layoutmlv3_sunday_sep3_v5

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: 0.1630
• Precision: 0.6867
• Recall: 0.7308
• F1: 0.7081
• Accuracy: 0.9570

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	8.33	100	0.5150	0.5139	0.4744	0.4933	0.8460
No log	16.67	200	0.2462	0.5053	0.6154	0.5549	0.9387
No log	25.0	300	0.1973	0.6471	0.7051	0.6748	0.9536
No log	33.33	400	0.1617	0.6667	0.7179	0.6914	0.9520
0.3718	41.67	500	0.1630	0.6867	0.7308	0.7081	0.9570
0.3718	50.0	600	0.2247	0.5106	0.6154	0.5581	0.9073
0.3718	58.33	700	0.3364	0.5393	0.6154	0.5749	0.8907
0.3718	66.67	800	0.1783	0.5435	0.6410	0.5882	0.9454
0.3718	75.0	900	0.2255	0.5263	0.6410	0.5780	0.9305
0.0196	83.33	1000	0.2781	0.5158	0.6282	0.5665	0.9123

Framework versions

• Transformers 4.33.0.dev0
• Pytorch 2.0.1+cu118
• Datasets 2.14.4
• Tokenizers 0.13.3