Custom CUDA Kernels for BatchGen Project
Project description
MoE Gen Kernel
This project is MoE Gen Kernel, providing CUDA implementations for MoE Model.
Build Instructions
Make sure Python and required dependencies are installed. Before building, ensure all submodules are initialized by running the following command in the project root directory:
git submodule update --init
Then, run the following command in the project root directory:
bash build.sh 3.10 12.8
After building, a Python wheel file (.whl) will be generated for installation and distribution in the dist directory.
Upload to PyPI
To upload the generated wheel file to PyPI, use the following command:
twine upload dist/mgn_kernel-*.whl --repository-url https://test.pypi.org/legacy/
License
See the LICENSE file for details.
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distributions
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file test_kernel_0ab602a9-0.1.4-cp39-abi3-manylinux2014_x86_64.whl.
File metadata
- Download URL: test_kernel_0ab602a9-0.1.4-cp39-abi3-manylinux2014_x86_64.whl
- Upload date:
- Size: 596.1 kB
- Tags: CPython 3.9+
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.1.0 CPython/3.10.12
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
f506efee9ba04d2faced3de369d0e750147cf8fb87b81a9643705abdb31cb511
|
|
| MD5 |
3b7eba6363f701606dc4e9311bb6f458
|
|
| BLAKE2b-256 |
f03aed8037fb381ebb1928040efcbf4a657e7ecb4e784cfb216b19b3e3529740
|
