Skip to content

GitLab

"vscode:/vscode.git/clone" did not exist on "41276b6c3704cd3ce93d2b95c0a1ae1937372b53"
Unverified Commit 75ea54a1 authored by Steven Liu's avatar Steven Liu Committed by GitHub
Browse files

[docs] Kandinsky guide (#4555) 

* kandinsky 2.1 first draft

* add kandinsky 2.2

* fix identical section headers

* try hfoptions syntax

* add img2img

* add inpaint

* add interpolate

* fix tag

* more cleanups

* typo

* update hfoptions id

* align hfoptions tags
parent c0f05826
Show whitespace changes
Inline Side-by-side
Showing with 740 additions and 717 deletions
docs/source/en/_toctree.yml
View file @ 75ea54a1
......@@ -40,6 +40,8 @@
title: Push files to the Hub
title: Loading & Hub
- sections:
- local: using-diffusers/pipeline_overview
title: Overview
- local: using-diffusers/unconditional_image_generation
title: Unconditional image generation
- local: using-diffusers/conditional_image_generation
......@@ -70,6 +72,8 @@
title: Overview
- local: using-diffusers/sdxl
title: Stable Diffusion XL
- local: using-diffusers/kandinsky
title: Kandinsky
- local: using-diffusers/controlnet
title: ControlNet
- local: using-diffusers/shap-e
......@@ -241,7 +245,7 @@
- local: api/pipelines/pix2pix
title: InstructPix2Pix
- local: api/pipelines/kandinsky
title: Kandinsky
title: Kandinsky 2.1
- local: api/pipelines/kandinsky_v22
title: Kandinsky 2.2
- local: api/pipelines/latent_consistency_models
......
docs/source/en/api/pipelines/kandinsky.md
View file @ 75ea54a1
......@@ -7,462 +7,60 @@ an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express o
specific language governing permissions and limitations under the License.
-->
# Kandinsky
# Kandinsky 2.1
## Overview
Kandinsky 2.1 is created by [Arseniy Shakhmatov](https://github.com/cene555), [Anton Razzhigaev](https://github.com/razzant), [Aleksandr Nikolich](https://github.com/AlexWortega), [Igor Pavlov](https://github.com/boomb0om), [Andrey Kuznetsov](https://github.com/kuznetsoffandrey) and [Denis Dimitrov](https://github.com/denndimitrov).
Kandinsky inherits best practices from [DALL-E 2](https://huggingface.co/papers/2204.06125) and [Latent Diffusion](https://huggingface.co/docs/diffusers/api/pipelines/latent_diffusion), while introducing some new ideas.
The description from it's GitHub page is:
It uses [CLIP](https://huggingface.co/docs/transformers/model_doc/clip) for encoding images and text, and a diffusion image prior (mapping) between latent spaces of CLIP modalities. This approach enhances the visual performance of the model and unveils new horizons in blending images and text-guided image manipulation.
*Kandinsky 2.1 inherits best practicies from Dall-E 2 and Latent diffusion, while introducing some new ideas. As text and image encoder it uses CLIP model and diffusion image prior (mapping) between latent spaces of CLIP modalities. This approach increases the visual performance of the model and unveils new horizons in blending images and text-guided image manipulation.*
The Kandinsky model is created by [Arseniy Shakhmatov](https://github.com/cene555), [Anton Razzhigaev](https://github.com/razzant), [Aleksandr Nikolich](https://github.com/AlexWortega), [Igor Pavlov](https://github.com/boomb0om), [Andrey Kuznetsov](https://github.com/kuznetsoffandrey) and [Denis Dimitrov](https://github.com/denndimitrov). The original codebase can be found [here](https://github.com/ai-forever/Kandinsky-2)
## Usage example
In the following, we will walk you through some examples of how to use the Kandinsky pipelines to create some visually aesthetic artwork.
### Text-to-Image Generation
For text-to-image generation, we need to use both [`KandinskyPriorPipeline`] and [`KandinskyPipeline`].
The first step is to encode text prompts with CLIP and then diffuse the CLIP text embeddings to CLIP image embeddings,
as first proposed in [DALL-E 2](https://cdn.openai.com/papers/dall-e-2.pdf).
Let's throw a fun prompt at Kandinsky to see what it comes up with.
```py
prompt = "A alien cheeseburger creature eating itself, claymation, cinematic, moody lighting"
```
First, let's instantiate the prior pipeline and the text-to-image pipeline. Both
pipelines are diffusion models.
```py
from diffusers import DiffusionPipeline
import torch
pipe_prior = DiffusionPipeline.from_pretrained("kandinsky-community/kandinsky-2-1-prior", torch_dtype=torch.float16)
pipe_prior.to("cuda")
t2i_pipe = DiffusionPipeline.from_pretrained("kandinsky-community/kandinsky-2-1", torch_dtype=torch.float16)
t2i_pipe.to("cuda")
```
<Tip warning={true}>
By default, the text-to-image pipeline use [`DDIMScheduler`], you can change the scheduler to [`DDPMScheduler`]
```py
scheduler = DDPMScheduler.from_pretrained("kandinsky-community/kandinsky-2-1", subfolder="ddpm_scheduler")
t2i_pipe = DiffusionPipeline.from_pretrained(
"kandinsky-community/kandinsky-2-1", scheduler=scheduler, torch_dtype=torch.float16
)
t2i_pipe.to("cuda")
```
</Tip>
Now we pass the prompt through the prior to generate image embeddings. The prior
returns both the image embeddings corresponding to the prompt and negative/unconditional image
embeddings corresponding to an empty string.
```py
image_embeds, negative_image_embeds = pipe_prior(prompt, guidance_scale=1.0).to_tuple()
```
<Tip warning={true}>
The text-to-image pipeline expects both `image_embeds`, `negative_image_embeds` and the original
`prompt` as the text-to-image pipeline uses another text encoder to better guide the second diffusion
process of `t2i_pipe`.
By default, the prior returns unconditioned negative image embeddings corresponding to the negative prompt of `""`.
For better results, you can also pass a `negative_prompt` to the prior. This will increase the effective batch size
of the prior by a factor of 2.
```py
prompt = "A alien cheeseburger creature eating itself, claymation, cinematic, moody lighting"
negative_prompt = "low quality, bad quality"
image_embeds, negative_image_embeds = pipe_prior(prompt, negative_prompt, guidance_scale=1.0).to_tuple()
```
</Tip>
Next, we can pass the embeddings as well as the prompt to the text-to-image pipeline. Remember that
in case you are using a customized negative prompt, that you should pass this one also to the text-to-image pipelines
with `negative_prompt=negative_prompt`:
```py
image = t2i_pipe(
prompt, image_embeds=image_embeds, negative_image_embeds=negative_image_embeds, height=768, width=768
).images[0]
image.save("cheeseburger_monster.png")
```
One cheeseburger monster coming up! Enjoy!
![img](https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/kandinsky-docs/cheeseburger.png)
The original codebase can be found at [ai-forever/Kandinsky-2](https://github.com/ai-forever/Kandinsky-2).
<Tip>
We also provide an end-to-end Kandinsky pipeline [`KandinskyCombinedPipeline`], which combines both the prior pipeline and text-to-image pipeline, and lets you perform inference in a single step. You can create the combined pipeline with the [`~AutoPipelineForText2Image.from_pretrained`] method
```python
from diffusers import AutoPipelineForText2Image
import torch
pipe = AutoPipelineForText2Image.from_pretrained(
"kandinsky-community/kandinsky-2-1", torch_dtype=torch.float16
)
pipe.enable_model_cpu_offload()
```
Under the hood, it will automatically load both [`KandinskyPriorPipeline`] and [`KandinskyPipeline`]. To generate images, you no longer need to call both pipelines and pass the outputs from one to another. You only need to call the combined pipeline once. You can set different `guidance_scale` and `num_inference_steps` for the prior pipeline with the `prior_guidance_scale` and `prior_num_inference_steps` arguments.
```python
prompt = "A alien cheeseburger creature eating itself, claymation, cinematic, moody lighting"
negative_prompt = "low quality, bad quality"
image = pipe(prompt=prompt, negative_prompt=negative_prompt, prior_guidance_scale =1.0, guidance_scacle = 4.0, height=768, width=768).images[0]
```
</Tip>
