layoutLMv3-finetuned-confluence

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

• Loss: 1.1354
• Precision: 0.8992
• Recall: 0.9126
• F1: 0.9058
• Accuracy: 0.8578

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: 2500

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	8.33	250	0.9563	0.8807	0.9056	0.8930	0.8505
0.0199	16.67	500	1.0827	0.8792	0.9041	0.8915	0.8393
0.0199	25.0	750	1.0539	0.8834	0.9036	0.8934	0.8493
0.0048	33.33	1000	1.1217	0.8944	0.9131	0.9036	0.8583
0.0048	41.67	1250	1.1195	0.9004	0.9071	0.9037	0.8616
0.0025	50.0	1500	1.1927	0.8923	0.9056	0.8989	0.8467
0.0025	58.33	1750	1.1155	0.9017	0.9116	0.9066	0.8640
0.0008	66.67	2000	1.1871	0.8971	0.9056	0.9014	0.8395
0.0008	75.0	2250	1.1709	0.9007	0.9106	0.9056	0.8420
0.0006	83.33	2500	1.1354	0.8992	0.9126	0.9058	0.8578

Framework versions

• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2