Evan Hennis

Evan Hennis is a Google Developer Expert in machine learning and an international speaker. He has an undergraduate degree in computer science from Iowa State University and a master's degree in computer science with a specialization in machine learning from Georgia Tech. He has spent over sixteen years in software development, working across multiple languages and domains.

projects by Evan Hennis

Deep Learning for Basketball Scores Prediction

3 weeks · 4-6 hours per week average · INTERMEDIATE

In this liveProject series, you’ll step into the role of a data scientist trying to predict the results of NCAA college basketball games. Your client’s favorite team is competing, and he wants to know how close their games will be. Each liveProject in this series covers a different aspect of the machine learning pipeline from creating the initial model to deploying it to the web and Android for your client’s easy use.

Deploy a Predictor on Android

1 week · 4-6 hours per week · INTERMEDIATE

In this liveProject, you’ll create an Android application that can run a pretrained basketball predictor deep learning model for the easy use of your client. Your challenges will include converting the DataFrames into JavaScript arrays, converting your model into a TensorFlow Lite model, and finally packaging the model inside a working Android application.

Deploy a Predictor on the Web

1 week · 4-6 hours per week · INTERMEDIATE

In this liveProject, you’ll deploy a pretrained basketball predictor deep learning model onto the web for easy use by clients. You’ll utilize the powerful TensorFlow.js framework to ensure the model works in the browser, as well as converting DataFrames into JavaScript arrays, and building a simple website around the model and data sets.

Create a Neural Network

1 week · 4-6 hours per week · INTERMEDIATE

In this liveProject, you’ll use Keras to create a deep learning model for predicting basketball scores. Once your model is created, you’ll train it up on sample data and then validate your results to ensure it’s still accurate when applied to data from the real world.