metadata
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- f1
model-index:
- name: distilbert-base-uncased_fold_1_binary
results: []
distilbert-base-uncased_fold_1_binary
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: 1.1222
- F1: 0.7596
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 | 288 | 0.4130 | 0.7517 |
| 0.3938 | 2.0 | 576 | 0.4260 | 0.7330 |
| 0.3938 | 3.0 | 864 | 0.5000 | 0.7488 |
| 0.19 | 4.0 | 1152 | 0.7415 | 0.7487 |
| 0.19 | 5.0 | 1440 | 0.8994 | 0.7397 |
| 0.0903 | 6.0 | 1728 | 0.9835 | 0.7386 |
| 0.0392 | 7.0 | 2016 | 1.1222 | 0.7596 |
| 0.0392 | 8.0 | 2304 | 1.2018 | 0.7314 |
| 0.0234 | 9.0 | 2592 | 1.2691 | 0.7330 |
| 0.0234 | 10.0 | 2880 | 1.2972 | 0.7496 |
| 0.0182 | 11.0 | 3168 | 1.4606 | 0.7492 |
| 0.0182 | 12.0 | 3456 | 1.4766 | 0.7361 |
| 0.006 | 13.0 | 3744 | 1.4888 | 0.7500 |
| 0.0057 | 14.0 | 4032 | 1.5684 | 0.7298 |
| 0.0057 | 15.0 | 4320 | 1.5354 | 0.7509 |
| 0.0058 | 16.0 | 4608 | 1.7733 | 0.7436 |
| 0.0058 | 17.0 | 4896 | 1.5695 | 0.7512 |
| 0.0089 | 18.0 | 5184 | 1.6593 | 0.7430 |
| 0.0089 | 19.0 | 5472 | 1.7092 | 0.7444 |
| 0.0048 | 20.0 | 5760 | 1.7206 | 0.7374 |
| 0.002 | 21.0 | 6048 | 1.7440 | 0.7343 |
| 0.002 | 22.0 | 6336 | 1.7582 | 0.7347 |
| 0.0006 | 23.0 | 6624 | 1.7294 | 0.7472 |
| 0.0006 | 24.0 | 6912 | 1.7454 | 0.7365 |
| 0.0001 | 25.0 | 7200 | 1.7395 | 0.7429 |
Framework versions
- Transformers 4.21.0
- Pytorch 1.12.0+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1