Reputation: 153
So I have been writing a code to standardize the elements of a matrix and the function I used is as follows:
def preprocess(Data):
if stdn ==True:
st=np.empty((Data.shape[0],Data.shape[1]))
for i in xrange(0,Data.shape[0]):
st[i,0]=Data[i,0]
for i in xrange(1,Data.shape[1]):
st[:,i]=((Data[:,i]-np.min(Data[:,i]))/(np.ptp(Data[:,i])))
np.random.shuffle(st)
return st
else:
return Data
It works very well outside the class but when used inside of it it gives me this error:
AttributeError: 'tuple' object has no attribute 'shape'
Any idea on how I can fix it??
P.S. This is a KNN classification code.
Upvotes: 15
Views: 94896
Reputation: 23449
.shape
is an attribute of numpy ndarrays and tuples don't have such attributes but it's possible to call numpy.shape
on a tuple to get its "shape".
import numpy as np
sh = np.shape(Data)
In general (not for OP), it's more useful to get the length of tuples:
len(Data)
Upvotes: 0
Reputation: 1136
According to the error you posted, Data
is of type tuple and there is no attribute shape
defined for data. You could try casting Data
when you call your preprocess
function, e.g.:
preprocess(numpy.array(Data))
Upvotes: 15