import gradio as gr
import torch
from finetuning import FineTunedModel
from StableDiffuser import StableDiffuser
from train import train
import os
model_map = {'Van Gogh' : 'models/vangogh.pt',
'Pablo Picasso': 'models/pablopicasso.pt',
'Car' : 'models/car.pt',
'Garbage Truck': 'models/garbagetruck.pt',
'French Horn': 'models/frenchhorn.pt',
'Kilian Eng' : 'models/kilianeng.pt',
'Thomas Kinkade' : 'models/thomaskinkade.pt',
'Tyler Edlin' : 'models/tyleredlin.pt',
'Kelly McKernan': 'models/kellymckernan.pt',
'Rembrandt': 'models/rembrandt.pt' }
ORIGINAL_SPACE_ID = 'baulab/Erasing-Concepts-In-Diffusion'
SPACE_ID = os.getenv('SPACE_ID')
SHARED_UI_WARNING = f'''## Attention - Training using the ESD-u method does not work in this shared UI. You can either duplicate and use it with a gpu with at least 40GB, or clone this repository to run on your own machine.
'''
class Demo:
def __init__(self) -> None:
self.training = False
self.generating = False
self.diffuser = StableDiffuser(scheduler='DDIM').to('cuda').eval().half()
with gr.Blocks() as demo:
self.layout()
demo.queue(concurrency_count=5).launch()
def layout(self):
with gr.Row():
if SPACE_ID == ORIGINAL_SPACE_ID:
self.warning = gr.Markdown(SHARED_UI_WARNING)
with gr.Row():
with gr.Tab("Test") as inference_column:
with gr.Row():
self.explain_infr = gr.Markdown(interactive=False,
value='This is a demo of [Erasing Concepts from Stable Diffusion](https://erasing.baulab.info/). To try out a model where a concept has been erased, select a model and enter any prompt. For example, if you select the model "Van Gogh" you can generate images for the prompt "A portrait in the style of Van Gogh" and compare the erased and unerased models. We have also provided several other pre-fine-tuned models with artistic styles and objects erased (Check out the "ESD Model" drop-down). You can also train and run your own custom models. Check out the "train" section for custom erasure of concepts.')
with gr.Row():
with gr.Column(scale=1):
self.prompt_input_infr = gr.Text(
placeholder="Enter prompt...",
label="Prompt",
info="Prompt to generate"
)
with gr.Row():
self.model_dropdown = gr.Dropdown(
label="ESD Model",
choices= list(model_map.keys()),
value='Van Gogh',
interactive=True
)
self.seed_infr = gr.Number(
label="Seed",
value=42
)
with gr.Column(scale=2):
self.infr_button = gr.Button(
value="Generate",
interactive=True
)
with gr.Row():
self.image_new = gr.Image(
label="ESD",
interactive=False
)
self.image_orig = gr.Image(
label="SD",
interactive=False
)
with gr.Tab("Train") as training_column:
with gr.Row():
self.explain_train= gr.Markdown(interactive=False,
value='In this part you can erase any concept from Stable Diffusion. Enter a prompt for the concept or style you want to erase, and select ESD-x if you want to focus erasure on prompts that mention the concept explicitly. [NOTE: ESD-u is currently unavailable in this space. But you can duplicate the space and run it on GPU with VRAM >40GB for enabling ESD-u]. With default settings, it takes about 15 minutes to fine-tune the model; then you can try inference above or download the weights. The training code used here is slightly different than the code tested in the original paper. Code and details are at [github link](https://github.com/rohitgandikota/erasing).')
with gr.Row():
with gr.Column(scale=3):
self.prompt_input = gr.Text(
placeholder="Enter prompt...",
label="Prompt to Erase",
info="Prompt corresponding to concept to erase"
)
choices = ['ESD-x']
if torch.cuda.get_device_properties(0).total_memory * 1e-9 >= 40:
choices.append('ESD-u')
self.train_method_input = gr.Dropdown(
choices=choices,
value='ESD-x',
label='Train Method',
info='Method of training'
)
self.neg_guidance_input = gr.Number(
value=1,
label="Negative Guidance",
info='Guidance of negative training used to train'
)
self.iterations_input = gr.Number(
value=150,
precision=0,
label="Iterations",
info='iterations used to train'
)
self.lr_input = gr.Number(
value=1e-5,
label="Learning Rate",
info='Learning rate used to train'
)
with gr.Column(scale=1):
self.train_status = gr.Button(value='', variant='primary', label='Status', interactive=False)
self.train_button = gr.Button(
value="Train",
)
self.download = gr.Files()
self.infr_button.click(self.inference, inputs = [
self.prompt_input_infr,
self.seed_infr,
self.model_dropdown
],
outputs=[
self.image_new,
self.image_orig
]
)
self.train_button.click(self.train, inputs = [
self.prompt_input,
self.train_method_input,
self.neg_guidance_input,
self.iterations_input,
self.lr_input
],
outputs=[self.train_button, self.train_status, self.download, self.model_dropdown]
)
def train(self, prompt, train_method, neg_guidance, iterations, lr, pbar = gr.Progress(track_tqdm=True)):
if self.training:
return [gr.update(interactive=True, value='Train'), gr.update(value='Someone else is training... Try again soon'), None, gr.update()]
if train_method == 'ESD-x':
modules = ".*attn2$"
frozen = []
elif train_method == 'ESD-u':
modules = "unet$"
frozen = [".*attn2$", "unet.time_embedding$", "unet.conv_out$"]
elif train_method == 'ESD-self':
modules = ".*attn1$"
frozen = []
randn = torch.randint(1, 10000000, (1,)).item()
save_path = f"models/{randn}_{prompt.lower().replace(' ', '')}.pt"
self.training = True
train(prompt, modules, frozen, iterations, neg_guidance, lr, save_path)
self.training = False
torch.cuda.empty_cache()
model_map['Custom'] = save_path
return [gr.update(interactive=True, value='Train'), gr.update(value='Done Training! \n Try your custom model in the "Test" tab'), save_path, gr.Dropdown.update(choices=list(model_map.keys()), value='Custom')]
def inference(self, prompt, seed, model_name, pbar = gr.Progress(track_tqdm=True)):
seed = seed or 42
generator = torch.manual_seed(seed)
model_path = model_map[model_name]
checkpoint = torch.load(model_path)
finetuner = FineTunedModel.from_checkpoint(self.diffuser, checkpoint).eval().half()
torch.cuda.empty_cache()
images = self.diffuser(
prompt,
n_steps=50,
generator=generator
)
orig_image = images[0][0]
torch.cuda.empty_cache()
generator = torch.manual_seed(seed)
with finetuner:
images = self.diffuser(
prompt,
n_steps=50,
generator=generator
)
edited_image = images[0][0]
del finetuner
torch.cuda.empty_cache()
return edited_image, orig_image
demo = Demo()