dart-trec-learner 3.1.0

T-REC learner libs

TREC Learner

For production config, set ENV to production

How to Run

1. Install requeriments

   • python3 -m venv .venv
   • source ./.venv/bin/activate
   • pip install -r requirements.txt

2. Config parameters

   • cp .env.example .env
   • cp pytest.ini.example pytest.ini
   • fill .env and pytest.ini

3. Run learner

   • ./run.sh

How to Setup

All application variables are exposed in src/config.py. Container variables are exposed in .env:

• Learner Options:

  Option	Description
  LEARNER_DAYS_INTERVAL	Learner interval in days
  LEARNER_ROOT_MODEL_DIR	Temporary root dir used during trainning
  LEARNER_MODEL_DIR	Temporary dir used store model files

• Connection Options:

  Option	Description
  MONGODB_URI	Mongo URI (without database name)
  MONGODB_DATABASE	Mongo database with PLM data
  MONGODB_COLLECTION	Mongo collection with issue-document data
  MINIO_HOST	MinIO host (without http://)
  MINIO_ACCESS_KEY	MinIO Key
  MINIO_SECRET_KEY	MinIO Secret
  MINIO_RECOMMENDER_BUCKET	MinIO recommender bucket

• Scheduler Options:

  Option	Description
  JOB_FREQUENCY	Frequency to run job (seconds interval for dev mode)
  JOB_FREQUENCY	Frequency to run job (hour of day for production mode)