Modelpublic

buelfhood/conplag1_codet5_ep30_bs16_lr2e-05_l512_s42_ppy_loss

sourceHugging Faceapache-2.0updated 11mo agoView on Hugging Face

0likes42downloads

Model Card

conplag1codet5ep30bs16lr2e-05l512s42ppyloss

This model is a fine-tuned version of Salesforce/codet5-small on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 0.4776
—Accuracy: 0.7664
—Recall: 0.6842
—Precision: 0.5652
—F1: 0.6190
—F Beta Score: 0.6426

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: 2e-05
—trainbatchsize: 16
—evalbatchsize: 16
—seed: 42
—optimizer: Use OptimizerNames.ADAMWTORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
—lrschedulertype: linear
—num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Recall	Precision	F1	F Beta Score
0.7264	1.0	40	0.6241	0.5620	0.7105	0.3553	0.4737	0.5433
0.7518	2.0	80	0.6004	0.6131	0.7368	0.3944	0.5138	0.5815
0.5962	3.0	120	0.5764	0.7080	0.7368	0.4828	0.5833	0.6341
0.5816	4.0	160	0.5301	0.7372	0.7368	0.5185	0.6087	0.6523
0.4856	5.0	200	0.4776	0.7664	0.6842	0.5652	0.6190	0.6426
0.4802	6.0	240	0.4909	0.7518	0.7105	0.54	0.6136	0.6476
0.3683	7.0	280	0.5882	0.5693	0.9211	0.3846	0.5426	0.6445
0.4059	8.0	320	0.5196	0.8248	0.5263	0.7692	0.625	0.5830

Framework versions

—Transformers 4.51.3
—Pytorch 2.8.0+cu128
—Datasets 3.1.0
—Tokenizers 0.21.4