SageMaker Image Classification inference

Make sure you have trained your model as outlined here

Using the Console

On the training job page, click create model. Give the model a name, leave all other options as default and create model Create model

On the next page, click create endpoint, add options as needed on the next page. Create endpoint

Using the boto3 python SDK

Create Model

import boto3
from time import gmtime, strftime

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

model_name="DEMO--classification-model" + time.strftime('-%Y-%m-%d-%H-%M-%S', time.gmtime())
print(model_name)
info = sage.describe_training_job(TrainingJobName=job_name)
model_data = info['ModelArtifacts']['S3ModelArtifacts']
print(model_data)

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

primary_container = {
    'Image': hosting_image,
    'ModelDataUrl': model_data,
}

create_model_response = sage.create_model(
    ModelName = model_name,
    ExecutionRoleArn = role,
    PrimaryContainer = primary_container)

print(create_model_response['ModelArn'])

Create Endpoint configuration

from time import gmtime, strftime

timestamp = time.strftime('-%Y-%m-%d-%H-%M-%S', time.gmtime())
endpoint_config_name = 'DEMO-imageclassification' + '-epc-' + timestamp
endpoint_config_response = sage.create_endpoint_config(
    EndpointConfigName = endpoint_config_name,
    ProductionVariants=[{
        'InstanceType':'ml.m4.xlarge',
        'InitialInstanceCount':1,
        'ModelName':model_name,
        'VariantName':'AllTraffic'}])

print('Endpoint configuration name: {}'.format(endpoint_config_name))
print('Endpoint configuration arn:  {}'.format(endpoint_config_response['EndpointConfigArn']))

Create Endpoint

# get the status of the endpoint
response = sagemaker.describe_endpoint(EndpointName=endpoint_name)
status = response['EndpointStatus']
print('EndpointStatus = {}'.format(status))


# wait until the status has changed
sagemaker.get_waiter('endpoint_in_service').wait(EndpointName=endpoint_name)


# print the status of the endpoint
endpoint_response = sagemaker.describe_endpoint(EndpointName=endpoint_name)
status = endpoint_response['EndpointStatus']
print('Endpoint creation ended with EndpointStatus = {}'.format(status))

if status != 'InService':
    raise Exception('Endpoint creation failed.')

Perform inference

import boto3
import json
import numpy as np

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

with open(file_name, 'rb') as f:
    payload = f.read()
    payload = bytearray(payload)
response = runtime.invoke_endpoint(EndpointName=endpoint_name,
                                   ContentType='application/x-image',
                                   Body=payload)
result = response['Body'].read()
# result will be in json format and convert it to ndarray
result = json.loads(result)

Updated on 07 Feb 2020