The Kandinsky model works extremely well with creative prompts. Here is some of the amazing art that can be created using the exact same process but with different prompts.
```python
prompt = "bird eye view shot of a full body woman with cyan light orange magenta makeup, digital art, long braided hair her face separated by makeup in the style of yin Yang surrealism, symmetrical face, real image, contrasting tone, pastel gradient background"
```
![img](https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/kandinsky-docs/hair.png)
```python
prompt = "A car exploding into colorful dust"
```
![img](https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/kandinsky-docs/dusts.png)
```python
prompt = "editorial photography of an organic, almost liquid smoke style armchair"
```
![img](https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/kandinsky-docs/smokechair.png)
```python
prompt = "birds eye view of a quilted paper style alien planet landscape, vibrant colours, Cinematic lighting"
```
![img](https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/kandinsky-docs/alienplanet.png)
### Text Guided Image-to-Image Generation
The same Kandinsky model weights can be used for text-guided image-to-image translation. In this case, just make sure to load the weights using the [`KandinskyImg2ImgPipeline`] pipeline.
**Note**: You can also directly move the weights of the text-to-image pipelines to the image-to-image pipelines
without loading them twice by making use of the [`~DiffusionPipeline.components`] function as explained [here](#converting-between-different-pipelines).
Let's download an image.
```python
from PIL import Image
import requests
from io import BytesIO
# download image
url = "https://raw.githubusercontent.com/CompVis/stable-diffusion/main/assets/stable-samples/img2img/sketch-mountains-input.jpg"
response = requests.get(url)
original_image = Image.open(BytesIO(response.content)).convert("RGB")
original_image = original_image.resize((768, 512))
```
![img](https://raw.githubusercontent.com/CompVis/stable-diffusion/main/assets/stable-samples/img2img/sketch-mountains-input.jpg)
```python
import torch
from diffusers import KandinskyImg2ImgPipeline, KandinskyPriorPipeline
# create prior
pipe_prior = KandinskyPriorPipeline.from_pretrained(
"kandinsky-community/kandinsky-2-1-prior", torch_dtype=torch.float16
)
pipe_prior.to("cuda")
# create img2img pipeline
pipe = KandinskyImg2ImgPipeline.from_pretrained("kandinsky-community/kandinsky-2-1", torch_dtype=torch.float16)
pipe.to("cuda")
prompt = "A fantasy landscape, Cinematic lighting"
negative_prompt = "low quality, bad quality"
image_embeds, negative_image_embeds = pipe_prior(prompt, negative_prompt).to_tuple()
out = pipe(
prompt,
image=original_image,
image_embeds=image_embeds,
negative_image_embeds=negative_image_embeds,
height=768,
width=768,
strength=0.3,
)
out.images[0].save("fantasy_land.png")
```
![img](https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/kandinsky-docs/img2img_fantasyland.png)
<Tip>
You can also use the [`KandinskyImg2ImgCombinedPipeline`] for end-to-end image-to-image generation with Kandinsky 2.1
```python
from diffusers import AutoPipelineForImage2Image
import torch
import requests
from io import BytesIO
from PIL import Image
import os
pipe = AutoPipelineForImage2Image.from_pretrained("kandinsky-community/kandinsky-2-1", torch_dtype=torch.float16)
pipe.enable_model_cpu_offload()
prompt = "A fantasy landscape, Cinematic lighting"
negative_prompt = "low quality, bad quality"
url = "https://raw.githubusercontent.com/CompVis/stable-diffusion/main/assets/stable-samples/img2img/sketch-mountains-input.jpg"
response = requests.get(url)
original_image = Image.open(BytesIO(response.content)).convert("RGB")
original_image.thumbnail((768, 768))
Check out the [Kandinsky Community](https://huggingface.co/kandinsky-community) organization on the Hub for the official model checkpoints for tasks like text-to-image, image-to-image, and inpainting.
image = pipe(prompt=prompt, image=original_image, strength=0.3).images[0]
```
</Tip>
### Text Guided Inpainting Generation
You can use [`KandinskyInpaintPipeline`] to edit images. In this example, we will add a hat to the portrait of a cat.
```py
from diffusers import KandinskyInpaintPipeline, KandinskyPriorPipeline
from diffusers.utils import load_image
import torch
import numpy as np
pipe_prior = KandinskyPriorPipeline.from_pretrained(
"kandinsky-community/kandinsky-2-1-prior", torch_dtype=torch.float16
)
pipe_prior.to("cuda")
prompt = "a hat"
prior_output = pipe_prior(prompt)
pipe = KandinskyInpaintPipeline.from_pretrained("kandinsky-community/kandinsky-2-1-inpaint", torch_dtype=torch.float16)
pipe.to("cuda")
init_image = load_image(
"https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main" "/kandinsky/cat.png"
)
mask = np.zeros((768, 768), dtype=np.float32)
# Let's mask out an area above the cat's head
mask[:250, 250:-250] = 1
out = pipe(
prompt,
image=init_image,
mask_image=mask,
**prior_output,
height=768,
width=768,
num_inference_steps=150,
)
image = out.images[0]
image.save("cat_with_hat.png")
```
![img](https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/kandinsky-docs/inpaint_cat_hat.png)
<Tip>
To use the [`KandinskyInpaintCombinedPipeline`] to perform end-to-end image inpainting generation, you can run below code instead
```python
from diffusers import AutoPipelineForInpainting
pipe = AutoPipelineForInpainting.from_pretrained("kandinsky-community/kandinsky-2-1-inpaint", torch_dtype=torch.float16)
pipe.enable_model_cpu_offload()
image = pipe(prompt=prompt, image=original_image, mask_image=mask).images[0]
```
</Tip>
🚨🚨🚨 __Breaking change for Kandinsky Mask Inpainting__ 🚨🚨🚨
We introduced a breaking change for Kandinsky inpainting pipeline in the following pull request: https://github.com/huggingface/diffusers/pull/4207. Previously we accepted a mask format where black pixels represent the masked-out area. This is inconsistent with all other pipelines in diffusers. We have changed the mask format in Knaindsky and now using white pixels instead.
Please upgrade your inpainting code to follow the above. If you are using Kandinsky Inpaint in production. You now need to change the mask to:
```python
# For PIL input
import PIL.ImageOps
mask = PIL.ImageOps.invert(mask)
# For PyTorch and Numpy input
mask = 1 - mask
```
### Interpolate
The [`KandinskyPriorPipeline`] also comes with a cool utility function that will allow you to interpolate the latent space of different images and texts super easily. Here is an example of how you can create an Impressionist-style portrait for your pet based on "The Starry Night".
Note that you can interpolate between texts and images - in the below example, we passed a text prompt "a cat" and two images to the `interplate` function, along with a `weights` variable containing the corresponding weights for each condition we interplate.
```python
from diffusers import KandinskyPriorPipeline, KandinskyPipeline
from diffusers.utils import load_image
import PIL
import torch
pipe_prior = KandinskyPriorPipeline.from_pretrained(
"kandinsky-community/kandinsky-2-1-prior", torch_dtype=torch.float16
)
pipe_prior.to("cuda")
img1 = load_image(
"https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main" "/kandinsky/cat.png"
)
img2 = load_image(
"https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main" "/kandinsky/starry_night.jpeg"
)
# add all the conditions we want to interpolate, can be either text or image
images_texts = ["a cat", img1, img2]
# specify the weights for each condition in images_texts
weights = [0.3, 0.3, 0.4]
# We can leave the prompt empty
prompt = ""
prior_out = pipe_prior.interpolate(images_texts, weights)
pipe = KandinskyPipeline.from_pretrained("kandinsky-community/kandinsky-2-1", torch_dtype=torch.float16)
pipe.to("cuda")
image = pipe(prompt, **prior_out, height=768, width=768).images[0]
image.save("starry_cat.png")
