This was extracted (@ 2024-04-17 22:10) from a list of minutes
which have been approved by the Board.
Please Note
The Board typically approves the minutes of the previous meeting at the
beginning of every Board meeting; therefore, the list below does not
normally contain details from the minutes of the most recent Board meeting.
WARNING: these pages may omit some original contents of the minutes.
Meeting times vary, the exact schedule is available to ASF Members and Officers, search for "calendar" in the Foundation's private index page (svn:foundation/private-index.html).
Report was filed, but display is awaiting the approval of the Board minutes.
## Description: Mahout is a distributed linear algebra framework and mathematically expressive DSL designed to let mathematicians, statisticians, and data scientists quickly implement their own algorithms. ## Project Status: Current project status: Ongoing Issues for the board: None at this time. A couple Directors showed interest in some details of project management during our recent period of revival, but the PMC feels we're on top of things now. ## Membership Data: Apache Mahout was founded 2010-04-20 (13 years ago) There are currently 28 committers and 10 PMC members in this project. The Committer-to-PMC ratio is roughly 7:3. Community changes, past quarter: - No new PMC members. Last addition was Shannon Quinn on 2023-02-12. - No new committers. Last addition was Jowanza Joseph on 2023-03-02. ## Project Activity: In our community meetings this quarter we prioritized quantum compute as a new back end, given the affinity between our matrix math focus and the arithmetic performed by quantum logic gates. We have a few new interested collaborators on the lists and in our ASF Slack channel. ## Community Health: Core team is in touch with each other and we have been consistent with community meetings (https://mahout.apache.org/minutes/2023/).
## Description: Mahout is a distributed linear algebra framework and mathematically expressive DSL designed to let mathematicians, statisticians, and data scientists quickly implement their own algorithms. ## Project Status: Current project status: Ongoing Issues for the board: None ## Membership Data: Apache Mahout was founded 2010-04-20 (13 years ago) There are currently 28 committers and 10 PMC members in this project. The Committer-to-PMC ratio is roughly 7:3. Community changes, past quarter: - No new PMC members. Last addition was Shannon Quinn on 2023-02-12. - No new committers. Last addition was Jowanza Joseph on 2023-03-02. ## Project Activity: In our community meetings this quarter we identified some work items that could benefit to and from some grad student projects, as well as some promising new compute platforms to prove out. ## Community Health: Core team is in touch with each other and we have been consistent with community meetings (https://mahout.apache.org/minutes/2023/).
@Sander: follow up about PMC removal process
## Description: Apache Mahout is a distributed linear algebra framework with a math DSL designed to let mathematicians, statisticians, and data scientists quickly implement their own algorithms. ## Project Status: Current project status: A slow quarter with team members' family obligations Issues for the board: None ## Membership Data: Apache Mahout was founded 2010-04-20 (13 years ago) There are currently 29 committers and 11 PMC members in this project. The Committer-to-PMC ratio is roughly 8:3. Community changes, past quarter: - No new PMC members. Last addition was Shannon Quinn on 2023-02-12. - No new committers. Last addition was Jowanza Joseph on 2023-03-02. ## Project Activity: Slowed on community meetings and project planning, but we have a structure in place. ## Community Health: Similar to last quarter, same core team is in touch with each other and we will pick back up on community calls this month.
## Description: Apache Mahout is a distributed linear algebra framework with a mathematically expressive Scala DSL designed to let mathematicians, statisticians, and data scientists quickly implement their own algorithms. ## Issues: There are no issues requiring board attention. ## Membership Data: Apache Mahout was founded 2010-04-20 (13 years ago) There are currently 29 committers and 11 PMC members in this project. The Committer-to-PMC ratio is roughly 8:3. Community changes, past quarter: - Shannon Quinn was added to the PMC on 2023-02-12 - Jowanza Joseph was added as committer on 2023-03-02 ## Project Activity: Revamping the build and release process underway, point release targeted for this month. Web site refreshed with improved documentation. Monthly community meetings resumed, minutes published to home page. ## Community Health: * 19 issues opened in JIRA, past quarter (1800% increase) * 10 issues closed in JIRA, past quarter (900% increase) * 30 commits in the past quarter (900% increase) * 5 code contributors in the past quarter (150% increase) * 10 PRs opened on GitHub, past quarter (900% increase) * 10 PRs closed on GitHub, past quarter (900% increase) * Mail totals across commits, issues, dev, and user: * Past quarter: 234 * Prev quarter: 22 * Health score: Healthy (6.74)
## Description: Apache Mahout is a distributed linear algebra framework with a mathematically expressive Scala DSL designed to let mathematicians, statisticians, and data scientists quickly implement their own algorithms. ## Issues: There are no issues requiring board attention. ## Membership Data: Apache Mahout was founded 2010-04-20 (13 years ago) There are currently 29 committers and 11 PMC members in this project. The Committer-to-PMC ratio is roughly 3:1. Community changes, past quarter: - New committer Jowanza Joseph accepted - New PMC member Shannon Quinn (squinn) - Emeritus PMC Drew Farris (drew) ## Project Activity: The project has been slow due to personal commitments for some time, while continuing to debate direction. We are renewing recruiting efforts and committing to some core improvements and additions, such as supporting Python in the math DSL instead of continuing with Scala. We are moving back to a monthly community meeting to include new contributors as well as to establish momentum around concrete plans. Short-term improvements include documentation and website fixes, along with outlines for plans. Mid-term (six months) target will be to have a plan in place for a Python DSL along with other features such as new data sources and indexers. Long-term (nine months plus) could include new back-end compute platforms such as Ray. ## Community Health: More activity across the board since previous quarter New JIRAs filed for doc improvements A few code comment doc PRs merged Website publishing fixed
No report was submitted.
@Roman: pursue a PMC roll call for Mahout
WHEREAS, the Board of Directors heretofore appointed Trevor Grant (rawkintrevo) to the office of Vice President, Apache Mahout, and WHEREAS, the Board of Directors is in receipt of the resignation of Trevor Grant from the office of Vice President, Apache Mahout, and WHEREAS, the Project Management Committee of the Apache Mahout project has chosen by vote to recommend Andrew Musselman (akm) as the successor to the post; NOW, THEREFORE, BE IT RESOLVED, that Trevor Grant is relieved and discharged from the duties and responsibilities of the office of Vice President, Apache Mahout, and BE IT FURTHER RESOLVED, that Andrew Musselman be and hereby is appointed to the office of Vice President, Apache Mahout, to serve in accordance with and subject to the direction of the Board of Directors and the Bylaws of the Foundation until death, resignation, retirement, removal or disqualification, or until a successor is appointed. Special Order 7A, Change the Apache Mahout Project Chair, was approved by Unanimous Vote of the directors present.
## Description: The mission of Mahout is the creation and maintenance of software related to Scalable machine learning library ## Issues: The project has not released for sometime and has done a self inventory and decided: 1. Pivot the direction of the project 2. Current PMC Chair to step down ## Membership Data: Apache Mahout was founded 2010-04-20 (12 years ago) There are currently 28 committers and 11 PMC members in this project. The Committer-to-PMC ratio is roughly 7:3. Community changes, past quarter: - No new PMC members. Last addition was Trevor Grant on 2017-02-03. - No new committers. Last addition was Christofer Dutz on 2020-06-08. ## Project Activity: 14.1 was released on 2020-10-07. 0.14.0 was released on 2019-03-05. 0.13.0 was released on 2017-04-17. We are overdue for a new release, however we've decided to pivot the project from "ML-on-Spark" to a different paradigm. ## Community Health: To prior notes on requesting comment on time since last release (and subsequently, lack of any code contribution over the last quarter): We have taken some inventory- and found ourselves at a cross road, our options: 1. Complete a long overdue refactoring of the code base to move the project from Apache Spark < v2.3 compatible to Spark v3+ compatible OR 2. Pivot the project in a new direction. After discussion among active PMCs and soliciting feedback on dev@ and user@ we decided option 2 was much better for the long term health of the project.
## Description: The mission of Mahout is the creation and maintenance of software related to Scalable machine learning library ## Issues: There are no specific issues the board needs to be aware of at this time. ## Membership Data: Apache Mahout was founded 2010-04-20 (12 years ago) There are currently 28 committers and 11 PMC members in this project. The Committer-to-PMC ratio is roughly 7:3. Community changes, past quarter: - No new PMC members. Last addition was Trevor Grant on 2017-02-03. - No new committers. Last addition was Christofer Dutz on 2020-06-08. ## Project Activity: 14.1 was released on 2020-10-07 There has been discussion of what direction to take the project next (specifically to update dependencies for Apache Spark 3+ or some other direction.) Also (post pandemic) life has drawn the attention of some of the more active committers. It has been a long while since any PMC or committers have been added, we are being mindful of 'who can we attract' to the project with respect to ongoing though exercises of 'which direction do we go next'. It's also worth noting that while we have solicited feed back on thoughts of going in other directions on user@ and dev@ we haven't actually held many (if any) discussions there, but we also haven't had them anywhere else either- just the occasional Slack DM or text message one-off between active PMC members, and where there is interest, reflection back to the list. ## Community Health: Mahout continues to exist as a sleepy little project: * dev@mahout.apache.org had a 200% increase in traffic in the past quarter (9 emails compared to 3) * user@mahout.aache.org had a 300% increase in traffic in the past quarter (8 emails compared to 2) * 1 commit in the past quarter (100% increase) * 1 code contributor in the past quarter (100% increase) Those statistics (while all positive, continue to speak to the sleepiness of the project. However, in addition to those statistics- two of the PMC members (akm and rawkintrevo) wrote an article that was published in Linux Magazine Germany, as well as republished in ADMIN Magazine[1] and is scheduled to also republished in AM70 Admin magazine. https://www.admin-magazine.com/HPC/Articles/Distributed-Linear-Algebra-with-Mahout
## Description: The mission of Mahout is the creation and maintenance of software related to Scalable machine learning library ## Issues: There are no issues requiring board attention. ## Membership Data: Apache Mahout was founded 2010-04-20 (12 years ago) There are currently 28 committers and 11 PMC members in this project. The Committer-to-PMC ratio is roughly 7:3. Community changes, past quarter: - No new PMC members. Last addition was Trevor Grant on 2017-02-03. - No new committers. Last addition was Christofer Dutz on 2020-06-08. ## Project Activity: 14.1 Was released 2020-10-07. We've cleaned up the releases area thanks to a note from Sebb. ## Community Health: dev@ and user@ have both once again seen dramatic upticks in email traffic. But the code base has had little to no activity, which is mainly a reflection of there really isn't that much changing in the world of distributed linear algebra. Comitters/PMC are over extended on other projects at the moment, but next steps should be updating to run with more modern versions of Apache Spark, and also easing the learning curve for new members.
No report was submitted.
## Description: The mission of Mahout is the creation and maintenance of software related to Scalable machine learning library ## Issues: The project has no specific issues other than Matrix Math on large distributed matrices is a bit of a niche problem which makes it difficult to attract users and contributors, PMC is working on this- no outside assistance is requested at this time. ## Membership Data: Apache Mahout was founded 2010-04-20 (12 years ago) There are currently 28 committers and 11 PMC members in this project. The Committer-to-PMC ratio is roughly 7:3. Community changes, past quarter: - No new PMC members. Last addition was Trevor Grant on 2017-02-03. - No new committers. Last addition was Christofer Dutz on 2020-06-08. ## Project Activity: 14.1 was released on 2020-10-07. Since the last report there has been sporadic activity on and a feature branch cut for Python bindings. There was a marked increase in mailing list activity due to discussions around Log4j vulnerability (we're OK since we're still on 1.x), as well as attempts to reboot the community calls. ## Community Health: As stated in the last section mailing list activity was up - 200% on dev@ and 300% on user@ however take these metrics with a grain of salt as they more tell how slow mailing list traffic was last quarter. PMC have been having discussions on how to re-invigorate the project and attract new users / committers as well as planning talks that will give us exposure at upcoming conferences.
