Add image2image example script.

README.md

@@ -23,7 +23,8 @@ More precisely, 🤗 Diffusers offers:
 - State-of-the-art diffusion pipelines that can be run in inference with just a couple of lines of code (see [src/diffusers/pipelines](https://github.com/huggingface/diffusers/tree/main/src/diffusers/pipelines)).
 - Various noise schedulers that can be used interchangeably for the prefered speed vs. quality trade-off in inference (see [src/diffusers/schedulers](https://github.com/huggingface/diffusers/tree/main/src/diffusers/schedulers)).
 - Multiple types of models, such as UNet, can be used as building blocks in an end-to-end diffusion system (see [src/diffusers/models](https://github.com/huggingface/diffusers/tree/main/src/diffusers/models)).
-- Training examples to show how to train the most popular diffusion models (see [examples](https://github.com/huggingface/diffusers/tree/main/examples)).
+- Training examples to show how to train the most popular diffusion models (see [examples/training](https://github.com/huggingface/diffusers/tree/main/examples/training)).
+- Inference examples to show how to create custom pipelines for advanced tasks such as image2image, in-painting (see [examples/inference](https://github.com/huggingface/diffusers/tree/main/examples/inference))
 
 ## Quickstart
 

examples/inference/image_to_image.py

@@ -0,0 +1,161 @@
+import inspect
+from typing import List, Optional, Union
+
+import numpy as np
+import torch
+
+import PIL
+from diffusers import AutoencoderKL, DDIMScheduler, DiffusionPipeline, PNDMScheduler, UNet2DConditionModel
+from diffusers.pipelines.stable_diffusion import StableDiffusionSafetyChecker
+from tqdm.auto import tqdm
+from transformers import CLIPFeatureExtractor, CLIPTextModel, CLIPTokenizer
+
+
+def preprocess(image):
+    w, h = image.size
+    w, h = map(lambda x: x - x % 32, (w, h))  # resize to integer multiple of 32
+    image = image.resize((w, h), resample=PIL.Image.LANCZOS)
+    image = np.array(image).astype(np.float32) / 255.0
+    image = image[None].transpose(0, 3, 1, 2)
+    image = torch.from_numpy(image)
+    return 2.0 * image - 1.0
+
+
+class StableDiffusionImg2ImgPipeline(DiffusionPipeline):
+    def __init__(
+        self,
+        vae: AutoencoderKL,
+        text_encoder: CLIPTextModel,
+        tokenizer: CLIPTokenizer,
+        unet: UNet2DConditionModel,
+        scheduler: Union[DDIMScheduler, PNDMScheduler],
+        safety_checker: StableDiffusionSafetyChecker,
+        feature_extractor: CLIPFeatureExtractor,
+    ):
+        super().__init__()
+        scheduler = scheduler.set_format("pt")
+        self.register_modules(
+            vae=vae,
+            text_encoder=text_encoder,
+            tokenizer=tokenizer,
+            unet=unet,
+            scheduler=scheduler,
+            safety_checker=safety_checker,
+            feature_extractor=feature_extractor,
+        )
+
+    @torch.no_grad()
+    def __call__(
+        self,
+        prompt: Union[str, List[str]],
+        init_image: torch.FloatTensor,
+        strength: float = 0.8,
+        num_inference_steps: Optional[int] = 50,
+        guidance_scale: Optional[float] = 7.5,
+        eta: Optional[float] = 0.0,
+        generator: Optional[torch.Generator] = None,
+        output_type: Optional[str] = "pil",
+    ):
+
+        if isinstance(prompt, str):
+            batch_size = 1
+        elif isinstance(prompt, list):
+            batch_size = len(prompt)
+        else:
+            raise ValueError(f"`prompt` has to be of type `str` or `list` but is {type(prompt)}")
+
+        # set timesteps
+        accepts_offset = "offset" in set(inspect.signature(self.scheduler.set_timesteps).parameters.keys())
+        extra_set_kwargs = {}
+        offset = 0
+        if accepts_offset:
+            offset = 1
+            extra_set_kwargs["offset"] = 1
+
+        self.scheduler.set_timesteps(num_inference_steps, **extra_set_kwargs)
+
+        # encode the init image into latents and scale the latents
+        init_latents = self.vae.encode(init_image.to(self.device)).sample()
+        init_latents = 0.18215 * init_latents
+
+        # prepare init_latents noise to latents
+        init_latents = torch.cat([init_latents] * batch_size)
+
+        # get the original timestep using init_timestep
+        init_timestep = int(num_inference_steps * strength) + offset
+        init_timestep = min(init_timestep, num_inference_steps)
+        timesteps = self.scheduler.timesteps[-init_timestep]
+        timesteps = torch.tensor([timesteps] * batch_size, dtype=torch.long, device=self.device)
+
+        # add noise to latents using the timesteps
+        noise = torch.randn(init_latents.shape, generator=generator, device=self.device)
+        init_latents = self.scheduler.add_noise(init_latents, noise, timesteps)
+
+        # get prompt text embeddings
+        text_input = self.tokenizer(
+            prompt,
+            padding="max_length",
+            max_length=self.tokenizer.model_max_length,
+            truncation=True,
+            return_tensors="pt",
+        )
+        text_embeddings = self.text_encoder(text_input.input_ids.to(self.device))[0]
+
+        # here `guidance_scale` is defined analog to the guidance weight `w` of equation (2)
+        # of the Imagen paper: https://arxiv.org/pdf/2205.11487.pdf . `guidance_scale = 1`
+        # corresponds to doing no classifier free guidance.
+        do_classifier_free_guidance = guidance_scale > 1.0
+        # get unconditional embeddings for classifier free guidance
+        if do_classifier_free_guidance:
+            max_length = text_input.input_ids.shape[-1]
+            uncond_input = self.tokenizer(
+                [""] * batch_size, padding="max_length", max_length=max_length, return_tensors="pt"
+            )
+            uncond_embeddings = self.text_encoder(uncond_input.input_ids.to(self.device))[0]
+
+            # For classifier free guidance, we need to do two forward passes.
+            # Here we concatenate the unconditional and text embeddings into a single batch
+            # to avoid doing two forward passes
+            text_embeddings = torch.cat([uncond_embeddings, text_embeddings])
+
+        # prepare extra kwargs for the scheduler step, since not all schedulers have the same signature
+        # eta (η) is only used with the DDIMScheduler, it will be ignored for other schedulers.
+        # eta corresponds to η in DDIM paper: https://arxiv.org/abs/2010.02502
+        # and should be between [0, 1]
+        accepts_eta = "eta" in set(inspect.signature(self.scheduler.step).parameters.keys())
+        extra_step_kwargs = {}
+        if accepts_eta:
+            extra_step_kwargs["eta"] = eta
+
+        latents = init_latents
+        t_start = max(num_inference_steps - init_timestep + offset, 0)
+        for i, t in tqdm(enumerate(self.scheduler.timesteps[t_start:])):
+            # expand the latents if we are doing classifier free guidance
+            latent_model_input = torch.cat([latents] * 2) if do_classifier_free_guidance else latents
+
+            # predict the noise residual
+            noise_pred = self.unet(latent_model_input, t, encoder_hidden_states=text_embeddings)["sample"]
+
+            # perform guidance
+            if do_classifier_free_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 = self.scheduler.step(noise_pred, t, latents, **extra_step_kwargs)["prev_sample"]
+
+        # scale and decode the image latents with vae
+        latents = 1 / 0.18215 * latents
+        image = self.vae.decode(latents)
+
+        image = (image / 2 + 0.5).clamp(0, 1)
+        image = image.cpu().permute(0, 2, 3, 1).numpy()
+
+        # run safety checker
+        safety_cheker_input = self.feature_extractor(self.numpy_to_pil(image), return_tensors="pt").to(self.device)
+        image, has_nsfw_concept = self.safety_checker(images=image, clip_input=safety_cheker_input.pixel_values)
+
+        if output_type == "pil":
+            image = self.numpy_to_pil(image)
+
+        return {"sample": image, "nsfw_content_detected": has_nsfw_concept}

