scikit-learn LinearRegression IndexError

Question

I am working on a LinearRegression model to fill the null values for the feature Rupeepersqft. When I run the code, I am receiving this error:

IndexError                                Traceback (most recent call last)
<ipython-input-20-33d4e6d2998e> in <module>()
      1 test_data = data_with_null.iloc[:,:3]
----> 2 Rupeepersqft_predicted['Rupeepersqft'] = pd.DataFrame(linreg.predict(test_data))

IndexError: only integers, slices (`:`), ellipsis (`...`), numpy.newaxis (`None`) and integer or boolean arrays are valid indices

This is the code which gives me the error:

from sklearn.linear_model import LinearRegression
linreg = LinearRegression()

data_with_null = data2[['Price (Lakhs)','Area','Area Type','Rupeepersqft','Condition','Purchase Type','Real Estate Regulation Act']].dropna()
data_without_null =  data_with_null.dropna()

train_data_x = data_without_null.iloc[:,:3]
train_data_y = data_without_null.iloc[:,3]

linreg.fit(train_data_x, train_data_y)

test_data = data_with_null.iloc[:,:3]
Rupeepersqft_predicted['Rupeepersqft'] = pd.DataFrame(linreg.predict(test_data))

data_with_null.Rupeepersqft.fillna(Rupeepersqft_predicted, inplace=True)

This is how the data looks like:

Can anyone help me out with this?

Rupeepersqft_predicted['Rupeepersqft'] would be ok if it is selecting a column from a dataframe (of value from a dict). But apparently it is an array, which doesn't accept string indexing (unless it's a structured array). — hpaulj
– hpaulj, Commented Nov 18, 2021 at 19:50

Michael · Accepted Answer · 2021-11-19 08:32:55Z

2

To assign values to a column in Pandas.DataFrame you should use the locators, i.e., loc and iloc (for array-like manipulations), so to fix your issue try changing the

Rupeepersqft_predicted['Rupeepersqft'] = pd.DataFrame(linreg.predict(test_data))

to:

Rupeepersqft_predicted.loc[:, 'Rupeepersqft'] = pd.DataFrame(linreg.predict(test_data))

which will chose all the rows (the :), and the column Rupeepersqft, and assign whatever values you have on the right.

or by using the iloc:

Rupeepersqft_predicted.iloc[:, 1] = pd.DataFrame(linreg.predict(test_data))

to assign it to the all rows (again by : operator) of the 1st column of the DataFrame.

Just make sure the values on the right are of the same length as the column you try to assign it to.

More on Pandas you can find in this book.

Cheers

edited Nov 19, 2021 at 8:32

answered Nov 19, 2021 at 8:24

Michael

2,3896 gold badges27 silver badges45 bronze badges

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

1 Comment

Michael Over a year ago

You are welcome, and I'm really happy it helped you. Please consider accepting this answer (by clicking the "v" next to the answer) to let other people, which encounter the same issue, know that it solves it. Cheers mate.

Collectives™ on Stack Overflow

scikit-learn LinearRegression IndexError

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related