sid0402 / ipl-price-predictor

A machine learning project to predict the prices that cricketers in the IPL Player Auction will receive.

IPL-Price-Predictor

A machine learning project to determine cricket player prices in the Indian Premier League annual auctions. The dataset has been built by scraping data from the howstat website. A Support Vector Regression (SVR) model has been developed to predict prices for batsman and wicketkeepers, and a Linear Regression model has been built to predict prices for bowlers and all-rounders. The entire project has been built using Python and its libraries.

Installation

The requirements.txt file should list all Python libraries that your notebooks depend on, and they will be installed using:

pip install -r requirements.txt

Usage

To build the dataset "AuctionId.csv", run the following (in the specified order) on terminal:

python scrapers/id_scraper.py
python scrapers/stats_scraper.py
python scrapers/stats_analyzer.py

To process the dataset, run the following on terminal:

python ml/data_processing.py

The two files, ml/train.py and ml/train_bowl.py builds the batting/wicketkeeper model and bowler/allrounder models respectively.

Datasets

The "IPLPlayerAuctionData.xlsx" dataset was taken from Kaggle. This dataset was upgraded by scraping from howstat.com, and the upgraded dataset is "AuctionId.csv". The following features were scraped:

1. Career Statistics (Matches, Innings batted, Aggregate runs, Average, Highest Score, 30s, 50s, 100s, 4s, 6s, Scoring Rate, Overs Bowled, Wickets, 3 wickets in an innings, Economy Rate, Bowling Strike Rate)
2. Career statistics until the auction year(Matches, Aggregate runs, Batting Average, Scoring Rate, Wickets taken, Economy Rate, Bowling Strike Rate, Bowling average)
3. Last Season Statistics (Matches, Runs scored, Highest Score, Batting Average, Scoring Rate, Overs bowled, Wickets taken, Bowling average, Bowling Strike Rate)