Skip to main content

Inkscape + matplotlib's svg = one strange love child

Heh, heh. Just when my analysis was coming along at a nice clip...

Here is part of figure1 (png shot of)



Here is part of fig2


And here is what happens when I copy fig2 and paste it into fig1


Note that the above figure should look exactly like fig2...
I LOVE Inkscape, I LOVE matplotlib, and I'm attached to my toolchain, I hope this can be fixed.

UPDATE:
  1. See this thread
  2. Mike has fixed the problem is SVN - you should replace the backend_svg.py with the one linked in Mike's post.

Comments

  1. Got a reply from Mike on the matplotlib-users mailing list:


    When matplotlib outputs an SVG, each unique character is assigned a numeric id (these are just assigned in order), and inserted as a "def", and then "use"d (referenced) wherever they are used.

    When you paste on SVG into another, those names clash, and Inkscape is pulling in the wrong characters when it goes to draw.

    Now, my gut feels that this is actually a bug in Inkscape -- pasting of referenced objects from one file into another should reassign new ids. However, I'm not an SVG expert, so I could be wrong, but I think I will take this question over to the Inkscape mailing list anyway.

    That said, there are probably some workarounds that matplotlib could make (using a hash of the character's content as the id, for instance). I'll look into that and reply when progress has been made.

    Mike

    ReplyDelete
  2. SVG is a great basis for cooperation between designers and coders (of for example web apps). Now SVG is becoming more mainstream, one of the advantages of being an open standard becomes more visible: easier cooperation. One of the 'disadvantages' however is that certain implied constraints ON the content, or allowed errors IN the content for one program could easily give trouble in the next program. Number of eyeballs with network effect ++. And this is 'only' a visual example of such a thing. If you also care about semantics or code readability, there's more to gain from SVG, plus more for SVG programs to actually make you enjoy it fully.

    ReplyDelete
  3. Save the figures as PDF and open them in Inkscape. The text is saved as real text, rather than as paths.

    Unfortunately, the image is not structured as nicely with groups anymore...

    ReplyDelete

Post a Comment

Popular posts from this blog

Python: Multiprocessing: passing multiple arguments to a function

Write a wrapper function to unpack the arguments before calling the real function. Lambda won't work, for some strange un-Pythonic reason.


import multiprocessing as mp def myfun(a,b): print a + b def mf_wrap(args): return myfun(*args) p = mp.Pool(4) fl = [(a,b) for a in range(3) for b in range(2)] #mf_wrap = lambda args: myfun(*args) -> this sucker, though more pythonic and compact, won't work p.map(mf_wrap, fl)

Flowing text in inkscape (Poster making)

You can flow text into arbitrary shapes in inkscape. (From a hint here).

You simply create a text box, type your text into it, create a frame with some drawing tool, select both the text box and the frame (click and shift) and then go to text->flow into frame.

UPDATE:

The omnipresent anonymous asked:
Trying to enter sentence so that text forms the number three...any ideas?
The solution:
Type '3' using the text toolConvert to path using object->pathSize as necessaryRemove fillUngroupType in actual text in new text boxSelect the text and the '3' pathFlow the text

Calculating confidence intervals: straight Python is as good as scipy.stats.scoreatpercentile

UPDATE:
I would say the most efficient AND readable way of working out confidence intervals from bootstraps is:

numpy.percentile(r,[2.5,50,97.5],axis=1)

Where r is a n x b array where n are different runs (e.g different data sets) and b are the individual bootstraps within a run. This code returns the 95% CIs as three numpy arrays.


Confidence intervals can be computed by bootstrapping the calculation of a descriptive statistic and then finding the appropriate percentiles of the data. I saw that scipy.stats has a built in percentile function and assumed that it would work really fast because (presumably) the code is in C. I was using a simple minded Python/Numpy implementation by first sorting and then picking the appropriate percentile data. I thought this was going to be inefficient timewise and decided that using scipy.stats.scoreatpercentile was going to be blazing fast because
It was native C It was vectorized - I could compute the CIs for multiple bootstrap runs at the same time …