layoutlmv3-base-sroie

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

• Loss: 0.0639
• Precision: 0.9236
• Recall: 0.9625
• F1: 0.9427
• Accuracy: 0.9821

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: 2e-06
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• training_steps: 3000

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	2.5	100	0.1464	0.9081	0.8488	0.8775	0.9645
No log	5.0	200	0.0821	0.9322	0.9294	0.9308	0.9791
No log	7.5	300	0.0746	0.9204	0.9469	0.9335	0.9796
No log	10.0	400	0.0685	0.9213	0.9506	0.9357	0.9802
0.1644	12.5	500	0.0657	0.9192	0.9586	0.9385	0.9809
0.1644	15.0	600	0.0678	0.9071	0.9649	0.9351	0.9796
0.1644	17.5	700	0.0636	0.9242	0.9625	0.9430	0.9822
0.1644	20.0	800	0.0643	0.9238	0.9609	0.9420	0.9819
0.1644	22.5	900	0.0620	0.9254	0.9629	0.9438	0.9824
0.0331	25.0	1000	0.0639	0.9236	0.9625	0.9427	0.9821
0.0331	27.5	1100	0.0632	0.9249	0.9639	0.9440	0.9825
0.0331	30.0	1200	0.0619	0.9268	0.9615	0.9439	0.9825
0.0331	32.5	1300	0.0640	0.9216	0.9665	0.9435	0.9823
0.0331	35.0	1400	0.0653	0.9201	0.9665	0.9428	0.9820

Framework versions

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