hubert-base-ls960-finetuned-bmd-V1-20241110_112632-LOSO-section-out1

This model is a fine-tuned version of facebook/hubert-base-ls960 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.0562
• Accuracy: 0.4828
• F1: 0.4858

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: 1968
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.3
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	0.8571	3	1.0983	0.3103	0.2990
No log	2.0	7	1.0947	0.4828	0.4047
1.0979	2.8571	10	1.0935	0.3793	0.2808
1.0979	4.0	14	1.0766	0.4138	0.3723
1.0979	4.8571	17	1.0710	0.4483	0.3853
1.064	6.0	21	1.0185	0.4483	0.4447
1.064	6.8571	24	1.0051	0.4483	0.4
1.064	8.0	28	1.0496	0.4828	0.4425
0.9118	8.8571	31	1.0056	0.5172	0.5110
0.9118	10.0	35	1.1071	0.5172	0.5232
0.9118	10.8571	38	1.0749	0.4828	0.4849
0.8278	12.0	42	1.0614	0.4483	0.4537
0.8278	12.8571	45	1.0562	0.4828	0.4858

Framework versions

• Transformers 4.46.2
• Pytorch 2.5.0+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3