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
0.518	1.0	68	0.4692	0.0	0.5	0.1292
0.4671	2.0	136	0.4422	0.0347	0.5089	0.1433
0.4148	3.0	204	0.4111	0.1912	0.5718	0.2135
0.3962	4.0	272	0.3951	0.2829	0.6088	0.2360
0.3626	5.0	340	0.3708	0.3675	0.6318	0.3118
0.3204	6.0	408	0.3672	0.4323	0.6702	0.3427
0.328	7.0	476	0.3593	0.4433	0.6741	0.3258
0.2809	8.0	544	0.3395	0.5188	0.7055	0.3708
0.2753	9.0	612	0.3416	0.5528	0.7150	0.3989
0.2938	10.0	680	0.3259	0.5776	0.7205	0.4354
0.2326	11.0	748	0.3206	0.6072	0.7377	0.4466
0.2209	12.0	816	0.3225	0.6072	0.7426	0.4326
0.2015	13.0	884	0.3261	0.5976	0.7392	0.4466
0.2056	14.0	952	0.3290	0.5939	0.7341	0.4213
0.1839	15.0	1020	0.3305	0.5920	0.7341	0.4326
0.1822	16.0	1088	0.3323	0.5997	0.7383	0.4326

Kuongan/xlm-roberta-base-hau-noaug

xlm-roberta-base-hau-noaug