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: 0.2409
• Validation Loss: 0.6905
• Train Overall Precision: 0.7213
• Train Overall Recall: 0.7858
• Train Overall F1: 0.7522
• Train Overall Accuracy: 0.7998
• 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': np.float32(3e-05), 'decay': 0.0, 'beta_1': np.float32(0.9), 'beta_2': np.float32(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
1.6885	1.3544	0.2506	0.2534	0.2520	0.5600	0
1.1672	0.8949	0.5489	0.5996	0.5731	0.7115	1
0.7766	0.7345	0.6327	0.7476	0.6854	0.7633	2
0.5690	0.6435	0.6684	0.7727	0.7168	0.7936	3
0.4387	0.6493	0.7180	0.7792	0.7474	0.7962	4
0.3470	0.6380	0.7147	0.7893	0.7501	0.8020	5
0.2901	0.6569	0.7317	0.7868	0.7582	0.8059	6
0.2409	0.6905	0.7213	0.7858	0.7522	0.7998	7

Framework versions

• Transformers 4.46.0
• TensorFlow 2.18.0
• Datasets 3.0.2
• Tokenizers 0.20.1