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

Time for a new release - scipy 0.18? #1169

Closed
matthew-brett opened this Issue Jan 13, 2017 · 12 comments

Comments

Projects
None yet
4 participants
@matthew-brett
Member

matthew-brett commented Jan 13, 2017

I just set of a wheel build, and noticed that the current release of dipy errors out with the current release of scipy:

======================================================================
ERROR: Failure: ImportError (cannot import name nanmean)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/venv/local/lib/python2.7/site-packages/nose/loader.py", line 418, in loadTestsFromName
    addr.filename, addr.module)
  File "/venv/local/lib/python2.7/site-packages/nose/importer.py", line 47, in importFromPath
    return self.importFromDir(dir_path, fqname)
  File "/venv/local/lib/python2.7/site-packages/nose/importer.py", line 94, in importFromDir
    mod = load_module(part_fqname, fh, filename, desc)
  File "/venv/local/lib/python2.7/site-packages/dipy/reconst/sfm.py", line 25, in <module>
    from scipy.stats import nanmean
ImportError: cannot import name nanmean

https://travis-ci.org/MacPython/dipy-wheels/jobs/191741388#L413

Time for a new release?

@arokem

This comment has been minimized.

Member

arokem commented Jan 14, 2017

Past time for a release, for sure.

If we want to handle different versions, we could (should?) do what we did here:

#1166

@jchoude

This comment has been minimized.

Contributor

jchoude commented Feb 6, 2017

Also, the interpolation code in dipy.tracking.interpolation errors out when using Numpy 1.12, because the Deprecation Warning was finally changed to an Error (see https://docs.scipy.org/doc/numpy/release.html#deprecationwarning-to-error)

Do we have a buildbot using the 1.12 release yet?

@jchoude

This comment has been minimized.

Contributor

jchoude commented Feb 6, 2017

Also, my previous message could also be linked with #728

@arokem

This comment has been minimized.

@jchoude

This comment has been minimized.

Contributor

jchoude commented Feb 6, 2017

No, you would get an error like:

File "dipy/tracking/local/interpolation.pyx", line 116, in dipy.tracking.local.interpolation.nearestneighbor_interpolate (dipy/tracking/local/interpolation.c:2847)
    return data[index]

IndexError: only integers, slices (`:`), ellipsis (`...`), numpy.newaxis (`None`) and integer or boolean arrays are valid indices

Casting the index to an int fixes the issue. I'm testing the code right now.

@arokem

This comment has been minimized.

Member

arokem commented Feb 6, 2017

@jchoude

This comment has been minimized.

Contributor

jchoude commented Feb 6, 2017

I'm not sure if it is happening only on 2.7. It was reported to me by a end user. I don't have his exact setup to replicate, but on my tests on 2.7 it fixes the issue.

For my knowledge, what would cause the behavior to differ for Python >= 3?

@matthew-brett

This comment has been minimized.

Member

matthew-brett commented Feb 6, 2017

Sorry - I should have renamed them - but I believe all the 'py26' bots are in fact using Python 2.7.

@arokem

This comment has been minimized.

Member

arokem commented Feb 6, 2017

@jchoude : seems like I am wrong. I just confirmed that this would fail with that error message on both Python 2.7 and 3.5.

@matthew-brett : thanks!

@fmorency

This comment has been minimized.

fmorency commented May 16, 2017

+1

@arokem

This comment has been minimized.

Member

arokem commented May 16, 2017

Yeah - we're working on it. I think that all we have left for the release are the following:

#1245
#1241
#1243
#1164
#1156

I think we should postpone merging the LPCA PR (#1233) until after the release, given that it looks like there might be more work to do there.

@arokem

This comment has been minimized.

Member

arokem commented Jun 27, 2017

🎉

@arokem arokem closed this Jun 27, 2017

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment