Configuration file

Configuration file

An example configuration file is available at config.example.json

The composition of the file must be structured as follows:

• qa_engine: object containing all the engine configurations. Formed by:
  • models_available: list of algorithms and models for which system support exists. Every object must be composed as follows:
    • from: name of the general python module which makes accessible the various QA-Integrator-Model. Normally qa_integrator.models
    • name: name of the python module just created in the from module which implements the QA-Integrator-Model.
    • api_name: name to be used outside the API to use the corresponding QA-Model.
  • predictions_root_dir: a directory relative to the project root that will contain all the support file of prediction requests.
  • trainings_root_dir: a directory relative to the project root that will contain all the files to support training requests.
  • predictions_thread_pool_executor_max_workers: the maximum number of Workers to use for the ThreadPool which performs the various operations.
• utils_auto_delete_cronjob: object containing all the configurations concerning the Deletion Cron Job. Formed by:
  • cronjob_name: name of the cronjob to be created through the operating system.
  • job_every_n_minutes: execution time of the cronjob, expressed in minutes.
  • delete_older_than_minutes_predictions: minimum time of existence of the prediction requests after they have been completed, expressed in minutes.
  • delete_older_than_minutes_trainings: minimum time of existence of training requests after they have been completed, expressed in minutes.
  • log_file_name: name of the log file in which the operations performed by the cronjob are saved.