Aller au contenu
ModelPrésentéExterne
Télécharger

Audio Cls Hhd

This model is a fine-tuned version of Kkonjeong/wav2vec2-base-korean on an unknown dataset. It achieves the following results on the evaluation set: Loss: 0.9499 Accuracy: 0.8739

Sourcehuggingface.co/alwaysnotavailable/audio_cls_hhd/tree/cb86a002e33a311b57bd56840a06a9b0a0fad81e
TypeOtherModèle de baseKkonjeong/wav2vec2-base-koreanVersionRevision cb86a00

Source du modèle

Type
Other
Taille du fichier
361 MB
Modèle de base
Kkonjeong/wav2vec2-base-korean
Version
vRevision cb86a00
Mis à jour
19 sept. 2026
Téléchargements
3
Source
Hugging Face

Description de la source

audio_cls_hhd

This model is a fine-tuned version of Kkonjeong/wav2vec2-base-korean on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.9499
  • Accuracy: 0.8739

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: 0.0001
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 32
  • optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • num_epochs: 10
  • mixed_precision_training: Native AMP

Training results

Training LossEpochStepValidation LossAccuracy
No log1.0152.56640.1597
No log2.0302.37000.2269
No log3.0452.26250.3361
No log4.0601.89200.5966
No log5.0751.61110.6975
No log6.0901.37860.8067
No log7.01051.20790.8319
No log8.01201.08530.8403
No log9.01350.98450.8571
No log10.01500.94990.8739

Framework versions

  • Transformers 4.57.3
  • Pytorch 2.9.0+cu126
  • Datasets 4.0.0
  • Tokenizers 0.22.1

Sources

1 source
Hugging FacePrincipalVersion cb86a002e33a311b57bd56840a06a9b0a0fad81e
huggingface.co/alwaysnotavailable/audio_cls_hhd/tree/cb86a002e33a311b57bd56840a06a9b0a0fad81e

Vérifié 19 sept.

Artefacts du modèle

1 artefact

model.safetensors

safetensors · 361 MB · SHA-256 9e7d3f872ea9…c39c · Hugging Face

Télécharger

Extraits de sources

2 extraits
Readme19 sept.huggingface.co/alwaysnotavailable/audio_cls_hhd

--- library_name: transformers base_model: Kkonjeong/wav2vec2-base-korean tags: - generated_from_trainer metrics: - accuracy model-index: - name: audio_cls_hhd 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_cls_hhd This model is a fine-tuned version of [Kkonjeong/wav2vec2-base-korean](https://huggingface.co/Kkonjeong/wav2vec2-base-korean) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.9499 - Accuracy: 0.8739 ## 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: 0.0001 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 32 - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: linear - num_epochs: 10 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | |:-------------:|:-----:|:----:|:---------------:|:--------:| | No log | 1.0 | 15 | 2.5664 | 0.1597 | | No log | 2.0 | 30 | 2.3700 | 0.2269 | | No log | 3.0 | 45 | 2.2625 | 0.3361 | | No log | 4.0 | 60 | 1.8920 | 0.5966 | | No log | 5.0 | 75 | 1.6111 | 0.6975 | | No log | 6.0 | 90 | 1.3786 | 0.8067 | | No log | 7.0 | 105 | 1.2079 | 0.8319 | | No log | 8.0 | 120 | 1.0853 | 0.8403 | | No log | 9.0 | 135 | 0.9845 | 0.8571 | | No log | 10.0 | 150 | 0.9499 | 0.8739 | ### Framework versions - Transformers 4.57.3 - Pytorch 2.9.0+cu126 - Datasets 4.0.0 - Tokenizers 0.22.1

Note19 sept.huggingface.co/alwaysnotavailable/audio_cls_hhd

Source context: 3 downloads · 0 likes · Pipeline audio-classification · Library transformers · Repo alwaysnotavailable/audio_cls_hhd