Blog

I attended my first I/O this year, and it was a really good experience. While there, I Tweeted about some of the announcements in real-time. You can check them out on my twitter page.

But I wanted to dig a bit deeper and give more detail about news related to Android development that came out of I/O. There are other developments and Android-related announcements, but here I’ll only mention one connected specifically to Android development.

Continue reading Google I/O 2019: News and Announcements for Android Developers

Through the first 3 parts of our federated learning (FL) demo project, we’ve set up a system wherein machine learning (ML) models is trained using FL. Put simply, a generic model is created at the server. The model is then sent to the clients for training, and then sent back to the server.

Check out the previous 3 parts in the project to get caught up:

Continue reading Federated Learning Demo in Python (Part 4): Working with Mobile Devices

In my previous article that examined classification with TensorFlow, I covered the basics details of how to perform linear classification with TensorFlow’s estimator API. You can read that blog post here:

For part two, I’m going to cover how we can tackle classification with a dense neural network. I’ll be using the same dataset and the same amount of input columns to train the model, but instead of using TensorFlow’s LinearClassifier, I’ll instead be using DNNClassifier. We’ll also compare the two methods.

Continue reading Classification with TensorFlow and Dense Neural Networks

In this blog post, we’ll be learning how to automate the Chrome Dinosaur Game using neural networks with TensorFlow.js. If you haven’t played it before, it’s a side scrolling game available offline (for when Chrome or your Internet crashes) where you control a 2D dinosaur and have to jump and duck to avoid obstacles. Give it a shot here:

Continue reading Using TensorFlow.js to Automate the Chrome Dinosaur Game

When building machine learning models, we often tend to focus on the more glamorous aspects of machine learning, such as finding the best architecture and tuning hyperparameters, that we fail to recognize that our dataset is the lifeblood of our entire pipeline.

Continue reading Building a Data Pipeline with TensorFlow

Feature engineering is one of the most important skills needed in data science and machine learning. It has a major influence on the performance of machine learning models and even the quality of insights derived during exploratory data analysis (EDA).

In this article, we’re going to learn some important techniques and tools that will help you properly extract, prepare, and engineer features from your dataset.

Continue reading A Practical Guide to Feature Engineering in Python

The Open Neural Network Exchange (ONNX) is an open standard for representing machine learning models. With ONNX, AI developers can more easily move models between state-of-the-art tools and choose combinations that are best for them. ONNX is developed and supported by a community of partners including AWS, Facebook OpenSource, Microsoft, AMD, IBM, Intel AI, etc.

Continue reading Building an image recognition React app using ONNX.js

Deep neural networks perform incredibly well on computer vision tasks such as classification, object detection, and segmentation, but what do they consider before performing these tasks, and what does it take to make these decisions?

Interpreting neural network decision-making is an ongoing area of research, and it’s quite an important concept to understand. Neural networks are used in the real-world, so we can’t treat them like black boxes—we need to learn what they interpret, how they interpret, and what information each layer/channel in a neural network has learned.

Continue reading Class activation maps: Visualizing neural network decision-making

At the turn of this new decade, if there’s one field that has the potential to completely change our way of interacting with smartphones, it’s augmented reality (AR Glasses: are you reading this?). Google has been pushing ahead with its own platform—ARCore—that enables developers to create AR experiences on Android, iOS, Unity, and more.

Continue reading ARCore and Sceneform in Android — Performing Gestures and Collisions on Transformable Nodes