If you’ve tried deploying your trained deep learning models on Android, you must have heard about TensorFlow Lite, the lite version of TensorFlow built for mobile deployment.
As a quick overview, it supports most of the basic operators; in simple terms, you can use it to do classification, object detection, semantic segmentation, and most of the text synthesis operations.
Continue reading “Profiling TensorFlow Lite models for Android”
In this article, let’s look at how you can use TensorFlow Model Maker to create a custom text classification model. Currently, the TF Lite model maker supports image classification, question answering, and text classification models. It uses transfer learning for shortening the amount of time required to build TF Lite models.
Continue reading “TensorFlow Lite Text Classification Models with Model Maker”
Computer vision continues to be one of the fastest-growing fields in machine learning. Both web and mobile developers are harnessing the power of cameras as sensors and the hardware behind them to create unique and impactful user experiences.
Materials can seem intimidating to get the hang of — so many interconnected nodes, math operations, and procedural textures. I know it seemed like magic to me when I first started, but beneath this complexity are some very simple principles, and understanding these will enable you to create your own relatively complex materials in a short space of time.
Continue reading “How to Make a Space Wormhole in Lens Studio Using the Material Editor”
I’ve been working on a personal project recently, in which I need to know the main topic of a user’s generated text. So I made a tag system, which means that the user can tag his post based on the topic.
I thought that this would suffice to classify the posts based on topics, but I was wrong.
Continue reading “Easy Topic Classifier on iOS with Apple’s Natural Language Framework”
This blog is the sixth blog in the series and a follow-up to my previous blog post on running TensorFlow Lite image classification models in Python. If you haven’t read that post, you can read it here:
This blog post assumes that you already have a trained TFLite model on hand. If you don’t or need to build one, you can take a look at the blog post that I have written here:
Continue reading Running TensorFlow Lite Object Detection Models in Python
Following up from my earlier blogs on training and using TensorFlow models on the edge in Python, in this seventh blog in the series; I wanted to cover a topic that’s generally not talked about enough—optimizing the performance and latency of your TensorFlow models.
Continue reading “Optimizing the performance of TensorFlow models for the edge”
In my previous blog on using Room with Android, I outlined how one can use Room to effectively store app data after a user has exited your app for a better user experience. If you haven’t read this blog, you can read it here:
But what about cases where your app is running and its state is changed—for example, when the phone’s orientation is changed from portrait to landscape and vice versa?
Continue reading “Using ViewModels in Android to preserve your running app’s state”
Audio source separation is a wonderful use case where every one of us can easily relate with and in this article we are going to deep dive into technical implementations of the solution where we will build an Android app with the TensorFlow model performing audio source separation.
Continue reading “Audio Source Separation on Android with TensorFlow 2.0”
