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: 0.2090
• Precision: 0.9465
• Recall: 0.9543
• F1: 0.9504
• Accuracy: 0.9584

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	1.56	250	1.0347	0.6965	0.7695	0.7312	0.7861
1.4031	3.12	500	0.5641	0.8491	0.8720	0.8604	0.8744
1.4031	4.69	750	0.3899	0.8810	0.9087	0.8946	0.9138
0.4005	6.25	1000	0.3025	0.9202	0.9319	0.9260	0.9355
0.4005	7.81	1250	0.2641	0.9211	0.9349	0.9279	0.9402
0.2161	9.38	1500	0.2406	0.9277	0.9416	0.9346	0.9474
0.2161	10.94	1750	0.2250	0.9343	0.9469	0.9405	0.9516
0.1474	12.5	2000	0.2238	0.9415	0.9513	0.9464	0.9542
0.1474	14.06	2250	0.2128	0.9451	0.9536	0.9493	0.9571
0.1128	15.62	2500	0.2090	0.9465	0.9543	0.9504	0.9584

Framework versions

• Transformers 4.30.2
• Pytorch 2.0.1
• Datasets 2.13.1
• Tokenizers 0.13.3