Placeholder Image

Subtitles section Play video

  • JOANA CARRASQUEIRA: Hello, everybody.

  • My name is Joana Carrasqueira, and I'm a program manager

  • for TensorFlow.

  • Today, I'm very excited to bring you

  • the greatest and the latest about our community.

  • As you can see, our community is growing every day.

  • And here is a snapshot of the commits to TensorFlow

  • since we open sourced the technology in 2015.

  • Since then, our project has received 142,000 stars.

  • How impressive.

  • And we couldn't have gotten this far without you--

  • the community.

  • I'm truly impressed by all our contributions and hard work.

  • So thank you so much for all your efforts.

  • Now let's have a more in-depth look at the different groups

  • that form our community.

  • Our vibrant community includes code and non-code contributors,

  • community managers, and students.

  • But also we onboarded 148 machine

  • learning Google Developer experts

  • who are a part of the Google Developer Expert Network.

  • We also established 73 TensorFlow user groups

  • worldwide.

  • We formed 11 special interest groups and welcome

  • 1,375 developers at the developers

  • at TensorFlow mailing list.

  • And just last year, we welcomed 20 students

  • at the Summer of Code.

  • Furthermore, our documentation has been

  • translated to 12 languages.

  • And on GitHub we have nearly 2,500 contributors

  • and nearly 9,000 commits.

  • How impressive, again.

  • What a fantastic work.

  • Altogether, we've built an open source project

  • that we can be very proud of.

  • I'm going to give you an overview

  • on how you can get involved in the TensorFlow community.

  • Firstly, the TensorFlow user groups,

  • they aim to upscale contributions

  • to TensorFlow with a local focus.

  • Some examples of the initiatives conducted by these groups,

  • they include events, meet-ups, and doc sprints.

  • The machine learning GDEs, they are experts and practitioners

  • from the community who support with outreach initiatives.

  • And finally, the Special Interest Groups, the SIGs.

  • They're focus groups contributing to TensorFlow

  • to certain parts in more specific projects.

  • Right now, we have 73 TensorFlow user groups distributed

  • all across the world.

  • And so far, these groups have reached more than 100,000

  • developers worldwide.

  • In your country, if you don't have a TensorFlow user

  • group yet, please feel free to reach out to us.

  • We can help you with the resources

  • and share best practices from other parts of the world.

  • These events are a great way to engage with other developers

  • and with the community.

  • We also strongly encourage students

  • to engage in these initiatives and learn with more experienced

  • developers.

  • After the TensorFlow roadshow in San Paolo

  • last October, we also established the first two

  • TensorFlow user groups in LATAM, in San Paolo, in Brazil,

  • and Argentina.

  • We are extremely happy to welcome these developers

  • to the TensorFlow community.

  • Also, in collaboration with our machine

  • learning ecosystem teams, we enable contributions

  • from the local communities.

  • And I'm very excited to announce that this year we

  • are going to do a series of TensorFlow bootcamps

  • and doc sprints in Latin America.

  • Our goal is to really get closer to you, to the machine learning

  • and the TensorFlow communities.

  • So we look forward to stay connected

  • with the communities in Colombia, Argentina, Chile,

  • and Mexico.

  • Documentation is also the backbone

  • off an open source project.

  • And so far, we have 268 tutorials and notebooks

  • translated across 12 different languages.

  • Additionally, we have over 40 translation

  • pull requests that are still awaiting review.

  • We really want to provide a good experience to our contributors.

  • And in order to achieve this goal,

  • we'll continue to support translations

  • of our documentation into several different languages.

  • Just last year, the machine learning GDEs,

  • they gave more than 100,000 talks.

  • And they published 295 articles that reached out more than half

  • a million developers.

  • Our machine learning GDEs, they play a critical role

  • in advancing machine learning education worldwide.

  • And if you'd like to be a part of this brilliant group,

  • please get in touch with us.

  • Just to highlight a few events where our machine learning GDEs

  • spoke at last year.

  • We have a few pictures, as you can see here,

  • and they refer to the African Build

  • Up Event, the Explore Machine Learning India Day,

  • and the doc sprints in Korea.

  • For bigger projects in which we have to work as a team,

  • we created the TensorFlow Special Interest Groups.

  • This is a program that organizes contributors into focus stream

  • areas of work.

  • These groups, they take ownership of specific areas

  • to add and maintain new features to TensorFlow.

  • It all started with a SIG build.

  • And right now, we have 11 different SIGs.

  • The contributors are welcome to join the SIGs

  • based on the parts of TensorFlow that they care the most.

  • As I mentioned before, our ecosystem

  • includes the SIG add-ons, Build IO, Networking, JVM, and Rust,

  • and Micro, who are community-led open source projects.

  • Whereas Keras, MLIR, Swift, and TensorBoard

  • are Google-led with an open source philosophy.

  • Each individual SIG meets on a monthly basis.

  • And if you want to be part of a SIG,

  • you can head to GitHub for more information on how to join.

  • You can also request a new SIG.

  • And we are here to help you and guide you through the process.

  • There are many ways that you can contribute to a SIG.

  • And these are just a few examples of specific projects

  • that are looking for contributors in 2020.

  • A great way to get feedback on a large project

  • is through the request for comment or RFC process.

  • It is the primary way we communicate our design

  • rationale and receive feedback from the community.

  • Anyone can create or comment on an RFC.

  • We have started tweeting our RFCs.

  • So please, keep an eye at the TensorFlow Twitter account.

  • We manage our RFC process on GitHub.

  • And there you can comment on an open RFC or propose a new RFC.

  • If you want to propose a design, we strongly

  • encourage you to socialize your idea, recruit the sponsor,

  • write your RFC, respond to feedback, and implement it.

  • And if you are interested in learning more about TensorFlow,

  • you may want to consider doing one

  • of our fantastic online courses, either on Coursera or Udacity.

  • We are proud to also announce that our university

  • program includes new courses about TensorFlow

  • at MIT, Imperial College, and University of Hong Kong,

  • just to mention a few.

  • And lastly, as announced this morning,

  • we are launching the TensorFlow Certificate Program.

  • It is intended for individuals who

  • want to advance their deep learning proficiency

  • and want to demonstrate experience

  • in building practical business applications using TensorFlow.

  • Head to tensorflow.org/certificate

  • for more information.

  • And then lastly, please follow us

  • on social media for the latest news

  • and content about TensorFlow and the community.

  • On Twitter, you can get the latest news about TensorFlow.

  • Head to the blog for more technical use cases

  • and best practices.

  • And then our YouTube channel has more educational resources

  • that will help you through your journey.

  • Get in touch and get involved with the community.

  • Thank you so much.

  • [MUSIC PLAYING]

JOANA CARRASQUEIRA: Hello, everybody.

Subtitles and vocabulary

Click the word to look it up Click the word to find further inforamtion about it