metadata
license: apache-2.0
base_model: facebook/wav2vec2-base-960h
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: exp_1715080677
results: []
exp_1715080677
This model is a fine-tuned version of facebook/wav2vec2-base-960h on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4691
- Accuracy: 0.8862
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: 1e-05
- train_batch_size: 8
- eval_batch_size: 8
- 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 |
|---|---|---|---|---|
| 0.6743 | 1.0 | 312 | 0.6215 | 0.7933 |
| 0.563 | 2.0 | 624 | 0.5459 | 0.8349 |
| 0.4834 | 3.0 | 936 | 0.5464 | 0.8606 |
| 0.3928 | 4.0 | 1248 | 0.4926 | 0.8782 |
| 0.3749 | 5.0 | 1560 | 0.4566 | 0.8846 |
| 0.5111 | 6.0 | 1872 | 0.4765 | 0.8734 |
| 0.2556 | 7.0 | 2184 | 0.4317 | 0.8958 |
| 0.2757 | 8.0 | 2496 | 0.4220 | 0.8926 |
| 0.342 | 9.0 | 2808 | 0.4290 | 0.8958 |
| 0.4211 | 10.0 | 3120 | 0.4691 | 0.8862 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1