learning_rate: 1e-05
train_batch_size: 4
eval_batch_size: 4
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_ratio: 0.1
lr_scheduler_warmup_steps: 500
num_epochs: 40
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1	Balanced Accuracy
0.8192	1.0	2902	0.8489	0.6340	0.6084	0.6340	0.6083	0.4798
0.8388	2.0	5804	0.7590	0.6857	0.6758	0.6857	0.6604	0.5380
0.6252	3.0	8706	0.8490	0.6930	0.6947	0.6930	0.6756	0.5712
0.7868	4.0	11608	0.9270	0.6930	0.6987	0.6930	0.6836	0.5969
0.5892	5.0	14510	0.8853	0.7019	0.7048	0.7019	0.6998	0.6292
0.719	6.0	17412	1.0641	0.7043	0.6994	0.7043	0.7010	0.6249
0.6805	7.0	20314	1.2283	0.7126	0.7139	0.7126	0.7043	0.6181
0.7471	8.0	23216	1.3237	0.7057	0.6988	0.7057	0.6983	0.6102
0.6011	9.0	26118	1.3071	0.7085	0.6965	0.7085	0.6944	0.5885
0.5917	10.0	29020	1.5666	0.6888	0.6925	0.6888	0.6901	0.6241
0.5057	11.0	31922	1.5296	0.7112	0.7027	0.7112	0.6877	0.5637

chechdop/xlm-roberta-sentiment-shp-tokenizer-vocab

xlm-roberta-sentiment-shp-tokenizer-vocab