<!-- 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. -->
asd_pron_w2v_clf_acc_balanced_xlsr_loss
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.5054
- Accuracy: 0.6950
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: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 8
- 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: 0.1
- num_epochs: 50
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 0.98 | 46 | 1.0989 | 0.3283 |
| No log | 1.98 | 92 | 1.0973 | 0.3317 |
| No log | 2.98 | 138 | 1.1106 | 0.3767 |
| No log | 3.98 | 184 | 1.0131 | 0.5550 |
| 1.0474 | 4.98 | 230 | 0.7994 | 0.6500 |
| 1.0474 | 5.98 | 276 | 0.8775 | 0.6633 |
| 1.0474 | 6.98 | 322 | 0.8374 | 0.7017 |
| 1.0474 | 7.98 | 368 | 1.1736 | 0.6900 |
| 0.4029 | 8.98 | 414 | 1.7477 | 0.6467 |
| 0.4029 | 9.98 | 460 | 1.4440 | 0.6917 |
| 0.4029 | 10.98 | 506 | 1.7208 | 0.6783 |
| 0.4029 | 11.98 | 552 | 1.9574 | 0.5933 |
| 0.4029 | 12.98 | 598 | 1.9526 | 0.6467 |
| 0.1003 | 13.98 | 644 | 1.4302 | 0.7317 |
| 0.1003 | 14.98 | 690 | 1.4979 | 0.7417 |
| 0.1003 | 15.98 | 736 | 1.8135 | 0.7033 |
| 0.1003 | 16.98 | 782 | 1.4752 | 0.7533 |
| 0.0477 | 17.98 | 828 | 1.9857 | 0.6767 |
| 0.0477 | 18.98 | 874 | 2.2540 | 0.6900 |
| 0.0477 | 19.98 | 920 | 2.2338 | 0.6333 |
| 0.0477 | 20.98 | 966 | 1.7913 | 0.7483 |
| 0.0217 | 21.98 | 1012 | 1.9875 | 0.7200 |
| 0.0217 | 22.98 | 1058 | 2.0505 | 0.7150 |
| 0.0217 | 23.98 | 1104 | 2.0233 | 0.7150 |
| 0.0217 | 24.98 | 1150 | 2.5344 | 0.6700 |
| 0.0217 | 25.98 | 1196 | 1.8976 | 0.7233 |
| 0.0148 | 26.98 | 1242 | 2.2131 | 0.7083 |
| 0.0148 | 27.98 | 1288 | 2.4018 | 0.6817 |
| 0.0148 | 28.98 | 1334 | 2.3855 | 0.6933 |
| 0.0148 | 29.98 | 1380 | 2.4776 | 0.6967 |
| 0.0153 | 30.98 | 1426 | 2.3106 | 0.6983 |
| 0.0153 | 31.98 | 1472 | 2.5678 | 0.6567 |
| 0.0153 | 32.98 | 1518 | 2.3137 | 0.7100 |
| 0.0153 | 33.98 | 1564 | 2.4556 | 0.6867 |
| 0.007 | 34.98 | 1610 | 2.5237 | 0.6850 |
| 0.007 | 35.98 | 1656 | 2.2113 | 0.7283 |
| 0.007 | 36.98 | 1702 | 2.2821 | 0.7017 |
| 0.007 | 37.98 | 1748 | 2.4137 | 0.7033 |
| 0.007 | 38.98 | 1794 | 2.2340 | 0.7233 |
| 0.0079 | 39.98 | 1840 | 2.4221 | 0.6967 |
| 0.0079 | 40.98 | 1886 | 2.3444 | 0.7133 |
| 0.0079 | 41.98 | 1932 | 2.3962 | 0.7033 |
| 0.0079 | 42.98 | 1978 | 2.6801 | 0.6817 |
| 0.0062 | 43.98 | 2024 | 2.4529 | 0.7033 |
| 0.0062 | 44.98 | 2070 | 2.4129 | 0.7083 |
| 0.0062 | 45.98 | 2116 | 2.4010 | 0.7067 |
| 0.0062 | 46.98 | 2162 | 2.6413 | 0.6900 |
| 0.0032 | 47.98 | 2208 | 2.5771 | 0.6917 |
| 0.0032 | 48.98 | 2254 | 2.5157 | 0.6933 |
| 0.0032 | 49.98 | 2300 | 2.5054 | 0.6950 |
Framework versions
- Transformers 4.13.0
- Pytorch 2.0.0+cu118
- Datasets 2.14.4
- Tokenizers 0.10.3
