models

This model is a fine-tuned version of microsoft/layoutlmv3-base on the data_loader dataset. It achieves the following results on the evaluation set:

• Loss: 0.1595
• Precision: 0.8940
• Recall: 0.9169
• F1: 0.9053
• Accuracy: 0.9744

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	2.5	100	0.1926	0.7730	0.8274	0.7993	0.9452
No log	5.0	200	0.1342	0.8285	0.8708	0.8491	0.9583
No log	7.5	300	0.1217	0.8758	0.9015	0.8885	0.9693
No log	10.0	400	0.1157	0.9082	0.9233	0.9157	0.9769
0.15	12.5	500	0.1310	0.9011	0.9092	0.9052	0.9744
0.15	15.0	600	0.1583	0.8682	0.9015	0.8846	0.9693
0.15	17.5	700	0.1628	0.8867	0.9105	0.8984	0.9724
0.15	20.0	800	0.1594	0.8945	0.9220	0.9081	0.9749
0.15	22.5	900	0.1579	0.8940	0.9169	0.9053	0.9744
0.0047	25.0	1000	0.1595	0.8940	0.9169	0.9053	0.9744

Framework versions

• Transformers 4.40.0.dev0
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2