-
Notifications
You must be signed in to change notification settings - Fork 27
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Python 3.12 version bump #528
base: develop
Are you sure you want to change the base?
Conversation
PR-Agent was enabled for this repository. To continue using it, please link your git user with your CodiumAI identity here. PR Reviewer Guide 🔍Here are some key observations to aid the review process:
|
PR-Agent was enabled for this repository. To continue using it, please link your git user with your CodiumAI identity here. PR Code Suggestions ✨Explore these optional code suggestions:
|
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## develop #528 +/- ##
===========================================
- Coverage 89.84% 89.48% -0.37%
===========================================
Files 93 93
Lines 6098 6094 -4
===========================================
- Hits 5479 5453 -26
- Misses 619 641 +22
... and 5 files with indirect coverage changes Continue to review full report in Codecov by Sentry.
|
@@ -413,15 +409,7 @@ def sqrtm(self, tensor: tf.Tensor, dtype, rtol=1e-05, atol=1e-08) -> Tensor: | |||
# ~~~~~~~~~~~~~~~~~ | |||
|
|||
def DefaultEuclideanOptimizer(self) -> tf.keras.optimizers.legacy.Optimizer: | |||
use_legacy = Version(metadata.distribution("tensorflow").version) < Version("2.16.0") |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
i know we drop 2.15 in this PR, but do we happen to know if the Adam optimizer is still slow on MacOS? I can't find any relevant issues or documentation.
I'm ok to just say this will be slow on Mac users for the time being if this upgrade is fairly critical. I definitely like how much easier installation is.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Based on running the tests that work on >=2.15.0
(meaning actual optimization tests are skipped) they are roughly the same. 0:03:02
for 2.15.0
and 0:03:09
for 2.18.0
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
That is to say, preliminary profiling suggests they're similar but we really wouldn't know until the optimization itself is fixed. Considering tensorflow-macos
is no longer the suggested installation for macos I think it should be fine
Skips tests if the backend is tensorflow and the version is ``>2.15.0``. | ||
""" | ||
if math.backend_name == "tensorflow": | ||
if Version(metadata.distribution("tensorflow").version) > Version("2.15.0"): |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
it always is now, right?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Right now yes. I'm still debating on having 2.15.0 be the lowest but then you have to jump through some hoops to have tensorflow-macos
install when tf == 2.15.0
but not install if tf>2.15.0
.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
hmm. we haven't released since March - @elib20 is there a plan to make another official MrMustard version? I'm wondering how bad it would be to just say "if using MrMustard 7.4+, then you need TF 2.16+"
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think the next planned release is once lab dev is complete. Seeing as that would be a massive update from the previous release it's completely fine to mention in the CHANGELOG that we just bumped the version
User description
Context: Support for python 3.12. This includes bumping the tensorflow version.
Description of the Change: Updated the pyproject.toml to support python 3.12 and bumped the tensorflow version to min 2.16.0. Removed seemingly unnecessary dependencies. Removed macOS warning about switching to tensorflow 2.15.0.
Note:
test_training
andtest_opt
fail withtf > 2.15.0
. For now these tests are skipped and are to be fixed in the future.PR Type
enhancement, dependencies
Description
classifiers
andpython
version range inpyproject.toml
.Changes walkthrough 📝
pyproject.toml
Update Python and TensorFlow version support in pyproject.toml
pyproject.toml