Feature Store

Tags
Dataiku DSS Core Designer, Dataiku DSS ML Practitioner, Dataiku DSS Adv Designer, Registered Posts: 6

Hi everyone,

I'm currently ussing Dataiku 11.3, and when reading about the Feature Store i find that it saves the datasets to share with others. So, my question is if the datasets or features are saved in the storage space that Dataiku has or does it work like a logic layer (only shows the feature but is stored in my database)?

Welcome!

It looks like you're new here. Sign in or register to get started.

Answers

  • Neuron 2020, Neuron, Registered, Dataiku Frontrunner Awards 2021 Finalist, Neuron 2021, Neuron 2022, Dataiku Frontrunner Awards 2021 Participant, Neuron 2023 Posts: 330 Neuron

    Hi @Vale_NCH
    ,

    Any type of dataset you have available within Dataiku can be a feature dataset. You could choose to store your feature datasets as a text file in a folder managed by Dataiku (e.g. using the filesystem_managed connection) or yes, you could store your feature datasets in a Dataiku dataset backed by a database. We do the latter and store our feature datasets in Snowflake.

    Marlan

  • Dataiku DSS Core Designer, Dataiku DSS ML Practitioner, Dataiku DSS Adv Designer, Registered Posts: 6

    Thank you @Marlan
    ! and how do i choose between the two options?

    Tanks for your answer.

  • Neuron 2020, Neuron, Registered, Dataiku Frontrunner Awards 2021 Finalist, Neuron 2021, Neuron 2022, Dataiku Frontrunner Awards 2021 Participant, Neuron 2023 Posts: 330 Neuron

    Hi @Vale_NCH
    , well, refreshing feature tables and joining feature tables to other tables will be much faster in a database than in text files stored on a server. This may not matter much if your tables are small. But for larger tables, a database is definitely the way to go. if you have the option to use datasets backed by a database, I'd definitely do that.

    Marlan

Welcome!

It looks like you're new here. Sign in or register to get started.

Welcome!

It looks like you're new here. Sign in or register to get started.