codice_fiscale

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

• Loss: 0.2024
• Precision: 0.8316
• Recall: 0.5374
• F1: 0.6529
• Accuracy: 0.9405

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	4	1.0541	0.0	0.0	0.0	0.8445
No log	2.0	8	0.6374	0.0	0.0	0.0	0.8445
No log	3.0	12	0.5150	0.0	0.0	0.0	0.8445
No log	4.0	16	0.4235	0.0	0.0	0.0	0.8445
No log	5.0	20	0.3564	0.5	0.0850	0.1453	0.8667
No log	6.0	24	0.3024	0.5	0.0850	0.1453	0.8667
No log	7.0	28	0.2609	0.6835	0.1837	0.2895	0.8796
No log	8.0	32	0.2299	0.8264	0.4048	0.5434	0.9085
No log	9.0	36	0.2104	0.7826	0.4898	0.6025	0.9280
No log	10.0	40	0.2024	0.8316	0.5374	0.6529	0.9405

Framework versions

• Transformers 4.40.2
• Pytorch 2.2.1+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1