grecosalvatore/drift-lens

Drift-Lens: an Unsupervised Drift Detection Framework for Deep Learning Classifiers on Unstructured Data

41
/ 100
Emerging

DriftLens helps machine learning engineers or MLOps practitioners continuously monitor deep learning models that process unstructured data like images, audio, or text. It takes the model's output embeddings and predicted labels, then detects and characterizes when the real-world data distribution shifts away from what the model was trained on. This allows for proactive model retraining or intervention before performance degrades.

Use this if you need to automatically detect unexpected changes in your real-time input data that could cause your deployed deep learning classifiers to perform poorly.

Not ideal if you are working with traditional machine learning models or structured (tabular) data, or if you need to detect drift in the model's predictions rather than its input data representations.

MLOps Deep Learning Monitoring Data Quality Model Reliability Unstructured Data
No Package No Dependents
Maintenance 10 / 25
Adoption 6 / 25
Maturity 16 / 25
Community 9 / 25

How are scores calculated?

Stars

16

Forks

2

Language

Jupyter Notebook

License

MIT

Last pushed

Feb 11, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/grecosalvatore/drift-lens"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.