hubert-base-superb-er-3kfoldfull20-V2-finetuned-bmd-20250824_121113-LOSO-section-out1

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

Model description

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

learning_rate: 2e-05
train_batch_size: 8
eval_batch_size: 8
seed: 1968
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
lr_scheduler_warmup_ratio: 0.1
num_epochs: 20
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	1.0	3	1.1067	0.3333	0.1667
No log	2.0	6	1.1035	0.3333	0.1667
No log	3.0	9	1.0988	0.375	0.2445
1.1017	4.0	12	1.0946	0.4583	0.3644
1.1017	5.0	15	1.0908	0.4583	0.4144
1.1017	6.0	18	1.0870	0.4583	0.4273
1.0899	7.0	21	1.0826	0.4583	0.4472
1.0899	8.0	24	1.0774	0.4167	0.4151
1.0899	9.0	27	1.0704	0.4583	0.4750
1.073	10.0	30	1.0613	0.5	0.5162
1.073	11.0	33	1.0502	0.5833	0.5874
1.073	12.0	36	1.0368	0.5833	0.5874
1.073	13.0	39	1.0211	0.625	0.6242
1.0468	14.0	42	1.0042	0.6667	0.6583
1.0468	15.0	45	0.9882	0.6667	0.6631
1.0468	16.0	48	0.9770	0.625	0.6114
1.0055	17.0	51	0.9722	0.5833	0.5756
1.0055	18.0	54	0.9654	0.625	0.6144
1.0055	19.0	57	0.9563	0.5833	0.5569
0.9735	20.0	60	0.9530	0.5833	0.5569