learning_rate: 2e-05
train_batch_size: 32
eval_batch_size: 32
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: cosine
lr_scheduler_warmup_steps: 100
num_epochs: 20

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc	Accuracy
0.4026	1.0	178	0.3807	0.1882	0.5574	0.2535
0.3433	2.0	356	0.3388	0.3055	0.5988	0.3324
0.3018	3.0	534	0.2926	0.4564	0.6849	0.4768
0.2482	4.0	712	0.2684	0.5884	0.7582	0.5190
0.2163	5.0	890	0.2530	0.5982	0.7719	0.5472
0.1771	6.0	1068	0.2235	0.6247	0.7820	0.5944
0.1559	7.0	1246	0.1981	0.7041	0.8187	0.6810
0.1295	8.0	1424	0.1847	0.7338	0.8383	0.7183
0.1067	9.0	1602	0.1720	0.7632	0.8591	0.7592
0.0824	10.0	1780	0.1504	0.8117	0.8735	0.7979
0.0669	11.0	1958	0.1522	0.8178	0.8812	0.8042
0.0525	12.0	2136	0.1467	0.8342	0.8903	0.8268
0.0449	13.0	2314	0.1418	0.8565	0.9067	0.8458
0.0378	14.0	2492	0.1416	0.8497	0.9008	0.8401
0.0383	15.0	2670	0.1382	0.8670	0.9165	0.8458
0.0324	16.0	2848	0.1351	0.8664	0.9151	0.8620
0.0322	17.0	3026	0.1363	0.8786	0.9184	0.8683
0.028	18.0	3204	0.1329	0.8870	0.9249	0.8704
0.028	19.0	3382	0.1337	0.8842	0.9209	0.8718
0.0299	20.0	3560	0.1325	0.8834	0.9202	0.8711

Kuongan/xlm-roberta-base-amh-finetuned

xlm-roberta-base-amh-finetuned