update model card README.md

README.md

@@ -19,12 +19,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [neuralsentry/starencoder-git-commits-mlm](https://huggingface.co/neuralsentry/starencoder-git-commits-mlm) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1191
-- Accuracy: 0.9703
-- Precision: 0.9769
-- Recall: 0.96
-- F1: 0.9684
-- Roc Auc: 0.9698
+- Loss: 0.2327
+- Accuracy: 0.9274
+- Precision: 0.9691
+- Recall: 0.9199
+- F1: 0.9438
+- Roc Auc: 0.9310
 
 ## Model description
 
@@ -55,15 +55,7 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1     | Roc Auc |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|:-------:|
-| 0.3716        | 0.33  | 66   | 0.2018          | 0.9296   | 0.9368    | 0.9133 | 0.9249 | 0.9288  |
-| 0.1745        | 0.67  | 132  | 0.1468          | 0.9533   | 0.9711    | 0.9293 | 0.9498 | 0.9522  |
-| 0.1346        | 1.0   | 198  | 0.1091          | 0.9657   | 0.9761    | 0.951  | 0.9634 | 0.9650  |
-| 0.0917        | 1.33  | 264  | 0.1294          | 0.9647   | 0.9790    | 0.946  | 0.9622 | 0.9638  |
-| 0.0877        | 1.67  | 330  | 0.1090          | 0.9668   | 0.9619    | 0.9683 | 0.9651 | 0.9669  |
-| 0.0731        | 2.0   | 396  | 0.1042          | 0.9688   | 0.9746    | 0.9593 | 0.9669 | 0.9684  |
-| 0.0342        | 2.33  | 462  | 0.1291          | 0.9692   | 0.9686    | 0.9663 | 0.9675 | 0.9690  |
-| 0.0375        | 2.67  | 528  | 0.1202          | 0.9706   | 0.9753    | 0.9623 | 0.9688 | 0.9702  |
-| 0.0342        | 3.0   | 594  | 0.1191          | 0.9703   | 0.9769    | 0.96   | 0.9684 | 0.9698  |
+| 0.2872        | 1.57  | 66   | 0.2017          | 0.9177   | 0.9439    | 0.9312 | 0.9375 | 0.9111  |
 
 
 ### Framework versions