lmv2-g-recp-992-doc-09-09

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: 0.2241
• Purchase Time Precision: 0.872
• Purchase Time Recall: 0.8516
• Purchase Time F1: 0.8617
• Purchase Time Number: 128
• Receipt Date Precision: 0.8713
• Receipt Date Recall: 0.8817
• Receipt Date F1: 0.8765
• Receipt Date Number: 169
• Sub Total Precision: 0.8211
• Sub Total Recall: 0.7091
• Sub Total F1: 0.7610
• Sub Total Number: 110
• Supplier Address Precision: 0.7009
• Supplier Address Recall: 0.7193
• Supplier Address F1: 0.7100
• Supplier Address Number: 114
• Supplier Name Precision: 0.7442
• Supplier Name Recall: 0.7191
• Supplier Name F1: 0.7314
• Supplier Name Number: 267
• Tip Amount Precision: 0.6667
• Tip Amount Recall: 1.0
• Tip Amount F1: 0.8
• Tip Amount Number: 2
• Total Precision: 0.8436
• Total Recall: 0.8251
• Total F1: 0.8343
• Total Number: 183
• Total Tax Amount Precision: 0.8361
• Total Tax Amount Recall: 0.7846
• Total Tax Amount F1: 0.8095
• Total Tax Amount Number: 65
• Overall Precision: 0.8067
• Overall Recall: 0.7842
• Overall F1: 0.7953
• Overall Accuracy: 0.9728

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	Purchase Time Precision	Purchase Time Recall	Purchase Time F1	Purchase Time Number	Receipt Date Precision	Receipt Date Recall	Receipt Date F1	Receipt Date Number	Sub Total Precision	Sub Total Recall	Sub Total F1	Sub Total Number	Supplier Address Precision	Supplier Address Recall	Supplier Address F1	Supplier Address Number	Supplier Name Precision	Supplier Name Recall	Supplier Name F1	Supplier Name Number	Tip Amount Precision	Tip Amount Recall	Tip Amount F1	Tip Amount Number	Total Precision	Total Recall	Total F1	Total Number	Total Tax Amount Precision	Total Tax Amount Recall	Total Tax Amount F1	Total Tax Amount Number	Overall Precision	Overall Recall	Overall F1	Overall Accuracy
0.9017	1.0	793	0.3748	0.0	0.0	0.0	128	0.5	0.0710	0.1244	169	0.0	0.0	0.0	110	0.4632	0.5526	0.504	114	0.3724	0.2022	0.2621	267	0.0	0.0	0.0	2	0.7387	0.4481	0.5578	183	0.0	0.0	0.0	65	0.4637	0.2033	0.2827	0.9330
0.2651	2.0	1586	0.2025	0.8	0.8438	0.8213	128	0.8274	0.8225	0.8249	169	0.4	0.0182	0.0348	110	0.5329	0.7105	0.6090	114	0.5886	0.6592	0.6219	267	0.0	0.0	0.0	2	0.5720	0.8470	0.6828	183	1.0	0.0308	0.0597	65	0.6424	0.6387	0.6406	0.9624
0.1403	3.0	2379	0.1585	0.8248	0.8828	0.8528	128	0.7897	0.9112	0.8462	169	0.7054	0.7182	0.7117	110	0.5931	0.7544	0.6641	114	0.6288	0.6217	0.6252	267	0.0	0.0	0.0	2	0.7877	0.7705	0.7790	183	0.8276	0.7385	0.7805	65	0.7220	0.7582	0.7397	0.9683
0.0935	4.0	3172	0.1771	0.7891	0.7891	0.7891	128	0.6474	0.7278	0.6852	169	0.8205	0.5818	0.6809	110	0.6074	0.7193	0.6586	114	0.6548	0.6891	0.6715	267	0.0	0.0	0.0	2	0.8476	0.7596	0.8012	183	0.75	0.2308	0.3529	65	0.7108	0.6821	0.6962	0.9648
0.0684	5.0	3965	0.1552	0.9237	0.8516	0.8862	128	0.8362	0.8757	0.8555	169	0.7629	0.6727	0.7150	110	0.6029	0.7193	0.6560	114	0.7167	0.6442	0.6785	267	0.0	0.0	0.0	2	0.8128	0.8306	0.8216	183	0.7937	0.7692	0.7813	65	0.7731	0.7582	0.7656	0.9696
