Architecting Production-ready ML Models Using Google Cloud ML Engine

Building machine learning models with Python and a machine learning framework is the first step toward constructing an enterprise-grade ML architecture, but there are still two major challenges to overcome: training the model with enough computing power to produce the best possible model and then making that model available to users who aren't data scientists or even Python users. Architecting Production-Ready Machine Learning Models is the title of this course. You'll learn how to do on-cloud distributed training and hyperparameter tweaking using the Google Cloud ML Engine, as well as how to make your ML models available for use in prediction via simple HTTP queries.

First, you'll learn how to use the ML Engine with XGBoost models. XGBoost is a machine learning framework that uses the Ensemble Learning approach to create a single, strong model by merging numerous weak learners. Then you'll see how simple it is to migrate serialized models from on-premise to GCP. You'll create a basic model in scikit-learn, the most popular traditional machine learning framework, and then serialize and upload it to the GCP using the ML Engine. Finally, you'll learn how to use TensorFlow, one of the most popular libraries for deep learning applications, to harness the full potential of distributed training, hyperparameter tweaking, and prediction.

You'll learn how the TF CONFIG JSON environment variable is utilized to convey state information and improve the training and hyperparameter tuning process. After completing this course, you'll have the skills and understanding of the Google Cloud ML Engine required to reap the full benefits of distributed training and make batch and online prediction available to your client apps via simple HTTP queries.

Table of contents:

• Course Overview
• Introducing the Google Cloud ML Engine
• Deploying XGBoost Models to the Cloud ML Engine
• Deploying Scikit-learn Models to the Cloud ML Engine
• Deploying TensorFlow Models to the Cloud ML Engine

Level: Intermediate

Enroll here: https://www.pluralsight.com/courses/google-ml-engine-architecting-production-ready-models