metadata
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: wav2vec2-base-intent-classification-ori
results: []
wav2vec2-base-intent-classification-ori
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: 0.6214
- Accuracy: 0.8542
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: 3e-05
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 45
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.1969 | 1.0 | 28 | 2.1823 | 0.1667 |
| 2.1258 | 2.0 | 56 | 2.1160 | 0.2708 |
| 2.1138 | 3.0 | 84 | 2.1125 | 0.3125 |
| 2.038 | 4.0 | 112 | 2.0132 | 0.2708 |
| 2.0267 | 5.0 | 140 | 2.0051 | 0.3333 |
| 1.8788 | 6.0 | 168 | 1.8372 | 0.4167 |
| 1.8875 | 7.0 | 196 | 1.9644 | 0.2917 |
| 1.7334 | 8.0 | 224 | 1.5419 | 0.5625 |
| 1.3128 | 9.0 | 252 | 1.5499 | 0.4792 |
| 1.1863 | 10.0 | 280 | 1.4967 | 0.4792 |
| 1.2418 | 11.0 | 308 | 1.3421 | 0.5625 |
| 1.2286 | 12.0 | 336 | 1.1493 | 0.5625 |
| 1.087 | 13.0 | 364 | 1.2001 | 0.5625 |
| 0.7581 | 14.0 | 392 | 1.2114 | 0.6042 |
| 0.7801 | 15.0 | 420 | 0.8873 | 0.7292 |
| 0.6041 | 16.0 | 448 | 1.0526 | 0.75 |
| 0.4093 | 17.0 | 476 | 0.8694 | 0.6875 |
| 0.36 | 18.0 | 504 | 0.6712 | 0.7917 |
| 0.3617 | 19.0 | 532 | 0.7221 | 0.7708 |
| 0.2808 | 20.0 | 560 | 0.5851 | 0.8333 |
| 0.192 | 21.0 | 588 | 0.5821 | 0.8125 |
| 0.1924 | 22.0 | 616 | 0.5993 | 0.7917 |
| 0.3129 | 23.0 | 644 | 0.6615 | 0.7708 |
| 0.1542 | 24.0 | 672 | 0.8268 | 0.7292 |
| 0.1038 | 25.0 | 700 | 0.4629 | 0.875 |
| 0.0749 | 26.0 | 728 | 0.5098 | 0.8542 |
| 0.043 | 27.0 | 756 | 0.5493 | 0.8333 |
| 0.0521 | 28.0 | 784 | 0.5119 | 0.8542 |
| 0.0411 | 29.0 | 812 | 0.5280 | 0.875 |
| 0.04 | 30.0 | 840 | 0.5243 | 0.875 |
| 0.0341 | 31.0 | 868 | 0.5478 | 0.875 |
| 0.0313 | 32.0 | 896 | 0.5489 | 0.875 |
| 0.0271 | 33.0 | 924 | 0.5563 | 0.875 |
| 0.0261 | 34.0 | 952 | 0.5735 | 0.875 |
| 0.0223 | 35.0 | 980 | 0.5748 | 0.8542 |
| 0.0235 | 36.0 | 1008 | 0.6004 | 0.8542 |
| 0.0229 | 37.0 | 1036 | 0.6360 | 0.8542 |
| 0.0935 | 38.0 | 1064 | 0.6190 | 0.8542 |
| 0.0215 | 39.0 | 1092 | 0.6138 | 0.8542 |
| 0.0237 | 40.0 | 1120 | 0.6231 | 0.8542 |
| 0.0219 | 41.0 | 1148 | 0.6197 | 0.8542 |
| 0.0236 | 42.0 | 1176 | 0.6207 | 0.8542 |
| 0.021 | 43.0 | 1204 | 0.6189 | 0.8542 |
| 0.021 | 44.0 | 1232 | 0.6204 | 0.8542 |
| 0.0217 | 45.0 | 1260 | 0.6214 | 0.8542 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.11.0
- Datasets 2.1.0
- Tokenizers 0.12.1