83 Folgen

  1. Code generation

    Vom: 4.6.2021

  2. Why is autograd so complicated

    Vom: 3.6.2021

  3. __torch_function__

    Vom: 2.6.2021

  4. TensorIterator

    Vom: 1.6.2021

  5. native_functions.yaml

    Vom: 28.5.2021

  6. Serialization

    Vom: 27.5.2021

  7. Continuous integration

    Vom: 26.5.2021

  8. Stacked diffs and ghstack

    Vom: 25.5.2021

  9. Shared memory

    Vom: 24.5.2021

  10. Automatic mixed precision

    Vom: 21.5.2021

  11. Conjugate views

    Vom: 20.5.2021

  12. History and constraints of Tensor

    Vom: 19.5.2021

  13. How new operators are authored

    Vom: 18.5.2021

  14. The life and death of Variable

    Vom: 17.5.2021

  15. Backend extensibility

    Vom: 14.5.2021

  16. The road to structured kernels

    Vom: 13.5.2021

  17. Functionalization

    Vom: 12.5.2021

  18. Just enough CUDA to be dangerous

    Vom: 11.5.2021

  19. Inference mode

    Vom: 10.5.2021

  20. Vectorization

    Vom: 7.5.2021

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