0.0491	6.0	4758	0.1702	0.8760	0.8828	0.8794	128	0.8352	0.8698	0.8522	169	0.8056	0.7909	0.7982	110	0.5894	0.7807	0.6717	114	0.6844	0.6742	0.6792	267	0.0	0.0	0.0	2	0.8778	0.8634	0.8705	183	0.9074	0.7538	0.8235	65	0.7757	0.7929	0.7842	0.9703
0.0472	7.0	5551	0.2037	0.8952	0.8672	0.8810	128	0.8876	0.8876	0.8876	169	0.8	0.7273	0.7619	110	0.6557	0.7018	0.6780	114	0.7953	0.6404	0.7095	267	0.0	0.0	0.0	2	0.8795	0.7978	0.8367	183	0.9394	0.4769	0.6327	65	0.8278	0.7408	0.7819	0.9701
0.0361	8.0	6344	0.1862	0.875	0.8203	0.8468	128	0.7978	0.8402	0.8184	169	0.7739	0.8091	0.7911	110	0.6512	0.7368	0.6914	114	0.6906	0.6854	0.6880	267	0.0	0.0	0.0	2	0.8486	0.8579	0.8533	183	0.6780	0.6154	0.6452	65	0.7612	0.7707	0.7659	0.9701
0.0318	9.0	7137	0.1889	0.9	0.8438	0.8710	128	0.8743	0.8639	0.8690	169	0.875	0.6364	0.7368	110	0.6417	0.6754	0.6581	114	0.6914	0.6966	0.6940	267	0.0	0.0	0.0	2	0.7833	0.8689	0.8238	183	0.7797	0.7077	0.7419	65	0.7772	0.7630	0.7701	0.9697
0.3481	10.0	7930	0.7581	0.0	0.0	0.0	128	0.0	0.0	0.0	169	0.0	0.0	0.0	110	0.0	0.0	0.0	114	0.0	0.0	0.0	267	0.0	0.0	0.0	2	0.0	0.0	0.0	183	0.0	0.0	0.0	65	0.0	0.0	0.0	0.8967
0.7157	11.0	8723	0.7634	0.0	0.0	0.0	128	0.0	0.0	0.0	169	0.0	0.0	0.0	110	0.0	0.0	0.0	114	0.0	0.0	0.0	267	0.0	0.0	0.0	2	0.0	0.0	0.0	183	0.0	0.0	0.0	65	0.0	0.0	0.0	0.8967
0.7136	12.0	9516	0.7611	0.0	0.0	0.0	128	0.0	0.0	0.0	169	0.0	0.0	0.0	110	0.0	0.0	0.0	114	0.0	0.0	0.0	267	0.0	0.0	0.0	2	0.0	0.0	0.0	183	0.0	0.0	0.0	65	0.0	0.0	0.0	0.8967
0.1095	13.0	10309	0.1744	0.8284	0.8672	0.8473	128	0.8531	0.8935	0.8728	169	0.7717	0.6455	0.7030	110	0.5662	0.6754	0.6160	114	0.6424	0.6929	0.6667	267	0.0	0.0	0.0	2	0.8211	0.8525	0.8365	183	0.8214	0.7077	0.7603	65	0.7428	0.7678	0.7551	0.9698
0.0316	14.0	11102	0.1812	0.8943	0.8594	0.8765	128	0.8409	0.8757	0.8580	169	0.8415	0.6273	0.7188	110	0.5714	0.6667	0.6154	114	0.6279	0.7079	0.6655	267	1.0	0.5	0.6667	2	0.8256	0.8798	0.8519	183	0.8136	0.7385	0.7742	65	0.7495	0.7726	0.7609	0.9703
0.0226	15.0	11895	0.2132	0.8843	0.8359	0.8594	128	0.8476	0.8225	0.8348	169	0.7525	0.6909	0.7204	110	0.5804	0.7281	0.6459	114	0.6679	0.6929	0.6801	267	0.2	0.5	0.2857	2	0.8571	0.8525	0.8548	183	0.4835	0.6769	0.5641	65	0.7297	0.7620	0.7455	0.9672
0.0241	16.0	12688	0.1962	0.8984	0.8984	0.8984	128	0.8613	0.8817	0.8713	169	0.6615	0.7818	0.7167	110	0.6	0.7368	0.6614	114	0.6431	0.7154	0.6773	267	0.0833	0.5	0.1429	2	0.8795	0.7978	0.8367	183	0.7727	0.7846	0.7786	65	0.7401	0.7929	0.7656	0.9709
0.0155	17.0	13481	0.1995	0.8906	0.8906	0.8906	128	0.8678	0.8935	0.8805	169	0.7438	0.8182	0.7792	110	0.6042	0.7632	0.6744	114	0.6193	0.7678	0.6856	267	1.0	0.5	0.6667	2	0.8325	0.8689	0.8503	183	0.8644	0.7846	0.8226	65	0.7467	0.8266	0.7846	0.9696
0.0165	18.0	14274	0.2402	0.8966	0.8125	0.8525	128	0.8293	0.8047	0.8168	169	0.8118	0.6273	0.7077	110	0.5766	0.6930	0.6295	114	0.7220	0.6517	0.6850	267	1.0	1.0	1.0	2	0.8603	0.8415	0.8508	183	0.7826	0.5538	0.6486	65	0.7773	0.7264	0.7510	0.9683
