The modules found in this project allow one to create an actual-vs-predicted plot using seaborn and LaTeX. Furthermore, they allow one to print the regression scores on the plot and add a hue to the markers according to a QGIS palette for the Local Climate Zone (LCZ) variable.
-
Clone GitHub repository and change directory:
git clone https://github.com/eliocpereira95/fancy_plotter_actual_vs_predicted.git cd fancy_plotter_actual_vs_predicted
-
Install LaTeX
In the case of Windows, you would need to install MiKTeX.
In the case of a Debian-based Linux distribution (such as Ubuntu), you do:
sudo apt-get install -y dvipng texlive-latex-extra texlive-fonts-recommended cm-super
-
Make sure to initialize
conda
using the command below. After that, open a new terminal which would then have the baseconda
virtual environment activated by default.conda init
-
Create a dedicated
conda
virtual environment, activate it and install the required packages:conda create -n fancy_plotter python=3.12 conda activate fancy_plotter pip install -r requirements.txt
Have a look at the Jupyter notebook examply.ipynb
to undestand how to
use the modules. Be sure to use the created virtual environment as kernel to run it.