Author: Fritz

In a previous article, I focused on transfer learning scenarios with Core ML, and in particular we saw how to create a new model on an iOS device, import embedding weights from a previously-trained model, and train the rest of the layers on-device, using private and local data:

Moving forward in my long journey towards developing a Swift federated learning infrastructure, this time I’ve investigated how to train, from scratch on iOS devices, a little bit more complex model architecture—a CNN.

Continue reading “Fully Training a Core ML Convolutional Neural Network On-Device”

In my last post, I covered some of the difficulties of working with custom neural networks in Lens Studio. This time, I’ll show you what it looks like, with a step-by-step guide for building and implementing your own custom neural network in a Snap Lens.

Continue reading “Building a Custom Face Mask Snapchat Lens with Fritz AI and Lens Studio”

In this blog post, I’ll be covering the steps to create an Android app that’s able to detect and count objects from the Camera Preview in real time.

There are tons of real-time object detection tutorials out there, so you might be wondering, How is this one any different? One key difference is that, instead of just detecting objects, we’ll also be counting objects in a given frame, opening up a much wider range of possible applications.

Continue reading “Identifying and Counting Items in Real-Time with Fritz Object Detection for Android”

Reinforcement learning is a type of machine learning in which an agent learns to behave in an unknown environment by performing actions and seeing the ensuing results. The agent’s objective is to learn to act in ways that maximizes expected long-term rewards.

When it comes to reinforcement learning, there is no expected outcome—the agent makes the best decision based on its knowledge, for which it is either rewarded or penalized. As such, it is bound to learn from past experiences.

Continue reading “Introduction to Reinforcement Learning”

We’ve talked with more than 20 amazingly talented Snapchat Lens Creators over the last few months, and while they’ve been unanimously excited about the potential of creating Lenses with custom machine learning models, there remains an recurring caveat attached to the end of that sentiment—when I can get a grasp of the ML side of things.

Machine learning is tough—especially when you don’t have much in the way of knowledge or previous experience. And the same thing (so far) goes for working with ML in Lens Studio.

Continue reading “Machine Learning Primer for Snapchat Lens Creators”

In a normal machine learning problem, we get an observation for which we predict an outcome, irrespective of the time factor. In some of these cases, future outcomes are being predicted, but that treats all the past observations equally, with little or no significant difference.

However, a time series dataset is entirely different. Time series tasks add a “time dimension”, and also have an explicit order of dependence between the observations. To put it simply: A time series is a sequence of observations taken sequentially in time.

Continue reading “Time Series Forecasting in Machine Learning”