Breast Cancer Prediction Project

This project uses machine learning to predict whether a tumor is benign or malignant based on its characteristics.

Overview

The project includes:

• Cancer_Data.csv: The dataset used for training the model.
• train.py: Python script to train the machine learning model (Logistic Regression) and save the trained model and scaler.
• cancer_pred.py: Python script to get feature inputs from the terminal and make a prediction using the saved model.
• app.py: A Streamlit web application for interactive predictions.
• breast_cancer_model.joblib: The saved trained machine learning model.
• breast_cancer_scaler.joblib: The saved scaler used for feature preprocessing.
• requirements.txt: Lists the Python libraries required to run the project.
• .gitignore: Specifies files that Git should ignore.

Usage

Training the Model

Run the train.py script to train the model and save the necessary files:

python train.py

Terminal Prediction

Run the cancer_pred.py script to get a prediction by entering feature values in the terminal:

python cancer_pred.py

Web Application

Run the app.py script using Streamlit to interact with the web application:

streamlit run app.py

Data Source

The project uses the Breast Cancer dataset (contained in Cancer_Data.csv).

This dataset is publicly available on Kaggle: Benign and malignant Cancer Dataset on Kaggle

Please refer to the Kaggle page for more information about the dataset, its attributes, and any applicable licenses.

Model

A Logistic Regression model was used for prediction.

Libraries

• Streamlit
• Pandas
• Scikit-learn
• Joblib

Author

Joe Mathew