metadata
license: mit
base_model: facebook/w2v-bert-2.0
tags:
- generated_from_trainer
model-index:
- name: wav2vec-reptiles
results: []
wav2vec-reptiles
This model is a fine-tuned version of facebook/w2v-bert-2.0 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 484.9289
- Pcc Accuracy: -0.1604
- Pcc Fluency: -0.1393
- Pcc Total Score: -0.1591
- Pcc Content: -0.1544
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: 5e-05
- train_batch_size: 4
- eval_batch_size: 6
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.4
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Pcc Accuracy | Pcc Fluency | Pcc Total Score | Pcc Content |
|---|---|---|---|---|---|---|---|
| 3175.5941 | 1.07 | 500 | 2802.1936 | -0.2863 | -0.2729 | -0.3001 | -0.2745 |
| 1733.457 | 2.13 | 1000 | 2440.8833 | -0.2827 | -0.2779 | -0.2959 | -0.2787 |
| 1890.6879 | 3.2 | 1500 | 1470.4958 | -0.2806 | -0.2763 | -0.2933 | -0.2772 |
| 470.8979 | 4.27 | 2000 | 565.3928 | -0.2658 | -0.2589 | -0.2764 | -0.2621 |
| 881.7893 | 5.34 | 2500 | 501.9731 | -0.2331 | -0.2204 | -0.2394 | -0.2285 |
| 379.352 | 6.4 | 3000 | 497.4395 | -0.2040 | -0.1871 | -0.2068 | -0.1982 |
| 378.5915 | 7.47 | 3500 | 491.6927 | -0.1783 | -0.1590 | -0.1789 | -0.1726 |
| 539.6395 | 8.54 | 4000 | 487.6133 | -0.1639 | -0.1434 | -0.1631 | -0.1582 |
| 319.019 | 9.61 | 4500 | 484.9289 | -0.1604 | -0.1393 | -0.1591 | -0.1544 |
Framework versions
- Transformers 4.37.0
- Pytorch 2.1.2
- Datasets 2.17.1
- Tokenizers 0.15.1