lichess-time-spent

Extracting time spent playing on Lichess by player, and analysing it.

Extracting time spent playing

To extract the data, you need a working rust installation.

Then in a terminal:

• Clone the repository: git clone https://github.com/kraktus/lichess-time-spent

• Enter the lichess-time-spent directory then run: cargo run --release -- <PATH_TO_PGN> <NUMBER_OF_GAMES_IN_PGN>

PATH_TO_PGN can lead to a compressed file that will be decompressed on the fly. You can use database.lichess.org to download compressed versions of Lichess rated games.

NUMBER_OF_GAMES_IN_PGN is just used for the progress bar and compute approximate duration of operation. You can use any number if you don't know or care. The results are stored in time-spent.csv put in the current directory.

Data analysis

Some data analysis can be found in data-analysis.ipynb. To run it:

• Create a venv: python3 -m venv venv
• Turn it on: source venv/bin/activate
• Install the dependencies: pip3 install -r requirements.txt
• In data-analysis.ipynb, edit the line df = pd.read_csv(\"time-spent-2023-01.csv\",dtype=dtypes), replacing time-spent-2023-01.csv with the name of your CSV file (time-spent.csv if you have not renamed it).
• Run jupyter notebook which will open jupyter in the browser.

Proofreading

To produce a PDF file for easier proof-reading:

• First make sure to save the document (checkpoint) in jupyter UI
• Then run jupyter nbconvert data-analysis.ipynb --no-input --to pdf && open data-analysis.pdf

Export

Produce the figures

jupyter nbconvert data-analysis.ipynb --to python && python3 -O data-analysis.py

Produce the markdown

jupyter nbconvert data-analysis.ipynb --no-input --to markdown