Build, Train and Deploy ML Models with Keras on Google Cloud

Google Cloud via Coursera

Coursera based on 2,784 ratings

Likes

Overview

"Build, Train and Deploy ML Models with Keras on Google Cloud" is a hands-on, practical course designed to guide learners through the complete machine learning (ML) lifecycle using Keras and Google Cloud Platform (GCP). Ideal for beginners and intermediate learners, the course introduces foundational ML concepts while providing real-world implementation experience. Learners will start by understanding the basics of building neural networks with Keras, a user-friendly deep learning API. The course then moves on to training models efficiently using Google Cloud’s...

Syllabus

Introduction to the Course

This module provides an overview of the course and its objectives.

Introduction to the TensorFlow Ecosystem

This module introduces the TensorFlow framework and previews its main components as well as the overall API hierarchy.

Design and Build an Input Data Pipeline

Data is the a crucial component of a machine learning model. Collecting the right data is not enough. You also need to make sure you put the right processes in place to clean, analyze and transform the data, as needed, so that the model can take the most signal of it as possible. In this module we discuss training on large datasets with tf.data, working with in-memory files, and how to get the data ready for training. Then we discuss embeddings, and end with an overview of scaling data with tf.keras preprocessing layers.

Building Neural Networks with the TensorFlow and Keras API

In this module, we discuss activation functions and how they are needed to allow deep neural networks to capture nonlinearities of the data. We then provide an overview of Deep Neural Networks using the Keras Sequential and Functional APIs. Next we describe model subclassing, which offers greater flexibility in model building. The module ends with a lesson on regularization.