ser-model-microsoft

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: 0.2762
• Precision: 0.6
• Recall: 0.9
• F1: 0.7200
• Accuracy: 0.925

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: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	5.0	10	0.9255	0.0	0.0	0.0	0.7
No log	10.0	20	0.6668	0.4444	0.4	0.4211	0.7875
No log	15.0	30	0.4304	0.6667	0.8	0.7273	0.85
No log	20.0	40	0.4050	0.6667	0.8	0.7273	0.85
No log	25.0	50	0.5639	0.8	0.8	0.8000	0.8125
No log	30.0	60	0.2429	0.8	0.8	0.8000	0.925
No log	35.0	70	0.4434	0.6667	0.8	0.7273	0.8625
No log	40.0	80	0.2817	0.6	0.9	0.7200	0.925
No log	45.0	90	0.2784	0.6	0.9	0.7200	0.925
No log	50.0	100	0.2762	0.6	0.9	0.7200	0.925

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.1