Outer Web | Explore

/explore

Click through on any links that interest you or select the planets on the right to continue exploring the Outer Web.

You are here		pytorch.org Introducing PyTorch Fully Sharded Data Parallel (FSDP) API \| PyTorch
\|	\|	research.google General and Scalable Parallelization for Neural Networks	3.8 parsecs away Travel
\|	\|	Posted by Yuanzhong Xu and Yanping Huang, Software Engineers; Google Research, Brain Team Scaling neural networks, whether it be the amount of trai...	3.8 parsecs away Travel
\|	\|	github.com GitHub - mosaicml/composer: Supercharge Your Model Training	3.4 parsecs away Travel
\|	\|	Supercharge Your Model Training. Contribute to mosaicml/composer development by creating an account on GitHub.	3.4 parsecs away Travel
\|	\|	dev-discuss.pytorch.org TorchDynamo Update 9: Making DDP Work with TorchDynamo - compiler - PyTorch Developer Mailing List	5.2 parsecs away Travel
\|	\|	TL;DR: Previously, torchdynamo interrupted compute-communication overlap in DDP to a sufficient degree that DDP training with dynamo was up to 25% slower than DDP training with eager. We modified dynamo to add additional...	5.2 parsecs away Travel
\|	\|	ssir.org How AI-Powered Nonprofits Are Building the Next Generation of Education Tools	19.4 parsecs away Travel
\|		AI is being used in exciting ways to bridge educational divides, and AI-powered nonprofits are creating a roadmap for what the future of education may hold.	19.4 parsecs away Travel