Closed Feez closed 3 years ago
@Feez I tried doing that a while back and it didn't seem to work. Though maybe things have changed since then?
It may be easier to use something like randint: https://docs.python.org/3/library/random.html#functions-for-integers
import random
min, max = 0, 9999999
seed = random.randint(min, max)
print(seed)
params.seed = seed
Could it be added to print the seed used? Useful for when want to reproduce the result of a random seed after-the-fact
I believe it is as simple as
But I not familiar enough with Torch or ML libraries in general to know if this is the correct seed and/or if there are side effects to this.