A tensor-valued autograd engine for Python
Project description
cudagrad
CUDA C++ strided float tensor automatic differentiation engine with Python bindings
Install
Available on PyPI as a source distribution, requires cmake and optionally nvcc if available:
pip install cudagrad
Examples
The following examples were written purely in Python using only cudagrad.Tensor for learning:
OR
0.52 seconds (59.5% faster than torch)
XOR
4.5 seconds (39.2% faster than torch)
MOONS
14.25 seconds (5.8% slower than torch)
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
cudagrad-0.2.2.tar.gz
(4.8 MB
view details)
File details
Details for the file cudagrad-0.2.2.tar.gz.
File metadata
- Download URL: cudagrad-0.2.2.tar.gz
- Upload date:
- Size: 4.8 MB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.1.1 CPython/3.11.9
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
1076e666232a2ad68d85e63438077362210cc0bef0ba649dd94aae580858e112
|
|
| MD5 |
ad6d8d3f4484888e4ac3e2d7b9798a63
|
|
| BLAKE2b-256 |
d91ebdf6655fa03ca03eb0d13ee2993bee94b9133f0bafa329f9b00d06f7ef07
|
