metadata
license: apache-2.0
base_model: facebook/wav2vec2-base
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: wav2vec2-base-finetuned-ravdess
results: []
wav2vec2-base-finetuned-ravdess
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: 0.9291
- Accuracy: 0.7431
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: 0.1
- num_epochs: 50
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 1.0 | 9 | 1.8542 | 0.375 |
| 1.8624 | 2.0 | 18 | 1.7931 | 0.4410 |
| 1.8211 | 3.0 | 27 | 1.7181 | 0.4479 |
| 1.7355 | 4.0 | 36 | 1.6184 | 0.4861 |
| 1.6208 | 5.0 | 45 | 1.5383 | 0.4965 |
| 1.4999 | 6.0 | 54 | 1.4630 | 0.5069 |
| 1.4016 | 7.0 | 63 | 1.4161 | 0.5278 |
| 1.3028 | 8.0 | 72 | 1.3513 | 0.5312 |
| 1.2202 | 9.0 | 81 | 1.3359 | 0.5625 |
| 1.1506 | 10.0 | 90 | 1.3518 | 0.5625 |
| 1.1506 | 11.0 | 99 | 1.2043 | 0.6076 |
| 1.0749 | 12.0 | 108 | 1.2926 | 0.5833 |
| 0.993 | 13.0 | 117 | 1.1970 | 0.6181 |
| 0.8961 | 14.0 | 126 | 1.1593 | 0.6285 |
| 0.8062 | 15.0 | 135 | 1.1184 | 0.6424 |
| 0.7547 | 16.0 | 144 | 1.1318 | 0.6285 |
| 0.7192 | 17.0 | 153 | 1.0872 | 0.6493 |
| 0.65 | 18.0 | 162 | 1.0737 | 0.6528 |
| 0.6142 | 19.0 | 171 | 1.0677 | 0.6597 |
| 0.5845 | 20.0 | 180 | 1.0220 | 0.6806 |
| 0.5845 | 21.0 | 189 | 0.9968 | 0.6840 |
| 0.5217 | 22.0 | 198 | 0.9864 | 0.6840 |
| 0.4798 | 23.0 | 207 | 0.9708 | 0.6840 |
| 0.4501 | 24.0 | 216 | 1.0981 | 0.6632 |
| 0.4339 | 25.0 | 225 | 1.0536 | 0.6806 |
| 0.4274 | 26.0 | 234 | 0.9387 | 0.6979 |
| 0.3742 | 27.0 | 243 | 0.9879 | 0.6979 |
| 0.3747 | 28.0 | 252 | 0.9773 | 0.6979 |
| 0.3389 | 29.0 | 261 | 0.9257 | 0.7361 |
| 0.3213 | 30.0 | 270 | 0.9292 | 0.7049 |
| 0.3213 | 31.0 | 279 | 0.9555 | 0.7153 |
| 0.3007 | 32.0 | 288 | 0.9733 | 0.7083 |
| 0.2681 | 33.0 | 297 | 1.0336 | 0.6979 |
| 0.2576 | 34.0 | 306 | 1.0443 | 0.6875 |
| 0.2561 | 35.0 | 315 | 0.9261 | 0.7292 |
| 0.2575 | 36.0 | 324 | 1.0833 | 0.6771 |
| 0.2483 | 37.0 | 333 | 0.9775 | 0.7083 |
| 0.2108 | 38.0 | 342 | 0.8911 | 0.7326 |
| 0.218 | 39.0 | 351 | 1.0301 | 0.6840 |
| 0.1927 | 40.0 | 360 | 0.9935 | 0.7014 |
| 0.1927 | 41.0 | 369 | 0.9619 | 0.7292 |
| 0.1945 | 42.0 | 378 | 1.0197 | 0.6944 |
| 0.1926 | 43.0 | 387 | 0.9291 | 0.7431 |
| 0.1799 | 44.0 | 396 | 1.0348 | 0.6875 |
| 0.1768 | 45.0 | 405 | 0.9899 | 0.7153 |
| 0.1728 | 46.0 | 414 | 1.0284 | 0.7049 |
| 0.1713 | 47.0 | 423 | 1.0527 | 0.7049 |
| 0.1638 | 48.0 | 432 | 1.0001 | 0.7083 |
| 0.159 | 49.0 | 441 | 0.9931 | 0.7153 |
| 0.1637 | 50.0 | 450 | 1.0015 | 0.7118 |
Framework versions
- Transformers 4.31.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.4
- Tokenizers 0.13.3