当列在python pandas中具有对象dtype时，如何基于“列A”中的值填充“列B”？

Question

I have a CSV file which I imported as a pandas dataframe. I want to create and fill up a column based on some specific terms in another column. The column that has all those values is an object dtype. It has values like:

ABC|MNO - 2017 - Trial|1|Random|xyz|RUN|Google|1x1|A10001-21|SD|GH|PRIME - 2017 - Big - This is For Example 

The code I was using is:

def new(row):
  if row.str.contains("PRIME"):
      return 'A'
  if row.str.contains("Random"):
      return 'B'
  if row.str.contains("Google"):
      return 'C'

df['X'] = df['Y'].apply (lambda row: new (row))

This code is giving me following error:

AttributeError: 'str' object has no attribute 'str'

I think it is because Column X has Object dtype.

I tried converting it to a string using the code:

df['Y'] = df['Y'].astype('str')

but it doesn't work. Then I tried splitting it using the following code:

df['Y_new'] = df['Y'].str.split(r'([A-Z][^\.!?]*[\.!?])')

But it converted all the values to NaN . How should I do this?

Answer 1

Try doing this:

    def new(row):
      if row.contains("PRIME"):
         return 'A'
      if row.contains("Random"):
         return 'B'
      if row.contains("Google"):
         return 'C'