metadata
library_name: transformers
license: apache-2.0
base_model: facebook/wav2vec2-base
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: esc50-wav2vec2-base
results: []
esc50-wav2vec2-base
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: 1.5005
- Accuracy: 0.765
- F1 Macro: 0.7567
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 225
- num_epochs: 10.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro |
|---|---|---|---|---|---|
| 3.7525 | 1.0 | 225 | 3.6525 | 0.21 | 0.1352 |
| 3.2216 | 2.0 | 450 | 3.1632 | 0.315 | 0.2548 |
| 2.7897 | 3.0 | 675 | 2.7468 | 0.465 | 0.4172 |
| 2.3272 | 4.0 | 900 | 2.3903 | 0.575 | 0.5374 |
| 2.0279 | 5.0 | 1125 | 2.1068 | 0.655 | 0.6167 |
| 1.6353 | 6.0 | 1350 | 1.8588 | 0.665 | 0.6386 |
| 1.405 | 7.0 | 1575 | 1.6419 | 0.735 | 0.7068 |
| 1.1893 | 8.0 | 1800 | 1.5005 | 0.765 | 0.7567 |
| 1.0534 | 9.0 | 2025 | 1.4371 | 0.735 | 0.7259 |
| 0.9821 | 10.0 | 2250 | 1.4061 | 0.755 | 0.7381 |
Framework versions
- Transformers 4.56.1
- Pytorch 2.8.0+cu128
- Datasets 2.19.0
- Tokenizers 0.22.0