## Description: The mission of Mahout is the creation and maintenance of software related to Scalable machine learning library ## Issues: No issues at this time. ## Membership Data: Apache Mahout was founded 2010-04-20 (11 years ago) There are currently 28 committers and 11 PMC members in this project. The Committer-to-PMC ratio is roughly 7:3. Community changes, past quarter: - No new PMC members. Last addition was Trevor Grant on 2017-02-03. - No new committers. Last addition was Christofer Dutz on 2020-06-08. ## Project Activity: Recent releases: 14.1 was released on 2020-10-07. 0.14.0 was released on 2019-03-05. 0.13.0 was released on 2017-04-17. Continues work on Python Bindings. Py4j+Scala+Pyton don't play nice. We're getting close pretty sure we are down to a Java versioning issue. (Java in Docker container is v1.11 which has known issues, including the error message we're seeing with JARs compiled with v1.8) ## Community Health: Little to no action on main branch- most free cycles were on pymahout feature. Once we solve Java version issue and have working prototype we'll merge that to a feature branch and show more activity. Comm. Health Statistics: dev@mahout.apache.org had a 75% decrease in traffic in the past quarter (3 emails compared to 12) issues@mahout.apache.org had a 100% decrease in traffic in the past quarter (0 emails compared to 18) 0 commits in the past quarter (-100% change) 0 code contributors in the past quarter (-100% change) 0 PRs opened on GitHub, past quarter (-100% change) 0 PRs closed on GitHub, past quarter (-100% change)
## Description: The mission of Mahout is the creation and maintenance of software related to Scalable machine learning library ## Issues: There are no issues requiring board attention. ## Membership Data: Apache Mahout was founded 2010-04-20 (11 years ago) There are currently 28 committers and 11 PMC members in this project. The Committer-to-PMC ratio is roughly 7:3. Community changes, past quarter: - No new PMC members. Last addition was Trevor Grant on 2017-02-03. - No new committers. Last addition was Christofer Dutz on 2020-06-08. ## Project Activity: Recent activity has focused on making Mahout more accessible to new users. This has been accomplished via * Getting started Docker container which features Apache Zeppelin with a Apache Mahout + Apache Spark interpreter and example notebooks * Continued work on Python bindings Recent releases: 14.1 was released on 2020-10-07. 0.14.0 was released on 2019-03-05. 0.13.0 was released on 2017-04-17 ## Community Health: It is somewhat concerning to see our community health score has fallen, as we felt there was an uptick in "real activity" over the last quarter. We continue to be on the look out for new contributors/committers to "fill the pipe". Potentially useful observations on community health: dev@mahout.apache.org had a 68% decrease in traffic in the past quarter (12 emails compared to 37) 0 issues opened in JIRA, past quarter (-100% change) 6 issues closed in JIRA, past quarter (100% increase) 3 commits in the past quarter (-75% change) 1 code contributor in the past quarter (-66% change) 3 PRs opened on GitHub, past quarter (200% increase) 4 PRs closed on GitHub, past quarter (300% increase)
## Description: The mission of Mahout is the creation and maintenance of software related to Scalable machine learning library ## Issues: There are no issues requiring board attention. ## Membership Data: Apache Mahout was founded 2010-04-20 (11 years ago) There are currently 28 committers and 11 PMC members in this project. The Committer-to-PMC ratio is roughly 7:3. Community changes, past quarter: - No new PMC members. Last addition was Trevor Grant on 2017-02-03. - No new committers. Last addition was Christofer Dutz on 2020-06-08. ## Project Activity: The statistics related the project tell a story of sharply decreased attention, however this does not paint an accurate picture. As Data Science as a phenomenon has shifted away from the Java ecosystem and Scala wanes in popularity in general- we believe now more than ever the importance of developing a Python interface to Apache Mahout. While the Java components were shockingly easy to incorporate, the Scala portions have proven more... troublesome. However, we are still working along as we are able to develop a prototype that will allow us to itemize the work via JIRA tickets, and assign out. Aside from the work on Python bindings fork, little has been accomplished on the actual code base. Finally, we've had a new contributor who spoke at ApacheCon@Home who donated a Ridge Regression algorithm to the library. ## Community Health: The community is still strong in spite of the the story the statistics tell. I will restate, that most of the actual coding has been toying with a prototype of Python bindings, which the active PMC members feel like is the best use of their time for the future of the project. Also- the community calls which started before the holidays, were never able to regain momentum in the New Year, a trend we can hopefully reverse, however again, there isn't much to talk about, since most of the work is on the Python bindings. We do note the exorbitant amount of time since a PMC or committer was added, and realize a close second priority to composing Python bindings would be focusing on community health (specifically, a strategy to attract and retain "new blood"). That said, we are hoping introduction of Python bindings will open us up to an entire new world of potential users, some of which we hope will graduate to contributors, from which we will readily grant commit bits, and those who show long term interest and dedication will happily be welcomed as PMC.
## Description: The mission of Mahout is the creation and maintenance of software related to Scalable machine learning library ## Issues: Nothing requiring board attention at this time. ## Membership Data: Apache Mahout was founded 2010-04-20 (11 years ago) There are currently 28 committers and 11 PMC members in this project. The Committer-to-PMC ratio is roughly 7:3. Community changes, past quarter: - No new PMC members. Last addition was Trevor Grant on 2017-02-03. - No new committers. Last addition was Christofer Dutz on 2020-06-08. ## Project Activity: Since last board report Trevor Grant has taken over as PMC Chair and initiated weekly status call meetings, with minutes posted to mahout.apache.org and posted back to the mailing list. Also- the community has taken up an initiative to begin releasing Python bindings, and hope to include this in the next release. ## Community Health: We are MUCH healthier than we have been for some time, due alone to our ability to execute builds. This isn't really reflected in the statistics, but is a huge boon for the project. Secondly, after Trevor Grant took over as project chair and began hosting weekly meetings this has negatively impacted mailing list activity as often interested parties will discuss their plans and get feedback on a weekly call whose minutes are reported back- however the entire thread is not archived on the list (decreased mailing list chatter). We are starting to see more action in meaningful PRs and large initiatives, such as Python bindings, Zeppelin+Mahout Getting Started Docker containers, and others are of course still discussed on the list as well as at community meetings. An interesting bit- is that the opened and closed JIRA tickets are greater than open and closed PRs. This is due to some JIRA pruning and deleting old spammy JIRA tickets (from over the prior quarter- one of the first topics of the weekly community call meetings). Issues mailing list was also up due to this. We have resumed a focus on the hunt to bring in fresh committers to our community and have promising leads from ApacheCon and other sources. We have also as a project begun to re-envsion ourselves from just anotherML lib to distributed statistics, a niche exploitation strategy that we hope will help us attract more interest. In this vein- Apache Mahout was used in an example in a new O'Reilly book and we hope that will also help us with this rebranding (using DS-SVD to decompose COVID lung scans). Finally, after a big push to release, and with holidays and other life events of some of the main committers, we all just took a breather. And still the project has healthy statistics. We look forward to some great progression in 2021. - dev@mahout.apache.org had a 48% decrease in traffic in the past quarter (64 emails compared to 123) - issues@mahout.apache.org had a 76% increase in traffic in the past quarter (92 emails compared to 52) - user@mahout.apache.org had a 70% decrease in traffic in the past quarter (4 emails compared to 13) - 8 issues opened in JIRA, past quarter (-50% decrease) - 9 issues closed in JIRA, past quarter (125% increase) - 15 commits in the past quarter (-63% decrease) - 3 code contributors in the past quarter (-40% decrease) - 5 PRs opened on GitHub, past quarter (-16% decrease) - 6 PRs closed on GitHub, past quarter (100% increase)
WHEREAS, the Board of Directors heretofore appointed Andrew Musselman (akm) to the office of Vice President, Apache Mahout, and WHEREAS, the Board of Directors is in receipt of the resignation of Andrew Musselman from the office of Vice President, Apache Mahout, and WHEREAS, the Project Management Committee of the Apache Mahout project has chosen by vote to recommend Trevor Grant (rawkintrevo) as the successor to the post; NOW, THEREFORE, BE IT RESOLVED, that Andrew Musselman is relieved and discharged from the duties and responsibilities of the office of Vice President, Apache Mahout, and BE IT FURTHER RESOLVED, that Trevor Grant be and hereby is appointed to the office of Vice President, Apache Mahout, to serve in accordance with and subject to the direction of the Board of Directors and the Bylaws of the Foundation until death, resignation, retirement, removal or disqualification, or until a successor is appointed. Special Order 7A, Change the Apache Mahout Project Chair, was approved by Unanimous Vote of the directors present.
## Description: The mission of Mahout is the creation and maintenance of software related to Scalable machine learning library ## Issues: No issues to report. ## Membership Data: Apache Mahout was founded 2010-04-20 (10 years ago) There are currently 28 committers and 11 PMC members in this project. The Committer-to-PMC ratio is roughly 7:3. Community changes, past quarter: - No new PMC members. Last addition was Trevor Grant on 2017-02-03. - No new committers. Last addition was Chris Dutz on 2020-06-08. ## Project Activity: New release of 14.1 this month, with extensive refactoring of the build system by new committer Chris Dutz. Talks at Apachecon @ Home: 1. A Data Scientist First-Time Mahout Experience: Tips and Takeaways * Jose Francisco Hernandez Santa Cruz 2. Modern Recommenders with Mahout * Patrick (Pat) Ferrel 3. Mahout and Kubeflow Together At Last * Trevor Grant 4. Apache Mahout on Zeppelin * Andrew Musselman 5. The Long and Winding Road to Becoming A Mahout Committer * Trevor Grant, Andrew Musselman, Pat Ferrel 6. Mahout: State of the Matrix * Trevor Grant ## Community Health: We have had a good quarter in terms of engagement and technical progress. The stats here show a lot of activity around build restructuring and release, as well as a consistent amount of code contributors. * Community Health Score (Chi): 4.70 (Healthy) * dev@mahout.apache.org had a 30% decrease in traffic in the past quarter (124 emails compared to 175) * user@mahout.apache.org had a 116% increase in traffic in the past quarter (13 emails compared to 6) * 15 issues opened in JIRA, past quarter (114% increase) * 3 issues closed in JIRA, past quarter (200% increase) * 31 commits in the past quarter (24% increase) * 5 code contributors in the past quarter (no change) * 5 PRs opened on GitHub, past quarter (-28% decrease) * 2 PRs closed on GitHub, past quarter (-93% decrease) We hope to interest new contributors in some documentation and tutorial creation in the coming quarter. After just over two years of chairing the project, Andrew Musselman (akm@a.o) is resigning and the PMC has approved nomination of Trevor Grant (rawkintrevo@a.o) to take the position. This notice is before the Board this month; thank you in advance for attending to the change.
