Getting Started with PyBIRD AI

Note that there is also a video tutorial that goes through these steps and further steps guided by the application in order to make a BIRD implementation with a database and executable transformation rules at https://www.youtube.com/playlist?list=PLsfY4jY5qjfanjEzO4Tm5KQCPdn-kHa27 . This tutorial is very slightly out of date as it shows using the BIRD input layer, whereas the LDM is the default choice for datamodel currently. Otherwise the tutorial is still completely relevant.

Step 1

If working on a local machine (instead of github codespaces on the web), download the repository (develop branch) or clone the repository and switch to the develop branch

for example in a command line window or a terminal (such as VSCode terminal ):

git clone https://github.com/eclipse-efbt/efbt.git

git checkout develop

If using Github codespaces you can open the codespace directly from the github page after choosing the develop branch.

Currently the process is better tested with the BIRD LDM and so this is the default. We recommend trying the BIRD LDM first.

If you are using the Enriched Input Layer as opposed to the BIRD LDM (which is the default) then you should change the variable ldm_or_il in the file birds_nest\pybirdai\context\context.py to equal ‘il’

Step 2 - Setup environment

Install Python if not installed

Install an editor (tested with VSCode on windows and Github codespaces which is VSCode on Linux in the cloud in a web-browser)

Run these commands in the terminal:

python -m pip install Django

pip install pyecore

pip install unidecode

change to the birds_nest directory:

cd birds_nest

python manage.py runserver

go to the website link shown, add /pybirdai to the link

http://127.0.0.1:8000/pybirdai (don’t forget the /pybirdai at the end or you will see a 404 page not found error)

Go through the steps in order on the webpage. Not that when there are multiple 'buttons' you should go through them in order.