metadata
license: apache-2.0
base_model: facebook/wav2vec2-base
tags:
- generated_from_trainer
model-index:
- name: pic_asr-scr_w2v2-base_002
results: []
pic_asr-scr_w2v2-base_002
This model is a fine-tuned version of facebook/wav2vec2-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.9265
- Per: 0.1166
- Pcc: 0.6564
- Ctc Loss: 0.4120
- Mse Loss: 0.8926
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: 16
- eval_batch_size: 1
- seed: 2222
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 749
- training_steps: 7490
Training results
| Training Loss | Epoch | Step | Validation Loss | Per | Pcc | Ctc Loss | Mse Loss |
|---|---|---|---|---|---|---|---|
| 11.7735 | 1.0 | 749 | 4.6360 | 0.9994 | 0.5280 | 3.7649 | 1.0658 |
| 3.8371 | 2.0 | 1498 | 2.1765 | 0.4402 | 0.6444 | 1.3986 | 0.8505 |
| 1.5597 | 3.0 | 2247 | 1.4609 | 0.1505 | 0.6303 | 0.5347 | 0.8698 |
| 1.0295 | 4.0 | 2996 | 1.3057 | 0.1356 | 0.6506 | 0.4765 | 0.8171 |
| 0.6253 | 5.0 | 3745 | 1.4002 | 0.1293 | 0.6472 | 0.4443 | 0.9477 |
| 0.2398 | 6.0 | 4494 | 1.0775 | 0.1251 | 0.6506 | 0.4249 | 0.8004 |
| -0.1329 | 7.0 | 5243 | 1.1855 | 0.1213 | 0.6509 | 0.4299 | 0.9175 |
| -0.4801 | 8.0 | 5992 | 1.2405 | 0.1194 | 0.6515 | 0.4101 | 1.0046 |
| -0.7409 | 9.0 | 6741 | 0.9340 | 0.1169 | 0.6554 | 0.4097 | 0.8839 |
| -0.8966 | 10.0 | 7490 | 0.9265 | 0.1166 | 0.6564 | 0.4120 | 0.8926 |
Framework versions
- Transformers 4.38.1
- Pytorch 2.0.1
- Datasets 2.16.1
- Tokenizers 0.15.2