PyTorch Developer Podcast
A podcast by Edward Yang, Team PyTorch

Categories:
83 Episodes
-
Weak references
Published: 6/6/2022 -
Strides
Published: 5/30/2022 -
AOTAutograd
Published: 5/9/2022 -
Dispatcher questions with Sherlock
Published: 5/2/2022 -
New CI
Published: 4/25/2022 -
Python exceptions
Published: 4/17/2022 -
Torch vs ATen APIs
Published: 4/11/2022 -
All about NVIDIA GPUs
Published: 9/24/2021 -
Tensor subclasses and Liskov substitution principle
Published: 9/16/2021 -
Half precision
Published: 9/10/2021 -
DataLoader with multiple workers leaks memory
Published: 9/1/2021 -
Batching
Published: 8/18/2021 -
Multiple dispatch in __torch_function__
Published: 8/10/2021 -
Multithreading
Published: 8/3/2021 -
Asynchronous versus synchronous execution
Published: 7/27/2021 -
gradcheck
Published: 7/23/2021 -
torch.use_deterministic_algorithms
Published: 7/21/2021 -
Reference counting
Published: 7/20/2021 -
Memory layout
Published: 7/13/2021 -
pytorch-probot
Published: 7/12/2021
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.