All Posts By

Christina Harter

Adlik 0.2.0 Bear Release Now Available!

By Blog

Adlik, an LF AI & Data Foundation Incubation-Stage Project, has released version 0.2.0, called Bear. We’re thrilled to see a release from this community who has been hard at work the past few months! Adlik is a toolkit for accelerating deep learning inference, which provides an overall support for bringing trained models into production and eases the learning curves for different kinds of inference frameworks. In Adlik, Model Optimizer and Model Compiler delivers optimized and compiled models for a certain hardware environment, and Serving Engine provides deployment solutions for cloud, edge and device.

In version 0.2.0, Adlik enhances features, increases useability, and addresses miscellaneous bug fixes. A few of the release highlights include the following:

New Compiler

  • Support DAG generation for end-to-end compilation of models with different representation 
  • Source representation: H5, Ckpt, Pb, Pth, Onnx and SavedModel
  • Target representation: SavedModel, OpenVINO IR, TensorRT Plan and Tflite 
  • Support model quantization for TfLite and TensorRT
  • Int8 quantization for TfLite
  • Int8 and fp16 quantization for TensorRT

Inference Engine

  • Support hybrid scheduling of ML and DL inference jobs
  • Support image based deployment of Adlik compiler and inference engine in cloud native environment
  • Deployment and functions have been tested in docker (V19.03.12) and Kubernetes (V1.13)
  • Support Adlik running in RaspberryPi and JetsonNano
  • Support the newest version of OpenVINO (2021.1.110) and TensorFlow (2.3.1)

Benchmark Test

  • Support benchmark test for models including ResNet-50, Inception V3, Yolo V3 and Bert with devices and 5 runtimes supported by Adlik

Out of these features, one important goal of this release is to make Adlik, especially Inference Engine, easier for developers to use. In addition, Inference Engine in this release is the first lightweight inference engine that supports hybrid scheduling of ML and DL inference jobs. So far it has been integrated in ZTE base station products. See the release note here.

The Adlik team expressed a special thank you to all of the contributors for their extra hard work.

The Adlik Project invites you to adopt or upgrade to Bear, version 0.2.0, and welcomes feedback. To learn more about the Adlik 0.2.0 release, check out the full release notes. Want to get involved with Adlik? Be sure to join the Adlik-Announce and Adlik Technical-Discuss mailing lists to join the community and stay connected on the latest updates.

Congratulations to the Adlik team! We look forward to continued growth and success as part of the LF AI & Data Foundation. To learn about how to host an open source project with us, visit the LF AI & Data website.

Adlik Key Links

LF AI & Data Resources

 

LF AI & Data Day-Shenzhen, Recap

By Blog

The 2020 LF AI & DATA Day – Shenzhen, was held on Nov. 7, 2020, at Shenzhen Pengcheng Lab. It was a remarkable day with presentations and a panel discussion from some of the most prestigious communities and organizations such as Pengcheng Lab, OpenI Community, LF AI & DATA Foundation, MindSpore Community, Huawei, Tencent, Baidu, ZTE, Zilliz and Didi. This event has attracted many audiences both online and offline.

Opening Speech

Representatives from Pengcheng Lab, OpenI Community and LF AI & DATA Foundation gave the opening speech.  At the end of the opening speech, an announcement of tripartite strategic collaboration between Pengcheng Lab, OpenI Qizhi Community and LF AI & DATA Foundation was made. This collaboration intends to further strengthen the global efforts on open-source AI. 

Starlord, Chair of LF AI & DATA Foundation Governing Board in his opening speech:

2020 is an extraordinary year and a good starting point for China’s open source. Starlord then shared the evolution of AI and big data. He believes that the change from structured data to the Internet era is the reason why the foundation is now expanding from AI to AI and data​​. To complete such a huge data processing and analysing, open-source cooperation must be strengthened. He then urges more developers to join the LF AI & DATA family to advance open-source movement. 

MindSpore open source community operation leader Zhipeng Huang shared recent achievements of MindSpore. Although only open sourced at the end of March this year, MindSpore has achieved brilliant success so far. 

In his speech, Zhipeng Huang emphasized his hope to promote the construction of a new AI Native Programming industry and ecosystem through collaboration with LF AI & Data Foundation and OpenI community. In order to build an AI native programming framework system MindSpore community has also launched the MLWorkflow & Interop committee in the LF AI & Data Foundation.

