tennis-dataset

This repo allow both scraping flashscore website and merging every CSV file into a POSTRESQL DB

• flashscore scraper that fill database for each new match with statistics / odds / live-score
• CSV merger that fill database for each match line

the database setup contains 9 tables :

• heads
  • combine
  • commonopponent
  • atpworldtour
  • flashscore
  • csv
  • players
  • weather
    • gps

flashscore scraper

this scraper crawl the entire flashscore.com website, more precisely all the ATP tournament begin 2001 and the player page

• The tools are the puppeter module and sequelize for the orm

the prod architecture

CSV merger

this tools merge all the csv file dataset grab into github repo to the database

how to use

exec the node index.py on each app sub-folder

Machine Learning prediction

https://github.com/panaC/machine-learning/blob/master/notebook/tennis/tennis%20pre-match.ipynb