83 Épisodes

  1. Weak references

    Publié: 06/06/2022

  2. Strides

    Publié: 30/05/2022

  3. AOTAutograd

    Publié: 09/05/2022

  4. Dispatcher questions with Sherlock

    Publié: 02/05/2022

  5. New CI

    Publié: 25/04/2022

  6. Python exceptions

    Publié: 17/04/2022

  7. Torch vs ATen APIs

    Publié: 11/04/2022

  8. All about NVIDIA GPUs

    Publié: 24/09/2021

  9. Tensor subclasses and Liskov substitution principle

    Publié: 16/09/2021

  10. Half precision

    Publié: 10/09/2021

  11. DataLoader with multiple workers leaks memory

    Publié: 01/09/2021

  12. Batching

    Publié: 18/08/2021

  13. Multiple dispatch in __torch_function__

    Publié: 10/08/2021

  14. Multithreading

    Publié: 03/08/2021

  15. Asynchronous versus synchronous execution

    Publié: 27/07/2021

  16. gradcheck

    Publié: 23/07/2021

  17. torch.use_deterministic_algorithms

    Publié: 21/07/2021

  18. Reference counting

    Publié: 20/07/2021

  19. Memory layout

    Publié: 13/07/2021

  20. pytorch-probot

    Publié: 12/07/2021

2 / 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