(no title)
SatvikBeri | 2 days ago
You can profile memory usage line by line in detail pretty easily with tools like @timeit or @btime
In practice I've found it pretty easy to get inner loops down to few or 0 allocations when needed for parallelization
No comments yet.