Hands-On Artificial Intelligence for Beginners

Training on GCP 

Google has made the entire deep learning training / deployment process streamlined and simple by allowing us to train models, store them, and deploy them with minimal code. Before we start training in the cloud, let's train our model locally to ensure that everything is working as intended. First, we need to set some environment variables. First and foremost, we'll have to put our files in a particular structure to train with Cloud ML. Look for the training folder in the chapter GitHub folder, and you will find the correct file structure. The __init__.py file that you see there will simply tell GCP that our file is an executable Python program. 

First we'll define a job directory, which should be the folder where your simple_classifier.py ...