my_birdcall_model
This model is a fine-tuned version of facebook/wav2vec2-base on the audiofolder dataset. It achieves the following results on the evaluation set:
- Loss: 4.1044
- Accuracy: 0.0953
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: 15
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 4.925 | 1.0 | 153 | 4.9054 | 0.0194 |
| 4.6287 | 2.0 | 306 | 4.5973 | 0.0356 |
| 4.4783 | 3.0 | 459 | 4.4897 | 0.0478 |
| 4.3947 | 4.0 | 612 | 4.4000 | 0.0576 |
| 4.3246 | 5.0 | 765 | 4.3330 | 0.0681 |
| 4.2475 | 6.0 | 918 | 4.2791 | 0.0769 |
| 4.1713 | 7.0 | 1071 | 4.2455 | 0.0809 |
| 4.1255 | 8.0 | 1224 | 4.2108 | 0.0765 |
| 4.0992 | 9.0 | 1377 | 4.1849 | 0.0820 |
| 4.0128 | 10.0 | 1530 | 4.1478 | 0.0914 |
| 3.9299 | 11.0 | 1683 | 4.1618 | 0.0865 |
| 3.9218 | 12.0 | 1836 | 4.1216 | 0.0916 |
| 3.8447 | 13.0 | 1989 | 4.1305 | 0.0942 |
| 3.8873 | 14.0 | 2142 | 4.1120 | 0.0942 |
| 3.8107 | 15.0 | 2295 | 4.1044 | 0.0953 |
Framework versions
- Transformers 4.38.0
- Pytorch 2.3.0+cu121
- Datasets 2.15.0
- Tokenizers 0.15.0
- Downloads last month
- 28