From the course: Serverless and Microservices for AWS

Unlock the full course today

Join today to access over 22,600 courses taught by industry experts or purchase this course individually.

Serverless machine learning

Serverless machine learning - Amazon Web Services (AWS) Tutorial

From the course: Serverless and Microservices for AWS

Start my 1-month free trial

Serverless machine learning

“

- We've looked at event driven and message based coordination of the serverless components that make up a serverless application. In some cases, more complex coordination of Lambda functions is required. AWS step functions are designed to address this need. In our public transit inside dashboard, we have a feature that characterizes delays. Are they train related or track related, and are they likely to create knock on delays or be repeated by other trains on subsequent trips. Let's focus on the orchestration of the training process. Model training requires a training data set to be available and the training itself takes time, may even be measured in hours. Retraining the model is required periodically. We're generating data constantly and patterns change in transit over time. So, for example, let's assume we retrain the model once a day. There are a few steps that are needed. First, is there new training data available?…

Contents