luma-ml 0.3.2

Custom-made powerful and light Python machine learning module

LUMA

LUMA is a powerful and flexible Python module designed to simplify and streamline various machine learning tasks. It is specifically created to enhance the ease of building, training, and deploying machine learning models while offering extensive customization options for data scientists and developers.

luma.classifier

The luma.classifier submodule is a comprehensive toolkit for building, training, and evaluating classification models. It provides a wide range of classification algorithms, including naive bayes, decision trees, and support vector machines.

luma.clustering

The luma.clustering submodule focuses on unsupervised machine learning tasks, specifically clustering. It encompasses algorithms such as K-Means, hierarchical clustering, etc. It simplifies cluster creation, analysis, and visualization, allowing users to gain insights from their data without the need for labels.

luma.core

The luma.core submodule serves as the foundational backbone for the entire LUMA framework. It provides essential data structures and utility functions that are used throughout the LUMA ecosystem.

luma.ensemble

The luma.ensemble submodule empowers users to harness the strength of ensemble learning for improved model performance. Ensemble learning combines the predictions of multiple base models to enhance overall accuracy and robustness. The submodule includes popular ensemble methods such as Random Forests.

luma.interface

The luma.interface submodule contains files that define protocols and custom data types used internally within the luma framework. These files are not intended for direct external use but play a crucial role in the functionality and communication between various luma components.

luma.metric

The luma.metric submodule provides a rich collection of performance metrics for evaluating machine learning models. It includes metrics for classification tasks like ROC-AUC, log-loss, and confusion matrices. For regression tasks, it offers metrics such as mean squared error (MSE) and R-squared. These metrics are essential for assessing model quality and guiding model selection.

luma.migrate

The luma.migrate submodule is specifically designed to facilitate the import and export of machine learning models within the LUMA framework. This submodule is crucial for preserving and transferring the state of models across different platforms and environments, enhancing the portability and scalability of machine learning solutions.

luma.model_selection

The luma.model_selection submodule streamlines the process of selecting the best machine learning model and optimizing hyperparameters. It offers tools for hyperparameter tuning, cross-validation, and model selection, enabling users to find the optimal model configuration for their specific task.

luma.pipe

The luma.pipe submodule is dedicated to creating and managing machine learning pipelines, streamlining the process from data preprocessing to model evaluation. It offers a seamless interface for combining different stages of machine learning workflows into a coherent and efficient pipeline. This includes integrating preprocessing steps, model fitting, and post-processing tasks into a unified workflow.

luma.preprocessing

The luma.preprocessing submodule includes a variety of data preprocessing functions to ensure data is properly prepared for machine learning tasks. It covers tasks like feature scaling, one-hot encoding, handling missing values, and data splitting. Proper data preprocessing is crucial for model performance and accuracy.

luma.reduction

The luma.reduction submodule specializes in dimensionality reduction techniques. It provides methods for feature selection and extraction, reducing the dimensionality of high-dimensional datasets. This not only improves model performance but also reduces computational time and complexity.

luma.regressor

The luma.regressor submodule is tailored for regression tasks. It offers a comprehensive range of regression algorithms, such as linear regression, decision tree regression, and support vector regression. Additionally, it includes a suite of regression-specific evaluation metrics to assess model accuracy and performance.

luma.visual

The luma.visual submodule simplifies model visualization. It includes tools for plotting data, visualizing decision boundaries, and creating performance charts. These visualization aids help users gain insights from their machine learning models and communicate results effectively.

Latest Version

0.3.2

Dependencies

NumPy, SciPy, Matplotlib, Seaborn

Python Version

Python 3.10 or later

Document

LUMA Notion Document