layoutlmv3-finetuned-cord_100

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: 0.2879
• Precision: 0.9258
• Recall: 0.9384
• F1: 0.9321
• Accuracy: 0.9474

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	4.17	250	0.6982	0.7974	0.8017	0.7995	0.8265
0.9999	8.33	500	0.4214	0.8603	0.8754	0.8678	0.8936
0.9999	12.5	750	0.2820	0.9081	0.9164	0.9123	0.9364
0.1912	16.67	1000	0.2710	0.9147	0.9293	0.9220	0.9389
0.1912	20.83	1250	0.2748	0.9125	0.9271	0.9197	0.9406
0.061	25.0	1500	0.2612	0.9220	0.9339	0.9279	0.9474
0.061	29.17	1750	0.2731	0.9300	0.9384	0.9342	0.9478
0.0275	33.33	2000	0.2824	0.9279	0.9384	0.9331	0.9487
0.0275	37.5	2250	0.2886	0.9242	0.9362	0.9302	0.9457
0.0168	41.67	2500	0.2879	0.9258	0.9384	0.9321	0.9474

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0