layoutlm-funsd

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:

• Loss: 0.6656
• Answer: {'precision': 0.7408637873754153, 'recall': 0.826946847960445, 'f1': 0.7815420560747665, 'number': 809}
• Header: {'precision': 0.2992125984251969, 'recall': 0.31932773109243695, 'f1': 0.30894308943089427, 'number': 119}
• Question: {'precision': 0.7724077328646749, 'recall': 0.8253521126760563, 'f1': 0.7980027235587837, 'number': 1065}
• Overall Precision: 0.7315
• Overall Recall: 0.7958
• Overall F1: 0.7623
• Overall Accuracy: 0.8127

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:

• learning_rate: 3e-05
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 15
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Answer	Header	Question	Overall Precision	Overall Recall	Overall F1	Overall Accuracy
1.8138	1.0	10	1.6087	{'precision': 0.03215434083601286, 'recall': 0.037082818294190356, 'f1': 0.03444316877152698, 'number': 809}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119}	{'precision': 0.20825335892514396, 'recall': 0.20375586854460093, 'f1': 0.2059800664451827, 'number': 1065}	0.1251	0.1239	0.1245	0.3836
1.4355	2.0	20	1.2532	{'precision': 0.22752043596730245, 'recall': 0.20642768850432633, 'f1': 0.21646143875567075, 'number': 809}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119}	{'precision': 0.4494296577946768, 'recall': 0.5549295774647888, 'f1': 0.49663865546218483, 'number': 1065}	0.3699	0.3803	0.3751	0.5848
1.106	3.0	30	0.9895	{'precision': 0.49122807017543857, 'recall': 0.553770086526576, 'f1': 0.5206275421266705, 'number': 809}	{'precision': 0.038461538461538464, 'recall': 0.008403361344537815, 'f1': 0.013793103448275862, 'number': 119}	{'precision': 0.57109375, 'recall': 0.6863849765258216, 'f1': 0.623454157782516, 'number': 1065}	0.5320	0.5921	0.5604	0.6946
0.8611	4.0	40	0.8055	{'precision': 0.6096807415036045, 'recall': 0.7317676143386898, 'f1': 0.6651685393258427, 'number': 809}	{'precision': 0.19148936170212766, 'recall': 0.07563025210084033, 'f1': 0.10843373493975902, 'number': 119}	{'precision': 0.663527397260274, 'recall': 0.7276995305164319, 'f1': 0.6941334527541425, 'number': 1065}	0.6295	0.6904	0.6585	0.7575
0.684	5.0	50	0.7200	{'precision': 0.6620021528525296, 'recall': 0.7601977750309024, 'f1': 0.70771001150748, 'number': 809}	{'precision': 0.23529411764705882, 'recall': 0.16806722689075632, 'f1': 0.19607843137254902, 'number': 119}	{'precision': 0.6933010492332526, 'recall': 0.8065727699530516, 'f1': 0.7456597222222223, 'number': 1065}	0.6631	0.7496	0.7037	0.7868
0.5693	6.0	60	0.6933	{'precision': 0.6771488469601677, 'recall': 0.7985166872682324, 'f1': 0.7328417470221215, 'number': 809}	{'precision': 0.20202020202020202, 'recall': 0.16806722689075632, 'f1': 0.1834862385321101, 'number': 119}	{'precision': 0.7029787234042553, 'recall': 0.7755868544600939, 'f1': 0.7375, 'number': 1065}	0.6697	0.7486	0.7069	0.7882
0.4931	7.0	70	0.6542	{'precision': 0.6927138331573389, 'recall': 0.8108776266996292, 'f1': 0.7471526195899771, 'number': 809}	{'precision': 0.2689075630252101, 'recall': 0.2689075630252101, 'f1': 0.2689075630252101, 'number': 119}	{'precision': 0.729043183742591, 'recall': 0.8084507042253521, 'f1': 0.7666963490650045, 'number': 1065}	0.6894	0.7772	0.7307	0.8042