```
![img](https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/kandinsky-docs/starry_cat.png)
## Optimization
Running Kandinsky in inference requires running both a first prior pipeline: [`KandinskyPriorPipeline`]
and a second image decoding pipeline which is one of [`KandinskyPipeline`], [`KandinskyImg2ImgPipeline`], or [`KandinskyInpaintPipeline`].
The bulk of the computation time will always be the second image decoding pipeline, so when looking
into optimizing the model, one should look into the second image decoding pipeline.
When running with PyTorch < 2.0, we strongly recommend making use of [`xformers`](https://github.com/facebookresearch/xformers)
to speed-up the optimization. This can be done by simply running:
```py
from diffusers import DiffusionPipeline
import torch
t2i_pipe = DiffusionPipeline.from_pretrained("kandinsky-community/kandinsky-2-1", torch_dtype=torch.float16)
t2i_pipe.enable_xformers_memory_efficient_attention()
```
When running on PyTorch >= 2.0, PyTorch's SDPA attention will automatically be used. For more information on
PyTorch's SDPA, feel free to have a look at [this blog post](https://pytorch.org/blog/accelerated-diffusers-pt-20/).
To have explicit control , you can also manually set the pipeline to use PyTorch's 2.0 efficient attention:
```py
from diffusers.models.attention_processor import AttnAddedKVProcessor2_0
t2i_pipe.unet.set_attn_processor(AttnAddedKVProcessor2_0())
```
The slowest and most memory intense attention processor is the default `AttnAddedKVProcessor` processor.
We do **not** recommend using it except for testing purposes or cases where very high determistic behaviour is desired.
You can set it with:
```py
from diffusers.models.attention_processor import AttnAddedKVProcessor
t2i_pipe.unet.set_attn_processor(AttnAddedKVProcessor())
```
With PyTorch >= 2.0, you can also use Kandinsky with `torch.compile` which depending
on your hardware can significantly speed-up your inference time once the model is compiled.
To use Kandinsksy with `torch.compile`, you can do:
```py
t2i_pipe.unet.to(memory_format=torch.channels_last)
t2i_pipe.unet = torch.compile(t2i_pipe.unet, mode="reduce-overhead", fullgraph=True)
```
After compilation you should see a very fast inference time. For more information,
feel free to have a look at [Our PyTorch 2.0 benchmark](https://huggingface.co/docs/diffusers/main/en/optimization/torch2.0).
<Tip>
To generate images directly from a single pipeline, you can use [`KandinskyCombinedPipeline`], [`KandinskyImg2ImgCombinedPipeline`], [`KandinskyInpaintCombinedPipeline`].
These combined pipelines wrap the [`KandinskyPriorPipeline`] and [`KandinskyPipeline`], [`KandinskyImg2ImgPipeline`], [`KandinskyInpaintPipeline`] respectively into a single
pipeline for a simpler user experience
</Tip>
## Available Pipelines:
| Pipeline | Tasks |
|---|---|
| [pipeline_kandinsky.py](https://github.com/huggingface/diffusers/blob/main/src/diffusers/pipelines/kandinsky/pipeline_kandinsky.py) | *Text-to-Image Generation* |
| [pipeline_kandinsky_combined.py](https://github.com/huggingface/diffusers/blob/main/src/diffusers/pipelines/kandinsky2_2/pipeline_kandinsky_combined.py) | *End-to-end Text-to-Image, image-to-image, Inpainting Generation* |
| [pipeline_kandinsky_inpaint.py](https://github.com/huggingface/diffusers/blob/main/src/diffusers/pipelines/kandinsky/pipeline_kandinsky_inpaint.py) | *Image-Guided Image Generation* |
| [pipeline_kandinsky_img2img.py](https://github.com/huggingface/diffusers/blob/main/src/diffusers/pipelines/kandinsky/pipeline_kandinsky_img2img.py) | *Image-Guided Image Generation* |
### KandinskyPriorPipeline
## KandinskyPriorPipeline
[[autodoc]] KandinskyPriorPipeline
- all
- __call__
- interpolate
### KandinskyPipeline
## KandinskyPipeline
[[autodoc]] KandinskyPipeline
- all
- __call__
### KandinskyImg2ImgPipeline
## KandinskyCombinedPipeline
[[autodoc]] KandinskyImg2ImgPipeline
[[autodoc]] KandinskyCombinedPipeline
- all
- __call__
### KandinskyInpaintPipeline
## KandinskyImg2ImgPipeline
[[autodoc]] KandinskyInpaintPipeline
[[autodoc]] KandinskyImg2ImgPipeline
- all
- __call__
### KandinskyCombinedPipeline
## KandinskyImg2ImgCombinedPipeline
[[autodoc]] KandinskyCombinedPipeline
[[autodoc]] KandinskyImg2ImgCombinedPipeline
- all
- __call__
### KandinskyImg2ImgCombinedPipeline
## KandinskyInpaintPipeline
[[autodoc]] KandinskyImg2ImgCombinedPipeline
[[autodoc]] KandinskyInpaintPipeline
- all
- __call__
### KandinskyInpaintCombinedPipeline
## KandinskyInpaintCombinedPipeline
[[autodoc]] KandinskyInpaintCombinedPipeline
- all
......
docs/source/en/api/pipelines/kandinsky_v22.md
View file @ 75ea54a1
......@@ -9,348 +9,77 @@ specific language governing permissions and limitations under the License.
# Kandinsky 2.2
The Kandinsky 2.2 release includes robust new text-to-image models that support text-to-image generation, image-to-image generation, image interpolation, and text-guided image inpainting. The general workflow to perform these tasks using Kandinsky 2.2 is the same as in Kandinsky 2.1. First, you will need to use a prior pipeline to generate image embeddings based on your text prompt, and then use one of the image decoding pipelines to generate the output image. The only difference is that in Kandinsky 2.2, all of the decoding pipelines no longer accept the `prompt` input, and the image generation process is conditioned with only `image_embeds` and `negative_image_embeds`.
Kandinsky 2.1 is created by [Arseniy Shakhmatov](https://github.com/cene555), [Anton Razzhigaev](https://github.com/razzant), [Aleksandr Nikolich](https://github.com/AlexWortega), [Igor Pavlov](https://github.com/boomb0om), [Andrey Kuznetsov](https://github.com/kuznetsoffandrey) and [Denis Dimitrov](https://github.com/denndimitrov).
Same as with Kandinsky 2.1, the easiest way to perform text-to-image generation is to use the combined Kandinsky pipeline. This process is exactly the same as Kandinsky 2.1. All you need to do is to replace the Kandinsky 2.1 checkpoint with 2.2.
The description from it's GitHub page is:
```python
from diffusers import AutoPipelineForText2Image
import torch
*Kandinsky 2.2 brings substantial improvements upon its predecessor, Kandinsky 2.1, by introducing a new, more powerful image encoder - CLIP-ViT-G and the ControlNet support. The switch to CLIP-ViT-G as the image encoder significantly increases the model's capability to generate more aesthetic pictures and better understand text, thus enhancing the model's overall performance. The addition of the ControlNet mechanism allows the model to effectively control the process of generating images. This leads to more accurate and visually appealing outputs and opens new possibilities for text-guided image manipulation.*
pipe = AutoPipelineForText2Image.from_pretrained("kandinsky-community/kandinsky-2-2-decoder", torch_dtype=torch.float16)
pipe.enable_model_cpu_offload()
prompt = "A alien cheeseburger creature eating itself, claymation, cinematic, moody lighting"
negative_prompt = "low quality, bad quality"
image = pipe(prompt=prompt, negative_prompt=negative_prompt, prior_guidance_scale =1.0, height=768, width=768).images[0]
```
Now, let's look at an example where we take separate steps to run the prior pipeline and text-to-image pipeline. This way, we can understand what's happening under the hood and how Kandinsky 2.2 differs from Kandinsky 2.1.
First, let's create the prior pipeline and text-to-image pipeline with Kandinsky 2.2 checkpoints.
```python
from diffusers import DiffusionPipeline
import torch
pipe_prior = DiffusionPipeline.from_pretrained("kandinsky-community/kandinsky-2-2-prior", torch_dtype=torch.float16)
pipe_prior.to("cuda")
t2i_pipe = DiffusionPipeline.from_pretrained("kandinsky-community/kandinsky-2-2-decoder", torch_dtype=torch.float16)
t2i_pipe.to("cuda")
```
You can then use `pipe_prior` to generate image embeddings.
```python
prompt = "portrait of a women, blue eyes, cinematic"
