MK8D

Python package for "Mario Kart 8 Deluxe" livesplit data analysis. It processes Livesplit LSS files (XML format) to provide a cleaner way to analyze the data by re-shaping it to a dataframe. Pypi: https://pypi.org/project/MK8D/

This pip package is an improvement upon the mk8dLivesplit implementation. The main change is that it handles runs in a dataframe for easier parsing and filtering (sacrificing a bit of processing speed).

Features

• Streamlined conversion from LSS to CSV dataframe
• Combine different LSS files into one dataframe automatically
• Plot and compare runs
  • Full run
  • Individual Tracks

Basic Usage

Install the package with:

pip install MK8D

To convert an LSS file into a CSV dataframe, run:

MK8D_lss2csv INPUT_FOLDER OUTPUT_FOLDER OUTPUT_FILENAME

where INPUT_FOLDER is the path to the directory where the LSS files are stored, OUTPUT_FOLDER is the location in which we want our files to be exported to, and OUTPUT_FILENAME is the name of the CSV file to be exported.

For example, if run from the repository directory:

MK8D_trk2csv ./MK8D/dev/data ./MK8D/dev/data MK8D_trks.csv
K8D_run2csv ./MK8D/dev/data/MK8D_trks.csv ./MK8D/dev/data MK8D_runs.csv max

would take every LSS file found in the ./MK8D/dev/data and compile a dataframe to a CSV file, which would be exported to the same folder.

Dataframe Head

There are two dataframe types generated by the package: Tracks (trks), and Runs (runs).

Tracks

• ID: yy/mm/dd (file:rid)
• Track: MK8D track name
• Time: Time elapsed in the track/split (seconds)
• Version: Digital/Cartridge
• Items: Items/No Items
• Speed: 50cc, 100cc, 150cc, 200cc
• Category: Speedrun category (32 tracks, 48 tracks, etc)

Runs

• ID: yy/mm/dd (file:rid)
• Track: MK8D track name
• Time: Time elapsed in the track/split (seconds)
• Version: Digital/Cartridge
• Items: Items/No Items
• Speed: 50cc, 100cc, 150cc, 200cc
• Category: Speedrun category (32 tracks, 48 tracks, etc)
• Split: Human-readable split time
• Center metric: Statistic calculated for each track across runs (mean, median, min, max)
• Center offset: Time as compared to the "center" statistic of the track (mean, median, min, max)

Dependencies

Pandas, Numpy, Plotly, Plotly Express, Matplotlib, Colour, XMLToDict

To do

• Fix bug with tracks sorting
• Add attempt date support
• Create command line tool
• Auto-export plots
• Export violin plots
• Add tracks filter to violin plots
• Highlight PB
• Add stats to plots
• Create progress plot
• Make timings readable
• Fix the milliseconds formatting
• Create non-interactive versions of the plots
• Add stats to violins
• Add tracks order master file support

Author

Héctor M. Sánchez C.