layoutmlv3_thursday_oct4_v7

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: 0.2861
• Precision: 0.8352
• Recall: 0.7894
• F1: 0.8116
• Accuracy: 0.9586

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	1.12	100	0.2568	0.8574	0.7770	0.8152	0.9586
No log	2.25	200	0.2653	0.8268	0.7858	0.8058	0.9581
No log	3.37	300	0.2728	0.7982	0.7770	0.7874	0.9565
No log	4.49	400	0.2626	0.8569	0.7735	0.8130	0.9589
0.114	5.62	500	0.2861	0.8352	0.7894	0.8116	0.9586
0.114	6.74	600	0.2978	0.8205	0.7929	0.8065	0.9582
0.114	7.87	700	0.2942	0.8256	0.7876	0.8062	0.9584
0.114	8.99	800	0.2910	0.8420	0.7735	0.8063	0.9579
0.114	10.11	900	0.3028	0.8346	0.7770	0.8048	0.9574
0.0846	11.24	1000	0.2989	0.8318	0.7876	0.8091	0.9581

Framework versions

• Transformers 4.35.0.dev0
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.14.0