numpy, pandas, sklearn, mahjong, requests
make sure to python3 -m pip install numpy pandas sklearn mahjong requests before running
All the html files in htmls contains link to a game with Paifu
webparser.py iterates through all the html files in htmls to create [gameid].txt files
Those text files are the raw files for all the games
Mjlogreader3.py iterates through all the text files in mjlog and writes out csv files into csvs
As of now, mjML.py uses machines from sklearn to predict and uses Mean Squared Errors or Mean Absolute Errors to get score.
The score that it prints out is calculated as 1 - (score/6)
python3 Mjlogreader3.py to make csv files from mjlogs
Run "webparser.py" to get datas from htmls