Here is my code:
y = np.array([-3.44 , 1.16 , -0.81])
y = np.exp(y)
print(y)
for this block, I got the below result
[0.03206469 3.18993328 0.44485807]
However, when I add 3.91 to the list, the result changed
x = np.array([-3.44 , 1.16 , -0.81 , 3.91])
x = np.exp(x)
print(x)
result:
[3.20646853e-02 3.18993328e+00 4.44858066e-01 4.98989520e+01]
how can I prevent this change?
You can use np.set_printoptions:
>>> x = np.array([-3.44 , 1.16 , -0.81 , 3.91])
>>> x = np.exp(x)
>>> print(x)
[3.20646853e-02 3.18993328e+00 4.44858066e-01 4.98989520e+01]
>>> np.set_printoptions(suppress=True, precision=8)
>>> print(x)
[ 0.03206469 3.18993328 0.44485807 49.89895197]
Explanation:
Decimal rounding can not always be exactly represented in binary, so you will see some floating point inconsistencies. For example:
>>> 0.1 + 0.2
0.30000000000000004
In light of these inconsistencies, numpy by default represents floating points in scientific notation. The way I have shown above, only sets the printing option to the way you desire, but the actual representation inside the array does not change.
np.set_printoptions(suppress=True) suppresses the scientific notation, and suppresses the floating points to 8 decimal places by default, so technically the precision parameter was not needed in this case:
>>> np.set_printoptions(suppress=True)
>>> x
[ 0.03206469 3.18993328 0.44485807 49.89895197]
I added precision just in case you want a desired precision while printing.
>>> np.set_printoptions(suppress=True, precision=2)
>>> x
array([ 0.03, 3.19, 0.44, 49.9 ])
Learn more about floating point operations here:
