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: 0.2438
• Validation Loss: 0.6811
• Train Overall Precision: 0.7239
• Train Overall Recall: 0.7933
• Train Overall F1: 0.7570
• Train Overall Accuracy: 0.8147
• 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: {'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}
• training_precision: mixed_float16

Training results

Train Loss	Validation Loss	Train Overall Precision	Train Overall Recall	Train Overall F1	Train Overall Accuracy	Epoch
1.6796	1.3763	0.2522	0.3171	0.2810	0.5028	0
1.1052	0.8212	0.6124	0.6849	0.6466	0.7375	1
0.7212	0.7138	0.6471	0.7461	0.6931	0.7665	2
0.5566	0.6209	0.7053	0.7792	0.7404	0.8104	3
0.4273	0.6530	0.71	0.7837	0.7451	0.7999	4
0.3538	0.6343	0.7188	0.7913	0.7533	0.8134	5
0.2872	0.6603	0.7316	0.8013	0.7648	0.8153	6
0.2438	0.6811	0.7239	0.7933	0.7570	0.8147	7

Framework versions

• Transformers 4.35.0
• TensorFlow 2.14.0
• Datasets 2.14.6
• Tokenizers 0.14.1