Machine Learning with Apache Kafka without another Data Lake

preview_player
Показать описание
Streaming Analytics for IoT Data with Apache Kafka and Machine Learning - without the need for another data lake like Hadoop or Spark.

Tiered Storage in Kafka and other capabilities establish a simplified big data architecture without the need for another data lake. Implement new consumer applications, error-handling, compliance / regulatory processing, query and analyze existing events, model training with an event streaming platform as the central nervous system.

More resources:

Рекомендации по теме
Комментарии
Автор

Hi i just noticed you have just a few views... supreme presentation!

machinelearning
Автор

Hi Kai, good overview. One note: nowadays models are not just tensors (matrix of coefficients) or object code. They often come as Python code (as data-scientists use it). My question: what architectures / tools you recommend for production Kafka-based pipelines processing for real-time predictions, e.g. when it is required to execute multiple models (concurrently and/or sequentially) and make data transformations tasks among these steps? Thank you.

nikolayilyin