project.yaml Configuration¶

This document aims to explain the parameters used in the conf/project.yaml file and their effects.

Project Configuration¶

project_name¶

project_name: recommend-hq

• Description: The name of the project.
• Example: recommend-hq
• Constraints: Maximum length of 20 characters.

AWS Configuration¶

region_name¶

aws:
  region_name: ap-northeast-1

• Description: The AWS region name.
• Example: ap-northeast-1
• Constraints: Maximum length of 50 characters.

Personalize Configuration¶

personalize:
  item_schema_file_path: conf/schema/item_schema.json
  user_schema_file_path: conf/schema/user_schema.json
  interaction_schema_file_path: conf/schema/interaction_schema.json
  etl_configure_file_path: conf/etl/configure.json
  force_reprocess_lookback_days: 8
  log_lookback_days: 180

item_schema_file_path¶

• Description: The path to the item schema file.
• Example: conf/schema/item_schema.json

interaction_schema_file_path¶

• Description: The path to the interaction schema file.
• Example: conf/schema/interaction_schema.json

user_schema_file_path¶

• Description: The path to the user schema file.
• Example: conf/schema/user_schema.json

etl_configure_file_path¶

• Description: The path to the ETL configuration file.
• Example: conf/etl/configure.json

force_reprocess_lookback_days¶

• Description: The number of days to look back for force reprocessing.
• Example: 8

log_lookback_days¶

• Description: The number of days to look back for logs.
• Example: 180

Provisioned Concurrency Configuration (Optional)¶

Configure provisioned concurrency for APIs to ensure low-latency during high traffic by maintaining ready instances.

data_api¶

Configuration for data API.

data_api:
  min_capacity: 1
  max_capacity: 10
  utilization_target: 0.8

min_capacity¶

• Description: Minimum instances ready to handle requests.
• Example: 1
• Constraints: Min value 1.

max_capacity¶

• Description: Maximum instances during peak loads.
• Example: 10
• Constraints: Must be ≥ min_capacity.

utilization_target¶

• Description: Target utilization to optimize cost and performance.
• Example: 0.8
• Constraints: Value between 0.0 and 1.0.

recommentation_api¶

Configuration for recommentation API.

recommendation_api:
  min_capacity: 1
  max_capacity: 10
  utilization_target: 0.8

min_capacity¶

• Description: Minimum instances ready to handle requests.
• Example: 1
• Constraints: Min value 1.

max_capacity¶

• Description: Maximum instances during peak loads.
• Example: 10
• Constraints: Must be ≥ min_capacity.

utilization_target¶

• Description: Target utilization to optimize cost and performance.
• Example: 0.8
• Constraints: Value between 0.0 and 1.0.

Cronjob Configuration (Optional)¶

Configure the cronjob settings for training jobs. All scheduled events use the UTC+0 time zone and schedules are in AWS cron format. For more information, please refer to the AWS cron-based scheduling documentation.

update_dataset_group¶

training_schedule:
  update_dataset_group: "0 2 * * * *"

• Description: Schedule for updating the dataset group.
• Example: "0 2 * * * *" (Every Sunday at 02:00 UTC)

update_solution_version¶

training_schedule:
  update_solution_version:
    user_personalization: "0 3 * * * *"
    personalized_ranking: "0 4 * * * *"
    similar_items: "0 5 * * * *"

user_personalization¶

• Description: Schedule for user personalization recipe
• Example: "0 3 * * * *" (Every day at 03:00 UTC)

personalized_ranking¶

• Description: Schedule for personalized ranking recipe.
• Example: "0 4 * * * *" (Every day at 04:00 UTC)

similar_items¶

• Description: Schedule for similar items recipe.
• Example: "0 5 * * * *" (Every day at 05:00 UTC)