highland white terrier in the garden, high quality, detail rich, 8K', help='text prompt')
    parser.add_argument( '--num_image', type=int, default=1, help='how many images generated')
    parser.add_argument( '--output_dir', type=str, default='figures/output_results/', help='output directory for generated image')
    parser.add_argument( '--stage_a_tiled', action='store_true', help='whther or nor to use tiled decoding for stage a to save memory')
    parser.add_argument( '--pretrained_path', type=str, default='models/ultrapixel_t2i.safetensors', help='pretrained path of newly added paramter of UltraPixel')
    args = parser.parse_args()
    return args

def clear_image():
    return None
def load_message(height, width, seed, prompt, args, stage_a_tiled):
    args.height = height
    args.width = width
    args.seed  = seed
    args.prompt = prompt + ' rich detail, 4k, high quality'
    args.stage_a_tiled = stage_a_tiled
    return args
def get_image(height, width, seed, prompt, cfg, timesteps, stage_a_tiled):
    global args
    args = load_message(height, width, seed, prompt,  args, stage_a_tiled)
    torch.manual_seed(args.seed)
    random.seed(args.seed) 
    np.random.seed(args.seed)
    dtype = torch.bfloat16 if args.dtype == 'bf16' else torch.float

    captions = [args.prompt] * args.num_image
    height, width = args.height, args.width
    batch_size=1 
    height_lr, width_lr = get_target_lr_size(height / width, std_size=32)
    stage_c_latent_shape, stage_b_latent_shape = calculate_latent_sizes(height, width, batch_size=batch_size)
    stage_c_latent_shape_lr, stage_b_latent_shape_lr = calculate_latent_sizes(height_lr, width_lr, batch_size=batch_size)
   
    # Stage C Parameters
    extras.sampling_configs['cfg'] = 4
    extras.sampling_configs['shift'] = 1
    extras.sampling_configs['timesteps'] = 20
    extras.sampling_configs['t_start'] = 1.0
    extras.sampling_configs['sampler'] = DDPMSampler(extras.gdf)
    
    
    
    # Stage B Parameters
    extras_b.sampling_configs['cfg'] = 1.1
    extras_b.sampling_configs['shift'] = 1
    extras_b.sampling_configs['timesteps'] = 10
    extras_b.sampling_configs['t_start'] = 1.0

    for _, caption in enumerate(captions):

        
            batch = {'captions': [caption] * batch_size}
            #conditions = core.get_conditions(batch, models, extras, is_eval=True, is_unconditional=False, eval_image_embeds=False)
            #unconditions = core.get_conditions(batch, models, extras, is_eval=True, is_unconditional=True, eval_image_embeds=False)    
            
            conditions_b = core_b.get_conditions(batch, models_b, extras_b, is_eval=True, is_unconditional=False)
            unconditions_b = core_b.get_conditions(batch, models_b, extras_b, is_eval=True, is_unconditional=True)
            
            
            with torch.no_grad():
        
            
                models.generator.cuda()
                print('STAGE C GENERATION***************************')
                with torch.cuda.amp.autocast(dtype=dtype):
                    sampled_c = generation_c(batch, models, extras, core, stage_c_latent_shape, stage_c_latent_shape_lr, device)
                
                    
                    
                models.generator.cpu()
                torch.cuda.empty_cache()
                
                conditions_b = core_b.get_conditions(batch, models_b, extras_b, is_eval=True, is_unconditional=False)
                unconditions_b = core_b.get_conditions(batch, models_b, extras_b, is_eval=True, is_unconditional=True)
                conditions_b['effnet'] = sampled_c
                unconditions_b['effnet'] = torch.zeros_like(sampled_c)
                print('STAGE B + A DECODING***************************')
                
                with torch.cuda.amp.autocast(dtype=dtype):
                        sampled = decode_b(conditions_b, unconditions_b, models_b, stage_b_latent_shape, extras_b, device, stage_a_tiled=args.stage_a_tiled)
                
                torch.cuda.empty_cache()
                imgs = show_images(sampled)
                #for idx, img in enumerate(imgs):
                    #print(os.path.join(save_dir, args.prompt[:20]+'_' + str(cnt).zfill(5) + '.jpg'), idx)
                    #img.save(os.path.join(save_dir, args.prompt[:20]+'_' + str(cnt).zfill(5) + '.jpg'))
                    
    return imgs[0]           
    #print('finished! Results ')


with gr.Blocks() as demo:
        with gr.Column():
            with gr.Row():
                with gr.Column():
                    height = gr.Slider(value=2304, step=32, minimum=1536, maximum=4096, label='Height')
                    width = gr.Slider(value=4096, step=32, minimum=1536, maximum=5120, label='Width')
                    seed = gr.Number(value=123, step=1, label='Random Seed')
                    prompt = gr.Textbox(value='', max_lines=4, label='Text Prompt')
                    cfg = gr.Slider(value=4, step=0.1, minimum=3, maximum=10, label='CFG')
                    timesteps = gr.Slider(value=20, step=1, minimum=10, maximum=50, label='Timesteps')
                    stage_a_tiled = gr.Checkbox(value=False, label='Stage_a_tiled')
                    with gr.Row():
                       clear_button = gr.Button("Clear!")
                    polish_button = gr.Button("Submit!") 
                with gr.Column():
                    output_img = gr.Image(label='Output Image', sources=None)
        with gr.Column():
            prompt2 = gr.Textbox(
                value='''
                1. a happy cat
                2. a happy girl
                ''', label='Text prompt examples'
            )
        
        polish_button.click(get_image, inputs=[height, width, seed, prompt, cfg, timesteps, stage_a_tiled], outputs=output_img)           
        polish_button.click(clear_image, inputs=[], outputs=output_img)
        
if __name__ == "__main__":
   
    args = parse_args()
    device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
    
    config_file = args.config_c
    with open(config_file, "r", encoding="utf-8") as file:
        loaded_config = yaml.safe_load(file)
    
    core = WurstCoreC(config_dict=loaded_config, device=device, training=False)
    
    # SETUP STAGE B
    config_file_b = args.config_b
    with open(config_file_b, "r", encoding="utf-8") as file:
        config_file_b = yaml.safe_load(file)
        
    core_b = WurstCoreB(config_dict=config_file_b, device=device, training=False)
    
    extras = core.setup_extras_pre()
    models = core.setup_models(extras)
    models.generator.eval().requires_grad_(False)
    print("STAGE C READY")
    
    extras_b = core_b.setup_extras_pre()
    models_b = core_b.setup_models(extras_b, skip_clip=True)
    models_b = WurstCoreB.Models(
       **{**models_b.to_dict(), 'tokenizer': models.tokenizer, 'text_model': models.text_model}
    )
    models_b.generator.bfloat16().eval().requires_grad_(False)
    print("STAGE B READY")
    
    pretrained_path = args.pretrained_path    
    sdd = torch.load(pretrained_path, map_location='cpu')
    collect_sd = {}
    for k, v in sdd.items():
        collect_sd[k[7:]] = v
    
    models.train_norm.load_state_dict(collect_sd)
    models.generator.eval()
    models.train_norm.eval()
    
    
    demo.launch(
            debug=True, share=True, 
            #server_name='10.160.211.26', server_port=7867
            
        )