stable_diffusion_experiment / Era_s20_updt.py
nkanungo's picture
Update Era_s20_updt.py
752953f verified
# import transformers as t
# assert t.__version__=='4.25.1', "Transformers version should be as specified"
#
import torch
from diffusers import AutoencoderKL, LMSDiscreteScheduler, UNet2DConditionModel
#from huggingface_hub import notebook_login
# For video display:
from IPython.display import HTML
from matplotlib import pyplot as plt
from pathlib import Path
from PIL import Image
from torch import autocast
from torchvision import transforms as tfms
from tqdm.auto import tqdm
from transformers import CLIPTextModel, CLIPTokenizer, logging
import os
import io
#import base64
import torch.nn.functional as F
#from pytorch_grad_cam.utils.image import show_cam_on_image
torch.manual_seed(1)
#if not (Path.home()/'.cache/huggingface'/'token').exists(): notebook_login()
# Supress some unnecessary warnings when loading the CLIPTextModel
logging.set_verbosity_error()
# Set device
torch_device = "cuda" if torch.cuda.is_available() else "mps" if torch.backends.mps.is_available() else "cpu"
if "mps" == torch_device: os.environ['PYTORCH_ENABLE_MPS_FALLBACK'] = "1"
import sys,gc,traceback
import fastcore.all as fc
# %% ../nbs/11_initializing.ipynb 11
def clean_ipython_hist():
# Code in this function mainly copied from IPython source
if not 'get_ipython' in globals(): return
ip = get_ipython()
user_ns = ip.user_ns
ip.displayhook.flush()
pc = ip.displayhook.prompt_count + 1
for n in range(1, pc): user_ns.pop('_i'+repr(n),None)
user_ns.update(dict(_i='',_ii='',_iii=''))
hm = ip.history_manager
hm.input_hist_parsed[:] = [''] * pc
hm.input_hist_raw[:] = [''] * pc
hm._i = hm._ii = hm._iii = hm._i00 = ''
# %% ../nbs/11_initializing.ipynb 12
def clean_tb():
# h/t Piotr Czapla
if hasattr(sys, 'last_traceback'):
traceback.clear_frames(sys.last_traceback)
delattr(sys, 'last_traceback')
if hasattr(sys, 'last_type'): delattr(sys, 'last_type')
if hasattr(sys, 'last_value'): delattr(sys, 'last_value')
# %% ../nbs/11_initializing.ipynb 13
def clean_mem():
clean_tb()
clean_ipython_hist()
gc.collect()
torch.cuda.empty_cache()
clean_mem()
# Load the autoencoder model which will be used to decode the latents into image space.
vae = AutoencoderKL.from_pretrained("CompVis/stable-diffusion-v1-4", subfolder="vae")
# Load the tokenizer and text encoder to tokenize and encode the text.
tokenizer = CLIPTokenizer.from_pretrained("openai/clip-vit-large-patch14")
text_encoder = CLIPTextModel.from_pretrained("openai/clip-vit-large-patch14")
# The UNet model for generating the latents.
unet = UNet2DConditionModel.from_pretrained("CompVis/stable-diffusion-v1-4", subfolder="unet")
# The noise scheduler
scheduler = LMSDiscreteScheduler(beta_start=0.00085, beta_end=0.012, beta_schedule="scaled_linear", num_train_timesteps=1000)
# To the GPU we go!
vae = vae.to(torch_device)
text_encoder = text_encoder.to(torch_device)
unet = unet.to(torch_device);
embeds_folder = Path('./paintings_embed')
file_names = [path.name for path in embeds_folder.glob('*') if path.is_file()]
print(file_names)
style_names = [list(torch.load(embeds_folder/file).keys())[0] for file in file_names]
style_names
num_added_tokens = tokenizer.add_tokens(style_names)
added_tokens = list(map(tokenizer.added_tokens_encoder.get,style_names))
added_tokens,style_names
text_encoder.resize_token_embeddings(len(tokenizer))
text_encoder.text_model.embeddings.token_embedding
style_dict = {}
list_styles = [torch.load(embeds_folder/file) for file in file_names]
# for k,v in list_styles[0].items():
# print(k,v.shape)
style_dict = {style:embedding for each_style in list_styles for style,embedding in each_style.items()}
list(style_dict)
for token,style in zip(added_tokens,style_names):
text_encoder.text_model.embeddings.token_embedding.weight.data[token] = style_dict[style]
# #checking if we added the embeddings properly to text_encoder
# ft_dict = torch.load(embeds_folder/'fairy-tale-painting_embeds.bin')
# list(ft_dict.keys())[0]
# ft_dict['<fairy-tale-painting-style>'][:10]
clean_mem()
# text_encoder.get_input_embeddings()(torch.tensor(49408, device=torch_device))[:10]
# Prep Scheduler
def set_timesteps(scheduler, num_inference_steps):
scheduler.set_timesteps(num_inference_steps)
scheduler.timesteps = scheduler.timesteps.to(torch.float32) # minor fix to ensure MPS compatibility, fixed in diffusers PR 3925
def pil_to_latent(input_im):
# Single image -> single latent in a batch (so size 1, 4, 64, 64)
with torch.no_grad():
