Articles Fritz has written:

Enterprise Scale ML Jumpstart Kit — FastAI + RabbitMQ + Docker

Articles

As most ML practitioners realize, developing a predictive model in Jupyter Notebook and making the predictions with excel data may not help you build the predictive models required at enterprise scale. To build the model at such a scale, you will need to consider several requirements and use various tools/frameworks that are especially designed to meet the purpose of this expansion.

Continue reading Enterprise Scale ML Jumpstart Kit — FastAI + RabbitMQ + Docker

Data Pre-processing and Visualization for Machine Learning Models

Articles

The objective of data science projects is to make sense of data to people who are only interested in the insights of that data. There are multiple steps a Data Scientist/Machine Learning Engineer follows to provide these desired results. Data pre-processing (Cleaning, Formatting, Scaling, and Normalization) and data visualization through different plots are two very important steps that help in building machine learning models more accurately.

Continue reading Data Pre-processing and Visualization for Machine Learning Models

Deep Learning for Image Segmentation: U-Net Architecture

Articles

Basically, segmentation is a process that partitions an image into regions. It is an image processing approach that allows us to separate objects and textures in images. Segmentation is especially preferred in applications such as remote sensing or tumor detection in biomedicine.

There are many traditional ways of doing this. For example; point, line, and edge detection methods, thresholding, region-based, pixel-based clustering, morphological approaches, etc. Various methods have been developed for segmentation with convolutional neural networks (a common deep learning architecture), which have become indispensable in tackling more advanced challenges with image segmentation. In this post, we’ll take a closer look at one such architecture: u-net.

Continue reading Deep Learning for Image Segmentation: U-Net Architecture

Creating a TensorFlow Lite Object Detection Model using Google Cloud AutoML

Articles

Following up on my last blog post on training an image labeling model using Google Cloud AutoML (linked below), in this second blog post in the series; we’ll look into how to train yet another model to identify and locate objects within an image instead—an object detection model!

If you haven’t read my blog on image labeling, you can read it here:

Continue reading Creating a TensorFlow Lite Object Detection Model using Google Cloud AutoML

Creating a Style Transfer Snapchat Lens with Fritz AI and SnapML in Lens Studio

Articles

In 2015, Snapchat, the incredibly popular social content platform, added Lenses to their mobile app —augmented reality (AR) filters that give you big strange teeth, turn your face into an alpaca, or trigger digital brand-based experiences.

In addition to AR, the other core underlying technology in Lenses is mobile machine learning — neural networks running on-device that do things like create a precise map of your face or separate an image/video’s background from its foreground.

Continue reading Creating a Style Transfer Snapchat Lens with Fritz AI and SnapML in Lens Studio

Create and Publish Your First Instagram AR Filter Using Spark AR Studio

Articles

AR (augmented reality) is the next-gen thing. It provides an interactive experience of a real-world environment where the objects that reside in the real world are enhanced by computer-generated perceptual information, sometimes across multiple sensory modalities including visual, auditory, and haptic.

Facebook’s Spark AR Studio is a tool for creators to create, publish, and manage AR Apps. It is an open-source tool by Facebook and anyone with a Facebook account can easily access this studio. Nowadays, it is mostly used for creating AR filters for Facebook and Instagram.

Continue reading Create and Publish Your First Instagram AR Filter Using Spark AR Studio

Converting TensorFlow / Keras models built in Python to JavaScript

Articles

Python remains the most popular language for building and training machine/deep learning models. This is because of the numerous libraries and tools built around it, that enables developers and researchers to quickly build models.

But in terms of deployment of these models created in Python, there is a trend towards using a different language. Some of the reasons behind this are:

Continue reading Converting TensorFlow / Keras models built in Python to JavaScript

Building an iOS camera calculator with Core ML’s Vision and Tesseract OCR

Articles

Math might be scary, but it’s an essential part of everyday life. Wouldn’t it be cool if we could build an app, point our phone’s camera at an expression, and let the app compute the result? Whenever I’ve needed to use math, I’ve wished this was possible. Now, with advances in machine learning and vision recognition in iOS, this is doable.

Continue reading Building an iOS camera calculator with Core ML’s Vision and Tesseract OCR