layoutlm-funsd-tf

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

• Train Loss: 0.2595
• Validation Loss: 0.6704
• Train Overall Precision: 0.7379
• Train Overall Recall: 0.8038
• Train Overall F1: 0.7695
• Train Overall Accuracy: 0.8080
• Epoch: 7

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:

• optimizer: {'inner_optimizer': {'class_name': 'AdamWeightDecay', 'config': {'name': 'AdamWeightDecay', 'learning_rate': 3e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False, 'weight_decay_rate': 0.01}}, 'dynamic': True, 'initial_scale': 32768.0, 'dynamic_growth_steps': 2000}
• training_precision: mixed_float16

Training results

Train Loss	Validation Loss	Train Overall Precision	Train Overall Recall	Train Overall F1	Train Overall Accuracy	Epoch
1.7036	1.4227	0.2407	0.3337	0.2796	0.5026	0
1.1552	0.8765	0.5724	0.6944	0.6275	0.7181	1
0.7585	0.7119	0.6652	0.7406	0.7009	0.7820	2
0.5699	0.6705	0.6880	0.7592	0.7219	0.7822	3
0.4487	0.6814	0.7087	0.7556	0.7314	0.7798	4
0.3664	0.6761	0.7033	0.7958	0.7467	0.8027	5
0.3170	0.6953	0.7162	0.7903	0.7514	0.7984	6
0.2595	0.6704	0.7379	0.8038	0.7695	0.8080	7

Framework versions

• Transformers 4.32.0
• TensorFlow 2.12.0
• Datasets 2.14.4
• Tokenizers 0.13.3