Update app.py

app.py

@@ -39,15 +39,26 @@ else:
 
 @spaces.GPU(duration=150)
 def chat(message, history, temperature,do_sample, max_tokens):
+    prompt_template = """
+    You are a helpful Agricultural assistant for farmers. You are given the following input. Please complete the response briefly.
+    ## Question:
+    {}
+    
+    ## Response:
+    {}"""
     start_time = time.time()
     chat = []
-    for item in history:
-        chat.append({"role": "user", "content": item[0]})
-        if item[1] is not None:
-            chat.append({"role": "assistant", "content": item[1]})
-    chat.append({"role": "user", "content": message})
-    messages = tok.apply_chat_template(chat, tokenize=False, add_generation_prompt=True)
-    model_inputs = tok([messages], return_tensors="pt").to(device)
+    # for item in history:
+    #     chat.append({"role": "user", "content": item[0]})
+    #     if item[1] is not None:
+    #         chat.append({"role": "assistant", "content": item[1]})
+    # chat.append({"role": "user", "content": message})
+    # messages = tok.apply_chat_template(chat, tokenize=False, add_generation_prompt=True)
+    
+    model_inputs = tok(prompt_template.format(
+        message, #input
+        "" # response
+    ), return_tensors="pt").to(device)
     streamer = TextIteratorStreamer(
         tok, timeout=10.0, skip_prompt=True, skip_special_tokens=True
     )