layoutlmv3-large-model2a-router-2aa-vs-2ab

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

• Loss: 0.4612
• Accuracy: 0.8564
• Macro Precision: 0.8511
• Macro Recall: 0.8643
• Macro F1: 0.8537
• Weighted F1: 0.8577
• Precision Route To Model 2aa: 0.7743
• Recall Route To Model 2aa: 0.9033
• F1 Route To Model 2aa: 0.8338
• Precision Route To Model 2ab: 0.9279
• Recall Route To Model 2ab: 0.8252
• F1 Route To Model 2ab: 0.8735

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: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.ADAFACTOR and the args are: No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Macro Precision	Macro Recall	Macro F1	Weighted F1	Precision Route To Model 2aa	Recall Route To Model 2aa	F1 Route To Model 2aa	Precision Route To Model 2ab	Recall Route To Model 2ab	F1 Route To Model 2ab
0.3874	1.1849	500	0.4425	0.8165	0.8310	0.8412	0.8161	0.8178	0.6947	0.9633	0.8073	0.9673	0.7190	0.8249
0.1955	2.3699	1000	0.4614	0.8564	0.8511	0.8643	0.8537	0.8577	0.7743	0.9033	0.8338	0.9279	0.8252	0.8735

Framework versions

• Transformers 4.57.6
• Pytorch 2.10.0+cu128
• Tokenizers 0.22.2