vishnupsatish/microtensor

Deep learning library (training & inference) and GPT-2 in C++ from scratch.

/ 100

Emerging

This project provides a foundational deep learning library for building and training neural networks, including a GPT-2 model. It takes raw data and processes it through a custom-built deep learning framework to produce trained models or perform inference. This is intended for C++ developers who need to implement machine learning capabilities from scratch, especially in resource-constrained or performance-critical environments.

Use this if you are a C++ developer who needs a deep learning framework without external dependencies, requiring fine-grained control over model implementation and optimization.

Not ideal if you prefer using established, high-level deep learning frameworks like PyTorch or TensorFlow, or if you primarily work in Python.

deep-learning-engineering natural-language-processing machine-learning-inference custom-model-development performance-optimization

No Package No Dependents

Maintenance 10 / 25

Adoption 7 / 25

Maturity 13 / 25

Community 0 / 25

How are scores calculated?

Stars

Forks

—

Language

C++

License

MIT

Higher-rated alternatives

apache/tvm

Open Machine Learning Compiler Framework

uxlfoundation/oneDNN

oneAPI Deep Neural Network Library (oneDNN)

Tencent/ncnn

ncnn is a high-performance neural network inference framework optimized for the mobile platform

OpenMined/TenSEAL

A library for doing homomorphic encryption operations on tensors

iree-org/iree-turbine

IREE's PyTorch Frontend, based on Torch Dynamo.

Explore ML Frameworks

All categories Trending ML Framework directory Insights