Rename 'CLIPFeatureExtractor' class to 'CLIPImageProcessor' (#2732)

The 'CLIPFeatureExtractor' class name has been renamed to 'CLIPImageProcessor' in order to comply with future deprecation. This commit includes the necessary changes to the affected files.

src/diffusers/pipelines/versatile_diffusion/pipeline_versatile_diffusion_image_variation.py

@@ -19,7 +19,7 @@ import numpy as np
 import PIL
 import torch
 import torch.utils.checkpoint
-from transformers import CLIPFeatureExtractor, CLIPVisionModelWithProjection
+from transformers import CLIPImageProcessor, CLIPVisionModelWithProjection
 
 from ...models import AutoencoderKL, UNet2DConditionModel
 from ...schedulers import KarrasDiffusionSchedulers
@@ -48,7 +48,7 @@ class VersatileDiffusionImageVariationPipeline(DiffusionPipeline):
             A scheduler to be used in combination with `unet` to denoise the encoded image latents. Can be one of
             [`DDIMScheduler`], [`LMSDiscreteScheduler`], or [`PNDMScheduler`].
     """
-    image_feature_extractor: CLIPFeatureExtractor
+    image_feature_extractor: CLIPImageProcessor
     image_encoder: CLIPVisionModelWithProjection
     image_unet: UNet2DConditionModel
     vae: AutoencoderKL
@@ -56,7 +56,7 @@ class VersatileDiffusionImageVariationPipeline(DiffusionPipeline):
 
     def __init__(
         self,
-        image_feature_extractor: CLIPFeatureExtractor,
+        image_feature_extractor: CLIPImageProcessor,
         image_encoder: CLIPVisionModelWithProjection,
         image_unet: UNet2DConditionModel,
         vae: AutoencoderKL,

src/diffusers/pipelines/versatile_diffusion/pipeline_versatile_diffusion_text_to_image.py

@@ -17,7 +17,7 @@ from typing import Callable, List, Optional, Union
 
 import torch
 import torch.utils.checkpoint
-from transformers import CLIPFeatureExtractor, CLIPTextModelWithProjection, CLIPTokenizer
+from transformers import CLIPImageProcessor, CLIPTextModelWithProjection, CLIPTokenizer
 
 from ...models import AutoencoderKL, Transformer2DModel, UNet2DConditionModel
 from ...schedulers import KarrasDiffusionSchedulers
@@ -48,7 +48,7 @@ class VersatileDiffusionTextToImagePipeline(DiffusionPipeline):
             [`DDIMScheduler`], [`LMSDiscreteScheduler`], or [`PNDMScheduler`].
     """
     tokenizer: CLIPTokenizer
-    image_feature_extractor: CLIPFeatureExtractor
+    image_feature_extractor: CLIPImageProcessor
     text_encoder: CLIPTextModelWithProjection
     image_unet: UNet2DConditionModel
     text_unet: UNetFlatConditionModel

tests/pipelines/stable_unclip/test_stable_unclip_img2img.py

@@ -4,7 +4,7 @@ import unittest
 
 import torch
 from transformers import (
-    CLIPFeatureExtractor,
+    CLIPImageProcessor,
     CLIPTextConfig,
     CLIPTextModel,
     CLIPTokenizer,
@@ -36,7 +36,7 @@ class StableUnCLIPImg2ImgPipelineFastTests(PipelineTesterMixin, unittest.TestCas
 
         # image encoding components
 
-        feature_extractor = CLIPFeatureExtractor(crop_size=32, size=32)
+        feature_extractor = CLIPImageProcessor(crop_size=32, size=32)
 
         image_encoder = CLIPVisionModelWithProjection(
             CLIPVisionConfig(

tests/test_pipelines.py

@@ -31,7 +31,7 @@ import torch
 from parameterized import parameterized
 from PIL import Image
 from requests.exceptions import HTTPError
-from transformers import CLIPFeatureExtractor, CLIPModel, CLIPTextConfig, CLIPTextModel, CLIPTokenizer
+from transformers import CLIPImageProcessor, CLIPModel, CLIPTextConfig, CLIPTextModel, CLIPTokenizer
 
 from diffusers import (
     AutoencoderKL,
@@ -433,7 +433,7 @@ class CustomPipelineTests(unittest.TestCase):
     def test_download_from_git(self):
         clip_model_id = "laion/CLIP-ViT-B-32-laion2B-s34B-b79K"
 
-        feature_extractor = CLIPFeatureExtractor.from_pretrained(clip_model_id)
+        feature_extractor = CLIPImageProcessor.from_pretrained(clip_model_id)
         clip_model = CLIPModel.from_pretrained(clip_model_id, torch_dtype=torch.float16)
 
         pipeline = DiffusionPipeline.from_pretrained(