search

v923z / micropython-ulab

a numpy-like fast vector module for micropython, circuitpython, and their derivatives
https://micropython-ulab.readthedocs.io/en/latest
MIT License
416 stars 114 forks source link

add random module #654

Closed v923z closed 8 months ago

v923z commented 9 months ago

This PR implements the feature request https://github.com/v923z/micropython-ulab/issues/569. The random number generator supports the size and out keyword arguments. 

MicroPython v1.20.0-311-gea1a5e43d on 2023-07-23; linux [GCC 12.2.0] version
Use Ctrl-D to exit, Ctrl-E for paste mode
>>> from ulab import numpy as np
>>> rng = np.random.Generator()
>>> rng.random(12)
array([1.688118533849092e-10, 0.7703293013781235, 0.6074833036730274, ..., 0.2697689388821915, 0.6484892791277977, 0.6955782450470177], dtype=float64)
>>> rng.random(size=(5, 5))
array([[0.7724060907456224, 0.1599236337293051, 0.7339406510831528, 0.5274410894040502, 0.6714481027287427],
       [0.6492265779888077, 0.06322272656419292, 0.4808784435223349, 0.8659584576783896, 0.6510361989559311],
       [0.7035330373879642, 0.1143488863528934, 0.3228607724092158, 0.9696441515163603, 0.3353713223418759],
       [0.4890397567261386, 0.8449387836308111, 0.4607909039388876, 0.1653575696739978, 0.4966210702938057],
       [0.3707932026109159, 0.7451473233433786, 0.1918072763583848, 0.7469201739384557, 0.7688914647671239]], dtype=float64)
a = np.array([[2, 2], [3, 3]])
rng.random(out=a)
array([[0.7724060907456224, 0.1599236337293051],
       [0.7339406510831528, 0.5274410894040502]], dtype=float64)