audio_cls_1016

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: 0.8282
• Accuracy: 0.9328

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	2.6818	0.0588
No log	2.0	30	2.5366	0.1849
No log	3.0	45	2.2080	0.4538
No log	4.0	60	1.8926	0.6387
No log	5.0	75	1.6790	0.6723
No log	6.0	90	1.3159	0.8824
No log	7.0	105	1.0967	0.9076
No log	8.0	120	0.9650	0.9160
No log	9.0	135	0.8733	0.9160
No log	10.0	150	0.8282	0.9328

Framework versions

• Transformers 5.18.0
• Pytorch 2.11.0+cu130
• Datasets 4.8.5
• Tokenizers 0.23.2