model-caisse-2024-06-13

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: 1.5952
• Precision: 0.4889
• Recall: 0.5641
• F1: 0.5238
• Accuracy: 0.8098

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	12.5	100	1.1305	0.2727	0.2308	0.2500	0.7730
No log	25.0	200	1.0726	0.5	0.5128	0.5063	0.8160
No log	37.5	300	1.3312	0.4565	0.5385	0.4941	0.8037
No log	50.0	400	1.4504	0.4348	0.5128	0.4706	0.7975
0.5177	62.5	500	1.5561	0.4255	0.5128	0.4651	0.7914
0.5177	75.0	600	1.6232	0.4565	0.5385	0.4941	0.7975
0.5177	87.5	700	1.6746	0.4375	0.5385	0.4828	0.7853
0.5177	100.0	800	1.6196	0.4255	0.5128	0.4651	0.7914
0.5177	112.5	900	1.5931	0.4889	0.5641	0.5238	0.8098
0.0356	125.0	1000	1.5952	0.4889	0.5641	0.5238	0.8098

Framework versions

• Transformers 4.29.2
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.13.3