Update app.py

app.py

@@ -1,30 +1,35 @@
-import torch
-import re
+import torch 
+import re 
 import gradio as gr
-from transformers import AutoTokenizer,ViTFeatureExtractor,VisionEncoderDecoderModel
+from transformers import AutoTokenizer, ViTFeatureExtractor, VisionEncoderDecoderModel 
 
+    
 device = 'cpu'
 encoder_checkpoint = 'google/vit-base-patch16-224'
-decoder_checkpoint = 'gpt2'
-model_checkpoint = 'nlpconnect/vit-gpt2-image-captioning'
+decoder_checkpoint = 'surajp/gpt2-hindi'
+model_checkpoint = 'team-indain-image-caption/hindi-image-captioning'
 feature_extractor = ViTFeatureExtractor.from_pretrained(encoder_checkpoint)
+tokenizer = AutoTokenizer.from_pretrained(decoder_checkpoint)
 model = VisionEncoderDecoderModel.from_pretrained(model_checkpoint).to(device)
 
 
-def predict(image,max_length=64,num_beams=4):
+
+def predict(image,max_length=64, num_beams=4):
   image = image.convert('RGB')
-  image = feature_extractor(image,return_tensor='pt').pixel_values.to(device)
-  clean_text = lambda x: x.replace('<|endoftext|>','').split('\n')[0]  
+  image = feature_extractor(image, return_tensors="pt").pixel_values.to(device)
+  clean_text = lambda x: x.replace('<|endoftext|>','').split('\n')[0]
   caption_ids = model.generate(image, max_length = max_length)[0]
   caption_text = clean_text(tokenizer.decode(caption_ids))
-  return caption_text
-  
+  return caption_text 
+
+
+
+input = gr.inputs.Image(label="Image to search", type = 'pil', optional=False)
+output = gr.outputs.Textbox(type="auto",label="Captions")
 
-input = gr.inputs.Image(label='Image to generate caption',type = 'pil', optional=False)
-output = gr.outputs.Textbox(type="auto",label="Caption")
 
-article = "This is an Image captioning application created by Shreyas Dixit"
-title = "Image Captioning"
+article = "This HuggingFace Space presents a demo for Image captioning in Hindi built with VIT Encoder and GPT2 Decoder"
+title = "Hindi Image Captioning System"
 
 interface = gr.Interface(
         fn=predict,