ZARZA We are Zarza, the prestigious firm behind major projects in information technology.
Listen "Double backwards"
Episode Synopsis
Double backwards is PyTorch's way of implementing higher order differentiation. Why might you want it? How does it work? What are some of the weird things that happen when you do this?Further reading.Epic PR that added double backwards support for convolution initially https://github.com/pytorch/pytorch/pull/1643
More episodes of the podcast PyTorch Developer Podcast
Compiler collectives
04/08/2024
TORCH_TRACE and tlparse
29/04/2024
Higher order operators
21/04/2024
Inductor - Post-grad FX passes
12/04/2024
CUDA graph trees
24/03/2024
Min-cut partitioner
17/03/2024
AOTInductor
02/03/2024
Tensor subclasses and PT2
24/02/2024
Compiled autograd
19/02/2024
PT2 extension points
05/02/2024
In God we trust