yrmo / cudagrad

CUDA C++ strided float tensor automatic differentiation engine with Python bindings
MIT License
0 stars 0 forks source link

cudagrad

CUDA C++ strided float tensor automatic differentiation engine with Python bindings

RepositoryDocumentation

Install

This project is available on PyPI, but requires cmake and nvcc to be available at installation time:

pip install cudagrad

Examples

The following examples were written purely in Python using only cudagrad.Tensor for learning

Neuron

OR

/examples/or.py (in 0.57 seconds)

Multilayer perceptron (MLP)

XOR

/examples/xor.py (in 4.83 seconds)

Two moons

/examples/moons.py (in 12.79 seconds)

Convolutional neural network (CNN) WIP!

MNIST

/examples/mnist.py