I have a simple text file composed of 8 columns and I read it with loadtxt function. I want to plot as y-axis column2-column5 and as x-axis column1-column4 divided by cos(column2-column5) so I put this commands
>>> y = data[:,2] - data[:,5]
>>> x = (data[:,1] - data[:,4])/cos(y)
and it gave this error
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
NameError: name 'cos' is not defined
What is the problem?
You probably want to use numpy arrays. Then you can use element wise division. Additionally numpy provides all necessary math functions.
import numpy as np
d = np.asarray(data)
y = d[:,2] - d[:,5]
x = (d[:,1] - d[:,4])/np.cos(y)
loadtxt and 2D indexing I'm sure the variables already are numpy arrays.
loadtxt function he is using, np.asarray makes sure that we will have ndarrays. If it is already one, nothing happens.matplotlib so it's surely numpy loadtxt in question. What else can you index like data[:,2]? (not a rhetorical question)cos is in math module (as well as other mathematical libraries like numpy etc.)
import math
math.cos(3)
for numpy
import numpy
numpy.cos( np.array( [ [1,2], [3,4] ] ) )
math.cos won't work for their use case