layout2_large

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

• Loss: 5.2238

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
3.6695	0.22	50	2.4612
2.3221	0.44	100	1.9209
2.1651	0.66	150	1.7838
1.7644	0.88	200	1.9484
1.8272	1.11	250	1.7739
1.2938	1.33	300	1.7310
1.477	1.55	350	1.7254
1.4623	1.77	400	1.4813
1.2976	1.99	450	1.8113
0.8746	2.21	500	2.0039
0.8223	2.43	550	2.5449
1.0367	2.65	600	2.3888
1.0897	2.88	650	1.8641
0.6269	3.1	700	3.3845
0.7064	3.32	750	3.5282
0.8204	3.54	800	2.2839
0.9558	3.76	850	3.1004
0.8255	3.98	900	2.5468
0.778	4.2	950	4.6735
0.8192	4.42	1000	3.7428
0.798	4.65	1050	3.7837
0.5329	4.87	1100	3.2764
0.7568	5.09	1150	3.5623
0.5929	5.31	1200	3.1992
0.4427	5.53	1250	4.6837
0.7004	5.75	1300	2.6854
0.4315	5.97	1350	3.0840
0.2394	6.19	1400	4.3777
0.4605	6.42	1450	4.0313
0.4959	6.64	1500	3.3943
0.618	6.86	1550	3.4803
0.254	7.08	1600	4.0659
0.2255	7.3	1650	3.5000
0.3048	7.52	1700	4.0518
0.5369	7.74	1750	3.9500
0.3641	7.96	1800	3.7982
0.1725	8.19	1850	3.8756
0.3369	8.41	1900	4.1205
0.3522	8.63	1950	4.3853
0.4681	8.85	2000	3.2152
0.0857	9.07	2050	4.1299
0.36	9.29	2100	3.5658
0.3325	9.51	2150	3.8848
0.1645	9.73	2200	3.8714
0.2046	9.96	2250	3.9900
0.1267	10.18	2300	4.6510
0.2284	10.4	2350	4.1686
0.173	10.62	2400	4.0575
0.219	10.84	2450	4.2034
0.1107	11.06	2500	4.5646
0.0916	11.28	2550	4.9722
0.1358	11.5	2600	5.0297
0.0681	11.73	2650	5.2933
0.0447	11.95	2700	5.4102
0.0655	12.17	2750	5.8173
0.0694	12.39	2800	6.3645
0.1717	12.61	2850	5.4551
0.0197	12.83	2900	5.3129
0.0319	13.05	2950	5.2423
0.0	13.27	3000	5.3664
0.124	13.5	3050	5.2445
0.0025	13.72	3100	5.5092
0.1243	13.94	3150	5.3060
0.071	14.16	3200	5.3991
0.0301	14.38	3250	5.2129
0.0059	14.6	3300	5.3309
0.1027	14.82	3350	4.6790
0.0001	15.04	3400	5.4300
0.0	15.27	3450	5.4420
0.0046	15.49	3500	5.7107
0.0488	15.71	3550	5.3521
0.0797	15.93	3600	4.9206
0.06	16.15	3650	4.6152
0.0295	16.37	3700	4.6441
0.0126	16.59	3750	4.8442
0.0	16.81	3800	4.8838
0.0004	17.04	3850	5.0888
0.0	17.26	3900	5.1360
0.0003	17.48	3950	5.4019
0.0583	17.7	4000	5.0542
0.0	17.92	4050	4.9908
0.0001	18.14	4100	5.1099
0.0194	18.36	4150	5.1146
0.0002	18.58	4200	5.2076
0.0	18.81	4250	5.3454
0.0249	19.03	4300	5.3082
0.0329	19.25	4350	5.2274
0.0	19.47	4400	5.2129
0.0	19.69	4450	5.2155
0.0009	19.91	4500	5.2238

Framework versions

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