GENERATE_PLOTS

# GENERATE EXPERIMENTAL EVALUATION PLOTS

## Install matplotlib
The following command installs matplotlib via the pip package manager.
  $> pip3 install --user matplotlib
On Ubuntu20, it is required to install the package via apt.
Otherwise, matplotlib fails to open the windows in which the plots are shown.
Run the following command as sudo:
  $> apt -y install python3-matplotlib

## Download data
The archive containing the output of each tool can be downloaded from:
  https://drive.google.com/file/d/1TLCOi7E_NEoH_KWgh1SZWIgzK0akWuR4/view?usp=sharing

Extract the content of such archive using:
  $> tar -xzpf infocomp2021_expeval_output.tar.gz
This should create a directory named "infocomp2021_expeval_output" of size 713 MB.

## Generate the plots.
Let $OUTPUT be the directory created at the previous point.
The python script "scripts/plot_scatter.py" generates the plots.
Each plot compares F3 with one other tool on the benchmarks of one category.
The following command generates the plots for all categories.
  $> python3 ./scripts/plot_scatter.py -ls -ns -its -ta -tts -hs -in $OUTPUT

The script prints on stdout the name of the category being shown and the number of instances solved by each tool.