End of training

README.md

@@ -15,13 +15,13 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [deepset/gbert-base](https://huggingface.co/deepset/gbert-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1516
-- Hard: {'precision': 0.6638023630504833, 'recall': 0.7696139476961394, 'f1': 0.7128027681660899, 'number': 803}
-- Soft: {'precision': 0.6542553191489362, 'recall': 0.7935483870967742, 'f1': 0.7172011661807581, 'number': 155}
-- Overall Precision: 0.6622
-- Overall Recall: 0.7735
-- Overall F1: 0.7135
-- Overall Accuracy: 0.9526
+- Loss: 0.1391
+- Hard: {'precision': 0.6772616136919315, 'recall': 0.760989010989011, 'f1': 0.7166882276843466, 'number': 364}
+- Soft: {'precision': 0.6883116883116883, 'recall': 0.803030303030303, 'f1': 0.7412587412587411, 'number': 66}
+- Overall Precision: 0.6790
+- Overall Recall: 0.7674
+- Overall F1: 0.7205
+- Overall Accuracy: 0.9533
 
 ## Model description
 
@@ -50,13 +50,13 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Hard                                                                                                     | Soft                                                                                                      | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
-|:-------------:|:-----:|:----:|:---------------:|:--------------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:|
-| No log        | 1.0   | 158  | 0.1602          | {'precision': 0.5013054830287206, 'recall': 0.7173100871731009, 'f1': 0.5901639344262294, 'number': 803} | {'precision': 0.47639484978540775, 'recall': 0.7161290322580646, 'f1': 0.5721649484536083, 'number': 155} | 0.4971            | 0.7171         | 0.5872     | 0.9375           |
-| No log        | 2.0   | 316  | 0.1340          | {'precision': 0.600802407221665, 'recall': 0.7459526774595268, 'f1': 0.6655555555555556, 'number': 803}  | {'precision': 0.605, 'recall': 0.7806451612903226, 'f1': 0.6816901408450703, 'number': 155}               | 0.6015            | 0.7516         | 0.6682     | 0.9476           |
-| No log        | 3.0   | 474  | 0.1315          | {'precision': 0.6577825159914712, 'recall': 0.7683686176836861, 'f1': 0.7087880528431935, 'number': 803} | {'precision': 0.6631016042780749, 'recall': 0.8, 'f1': 0.7251461988304094, 'number': 155}                 | 0.6587            | 0.7735         | 0.7115     | 0.9522           |
-| 0.1497        | 4.0   | 632  | 0.1456          | {'precision': 0.6789989118607181, 'recall': 0.7770859277708593, 'f1': 0.7247386759581882, 'number': 803} | {'precision': 0.5970873786407767, 'recall': 0.7935483870967742, 'f1': 0.6814404432132964, 'number': 155}  | 0.664             | 0.7797         | 0.7172     | 0.9525           |
-| 0.1497        | 5.0   | 790  | 0.1516          | {'precision': 0.6638023630504833, 'recall': 0.7696139476961394, 'f1': 0.7128027681660899, 'number': 803} | {'precision': 0.6542553191489362, 'recall': 0.7935483870967742, 'f1': 0.7172011661807581, 'number': 155}  | 0.6622            | 0.7735         | 0.7135     | 0.9526           |
+| Training Loss | Epoch | Step | Validation Loss | Hard                                                                                                     | Soft                                                                                                    | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:--------------------------------------------------------------------------------------------------------:|:-------------------------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:|
+| No log        | 1.0   | 178  | 0.1284          | {'precision': 0.5326315789473685, 'recall': 0.695054945054945, 'f1': 0.6030989272943981, 'number': 364}  | {'precision': 0.5, 'recall': 0.5757575757575758, 'f1': 0.5352112676056339, 'number': 66}                | 0.5281            | 0.6767         | 0.5933     | 0.9463           |
+| No log        | 2.0   | 356  | 0.1157          | {'precision': 0.6073059360730594, 'recall': 0.7307692307692307, 'f1': 0.6633416458852868, 'number': 364} | {'precision': 0.631578947368421, 'recall': 0.7272727272727273, 'f1': 0.676056338028169, 'number': 66}   | 0.6109            | 0.7302         | 0.6653     | 0.9519           |
+| 0.1468        | 3.0   | 534  | 0.1286          | {'precision': 0.6846153846153846, 'recall': 0.7335164835164835, 'f1': 0.7082228116710876, 'number': 364} | {'precision': 0.6582278481012658, 'recall': 0.7878787878787878, 'f1': 0.7172413793103448, 'number': 66} | 0.6802            | 0.7419         | 0.7097     | 0.9547           |
+| 0.1468        | 4.0   | 712  | 0.1383          | {'precision': 0.6799007444168734, 'recall': 0.7527472527472527, 'f1': 0.7144719687092568, 'number': 364} | {'precision': 0.6582278481012658, 'recall': 0.7878787878787878, 'f1': 0.7172413793103448, 'number': 66} | 0.6763            | 0.7581         | 0.7149     | 0.9544           |
+| 0.1468        | 5.0   | 890  | 0.1391          | {'precision': 0.6772616136919315, 'recall': 0.760989010989011, 'f1': 0.7166882276843466, 'number': 364}  | {'precision': 0.6883116883116883, 'recall': 0.803030303030303, 'f1': 0.7412587412587411, 'number': 66}  | 0.6790            | 0.7674         | 0.7205     | 0.9533           |
 
 
 ### Framework versions