Before the end of the morning session, the guests all came one stage for a panel discussion on how to better build an open-source community. The speakers all agreed that open source is not self-innovation, but external innovation, and working together with others can achieve the result of 1+1>2; the open-source community is a global community, and we need to use the power of global developers to promote greater tech innovation.

In the afternoon session, Ibrahim Haddad (VP of the Linux Foundation) analyzed the ecosystem challenges faced by the open-source community and introduced the development history and future plans of the LF AI & DATA Foundation in detail. He also stated that he will work hard to establish and support an open and growing ecosystem of open-source artificial intelligence, data, and analysis projects. 

Dongwei Jiang, a senior algorithm engineer at Didi, talked about the speech processing research and application in the field. Athena’s unique acoustics, language models, and decoding can help users solve more tasks such as voice conversion and ASR.

Hai Jin, R&D Director of Zilliz, shared Zilliz’s progress in unstructured data services.

Milvus, an open-source similarity search engine for massive-scale feature vectors has achieved remarkable results since it open sourced. It joined LF AI & DATA as an incubation project in April 2020. 

It has gained more than 400 enterprise users and many more individual users so far. He hopes with the ease of use and versatility of the Milvus, it can further help more companies implement AI applications.

Wen Ouyang, a senior R&D engineer from Tencent covered the recent optimization efforts of Angel. 

Angel is a full-stack machine learning platform open sourced by Tencent, and its functional features cover all stages of machine learning: feature engineering, model training, Hyperparameter adjustment, and model service. 

Angel is the first project in China to graduate from the LF AI & DATA Foundation. After years of development, Angel has been recognized by many developers.

ZTE AI Platform System Architect Bo Tang discussed the data security issues and current industrial practices of federation learning. 

As a decentralized solution, federated learning can solve the problem of user data security to a large extent. He then shared with us ZTE’s current efforts and achievements in this field.

Ti Zhou, Senior Architect of Baidu shared PaddlePaddle/EDL’s deep learning practice. In the future, Baidu will continue to optimize user experience and hope that more developers will join them to build a better deep learning platform together.

Thanks everyone for joining the event, we are very happy to see so many people interested in opensource AI. We hope to see you at the next LF AI & DATA Day soon!

LF AI & Data Day is a regional, one-day event hosted and organized by local members with support from LF AI & Data, its members, and projects. If you are interested in hosting an LF AI & Data Day, please email info@lfaidata.foundation to discuss.

LF AI & Data Resources

FEAST Joins LF AI & Data as New Incubation Project

By Blog

LF AI & Data Foundation—the organization building an ecosystem to sustain open source innovation in artificial intelligence (AI), machine learning (ML), deep learning (DL), and Data open source projects—today is announcing FEAST as its latest Incubation Project. Feast (Feature Store) is an open source feature store for machine learning. 

Today, teams running operational machine learning systems are faced with many technical and organizational challenges:

  1. Models don’t have a consistent view of feature data and are tightly coupled to data infrastructure.
  2. Deploying new features in production is difficult.
  3. Feature leakage decreases model accuracy.
  4. Features aren’t reused across projects.
  5. Operational teams can’t monitor the quality of data served to models.

Developed collaboratively between Gojek and Google Cloud in 2018, Feast was open sourced in early 2019. The project sets out to address these challenges as follows:

  1. Providing a single data access layer that decouples models from the infrastructure used to generate, store, and serve feature data.
  2. Decoupling the creation of features from the consumption of features through a centralized store, thereby allowing teams to ship features into production with minimal engineering support.
  3. Providing point-in-time correct retrieval of feature data for both model training and online serving.
  4. Encouraging reuse of features by allowing organizations to build a shared foundation of features.
  5. Providing data-centric operational monitoring that ensures operational teams can run production machine learning systems confidently at scale.

“Feast was created to address the data challenges we faced at Gojek while scaling machine learning for ride-hailing, food delivery, digital payments, fraud detection, and a myriad of other use cases” said Willem Pienaar, creator of Feast. “After open sourcing the project we’ve seen an explosion of demand for the software, leading to strong adoption and community growth. Entering the LF AI & Data Foundation is an important step for us toward decentralized governance and wider industry adoption and development.” 

Jeremy Lewi, Kubeflow founder, said “Feast entering the LF AI & Data Foundation is both a major milestone for the project and recognition of the strides the project has made toward solving some of the hardest problems in productionizing data for machine learning. Technologies like Feast have the potential to shape the machine learning stack of the future, and with its incubation in LF AI & Data, the project now has the ideal environment to expand its community in building a best-in-class open source feature store.”

