neuralfabric 0.7.1

A from-scratch machine learning and deep learning framework, from tensors to transformers.

NeuralFabric

A from-scratch machine learning and deep learning framework, from tensors to transformers.

Features

• Tensor engine built on NumPy
• Automatic differentiation (Autograd)
• Linear Regression
• Logistic Regression
• Train-test split utility
• Modern Python packaging
• Type-safe codebase with MyPy
• Ruff, Black, Pytest, and GitHub Actions integration

Installation

pip install -U neuralfabric

For development:

pip install -e ".[dev]"

Quick Start

from neuralfabric.linear_model import LinearRegression
from neuralfabric.model_selection import train_test_split

X_train, X_test, y_train, y_test = train_test_split(
    X,
    y,
    test_size=0.2,
)

model = LinearRegression()
model.fit(X_train, y_train)

predictions = model.predict(X_test)

Project Structure

src/neuralfabric/
├── core/               # Tensor + autograd engine
├── linear_model/       # Linear Regression, Logistic Regression
├── model_selection/    # train_test_split
└── utils/              # Utility functions

Roadmap

Completed

• Tensor implementation
• Automatic differentiation engine
• Linear Regression
• Logistic Regression
• Train-test split utility
• Unit testing
• CI/CD workflows
• PyPI publishing

Upcoming

• Ridge Regression
• Lasso Regression
• Elastic Net
• Decision Trees
• Random Forests
• Support Vector Machines
• K-Means Clustering
• PCA
• Neural Network API
• Optimizers (SGD, Adam)
• Transformer Architecture
• Documentation Website

Development

make dev
make test
make lint
make format
make build
make publish

License

MIT License. See LICENSE for details.