metadata
license: apache-2.0
base_model: facebook/wav2vec2-base
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: wav2vec2-base-nsc-demo-4
results: []
wav2vec2-base-nsc-demo-4
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: 1.0598
- Wer: 0.5177
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: 5e-05
- train_batch_size: 32
- 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_steps: 100
- num_epochs: 200
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 2.943 | 4.55 | 100 | 2.9614 | 1.0 |
| 2.909 | 9.09 | 200 | 2.9454 | 1.0 |
| 2.8673 | 13.64 | 300 | 2.9148 | 1.0 |
| 2.3216 | 18.18 | 400 | 1.9129 | 0.9218 |
| 0.8129 | 22.73 | 500 | 1.0840 | 0.6480 |
| 0.3488 | 27.27 | 600 | 1.0021 | 0.5721 |
| 0.2166 | 31.82 | 700 | 1.0291 | 0.5568 |
| 0.153 | 36.36 | 800 | 1.0672 | 0.5394 |
| 0.1186 | 40.91 | 900 | 1.0598 | 0.5177 |
Framework versions
- Transformers 4.33.0
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.13.3