Rami
/

multi-label-class-classification-on-github-issues

@@ -14,9 +14,9 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3836
-- Micro f1: 0.4888
-- Macro f1: 0.0304
 ## Model description
@@ -47,21 +47,21 @@ The following hyperparameters were used during training:
 | Training Loss | Epoch | Step | Validation Loss | Micro f1 | Macro f1 |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|
-| No log        | 1.0   | 4    | 0.6376          | 0.1870   | 0.0280   |
-| No log        | 2.0   | 8    | 0.5847          | 0.1961   | 0.0154   |
-| No log        | 3.0   | 12   | 0.5465          | 0.1967   | 0.0146   |
-| No log        | 4.0   | 16   | 0.5148          | 0.1982   | 0.0146   |
-| No log        | 5.0   | 20   | 0.4878          | 0.2915   | 0.0187   |
-| No log        | 6.0   | 24   | 0.4653          | 0.4655   | 0.0301   |
-| No log        | 7.0   | 28   | 0.4465          | 0.4862   | 0.0305   |
-| No log        | 8.0   | 32   | 0.4310          | 0.4884   | 0.0305   |
-| No log        | 9.0   | 36   | 0.4179          | 0.4894   | 0.0305   |
-| No log        | 10.0  | 40   | 0.4072          | 0.4894   | 0.0305   |
-| No log        | 11.0  | 44   | 0.3986          | 0.4893   | 0.0305   |
-| No log        | 12.0  | 48   | 0.3921          | 0.4893   | 0.0305   |
-| No log        | 13.0  | 52   | 0.3875          | 0.4888   | 0.0304   |
-| No log        | 14.0  | 56   | 0.3847          | 0.4888   | 0.0304   |
-| No log        | 15.0  | 60   | 0.3836          | 0.4888   | 0.0304   |
 ### Framework versions

 This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the None dataset.
 It achieves the following results on the evaluation set:
+- Loss: 0.3785
+- Micro f1: 0.5031
+- Macro f1: 0.0317
 ## Model description
 | Training Loss | Epoch | Step | Validation Loss | Micro f1 | Macro f1 |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|
+| No log        | 1.0   | 4    | 0.6548          | 0.2035   | 0.0412   |
+| No log        | 2.0   | 8    | 0.5953          | 0.2747   | 0.0395   |
+| No log        | 3.0   | 12   | 0.5509          | 0.3474   | 0.0376   |
+| No log        | 4.0   | 16   | 0.5159          | 0.4098   | 0.0375   |
+| No log        | 5.0   | 20   | 0.4859          | 0.4842   | 0.0362   |
+| No log        | 6.0   | 24   | 0.4625          | 0.4875   | 0.0315   |
+| No log        | 7.0   | 28   | 0.4434          | 0.4893   | 0.0315   |
+| No log        | 8.0   | 32   | 0.4268          | 0.4906   | 0.0315   |
+| No log        | 9.0   | 36   | 0.4134          | 0.4920   | 0.0306   |
+| No log        | 10.0  | 40   | 0.4027          | 0.4943   | 0.0308   |
+| No log        | 11.0  | 44   | 0.3939          | 0.4952   | 0.0309   |
+| No log        | 12.0  | 48   | 0.3872          | 0.4986   | 0.0312   |
+| No log        | 13.0  | 52   | 0.3824          | 0.5025   | 0.0316   |
+| No log        | 14.0  | 56   | 0.3796          | 0.5033   | 0.0317   |
+| No log        | 15.0  | 60   | 0.3785          | 0.5031   | 0.0317   |
 ### Framework versions