Is there a way to select a random number from two different ranges at once?-CodePudding

So, basically, I have two ranges. "Narrow range" is any number from -0.5 to 0.3 (up to 5 decimal places), and the "Full range" is any number from -1 to 1. To select a random value, for each, I do the following:

 narrowrange=np.random.uniform(-.5,.3)
 fullrange=np.random.uniform(-1,1)

However, I would like to create a new range, which subtracts the narrowrange from the fullrange. In other words, I'd like to select random numbers from -1 to -0.5 and 0.3 to 1 (so excluding any numbers from the narrow range).

The only way I can think of doing this is by creating an if statement for the full range (if this generated # falls into the narrow range, then generate a # again). Anyone know how to do this without for loops/if statements?

CodePudding user response：

You can use np.random.choice with a list of random.uniform as parameter

np.random.choice([np.random.uniform(-1, -.5), np.random.uniform(.3, 1)])

If you would like to add weight as suggested in the comments you can use the p parameter

total = -.5   1   1 - .3
p = [(-.5   1) / total, (1 - .3) / total]
np.random.choice([np.random.uniform(-1, -.5), np.random.uniform(.3, 1)], p=p)

CodePudding user response：

I can't think of how to do it without an if statement, but I can do it without the risk of multiple tries.

Your full range of values is 1.2 (-1 to -0.5 is a range of 0.5, and 0.3 to 1 is a range of 0.7, so that's 1.2 in total).

I would suggest picking a random number in the range 1.2, and then shifting that value into one of the two ranges based on where it lies:

unshifted = np.random.uniform(0, 1.2)
if unshifted <= 0.5:
    shifted = unshifted - 1
else:
    shifted = unshifted - 0.2