Python：插入似然函数的值范围始终返回相同的值

Question

I am writing a program to calculate a likelihood function. The parameter I am trying to fit is I. This is my function:

#loglikelihood
def like_xe(I):
    model=(0.1535*(Z/A)*((charge*c/v_array)**2))*(np.log((2*m_e*(gamma_array**2)*(v_array**2)*(w_array)/((I*1.6E-6)**2)))-(2*(v_array/c_speed)**2))
return np.sum(((-0.5*np.log(2*np.pi*xe_un))-((xe_sp-(model))/(2*xe_un**2))))

The "arrays" are all data arrays, all of length 50. Then, I want to calculate the likelihood function for 1000 values of I between 0.0001 and 0.001, and from this data extract the maximum value of the likelihood function.

I_list=np.arange(1E-4,0.001,1000)
like_val=[0 for like in range(len(I_list))]
for like in range(len(I_list)):
    like_val[like]=like_xe(I_list[like])

This only returns [-169.58003268336941]. Obviously, this is incorrect. What is the matter with how I am approaching this problem? Is it possibly because it only reads the return statement once?

EDIT: Changed the np.arange to np.linspace. This creates an array of the correct size, however, it is entirely full of the value -169.58003268336941.

Answer 1

np.arange takes start, stop and step. A step of 1000 between 1e-4 and 0.0001 means 1 value.

>> np.arange(1E-4,0.001,1000)
array([ 0.0001])

replace this with np.linspace and you should get a better iteration:

>> np.linspace(1E-4,0.001,10)
array([ 0.0001,  0.0002,  0.0003,  0.0004,  0.0005,  0.0006,  0.0007,
    0.0008,  0.0009,  0.001 ])

(set to 10 rather than 1000 for brevity)

https://docs.scipy.org/doc/numpy/reference/generated/numpy.linspace.html