|
--- |
|
license: apache-2.0 |
|
base_model: facebook/wav2vec2-base |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- wer |
|
model-index: |
|
- name: w2v2-base-pretrained_lr1e-4_at0.4_da1 |
|
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. --> |
|
|
|
# w2v2-base-pretrained_lr1e-4_at0.4_da1 |
|
|
|
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: 6.1307 |
|
- Wer: 1.0580 |
|
|
|
## 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: 0.0001 |
|
- 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: 1000 |
|
- num_epochs: 60 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Wer | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:| |
|
| 14.7571 | 8.06 | 500 | 3.6451 | 1.0 | |
|
| 1.7406 | 16.13 | 1000 | 3.2148 | 1.0448 | |
|
| 0.2001 | 24.19 | 1500 | 4.3045 | 1.0345 | |
|
| 0.0734 | 32.26 | 2000 | 5.3205 | 1.0516 | |
|
| 0.0402 | 40.32 | 2500 | 5.5110 | 1.0477 | |
|
| 0.0254 | 48.39 | 3000 | 5.8453 | 1.0554 | |
|
| 0.0161 | 56.45 | 3500 | 6.1307 | 1.0580 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.35.0 |
|
- Pytorch 2.0.0 |
|
- Datasets 2.14.6 |
|
- Tokenizers 0.14.1 |
|
|
