cleanlab/label-errors
🛠️ Corrected Test Sets for ImageNet, MNIST, CIFAR, Caltech-256, QuickDraw, IMDB, Amazon Reviews, 20News, and AudioSet
This project helps machine learning researchers and practitioners evaluate their models more accurately by providing tools to identify and correct mislabeled examples in popular benchmark test datasets like ImageNet, MNIST, and CIFAR. It takes original test data and labels as input, along with model predictions, and outputs identified label errors and corrected labels. This is for anyone training and testing machine learning models who needs reliable evaluation metrics.
187 stars.
Use this if you need to ensure the quality and integrity of your model's evaluation by identifying and correcting label errors in standard ML benchmark test sets.
Not ideal if you are looking for a fully pre-corrected, single-file test set for immediate download without any customization options.
Stars
187
Forks
11
Language
—
License
Apache-2.0
Category
Last pushed
Dec 16, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/cleanlab/label-errors"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
open-edge-platform/datumaro
Dataset Management Framework, a Python library and a CLI tool to build, analyze and manage...
explosion/ml-datasets
🌊 Machine learning dataset loaders for testing and example scripts
webdataset/webdataset
A high-performance Python-based I/O system for large (and small) deep learning problems, with...
tensorflow/datasets
TFDS is a collection of datasets ready to use with TensorFlow, Jax, ...
mlcommons/croissant
Croissant is a high-level format for machine learning datasets that brings together four rich layers.