0.4267	8.0	80	0.6503	{'precision': 0.7034700315457413, 'recall': 0.826946847960445, 'f1': 0.7602272727272728, 'number': 809}	{'precision': 0.275, 'recall': 0.2773109243697479, 'f1': 0.27615062761506276, 'number': 119}	{'precision': 0.7510656436487638, 'recall': 0.8272300469483568, 'f1': 0.7873100983020554, 'number': 1065}	0.7054	0.7943	0.7472	0.8070
0.3872	9.0	90	0.6552	{'precision': 0.7311111111111112, 'recall': 0.8133498145859085, 'f1': 0.770040959625512, 'number': 809}	{'precision': 0.29906542056074764, 'recall': 0.2689075630252101, 'f1': 0.28318584070796454, 'number': 119}	{'precision': 0.7558239861949957, 'recall': 0.8225352112676056, 'f1': 0.787769784172662, 'number': 1065}	0.7230	0.7858	0.7531	0.8113
0.3651	10.0	100	0.6531	{'precision': 0.7281659388646288, 'recall': 0.8244746600741656, 'f1': 0.7733333333333332, 'number': 809}	{'precision': 0.29838709677419356, 'recall': 0.31092436974789917, 'f1': 0.3045267489711935, 'number': 119}	{'precision': 0.756872852233677, 'recall': 0.8272300469483568, 'f1': 0.7904890085240016, 'number': 1065}	0.7191	0.7953	0.7553	0.8144
0.3186	11.0	110	0.6525	{'precision': 0.7268770402611534, 'recall': 0.8257107540173053, 'f1': 0.773148148148148, 'number': 809}	{'precision': 0.312, 'recall': 0.3277310924369748, 'f1': 0.31967213114754095, 'number': 119}	{'precision': 0.7627705627705628, 'recall': 0.8272300469483568, 'f1': 0.7936936936936936, 'number': 1065}	0.7221	0.7968	0.7576	0.8131
0.303	12.0	120	0.6564	{'precision': 0.7306843267108167, 'recall': 0.8182941903584673, 'f1': 0.7720116618075801, 'number': 809}	{'precision': 0.3333333333333333, 'recall': 0.31932773109243695, 'f1': 0.3261802575107296, 'number': 119}	{'precision': 0.7755102040816326, 'recall': 0.8206572769953052, 'f1': 0.7974452554744526, 'number': 1065}	0.7331	0.7898	0.7604	0.8127
0.2847	13.0	130	0.6678	{'precision': 0.7435320584926884, 'recall': 0.8170580964153276, 'f1': 0.7785630153121318, 'number': 809}	{'precision': 0.29545454545454547, 'recall': 0.3277310924369748, 'f1': 0.3107569721115538, 'number': 119}	{'precision': 0.7697022767075307, 'recall': 0.8253521126760563, 'f1': 0.7965564114182148, 'number': 1065}	0.7300	0.7923	0.7599	0.8106
0.2689	14.0	140	0.6648	{'precision': 0.7398015435501654, 'recall': 0.8294190358467244, 'f1': 0.782051282051282, 'number': 809}	{'precision': 0.3046875, 'recall': 0.3277310924369748, 'f1': 0.31578947368421056, 'number': 119}	{'precision': 0.7748460861917327, 'recall': 0.8272300469483568, 'f1': 0.8001816530426885, 'number': 1065}	0.7325	0.7983	0.7640	0.8123
0.2642	15.0	150	0.6656	{'precision': 0.7408637873754153, 'recall': 0.826946847960445, 'f1': 0.7815420560747665, 'number': 809}	{'precision': 0.2992125984251969, 'recall': 0.31932773109243695, 'f1': 0.30894308943089427, 'number': 119}	{'precision': 0.7724077328646749, 'recall': 0.8253521126760563, 'f1': 0.7980027235587837, 'number': 1065}	0.7315	0.7958	0.7623	0.8127

Framework versions

• Transformers 4.57.3
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1