run_model_training_job.py

import sagemaker as sage

# this line of code requires iam:GetRole permissions
role = sage.get_execution_role()

# initialize a SageMaker session and use it to get account and region info
# you can also do this using boto3 if you so prefer
# use that to get your home ECR registry, and from there your training image
sess = sage.Session()
account = sess.boto_session.client('sts').get_caller_identity()['Account']
region = sess.boto_session.region_name
image = '{}.dkr.ecr.{}.amazonaws.com/quiltdata/sagemaker-demo'.format(account, region)

# create an estimator and fit it
# replace "s3://quilt-example" in `output_path` with a path into bucket you have access to
clf = sage.estimator.Estimator(image,
                               role, 1, 'ml.c4.2xlarge',
                               output_path="s3://quilt-example/quilt/quilt_sagemaker_demo/model",
                               sagemaker_session=sess)
clf.fit()