Overview
The Machine Learning App is designed to provide a user-friendly interface for uploading data, selecting machine learning algorithms, training models, and evaluating their performance. I am planning to add more features for supervised and unsupervised classification and include scikit-learn preprocessing methods, this project is still in development phase
Technology Stack
- Python: Programming language for backend logic.
- Streamlit: Web application framework for data apps.
- Scikit-learn: Machine learning library for Python.
- Pandas: Data manipulation library for Python.
Features
Data Handling
- Upload CSV: Upload datasets in CSV format for analysis and modeling.
- Missing Values Handling: Options to ignore or fill missing values.
Model Selection and Training
- Select Columns: Choose columns for modeling.
- Select Target: Choose the target variable for prediction.
- Algorithm Selection: Choose from various classification and regression algorithms.
- Parameter Tuning: Manually set or auto-tune parameters.
Model Evaluation
- Train-Test Split: Options for random or shuffled split of data.
- Performance Metrics: Display accuracy, RMSE, and confusion matrix.
Cross-Validation
- Perform cross-validation and display average scores.
User Interface
Responsive Design
- The application features a responsive design ensuring usability across various devices.
User-friendly Interface
- Clean and intuitive interface for easy navigation and model building.
