layoutlmv3-finetuned-arabic_fullname

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: 0.0196
• Precision: 0.0242
• Recall: 1.0
• F1: 0.0472
• Accuracy: 0.8561

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: 10
• eval_batch_size: 10
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• training_steps: 2500

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	6.7568	250	0.0133	0.1352	1.0	0.2382	0.9772
0.0316	13.5135	500	0.0164	0.0185	1.0	0.0364	0.8111
0.0316	20.2703	750	0.0177	0.0173	1.0	0.0341	0.7981
0.0007	27.0270	1000	0.0180	0.0652	1.0	0.1225	0.9489
0.0007	33.7838	1250	0.0188	0.0798	1.0	0.1478	0.9589
0.0007	40.5405	1500	0.0175	0.0238	1.0	0.0465	0.8540
0.0007	47.2973	1750	0.0189	0.0110	1.0	0.0218	0.6807
0.0003	54.0541	2000	0.0193	0.0110	1.0	0.0218	0.6805
0.0003	60.8108	2250	0.0195	0.0240	1.0	0.0469	0.8553
0.0	67.5676	2500	0.0196	0.0242	1.0	0.0472	0.8561

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.3.1
• Tokenizers 0.21.0