Machine Learning Engineering – Speeding up training using GPU & TPU (FREE)
Overview
This is part of Machine Learning Engineering and DevOps Learning Series.
In this series we will demonstrate a neural network to process images (CIFAR). Using CPUs the training takes a long time to run (~10 mins). We will show how how switching to GPU cuts down the training time to just few seconds (under a min).
Also we will show how to utilize TPU system
This is a FREE class!
On-Demand
Can’t make it to live session? No worries. Go ahead and register; we will send you the session recording .
See below for past session recordings & notes
What you will learn
- Sample TF2 program to process CIFAR image dataset
- Using free GPU/TPU on Google Colab
- Speeding up training by using GPU
- How to use TPU
Intended Audience
Developers, DevOps, Data Scientists
Prerequisites
- Must have : Development experience
- Nice to have: Python knowledge
What to Bring
- Please bring a reasonably modern laptop (Corporate laptops with overly restrictive firewalls may not work well; Personal laptops are recommended)
- Need to have a Machine Learning Environment setup on your laptop. Please follow this guide
- [nice to have] download our docker image elephantscale/es-training
Session Recording
Session Notes
Presenters
Mike Kane
Mike Kane is a Senior Data Scientist specializing in Natural Language Processing. He has a passion for Deep Learning, and has professionally trained teams from multiple Fortune 100 companies in ML and AI
Linkedin : https://www.linkedin.com/in/mikekane2/
Github : https://github.com/mike-kane