Getting started with a Domain dataset group (SDK for Python (Boto3))

Use the following code to create a schema in Amazon Personalize. Replace gs-domain-interactions-schema with a name for the schema.

import json
schema = {
  "type": "record",
  "name": "Interactions",
  "namespace": "com.amazonaws.personalize.schema",
  "fields": [
      {
          "name": "USER_ID",
          "type": "string"
      },
      {
          "name": "ITEM_ID",
          "type": "string"
      },
      {
          "name": "EVENT_TYPE",
          "type": "string"
      },
      {
          "name": "TIMESTAMP",
          "type": "long"
      }
  ],
  "version": "1.0"
}

create_interactions_schema_response = personalize.create_schema(
    name='gs-domain-interactions-schema',
    schema=json.dumps(schema),
    domain='VIDEO_ON_DEMAND'
)

interactions_schema_arn = create_interactions_schema_response['schemaArn']
print(json.dumps(create_interactions_schema_response, indent=2))

Create a dataset group with the following code. Replace dataset group name with a name for the dataset group.

response = personalize.create_dataset_group(
  name = 'dataset group name',
  domain = 'VIDEO_ON_DEMAND'
)
dsg_arn = response['datasetGroupArn']

description = personalize.describe_dataset_group(datasetGroupArn = dsg_arn)['datasetGroup']

print('Name: ' + description['name'])
print('ARN: ' + description['datasetGroupArn'])
print('Status: ' + description['status'])

Create an Item interactions dataset in your new dataset group with the following code. Give the dataset a name and provide the schema_arn and dataset_group_arn from the previous steps.

response = personalize.create_dataset(
    name = 'interactions-dataset-name',
    schemaArn = interactions_schema_arn,
    datasetGroupArn = dsg_arn,
    datasetType = 'INTERACTIONS'
)

dataset_arn = response['datasetArn']

Import your data with a dataset import job with the following code. The code uses the describe_dataset_import_job method to track the status of the job.

Pass the following as parameters: a name for the job, the dataset_arn from the previous step, the Amazon S3 bucket path (s3://bucket name/folder name/ratings.csv) where you stored the training data, and your IAM service role's ARN. You created this role as part of the Getting started prerequisites. Amazon Personalize needs permission to access the bucket. For information on granting access, see Giving Amazon Personalize access to Amazon S3 resources.

import time
response = personalize.create_dataset_import_job(
    jobName = 'JobName',
    datasetArn = 'dataset_arn',
    dataSource = {'dataLocation':'s3://amzn-s3-demo-bucket/filename.csv'},
    roleArn = 'role_arn'
)

dataset_interactions_import_job_arn = response['datasetImportJobArn']

description = personalize.describe_dataset_import_job(
    datasetImportJobArn = dataset_interactions_import_job_arn)['datasetImportJob']

print('Name: ' + description['jobName'])
print('ARN: ' + description['datasetImportJobArn'])
print('Status: ' + description['status'])

max_time = time.time() + 3*60*60 # 3 hours
while time.time() < max_time:
    describe_dataset_import_job_response = personalize.describe_dataset_import_job(
        datasetImportJobArn = dataset_interactions_import_job_arn
    )
    status = describe_dataset_import_job_response["datasetImportJob"]['status']
    print("Interactions DatasetImportJob: {}".format(status))
    
    if status == "ACTIVE" or status == "CREATE FAILED":
        break
        
    time.sleep(60)