Chung835/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: 0.4093
• Validation Loss: 0.6195
• Train Overall Precision: 0.7228
• Train Overall Recall: 0.7928
• Train Overall F1: 0.7562
• Train Overall Accuracy: 0.8145
• Epoch: 6

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': {'module': 'keras.optimizers.legacy', 'class_name': 'Adam', 'config': {'name': 'Adam', 'learning_rate': 2.9999999242136255e-05, 'decay': 0.01, 'beta_1': 0.8999999761581421, 'beta_2': 0.9990000128746033, 'epsilon': 1e-07, 'amsgrad': False}, 'registered_name': None}, '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
1.7014	1.4461	0.2258	0.2479	0.2363	0.5036	0
1.2189	0.9465	0.5340	0.5986	0.5645	0.7065	1
0.8423	0.7706	0.6196	0.7095	0.6615	0.7561	2
0.6432	0.6792	0.6762	0.7501	0.7112	0.7850	3
0.5343	0.6767	0.6774	0.7471	0.7106	0.7844	4
0.4602	0.6232	0.7094	0.7878	0.7466	0.8101	5
0.4093	0.6195	0.7228	0.7928	0.7562	0.8145	6

Framework versions

• Transformers 4.52.4
• TensorFlow 2.19.0
• Datasets 3.6.0
• Tokenizers 0.21.1