lmv2-g-pan-143doc-06-12

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: 0.0443
• Dob Precision: 1.0
• Dob Recall: 1.0
• Dob F1: 1.0
• Dob Number: 27
• Fname Precision: 1.0
• Fname Recall: 0.9643
• Fname F1: 0.9818
• Fname Number: 28
• Name Precision: 0.9630
• Name Recall: 0.9630
• Name F1: 0.9630
• Name Number: 27
• Pan Precision: 1.0
• Pan Recall: 1.0
• Pan F1: 1.0
• Pan Number: 26
• Overall Precision: 0.9907
• Overall Recall: 0.9815
• Overall F1: 0.9860
• Overall Accuracy: 0.9978

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	Dob Precision	Dob Recall	Dob F1	Dob Number	Fname Precision	Fname Recall	Fname F1	Fname Number	Name Precision	Name Recall	Name F1	Name Number	Pan Precision	Pan Recall	Pan F1	Pan Number	Overall Precision	Overall Recall	Overall F1	Overall Accuracy
1.274	1.0	114	0.9098	0.9310	1.0	0.9643	27	0.1481	0.1429	0.1455	28	0.1639	0.3704	0.2273	27	0.8125	1.0	0.8966	26	0.4497	0.6204	0.5214	0.9143
0.7133	2.0	228	0.5771	0.9310	1.0	0.9643	27	0.2093	0.3214	0.2535	28	0.6562	0.7778	0.7119	27	0.9630	1.0	0.9811	26	0.6336	0.7685	0.6946	0.9443
0.4593	3.0	342	0.4018	0.9643	1.0	0.9818	27	0.8276	0.8571	0.8421	28	0.9259	0.9259	0.9259	27	1.0	1.0	1.0	26	0.9273	0.9444	0.9358	0.9655
0.3011	4.0	456	0.2638	0.9643	1.0	0.9818	27	1.0	0.9286	0.9630	28	0.9259	0.9259	0.9259	27	0.9630	1.0	0.9811	26	0.9630	0.9630	0.9630	0.9811
0.2209	5.0	570	0.2108	0.9643	1.0	0.9818	27	0.8621	0.8929	0.8772	28	0.9286	0.9630	0.9455	27	0.9286	1.0	0.9630	26	0.9204	0.9630	0.9412	0.9811
0.1724	6.0	684	0.1671	0.9643	1.0	0.9818	27	0.9286	0.9286	0.9286	28	0.8667	0.9630	0.9123	27	0.8966	1.0	0.9455	26	0.9130	0.9722	0.9417	0.9844
0.1285	7.0	798	0.1754	0.9643	1.0	0.9818	27	0.8929	0.8929	0.8929	28	0.9630	0.9630	0.9630	27	0.9630	1.0	0.9811	26	0.9455	0.9630	0.9541	0.9788
0.0999	8.0	912	0.1642	0.9643	1.0	0.9818	27	0.9615	0.8929	0.9259	28	0.9630	0.9630	0.9630	27	0.9630	1.0	0.9811	26	0.9630	0.9630	0.9630	0.9811
0.0862	9.0	1026	0.1417	0.9643	1.0	0.9818	27	0.8966	0.9286	0.9123	28	0.8966	0.9630	0.9286	27	0.9630	1.0	0.9811	26	0.9292	0.9722	0.9502	0.9788
0.0722	10.0	1140	0.1317	0.9643	1.0	0.9818	27	0.9630	0.9286	0.9455	28	0.9286	0.9630	0.9455	27	0.9630	1.0	0.9811	26	0.9545	0.9722	0.9633	0.9822
0.0748	11.0	1254	0.1220	0.9643	1.0	0.9818	27	1.0	0.8929	0.9434	28	1.0	0.9630	0.9811	27	0.9286	1.0	0.9630	26	0.9720	0.9630	0.9674	0.9833
0.0549	12.0	1368	0.1157	0.9643	1.0	0.9818	27	0.8966	0.9286	0.9123	28	0.8667	0.9630	0.9123	27	0.8966	1.0	0.9455	26	0.9052	0.9722	0.9375	0.9811
