spark-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Nick Pentreath (JIRA)" <>
Subject [jira] [Commented] (SPARK-18813) MLlib 2.2 Roadmap
Date Thu, 23 Feb 2017 13:02:44 GMT


Nick Pentreath commented on SPARK-18813:

Thanks for this Joseph and everyone for the comments & discussion (I also responded briefly
on @dev list).

>From my perspective, I think the highest priorities are:

# *ML / MLlib Parity* - this is sort of a weight around our necks currently, and once it is
finally done it can free up bandwidth and mind space for how to take Spark ML forward. Until
it is complete, I feel that almost everything else that is not a critical bug fix or missing
feature will not (and probably should not) receive much attention;
# *Python / R / Scala parity* - I think this is obvious
# *Persistence completion & backward compat* - again obvious

In terms of the "new features & improvements" stuff, I tend to agree with some of the
items mentioned by Yanbo:

# *Improve scalability & usability of feature transformers* - also fixing any big issues
here (e.g. and OHE issue such as
# *Set initial model* - useful and a lot of the initial work in the PR has been done so actually
not that much blocking this going in
# *GBT improvements* - I think it is safe to say that GBT/GBM is a core algorithm and we should
aim to be able to provide comparable functionality to scikit-learn, XGBoost & LightGBM
in terms of scalability and accuracy. Even if this means re-writing much of the implementation!

FYI, I don't really believe the local model stuff is high priority - not because it is not
critically important and often requested, but because it is such a complex thing, and Spark
ML is so tightly integrated with Spark SQL runtime currently, that I think we'd need to come
up with a pretty detailed plan here for it to make its way anywhere near the medium-term roadmap...
Added to this the community is moving ahead with their own solutions fairly rapidly, so that
perhaps it will not be as important in the near future (e.g. MLeap and other serving frameworks,
standards such as PMML & PFA, etc).

Apart from a few JIRAs / PRs that were around before that I need to re-shepherd, I'll be trying
to focus on the parity stuff for remainder of {{2.2}} dev cycle - for my own outstanding ALS
PR and others. I'll work through the major items and either shepherd or help shepherd them
over the line.

