Update image_summary.py

image_summary.py

@@ -3,10 +3,17 @@ import gradio as gr
 from PIL import Image
 import requests
 
-from transformers import ViTFeatureExtractor
-feature_extractor = ViTFeatureExtractor()
+# from transformers import ViTFeatureExtractor
+# feature_extractor = ViTFeatureExtractor()
+from transformers import ViTImageProcessor
+feature_extractor = ViTImageProcessor()
+
+
 # or, to load one that corresponds to a checkpoint on the hub:
-feature_extractor = ViTFeatureExtractor.from_pretrained("google/vit-base-patch16-224")
+# feature_extractor = ViTFeatureExtractor.from_pretrained("google/vit-base-patch16-224")
+feature_extractor = ViTImageProcessor.from_pretrained("google/vit-base-patch16-224")
+
+
 
 from transformers import VisionEncoderDecoderModel
 # initialize a vit-bert from a pretrained ViT and a pretrained BERT model. Note that the cross-attention layers will be randomly initialized
@@ -21,7 +28,10 @@ model = VisionEncoderDecoderModel.from_pretrained("./vit-bert")
 #####################
 from transformers import AutoTokenizer
 repo_name = "ydshieh/vit-gpt2-coco-en"
-feature_extractor = ViTFeatureExtractor.from_pretrained(repo_name)
+# feature_extractor = ViTFeatureExtractor.from_pretrained(repo_name)
+feature_extractor = ViTImageProcessor.from_pretrained(repo_name)
+
+
 tokenizer = AutoTokenizer.from_pretrained(repo_name)
 model = VisionEncoderDecoderModel.from_pretrained(repo_name)