codet5p_sql_finetuned

This model is a fine-tuned version of Salesforce/codet5p-770m on an unknown dataset. It achieves the following results on the evaluation set:

• Execution Match: 0.2771
• Loss: 0.0198
• Exact Match: 0.2514
• Invalid Sql: 0.1771

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Execution Match	Validation Loss	Exact Match	Invalid Sql
0.0139	1.0	788	0.24	0.0185	0.2057	0.1543
0.0085	2.0	1576	0.2486	0.0170	0.22	0.14
0.0066	3.0	2364	0.2486	0.0155	0.2286	0.18
0.0053	4.0	3152	0.2543	0.0153	0.2343	0.1771
0.0052	5.0	3940	0.2543	0.0149	0.2343	0.2171

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1