Panant/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: 0.2531
• Validation Loss: 0.6684
• Train Overall Precision: 0.7060
• Train Overall Recall: 0.8013
• Train Overall F1: 0.7506
• Train Overall Accuracy: 0.8087
• 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
1.6990	1.3937	0.2342	0.2454	0.2396	0.5233	0
1.1531	0.8728	0.5680	0.6327	0.5986	0.7422	1
0.7533	0.7316	0.6481	0.7200	0.6822	0.7642	2
0.5825	0.6844	0.6840	0.7592	0.7196	0.7896	3
0.4667	0.6354	0.7123	0.7812	0.7452	0.8108	4
0.3650	0.6304	0.7208	0.7797	0.7491	0.8153	5
0.2901	0.6709	0.7158	0.7973	0.7543	0.8036	6
0.2531	0.6684	0.7060	0.8013	0.7506	0.8087	7

Framework versions

• Transformers 4.34.0
• TensorFlow 2.13.0
• Datasets 2.14.5
• Tokenizers 0.14.1