The list is currently pretty long:
https://github.com/scikit-learn/scikit-learn/issues?q=is%3Aopen+is%3Aissue+milestone%3A0.17

I'd like to clean up the run of the test-suite. There are currently many 
deprecation warnings: 
https://github.com/scikit-learn/scikit-learn/issues/5089
We should remove the multi-core support on LDA if it usually makes the 
runtime longer: https://github.com/scikit-learn/scikit-learn/issues/5118
There are random segfaults in sparse PCA: 
https://github.com/scikit-learn/scikit-learn/issues/5013
We might want to put a warning into PLS if our results are still not 
consistent with reference implementations: 
https://github.com/scikit-learn/scikit-learn/issues/2821
We use 3.5 deprecated inspection: 
https://github.com/scikit-learn/scikit-learn/issues/5281
There seem to be bugs in SGD 
https://github.com/scikit-learn/scikit-learn/issues/5246
and Naive Bayes https://github.com/scikit-learn/scikit-learn/issues/5136
that could be easily fixed.

There are some near-ready PRs:
https://github.com/scikit-learn/scikit-learn/pulls?q=is%3Aopen+is%3Apr+milestone%3A0.17
The pipeline inverse_transform 1d thing should be deprecated: 
https://github.com/scikit-learn/scikit-learn/pull/5065 [this needs a 
two-line fix]
We should value error if someone gives floats to a classifier: 
https://github.com/scikit-learn/scikit-learn/pull/5084

I'll try to get to some of the easy fixes / bugs this week, but I'm 
still catching up on email.

Andy

On 09/21/2015 04:43 AM, Gilles Louppe wrote:
> Hi Olivier,
>
> It seems the 3 PRs you mentioned are now closed/merged. Are there
> other blocking PRs you need us to look at before freezing for the
> release?
>
> Cheers,
> Gilles
>
> On 4 September 2015 at 12:16, Olivier Grisel <olivier.gri...@ensta.org> wrote:
>> Hi all,
>>
>> It's been a while since we have not made a release. I plan to cut the
>> 0.17.X branch to prepare a first beta next week. Then 2 weeks after
>> that we can release either a new beta or the final 0.17.0 based on
>> feedback and if there is no identified blocker or major regression
>> from 0.16.1.
>>
>> I would like the following to get in for the release to make it easier
>> for people who experiment the multiprocessing crash under OSX (they
>> would just have to use Python 3.4 to get rid of the crash):
>>
>> https://github.com/scikit-learn/scikit-learn/pull/5199
>>
>> Andreas also raised that as 0.17 will be the first release to include
>> Latent Dirichlet Allocation, we should rename the sklearn.lda package
>> and models to emphasize they are about Linear Discriminant Analysis to
>> avoid confusion. This was attempted in
>> https://github.com/scikit-learn/scikit-learn/pull/4421 but work is
>> needed to rebase the renaming on the current state of master. I can do
>> that if nobody does it in the mean time.
>>
>> There are other maintenance PRs that I would like to see part of this
>> release such as:
>>
>> https://github.com/scikit-learn/scikit-learn/pull/5152
>>
>> Small bug fixes tagged with
>> https://github.com/scikit-learn/scikit-learn/milestones/0.17 can be
>> back-ported into the future 0.17.X branch to make it into the final
>> 0.17.0 release.
>>
>> Le me know if you have any comment on this release process.
>>
>> Best,
>>
>> --
>> Olivier
>> http://twitter.com/ogrisel - http://github.com/ogrisel
>>
>> ------------------------------------------------------------------------------
>> _______________________________________________
>> Scikit-learn-general mailing list
>> Scikit-learn-general@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
> ------------------------------------------------------------------------------
> _______________________________________________
> Scikit-learn-general mailing list
> Scikit-learn-general@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general


------------------------------------------------------------------------------
_______________________________________________
Scikit-learn-general mailing list
Scikit-learn-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general

Reply via email to