learning_rate: 0.001
train_batch_size: 16
eval_batch_size: 16
seed: 42
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: cosine
lr_scheduler_warmup_steps: 0.1
num_epochs: 50
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Macro Acc	Sensitivity	Specificity
No log	0.6667	10	0.7013	0.4167	0.5	1.0	0.0
No log	1.3333	20	0.7200	0.4167	0.5	1.0	0.0
No log	2.0	30	0.6943	0.5833	0.5	0.0	1.0
No log	2.6667	40	0.7045	0.375	0.45	0.9	0.0
0.6992	3.3333	50	0.6906	0.6042	0.5464	0.2	0.8929
0.6992	4.0	60	0.6886	0.6042	0.525	0.05	1.0
0.6992	4.6667	70	0.6845	0.5833	0.5	0.0	1.0
0.6992	5.3333	80	0.7424	0.4167	0.5	1.0	0.0
0.6992	6.0	90	0.6946	0.5208	0.5536	0.75	0.3571
0.7044	6.6667	100	0.7162	0.4167	0.5	1.0	0.0
0.7044	7.3333	110	0.7216	0.4167	0.5	1.0	0.0
0.7044	8.0	120	0.6949	0.4583	0.5143	0.85	0.1786
0.7044	8.6667	130	0.6893	0.5833	0.5	0.0	1.0
0.7044	9.3333	140	0.7238	0.4167	0.5	1.0	0.0
0.6956	10.0	150	0.7030	0.4167	0.5	1.0	0.0
0.6956	10.6667	160	0.7128	0.4167	0.5	1.0	0.0
0.6956	11.3333	170	0.6887	0.5833	0.5	0.0	1.0
0.6956	12.0	180	0.6897	0.5833	0.5	0.0	1.0
0.6956	12.6667	190	0.6972	0.4167	0.5	1.0	0.0
0.6948	13.3333	200	0.6983	0.4167	0.5	1.0	0.0
0.6948	14.0	210	0.7059	0.4167	0.5	1.0	0.0
0.6948	14.6667	220	0.6987	0.4167	0.4929	0.95	0.0357
0.6948	15.3333	230	0.7160	0.4167	0.5	1.0	0.0
0.6948	16.0	240	0.7027	0.4167	0.5	1.0	0.0
0.6932	16.6667	250	0.7034	0.4167	0.5	1.0	0.0
0.6932	17.3333	260	0.6989	0.4375	0.5179	1.0	0.0357
0.6932	18.0	270	0.6892	0.5833	0.5	0.0	1.0
0.6932	18.6667	280	0.7179	0.4167	0.5	1.0	0.0
0.6932	19.3333	290	0.6932	0.5625	0.5893	0.75	0.4286
0.7030	20.0	300	0.6855	0.5833	0.5	0.0	1.0
0.7030	20.6667	310	0.6952	0.4167	0.4929	0.95	0.0357
0.7030	21.3333	320	0.6865	0.5833	0.5	0.0	1.0
0.7030	22.0	330	0.6861	0.5833	0.5	0.0	1.0
0.7030	22.6667	340	0.6844	0.5833	0.5	0.0	1.0
0.6963	23.3333	350	0.6927	0.5417	0.4786	0.1	0.8571
0.6963	24.0	360	0.6957	0.4167	0.4929	0.95	0.0357
0.6963	24.6667	370	0.6908	0.5833	0.5	0.0	1.0
0.6963	25.3333	380	0.6992	0.4167	0.5	1.0	0.0
0.6963	26.0	390	0.6973	0.4375	0.5179	1.0	0.0357
0.6917	26.6667	400	0.7041	0.4167	0.5	1.0	0.0
0.6917	27.3333	410	0.6976	0.4167	0.4929	0.95	0.0357
0.6917	28.0	420	0.6914	0.6042	0.525	0.05	1.0
0.6917	28.6667	430	0.6855	0.5833	0.5	0.0	1.0
0.6917	29.3333	440	0.6885	0.5833	0.5	0.0	1.0
0.6960	30.0	450	0.6936	0.4583	0.45	0.4	0.5
0.6960	30.6667	460	0.7028	0.4167	0.5	1.0	0.0
0.6960	31.3333	470	0.6977	0.4375	0.5179	1.0	0.0357
0.6960	32.0	480	0.6961	0.4167	0.4929	0.95	0.0357
0.6960	32.6667	490	0.6940	0.4167	0.45	0.65	0.25
0.6936	33.3333	500	0.6895	0.5833	0.5	0.0	1.0
0.6936	34.0	510	0.6896	0.5833	0.5	0.0	1.0
0.6936	34.6667	520	0.6915	0.6042	0.525	0.05	1.0
0.6936	35.3333	530	0.6916	0.625	0.55	0.1	1.0
0.6936	36.0	540	0.6894	0.5833	0.5	0.0	1.0
0.6918	36.6667	550	0.6906	0.6042	0.525	0.05	1.0
0.6918	37.3333	560	0.6909	0.6042	0.525	0.05	1.0
0.6918	38.0	570	0.6925	0.5208	0.4679	0.15	0.7857
0.6918	38.6667	580	0.6941	0.4167	0.45	0.65	0.25
0.6918	39.3333	590	0.6949	0.4167	0.4857	0.9	0.0714
0.6928	40.0	600	0.6957	0.4167	0.4929	0.95	0.0357
0.6928	40.6667	610	0.6989	0.4375	0.5179	1.0	0.0357
0.6928	41.3333	620	0.6982	0.4375	0.5179	1.0	0.0357
0.6928	42.0	630	0.6974	0.4167	0.4929	0.95	0.0357
0.6928	42.6667	640	0.6979	0.4167	0.4929	0.95	0.0357
0.6903	43.3333	650	0.6977	0.4167	0.4929	0.95	0.0357
0.6903	44.0	660	0.6978	0.4167	0.4929	0.95	0.0357
0.6903	44.6667	670	0.6974	0.4167	0.4929	0.95	0.0357
0.6903	45.3333	680	0.6972	0.4167	0.4929	0.95	0.0357
0.6903	46.0	690	0.6972	0.4167	0.4929	0.95	0.0357
0.6899	46.6667	700	0.6967	0.4167	0.4929	0.95	0.0357
0.6899	47.3333	710	0.6965	0.4167	0.4929	0.95	0.0357
0.6899	48.0	720	0.6965	0.4167	0.4929	0.95	0.0357
0.6899	48.6667	730	0.6965	0.4167	0.4929	0.95	0.0357
0.6899	49.3333	740	0.6965	0.4167	0.4929	0.95	0.0357
0.6893	50.0	750	0.6965	0.4167	0.4929	0.95	0.0357

Framework versions

Transformers 5.3.0
Pytorch 2.10.0+cu128
Datasets 4.0.0
Tokenizers 0.22.2

Srutid/wav2vec2-base-dm32

wav2vec2-base-dm32

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Srutid/
wav2vec2-base-dm32