dtorber
/

bert-base-cased

@@ -18,15 +18,14 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [google-bert/bert-base-cased](https://huggingface.co/google-bert/bert-base-cased) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.2458
-- F1 Macro: 0.8817
-- F1: 0.9130
-- F1 Neg: 0.8503
-- Acc: 0.89
-- Prec: 0.9130
-- Recall: 0.9130
-- Mcc: 0.7634
-- Millor Epoca: 6
 ## Model description
@@ -49,28 +48,25 @@ The following hyperparameters were used during training:
 - train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
-- distributed_type: multi-GPU
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 7
 - mixed_precision_training: Native AMP
 ### Training results
-| Training Loss | Epoch | Step  | Validation Loss | F1 Macro | F1     | F1 Neg | Acc    | Prec   | Recall | Mcc    | Millor Epoca |
-|:-------------:|:-----:|:-----:|:---------------:|:--------:|:------:|:------:|:------:|:------:|:------:|:------:|:------------:|
-| 0.2675        | 1.0   | 3801  | 0.5655          | 0.8736   | 0.9115 | 0.8357 | 0.885  | 0.8876 | 0.9368 | 0.7498 | 1            |
-| 0.1016        | 2.0   | 7602  | 0.7133          | 0.8640   | 0.8926 | 0.8354 | 0.87   | 0.9351 | 0.8538 | 0.7337 | 1            |
-| 0.041         | 3.0   | 11403 | 1.0065          | 0.8702   | 0.9101 | 0.8303 | 0.8825 | 0.8815 | 0.9407 | 0.7442 | 1            |
-| 0.0211        | 4.0   | 15204 | 1.2972          | 0.8736   | 0.9115 | 0.8357 | 0.885  | 0.8876 | 0.9368 | 0.7498 | 1            |
-| 0.0108        | 5.0   | 19005 | 1.3592          | 0.8677   | 0.9080 | 0.8273 | 0.88   | 0.8810 | 0.9368 | 0.7387 | 1            |
-| 0.002         | 6.0   | 22806 | 1.2458          | 0.8817   | 0.9130 | 0.8503 | 0.89   | 0.9130 | 0.9130 | 0.7634 | 6            |
-| 0.0           | 7.0   | 26607 | 1.3670          | 0.8766   | 0.9133 | 0.8399 | 0.8875 | 0.8910 | 0.9368 | 0.7554 | 6            |
 ### Framework versions
-- Transformers 4.39.3
 - Pytorch 2.3.0+cu121
 - Datasets 2.18.0
-- Tokenizers 0.15.2

 This model is a fine-tuned version of [google-bert/bert-base-cased](https://huggingface.co/google-bert/bert-base-cased) on the None dataset.
 It achieves the following results on the evaluation set:
+- Loss: 0.4624
+- F1 Macro: 0.9016
+- F1: 0.9343
+- F1 Neg: 0.8689
+- Acc: 0.9125
+- Prec: 0.9432
+- Recall: 0.9257
+- Mcc: 0.8036
 ## Model description
 - train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
+- num_epochs: 5
 - mixed_precision_training: Native AMP
 ### Training results
+| Training Loss | Epoch | Step | Validation Loss | F1 Macro | F1     | F1 Neg | Acc    | Prec   | Recall | Mcc    |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:------:|:------:|:------:|:------:|:------:|
+| No log        | 1.0   | 474  | 0.3773          | 0.8286   | 0.8676 | 0.7896 | 0.8375 | 0.9064 | 0.8320 | 0.6623 |
+| 0.4662        | 2.0   | 948  | 0.4197          | 0.8638   | 0.9070 | 0.8205 | 0.8775 | 0.8819 | 0.9336 | 0.7305 |
+| 0.3045        | 3.0   | 1422 | 0.4950          | 0.8808   | 0.9219 | 0.8397 | 0.895  | 0.8794 | 0.9688 | 0.7711 |
+| 0.2105        | 4.0   | 1896 | 0.4999          | 0.8817   | 0.9175 | 0.8459 | 0.8925 | 0.9019 | 0.9336 | 0.7644 |
+| 0.1246        | 5.0   | 2370 | 0.5317          | 0.8889   | 0.9198 | 0.8581 | 0.8975 | 0.9216 | 0.9180 | 0.7779 |
 ### Framework versions
+- Transformers 4.40.1
 - Pytorch 2.3.0+cu121
 - Datasets 2.18.0
+- Tokenizers 0.19.1