latent = vae.encode(tfms.ToTensor()(input_im).unsqueeze(0).to(torch_device)*2-1) # Note scaling
return 0.18215 * latent.latent_dist.sample()
def latents_to_pil(latents):
# bath of latents -> list of images
latents = (1 / 0.18215) * latents
with torch.no_grad():
image = vae.decode(latents).sample
image = (image / 2 + 0.5).clamp(0, 1)
image = image.detach().cpu().permute(0, 2, 3, 1).numpy()
images = (image * 255).round().astype("uint8")
pil_images = [Image.fromarray(image) for image in images]
return pil_images
# Access the embedding layer
token_emb_layer = text_encoder.text_model.embeddings.token_embedding
token_emb_layer # Vocab size 49408, emb_dim 768
pos_emb_layer = text_encoder.text_model.embeddings.position_embedding
position_ids = text_encoder.text_model.embeddings.position_ids[:, :77]
position_embeddings = pos_emb_layer(position_ids)
print(position_embeddings.shape)
def get_output_embeds(input_embeddings):
# CLIP's text model uses causal mask, so we prepare it here:
bsz, seq_len = input_embeddings.shape[:2]
causal_attention_mask = text_encoder.text_model._build_causal_attention_mask(bsz, seq_len, dtype=input_embeddings.dtype)
# Getting the output embeddings involves calling the model with passing output_hidden_states=True
# so that it doesn't just return the pooled final predictions:
encoder_outputs = text_encoder.text_model.encoder(
inputs_embeds=input_embeddings,
attention_mask=None, # We aren't using an attention mask so that can be None
causal_attention_mask=causal_attention_mask.to(torch_device),
output_attentions=None,
output_hidden_states=True, # We want the output embs not the final output
return_dict=None,
)
# We're interested in the output hidden state only
output = encoder_outputs[0]
# There is a final layer norm we need to pass these through
output = text_encoder.text_model.final_layer_norm(output)
# And now they're ready!
return output
#Generating an image with these modified embeddings
def generate_with_embs_custom(text_embeddings,seed):
height = 512 # default height of Stable Diffusion
width = 512 # default width of Stable Diffusion
num_inference_steps = 30 # Number of denoising steps
guidance_scale = 7.5 # Scale for classifier-free guidance
generator = torch.manual_seed(seed) # Seed generator to create the inital latent noise
batch_size = 1
max_length = text_embeddings.shape[1]
uncond_input = tokenizer(
[""] * batch_size, padding="max_length", max_length=max_length, return_tensors="pt"
)
with torch.no_grad():
uncond_embeddings = text_encoder(uncond_input.input_ids.to(torch_device))[0]
text_embeddings = torch.cat([uncond_embeddings, text_embeddings])
# Prep Scheduler
set_timesteps(scheduler, num_inference_steps)
# Prep latents
latents = torch.randn(
(batch_size, unet.in_channels, height // 8, width // 8),
generator=generator,
)
latents = latents.to(torch_device)
latents = latents * scheduler.init_noise_sigma
# Loop
for i, t in tqdm(enumerate(scheduler.timesteps), total=len(scheduler.timesteps)):
# expand the latents if we are doing classifier-free guidance to avoid doing two forward passes.
latent_model_input = torch.cat([latents] * 2)
sigma = scheduler.sigmas[i]
latent_model_input = scheduler.scale_model_input(latent_model_input, t)
# predict the noise residual
with torch.no_grad():
noise_pred = unet(latent_model_input, t, encoder_hidden_states=text_embeddings)["sample"]
# perform guidance
noise_pred_uncond, noise_pred_text = noise_pred.chunk(2)
noise_pred = noise_pred_uncond + guidance_scale * (noise_pred_text - noise_pred_uncond)
# compute the previous noisy sample x_t -> x_t-1
latents = scheduler.step(noise_pred, t, latents).prev_sample
return latents_to_pil(latents)[0]
# ref_image = Image.open('C:/Users/shivs/Downloads/lg.jpg').resize((512,512))
# ref_latent = pil_to_latent(ref_image)
## Guidance through Custom Loss Function
def custom_loss(latent,ref_latent):
error = F.mse_loss(0.5*latent,0.8*ref_latent)
return error
class Styles_paintings():
def __init__(self,prompt):
self.output_styles = []
self.prompt = prompt
self.style_names = list(style_dict)
self.seeds = [1024+i for i in range(len(self.style_names))]
def generate_styles(self):
#print('The Values are ', list(style_dict)[0])
for seed,style_name in zip(self.seeds,self.style_names):
# Tokenize
prompt = f'{self.prompt} in the style of {style_name}'
text_input = tokenizer(prompt, padding="max_length", max_length=tokenizer.model_max_length, truncation=True, return_tensors="pt")
