JoseBambora
/

mistral_retrained

Generated from Trainer

Model card Files Files and versions Metrics Training metrics Community

JoseBambora commited on May 7

Commit

fd7516b

•

1 Parent(s): 34d7c58

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -24,7 +24,7 @@ It achieves an accuracy of 96.4% for 2500 test cases. This test cases were insid
 The dataset used for training only had 900 cases from the training part of the original dataset, so that the training phase did not take too much time.
 It is important to mention that the data for train is totally balanced (450 positive reviews, 450 negative reviews).
-The evaluation data is just 100 cases from the training part of the original dataset.
 ## Training procedure

 The dataset used for training only had 900 cases from the training part of the original dataset, so that the training phase did not take too much time.
 It is important to mention that the data for train is totally balanced (450 positive reviews, 450 negative reviews).
+The evaluation data is just 100 different cases from the training part of the original dataset.
 ## Training procedure