0.0721	19.0	15067	0.2718	0.3506	0.6328	0.4513	128	0.7268	0.7870	0.7557	169	0.7742	0.4364	0.5581	110	0.5271	0.5965	0.5597	114	0.5294	0.5056	0.5172	267	0.0	0.0	0.0	2	0.7526	0.7978	0.7745	183	0.7414	0.6615	0.6992	65	0.5881	0.6301	0.6084	0.9564
0.0136	20.0	15860	0.2213	0.8651	0.8516	0.8583	128	0.8555	0.8757	0.8655	169	0.8191	0.7	0.7549	110	0.6103	0.7281	0.664	114	0.6977	0.6742	0.6857	267	1.0	0.5	0.6667	2	0.8571	0.8197	0.8380	183	0.7656	0.7538	0.7597	65	0.7760	0.7678	0.7719	0.9697
0.0111	21.0	16653	0.2241	0.872	0.8516	0.8617	128	0.8713	0.8817	0.8765	169	0.8211	0.7091	0.7610	110	0.7009	0.7193	0.7100	114	0.7442	0.7191	0.7314	267	0.6667	1.0	0.8	2	0.8436	0.8251	0.8343	183	0.8361	0.7846	0.8095	65	0.8067	0.7842	0.7953	0.9728
0.011	22.0	17446	0.2206	0.7770	0.8984	0.8333	128	0.8270	0.9053	0.8644	169	0.8586	0.7727	0.8134	110	0.5985	0.6930	0.6423	114	0.6618	0.6742	0.6679	267	0.0	0.0	0.0	2	0.8870	0.8579	0.8722	183	0.7391	0.7846	0.7612	65	0.7579	0.7900	0.7736	0.9697
0.0104	23.0	18239	0.2571	0.9310	0.8438	0.8852	128	0.875	0.8698	0.8724	169	0.8316	0.7182	0.7707	110	0.6417	0.6754	0.6581	114	0.7386	0.6667	0.7008	267	0.1429	0.5	0.2222	2	0.8579	0.8579	0.8579	183	0.7812	0.7692	0.7752	65	0.8018	0.7678	0.7844	0.9705
0.0132	24.0	19032	0.2252	0.8810	0.8672	0.8740	128	0.8297	0.8935	0.8604	169	0.7607	0.8091	0.7841	110	0.6074	0.7193	0.6586	114	0.6578	0.7416	0.6972	267	0.3333	1.0	0.5	2	0.8659	0.8470	0.8564	183	0.7966	0.7231	0.7581	65	0.7557	0.8044	0.7793	0.9717
0.0114	25.0	19825	0.2303	0.8917	0.8359	0.8629	128	0.8947	0.9053	0.9000	169	0.8144	0.7182	0.7633	110	0.6296	0.7456	0.6827	114	0.6937	0.7041	0.6989	267	1.0	0.5	0.6667	2	0.8533	0.8579	0.8556	183	0.8913	0.6308	0.7387	65	0.7912	0.7813	0.7862	0.9705
0.0121	26.0	20618	0.2485	0.8810	0.8672	0.8740	128	0.8793	0.9053	0.8921	169	0.8667	0.7091	0.7800	110	0.5926	0.7018	0.6426	114	0.7446	0.6442	0.6908	267	0.25	0.5	0.3333	2	0.8361	0.8361	0.8361	183	0.7581	0.7231	0.7402	65	0.7910	0.7659	0.7783	0.9705
0.0124	27.0	21411	0.2280	0.8504	0.8438	0.8471	128	0.8391	0.8639	0.8513	169	0.8119	0.7455	0.7773	110	0.6435	0.6491	0.6463	114	0.6259	0.6891	0.6560	267	0.4	1.0	0.5714	2	0.8548	0.8689	0.8618	183	0.8627	0.6769	0.7586	65	0.7588	0.7697	0.7642	0.9702
0.0111	28.0	22204	0.2728	0.8917	0.8359	0.8629	128	0.8704	0.8343	0.8520	169	0.9059	0.7	0.7897	110	0.5833	0.6754	0.6260	114	0.6618	0.6816	0.6716	267	1.0	0.5	0.6667	2	0.8713	0.8142	0.8418	183	0.8837	0.5846	0.7037	65	0.7806	0.7437	0.7617	0.9692
0.0079	29.0	22997	0.2596	0.8661	0.8594	0.8627	128	0.8817	0.8817	0.8817	169	0.7436	0.7909	0.7665	110	0.616	0.6754	0.6444	114	0.6794	0.6667	0.6730	267	1.0	1.0	1.0	2	0.8681	0.8634	0.8658	183	0.8727	0.7385	0.8	65	0.7786	0.7794	0.7790	0.9705
0.0076	30.0	23790	0.2476	0.8088	0.8594	0.8333	128	0.8889	0.8994	0.8941	169	0.7909	0.7909	0.7909	110	0.6397	0.7632	0.6960	114	0.6727	0.6929	0.6827	267	0.3333	1.0	0.5	2	0.8641	0.8689	0.8665	183	0.6512	0.8615	0.7417	65	0.7591	0.8073	0.7824	0.9705

Framework versions

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