deleting rows with nan

Question

I'm reading a dat file and loading the values into a numpy array. This 2D array has some nan values in the second column:

myarray = array([[ 0.        ,       nan],
                 [ 0.00057525,         4],
                 [ 0.53212512,         5],
                 [ 0.53268938,       nan]])

I need to get rid off the rows with nan values. This is what I'm trying

import numpy as np
myarray = np.loadtxt(filename, skiprows=30, usecols=(0,1))
intensity = myarray[:,1]
for row in intensity:
    if np.isnan(intensity):
        np.delete(myarray, (0), axis=0)

print myarray

But I'm getting this error:

ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()

Any advices in how to proceed?

Thanks in advance