This code accompanies my blogpost on color optimization.
First, install the requirements:
pip install -r requirements.txt
or simply
pip install colormath numpy click
Then, you can use the command-line tool as follows:
python coloropt.py WEIGHT_VECTOR INITIAL_HUES --c_from 0 --c_to 100 --l_from 0 --l_to 100 --logfir logs
where WEIGHT_VECTOR is a parameters for the objective function and INITIAL_HUES are initial guesses for the hue values for the optimal colors. Example usage:
python coloropt.py 500 100 75 50 25 100 75 50 25 10 25 10 20 10 1000 100 50 25 10 30 10 350 150 100 50 25 200 100 50 25 10 40 200 240 360 --c_from 50 --c_to 75 --l_from 40 --l_to 75 --logdir logs
You are free to use the pallettes and the code without attribution in commercial projects (specifically, assume public domain license on the colors). I would love if you drop an email if you find them useful.