Update app.py

app.py

@@ -26,33 +26,33 @@ model = AutoModelForCausalLM.from_pretrained(model_name,
                                              device_map="auto")
 model.eval()
 
-def format_history(msg: str, history: list[list[str, str]], system_prompt: str):
-    chat_history = system_prompt
-    for query, response in history:
-        chat_history += f"\nUser: {query}\nAssistant: {response}"
-    chat_history += f"\nUser: {msg}\nAssistant:"
-    return chat_history
-
 @spaces.GPU(duration=90)
-def generate(msg: str, 
-            history: list[list[str, str]], 
-            system_prompt: str,
-            max_new_tokens: int = 1024,
-            temperature: float = 0.6,
-            top_p: float = 0.9,
-            top_k: int = 50,
-            repetition_penalty: float = 1.2,) -> Iterator[str]:
-    chat_history = format_history(msg, history, system_prompt)
-    
-    # Tokenize the input prompt
-    input_ids = tokenizer(chat_history, return_tensors="pt").to("cuda")
-    streamer = TextIteratorStreamer(tokenizer, timeout=20.0, skip_prompt=True, skip_special_tokens=True)
-    
-    # Generate a response using the model
-    # outputs = model.generate(inputs["input_ids"], max_length=500, pad_token_id=tokenizer.eos_token_id)
+def generate(
+    message: str,
+    chat_history: list[tuple[str, str]],
+    max_new_tokens: int = 1024,
+    temperature: float = 0.6,
+    top_p: float = 0.9,
+    top_k: int = 50,
+    repetition_penalty: float = 1.2,
+) -> Iterator[str]:
+    conversation = []
+    for user, assistant in chat_history:
+        conversation.extend(
+            [
+                {"role": "user", "content": user},
+                {"role": "assistant", "content": assistant},
+            ]
+        )
+    conversation.append({"role": "user", "content": message})
+
+    input_ids = tokenizer.apply_chat_template(conversation, add_generation_prompt=True, return_tensors="pt")
+    if input_ids.shape[1] > MAX_INPUT_TOKEN_LENGTH:
+        input_ids = input_ids[:, -MAX_INPUT_TOKEN_LENGTH:]
+        gr.Warning(f"Trimmed input from conversation as it was longer than {MAX_INPUT_TOKEN_LENGTH} tokens.")
+    input_ids = input_ids.to(model.device)
 
-    # Decode the response back to a string
-    # response = tokenizer.decode(outputs[:, inputs["input_ids"].shape[-1]:][0], skip_special_tokens=True)
+    streamer = TextIteratorStreamer(tokenizer, timeout=20.0, skip_prompt=True, skip_special_tokens=True)
     generate_kwargs = dict(
         {"input_ids": input_ids},
         streamer=streamer,
@@ -66,9 +66,7 @@ def generate(msg: str,
     )
     t = Thread(target=model.generate, kwargs=generate_kwargs)
     t.start()
-    
-    # Yield the generated response
-    #yield response
+
     outputs = []
     for text in streamer:
         outputs.append(text)