将 float64 转换为 1 作为 int，将 naan 转换为 0 作为 int

Question

I am predicting loans, and have a column with number of months since last delinquency. There are many missing values because many people are not delinquent. I cannot fill in zeros, as that would mean they were delinquent. I would like to convert any numeric values to '1', and any Naan to '0'. The column is a float64.

I have tried converting float64 to int(), train['X25'] = int(train['X25']) but got TypeError: cannot convert the series to <class 'int'>

What it looks like

Loan Number Months Since Delinquency
 1           Naan 
 2            0
 3            3

What I want it to look like

   Loan      Delinquency
         1            0
         2            1
         3            1

Answer 1

您应该使用df['Delinquency '] = df['Months Since Delinquency'].isna()来获取布尔列

Answer 2

这应该工作

df['Delinquency'] = df['Months Since Delinquency'].notnull().astype(int)