I developed an end-to-end Machine Learning web app that visualizes actual data vs future predictions using Logistic Regression, all wrapped in an interactive Streamlit dashboard.
An interactive Streamlit-based Machine Learning dashboard that demonstrates Logistic Regression classification, model evaluation, and visualization of actual vs future data.
This project is designed to showcase end-to-end ML workflow — from data loading and preprocessing to model training, evaluation, and visualization — all inside a clean web interface.
- 📁 Load and preview actual dataset and future dataset
- 🔢 Automatic categorical encoding using
LabelEncoder - ⚖️ Feature scaling with
StandardScaler - 🤖 Logistic Regression model training
- 📈 Model performance evaluation:
- Accuracy score
- Confusion Matrix
- 📉 ROC Curve visualization with AUC score
- 🎨 Custom UI styling using CSS
- ⚡ Cached data loading for better performance
- Python
- Streamlit
- Pandas & NumPy
- Scikit-learn
- Matplotlib
- Load actual and future datasets
- Encode categorical variables
- Select features and target variable
- Split data into training and testing sets
- Apply feature scaling
- Train Logistic Regression model
- Evaluate model using accuracy & confusion matrix
- Visualize performance using ROC curve