metadata
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: wav2vec2-base-intent-classification-ori-f1
results: []
wav2vec2-base-intent-classification-ori-f1
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: 0.7879
- Accuracy: 0.8125
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 |
|---|---|---|---|---|
| 2.1936 | 1.0 | 28 | 2.1721 | 0.2708 |
| 2.1251 | 2.0 | 56 | 2.1119 | 0.2708 |
| 2.1113 | 3.0 | 84 | 2.0547 | 0.2708 |
| 1.9978 | 4.0 | 112 | 2.0695 | 0.3125 |
| 1.9134 | 5.0 | 140 | 1.7726 | 0.5 |
| 1.6926 | 6.0 | 168 | 1.6564 | 0.4167 |
| 1.7755 | 7.0 | 196 | 1.5987 | 0.4375 |
| 1.5199 | 8.0 | 224 | 1.5848 | 0.3958 |
| 1.2484 | 9.0 | 252 | 1.5180 | 0.5 |
| 1.114 | 10.0 | 280 | 1.5078 | 0.4375 |
| 1.281 | 11.0 | 308 | 1.3506 | 0.4792 |
| 1.2449 | 12.0 | 336 | 1.2551 | 0.5208 |
| 1.1664 | 13.0 | 364 | 1.6051 | 0.4792 |
| 0.9792 | 14.0 | 392 | 1.2325 | 0.5417 |
| 0.8464 | 15.0 | 420 | 1.1644 | 0.4583 |
| 0.6899 | 16.0 | 448 | 1.0157 | 0.7083 |
| 0.6696 | 17.0 | 476 | 1.1179 | 0.6042 |
| 0.7003 | 18.0 | 504 | 1.0003 | 0.625 |
| 0.5867 | 19.0 | 532 | 0.8064 | 0.7292 |
| 0.539 | 20.0 | 560 | 0.8617 | 0.6875 |
| 0.5039 | 21.0 | 588 | 1.0144 | 0.6458 |
| 0.4377 | 22.0 | 616 | 0.8810 | 0.6458 |
| 0.4507 | 23.0 | 644 | 0.8612 | 0.7292 |
| 0.274 | 24.0 | 672 | 0.8041 | 0.7083 |
| 0.3679 | 25.0 | 700 | 0.9090 | 0.6667 |
| 0.2737 | 26.0 | 728 | 0.7768 | 0.75 |
| 0.2019 | 27.0 | 756 | 0.7486 | 0.7708 |
| 0.1841 | 28.0 | 784 | 0.9169 | 0.75 |
| 0.1237 | 29.0 | 812 | 0.6935 | 0.7917 |
| 0.1174 | 30.0 | 840 | 0.7879 | 0.8125 |
| 0.1159 | 31.0 | 868 | 0.7981 | 0.7917 |
| 0.0878 | 32.0 | 896 | 0.8157 | 0.7917 |
| 0.0884 | 33.0 | 924 | 0.8122 | 0.7708 |
| 0.0627 | 34.0 | 952 | 0.8276 | 0.7917 |
| 0.0549 | 35.0 | 980 | 0.8650 | 0.7708 |
| 0.0565 | 36.0 | 1008 | 0.8582 | 0.7708 |
| 0.0533 | 37.0 | 1036 | 0.9108 | 0.7708 |
| 0.0439 | 38.0 | 1064 | 0.9221 | 0.7708 |
| 0.0435 | 39.0 | 1092 | 0.8844 | 0.7708 |
| 0.0517 | 40.0 | 1120 | 0.8865 | 0.7708 |
| 0.0411 | 41.0 | 1148 | 0.9118 | 0.7708 |
| 0.045 | 42.0 | 1176 | 0.9109 | 0.7708 |
| 0.0455 | 43.0 | 1204 | 0.9194 | 0.7708 |
| 0.0435 | 44.0 | 1232 | 0.9221 | 0.7708 |
| 0.0422 | 45.0 | 1260 | 0.9215 | 0.7708 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.11.0
- Datasets 2.1.0
- Tokenizers 0.12.1