A few days ago, the fourth PaddlePaddle Hackathon, sponsored by the National Engineering Research Center for Deep Learning Technology and Application and undertaken by Paddle, has officially started and is continuing to recruit developers. This competition has specially set up a TVM task area. Fei Paddle and Imagination jointly released a number of TVM-based development tasks, inviting developers to understand and participate in the joint construction of deep learning open source projects, aiming to encourage developers to participate in the joint development of deep learning open source projects. establish.
In order to allow developers to fully understand the development of flying paddles and get started with development tasks faster, the Hackathon Organizing Committee has launched TVM hard-core technology sharing and open source experience exchange. At 19:00 on the evening of March 27, the senior software engineer of Imagination China Lab was invited to participate in the fourth live broadcast of the PaddlePaddle Hackathon series. He will share the explanation and optimization of the new operator of TVM, and how to use it in Imagination NNA ( AI accelerator) to complete the deployment of the 3D model of the flying propeller. The live broadcast will be broadcast on the Fei Paddle Video Account and the live broadcast room of Station B, so stay tuned.
Time: Monday, March 27 at 19:00
Guest: Lu Zhuoran, Senior Software Engineer, Imagination China Lab
Sharing points:
● TVM new operator and optimization
● Imagination related product line introduction
● How to deploy the 3D model of flying propeller on Imagination NNA
● Imagination Paddlepaddle Model Zoo project
Development task introduction
The TVM mission activities are mainly divided into the following categories:
1) No.226-No.230: Add TVM operators to the Paddle framework (advanced question 1/basic question 4)
Task link:
https://github.com/PaddlePaddle/community/blob/master/hackthon_4th/【PaddlePaddle%20Hackathon%204】%20 ecological partner open source contribution task collection.md#task227
2) No.231-No.234: Improve operator support or fix problems for the front end of TVM PaddlePaddle (basic question 4/advanced question 1)
Task link:
https://github.com/PaddlePaddle/community/blob/master/hackthon_4th/【PaddlePaddle%20Hackathon%204】%20 ecological partner open source contribution task collection.md#task232
3) No.235: Add support for the PaddleSlim quantization model to the TVM PaddlePaddle front end (advanced questions)
Task link:
https://github.com/PaddlePaddle/community/blob/master/hackthon_4th/【PaddlePaddle%20Hackathon%204】%20 ecological partner open source contribution task collection.md#task235
4) No.181: Complete TVM access to FastDeploy, and verify the correctness on the PP-YOLOE model
Task link:
https://github.com/PaddlePaddle/community/blob/master/hackthon_4th/【PaddlePaddle%20Hackathon%204】%20 model kit open source contribution task collection.md#task181
Welcome to scan the code to register
More PaddlePadddle Hackathon details are as follows:
END
Welcome to join Imagination GPU and artificial intelligence communication group 2
Join the group, please add the editor WeChat: eetrend89
(Please add company name and title)
recommended reading
Dialogue with the Chairman of Imagination China: Using GPU as the fulcrum to strengthen software and hardware collaboration to facilitate digital transformation
Imagination Technologies is a UK-based company dedicated to the research and development of chips and software intellectual property (IP). Products based on Imagination IP are used in the phones, cars, homes and workplaces of billions of people around the world. For more information on cutting-edge technologies such as the Internet of Things, smart wearables, communications, automotive electronics, and graphics and image development, welcome to Imagination Tech!