Built-in dataset#


get_public_dataset will automatically download the public dataset specified by name and return it as a TSDataset instance.

In this guide, we will help you learn how to use this feature.


To begin our guide, you should prepare your environment.

!pip install --pre --upgrade bigdl-chronos[pytorch]

How to use#

You can select a dataset by specifying the name.

more dataset and TSDataset info, pelase refer to here.


We provide automatic download support for datasets other than tsinghua_electricity.

Please download tsinghua_electricity manually and move the downloaded csv file to the specified path, the default path is ~/.chronos/dataset/.

[ ]:
from import get_public_dataset
train, _, test = get_public_dataset(name='nyc_taxi')