metadata
license: apache-2.0
base_model: facebook/wav2vec2-base
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: w2v2-base-pretrained_lr5e-5_at0.8_da1
results: []
w2v2-base-pretrained_lr5e-5_at0.8_da1
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: 1.6691
- Wer: 0.1858
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: 500
- training_steps: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 14.0341 | 5.43 | 250 | 3.5667 | 1.0 |
| 3.138 | 10.87 | 500 | 3.1976 | 1.0 |
| 3.0471 | 16.3 | 750 | 3.0452 | 1.0 |
| 1.3116 | 21.74 | 1000 | 1.0225 | 0.3358 |
| 0.1806 | 27.17 | 1250 | 1.2151 | 0.2533 |
| 0.1065 | 32.61 | 1500 | 1.2673 | 0.2384 |
| 0.0707 | 38.04 | 1750 | 1.4184 | 0.1888 |
| 0.0547 | 43.48 | 2000 | 1.6087 | 0.1901 |
| 0.0471 | 48.91 | 2250 | 1.4537 | 0.1880 |
| 0.0381 | 54.35 | 2500 | 1.6858 | 0.1845 |
| 0.0307 | 59.78 | 2750 | 1.5607 | 0.1961 |
| 0.0269 | 65.22 | 3000 | 1.6716 | 0.1862 |
| 0.024 | 70.65 | 3250 | 1.5554 | 0.1884 |
| 0.0199 | 76.09 | 3500 | 1.7169 | 0.1850 |
| 0.0187 | 81.52 | 3750 | 1.6661 | 0.1858 |
| 0.0176 | 86.96 | 4000 | 1.6691 | 0.1858 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.0.0
- Datasets 2.14.6
- Tokenizers 0.14.1