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	0.4172	0.8484	0.6957	0.4805	0.4990
0.497	2.0	692	0.3968	0.8578	0.7050	0.5336	0.5672
0.3758	3.0	1038	0.3953	0.8567	0.7163	0.5266	0.5627
0.3758	4.0	1384	0.3873	0.8574	0.7102	0.5456	0.5754
0.3707	5.0	1730	0.3850	0.8608	0.7230	0.5509	0.5847
0.3635	6.0	2076	0.3887	0.8597	0.7348	0.5497	0.5848
0.3635	7.0	2422	0.3791	0.8604	0.6993	0.5466	0.5875
0.363	8.0	2768	0.3768	0.8619	0.7200	0.5556	0.5937
0.3553	9.0	3114	0.3732	0.8619	0.6978	0.5712	0.6092
0.3553	10.0	3460	0.3880	0.8570	0.7013	0.5603	0.5806
0.3612	11.0	3806	0.3748	0.8608	0.7050	0.5532	0.5943
0.3512	12.0	4152	0.3771	0.8619	0.7057	0.5904	0.6156
0.3512	13.0	4498	0.3745	0.8626	0.7205	0.5584	0.5948
0.3542	14.0	4844	0.3718	0.8608	0.6970	0.5579	0.5971
0.3484	15.0	5190	0.3722	0.8619	0.7085	0.5576	0.5975
0.352	16.0	5536	0.3732	0.8600	0.7223	0.5376	0.5792
0.352	17.0	5882	0.3710	0.8600	0.6859	0.5845	0.6136

trhgquan/visobert-human-tl-seg-seed-24

visobert-human-tl-seg-seed-24