Upload TFBertForMaskedLM

README.md

@@ -14,8 +14,8 @@ probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [microsoft/graphcodebert-base](https://huggingface.co/microsoft/graphcodebert-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Train Loss: 0.2343
-- Train Accuracy: 0.9539
+- Train Loss: 0.1757
+- Train Accuracy: 0.9637
 - Epoch: 9
 
 ## Model description
@@ -42,16 +42,16 @@ The following hyperparameters were used during training:
 
 | Train Loss | Train Accuracy | Epoch |
 |:----------:|:--------------:|:-----:|
-| 0.9079     | 0.8991         | 0     |
-| 0.5494     | 0.9291         | 1     |
-| 0.5046     | 0.9297         | 2     |
-| 0.4773     | 0.9309         | 3     |
-| 0.4474     | 0.9329         | 4     |
-| 0.4100     | 0.9352         | 5     |
-| 0.3676     | 0.9384         | 6     |
-| 0.3226     | 0.9430         | 7     |
-| 0.2766     | 0.9482         | 8     |
-| 0.2343     | 0.9539         | 9     |
+| 1.0763     | 0.8934         | 0     |
+| 0.5565     | 0.9292         | 1     |
+| 0.5056     | 0.9298         | 2     |
+| 0.4761     | 0.9317         | 3     |
+| 0.4387     | 0.9346         | 4     |
+| 0.3948     | 0.9375         | 5     |
+| 0.3434     | 0.9418         | 6     |
+| 0.2872     | 0.9474         | 7     |
+| 0.2297     | 0.9550         | 8     |
+| 0.1757     | 0.9637         | 9     |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:984c51eb3e11d81b5f62756527cdd7ce177d40442889846107cb18c18e519036
+oid sha256:5e2801a968f28d1e40b8f7c9d0e329681c69945da52d20361cdc2e9420f1572b
 size 655073724