Skip to content

Latest commit

 

History

History
39 lines (24 loc) · 1.11 KB

readme.MD

File metadata and controls

39 lines (24 loc) · 1.11 KB
Raw

This is a GCP repo for MLOps pipelines using feature store

It's needed a GCP project on the Google Cloud Platform. Login into: https://console.cloud.google.com/home/

These MLOps pipelines are using Kubeflow and Feature Store

The data used for this project is public and here is the link: https://www.kaggle.com/prajitdatta/movielens-100k-dataset

GCP services to be enabled/used:

  1. Google Vertex AI
  2. Notebooks API
  3. Google Cloud Storage
  4. Google Deployment Manager API

Steps:

Note: Important to configure/install Google SDK previously and deploy all resources on same region (for example: europe-west4 - Netherlands which has most of the services available)

  1. Deploy workbench Vertex AI with following configuration (including 1 GPU)

alt text

  1. Once deployed workbench machine then open jupyterlab environment
  2. Open terminal and pull code from github
  3. Open iPython notebook locally on workbench environment
  4. Run each cell as it's written code on notebook

Authors:

  • Lucho Farje
  • Pawan Poojary

CapGemini Insights & Data :sparkles: :rocket: :octocat: