If you want to build the packages for scenarios all you need to do is run some
scripts. Clone the directory and cd to the root diectory. From here all
commands are to be executed from the console.
To run the script, make sure the requirements e.g. via pip
pip install -U -r requirements.txt
In addition for some plots plotly-orca. Installation instructions can
found here: orca
To run the build script the required raw data needs to be downloaded. Generally,
the data will be downloaded automatically. You have to create an oemof-raw-data
directory in your home folder:
mkdir /home/user/oemof-raw-data
If raw data does not exist run:
python scripts/prepare.py
To generate sensitivities on existing scenario toml files run (adapt):
python scripts/generate_sensitivities.py
To build the package locally run the python script
python scripts/build.py
This will initialize all directories if they don't exist, (download raw data), create the meta-data file. The output data are stored under:
/datapackages
The datapackages are build with the provided .toml-files provided in the
/scenarios folder. You can adapt the configuration files or add new ones to
your needs. All datapackages are build in parallel. If you want to build single
datapackages you can use the build.py function.
The packages will be build with the multiprocessing package. Make sure that all files exis in the raw data path, for example by running the building process for only one file without multiprocessing. Sometimes errors occure in combination with downloading and unzipping multiple file with multiprocessing.
To compute the datapackages run:
python scripts/compute.py
This will create a results directory with all results.
The provided datapackages in the datapackage directory of this repository
may be outdated. Therefore you should rebuild the datapackages based on the provided
scripts/build.py script.
Plots and tables can be generated with the following scripts:
python documentation/generate_figures.py
python documentation/paper_figures.py
python documentation/generate_tables.py
Note for these scripts to work, you need to have a result directory and the directory with input datapackages on this level with results in the correct format.