MLOps using VertexAI: Beyond Model Training
Shadab Hussain (~shadab5) |
Description:
MLOps can be defined as a set of practices that help you deploy and manage machine learning models in production environments. Vertex Pipelines, which let you orchestrate and automate your ML workflows, are the backbone of the Vertex AI MLOps story. In this session, learn how to use Vertex Pipelines to build an end-to-end ML workflow using prebuilt components (pipeline steps) for Vertex AI services, visualize and compare pipeline run results, build your own custom components from Python functions, pass artifacts and parameters between pipeline steps, access and query metadata about your pipeline runs, and schedule recurring pipeline runs.
Prerequisites:
Understanding of Python, ML, and workflow.
Content URLs:
https://github.com/techwithshadab/
Speaker Info:
Shadab, a certified AWS Machine Learning Specialist and Google Cloud Profession Machine Learning professional, is a Senior Associate – MLOps at TheMathCompany. He formerly worked as a Developer Advocate (Data Science & Machine Learning) at the London Stock Exchange Group and has over five years of experience in ML, NLP, CV, Analytics, and building end-to-end scalable & reproducible ML pipelines on different clouds. He has published research papers at National and International conferences and is presently investigating use cases of Quantum Computing & Machine Learning in Finance and Healthcare. He is also authoring a book titled Financial Modeling using Quantum Computing, which will be published in July 2023. He is regularly invited as a speaker and attendee at conferences, user groups, and other events around the world. Apart from work, you will find him evangelizing about topics ranging from Quantum, Explainable AI, to MLOps.
Speaker Links:
https://medium.com/@techwithshadab https://des.analyticsindiamag.com/speaker/shadab-hussain/