Developing and Deploying ML Models with Java
summary
Demonstrates deploying and training real models end-to-end using TensorFlow for Java, DJL and a microservice stack on Micronaut. Shows a concrete example: trains ResNet15 on a shoe dataset (four labels: boots, sandals, shoes, slippers) with batch size 32, 2 epochs, 80/20 train/test split, uses softmax cross-entropy loss and reports accuracy/loss during training. Provides a Micronaut endpoint for inference, discusses model repositories, container/Kubernetes deployment and serverless options; live demo starts training but cuts off before completion.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 0:40Welcome
- 2:40What is ML
- 9:22ML frameworks
- 9:47ML Engineering
- 13:03Cloud Native Technology
- 16:45Architecture
- 18:51Demo
related talks
47mML in Java, YES it's possible!
46mML in Java, YES it's possible!
01h 05mNatively Cloud Native: Building Agile Microservices
51mIntegrate Machine Learning into Your Spring Application in Less than an Hour
02h 37mDeveloping Cloud-Native Java AI applications with DJL and LangChain4J
44mDeep Learning with DeepLearning4J and Spring Boot
34mBuilding Machine Learning-powered Applications with JS
42mDeploying ML at Scale with Kubernetes and TFX