NLCPy : NumPy-like API accelerated with SX-Aurora TSUBASA
NLCPy is a library for accelerating performance of Python scripts using NumPy on SX-Aurora TSUBASA. Python programmers can use this library on Linux/x86 of SX-Aurora TSUBASA. NLCPy's API is designed based on NumPy's one. The current version provides a subset of NumPy's API.
Requirements
Before the installation, the following components are required to be installed on your x86 Node of SX-Aurora TSUBASA.
-
- required NEC C/C++ compiler version: >= 5.4.0
- required NLC version: >= 3.0.0
-
Alternative VE Offloading (AVEO)
-
required version: >= 3.6.0
-
If you install NLCPy from wheel, the runtime packages of Alternative VE Offloading(AVEO) are required.
# yum install veoffload-aveo veoffload-aveorun-ve1 veoffload-aveorun-ve3 -
If you install NLCPy from source, the development packages of Alternative VE Offloading(AVEO) are required.
# yum install veoffload-aveo-devel veoffload-aveorun-ve1-devel veoffload-aveorun-ve3-devel
-
-
veosinfo3
# yum install veosinfo3 -
Python - required version: 3.12
-
NumPy - required version: >= v1.26.0
Install from wheel
You can install NLCPy by executing either of following commands.
-
Install from PyPI
$ pip install nlcpy -
Install from your local computer
-
Download the wheel package from GitHub.
-
Put the wheel package to your any directory.
-
Install the local wheel package via pip command.
$ pip install <path_to_wheel>
-
The shared objects for Vector Engine, which are included in the wheel package, are tested by using NEC C/C++ Compiler Version 5.0.0 and NumPy v1.19.5.
Install from source (with building)
Before building source files, please install following packages.
$ pip install numpy cython wheel
$ sudo yum install veosinfo3-devel
$ sudo yum install veoffload-aveo-devel veoffload-aveorun-ve1-devel veoffload-aveorun-ve3-devel
And, entering these commands in the environment where gcc and ncc commands are available.
$ git clone https://github.com/SX-Aurora/nlcpy.git
$ cd nlcpy
$ python3 setup.py build_ext --targ ve1,ve3,vh
$ python3 setup.py intall --targ ve1,ve3,vh
Documentation
License
The BSD-3-Clause license (see LICENSE file).
NLCPy is derived from NumPy, CuPy, PyVEO, and numpydoc (see LICENSE_DETAIL/LICENSE_DETAIL file).
Metadata
Release files for nlcpy-ve1-kernel 3.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| nlcpy_ve1_kernel-3.1.0-py3-none-any.whl | Python 3 | none | any | Details for nlcpy_ve1_kernel-3.1.0-py3-none-any.whl |
Release files / nlcpy_ve1_kernel-3.1.0-py3-none-any.whl
| Download URL | nlcpy_ve1_kernel-3.1.0-py3-none-any.whl |
|---|---|
| Size | 82.4 MB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
e7285619d7aa7ad9b8546aae6c55ace00ca150ae3ad010818a1655ea6b918ae0
|
|
BLAKE2b-256 checksum How to use checksums |
3a38ad8b24e87d9cc1d487ed4c6ee3744ed5e13680c3f654b4907ec261ff0758
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.12.13
|
Security for nlcpy-ve1-kernel 3.1.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.