lmv2-g-w2-300-doc-09-08

This model is a fine-tuned version of microsoft/layoutlmv2-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0262
• Control Number Precision: 1.0
• Control Number Recall: 1.0
• Control Number F1: 1.0
• Control Number Number: 17
• Ein Precision: 1.0
• Ein Recall: 0.9833
• Ein F1: 0.9916
• Ein Number: 60
• Employee’s Address Precision: 0.9667
• Employee’s Address Recall: 0.9831
• Employee’s Address F1: 0.9748
• Employee’s Address Number: 59
• Employee’s Name Precision: 0.9833
• Employee’s Name Recall: 1.0
• Employee’s Name F1: 0.9916
• Employee’s Name Number: 59
• Employee’s Ssn Precision: 0.9836
• Employee’s Ssn Recall: 1.0
• Employee’s Ssn F1: 0.9917
• Employee’s Ssn Number: 60
• Employer’s Address Precision: 0.9833
• Employer’s Address Recall: 0.9672
• Employer’s Address F1: 0.9752
• Employer’s Address Number: 61
• Employer’s Name Precision: 0.9833
• Employer’s Name Recall: 0.9833
• Employer’s Name F1: 0.9833
• Employer’s Name Number: 60
• Federal Income Tax Withheld Precision: 1.0
• Federal Income Tax Withheld Recall: 1.0
• Federal Income Tax Withheld F1: 1.0
• Federal Income Tax Withheld Number: 60
• Medicare Tax Withheld Precision: 1.0
• Medicare Tax Withheld Recall: 1.0
• Medicare Tax Withheld F1: 1.0
• Medicare Tax Withheld Number: 60
• Medicare Wages Tips Precision: 1.0
• Medicare Wages Tips Recall: 1.0
• Medicare Wages Tips F1: 1.0
• Medicare Wages Tips Number: 60
• Social Security Tax Withheld Precision: 1.0
• Social Security Tax Withheld Recall: 0.9836
• Social Security Tax Withheld F1: 0.9917
• Social Security Tax Withheld Number: 61
• Social Security Wages Precision: 0.9833
• Social Security Wages Recall: 1.0
• Social Security Wages F1: 0.9916
• Social Security Wages Number: 59
• Wages Tips Precision: 1.0
• Wages Tips Recall: 0.9836
• Wages Tips F1: 0.9917
• Wages Tips Number: 61
• Overall Precision: 0.9905
• Overall Recall: 0.9905
• Overall F1: 0.9905
• Overall Accuracy: 0.9973

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: 4e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: constant
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Control Number Precision	Control Number Recall	Control Number F1	Control Number Number	Ein Precision	Ein Recall	Ein F1	Ein Number	Employee’s Address Precision	Employee’s Address Recall	Employee’s Address F1	Employee’s Address Number	Employee’s Name Precision	Employee’s Name Recall	Employee’s Name F1	Employee’s Name Number	Employee’s Ssn Precision	Employee’s Ssn Recall	Employee’s Ssn F1	Employee’s Ssn Number	Employer’s Address Precision	Employer’s Address Recall	Employer’s Address F1	Employer’s Address Number	Employer’s Name Precision	Employer’s Name Recall	Employer’s Name F1	