'Using numpy default_rng with Dask?
I'd like to know how I can use a numpy default random number generator with Dask.
The documentation for e.g. dask.array.random.normal has a note saying :
New code should use the normal method of a default_rng() instance instead
I assume that is because the documentation is copied from numpy. In numpy they introduced a default random number generator so that one can have rngs with different seeds (in contrast to the old global numpy.random.seed) but I can't figure out how to use one with the Dask random functions.
Solution 1:[1]
There is an issue in Dask for a feature request on this specific topic, where you can check if there are any updates.
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|---|
| Solution 1 | Joan Rue |