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: 0.2479
• Validation Loss: 0.6865
• Train Overall Precision: 0.7469
• Train Overall Recall: 0.8098
• Train Overall F1: 0.7771
• Train Overall Accuracy: 0.8111
• 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': {'module': 'transformers.optimization_tf', 'class_name': 'AdamWeightDecay', 'config': {'name': 'AdamWeightDecay', 'learning_rate': 2.9999999242136255e-05, 'decay': 0.0, 'beta_1': 0.8999999761581421, 'beta_2': 0.9990000128746033, 'epsilon': 1e-07, 'amsgrad': False, 'weight_decay_rate': 0.01}, 'registered_name': 'AdamWeightDecay'}, '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
1.7068	1.4323	0.2302	0.2604	0.2444	0.5097	0
1.1785	0.8879	0.5487	0.6553	0.5973	0.7149	1
0.7570	0.7017	0.6315	0.7411	0.6819	0.7810	2
0.5598	0.6353	0.6893	0.7747	0.7295	0.7954	3
0.4407	0.6282	0.7144	0.7842	0.7477	0.8015	4
0.3450	0.6653	0.7174	0.7822	0.7484	0.8036	5
0.2758	0.7178	0.7002	0.7863	0.7407	0.7920	6
0.2479	0.6865	0.7469	0.8098	0.7771	0.8111	7

Framework versions

• Transformers 4.38.2
• TensorFlow 2.13.1
• Datasets 2.20.0
• Tokenizers 0.15.2