Employer’s Name Number	Federal Income Tax Withheld Precision	Federal Income Tax Withheld Recall	Federal Income Tax Withheld F1	Federal Income Tax Withheld Number	Medicare Tax Withheld Precision	Medicare Tax Withheld Recall	Medicare Tax Withheld F1	Medicare Tax Withheld Number	Medicare Wages Tips Precision	Medicare Wages Tips Recall	Medicare Wages Tips F1	Medicare Wages Tips Number	Social Security Tax Withheld Precision	Social Security Tax Withheld Recall	Social Security Tax Withheld F1	Social Security Tax Withheld Number	Social Security Wages Precision	Social Security Wages Recall	Social Security Wages F1	Social Security Wages Number	Wages Tips Precision	Wages Tips Recall	Wages Tips F1	Wages Tips Number	Overall Precision	Overall Recall	Overall F1	Overall Accuracy
1.7717	1.0	240	0.9856	0.0	0.0	0.0	17	0.9206	0.9667	0.9431	60	0.6824	0.9831	0.8056	59	0.2333	0.5932	0.3349	59	0.9836	1.0	0.9917	60	0.7609	0.5738	0.6542	61	0.3654	0.3167	0.3393	60	0.0	0.0	0.0	60	0.8194	0.9833	0.8939	60	0.6064	0.95	0.7403	60	0.5050	0.8361	0.6296	61	0.0	0.0	0.0	59	0.5859	0.9508	0.725	61	0.5954	0.6649	0.6282	0.9558
0.5578	2.0	480	0.2957	0.8462	0.6471	0.7333	17	0.9831	0.9667	0.9748	60	0.9048	0.9661	0.9344	59	0.8358	0.9492	0.8889	59	0.9836	1.0	0.9917	60	0.8125	0.8525	0.8320	61	0.8462	0.9167	0.8800	60	0.9672	0.9833	0.9752	60	0.9524	1.0	0.9756	60	0.9194	0.95	0.9344	60	0.9833	0.9672	0.9752	61	0.9508	0.9831	0.9667	59	0.9516	0.9672	0.9593	61	0.9212	0.9512	0.9359	0.9891
0.223	3.0	720	0.1626	0.5	0.6471	0.5641	17	0.9667	0.9667	0.9667	60	0.9355	0.9831	0.9587	59	0.9672	1.0	0.9833	59	0.9836	1.0	0.9917	60	0.8769	0.9344	0.9048	61	0.9508	0.9667	0.9587	60	0.9833	0.9833	0.9833	60	0.9836	1.0	0.9917	60	0.8769	0.95	0.912	60	1.0	0.9836	0.9917	61	0.9355	0.9831	0.9587	59	0.9516	0.9672	0.9593	61	0.9370	0.9688	0.9526	0.9923
0.1305	4.0	960	0.1025	0.9444	1.0	0.9714	17	0.9831	0.9667	0.9748	60	0.9194	0.9661	0.9421	59	0.9508	0.9831	0.9667	59	0.9836	1.0	0.9917	60	0.9219	0.9672	0.944	61	0.9667	0.9667	0.9667	60	0.9833	0.9833	0.9833	60	0.9524	1.0	0.9756	60	0.8906	0.95	0.9194	60	0.9833	0.9672	0.9752	61	0.9355	0.9831	0.9587	59	0.9516	0.9672	0.9593	61	0.9511	0.9756	0.9632	0.9947
0.0852	5.0	1200	0.0744	0.7391	1.0	0.85	17	0.9831	0.9667	0.9748	60	0.9667	0.9831	0.9748	59	0.9833	1.0	0.9916	59	0.9836	1.0	0.9917	60	0.9344	0.9344	0.9344	61	1.0	0.9833	0.9916	60	0.9365	0.9833	0.9593	60	0.9677	1.0	0.9836	60	0.95	0.95	0.9500	60	0.9836	0.9836	0.9836	61	0.9667	0.9831	0.9748	59	0.9833	0.9672	0.9752	61	0.9626	0.9783	0.9704	0.9953
0.0583	6.0	1440	0.0554	0.7727	1.0	0.8718	17	0.9831	0.9667	0.9748	60	0.9667	0.9831	0.9748	59	0.9833	1.0	0.9916	59	0.9836	1.0	0.9917	60	0.9048	0.9344	0.9194	61	1.0	0.9833	0.9916	60	1.0	0.9833	0.9916	60	0.9833	0.9833	0.9833	60	0.9344	0.95	0.9421	60	1.0	0.9672	0.9833	61	0.9667	0.9831	0.9748	59	0.9833	0.9672	0.9752	61	0.9677	0.9756	0.9716	0.9957
