Author: Fritz

As an ML engineer or data scientist, sometimes you inevitably find yourself in a situation where you have hundreds of records for one class label and thousands of records for another class label.

Upon training your model you obtain an accuracy above 90%. You then realize that the model is predicting everything as if it’s in the class with the majority of records.

Excellent examples of this are fraud detection problems and churn prediction problems, where the majority of the records are in the negative class. What do you do in such a scenario? That will be the focus of this post.

Continue reading “Dealing with Imbalanced Data in Machine Learning”

There was no shortage of excellent computer vision research presented during the recently-concluded 2020 CVPR conference. In this article, we’ll take a focused look at a couple of those that touched or have implications for mobile or edge-related tasks.

While not all of these papers connect directly to mobile-first applications, their implications for mobile ML are significant. They push forward ML tasks commonly performed on mobile and edge devices, so their advancement is crucial in pushing the industry forward.

Continue reading “CVPR 2020: Research with Mobile ML Implications”

Oftentimes, sci-fi movie storylines start with a robot that realizes its role, and the roles of other similar robots in a colony.

The plot then progresses as a form of collaboration between these robots, as they deviate from the human-designed objectives, ignoring human safety to some evil control conspiracies.

Continue reading “Developing AI consciousness in connected autonomous vehicles (CAVs)”

Welcome back to my series on building image effects for Android using OpenCV. In this tutorial, which is part 5 of the series, we’re going to build an image pattern—filling a large image with a pattern created by repeating a small image.

The GitHub project for the series is available on this page. The project of this tutorial is available inside the Part 5 folder under the project.

Continue reading “Image Effects for Android using OpenCV: Image Patterns”

Federated learning is a new way of training a machine learning using distributed data that is not centralized in a server. It works by training a generic (shared) model with a given user’s private data, without having direct access to such data.

For a deeper dive into how this works, I’d encourage you to check out my previous blog post, which provides a high-level overview, as well as an in depth look at Google’s research.

Continue reading “Breaking Privacy in Federated Learning”

If you’ve ever considered working with the new Core ML framework in iOS, you might have seen a demo video like this:

Continue reading “Core ML Simplified with Lumina”

For a major launch such as this, Apple normally holds a press conference in their Cupertino headquarters. However, due to growing health concerns and widespread retail shutdowns, the company opted instead for an online-only launch for their latest device.

Sporting a powerful 8-core processor and LiDAR scanner, the iPad Pro looks more like a computer replacement than ever before.

Continue reading “The New iPad Pro’s LiDAR Sensor Changes the Augmented Reality Game”

As we’ve been working through this series introducing SwiftUI, we’ve so far made an app that allows us to view our team from a list and dive into specific members’ detailed profiles.

However, all our users can do with the app is view dynamic data in a very static UI. The only interaction we’ve provided is the navigation into a team member’s detailed profile.

Continue reading SwiftUI: Interactivity, State Bindings, & Simple Animations