Deep Java Library examples
The repository contains the source code of the examples for Deep Java Library (DJL) - an framework-agnostic Java API for deep learning.
Inference examples
Malicious URL Detector
An example application detects malicious urls based on a trained Character Level CNN model.
Pneumonia Detection with TensorFlow engine
An example application detects Pneumonia based on X-ray images using a trained Keras model.
Live Object Detection
An example application detects live objects from web camera.
Online DoodleDraw game powered by DJL
A web based DoodleDraw game built with DJL.
Run deep learning in browser
A web application that runs DJL code in browser.
Training examples
Train footwear classification with DJL
An example application trains footwear classification model using DJL.
Visualizing Training with DJL
An example application features a web UI to track and visualize metrics such as loss and accuracy.
Android
Doodle draw app with PyTorch mode
A Doodle draw android game that is built with PyTorch model.
Build Android app with MXNet engine
A template that can be used to build Android applications with MXNet engine.
Face detection made easy with DJL
An example that shows how to build deep learning android app with ease.
AWS services
AWS Kinesis Video Streams
An example application that reads the output of a KVS Stream.
Serverless Model Serving with AWS Lambda
An example application that serves deep learning model with AWS Lambda.
Model Serving on AWS Elastic Beanstalk
Build a micro service to deploy on AWS Elastic Beanstalk.
Model Serving on AWS Elastic Beanstalk
Build a micro service to deploy on AWS Elastic Beanstalk.
Low cost inference with AWS Inferentia
An example application that runs low cost/high performance inference with AWS Inferentia.
Big data integration
Spark Image Classification
Contains Spark image classification demos.
Apache Beam CTR Prediction
An example application using Apache Beam to predict the click-through rate for online advertisements.
Apache Flink Sentiment Analysis
An example using Apache Flink to run sentiment analysis.
DJL Component in Apache Camel
An example application that demonstrates simple HTTP-service to classify images using Zoo Model.
Other demos
Use multiple engines in single JVM
An example application that runs multiple deep learning frameworks in one Java Process.
Use graalvm to speed up your deep learning application
An example application that demonstrates compile DJL apps into native executables.
Deploy DJL models on guarkus
An example application that serves deep learning models using quarkus.