0.0431	7.0	1680	0.0471	0.9444	1.0	0.9714	17	0.9831	0.9667	0.9748	60	0.9016	0.9322	0.9167	59	0.95	0.9661	0.9580	59	0.9836	1.0	0.9917	60	0.8676	0.9672	0.9147	61	0.9831	0.9667	0.9748	60	1.0	0.9833	0.9916	60	1.0	1.0	1.0	60	0.9516	0.9833	0.9672	60	0.9836	0.9836	0.9836	61	0.9831	0.9831	0.9831	59	0.9833	0.9672	0.9752	61	0.9625	0.9756	0.9690	0.9947
0.0314	8.0	1920	0.0359	1.0	1.0	1.0	17	0.9831	0.9667	0.9748	60	0.9355	0.9831	0.9587	59	0.9833	1.0	0.9916	59	0.9836	1.0	0.9917	60	0.9516	0.9672	0.9593	61	1.0	0.9667	0.9831	60	0.9833	0.9833	0.9833	60	1.0	1.0	1.0	60	0.9516	0.9833	0.9672	60	1.0	0.9836	0.9917	61	0.9831	0.9831	0.9831	59	0.9672	0.9672	0.9672	61	0.9771	0.9824	0.9797	0.9969
0.0278	9.0	2160	0.0338	0.8947	1.0	0.9444	17	0.9833	0.9833	0.9833	60	0.9355	0.9831	0.9587	59	0.9667	0.9831	0.9748	59	1.0	1.0	1.0	60	0.9365	0.9672	0.9516	61	0.9672	0.9833	0.9752	60	1.0	0.9833	0.9916	60	1.0	1.0	1.0	60	0.9516	0.9833	0.9672	60	1.0	0.9836	0.9917	61	0.9667	0.9831	0.9748	59	0.9672	0.9672	0.9672	61	0.9705	0.9837	0.9771	0.9965
0.0231	10.0	2400	0.0332	0.9444	1.0	0.9714	17	0.9831	0.9667	0.9748	60	0.9508	0.9831	0.9667	59	0.9048	0.9661	0.9344	59	0.9836	1.0	0.9917	60	0.9667	0.9508	0.9587	61	0.9667	0.9667	0.9667	60	1.0	0.9833	0.9916	60	0.9836	1.0	0.9917	60	0.9365	0.9833	0.9593	60	1.0	0.9672	0.9833	61	0.9831	0.9831	0.9831	59	0.9833	0.9672	0.9752	61	0.9690	0.9769	0.9730	0.9964
0.0189	11.0	2640	0.0342	1.0	1.0	1.0	17	0.9667	0.9667	0.9667	60	0.8657	0.9831	0.9206	59	0.9833	1.0	0.9916	59	0.9836	1.0	0.9917	60	0.8594	0.9016	0.88	61	1.0	0.9833	0.9916	60	0.9833	0.9833	0.9833	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	1.0	0.9836	0.9917	61	0.9833	1.0	0.9916	59	0.9516	0.9672	0.9593	61	0.964	0.9810	0.9724	0.9958
0.0187	12.0	2880	0.0255	1.0	1.0	1.0	17	0.9667	0.9667	0.9667	60	0.9508	0.9831	0.9667	59	0.9833	1.0	0.9916	59	0.9836	1.0	0.9917	60	0.9667	0.9508	0.9587	61	1.0	0.9833	0.9916	60	0.9672	0.9833	0.9752	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	1.0	0.9836	0.9917	61	0.9833	1.0	0.9916	59	0.9833	0.9672	0.9752	61	0.9824	0.9851	0.9837	0.9976
0.0126	13.0	3120	0.0257	1.0	1.0	1.0	17	0.9667	0.9667	0.9667	60	0.9344	0.9661	0.95	59	0.8889	0.9492	0.9180	59	0.9836	1.0	0.9917	60	0.8788	0.9508	0.9134	61	1.0	0.9833	0.9916	60	1.0	1.0	1.0	60	0.9836	1.0	0.9917	60	1.0	1.0	1.0	60	1.0	0.9672	0.9833	61	0.9508	0.9831	0.9667	59	1.0	0.9836	0.9917	61	0.9652	0.9796	0.9724	0.9971
0.012	14.0	3360	0.0227	1.0	1.0	1.0	17	0.9667	0.9667	0.9667	60	0.9516	1.0	0.9752	59	0.9833	1.0	0.9916	59	0.9836	1.0	0.9917	60	0.9194	0.9344	0.9268	61	1.0	0.9833	0.9916	60	0.9672	0.9833	0.9752	60	1.0	0.9833	0.9916	60	1.0	1.0	1.0	60	0.9836	0.9836	0.9836	61	0.9833	1.0	0.9916	59	1.0	0.9836	0.9917	61	0.9784	0.9851	0.9817	0.9977
