https://github.com/chanlumerico/luma
A Comprehensive Python Module for Machine Learning and Data Science
https://github.com/chanlumerico/luma
framework machine-learning module python
Last synced: 8 months ago
JSON representation
A Comprehensive Python Module for Machine Learning and Data Science
- Host: GitHub
- URL: https://github.com/chanlumerico/luma
- Owner: ChanLumerico
- License: gpl-3.0
- Created: 2023-11-07T12:55:25.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2024-05-19T19:57:21.000Z (almost 2 years ago)
- Last Synced: 2024-05-19T20:46:11.919Z (almost 2 years ago)
- Topics: framework, machine-learning, module, python
- Language: Python
- Homepage:
- Size: 5.32 MB
- Stars: 2
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
- Code of conduct: CODE_OF_CONDUCT.md
Awesome Lists containing this project
README
A Comprehensive Python Module for Machine Learning and Data Science
## About
Luma is a comprehensive, user-friendly Python library designed for both beginners
and advanced users in the field of machine learning and data science. It provides
a wide range of tools and functionalities to streamline the process of data analysis,
model building, evaluation, and deployment.
### Purpose
Luma is built for an educational purpose, focused on implementing various machine learning algorithms and models **from scratch** solely depending on low-level libraries such as `NumPy`.
### Key Features
- **Easy Data Handling**: Simplify data preprocessing, transformation, and visualization.
- **Model Building**: Access a variety of machine learning algorithms and models.
- **Model Evaluation**: Utilize robust tools for model validation and tuning.
## Packages
| Name | Description |
| --- | --- |
| `luma.classifier` | Toolkit for classification models including various algorithms. |
| `luma.clustering` | Focuses on unsupervised learning and clustering algorithms. |
| `luma.core` | Foundational backbone providing essential data structures and utilities. |
| `luma.ensemble` | Ensemble learning methods for improved model performance. |
| `luma.extension` | Various extensions for Luma development. Not for end-users. |
| `luma.interface` | Protocols and custom data types for internal use within Luma. |
| `luma.metric` | Performance metrics for evaluating machine learning models. |
| `luma.migrate` | Import and export of machine learning models within Luma. |
| `luma.model_selection` | Tools for model selection and hyperparameter tuning. |
| `luma.neural` [🔗](https://github.com/ChanLumerico/luma-neural) | Deep learning models and neural network utilities. A dedicated DL package for Luma. |
| `luma.pipe` | Creating and managing machine learning pipelines. |
| `luma.preprocessing` | Data preprocessing functions for machine learning tasks. |
| `luma.reduction` | Dimensionality reduction techniques for high-dimensional datasets. |
| `luma.regressor` | Comprehensive range of regression algorithms. |
| `luma.visual` | Tools for model visualization and data plotting. |
---
## Getting Started
### Installation
To get started with Luma, install the package using `pip`:
```bash
pip install luma-ml
```
Or for a specific version,
```bash
pip install luma-ml==[any_version]
```
### Import
After installation, import Luma in your Python script to access its features:
```python
import luma
```
### Acceleration
Luma supports `MLX` based `NumPy` acceleration on **Apple Silicon**. By importing Luma’s neural package, it will automatically detect Apple’s Metal Performance Shader(MPS) availability and directly apply MLX acceleration for all execution flows and operations using `luma.neural`.
```python
import luma.neural
```
Otherwise, the default CPU based operation is applied.
For more details, please refer to the link 🔗 shown at Luma’s neural package description.
---
## Others
### Contribution
Luma is an open-source project, and we welcome contributions from the community. 😃
Whether you're interested in fixing bugs, adding new features, or improving documentation, your help is appreciated.
### License
Luma is released under the GPL-3.0 License. See `LICENSE` file for more details.
### Inspired By
Luma is inspired by these libraries:
### Specifications
| | Description |
| --- | --- |
| Latest Version | 1.2.5 |
| Lines of Code | ~40.1K |
| Dependencies | NumPy, SciPy, Pandas, Matplotlib, Seaborn, MLX(Optional) |