Dr. Ibrahim Haddad, Executive Director of LF AI & Data, said: “We are very excited to welcome FEAST to LF AI & Data and help it thrive in a vendor-neutral environment under an open governance model. With the addition of FEAST, we are increasing the number of hosted projects under the Data category and look forward to tighter collaboration between our data projects and all other projects to drive innovation in data, analytics, and AI open source technologies.” 

LF AI & Data supports projects via a wide range of services, and the first step is joining as an Incubation Project.  LF AI & Data will support the neutral open governance for FEAST to help foster the growth of the project. Check out the Documentation to start working with FEAST today. Learn more about FEAST on their GitHub and be sure to join the FEAST-Announce and FEAST-Technical-Discuss mail lists to join the community and stay connected on the latest updates. 

A warm welcome to FEAST! We look forward to the project’s continued growth and success as part of the LF AI & Data Foundation. To learn about how to host an open source project with us, visit the LF AI & Data website.

FEAST Key Links

LF AI & Data Resources

Thanks for Joining us at OSS+ELC EU 2020!

By Blog

Thank you to everyone who joined us last week at Open Source Summit + Embedded Linux Conference Europe 2020! 

The event was held virtually this year, with almost 2,500 registrants from 85 countries

AI/ML/DL Track

The AI/ML/DL Track was well attended, with several AI industry leaders sharing knowledge and insights, and engaging with attendees in several great discussions during and after the sessions.

Miss the sessions? If you registered for OSS+ELC EU, you can always view all of the session recordings on demand on the virtual event platform for 30 days after the event. Or check back on the OSS EU website for a link to all of the sessions recordings on YouTube soon.

LF AI Virtual Booth

The LF AI Foundation virtual booth was also a hit! We had several people visit the booth and interact with our LF AI community members who were staffing the booth. Visitors were also able to download resources to learn more about LF AI.

LF AI Mini Summit

On Thursday, October 29, we hosted our LF AI Mini Summit. Our presenters covered the latest updates from the Foundation, Technical Advisory Council, Trusted AI Committee, and more. We’re looking forward to more engagement opportunities with the community in the future. 

Miss the Summit? If you registered for OSS EU, you can always view the Summit recording on demand on the virtual event platform for 30 days after the event. Or check back on the OSS EU website for a link to all of the sessions recordings on YouTube soon.

LF AI & Data Announcement

On Monday, October 26, during OSS EU, LF AI announced the formation of the LF AI & Data Foundation.

The LF AI & Data Foundation will build and support an open community and a growing ecosystem of open source AI, data and analytics projects, by accelerating development and innovation, enabling collaboration and the creation of new opportunities for all the members of the community.

As one entity under the Linux Foundation, this consolidated and focused effort will enable additional collaboration and integration in the space of AI/ML/DL and Data. With the creation of LF AI & Data, both communities will now support a growing ecosystem of artificial intelligence, machine learning, deep learning and data technologies.

Want to get involved with the LF AI & Data Foundation? Be sure to subscribe to our mailing lists to join the community and stay connected on the latest updates. 

LF AI & Data Resources

New AI & Data Foundation Combines Industry’s Fastest-Growing Open Source Developments in Artificial Intelligence and Open Data

By Blog, Press Release

The Linux Foundation’s AI Foundation & ODPi merge to support growing portfolio of technologies and drive open source collaboration across AI and data

LF AI Foundation (LF AI), the organization building an ecosystem to enable and sustain open source innovation in artificial intelligence (AI), machine learning (ML), and deep learning (DL), and ODPi, a nonprofit organization accelerating the open ecosystem of big data solutions, today announced they will come together under the new LF AI & Data Foundation. The LF AI & Data Foundation will build and support an open community and a growing ecosystem of open source AI, data and analytics projects, by accelerating development and innovation, enabling collaboration and the creation of new opportunities for all the members of the community.

As one entity under the Linux Foundation, this consolidated and focused effort will enable additional collaboration and integration in the space of AI/ML/DL and Data. With the creation of LF AI & Data, both communities will now support a growing ecosystem of artificial intelligence, machine learning, deep learning and data technologies. AI and Data are inseparable and codependent on each other. Combining efforts in both spaces will bring developers and projects under a single roof, orchestrated by a single Technical Advisory Council and several committees (Trusted AI, BI & AI), to work together towards building the open source AI & Data ecosystem and accelerating development and innovation. Hosting projects under a single umbrella enables closer collaboration, integration, and interoperability across projects and is a proven recipe for building strong open ecosystems. At the same time, it will provide a unified guidance for end users on tools, interoperability, integration, standards, and the future of AI, Data and Analytics as its use continues to grow in every industry. Furthemore, as member driven organizations, joining forces under LF AI & Data will allow greater efficiency for members across the various services we offer to our hosted projects. 

