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 [dathi103/gbert-job](https://huggingface.co/dathi103/gbert-job) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1459
-- Hard: {'precision': 0.6588983050847458, 'recall': 0.7745952677459527, 'f1': 0.7120778477389812, 'number': 803}
-- Soft: {'precision': 0.7010869565217391, 'recall': 0.832258064516129, 'f1': 0.761061946902655, 'number': 155}
-- Overall Precision: 0.6658
-- Overall Recall: 0.7839
-- Overall F1: 0.7200
-- Overall Accuracy: 0.9560
+- Loss: 0.1383
+- Hard: {'precision': 0.6893203883495146, 'recall': 0.7802197802197802, 'f1': 0.7319587628865979, 'number': 364}
+- Soft: {'precision': 0.6329113924050633, 'recall': 0.7575757575757576, 'f1': 0.6896551724137931, 'number': 66}
+- Overall Precision: 0.6802
+- Overall Recall: 0.7767
+- Overall F1: 0.7253
+- Overall Accuracy: 0.9570
 
 ## 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  | 0.1363          | {'precision': 0.5074626865671642, 'recall': 0.6774595267745953, 'f1': 0.5802666666666667, 'number': 803} | {'precision': 0.5432692307692307, 'recall': 0.7290322580645161, 'f1': 0.6225895316804407, 'number': 155} | 0.5133            | 0.6858         | 0.5871     | 0.9451           |
-| No log        | 2.0   | 316  | 0.1282          | {'precision': 0.6240913811007269, 'recall': 0.7484433374844334, 'f1': 0.680634201585504, 'number': 803}  | {'precision': 0.59, 'recall': 0.7612903225806451, 'f1': 0.6647887323943662, 'number': 155}               | 0.6182            | 0.7505         | 0.6780     | 0.9504           |
-| No log        | 3.0   | 474  | 0.1240          | {'precision': 0.653475935828877, 'recall': 0.7608966376089664, 'f1': 0.7031070195627158, 'number': 803}  | {'precision': 0.6597938144329897, 'recall': 0.8258064516129032, 'f1': 0.7335243553008597, 'number': 155} | 0.6546            | 0.7714         | 0.7082     | 0.9550           |
-| 0.1378        | 4.0   | 632  | 0.1341          | {'precision': 0.6843267108167771, 'recall': 0.772104607721046, 'f1': 0.7255705090696314, 'number': 803}  | {'precision': 0.6822916666666666, 'recall': 0.8451612903225807, 'f1': 0.7550432276657061, 'number': 155} | 0.6840            | 0.7839         | 0.7305     | 0.9578           |
-| 0.1378        | 5.0   | 790  | 0.1459          | {'precision': 0.6588983050847458, 'recall': 0.7745952677459527, 'f1': 0.7120778477389812, 'number': 803} | {'precision': 0.7010869565217391, 'recall': 0.832258064516129, 'f1': 0.761061946902655, 'number': 155}   | 0.6658            | 0.7839         | 0.7200     | 0.9560           |
+| Training Loss | Epoch | Step | Validation Loss | Hard                                                                                                     | Soft                                                                                                    | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:--------------------------------------------------------------------------------------------------------:|:-------------------------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:|
+| No log        | 1.0   | 178  | 0.1186          | {'precision': 0.5478260869565217, 'recall': 0.6923076923076923, 'f1': 0.6116504854368932, 'number': 364} | {'precision': 0.5070422535211268, 'recall': 0.5454545454545454, 'f1': 0.5255474452554744, 'number': 66} | 0.5424            | 0.6698         | 0.5994     | 0.9511           |
+| No log        | 2.0   | 356  | 0.1073          | {'precision': 0.6247030878859857, 'recall': 0.7225274725274725, 'f1': 0.6700636942675159, 'number': 364} | {'precision': 0.573170731707317, 'recall': 0.7121212121212122, 'f1': 0.6351351351351352, 'number': 66}  | 0.6163            | 0.7209         | 0.6645     | 0.9572           |
+| 0.1381        | 3.0   | 534  | 0.1310          | {'precision': 0.6836734693877551, 'recall': 0.7362637362637363, 'f1': 0.7089947089947091, 'number': 364} | {'precision': 0.5632183908045977, 'recall': 0.7424242424242424, 'f1': 0.6405228758169934, 'number': 66} | 0.6618            | 0.7372         | 0.6975     | 0.9574           |
+| 0.1381        | 4.0   | 712  | 0.1350          | {'precision': 0.7019704433497537, 'recall': 0.782967032967033, 'f1': 0.7402597402597403, 'number': 364}  | {'precision': 0.6024096385542169, 'recall': 0.7575757575757576, 'f1': 0.6711409395973154, 'number': 66} | 0.6851            | 0.7791         | 0.7291     | 0.9564           |
+| 0.1381        | 5.0   | 890  | 0.1383          | {'precision': 0.6893203883495146, 'recall': 0.7802197802197802, 'f1': 0.7319587628865979, 'number': 364} | {'precision': 0.6329113924050633, 'recall': 0.7575757575757576, 'f1': 0.6896551724137931, 'number': 66} | 0.6802            | 0.7767         | 0.7253     | 0.9570           |
 
 
 ### Framework versions