layoutlmv3-finetuned-sroie-funsd

This model is a fine-tuned version of layoutlmv3 on the mp-02/sroie-funsd dataset. It achieves the following results on the evaluation set:

• Loss: 0.1436
• Precision: 0.8236
• Recall: 0.8143
• F1: 0.8189
• Accuracy: 0.9670

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: 7e-06
• 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: 2500

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	2.5773	250	0.1947	0.7126	0.6572	0.6838	0.9480
0.396	5.1546	500	0.1317	0.7733	0.7599	0.7665	0.9607
0.396	7.7320	750	0.1302	0.7963	0.7874	0.7918	0.9651
0.101	10.3093	1000	0.1165	0.8121	0.8047	0.8084	0.9713
0.101	12.8866	1250	0.1226	0.8219	0.8077	0.8148	0.9689
0.0598	15.4639	1500	0.1353	0.8192	0.8113	0.8152	0.9676
0.0598	18.0412	1750	0.1373	0.8271	0.8154	0.8212	0.9689
0.0405	20.6186	2000	0.1436	0.8236	0.8143	0.8189	0.9670
0.0405	23.1959	2250	0.1447	0.8260	0.8184	0.8222	0.9683
0.033	25.7732	2500	0.1491	0.8157	0.8174	0.8166	0.9665

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu118
• Datasets 2.21.0
• Tokenizers 0.19.1