Output_LayoutLMv3_v6

This model is a fine-tuned version of microsoft/layoutlmv3-large on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1308
• Precision: 0.7788
• Recall: 0.8
• F1: 0.7892
• Accuracy: 0.9637

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: 3e-07
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• training_steps: 3000

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.61	100	0.5057	0.0	0.0	0.0	0.8962
No log	3.23	200	0.3746	0.0	0.0	0.0	0.8962
No log	4.84	300	0.2979	0.2143	0.0273	0.0484	0.9014
No log	6.45	400	0.2474	0.4444	0.1455	0.2192	0.9135
0.4794	8.06	500	0.2189	0.5	0.3	0.3750	0.9291
0.4794	9.68	600	0.2031	0.5301	0.4	0.4560	0.9325
0.4794	11.29	700	0.1916	0.6	0.4636	0.5231	0.9377
0.4794	12.9	800	0.1788	0.6364	0.5727	0.6029	0.9412
0.4794	14.52	900	0.1728	0.6796	0.6364	0.6573	0.9464
0.184	16.13	1000	0.1644	0.7010	0.6182	0.6570	0.9481
0.184	17.74	1100	0.1593	0.75	0.7091	0.7290	0.9567
0.184	19.35	1200	0.1520	0.7714	0.7364	0.7535	0.9602
0.184	20.97	1300	0.1420	0.7778	0.7636	0.7706	0.9619
0.184	22.58	1400	0.1427	0.7925	0.7636	0.7778	0.9637
0.1278	24.19	1500	0.1361	0.7727	0.7727	0.7727	0.9619
0.1278	25.81	1600	0.1342	0.8019	0.7727	0.7870	0.9654
0.1278	27.42	1700	0.1310	0.8056	0.7909	0.7982	0.9671
0.1278	29.03	1800	0.1290	0.7857	0.8	0.7928	0.9654
0.1278	30.65	1900	0.1268	0.7946	0.8091	0.8018	0.9671
0.0999	32.26	2000	0.1229	0.7768	0.7909	0.7838	0.9637
0.0999	33.87	2100	0.1305	0.8056	0.7909	0.7982	0.9654
0.0999	35.48	2200	0.1349	0.8241	0.8091	0.8165	0.9689
0.0999	37.1	2300	0.1327	0.8018	0.8091	0.8054	0.9654
0.0999	38.71	2400	0.1289	0.8018	0.8091	0.8054	0.9654
0.0833	40.32	2500	0.1274	0.8018	0.8091	0.8054	0.9654
0.0833	41.94	2600	0.1279	0.8018	0.8091	0.8054	0.9654
0.0833	43.55	2700	0.1295	0.8018	0.8091	0.8054	0.9654
0.0833	45.16	2800	0.1306	0.7788	0.8	0.7892	0.9637
0.0833	46.77	2900	0.1312	0.7788	0.8	0.7892	0.9637
0.0749	48.39	3000	0.1308	0.7788	0.8	0.7892	0.9637

Framework versions

• Transformers 4.29.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.13.3