Why are RNGs important? What is the generator concept? How do PyTorch's CPU and CUDA RNGs differ? What are some of the reasons why Philox is a good RNG for CUDA? Why doesn't the generator class have virtual methods for getting random numbers? What's with the next normal double and what does it have to do with Box Muller transform? What's up with csprng?Further reading.<ul><li>CUDAGeneratorImpl has good notes about CUDA graph interaction and pointers to all of the rest of the stuff <a href="https://github.com/pytorch/pytorch/blob/1dee99c973fda55e1e9cac3d50b4d4982b6c6c26/aten/src/ATen/CUDAGeneratorImpl.h">https://github.com/pytorch/pytorch/blob/1dee99c973fda55e1e9cac3d50b4d4982b6c6c26/aten/src/ATen/CUDAGeneratorImpl.h</a></li><li>Transform uniformly distributed random numbers to other distributions with <a href="https://github.com/pytorch/pytorch/blob/master/aten/src/ATen/core/TransformationHelper.h">https://github.com/pytorch/pytorch/blob/master/aten/src/ATen/core/TransformationHelper.h</a></li><li>torchcsprng <a href="https://github.com/pytorch/csprng">https://github.com/pytorch/csprng</a></li></ul>

TensorAccessor

The PyTorch Developer Podcast is a place for the PyTorch dev team to do bite sized (10-20 min) topics about all sorts of internal development topics in PyTorch.