Update app.py

app.py

@@ -44,7 +44,10 @@ def respond(
             messages.append({"role": "assistant", "content": val[1]})
 
     if image:
-        # Process image if provided
+        # Load and process the image
+        if isinstance(image, str):
+            image = Image.open(image)
+
         image_tensor = process_images([image], image_processor, model.config)
         image_tensor = [_image.to(dtype=torch.float16, device=device) for _image in image_tensor]
 
@@ -71,8 +74,23 @@ def respond(
     else:
         messages.append({"role": "user", "content": message})
 
-        # Simulating a response as `client.chat_completion` is not defined in this script
-        response = "Simulated response based on input: " + message
+        conv_template = "qwen_1_5"
+        conv = copy.deepcopy(conv_templates[conv_template])
+        conv.append_message(conv.roles[0], message)
+        conv.append_message(conv.roles[1], None)
+        prompt_question = conv.get_prompt()
+
+        input_ids = tokenizer(prompt_question, return_tensors="pt", max_length=max_tokens, truncation=True).to(device)
+
+        cont = model.generate(
+            input_ids,
+            do_sample=True,
+            temperature=temperature,
+            max_new_tokens=max_tokens,
+            top_p=top_p,
+        )
+
+        response = tokenizer.batch_decode(cont, skip_special_tokens=True)[0]
 
     yield response