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: 0.6945
• Answer: {'precision': 0.7029379760609358, 'recall': 0.7985166872682324, 'f1': 0.7476851851851851, 'number': 809}
• Header: {'precision': 0.3656716417910448, 'recall': 0.4117647058823529, 'f1': 0.3873517786561265, 'number': 119}
• Question: {'precision': 0.7918552036199095, 'recall': 0.8215962441314554, 'f1': 0.8064516129032258, 'number': 1065}
• Overall Precision: 0.7275
• Overall Recall: 0.7878
• Overall F1: 0.7564
• Overall Accuracy: 0.8052

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
1.8328	1.0	10	1.6269	{'precision': 0.012106537530266344, 'recall': 0.012360939431396786, 'f1': 0.012232415902140673, 'number': 809}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119}	{'precision': 0.2, 'recall': 0.15492957746478872, 'f1': 0.17460317460317462, 'number': 1065}	0.1060	0.0878	0.0960	0.3555
1.4745	2.0	20	1.2810	{'precision': 0.1425339366515837, 'recall': 0.1557478368355995, 'f1': 0.1488481984642646, 'number': 809}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119}	{'precision': 0.44803982576228996, 'recall': 0.676056338028169, 'f1': 0.5389221556886227, 'number': 1065}	0.3394	0.4245	0.3772	0.5679
1.0973	3.0	30	0.9149	{'precision': 0.48193916349809884, 'recall': 0.6266996291718171, 'f1': 0.5448683503492746, 'number': 809}	{'precision': 0.06521739130434782, 'recall': 0.025210084033613446, 'f1': 0.03636363636363636, 'number': 119}	{'precision': 0.5712250712250713, 'recall': 0.7530516431924883, 'f1': 0.6496557310652086, 'number': 1065}	0.5244	0.6583	0.5838	0.7085
0.8318	4.0	40	0.7626	{'precision': 0.5760151085930123, 'recall': 0.754017305315204, 'f1': 0.6531049250535332, 'number': 809}	{'precision': 0.21621621621621623, 'recall': 0.13445378151260504, 'f1': 0.16580310880829016, 'number': 119}	{'precision': 0.6803418803418804, 'recall': 0.7474178403755869, 'f1': 0.7123042505592841, 'number': 1065}	0.6175	0.7135	0.6620	0.7648
0.6741	5.0	50	0.7031	{'precision': 0.630457933972311, 'recall': 0.7317676143386898, 'f1': 0.6773455377574371, 'number': 809}	{'precision': 0.29473684210526313, 'recall': 0.23529411764705882, 'f1': 0.2616822429906542, 'number': 119}	{'precision': 0.6975060337892196, 'recall': 0.8140845070422535, 'f1': 0.7512998266897747, 'number': 1065}	0.6531	0.7461	0.6965	0.7841
0.5677	6.0	60	0.6814	{'precision': 0.6348884381338742, 'recall': 0.7737948084054388, 'f1': 0.6974930362116991, 'number': 809}	{'precision': 0.3125, 'recall': 0.21008403361344538, 'f1': 0.25125628140703515, 'number': 119}	{'precision': 0.7495479204339964, 'recall': 0.7784037558685446, 'f1': 0.7637033625057578, 'number': 1065}	0.6814	0.7426	0.7107	0.7808
0.4939	7.0	70	0.6524	{'precision': 0.6795698924731183, 'recall': 0.7812113720642769, 'f1': 0.7268545140885566, 'number': 809}	{'precision': 0.3148148148148148, 'recall': 0.2857142857142857, 'f1': 0.29955947136563876, 'number': 119}	{'precision': 0.76657824933687, 'recall': 0.8140845070422535, 'f1': 0.7896174863387978, 'number': 1065}	0.7068	0.7692	0.7367	0.7968
