hubert-base-superb-er-3kfoldfull40-finetuned-bmd-20250824_101658-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:

• Loss: 0.6934
• Accuracy: 0.6667
• F1: 0.6458

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: 3e-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: 40
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	1.0	3	1.1019	0.3333	0.1720
No log	2.0	6	1.1002	0.3333	0.1667
No log	3.0	9	1.0970	0.3333	0.1667
1.0998	4.0	12	1.0926	0.4167	0.3259
1.0998	5.0	15	1.0870	0.5417	0.4938
1.0998	6.0	18	1.0796	0.5417	0.4938
1.0876	7.0	21	1.0663	0.5833	0.5352
1.0876	8.0	24	1.0395	0.5417	0.5222
1.0876	9.0	27	1.0070	0.5417	0.4444
1.0392	10.0	30	0.9650	0.625	0.5944
1.0392	11.0	33	0.9296	0.625	0.5944
1.0392	12.0	36	0.9093	0.6667	0.6258
1.0392	13.0	39	0.9176	0.5833	0.4722
0.951	14.0	42	0.8967	0.5417	0.4474
0.951	15.0	45	0.8583	0.5833	0.4889
0.951	16.0	48	0.8313	0.625	0.5017
0.8834	17.0	51	0.8967	0.5417	0.4444
0.8834	18.0	54	0.8661	0.625	0.5560
0.8834	19.0	57	0.8021	0.625	0.5017
0.8182	20.0	60	0.7866	0.625	0.5017
0.8182	21.0	63	0.7642	0.625	0.5017
0.8182	22.0	66	0.7667	0.625	0.5017
0.8182	23.0	69	0.7610	0.625	0.5017
0.7367	24.0	72	0.7243	0.75	0.7218
0.7367	25.0	75	0.7803	0.6667	0.5910
0.7367	26.0	78	0.7885	0.625	0.5591
0.7063	27.0	81	0.6803	0.75	0.7222
0.7063	28.0	84	0.6617	0.75	0.7222
0.7063	29.0	87	0.7135	0.625	0.5873
0.6549	30.0	90	0.7171	0.625	0.5915
0.6549	31.0	93	0.6819	0.6667	0.6458
0.6549	32.0	96	0.6974	0.625	0.5915
0.6549	33.0	99	0.6980	0.625	0.5915
0.6135	34.0	102	0.6699	0.6667	0.6458
0.6135	35.0	105	0.6568	0.7083	0.6963
0.6135	36.0	108	0.6712	0.7083	0.6963
0.6178	37.0	111	0.6867	0.6667	0.6458
0.6178	38.0	114	0.6959	0.6667	0.6458
0.6178	39.0	117	0.6971	0.625	0.5915
0.5819	40.0	120	0.6934	0.6667	0.6458

Framework versions

• Transformers 4.55.2
• Pytorch 2.8.0+cu126
• Datasets 3.6.0
• Tokenizers 0.21.4