our-dataset

This model is a fine-tuned version of microsoft/layoutlmv3-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.0167
• Precision: 0.7541
• Recall: 0.6479
• F1: 0.6970
• Accuracy: 0.7975

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: 1000

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	4.55	100	1.2849	0.5789	0.4648	0.5156	0.6456
No log	9.09	200	1.0959	0.6724	0.5493	0.6047	0.7215
No log	13.64	300	1.1048	0.6833	0.5775	0.6260	0.7342
No log	18.18	400	1.0442	0.7541	0.6479	0.6970	0.7848
0.488	22.73	500	1.0966	0.7333	0.6197	0.6718	0.7722
0.488	27.27	600	1.0650	0.75	0.6338	0.6870	0.7848
0.488	31.82	700	0.9722	0.7742	0.6761	0.7218	0.8101
0.488	36.36	800	1.0596	0.7541	0.6479	0.6970	0.7975
0.488	40.91	900	0.9996	0.7541	0.6479	0.6970	0.7975
0.0298	45.45	1000	1.0167	0.7541	0.6479	0.6970	0.7975

Framework versions

• Transformers 4.31.0.dev0
• Pytorch 1.12.0+cu102
• Datasets 2.13.1
• Tokenizers 0.13.3