Update app.py

app.py

@@ -61,10 +61,6 @@ tokenizer.pad_token = tokenizer.unk_token
 
 feature_extractor = ViTImageProcessor.from_pretrained(encoder_name_or_path)
 
-# cap_model = VisionEncoderDecoderModel.from_pretrained("model")
-
-# tokenizer = AutoTokenizer.from_pretrained("model")
-
 
 
 
@@ -98,16 +94,8 @@ device = "cuda" if torch.cuda.is_available() else "cpu"
 
 def predict_event(image):
     
-
-    # img =  Image.open(image).convert("RGB")
-
     generated_caption = tokenizer.decode(model.generate(feature_extractor(image, return_tensors="pt").pixel_values.to(device))[0])
 
-    # caption_vitgpt = generate_caption(model, image)
-    #caption_vitgpt = generate_caption(feature_extractor, cap_model, image, tokenizer)
-    # preds = [pred.strip() for pred in generated_caption]
-    # return preds
-
     return '\033[96m' +generated_caption+ '\033[0m'
 
 
@@ -131,9 +119,10 @@ iface = gr.Interface(predict_event,
                          # gr.Image(type="pil"),
                          outputs=["text"] )
                         
-    
 
 iface.launch()
+
+
 # app = gr.mount_gradio_app(app, iface, path="/")
 
 # iface.launch(server_name = "0.0.0.0", server_port = 8001,share=True)