|
--- |
|
library_name: transformers |
|
license: apache-2.0 |
|
base_model: facebook/wav2vec2-xls-r-1b |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: wav2vec2-1b-E30 |
|
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 |
|
|
|
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: 0.7150 |
|
- Cer: 16.9467 |
|
|
|
## 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 | |
|
|:-------------:|:------:|:----:|:---------------:|:-------:| |
|
| 9.1231 | 0.8602 | 200 | 2.2073 | 46.5049 | |
|
| 1.2107 | 1.7204 | 400 | 1.1967 | 26.6095 | |
|
| 0.7167 | 2.5806 | 600 | 0.9255 | 21.9807 | |
|
| 0.4842 | 3.4409 | 800 | 0.8168 | 19.2375 | |
|
| 0.3261 | 4.3011 | 1000 | 0.7150 | 16.9467 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.45.2 |
|
- Pytorch 2.3.1.post100 |
|
- Datasets 2.19.1 |
|
- Tokenizers 0.20.1 |
|
|
