CODEWITHSUNDEEP
pythonpandasindexing
Taylrl
Taylrl

Reputation: 3919

iloc giving 'IndexError: single positional indexer is out-of-bounds'

I am trying to encode some information to read into a Machine Learning model using the following

import numpy as np
import pandas as pd
import matplotlib.pyplot as py

Dataset = pd.read_csv('filename.csv', sep = ',')

X = Dataset.iloc[:,:-1].values
Y = Dataset.iloc[:,18].values
            
from sklearn.preprocessing import LabelEncoder, OneHotEncoder
labelencoder_X = LabelEncoder()
X[:, 0] = labelencoder_X.fit_transform(X[:, 0])
onehotencoder = OneHotEncoder(categorical_features = [0])
X = onehotencoder.fit_transform(X).toarray()

however I am getting an error that reads

IndexError: single positional indexer is out-of-bounds

Upvotes: 115

Views: 580676

Answers (6)

cottontail
cottontail

Reputation: 23041

To avoid this error altogether, you can use -1 to index the last column:

X = Dataset.iloc[:,:-1].values   # <--- all but the last column
Y = Dataset.iloc[:,-1].values    # <--- the last column

or index the last column and select it:

X = Dataset[Dataset.columns[:-1]].values   # <--- all but the last column
Y = Dataset[Dataset.columns[-1]].values    # <--- the last column

Another way is to pop the last column from the dataframe; pop() removes a column out of the dataframe it's called on.

X = Dataset.copy()                         # <--- copy Dataset into X
Y = X.pop(X.columns[-1])                   # <--- pop last column out of X

Upvotes: 0

Ravi kumar
Ravi kumar

Reputation: 330

Mostly other members did explained the issue and the solution as this code block is trying to access the element which is not exist.

However, for sanity before applying iloc, make sure the dataframe indexes are in proper order. For this you can simply execute this,

df.reset_index(inplace=True)

 (Or)

df = df.reset_index()

Upvotes: 3

Poornima Devi
Poornima Devi

Reputation: 151

This issue is seen when the input file (csv in this case) itself is empty somehow. Again it boils down to the same issue where it is trying to access the data from a column that doesn't exist.

Upvotes: 0

questionto42
questionto42

Reputation: 9522

It does not help for the solution of the question here, but whoever might come here for the error and not for the example, I had this error IndexError: single positional indexer is out-of-bounds when I tried to find a row in a dataframe2 while looping over the rows of dataframe1, using many criteria in the filter of dataframe2, and adding each found row to a new empty dataframe3 (Do not ask me why!). One of the values in the row was a "nan" value both in dataframe1 and dataframe2. I could not filter anymore nor add a new row.

Solution:

dataframe1.fillna("nan") # or whatever you want as a fill value
dataframe2.fillna("nan")

and the script ran through without the error.

Upvotes: 5

Nicolas Gervais
Nicolas Gervais

Reputation: 36594

This happens when you index a row/column with a number that is larger than the dimensions of your dataframe. For instance, getting the eleventh column when you have only three. 

import pandas as pd

df = pd.DataFrame({'Name': ['Mark', 'Laura', 'Adam', 'Roger', 'Anna'],
                   'City': ['Lisbon', 'Montreal', 'Lisbon', 'Berlin', 'Glasgow'],
                   'Car': ['Tesla', 'Audi', 'Porsche', 'Ford', 'Honda']})

You have 5 rows and three columns:

    Name      City      Car
0   Mark    Lisbon    Tesla
1  Laura  Montreal     Audi
2   Adam    Lisbon  Porsche
3  Roger    Berlin     Ford
4   Anna   Glasgow    Honda

Let's try to index the eleventh column (it doesn't exist):

df.iloc[:, 10] # there is obviously no 11th column

IndexError: single positional indexer is out-of-bounds

If you are a beginner with Python, remember that df.iloc[:, 10] would refer to the eleventh column.

Upvotes: 33

slonopotam
slonopotam

Reputation: 1710

This error is caused by:

Y = Dataset.iloc[:,18].values

Indexing is out of bounds here most probably because there are less than 19 columns in your Dataset, so column 18 does not exist. The following code you provided doesn't use Y at all, so you can just comment out this line for now.

Upvotes: 128

Related Questions