Apache Deep Learning 302
summary
Shows working pipelines that push data from Apache NiFi into Apache Pulsar and Apache Flink while calling models via MXNet Multi-Model Server and DJL; demos include PDF-to-text with Apache Tika, entity extraction with OpenNLP, sentiment and BERT QA, and image classification using ResNet-50 on NVIDIA Xavier/Jetson. Live code runs use NiFi processors to fetch, split and validate RSS/XML, call model servers, stream results to Pulsar, and post alerts to Slack; source code and slides are on GitHub.
AI-written summary · about
chapters (9)
- 0:00Introduction to Apache Deep Learning 302
- 1:57Target audience and tools overview
- 8:13Deep learning frameworks and model servers
- 18:26NiFi for data pipeline orchestration
- 25:27Deep learning demo with camera images
- 34:37Processing results and metadata
- 42:25Security and NiFi REST API
- 45:56Running deep learning in Spark
- 52:17Future talks and closing
related talks
41mAdding Generative AI to Real Time Streaming Pipelines
49mData pipelines: building an efficient instrument to make custom workflow
39mOperating Deep Learning Pipelines Anywhere Using Kubeflow
52mMachine Learning Data Pipelines
45mReal-time, Streaming Advanced Analytics....
43mSentiment Analysis in Action: Building Your Real-time Pipeline
27mBuild ML Enhanced Event Streaming Apps with Java Microservices
33mDeploying and Scaling Spark ML and Tensorflow AI Models