create_docker_3.py

# Create docker image
from azureml.core.image import Image, ContainerImage

image_config = ContainerImage.image_configuration(runtime= "python",
                                 execution_script=config['docker']['path_scoring_script'],
                                 conda_file=config['docker']['conda_env_file'],
                                 tags = {'area': "diabetes", 'type': "regression"},
                                 description = "test")

image = Image.create(name = config['docker']['docker_image_name'],
                     # this is the model object 
                     models = [model],
                     image_config = image_config, 
                     workspace = ws)

image.wait_for_creation(show_output = True)