Skip to content

Latest commit

 

History

History
30 lines (23 loc) · 1.43 KB

plot_graphs.md

File metadata and controls

30 lines (23 loc) · 1.43 KB

How to plot graphs comparing runs ?

Plotting graphs is used only for multi-run experiments.

cd /home/binsec-ase/artefact/plotting_graphs

Plotting experiment results to get paper graphs

Since the full experiment considered in this paper takes more than a week with 40 CPU cores to run, we simply provide the .csv file aggregating the raw results of said experiment to showcase its transformation into the graphs in the paper. You will find it in plotting_graphs/records_paper.

Simply launch

python3 plot_graphs.py ./records_paper/raw_paper_stats.csv paper

to get the graphs in plotting_graphs/figures, prefixed by paper. The analysis time graphs are labelled AT, the explored paths are labelled EP, the average solving time per query graphs are labelled ST and the number of query sent graph is prefixed QS. Graphs without the Forking technique which highlight the differences between our variants are labelled zoom.

Plotting your own results

To plot your own results instead, use the aggregating script. For instance, if you want to get the results from a multi-run called check_7.3, use the following command.

bash concat_and_move_csv.sh check_7.3

It will aggregate each individual .csv files into a big one and place it in plotting_graphs/records_artefact/ with the name raw_check_7.3.csv.

Then launch the plotting script

python3 plot_graphs.py ./records_artefact/raw_check_7.3.csv check_7.3