1

I am working on the following code:

np.random.seed(6233)
np.random.choice(20, 20)
u1 = np.random.normal(loc=0.0, scale=25.0, size=20)
u2 = np.random.normal(loc=0.0, scale=5.0, size=20) 
idu1 = np.random.choice(len(u1), 20)
idu2 = np.random.choice(len(u2), 20)
idu = [idu1, idu2]

I know it is not correct but my goal is to create a variable u which will be as follows:

u = (np.random.normal(loc=0.0, scale=25.0, size=20))*0.3 + (np.random.normal(loc=0.0, scale=5.0, size=20)*0.7)

I tried to create two other variables u1 and u2 and then convert each of them into one-dimensional array (because I though I might need to use choice(a, size, replace=True, p) as follows:

idu = [idu1, idu2]
probabilities = [0.3, 0.7]
choice(idu, size=20, replace=True, p=probabilities)

My problem is that I have no knowledge how I am supposed to mix two normal distributions by assigning probabilities 0.3 and 0.7 and create a variable that will have a combinations of the two with size=20, because I get a syntax error. Please any help will be appreciated!

Samuel Dion-Girardeau
  • 2,790
  • 1
  • 29
  • 37
Varou37
  • 9
  • 3
  • Can you describe what the syntax error is? Also I submitted an edit because the code was indented in such a way that it wasn't valid Python, but I suspect this is not the original issue, but merely an error copying the code over in Stack Overflow's editor. – Samuel Dion-Girardeau Feb 05 '20 at 04:15
  • 1
    Thanks for your comment and suggested editing, I am new here and to Python.It seems that I was writing something incorrect but I used what you suggested and now the error is no more but I am not sure if the probabilities assigned to u sum up to 1. – Varou37 Feb 05 '20 at 04:39
  • Ok, in this case be careful, because in Python, whitespace is significant! As for your problem: I suggest you edit your question to detail your specific problem now, i.e. what you are expecting vs. what you are getting. Otherwise it's hard for anyone to help you! – Samuel Dion-Girardeau Feb 06 '20 at 00:25

0 Answers0