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: 0.5735
• Accuracy: 0.7238
• Precision: 0.0709
• Recall: 0.6786
• F1 Score: 0.1284
• F2 Score: 0.25
• Gmean: 0.7015

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	72	0.5482	0.7152	0.0388	0.3571	0.0699	0.1351	0.5093
0.8958	2.0	144	0.5244	0.7409	0.0504	0.4286	0.0902	0.1714	0.5672
0.7174	3.0	216	0.5339	0.7334	0.0632	0.5714	0.1139	0.2192	0.6496
0.7174	4.0	288	0.5268	0.7355	0.0706	0.6429	0.1272	0.2452	0.6890
0.6391	5.0	360	0.5969	0.7034	0.0632	0.6429	0.1150	0.2267	0.6734
0.6017	6.0	432	0.5509	0.7206	0.0669	0.6429	0.1212	0.2362	0.6817
0.607	7.0	504	0.5780	0.7141	0.0686	0.6786	0.1246	0.2442	0.6967
0.607	8.0	576	0.6406	0.6809	0.0645	0.7143	0.1183	0.2370	0.6969
0.5856	9.0	648	0.5832	0.7131	0.0714	0.7143	0.1299	0.2551	0.7137
0.5522	10.0	720	0.5944	0.7131	0.0714	0.7143	0.1299	0.2551	0.7137
0.5522	11.0	792	0.5861	0.7227	0.0706	0.6786	0.1279	0.2493	0.7009
0.5635	12.0	864	0.5801	0.7227	0.0706	0.6786	0.1279	0.2493	0.7009
0.5709	13.0	936	0.5526	0.7355	0.0739	0.6786	0.1333	0.2575	0.7073
0.5488	14.0	1008	0.5744	0.7248	0.0712	0.6786	0.1288	0.2507	0.7020
0.5488	15.0	1080	0.5735	0.7238	0.0709	0.6786	0.1284	0.25	0.7015

Framework versions

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