Create interpolation_methods.py

[RutgerK]

Added an example as requested at:
http://stackoverflow.com/questions/14722540/smoothing-between-pixels-of-imagesc-imshow-in-matlab-like-the-matplotlib-imshow/14728122#comment29802965_14728122

[mdboom]

Thanks for a contribution! A few comments. The examples/api directory is going away long term -- this should instead be in one of the more specific directories, probably examples/images_contours_and_fields.

This file should be PEP8 clean. (Use the pep8 tool to find violations).

[RutgerK]

@mdboom, thanks for pointing that out, i didnt know about the pep8 tool. I have used 'pep8 1.4.6' and it's (should be) clean now.

edit: After the changes the Travis build fails for Python 3.3, the others are fine.

ERROR: While executing gem ... (Gem::RemoteFetcher::FetchError)
Errno::ETIMEDOUT: Connection timed out - connect(2) (https://rubygems.org/gems/formatador-0.2.4.gem)
The command "gem install travis-artifacts" failed and exited with 1 during before_script.

Your build has been stopped.

[pelson]

Triggered a re-build. Hopefully that will do the trick.

[pelson]

How about for ax, interp_method in zip(axes, methods):?

[pelson]

LGTM 👍 after a possible improvement to the iteration. This should certainly make it into v1.4.x.

[pelson]

@RutgerK - I'm keen to merge this, any chance you would be willing to update the example re my comment?

[RutgerK]

@pelson, sorry, i misunderstood how Github works and thought that you already applied the change. I saw the email and thought it was fine so i didn't respond right away. I hope its does the trick...

[pelson]

Think this needs to be removed, and the lines 31 and 32 use the ax variable from the loop.

[pelson]

sorry, i misunderstood how Github works and thought that you already applied the change

No probs. Welcome to github & matplotlib!

[RutgerK]

@pelson, you're right. But the axes array needs to be flattened somehow, otherwise the zipping doesn't work for me with Python 2.7 & mpl 1.3, see:
http://nbviewer.ipython.org/gist/RutgerK/8420591

We could use either axes.flat or axes.ravel(), and do it outside or inside the zip. I have no preference, although inside does seem a bit obscure.

[pelson]

Squeeze might do the trick: http://matplotlib.org/api/pyplot_api.html#matplotlib.pyplot.subplots

[RutgerK]

Sqeeuze is True by default, it would only affect when the array of axes is Mx1 or 1xN, which does not apply in this case.

Could be a nice addition though, a sort of super-squeeze which forces a 1D array as results. Although for most cases axes.flat isn't a problem.