ruby/tensorflow_privacy
1
0
Fork 0
forked from 626_privacy/tensorflow_privacy
Code Pull requests Activity

Update the conventions used to specify the version of a Python dependency.

Browse source 
* For packages that have a stable release, we use a version that is compatible with that release (e.g. `~=x.y`).
* For packages that do not have a stable release, we use a version that matches a release that has been tested (e.g. `==x.y.z`).

Additionally, added documentation to the requirements.txt.

PiperOrigin-RevId: 468720651
This commit is contained in:
Galen Andrew 2022-08-19 09:37:38 -07:00 committed by A. Unique TensorFlower
parent d6ad59226d
commit 7fe491f7a4
2 changed files with 44 additions and 21 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

45
requirements.txt
View file

@ -1,14 +1,37 @@
absl-py~=1.0.0 # Requirements for Tensorflow Privacy.
attrs~=21.4.0 #
dm-tree~=0.1.7 # If you add a dependency here, also add the dependency to setup.py.
dp-accounting~=0.1.2 #
matplotlib~=3.3.4 # * For packages that have a stable release, we use a version that is
mpmath~=1.2.1 # compatible with that release (e.g. `~=x.y`). See
numpy~=1.23.2 # https://peps.python.org/pep-0440/#compatible-release for more information.
pandas~=1.4.3 # * For packages that do not have a stable release, we use a version that
scikit-learn~=1.0.2 # matches a release that has been tested (e.g. `==x.y.z`). See
scipy~=1.7.1 # https://peps.python.org/pep-0440/#version-matching for more information.
tensorflow-datasets~=4.5.2 #
# This assumes that the packages follows Semantic Versioning, see
# https://semver.org/. If a package follows a different versioning scheme or
# requires unique handling, we use a different version specifier and comment the
# versioning scheme or reasoning.
#
# Note: As of 2022-08-17 there is bug in `pip` when multiple packages use the
# compatible release operator `~=` to specify a version and one of those
# versions ends in `0`. See https://github.com/pypa/pip/issues/9613 for more
# information. In this case, use the equivalent clause `>=x.0,==x.*` instead of
# `~=x.0`.
absl-py>=1.0,==1.*
attrs~=21.4
dm-tree==0.1.7
dp-accounting==0.1.2
matplotlib~=3.3
mpmath~=1.2
numpy~=1.23
pandas~=1.4
scikit-learn>=1.0,==1.*
scipy~=1.7
tensorflow-datasets~=4.5
tensorflow-estimator~=2.4 tensorflow-estimator~=2.4
tensorflow-probability~=0.15 tensorflow-probability~=0.15
tensorflow~=2.4 tensorflow~=2.4

20
setup.py
View file

@ -27,16 +27,16 @@ setup(
url='https://github.com/tensorflow/privacy', url='https://github.com/tensorflow/privacy',
license='Apache-2.0', license='Apache-2.0',
install_requires=[ install_requires=[
'absl-py~=1.0.0', 'absl-py~=1.0',
'attrs~=21.4.0', 'attrs~=21.4',
'dm-tree~=0.1.7', 'dm-tree==0.1.7',
'dp-accounting~=0.1.2', 'dp-accounting==0.1.2',
'matplotlib~=3.3.4', 'matplotlib~=3.3',
'numpy~=1.23.2', 'numpy~=1.23',
'pandas~=1.4.3', 'pandas~=1.4',
'scikit-learn~=1.0.2', 'scikit-learn~=1.0',
'scipy~=1.7.1', 'scipy~=1.7',
'tensorflow-datasets~=4.5.2', 'tensorflow-datasets~=4.5',
'tensorflow-estimator~=2.4', 'tensorflow-estimator~=2.4',
'tensorflow-probability~=0.15', 'tensorflow-probability~=0.15',
'tensorflow~=2.4', 'tensorflow~=2.4',