Quick Start

This section shows how to load your first dataset using the unified get_<dataset>() API.

Basic example

Load any dataset in a familiar format (e.g., a pandas DataFrame) with a single function call:

import dataset_hub

dataset = dataset_hub.classification.get_titanic() # pd.DataFrame

After calling the function, a link to the dataset page will appear in the log output. It contains detailed information about the dataset (description, columns, target, etc.).

Next steps

You can proceed in two ways:

• Start your own experiments with the dataset (split into train and test, build validation, perform feature engineering, build models, etc.)

• Use a ready-made baseline provided in the documentation. For example, here is a simple baseline for Titanic:

Available datasets

See the full list here: Datasets

Or jump directly to the quick examples:

• Titanic

• Iris

• California Housing Prices

• Household Electric Power Consumption