learning_rate: 2e-05
train_batch_size: 32
eval_batch_size: 32
seed: 42
optimizer: Use adamw_torch 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: 100
num_epochs: 20

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc	Accuracy
0.4203	1.0	111	0.4120	0.0623	0.5157	0.1757
0.3786	2.0	222	0.3952	0.1658	0.5470	0.2280
0.3598	3.0	333	0.3470	0.3744	0.6330	0.3666
0.3057	4.0	444	0.3239	0.4087	0.6530	0.4155
0.2807	5.0	555	0.3067	0.4417	0.6748	0.4341
0.2362	6.0	666	0.3224	0.5196	0.7052	0.4476
0.2233	7.0	777	0.3059	0.5451	0.7174	0.4443
0.2013	8.0	888	0.3206	0.5278	0.7118	0.4358
0.1802	9.0	999	0.3090	0.5489	0.7240	0.4645
0.1659	10.0	1110	0.3262	0.5385	0.7222	0.4561
0.1437	11.0	1221	0.3318	0.5495	0.7259	0.4628
0.1271	12.0	1332	0.3471	0.5510	0.7294	0.4409
0.1234	13.0	1443	0.3527	0.5591	0.7284	0.4527
0.1172	14.0	1554	0.3566	0.5613	0.7314	0.4459
0.1084	15.0	1665	0.3601	0.5564	0.7255	0.4527
0.1021	16.0	1776	0.3596	0.5568	0.7238	0.4527
0.0925	17.0	1887	0.3648	0.5620	0.7274	0.4544
0.0826	18.0	1998	0.3654	0.5606	0.7261	0.4510
0.0841	19.0	2109	0.3660	0.5559	0.7245	0.4544
0.0891	20.0	2220	0.3662	0.5548	0.7236	0.4510

Kuongan/xlm-roberta-base-amh-noaug

xlm-roberta-base-amh-noaug