0.0119	15.0	3600	0.0284	1.0	1.0	1.0	17	1.0	1.0	1.0	60	0.9355	0.9831	0.9587	59	0.9833	1.0	0.9916	59	1.0	1.0	1.0	60	0.9167	0.9016	0.9091	61	0.9661	0.95	0.9580	60	0.9833	0.9833	0.9833	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	1.0	0.9836	0.9917	61	0.9833	1.0	0.9916	59	1.0	0.9836	0.9917	61	0.9810	0.9824	0.9817	0.9965
0.0103	16.0	3840	0.0289	0.9444	1.0	0.9714	17	0.9672	0.9833	0.9752	60	0.9344	0.9661	0.95	59	0.9833	1.0	0.9916	59	1.0	1.0	1.0	60	0.8088	0.9016	0.8527	61	0.9667	0.9667	0.9667	60	0.9833	0.9833	0.9833	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	1.0	0.9836	0.9917	61	0.9833	1.0	0.9916	59	1.0	0.9836	0.9917	61	0.9666	0.9810	0.9737	0.9963
0.01	17.0	4080	0.0305	0.8947	1.0	0.9444	17	1.0	0.9833	0.9916	60	0.9355	0.9831	0.9587	59	0.9516	1.0	0.9752	59	0.9836	1.0	0.9917	60	0.9355	0.9508	0.9431	61	0.9833	0.9833	0.9833	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	0.8955	1.0	0.9449	60	1.0	0.9836	0.9917	61	0.9833	1.0	0.9916	59	1.0	0.9836	0.9917	61	0.9694	0.9891	0.9792	0.9961
0.0082	18.0	4320	0.0256	1.0	1.0	1.0	17	1.0	0.9833	0.9916	60	0.9508	0.9831	0.9667	59	0.9833	1.0	0.9916	59	0.9836	1.0	0.9917	60	0.8636	0.9344	0.8976	61	0.9831	0.9667	0.9748	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	1.0	0.9836	0.9917	61	0.9833	1.0	0.9916	59	1.0	0.9836	0.9917	61	0.9785	0.9864	0.9824	0.9970
0.0059	19.0	4560	0.0255	1.0	1.0	1.0	17	1.0	0.9833	0.9916	60	0.9667	0.9831	0.9748	59	0.9833	1.0	0.9916	59	0.9836	1.0	0.9917	60	0.9355	0.9508	0.9431	61	0.9833	0.9833	0.9833	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	1.0	0.9836	0.9917	61	0.9833	1.0	0.9916	59	1.0	0.9836	0.9917	61	0.9865	0.9891	0.9878	0.9974
0.0078	20.0	4800	0.0293	1.0	1.0	1.0	17	1.0	0.9833	0.9916	60	0.9508	0.9831	0.9667	59	0.9833	1.0	0.9916	59	0.9836	1.0	0.9917	60	0.9	0.8852	0.8926	61	0.9661	0.95	0.9580	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	1.0	0.9836	0.9917	61	0.9833	1.0	0.9916	59	1.0	0.9836	0.9917	61	0.9810	0.9810	0.9810	0.9966
0.009	21.0	5040	0.0264	1.0	1.0	1.0	17	1.0	0.9833	0.9916	60	0.9206	0.9831	0.9508	59	0.9667	0.9831	0.9748	59	0.9836	1.0	0.9917	60	0.8889	0.9180	0.9032	61	0.9672	0.9833	0.9752	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	0.9836	0.9836	0.9836	61	0.9831	0.9831	0.9831	59	1.0	0.9836	0.9917	61	0.9745	0.9837	0.9791	0.9969
0.0046	22.0	5280	0.0271	1.0	1.0	1.0	17	1.0	0.9833	0.9916	60	0.9355	0.9831	0.9587	59	0.9667	0.9831	0.9748	59	0.9836	1.0	0.9917	60	0.9032	0.9180	0.9106	61	0.9672	0.9833	0.9752	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	1.0	0.9836	0.9917	61	0.9833	1.0	0.9916	59	1.0	0.9836	0.9917	61	0.9784	0.9851	0.9817	0.9970