negative_prompt = "low quality, bad quality"
image_embeds, negative_image_embeds = pipe_prior(prompt, guidance_scale=1.0).to_tuple()
```
Now you can pass these embeddings to the text-to-image pipeline. When using Kandinsky 2.2 you don't need to pass the `prompt` (but you do with the previous version, Kandinsky 2.1).
```
image = t2i_pipe(image_embeds=image_embeds, negative_image_embeds=negative_image_embeds, height=768, width=768).images[
0
]
image.save("portrait.png")
```
![img](https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinskyv22/%20blue%20eyes.png)
We used the text-to-image pipeline as an example, but the same process applies to all decoding pipelines in Kandinsky 2.2. For more information, please refer to our API section for each pipeline.
### Text-to-Image Generation with ControlNet Conditioning
In the following, we give a simple example of how to use [`KandinskyV22ControlnetPipeline`] to add control to the text-to-image generation with a depth image.
First, let's take an image and extract its depth map.
```python
from diffusers.utils import load_image
img = load_image(
"https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinskyv22/cat.png"
).resize((768, 768))
```
![img](https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinskyv22/cat.png)
We can use the `depth-estimation` pipeline from transformers to process the image and retrieve its depth map.
```python
import torch
import numpy as np
from transformers import pipeline
from diffusers.utils import load_image
def make_hint(image, depth_estimator):
image = depth_estimator(image)["depth"]
image = np.array(image)
image = image[:, :, None]
image = np.concatenate([image, image, image], axis=2)
detected_map = torch.from_numpy(image).float() / 255.0
hint = detected_map.permute(2, 0, 1)
return hint
depth_estimator = pipeline("depth-estimation")
hint = make_hint(img, depth_estimator).unsqueeze(0).half().to("cuda")
```
Now, we load the prior pipeline and the text-to-image controlnet pipeline
```python
from diffusers import KandinskyV22PriorPipeline, KandinskyV22ControlnetPipeline
pipe_prior = KandinskyV22PriorPipeline.from_pretrained(
"kandinsky-community/kandinsky-2-2-prior", torch_dtype=torch.float16
)
pipe_prior = pipe_prior.to("cuda")
pipe = KandinskyV22ControlnetPipeline.from_pretrained(
"kandinsky-community/kandinsky-2-2-controlnet-depth", torch_dtype=torch.float16
)
pipe = pipe.to("cuda")
```
We pass the prompt and negative prompt through the prior to generate image embeddings
```python
prompt = "A robot, 4k photo"
negative_prior_prompt = "lowres, text, error, cropped, worst quality, low quality, jpeg artifacts, ugly, duplicate, morbid, mutilated, out of frame, extra fingers, mutated hands, poorly drawn hands, poorly drawn face, mutation, deformed, blurry, dehydrated, bad anatomy, bad proportions, extra limbs, cloned face, disfigured, gross proportions, malformed limbs, missing arms, missing legs, extra arms, extra legs, fused fingers, too many fingers, long neck, username, watermark, signature"
generator = torch.Generator(device="cuda").manual_seed(43)
image_emb, zero_image_emb = pipe_prior(
prompt=prompt, negative_prompt=negative_prior_prompt, generator=generator
).to_tuple()
```
Now we can pass the image embeddings and the depth image we extracted to the controlnet pipeline. With Kandinsky 2.2, only prior pipelines accept `prompt` input. You do not need to pass the prompt to the controlnet pipeline.
```python
images = pipe(
image_embeds=image_emb,
negative_image_embeds=zero_image_emb,
hint=hint,
num_inference_steps=50,
generator=generator,
height=768,
width=768,
).images
images[0].save("robot_cat.png")
```
The output image looks as follow:
![img](https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinskyv22/robot_cat_text2img.png)
### Image-to-Image Generation with ControlNet Conditioning
Kandinsky 2.2 also includes a [`KandinskyV22ControlnetImg2ImgPipeline`] that will allow you to add control to the image generation process with both the image and its depth map. This pipeline works really well with [`KandinskyV22PriorEmb2EmbPipeline`], which generates image embeddings based on both a text prompt and an image.
For our robot cat example, we will pass the prompt and cat image together to the prior pipeline to generate an image embedding. We will then use that image embedding and the depth map of the cat to further control the image generation process.
We can use the same cat image and its depth map from the last example.
```python
import torch
import numpy as np
from diffusers import KandinskyV22PriorEmb2EmbPipeline, KandinskyV22ControlnetImg2ImgPipeline
from diffusers.utils import load_image
from transformers import pipeline
img = load_image(
"https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main" "/kandinskyv22/cat.png"
).resize((768, 768))
def make_hint(image, depth_estimator):
image = depth_estimator(image)["depth"]
image = np.array(image)
image = image[:, :, None]
image = np.concatenate([image, image, image], axis=2)
detected_map = torch.from_numpy(image).float() / 255.0
hint = detected_map.permute(2, 0, 1)
return hint
depth_estimator = pipeline("depth-estimation")
hint = make_hint(img, depth_estimator).unsqueeze(0).half().to("cuda")
pipe_prior = KandinskyV22PriorEmb2EmbPipeline.from_pretrained(
"kandinsky-community/kandinsky-2-2-prior", torch_dtype=torch.float16
)
pipe_prior = pipe_prior.to("cuda")
pipe = KandinskyV22ControlnetImg2ImgPipeline.from_pretrained(
"kandinsky-community/kandinsky-2-2-controlnet-depth", torch_dtype=torch.float16
)
pipe = pipe.to("cuda")
prompt = "A robot, 4k photo"
negative_prior_prompt = "lowres, text, error, cropped, worst quality, low quality, jpeg artifacts, ugly, duplicate, morbid, mutilated, out of frame, extra fingers, mutated hands, poorly drawn hands, poorly drawn face, mutation, deformed, blurry, dehydrated, bad anatomy, bad proportions, extra limbs, cloned face, disfigured, gross proportions, malformed limbs, missing arms, missing legs, extra arms, extra legs, fused fingers, too many fingers, long neck, username, watermark, signature"
generator = torch.Generator(device="cuda").manual_seed(43)
# run prior pipeline
img_emb = pipe_prior(prompt=prompt, image=img, strength=0.85, generator=generator)
negative_emb = pipe_prior(prompt=negative_prior_prompt, image=img, strength=1, generator=generator)
# run controlnet img2img pipeline
images = pipe(
image=img,
strength=0.5,
image_embeds=img_emb.image_embeds,
negative_image_embeds=negative_emb.image_embeds,
hint=hint,
num_inference_steps=50,
generator=generator,
height=768,
width=768,
).images
images[0].save("robot_cat.png")
```
Here is the output. Compared with the output from our text-to-image controlnet example, it kept a lot more cat facial details from the original image and worked into the robot style we asked for.
![img](https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinskyv22/robot_cat.png)
## Optimization
Running Kandinsky in inference requires running both a first prior pipeline: [`KandinskyPriorPipeline`]
and a second image decoding pipeline which is one of [`KandinskyPipeline`], [`KandinskyImg2ImgPipeline`], or [`KandinskyInpaintPipeline`].
The bulk of the computation time will always be the second image decoding pipeline, so when looking
into optimizing the model, one should look into the second image decoding pipeline.
When running with PyTorch < 2.0, we strongly recommend making use of [`xformers`](https://github.com/facebookresearch/xformers)
to speed-up the optimization. This can be done by simply running:
```py
from diffusers import DiffusionPipeline
import torch
t2i_pipe = DiffusionPipeline.from_pretrained("kandinsky-community/kandinsky-2-2-decoder", torch_dtype=torch.float16)
t2i_pipe.enable_xformers_memory_efficient_attention()
```
When running on PyTorch >= 2.0, PyTorch's SDPA attention will automatically be used. For more information on
