|
--- |
|
license: apache-2.0 |
|
base_model: facebook/wav2vec2-base-960h |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: audioclass-alpha |
|
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. --> |
|
|
|
# audioclass-alpha |
|
|
|
This model is a fine-tuned version of [facebook/wav2vec2-base-960h](https://huggingface.co/facebook/wav2vec2-base-960h) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.0998 |
|
- Accuracy: 0.9660 |
|
|
|
## 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: 4 |
|
- total_train_batch_size: 64 |
|
- 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 | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:| |
|
| 3.4273 | 1.0 | 62 | 3.4237 | 0.0431 | |
|
| 3.4084 | 2.0 | 124 | 3.3905 | 0.1179 | |
|
| 3.3119 | 3.0 | 186 | 3.2305 | 0.2789 | |
|
| 3.0335 | 4.0 | 248 | 2.8316 | 0.3537 | |
|
| 2.5734 | 5.0 | 310 | 2.3766 | 0.4308 | |
|
| 2.1768 | 6.0 | 372 | 1.9373 | 0.5760 | |
|
| 1.8634 | 7.0 | 434 | 1.6130 | 0.6712 | |
|
| 1.6591 | 8.0 | 496 | 1.3387 | 0.7347 | |
|
| 1.3715 | 9.0 | 558 | 1.1461 | 0.7868 | |
|
| 1.1765 | 10.0 | 620 | 0.9773 | 0.8027 | |
|
| 1.0398 | 11.0 | 682 | 0.7819 | 0.8481 | |
|
| 0.845 | 12.0 | 744 | 0.7010 | 0.8549 | |
|
| 0.7978 | 13.0 | 806 | 0.6215 | 0.8662 | |
|
| 0.6727 | 14.0 | 868 | 0.5388 | 0.8707 | |
|
| 0.6023 | 15.0 | 930 | 0.4660 | 0.8844 | |
|
| 0.6219 | 16.0 | 992 | 0.4607 | 0.8844 | |
|
| 0.4981 | 17.0 | 1054 | 0.3918 | 0.8889 | |
|
| 0.4693 | 18.0 | 1116 | 0.3890 | 0.8753 | |
|
| 0.421 | 19.0 | 1178 | 0.3288 | 0.8866 | |
|
| 0.4219 | 20.0 | 1240 | 0.3367 | 0.8934 | |
|
| 0.3775 | 21.0 | 1302 | 0.3176 | 0.8866 | |
|
| 0.3618 | 22.0 | 1364 | 0.3077 | 0.9002 | |
|
| 0.3046 | 23.0 | 1426 | 0.3206 | 0.9093 | |
|
| 0.3092 | 24.0 | 1488 | 0.2413 | 0.9320 | |
|
| 0.2793 | 25.0 | 1550 | 0.2777 | 0.9252 | |
|
| 0.3082 | 26.0 | 1612 | 0.2795 | 0.9274 | |
|
| 0.2147 | 27.0 | 1674 | 0.2467 | 0.9388 | |
|
| 0.1996 | 28.0 | 1736 | 0.2538 | 0.9388 | |
|
| 0.3012 | 29.0 | 1798 | 0.1885 | 0.9501 | |
|
| 0.2203 | 30.0 | 1860 | 0.1916 | 0.9524 | |
|
| 0.2663 | 31.0 | 1922 | 0.2053 | 0.9501 | |
|
| 0.1942 | 32.0 | 1984 | 0.1698 | 0.9524 | |
|
| 0.1852 | 33.0 | 2046 | 0.1689 | 0.9569 | |
|
| 0.2856 | 34.0 | 2108 | 0.1276 | 0.9615 | |
|
| 0.2095 | 35.0 | 2170 | 0.1376 | 0.9592 | |
|
| 0.1791 | 36.0 | 2232 | 0.1346 | 0.9615 | |
|
| 0.1528 | 37.0 | 2294 | 0.1452 | 0.9569 | |
|
| 0.1584 | 38.0 | 2356 | 0.1326 | 0.9524 | |
|
| 0.1868 | 39.0 | 2418 | 0.1309 | 0.9615 | |
|
| 0.1764 | 40.0 | 2480 | 0.1314 | 0.9592 | |
|
| 0.1877 | 41.0 | 2542 | 0.1546 | 0.9546 | |
|
| 0.1319 | 42.0 | 2604 | 0.1505 | 0.9546 | |
|
| 0.1196 | 43.0 | 2666 | 0.1386 | 0.9569 | |
|
| 0.1606 | 44.0 | 2728 | 0.1305 | 0.9569 | |
|
| 0.2146 | 45.0 | 2790 | 0.1088 | 0.9615 | |
|
| 0.1377 | 46.0 | 2852 | 0.1167 | 0.9592 | |
|
| 0.1544 | 47.0 | 2914 | 0.1027 | 0.9637 | |
|
| 0.1396 | 48.0 | 2976 | 0.0998 | 0.9660 | |
|
| 0.1702 | 49.0 | 3038 | 0.0991 | 0.9660 | |
|
| 0.1555 | 50.0 | 3100 | 0.0980 | 0.9660 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.0.dev0 |
|
- Pytorch 2.1.0+cu118 |
|
- Datasets 2.14.6 |
|
- Tokenizers 0.14.1 |
|
|
