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
0.4176	1.0	78	0.2861	0.0	0.5	0.4940
0.2848	2.0	156	0.2879	0.0	0.5	0.4940
0.2775	3.0	234	0.2833	0.0	0.5	0.4940
0.2709	4.0	312	0.2631	0.0392	0.5106	0.5060
0.2349	5.0	390	0.2428	0.1379	0.5491	0.5582
0.1911	6.0	468	0.2338	0.2897	0.6182	0.5783
0.1711	7.0	546	0.2026	0.3843	0.6657	0.6305
0.1439	8.0	624	0.1988	0.3878	0.6753	0.6305
0.1212	9.0	702	0.2118	0.4548	0.7086	0.6506
0.1051	10.0	780	0.2096	0.4055	0.6682	0.6426
0.0989	11.0	858	0.2140	0.5377	0.7450	0.6426
0.0859	12.0	936	0.2148	0.5165	0.7336	0.6707
0.0716	13.0	1014	0.2178	0.5613	0.7600	0.6747
0.0645	14.0	1092	0.2155	0.5161	0.7311	0.6787
0.0679	15.0	1170	0.2178	0.5612	0.7494	0.6867
0.0554	16.0	1248	0.2145	0.5424	0.7449	0.6948
0.0609	17.0	1326	0.2121	0.5536	0.7499	0.6867

Kuongan/xlm-roberta-base-ukr-noaug

xlm-roberta-base-ukr-noaug