Update app.py

app.py

@@ -59,7 +59,7 @@ Age=""
 chat_log_name =""
 
 from llama_cpp import Llama
-llm = Llama(model_path=model_file, model_type="mistral",n_ctx = 2048)
+llm = Llama(model_path=model_file, model_type="mistral",n_gpu_layers=-1,n_ctx = 2048)
 
 def get_drive_service():
     credentials = service_account.Credentials.from_service_account_file(
@@ -146,17 +146,19 @@ def generate(prompt, history):
 
         response = ""
 
-#        while(len(response) < 1):
- #           output = llm(context, max_tokens=400, stop=["Nurse:"], echo=False)
- #           response = output["choices"][0]["text"]
-  #          response = response.strip()
+        while(len(response) < 1):
+            output = llm(context, max_tokens=400, stop=["Nurse:"], echo=False)
+            response = output["choices"][0]["text"]
+            response = response.strip()
+            yield  response
 
-        for output in llm(input, stream=True, max_tokens=100, ):
-            piece = output['choices'][0]['text']
-            response += piece
-            chatbot[-1] = (chatbot[-1][0], response)
 
-            yield  response
+      #  for output in llm(input, stream=True, max_tokens=100, ):
+        #    piece = output['choices'][0]['text']
+       #     response += piece
+       #     chatbot[-1] = (chatbot[-1][0], response)
+
+       #     yield  response
 
         context += response
         print (context)