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: 0.4674
• Accuracy: 0.7824
• Precision: 0.0804
• Recall: 0.6957
• F1 Score: 0.1441
• F2 Score: 0.2749
• Gmean: 0.7388

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	70	0.6435	0.6495	0.0450	0.6087	0.0838	0.1737	0.6293
0.769	2.0	140	0.4895	0.7995	0.0824	0.6522	0.1463	0.2737	0.7239
0.6429	3.0	210	0.4972	0.7950	0.0806	0.6522	0.1435	0.2698	0.7218
0.6429	4.0	280	0.5368	0.7755	0.0739	0.6522	0.1327	0.2542	0.7127
0.6276	5.0	350	0.4395	0.8064	0.0805	0.6087	0.1421	0.2632	0.7029
0.5782	6.0	420	0.4263	0.8167	0.0898	0.6522	0.1579	0.2896	0.7318
0.5782	7.0	490	0.4501	0.7973	0.0815	0.6522	0.1449	0.2717	0.7228
0.5599	8.0	560	0.4610	0.7950	0.0806	0.6522	0.1435	0.2698	0.7218
0.5624	9.0	630	0.5381	0.7595	0.0807	0.7826	0.1463	0.2857	0.7706
0.5356	10.0	700	0.4899	0.7755	0.0780	0.6957	0.1404	0.2694	0.7355
0.5356	11.0	770	0.5032	0.7675	0.0794	0.7391	0.1435	0.2778	0.7535
0.5576	12.0	840	0.4943	0.7709	0.0806	0.7391	0.1453	0.2805	0.7553
0.5145	13.0	910	0.4752	0.7801	0.0837	0.7391	0.1504	0.2881	0.7599
0.5145	14.0	980	0.4725	0.7824	0.0846	0.7391	0.1518	0.2901	0.7610
0.531	15.0	1050	0.4674	0.7824	0.0804	0.6957	0.1441	0.2749	0.7388

Framework versions

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