Skip to content

Deployment of PyPSA-SPICE-Vis app

The easiest method to deploy is using git-lfs to manage the data and simply using streamlit's community cloud.

  1. Download git-lfs:
download git-lfs directly from https://git-lfs.com/.

sudo apt-get install git-lfs

brew install git-lfs

  1. To deploy the app, create a branch of the repo. This will also allow you to make changes like default values for country and scenario, graph setting, etc.

  2. Install git-lfs via the following command:

installing git-lfs
git lfs install

  1. Add the results data files inside this branch. These data files will be tracked using git-lfs. Note: add only the CSV files but do keep the same folder structure as that of pypsa-spice results.

  2. Add path to this results folder inside pypsa-spice-vis/setting/initial_project_01_deploy.yaml. Note: the pypsa-spice-vis/setting/initial.yamlis ignored by git and used for local deployment.

  3. In pypsa-spice-vis/main.py, make DEPLOY == True. By default, False for local run.

  4. Initialise and track the result files using git-lfs. You will have to add all csvs to be tracked with lfs tracking system with following commands based on your operation system:

git lfs install
git lfs track "*.csv"
git add .gitattributes
git commit -m "Track all CSV files with Git LFS"
find . -name "*.csv" -exec git add {} \; 

git lfs install
git lfs track "*.csv"
git add .gitattributes
git commit -m "Track all CSV files with Git LFS"
for /r %i in (*.csv) do git add "%i"

After adding all csv files, you can initialise and commit tracked files using git-lfs:

committing all tracked files
git commit -m "Add all CSV files"
git push

Now the repository can be linked simply to Streamlit's deployment system.