benchmark update

demo.py

@@ -252,11 +252,11 @@ with st.expander("See more explanation"):
 
     """)
 
-st.info('''Updated on Dec. 4th, 2022\n
-The current version was trained on 1,333 sentences and tested on 323 sentences. It achieved the following benchmark:
-- Micro F1 = 68.94 
-- Micro Precision = 71.17
-- Micro Recall = 66.84
+st.info('''Updated on Jan. 4th, 2023\n
+The current version was trained on 2,430 sentences and tested on 431 sentences. It achieved the following benchmark:
+- Micro F1 = .75 
+- Micro Precision = .74
+- Micro Recall = .75
 I expect that the model's performance improves as the annotated dataset gets larger.
 ''')