MLHelper

A Python toolkit for streamlining machine learning workflows, from data preprocessing to model evaluation.

Features

• Data Preprocessing: Handle missing values, encode categorical variables, scale numerical features, and identify outliers
• Feature Engineering: Create polynomial features, interaction terms, bin numerical features, and extract date features
• Model Evaluation: Evaluate model performance with various metrics and cross-validation techniques
• Visualization: Visualize feature importance, correlations, distributions, and model performance
• Model Helpers: Split data, perform hyperparameter tuning, and save/load models

Installation

pip install git+https://github.com/lognjen/MLHelper.git

Module Overview

• data_processing.py - Contains the DataPreprocessor class with methods for handling missing values, encoding, scaling, and removing outliers
• feature_engineering.py - Contains the FeatureEngineering class with methods for polynomial features, interaction terms, binning, and date features
• model_evaluation.py - Contains the ModelEvaluation class with methods for evaluating models and cross-validation
• visualization.py - Contains the Visualization class with methods for plotting feature importance, correlations, and distributions
• model_helper.py - Contains the ModelHelper class with methods for data splitting, grid search, and model persistence

Requirements

• Python 3.7+
• NumPy
• Pandas
• Scikit-learn
• Matplotlib
• Seaborn

License

This project is licensed under the MIT License - see the LICENSE file for details.