## Description: Apache Mahout is an environment for quickly creating scalable performant machine learning applications. ## Issues: Nothing requiring board attention at this time. ## Membership Data: Apache Mahout was founded 2010-04-20 (10 years ago) There are currently 28 committers and 11 PMC members in this project. The Committer-to-PMC ratio is roughly 7:3. Community changes, past quarter: - No new PMC members. Last addition was Trevor Grant on 2017-02-03. - Christofer Dutz was added as committer on 2020-06-08 ## Project Activity: With volunteer effort from Chris Dutz we have refactored and modernized the build structure, and we were able to push a release candidate for 14.1 to repository.a.o with simple maven release plugin commands. Two bugs were discovered in the RC which requires another build, but we expect to have our release out this month. ## Community Health: Per reporter, 5.11 (Healthy) Notable mailing list trends: dev@mahout.apache.org had a 52% increase in traffic in the past quarter (178 emails compared to 117) issues@mahout.apache.org had a 76% decrease in traffic in the past quarter (45 emails compared to 187) JIRA activity: 7 issues opened in JIRA, past quarter (-65% decrease) 1 issue closed in JIRA, past quarter (-91% decrease) Commit activity: 25 commits in the past quarter (-71% decrease) 5 code contributors in the past quarter (66% increase) GitHub PR activity: 7 PRs opened on GitHub, past quarter (16% increase) 33 PRs closed on GitHub, past quarter (560% increase)
## Description: Apache Mahout is an environment for quickly creating scalable performant machine learning applications. ## Issues: No changes since last report. ## Membership Data: Apache Mahout was founded 2010-04-20 (10 years ago) There are currently 27 committers and 12 PMC members in this project. The Committer-to-PMC ratio is 9:4. Community changes, past quarter: - No new PMC members. Last addition was Trevor Grant on 2017-02-03. - No new committers. Last addition was Holden Karau on 2017-07-12. ## Project Activity: The team are still working on a 0.14 release; we will be requesting help from the builds@a.o list and a known-good Maven user from the roster in the next month. We have included two new collaborators (Joe Olson, Tom Liakos) in discussions on refactoring the build tools. ## Community Health: (3.06 per Reporter.a.o) Notable mailing list trends: - dev@mahout.apache.org had a 37% decrease in traffic in the past quarter (118 emails compared to 186): - issues@mahout.apache.org had a 87% increase in traffic in the past quarter (191 emails compared to 102): JIRA activity: - 20 issues opened in JIRA, past quarter (53% increase) - 12 issues closed in JIRA, past quarter (140% increase) Commit activity: - 87 commits in the past quarter (-55% decrease) - 3 code contributors in the past quarter (-40% decrease) GitHub PR activity: - 6 PRs opened on GitHub, past quarter (-50% decrease) - 5 PRs closed on GitHub, past quarter (-44% decrease)
@Justin: look into helping Mahout perform a release
## Description: Apache Mahout is an environment for quickly creating scalable performant machine learning applications. ## Issues: There are no issues requiring board attention. ## Membership Data: Apache Mahout was founded 2010-04-20 (10 years ago) There are currently 27 committers and 12 PMC members in this project. The Committer-to-PMC ratio is 9:4. Community changes, past quarter: - No new PMC members. Last addition was Trevor Grant on 2017-02-03. - No new committers. Last addition was Holden Karau on 2017-07-12. ## Project Activity: The team is working on a point release, v14.1. There are some continued issues which are drawing out this release, and the team has resumed weekly sessions to resolve. There are additional contributions from new team members which are queued up for a .2 release. Point release deployed artifacts are cross-compiled for Scala 2.12 and 2.11, with several other dependency upgrades. - Last release was 0.14.0 on Wednesday, March 6, 2019 ## Community Health: (4.70 per Reporter.a.o) Notable mailing list trends: - dev@mahout.apache.org had a 3820% increase in traffic in the past quarter (196 emails compared to 5): - issues@mahout.apache.org had a big increase in traffic in the past quarter (102 emails compared to 0) JIRA activity: - 13 issues opened in JIRA, past quarter (1300% increase) - 5 issues closed in JIRA, past quarter (500% increase) Commit activity: - 196 commits in the past quarter (752% increase) - 5 code contributors in the past quarter (150% increase) GitHub PR activity: - 12 PRs opened on GitHub, past quarter (1200% increase) - 9 PRs closed on GitHub, past quarter (800% increase)
## Description: Apache Mahout is an environment for quickly creating scalable performant machine learning applications. ## Issues: There are no issues requiring board attention at this time. ## Activity: The team is working on a point release, v14.1, to resolve missing binary artifacts from the 0.14.0 release. There are some tough issues which are drawing out this release, and the team is actively recruiting people with experience fixing errors in Maven configs. ## PMC changes: - Currently 14 PMC members. - No new PMC members added in the last 3 months - Last PMC addition was Trevor Grant on Fri Feb 03 2017 ## Committer base changes: - Currently 28 committers. - No new committers added in the last 3 months - Last committer addition was Holden Karau at Wed Jul 12 2017 ## Releases: - Last release was 0.14.0 on Wednesday, March 6, 2019 ## Mailing list activity: - Nothing significant in the figures ## JIRA activity: - Nothing significant in the figures
## Description: Apache Mahout is an environment for quickly creating scalable performant machine learning applications. ## Issues: There are no issues requiring board attention at this time. ## Activity: The team is working on a point release, v14.1, to resolve missing binary artifacts from the 0.14.0 release. ## Presentations and Talks Josh Kalina, “Portfolio theory with Apache,” Apache Roadshow Chicago, IL, May 14 ## PMC changes: - Currently 14 PMC members. - No new PMC members added in the last 3 months - Last PMC addition was Trevor Grant on Fri Feb 03 2017 ## Committer base changes: - Currently 28 committers. - No new committers added in the last 3 months - Last committer addition was Holden Karau at Wed Jul 12 2017 ## Releases: - Last release was 0.14.0 on Wednesday, March 6, 2019 ## Mailing list activity: - Nothing significant in the figures ## JIRA activity: - 4 JIRA tickets created in the last 3 months - 3 JIRA tickets closed/resolved in the last 3 months
## Description: - Apache Mahout is an environment for quickly creating scalable performant machine learning applications. ## Issues: - There are no issues requiring board attention at this time. ## Activity: - The project released version 0.14.0 on March 4, and is working on a point release this month. ## Health report: - Project health increased with our last release, and the project team is working on community efforts including conference talks and networking for committers. ## PMC changes: - Currently 14 PMC members. - No new PMC members added in the last 3 months - Last PMC addition was Trevor Grant on Fri Feb 03 2017 ## Committer base changes: - Currently 28 committers. - No new committers added in the last 3 months - Last committer addition was Holden Karau at Wed Jul 12 2017 ## Releases: - 0.14.0 was released on Mon Mar 04 2019 ## JIRA activity: - 10 JIRA tickets created in the last 3 months - 2 JIRA tickets closed/resolved in the last 3 months
## Description: Apache Mahout is an environment for quickly creating scalable performant machine learning applications. ## Issues: No changes since last report. ## Activity: The team are in the middle of a 0.14 release; first RC is being tested and kinks ranging from the move to gitbox and left-over items that need to be adjusted in Jenkins are being shaken out. The PMC heard the advice from the board to consider adding committers and PMC members; with the holidays and work pressures there has been less time than usual to work on community efforts but we would like to grow the team this quarter. Some conference activity at FOSDEM for example will be one route, continued meetup sessions will be another. ## PMC changes: - Currently 14 PMC members - No new PMC members added in the last 3 months - Last PMC addition was Trevor Grant on Fri Feb 03 2017 ## Committer base changes: - Currently 28 committers. - No new committers added in the last 3 months - Last committer addition was Holden Karau at Wed Jul 12 2017 ## Releases: - Last release was 0.13.0 on Sun Apr 16 2017 - Current release for 0.14.0 is underway this week
## Description:
Apache Mahout is an environment for quickly creating scalable performant
machine learning applications.
## Issues:
The board passed Andrew Palumbo’s resignation as PMC chair with Andrew
Musselman taking over as the PMC chair.
## Activity:
The team has begun holding weekly working sessions toward a 0.14 release, to
refocus on a large refactoring effort.
## Presentations and Talks:
- “Matrix Math at Scale with Apache Mahout and Spark”: workshop at Open
Source Summit, Vancouver, BC, Canada, August 28 (Slides at https://
events.linuxfoundation.org/wp-content/uploads/2017/11/Workshop-Matrix-Math
-at-Scale-with-Apache-Mahout-and-Spark-Andrew-Musselman-Apache-Mahout.pdf)
## PMC changes:
- Currently 14 PMC members
- No new PMC members added in the last 3 months
- Last PMC addition was Trevor Grant on Fri Feb 03 2017
## Committer base changes:
- Currently 28 committers
- No new committers added in the last 3 months
- Last committer addition was Holden Karau at Wed Jul 12 2017
## Releases:
- Last release was 0.13.0 on Sun Apr 16 2017
## Mailing list activity:
- Nothing significant in the figures
## JIRA activity:
- 4 JIRA tickets created in the last 3 months
- 0 JIRA tickets closed/resolved in the last 3 months
WHEREAS, the Board of Directors heretofore appointed Andrew Palumbo (apalumbo) to the office of Vice President, Apache Mahout, and WHEREAS, the Board of Directors is in receipt of the resignation of Andrew Palumbo from the office of Vice President, Apache Mahout, and WHEREAS, the Project Management Committee of the Apache Mahout project has chosen by vote to recommend Andrew Musselman (akm) as the successor to the post; NOW, THEREFORE, BE IT RESOLVED, that Andrew Palumbo is relieved and discharged from the duties and responsibilities of the office of Vice President, Apache Mahout, and BE IT FURTHER RESOLVED, that Andrew Musselman be and hereby is appointed to the office of Vice President, Apache Mahout, to serve in accordance with and subject to the direction of the Board of Directors and the Bylaws of the Foundation until death, resignation, retirement, removal or disqualification, or until a successor is appointed. Special Order 7B, Change the Apache Mahout Project Chair, was approved by Unanimous Vote of the directors present.
## Description: Apache Mahout is an environment for quickly creating scalable performant machine learning applications. ## Issues: A change of chair resolution is before the board with Andrew Palumbo’s resignation as PMC chair; with Andrew Musselman taking over as the PMC chair. ## Activity: Working towards a 0.14.0 release this Summer. The Primary sticking point is building in Maven for release to multiple versions of Scala, which requires major restructuring of the poms. The Website has been slightly restructures so as to leave old URL pointing to the right pages even though the new site uses Jekyll. Periodic blog posts are being solicited and are in process. ## PMC changes: - Currently 14 PMC members. - No new PMC members added in the last 3 months - Last PMC addition was Trevor Grant on Fri Feb 03 2017 ## Committer base changes: - Currently 28 committers. - No new committers added in the last 3 months - Last committer addition was Holden Karau at Wed Jul 12 2017 ## Releases: - Last release was 0.13.0 on Sun Apr 16 2017 ## JIRA activity: - 20 JIRA tickets created in the last 3 months - 13 JIRA tickets closed/resolved in the last 3 months
## Description: Apache Mahout is an environment for quickly creating scalable performant machine learning applications. ## Issues: There are no issues requiring board attention at this time ## Activity: Activity this quarter has been low. The team hit a larger blocker with a very ambitious multi-artifact release; while upgrading scala and spark versions shifting IP restrictions and life for an all volunteer team has made it difficult to get over this hump. As well the the team, spread very thin in the past quarters, took on several other large tasks, leaving all overworked. The team has been considering plan for a release to become Spark 2.x/scala 2_11.x compliant: https://lists.apache.org/list.html?dev@mahout.apache.org:2018-3 ## PMC changes: - Currently 14 PMC members. - No new PMC members added in the last 3 months - Last PMC addition was Trevor Grant on Sat Feb 04 2017 ## Committer base changes: - Currently 28 committers. - No new committers added in the last 3 months - Last committer addition was Holden Karau at Wed Jul 12 2017 ## Releases: - Last release was 0.13.0 on Mon Apr 17 2017 ## Mailing list activity: Mailing list activity has slowed compared to relatively steadily over the last quarters. We have established a #mahout channel on https://the-asf.slack.com/ in hopes of reaching more people. - dev@mahout.apache.org: - 902 subscribers (down -7 in the last 3 months): - 46 emails sent to list (49 in previous quarter) - general@mahout.apache.org: - 10 subscribers (up 0 in the last 3 months): - 3 emails sent to list (0 in previous quarter) - issues@mahout.apache.org: - 15 subscribers (up 0 in the last 3 months): - 22 emails sent to list (125 in previous quarter) - user@mahout.apache.org: - 1750 subscribers (down -11 in the last 3 months): - 11 emails sent to list (19 in previous quarter) ## JIRA activity: - 2 JIRA tickets created in the last 3 months - 2 JIRA tickets closed/resolved in the last 3 months ## Talks and Publications “Apache Mahout.” Author: Andrew Musselman. In: Sakr S., Zomaya A. (eds) Encyclopedia of Big Data Technologies. Springer, Cham. February 26, 2018. https://link.springer.com/referenceworkentry/10.1007/978-3-319-63962-8_144-1 “Matrix Math at Scale with Apache Mahout and Spark,” Andrew Musselman; workshop at ODSC East, Boston, May 2nd 2018. https://odsc.com/training/portfolio/matrix-math-scale-apache-mahout-spark The Magnificent Modular Mahout:An extensible library for distributed math and HPC. Trevor Grant, HPC, Big Data, and Data Science track, Fosdem 2018.
