83 Episodes

  1. Code generation

    Published: 4/06/2021

  2. Why is autograd so complicated

    Published: 3/06/2021

  3. __torch_function__

    Published: 2/06/2021

  4. TensorIterator

    Published: 1/06/2021

  5. native_functions.yaml

    Published: 28/05/2021

  6. Serialization

    Published: 27/05/2021

  7. Continuous integration

    Published: 26/05/2021

  8. Stacked diffs and ghstack

    Published: 25/05/2021

  9. Shared memory

    Published: 24/05/2021

  10. Automatic mixed precision

    Published: 21/05/2021

  11. Conjugate views

    Published: 20/05/2021

  12. History and constraints of Tensor

    Published: 19/05/2021

  13. How new operators are authored

    Published: 18/05/2021

  14. The life and death of Variable

    Published: 17/05/2021

  15. Backend extensibility

    Published: 14/05/2021

  16. The road to structured kernels

    Published: 13/05/2021

  17. Functionalization

    Published: 12/05/2021

  18. Just enough CUDA to be dangerous

    Published: 11/05/2021

  19. Inference mode

    Published: 10/05/2021

  20. Vectorization

    Published: 7/05/2021

4 / 5

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.

Visit the podcast's native language site