> MLlib 2.2 Roadmap
> -----------------
>                 Key: SPARK-18813
>                 URL:
>             Project: Spark
>          Issue Type: Umbrella
>          Components: ML, MLlib
>            Reporter: Joseph K. Bradley
>            Priority: Blocker
>              Labels: roadmap
> *PROPOSAL: This includes a proposal for the 2.2 roadmap process for MLlib.*
> The roadmap process described below is significantly updated since the 2.1 roadmap [SPARK-15581].
 Please refer to [SPARK-15581] for more discussion on the basis for this proposal, and comment
in this JIRA if you have suggestions for improvements.
> h1. Roadmap process
> This roadmap is a master list for MLlib improvements we are working on during this release.
 This includes ML-related changes in PySpark and SparkR.
> *What is planned for the next release?*
> * This roadmap lists issues which at least one Committer has prioritized.  See details
below in "Instructions for committers."
> * This roadmap only lists larger or more critical issues.
> *How can contributors influence this roadmap?*
> * If you believe an issue should be in this roadmap, please discuss the issue on JIRA
and/or the dev mailing list.  Make sure to ping Committers since at least one must agree to
shepherd the issue.
> * For general discussions, use this JIRA or the dev mailing list.  For specific issues,
please comment on those issues or the mailing list.
> * Vote for & watch issues which are important to you.
> ** MLlib, sorted by: [Votes |]
or [Watchers |]
> ** SparkR, sorted by: [Votes |]
or [Watchers |]
> h2. Target Version and Priority
> This section describes the meaning of Target Version and Priority.  _These meanings have
been updated in this proposal for the 2.2 process._
> || Category | Target Version | Priority | Shepherd | Put on roadmap? | In next release?
> | [1 |]
| next release | Blocker | *must* | *must* | *must* |
> | [2 |]
| next release | Critical | *must* | yes, unless small | *best effort* |
> | [3 |]
| next release | Major | *must* | optional | *best effort* |
> | [4 |]
| next release | Minor | optional | no | maybe |
> | [5 |]
| next release | Trivial | optional | no | maybe |
> | [6 |"In%20Progress"%2C%20Reopened)%20AND%20component%20in%20(GraphX%2C%20ML%2C%20MLlib%2C%20SparkR)%20AND%20"Target%20Version%2Fs"%20in%20(EMPTY)%20AND%20Shepherd%20not%20in%20(EMPTY)%20ORDER%20BY%20priority%20DESC]
| (empty) | (any) | yes | no | maybe |
> | [7 |]
| (empty) | (any) | no | no | maybe |
> The *Category* in the table above has the following meaning:
> 1. A committer has promised to see this issue to completion for the next release.  Contributions
*will* receive attention.
> 2-3. A committer has promised to see this issue to completion for the next release. 
Contributions *will* receive attention.  The issue may slip to the next release if development
is slower than expected.
> 4-5. A committer has promised interest in this issue.  Contributions *will* receive attention.
 The issue may slip to another release.
> 6. A committer has promised interest in this issue and should respond, but no promises
are made about priorities or releases.
> 7. This issue is open for discussion, but it needs a committer to promise interest to
> h1. Instructions
> h2. For contributors
> Getting started
> * Please read carefully. Code style, documentation,
and unit tests are important.
> * If you are a first-time contributor, please always start with a small [starter task|]
rather than a larger feature.
> Coordinating on JIRA
> * Never work silently. Let everyone know on the corresponding JIRA page when you start
work. This is to avoid duplicate work. For small patches, you do not need to get the JIRA
assigned to you to begin work.
> * For medium/large features or features with dependencies, please get assigned first
before coding and keep the ETA updated on the JIRA. If there is no activity on the JIRA page
for a certain amount of time, the JIRA should be released for other contributors.
> * Do not claim multiple (>3) JIRAs at the same time. Try to finish them one after
> * Do not set these fields: Target Version, Fix Version, or Shepherd.  Only Committers
should set those.
> Writing and reviewing PRs
> * Remember to add the `@Since("VERSION")` annotation to new public APIs.
> * *Please review others' PRs ( Code review greatly
helps to improve others' code as well as yours.*
> h2. For Committers
> Adding to this roadmap
> * You can update the roadmap by (a) adding issues to this list and (b) setting Target
Versions.  Only Committers may make these changes.
> * *If you add an issue to this roadmap or set a Target Version, you _must_ assign yourself
or another Committer as Shepherd.*
> * This list should be actively managed during the release.
> * If you target a significant item for the next release, please list the item on this
> * If you commit to shepherding a new public API, you implicitly commit to shepherding
the follow-up issues as well (Python/R APIs, docs).
> Creating JIRA issues
> * Try to break down big features into small and specific JIRA tasks and link them properly.
> * Add a "starter" label to starter tasks.
> * Put a rough time estimate for medium/big features and track the progress.
> * Set Priority carefully.  Priority should not be mixed with size of effort for implementation.
> Managing JIRA issues and PRs
> * Please add yourself to the Shepherd field on JIRA if you start reviewing a PR.
> * If the code looks good to you, please comment "LGTM". For non-trivial PRs, please ping
a Committer experienced with the relevant code to make a final pass.
> Follow-up issues: *After merging a PR, create and link the necessary follow-up JIRAs.*
> * For a new Scala/Java API
> ** Create issues for adding analogous Python and R APIs
> ** Create issues for adding example code and documentation
> * For a new Python/R API
> ** Create issues for adding example code and documentation
> h1. Roadmap for this release
> This roadmap only includes larger, more critical tasks targeted at the next release.
 To find all issues targeted for the next release, use the links listed below.
> Notes
> * We will prioritize API parity, bug fixes, and improvements over new features.
> * The RDD-based API (`spark.mllib`) is in maintenance mode now.  We will accept bug fixes
for it, but new features, APIs, and improvements will only be added to the DataFrame-based
API (``).
> *WIP: This section is still being updated, pending confirmation of the Roadmap Process
described above.*
> h2. Critical feature parity in DataFrame-based API
> * [SPARK-14501]: Frequent Pattern Mining
> * [SPARK-14709]: linear SVM
> * [SPARK-15784]: Power Iteration Clustering (PIC)
> The umbrella JIRA for feature parity is [SPARK-4591], though it includes items not targeted
for 2.2.
> h2. Persistence
> * Complete persistence within MLlib
> ** Python tuning (SPARK-13786)
> * MLlib in R format: compatibility with other languages (SPARK-15572)
> * Impose backwards compatibility for persistence (SPARK-15573)
> h2. SparkR
> * Release SparkR on CRAN [SPARK-15799]
> h2. Other prioritized issues: links for searching JIRA
> This section provides links to help people identify smaller patches targeted at the next
release, as well as patches for major areas within MLlib.
> * [All MLlib, SparkR, GraphX JIRAs with Target Version 2.2 |!%3D%202.2.0)%20ORDER%20BY%20priority]
> * [MLlib, SparkR, GraphX Umbrella JIRAs (regardless of Target Version) |]
> * [MLlib, SparkR, GraphX bugs |"In%20Progress"%2C%20Reopened)%20AND%20component%20in%20(GraphX%2C%20ML%2C%20MLlib%2C%20SparkR)%20ORDER%20BY%20priority%20DESC]
> h1. Long-term roadmap
> This section lists long-term or constant efforts.  For example, Python/R API parity with
Scala/Java will always be a priority, but we do not promise exact parity with each release.
> h2. Python and R feature parity
> Python feature parity: The main goal of the Python API is to have feature parity with
the Scala/Java API. You can find a [complete list of Python MLlib issues targeted at the next
release here|"In%20Progress"%2C%20Reopened)%20AND%20component%20in%20(ML%2C%20MLlib)%20AND%20component%20in%20(PySpark)%20AND%20"Target%20Version%2Fs"%20%3D%202.2.0%20ORDER%20BY%20priority%20DESC].
> R feature parity: We are building towards feature parity in SparkR as well. You can find
a [complete list of SparkR MLlib issues targeted at the next release here|"In%20Progress"%2C%20Reopened)%20AND%20component%20in%20(ML%2C%20MLlib)%20AND%20component%20in%20(SparkR)%20AND%20"Target%20Version%2Fs"%20%3D%202.2.0%20ORDER%20BY%20priority%20DESC].

This message was sent by Atlassian JIRA

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message