Update app.py

app.py

@@ -301,7 +301,9 @@ def generateOutput(image_path, audio_path, context_text, question, max_length=3)
         f"\n<|question|>: {question} \n<|end|>\n<|assistant|>\n"
     )
     print("input_text:", input_text)
-    start_tokens = tokenizer(input_text, padding=True, truncation=True, max_length=1024, return_tensors="pt")['input_ids'].to(device)
+    tokens = tokenizer(input_text, padding=True, truncation=True, max_length=1024, return_tensors="pt")
+    start_tokens = tokens['input_ids'].to(device)
+    # start_tokens = tokenizer(input_text, padding=True, truncation=True, max_length=1024, return_tensors="pt")['input_ids'].to(device)
     attention_mask = tokens['attention_mask'].to(device)
     # base_phi_model.generate(start_tokens, max_length=2, do_sample=False, pad_token_id=tokenizer.pad_token_id)