woongbinchoi / english-premier-league-prediction Goto Github PK

View Code? Open in Web Editor NEW

58.0 3.0 25.0 7.31 MB

Apply machine learning to predict English Premier League soccer match.

License: MIT License

Python 100.00%

soccer machine-learning premier-league

english-premier-league-prediction's Introduction

English-Premier-League-Prediction

Apply machine learning to predict English Premier League soccer match.

Demo App

To Run

python3 model.py

Warning: Few python packages required to run the script. Install them all, or use a 3rd party IDE (such as spyder) that pre-installs these packages by default. Few of the packages required are:

brew cask install chromedriver
brew install libomp
pip3 install xgboost
pip3 install selenium

Scripts

1. clean_data.py

Includes necessary helper functions to process raw data

2. current_status.py

Collects and adds more details to the processed raw data
current/past standings, goals for/against/differences, etc.

3. match_history.py

Collects the latest match results

4. rankings.py

Calculate league points and generate standings

5. sofifa_scraper.py

Scrape overall team stat from FIFA

6. predict.py

With using processed data, train a ML model to predict future results

7. model.py

I/O file where the functions from the above files are actually executed

Data

1. data/raw/OVAs (directory)

scraped overall team stat data

2. data/cleaned/standings (directory)

historical standing results calculated in rankings.py

3. data/raw/results (directory)

manually collected historical data of match outcomes
latest match outcomes of the current season

4. data/cleaned/results (directory)

data extracted from data/raw/results

5. data/train_data/results (directory)

data processed from data/cleaned/results

6. data/statistics (directory)

data/statistics/round_rankings (directory)
- standings calculated based on the predicted match outcomes
- each file in the directory has a date included in its name. It provides predicted standing outcomes at the denoted date
data/statistics/prediction_ranking.csv
- predicted standing at the end of the season
data/statistics/prediction_result.csv
- individual predicted match outcomes
data/statistics/round_rankings_summary.csv
- predicted standing summary over the course of the season

7. data/statistics/best_clf.joblib

disk cache of classifier that gives the best accuracy of prediction

8. data/database.db

sql database that stores previous match outcomes, predicted match results and predicted standings

9. data/train_data/final.csv

csv file used for training a model and making predictions

10. data/statistics/model_confidence.csv

list of grid searched classifiers and its confidence score

english-premier-league-prediction's People

Contributors

Stargazers

Watchers

english-premier-league-prediction's Issues

Proxy URL issue

Hi,
I try to run the 'sofifa_scarper.py' function. but I have an issue with URL from lines.

    # Start scraping.
    browser = webdriver.Chrome(ChromeDriverManager().install())
    browser.get("https://sofifa.com/teams?type=all&lg%5B%5D=13&lg%5B%5D=14")

Traceback (most recent call last):
File "c:\Users\hosseinv\PycharmProjects\English-Premier-League-Prediction-master\sofifa_scraper.py", line 134, in
scrape_team_ova_all("data/OVAs", 2006, 2019)
File "c:\Users\hosseinv\PycharmProjects\English-Premier-League-Prediction-master\sofifa_scraper.py", line 118, in scrape_team_ova_all
scrape_team_ova(from_year, to_year, path)
File "c:\Users\hosseinv\PycharmProjects\English-Premier-League-Prediction-master\sofifa_scraper.py", line 20, in scrape_team_ova
browser = webdriver.Chrome(ChromeDriverManager().install())
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "c:\Users\hosseinv\PycharmProjects\English-Premier-League-Prediction-master\venv\Lib\site-packages\selenium\webdriver\chrome\webdriver.py", line 80, in init
super().init(
File "c:\Users\hosseinv\PycharmProjects\English-Premier-League-Prediction-master\venv\Lib\site-packages\selenium\webdriver\chromium\webdriver.py", line 105, in init
command_executor=ChromiumRemoteConnection(
^^^^^^^^^^^^^^^^^^^^^^^^^
File "c:\Users\hosseinv\PycharmProjects\English-Premier-League-Prediction-master\venv\Lib\site-packages\selenium\webdriver\chromium\remote_connection.py", line 31, in init
super().init(remote_server_addr, keep_alive, ignore_proxy=ignore_proxy)
File "c:\Users\hosseinv\PycharmProjects\English-Premier-League-Prediction-master\venv\Lib\site-packages\selenium\webdriver\remote\remote_connection.py", line 180, in init
self._conn = self._get_connection_manager()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "c:\Users\hosseinv\PycharmProjects\English-Premier-League-Prediction-master\venv\Lib\site-packages\selenium\webdriver\remote\remote_connection.py", line 151, in _get_connection_manager
return urllib3.ProxyManager(self._proxy_url, **pool_manager_init_args)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "c:\Users\hosseinv\PycharmProjects\English-Premier-League-Prediction-master\venv\Lib\site-packages\urllib3\poolmanager.py", line 481, in init
raise ProxySchemeUnknown(proxy.scheme)
urllib3.exceptions.ProxySchemeUnknown: Proxy URL had no scheme, should start with http:// or https://

and also, I can not install these packages by pycharm with python version 3.11.
how should install those, if this issue is related to this part?!

brew cask install chromedriver
brew install libomp

No connection could be made because the target machine actively refused it

Hey, I am working on a similar task and wanted to test your code to see if I could obtain any insights that might help me with the task.

I get a "No connection could be made because the target machine actively refused it" error. This is accompanied by a 404 in the browser. Do you have any idea why this is the case?

Regards, Ingo