Metadata-Version: 2.1
Name: unrollcuda
Version: 0.0.2
Summary: Loop unrolling and batching for CUDA
Project-URL: Homepage, https://github.com/format37/unrollcuda
Project-URL: Bug Tracker, https://github.com/format37/unrollcuda/issues
Author-email: Aleksei Iurasov <format37@gmail.com>
License-File: LICENSE
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Requires-Python: >=3.7
Requires-Dist: numpy
Description-Content-Type: text/markdown

# unrollcuda
Loop unrolling and batching for CUDA  
The core idea of this solution is to give a way to solve the following tasks:  
1. Use Loop unrolling to compute in CUDA any size and any count of dimensions array  
2. Use Batching to compute any size array, even if it s big that can't be fitted in GPU memory  