metadata
license: apache-2.0
base_model: facebook/wav2vec2-large
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: ft-wav2vec2-with-minds
results: []
ft-wav2vec2-with-minds
This model is a fine-tuned version of facebook/wav2vec2-large on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0732
- Accuracy: 0.9822
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: 120
- eval_batch_size: 120
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 480
- 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 |
|---|---|---|---|---|
| 2.0814 | 1.0 | 9 | 2.0883 | 0.1143 |
| 2.064 | 2.0 | 18 | 2.0619 | 0.1678 |
| 2.0232 | 3.0 | 27 | 1.9712 | 0.2709 |
| 1.861 | 4.0 | 36 | 1.7455 | 0.3880 |
| 1.6003 | 5.0 | 45 | 1.5115 | 0.4724 |
| 1.4972 | 6.0 | 54 | 1.2623 | 0.5998 |
| 1.2332 | 7.0 | 63 | 1.0138 | 0.6935 |
| 1.081 | 8.0 | 72 | 0.8169 | 0.7601 |
| 0.9925 | 9.0 | 81 | 0.7757 | 0.7873 |
| 0.8516 | 10.0 | 90 | 0.6470 | 0.8163 |
| 0.7544 | 11.0 | 99 | 0.7208 | 0.7873 |
| 0.7006 | 12.0 | 108 | 0.5074 | 0.8557 |
| 0.591 | 13.0 | 117 | 0.4326 | 0.8782 |
| 0.5155 | 14.0 | 126 | 0.3707 | 0.9053 |
| 0.4715 | 15.0 | 135 | 0.3116 | 0.9091 |
| 0.4461 | 16.0 | 144 | 0.3167 | 0.9138 |
| 0.445 | 17.0 | 153 | 0.2963 | 0.9250 |
| 0.3899 | 18.0 | 162 | 0.2499 | 0.9353 |
| 0.3656 | 19.0 | 171 | 0.2756 | 0.9194 |
| 0.3255 | 20.0 | 180 | 0.2280 | 0.9297 |
| 0.2756 | 21.0 | 189 | 0.2178 | 0.9438 |
| 0.3119 | 22.0 | 198 | 0.1858 | 0.9513 |
| 0.2595 | 23.0 | 207 | 0.1794 | 0.9475 |
| 0.2713 | 24.0 | 216 | 0.1737 | 0.9466 |
| 0.2336 | 25.0 | 225 | 0.1758 | 0.9531 |
| 0.2359 | 26.0 | 234 | 0.1690 | 0.9485 |
| 0.2229 | 27.0 | 243 | 0.1336 | 0.9606 |
| 0.2145 | 28.0 | 252 | 0.1338 | 0.9700 |
| 0.1986 | 29.0 | 261 | 0.1525 | 0.9625 |
| 0.1811 | 30.0 | 270 | 0.1415 | 0.9653 |
| 0.165 | 31.0 | 279 | 0.1208 | 0.9672 |
| 0.1755 | 32.0 | 288 | 0.1266 | 0.9634 |
| 0.175 | 33.0 | 297 | 0.1269 | 0.9672 |
| 0.149 | 34.0 | 306 | 0.1072 | 0.9728 |
| 0.1606 | 35.0 | 315 | 0.1183 | 0.9738 |
| 0.161 | 36.0 | 324 | 0.1009 | 0.9719 |
| 0.1533 | 37.0 | 333 | 0.1000 | 0.9728 |
| 0.1239 | 38.0 | 342 | 0.1109 | 0.9691 |
| 0.1353 | 39.0 | 351 | 0.0905 | 0.9775 |
| 0.1287 | 40.0 | 360 | 0.0920 | 0.9738 |
| 0.223 | 41.0 | 369 | 0.0855 | 0.9775 |
| 0.1302 | 42.0 | 378 | 0.0748 | 0.9794 |
| 0.1249 | 43.0 | 387 | 0.0732 | 0.9822 |
| 0.1552 | 44.0 | 396 | 0.0688 | 0.9822 |
| 0.098 | 45.0 | 405 | 0.0777 | 0.9766 |
| 0.1459 | 46.0 | 414 | 0.0634 | 0.9813 |
| 0.1267 | 47.0 | 423 | 0.0653 | 0.9822 |
| 0.149 | 48.0 | 432 | 0.0709 | 0.9794 |
| 0.1135 | 49.0 | 441 | 0.0660 | 0.9813 |
| 0.118 | 50.0 | 450 | 0.0652 | 0.9813 |
Framework versions
- Transformers 4.35.2
- Pytorch 1.12.1+cu116
- Datasets 2.15.0
- Tokenizers 0.15.2