Infrastructure, Distributed Training & Scaling · Problem 1 of 4
Implement a function computing total per-GPU memory for a given model/parallelism config.
Implement the function/class skeleton in the editor. Any correct approach is accepted.
def memory_report(num_params, dp=1, tp=1, pp=1, zero_stage=0, bytes_weight=2, bytes_grad=2, bytes_optim=12):
raise NotImplementedErrorReady when you are
Submit your solution and a structured review appears here — verdict, score, and concrete feedback. Any correct approach passes.
Implement a function computing total per-GPU memory for a given model/parallelism config.
Implement the function/class skeleton in the editor. Any correct approach is accepted.