ml-qol 0.1.2

An ML toolkit package that provides quality-of-life features

ML QOL

ML QOL is a Python package that provides helper functions and quality-of-life features for machine learning tasks

Features

• Automated hyperparameter mapping for different models such as CatBoost and LightGBM
• Data handling functions for managing dates, NaN values, and more
• Feature engineering functions such as combining features together or adding target encoded features
• Fast and easy way to train and compare different models and their performance, e.g, feature importance, confusion matrix
• Perform folded training and gathering averaged predictions
• Perform weighted ensembling with different types of models

Dependencies

This package relies on the following Python libraries:

• pandas
• numpy
• scikit-learn
• lightgbm
• catboost
• xgboost
• matplotlib
• seabron

You can install them via pip:

pip install pandas numpy scikit-learn lightgbm catboost xgboost matplotlib seaborn

Installation

Using pip

pip install ml-qol

Quick Start

from ml_qol import train_model

# Train a model
model, model_info = train_model('lightgbm', 'regression', train_data=train_df, target_col='price')

# Show feature importances
model.plot_importance()

# Use for inference
predictions = model.predict(test_df)
print(predictions)

Resources

• GitHub repository: (https://github.com/mashrursakif/ml-qol)
• Documentation and examples: (https://github.com/mashrursakif/ml-qol/tree/main/examples/)

License

MIT

Author

Developed by Mashrur Sakif Souherdo - GitHub