Update app.py

app.py

@@ -35,13 +35,7 @@ def generate_caption(image_processor, model, tokenizer,device, image):
     model.eval()
     # Generate caption
     with torch.no_grad():
-      output = model.generate(
-          pixel_values=inputs ,
-          max_length=1000,# Adjust the maximum length of the generated caption as needed
-          num_beams=4,    # Adjust the number of beams for beam search decoding
-          early_stopping=True , # Enable early stopping to stop generation when all beams finished
-          batch_size=1
-      )
+      output = model.generate(pixel_values=inputs)
       # Decode the generated caption
       caption = tokenizer.decode(output[0], skip_special_tokens=True)
     return caption