> ## 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.

# Ray

> Run StatsForecast distributedly on top of Ray.

StatsForecast works on top of Spark, Dask, and Ray through
[Fugue](https://github.com/fugue-project/fugue/). StatsForecast will
read the input DataFrame and use the corresponding engine. For example,
if the input is a Ray Dataset, StatsForecast will use the existing Ray
instance to run the forecast.

A benchmark (with older syntax) can be found
[here](https://www.anyscale.com/blog/how-nixtla-uses-ray-to-accurately-predict-more-than-a-million-time-series)
where we forecasted one million timeseries in under half an hour.

## Installation

As long as Ray is installed and configured, StatsForecast will be able
to use it. If executing on a distributed Ray cluster, make use the
`statsforecast` library is installed across all the workers.

## StatsForecast on Pandas

Before running on Ray, it’s recommended to test on a smaller Pandas
dataset to make sure everything is working. This example also helps show
the small differences when using Ray.

```python theme={null}
from statsforecast.core import StatsForecast
from statsforecast.models import AutoARIMA, AutoETS
from statsforecast.utils import generate_series
```

```python theme={null}
n_series = 4
horizon = 7

series = generate_series(n_series)

sf = StatsForecast(
    models=[AutoETS(season_length=7)],
    freq='D',
)
sf.forecast(df=series, h=horizon).head()
```

|   | unique\_id | ds         | AutoETS  |
| - | ---------- | ---------- | -------- |
| 0 | 0          | 2000-08-10 | 5.261609 |
| 1 | 0          | 2000-08-11 | 6.196357 |
| 2 | 0          | 2000-08-12 | 0.282309 |
| 3 | 0          | 2000-08-13 | 1.264195 |
| 4 | 0          | 2000-08-14 | 2.262453 |

## Executing on Ray

To run the forecasts distributed on Ray, just pass in a Ray Dataset
instead.

```python theme={null}
import ray
import logging
```

```python theme={null}
ray.init(logging_level=logging.ERROR)

series['unique_id'] = series['unique_id'].astype(str)
ctx = ray.data.context.DatasetContext.get_current()
ctx.use_streaming_executor = False
ray_series = ray.data.from_pandas(series).repartition(4)
```

```python theme={null}
sf.forecast(df=ray_series, h=horizon).take(5)
```