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: 0.6826
• Answer: {'precision': 0.7305524239007892, 'recall': 0.8009888751545118, 'f1': 0.7641509433962265, 'number': 809}
• Header: {'precision': 0.3643410852713178, 'recall': 0.3949579831932773, 'f1': 0.3790322580645162, 'number': 119}
• Question: {'precision': 0.7896613190730838, 'recall': 0.831924882629108, 'f1': 0.8102423411065386, 'number': 1065}
• Overall Precision: 0.7395
• Overall Recall: 0.7933
• Overall F1: 0.7654
• Overall Accuracy: 0.8177

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
1.8197	1.0	10	1.6129	{'precision': 0.02245508982035928, 'recall': 0.018541409147095178, 'f1': 0.020311442112389978, 'number': 809}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119}	{'precision': 0.26639344262295084, 'recall': 0.18309859154929578, 'f1': 0.21702838063439067, 'number': 1065}	0.15	0.1054	0.1238	0.3471
1.459	2.0	20	1.2609	{'precision': 0.2609673790776153, 'recall': 0.2867737948084054, 'f1': 0.27326266195524146, 'number': 809}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119}	{'precision': 0.4169329073482428, 'recall': 0.49014084507042255, 'f1': 0.4505826499784203, 'number': 1065}	0.3522	0.3783	0.3648	0.5855
1.105	3.0	30	0.9633	{'precision': 0.48520710059171596, 'recall': 0.6081582200247219, 'f1': 0.53976961053209, 'number': 809}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119}	{'precision': 0.5333333333333333, 'recall': 0.676056338028169, 'f1': 0.5962732919254659, 'number': 1065}	0.5107	0.6081	0.5552	0.7051
0.8404	4.0	40	0.8047	{'precision': 0.5744274809160306, 'recall': 0.7441285537700866, 'f1': 0.6483575659666129, 'number': 809}	{'precision': 0.08333333333333333, 'recall': 0.03361344537815126, 'f1': 0.04790419161676646, 'number': 119}	{'precision': 0.635036496350365, 'recall': 0.7352112676056338, 'f1': 0.6814621409921671, 'number': 1065}	0.5964	0.6969	0.6428	0.7578
0.6838	5.0	50	0.7318	{'precision': 0.6411637931034483, 'recall': 0.7354758961681088, 'f1': 0.6850892343120323, 'number': 809}	{'precision': 0.2236842105263158, 'recall': 0.14285714285714285, 'f1': 0.17435897435897438, 'number': 119}	{'precision': 0.6717011128775835, 'recall': 0.7934272300469484, 'f1': 0.7275075333620319, 'number': 1065}	0.6441	0.7311	0.6848	0.7834
0.5808	6.0	60	0.7147	{'precision': 0.6506276150627615, 'recall': 0.7688504326328801, 'f1': 0.7048158640226628, 'number': 809}	{'precision': 0.32894736842105265, 'recall': 0.21008403361344538, 'f1': 0.25641025641025644, 'number': 119}	{'precision': 0.7051826677994902, 'recall': 0.7793427230046949, 'f1': 0.7404103479036575, 'number': 1065}	0.6686	0.7411	0.7030	0.7825
0.5061	7.0	70	0.6761	{'precision': 0.68, 'recall': 0.7775030902348579, 'f1': 0.7254901960784315, 'number': 809}	{'precision': 0.3274336283185841, 'recall': 0.31092436974789917, 'f1': 0.3189655172413793, 'number': 119}	{'precision': 0.7265692175408427, 'recall': 0.7934272300469484, 'f1': 0.7585278276481149, 'number': 1065}	0.6865	0.7582	0.7206	0.7993
0.4467	8.0	80	0.6618	{'precision': 0.676130389064143, 'recall': 0.7948084054388134, 'f1': 0.7306818181818182, 'number': 809}	{'precision': 0.27184466019417475, 'recall': 0.23529411764705882, 'f1': 0.2522522522522523, 'number': 119}	{'precision': 0.7289719626168224, 'recall': 0.8056338028169014, 'f1': 0.7653880463871543, 'number': 1065}	0.6853	0.7672	0.7240	0.8040
0.399	9.0	90	0.6648	{'precision': 0.6933911159263272, 'recall': 0.7911001236093943, 'f1': 0.7390300230946881, 'number': 809}	{'precision': 0.3103448275862069, 'recall': 0.3025210084033613, 'f1': 0.30638297872340425, 'number': 119}	{'precision': 0.7459366980325064, 'recall': 0.8187793427230047, 'f1': 0.7806624888093106, 'number': 1065}	0.7011	0.7767	0.7370	0.8099
0.3777	10.0	100	0.6685	{'precision': 0.7158962795941376, 'recall': 0.7849196538936959, 'f1': 0.7488207547169812, 'number': 809}	{'precision': 0.3162393162393162, 'recall': 0.31092436974789917, 'f1': 0.3135593220338983, 'number': 119}	{'precision': 0.7574978577549272, 'recall': 0.8300469483568075, 'f1': 0.7921146953405018, 'number': 1065}	0.7167	0.7807	0.7474	0.8111
0.326	11.0	110	0.6740	{'precision': 0.7254464285714286, 'recall': 0.8034610630407911, 'f1': 0.7624633431085045, 'number': 809}	{'precision': 0.3356643356643357, 'recall': 0.40336134453781514, 'f1': 0.366412213740458, 'number': 119}	{'precision': 0.7606614447345518, 'recall': 0.8206572769953052, 'f1': 0.7895212285456188, 'number': 1065}	0.7185	0.7888	0.7520	0.8130
0.307	12.0	120	0.6741	{'precision': 0.7319004524886877, 'recall': 0.799752781211372, 'f1': 0.7643236857649144, 'number': 809}	{'precision': 0.3548387096774194, 'recall': 0.3697478991596639, 'f1': 0.36213991769547327, 'number': 119}	{'precision': 0.7879325643300799, 'recall': 0.8338028169014085, 'f1': 0.8102189781021898, 'number': 1065}	0.7396	0.7923	0.7650	0.8132
0.2926	13.0	130	0.6789	{'precision': 0.7275784753363229, 'recall': 0.8022249690976514, 'f1': 0.7630805408583187, 'number': 809}	{'precision': 0.3409090909090909, 'recall': 0.37815126050420167, 'f1': 0.3585657370517928, 'number': 119}	{'precision': 0.7806167400881058, 'recall': 0.831924882629108, 'f1': 0.8054545454545454, 'number': 1065}	0.7318	0.7928	0.7611	0.8147
0.278	14.0	140	0.6796	{'precision': 0.723404255319149, 'recall': 0.7985166872682324, 'f1': 0.7591069330199766, 'number': 809}	{'precision': 0.35384615384615387, 'recall': 0.3865546218487395, 'f1': 0.3694779116465864, 'number': 119}	{'precision': 0.7834507042253521, 'recall': 0.8356807511737089, 'f1': 0.8087233075874602, 'number': 1065}	0.7327	0.7938	0.7620	0.8170
0.2745	15.0	150	0.6826	{'precision': 0.7305524239007892, 'recall': 0.8009888751545118, 'f1': 0.7641509433962265, 'number': 809}	{'precision': 0.3643410852713178, 'recall': 0.3949579831932773, 'f1': 0.3790322580645162, 'number': 119}	{'precision': 0.7896613190730838, 'recall': 0.831924882629108, 'f1': 0.8102423411065386, 'number': 1065}	0.7395	0.7933	0.7654	0.8177

Framework versions

• Transformers 4.57.1
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1