“LF AI has been growing at the rate of one new project per month, including several data projects. It is a natural move to bring together the open AI and data communities to enable better interoperability and capabilities across all of our hosted projects and to enable closer collaboration, which has been a proven recipe for building a strong open ecosystem. It will also provide our members greater cost efficiency when supporting our projects,” said Dr. Ibrahim Haddad, LF AI Executive Director. “We look forward to supporting innovation in the open source ecosystem focused on AI, Data and Analytics.”

“Over the past 5 years, ODPi has been a part of driving standardization and consolidation in the Big Data and Hadoop ecosystem, as well as becoming a focal point for the data challenges of the enterprises of today such as metadata, governance, and data science,” said John Mertic, Director, ODPi. “Coming together to form LF AI & Data is the next natural step in this mission, enabling greater interoperability to drive both innovation and sustainable growth for key projects.”

ODPi and its projects Egeria and OpenDS4All will become hosted projects under the LF AI & Data Foundation, with BI and AI becoming a committee within the foundation. They will maintain their current open technical governance model, establish collaboration with other hosted projects in LF AI & Data via the efforts of the its Technical Advisory Council, and benefit from a host of services offered to facilitate collaboration and increased adoption. 

Charles “Starlord” Xie, Chairperson, LF AI Governing Board, said: “Today’s announcement of LF AI & Data is very exciting news for the open source AI ecosystem. LF AI & Data brings together companies, projects and communities focusing on AI and Data under a single organization to foster collaboration and integration across projects. We look forward to the months and years to come as we anticipate significant growth in our project portfolio and the development of many collaboration opportunities.”

Craig Rubendall, Chairperson, ODPi, said: “The ODPi Board of Directors and technical community are excited to come together with LF AI to form LF AI & Data. This joint foundation will enable the key open source projects our industry depends on to have a sustainable home, which will drive further innovation and collaboration.”

Jim Spohrer, Director, IBM Center for Open Source Data and AI Technologies and LF AI Technical Advisory Council Chairperson, said: “If you are committed to advancing open source AI and data as an industry-standard infrastructure and democratizing access to AI, there is a great opportunity to contribute within the community. Join our bi-weekly TAC meetings and connect with us on the LFAI slack channel.”

LF AI launched two years ago with nine founding members and one hosted project, and today has 25 members and 20 technical projects. ODPi launched five years ago, and is supported by a strong roster of industry-leading members. Together, under the LF AI & Data Foundation, there are 22 projects supported by more than 60 companies, 20 universities and more than 1,300 active developers contributing to these projects. 

New members of LF AI & Data Foundation include aivancity School for Technology, Business & Society, AlphaBravo, Broadcom, Cloudera, Databricks, Index Analytics, ING Bank, OpenI, Precisely, Peng Cheng Laboratory, SAS Institute, and the Shanghai OpenSource Information Technology Association.

“The future of open source is directly linked to the AI ecosystem and a multitude of data communities. Therefore, it is a natural pairing for LF AI and ODPi to join forces and Cloudera is thrilled to be aiding their efforts in growing the project portfolio to encourage innovation and enable stronger interoperability and collaboration.” – Arun Murthy, CPO, Cloudera.

“Both fresh ingredients and a wonderful recipe are important for a delicious meal. Index Analytics is happy to join this feast as AI (Recipe) and Data (Ingredients) converge!” – Cupid Chan, CTO, Index Analytics

“Precisely is proud to be a member of the Linux Foundation AI & Data to help accelerate the growth and adoption of artificial Intelligence (AI). At Precisely we are committed to delivering the trusted data that is required to enable trusted AI and analytics. We believe that providing data integrity – data with accuracy, consistency, and context – help enable organizations to accelerate their adoption of AI and drive better business decisions.” – Tendü Yoğurtçu, PhD, Chief Technology Officer, Precisely

For more information and to get involved please visit: https://lfai.foundation/about/join/. If you’re interested in hosting a project, please review the Proposal and Hosting Process and check out the Hosting Requirements. For questions, please email info@lfaidata.foundation.

