Channels
random
linen
statsforecast
hierarchicalforecast
neural-forecast
mlforecast
general
timegpt
Powered by
#general
Title
v
Valeriy
12/09/2022, 3:39 PM
Hi
@Max (Nixtla)
@fede
is Nixtla using Spark? From what I recollect Fugue was mentioned I presume Nixtla looked at Spark at some point and decided to use / not to use it. Was wondering if Spark is suitable for building forecasting system?
h
Han Wang
12/09/2022, 5:26 PM
Hi
@Valeriy
fugue is an abstraction on top of spark dask as ray, Nixtla uses Fugue to interact with these backends. We achieved amazing result on Spark, please read this article
https://towardsdatascience.com/distributed-forecast-of-1m-time-series-in-under-15-minutes-with-spark-nixtla-and-fugue-e9892da6fd5c
👍 1
and more importantly, it's easy to use. With Statsforecast interface, it is trivial to scale out on Spark, here is a code example
https://github.com/Nixtla/statsforecast/blob/8ea8c9e0d72b8e2b2edb4fb18a81ef5f713ec8f0/nbs/examples/Prophet_spark_m5.ipynb#L112
👍 1
v
Valeriy
12/10/2022, 11:05 AM
Hi
@Han Wang
that’s awesome thanks for sharing these links.
Post