layoutlmv3-finetuned-cord_100

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

• Loss: 1.9357
• Precision: 0.5961
• Recall: 0.6826
• F1: 0.6364
• Accuracy: 0.6846

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	250.0	250	1.5298	0.5778	0.6781	0.6240	0.6825
0.6654	500.0	500	1.6175	0.5942	0.6849	0.6363	0.6880
0.6654	750.0	750	1.7087	0.5947	0.6841	0.6363	0.6876
0.0208	1000.0	1000	1.7729	0.5948	0.6834	0.6360	0.6859
0.0208	1250.0	1250	1.8273	0.5949	0.6826	0.6358	0.6851
0.0099	1500.0	1500	1.8693	0.5957	0.6826	0.6362	0.6846
0.0099	1750.0	1750	1.8969	0.5950	0.6819	0.6355	0.6842
0.0066	2000.0	2000	1.9196	0.5972	0.6826	0.6371	0.6842
0.0066	2250.0	2250	1.9312	0.5946	0.6819	0.6353	0.6838
0.0054	2500.0	2500	1.9357	0.5961	0.6826	0.6364	0.6846

Framework versions

• Transformers 4.34.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.14.1