Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Deprecate support for (n, 1)-shaped error arrays in errorbar(). #13209

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
Merged
merged 1 commit into from
Feb 23, 2019
Merged

Deprecate support for (n, 1)-shaped error arrays in errorbar(). #13209

merged 1 commit into from
Feb 23, 2019

Conversation

anntzer
Copy link
Contributor

@anntzer anntzer commented Jan 17, 2019

errorbar() documents that it supports the following shapes for error
arrays:

- scalar: Symmetric +/- values for all data points.
- shape(N,): Symmetric +/-values for each data point.
- shape(2,N): Separate - and + values for each bar. First row
    contains the lower errors, the second row contains the
    upper errors.
- *None*: No errorbar.

Actually it also supports (N, 1)-shaped arrays (treating them as shape
(N,)), but this support is broken for N=2 at least since Matplotlib
1.5:

N = 2; errorbar(range(N), range(N), np.arange(N).reshape((N, 1)))
...
ValueError: In safezip, len(args[0])=2 but len(args[1])=1

(This works for other values of N.)

Instead of further maintaining code to handle that case, just deprecate
it (no one has apparently noticed the N=2 bug, suggesting that this
(mis)feature is not much used anyways).

PR Summary

PR Checklist

  • Has Pytest style unit tests
  • Code is Flake 8 compliant
  • New features are documented, with examples if plot related
  • Documentation is sphinx and numpydoc compliant
  • Added an entry to doc/users/next_whats_new/ if major new feature (follow instructions in README.rst there)
  • Documented in doc/api/api_changes.rst if API changed in a backward-incompatible way

@anntzer
Deprecate support for (n, 1)-shaped error arrays in errorbar().
4fc30a1 
errorbar() documents that it supports the following shapes for error
arrays:

    - scalar: Symmetric +/- values for all data points.
    - shape(N,): Symmetric +/-values for each data point.
    - shape(2,N): Separate - and + values for each bar. First row
        contains the lower errors, the second row contains the
        upper errors.
    - *None*: No errorbar.

Actually it also supports (N, 1)-shaped arrays (treating them as shape
(N,)), but this support is broken for N=2 *at least* since Matplotlib
1.5:

    N = 2; errorbar(range(N), range(N), np.arange(N).reshape((N, 1)))
    ...
    ValueError: In safezip, len(args[0])=2 but len(args[1])=1

(This works for other values of N.)

Instead of further maintaining code to handle that case, just deprecate
it (no one has apparently noticed the N=2 bug, suggesting that this
(mis)feature is not much used anyways).
@QuLogic QuLogic added this to the v3.1 milestone Jan 19, 2019
@anntzer anntzer mentioned this pull request Jan 25, 2019
Closed
@jklymak
Copy link
Member

jklymak commented Feb 3, 2019

I think this is OK. Not sure though if there is a community that has Nx1 matrices. I probably did for the first year or so after switching from Matlab...

@anntzer
Copy link
Contributor Author

anntzer commented Feb 3, 2019

And then you saw the light? :p
(My experience with dealing with MATLAB code (still on a regular basis) is that half of the people use rows and the other half use columns, so you never know (well plenty of MATLAB functions "correctly" infer the intent by checking for dimensions equal to 1, but still :/).)

@jklymak
Copy link
Member

jklymak commented Feb 3, 2019

Well I am now used to the numpy way, but when I took matrix algebra there was a difference between an Nx1 and a 1xN vector that is less explicit in numpy.

@anntzer
Copy link
Contributor Author

anntzer commented Feb 3, 2019

In linear algebra, yes, but most often the data we're plotting is not a vector in a linear algebra sense (you're not intending to matrix-multiply it by some matrix, for example); they're just... a list (in the non-technical sense) of data points.
Also, MATLAB will happily plot both 1xn and nx1 data (including one against the other) so if anything it's MATLAB that's sloppier there...

@tacaswell tacaswell merged commit 5ff12ef into matplotlib:master Feb 23, 2019
meeseeksmachine pushed a commit to meeseeksmachine/matplotlib that referenced this pull request Feb 23, 2019
@tacaswell
Backport PR matplotlib#13209: Deprecate support for (n, 1)-shaped err…
5dc0474 
…or arrays in errorbar().
@meeseeksmachine meeseeksmachine mentioned this pull request Feb 23, 2019
Merged
QuLogic added a commit that referenced this pull request Feb 23, 2019
@QuLogic
Merge pull request #13501 from meeseeksmachine/auto-backport-of-pr-13…
d966e54 
…209-on-v3.1.x

Backport PR #13209 on branch v3.1.x (Deprecate support for (n, 1)-shaped error arrays in errorbar().)
@anntzer anntzer deleted the errorbar branch February 23, 2019 21:10
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@timhoffm timhoffm timhoffm approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
v3.1.0
Development

Successfully merging this pull request may close these issues.
5 participants
@anntzer @jklymak @timhoffm @tacaswell @QuLogic