classifier_arabic_sl
This model is a fine-tuned version of facebook/wav2vec2-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.1488
- Accuracy: 0.8010
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: 3e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.01
- training_steps: 2000
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 1.0 | 44 | 4.3814 | 0.0417 |
| No log | 2.0 | 88 | 4.0491 | 0.0562 |
| No log | 3.0 | 132 | 3.7968 | 0.1188 |
| No log | 4.0 | 176 | 3.5526 | 0.2055 |
| 4.0183 | 5.0 | 220 | 3.3481 | 0.3210 |
| 4.0183 | 6.0 | 264 | 3.1680 | 0.3900 |
| 4.0183 | 7.0 | 308 | 3.0116 | 0.4783 |
| 4.0183 | 8.0 | 352 | 2.8638 | 0.5554 |
| 4.0183 | 9.0 | 396 | 2.7116 | 0.6132 |
| 3.1421 | 10.0 | 440 | 2.6048 | 0.6404 |
| 3.1421 | 11.0 | 484 | 2.4877 | 0.6613 |
| 3.1421 | 12.0 | 528 | 2.3655 | 0.6902 |
| 3.1421 | 13.0 | 572 | 2.2702 | 0.7175 |
| 2.5637 | 14.0 | 616 | 2.1865 | 0.7207 |
| 2.5637 | 15.0 | 660 | 2.1042 | 0.7095 |
| 2.5637 | 16.0 | 704 | 2.0213 | 0.7255 |
| 2.5637 | 17.0 | 748 | 1.9595 | 0.7095 |
| 2.5637 | 18.0 | 792 | 1.8877 | 0.7255 |
| 2.1297 | 19.0 | 836 | 1.8018 | 0.7432 |
| 2.1297 | 20.0 | 880 | 1.7479 | 0.7512 |
| 2.1297 | 21.0 | 924 | 1.6810 | 0.7657 |
| 2.1297 | 22.0 | 968 | 1.6300 | 0.7592 |
| 1.789 | 23.0 | 1012 | 1.5904 | 0.7592 |
| 1.789 | 24.0 | 1056 | 1.5613 | 0.7657 |
| 1.789 | 25.0 | 1100 | 1.5007 | 0.7689 |
| 1.789 | 26.0 | 1144 | 1.4766 | 0.7689 |
| 1.789 | 27.0 | 1188 | 1.4231 | 0.7817 |
| 1.5387 | 28.0 | 1232 | 1.3655 | 0.7961 |
| 1.5387 | 29.0 | 1276 | 1.3699 | 0.7849 |
| 1.5387 | 30.0 | 1320 | 1.3354 | 0.7785 |
| 1.5387 | 31.0 | 1364 | 1.3073 | 0.7785 |
| 1.3525 | 32.0 | 1408 | 1.2796 | 0.7849 |
| 1.3525 | 33.0 | 1452 | 1.2644 | 0.7978 |
| 1.3525 | 34.0 | 1496 | 1.2487 | 0.7865 |
| 1.3525 | 35.0 | 1540 | 1.2364 | 0.7865 |
| 1.3525 | 36.0 | 1584 | 1.2158 | 0.7881 |
| 1.2188 | 37.0 | 1628 | 1.2001 | 0.7929 |
| 1.2188 | 38.0 | 1672 | 1.1916 | 0.7945 |
| 1.2188 | 39.0 | 1716 | 1.1933 | 0.7817 |
| 1.2188 | 40.0 | 1760 | 1.1778 | 0.7929 |
| 1.1275 | 41.0 | 1804 | 1.1622 | 0.7978 |
| 1.1275 | 42.0 | 1848 | 1.1598 | 0.7881 |
| 1.1275 | 43.0 | 1892 | 1.1557 | 0.7961 |
| 1.1275 | 44.0 | 1936 | 1.1461 | 0.7961 |
| 1.1275 | 45.0 | 1980 | 1.1488 | 0.8026 |
| 1.0894 | 45.45 | 2000 | 1.1488 | 0.8010 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.2.1
- Datasets 2.18.0
- Tokenizers 0.13.3
- Downloads last month
- 1