Helpful Resources

About LF AI & Data Foundation

The LF AI & Data Foundation, a Linux Foundation project, accelerates and sustains the growth of Artificial Intelligence (AI), Data Management, Machine Learning (ML), Deep Learning (DL) and Data open source projects. Backed by many of the world’s largest technology leaders, LF AI & Data is a neutral space for harmonization and ecosystem engagement to advance AI and Data innovation. To get involved with the LF AI & Data Foundation, please visit https://lfaidata.foundation.

Thank you ONNX & Amazon for Hosting a Great LF AI Day!

By Blog

A big thank you to Amazon and ONNX for hosting a great virtual meetup! The LF AI Day ONNX Community Virtual Meetup was held on October 14, 2020 and was a great success with over 100 attendees joining live. 

The meetup included ONNX Community updates, partner/end-user stories, and SIG/WG updates. The virtual meetup was an opportunity to connect with and hear from people working with ONNX across a variety of groups. A special thank you to Sheng Zha from Amazon for working closely with the ONNX Technical Steering Committee, SIGs, and ONNX community to curate the content.

Missed the meetup? Check out all of the presentations and recordings here.

This meetup took on a virtual format but we look forward to connecting again at another event in person soon. LF AI Day is a regional, one-day event hosted and organized by local members with support from LF AI, its members, and projects. If you are interested in hosting an LF AI Day please email info@lfai.foundation to discuss.

ONNX, an LF AI Foundation Graduated Project, is an open format to represent deep learning models. With ONNX, AI developers can more easily move models between state-of-the-art tools and choose the combination that is best for them.  Be sure to join the ONNX-Announce mailing list to join the community and stay connected on the latest updates. You can join technical discussions on GitHub and more conversations with the community on LF AI Slack’s ONNX channels.

ONNX Key Links

LF AI Resources

Thank you for Joining the LF AI Foundation at ONES 2020!

By Blog

Thank you to everyone who joined us last week at Open Networking & Edge Summit 2020! 

The event was held virtually this year, with over 1,300 people from 71 countries attending

The virtual LF AI Foundation booth was a hit! We had over 60 people visit our booth in the Gold & Bronze Hall and interact with our LF AI community members who were staffing the booth. Visitors were also able to download resources to learn more about each of LF AI’s projects, and they were able to give us insight into what type of projects they want to see in the future through our attendee survey. We were also able to engage with attendees in several great discussions, both at the booth and in the official ONES 2020 Slack workspace.

The session “The Making of 5G with AI & Open Source, presented by LF AI community member Mazin Gilbert, Vice President of Technology & Innovation at AT&T, was well attended! Attendees were able to hear how the marriage of 5G with edge cloud will enable next generation experiences like holograms for gatherings and meetings, mobile and untethered xR experiences for gaming and remote surgery, and immersive experiences for digital shopping.

Miss the session? If you registered for ONES, you can always view all of the session recordings on demand on the virtual event platform anytime. Or check back on the ONES website for a link to all of the sessions recordings on YouTube soon.

The LF AI Foundation mission is to build and support an open AI community, and drive open source innovation in the AI, ML, and DL domains by enabling collaboration and the creation of new opportunities for all the members of the community.

Want to get involved with the LF AI Foundation? Be sure to subscribe to our mailing lists to join the community and stay connected on the latest updates. 

LF AI Resources

Sparklyr 1.4.0 Release Now Available!

By Blog

Sparklyr, an LF AI Foundation Incubation Project, has released version 1.4.0! Sparklyr is an R Language package that lets you analyze data in Apache Spark, the well-known engine for big data processing, while using familiar tools in R. The R Language is widely used by data scientists and statisticians around the world and is known for its advanced features in statistical computing and graphics. 

In version 1.4.0, sparklyr adds a variety of improvements. Highlights include:

  • This release features efficient and parallelizable weighted sampling methods for Spark data frames. The technique being utilized by those methods is known as “exponential variates” (see https://blogs.rstudio.com/ai/posts/2020-07-29-parallelized-sampling)
  • Tidyr verbs such as `pivot_wider`, `pivot_longer`, `nest`, `unnest`, `separate`, `unite`, and `fill` now have specialized implementations in `sparklyr` for working with Spark data frames
  • Support for newly introduced higher-order functions in Spark 3.0 (e.g., `array_sort`, `map_filter`, `map_zip_with`, and many others)
  • Dplyr-related improvements:
    • All higher-order functions and sampling methods are made directly accessible through `dplyr` verbs
    • Made `grepl` part of the `dplyr` interface for Spark data frames
    • Thanks to a pull request from @wkdavis, `dplyr::inner_join`, `dplyr::left_join`, `dplyr::right_join`, and  `dplyr::full_join` now correctly replace `’.’` with `’_’` in the `suffix` parameter when working with Spark data frames (see https://github.com/sparklyr/sparklyr/issues/2648)
  • Support for newly introduced functionalities in Spark 3.0
    • Thanks to a pull request from @zero323, the `RobustScaler` functionality in Spark 3.0 is now supported in sparklyr through `ft_robust_scaler`
    • RAPIDS GPU acceleration plugin can now be enabled with `spark_connect(…, package = “rapids”)` and configured with `spark_config` options prefixed with “spark.rapids.”

