On Mon, Oct 1, 2012 at 5:35 PM, Lars Buitinck wrote:
> 2012/10/1 Rob Zinkov :
> > It might be possible to use some networkx fileformat, since I know
> network
> > exports to d3.
>
> Doesn't networkx have converters to/from scipy.sparse?
>
yes it does. http://networkx.lanl.gov/reference/convert.h
2012/10/1 Rob Zinkov :
> It might be possible to use some networkx fileformat, since I know network
> exports to d3.
Doesn't networkx have converters to/from scipy.sparse?
--
Lars Buitinck
Scientific programmer, ILPS
University of Amsterdam
--
It might be possible to use some networkx fileformat, since I know network
exports to d3.
On Oct 1, 2012 12:30 PM, "Peter Prettenhofer"
wrote:
> Hi Andy,
>
> the problem is: when it comes to visualization I'm a cargo-cult
> programmer [1] - I basically took the Tree Layout example from the D3
> g
Hi Andy,
the problem is: when it comes to visualization I'm a cargo-cult
programmer [1] - I basically took the Tree Layout example from the D3
gallery, created the JSON representation along the lines of the
example and added some minor modifications.
> How much does the JSON differ from the dot f
Hey Peter.
That is pretty cool! Thanks for sharing!
I think it would be great if we could make this more accessible.
How much does the JSON differ from the dot file?
Was it not possible to read the dot with D3? Or did you just add extra
information?
Cheers,
Andy
---
+1
On Oct 1, 2012 11:49 AM, "Peter Prettenhofer"
wrote:
> Hi all,
>
> I recently wrote a decision tree viewer in D3 [1] that allows
> interactive exploring of decision trees - even though its quite a hack
> (JS is my nemesis) I thought it may be of use for other people as
> well. You can find the
Hi all,
I recently wrote a decision tree viewer in D3 [1] that allows
interactive exploring of decision trees - even though its quite a hack
(JS is my nemesis) I thought it may be of use for other people as
well. You can find the source code in this gist [2] and a demo for
iris and boston here [3]
