Skip to content
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.

Sign up for GitHub

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

Jump to bottom

DOC: Improvements and additions to the matrix_normal doc #5517

Merged
merged 1 commit into from Nov 22, 2015
Merged

DOC: Improvements and additions to the matrix_normal doc #5517

merged 1 commit into from Nov 22, 2015

Conversation

drpeteb
Copy link
Contributor

@drpeteb drpeteb commented Nov 22, 2015

For #5478

  • Documented the relationship with the multivariate normal distribution.
  • Some small clarifications made and typos fixed.

@drpeteb drpeteb mentioned this pull request Nov 22, 2015
Closed
@drpeteb
Copy link
Contributor Author

drpeteb commented Nov 22, 2015

Dammit. I thought I'd rebased those first three commits away. Is this messy history a problem?

ev-br
ev-br reviewed Nov 22, 2015
View reviewed changes
scipy/stats/_multivariate.py
`multivariate_normal` distribution. Specifically, :math:`\mathrm{Vec}(X)`
(the vectorisation of :math:`X`) has a multivariate normal distribution
with mean :math:`\mathrm{Vec}(M)` and covariance :math:`V \otimes U`
(where :math:`\otimes` is the Kronecker product).
Copy link
Member

Choose a reason for hiding this comment

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

Having this paragraph is better then not having it, but could you reword it a bit. Specifically, "vectorization" is heavily used for something different in numpy land (whether it's -s- or -z- :-). Would also be good to mention that this is equivalent but algorithmically inefficient.

@ev-br
Copy link
Member

ev-br commented Nov 22, 2015

LGTM modulo a nitpick.
Extra commits would be good to squash (git rebase -I HEAD~4), but we could do it if you are not easy with it.

@ev-br ev-br added scipy.stats Documentation Issues related to the SciPy documentation. Also check https://github.com/scipy/scipy.org labels Nov 22, 2015
@drpeteb
ENH: Added a method to get the equivalent multivariate normal
9be75a0 
Doc fix

Fixed a silly docstring error

Removed equivalent_multivariate

DOC: Improved documentation of matrix_normal.

Slight rewording. Remove use of 'vectorisation'.
@ev-br ev-br added this to the 0.17.0 milestone Nov 22, 2015
argriffing added a commit that referenced this pull request Nov 22, 2015
@argriffing
Merge pull request #5517 from drpeteb/enh-matrix-normal-doc
8c724a9 
DOC: Improvements and additions to the matrix_normal doc
@argriffing argriffing merged commit 8c724a9 into scipy:master Nov 22, 2015
@argriffing
Copy link
Contributor

Thanks @drpeteb!

BTW I checked out this branch locally to run $ python runtests.py --doctests -s stats but this has 94 test errors and 33 test failures and popped up some plot windows. I assume none of that is due to this PR, but rather that scipy docstring examples are not really interpretable as doctests.

@ev-br
Copy link
Member

ev-br commented Nov 22, 2015

@argriffing you probably want --refguide-check not --doctests.

@argriffing
Copy link
Contributor

Is --doctests still a thing? Should it be removed from runtests.py?

@ev-br
Copy link
Member

ev-br commented Nov 22, 2015

My experience with --doctests matches what you wrote above, so maybe we're better off aliasing it to --refguide-check ( which does run doctests with a modified checker).

@argriffing
Copy link
Contributor

With --refguide-check I still get a handful of errors, mostly involving passing floats when integers are expected. I assume this is because I am using a newish numpy version that must not be tested with scipy on TravisCI? Also there is a SuperLU error.

@ev-br
Copy link
Member

ev-br commented Nov 22, 2015

I still get a handful of errors,

Would be good to fix I guess (unless these are numpy 1.10.1 regressions which are going to be fixed in 1.10.2?)

@argriffing
Copy link
Contributor

maybe we're better off aliasing it to --refguide-check

I looked at this... maybe it would make sense if refguide_check.py gets an --only-doctests flag or something. Or maybe if it's renamed to doc_check.py with --refguide, --doctests, and --tutorials flags.

@ev-br ev-br mentioned this pull request Dec 19, 2015
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
Documentation Issues related to the SciPy documentation. Also check https://github.com/scipy/scipy.org scipy.stats
Projects
None yet
Milestone
0.17.0
Development

Successfully merging this pull request may close these issues.

None yet
3 participants
@drpeteb @ev-br @argriffing