Ir para o conteúdo
ModelListadosExterno
Baixar

Red Round Button Big 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: 0.0079 Recall Pressed: 0.9973 Precision Pressed:...

Fontehuggingface.co/ramen-noodels/red_round_button_big_click_detector/tree/1b4a473176bdc8bd6c59b65cd85ba7ab1712b8e0
TipoOtherModelo baseMIT/ast-finetuned-audioset-10-10-0.4593VersãoRevision 1b4a473Bnota

Fonte do modelo

Tipo
Other
Tamanho do arquivo
326 MB
Modelo base
MIT/ast-finetuned-audioset-10-10-0.4593
Versão
vRevision 1b4a473
Atualizado
5 de out. de 2026
Downloads
4
Fonte
Hugging Face

Descrição da fonte

audio_red_round_button_big_val_unnormalized

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: 0.0079
  • Recall Pressed: 0.9973
  • Precision Pressed: 0.9945
  • F1 Pressed: 0.9959
  • False Negative Rate: 0.0027

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
0.01011.0810.00790.99730.99450.99590.0027
0.00542.01620.00650.99730.99450.99590.0027
0.00613.02430.00820.99730.99180.99450.0027
0.00624.03240.01320.99730.99180.99450.0027
0.00435.04050.00710.99730.99180.99450.0027
0.00086.04860.01120.99730.99180.99450.0027
0.00037.05670.01240.99730.99180.99450.0027
0.00038.06480.01310.99730.99180.99450.0027
0.00019.07290.01300.99730.99180.99450.0027
0.000110.08100.01300.99730.99180.99450.0027

Framework versions

  • Transformers 5.8.1
  • Pytorch 2.7.1+cu118
  • Datasets 4.8.5
  • Tokenizers 0.22.2

Fontes

1 fonte
Hugging FacePrimárioVersão 1b4a473176bdc8bd6c59b65cd85ba7ab1712b8e0
huggingface.co/ramen-noodels/red_round_button_big_click_detector/tree/1b4a473176bdc8bd6c59b65cd85ba7ab1712b8e0

Verificado 22 de ago.

Artefatos de modelo

1 artefato

model.safetensors

safetensors · 326 MB · SHA-256 a2a0de502720…fb51 · Hugging Face

Download file

Trechos de fonte

2 trechos
Readme22 de ago.huggingface.co/ramen-noodels/red_round_button_big_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: audio_red_round_button_big_val_unnormalized 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_red_round_button_big_val_unnormalized 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: 0.0079 - Recall Pressed: 0.9973 - Precision Pressed: 0.9945 - F1 Pressed: 0.9959 - False Negative Rate: 0.0027 ## 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 | |:-------------:|:-----:|:----:|:---------------:|:--------------:|:-----------------:|:----------:|:-------------------:| | 0.0101 | 1.0 | 81 | 0.0079 | 0.9973 | 0.9945 | 0.9959 | 0.0027 | | 0.0054 | 2.0 | 162 | 0.0065 | 0.9973 | 0.9945 | 0.9959 | 0.0027 | | 0.0061 | 3.0 | 243 | 0.0082 | 0.9973 | 0.9918 | 0.9945 | 0.0027 | | 0.0062 | 4.0 | 324 | 0.0132 | 0.9973 | 0.9918 | 0.9945 | 0.0027 | | 0.0043 | 5.0 | 405 | 0.0071 | 0.9973 | 0.9918 | 0.9945 | 0.0027 | | 0.0008 | 6.0 | 486 | 0.0112 | 0.9973 | 0.9918 | 0.9945 | 0.0027 | | 0.0003 | 7.0 | 567 | 0.0124 | 0.9973 | 0.9918 | 0.9945 | 0.0027 | | 0.0003 | 8.0 | 648 | 0.0131 | 0.9973 | 0.9918 | 0.9945 | 0.0027 | | 0.0001 | 9.0 | 729 | 0.0130 | 0.9973 | 0.9918 | 0.9945 | 0.0027 | | 0.0001 | 10.0 | 810 | 0.0130 | 0.9973 | 0.9918 | 0.9945 | 0.0027 | ### Framework versions - Transformers 5.8.1 - Pytorch 2.7.1+cu118 - Datasets 4.8.5 - Tokenizers 0.22.2

Note22 de ago.huggingface.co/ramen-noodels/red_round_button_big_click_detector

Source context: 4 downloads · 0 likes · Pipeline audio-classification · Library transformers · Repo ramen-noodels/red_round_button_big_click_detector