Riku Wav2Vec
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: 0.9405 Accuracy: 0.5588
Model source
- Kind
- Other
- File size
- 361 MB
- Base model
- facebook/wav2vec2-base
- Version
- vRevision 6d662ef
- Updated
- Sep 13, 2026
- Downloads
- 7
- Source
- Hugging Face
Source description
Riku_Wav2Vec
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: 0.9405
- Accuracy: 0.5588
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: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 40
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.0971 | 0.9917 | 30 | 1.1029 | 0.3627 |
| 1.0806 | 1.9835 | 60 | 1.0783 | 0.4510 |
| 1.0378 | 2.9752 | 90 | 1.0542 | 0.4314 |
| 1.0563 | 4.0 | 121 | 1.0302 | 0.4902 |
| 1.0073 | 4.9917 | 151 | 1.2232 | 0.4412 |
| 1.0167 | 5.9835 | 181 | 1.1031 | 0.3824 |
| 1.0349 | 6.9752 | 211 | 1.0827 | 0.3824 |
| 1.0321 | 8.0 | 242 | 1.0226 | 0.4314 |
| 1.023 | 8.9917 | 272 | 1.0016 | 0.4608 |
| 0.9698 | 9.9835 | 302 | 1.0380 | 0.4118 |
| 0.9705 | 10.9752 | 332 | 1.0293 | 0.4020 |
| 0.9894 | 12.0 | 363 | 0.9970 | 0.4608 |
| 0.9792 | 12.9917 | 393 | 1.0357 | 0.3824 |
| 0.9743 | 13.9835 | 423 | 1.0314 | 0.4510 |
| 0.9926 | 14.9752 | 453 | 0.9906 | 0.4510 |
| 0.9766 | 16.0 | 484 | 1.0058 | 0.4020 |
| 0.9501 | 16.9917 | 514 | 1.0387 | 0.3529 |
| 0.9555 | 17.9835 | 544 | 0.9945 | 0.4804 |
| 0.9454 | 18.9752 | 574 | 1.0198 | 0.4706 |
| 0.9405 | 20.0 | 605 | 1.0239 | 0.4510 |
| 0.939 | 20.9917 | 635 | 0.9818 | 0.4902 |
| 0.9415 | 21.9835 | 665 | 0.9706 | 0.4706 |
| 0.9322 | 22.9752 | 695 | 1.0117 | 0.4510 |
| 0.9477 | 24.0 | 726 | 0.9799 | 0.4412 |
| 0.9348 | 24.9917 | 756 | 1.0015 | 0.4412 |
| 0.9502 | 25.9835 | 786 | 0.9883 | 0.4902 |
| 0.9018 | 26.9752 | 816 | 1.0123 | 0.5098 |
| 0.9141 | 28.0 | 847 | 0.9734 | 0.5098 |
| 0.9003 | 28.9917 | 877 | 0.9738 | 0.5294 |
| 0.8835 | 29.9835 | 907 | 0.9396 | 0.5294 |
| 0.8848 | 30.9752 | 937 | 0.9741 | 0.5098 |
| 0.8941 | 32.0 | 968 | 0.9719 | 0.5196 |
| 0.8651 | 32.9917 | 998 | 0.9700 | 0.4902 |
| 0.826 | 33.9835 | 1028 | 0.9783 | 0.5196 |
| 0.8421 | 34.9752 | 1058 | 0.9607 | 0.5294 |
| 0.8372 | 36.0 | 1089 | 0.9448 | 0.5392 |
| 0.8647 | 36.9917 | 1119 | 0.9314 | 0.5392 |
| 0.8074 | 37.9835 | 1149 | 0.9375 | 0.5294 |
| 0.8106 | 38.9752 | 1179 | 0.9404 | 0.5490 |
| 0.8217 | 39.6694 | 1200 | 0.9405 | 0.5588 |
Framework versions
- Transformers 4.46.3
- Pytorch 2.4.1+cu121
- Datasets 3.1.0
- Tokenizers 0.20.3
Sources
1 source
Sources
1 sourceVerified Sep 13
Model artifacts
1 artifact
Model artifacts
1 artifactSource excerpts
2 excerpts
Source excerpts
2 excerpts--- library_name: transformers license: apache-2.0 base_model: facebook/wav2vec2-base tags: - generated_from_trainer metrics: - accuracy model-index: - name: Riku_Wav2Vec 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. --> # Riku_Wav2Vec This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.9405 - Accuracy: 0.5588 ## 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: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: linear - lr_scheduler_warmup_ratio: 0.1 - num_epochs: 40 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | |:-------------:|:-------:|:----:|:---------------:|:--------:| | 1.0971 | 0.9917 | 30 | 1.1029 | 0.3627 | | 1.0806 | 1.9835 | 60 | 1.0783 | 0.4510 | | 1.0378 | 2.9752 | 90 | 1.0542 | 0.4314 | | 1.0563 | 4.0 | 121 | 1.0302 | 0.4902 | | 1.0073 | 4.9917 | 151 | 1.2232 | 0.4412 | | 1.0167 | 5.9835 | 181 | 1.1031 | 0.3824 | | 1.0349 | 6.9752 | 211 | 1.0827 | 0.3824 | | 1.0321 | 8.0 | 242 | 1.0226 | 0.4314 | | 1.023 | 8.9917 | 272 | 1.0016 | 0.4608 | | 0.9698 | 9.9835 | 302 | 1.0380 | 0.4118 | | 0.9705 | 10.9752 | 332 | 1.0293 | 0.4020 | | 0.9894 | 12.0 | 363 | 0.9970 | 0.4608 | | 0.9792 | 12.9917 | 393 | 1.0357 | 0.3824 | | 0.9743 | 13.9835 | 423 | 1.0314 | 0.4510 | | 0.9926 | 14.9752 | 453 | 0.9906 | 0.4510 | | 0.9766 | 16.0 | 484 | 1.0058 | 0.4020 | | 0.9501 | 16.9917 | 514 | 1.0387 | 0.3529 | | 0.9555 | 17.9835 | 544 | 0.9945 | 0.4804 | | 0.9454 | 18.9752 | 574 | 1.0198 | 0.4706 | | 0.9405 | 20.0 | 605 | 1.0239 | 0.4510 | | 0.939 | 20.9917 | 635 | 0.9818 | 0.4902 | | 0.9415 | 21.9...
Source context: 7 downloads · 0 likes · Pipeline audio-classification · Library transformers · Repo penkichiai/Riku_Wav2Vec
