|
--- |
|
license: apache-2.0 |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: Audio_CREMA |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# Audio_CREMA |
|
|
|
This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.8274 |
|
- Accuracy: 0.7909 |
|
- Weighted f1: 0.7913 |
|
- Micro f1: 0.7909 |
|
- Macro f1: 0.7909 |
|
- Weighted recall: 0.7909 |
|
- Micro recall: 0.7909 |
|
- Macro recall: 0.7945 |
|
- Weighted precision: 0.8014 |
|
- Micro precision: 0.7909 |
|
- Macro precision: 0.7976 |
|
|
|
## 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 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 15 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Weighted f1 | Micro f1 | Macro f1 | Weighted recall | Micro recall | Macro recall | Weighted precision | Micro precision | Macro precision | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:-----------:|:--------:|:--------:|:---------------:|:------------:|:------------:|:------------------:|:---------------:|:---------------:| |
|
| 1.0002 | 1.0 | 55 | 1.0265 | 0.5477 | 0.5159 | 0.5477 | 0.5169 | 0.5477 | 0.5477 | 0.5486 | 0.5338 | 0.5477 | 0.5341 | |
|
| 0.8613 | 2.0 | 110 | 0.9630 | 0.5795 | 0.5540 | 0.5795 | 0.5558 | 0.5795 | 0.5795 | 0.5825 | 0.5737 | 0.5795 | 0.5718 | |
|
| 0.7676 | 3.0 | 165 | 0.8474 | 0.6659 | 0.6655 | 0.6659 | 0.6624 | 0.6659 | 0.6659 | 0.6629 | 0.6746 | 0.6659 | 0.6713 | |
|
| 0.6886 | 4.0 | 220 | 0.9269 | 0.6318 | 0.6203 | 0.6318 | 0.6198 | 0.6318 | 0.6318 | 0.6351 | 0.6581 | 0.6318 | 0.6506 | |
|
| 0.6536 | 5.0 | 275 | 0.7114 | 0.7341 | 0.7364 | 0.7341 | 0.7350 | 0.7341 | 0.7341 | 0.7360 | 0.7472 | 0.7341 | 0.7424 | |
|
| 0.4429 | 6.0 | 330 | 0.7026 | 0.7432 | 0.7419 | 0.7432 | 0.7406 | 0.7432 | 0.7432 | 0.7425 | 0.7417 | 0.7432 | 0.7399 | |
|
| 0.3755 | 7.0 | 385 | 0.6925 | 0.7682 | 0.7679 | 0.7682 | 0.7680 | 0.7682 | 0.7682 | 0.7717 | 0.7743 | 0.7682 | 0.7712 | |
|
| 0.3603 | 8.0 | 440 | 0.7445 | 0.7591 | 0.7608 | 0.7591 | 0.7604 | 0.7591 | 0.7591 | 0.7610 | 0.7740 | 0.7591 | 0.7716 | |
|
| 0.296 | 9.0 | 495 | 0.7235 | 0.7614 | 0.7577 | 0.7614 | 0.7590 | 0.7614 | 0.7614 | 0.7669 | 0.7718 | 0.7614 | 0.7685 | |
|
| 0.2854 | 10.0 | 550 | 0.6988 | 0.7818 | 0.7832 | 0.7818 | 0.7824 | 0.7818 | 0.7818 | 0.7840 | 0.7923 | 0.7818 | 0.7891 | |
|
| 0.2655 | 11.0 | 605 | 0.7530 | 0.7568 | 0.7526 | 0.7568 | 0.7539 | 0.7568 | 0.7568 | 0.7618 | 0.7632 | 0.7568 | 0.7605 | |
|
| 0.1359 | 12.0 | 660 | 0.7503 | 0.7955 | 0.7974 | 0.7955 | 0.7972 | 0.7955 | 0.7955 | 0.7997 | 0.8110 | 0.7955 | 0.8069 | |
|
| 0.1258 | 13.0 | 715 | 0.8318 | 0.7659 | 0.7634 | 0.7659 | 0.7638 | 0.7659 | 0.7659 | 0.7710 | 0.7808 | 0.7659 | 0.7767 | |
|
| 0.0731 | 14.0 | 770 | 0.8758 | 0.7727 | 0.7718 | 0.7727 | 0.7715 | 0.7727 | 0.7727 | 0.7766 | 0.7883 | 0.7727 | 0.7846 | |
|
| 0.0676 | 15.0 | 825 | 0.8274 | 0.7909 | 0.7913 | 0.7909 | 0.7909 | 0.7909 | 0.7909 | 0.7945 | 0.8014 | 0.7909 | 0.7976 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.18.0 |
|
- Pytorch 1.11.0 |
|
- Datasets 2.1.0 |
|
- Tokenizers 0.12.1 |
|
|