VersatileDiffusion: fix input processing (huggingface#1568)

* fix versatile diffusion input

* merge main

* `make fix-copies`

Co-authored-by: anton- <[email protected]>

Author: LukasStruppek

src/diffusers/pipelines/paint_by_example/pipeline_paint_by_example.py

@@ -271,7 +271,8 @@ def check_inputs(self, image, height, width, callback_steps):
             and not isinstance(image, list)
         ):
             raise ValueError(
-                f"`image` has to be of type `torch.FloatTensor` or `PIL.Image.Image` or `list` but is {type(image)}"
+                "`image` has to be of type `torch.FloatTensor` or `PIL.Image.Image` or `List[PIL.Image.Image]` but is"
+                f" {type(image)}"
             )
 
         if height % 8 != 0 or width % 8 != 0:

src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion_image_variation.py

@@ -240,7 +240,8 @@ def check_inputs(self, image, height, width, callback_steps):
             and not isinstance(image, list)
         ):
             raise ValueError(
-                f"`image` has to be of type `torch.FloatTensor` or `PIL.Image.Image` or `list` but is {type(image)}"
+                "`image` has to be of type `torch.FloatTensor` or `PIL.Image.Image` or `List[PIL.Image.Image]` but is"
+                f" {type(image)}"
             )
 
         if height % 8 != 0 or width % 8 != 0:

src/diffusers/pipelines/versatile_diffusion/pipeline_versatile_diffusion_image_variation.py

@@ -134,6 +134,9 @@ def normalize_embeddings(encoder_output):
             embeds = embeds / torch.norm(embeds_pooled, dim=-1, keepdim=True)
             return embeds
 
+        if isinstance(prompt, torch.Tensor) and len(prompt.shape) == 4:
+            prompt = [p for p in prompt]
+
         batch_size = len(prompt) if isinstance(prompt, list) else 1
 
         # get prompt text embeddings
@@ -212,9 +215,17 @@ def prepare_extra_step_kwargs(self, generator, eta):
             extra_step_kwargs["generator"] = generator
         return extra_step_kwargs
 
+    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion_image_variation.StableDiffusionImageVariationPipeline.check_inputs
     def check_inputs(self, image, height, width, callback_steps):
-        if not isinstance(image, PIL.Image.Image) and not isinstance(image, torch.Tensor):
-            raise ValueError(f"`image` has to be of type `PIL.Image.Image` or `torch.Tensor` but is {type(image)}")
+        if (
+            not isinstance(image, torch.Tensor)
+            and not isinstance(image, PIL.Image.Image)
+            and not isinstance(image, list)
+        ):
+            raise ValueError(
+                "`image` has to be of type `torch.FloatTensor` or `PIL.Image.Image` or `List[PIL.Image.Image]` but is"
+                f" {type(image)}"
+            )
 
         if height % 8 != 0 or width % 8 != 0:
             raise ValueError(f"`height` and `width` have to be divisible by 8 but are {height} and {width}.")