0.0444	13.0	1482	0.1198	0.9643	1.0	0.9818	27	1.0	0.8929	0.9434	28	0.9630	0.9630	0.9630	27	0.9630	1.0	0.9811	26	0.9720	0.9630	0.9674	0.9811
0.0371	14.0	1596	0.1082	0.9643	1.0	0.9818	27	0.8966	0.9286	0.9123	28	0.8966	0.9630	0.9286	27	0.7879	1.0	0.8814	26	0.8824	0.9722	0.9251	0.9833
0.036	15.0	1710	0.1257	0.9643	1.0	0.9818	27	0.9630	0.9286	0.9455	28	0.9630	0.9630	0.9630	27	0.8966	1.0	0.9455	26	0.9459	0.9722	0.9589	0.9800
0.0291	16.0	1824	0.0930	0.9643	1.0	0.9818	27	0.9643	0.9643	0.9643	28	0.9643	1.0	0.9818	27	0.8667	1.0	0.9286	26	0.9386	0.9907	0.9640	0.9900
0.0267	17.0	1938	0.0993	0.9643	1.0	0.9818	27	0.9286	0.9286	0.9286	28	0.9286	0.9630	0.9455	27	0.9286	1.0	0.9630	26	0.9375	0.9722	0.9545	0.9844
0.023	18.0	2052	0.1240	0.9643	1.0	0.9818	27	0.7941	0.9643	0.8710	28	0.9643	1.0	0.9818	27	0.8387	1.0	0.9123	26	0.8843	0.9907	0.9345	0.9800
0.0379	19.0	2166	0.1154	0.9643	1.0	0.9818	27	1.0	0.9286	0.9630	28	0.9286	0.9630	0.9455	27	0.9286	1.0	0.9630	26	0.9545	0.9722	0.9633	0.9833
0.0199	20.0	2280	0.1143	0.9643	1.0	0.9818	27	1.0	0.9286	0.9630	28	0.8966	0.9630	0.9286	27	0.8667	1.0	0.9286	26	0.9292	0.9722	0.9502	0.9844
0.0256	21.0	2394	0.1175	0.9643	1.0	0.9818	27	0.8667	0.9286	0.8966	28	0.9286	0.9630	0.9455	27	0.9286	1.0	0.9630	26	0.9211	0.9722	0.9459	0.9811
0.0388	22.0	2508	0.0964	0.9643	1.0	0.9818	27	0.8966	0.9286	0.9123	28	0.9310	1.0	0.9643	27	0.8966	1.0	0.9455	26	0.9217	0.9815	0.9507	0.9855
0.0334	23.0	2622	0.1186	0.9643	1.0	0.9818	27	1.0	0.9286	0.9630	28	1.0	0.9630	0.9811	27	0.8966	1.0	0.9455	26	0.9633	0.9722	0.9677	0.9833
0.0134	24.0	2736	0.1193	0.9643	1.0	0.9818	27	0.9630	0.9286	0.9455	28	1.0	0.9630	0.9811	27	0.9286	1.0	0.9630	26	0.9633	0.9722	0.9677	0.9822
0.0157	25.0	2850	0.1078	1.0	1.0	1.0	27	0.9259	0.8929	0.9091	28	0.9286	0.9630	0.9455	27	0.8966	1.0	0.9455	26	0.9369	0.9630	0.9498	0.9833
0.0157	26.0	2964	0.0758	1.0	1.0	1.0	27	0.8929	0.8929	0.8929	28	1.0	1.0	1.0	27	0.8966	1.0	0.9455	26	0.9459	0.9722	0.9589	0.9911
0.0096	27.0	3078	0.0766	1.0	1.0	1.0	27	0.8929	0.8929	0.8929	28	1.0	1.0	1.0	27	0.8966	1.0	0.9455	26	0.9459	0.9722	0.9589	0.9889
0.0135	28.0	3192	0.0443	1.0	1.0	1.0	27	1.0	0.9643	0.9818	28	0.9630	0.9630	0.9630	27	1.0	1.0	1.0	26	0.9907	0.9815	0.9860	0.9978
0.012	29.0	3306	0.1153	0.9643	1.0	0.9818	27	0.8966	0.9286	0.9123	28	0.8667	0.9630	0.9123	27	0.8966	1.0	0.9455	26	0.9052	0.9722	0.9375	0.9822
0.0069	30.0	3420	0.1373	0.9643	1.0	0.9818	27	0.8966	0.9286	0.9123	28	0.9286	0.9630	0.9455	27	0.8966	1.0	0.9455	26	0.9211	0.9722	0.9459	0.9777

Framework versions

• Transformers 4.20.0.dev0
• Pytorch 1.11.0+cu113
• Datasets 2.2.2
• Tokenizers 0.12.1