## Description: Apache Mahout is an environment for quickly creating scalable performant machine learning applications. ## Issues: There are no issues requiring board attention at this time ## Activity: Activity this quarter has been low. The team hit a larger blocker with a very ambitious multi-artifact release; while upgrading scala and spark versions shifting IP restrictions and life for an all volunteer team has made it difficult to get over this hump. As well the the team, spread very thin in the past quarters, took on several other large tasks, leaving all overworked. The team has been considering plan for a release to become Spark 2.x/scala 2_11.x compliant: https://lists.apache.org/list.html?dev@mahout.apache.org:2018-3 ## PMC changes: - Currently 14 PMC members. - No new PMC members added in the last 3 months - Last PMC addition was Trevor Grant on Sat Feb 04 2017 ## Committer base changes: - Currently 28 committers. - No new committers added in the last 3 months - Last committer addition was Holden Karau at Wed Jul 12 2017 ## Releases: - Last release was 0.13.0 on Mon Apr 17 2017 ## Mailing list activity: Mailing list activity has slowed compared to relatively steadily over the last quarters. We have established a #mahout channel on https://the-asf.slack.com/ in hopes of reaching more people. - dev@mahout.apache.org: - 902 subscribers (down -7 in the last 3 months): - 46 emails sent to list (49 in previous quarter) - general@mahout.apache.org: - 10 subscribers (up 0 in the last 3 months): - 3 emails sent to list (0 in previous quarter) - issues@mahout.apache.org: - 15 subscribers (up 0 in the last 3 months): - 22 emails sent to list (125 in previous quarter) - user@mahout.apache.org: - 1750 subscribers (down -11 in the last 3 months): - 11 emails sent to list (19 in previous quarter) ## JIRA activity: - 2 JIRA tickets created in the last 3 months - 2 JIRA tickets closed/resolved in the last 3 months ## Talks and Publications “Apache Mahout.” Author: Andrew Musselman. In: Sakr S., Zomaya A. (eds) Encyclopedia of Big Data Technologies. Springer, Cham. February 26, 2018. https://link.springer.com/referenceworkentry/10.1007/978-3-319-63962-8_144-1 “Matrix Math at Scale with Apache Mahout and Spark,” Andrew Musselman; workshop at ODSC East, Boston, May 2nd 2018. https://odsc.com/training/portfolio/matrix-math-scale-apache-mahout-spark The Magnificent Modular Mahout:An extensible library for distributed math and HPC. Trevor Grant, HPC, Big Data, and Data Science track, Fosdem 2018, Brussels.
@Phil: pursue a report for Mahout for next month
Apache Mahout Board Report, Jan 2018
Apache Mahout is an environment for quickly creating scalable performant
machine learning applications.
## Issues:
- None
## Activity:
- 0.13.1 release in the works, though a code freeze has been temporarily
lifted. 0.13.1 is a multi-artifact release extending 0.13.0 to all
combinations of Spark from 1.6 - 2.x and 2.10, scala 2.11
- Continuing work on building out an algorithm library and continued native
optimizations.
- A More modern website has been designed and deployed
- David Miller, Creator of Start Bootstrap has agreed to do the site redesign
pro-bono.
Work is ongoing to fix minor errors on the new website; broken Links, etc. A
new logo is being considered.
## Health report:
- The health of the project is good with a devoted team of committers.
## PMC changes:
- Currently 14 PMC members.
- No new PMC members added in the last 3 months
- Last PMC addition was Trevor Grant on Sat Feb 04 2017
- PMC member Benson Margulies has changed his status to PMC Emeritus
## Committer base changes:
- Currently 28 committers.
- New commmitters:
- Holden Karau was added as a committer on Wed Jul 12 2017
- Dustin VanStee was added as a committer on Tue Jun 20 2017
## Releases:
- Last release was 0.13.0 on Mon Apr 17 2017
## Mailing list activity:
- dev@mahout.apache.org:
- 912 subscribers (down -6 in the last 3 months):
- 53 emails sent to list (110 in previous quarter)
- issues@mahout.apache.org:
- 15 subscribers (down -1 in the last 3 months):
- 132 emails sent to list (111 in previous quarter)
- user@mahout.apache.org:
- 1761 subscribers (down -14 in the last 3 months):
- 19 emails sent to list (43 in previous quarter)
Again we are seeing a dip in user@mahout.apache.org emails and
dev@mahout.apache.org. We will also continue to monitor these.
Apache Mahout is an environment for quickly creating scalable performant machine learning applications. ## Issues: - None ## Activity: * 0.13.1 release in the works, though a code freeze has been temporarily lifted. 0.13.1 is a multi-artifact release extending 0.13.0 to all combinations of Spark from 1.6 - 2.x and 2.10, scala 2.11 * Current work is on building out an algorithm library and continued native optimizations. * More work on a modern Website * A designer has been found. * David Miller, Creator of Start Bootstrap has agreed to do a site redesign pro-bono. * Work is ongoing to update the way the website is built and deployed. We are working with the Apache infrastructure team to move from a custom process to a more standardized way of deploying the website using pre-built deployment templates. * Google Summer of Code - We have enthusiastically accepted Aditya Sarma’s proposal to add the DBSCAN clustering algorithm, and additionally an alternate implementation of the DBSCAN algorithm which reduces complexity from O(n^2) to O(log(n) * n). * GSoC experience [Aditya] - I proposed to add an distributed DBSCAN implementation on the lines of the paper “A new scalable parallel DBSCAN algorithm using the disjoint-set data structure” authored by Md. Mostofa Ali Patwary, Diana Palsetia, Ankit Agrawal, Wei-keng Liao, Fredrik Manne, Alok Choudhary of Northwestern University. But it turned out that the distribution strategy that they have adopted does not fit well with Mahout’s underlying framework. So, I contributed the Sequential algorithm and am working on completing the RTree module (which can be used by both the sequential as well as the distributed algorithm). In the meanwhile, I got in touch with a professor from the Barcelona Supercomputing Center and her group worked on an approximate dbscan algorithm that scaled well. (As an aside, I’m planning to work on making Mahout accessible to newcomers along with Trevor) * GSoC Student Aditya Sarma passed with the mentoring of Trevor Grant. ## Health report: - The health of the project is good with a devoted team of committers. ## PMC changes: - Currently 14 PMC members. - No new PMC members added in the last 3 months - Last PMC addition was Trevor Grant on Sat Feb 04 2017 - PMC member Benson Margulies has changed his status to PMC Emeritus ## Committer base changes: - Currently 28 committers. - New commmitters: - Holden Karau was added as a committer on Wed Jul 12 2017 - Dustin VanStee was added as a committer on Tue Jun 20 2017 ## Releases: - Last release was 0.13.0 on Mon Apr 17 2017 ## External Events Eigenfaces for Realtime Facial Recognition Scott Cote, Trevor Grant. Lucene Revolution. Las Vegas, NV- September 15, 2017. Do I Know You? Realtime Facial Recognition with an Apache Stack. Trevor Grant. Flink Forward. Berlin, DE - September 12, 2017. Using Open Source AI with Drones to identify humans… Friendly Cylons 1.0… Trevor Grant, who did not have editing privileges on the title or abstract which is why it seems so hokey. Data and Cognitive Developers Meetup. New York, NY - September 25- 2017. Open Source AI - Roll Your Own Cylon. Trevor Grant Chicago Hadoop Users Group (CHUG) / Chicago Apache Flink Meetup (CHAF) Joint Meetup. Chicago, IL - August 24, 2017. Weekend Project: Real World AirBnB Data Science and Pricing Bot. Trevor Grant, Andrew Weiner. Berlin Buzzwords 2017. https://berlinbuzzwords.de/17/session/weekend-project-real-world-airbnb-data-s cience-and-pricing-bot. Introduction to Online Machine Learning Algorithms. Trevor Grant, Dataworks Summit, San Jose, CA - https://dataworkssummit.com/san-jose-2017/sessions/introduction-to-online-mach ine-learning-algorithms/. Success at Apache: All My Roads Led to Apache, Pat Ferrel: https://blogs.apache.org/foundation/entry/success-at-apache-all-my Apache Mahout: Distributed Matrix Math for Machine Learning. Andrew Musselman, Seattle Data/Analytics/Machine Learning Meetup, Seattle, WA - October 17, 2017. Distributed Evolution of Spiking Neuron Models on Apache Mahout for Time Series Analysis. Andrew Palumbo, Annual Symposium on Biomathematics and Ecology: Education and Research, Illinois State University, Bloomington Illinois, October 8, 2017. Open Source Artificial Intelligence in a Biological/Ecological Context. Trevor Grant, Annual Symposium on Biomathematics and Ecology: Education and Research, Illinois State University, Bloomington Illinois, October 8, 2017. ## Question asked by board to clarify from March’s board report: * AWS has been sending emails to private@mahout.apache.org RE: a small (~16$) balance. This is due to Amazon donating 1000$ of cluster time to a project member, who has since taken a position with a different organization. The 1000$ was on a now discontinued corporate card. We are actively working on getting the situation worked out (the usual large corporate SNAFU keeps this fix at a snail’s pace), and getting more compute time donated from AWS. * Resolution: Balance has been paid, and the account moved to an active credit card. ## Mailing list activity: - dev@mahout.apache.org: - 918 subscribers (down -1 in the last 3 months): - 256 emails sent to list (582 in previous quarter) - issues@mahout.apache.org: - 16 subscribers (up 16 in the last 3 months): - 84 emails sent to list (0 in previous quarter) We’ve moved all Jira (including Github linked) comments from dev@mahout.apache.org to issues@mahout.apache.org, in order to reduce noise on dev@mahout.apache.org and to facilitate discussion on the list. This move however does not account for the full dip in dev@mahout.apache.org emails over the summer (582 to 256). We will be monitoring the activity on this list. - user@mahout.apache.org: - 1783 subscribers (down -8 in the last 3 months): - 41 emails sent to list (155 in previous quarter) As well we can see a dip in user@mahout.apache.org emails. We will also continue to monitor this.
Apache Mahout is an environment for quickly creating scalable performant machine learning applications. ## Issues: - None ## Activity: - 0.13.1 release in the works extends 0.13.0 to Spark 2.x and scala 2.11 Current work is on building out an algorithm library and continued native optimizations. - More work on a modern Website - A designer has been found. - Google Summer of Code - We have enthusiastically accepted Aditya Sarma’s proposal to add the DBSCAN clustering algorithm, and additionally an alternate implementation of the DBSCAN algorithm which reduces complexity from O(n^2) to O(log(n) * n). ## Health report: - The health of the project is good with a devoted team of committers. ## PMC changes: - Currently 15 PMC members. ## Committer base changes: New Committers this quarter: - Dustin VanStee was made committer on Jun 19, 2017 - Holden Karu was made a committer on Jul 11, 2017 - Currently 29 committers. ## External Events - Eigenfaces for Realtime Facial Recognition Scott Cote, Trevor Grant. Lucene Revolution. Las Vegas, NV- September 15, 2017. - INTRODUCTION TO ONLINE MACHINE LEARNING ALGORITHMS Trevor Grant. Dataworks Summit. San Jose, CA- June 15, 2007 - Distributed and Native Hybrid optimizations for Machine Learning Workloads Suneel Marthi. Berlin Buzzwords. Berlin, Germany- June 12, 2017 - Apache Mahout: Distributed Matrix Math for Machine Learning Andrew Musselman. MLConf. Seattle, WA- May 19, 2017 - An Apache Based Intelligent IoT Stack for Transportation Trevor Grant, Joe Olsen. ApacheCon IoT. Miami, FL- May 18, 2017 - Apache Mahout: An Extendable Machine Learning Framework for Spark and Flink Trevor Grant. Apache Big Data. Miami, FL- May 16, 2017 - APACHE MAHOUT’S NEW RECOMMENDER ALGORITHM AND USING GPUS TO SPEED MODEL CREATION Pat Ferrel, Andy Palumbo. GPU Technology Conference. Silicon Valley, CA- May 11, 2017 - EXTENDING MAHOUT-SAMSARA LINEAR ALGEBRA DSL TO SUPPORT GPU CLUSTERS Suneel Marthi, Trevor Grant. GPU Technology Conference. Silicon Valley, CA- May 11, 2017 ## Question asked by board to clarify from last quarter’s report: - AWS has been sending emails to private@mahout.apache.org RE: a small (~16$) balance. This is due to Amazon donating 1000$ of cluster time to a project member, who has since taken a position with a different organization. The 1000$ was on a now discontinued corporate card. We are actively working on getting the situation worked out (the usual large corporate SNAFU keeps this fix at a snail’s pace), and getting more compute time donated from AWS.
