This project is a machine learning classifier for predicting whether a bank customer is likely to churn (leave) or not. It includes a Streamlit web application that allows users to interact with the predictive model and visualize the results.
Click here to view the prediction app in your web browser.
Here are some pictures of what the app looks like:
1. Prediction Page
2. Visualization Page
List the prerequisites that users need to have installed or set up before using the project.
python>=3.8
requirements.txt
To use my application, follow this steps below to successfully install and run the program.
# Clone the repository
git clone https://github.com/juliusmarkwei/Customer-Churn-EDA-Balancing-and-ML.git
# Change directory
cd Customer-Churn-EDA-Balancing-and-ML/
# Install dependencies
pip install -r requirements.txt
Carefully type the command below in your teminal of the "Customer-Churn-EDA-Balancing-and-ML/" directory to run the app.
# Run the Streamlit app
streamlit run app.py
Our machine learning model was trained using a dataset containing [describe your dataset]. The training process involved the following steps:
-
Data Preprocessing: We performed data cleaning, handled missing values, and encoded categorical features as part of data preparation.
-
Model Selection: We selected the Random Forest model for the prediciton app after evaluation as the base model due to its suitability for our problem.
-
Model Evaluation: The model's performance was evaluated using metrics accuracy and F1-score. Cross-validation was used to assess its generalization ability.
-
Hyperparameter Tuning: We fine-tuned the model's hyperparameters to optimize performance.
For detailed information on the model training process, please refer to the training notebook.
We welcome contributions to improve this project! Whether it's bug reports, feature suggestions, or code contributions, we appreciate your help.
-
Reporting Issues: If you encounter a problem or have a suggestion, open an issue with details.
-
Making Pull Requests: Feel free to submit pull requests for fixes or enhancements. Follow common coding standards and provide clear descriptions for your changes.
Thank you for your contributions!
This project is licensed under the MIT License - see the LICENSE file for details.