# How to split data in ML Multimetric job

**URL:** <https://discuss.elastic.co/t/how-to-split-data-in-ml-multimetric-job/162023>\
**Category:** Elasticsearch\
**Tags:** elastic-stack-machine-learning\
**Created:** [December 24, 2018, 7:02pm UTC](https://discuss.elastic.co/t/how-to-split-data-in-ml-multimetric-job/162023 "2018-12-24T19:02:01Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![sesha](https://avatars.discourse-cdn.com/v4/letter/s/eb9ed0/32.png) [@sesha](https://discuss.elastic.co/u/sesha)\
**Post date:** [December 24, 2018, 7:02pm UTC](https://discuss.elastic.co/t/how-to-split-data-in-ml-multimetric-job/162023/1 "2018-12-24T19:02:01Z")

</div>

I was trying to do Multimetric ml job with sample data(kibana\_sample\_data\_flights), to get anomalies for flights from different Origin(Origin, OriginCountry, OriginCityName, OriginAirportID ..) to the destination ( Dest, DestCountry, DestCityName, DestAirportID ) for each Carrier for metric average AvgTicketPrice and event rate.  
I know that should use split. But how do I split by Origin, Dest, Carrier for each metric (AvgTicketPrice, eventrate).

I have following doubts

- How to split **Origin** (Origin, OriginCountry, OriginCityName, OriginAirportID ..) fields.
- How to split **Dest** ( Dest, DestCountry, DestCityName, DestAirportID ).
- How to split **Origin** (Origin, OriginCountry, OriginCityName, OriginAirportID ..) to **Dest** ( Dest, DestCountry, DestCityName, DestAirportID ) by **Carrier** for metric (mean of AvgTicketPrice).

```auto
{
    "FlightNum": "Y56DWSH",
    "DestCountry": "RU",
    "OriginWeather": "Cloudy",
    "OriginCityName": "Durban",
    "AvgTicketPrice": 404.7860204873357,
    "DistanceMiles": 6736.8698449714075,
    "FlightDelay": false,
    "DestWeather": "Sunny",
    "Dest": "Olenya Air Base",
    "FlightDelayType": "No Delay",
    "OriginCountry": "ZA",
    "dayOfWeek": 0,
    "DistanceKilometers": 10841.941063785665,
    "timestamp": "2018-12-24T18:22:55",
    "DestLocation": {
      "lat": "68.15180206",
      "lon": "33.46390152"
    },
    "DestAirportID": "XLMO",
    "Carrier": "ES-Air",
    "Cancelled": false,
    "FlightTimeMin": 602.3300590992036,
    "Origin": "King Shaka International Airport",
    "OriginLocation": {
      "lat": "-29.61444444",
      "lon": "31.11972222"
    },
    "DestRegion": "RU-MUR",
    "OriginAirportID": "DUR",
    "OriginRegion": "SE-BD",
    "DestCityName": "Olenegorsk",
    "FlightTimeHour": 10.03883431832006,
    "FlightDelayMin": 0
  }

```

---

<div class="post-metadata">

**Author:** ![richcollier](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/richcollier/32/115035_2.png) [@richcollier](https://discuss.elastic.co/u/richcollier)\
**Post date:** [December 25, 2018, 1:23pm UTC](https://discuss.elastic.co/t/how-to-split-data-in-ml-multimetric-job/162023/2 "2018-12-25T13:23:15Z")

</div>

The multi-mettic UI only allows one split at a time. If you want to do more splits you can make separate jobs.

---

<div class="post-metadata">

**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\
**Post date:** [January 22, 2019, 1:23pm UTC](https://discuss.elastic.co/t/how-to-split-data-in-ml-multimetric-job/162023/3 "2019-01-22T13:23:23Z")

</div>

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.
