add StableDiffusionXLKDiffusionPipeline (huggingface#6447)

---------

Co-authored-by: yiyixuxu <yixu310@gmail,com>
Co-authored-by: Patrick von Platen <[email protected]>

docs/source/en/_toctree.yml

@@ -333,6 +333,8 @@
         title: Latent upscaler
       - local: api/pipelines/stable_diffusion/upscale
         title: Super-resolution
+      - local: api/pipelines/stable_diffusion/k_diffusion
+        title: K-Diffusion
       - local: api/pipelines/stable_diffusion/ldm3d_diffusion
         title: LDM3D Text-to-(RGB, Depth), Text-to-(RGB-pano, Depth-pano), LDM3D Upscaler
       - local: api/pipelines/stable_diffusion/adapter

docs/source/en/api/pipelines/stable_diffusion/k_diffusion.md

@@ -0,0 +1,27 @@
+<!--Copyright 2023 The HuggingFace Team. All rights reserved.
+
+Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
+the License. You may obtain a copy of the License at
+
+http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
+an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
+specific language governing permissions and limitations under the License.
+-->
+
+# K-Diffusion
+
+[k-diffusion](https://github.com/crowsonkb/k-diffusion) is a popular library created by [Katherine Crowson](https://github.com/crowsonkb/). We provide `StableDiffusionKDiffusionPipeline` and `StableDiffusionXLKDiffusionPipeline` that allow you to run Stable DIffusion with samplers from k-diffusion. 
+
+Note that most the samplers from k-diffusion are implemented in Diffusers and we recommend using existing schedulers. You can find a mapping between k-diffusion samplers and schedulers in Diffusers [here](https://huggingface.co/docs/diffusers/api/schedulers/overview)
+
+
+## StableDiffusionKDiffusionPipeline
+
+[[autodoc]] StableDiffusionKDiffusionPipeline
+
+
+## StableDiffusionXLKDiffusionPipeline
+
+[[autodoc]] StableDiffusionXLKDiffusionPipeline

src/diffusers/__init__.py

@@ -316,7 +316,7 @@
     ]
 
 else:
-    _import_structure["pipelines"].extend(["StableDiffusionKDiffusionPipeline"])
+    _import_structure["pipelines"].extend(["StableDiffusionKDiffusionPipeline", "StableDiffusionXLKDiffusionPipeline"])
 
 try:
     if not (is_torch_available() and is_transformers_available() and is_onnx_available()):
@@ -668,7 +668,7 @@
     except OptionalDependencyNotAvailable:
         from .utils.dummy_torch_and_transformers_and_k_diffusion_objects import *  # noqa F403
     else:
-        from .pipelines import StableDiffusionKDiffusionPipeline
+        from .pipelines import StableDiffusionKDiffusionPipeline, StableDiffusionXLKDiffusionPipeline
 
     try:
         if not (is_torch_available() and is_transformers_available() and is_onnx_available()):

src/diffusers/pipelines/__init__.py

@@ -265,7 +265,10 @@
 
     _dummy_objects.update(get_objects_from_module(dummy_torch_and_transformers_and_k_diffusion_objects))
 else:
-    _import_structure["stable_diffusion_k_diffusion"] = ["StableDiffusionKDiffusionPipeline"]
+    _import_structure["stable_diffusion_k_diffusion"] = [
+        "StableDiffusionKDiffusionPipeline",
+        "StableDiffusionXLKDiffusionPipeline",
+    ]
 try:
     if not is_flax_available():
         raise OptionalDependencyNotAvailable()
@@ -491,7 +494,10 @@
         except OptionalDependencyNotAvailable:
             from ..utils.dummy_torch_and_transformers_and_k_diffusion_objects import *
         else:
-            from .stable_diffusion_k_diffusion import StableDiffusionKDiffusionPipeline
+            from .stable_diffusion_k_diffusion import (
+                StableDiffusionKDiffusionPipeline,
+                StableDiffusionXLKDiffusionPipeline,
+            )
 
         try:
             if not is_flax_available():

src/diffusers/pipelines/stable_diffusion_k_diffusion/__init__.py

@@ -30,6 +30,7 @@
     _dummy_objects.update(get_objects_from_module(dummy_torch_and_transformers_and_k_diffusion_objects))
 else:
     _import_structure["pipeline_stable_diffusion_k_diffusion"] = ["StableDiffusionKDiffusionPipeline"]
+    _import_structure["pipeline_stable_diffusion_xl_k_diffusion"] = ["StableDiffusionXLKDiffusionPipeline"]
 
 if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
     try:
@@ -45,6 +46,7 @@
         from ...utils.dummy_torch_and_transformers_and_k_diffusion_objects import *
     else:
         from .pipeline_stable_diffusion_k_diffusion import StableDiffusionKDiffusionPipeline
+        from .pipeline_stable_diffusion_xl_k_diffusion import StableDiffusionXLKDiffusionPipeline
 
 else:
     import sys

src/diffusers/pipelines/stable_diffusion_k_diffusion/pipeline_stable_diffusion_k_diffusion.py

@@ -134,7 +134,15 @@ def __init__(
     def set_scheduler(self, scheduler_type: str):
         library = importlib.import_module("k_diffusion")
         sampling = getattr(library, "sampling")
-        self.sampler = getattr(sampling, scheduler_type)
+        try:
+            self.sampler = getattr(sampling, scheduler_type)
+        except Exception:
+            valid_samplers = []
+            for s in dir(sampling):
+                if "sample_" in s:
+                    valid_samplers.append(s)
+
+            raise ValueError(f"Invalid scheduler type {scheduler_type}. Please choose one of {valid_samplers}.")
 
     # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline._encode_prompt
     def _encode_prompt(