CODEWITHSUNDEEP
pythonmathnumpyexponential
Raaj
Raaj

Reputation: 1200

python exponential function - incorrect exponential values

I have a range of negative values : -30 to -80 Let's say 

array_values = [-30, -31, -32, -33, -34, -35, -36, -37, -38, -39, -40, -41, -42, -43, -44, -45, -46, -47, -48, -49, -50, -51, -52, -53, -54, -55, -56, -57, -58, -59, -60, -61, -62, -63, -64, -65, -66, -67, -68, -69, -70, -71, -72, -73, -74, -75, -76, -77, -78, -79, -80]

I'm evaluating this expression : |(-k).exp(array_values[idx]/k)| for idx in array_values

k=35

The problem I'm having is that, there's no smooth exponential curve I'm getting after plotting this expression.

So I thought I'd check with just exp(array_values[idx]/k) Here are a bunch of values I get from the calculator and python(both math.exp() and numpy.exp() respectively)

exp(-30/35): Calculator: 0.42437284567   Python:0.36787944117144233
exp(-31/35):             0.4124194827           0.36787944117144233                 
exp(-32/35):             0.40080281159          0.36787944117144233
exp(-33/35):             0.38951334871          0.36787944117144233
exp(-34/35):             0.38754187754          0.36787944117144233

Am I doing something wrong, or is there a bug in the function? My code line for this is 

expos = [math.exp(-i/35) for i in array_values]

Upvotes: 0

Views: 3777

Answers (4)

user1196549
user1196549

Reputation:

As you can check, 0.36787944117144233 is just e^(-1), and you can infer that you are passing the floor of the fractional arguments (because you are performing an integer division).

Nose-on-your-face bugs of such a kind in fundamental functions are highly unlikely.

Upvotes: 1

Cesar
Cesar

Reputation: 717

You can do:

expos = [math.exp(float(-i)/35) for i in array_values]

The integer division returns integer values for example:

>>> 1/2
0

but a float value:

>>> 1.0/2
0.5

Upvotes: 0

igavriil
igavriil

Reputation: 1021

Dividing int with int results in int. So

array_values = [-30, -31, -32, -33, -34]
test = [-i/35 for i in array_values]
>>> [0, 0, 0, 0, 0]
expos = [math.exp(-i/35) for i in array_values]
>>> [1.0, 1.0, 1.0, 1.0, 1.0]

If you convert one of the numbers in the division to float will produce the expected results.

test = [float(-i)/35 for i in array_values]
>>> [0.8571428571428571, 0.8857142857142857, 0.9142857142857143, 0.9428571428571428, 0.9714285714285714]

expos = [math.exp(-float(i)/35) for i in array_values]
#or expos = [math.exp(-i/35.0) for i in array_values]
>>> [2.3564184423836605, 2.4247157128777834, 2.494992477791509, 2.5673061098152665, 2.6417156445012626]

Upvotes: 2

Selcuk
Selcuk

Reputation: 59444

You are converting your values to positive by putting a minus in front of an already negative value. You should also divide it to 35.0 so that it returns a float instead of an int. This works:

import math
array_values = [-30, -31, -32, -33, -34, -35, -36]
expos = [math.exp(i/35.0) for i in array_values]
print expos

result:

[0.42437284567695, 0.4124194827001579, 0.40080281159210923, 0.3895133487108618, 0.37854187754140095, 0.36787944117144233, 0.35751733497916927]

Upvotes: 1

Related Questions