learning_rate: 2e-05
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: linear
lr_scheduler_warmup_ratio: 0.06
num_epochs: 30
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision Macro	Recall Macro	F1 Macro	Precision Weighted	Recall Weighted	F1 Weighted
2.237	1.0	60	2.2047	0.1881	0.0530	0.125	0.0598	0.0798	0.1881	0.0900
2.0726	2.0	120	2.0602	0.2947	0.2057	0.2513	0.2044	0.2474	0.2947	0.2444
1.9488	3.0	180	1.8863	0.3793	0.3029	0.3348	0.3007	0.3339	0.3793	0.3332
1.6403	4.0	240	1.8497	0.3417	0.3667	0.3273	0.2970	0.4017	0.3417	0.3225
1.1127	5.0	300	1.9216	0.4013	0.4429	0.3677	0.3477	0.4456	0.4013	0.3764
0.761	6.0	360	2.0790	0.3542	0.3657	0.3428	0.3298	0.3949	0.3542	0.3532
0.5245	7.0	420	2.2272	0.3981	0.4136	0.3682	0.3675	0.4195	0.3981	0.3854
0.3438	8.0	480	2.4195	0.3887	0.4230	0.3895	0.3922	0.4372	0.3887	0.3988
0.216	9.0	540	2.5877	0.4138	0.4259	0.4192	0.4080	0.4500	0.4138	0.4198
0.093	10.0	600	2.9354	0.4169	0.4493	0.3994	0.3987	0.4470	0.4169	0.4130
0.0766	11.0	660	3.0481	0.4138	0.4370	0.4115	0.4170	0.4465	0.4138	0.4215
0.0515	12.0	720	3.4067	0.4389	0.4702	0.4234	0.4185	0.4801	0.4389	0.4318
0.0375	13.0	780	3.6417	0.4075	0.4323	0.3918	0.4003	0.4542	0.4075	0.4169
0.0167	14.0	840	3.7722	0.4138	0.4106	0.4076	0.3942	0.4326	0.4138	0.4091
0.0072	15.0	900	3.9363	0.4138	0.4378	0.4062	0.4102	0.4488	0.4138	0.4185

Kudod/ROBerta-distortion-fold-4-v4

ROBerta-distortion-fold-4-v4