layoutlmv3-finetuned-cord_100

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

• Loss: 0.3755
• Precision: 0.8789
• Recall: 0.8825
• F1: 0.8807
• Accuracy: 0.8920

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: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• training_steps: 1000

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	0.25	100	1.2179	0.5929	0.6203	0.6063	0.7059
No log	0.5	200	0.8495	0.8007	0.7808	0.7906	0.8147
No log	0.75	300	0.6782	0.8353	0.8180	0.8266	0.8339
No log	1.0	400	0.5858	0.8430	0.8304	0.8367	0.8417
0.9914	1.25	500	0.5506	0.8573	0.8395	0.8483	0.8490
0.9914	1.5	600	0.4891	0.8670	0.8519	0.8594	0.8554
0.9914	1.75	700	0.4376	0.8684	0.8569	0.8626	0.8628
0.9914	2.0	800	0.3956	0.8768	0.8768	0.8768	0.8820
0.9914	2.25	900	0.3797	0.8798	0.8776	0.8787	0.8852
0.4483	2.5	1000	0.3755	0.8789	0.8825	0.8807	0.8920

Framework versions

• Transformers 4.46.3
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.20.3