CODEWITHSUNDEEP
pythonnumpy
user3037208
user3037208

Reputation: 11

Planck's Energy Density Python

I want to plot the graphs for different temperatures and frequencies in Python, I defined a function, but when I want to plot it, it shows me an error.

def planck(v,T):
    h=6.62606957*(10**-34.0)
    c=3*(10**8.0)
    k=1.3806488*(10**-23.0)

    x=(h*8*pi/c**3.0)
    y=v**3
    exponente = (h*v/k*T)
    ex = math.exp(exponente)-1

    PLANCK=(x*y)*(ex**-1)
    return PLANCK

x0, xf, dx = 800,2*(10**8),1000  
X = arange(x0, xf, dx)
print X

P1=planck(X, 3000)

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-16-034ad82af010> in <module>()
----> 1 P1=planck(X, 3000)

<ipython-input-11-e52b9512e92c> in planck(v, T)
      7     y=v**3
      8     exponente = (h*v/k*T)
----> 9     ex = math.exp(exponente)-1
     10 
     11     PLANCK=(x*y)*(ex**-1)

 TypeError: only length-1 arrays can be converted to Python scalars

And then, if I just use exp instead of math.exp, the graph results a constant.

Upvotes: 1

Views: 2623

Answers (3)

Prashant Kumar
Prashant Kumar

Reputation: 22659

@Sven Marnach got the programming bit right, you need to use Numpy exp.

As for why you get the constants, you've written your function using SI units. As a result, your input must be in SI units.

You want to divide X by 1e9 if your wavelengths were in nanometers. As it stands, you are passing wavelengths of ~800 meters.

Upvotes: 0

Gael Goret
Gael Goret

Reputation: 62

Compute the constant values once, then hard-code them, if you need some speed...

Upvotes: -1

unutbu
unutbu

Reputation: 880707

In addition to using numpy.exp instead of math.exp, you also need to specify the dtype of X to be float rather than int32. With int32, the largest number representable is 

In [46]: np.iinfo('int32').max
Out[46]: 2147483647

and y=v**3 exceeds this value:

In [38]: v
Out[38]: 
array([      800,      1800,      2800, ..., 199997800, 199998800,
       199999800])
In [37]: y
Out[37]: 
array([  512000000, -2147483648, -2147483648, ..., -2147483648,
       -2147483648, -2147483648])

Thus the reason why your values look like constants is because of arithmetic overflow.

So change 

X = np.arange(x0, xf, dx)

to

X = np.arange(x0, xf, dx, dtype='float')

import matplotlib.pyplot as plt
import numpy as np
pi = np.pi

def planck(v,T):
    h=6.62606957*(10**-34.0)
    c=3*(10**8.0)
    k=1.3806488*(10**-23.0)

    x=(h*8*pi/c**3.0)
    y=v**3
    exponente = (h*v/k*T)
    ex = np.exp(exponente)-1

    PLANCK=(x*y)*(ex**-1)
    return PLANCK

x0, xf, dx = 800,2*(10**8),1000  
X = np.arange(x0, xf, dx, dtype='float')

P1 = planck(X, 3000)
plt.plot(X, P1)
plt.show()

enter image description here

Upvotes: 6

Related Questions