|
--- |
|
license: mit |
|
base_model: facebook/w2v-bert-2.0 |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- wer |
|
model-index: |
|
- name: w2v-bert-2.0-tamil-gpu-custom.v1 |
|
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. --> |
|
|
|
# w2v-bert-2.0-tamil-gpu-custom.v1 |
|
|
|
This model is a fine-tuned version of [facebook/w2v-bert-2.0](https://huggingface.co/facebook/w2v-bert-2.0) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: nan |
|
- Wer: 1.0 |
|
|
|
## 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: 4.43567e-05 |
|
- train_batch_size: 16 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 2 |
|
- total_train_batch_size: 32 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 500 |
|
- num_epochs: 10 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Wer | |
|
|:-------------:|:-----:|:-----:|:---------------:|:------:| |
|
| 2.3141 | 0.25 | 300 | inf | 0.3486 | |
|
| 0.2064 | 0.5 | 600 | inf | 0.3516 | |
|
| 0.1763 | 0.75 | 900 | inf | 0.2858 | |
|
| 0.1673 | 1.0 | 1200 | inf | 0.2929 | |
|
| 0.5517 | 1.25 | 1500 | inf | 0.5617 | |
|
| 0.7415 | 1.49 | 1800 | inf | 0.4608 | |
|
| 0.7446 | 1.74 | 2100 | inf | 0.4608 | |
|
| 0.7467 | 1.99 | 2400 | inf | 0.4608 | |
|
| 0.7447 | 2.24 | 2700 | inf | 0.4608 | |
|
| 0.7505 | 2.49 | 3000 | inf | 0.4608 | |
|
| 0.7469 | 2.74 | 3300 | inf | 0.4608 | |
|
| 0.7449 | 2.99 | 3600 | inf | 0.4608 | |
|
| 0.7487 | 3.24 | 3900 | inf | 0.4608 | |
|
| 0.7472 | 3.49 | 4200 | inf | 0.4608 | |
|
| 0.747 | 3.74 | 4500 | inf | 0.4608 | |
|
| 0.7462 | 3.99 | 4800 | inf | 0.4608 | |
|
| 0.7486 | 4.23 | 5100 | inf | 0.4608 | |
|
| 0.7503 | 4.48 | 5400 | inf | 0.4608 | |
|
| 0.7424 | 4.73 | 5700 | inf | 0.4608 | |
|
| 0.746 | 4.98 | 6000 | inf | 0.4608 | |
|
| 0.7518 | 5.23 | 6300 | inf | 0.4608 | |
|
| 0.7442 | 5.48 | 6600 | inf | 0.4608 | |
|
| 0.7466 | 5.73 | 6900 | inf | 0.4608 | |
|
| 0.7468 | 5.98 | 7200 | inf | 0.4608 | |
|
| 0.7542 | 6.23 | 7500 | inf | 0.4608 | |
|
| 0.748 | 6.48 | 7800 | inf | 0.4608 | |
|
| 0.7453 | 6.72 | 8100 | inf | 0.4608 | |
|
| 0.74 | 6.97 | 8400 | inf | 0.4608 | |
|
| 1.2386 | 7.22 | 8700 | nan | 1.0 | |
|
| 0.0 | 7.47 | 9000 | nan | 1.0 | |
|
| 0.0 | 7.72 | 9300 | nan | 1.0 | |
|
| 0.0 | 7.97 | 9600 | nan | 1.0 | |
|
| 0.0 | 8.22 | 9900 | nan | 1.0 | |
|
| 0.0 | 8.47 | 10200 | nan | 1.0 | |
|
| 0.0 | 8.72 | 10500 | nan | 1.0 | |
|
| 0.0 | 8.97 | 10800 | nan | 1.0 | |
|
| 0.0 | 9.22 | 11100 | nan | 1.0 | |
|
| 0.0 | 9.46 | 11400 | nan | 1.0 | |
|
| 0.0 | 9.71 | 11700 | nan | 1.0 | |
|
| 0.0 | 9.96 | 12000 | nan | 1.0 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.39.3 |
|
- Pytorch 2.1.2+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
|
|