codet5-base-sql-create-context

This model is a fine-tuned version of Salesforce/codet5-base on b-mc2/sql-create-context dataset. It achieves the following results on the evaluation set:

• Loss: 0.0036
• Exact Match: 0.2451
• Rouge2: 0.8839
• Bleu: 75.6780

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 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	Validation Loss	Exact Match	Rouge2	Bleu
0.0071	1.0	7858	0.0047	0.2360	0.8808	75.1891
0.0041	2.0	15716	0.0039	0.2393	0.8820	75.4061
0.0028	3.0	23574	0.0037	0.2447	0.8832	75.5863
0.0019	4.0	31432	0.0035	0.2469	0.8837	75.7130
0.0013	5.0	39290	0.0036	0.2451	0.8839	75.6780

Framework versions

• Transformers 4.50.3
• Pytorch 2.5.1+cu121
• Datasets 3.5.0
• Tokenizers 0.21.1