Content at a glance: The TVM offline gathering, which has been postponed since December last year, is finally here! We chose Shanghai as the first stop and invited 4 lecturers to share their TVM-related development experience based on their work practices. We look forward to meeting you offline~

Keywords: 2023 Meet TVM offline activities

Since its open source release in 2017, Apache TVM has attracted widespread attention and discussion in the industry for simplifying model training and inference deployment, especially in the past two to three years.The rapid rise of various hardware architectures in China has made machine learning compilation a hot topic again in the Chinese technology community.

After intensive preparations,The "2023 Meet TVM: First Gathering of the New Year" offline event hosted by the MLC.AI community will be held in Shanghai on March 4.

An Easter egg

Chen Tianqi, the main inventor of TVM and a famous young scholar in the field of machine learning, also prepared a mysterious Easter egg for everyone.At that time, friends are welcome to sign up and participate to discuss the present and future of machine learning compilation!

2023 Meet TVM Shanghai Event Information

⏰  time:March 4 (Saturday) 14:00-17:30

Place:Wujiaochang Innovation and Entrepreneurship College, Yangpu District, Shanghai (near Jiangwan Stadium Station, Metro Line 10)

Number of people:50 (Onsite seats are limited, please register as early as possible)

Sign up:

Offline participation:Click the link to jump to the event registration https://www.huodongxing.com/event/2691307342300

Online participation:HyperAI video account online live broadcast

Event Schedule:

Brief introduction of guests and content

Share topic:TVM and the Development of Machine Learning Compilation

Contents:Machine learning compilers have also been continuously improving their deployment performance in recent years, and TVM has always hoped to achieve performance improvements through "automation" technology. However, with the addition of new hardware, the development of compilers has also encountered many challenges including performance and versatility. The Unity design (TensorIR+Relax) currently launched by the TVM community attempts to fundamentally improve the versatility and customization capabilities of TVM and provide a more comprehensive infrastructure.
Watch this sharing session and you will learn:

1. The development process and main technical points of machine learning compilation (MLC)
2. Technical Development of TVM Core Components
3. TVM Unity design
4. TVM’s future plans and opportunities

Share topic:Use TVM to compile and reason about models on edge devices

Contents:In order to facilitate model deployment on edge devices and improve performance, we use TVM to implement it on different backends to support model compilation and deployment on different edge devices.

Watch this sharing session and you will learn:

1. Practice of using TVM for inference on edge devices

2. Practice of using TVM auto-scheduler to optimize the model

3. Preliminary practice of using TVM backend development

Share topic:DSA AI compiler built on TVM

Contents:TVM was open sourced in 2017 and natively supports GPU and CPU. How to build an AI compiler for DSA/NPU based on TVM? This sharing will try to answer this question through our practice.

Watch this sharing session and you will learn:

1. Best practices for leveraging native TVM technical capabilities on DSA

2. Vectorization and Tensorization Using TVM

3. Generate custom operators using TVM
4. Develop efficient engineering management strategies, such as self-developed and open source code management, CI/CD, etc.

Share topic:Extending SYCL backend for TVM

Contents:We have expanded the cross-heterogeneous programming model SYCL for TVM and added a path for SYCL code generation and automatic tuning. We have expanded the hardware support for NPU for SYCL and added support for instruction compilation and runtime adaptation for domestic Cambrian NPU. In the future, we will further open up the full-stack technology path from TVM to SYCL and then to other domestic NPU instruction compilation.

Watch this sharing session and you will learn:

1. The process of integrating SYCL into TVM as a new backend

2. Adaptation of SYCL in TVM automatic tuning

3. The process of extending Cambrian hardware support for SYCL

4. Introduce future development and optimization directions

Introduction to the organizers and partners

As the organizer of this event, MLC.AI Community (https://mlc.ai/) was founded in June 2022, and led by Chen Tianqi, the main inventor of Apache TVM and a well-known young scholar in the field of machine learning, the team launched the MLC online course, which systematically introduced the key elements and core concepts of machine learning compilation.

In November 2022, with the joint efforts of MLC.AI community volunteers,The first complete TVM Chinese document was launched and successfully hosted on the HyperAI official website.It further provides domestic developers who are interested in machine learning compilation with the basic settings - documentation - to access and learn a new technology.

In 2023, the community will launch the "2023 Meet TVM" series of activities in many cities across the country, and welcomes enterprises and community partners to participate in co-creation.

MLC Online Courses:https://mlc.ai/

TVM Chinese Documentation:https://tvm.hyper.ai/

Shanghai Wujiaochang Innovation and Entrepreneurship College was jointly established by the Yangpu District Government, Fudan University, Tongji University and other famous universities to accelerate the construction of an internationally competitive innovation and entrepreneurship ecosystem and improve the efficiency of resource allocation. Wujiaochang Innovation and Entrepreneurship College is a non-profit education and service organization, and also an innovation and entrepreneurship university without walls and non-traditional teaching in the context of the new era.

Visit the official website:http://www.spii.net.cn/

China's leading artificial intelligence and high-performance computing community,We are committed to providing high-quality public resources in the field of data science to domestic developers.So far, it has provided domestic download nodes for more than 1,200 public data sets, supported more than 300 artificial intelligence and high-performance computing related term queries, hosted the complete TVM Chinese documentation, and will soon launch multiple basic and popular tutorials.

Visit the official website:https://orion.hyper.ai/

OpenBayes is a leading high-performance computing service provider in China. By grafting classic software ecosystems and machine learning models onto new-generation heterogeneous chips, it provides faster and easier-to-use data science computing products for industrial enterprises and university scientific research. Its products have been adopted by dozens of large industrial scenarios or leading scientific research institutes.

Visit the official website:https://openbayes.com/

Considering the venue space of this event,We have only 50 places available.It is recommended that you register as early as possible to secure your seat.

Offline participation:Click the link to jump to the event registration https://www.huodongxing.com/event/2691307342300

Online participation:HyperAI video account online live broadcast

The 2023 Meet TVM series of events is now open. See you in Shanghai from 14:00 to 17:30 on March 4!

-- over--