results_v3

This model is a fine-tuned version of Salesforce/codet5-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7503
• Accuracy: 0.6336
• Precision: 0.1607
• Recall: 0.9
• F1 Score: 0.2727
• F2 Score: 0.4688
• Gmean: 0.7419

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: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1 Score	F2 Score	Gmean
No log	1.0	57	0.9026	0.4809	0.0735	0.5	0.1282	0.2315	0.4896
0.7464	2.0	114	0.7800	0.5725	0.1034	0.6	0.1765	0.3061	0.5849
0.7464	3.0	171	0.7762	0.5878	0.1452	0.9	0.25	0.4412	0.7112
0.6653	4.0	228	0.8069	0.5725	0.1406	0.9	0.2432	0.4327	0.7006
0.6653	5.0	285	0.7910	0.5954	0.1475	0.9	0.2535	0.4455	0.7164
0.6437	6.0	342	0.7414	0.6412	0.1636	0.9	0.2769	0.4737	0.7469
0.6437	7.0	399	0.8038	0.5954	0.1475	0.9	0.2535	0.4455	0.7164
0.6328	8.0	456	0.6908	0.6794	0.18	0.9	0.3	0.5000	0.7714
0.5966	9.0	513	0.7782	0.6183	0.1552	0.9	0.2647	0.4592	0.7318
0.5966	10.0	570	0.7343	0.6565	0.1698	0.9	0.2857	0.4839	0.7568
0.5819	11.0	627	0.7569	0.6260	0.1579	0.9	0.2687	0.4639	0.7369
0.5819	12.0	684	0.7650	0.6183	0.1552	0.9	0.2647	0.4592	0.7318
0.5643	13.0	741	0.7711	0.6260	0.1579	0.9	0.2687	0.4639	0.7369
0.5643	14.0	798	0.7470	0.6336	0.1607	0.9	0.2727	0.4688	0.7419
0.569	15.0	855	0.7503	0.6336	0.1607	0.9	0.2727	0.4688	0.7419

Framework versions

• PEFT 0.14.0
• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.3.1
• Tokenizers 0.21.0