xlm-roberta-large-tir-finetuned-augmentation-LUNAR

This model is a fine-tuned version of FacebookAI/xlm-roberta-large 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: 20

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc	Accuracy
0.3763	1.0	215	0.3776	0.1438	0.5606	0.2747
0.4229	2.0	430	0.3735	0.1737	0.5735	0.3201
0.3405	3.0	645	0.3376	0.2403	0.5967	0.3888
0.3104	4.0	860	0.3253	0.2997	0.6069	0.4237
0.2814	5.0	1075	0.3238	0.3574	0.6352	0.4808
0.2466	6.0	1290	0.3334	0.3758	0.6471	0.4668
0.244	7.0	1505	0.3430	0.3920	0.6526	0.4622
0.2405	8.0	1720	0.3531	0.3907	0.6528	0.4400
0.1746	9.0	1935	0.3648	0.3995	0.6543	0.4703
0.1659	10.0	2150	0.3761	0.3967	0.6565	0.4587
0.1382	11.0	2365	0.3965	0.4261	0.6664	0.4400
0.1077	12.0	2580	0.4017	0.4170	0.6549	0.4389
0.1126	13.0	2795	0.4171	0.4404	0.6663	0.4424
0.0827	14.0	3010	0.4184	0.4572	0.6747	0.4412
0.0768	15.0	3225	0.4349	0.4633	0.6776	0.4412
0.0647	16.0	3440	0.4417	0.4663	0.6786	0.4307
0.0577	17.0	3655	0.4457	0.4605	0.6758	0.4272
0.0478	18.0	3870	0.4480	0.4589	0.6734	0.4284
0.0661	19.0	4085	0.4497	0.4586	0.6744	0.4377
0.0488	20.0	4300	0.4497	0.4580	0.6734	0.4331