@Rich: help resolve billing issue with AWS
Apache Mahout Board Report, May 2017 Apache Mahout is an environment for quickly creating scalable performant machine learning applications. ## Issues: - None ## Activity: Mahout released its benchmark 0.13.0 release with GPU and multi-threaded native solvers using OpenCL, OpenMP (ViennaCL), and CUDA (NVIDIA) in the works. An intuitive Algorithm Development Framework was also released in 0.13.0 based on the sk-learn model. Current work is on building out an algorithm library and continued native optimizations. New more modern Website Google Summer of Code - We have enthusiastically accepted Aditya Sarma’s proposal to add the DBSCAN clustering algorithm, and additionally an alternate implementation of the DBSCAN algorithm which reduces complexity from O(n^2) to O(log(n) * n). ## Health report: - The health of the project is good with a devoted team of committers. ## PMC changes: - Currently 15 PMC members. - Last PMC addition was Trevor Grant on Feb 4 2017 ## Committer base changes: - Nikolai Sakarnykh was added as a committer on April 21, 2017 - Currently 27 committers. ## External Events APACHE MAHOUT'S NEW RECOMMENDER ALGORITHM AND USING GPUS TO SPEED MODEL CREATION Pat Ferrel, Andy Palumbo. GPU Technology Conference. Silicon Valley, CA- May 11, 2017 EXTENDING MAHOUT-SAMSARA LINEAR ALGEBRA DSL TO SUPPORT GPU CLUSTERS Suneel Marthi, Trevor Grant. GPU Technology Conference. Silicon Valley, CA- May 11, 2017 Apache Mahout: An Extendable Machine Learning Framework for Spark and Flink Trevor Grant. Apache Big Data. Miami, FL- May 16, 2017 An Apache Based Intelligent IoT Stack for Transportation Trevor Grant, Joe Olsen. ApacheCon IoT. Miami, FL- May 18, 2017 Apache Mahout: Distributed Matrix Math for Machine Learning Andrew Musselman. MLConf. Seattle, WA- May 19, 2017
No report was submitted.
## Description:
Apache Mahout is an environment for quickly creating scalable performant
machine learning applications.
## Issues:
None
## Activity:
- The Team is currently in the process of putting together a milestone 0.13.0
release.
- Work is presently focused on adding support for Visualization, GPU and
native optimization.
- Sebastian Schelter presented a poster at Machine Learning Systems Workshop,
NIPS 2016 Dec 10, 2016 “Samsara: Declarative Machine Learning on
Distributed Dataflow Systems” - https://ssc.io/pdf/poster-mlsystems.pdf
- Andrew Palumbo presented “Apache Mahout: Beyond MapReduce” at the Orange
County Big Data Meetup, October, 2016.
- Trevor Grant presented: “Apache Mahout?! What’s Next!” At
Chicago Hadoop Users Group, October 2016
Seattle Data Science Meetup, December 2016
San Diego Big Data Meetup, December 2016
Austin Data Meetup, December 2016
DFW Data Science Meetup, December 2016
- Andrew Musselman presented: “Apache Mahout?! What’s Next!” at Seattle Data
Science Meetup, December 2016
- Suneel Marthi presented: “Native and Distributed Machine Learning with
Apache Mahout” Apache Big Data Europe 2016, Nov 13 2016, Seville, Spain
## Health report:
- The project has a dedicated team of voluntary committers.
## PMC changes:
- Currently 14 PMC members.
- No new PMC members added in the last 3 months
- Last PMC addition was Stevo Slavić on Tue Apr 21 2015
## Committer base changes:
- Currently 26 committers.
- No new committers added in the last 3 months
- Last committer addition was Trevor Grant at Tue May 24 2016
## Releases:
- Last release was 0.12.2 on Sun Jun 12 2016
## JIRA activity:
- 16 JIRA tickets created in the last 3 months
- 15 JIRA tickets closed/resolved in the last 3 months
Apache Mahout is an environment for quickly creating scalable performant machine learning applications. ## Issues: - None ## Activity: 1. Work is presently focused on adding support for Visualization, GPU and native optimization 2. Suneel Marthi and Trevor Grant did a Mahout on Flink talk at Flink Forward 2016, Berlin, Germany - September 13, 2016 3. Suneel Marthi did a Mahout talk at Department of Theoretical Physics, Fritz-Haber Institut der Max Planck Gessellschaft, Berlin, Germany - September 16, 2016 4. Suneel Marthi did a ‘Distributed Machine Learning with Apache Mahout’ talk at Big Data Ignite, Grand Rapids, Michigan - September 30, 2016 5. Upcoming Apache Mahout talk at Apache Big Data Europe, Seville, Spain - Nov 2016 6. Team presently working on 0.13.0 release planned for Oct 2016. ## Health report: - The health of the project is good with a devoted team of committers. ## PMC changes: - Currently 14 PMC members. - Last PMC addition was Stevo Slavić on Tue Apr 21 2015. ## Committer base changes: - Currently 26 committers. ## Releases: - Mahout 0.12.2 released on June 12, 2016
## Description: Apache Mahout is an environment for quickly creating scalable performant machine learning applications. ## Issues: - None ## Activity: 1. Work is presently focused on adding support for Visualization and Native optimization. 2. Suneel Marthi did talks on Apache Mahout at Apache Big Data 2016, Vancouver [1] and MapR BigData EveryWhere, Washington DC [2]. 3. Integration of Mahout with Apache Zeppelin being worked on by Trevor Grant [3]. 4. Presently working towards 0.13.0 release that would add native optimizations. ## Health report: - The health of the project is good with a devoted team of committers. ## PMC changes: - Currently 14 PMC members. - Last PMC addition was Stevo Slavić on Tue Apr 21 2015. ## Committer base changes: - Currently 26 committers. - Trevor Grant was added as a committer on Tue May 24 2016. ## Releases: - 0.12.1 was released on Wed May 18 2016. - 0.12.2 was released on Mon Jun 13 2016. ## JIRA activity: - 46 JIRA tickets created in the last 3 months. - 26 JIRA tickets closed/resolved in the last 3 months. [1]http://events.linuxfoundation.org/events/apache-big-data-north-america/program/schedule [2]http://www.bigdataeverywhere.com/dcarea-hadoop-conference-2016/#t0 [3]https://trevorgrant.org/2016/05/19/visualizing-apache-mahout-in-r-via-apache-zeppelin-incubating/
WHEREAS, the Board of Directors heretofore appointed Suneel Marthi (smarthi) to the office of Vice President, Apache Mahout, and WHEREAS, the Board of Directors is in receipt of the resignation of Suneel Marthi from the office of Vice President, Apache Mahout, and WHEREAS, the Project Management Committee of the Apache Mahout project has chosen by vote to recommend Andrew Palumbo (apalumbo) as the successor to the post; NOW, THEREFORE, BE IT RESOLVED, that Suneel Marthi is relieved and discharged from the duties and responsibilities of the office of Vice President, Apache Mahout, and BE IT FURTHER RESOLVED, that Andrew Palumbo be and hereby is appointed to the office of Vice President, Apache Mahout, to serve in accordance with and subject to the direction of the Board of Directors and the Bylaws of the Foundation until death, resignation, retirement, removal or disqualification, or until a successor is appointed. Special Order 7H, Change the Apache Mahout Project Chair, was approved by Unanimous Vote of the directors present.
The goal of Apache Mahout project is to build an environment for quickly
creating scalable performant machine learning applications.
Activity:
- New Apache Mahout book - “Apache Mahout: Beyond MapReduce” authored by
Mahout committers - Dmitriy Lyubimov and Andrew Palumbo, published by
Createspace on February 18, 2016 (1)
- Apache Mahout 0.11.2 was released on March 11, 2016, this release
introduced major performance enhancements for linear algebra computations
and also supports Apache Spark 1.5.2.
- Apache Mahout 0.12.0 was released on April 11, 2016.
This release adds Apache Flink as an execution engine to Mahout Samsara.
With the milestone 0.12.0 release, Mahout now supports Spark, Flink and
H2O.
- Suneel Marthi will be doing a talk on the new Mahout Distributed Linear
Algebra at Apache Big Data, Vancouver on May 11, 2016 (2)
PMC changes:
- Currently 14 PMC members.
- No new PMC members added in the last 3 months
- Last PMC addition was Stevo Slavić on Tue Apr 21 2015
Committer base changes:
- Currently 25 committers.
- No new committers added in the last 3 months
- Last committer addition was Anand Avati on Thu Apr 23 2015
Releases:
- Mahout 0.11.2 was released on Fri Mar 11 2016
- Mahout 0.12.0 was released on Mon Apr 11 2016
Issues:
None
JIRA activity:
- 34 JIRA tickets created in the last 3 months
- 71 JIRA tickets closed/resolved in the last 3 months
Mailing list activity:
- dev@mahout.apache.org:
- 947 subscribers (down -6 in the last 3 months):
- 587 emails sent to list (434 in previous quarter)
- user@mahout.apache.org:
- 1878 subscribers (down -16 in the last 3 months):
- 141 emails sent to list (114 in previous quarter)
[1] http://www.amazon.com/Apache-Mahout-MapReduce-Dmitriy-Lyubimov/dp/1523775785
[2] http://events.linuxfoundation.org/events/apache-big-data-north-america/program/schedule
The goal of Apache Mahout project is to build an environment for quickly creating scalable performant machine learning applications. Activity: Apache Mahout 0.11.1 was released on Nov 6, 2015. This release supports Spark 1.4+ and has major performance improvements for vector and matrix operations. Sebastian Schelter presented the new Mahout distributed linear algebra framework at Flink Forward, Berlin On October 12, 2015. [1] Present activity is restricted to finalizing the Flink - Mahout integration which would be Mahout 1.0 release and to bolster the performance of the backend linear algebra by rebasing the code with alternate native implementations. PMC changes: - Currently 14 PMC members. - No new PMC members added in the last 3 months - Last PMC addition was Stevo Slavić on Tue Apr 21 2015 Committer base changes: - Currently 25 committers. - No new committers added in the last 3 months - Last committer addition was Anand Avati at Thu Apr 23 2015 Releases: - Mahout 0.11.1 was released on Fri Nov 06 2015 Issues: Decline in the project user and developer base over the past 2 years, in large part due to the availability of competing Machine Learning libraries with very active developer teams backed by organizations. Its hard to sustain a Machine Learning project on voluntary basis with no dedicated resources and yet be relevant with changing times and increasing competition. In the past, there was some promise of dedicated resources from organizations but nothing promising enough. JIRA activity: - 22 JIRA tickets created in the last 3 months - 35 JIRA tickets closed/resolved in the last 3 months [1] https://www.youtube.com/watch?v=Uh92PK0K0mA
The goal of Apache Mahout project is to build an environment for quickly creating scalable performant machine learning applications. ISSUES FOR BOARD'S ATTENTION None at this time. RELEASES - 0.10.2 was released on Aug 6, 2015 - 0.11.0 was released on Aug 7, 2015 ACTIVITY No new PMC members or Committers added in the last 3 months. Last PMC addition was Stevo Slavic on April 21, 2015. Sebastian Schelter will be presenting the new Mahout-Samsara Linear Algebra framework at the upcoming Flink Forward conference in Berlin on October 12, 2015. [1] 0.10.2 was released on Aug 6, 2015. This release had major optimizations and performance improvements to the new Samsara Linear Algebra backend. 0.11.0 was released on Aug 7, 2015. This release makes Mahout compatible with Spark 1.3.1. Mahout 0.11.0 has been integrated with Apache BigTop 1.0.1. Integration of Apache Mahout with Apache Flink is presently in the works and is being done in collaboration with TU Berlin and Data Artisans. Apache Mahout has been recognized as one of the 5 Big Data Open Source projects to watch out for in a ZDNet article dated Aug 21, 2015. [2] STATS 25 committers 14 PMC members 19 JIRA tickets created in last 3 months 30 JIRA tickets closed/resolved in last 3 months [1]http://www.flink-forward.org [2]http://www.zdnet.com/article/five-open-source-big-data-projects-to-watch/
DESCRIPTION: The goal of Apache Mahout project is to build an environment for quickly creating scalable performant machine learning applications. ACTIVITY: - Apache Mahout’s next generation 0.10.0 was released on April 11, 2015. - Apache Mahout 0.10.1 was released on May 31, 2015. This was a minor bug fix release following 0.10.0. - Apache Mahout now supports scalable Machine Learning on Spark, H2O and MapReduce. - The project has been working closely with Apache BigTop to integrate Apache Mahout into BigTop following a release. - Integration of Apache Mahout with Apache Flink is in the works and is being done in collaboration with Data Artisans and TU Berlin. - Anand Avati was added as a new committer. - Stevo Slavic was added as a PMC member. - Team presently working on 0.10.2 release, planned for the week of July 10, 2015. ISSUES: - Lately most design and tech discussions have been happening off the dev@ mailing lists, the PMC is well aware of the issue and working on addressing that. PMC/Committership changes: - Currently 25 committers and 14 PMC members in the project. - Stevo Slavić was added to the PMC on Fri May 08 2015 - Anand Avati was added as a committer on Thu Apr 23 2015 RELEASES: - 0.10.1 was released on Sun May 31 2015 - 0.10.0 was released on Sat Apr 11 2015 MAILING LIST ACTIVITY: - dev@mahout.apache.org: - 977 subscribers (down -8 in the last 3 months): - 1324 emails sent to list (1419 in previous quarter) - user@mahout.apache.org: - 1933 subscribers (down -10 in the last 3 months): - 243 emails sent to list (252 in previous quarter) - general@mahout.apache.org: - 10 subscribers (up 0 in the last 3 months): - 0 emails sent to list (0 in previous quarter) JIRA ACTIVITY: - 85 JIRA tickets created in the last 3 months - 74 JIRA tickets closed/resolved in the last 3 months
WHEREAS, the Board of Directors heretofore appointed Grant Ingersoll to the office of Vice President, Apache Mahout, and WHEREAS, the Board of Directors is in receipt of the resignation of Grant Ingersoll from the office of Vice President, Apache Mahout, and WHEREAS, the Project Management Committee of the Apache Mahout project has chosen by vote to recommend Suneel Marthi as the successor to the post; NOW, THEREFORE, BE IT RESOLVED, that Grant Ingersoll is relieved and discharged from the duties and responsibilities of the office of Vice President, Apache Mahout, and BE IT FURTHER RESOLVED, that Suneel Marthi be and hereby is appointed to the office of Vice President, Apache Mahout, to serve in accordance with and subject to the direction of the Board of Directors and the Bylaws of the Foundation until death, resignation, retirement, removal or disqualification, or until a successor is appointed. Special Order 7G, Change the Apache Mahout Chair, was approved by Unanimous Vote of the directors present.
