metadata
license: apache-2.0
base_model: facebook/wav2vec2-base
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: shimalabaudio-20240425
results: []
shimalabaudio-20240425
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: 0.7193
- Accuracy: 0.5385
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
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.6884 | 1.0 | 4 | 0.6969 | 0.3846 |
| 0.6795 | 2.0 | 8 | 0.7070 | 0.3846 |
| 0.604 | 3.0 | 12 | 0.7304 | 0.3846 |
| 0.6326 | 4.0 | 16 | 0.7299 | 0.4615 |
| 0.6014 | 5.0 | 20 | 0.7191 | 0.4615 |
| 0.5396 | 6.0 | 24 | 0.7744 | 0.3077 |
| 0.599 | 7.0 | 28 | 0.7345 | 0.4615 |
| 0.8114 | 8.0 | 32 | 0.7081 | 0.5385 |
| 0.7966 | 9.0 | 36 | 0.7111 | 0.5385 |
| 0.6202 | 10.0 | 40 | 0.7193 | 0.5385 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.2.0
- Datasets 2.12.0
- Tokenizers 0.15.1