165

Using numpy, how can I do the following:

ln(x)

Is it equivalent to:

np.log(x)

I apologise for such a seemingly trivial question, but my understanding of the difference between log and ln is that ln is logspace e?

tsherwen
  • 1,076
  • 16
  • 21
user1220022
  • 11,167
  • 19
  • 41
  • 57

6 Answers6

259

np.log is ln, whereas np.log10 is your standard base 10 log.

user16714199
  • 228
  • 6
  • 19
JoshAdel
  • 66,734
  • 27
  • 141
  • 140
25

Correct, np.log(x) is the Natural Log (base e log) of x.

For other bases, remember this law of logs: log-b(x) = log-k(x) / log-k(b) where log-b is the log in some arbitrary base b, and log-k is the log in base k, e.g.

here k = e

l = np.log(x) / np.log(100)

and l is the log-base-100 of x

kaveman
  • 4,339
  • 25
  • 44
22

I usually do like this:

from numpy import log as ln

Perhaps this can make you more comfortable.

Vincent
  • 329
  • 2
  • 2
1

Numpy seems to take a cue from MATLAB/Octave and uses log to be "log base e" or ln. Also like MATLAB/Octave, Numpy does not offer a logarithmic function for an arbitrary base.

If you find log confusing you can create your own object ln that refers to the numpy.log function:

>>> import numpy as np
>>> from math import e
>>> ln = np.log  # assign the numpy log function to a new function called ln
>>> ln(e)
1.0
bfris
  • 5,272
  • 1
  • 20
  • 37
-3
from numpy.lib.scimath import logn
from math import e

#using: x - var
logn(e, x)
outoftime
  • 715
  • 7
  • 21
-3

You could simple just do the reverse by making the base of log to e.

import math

e = 2.718281

math.log(e, 10) = 2.302585093
ln(10) = 2.30258093
Simas Joneliunas
  • 2,890
  • 20
  • 28
  • 35
Raviole
  • 1
  • 1
    note `math.e` exists and `math.log` takes the base 2nd. so `math.log(10, math.e)` is correct, while the above would actually return ~0.43... – Alex Moore-Niemi Jan 15 '21 at 03:21