Apache Mahout has implementations of a wide range of machine learning and data mining algorithms: clustering, classification, collaborative filtering and frequent pattern mining. Project Status -------------- The project continues to have a large and active user base. The project now has integrations with Spark and H2O execution engines, this is in addition to the traditional MapReduce. Integration with Apache Flink is next on the cards with a possibly dedicated resource available from the Flink community to work with Mahout. The new integrations with H2O and Spark engines extend Mahout Machine Learning to other more popular Big Data platforms. Community --------- * We have added 3 new PMC members: Pat Ferrel, Andrew Musselman and Andrew Palumbo There is a healthy committer base to the project that are actively working on the project on a voluntary basis. There is no dedicated full time resource available for the project yet as most large scale Machine Learning libraries cannot be built and sustained on voluntary contributions. Community Objectives -------------------- The project has an active committer base and there’s a renewed interest in the project with the new Scala based Engine agnostic distributed linear algebra library with bindings for Spark, H2O and Flink in the future. The project got a shot in the arm with backing from Apache BigTop community and we are looking to keep that momentum going for future releases. The project is targeting more frequent minor releases and a major release once every quarter. While the 0.10.0 release is targeted for the week of April 7-11 2015, a subsequent 0.10.1 release is planned in the subsequent releases. Releases -------- The team is working towards Mahout 0.10.0 release targeted for the week of April 7-11 in time for ApacheCon North America 2015. Issues ------ None now.
=== Apache Mahout Status Report: February (missed January) 2015 === ----- Apache Mahout has implementations of a wide range of machine learning and data mining algorithms: clustering, classification, collaborative filtering and frequent pattern mining Project Status -------------- The project continues to have a large and active user base. Development continues by a small number of dedicated individuals. The PMC is reviewing how we can improve contributions as well as exploring other options to make sure the project remains viable to the user base. Community --------- * As per the status, the main issue is we have only 2-3 committers who are contributing on a regular basis. While they are doing good work, it is concerning from a sustainment issue. We are discussing as a PMC how to rectify this situation. The main issue is that developing machine learning libraries is involved process that is hard to do on a part time basis and we have yet to find anyone that can be dedicated full time to the project. Community Objectives -------------------- Identify next steps for either growing the list of active committers or finding an appropriate home for the code that exists (attic or elsewhere). Releases -------- The migration to Spark is still ongoing and no new releases are planned at this time. Issues ------ See above.
No report was submitted.
Apache Mahout has implementations of a wide range of machine learning and data mining algorithms: clustering, classification, collaborative filtering and frequent pattern mining Project Status -------------- The project continues to have a large and active user base. Development continues at a steady pace. Community --------- * The main issue concerning the community right now is the addition of new contributions from 0xData and the integration of Mahout with Scala/Spark. Community Objectives -------------------- Our goal is to build scalable machine learning libraries. See the Issues section below for the debate in the community about our objectives. Releases -------- The migration to Spark is still ongoing and no new releases are planned at this time. Issues ------ The community is still actively working on converting the codebase to Scala and Spark. The number of devs contributing is still small, but it is sustained.
Apache Mahout has implementations of a wide range of machine learning and data mining algorithms: clustering, classification, collaborative filtering and frequent pattern mining Project Status -------------- The project continues to have a large and active user base. While the developer base has continued to grow, there is a very active and healthy debate going on about where Mahout goes next. We have worked through many of these issues, but are not out of the proverbial woods just yet. Community --------- * Andrew Palumbo and Pat Ferrel are new committers * Dmitriy Lyubimov has resigned from the PMC * The main issue concerning the community right now is the addition of new contributions from 0xData and the integration of Mahout with Scala/Spark. Community Objectives -------------------- Our goal is to build scalable machine learning libraries. See the Issues section below for the debate in the community about our objectives. Releases -------- In addition to an ongoing debate on Mahout's future, the community is actively working on integrating Mahout with Scala/Spark, and bringing in new code and committers to update the core project. Issues ------ For the most part, the community has gotten back to work by adding a couple of new committers and pursuing the path of Scala support. While there is still not a huge developer base, people are contributing and working through the issues.
Apache Mahout has implementations of a wide range of machine learning and data mining algorithms: clustering, classification, collaborative filtering and frequent pattern mining Project Status -------------- The project continues to have a large and active user base. While the developer base has continued to grow, there is a very active and healthy debate going on about where Mahout goes next. Please see the Issues section below for more details. Community --------- * Andrew Musselman was voted in as new committer. * No changes to the PMC in the reporting period. * The main issue concerning the community right now is the addition of new contributions from 0xData and the integration of Mahout with Spark. Community Objectives -------------------- Our goal is to build scalable machine learning libraries. See the Issues section below for the debate in the community about our objectives. Releases -------- In addition to an ongoing debate on Mahout's future, the community is actively working on integrating Mahout with Scala/Spark, and bringing in new code and committers to update the core project. A lot of work on improving documentation has been done. The project has finished the move from the wiki to Apache CMS, redesigned the project website and is in the process of updating all pages. Issues ------ The Mahout community is at a crossroads in terms of where to go next. While the project has a broad number of users and interested parties, most committers are trying to maintain the code base on a purely part time basis, when the amount of work to sustain these users clearly points to it needing to be full time. Furthermore, much of our original code base is written for Hadoop MapReduce 1.0, which many in the community have come to realize is not well-suited for solving the kinds of problems that Mahout has set out to solve. There have been several lengthy discussions and prototypes going on to work out next directions along the lines of the Spark and 0xData contributions (there are numerous threads on the dev@mahout.a.o mailing list.) The PMC does not think this requires Board intervention at this time as the debate is, as far as we can tell, healthy. We do, however, expect that this debate will take some time to resolve and may mean we won't be shipping a 1.0 release any time soon. We will keep the Board apprised of our next steps as we work through the process.
Apache Mahout has implementations of a wide range of machine learning and data mining algorithms: clustering, classification, collaborative filtering and frequent pattern mining Project Status -------------- The project continues to have a large and active user base and the developer base continues to grow, as well. Community --------- * On November 28th Frank Scholten was voted in as new committer. * No changes to the PMC in the reporting period. * With Suneel Marthi now working full time on the project there has been a flurry of patches reviewed and committed. * The project has moved to Apache CMS, is in the process of tidying most of the wiki based documentation. * After a small Hackathon in Berlin pre-Christmas activity has been steady even during the holiday season. Community Objectives -------------------- With most committers not working on Mahout full time there is always a lack of time on lists as well as when it comes to dealing with patches submitted quickly. The current goal is to grow the committer base to deal with that issue. As for students that would like to contribute the problem remains that the most interesting work seems to be adding new algorithms and implementations. It remains a challenge to motivate those interested in contributing to work on getting existing implementations stable, improving documentation and reviewing incoming patches. Releases -------- The community is actively working on getting the 0.9 release out the door with just one scaling issue remaining the the k-means++ code newly added as part of the 0.8 release (June 2013). This is supposed to be the last release before 1.0. Issues ------ There are no issues requiring board attention at this time.
Apache Mahout has implementations of a wide range of machine learning and data mining algorithms: clustering, classification, collaborative filtering and frequent pattern mining Project Status -------------- The project continues to have a large and active user base and the developer base continues to grow, as well. Suneel Marthi was added to the PMC. Community --------- The third quarter of 2013 has seen continued activity on par with the last report. We are primarily working on 0.9 release, some new recommendation integration with Solr. The user list is quite active with a mix of new and experienced users. No new committers have been added since the last report. If all goes well one of the committers will be having her first baby early April 2014. Patches/commits from her will need some extra careful review from the community. [Disclaimer: Due to timing issues this amendment was added to the report after it was submitted by the committer in question. Sorry for the additional noise.] Community Objectives -------------------- Our main focus is on cleanup and preparation of 0.9 and 1.0 releases, as well as the usual bug fixes. Releases -------- None since last report. Next likely one is sometime between Nov. '13 and Jan. '14. Issues ------ There are no issues requiring board attention at this time.
WHEREAS, the Board of Directors heretofore appointed Jake Mannix to the office of Vice President, Apache Mahout, and WHEREAS, the Board of Directors is in receipt of the resignation of Jake Mannix from the office of Vice President, Apache Mahout, and WHEREAS, the Project Management Committee of the Apache Mahout project has chosen by vote to recommend Grant Ingersoll as the successor to the post; NOW, THEREFORE, BE IT RESOLVED, that Jake Mannix is relieved and discharged from the duties and responsibilities of the office of Vice President, Apache Mahout, and BE IT FURTHER RESOLVED, that Grant Ingersoll be and hereby is appointed to the office of Vice President, Apache Mahout, to serve in accordance with and subject to the direction of the Board of Directors and the Bylaws of the Foundation until death, resignation, retirement, removal or disqualification, or until a successor is appointed. Special Order 7C, Change the Apache Mahout Project Chair, was approved by Unanimous Vote of the directors present.
Apache Mahout has implementations of a wide range of machine learning and data mining algorithms: clustering, classification, collaborative filtering and frequent pattern mining Project Status -------------- The project continues to have a large and active user base. With the book Mahout in Action it has become simpler for beginners to get started using the project. Community --------- The third quarter of 2013 was seen continued activity on par with the second quarter, with our first new release in more than a year, a new committer, and our first live google hangout for users and developers. There has been continued effort fixing bugs and reviewing contributor patches, especially with the recent release. We added one new committer to the project: Ellen Friedman There is a SF-Bay Area Mahout MeetUp scheduled for August 27 in Redwood City. Sebastian Schelter will be the main speaker, talking about new directions with Mahout recommendation. Grant Ingersoll, Ted Dunning and Ellen Friedman be there to do a short introduction for the meet-up and update on the 0.8 release. Community Objectives -------------------- Discussions regarding the 0.9 planning and 1.0 release has continued on the mailing list, revolving significantly around what features/algorithms will be supported in 1.0 and onward, with an eye toward streamlining the scope of the project to not contain as many rarely used / unsupported algorithms. The PMC and especially the PMC Chair apologize for missing the last several Board Reports, and we have discussed internally as a PMC the need for a new PMC chair who is a bit more "bureaucratically minded", and with several experienced volunteers stepping forward, we should be calling a vote and moving forward with this by the end of August. Releases -------- Mahout 0.8 was released in July, see below for details, and https://cwiki.apache.org/confluence/display/MAHOUT/Release+0.8 for release notes. Code ---- The 0.8 release contains one significant new algorithm implementation, Streaming K-Means ( MAHOUT-1154 ), as well as numerous performance enhancements and API improvements to the core linear algebra library and many bugfixes. Additionally, two new directions have started up, regarding visualization of recommender and co-occurrence calculations (http://s.apache.org/mahout_viz_thread); and creating a scala DSL for some Mahout calculations (http://s.apache.org/mahout_scala_dsl). Both of these are at the design and prototyping phase, but seem promising. Issues ------ There are no issues requiring board attention at this time.
