Update app.py

app.py

@@ -28,17 +28,10 @@ def generater(message, history, temperature, top_p, top_k):
         prompt += model.config["promptTemplate"].format(user_message)
         prompt += assistant_message + "</s>"
     prompt += model.config["promptTemplate"].format(message)
-
-    try:
-        generated_tokens = model.generate(prompt=prompt, temp=temperature, top_k=top_k, top_p=top_p, max_tokens=max_new_tokens, streaming=True)
-        
-        # Convert the generated tokens to a string
-        generated_text = model.decode(generated_tokens, skip_special_tokens=True)[0]
-        return generated_text
-
-    except Exception as e:
-        print("Error during generation:", str(e))
-        return "An error occurred during generation."
+    outputs = []    
+    for token in model.generate(prompt=prompt, temp=temperature, top_k = top_k, top_p = top_p, max_tokens = max_new_tokens, streaming=True):
+        outputs.append(token)
+        yield "".join(outputs)