oneDAL is a C++ and DPC++ library for accelerated machine learning routines on tabular data, governed by the UXL Foundation.
Open source
24 alternativesCleanlab
Cleanlab provides a Python library for data quality and machine learning with messy real world data and labels.
About Cleanlab
Cleanlab is an open-source Python library for finding and fixing data and label problems in machine learning datasets. It uses an existing model's predictions to estimate issues such as label errors, outliers, and duplicates, and it can check text, audio, image, and tabular data.
The library also supports training. Its CleanLearning wrapper trains a more robust version of a classifier on noisy labels. For data labeled by several annotators, it can infer consensus and annotator quality, and it can suggest which items to label or relabel next. Dedicated functionality covers classification, multi-label and token classification, regression, image segmentation, object detection, and outlier detection. The source states that other ML tasks can still benefit if the library is appropriately applied.
Key features
- Detects label errors, outliers, and duplicates in text, audio, image, and tabular datasets
- Estimates dataset problems using predictions from your existing models
- Trains a robust version of a classifier that handles noisy labels
- Infers consensus and annotator quality for multi-annotator labeled data
- Suggests which data to label or relabel next through active learning
- Provides dedicated functionality for segmentation and object detection tasks
Best for: Machine learning practitioners who want to find and correct label and data-quality problems in existing datasets.
Good for
- The source states it works with any dataset and any model, including PyTorch, TensorFlow, scikit-learn, and XGBoost.
- The project is open source and hosted as a public GitHub repository.
- The package supports Linux, macOS, and Windows.
Watch out for
- It requires Python 3.10 or later.
- Dedicated functionality covers the listed task types; other ML tasks depend on being appropriately applied, according to the source.
Cleanlab at a glance
- Website
- github.com/cleanlab/cleanlab
- Category
- AI & Machine Learning › AI Data & Analytics
- Type
- Open-source project
- Pricing
- Not confirmed. Check the site
- Open source
- Yes · Source code
- Language
- English
- Last checked
- Oct 6, 2026 · Online when checked
- Listed
- Oct 5, 2026
Cleanlab alternatives
Tools and sites you could use instead, ranked by how closely they match.
This project profiles data quality and supports exploratory analysis of Pandas and Spark DataFrames.
Open source
24 alternatives
Lance provides an open lakehouse format for multimodal AI, with vector indexing and data versioning.
Open source
24 alternatives
Optimus supports agile data preparation workflows across Pandas, Dask, cuDF, Vaex, and PySpark.
Open source
24 alternatives
BambooAI analyzes data with a language model in a persistent Python kernel and supports web and Docker workflows.
Open source
24 alternatives
DVC tracks data versions and machine learning experiments for collaborative development.
Open source
24 alternativesPopular in AI Data & Analytics
Well-known listings from the same topic.
Dotgetdot.aiForvioforvio.comFluidAI Medicalfluidai.mdApache Zeppelinzeppelin.apache.orgSola Securitysola.securityQuadraticquadratichq.com
Cleanlab FAQ
Which Python versions does Cleanlab support?
The package runs on Python 3.10 and later.
How can Cleanlab be installed?
The source lists installation via uv, pip, or conda.
Is something wrong or out of date?
Tell us and an editor will check it.
Own this website?
Claim this listing
Claim the listing and suggest updates yourself. An editor checks every claim first.