This is a task-oriented test frame work for learning algorithms. We provide two API version: numpy
& jax
. You can import them as submodules. For example, if you want to use jax
version:
git submodule add -b jax https://github.com/tibless/lrkit.git ./plugins/lrkit
# git submodule add -b numpy https://github.com/tibless/lrkit.git ./plugins/lrkit # use numpy version
git submodule sync
then, you should make sure there is a __init__.py
under the plugins/
folder & write setup.py
for current project. For example, put this file under your project folder:
from setuptools import setup, find_packages
setup(
name='myproject',
version='0.1',
packages=find_packages(),
include_package_data=True,
install_requires=[
'numpy',
'toml',
'tabulate',
],
)
then install by pip under Debug:
pip install -e .
- numpy
- toml
- tabulate