PyTorch Developer Podcast
En podkast av Edward Yang, Team PyTorch
83 Episoder
-
Code generation
Publisert: 4.6.2021 -
Why is autograd so complicated
Publisert: 3.6.2021 -
__torch_function__
Publisert: 2.6.2021 -
TensorIterator
Publisert: 1.6.2021 -
native_functions.yaml
Publisert: 28.5.2021 -
Serialization
Publisert: 27.5.2021 -
Continuous integration
Publisert: 26.5.2021 -
Stacked diffs and ghstack
Publisert: 25.5.2021 -
Shared memory
Publisert: 24.5.2021 -
Automatic mixed precision
Publisert: 21.5.2021 -
Conjugate views
Publisert: 20.5.2021 -
History and constraints of Tensor
Publisert: 19.5.2021 -
How new operators are authored
Publisert: 18.5.2021 -
The life and death of Variable
Publisert: 17.5.2021 -
Backend extensibility
Publisert: 14.5.2021 -
The road to structured kernels
Publisert: 13.5.2021 -
Functionalization
Publisert: 12.5.2021 -
Just enough CUDA to be dangerous
Publisert: 11.5.2021 -
Inference mode
Publisert: 10.5.2021 -
Vectorization
Publisert: 7.5.2021
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.
