Update app.py

app.py

@@ -2,7 +2,7 @@ import torch
 import re
 import gradio as gr
 from pathlib import Path
-from transformers import AutoTokenizer, AutoFeatureExtractor, VisionEncoderDecoderModel
+from transformers import GPT2Tokenizer, AutoFeatureExtractor, VisionEncoderDecoderModel
 
 def predict(image, max_length=64, num_beams=4):
     image = image.convert('RGB')
@@ -21,7 +21,7 @@ model.to(device)
 print("Loaded model")
 feature_extractor = AutoFeatureExtractor.from_pretrained("google/vit-base-patch16-224-in21k")
 print("Loaded feature_extractor")
-tokenizer = AutoTokenizer.from_pretrained(model_path)
+tokenizer = GPT2Tokenizer.from_pretrained(model_path)
 print("Loaded tokenizer")
 title = "Hindi Image Captioning"
 description = ""