PAPIs Latam 2019

For real-world ML systems, it is crucial to have scalable and flexible platforms to build ML workflows. In this workshop, we will demonstrate how to build an ML DevOps pipeline using Kubeflow and TensorFlow Extended (TFX). Kubeflow is a flexible environment to implement ML workflows on top of Kubernetes - an open-source platform for managing containerized workloads and services, which can be deployed either on-premises or on a Cloud platform. TFX has a special integration with Kubeflow and provides tools for data pre-processing, model training, evaluation, deployment, and monitoring. In our workshop, we will demonstrate a pipeline for training and deploying an RNN-based Recommender System model using Kubeflow.