PyTorch's SDPA, feel free to have a look at [this blog post](https://pytorch.org/blog/accelerated-diffusers-pt-20/).
To have explicit control , you can also manually set the pipeline to use PyTorch's 2.0 efficient attention:
```py
from diffusers.models.attention_processor import AttnAddedKVProcessor2_0
t2i_pipe.unet.set_attn_processor(AttnAddedKVProcessor2_0())
```
The slowest and most memory intense attention processor is the default `AttnAddedKVProcessor` processor.
We do **not** recommend using it except for testing purposes or cases where very high determistic behaviour is desired.
You can set it with:
```py
from diffusers.models.attention_processor import AttnAddedKVProcessor
t2i_pipe.unet.set_attn_processor(AttnAddedKVProcessor())
```
With PyTorch >= 2.0, you can also use Kandinsky with `torch.compile` which depending
on your hardware can significantly speed-up your inference time once the model is compiled.
To use Kandinsksy with `torch.compile`, you can do:
```py
t2i_pipe.unet.to(memory_format=torch.channels_last)
t2i_pipe.unet = torch.compile(t2i_pipe.unet, mode="reduce-overhead", fullgraph=True)
```
After compilation you should see a very fast inference time. For more information,
feel free to have a look at [Our PyTorch 2.0 benchmark](https://huggingface.co/docs/diffusers/main/en/optimization/torch2.0).
The original codebase can be found at [ai-forever/Kandinsky-2](https://github.com/ai-forever/Kandinsky-2).
<Tip>
To generate images directly from a single pipeline, you can use [`KandinskyV22CombinedPipeline`], [`KandinskyV22Img2ImgCombinedPipeline`], [`KandinskyV22InpaintCombinedPipeline`].
These combined pipelines wrap the [`KandinskyV22PriorPipeline`] and [`KandinskyV22Pipeline`], [`KandinskyV22Img2ImgPipeline`], [`KandinskyV22InpaintPipeline`] respectively into a single
pipeline for a simpler user experience
Check out the [Kandinsky Community](https://huggingface.co/kandinsky-community) organization on the Hub for the official model checkpoints for tasks like text-to-image, image-to-image, and inpainting.
</Tip>
## Available Pipelines:
| Pipeline | Tasks |
|---|---|
| [pipeline_kandinsky2_2.py](https://github.com/huggingface/diffusers/blob/main/src/diffusers/pipelines/kandinsky2_2/pipeline_kandinsky2_2.py) | *Text-to-Image Generation* |
| [pipeline_kandinsky2_2_combined.py](https://github.com/huggingface/diffusers/blob/main/src/diffusers/pipelines/kandinsky2_2/pipeline_kandinsky2_2_combined.py) | *End-to-end Text-to-Image, image-to-image, Inpainting Generation* |
| [pipeline_kandinsky2_2_inpaint.py](https://github.com/huggingface/diffusers/blob/main/src/diffusers/pipelines/kandinsky2_2/pipeline_kandinsky2_2_inpaint.py) | *Image-Guided Image Generation* |
| [pipeline_kandinsky2_2_img2img.py](https://github.com/huggingface/diffusers/blob/main/src/diffusers/pipelines/kandinsky2_2/pipeline_kandinsky2_2_img2img.py) | *Image-Guided Image Generation* |
| [pipeline_kandinsky2_2_controlnet.py](https://github.com/huggingface/diffusers/blob/main/src/diffusers/pipelines/kandinsky2_2/pipeline_kandinsky2_2_controlnet.py) | *Image-Guided Image Generation* |
| [pipeline_kandinsky2_2_controlnet_img2img.py](https://github.com/huggingface/diffusers/blob/main/src/diffusers/pipelines/kandinsky2_2/pipeline_kandinsky2_2_controlnet_img2img.py) | *Image-Guided Image Generation* |
## KandinskyV22PriorPipeline
### KandinskyV22Pipeline
[[autodoc]] KandinskyV22Pipeline
[[autodoc]] KandinskyV22PriorPipeline
- all
- __call__
- interpolate
### KandinskyV22ControlnetPipeline
## KandinskyV22Pipeline
[[autodoc]] KandinskyV22ControlnetPipeline
[[autodoc]] KandinskyV22Pipeline
- all
- __call__
### KandinskyV22ControlnetImg2ImgPipeline
## KandinskyV22CombinedPipeline
[[autodoc]] KandinskyV22ControlnetImg2ImgPipeline
[[autodoc]] KandinskyV22CombinedPipeline
- all
- __call__
### KandinskyV22Img2ImgPipeline
## KandinskyV22ControlnetPipeline
[[autodoc]] KandinskyV22Img2ImgPipeline
[[autodoc]] KandinskyV22ControlnetPipeline
- all
- __call__
### KandinskyV22InpaintPipeline
## KandinskyV22PriorEmb2EmbPipeline
[[autodoc]] KandinskyV22InpaintPipeline
[[autodoc]] KandinskyV22PriorEmb2EmbPipeline
- all
- __call__
- interpolate
### KandinskyV22PriorPipeline
## KandinskyV22Img2ImgPipeline
[[autodoc]] KandinskyV22PriorPipeline
[[autodoc]] KandinskyV22Img2ImgPipeline
- all
- __call__
- interpolate
### KandinskyV22PriorEmb2EmbPipeline
## KandinskyV22Img2ImgCombinedPipeline
[[autodoc]] KandinskyV22PriorEmb2EmbPipeline
[[autodoc]] KandinskyV22Img2ImgCombinedPipeline
- all
- __call__
- interpolate
### KandinskyV22CombinedPipeline
## KandinskyV22ControlnetImg2ImgPipeline
[[autodoc]] KandinskyV22CombinedPipeline
[[autodoc]] KandinskyV22ControlnetImg2ImgPipeline
- all
- __call__
### KandinskyV22Img2ImgCombinedPipeline
## KandinskyV22InpaintPipeline
[[autodoc]] KandinskyV22Img2ImgCombinedPipeline
[[autodoc]] KandinskyV22InpaintPipeline
- all
- __call__
### KandinskyV22InpaintCombinedPipeline
## KandinskyV22InpaintCombinedPipeline
[[autodoc]] KandinskyV22InpaintCombinedPipeline
- all
......
docs/source/en/using-diffusers/kandinsky.md 0 → 100644
View file @ 75ea54a1
# Kandinsky
[[open-in-colab]]
The Kandinsky models are a series of multilingual text-to-image generation models. The Kandinsky 2.0 model uses two multilingual text encoders and concatenates those results for the UNet.
[Kandinsky 2.1](../api/pipelines/kandinsky) changes the architecture to include an image prior model ([`CLIP`](https://huggingface.co/docs/transformers/model_doc/clip)) to generate a mapping between text and image embeddings. The mapping provides better text-image alignment and it is used with the text embeddings during training, leading to higher quality results. Finally, Kandinsky 2.1 uses a [Modulating Quantized Vectors (MoVQ)](https://huggingface.co/papers/2209.09002) decoder - which adds a spatial conditional normalization layer to increase photorealism - to decode the latents into images.
[Kandinsky 2.2](../api/pipelines/kandinsky_v22) improves on the previous model by replacing the image encoder of the image prior model with a larger CLIP-ViT-G model to improve quality. The image prior model was also retrained on images with different resolutions and aspect ratios to generate higher-resolution images and different image sizes.
This guide will show you how to use the Kandinsky models for text-to-image, image-to-image, inpainting, interpolation, and more.
Before you begin, make sure you have the following libraries installed:
```py
# uncomment to install the necessary libraries in Colab
#!pip install transformers accelerate safetensors
```
<Tip warning={true}>
Kandinsky 2.1 and 2.2 usage is very similar! The only difference is Kandinsky 2.2 doesn't accept `prompt` as an input when decoding the latents. Instead, Kandinsky 2.2 only accepts `image_embeds` during decoding.
</Tip>
## Text-to-image
To use the Kandinsky models for any task, you always start by setting up the prior pipeline to encode the prompt and generate the image embeddings. The prior pipeline also generates `negative_image_embeds` that correspond to the negative prompt `""`. For better results, you can pass an actual `negative_prompt` to the prior pipeline, but this'll increase the effective batch size of the prior pipeline by 2x.
<hfoptions id="text-to-image">
<hfoption id="Kandinsky 2.1">
```py
from diffusers import KandinskyPriorPipeline, KandinskyPipeline
import torch
prior_pipeline = KandinskyPriorPipeline.from_pretrained("kandinsky-community/kandinsky-2-1-prior", torch_dtype=torch.float16).to("cuda")
pipeline = KandinskyPipeline.from_pretrained("kandinsky-community/kandinsky-2-1", torch_dtype=torch.float16).to("cuda")
prompt = "A alien cheeseburger creature eating itself, claymation, cinematic, moody lighting"
negative_prompt = "low quality, bad quality" # optional to include a negative prompt, but results are usually better
image_embeds, negative_image_embeds = prior_pipeline(prompt, negative_prompt, guidance_scale=1.0).to_tuple()
```
