> ## Documentation Index
> Fetch the complete documentation index at: https://nixtlaverse.nixtla.io/llms.txt
> Use this file to discover all available pages before exploring further.

> Datasets for time series forecasting

# datasetsforecast

## Install

```sh theme={null}
pip install datasetsforecast
```

## Datasets

* [Favorita](./favorita.html)
* [Hierarchical](./hierarchical.html)
* [Longhorizon](./long_horizon.html)
* [M3](./m3.html)
* [M4](./m4.html)
* [M5](./m5.html)
* [PHM2008](./phm2008.html)

## How to use

All the modules have a `load` method which you can use to load the
dataset for a specific group. If you don’t have the data locally it will
be downloaded for you.

```python theme={null}
from datasetsforecast.phm2008 import PHM2008
```

```python theme={null}
train_df, test_df = PHM2008.load(directory='data', group='FD001')
train_df.shape, test_df.shape
```

((20631, 17), (13096, 17))