Closed wreise closed 4 years ago
@wreise please, if you have time, could you take this opportunity to look for a fix for #408?
@wreise I noticed the following concerning jupyter notebooks: in the classifying shapes and persistence homology of graphs notebooks, a gif is linked by means of the full URL https://raw.githubusercontent.com/giotto-ai/giotto-tda/master/examples/images/vietoris_rips_point_cloud.gif. This does not seem to display in GitHub (though URLs sending to Medium did in the past!). On the other hand, in #458 @lewtun uses the relative path images/time-delay-embedding.gif
and this displays nicely in GitHub.
Ultimately, the final approach must be such that these GIFs are displayed in the API reference pages, but it would be a bonus if we can also display them in GitHub.
@wreise please, if you have time, could you take this opportunity to look for a fix for #408?
I don't know what helped, but, locally, i do not face the same issue anymore.
@wreise I noticed the following concerning jupyter notebooks: in the classifying shapes and persistence homology of graphs notebooks, a gif is linked by means of the full URL https://raw.githubusercontent.com/giotto-ai/giotto-tda/master/examples/images/vietoris_rips_point_cloud.gif. This does not seem to display in GitHub (though URLs sending to Medium did in the past!). On the other hand, in #458 @lewtun uses the relative path
images/time-delay-embedding.gif
and this displays nicely in GitHub.
I changed the urls, let's see if it works. On the other hand, i had to name the messages explicitly.
Thanks @wreise!
On the other hand, i had to name the messages explicitly.
This is not great because those messages are displayed (remember "SegmentLocal"?). Why did it seem necessary?
This is not great because those messages are displayed (remember "SegmentLocal"?). Why did it seem necessary?
How could i have missed that :facepalm:. If you leave them empty, they are referenced automatically image1
, image2
etc. While it should have been fixed in a previous release of sphinx, my experience was that the counter k
in imagek
restarts at 0, in a new notebook cell, and, for example, every image1
(first image in a cell) becomes replaced with the last image1
from the notebook.
Indeed, it looks ugly. What would you suggest?
What happens if you put a single whitespace? I.e. ![ ](URL)
instead of [](URL)
?
What happens if you put a single whitespace? I.e.
![ ](URL)
instead of[](URL)
?
it doesn't work, no (even with a variable number of spaces)
Reference issues/PRs
Types of changes
Description Fixes biblio and rst-references problems encountered when compiling docs. Fixes #408.
To do:
Screenshots (if appropriate)
Any other comments?
Checklist
flake8
to check my Python changes.pytest
to check this on Python tests.