hi kc,
it's not in scikit learn but we use these quite routinely alongside
scikit-learn.
https://github.com/scikit-learn/scikit-learn/pull/2730
here is also a set a notebook showing manifold extraction using diffusion
embedding. the notebook is a little out of date with respect to the code.
it also contains a basic procrustean alignment as well.
http://nbviewer.ipython.org/urls/dl.dropbox.com/s/qzavpq8pchbh97o/DiffusionMaps-Comparison.ipynb?dl=0
i think the notebook is quite convincing how the diffusion embedding
methods are superior to the other scikit-learn manifold embedders at least
for trivial cases.
for full inclusion, besides the powers that be being convinced that this is
a relatively small change and that it's performance outweighs the other
manifold learning methods, the PR definitely needs to be cleaned up,
together with full support for sparse matrices. thus far i have been happy
to just keep the PR and use the code for my needs.
cheers,
satra
On Tue, Sep 16, 2014 at 8:27 PM, <kevn...@phutureshock.com> wrote:
> Has anyone worked on the problem of manifold alignment?
>
> http://en.wikipedia.org/wiki/Manifold_alignment
>
> as described in papers like:
>
> "Manifold Alignment without Correspondence"
> http://ijcai.org/papers09/Papers/IJCAI09-214.pdf
>
> or
>
> "Data Fusion and Multi-Cue Data Matching by Diffusion Maps"
> http://people.ee.duke.edu/~lcarin/data_fusion_by_diffusion_maps.pdf
>
> or in CH. 5 of "Manifold Learning Theory and Applications"
>
> If not, is anyone interested in working on this? I would be happy to help.
>
> Best,
> kc
>
>
>
>
> ------------------------------------------------------------------------------
> Want excitement?
> Manually upgrade your production database.
> When you want reliability, choose Perforce
> Perforce version control. Predictably reliable.
>
> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
> _______________________________________________
> Scikit-learn-general mailing list
> Scikit-learn-general@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
>
------------------------------------------------------------------------------
Want excitement?
Manually upgrade your production database.
When you want reliability, choose Perforce
Perforce version control. Predictably reliable.
http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
_______________________________________________
Scikit-learn-general mailing list
Scikit-learn-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
