SettingWithCopyWarning 尝试对 pandas dataframe 中的用户进行分类时

Question

I am getting a SettingWithCopyWarning when trying to classify users in pandas dataframe,

I have a Dataframe that contains a 'user_id' column, that if it is contained in a determined list, it will give me a value and if not, then it will give me another

In this case, I am trying to create a new column that given a determined condition, (if the id is in a list) the it will give me a 'male' string as a value in the new column, but if the id is not in the list it will give me a female, this is what I am trying:

def select_user_type(df, male_list):
    types = {
        1: 'female',
        2: 'male'
    }
    search_column = 'user_id'
    df['user_sex'] = df.apply(lambda x: types[2] if x[search_column] in male_list else types[1], axis=1)
    return df

The error I get is the following

SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

Example:

male_list = [354, 899] #Example

output:

name    user_id   user_sex
Jane    890       female
Jean    899       male
Rita    708       female
John    354       male

Answer 1

Seems to work for me without issues. However lets try;

def select_user_type(df, male_list):
    types = {
        1: 'female',
        2: 'male'
    }
    search_column = 'user_id'
    df = df.assign(user_sex=df.apply(lambda x: types[2] if x[search_column] in male_list else types[1], axis=1))
    return df