learning_rate: 2e-05
train_batch_size: 16
eval_batch_size: 16
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
0.4902	1.0	176	0.4605	0.5127	0.6705	0.3348
0.3638	2.0	352	0.4028	0.6279	0.7349	0.4231
0.2784	3.0	528	0.3874	0.6829	0.7651	0.4744
0.1717	4.0	704	0.3967	0.7436	0.8069	0.5199
0.1063	5.0	880	0.4247	0.7453	0.7992	0.5299
0.0744	6.0	1056	0.4844	0.7536	0.8159	0.5285
0.0359	7.0	1232	0.5120	0.7775	0.8318	0.5670
0.0306	8.0	1408	0.5876	0.7641	0.8250	0.5427
0.0164	9.0	1584	0.5693	0.7799	0.8296	0.5883
0.0101	10.0	1760	0.6193	0.7817	0.8309	0.5840
0.0042	11.0	1936	0.6293	0.7853	0.8397	0.5855
0.0025	12.0	2112	0.6529	0.7880	0.8406	0.5840
0.0056	13.0	2288	0.6672	0.7843	0.8416	0.5726
0.0022	14.0	2464	0.6656	0.7865	0.8408	0.5812
0.0011	15.0	2640	0.6695	0.7859	0.8427	0.5826

Kuongan/CS221-deberta-v3-large-finetuned-augmentation

CS221-deberta-v3-large-finetuned-augmentation