learning_rate: 5e-05
train_batch_size: 4
eval_batch_size: 8
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
num_epochs: 10
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss
4.4903	0.4425	100	4.2465
3.8673	0.8850	200	3.6497
3.2619	1.3274	300	3.1490
2.9801	1.7699	400	2.9969
2.0238	2.2124	500	2.5302
1.8018	2.6549	600	2.3716
1.5673	3.0973	700	2.4482
1.308	3.5398	800	2.3696
1.3341	3.9823	900	2.4187
1.0219	4.4248	1000	2.5895
0.8824	4.8673	1100	2.8213
0.7421	5.3097	1200	2.8942
0.4557	5.7522	1300	3.1607
0.3511	6.1947	1400	3.3976
0.4014	6.6372	1500	3.3107
0.5716	7.0796	1600	3.3122
0.3662	7.5221	1700	3.3131
0.2312	7.9646	1800	3.4253
0.225	8.4071	1900	3.6089
0.3573	8.8496	2000	3.6146
0.0989	9.2920	2100	3.6891
0.2683	9.7345	2200	3.7012

guo1006/layoutlmv2-base-uncased-finetuned-docvqa_1200_examples

layoutlmv2-base-uncased-finetuned-docvqa_1200_examples