|
--- |
|
license: apache-2.0 |
|
base_model: distilbert-base-multilingual-cased |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- precision |
|
- recall |
|
- f1 |
|
- accuracy |
|
model-index: |
|
- name: BERT_B01 |
|
results: [] |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# BERT_B01 |
|
|
|
This model is a fine-tuned version of [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.6902 |
|
- Precision: 0.6636 |
|
- Recall: 0.6946 |
|
- F1: 0.6788 |
|
- Accuracy: 0.8776 |
|
|
|
## 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: 30 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| |
|
| 1.0749 | 1.0 | 47 | 0.9390 | 0.4203 | 0.3480 | 0.3807 | 0.7831 | |
|
| 0.6411 | 2.0 | 94 | 0.6110 | 0.5948 | 0.5392 | 0.5657 | 0.8452 | |
|
| 0.4786 | 3.0 | 141 | 0.5279 | 0.6784 | 0.6121 | 0.6435 | 0.8630 | |
|
| 0.3573 | 4.0 | 188 | 0.4972 | 0.6462 | 0.6382 | 0.6422 | 0.8691 | |
|
| 0.2824 | 5.0 | 235 | 0.4868 | 0.6339 | 0.6479 | 0.6408 | 0.8689 | |
|
| 0.2434 | 6.0 | 282 | 0.4970 | 0.6490 | 0.6561 | 0.6525 | 0.8715 | |
|
| 0.1854 | 7.0 | 329 | 0.5004 | 0.6578 | 0.6795 | 0.6685 | 0.8721 | |
|
| 0.1336 | 8.0 | 376 | 0.5091 | 0.6508 | 0.6768 | 0.6635 | 0.8736 | |
|
| 0.1186 | 9.0 | 423 | 0.5437 | 0.6340 | 0.6768 | 0.6547 | 0.8739 | |
|
| 0.103 | 10.0 | 470 | 0.5482 | 0.6570 | 0.6823 | 0.6694 | 0.8771 | |
|
| 0.0799 | 11.0 | 517 | 0.5620 | 0.6444 | 0.6781 | 0.6609 | 0.8752 | |
|
| 0.1045 | 12.0 | 564 | 0.5812 | 0.6557 | 0.6864 | 0.6707 | 0.8760 | |
|
| 0.0562 | 13.0 | 611 | 0.6009 | 0.6667 | 0.6850 | 0.6757 | 0.8780 | |
|
| 0.0637 | 14.0 | 658 | 0.5937 | 0.6707 | 0.6946 | 0.6824 | 0.8780 | |
|
| 0.0657 | 15.0 | 705 | 0.6017 | 0.6788 | 0.6946 | 0.6866 | 0.8789 | |
|
| 0.0371 | 16.0 | 752 | 0.6227 | 0.6858 | 0.6905 | 0.6881 | 0.8776 | |
|
| 0.0389 | 17.0 | 799 | 0.6476 | 0.6499 | 0.6919 | 0.6702 | 0.8767 | |
|
| 0.0461 | 18.0 | 846 | 0.6667 | 0.6556 | 0.7043 | 0.6790 | 0.8786 | |
|
| 0.0377 | 19.0 | 893 | 0.6515 | 0.6788 | 0.6919 | 0.6853 | 0.8793 | |
|
| 0.0364 | 20.0 | 940 | 0.6480 | 0.6791 | 0.7015 | 0.6901 | 0.8784 | |
|
| 0.0383 | 21.0 | 987 | 0.6646 | 0.6719 | 0.7070 | 0.6890 | 0.8802 | |
|
| 0.0173 | 22.0 | 1034 | 0.6724 | 0.6750 | 0.7029 | 0.6887 | 0.8793 | |
|
| 0.0613 | 23.0 | 1081 | 0.6779 | 0.6580 | 0.6988 | 0.6778 | 0.8778 | |
|
| 0.0578 | 24.0 | 1128 | 0.6847 | 0.6592 | 0.6864 | 0.6725 | 0.8767 | |
|
| 0.0201 | 25.0 | 1175 | 0.6714 | 0.6706 | 0.7001 | 0.6851 | 0.8791 | |
|
| 0.022 | 26.0 | 1222 | 0.6874 | 0.6667 | 0.6878 | 0.6770 | 0.8782 | |
|
| 0.0298 | 27.0 | 1269 | 0.6926 | 0.6675 | 0.6960 | 0.6815 | 0.8789 | |
|
| 0.03 | 28.0 | 1316 | 0.6895 | 0.6662 | 0.6974 | 0.6815 | 0.8784 | |
|
| 0.0216 | 29.0 | 1363 | 0.6888 | 0.6636 | 0.6946 | 0.6788 | 0.8780 | |
|
| 0.0236 | 30.0 | 1410 | 0.6902 | 0.6636 | 0.6946 | 0.6788 | 0.8776 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.33.3 |
|
- Pytorch 2.0.1+cu117 |
|
- Datasets 2.14.4 |
|
- Tokenizers 0.13.3 |
|
|
