|
|
--- |
|
|
license: apache-2.0 |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
metrics: |
|
|
- wer |
|
|
model-index: |
|
|
- name: mounir2 |
|
|
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. --> |
|
|
|
|
|
# mounir2 |
|
|
|
|
|
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: 0.8560 |
|
|
- Wer: 1 |
|
|
|
|
|
## 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 |
|
|
- gradient_accumulation_steps: 2 |
|
|
- total_train_batch_size: 16 |
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
|
- lr_scheduler_type: linear |
|
|
- lr_scheduler_warmup_steps: 3000 |
|
|
- training_steps: 5000 |
|
|
- mixed_precision_training: Native AMP |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Wer | |
|
|
|:-------------:|:-----:|:----:|:---------------:|:---:| |
|
|
| 26.8537 | 0.42 | 100 | 25.5127 | 1 | |
|
|
| 23.3042 | 0.85 | 200 | 20.5838 | 1 | |
|
|
| 13.5844 | 1.27 | 300 | 10.8923 | 1 | |
|
|
| 5.8285 | 1.7 | 400 | 4.5743 | 1 | |
|
|
| 3.6732 | 2.12 | 500 | 3.4595 | 1 | |
|
|
| 3.4969 | 2.55 | 600 | 3.3192 | 1 | |
|
|
| 3.7275 | 2.97 | 700 | 3.2367 | 1 | |
|
|
| 3.3092 | 3.4 | 800 | 3.1613 | 1 | |
|
|
| 3.1658 | 3.82 | 900 | 3.1019 | 1 | |
|
|
| 3.1157 | 4.25 | 1000 | 3.0578 | 1 | |
|
|
| 3.105 | 4.67 | 1100 | 3.0208 | 1 | |
|
|
| 3.0181 | 5.1 | 1200 | 2.9876 | 1 | |
|
|
| 3.0154 | 5.52 | 1300 | 2.9543 | 1 | |
|
|
| 2.9889 | 5.94 | 1400 | 2.9387 | 1 | |
|
|
| 2.9461 | 6.37 | 1500 | 2.9246 | 1 | |
|
|
| 2.9261 | 6.79 | 1600 | 2.9111 | 1 | |
|
|
| 2.919 | 7.22 | 1700 | 2.9049 | 1 | |
|
|
| 2.9235 | 7.64 | 1800 | 2.8974 | 1 | |
|
|
| 2.899 | 8.07 | 1900 | 2.8864 | 1 | |
|
|
| 2.9122 | 8.49 | 2000 | 2.8994 | 1 | |
|
|
| 2.8856 | 8.92 | 2100 | 2.8789 | 1 | |
|
|
| 2.8693 | 9.34 | 2200 | 2.8765 | 1 | |
|
|
| 2.9063 | 9.77 | 2300 | 2.8693 | 1 | |
|
|
| 2.8701 | 10.19 | 2400 | 2.8700 | 1 | |
|
|
| 2.9013 | 10.62 | 2500 | 2.8647 | 1 | |
|
|
| 2.8715 | 11.04 | 2600 | 2.8605 | 1 | |
|
|
| 2.8524 | 11.46 | 2700 | 2.8706 | 1 | |
|
|
| 2.8551 | 11.89 | 2800 | 2.8534 | 1 | |
|
|
| 2.8466 | 12.31 | 2900 | 2.8517 | 1 | |
|
|
| 2.8419 | 12.74 | 3000 | 2.8604 | 1 | |
|
|
| 2.8345 | 13.16 | 3100 | 2.8388 | 1 | |
|
|
| 2.827 | 13.59 | 3200 | 2.8220 | 1 | |
|
|
| 2.6233 | 14.01 | 3300 | 2.5103 | 1 | |
|
|
| 2.2908 | 14.44 | 3400 | 2.1460 | 1 | |
|
|
| 1.9386 | 14.86 | 3500 | 1.7383 | 1 | |
|
|
| 1.6341 | 15.29 | 3600 | 1.4527 | 1 | |
|
|
| 1.5684 | 15.71 | 3700 | 1.2669 | 1 | |
|
|
| 1.2713 | 16.14 | 3800 | 1.1512 | 1 | |
|
|
| 1.1983 | 16.56 | 3900 | 1.0776 | 1 | |
|
|
| 1.1806 | 16.99 | 4000 | 1.0261 | 1 | |
|
|
| 1.1456 | 17.41 | 4100 | 0.9729 | 1 | |
|
|
| 1.0942 | 17.83 | 4200 | 0.9407 | 1 | |
|
|
| 1.0452 | 18.26 | 4300 | 0.9172 | 1 | |
|
|
| 1.0082 | 18.68 | 4400 | 0.9006 | 1 | |
|
|
| 0.9748 | 19.11 | 4500 | 0.8902 | 1 | |
|
|
| 0.9672 | 19.53 | 4600 | 0.8742 | 1 | |
|
|
| 0.9737 | 19.96 | 4700 | 0.8738 | 1 | |
|
|
| 0.9567 | 20.38 | 4800 | 0.8639 | 1 | |
|
|
| 0.988 | 20.81 | 4900 | 0.8565 | 1 | |
|
|
| 0.9783 | 21.23 | 5000 | 0.8560 | 1 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.28.0 |
|
|
- Pytorch 2.0.1+cu117 |
|
|
- Datasets 2.12.0 |
|
|
- Tokenizers 0.13.3 |
|
|
|
