layout2

This model is a fine-tuned version of microsoft/layoutlmv2-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 4.9306

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: 5e-05
• train_batch_size: 4
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss
5.3459	0.22	50	4.5773
4.4535	0.44	100	4.3372
4.0847	0.66	150	3.7812
3.8993	0.88	200	3.5418
3.5723	1.11	250	3.4054
3.1247	1.33	300	3.1567
3.1067	1.55	350	2.8839
2.9002	1.77	400	2.9007
2.7197	1.99	450	3.1319
2.0518	2.21	500	2.8704
1.9376	2.43	550	2.5644
1.9025	2.65	600	2.5909
2.1491	2.88	650	2.2353
1.5984	3.1	700	2.2114
1.4801	3.32	750	2.5351
1.2501	3.54	800	2.2936
1.2647	3.76	850	2.2360
1.1896	3.98	900	2.3516
0.8484	4.2	950	2.9913
0.8924	4.42	1000	2.6146
0.8342	4.65	1050	3.0150
1.0721	4.87	1100	2.8298
0.6878	5.09	1150	2.9925
0.6807	5.31	1200	2.9388
0.6139	5.53	1250	3.0299
0.7792	5.75	1300	2.4786
0.6837	5.97	1350	3.0713
0.4388	6.19	1400	3.2049
0.5377	6.42	1450	3.1746
0.6028	6.64	1500	3.1708
0.6295	6.86	1550	3.4203
0.5441	7.08	1600	3.3876
0.4105	7.3	1650	3.4065
0.3848	7.52	1700	3.4103
0.507	7.74	1750	3.2958
0.3741	7.96	1800	3.9773
0.297	8.19	1850	4.1055
0.4131	8.41	1900	3.7588
0.3332	8.63	1950	3.6752
0.2608	8.85	2000	4.2848
0.2652	9.07	2050	3.6098
0.1967	9.29	2100	3.7971
0.3771	9.51	2150	3.6206
0.1677	9.73	2200	4.3747
0.1711	9.96	2250	4.2229
0.0949	10.18	2300	4.5390
0.1804	10.4	2350	4.3297
0.1313	10.62	2400	4.3467
0.2292	10.84	2450	4.4368
0.1342	11.06	2500	4.2101
0.1411	11.28	2550	4.2723
0.0907	11.5	2600	4.3821
0.1656	11.73	2650	4.3373
0.0757	11.95	2700	4.3297
0.1571	12.17	2750	4.1193
0.1015	12.39	2800	4.2492
0.0785	12.61	2850	4.3998
0.1003	12.83	2900	4.3994
0.0884	13.05	2950	4.4914
0.0261	13.27	3000	4.5869
0.1138	13.5	3050	4.4546
0.1481	13.72	3100	4.6197
0.0814	13.94	3150	4.4034
0.0079	14.16	3200	4.7623
0.0641	14.38	3250	4.5933
0.0499	14.6	3300	4.6543
0.1052	14.82	3350	4.2287
0.0775	15.04	3400	4.3898
0.0561	15.27	3450	4.5852
0.0443	15.49	3500	4.6139
0.006	15.71	3550	4.7572
0.1517	15.93	3600	4.5492
0.0473	16.15	3650	4.6196
0.0534	16.37	3700	4.5726
0.0069	16.59	3750	4.7234
0.0515	16.81	3800	4.7475
0.0187	17.04	3850	4.7945
0.0616	17.26	3900	4.7330
0.0385	17.48	3950	4.6939
0.0377	17.7	4000	4.7966
0.0039	17.92	4050	4.7507
0.027	18.14	4100	4.7653
0.0193	18.36	4150	4.7589
0.0318	18.58	4200	4.8174
0.0094	18.81	4250	4.8776
0.035	19.03	4300	4.9437
0.0118	19.25	4350	4.9520
0.0036	19.47	4400	4.9428
0.0082	19.69	4450	4.9414
0.004	19.91	4500	4.9306

Framework versions

• Transformers 4.32.0
• Pytorch 2.0.0+cu118
• Datasets 2.17.1
• Tokenizers 0.13.2