MMES ensamble ingestion and creation
This repository contains python script that daily generate the Multi-Model Ensable Sytem (MMES) from different sources. The software was originally developed for I-STORMS Web System and further improved within STREAM project. The software at the moment is tested on GNU/Linux environment running in other OS is not assured Also bash scripts are required to process each forecast and compute Ensamble. Bash scripts are in scripts folder
Sources must be defined in sources.json file (refer to sources_template.json to define your list).
- Create a python virtualenv and activate it.
python3 -m venv /path/to/new/virtual/environment/mmes
source activate /path/to/new/virtual/environment/mmes/bin/activate
-
Clone this repository in a convenient location.
-
Install requirements with
pip install -r requirements.txt
- Prepare the data directory structure according to [documentation](https://cnr-ismar.github.io/mmes/_build/html/index.html 5Manually edit config.json according to your needs
- From the main directory launch
python manage.py new
to create a new source config file
python manage.py mod
to edit existing source config file
python manage.py add
to add a new source to current sources config file
type
python mmes_creation.py
to create a new ensemble for current day and all variables
python mmes_creation.py -i
to run the program in a interactive manner: you will be prompted which source and which model to process and you can skyp the ensemble creation
python mmes_creation.py -v "waves" -d 20230530
to create a new ensemble only for variable "waves" and for specific date 30 may 2023. Useful to fill gaps in history
python mmes_creation.py --help
for available options