No report was submitted.
Report was not received and is expected next month.
No report was submitted.
AI: Doug to pursue a report for Mahout
Apache Mahout has implementations of a wide range of machine learning and data mining algorithms: clustering, classification, collaborative filtering and frequent pattern mining Project Status -------------- The project continues to have a large and active user base. With the book Mahout in Action it has become simpler for beginners to get started using the project. Community --------- The second quarter of 2013 was relatively more active, with many committers and PMC members fixing bugs, reviewing contributor patches, and slowly removing old dead code. We added four new committers to the project: Suneel Marthi, Dan Filimon, Gokhan Capan, and Stevo Slavic. There are a few committers who volunteered to become GSoC mentors. As for them it will be the first year participating as mentors on behalf of Mahout they will need some guidance on what the process looks like at the ASF. Community Objectives -------------------- Discussions regarding the 0.9 planning and 1.0 release happened in person among many of the core committers at Berlin Buzzwords, and has continued on the mailing list, revolving significantly around what features/algorithms will be supported in 1.0 and onward, with an eye toward streamlining the scope of the project to not contain as many rarely used / unsupported algorithms. The PMC and especially the PMC Chair apologize for missing the last two Board Reports, and we have discussed internally as a PMC whether we should make any changes and are working to make sure it doesn't happen again. Code ---- The upcoming 0.8 release contains one significant new algorithm implementation, Streaming K-Means ( MAHOUT-1154 ), as well as numerous performance enhancements and API improvements to the core linear algebra library and many bugfixes. Releases -------- No releases since the last report. 0.8 is targeted for the end of June, and currently bugfixes are the primary focus. Only two open issues remaining at the time of this writing ( http://s.apache.org/mahout_0.8_issues ) Issues ------ There are no issues requiring board attention at this time.
No report was submitted.
AI: Ross to pursue a report for Mahout
No report was submitted.
AI: Ross to pursue a report for Mahout
Apache Mahout provides implementations of machine learning algorithms (collaborative filtering, clustering, classification, and more) for large-scale data, mostly via Hadoop-based implementations. Issues: Sean Owen wishes to leave the Mahout PMC (but retain his commit rights), but this is the only issue which needs the Board attention. Current Activity: How has the community developed since the last report? In February: Originally planned for 0.8 release by March 8, but will be letting that slip forward a few weeks. Selection of Presentations, Articles and Outreach: * Ted Dunning on new fast streaming clustering (http://www.slideshare.net/tdunning/news-frommahout20130305) * Fast clustering at ACM http://www.slideshare.net/tdunning/acm-20130225 * Real time learning http://www.slideshare.net/tdunning/real-time-learning * MapR-Lucidworks on reflected intelligence http://www.slideshare.net/tdunning/mapr-lucidworks-joint-webinar * Ted Dunning at Strata on Mahout http://www.slideshare.net/tdunning/strata-newyork2012 * Ted Dunning on fast clustering at Oxford http://www.slideshare.net/tdunning/oxford-05oct2012 * MapR and Amex speak about large-scale analytics with Mahout http://www.slideshare.net/tdunning/customer-analysisatscalestrata10022012 * Overstock and Mahout http://www.wired.com/wiredenterprise/2012/12/mahout/ * Advanced Analytics in Mahout http://portfortune.wordpress.com/2012/12/05/advanced-analytics-in-hadoop-part-one * London Data Science http://datasciencelondon.org/tag/mahout/ * Mahout Updated in CDH 4.1 http://blog.cloudera.com/blog/2012/11/whats-new-in-cdh4-1-mahout/ Scientific publications based on Mahout * Sebastian Schelter, Sean Owen: Collaborative Filtering with Apache Mahout, Recommender Systems Challenge Workshop in conjunction with ACM RecSys 2012 http://ssc.io/wp-content/uploads/2013/02/cf-mahout.pdf * Sebastian Schelter, Christoph Boden, Volker Markl: Scalable Similarity-Based Neighborhood Methods with MapReduce, ACM Conference on Recommender Systems 2012, Dublin http://dl.acm.org/citation.cfm?id=2365984 http://ssc.io/wp-content/uploads/2012/06/rec11-schelter.pdf Code We were able to attract the developer of one of the leading scientific recommender libraries [http://mymedialite.net/] to port a few implementations to Mahout (https://issues.apache.org/jira/browse/MAHOUT-1106, https://issues.apache.org/jira/browse/MAHOUT-1089) However, new code contributions have slowed to a crawl, the number of commits in the past few months, compared to prior years: Feb 2013, 7 Jan 2013, 20 Dec 2012, 7 Feb 2012, 98 Jan 2012, 27 Dec 2011, 99 Feb 2011, 35 Jan 2011, 52 Dec 2010, 37 Feb 2010, 207 Jan 2010, 132 Dec 2009, 135 New Commercial Integrations * Predixion Readmission Insight, a "a preventable readmission healthcare solution" announced http://www.virtual-strategy.com/2013/03/05/predixion-software-wins-microsoft-health-users-group-innovation-award integration with Mahout, Greenplumb, Hive, and Microsoft's BI stack. * Overstock and Mahout http://www.wired.com/wiredenterprise/2012/12/mahout New Open Source Integrations * The recommendation and advertisement network http://www.plista.com/en has built an open source weblayer for Mahout's recommenders https://github.com/plista/kornakapi * Mahout seems to be the framework of choice for PredictionIO http://prediction.io/, an open source prediction server for software developers to create predictive features, such as personalization, recommendation and content discovery Mailing List Summary: User list discussions are currently focussed primarily on bug reporting and helping new users, but very little about future feature work. Developer Mailing List Posting: http://mail-archives.apache.org/mod_mbox/mahout-dev/ February 2013, 123 January 2013, 213 Dec 2012, 155 as compared to the same months in previous years: Feb 2012, 578 Jan 2012, 545 Dec 2011, 1079 and Feb 2011, 352 Jan 2011, 473 Dec 2010, 267 We've not had this low developer involvement since the first half of 2009. User Mailing List Posting http://mail-archives.apache.org/mod_mbox/mahout-user/ User list discussions are primarily in support of very new users, as well as bug reporting on released versions (0.6 and sometimes even 0.5), highlighting the need for 0.8 to be released. While the traffic to the user mailing list has gone down slightly from previous years: Feb 2012, 288 Jan 2012, 367 Feb 2011, 359 Jan 2011, 458 Feb 2010, 497 Jan 2010, 272 This is not a dramatic decrease, as there is still considerable interest in the user community. Summary: How has the project developed since the last report: A 1.0 release is not yet on the horizon. == Milestones == 1.) Working towards a 0.8 release 2.) Development on new, faster clustering code
No report was submitted.
No report was submitted.
WHEREAS, the Board of Directors heretofore appointed Jeff Eastman to the office of Vice President, Apache Mahout, and WHEREAS, the Board of Directors is in receipt of the resignation of Jeff Eastman from the office of Vice President, Apache Mahout, and WHEREAS, the Project Management Committee of the Apache Mahout project has chosen by vote to recommend Jake Mannix as the successor to the post; NOW, THEREFORE, BE IT RESOLVED, that Jeff Eastman is relieved and discharged from the duties and responsibilities of the office of Vice President, Apache Mahout, and BE IT FURTHER RESOLVED, that Jake Mannix be and hereby is appointed to the office of Vice President, Apache Mahout, to serve in accordance with and subject to the direction of the Board of Directors and the Bylaws of the Foundation until death, resignation, retirement, removal or disqualification, or until a successor is appointed. Special Order 7B, Change the Apache Mahout Project Chair, was approved by Unanimous Vote of the directors present.
Apache Mahout provides implementations of machine learning algorithms (collaborative filtering, clustering, classification, and more) for large-scale data, mostly via Hadoop-based implementations. ISSUES There are no issues requiring board attention at this time. CURRENT ACTIVITY Activity has remained high during the past 3 months. The user@mahout.a.o mailing list has 1448 current subscribers. The dev@mahout.a.o mailing list has 734 current subscribers. Now we are embarked upon a new 0.8 release. A goal of 0.8 is to continue clean up of existing functionality to improve consistency and improve user experience. In this release, some new additions to Mahout functionality are also planned. Code freeze for 0.8 is targeted for Nov 15. A 1.0 release is not yet on the horizon. COMMUNITY Jake Mannix has been elected to be the new Mahout PMC Chair. Paritosh Ranjan has been elected to the Mahout PMC. We have no new committers since our July report
Apache Mahout provides implementations of machine learning algorithms (collaborative filtering, clustering, classification, and more) for large-scale data, mostly via Hadoop-based implementations. ISSUES There are no issues requiring board attention at this time. CURRENT ACTIVITY Activity has remained high during the past 3 months. We completed our 0.7 release on June 16th that closed 63 JIRA issues. The user@mahout.a.o mailing list has 1379 current subscribers Now we are embarked upon a new 0.8 release. A goal of 0.8 is to continue clean up of existing functionality to improve consistency and improve user experience. In this release, some new additions to Mahout functionality are also planned. Code freeze for 0.8 is targeted for Nov 15. A 1.0 release is not yet on the horizon. COMMUNITY We have no new committers since our April report.
(Mahout)
Apache Mahout provides implementations of machine learning algorithms (collaborative filtering, clustering, classification, and more) for large-scale data, mostly via Hadoop-based implementations. ISSUES There are no issues requiring board attention at this time. CURRENT ACTIVITY Activity has remained high during the past 3 months. We completed our 0.6 release on Feb. 6th that closed 182 JIRA issues. The user@mahout.a.o mailing list has 1271 current subscribers The dev@mahout.a.o mailing list has 661 current subscribers Now we are embarked upon a new 0.7 release. The goal of 0.7 is to clean up and refactor existing functionality to improve consistency and improve user experience. Code freeze for 0.7 is targeted for May 15. A 1.0 release is not yet on the horizon. COMMUNITY We have two new committers since our January report: - Paritosh Ranjan - Tom Pierce MAHOUT DISTRIBUTIONS At least two commercially-supported Hadoop distributions now include Mahout in their offerings (Cloudera, MapR). We will keep an eye out to make sure they are distributed in accordance with Apache trademark guidelines. MAHOUT IN PRINT "Mahout in Action", Owen, Anil, Dunning & Friedman is being well received. (http://manning.com/owen/)
Apache Mahout provides implementations of machine learning algorithms (collaborative filtering, clustering, classification, and more) for large-scale data, mostly via Hadoop-based implementations. ISSUES There are no issues requiring board attention at this time. CURRENT ACTIVITY Activity has remained high during the past 3 months. We completed our 0.6 release on Feb. 6th that closed 182 JIRA issues. Now we are embarked upon a new 0.7 release. The goal of 0.7 is to clean up and refactor existing functionality to improve consistency and improve user experience. Code freeze for 0.7 is targeted for May 15. A 1.0 release is not yet on the horizon. COMMUNITY We have two new committers since our last report: - Paritosh Ranjan - Tom Pierce MAHOUT DISTRIBUTIONS Mahout now has multiple commercial distributions. MAHOUT IN PRINT "Mahout in Action", Owen, Anil, Dunning & Friedman is being well received. (http://manning.com/owen/)
AI: Shane ask PMC to update agenda
Apache Mahout provides implementations of machine learning algorithms (collaborative filtering, clustering, classification, and more) for large-scale data, mostly via Hadoop-based implementations. ISSUES There are no issues requiring board attention at this time. CURRENT ACTIVITY Activity has been high during the past 3 months as we have begun the release process for 0.6. Of 181 issues targeted for this release, there are only 4 remaining. Code freeze, originally targeted for Jan 1, is currently being delayed by these outstanding issues. We still expect to release 0.6 in the near future. A 1.0 release is not yet on the horizon. COMMUNITY There are no new committers since last report. Dmitriy Lyubimov has been elected a member of the Mahout PMC. MAHOUT DISTRIBUTIONS Mahout has been included in the Cloudera CDH3u2 release. (http://www.cloudera.com/blog/2011/11/cdh3u2-apache-mahout-integration) As with other commercial distributions we will keep an eye out to make sure it is distributed in accordance with Apache trademark guidelines. MAHOUT IN PRINT "Mahout in Action", Owen, Anil, Dunning & Friedman has been published and is being well received. (http://manning.com/owen/)
WHEREAS, the Board of Directors heretofore appointed Sean Owen to the office of Vice President, Apache Mahout, and WHEREAS, the Board of Directors is in receipt of the resignation of Sean Owen from the office of Vice President, Apache Mahout, and WHEREAS, the Project Management Committee of the Apache Mahout project has chosen by vote to recommend Jeff Eastman as the Successor to the post; NOW, THEREFORE, BE IT RESOLVED, that Sean Owen is relieved and discharged from the duties and responsibilities of the office of Vice President, Apache Mahout, and BE IT FURTHER RESOLVED, that Jeff Eastman be and hereby is appointed to the office of Vice President, Apache Mahout, to serve in accordance with and subject to the direction of the Board of Directors and the Bylaws of the Foundation until death, resignation, retirement, removal or disqualification, or until a successor is appointed. Special Order 7C, Resolution to Change the Apache Mahout Project Chair, was approved by Unanimous Vote of the directors present.
