KIS Log Analysis - VBS 2022

This repo provides a good code base for extracting and analyzing the logs from the different teams that participated to VBS 2022.

Hopefully, this repo will help to analyze also the logs from the future versions of VBS.

Data Preparation

First of all, run the following command. This will extract log data from 2022 teams and will prepare the environment. Some data are collected from the local logs of the systems, others from the public run file of the DRES server.

./prepare.sh
source venv/bin/activate

Generate a plot

A plot can be generated following the configuration written in a yaml configuration file. This is an example call to construct the time-recall table:

python plot.py time_recall_table --config config2022.yaml

You can use the --help argument to see all the available options. Note that, in order to optimize the data processing, two levels of cache are used. You can disable caching using the --no_result_cache and --no_log_cache options.

Add custom plots

Adding a custom plot is quite straightforward. These are the steps:

1. Inside the generate folder, create a .py file containing a class extending the Result class. In particular, you have to provide the methods:
   • _generate(): here you generate a nice view of the team log data, returning a Pandas dataframe. This dataframe is automatically cached and re-used (only if --no_result_cache is not set).
   • _render(df): it renders the dataframe into a graph or a table.
2. Expose this class in the generate/__init__.py file.
3. Add a corresponding entry in the yaml configuration file.

Look at the time_recall_table.py file for an example.

Contributors

Lucia Vadicamo - lucia.vadicamo@isti.cnr.it

Nicola Messina - nicola.messina@isti.cnr.it

Ladislav Peška - ladislav.peska@matfyz.cuni.cz