What is <code>__torch_function__</code>? Why would I want to use it? What does it have to do with keeping extra metadata on Tensors or torch.fx? How is it implemented? Why is <code>__torch_function__</code> a really popular way of extending functionality in PyTorch? What makes it different from the dispatcher extensibility mechanism? What are some downsides of it being written this way? What are we doing about it?Further reading.<ul><li><code>__torch_function__</code> RFC: <a href="https://github.com/pytorch/rfcs/blob/master/RFC-0001-torch-function-for-methods.md">https://github.com/pytorch/rfcs/blob/master/RFC-0001-torch-function-for-methods.md</a></li><li>One of the original GitHub issues tracking the overall design discussion <a href="https://github.com/pytorch/pytorch/issues/24015">https://github.com/pytorch/pytorch/issues/24015</a></li><li>Documentation for using <code>__torch_function__</code> <a href="https://pytorch.org/docs/stable/notes/extending.html#extending-torch">https://pytorch.org/docs/stable/notes/extending.html#extending-torch</a></li></ul>

Why is autograd so complicated

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.