Skip to content

DOC cleanup the roadmap #15332

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 6 commits into from
Nov 6, 2019
Merged

DOC cleanup the roadmap #15332

merged 6 commits into from
Nov 6, 2019

Conversation

adrinjalali
Copy link
Member

This cleans up the roadmap a bit. Removes the links to issues and PRs which are already solved, and is an effort to keep it up to date.

I'm happy to put things which are deleted in this PR back. I've mostly deleted them for us to talk about them if you still think they should be there.

ping @scikit-learn/core-devs

Copy link
Member

@NicolasHug NicolasHug left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks Adrin

made a few minor comments

doc/roadmap.rst Outdated
* Learners directly handling missing data :issue:`13911`
* Making sure meta-estimators are lenient towards missing data,
:issue:`15319`
* Non-trivial imputers
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I guess we can even remove it now unless there are other open issues/PR

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The iterative imputer still has a blocker issue.

doc/roadmap.rst Outdated
* Making sure meta-estimators are lenient towards missing data,
:issue:`15319`
* Non-trivial imputers
* Learners directly handling missing data
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

remove? Unless we want to support it for the tree module as well?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

That's what I was thinking when I didn't remove it. I also couldn't think of any other estimator which could easily support this.

#. Improved tracking of fitting

* Verbose is not very friendly and should use a standard logging library
:issue:`6929`
:issue:`6929`, :issue:`78`
* Callbacks or a similar system would facilitate logging and early stopping

#. Distributed parallelism
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

need to skip line, rendering is wrong

* Joblib can now plug onto several backends, some of them can distribute the
computation across computers
* However, we want to stay high level in scikit-learn
* Accept data which complies with ``__array_function__``
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm confused, isn't #15332 doing the exact opposite (i.e. not rely on __array_function__)?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

You mean Andy's PR? The idea is for now actively not to accept it, instead of giving odd errors if the user passes them, and then work towards possibly supporting them later.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

lol yeah sorry I meant #14702

@TomDLT
Copy link
Member

TomDLT commented Oct 22, 2019

#10463 has been addressed

Copy link
Member

@jnothman jnothman left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

At some point I think we wanted the numeric indexing to be useful for referencing. This breaks that. I'm not sure I care.

Are we trying to remove items that are resolved? should we be using strikethrough?

@glemaitre
Copy link
Member

glemaitre commented Oct 23, 2019

should we be using strikethrough?

+1

@adrinjalali
Copy link
Member Author

I think I have applied the comments.

doc/roadmap.rst Outdated
@@ -124,12 +135,12 @@ bottom.

#. Better interfaces for interactive development

* __repr__ and HTML visualisations of estimators :issue:`6323`
* |ss| __repr__ and HTML visualisations of estimators :issue:`6323` |se|
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

we've done repr but not HTML yet...

@NicolasHug
Copy link
Member

I think we should mention that the cross-out entries are "done" rather than "not relevant anymore".

Also, not strong opinion but I find it hard to read full crossed-out sentences. I'd be happier with a section at the end where we put the currently achieved goals (each bullet point would need more context though). But again, no strong opinion. LGTM apart from the html thing mentioned by Joel

@adrinjalali
Copy link
Member Author

Is this good now?

Copy link
Member

@amueller amueller left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

looks good apart from nitpicks.

@jnothman jnothman merged commit d98caae into scikit-learn:master Nov 6, 2019
@jnothman
Copy link
Member

jnothman commented Nov 6, 2019

Thanks @adrinjalali

@adrinjalali adrinjalali deleted the roadmap branch November 6, 2019 11:47
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

7 participants