examples/inference/readme.md

@@ -0,0 +1,50 @@
+# Inference Examples
+
+## Installing the dependencies
+
+Before running the scipts, make sure to install the library's dependencies:
+
+```bash
+pip install diffusers transformers ftfy
+```
+
+## Image-to-Image text-guided generation with Stable Diffusion
+
+The `image_to_image.py` script implements `StableDiffusionImg2ImgPipeline`. It lets you pass a text prompt and an initial image to condition the generation of new images. This example also showcases how you can write custom diffusion pipelines using `diffusers`!
+
+### How to use it
+
+
+```python
+from torch import autocast
+import requests
+from PIL import Image
+from io import BytesIO
+
+from image_to_image import StableDiffusionImg2ImgPipeline, preprocess
+
+# load the pipeline
+device = "cuda"
+pipe = StableDiffusionImg2ImgPipeline.from_pretrained(
+    "CompVis/stable-diffusion-v1-4",
+    revision="fp16", 
+    torch_dtype=torch.float16,
+    use_auth_token=True
+).to(device)
+
+# let's download an initial image
+url = "https://rg.gosu.cc/CompVis/stable-diffusion/main/assets/stable-samples/img2img/sketch-mountains-input.jpg"
+
+response = requests.get(url)
+init_image = Image.open(BytesIO(response.content)).convert("RGB")
+init_image = init_image.resize((768, 512))
+init_image = preprocess(init_image)
+
+prompt = "A fantasy landscape, trending on artstation"
+
+with autocast("cuda"):
+    images = pipe(prompt=prompt, init_image=init_image, strength=0.75, guidance_scale=7.5)["sample"]
+
+images[0].save("fantasy_landscape.png")
+```
+You can also run this example on colab [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/patil-suraj/Notebooks/blob/master/image_2_image_using_diffusers.ipynb)