learning_rate: 1e-05
train_batch_size: 32
eval_batch_size: 32
seed: 5151
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 200
num_epochs: 18

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Recall	Precision
0.1548	1.0	35	0.1456	0.7411	0.4571	0.5112	0.6227
0.1572	2.0	70	0.1354	0.7411	0.6588	0.6570	0.6607
0.1305	3.0	105	0.1212	0.7768	0.6402	0.6251	0.7194
0.1069	4.0	140	0.1155	0.8393	0.7857	0.7794	0.7930
0.0937	5.0	175	0.1216	0.8304	0.7764	0.7734	0.7798
0.0639	6.0	210	0.1257	0.8482	0.7899	0.7742	0.8125
0.0437	7.0	245	0.1610	0.8393	0.7614	0.7345	0.8195
0.0254	8.0	280	0.2101	0.8482	0.7842	0.7630	0.8197
0.0067	9.0	315	0.2555	0.8482	0.7899	0.7742	0.8125

belisards/posicao_tema

belisards/congretimbau