learning_rate: 2e-05
train_batch_size: 16
eval_batch_size: 32
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 150
num_epochs: 8

Training Loss	Epoch	Step	Validation Loss	Accuracy	Macro F1	Weighted F1
1.6133	1.0	97	1.5773	0.3667	0.2022	0.2967
1.5912	2.0	194	1.3944	0.5121	0.452	0.5033
1.3431	3.0	291	1.2270	0.5182	0.4756	0.515
1.1180	4.0	388	1.1103	0.5939	0.5359	0.6044
0.8877	5.0	485	1.1672	0.6152	0.5471	0.6123
0.7638	6.0	582	1.0736	0.6364	0.5816	0.6385
0.6304	7.0	679	1.1016	0.6364	0.585	0.6373
0.5446	8.0	776	1.1386	0.6333	0.5703	0.634

annasus10/xlmr-burmese-tone

xlmr-burmese-tone