Serving Scikit-learn Models at Scale




[ad_1]

Scikit-learn is a great tool for building your models. When it comes time to deploy them to prediction, scale up using Google Cloud ML Engine. In this episode of AI Adventures, Yufeng shows you how to set up your own deployment pipeline with scikit-learn so you can go back to focusing on tuning your model!

Associated blog post → http://bit.ly/2Mg70Wi
Kaggle kernel → http://bit.ly/2v31tMh
Cloud ML Engine with scikit-learn quickstart → http://bit.ly/2Mdtszp

Watch more AI Adventures → http://bit.ly/AIAdventures
Subscribe to the Google Cloud Platform channel → http://bit.ly/GCloudPlatform

#AIAdventures

Source


[ad_2]

Comment List

  • Google Cloud Platform
    December 7, 2020

    thanks for the tut, but i have a question: where to put the input.json? i tried upload in the same bucket as model, error

  • Google Cloud Platform
    December 7, 2020

    Normally data scientists create and deploy models for users who are not familiar with gcloud command line interface. They do this in a web app environment, where users can go online by typing the web address and submit data files for prediction and get results. How can we do that using GCP?

  • Google Cloud Platform
    December 7, 2020

    DO NOT MAKE A VIDEO IN FUTURE IF YOU JUST SPEAK ON THE GENERAL STEPS WITHOUT MENTIONING "HOW TO…" PART OF THOSE STEPS. JUST A WASTAGE OF TIME….

  • Google Cloud Platform
    December 7, 2020

    Hi Yufeng,

    I have a very basic requirement on Sklearn Pandas ML pipeline but I am not able to find any clear answer to this problem any where on the documentations.

    Let's say I have an Sklearn pipeline built using Pandas dataframe as input, with pandas functions as transformations etc. So obviously this pipeline will expect a pandas – dataframe during prediction. Does GCMLE already have capability to convert the key value paired JSON payload to Panda-DF via online-prediction mode ?

    Seems to be a very basic requirement if we talk about ML on Cloud, any pointers ?

  • Google Cloud Platform
    December 7, 2020

    Yo

  • Google Cloud Platform
    December 7, 2020

    Thnaks Yufeng, helpful deployment steps.

Write a comment