Output_LayoutLMv3_1

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: 0.2963
• Precision: 0.8017
• Recall: 0.8407
• F1: 0.8207
• Accuracy: 0.9724

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: 1e-05
• 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	2.27	100	0.1310	0.7254	0.7832	0.7532	0.9619
No log	4.55	200	0.1070	0.8333	0.8407	0.8370	0.9733
No log	6.82	300	0.1617	0.8421	0.8496	0.8458	0.9752
No log	9.09	400	0.1895	0.8145	0.7965	0.8054	0.9705
0.074	11.36	500	0.1689	0.8018	0.7876	0.7946	0.9686
0.074	13.64	600	0.1659	0.8291	0.8584	0.8435	0.9771
0.074	15.91	700	0.2043	0.8077	0.8363	0.8217	0.9705
0.074	18.18	800	0.2111	0.8151	0.8584	0.8362	0.9743
0.074	20.45	900	0.1938	0.8390	0.8761	0.8571	0.9762
0.0072	22.73	1000	0.1802	0.8319	0.8540	0.8428	0.98
0.0072	25.0	1100	0.2604	0.7940	0.8186	0.8061	0.9695
0.0072	27.27	1200	0.2550	0.8025	0.8451	0.8233	0.9733
0.0072	29.55	1300	0.2632	0.8077	0.8363	0.8217	0.9705
0.0072	31.82	1400	0.2865	0.8155	0.8407	0.8279	0.9714
0.0018	34.09	1500	0.2554	0.8253	0.8363	0.8308	0.9743
0.0018	36.36	1600	0.2763	0.8101	0.8496	0.8294	0.9743
0.0018	38.64	1700	0.2541	0.8197	0.8451	0.8322	0.9743
0.0018	40.91	1800	0.2785	0.8025	0.8628	0.8316	0.9724
0.0018	43.18	1900	0.2760	0.8059	0.8451	0.8251	0.9733
0.001	45.45	2000	0.2956	0.8155	0.8407	0.8279	0.9733
0.001	47.73	2100	0.2998	0.8017	0.8407	0.8207	0.9724
0.001	50.0	2200	0.3007	0.8017	0.8407	0.8207	0.9724
0.001	52.27	2300	0.3000	0.8017	0.8407	0.8207	0.9724
0.001	54.55	2400	0.3031	0.8017	0.8407	0.8207	0.9724
0.0002	56.82	2500	0.3174	0.7950	0.8407	0.8172	0.9714
0.0002	59.09	2600	0.3061	0.7950	0.8407	0.8172	0.9714
0.0002	61.36	2700	0.3059	0.7908	0.8363	0.8129	0.9705
0.0002	63.64	2800	0.2988	0.8017	0.8407	0.8207	0.9724
0.0002	65.91	2900	0.2972	0.8017	0.8407	0.8207	0.9724
0.0001	68.18	3000	0.2963	0.8017	0.8407	0.8207	0.9724

Framework versions

• Transformers 4.39.2
• Pytorch 2.2.2+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2