Skip to content

Switch backend to PyTensor #6365

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

Merged
merged 3 commits into from
Dec 5, 2022
Merged

Switch backend to PyTensor #6365

merged 3 commits into from
Dec 5, 2022

Conversation

michaelosthege
Copy link
Member

@michaelosthege michaelosthege commented Dec 3, 2022

What is this PR about?
Swapping the backend from Aesara to PyTensor.

Major / Breaking Changes

  • Backend changed from Aesara to PyTensor

Bugfixes / New features

  • None

Docs / Maintenance

  • None

@michaelosthege michaelosthege added the major Include in major changes release notes section label Dec 3, 2022
@review-notebook-app
Copy link

Check out this pull request on  ReviewNB

See visual diffs & provide feedback on Jupyter Notebooks.


Powered by ReviewNB

@michaelosthege michaelosthege added this to the v5.0.0 milestone Dec 3, 2022
@codecov
Copy link

codecov bot commented Dec 4, 2022

Codecov Report

Merging #6365 (f76f552) into main (5688555) will increase coverage by 0.00%.
The diff coverage is 97.80%.

Additional details and impacted files

Impacted file tree graph

@@           Coverage Diff           @@
##             main    #6365   +/-   ##
=======================================
  Coverage   94.72%   94.72%           
=======================================
  Files         132      132           
  Lines       26740    26741    +1     
=======================================
+ Hits        25330    25331    +1     
  Misses       1410     1410           
Impacted Files Coverage Δ
pymc/variational/inference.py 84.97% <ø> (ø)
pymc/step_methods/hmc/quadpotential.py 80.62% <75.00%> (ø)
pymc/math.py 70.04% <82.35%> (ø)
pymc/variational/approximations.py 87.67% <86.66%> (ø)
pymc/variational/opvi.py 87.08% <89.18%> (ø)
pymc/logprob/tensor.py 87.03% <90.90%> (ø)
pymc/tests/logprob/utils.py 54.87% <91.66%> (ø)
pymc/pytensorf.py 94.63% <94.73%> (ø)
pymc/tests/logprob/test_scan.py 93.51% <94.73%> (ø)
pymc/variational/updates.py 92.11% <95.00%> (ø)
... and 96 more

@michaelosthege michaelosthege marked this pull request as ready for review December 4, 2022 14:46
@michaelosthege michaelosthege changed the title Use pytensor Use PyTensor Dec 4, 2022
@michaelosthege michaelosthege requested a review from a team December 4, 2022 14:47
@ricardoV94 ricardoV94 requested a review from OriolAbril December 4, 2022 16:17
@michaelosthege michaelosthege changed the title Use PyTensor Switch backend to PyTensor Dec 4, 2022
@ricardoV94 ricardoV94 merged commit 5598298 into main Dec 5, 2022
@michaelosthege michaelosthege deleted the use-pytensor branch December 5, 2022 11:51
Copy link
Member

@OriolAbril OriolAbril left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

everything looks good, this transition is easier (will also be for notebooks that already run on aesara as its a rename only unlike v3-v4 where there were multiple breaking changes and improved features

@@ -74,7 +74,7 @@ Important modules to note are
a random variable distribution from a likelihood distribution.

* `logprob.py`: This contains the log probability logic for the distributions themselves.
The log probability calculation is deferred to Aesara
The log probability calculation is deferred to PyTensor
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think this is at least partially incorrect as "non-basic" distributions have their PPLs in pymc (logprob module)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
major Include in major changes release notes section
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants