A streamlit web-application allowing to plot colors on a CIE Diagram using the colour-science library.
The application allow to plot a single R-G-B color or a whole image using various methods. A lot of options allow you to customize how the final diagram looks.
Assuming you have :
- set the current working directory as desired
- installed poetry on your system
git clone https://github.com/MrLixm/streamlit-colour-plotting.git
cd streamlit-colour-plotting
poetry updateYou should now be able to run the application using the launcher :
./dev/run-streamlit.shYou can now start editing code in streamlit_colourplotting/.
It is recommended to put in place a file watcher or a pre-commit hook that would run black on all the files.
To use for local development (not on cloud).
STCP_DISABLE_SIZE_LIMITATIONS: to set with any non-empty value. Remove the 2048 size limitation for images upload.STCP_APP_LOG_LEVEL: set the python application logger level. Except a python log level in upper case. ex:WARNING
Streamlit use a single file as entry point that is re-run everytime a widget change on the web GUI.
This file is located in ./src/app.py. It is quite empty and
only import the python package streamlit_colourplotting which define everything.
A important concept to understand :
- streamlit re-run
app.pyfrom top to bottom everytime - but python cache imported package on their first import
This mean that anything in the global scope will be shared along users. This why the "config" system store everything in session_state instead of python global variables.
Note that for color manipulation, this application use cocoon which is mostly a convenient wrapper around colour library.
As streamlit cloud machine are limited in ressources, it's a good idea to check the memory profile of our app.
Once the streamlit app is started you can run the profile-running-streamlit.sh script and follow its intrsuction (make sure to launch it with the current directory being the root of this repo).
