layoutlmv3-violations-test

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

• Loss: 0.3685
• Precision: 0.9483
• Recall: 0.9116
• F1: 0.9296
• Accuracy: 0.9503

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
• training_steps: 1000

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	9.0909	100	0.2997	0.9543	0.9227	0.9382	0.9558
No log	18.1818	200	0.3729	0.9425	0.9061	0.9239	0.9448
No log	27.2727	300	0.3408	0.9543	0.9227	0.9382	0.9558
No log	36.3636	400	0.3566	0.9483	0.9116	0.9296	0.9503
0.0997	45.4545	500	0.3685	0.9483	0.9116	0.9296	0.9503
0.0997	54.5455	600	0.3736	0.9483	0.9116	0.9296	0.9503
0.0997	63.6364	700	0.3866	0.9483	0.9116	0.9296	0.9503
0.0997	72.7273	800	0.3990	0.9483	0.9116	0.9296	0.9503
0.0997	81.8182	900	0.4018	0.9483	0.9116	0.9296	0.9503
0.001	90.9091	1000	0.3979	0.9483	0.9116	0.9296	0.9503

Framework versions

• Transformers 4.42.1
• Pytorch 2.3.1+cu118
• Datasets 2.20.0
• Tokenizers 0.19.1