UIT-NO-PREPROCESSING-roberta-base-finetuned

This model is a fine-tuned version of FacebookAI/roberta-base on the None dataset. It achieves the following results on the evaluation set:

Model description

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

learning_rate: 2e-05
train_batch_size: 16
eval_batch_size: 16
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: 30

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc	Accuracy
0.5044	1.0	139	0.4405	0.4998	0.6667	0.3574
0.3828	2.0	278	0.3625	0.6850	0.7562	0.4477
0.334	3.0	417	0.3518	0.7075	0.7766	0.4603
0.2017	4.0	556	0.3930	0.7013	0.7635	0.4458
0.189	5.0	695	0.4013	0.7239	0.7921	0.4603
0.1366	6.0	834	0.3962	0.7295	0.7924	0.4838
0.0856	7.0	973	0.4296	0.7236	0.7887	0.4657
0.0668	8.0	1112	0.4475	0.7302	0.7934	0.4585
0.0522	9.0	1251	0.4935	0.7229	0.7920	0.4549
0.0369	10.0	1390	0.5097	0.7228	0.7885	0.4549
0.0338	11.0	1529	0.5174	0.7287	0.7953	0.4639
0.0237	12.0	1668	0.5585	0.7305	0.7943	0.4675
0.0168	13.0	1807	0.6014	0.7120	0.7808	0.4513
0.0153	14.0	1946	0.6455	0.7026	0.7715	0.4513
0.0103	15.0	2085	0.6343	0.7199	0.7918	0.4603
0.0083	16.0	2224	0.6500	0.7105	0.7767	0.4513
0.0058	17.0	2363	0.6804	0.7182	0.7846	0.4531
0.0069	18.0	2502	0.6682	0.7283	0.7944	0.4603
0.0044	19.0	2641	0.6832	0.7384	0.8002	0.4639
0.0067	20.0	2780	0.6881	0.7277	0.7926	0.4711
0.0031	21.0	2919	0.6951	0.7208	0.7888	0.4549
0.0031	22.0	3058	0.7029	0.7338	0.7983	0.4675
0.0028	23.0	3197	0.7135	0.7241	0.7908	0.4549
0.0028	24.0	3336	0.7142	0.7297	0.7943	0.4639
0.0034	25.0	3475	0.7192	0.7228	0.7913	0.4531
0.003	26.0	3614	0.7173	0.7227	0.7928	0.4603
0.0029	27.0	3753	0.7207	0.7291	0.7931	0.4585
0.0027	28.0	3892	0.7193	0.7250	0.7930	0.4621
0.0026	29.0	4031	0.7199	0.7255	0.7928	0.4603
0.0024	30.0	4170	0.7200	0.7255	0.7928	0.4603