metadata
license: apache-2.0
base_model: facebook/wav2vec2-base
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: SER_model_xapiens
results: []
SER_model_xapiens
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: 0.9329
- Accuracy: 0.6667
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: 64
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 100
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 1.0 | 1 | 1.6121 | 0.2167 |
| No log | 2.0 | 3 | 1.6109 | 0.2333 |
| No log | 3.0 | 5 | 1.6090 | 0.25 |
| No log | 4.0 | 6 | 1.6078 | 0.2667 |
| No log | 5.0 | 7 | 1.6058 | 0.2833 |
| No log | 6.0 | 9 | 1.6006 | 0.35 |
| 0.8024 | 7.0 | 11 | 1.5947 | 0.35 |
| 0.8024 | 8.0 | 12 | 1.5894 | 0.3667 |
| 0.8024 | 9.0 | 13 | 1.5809 | 0.35 |
| 0.8024 | 10.0 | 15 | 1.5609 | 0.3667 |
| 0.8024 | 11.0 | 17 | 1.5417 | 0.4 |
| 0.8024 | 12.0 | 18 | 1.5310 | 0.4167 |
| 0.8024 | 13.0 | 19 | 1.5147 | 0.4333 |
| 0.7692 | 14.0 | 21 | 1.4847 | 0.4833 |
| 0.7692 | 15.0 | 23 | 1.4592 | 0.5167 |
| 0.7692 | 16.0 | 24 | 1.4384 | 0.5 |
| 0.7692 | 17.0 | 25 | 1.4147 | 0.5 |
| 0.7692 | 18.0 | 27 | 1.3769 | 0.5167 |
| 0.7692 | 19.0 | 29 | 1.3544 | 0.5167 |
| 0.6858 | 20.0 | 30 | 1.3325 | 0.55 |
| 0.6858 | 21.0 | 31 | 1.3084 | 0.55 |
| 0.6858 | 22.0 | 33 | 1.2617 | 0.5667 |
| 0.6858 | 23.0 | 35 | 1.2359 | 0.6 |
| 0.6858 | 24.0 | 36 | 1.2245 | 0.5833 |
| 0.6858 | 25.0 | 37 | 1.2005 | 0.6333 |
| 0.6858 | 26.0 | 39 | 1.1695 | 0.6333 |
| 0.5744 | 27.0 | 41 | 1.1539 | 0.6 |
| 0.5744 | 28.0 | 42 | 1.1370 | 0.6333 |
| 0.5744 | 29.0 | 43 | 1.1075 | 0.65 |
| 0.5744 | 30.0 | 45 | 1.0617 | 0.6833 |
| 0.5744 | 31.0 | 47 | 1.0745 | 0.6167 |
| 0.5744 | 32.0 | 48 | 1.0653 | 0.5833 |
| 0.5744 | 33.0 | 49 | 1.0667 | 0.5833 |
| 0.4853 | 34.0 | 51 | 1.0611 | 0.6 |
| 0.4853 | 35.0 | 53 | 1.1032 | 0.6333 |
| 0.4853 | 36.0 | 54 | 1.0963 | 0.6167 |
| 0.4853 | 37.0 | 55 | 1.0748 | 0.6167 |
| 0.4853 | 38.0 | 57 | 1.0446 | 0.6333 |
| 0.4853 | 39.0 | 59 | 1.0205 | 0.6167 |
| 0.4276 | 40.0 | 60 | 1.0241 | 0.6333 |
| 0.4276 | 41.0 | 61 | 1.0499 | 0.6167 |
| 0.4276 | 42.0 | 63 | 1.0788 | 0.6167 |
| 0.4276 | 43.0 | 65 | 1.0358 | 0.6167 |
| 0.4276 | 44.0 | 66 | 1.0175 | 0.6 |
| 0.4276 | 45.0 | 67 | 1.0082 | 0.65 |
| 0.4276 | 46.0 | 69 | 0.9862 | 0.65 |
| 0.393 | 47.0 | 71 | 0.9763 | 0.6 |
| 0.393 | 48.0 | 72 | 0.9660 | 0.6 |
| 0.393 | 49.0 | 73 | 0.9506 | 0.6167 |
| 0.393 | 50.0 | 75 | 0.9394 | 0.65 |
| 0.393 | 51.0 | 77 | 0.9393 | 0.7 |
| 0.393 | 52.0 | 78 | 0.9463 | 0.65 |
| 0.393 | 53.0 | 79 | 0.9643 | 0.6167 |
| 0.3588 | 54.0 | 81 | 0.9750 | 0.6333 |
| 0.3588 | 55.0 | 83 | 0.9558 | 0.65 |
| 0.3588 | 56.0 | 84 | 0.9494 | 0.6333 |
| 0.3588 | 57.0 | 85 | 0.9450 | 0.6167 |
| 0.3588 | 58.0 | 87 | 0.9466 | 0.6 |
| 0.3588 | 59.0 | 89 | 0.9385 | 0.6 |
| 0.3372 | 60.0 | 90 | 0.9343 | 0.6667 |
| 0.3372 | 61.0 | 91 | 0.9309 | 0.6667 |
| 0.3372 | 62.0 | 93 | 0.9286 | 0.65 |
| 0.3372 | 63.0 | 95 | 0.9301 | 0.65 |
| 0.3372 | 64.0 | 96 | 0.9313 | 0.65 |
| 0.3372 | 65.0 | 97 | 0.9314 | 0.6667 |
| 0.3372 | 66.0 | 99 | 0.9327 | 0.6667 |
| 0.3246 | 66.6667 | 100 | 0.9329 | 0.6667 |
Framework versions
- Transformers 4.42.0.dev0
- Pytorch 2.3.0
- Datasets 2.19.2.dev0
- Tokenizers 0.19.1