Update app.py

app.py

@@ -39,11 +39,11 @@ import torch
 #     f1_metric.set(f1)
 
 feature_extractor = ViTImageProcessor.from_pretrained("model")
-print("feature_extractor--",feature_extractor)
+
 cap_model = VisionEncoderDecoderModel.from_pretrained("model")
-print("cap_model--",cap_model)
+
 tokenizer = AutoTokenizer.from_pretrained("model")
-print("tokenizer--",tokenizer)
+print("tokenizer --",tokenizer)
 
 device = "cuda" if torch.cuda.is_available() else "cpu"
 
@@ -63,12 +63,16 @@ def generate_caption(processor, model, image, tokenizer=None):
   # preds = [pred.strip() for pred in preds]
   # return preds
     inputs = processor(images=image, return_tensors="pt").to(device)
+    print("inputs",inputs)
     
     generated_ids = model.generate(pixel_values=inputs.pixel_values)
+    print("generated_ids",generated_ids)
 
     if tokenizer is not None:
+        print("tokenizer not null--",tokenizer)
         generated_caption = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
     else:
+        print("tokenizer null--",tokenizer)
         generated_caption = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]
    
     return generated_caption