Machine Learning Engineering

Summary

Machine Learning is all the rage today. Most ML courses focus on building models. However, taking the ML models to production, involves quite a bit of extra work, as illustrated diagram below.

This course will teach Machine Learning Engineering - the process of productionizing, monitoring and managing ML models.

We will use a cloud environment (Google Cloud or Amazon Cloud or Microsoft Cloud) for our deployment.

Prerequisites

• Some knowledge in Machine Learning or Deep Learning is highly recommended

• Some basic knowledge of Python is highly recommended.
• Our labs utilize Python language. But Python is a very easy language to learn. So even you don’t have previous exposure to Python, you will be able to complete the labs.

Duration

4 Days/Lecture & Lab

Audience

This course is designed for Data Scientists, DevOps, and Data Engineers.

Topics

• ML Engineering overview
• Overview of the AI capabilities of the Cloud ::Platform of choice
• Storing large data in the cloud
• Processing large data in the cloud using distributed tools
• Training models at scale, using GPUs on the cloud
• Deploying models as webservices
• Logging and tracing of model runtime
• Model metrics
• Setting up alerts
• A/B testing different models
• Updating newer model versions