Modelpublic

wandb/sourcecode-detection

sourceHugging Faceupdated 2y agoView on Hugging Face

0likes14downloads

Model Card

CodeBERTa-small-v1-sourcecode-detection-clf

This model is a fine-tuned version of huggingface/CodeBERTa-small-v1 on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 0.0171
—F1: 0.9975
—Accuracy: 0.9975
—Precision: 0.9975
—Recall: 0.9975

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: 0.0003
—trainbatchsize: 320
—evalbatchsize: 320
—seed: 2024
—optimizer: Use OptimizerNames.ADAMWTORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
—lrschedulertype: cosine
—lrschedulerwarmup_ratio: 0.1
—num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Accuracy	Precision	Recall
No log	0	0	0.6981	0.3337	0.5001	0.6162	0.5001
0.0294	0.1420	1000	0.0398	0.9947	0.9947	0.9947	0.9947
0.0076	0.2841	2000	0.0211	0.9968	0.9968	0.9968	0.9968
0.0053	0.4261	3000	0.0188	0.9973	0.9973	0.9973	0.9973
0.0056	0.5681	4000	0.0166	0.9976	0.9976	0.9976	0.9976
0.0044	0.7101	5000	0.0172	0.9975	0.9975	0.9975	0.9975
0.0009	0.8522	6000	0.0171	0.9975	0.9975	0.9975	0.9975
0.0052	0.9942	7000	0.0171	0.9975	0.9975	0.9975	0.9975

Framework versions

—Transformers 4.46.3
—Pytorch 2.5.1
—Datasets 3.1.0
—Tokenizers 0.20.3