layoutlmv3-finetuned-trf1-v2

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: 0.0871
• F1: 0.8814
• Precision: 0.8966
• Recall: 0.8667
• Accuracy: 0.9882

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-05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• training_steps: 2500
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Precision	Recall	Accuracy
0.0606	15.625	250	0.1710	0.7069	0.7321	0.6833	0.9647
0.0082	31.25	500	0.2174	0.8525	0.8387	0.8667	0.9681
0.0035	46.875	750	0.0848	0.8667	0.8667	0.8667	0.9882
0.0023	62.5	1000	0.0883	0.8667	0.8667	0.8667	0.9874
0.0017	78.125	1250	0.0860	0.8739	0.8814	0.8667	0.9874
0.0014	93.75	1500	0.0904	0.8739	0.8814	0.8667	0.9866
0.0012	109.375	1750	0.0883	0.8739	0.8814	0.8667	0.9874
0.001	125.0	2000	0.0882	0.8739	0.8814	0.8667	0.9874
0.0009	140.625	2250	0.0865	0.8814	0.8966	0.8667	0.9882
0.0009	156.25	2500	0.0871	0.8814	0.8966	0.8667	0.9882

Framework versions

• Transformers 4.57.3
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1