Downloads · 30 days
5
9% of all-time downloads
audio-abuse-feature
callmesan/audio-abuse-feature
audio-abuse-feature is a audio classification model from callmesan. Use it for the audio classification task on the model card, and read the license before you ship it in a product. It is set up for transformers.
should probably proofread and complete it, then remove this comment. --
All-time downloads
54
Public
Parameters
307M
1.2 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors1.2 GB · 100%
At a glance
- Task
- Audio Classification
- Library
- transformers
- Model type
- whisper
- Access
- Public
- Created
- Jun 16, 2024
- Updated
- Jun 16, 2024
- SHA
- f98a6487
Base models
Model card
From the Hugging Face model README
<!-- 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-abuse-feature
This model is a fine-tuned version of HariprasathSB/indic-whisper-vulnerable on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4489
- Accuracy: 0.8814
- Macro Precision: 0.8557
- Macro Recall: 0.8472
- Macro F1-score: 0.8513
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: 1e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.01
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Macro Precision | Macro Recall | Macro F1-score |
|---|---|---|---|---|---|---|---|
| 0.4633 | 0.4367 | 50 | 0.3753 | 0.8327 | 0.8321 | 0.8314 | 0.8317 |
| 0.345 | 0.8734 | 100 | 0.4170 | 0.8241 | 0.8612 | 0.8126 | 0.8150 |
| 0.2592 | 1.3100 | 150 | 0.3357 | 0.8512 | 0.8506 | 0.8502 | 0.8504 |
| 0.2097 | 1.7467 | 200 | 0.3142 | 0.8758 | 0.8757 | 0.8744 | 0.8749 |
| 0.1545 | 2.1834 | 250 | 0.3551 | 0.8721 | 0.8713 | 0.8718 | 0.8715 |
| 0.0829 | 2.6201 | 300 | 0.3916 | 0.8795 | 0.8797 | 0.8778 | 0.8786 |
| 0.0944 | 3.0568 | 350 | 0.4137 | 0.8721 | 0.8714 | 0.8730 | 0.8718 |
| 0.0416 | 3.4934 | 400 | 0.5350 | 0.8659 | 0.8677 | 0.8631 | 0.8646 |
| 0.0469 | 3.9301 | 450 | 0.5129 | 0.8733 | 0.8727 | 0.8726 | 0.8727 |
| 0.0247 | 4.3668 | 500 | 0.5543 | 0.8708 | 0.8713 | 0.8689 | 0.8698 |
| 0.0208 | 4.8035 | 550 | 0.5611 | 0.8696 | 0.8691 | 0.8688 | 0.8689 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.1.2
- Datasets 2.19.2
- Tokenizers 0.19.1
- Task
- Audio Classification
- Library
- transformers
- Type
- whisper
- Created
- Jun 16, 2024
- Updated
- Jun 16, 2024