|
--- |
|
library_name: transformers |
|
license: apache-2.0 |
|
base_model: facebook/wav2vec2-large-xlsr-53 |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- wer |
|
model-index: |
|
- name: xlsr-no-se-nmcpc |
|
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. --> |
|
|
|
# xlsr-no-se-nmcpc |
|
|
|
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.0008 |
|
- Wer: 0.2106 |
|
|
|
## 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: 0.0004 |
|
- 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: 132 |
|
- num_epochs: 100 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Wer | |
|
|:-------------:|:-------:|:----:|:---------------:|:------:| |
|
| 4.9119 | 4.2553 | 200 | 3.0710 | 1.0 | |
|
| 3.0199 | 8.5106 | 400 | 2.8618 | 1.0 | |
|
| 2.7311 | 12.7660 | 600 | 2.2436 | 0.9957 | |
|
| 1.7526 | 17.0213 | 800 | 0.5282 | 0.6234 | |
|
| 0.7948 | 21.2766 | 1000 | 0.1705 | 0.3894 | |
|
| 0.4432 | 25.5319 | 1200 | 0.1006 | 0.3149 | |
|
| 0.2996 | 29.7872 | 1400 | 0.0629 | 0.2660 | |
|
| 0.2087 | 34.0426 | 1600 | 0.0243 | 0.2532 | |
|
| 0.1572 | 38.2979 | 1800 | 0.0223 | 0.2362 | |
|
| 0.1222 | 42.5532 | 2000 | 0.0177 | 0.2404 | |
|
| 0.1003 | 46.8085 | 2200 | 0.0069 | 0.2298 | |
|
| 0.0897 | 51.0638 | 2400 | 0.0074 | 0.2234 | |
|
| 0.0686 | 55.3191 | 2600 | 0.0058 | 0.2149 | |
|
| 0.0649 | 59.5745 | 2800 | 0.0028 | 0.2064 | |
|
| 0.0481 | 63.8298 | 3000 | 0.0027 | 0.2064 | |
|
| 0.0504 | 68.0851 | 3200 | 0.0028 | 0.2085 | |
|
| 0.0416 | 72.3404 | 3400 | 0.0042 | 0.1936 | |
|
| 0.0317 | 76.5957 | 3600 | 0.0023 | 0.2021 | |
|
| 0.0326 | 80.8511 | 3800 | 0.0010 | 0.2043 | |
|
| 0.026 | 85.1064 | 4000 | 0.0009 | 0.2085 | |
|
| 0.0233 | 89.3617 | 4200 | 0.0009 | 0.2085 | |
|
| 0.0202 | 93.6170 | 4400 | 0.0008 | 0.2106 | |
|
| 0.0235 | 97.8723 | 4600 | 0.0008 | 0.2106 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.45.0.dev0 |
|
- Pytorch 2.4.0 |
|
- Datasets 2.21.0 |
|
- Tokenizers 0.19.1 |
|
|