array coding error in python

Question

I have a set of values

y_train=["User","Bot","User","Bot",.......]
y_pred=["Bot","User","User","Bot".........]

I want to generate an array which returns 1 if the values of y_train[i] and t_pred[i] dont match.Both y_train and y_pred consist of same no of values

That is array indicator should be:

indicator=[1,1,0,0..........]

I have tried

 indicator=0
 for i in range(len(y_train)):
      if y_train[i]!=y_pred[i]:
            indicator[i]=1
      else:
            indicator[i]=0

But error being shown is :

'int' object does not support item assignment

How could this be done? Thanks

idanshmu · Accepted Answer · 2014-01-16 12:12:34Z

3

indicator = map(lambda x,y: int(x != y), y_train, y_pred)

As commented, if indicator serves as a boolean array, you may remove the int conversion to generate a True & False array:

indicator = map(lambda x,y: x != y, y_train, y_pred)

or as suggested in the comment:

indicator =  map(operator.ne, y_train, y_pred)

edited Jan 16, 2014 at 12:12

answered Jan 16, 2014 at 10:20

idanshmu

5,3617 gold badges55 silver badges100 bronze badges

Sign up to request clarification or add additional context in comments.

4 Comments

volcano Over a year ago

int conversion excluded - the best answer.

idanshmu Over a year ago

If OP needs indicator values as True or False we may remove the conversion.

volcano Over a year ago

OP may be unaware of that option

glglgl Over a year ago

operator.ne is the same as lambda x,y: x != y.

Bartosz Marcinkowski · Accepted Answer · 2014-01-16 12:14:37Z

2

indicator = [t != p for t, p in zip(y_train, y_pred)]

edited Jan 16, 2014 at 12:14

answered Jan 16, 2014 at 10:04

Bartosz Marcinkowski

6,8914 gold badges41 silver badges72 bronze badges

2 Comments

volcano Over a year ago

Or just t==p, which is essentially integer in disguise

volcano Over a year ago

You don't need int, it just adds execution overhead. False and True are equivalent to 0 and 1. 1==True returns ... True and 5+True will evaluate to 6. Rule of the thumb - never convert when you don't have to (Python is not C and does not require strict explicit type castings)

Aswin Murugesh · Accepted Answer · 2014-01-16 10:03:35Z

1

You have declared indicator as an int. Try this:

indicator=[]
for i in range(len(y_train)):
  if y_train[i]!=y_pred[i]:
        indicator.append(1)
  else:
        indicator.append(0)

Another way to do this is:

indicator = []
for i,j in zip(y_train,y_pred):
    if i==j:
        indicator.append(0)
    else:
        indicator.append(1)

answered Jan 16, 2014 at 10:03

Aswin Murugesh

11.1k11 gold badges46 silver badges73 bronze badges

Comments

user2357112 · Accepted Answer · 2014-01-16 10:06:02Z

First, this is what you should do:

indicator = [1 if x != y else 0 for x, y in zip(y_train, y_pred)]
# or indicator = [int(x != y) for x, y in zip(y_train, y_pred)]

This is a list comprehension. It uses zip to go over the values of y_train and y_pred in parallel. For every pair of corresponding values in y_train and y_pred, indicator contains 1 if the values are unequal and 0 otherwise.

Now, here's what's wrong with your attempt. First, if you want indicator to be a list, don't make it an int:

#indicator=0
indicator = []

Second, you can't assign to indices past the end of a list. You can append the values instead:

for i in range(len(y_train)):
    if y_train[i]!=y_pred[i]:
#         indicator[i]=1
          indicator.append(1)
    else:
#         indicator[i]=0
          indicator.append(0)

Collectives™ on Stack Overflow

array coding error in python

4 Answers 4

4 Comments

2 Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

4 Comments

2 Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related