how to normalize a numpy array in python

Question

I have the following numpy array:

from sklearn.decomposition import PCA
from sklearn.preprocessing import normalize
import numpy as np

# Tracking 4 associate metrics
# Open TA's, Open SR's, Open SE's
associateMetrics = np.array([[111,  28,  21],
   [ 27,  17,  20],
   [ 79,  23,  17],
   [185, 125,  50],
   [155,  76,  32],
   [ 82,  24,  17],
   [127,  63,  33],
   [193,  91,  63],
   [107,  24,  17]])

Now, I want to normalize every 'column' so that the values are between 0 and 1. What I mean is that the values in the 1st column for example should be between 0 and 1.

How do i do this?

normed_matrix = normalize(associateMetrics, axis=1, norm='l1')

the above gives me rowwise normalization

I think i have my answer.I tried putting axis=0 and it worked. the documentation wasn't clear for me i guess...scikit-learn.org/stable/modules/generated/… — Rookie
– Rookie, Commented Feb 6, 2016 at 23:30

Rookie · Accepted Answer · 2016-02-06 23:52:37Z

2

I was able to do this using the following:

normalized_metrics = normalize(associateMetrics, axis=0, norm='l1')

answered Feb 6, 2016 at 23:52

Rookie

5,51714 gold badges47 silver badges66 bronze badges

Sign up to request clarification or add additional context in comments.

4 Comments

javad Over a year ago

An axis in scipy specifies the dimension you are working on. For 2D data (matrxi), axis = 0 means across rows and axis = 1 means across columns. docs.scipy.org/doc/numpy-1.10.1/glossary.html

Rookie Over a year ago

thank you! If I may ask an additional question. How would I multiply every column by different scalar values? What I want to do is assign different weights to each column. Should i do this after normalizaiotn? If yes, how? sorry I am a bit new to python

javad Over a year ago

You can use numpy.multiply for element-wise multiplication of the weight vector by your matrix. Perhaps np.multiply(v, m) where v is your vector and m is your matrix. docs.scipy.org/doc/numpy-1.10.1/reference/generated/…

javad Over a year ago

Sorry, for column-wise multiplication you might need np.multiply(v, m.transpose()).transpose()

Collectives™ on Stack Overflow

how to normalize a numpy array in python

1 Answer 1

4 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

4 Comments

Your Answer

Sign up or log in

Post as a guest

Related