Data Tools
Make test data with a known answer, or clean a CSV step by step. Runs in your browser, and your files never leave your device.
Test Data GeneratorMake realistic fake data for testing, up to 5 million rows as CSV, JSON or SQL INSERT, and mix in bad records on purpose.
Synthetic data for MLMake a CSV where the answer is known, such as a linear, logistic or if-then rule, with an answer key. Test regression, decision trees and random forests.
Data cleaning stepsPaste a CSV and add cleaning steps in order, like a Java stream: fill, drop duplicates, replace, cap, sort, bin. See the rows after each step.
About these tools
The test data tools make files for trying out software and models: fake records for testing a database or an import, and synthetic CSV files where the rule behind the numbers is known, so you can check whether a model finds it.
The data cleaning page takes a CSV and runs steps on it one after another, such as removing duplicates, filling missing values and sorting. It shows the row count after each step and gives the same steps as pandas code.