0.4355	8.0	80	0.6496	{'precision': 0.6659707724425887, 'recall': 0.788627935723115, 'f1': 0.7221279003961517, 'number': 809}	{'precision': 0.3055555555555556, 'recall': 0.2773109243697479, 'f1': 0.2907488986784141, 'number': 119}	{'precision': 0.7633851468048359, 'recall': 0.8300469483568075, 'f1': 0.7953216374269007, 'number': 1065}	0.6992	0.7802	0.7375	0.8021
0.3922	9.0	90	0.6662	{'precision': 0.6965442764578834, 'recall': 0.7972805933250927, 'f1': 0.7435158501440923, 'number': 809}	{'precision': 0.3170731707317073, 'recall': 0.3277310924369748, 'f1': 0.32231404958677684, 'number': 119}	{'precision': 0.7817028985507246, 'recall': 0.8103286384976526, 'f1': 0.7957584140156754, 'number': 1065}	0.7185	0.7762	0.7463	0.8034
0.3788	10.0	100	0.6630	{'precision': 0.7004357298474946, 'recall': 0.7948084054388134, 'f1': 0.7446438911407064, 'number': 809}	{'precision': 0.36283185840707965, 'recall': 0.3445378151260504, 'f1': 0.35344827586206895, 'number': 119}	{'precision': 0.7727674624226348, 'recall': 0.8206572769953052, 'f1': 0.795992714025501, 'number': 1065}	0.7206	0.7817	0.7499	0.8053
0.3263	11.0	110	0.6684	{'precision': 0.6940540540540541, 'recall': 0.7935723114956736, 'f1': 0.740484429065744, 'number': 809}	{'precision': 0.3333333333333333, 'recall': 0.35294117647058826, 'f1': 0.34285714285714286, 'number': 119}	{'precision': 0.7744755244755245, 'recall': 0.831924882629108, 'f1': 0.8021729289271163, 'number': 1065}	0.7153	0.7878	0.7498	0.8053
0.3098	12.0	120	0.6795	{'precision': 0.7033805888767721, 'recall': 0.7972805933250927, 'f1': 0.7473928157589804, 'number': 809}	{'precision': 0.359375, 'recall': 0.3865546218487395, 'f1': 0.3724696356275304, 'number': 119}	{'precision': 0.7887579329102448, 'recall': 0.8169014084507042, 'f1': 0.8025830258302583, 'number': 1065}	0.7267	0.7832	0.7539	0.8073
0.2976	13.0	130	0.6857	{'precision': 0.6913183279742765, 'recall': 0.7972805933250927, 'f1': 0.74052812858783, 'number': 809}	{'precision': 0.3524590163934426, 'recall': 0.36134453781512604, 'f1': 0.35684647302904565, 'number': 119}	{'precision': 0.7831858407079646, 'recall': 0.8309859154929577, 'f1': 0.806378132118451, 'number': 1065}	0.7199	0.7893	0.7530	0.8042
0.277	14.0	140	0.6918	{'precision': 0.7022900763358778, 'recall': 0.796044499381953, 'f1': 0.7462340672074159, 'number': 809}	{'precision': 0.36363636363636365, 'recall': 0.40336134453781514, 'f1': 0.38247011952191234, 'number': 119}	{'precision': 0.7848214285714286, 'recall': 0.8253521126760563, 'f1': 0.8045766590389016, 'number': 1065}	0.7243	0.7883	0.7549	0.8038
0.2706	15.0	150	0.6945	{'precision': 0.7029379760609358, 'recall': 0.7985166872682324, 'f1': 0.7476851851851851, 'number': 809}	{'precision': 0.3656716417910448, 'recall': 0.4117647058823529, 'f1': 0.3873517786561265, 'number': 119}	{'precision': 0.7918552036199095, 'recall': 0.8215962441314554, 'f1': 0.8064516129032258, 'number': 1065}	0.7275	0.7878	0.7564	0.8052

Framework versions

• Transformers 4.57.1
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1