Skip to content

salvirezwan/ML-Project-Cricket-Score-Predictor

Repository files navigation

Cricket Score Predictor

This is a Machine Learning Project that predicts the cricket score based on input parameters such as batting team, bowling team, city, current score, overs bowled, wickets out, and runs in the last 5 overs. I collaborated with my classmates Abeer and Rafin here. We have tried to modify and enhance further the project found in KNOWLEDGE DOCTOR YouTube channel. We altered the hyperparameters of Random Forest and XGBoost algorithms and compared the performances.

Features

  • Predicts the final score of a cricket match based on current match conditions.
  • Uses Bootstrap for responsive design and styling.
  • Validates user input to ensure meaningful predictions.

Technologies Used

  • HTML
  • CSS (Bootstrap)
  • JavaScript (jQuery)
  • Flask (for backend logic)
  • Python (for prediction logic)
  • Jupyter Notebook (for training the model)

Setup Instructions

  1. Clone the repository:

    git clone https://github.com/yourusername/cricket-score-predictor.git
    cd cricket-score-predictor
  2. Install required packages: flask, sci-kit learn, pandas, etc.

  3. Run app.py
    pipe = pickle.load(open('x.pkl', 'rb')) - Here, replace x with the name of the .pkl file which is generated by jupyter.

Customization

To add more teams or cities, update the dropdown options in the index.html file. To change the default teams or city, modify the JavaScript logic in the index.html file to set the desired default values. To adjust the prediction logic, update the backend logic in app.py.

Usage

Select the batting team, bowling team, and city from the dropdown menus. Enter the current score, overs bowled, wickets out, and runs scored in the last 5 overs. Click the "Predict Score" button.

Contributing

Contributions are welcome! Please fork the repository and submit a pull request for any features, bug fixes, or improvements.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors