From fb84c9548f8e4e15d2bece1d2d7834b31e0f7c2f Mon Sep 17 00:00:00 2001
From: hw <45089338+MorningForest@users.noreply.github.com>
Date: Wed, 26 Jul 2023 03:04:56 +0800
Subject: [PATCH] fix web_demo cache accelerate (#133)

---
 web_demo.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/web_demo.py b/web_demo.py
index bc9307c..f07bfdd 100644
--- a/web_demo.py
+++ b/web_demo.py
@@ -199,7 +199,7 @@ def combine_history(prompt):
 
 
 def main():
-    torch.cuda.empty_cache()
+    #torch.cuda.empty_cache()
     print("load model begin.")
     model, tokenizer = load_model()
     print("load model end.")
@@ -237,6 +237,7 @@ def main():
             message_placeholder.markdown(cur_response)
         # Add robot response to chat history
         st.session_state.messages.append({"role": "robot", "content": cur_response, "avatar": robot_avator})
+        torch.cuda.empty_cache()
         
 
 if __name__ == "__main__":