|
--- |
|
library_name: transformers |
|
license: apache-2.0 |
|
base_model: facebook/wav2vec2-xls-r-1b |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: wav2vec2-1b-E30_speed2 |
|
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. --> |
|
|
|
# wav2vec2-1b-E30_speed2 |
|
|
|
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-1b](https://huggingface.co/facebook/wav2vec2-xls-r-1b) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.5694 |
|
- Cer: 15.0846 |
|
|
|
## 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: 2 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 8 |
|
- 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: 50 |
|
- num_epochs: 5 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Cer | |
|
|:-------------:|:------:|:----:|:---------------:|:-------:| |
|
| 14.0019 | 0.2580 | 200 | 4.4372 | 96.3816 | |
|
| 2.5022 | 0.5160 | 400 | 2.0994 | 52.6786 | |
|
| 1.4318 | 0.7741 | 600 | 1.5913 | 40.9657 | |
|
| 1.1099 | 1.0321 | 800 | 1.2436 | 29.9930 | |
|
| 0.8555 | 1.2901 | 1000 | 1.0155 | 24.9530 | |
|
| 0.8201 | 1.5481 | 1200 | 0.9694 | 24.3715 | |
|
| 0.7405 | 1.8062 | 1400 | 0.9278 | 24.3480 | |
|
| 0.6497 | 2.0642 | 1600 | 0.9973 | 25.2056 | |
|
| 0.573 | 2.3222 | 1800 | 0.9353 | 23.7723 | |
|
| 0.5129 | 2.5802 | 2000 | 0.8295 | 21.0115 | |
|
| 0.4662 | 2.8383 | 2200 | 0.7473 | 18.8440 | |
|
| 0.4075 | 3.0963 | 2400 | 0.7551 | 20.4182 | |
|
| 0.3506 | 3.3543 | 2600 | 0.6929 | 18.4387 | |
|
| 0.3115 | 3.6123 | 2800 | 0.6474 | 17.4636 | |
|
| 0.3061 | 3.8703 | 3000 | 0.6699 | 17.4460 | |
|
| 0.2554 | 4.1284 | 3200 | 0.6221 | 16.2418 | |
|
| 0.2153 | 4.3864 | 3400 | 0.6394 | 16.9937 | |
|
| 0.1994 | 4.6444 | 3600 | 0.6119 | 15.9833 | |
|
| 0.1915 | 4.9024 | 3800 | 0.5694 | 15.0846 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.45.2 |
|
- Pytorch 2.3.1.post100 |
|
- Datasets 2.19.1 |
|
- Tokenizers 0.20.1 |
|
|