layoutlmv3-finetuned-cord_100

This model is a fine-tuned version of microsoft/layoutlmv3-base on the cord-layoutlmv3 dataset. It achieves the following results on the evaluation set:

• Loss: 2.5624
• Precision: 0.4115
• Recall: 0.5397
• F1: 0.4670
• Accuracy: 0.4351

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: 5
• eval_batch_size: 5
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• training_steps: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	0.06	10	3.8065	0.1637	0.2582	0.2003	0.2585
No log	0.12	20	3.4787	0.4661	0.3862	0.4224	0.3353
No log	0.19	30	3.2587	0.4332	0.4731	0.4522	0.3667
No log	0.25	40	3.0615	0.4144	0.4873	0.4479	0.3846
No log	0.31	50	2.9052	0.3993	0.5090	0.4475	0.4024
No log	0.38	60	2.7819	0.3876	0.5165	0.4429	0.4143
No log	0.44	70	2.6853	0.3891	0.5202	0.4452	0.4164
No log	0.5	80	2.6245	0.3942	0.5269	0.4510	0.4236
No log	0.56	90	2.5777	0.4056	0.5352	0.4614	0.4312
No log	0.62	100	2.5624	0.4115	0.5397	0.4670	0.4351

Framework versions

• Transformers 4.28.0
• Pytorch 2.0.1+cpu
• Datasets 2.12.0
• Tokenizers 0.13.3