Skip to content

Commit

Permalink
fix(api): update SD upscale pipeline
Browse files Browse the repository at this point in the history
  • Loading branch information
@ssube
ssube committed Feb 5, 2023
1 parent a2a0028 commit 49b3aa6
Showing 1 changed file with 52 additions and 80 deletions.
132 changes: 52 additions & 80 deletions api/onnx_web/diffusion/pipeline_onnx_stable_diffusion_upscale.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -3,14 +3,20 @@

import numpy as np
import torch

import PIL

from diffusers import DDPMScheduler, OnnxRuntimeModel, StableDiffusionUpscalePipeline
from diffusers.pipeline_utils import ImagePipelineOutput
from PIL import Image


logger = getLogger(__name__)

num_channels_latents = 4 # self.vae.config.latent_channels
unet_in_channels = 7 # self.unet.config.in_channels
# TODO: make this dynamic, from self.vae.config.latent_channels
num_channels_latents = 4

# TODO: make this dynamic, from self.unet.config.in_channels
unet_in_channels = 7

###
# This is based on a combination of the ONNX img2img pipeline and the PyTorch upscale pipeline:
Expand All @@ -22,10 +28,10 @@
def preprocess(image):
if isinstance(image, torch.Tensor):
return image
elif isinstance(image, Image.Image):
elif isinstance(image, PIL.Image.Image):
image = [image]

if isinstance(image[0], Image.Image):
if isinstance(image[0], PIL.Image.Image):
w, h = image[0].size
w, h = map(lambda x: x - x % 64, (w, h)) # resize to integer multiple of 32

Expand All @@ -52,20 +58,12 @@ def __init__(
scheduler: Any,
max_noise_level: int = 350,
):
super().__init__(
vae,
text_encoder,
tokenizer,
unet,
low_res_scheduler,
scheduler,
max_noise_level,
)
super().__init__(vae, text_encoder, tokenizer, unet, low_res_scheduler, scheduler, max_noise_level)

def __call__(
self,
prompt: Union[str, List[str]],
image: Union[torch.FloatTensor, Image.Image, List[Image.Image]],
image: Union[torch.FloatTensor, PIL.Image.Image, List[PIL.Image.Image]],
num_inference_steps: int = 75,
guidance_scale: float = 9.0,
noise_level: int = 20,
Expand All @@ -92,12 +90,9 @@ def __call__(

# 3. Encode input prompt
text_embeddings = self._encode_prompt(
prompt,
device,
num_images_per_prompt,
do_classifier_free_guidance,
negative_prompt,
prompt, device, num_images_per_prompt, do_classifier_free_guidance, negative_prompt
)
text_embeddings_dtype = torch.float32 # TODO: convert text_embeddings.dtype to torch dtype

# 4. Preprocess image
image = preprocess(image)
Expand All @@ -108,12 +103,8 @@ def __call__(
timesteps = self.scheduler.timesteps

# 5. Add noise to image
text_embeddings_dtype = torch.float32

noise_level = torch.tensor([noise_level], dtype=torch.long, device=device)
noise = torch.randn(
image.shape, generator=generator, device=device, dtype=text_embeddings_dtype
)
noise = torch.randn(image.shape, generator=generator, device=device, dtype=text_embeddings_dtype)
image = self.low_res_scheduler.add_noise(image, noise, noise_level)

batch_multiplier = 2 if do_classifier_free_guidance else 1
Expand All @@ -137,11 +128,11 @@ def __call__(
num_channels_image = image.shape[1]
if num_channels_latents + num_channels_image != unet_in_channels:
raise ValueError(
"Incorrect configuration settings! The config of `pipeline.unet`"
f" expects {unet_in_channels} but received `num_channels_latents`:"
f" {num_channels_latents} + `num_channels_image`: {num_channels_image} "
f" = {num_channels_latents+num_channels_image}. Please verify the"
" config of `pipeline.unet` or your `image` input."
"Incorrect configuration settings! The config of `pipeline.unet` expects"
f" {unet_in_channels} but received `num_channels_latents`: {num_channels_latents} +"
f" `num_channels_image`: {num_channels_image} "
f" = {num_channels_latents+num_channels_image}. Please verify the config of"
" `pipeline.unet` or your `image` input."
)

# 8. Prepare extra step kwargs. TODO: Logic should ideally just be moved out of the pipeline
Expand All @@ -152,16 +143,10 @@ def __call__(
with self.progress_bar(total=num_inference_steps) as progress_bar:
for i, t in enumerate(timesteps):
# expand the latents if we are doing classifier free guidance
latent_model_input = (
np.concatenate([latents] * 2)
if do_classifier_free_guidance
else latents
)
latent_model_input = np.concatenate([latents] * 2) if do_classifier_free_guidance else latents

# concat latents, mask, masked_image_latents in the channel dimension
latent_model_input = self.scheduler.scale_model_input(
latent_model_input, t
)
latent_model_input = self.scheduler.scale_model_input(latent_model_input, t)
latent_model_input = np.concatenate([latent_model_input, image], axis=1)

# timestep to tensor
Expand All @@ -178,19 +163,13 @@ def __call__(
# perform guidance
if do_classifier_free_guidance:
noise_pred_uncond, noise_pred_text = np.split(noise_pred, 2)
noise_pred = noise_pred_uncond + guidance_scale * (
noise_pred_text - noise_pred_uncond
)
noise_pred = noise_pred_uncond + guidance_scale * (noise_pred_text - noise_pred_uncond)

