metadata
license: apache-2.0
base_model: facebook/wav2vec2-base
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: CS337_finetune_wav2vec_base
results: []
CS337_finetune_wav2vec_base
This model is a fine-tuned version of facebook/wav2vec2-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.9758
- Accuracy: 0.5838
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: 0.0005
- 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: 20
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.9835 | 0.97 | 28 | 0.9852 | 0.6256 |
1.0709 | 1.98 | 57 | 0.9559 | 0.6193 |
2.2881 | 2.99 | 86 | 1.3049 | 0.5203 |
1.5153 | 4.0 | 115 | 1.7062 | 0.3604 |
1.5127 | 4.97 | 143 | 1.3469 | 0.4683 |
1.4081 | 5.98 | 172 | 1.1282 | 0.5063 |
1.5358 | 6.99 | 201 | 1.1630 | 0.5355 |
1.4543 | 8.0 | 230 | 1.6172 | 0.3426 |
1.4626 | 8.97 | 258 | 1.1863 | 0.5241 |
1.3419 | 9.98 | 287 | 1.3211 | 0.4898 |
1.2167 | 10.99 | 316 | 1.1622 | 0.5381 |
1.3479 | 12.0 | 345 | 1.1126 | 0.5355 |
1.2712 | 12.97 | 373 | 1.1934 | 0.5102 |
1.3099 | 13.98 | 402 | 1.0652 | 0.5241 |
1.1829 | 14.99 | 431 | 1.0393 | 0.5241 |
1.3834 | 16.0 | 460 | 1.0916 | 0.5508 |
1.1997 | 16.97 | 488 | 1.0677 | 0.5799 |
1.1304 | 17.98 | 517 | 0.9980 | 0.5863 |
1.0456 | 18.99 | 546 | 0.9734 | 0.5863 |
1.0446 | 19.48 | 560 | 0.9758 | 0.5838 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1