The power of open source projects is the aggregate contributions originating from different community members and organizations that collectively help drive the advancement of the projects and their roadmaps. The sparklyr community is a great example of this process and was instrumental in producing this release. The sparklyr team wanted to give a special THANK YOU to the following community members for their contributions via pull requests (listed in chronological order):

To learn more about the sparklyr 1.4.0 release, check out the full release notes. Want to get involved with sparklyr? Be sure to join the sparklyr-Announce and sparklyr Technical-Discuss mailing lists to join the community and stay connected on the latest updates.

Congratulations to the sparklyr team and we look forward to continued growth and success as part of the LF AI Foundation! To learn about hosting an open source project with us, visit the LF AI Foundation website.

sparklyr Key Links

LF AI Resources

LF AI Day ONNX Community Virtual Meetup – Fall 2020

By Blog

Amazon, ONNX, and the LF AI Foundation are pleased to sponsor the upcoming LF AI Day* – ONNX Community Virtual Meetup – Fall 2020, to be held via Zoom on October 14th.

ONNX, an LF AI Foundation Graduated Project, is an open format to represent deep learning models. With ONNX, AI developers can more easily move models between state-of-the-art tools and choose the combination that is best for them. 

The virtual meetup will cover ONNX Community updates, partner/end-user stories, and SIG/WG updates. Check back on the event website for the agenda. If you are using ONNX in your services and applications, building software or hardware that supports ONNX, or contributing to ONNX, you should attend! This is a great opportunity to connect with and hear from people working with ONNX across many companies. 

Registration is now open and the event is free to attend. Capacity will be 500 attendees. For up to date information on this virtual meetup, please visit the event website

Want to get involved with ONNX? Be sure to join the ONNX-Announce mailing list to join the community and stay connected on the latest updates. You can join technical discussions on GitHub and more conversations with the community on LF AI Slack’s ONNX channels.

Note: In order to ensure the safety of our event participants and staff due to the Novel Coronavirus situation (COVID-19) the ONNX Steering Committee decided to make this a virtual-only event via Zoom.*LF AI Day is a regional, one-day event hosted and organized by local members with support from LF AI and its Projects. Learn more about the LF AI Foundation here.

ONNX Key Links

LF AI Resources

NNStreamer 1.6.0 Release Now Available!

By Blog

NNStreamer, an LF AI Foundation Incubation-Stage Project, has released version 1.6.0. NNStreamer is a set of Gstreamer plugins that support ease and efficiency for Gstreamer developers adopting neural network models and neural network developers managing neural network pipelines and their filters.

In version 1.6.0, NNStreamer adds a variety of improvements; highlights include:

  • New hardware accelerators and neural network frameworks support added: Verisilicon-Vivante, Qualcomm-SNPE, NNFW-ONE-Runtime, and Tensorflow2-lite.
  • Data serialization support with flatbuf and protobuf.
  • Android APIs optimized (i.e., invoke latency in Galaxy S20: 2ms → 0.1ms)
  • Plug-and-play sub-plugins support for tensor-converters
  • Hardware acceleration configuration reworked: multiple candidates may be expressed and options may be altered in run-time.
  • Fixes, semantics updates, and minor features added after commercialization (Galaxy Watch 3 and a few “next-year” products and Tizen 6 releases).

The NNStreamer Project invites you to adopt or upgrade to version 1.6.0 in your application, and welcomes feedback. To learn more about the NNStreamer 1.6.0 release, check out the change log and full release notes. Want to get involved with NNStreamer? Be sure to join the  NNStreamer-Announce and NNStreamer-Technical-Discuss mail lists to join the community and stay connected on the latest updates.

Congratulations to the NNStreamer team! We look forward to continued growth and success as part of the LF AI Foundation. To learn about hosting an open source project with us, visit the LF AI Foundation website.

NNStreamer Key Links

LF AI Resources