pyqit.set_seed

pyqit.set_seed(seed: int) → int

Set the seed.

Seeds numpy and torch when it is installed.

Parameters:

seed (int) – Value applied to every supported RNG.

Returns:

The seed that was applied.

Return type:

int

Notes

This mutates global RNG state, the same contract as Lightning’s seed_everything. Trainer.fit calls it for you; call it yourself before constructing a model if you also want its initial weights to be reproducible, since those are drawn at construction time.