On 03/09/2015 10:44 PM, Joel Nothman wrote:
Congratulations! This has been a long time coming, and if not only for
the swathe of features it'll be great to see the documentation
improvements appearing on stable soon!
My thoughts on development priorities for the next release (and
ideally to focus on before GSoC eats everyone's brains):
We need to finalise some of last year's GSoC contributions.
Specifically, Issam (and perhaps more importantly ELM
<https://github.com/scikit-learn/scikit-learn/pull/3306> and MLP
<https://github.com/scikit-learn/scikit-learn/pull/3204>)
+1, especially for MLPs (I have been lazy on that front)
Even thought they are much more recent PRs, I think we should see that
the LDA, T-SNE and GP PRs get thorough and (comparably) swift PRs,
as they are awesome and are high quality contributions.
On the API front, I would really really love to see multiple metric
support in grid search working (after all, that issue was my first
contribution to scikit-learn), while reengineering cross-validation
for nesting also seems to be of interest and brings with it issues of
forwarding parameters like labels for LOLO and sample_weight in the
cross-validation/grid search apparatus.
+10
multiple metrics, data independent CV objects, and nesting EstimatorCV
sensibly in GridSearchCV are really important to me.
Unfortunately I'll be relatively indisposed over the next few months
due to movements in personal and work life. So thanks all for 0.16b
and I look forward to what 0.17 may bring!
You will be missed! Good luck with everything!
------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Scikit-learn-general mailing list
Scikit-learn-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