# compute the previous noisy sample x_t -> x_t-1
latents = self.scheduler.step(
noise_pred, t, latents, **extra_step_kwargs
).prev_sample
latents = self.scheduler.step(noise_pred, t, latents, **extra_step_kwargs).prev_sample

# call the callback, if provided
if i == len(timesteps) - 1 or (
(i + 1) > num_warmup_steps and (i + 1) % self.scheduler.order == 0
):
if i == len(timesteps) - 1 or ((i + 1) > num_warmup_steps and (i + 1) % self.scheduler.order == 0):
progress_bar.update()
if callback is not None and i % callback_steps == 0:
callback(i, t, latents)
Expand All @@ -214,14 +193,7 @@ def decode_latents(self, latents):
image = image.transpose((0, 2, 3, 1))
return image

def _encode_prompt(
self,
prompt,
device,
num_images_per_prompt,
do_classifier_free_guidance,
negative_prompt,
):
def _encode_prompt(self, prompt, device, num_images_per_prompt, do_classifier_free_guidance, negative_prompt):
batch_size = len(prompt) if isinstance(prompt, list) else 1

text_inputs = self.tokenizer(
Expand All @@ -232,34 +204,31 @@ def _encode_prompt(
return_tensors="pt",
)
text_input_ids = text_inputs.input_ids
untruncated_ids = self.tokenizer(
prompt, padding="longest", return_tensors="pt"
).input_ids

if untruncated_ids.shape[-1] >= text_input_ids.shape[-1] and not torch.equal(
text_input_ids, untruncated_ids
):
removed_text = self.tokenizer.batch_decode(
untruncated_ids[:, self.tokenizer.model_max_length - 1 : -1]
)
untruncated_ids = self.tokenizer(prompt, padding="longest", return_tensors="pt").input_ids

if untruncated_ids.shape[-1] >= text_input_ids.shape[-1] and not torch.equal(text_input_ids, untruncated_ids):
removed_text = self.tokenizer.batch_decode(untruncated_ids[:, self.tokenizer.model_max_length - 1 : -1])
logger.warning(
"The following part of your input was truncated because CLIP can only"
f" handle sequences up to {self.tokenizer.model_max_length} tokens:"
f" {removed_text}"
"The following part of your input was truncated because CLIP can only handle sequences up to"
f" {self.tokenizer.model_max_length} tokens: {removed_text}"
)

# if hasattr(text_inputs, "attention_mask"):
# attention_mask = text_inputs.attention_mask.to(device)
# else:
# attention_mask = None

# no positional arguments to text_encoder
text_embeddings = self.text_encoder(
input_ids=text_input_ids.int().to(device),
# TODO: is this needed?
# attention_mask=attention_mask,
)
text_embeddings = text_embeddings[0]

bs_embed, seq_len, _ = text_embeddings.shape
# duplicate text embeddings for each generation per prompt, using mps friendly method
text_embeddings = text_embeddings.repeat(1, num_images_per_prompt)
# TODO: is this needed?
# text_embeddings = text_embeddings.view(bs_embed * num_images_per_prompt, seq_len, -1)
text_embeddings = text_embeddings.reshape(bs_embed * num_images_per_prompt, seq_len, -1)

# get unconditional embeddings for classifier free guidance
if do_classifier_free_guidance:
Expand All @@ -268,17 +237,16 @@ def _encode_prompt(
uncond_tokens = [""] * batch_size
elif type(prompt) is not type(negative_prompt):
raise TypeError(
"`negative_prompt` should be the same type to `prompt`, but got"
f" {type(negative_prompt)} != {type(prompt)}."
f"`negative_prompt` should be the same type to `prompt`, but got {type(negative_prompt)} !="
f" {type(prompt)}."
)
elif isinstance(negative_prompt, str):
uncond_tokens = [negative_prompt]
elif batch_size != len(negative_prompt):
raise ValueError(
f"`negative_prompt`: {negative_prompt} has batch size"
f" {len(negative_prompt)}, but `prompt`: {prompt} has batch size"
f" {batch_size}. Please make sure that passed `negative_prompt`"
" matches the batch size of `prompt`."
f"`negative_prompt`: {negative_prompt} has batch size {len(negative_prompt)}, but `prompt`:"
f" {prompt} has batch size {batch_size}. Please make sure that passed `negative_prompt` matches"
" the batch size of `prompt`."
)
else:
uncond_tokens = negative_prompt
Expand All @@ -292,17 +260,21 @@ def _encode_prompt(
return_tensors="pt",
)

# if hasattr(uncond_input, "attention_mask"):
# attention_mask = uncond_input.attention_mask.to(device)
# else:
# attention_mask = None

uncond_embeddings = self.text_encoder(
input_ids=uncond_input.input_ids.int().to(device),
# TODO: is this needed?
# attention_mask=attention_mask,
)
uncond_embeddings = uncond_embeddings[0]

seq_len = uncond_embeddings.shape[1]
# duplicate unconditional embeddings for each generation per prompt, using mps friendly method
uncond_embeddings = uncond_embeddings.repeat(1, num_images_per_prompt)
# TODO: is this needed?
# uncond_embeddings = uncond_embeddings.view(batch_size * num_images_per_prompt, seq_len, -1)
uncond_embeddings = uncond_embeddings.reshape(batch_size * num_images_per_prompt, seq_len, -1)

# For classifier free guidance, we need to do two forward passes.
# Here we concatenate the unconditional and text embeddings into a single batch
Expand Down

0 comments on commit 49b3aa6

Please sign in to comment.