wag-advanced-analytics-utils 0.0.3

Package containing the common functions used in all the Advanced Analytics algorithms

README

This package contains all the common functions used to run the Advanced Analytics algorithms. We provide a brief explanation of each of the modules and the corresponding functions:

• athena_data_provider contains the class AthenaDataProvider, which is initialized by giving the parameters

  • aws_access_key_id
  • aws_secret_access_key
  • s3_staging_dir
  • region_name.

  The function read_query returns a DataFrame with the information requested by the query passed as argument.

• logger contains the class Logger, which logs important messages and prints them to the terminal or CloudWatch in AWS.

• metrics contains the class Metrics, which creates the metrics and pushes them to Datadog.

• s3_client contains the class S3Client, that is initialized using boto3.client and boto3.resource and the environment variable RESULT_BUCKET.

  The function upload_files uploads the DataFrame returned by the algorithm (output_df) to the given path (output_path) in AWS, for the chosen variant (variant – typically production or the name of the experiment, if testing new features).

---

If changes are made to this package, it has to be updated in PyPI by doing:

1. update the version number in the file pyproject.toml under the field version.

---

To use these modules in a Python script do the following:

• start by installing the package by running

  pip install wag_advanced_analytics_utils
  

  or

  pipenv install wag_advanced_analytics_utils
  

• import the desired class by typing

  from wag_advanced_analytics_utils.{module_name} import class_name
  

  where {module_name} is one of the modules of the package and class_name is (one of) the class contained in the module.