ser_model_adjusted_2023-03-03___2
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: 0.9107
- Accuracy: 0.7443
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: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.78 | 0.97 | 16 | 1.7021 | 0.2595 |
| 1.5997 | 1.97 | 32 | 1.4990 | 0.4313 |
| 1.5263 | 2.97 | 48 | 1.3821 | 0.4580 |
| 1.3081 | 3.97 | 64 | 1.2632 | 0.5 |
| 1.1996 | 4.97 | 80 | 1.2325 | 0.5115 |
| 1.2048 | 5.97 | 96 | 1.1371 | 0.5611 |
| 1.0209 | 6.97 | 112 | 1.0667 | 0.6145 |
| 1.0388 | 7.97 | 128 | 1.0138 | 0.6679 |
| 0.8895 | 8.97 | 144 | 0.9657 | 0.6947 |
| 0.8569 | 9.97 | 160 | 0.9107 | 0.7443 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.12.1
- Datasets 2.7.1
- Tokenizers 0.13.2