<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
distilbert-base-uncased_fold_2_binary_v1
This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.8833
- F1: 0.7841
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 25
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 |
|---|---|---|---|---|
| No log | 1.0 | 290 | 0.4060 | 0.8070 |
| 0.3981 | 2.0 | 580 | 0.4534 | 0.8072 |
| 0.3981 | 3.0 | 870 | 0.5460 | 0.7961 |
| 0.1985 | 4.0 | 1160 | 0.8684 | 0.7818 |
| 0.1985 | 5.0 | 1450 | 0.9009 | 0.7873 |
| 0.0844 | 6.0 | 1740 | 1.1529 | 0.7825 |
| 0.0329 | 7.0 | 2030 | 1.3185 | 0.7850 |
| 0.0329 | 8.0 | 2320 | 1.4110 | 0.7862 |
| 0.0109 | 9.0 | 2610 | 1.4751 | 0.7784 |
| 0.0109 | 10.0 | 2900 | 1.6276 | 0.7723 |
| 0.0071 | 11.0 | 3190 | 1.6779 | 0.7861 |
| 0.0071 | 12.0 | 3480 | 1.6258 | 0.7850 |
| 0.0041 | 13.0 | 3770 | 1.6324 | 0.7903 |
| 0.0109 | 14.0 | 4060 | 1.7563 | 0.7932 |
| 0.0109 | 15.0 | 4350 | 1.6740 | 0.7906 |
| 0.0079 | 16.0 | 4640 | 1.7468 | 0.7944 |
| 0.0079 | 17.0 | 4930 | 1.7095 | 0.7879 |
| 0.0067 | 18.0 | 5220 | 1.7293 | 0.7912 |
| 0.0021 | 19.0 | 5510 | 1.7875 | 0.7848 |
| 0.0021 | 20.0 | 5800 | 1.7462 | 0.7906 |
| 0.0026 | 21.0 | 6090 | 1.8549 | 0.7815 |
| 0.0026 | 22.0 | 6380 | 1.8314 | 0.7860 |
| 0.0021 | 23.0 | 6670 | 1.8577 | 0.7839 |
| 0.0021 | 24.0 | 6960 | 1.8548 | 0.7883 |
| 0.0001 | 25.0 | 7250 | 1.8833 | 0.7841 |
Framework versions
- Transformers 4.21.0
- Pytorch 1.12.0+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1
