layoutlmv3-final-v4-BI

This model is a fine-tuned version of microsoft/layoutlmv3-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.1471
• Precision: 0.5571
• Recall: 0.4728
• F1: 0.5115
• Accuracy: 0.4681

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 1e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	0.26	10	3.1260	0.1518	0.1149	0.1308	0.1478
No log	0.51	20	2.9269	0.1409	0.0471	0.0706	0.1592
No log	0.77	30	2.7697	0.2328	0.1102	0.1496	0.2116
No log	1.03	40	2.6158	0.4015	0.2895	0.3364	0.3455
No log	1.28	50	2.4704	0.4495	0.3486	0.3927	0.3784
No log	1.54	60	2.3492	0.4964	0.4070	0.4473	0.4195
No log	1.79	70	2.2643	0.5243	0.4369	0.4766	0.4416
No log	2.05	80	2.1982	0.5454	0.4588	0.4984	0.4574
No log	2.31	90	2.1600	0.5516	0.4688	0.5068	0.4637
No log	2.56	100	2.1471	0.5571	0.4728	0.5115	0.4681

Framework versions

• Transformers 4.30.0.dev0
• Pytorch 1.8.0+cu101
• Datasets 2.12.0
• Tokenizers 0.13.3