Medium

Implement toy data-parallel SGD with manual gradient all-reduce (torch

Infrastructure, Distributed Training & Scaling · Problem 2 of 4

Chapter 06Infrastructure, Distributed Training & Scaling

Implement toy data-parallel SGD with manual gradient all-reduce (torch

MediumProblem 2 / 4

Implement toy data-parallel SGD with manual gradient all-reduce (torch.distributed or a simulated ring).

Implement the function/class skeleton in the editor. Any correct approach is accepted.

Hints