Apache Mahout provides implementations of machine learning algorithms (collaborative filtering, clustering, classification, and more) for large-scale data, mostly via Hadoop-based implementations. ISSUES There are no issues requiring board attention at this time. CURRENT ACTIVITY Activity has been moderate during the past 3 months. There were no new releases, and the 0.6 release process is not yet begun, though will likely start within the next 2 months. Judging by Fixed issue count, 0.6 is about 60% as far along as previous releases. A 1.0 release is not yet on the horizon. MAHOUT DISTRIBUTIONS It appears that Mahout will be bundled with Cloudera soon. (https://groups.google.com/a/cloudera.org/group/cdh-user/ browse_thread/thread/5df8c1cb6d39288d?pli=1) As with other commercial distributions we'll keep an eye out to make sure it's distributed in accordance with Apache trademark guidelines. MAHOUT IN PRINT Mahout in Action has at last been published. http://manning.com/owen/
Bertrand notes that the community section is missing.
AI Bertrand: ask Mahout PMC chair for a community report next time
=== Apache Mahout Status Report: July 2011 === ISSUES There are no issues requiring board attention at this time. CURRENT ACTIVITY Apache Mahout 0.5 was released on May 27 2011. It resolved 137 issues: https://issues.apache.org/jira/secure/IssueNavigator.jspa?reset=true &jqlQuery=project+%3D+MAHOUT+AND+fixVersion+%3D+%220.5%22 The PMC plans an 0.6 release at the end of the year. The focus continues to be on polish and refinement in advance of a 1.0 release; A 1.0 release may come in mid 2012 but is not yet being planned. The community continues to grow steadily. The user and dev lists contained 793 and 470 subscribers, respectively, in January 2011. They now contain 983 and 557 respectively. We've seen healthy community activity around the world, including new talks at events from Berlin, Seoul, London and Chicago. The project has one area of significant new activity: graph mining and graph-related algorithms. For example, Mahout has a PageRank-like implementation now. MAHOUT PMC Sebastian Schelter was added to the PMC in May 2011. PROJECT BRANDING The project made changes to comply with Apache branding guidelines earlier in the year, but reconfirms that the site is in compliance with http://www.apache.org/foundation/marks/pmcs#checklist
=== Apache Mahout Status Report: April 2011 === ISSUES There are no issues requiring board attention at this time. CURRENT ACTIVITY The project expects to continue with an 0.5 release around May 2011. 115 issues have been resolved for 0.5, with 7 more planned before the release: https://issues.apache.org/jira/secure/IssueNavigator.jspa?reset=true &jqlQuery=project+%3D+MAHOUT+AND+fixVersion+%3D+%220.5%22 After that, we believe, will be a 1.0 release, though it is possible the PMC will elect to issue an interim 0.6 release later in the year. The focus will change to making the code base stable and 1.0-ready. NEW MAHOUTS Apache Mahout added Dmitriy Lyubimov and Shannon Quinn as new committers in February 2011. MAHOUT ON THE GO The community has recorded 12 talks on Mahout since the last release, a substantial increase in volume and diversity: https://cwiki.apache.org/MAHOUT/books-tutorials-and-talks.html MAHOUT IN PRINT The book "Mahout in Action", published by Manning, has been completed and will be published in July 2011. The book "Taming Text", also published by Manning, is also nearing completion and contains substantial coverage of Mahout and text clustering.
ISSUES There are no issues requiring board attention at this time. CURRENT ACTIVITY Apache Mahout released version 0.4 on October 31, 2010. 0.4 included changes related to 153 issues, summarized here: https://issues.apache.org/jira/browse/MAHOUT/fixforversion/12314396 It continues to change significantly and across the board, though a certain consistent scope and identity is confirming itself at this stage. It is a Java-based scalable data mining library that currently has much of its implementation based on Apache Hadoop 0.20.x. It currently covers, primarily, collaborative filtering, clustering, classification, frequent itemset mining, and some related and supporting algorithms. The project expects to continue with an 0.5 release around May 2011. The 57 issues to date that are resolved or are being worked on for 0.5 are: https://issues.apache.org/jira/secure/IssueNavigator.jspa?pid=12310751 &fixfor=12315255 After that, we believe, will be a 1.0 release. From 0.5, the focus will change to making the code base stable and 1.0-ready. MAHOUT IN ACTION The book "Mahout in Action", published by Manning, has been completed and will be published in February 2011.
ISSUES There are no issues requiring board attention at this time. CURRENT ACTIVITY The project is in "code freeze" leading up to a final 0.4 release planned for this week. The 150 issues resolved for this release can be viewed here: https://issues.apache.org/jira/secure/IssueNavigator.jspa? pid=12310751&fixfor=12314396 As of 0.4, the project will still be in a state of significant change and evolution. We still plan an 0.5 release in 6 months before contemplating a 1.0 release. However we believe the project's code base is beginning to stabilize, as relatively more effort is going into code cleanup, tests, polishing, removal of stale code. Judging by volume of mailing list messages and diversity of senders we have reason to believe usage of Apache Mahout is beginning to significantly expand. NEW COMMITTERS Sebastian Schelter was elected as a new committer in recognition of work on distributed recommender implementations. GOOGLE SUMMER OF CODE Mahout completed its GSoC projects. Two did not complete due to lack of student participation. Two completed successfully. One remains in progress. MAHOUT IN ACTION The book "Mahout in Action", published by Manning, has reached 15/16 chapters complete and will soon enter final review. PROJECT BRANDING We've reviewed the Apache Mahout home page (http://mahout.apache.org) just this week, per the e-mail request regarding branding. Project committer Robin Anil is addressing the following issues in this regard: - Add standard www.apache.org links to navigation - Ensure "TM" is used appropriate in names and logos - Add a DOAP file (we are having issues with the generator but that can be taken up offline)
Shane appreciates Mahout's being proactive on implementing the new branding policy.
=== Mahout Status Report: July 2010 === ISSUES There are no issues requiring board attention at this time. CURRENT ACTIVITY The project continues to target September, 2010 for release of version 0.4. This is unchanged since the last report. Recent activity in the project can be viewed here: https://issues.apache.org/jira/secure/IssueNavigator.jspa? pid=12310751&fixfor=12314396&resolution=1 WEBSITE The project's website at mahout.apache.org has been completely redesigned: http://mahout.apache.org/ GOOGLE SUMMER OF CODE As part of Google's Summer of Code program, Mahout is halfway through mentoring five projects. The projects will add or enhance capability in the specific areas of: - Boltzmann Machines - Support Vector Machines - Singular Value Decomposition for recommendations - Neural network with back propagation learning - Eigencuts spectral clustering MAHOUT IN ACTION The book "Mahout in Action", published by Manning, continues to be written and is in 2/3 completion review with the publisher. EXTERNAL EVENTS Mahout's recommender system was presented in the key note and two talks at the Berlin Buzzwords 2010 event.
Jim complemented the project on the format of their report.
ISSUES There are no issues requiring board attention at this time. CURRENT ACTIVITY The project continues to target September, 2010 for release of version 0.4. Recent activity in the project can be viewed here: https://issues.apache.org/jira/secure/IssueNavigator.jspa?pid=12310751&fixfor=12314396&resolution=1 In particular: - First real support for distributed recommenders has been released The project has completed migration of mailing lists and website to mahout.apache.org. GOOGLE SUMMER OF CODE As part of Google's Summer of Code program, Mahout has begun work mentoring five projects. The projects will add or enhance capability in the specific areas of: - Boltzmann Machines - Support Vector Machines - Singular Value Decomposition for recommendations - Neural network with back propagation learning - Eigencuts spectral clustering MAHOUT IN ACTION The book "Mahout in Action", published by Manning, continues to be written and is entering 2/3 completion review with the publisher.
=== Mahout Status Report: May 2010 === (This is the first report from Mahout as a top-level Apache project; previously it was a subproject of Apache Lucene. Mahout recently reported status with Lucene's special April report. We take the opportunity to summarize Mahout state and restate recent activity.) ISSUES There are no issues requiring board attention at this time. OVERVIEW Mahout's goal is to build scalable implementations of machine learning and data mining algorithms. "Scalable" means designed with exceptional scale in mind, for efficiency and low memory consumption, and in many cases means providing Hadoop-based implementations. The "machine learning" implemented to date has been primarily in the broad areas of: - Collaborative filtering / recommender engines - Clustering - Classification - Frequent item set mining - Evolutionary algorithms CURRENT ACTIVITY Mahout has created a release approximately every six months, most recently releasing version 0.3 in March 2010. The project remains in a state of rapid change and evolution, and looks to release 0.4 in September, 2010. Recent activity in the project can be viewed here: https://issues.apache.org/jira/secure/IssueNavigator.jspa? pid=12310751&fixfor=12314396&resolution=1 This month, Mahout will complete migration of website, mailing lists, SVN, and other information to reflect its status as a top-level project. GOOGLE SUMMER OF CODE Mahout will mentor five projects as part of Google's Summer of Code program. The projects will add or enhance capability in the specific areas of: - Boltzmann Machines - Support Vector Machines - Singular Value Decomposition for recommendations - Neural network with back propagation learning - Eigencuts spectral clustering MAHOUT IN ACTION The book "Mahout in Action", published by Manning, continues to be written and is approximately half complete. It has received some favorable feedback via Manning's early access program.
Great progress!
WHEREAS, the Board of Directors deems it to be in the best interests of the Foundation and consistent with the Foundation's purpose to establish a Project Management Committee charged with the creation and maintenance of open-source software related to a machine learning platform for distribution at no charge to the public. NOW, THEREFORE, BE IT RESOLVED, that a Project Management Committee (PMC), to be known as the "Apache Mahout Project", be and hereby is established pursuant to Bylaws of the Foundation; and be it further RESOLVED, that the Apache Mahout Project be and hereby is responsible for the creation and maintenance of software related to a machine learning platform; and be it further RESOLVED, that the office of "Vice President, Apache Mahout" be and hereby is created, the person holding such office to serve at the direction of the Board of Directors as the chair of the Apache Mahout Project, and to have primary responsibility for management of the projects within the scope of responsibility of the Apache Mahout Project; and be it further RESOLVED, that the persons listed immediately below be and hereby are appointed to serve as the initial members of the Apache Mahout Project: * Abdelhakim Deneche <adeneche@apache.org>> * Isabel Drost <isabel@apache.org> * Ted Dunning <tdunning@apache.org> * Jeff Eastman <jeastman@apache.org> * Drew Farris <drew@apache.org> * Grant Ingersoll <gsingers@apache.org> * Benson Margulies <bimargulies@apache.org> * Sean Owen <srowen@apache.org> * Robin Anil <robinanil@apache.org> * Jake Mannix <jmannix@apache.org> NOW, THEREFORE, BE IT FURTHER RESOLVED, that Sean Owen be appointed to the office of Vice President, Apache Mahout, to serve in accordance with and subject to the direction of the Board of Directors and the Bylaws of the Foundation until death, resignation, retirement, removal or disqualification, or until a successor is appointed; and be it further RESOLVED, that the initial Apache Mahout PMC be and hereby is tasked with the creation of a set of bylaws intended to encourage open development and increased participation in the Apache Mahout Project; and be it further RESOLVED, that the Apache Mahout Project be and hereby is tasked with the migration and rationalization of the Apache Lucene Mahout sub-project; and be it further RESOLVED, that all responsibilities pertaining to the Apache Lucene Mahout sub-project encumbered upon the Apache Lucene Project are hereafter discharged. Special Order 7A, Establish the Apache Mahout Project, was approved by Unanimous Vote of the directors present.