Aller au contenu
ModelPrésentéExterne
Télécharger

Ast Delta Xyz Button Click Detector

This model is a fine-tuned version of MIT/ast-finetuned-audioset-10-10-0.4593 on an unknown dataset. It achieves the following results on the evaluation set: Loss: 0.0225 Recall Pressed: 0.9791 Precision Pressed:...

Sourcehuggingface.co/ramen-noodels/ast_delta_xyz_button_click_detector/tree/f5670afdef6eb583002fe8e49b70ed3d8085a565
TypeOtherModèle de baseMIT/ast-finetuned-audioset-10-10-0.4593VersionRevision f5670afBClasse

Source du modèle

Type
Other
Taille du fichier
326 MB
Modèle de base
MIT/ast-finetuned-audioset-10-10-0.4593
Version
vRevision f5670af
Mis à jour
5 oct. 2026
Téléchargements
6
Source
Hugging Face

Description de la source

ast_delta_xyz_button_click_detector

This model is a fine-tuned version of MIT/ast-finetuned-audioset-10-10-0.4593 on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0225
  • Recall Pressed: 0.9791
  • Precision Pressed: 0.9880
  • F1 Pressed: 0.9835
  • False Negative Rate: 0.0209

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: 82
  • eval_batch_size: 82
  • seed: 42
  • optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • num_epochs: 10

Training results

Training LossEpochStepValidation LossRecall PressedPrecision PressedF1 PressedFalse Negative Rate
0.01941.01180.02290.97010.98480.97740.0299
0.00592.02360.02280.97310.98490.97900.0269
0.00823.03540.02240.98210.98210.98210.0179
0.00554.04720.02940.96120.99690.97870.0388
0.00885.05900.02250.97910.98800.98350.0209
0.00136.07080.03210.98210.97630.97920.0179
0.00407.08260.03180.97910.98500.98200.0209
0.00148.09440.03340.97910.98500.98200.0209
0.00029.010620.03370.97910.98500.98200.0209
0.000510.011800.03390.97910.98500.98200.0209

Framework versions

  • Transformers 5.7.0
  • Pytorch 2.7.1+cu118
  • Datasets 4.8.5
  • Tokenizers 0.22.2

Sources

1 source
Hugging FacePrincipalVersion f5670afdef6eb583002fe8e49b70ed3d8085a565
huggingface.co/ramen-noodels/ast_delta_xyz_button_click_detector/tree/f5670afdef6eb583002fe8e49b70ed3d8085a565

Vérifié 22 août

Artefacts du modèle

1 artefact

model.safetensors

safetensors · 326 MB · SHA-256 06cc0b690f2f…c5c2 · Hugging Face

Download file

Extraits de sources

2 extraits
Readme22 aoûthuggingface.co/ramen-noodels/ast_delta_xyz_button_click_detector

--- library_name: transformers license: bsd-3-clause base_model: MIT/ast-finetuned-audioset-10-10-0.4593 tags: - generated_from_trainer model-index: - name: ast_delta_xyz_button_click_detector 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. --> # ast_delta_xyz_button_click_detector This model is a fine-tuned version of [MIT/ast-finetuned-audioset-10-10-0.4593](https://huggingface.co/MIT/ast-finetuned-audioset-10-10-0.4593) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.0225 - Recall Pressed: 0.9791 - Precision Pressed: 0.9880 - F1 Pressed: 0.9835 - False Negative Rate: 0.0209 ## 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: 82 - eval_batch_size: 82 - seed: 42 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: linear - num_epochs: 10 ### Training results | Training Loss | Epoch | Step | Validation Loss | Recall Pressed | Precision Pressed | F1 Pressed | False Negative Rate | |:-------------:|:-----:|:----:|:---------------:|:--------------:|:-----------------:|:----------:|:-------------------:| | 0.0194 | 1.0 | 118 | 0.0229 | 0.9701 | 0.9848 | 0.9774 | 0.0299 | | 0.0059 | 2.0 | 236 | 0.0228 | 0.9731 | 0.9849 | 0.9790 | 0.0269 | | 0.0082 | 3.0 | 354 | 0.0224 | 0.9821 | 0.9821 | 0.9821 | 0.0179 | | 0.0055 | 4.0 | 472 | 0.0294 | 0.9612 | 0.9969 | 0.9787 | 0.0388 | | 0.0088 | 5.0 | 590 | 0.0225 | 0.9791 | 0.9880 | 0.9835 | 0.0209 | | 0.0013 | 6.0 | 708 | 0.0321 | 0.9821 | 0.9763 | 0.9792 | 0.0179 | | 0.0040 | 7.0 | 826 | 0.0318 | 0.9791 | 0.9850 | 0.9820 | 0.0209 | | 0.0014 | 8.0 | 944 | 0.0334 | 0.9791 | 0.9850 | 0.9820 | 0.0209 | | 0.0002 | 9.0 | 1062 | 0.0337 | 0.9791 | 0.9850 | 0.9820 | 0.0209 | | 0.0005 | 10.0 | 1180 | 0.0339 | 0.9791 | 0.9850 | 0.9820 | 0.0209 | ### Framework versions - Transformers 5.7.0 - Pytorch 2.7.1+cu118 - Datasets 4.8.5 - Tokenizers 0.22.2

Note22 aoûthuggingface.co/ramen-noodels/ast_delta_xyz_button_click_detector

Source context: 6 downloads · 0 likes · Pipeline audio-classification · Library transformers · Repo ramen-noodels/ast_delta_xyz_button_click_detector