Using Deeplearning4J in Android Applications
This tutorial demonstrates how the Java-based deep learning library Deeplearning4J can be used in Android Applications to build, train, and use a neural network. This application trains a simple neural network on the mobile device using a iris data set for iris flower type classification. The application has a simple UI to take measurements of petal length, petal width, sepal length, and sepal width from the user and it returns the probability that the measured iris belongs to one of three types (Iris serosa, Iris versicolor, and Iris virginica). A data set of 150 measurements is used to train the model, which takes anywhere from 5-30 seconds, depending on the device. A tutorial for the application can be found here.
For more information on Deeplearning4J, please visit http://deeplearning4j.org/