Classifying images using Amazon SageMaker

The SageMaker models that you have trained are now available to be used to predict objects in images. As we discussed at the beginning of the chapter, SageMaker offers a marketplace where you can use many models directly to perform your tasks:

1. Since we trained our own machine learning model, we will have to create a SageMaker model that can be used for predictions. The following code shows you how to generate a usable model in Amazon Sagemaker:

import boto3
from time import gmtime, strftime

sage = boto3.Session().client(service_name='sagemaker') 

model_name="example-full-image-classification-model"

info = sage.describe_training_job(TrainingJobName=job_name)
model_data = info['ModelArtifacts']['S3ModelArtifacts']

hosting_image = get_image_uri(boto3.Session().region_name, 'image-classification...