0.0087	23.0	5520	0.0278	0.9444	1.0	0.9714	17	1.0	0.9833	0.9916	60	0.9194	0.9661	0.9421	59	0.9667	0.9831	0.9748	59	0.9836	1.0	0.9917	60	0.8657	0.9508	0.9062	61	0.9836	1.0	0.9917	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	1.0	0.9836	0.9917	61	0.9833	1.0	0.9916	59	1.0	0.9836	0.9917	61	0.9733	0.9878	0.9805	0.9958
0.0054	24.0	5760	0.0276	0.9444	1.0	0.9714	17	1.0	0.9833	0.9916	60	0.95	0.9661	0.9580	59	0.9833	1.0	0.9916	59	0.9836	1.0	0.9917	60	0.9355	0.9508	0.9431	61	0.9831	0.9667	0.9748	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	0.9355	0.9667	0.9508	60	1.0	0.9836	0.9917	61	0.9833	1.0	0.9916	59	1.0	0.9836	0.9917	61	0.9784	0.9837	0.9811	0.9971
0.0057	25.0	6000	0.0260	1.0	1.0	1.0	17	1.0	0.9667	0.9831	60	0.9077	1.0	0.9516	59	0.95	0.9661	0.9580	59	0.9677	1.0	0.9836	60	0.9508	0.9508	0.9508	61	1.0	0.9833	0.9916	60	0.9833	0.9833	0.9833	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	1.0	0.9672	0.9833	61	0.9672	1.0	0.9833	59	1.0	0.9836	0.9917	61	0.9771	0.9837	0.9804	0.9971
0.0074	26.0	6240	0.0340	1.0	1.0	1.0	17	1.0	0.9833	0.9916	60	0.9180	0.9492	0.9333	59	0.9667	0.9831	0.9748	59	0.9836	1.0	0.9917	60	0.8906	0.9344	0.9120	61	0.9831	0.9667	0.9748	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	1.0	0.9836	0.9917	61	0.9833	1.0	0.9916	59	0.9836	0.9836	0.9836	61	0.9757	0.9824	0.9790	0.9959
0.0047	27.0	6480	0.0306	1.0	1.0	1.0	17	1.0	1.0	1.0	60	0.8923	0.9831	0.9355	59	0.9672	1.0	0.9833	59	1.0	1.0	1.0	60	0.9016	0.9016	0.9016	61	0.9667	0.9667	0.9667	60	0.9833	0.9833	0.9833	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	1.0	0.9672	0.9833	61	0.8551	1.0	0.9219	59	1.0	0.8525	0.9204	61	0.9624	0.9715	0.9669	0.9961
0.0052	28.0	6720	0.0262	1.0	1.0	1.0	17	1.0	0.9833	0.9916	60	0.9667	0.9831	0.9748	59	0.9833	1.0	0.9916	59	0.9836	1.0	0.9917	60	0.9833	0.9672	0.9752	61	0.9833	0.9833	0.9833	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	1.0	0.9836	0.9917	61	0.9833	1.0	0.9916	59	1.0	0.9836	0.9917	61	0.9905	0.9905	0.9905	0.9973
0.0033	29.0	6960	0.0320	0.9444	1.0	0.9714	17	1.0	0.9833	0.9916	60	0.8406	0.9831	0.9062	59	0.9672	1.0	0.9833	59	0.9836	1.0	0.9917	60	0.8852	0.8852	0.8852	61	0.9833	0.9833	0.9833	60	1.0	0.9667	0.9831	60	1.0	1.0	1.0	60	0.9833	0.9833	0.9833	60	1.0	0.9836	0.9917	61	0.9365	1.0	0.9672	59	1.0	0.9836	0.9917	61	0.9627	0.9796	0.9711	0.9960
0.0048	30.0	7200	0.0215	1.0	1.0	1.0	17	1.0	0.9833	0.9916	60	0.9672	1.0	0.9833	59	0.9833	1.0	0.9916	59	0.9836	1.0	0.9917	60	0.9833	0.9672	0.9752	61	1.0	0.9833	0.9916	60	0.9833	0.9833	0.9833	60	1.0	1.0	1.0	60	1.0	1.0	1.0	60	1.0	0.9672	0.9833	61	0.9672	1.0	0.9833	59	1.0	0.9836	0.9917	61	0.9891	0.9891	0.9891	0.9980

Framework versions

• Transformers 4.22.0.dev0
• Pytorch 1.12.1+cu113
• Datasets 2.2.2
• Tokenizers 0.12.1