End of training

README.md

@@ -15,14 +15,14 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [google/mt5-large](https://huggingface.co/google/mt5-large) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.9627
-- Loc: {'precision': 0.07002967359050445, 'recall': 0.13817330210772832, 'f1': 0.09294998030720757, 'number': 854}
-- Org: {'precision': 0.06141439205955335, 'recall': 0.1523076923076923, 'f1': 0.08753315649867373, 'number': 650}
-- Per: {'precision': 0.030874785591766724, 'recall': 0.07741935483870968, 'f1': 0.04414469650521153, 'number': 465}
-- Overall Precision: 0.0567
-- Overall Recall: 0.1285
-- Overall F1: 0.0787
-- Overall Accuracy: 0.3287
+- Loss: 0.5622
+- Loc: {'precision': 0.9222857142857143, 'recall': 0.9449648711943794, 'f1': 0.9334875650665124, 'number': 854}
+- Org: {'precision': 0.8973561430793157, 'recall': 0.8876923076923077, 'f1': 0.8924980665119876, 'number': 650}
+- Per: {'precision': 0.9014373716632443, 'recall': 0.9440860215053763, 'f1': 0.9222689075630252, 'number': 465}
+- Overall Precision: 0.9092
+- Overall Recall: 0.9259
+- Overall F1: 0.9175
+- Overall Accuracy: 0.9582
 
 ## Model description
 
@@ -47,14 +47,14 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 4
+- num_epochs: 20
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Loc                                                                                                         | Org                                                                                                         | Per                                                                                                          | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
-|:-------------:|:-----:|:----:|:---------------:|:-----------------------------------------------------------------------------------------------------------:|:-----------------------------------------------------------------------------------------------------------:|:------------------------------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:|
-| 3.8187        | 2.0   | 10   | 3.1219          | {'precision': 0.06360022714366836, 'recall': 0.13114754098360656, 'f1': 0.08565965583173997, 'number': 854} | {'precision': 0.05763688760806916, 'recall': 0.15384615384615385, 'f1': 0.08385744234800839, 'number': 650} | {'precision': 0.027879677182685254, 'recall': 0.08172043010752689, 'f1': 0.04157549234135668, 'number': 465} | 0.0515            | 0.1270         | 0.0732     | 0.2983           |
-| 3.2942        | 4.0   | 20   | 2.9627          | {'precision': 0.07002967359050445, 'recall': 0.13817330210772832, 'f1': 0.09294998030720757, 'number': 854} | {'precision': 0.06141439205955335, 'recall': 0.1523076923076923, 'f1': 0.08753315649867373, 'number': 650}  | {'precision': 0.030874785591766724, 'recall': 0.07741935483870968, 'f1': 0.04414469650521153, 'number': 465} | 0.0567            | 0.1285         | 0.0787     | 0.3287           |
+| Training Loss | Epoch | Step  | Validation Loss | Loc                                                                                                      | Org                                                                                                      | Per                                                                                                      | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
+|:-------------:|:-----:|:-----:|:---------------:|:--------------------------------------------------------------------------------------------------------:|:--------------------------------------------------------------------------------------------------------:|:--------------------------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:|
+| 0.1729        | 10.0  | 5000  | 0.4248          | {'precision': 0.9111361079865017, 'recall': 0.9484777517564403, 'f1': 0.9294320137693631, 'number': 854} | {'precision': 0.9027113237639554, 'recall': 0.8707692307692307, 'f1': 0.8864526233359435, 'number': 650} | {'precision': 0.9010309278350516, 'recall': 0.9397849462365592, 'f1': 0.92, 'number': 465}               | 0.9060            | 0.9208         | 0.9134     | 0.9584           |
+| 0.0068        | 20.0  | 10000 | 0.5622          | {'precision': 0.9222857142857143, 'recall': 0.9449648711943794, 'f1': 0.9334875650665124, 'number': 854} | {'precision': 0.8973561430793157, 'recall': 0.8876923076923077, 'f1': 0.8924980665119876, 'number': 650} | {'precision': 0.9014373716632443, 'recall': 0.9440860215053763, 'f1': 0.9222689075630252, 'number': 465} | 0.9092            | 0.9259         | 0.9175     | 0.9582           |
 
 
 ### Framework versions