Update app.py

app.py

@@ -120,7 +120,6 @@ def generate_caption_llava(image_path, caption_bool):
     
     inputs = processor(prompt, Image.open(image_path), return_tensors="pt").to(device)
 
-    # autoregressively complete prompt
     output = model.generate(**inputs, max_new_tokens=100)
 
     return processor.decode(output[0], skip_special_tokens=True)["generated_text"][len(text_prompt):]
@@ -128,7 +127,7 @@ def generate_caption_llava(image_path, caption_bool):
 @spaces.GPU
 def generate_answer_llava(image_path, question):
     text_prompt =f"[INST]  \n{question} [/INST]"
-    inputs = processor(prompt, Image.open(image_path), return_tensors="pt").to(device)
+    inputs = processor(text_prompt, Image.open(image_path), return_tensors="pt").to(device)
     output = model.generate(**inputs, max_new_tokens=100)
     return processor.decode(output[0], skip_special_tokens=True)["generated_text"][len(text_prompt):]