Author: Fritz

At WWDC18, Apple announced an exciting new addition to their machine learning frameworks, Create ML. Create ML gives developers an easy way to create custom, lightweight machine learning models within Xcode.

When combined with existing frameworks such as Core ML and Vision, iOS developers (even those without prior knowledge of machine learning) can create apps that use custom models they built themselves.

Continue reading “Making a “Pokédex” for iOS Using Create ML and Core ML with Vision”

While TensorFlow Lite is amazing, it has certain shortcomings that surface when you’re trying to run a model containing custom operations that aren’t supported by the framework.

Trying to load and run such a model often results in exceptions similar to the following :

As you can see, the model I was trying to load needed 3 operations (Normalize, ExtractFeatures, and Predict) that were missing from TensorFlow Lite’s default dependency.

Continue reading “Compiling a TensorFlow Lite Build with Custom Operations”

Many of your favorite mobile apps are currently using some form of neural networks and/or machine learning techniques to personalize your experience.

For example, Spotify and Apple Music, leaders in the music industry, use AI-powered recommendations for generating music suggestions.

Another shining example of the power of AI in apps is voice assistant tools such as Siri, Google Assistant, and Alexa. First, let’s define what mobile machine learning is and how it’s different than web-based ML.

Continue reading “5 Challenges for Developing Mobile Apps with AI & Machine Learning Capabilities”

The idea of being able to easily add powerful chatbots for improving user experience is appealing to React Native developers. A hybrid framework like React Native relies on tools that work cross-platform and reduce the mundane work of platform-dependent customizations and configurations.

Let’s talk about how you can bring the power of conversational AI to your React Native apps.

Continue reading “How to Make a Chatbot for your React Native App with Kommunicate”

In addition to good training data and the right model architecture, loss functions are one of the most important parts of training an accurate machine learning model.

For this post, I’d love to give developers an overview of some of the more advanced loss functions and how they can be used to improve the accuracy of models—or solve entirely new tasks.

For example, semantic segmentation models typically use a simple cross-categorical entropy loss function during training, but if we want to segment objects with many fine details like hair, adding a gradient loss function to the model can vastly improve results.

Continue reading “Research Guide: Advanced Loss Functions for Machine Learning Models”

While the concept of augmented reality (AR) has been around for decades, it has only recently become a popular technology with practical applications ranging from entertainment to marketing.

The tools for creating augmented reality experiences are constantly improving, and new platforms like Spark AR Studio are making AR more accessible to the social media generation.

Continue reading “How To Make And Share AR Effects”

Image classification is a cornerstone of computer vision, enabling machines to understand and label images accurately.

TensorFlow, a leading open-source framework, simplifies this process with powerful tools, pre-trained models, and APIs. Whether you’re a beginner or an expert, TensorFlow offers scalable solutions for tackling image classification challenges.

Continue reading “Image Classification with TensorFlow”