TVM: An End to End Deep Learning Compiler Stack
TVM: An End to End Deep Learning Compiler Stack
Tianqi Chen
A presentation from ApacheCon @Home 2020
https://apachecon.com/acah2020/
Apache(incubating) TVM is an open deep learning compiler stack for CPUs, GPUs, and specialized accelerators. It aims to close the gap between the productivity-focused deep learning frameworks, and the performance- or efficiency-oriented hardware backends. TVM provides the following main features: - Compilation of deep learning models in Keras, MXNet, PyTorch, Tensorflow, CoreML, DarkNet into minimum deployable modules on diverse hardware backends. - Infrastructure to automatic generate and optimize tensor operators on more backend with better performance. In this talk, I will cover the new developments in TVM in the past year around the areas of more backend, automation and model support.
Tianqi Chen received his PhD. from the Paul G. Allen School of Computer Science & Engineering at the University of Washington, working with Carlos Guestrin on the intersection of machine learning and systems. He has created three major learning systems that are widely adopted: XGBoost, TVM, and MXNet(co-creator). He is a recipient of the Google Ph.D. Fellowship in Machine Learning. He is currently the CTO of OctoML.
Видео TVM: An End to End Deep Learning Compiler Stack канала TheApacheFoundation
Tianqi Chen
A presentation from ApacheCon @Home 2020
https://apachecon.com/acah2020/
Apache(incubating) TVM is an open deep learning compiler stack for CPUs, GPUs, and specialized accelerators. It aims to close the gap between the productivity-focused deep learning frameworks, and the performance- or efficiency-oriented hardware backends. TVM provides the following main features: - Compilation of deep learning models in Keras, MXNet, PyTorch, Tensorflow, CoreML, DarkNet into minimum deployable modules on diverse hardware backends. - Infrastructure to automatic generate and optimize tensor operators on more backend with better performance. In this talk, I will cover the new developments in TVM in the past year around the areas of more backend, automation and model support.
Tianqi Chen received his PhD. from the Paul G. Allen School of Computer Science & Engineering at the University of Washington, working with Carlos Guestrin on the intersection of machine learning and systems. He has created three major learning systems that are widely adopted: XGBoost, TVM, and MXNet(co-creator). He is a recipient of the Google Ph.D. Fellowship in Machine Learning. He is currently the CTO of OctoML.
Видео TVM: An End to End Deep Learning Compiler Stack канала TheApacheFoundation
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
![TVM Conf 2020 - Tutorials - An Introduction to TVM](https://i.ytimg.com/vi/grlIOVGNNVw/default.jpg)
![Machine Learning magic for your web application with TensorFlow.js (Chrome Dev Summit 2019)](https://i.ytimg.com/vi/FDUfaYsFQrc/default.jpg)
![What is end-to-end deep learning? (C3W2L09)](https://i.ytimg.com/vi/ImUoubi_t7s/default.jpg)
![](https://i.ytimg.com/vi/kZVlsuPJeao/default.jpg)
![NLP for Developers: End to End Machine Learning | Rasa](https://i.ytimg.com/vi/w6SQ4-2tCA4/default.jpg)
![How to use Apache TVM to optimize your ML models](https://i.ytimg.com/vi/VK6cuGRH230/default.jpg)
![TVM Tutorial at FCRC [1/9]: TVM Overview](https://i.ytimg.com/vi/bqLfplA977o/default.jpg)
![Deep Learning - Computerphile](https://i.ytimg.com/vi/TJlAxW-2nmI/default.jpg)
![Design for Highly Flexible and Energy-Efficient Deep Neural Network Accelerators [Yu-Hsin Chen]](https://i.ytimg.com/vi/brhOo-_7NS4/default.jpg)
![Jeremy Howard: Deep Learning Frameworks - TensorFlow, PyTorch, fast.ai | AI Podcast Clips](https://i.ytimg.com/vi/XHyASP49ses/default.jpg)
![Apache TVM Developer Bootcamp #1: Overview - TVMCon 2021 Tutorial](https://i.ytimg.com/vi/ay_oOLsus2U/default.jpg)
!["TVM: An End to End Deep Learning Compiler Stack" by Thiery Moreau (OctoML)](https://i.ytimg.com/vi/JmWWAnHBDx0/default.jpg)
![TVM Stack: End to End Optimization for Deep Learning](https://i.ytimg.com/vi/DaCPJrTwT00/default.jpg)
![My Method to Learning ANY Song By Ear](https://i.ytimg.com/vi/bFtTvqASQ9k/default.jpg)
![Graph Convolutional Networks (GCNs) made simple](https://i.ytimg.com/vi/2KRAOZIULzw/default.jpg)
![Building High performance AI Training Systems with Gaudi](https://i.ytimg.com/vi/otoCxbZel1o/default.jpg)
![Apache Royale - moving a Flex app to Royale](https://i.ytimg.com/vi/O5WJ6nRPX40/default.jpg)
![Time Value of Money TVM Lesson/Tutorial Future/Present Value Formula Interest Annuities Perpetuities](https://i.ytimg.com/vi/m3azU7gYHc0/default.jpg)
![C4W1L06 Convolutions Over Volumes](https://i.ytimg.com/vi/KTB_OFoAQcc/default.jpg)
![Superoptimizing LLVM](https://i.ytimg.com/vi/Ux0YnVEaI6A/default.jpg)