input_ids = text_input.input_ids.to(torch_device)
# Get token embeddings
token_embeddings = token_emb_layer(input_ids)
# Combine with pos embs
input_embeddings = token_embeddings + position_embeddings
# Feed through to get final output embs
modified_output_embeddings = get_output_embeds(input_embeddings)
# And generate an image with this:
self.output_styles.append(generate_with_embs_custom(modified_output_embeddings,seed))
def generate_styles_with_custom_loss(self, image):
height = 512 # default height of Stable Diffusion
width = 512 # default width of Stable Diffusion
num_inference_steps = 10 #@param # Number of denoising steps
guidance_scale = 8 #@param # Scale for classifier-free guidance
batch_size = 1
custom_loss_scale = 200 #@param
print('image shape there is',image.size)
self.output_styles_with_custom_loss = []
#ref_image = Image.open('C:/Users/shivs/Downloads/ig.jpg').resize((512,512))
ref_latent = pil_to_latent(image)
for seed,style_name in zip(self.seeds,self.style_names):
# Tokenize
prompt = f'{self.prompt} in the style of {style_name}'
generator = torch.manual_seed(seed) # Seed generator to create the inital latent noise
print(f' the prompt is : {prompt} with seed value :{seed}')
# Prep text
text_input = tokenizer([prompt], padding="max_length", max_length=tokenizer.model_max_length, truncation=True, return_tensors="pt")
with torch.no_grad():
text_embeddings = text_encoder(text_input.input_ids.to(torch_device))[0]
# And the uncond. input as before:
max_length = text_input.input_ids.shape[-1]
uncond_input = tokenizer(
[""] * batch_size, padding="max_length", max_length=max_length, return_tensors="pt"
)
with torch.no_grad():
uncond_embeddings = text_encoder(uncond_input.input_ids.to(torch_device))[0]
text_embeddings = torch.cat([uncond_embeddings, text_embeddings])
# Prep Scheduler
set_timesteps(scheduler, num_inference_steps)
# Prep latents
latents = torch.randn(
(batch_size, unet.in_channels, height // 8, width // 8),
generator=generator,)
latents = latents.to(torch_device)
latents = latents * scheduler.init_noise_sigma
# Loop
for i, t in tqdm(enumerate(scheduler.timesteps), total=len(scheduler.timesteps)):
# expand the latents if we are doing classifier-free guidance to avoid doing two forward passes.
latent_model_input = torch.cat([latents] * 2)
sigma = scheduler.sigmas[i]
latent_model_input = scheduler.scale_model_input(latent_model_input, t)
# predict the noise residual
with torch.no_grad():
noise_pred = unet(latent_model_input, t, encoder_hidden_states=text_embeddings)["sample"]
# perform CFG
noise_pred_uncond, noise_pred_text = noise_pred.chunk(2)
noise_pred = noise_pred_uncond + guidance_scale * (noise_pred_text - noise_pred_uncond)
#### ADDITIONAL GUIDANCE ###
if i%5 == 0:
# Requires grad on the latents
latents = latents.detach().requires_grad_()
# Get the predicted x0:
latents_x0 = latents - sigma * noise_pred
#latents_x0 = scheduler.step(noise_pred, t, latents).pred_original_sample
# Decode to image space
#denoised_images = vae.decode((1 / 0.18215) * latents_x0).sample / 2 + 0.5 # range (0, 1)
# Calculate loss
loss = custom_loss(latents_x0,ref_latent) * custom_loss_scale
#loss = blue_loss(denoised_images) * blue_loss_scale
# Occasionally print it out
if i%10==0:
print(i, 'loss:', loss.item())
# Get gradient
cond_grad = torch.autograd.grad(loss, latents)[0]
# Modify the latents based on this gradient
latents = latents.detach() - cond_grad * sigma**2
# Now step with scheduler
latents = scheduler.step(noise_pred, t, latents).prev_sample
self.output_styles_with_custom_loss.append(latents_to_pil(latents)[0])
def generate_final_image(im1,in_prompt="an oil painting of an baby girl with flowers in a park"):
paintings = Styles_paintings(in_prompt)
paintings.generate_styles()
r_image = im1.resize((512,512))
print('image shape is',r_image.size)
paintings.generate_styles_with_custom_loss(r_image)
print(len(paintings.output_styles))
#return [paintings.output_styles[0]], [paintings.output_styles[1]],[paintings.output_styles[2]],[paintings.output_styles[3]],[paintings.output_styles[4]], [paintings.output_styles_with_custom_loss[0]],[paintings.output_styles_with_custom_loss[1]],[paintings.output_styles_with_custom_loss[2]],[paintings.output_styles_with_custom_loss[3]],[paintings.output_styles_with_custom_loss[4]]
return [paintings.output_styles[0]], [paintings.output_styles[1]],[paintings.output_styles_with_custom_loss[0]],[paintings.output_styles_with_custom_loss[1]]