tensor – Andrew Fairless, Ph.D.

What I Read: Accelerating PyTorch

By Andrew Fairless on September 6, 2023August 1, 2023

https://magazine.sebastianraschka.com/p/accelerating-pytorch-model-training Accelerating PyTorch Model TrainingUsing Mixed-Precision and Fully Sharded Data ParallelismSebastian Raschka, PhDJun 26, 2023 “…how to scale PyTorch model training with minimal code changes. The focus here is onContinue readingWhat I Read: Accelerating PyTorch

What I Read: Einstein Summation in Deep Learning

By Andrew Fairless on January 26, 2022December 11, 2021

https://rockt.github.io/2018/04/30/einsum Einsum is All you Need – Einstein Summation in Deep LearningTim Rocktäschel30/04/2018 – updated 02/05/2018 “If you are anything like me, you find it difficult to remember the namesContinue readingWhat I Read: Einstein Summation in Deep Learning

What I Read: How to Train Really Large Models

By Andrew Fairless on October 26, 2021October 16, 2021

https://lilianweng.github.io/lil-log/2021/09/24/train-large-neural-networks.html How to Train Really Large Models on Many GPUs?Sep 24, 2021Lilian Weng “How to train large and deep neural networks is challenging, as it demands a large amount ofContinue readingWhat I Read: How to Train Really Large Models

What I Read: Graph Theory Into New Dimensions

By Andrew Fairless on September 16, 2021September 4, 2021

https://www.quantamagazine.org/how-big-data-carried-graph-theory-into-new-dimensions-20210819/ How Big Data Carried Graph Theory Into New DimensionsStephen OrnesContributing WriterAugust 19, 2021 ““Now we know that the network is just the shadow of the thing,”… If a dataContinue readingWhat I Read: Graph Theory Into New Dimensions

What I Read: Feature Management

By Andrew Fairless on December 25, 2020December 13, 2020

https://medium.com/@jrodthoughts/the-architecture-used-at-linkedin-to-improve-feature-management-in-machine-learning-models-c7bd6ae54db The Architecture Used at LinkedIn to Improve Feature Management in Machine Learning ModelsThe new typed feature schema streamlined the reusability of features across thousands of machine learning models.Jesus RodriguezAprContinue readingWhat I Read: Feature Management

Tag: tensor