Now pass all the prompts and embeddings to the [`KandinskyPipeline`] to generate an image:
```py
image = pipeline(prompt, image_embeds=image_embeds, negative_prompt=negative_prompt, negative_image_embeds=negative_image_embeds, height=768, width=768).images[0]
```
<div class="flex justify-center">
<img class="rounded-xl" src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/kandinsky-docs/cheeseburger.png"/>
</div>
</hfoption>
<hfoption id="Kandinsky 2.2">
```py
from diffusers import KandinskyV22PriorPipeline, KandinskyV22Pipeline
import torch
prior_pipeline = KandinskyV22PriorPipeline.from_pretrained("kandinsky-community/kandinsky-2-2-prior", torch_dtype=torch.float16).to("cuda")
pipeline = KandinskyV22Pipeline.from_pretrained("kandinsky-community/kandinsky-2-2-decoder", torch_dtype=torch.float16).to("cuda")
prompt = "A alien cheeseburger creature eating itself, claymation, cinematic, moody lighting"
negative_prompt = "low quality, bad quality" # optional to include a negative prompt, but results are usually better
image_embeds, negative_image_embeds = prior_pipeline(prompt, guidance_scale=1.0).to_tuple()
```
Pass the `image_embeds` and `negative_image_embeds` to the [`KandinskyV22Pipeline`] to generate an image:
```py
image = pipeline(image_embeds=image_embeds, negative_image_embeds=negative_image_embeds, height=768, width=768).images[0]
```
<div class="flex justify-center">
<img class="rounded-xl" src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/kandinsky-text-to-image.png"/>
</div>
</hfoption>
</hfoptions>
🤗 Diffusers also provides an end-to-end API with the [`KandinskyCombinedPipeline`] and [`KandinskyV22CombinedPipeline`], meaning you don't have to separately load the prior and text-to-image pipeline. The combined pipeline automatically loads both the prior model and the decoder. You can still set different values for the prior pipeline with the `prior_guidance_scale` and `prior_num_inference_steps` parameters if you want.
Use the [`AutoPipelineForText2Image`] to automatically call the combined pipelines under the hood:
<hfoptions id="text-to-image">
<hfoption id="Kandinsky 2.1">
```py
from diffusers import AutoPipelineForText2Image
import torch
pipeline = AutoPipelineForText2Image.from_pretrained("kandinsky-community/kandinsky-2-1", torch_dtype=torch.float16).to("cuda")
pipeline.enable_model_cpu_offload()
prompt = "A alien cheeseburger creature eating itself, claymation, cinematic, moody lighting"
negative_prompt = "low quality, bad quality"
image = pipeline(prompt=prompt, negative_prompt=negative_prompt, prior_guidance_scale=1.0, guidance_scale = 4.0, height=768, width=768).images[0]
```
</hfoption>
<hfoption id="Kandinsky 2.2">
```py
from diffusers import AutoPipelineForText2Image
import torch
pipeline = AutoPipelineForText2Image.from_pretrained("kandinsky-community/kandinsky-2-2-decoder", torch_dtype=torch.float16).to("cuda")
pipeline.enable_model_cpu_offload()
prompt = "A alien cheeseburger creature eating itself, claymation, cinematic, moody lighting"
negative_prompt = "low quality, bad quality"
image = pipeline(prompt=prompt, negative_prompt=negative_prompt, prior_guidance_scale=1.0, guidance_scale = 4.0, height=768, width=768).images[0]
```
</hfoption>
</hfoptions>
## Image-to-image
For image-to-image, pass the initial image and text prompt to condition the image with to the pipeline. Start by loading the prior pipeline:
<hfoptions id="image-to-image">
<hfoption id="Kandinsky 2.1">
```py
import torch
from diffusers import KandinskyImg2ImgPipeline, KandinskyPriorPipeline
prior_pipeline = KandinskyPriorPipeline.from_pretrained("kandinsky-community/kandinsky-2-1-prior", torch_dtype=torch.float16, use_safetensors=True).to("cuda")
pipeline = KandinskyImg2ImgPipeline.from_pretrained("kandinsky-community/kandinsky-2-1", torch_dtype=torch.float16, use_safetensors=True).to("cuda")
```
</hfoption>
<hfoption id="Kandinsky 2.2">
```py
import torch
from diffusers import KandinskyV22Img2ImgPipeline, KandinskyPriorPipeline
prior_pipeline = KandinskyPriorPipeline.from_pretrained("kandinsky-community/kandinsky-2-2-prior", torch_dtype=torch.float16, use_safetensors=True).to("cuda")
pipeline = KandinskyV22Img2ImgPipeline.from_pretrained("kandinsky-community/kandinsky-2-2-decoder", torch_dtype=torch.float16, use_safetensors=True).to("cuda")
```
</hfoption>
</hfoptions>
Download an image to condition on:
```py
from PIL import Image
import requests
from io import BytesIO
# download image
url = "https://raw.githubusercontent.com/CompVis/stable-diffusion/main/assets/stable-samples/img2img/sketch-mountains-input.jpg"
response = requests.get(url)
original_image = Image.open(BytesIO(response.content)).convert("RGB")
original_image = original_image.resize((768, 512))
```
<div class="flex justify-center">
<img class="rounded-xl" src="https://raw.githubusercontent.com/CompVis/stable-diffusion/main/assets/stable-samples/img2img/sketch-mountains-input.jpg"/>
</div>
Generate the `image_embeds` and `negative_image_embeds` with the prior pipeline:
```py
prompt = "A fantasy landscape, Cinematic lighting"
negative_prompt = "low quality, bad quality"
image_embeds, negative_image_embeds = prior_pipeline(prompt, negative_prompt).to_tuple()
```
Now pass the original image, and all the prompts and embeddings to the pipeline to generate an image:
<hfoptions id="image-to-image">
<hfoption id="Kandinsky 2.1">
```py
image = pipeline(prompt, negative_prompt=negative_prompt, image=original_image, image_embeds=image_emebds, negative_image_embeds=negative_image_embeds, height=768, width=768, strength=0.3).images[0]
```
<div class="flex justify-center">
<img class="rounded-xl" src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/kandinsky-docs/img2img_fantasyland.png"/>
</div>
</hfoption>
<hfoption id="Kandinsky 2.2">
```py
image = pipeline(image=original_image, image_embeds=image_emebds, negative_image_embeds=negative_image_embeds, height=768, width=768, strength=0.3).images[0]
```
<div class="flex justify-center">
<img class="rounded-xl" src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/kandinsky-image-to-image.png"/>
</div>
</hfoption>
</hfoptions>
🤗 Diffusers also provides an end-to-end API with the [`KandinskyImg2ImgCombinedPipeline`] and [`KandinskyV22Img2ImgCombinedPipeline`], meaning you don't have to separately load the prior and image-to-image pipeline. The combined pipeline automatically loads both the prior model and the decoder. You can still set different values for the prior pipeline with the `prior_guidance_scale` and `prior_num_inference_steps` parameters if you want.
Use the [`AutoPipelineForImage2Image`] to automatically call the combined pipelines under the hood:
<hfoptions id="image-to-image">
<hfoption id="Kandinsky 2.1">
```py
from diffusers import AutoPipelineForImage2Image
import torch
import requests
from io import BytesIO
from PIL import Image
import os
pipeline = AutoPipelineForImage2Image.from_pretrained("kandinsky-community/kandinsky-2-1", torch_dtype=torch.float16, use_safetensors=True).to("cuda")
pipeline.enable_model_cpu_offload()
prompt = "A fantasy landscape, Cinematic lighting"
negative_prompt = "low quality, bad quality"
url = "https://raw.githubusercontent.com/CompVis/stable-diffusion/main/assets/stable-samples/img2img/sketch-mountains-input.jpg"
response = requests.get(url)
original_image = Image.open(BytesIO(response.content)).convert("RGB")
original_image.thumbnail((768, 768))
image = pipeline(prompt=prompt, image=original_image, strength=0.3).images[0]
```
</hfoption>
<hfoption id="Kandinsky 2.2">
```py
from diffusers import AutoPipelineForImage2Image
import torch
import requests
from io import BytesIO
from PIL import Image
import os
pipeline = AutoPipelineForImage2Image.from_pretrained("kandinsky-community/kandinsky-2-2-decoder", torch_dtype=torch.float16).to("cuda")
pipeline.enable_model_cpu_offload()
prompt = "A fantasy landscape, Cinematic lighting"
