While working on VIF in multiple regression I came across this problem

Question

def get_vif_factors(X):
    X_matrix = X.as_matrix()
    vif = [variance_inflation_factor(X_matrix, i ) for i in range(X_matrix.shape[1])]
    vif_factors = pd.DataFrame()
    vif_factors['column'] = X.columns
    vif_factors['VIF'] = vif

    return vif_factors

And I got a error of-

'DataFrame' object has no attribute 'as_matrix'

Answer 1

Hello, @Pawan,

The as_matrix method is deprecated since 0.23.0, so you should use to_numpy instead.

I am giving you an example regarding your doubt:

For example let's assume we only need to convert the subset ['col1', 'col2', 'col4'] of our original DataFrame to a Numpy array. In that case, you might have some legacy code relying on as_matrix to convert, which looks more or less like:

df.as_matrix(['col1', 'col2', 'col4'])

While converting the above code to to_numpy you cannot simply replace the function name like in:

df.to_numpy(['col1', 'col2', 'col4'])  # WRONG

because to_numpy does not accept a subset of columns as parameter. The solution, in that case, would be to do the selection first and apply to_numpy to the result, as in:

df[['col1', 'col2', 'col4']].to_numpy()  # CORRECT