:memo: new. model metrics update

README.md

@@ -58,16 +58,16 @@ We used heuristic parameters without separate hyperparameter tuning.
 ## Metrics
 
 For the dev set we got those results
-- size of the set: 700 examples from French Census  / 1200 from our own dataset
-- CER: 0.0636
-- WER: 0.1823
-- Loss: 0.6386
+- size of the set: 700 examples from French Census  / 1600 from our own dataset
+- CER: 0.0575
+- WER: 0.1651
+- Loss: 0.5768
 
 For the test set we got those results
 - size of the set: 730 examples from French Census  / 950 from our own dataset
-- CER: 0.09012
-- WER: 0.20446
-- Loss: 0.8547
+- CER: 0.09417
+- WER: 0.23485
+- Loss: 0.8700
 
 ### How to use