learning_rate: 0.0001
train_batch_size: 64
eval_batch_size: 64
seed: 42
optimizer: Use OptimizerNames.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_steps: 500
num_epochs: 50

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
No log	1.0	346	0.4126	0.8503	0.7036	0.4950	0.5081
0.4889	2.0	692	0.3922	0.8570	0.6838	0.5398	0.5648
0.3681	3.0	1038	0.3937	0.8578	0.7163	0.5351	0.5644
0.3681	4.0	1384	0.3841	0.8612	0.7138	0.5493	0.5819
0.3623	5.0	1730	0.3803	0.8604	0.7177	0.5594	0.5889
0.3531	6.0	2076	0.3853	0.8600	0.7154	0.5622	0.5874
0.3531	7.0	2422	0.3766	0.8615	0.7032	0.5593	0.5977
0.3511	8.0	2768	0.3781	0.8638	0.7112	0.5608	0.5921
0.3461	9.0	3114	0.3714	0.8623	0.6954	0.5633	0.5965
0.3461	10.0	3460	0.3797	0.8604	0.7156	0.5610	0.5816
0.3513	11.0	3806	0.3692	0.8623	0.7062	0.5686	0.6065
0.3409	12.0	4152	0.3738	0.8608	0.7001	0.5825	0.6045
0.3409	13.0	4498	0.3703	0.8675	0.7240	0.5729	0.6132
0.3423	14.0	4844	0.3690	0.8641	0.7081	0.5661	0.6042
0.3351	15.0	5190	0.3647	0.8649	0.7093	0.5740	0.6088
0.3401	16.0	5536	0.3697	0.8656	0.7381	0.5579	0.5959
0.3401	17.0	5882	0.3651	0.8638	0.7005	0.5794	0.6073
0.3278	18.0	6228	0.3691	0.8630	0.7236	0.5702	0.5959

trhgquan/visobert-human-tl-seed-24

visobert-human-tl-seed-24