Getting Started

How to get started with Skrub

1

Install Skrub

Run `pip install skrub` to install the library.

2

Import Modules

Import needed components, for example: `from skrub import TableReport, Cleaner`.

3

Generate Data Exploration Report

Create a report with `report = TableReport(df).render()` where `df` is your dataframe.

4

Build Preprocessing Pipeline

Chain Cleaner and TableVectorizer to clean and engineer features from your dataframe.

5

Validate Multi-Table Pipelines

Use MultiTableTransformer to build and validate pipelines involving multiple dataframes with hyperparameter tuning.