negative_prompt = "low quality, bad quality"
url = "https://raw.githubusercontent.com/CompVis/stable-diffusion/main/assets/stable-samples/img2img/sketch-mountains-input.jpg"
response = requests.get(url)
original_image = Image.open(BytesIO(response.content)).convert("RGB")
original_image.thumbnail((768, 768))
image = pipeline(prompt=prompt, image=original_image, strength=0.3).images[0]
```
</hfoption>
</hfoptions>
## Inpainting
<Tip warning={true}>
⚠️ The Kandinsky models uses ⬜️ **white pixels** to represent the masked area now instead of black pixels. If you are using [`KandinskyInpaintPipeline`] in production, you need to change the mask to use white pixels:
```py
# For PIL input
import PIL.ImageOps
mask = PIL.ImageOps.invert(mask)
# For PyTorch and NumPy input
mask = 1 - mask
```
</Tip>
For inpainting, you'll need the original image, a mask of the area to replace in the original image, and a text prompt of what to inpaint. Load the prior pipeline:
<hfoptions id="inpaint">
<hfoption id="Kandinsky 2.1">
```py
from diffusers import KandinskyInpaintPipeline, KandinskyPriorPipeline
from diffusers.utils import load_image
import torch
import numpy as np
prior_pipeline = KandinskyPriorPipeline.from_pretrained("kandinsky-community/kandinsky-2-1-prior", torch_dtype=torch.float16, use_safetensors=True).to("cuda")
pipeline = KandinskyInpaintPipeline.from_pretrained("kandinsky-community/kandinsky-2-1-inpaint", torch_dtype=torch.float16, use_safetensors=True).to("cuda")
```
</hfoption>
<hfoption id="Kandinsky 2.2">
```py
from diffusers import KandinskyV22InpaintPipeline, KandinskyV22PriorPipeline
from diffusers.utils import load_image
import torch
import numpy as np
prior_pipeline = KandinskyV22PriorPipeline.from_pretrained("kandinsky-community/kandinsky-2-2-prior", torch_dtype=torch.float16, use_safetensors=True).to("cuda")
pipeline = KandinskyV22InpaintPipeline.from_pretrained("kandinsky-community/kandinsky-2-2-decoder-inpaint", torch_dtype=torch.float16, use_safetensors=True).to("cuda")
```
</hfoption>
</hfoptions>
Load an initial image and create a mask:
```py
init_image = load_image("https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky/cat.png")
mask = np.zeros((768, 768), dtype=np.float32)
# mask area above cat's head
mask[:250, 250:-250] = 1
```
Generate the embeddings with the prior pipeline:
```py
prompt = "a hat"
prior_output = prior_pipeline(prompt)
```
Now pass the initial image, mask, and prompt and embeddings to the pipeline to generate an image:
<hfoptions id="inpaint">
<hfoption id="Kandinsky 2.1">
```py
image = pipeline(prompt, image=init_image, mask_image=mask, **prior_output, height=768, width=768, num_inference_steps=150).images[0]
```
<div class="flex justify-center">
<img class="rounded-xl" src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/kandinsky-docs/inpaint_cat_hat.png"/>
</div>
</hfoption>
<hfoption id="Kandinsky 2.2">
```py
image = pipeline(image=init_image, mask_image=mask, **prior_output, height=768, width=768, num_inference_steps=150).images[0]
```
<div class="flex justify-center">
<img class="rounded-xl" src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/kandinskyv22-inpaint.png"/>
</div>
</hfoption>
</hfoptions>
You can also use the end-to-end [`KandinskyInpaintCombinedPipeline`] and [`KandinskyV22InpaintCombinedPipeline`] to call the prior and decoder pipelines together under the hood. Use the [`AutoPipelineForInpainting`] for this:
<hfoptions id="inpaint">
<hfoption id="Kandinsky 2.1">
```py
import torch
from diffusers import AutoPipelineForInpainting
pipe = AutoPipelineForInpainting.from_pretrained("kandinsky-community/kandinsky-2-1-inpaint", torch_dtype=torch.float16)
pipe.enable_model_cpu_offload()
prompt = "a hat"
image = pipe(prompt=prompt, image=original_image, mask_image=mask).images[0]
```
</hfoption>
<hfoption id="Kandinsky 2.2">
```py
import torch
from diffusers import AutoPipelineForInpainting
pipe = AutoPipelineForInpainting.from_pretrained("kandinsky-community/kandinsky-2-2-decoder-inpaint", torch_dtype=torch.float16)
pipe.enable_model_cpu_offload()
prompt = "a hat"
image = pipe(prompt=prompt, image=original_image, mask_image=mask).images[0]
```
</hfoption>
</hfoptions>
## Interpolation
Interpolation allows you to explore the latent space between the image and text embeddings which is a cool way to see some of the prior model's intermediate outputs. Load the prior pipeline and two images you'd like to interpolate:
<hfoptions id="interpolate">
<hfoption id="Kandinsky 2.1">
```py
from diffusers import KandinskyPriorPipeline, KandinskyPipeline
from diffusers.utils import load_image
import PIL
import torch
prior_pipeline = KandinskyPriorPipeline.from_pretrained("kandinsky-community/kandinsky-2-1-prior", torch_dtype=torch.float16, use_safetensors=True).to("cuda")
img_1 = load_image("https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky/cat.png")
img_2 = load_image("https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky/starry_night.jpeg")
```
</hfoption>
<hfoption id="Kandinsky 2.2">
```py
from diffusers import KandinskyV22PriorPipeline, KandinskyV22Pipeline
from diffusers.utils import load_image
import PIL
import torch
prior_pipeline = KandinskyV22PriorPipeline.from_pretrained("kandinsky-community/kandinsky-2-2-prior", torch_dtype=torch.float16, use_safetensors=True).to("cuda")
img_1 = load_image("https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky/cat.png")
img_2 = load_image("https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky/starry_night.jpeg")
```
</hfoption>
</hfoptions>
<div class="flex gap-4">
<div>
<img class="rounded-xl" src="https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky/cat.png"/>
<figcaption class="mt-2 text-center text-sm text-gray-500">a cat</figcaption>
</div>
<div>
<img class="rounded-xl" src="https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky/starry_night.jpeg"/>
<figcaption class="mt-2 text-center text-sm text-gray-500">Van Gogh's Starry Night painting</figcaption>
</div>
</div>
Specify the text or images to interpolate, and set the weights for each text or image. Experiment with the weights to see how they affect the interpolation!
```py
images_texts = ["a cat", img1, img2]
weights = [0.3, 0.3, 0.4]
```
Call the `interpolate` function to generate the embeddings, and then pass them to the pipeline to generate the image:
<hfoptions id="interpolate">
<hfoption id="Kandinsky 2.1">
```py
# prompt can be left empty
prompt = ""
prior_out = prior_pipeline.interpolate(images_texts, weights)
pipeline = KandinskyPipeline.from_pretrained("kandinsky-community/kandinsky-2-1", torch_dtype=torch.float16, use_safetensors=True).to("cuda")
image = pipeline(prompt, **prior_out, height=768, width=768).images[0]
image
```
<div class="flex justify-center">
<img class="rounded-xl" src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/kandinsky-docs/starry_cat.png"/>
</div>
</hfoption>
<hfoption id="Kandinsky 2.2">
```py
# prompt can be left empty
prompt = ""
prior_out = prior_pipeline.interpolate(images_texts, weights)
pipeline = KandinskyV22Pipeline.from_pretrained("kandinsky-community/kandinsky-2-2-decoder", torch_dtype=torch.float16, use_safetensors=True).to("cuda")
image = pipeline(prompt, **prior_out, height=768, width=768).images[0]
image
```
<div class="flex justify-center">
<img class="rounded-xl" src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/kandinskyv22-interpolate.png"/>
</div>
</hfoption>
</hfoptions>
## ControlNet
<Tip warning={true}>
⚠️ ControlNet is only supported for Kandinsky 2.2!
</Tip>
ControlNet enables conditioning large pretrained diffusion models with additional inputs such as a depth map or edge detection. For example, you can condition Kandinsky 2.2 with a depth map so the model understands and preserves the structure of the depth image.
Let's load an image and extract it's depth map:
```py
from diffusers.utils import load_image
img = load_image(
