AWS Documentation » Amazon SageMaker » Developer Guide » Get Started » Step 2: Train a Model with a Built-in Algorithm and Deploy It » Step 2.4: Deploy the Model to Amazon SageMaker » Step 2.4.2: Deploy the Model with Batch Transform

Step 2.4.2: Deploy the Model with Batch Transform

To to get inference for an entire dataset, use batch transform. You can use batch transform in the console or by using the API.

For information about batch transforms, see Get Inferences for an Entire Dataset with Batch Transform. For an example that uses batch transform, see the batch transform sample notebook.

Deploy a Model with Batch Transform (API)

To create a batch transform job, you can use the Amazon SageMaker high-level Python library or the low-level SDK for Python (Boto 3).

Deploy a Model with Batch Transform (Amazon SageMaker High-level Python Library)

The following code creates a transform job using the Amazon SageMaker high-level Python library:

import boto3
import sagemaker
import json

input_key = 'kmeans_batch_example/input/valid-data.csv'
input_location = 's3://{}/{}'.format(bucket, input_key)
output_location = 's3://{}/kmeans_batch_example/output'.format(bucket)

### Convert the validation set NumPy array to a CSV file and upload it to Amazon S3
numpy.savetxt('valid-data.csv', valid_set[0], delimiter=',', fmt='%g')
s3_client = boto3.client('s3')
s3_client.upload_file('valid-data.csv', bucket, input_key)

# Initialize the transformer object
transformer =sagemaker.transformer.Transformer(
    base_transform_job_name='Batch-Transform',
    model_name=model_name,
    instance_count=1,
    instance_type='ml.c4.xlarge',
    output_path=output_location
    )
# Start a transform job
transformer.transform(input_location, content_type='text/csv', split_type='Line')
# Then wait until the transform job has completed
transformer.wait()

# Fetch validation result 
s3_client.download_file(bucket, 'kmeans_batch_example/output/valid-data.csv.out', 'valid-result')
with open('valid-result') as f:
    results = f.readlines()   
print("Sample transform result: {}".format(results[0]))

        

Deploy a Model with Batch Transform (Low-level SDK for Python (Boto 3))

To use batch transform, you need to create a transform job.

With the Low-level SDK for Python (Boto 3), you use the create_transform_job method to create a transform job. With this method, you do the following:

1. Import the required libraries (boto3) and the following classes from the time library: gtime and strftime .

2. Set the required variables: job_name, bucket, prefix, sm, and model_name. The sm variable allows you to use the Amazon SageMaker API and is especially important.

3. Write the request. For information, see CreateTransformJob.

   To avoid generating errors and to improve readability, use the variables that you created in Step when creating the request.

4. Use the request with the create_transform_job method to start a batch transform job. The following code shows how to use the create_transform_job operation.

import boto3
import sagemaker
import json
from urllib.parse import urlparse
from time import gmtime, strftime

batch_job_name = 'Batch-Transform-' + strftime("%Y-%m-%d-%H-%M-%S", gmtime())
input_location = 's3://{}/kmeans_batch_example/input'.format(bucket)
output_location = 's3://{}/kmeans_batch_example/output'.format(bucket)

### Convert the validation set NumPy array to a CSV file and upload it to Amazon S3
numpy.savetxt('valid-data.csv', valid_set[0], delimiter=',', fmt='%g')
s3_client = boto3.client('s3')
input_key = "{}/valid_data.csv".format(urlparse(input_location).path.lstrip('/'))
s3_client.upload_file('valid-data.csv', bucket, input_key)

### Create a transform job
sm = boto3.client('sagemaker')

request = \
{
    "TransformJobName": batch_job_name,
    "ModelName": model_name,
    "MaxConcurrentTransforms": 4,
    "MaxPayloadInMB": 6,
    "BatchStrategy": "MultiRecord",
    "TransformOutput": {
        "S3OutputPath": output_location
    },
    "TransformInput": {
        "DataSource": {
            "S3DataSource": {
                "S3DataType": "S3Prefix",
                "S3Uri": input_location 
            }
        },
        "ContentType": "text/csv",
        "SplitType": "Line",
        "CompressionType": "None"
    },
    "TransformResources": {
            "InstanceType": "ml.m4.xlarge",
            "InstanceCount": 1
    }
}

sm.create_transform_job(**request)

print("Created Transform job with name: ", batch_job_name)

### Wait until the job finishes
while(True):
    response = sm.describe_transform_job(TransformJobName=batch_job_name)
    status = response['TransformJobStatus']
    if  status == 'Completed':
        print("Transform job ended with status: " + status)
        break
    if status == 'Failed':
        message = response['FailureReason']
        print('Transform failed with the following error: {}'.format(message))
        raise Exception('Transform job failed') 
    print("Transform job is still in status: " + status)    
    time.sleep(30)    

### Fetch the transform output
output_key = "{}/valid_data.csv.out".format(urlparse(output_location).path.lstrip('/'))
s3_client.download_file(bucket, output_key, 'valid-result')
with open('valid-result') as f:
    results = f.readlines()   
print("Sample transform result: {}".format(results[0]))

                        

View the Details of a Transform Job (Low-level SDK for Python (Boto 3))

To see the details of a specific transform job, specify the name in a DescribeTransformJjobs requestand the corresponding describe_transform_jobs method:

response = sm.describe_transform_job(TransformJobName=job_name)

List Transform Jobs (Low-level SDK for Python (Boto 3))

To list previous transform jobs, specify the listing and sorting criteria and use the list_transform_jobs method:

import boto3
from time import gmtime, strftime

request = \
{
   "StatusEquals": "Completed",
   "SortBy": "CreationTime",
   "SortOrder": "Descending",
   "MaxResults": 20,
}

response = sm.list_transform_jobs(**request)

Tip

When using the list_transform_jobs method, you can specify any of the members in the request structure. For more information, see ListTransformJobs.

Next Step

Step 2.5: Validate the Model