"https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinskyv22/cat.png"
).resize((768, 768))
```
<div class="flex justify-center">
<img class="rounded-xl" src="https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinskyv22/cat.png"/>
</div>
Then you can use the `depth-estimation` [`~transformers.Pipeline`] from 🤗 Transformers to process the image and retrieve the depth map:
```py
import torch
import numpy as np
from transformers import pipeline
from diffusers.utils import load_image
def make_hint(image, depth_estimator):
image = depth_estimator(image)["depth"]
image = np.array(image)
image = image[:, :, None]
image = np.concatenate([image, image, image], axis=2)
detected_map = torch.from_numpy(image).float() / 255.0
hint = detected_map.permute(2, 0, 1)
return hint
depth_estimator = pipeline("depth-estimation")
hint = make_hint(img, depth_estimator).unsqueeze(0).half().to("cuda")
```
### Text-to-image [[controlnet-text-to-image]]
Load the prior pipeline and the [`KandinskyV22ControlnetPipeline`]:
```py
from diffusers import KandinskyV22PriorPipeline, KandinskyV22ControlnetPipeline
prior_pipeline = KandinskyV22PriorPipeline.from_pretrained(
"kandinsky-community/kandinsky-2-2-prior", torch_dtype=torch.float16, use_safetensors=True
)to("cuda")
pipeline = KandinskyV22ControlnetPipeline.from_pretrained(
"kandinsky-community/kandinsky-2-2-controlnet-depth", torch_dtype=torch.float16, use_safetensors=True
).to("cuda")
```
Generate the image embeddings from a prompt and negative prompt:
```py
prompt = "A robot, 4k photo"
negative_prior_prompt = "lowres, text, error, cropped, worst quality, low quality, jpeg artifacts, ugly, duplicate, morbid, mutilated, out of frame, extra fingers, mutated hands, poorly drawn hands, poorly drawn face, mutation, deformed, blurry, dehydrated, bad anatomy, bad proportions, extra limbs, cloned face, disfigured, gross proportions, malformed limbs, missing arms, missing legs, extra arms, extra legs, fused fingers, too many fingers, long neck, username, watermark, signature"
generator = torch.Generator(device="cuda").manual_seed(43)
image_emb, zero_image_emb = pipe_prior(
prompt=prompt, negative_prompt=negative_prior_prompt, generator=generator
).to_tuple()
```
Finally, pass the image embeddings and the depth image to the [`KandinskyV22ControlnetPipeline`] to generate an image:
```py
image = pipeline(image_embeds=image_emb, negative_image_embeds=zero_image_emb, hint=hint, num_inference_steps=50, generator=generator, height=768, width=768).images[0]
image
```
<div class="flex justify-center">
<img class="rounded-xl" src="https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinskyv22/robot_cat_text2img.png"/>
</div>
### Image-to-image [[controlnet-image-to-image]]
For image-to-image with ControlNet, you'll need to use the:
- [`KandinskyV22PriorEmb2EmbPipeline`] to generate the image embeddings from a text prompt and an image
- [`KandinskyV22ControlnetImg2ImgPipeline`] to generate an image from the initial image and the image embeddings
Process and extract a depth map of an initial image of a cat with the `depth-estimation` [`~transformers.Pipeline`] from 🤗 Transformers:
```py
import torch
import numpy as np
from diffusers import KandinskyV22PriorEmb2EmbPipeline, KandinskyV22ControlnetImg2ImgPipeline
from diffusers.utils import load_image
from transformers import pipeline
img = load_image(
"https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main" "/kandinskyv22/cat.png"
).resize((768, 768))
def make_hint(image, depth_estimator):
image = depth_estimator(image)["depth"]
image = np.array(image)
image = image[:, :, None]
image = np.concatenate([image, image, image], axis=2)
detected_map = torch.from_numpy(image).float() / 255.0
hint = detected_map.permute(2, 0, 1)
return hint
depth_estimator = pipeline("depth-estimation")
hint = make_hint(img, depth_estimator).unsqueeze(0).half().to("cuda")
```
Load the prior pipeline and the [`KandinskyV22ControlnetImg2ImgPipeline`]:
```py
prior_pipeline = KandinskyV22PriorEmb2EmbPipeline.from_pretrained(
"kandinsky-community/kandinsky-2-2-prior", torch_dtype=torch.float16, use_safetensors=True
).to("cuda")
pipeline = KandinskyV22ControlnetImg2ImgPipeline.from_pretrained(
"kandinsky-community/kandinsky-2-2-controlnet-depth", torch_dtype=torch.float16
).to("cuda")
```
Pass a text prompt and the initial image to the prior pipeline to generate the image embeddings:
```py
prompt = "A robot, 4k photo"
negative_prior_prompt = "lowres, text, error, cropped, worst quality, low quality, jpeg artifacts, ugly, duplicate, morbid, mutilated, out of frame, extra fingers, mutated hands, poorly drawn hands, poorly drawn face, mutation, deformed, blurry, dehydrated, bad anatomy, bad proportions, extra limbs, cloned face, disfigured, gross proportions, malformed limbs, missing arms, missing legs, extra arms, extra legs, fused fingers, too many fingers, long neck, username, watermark, signature"
generator = torch.Generator(device="cuda").manual_seed(43)
img_emb = pipe_prior(prompt=prompt, image=img, strength=0.85, generator=generator)
negative_emb = pipe_prior(prompt=negative_prior_prompt, image=img, strength=1, generator=generator)
```
Now you can run the [`KandinskyV22ControlnetImg2ImgPipeline`] to generate an image from the initial image and the image embeddings:
```py
image = pipeline(image=img, strength=0.5, image_embeds=img_emb.image_embeds, negative_image_embeds=negative_emb.image_embeds, hint=hint, num_inference_steps=50, generator=generator, height=768, width=768).images[0]
image
```
<div class="flex justify-center">
<img class="rounded-xl" src="https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinskyv22/robot_cat.png"/>
</div>
## Optimizations
Kandinsky is unique because it requires a prior pipeline to generate the mappings, and a second pipeline to decode the latents into an image. Optimization efforts should be focused on the second pipeline because that is where the bulk of the computation is done. Here are some tips to improve Kandinsky during inference.
1. Enable [xFormers](https://moon-ci-docs.huggingface.co/optimization/xformers) if you're using PyTorch < 2.0:
```diff
from diffusers import DiffusionPipeline
import torch
pipe = DiffusionPipeline.from_pretrained("kandinsky-community/kandinsky-2-1", torch_dtype=torch.float16)
+ pipe.enable_xformers_memory_efficient_attention()
```
2. Enable `torch.compile` if you're using PyTorch 2.0 to automatically use scaled dot-product attention (SDPA):
```diff
pipe.unet.to(memory_format=torch.channels_last)
+ pipe.unet = torch.compile(pipe.unet, mode="reduced-overhead", fullgraph=True)
pipe = DiffusionPipeline.from_pretrained("kandinsky-community/kandinsky-2-1", torch_dtype=torch.float16)
+ pipe.enable_xformers_memory_efficient_attention()
```
This is the same as explicitly setting the attention processor to use [`~models.attention_processor.AttnAddedKVProcessor2_0`]:
```py
from diffusers.models.attention_processor import AttnAddedKVProcessor2_0
pipe.unet.set_attn_processor(AttnAddedKVProcessor2_0())
```
3. Offload the model to the CPU with [`~KandinskyPriorPipeline.enable_model_cpu_offload`] to avoid out-of-memory errors:
```diff
from diffusers import DiffusionPipeline
import torch
pipe = DiffusionPipeline.from_pretrained("kandinsky-community/kandinsky-2-1", torch_dtype=torch.float16)
+ pipe.enable_model_cpu_offload()
```
4. By default, the text-to-image pipeline uses the [`DDIMScheduler`] but you can replace it with another scheduler like [`DDPMScheduler`] to see how that affects the tradeoff between inference speed and image quality:
```py
from diffusers import DDPMSCheduler
scheduler = DDPMScheduler.from_pretrained("kandinsky-community/kandinsky-2-1", subfolder="ddpm_scheduler")
pipe = DiffusionPipeline.from_pretrained("kandinsky-community/kandinsky-2-1", scheduler=scheduler, torch_dtype=torch.float16, use_safetensors=True).to("cuda")
```
\ No newline at end of file
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment