Pandas SettingWithCopyError

Asked Feb 09 '20 at 06:50

Active Feb 09 '20 at 07:09

Viewed 9 times

I still do not quite understand the SettingWithCopyError, none of the following seem to work:

df['year'] = df['date'].dt.year

df.loc[:,'year'] = df['date'].dt.year

df.loc[:,'year'] = df.loc[:,'date'].dt.year

pd.__version__ '0.25.3'

Thank you!

The complete code:

df = other_df[['id','date','price']]

def date_augment(df, col):
    field =  df[col]
    attr = ['Year', 'Month', 'Week', 'Day', 'Dayofweek', 'Dayofyear', 'Is_month_end', 'Is_month_start', 
        'Is_quarter_end', 'Is_quarter_start', 'Is_year_end', 'Is_year_start']
    for n in attr: df[n] = getattr(field.dt, n.lower())
    df['Elapsed'] = field.astype(np.int64) // 10 ** 9

date_augment(df, 'date')

edited Feb 09 '20 at 07:09

asked Feb 09 '20 at 06:50

Dmitriy Grankin

I think obviously problem is in code above `df['year'] = df['date'].dt.year`, solution is obviously `.copy` – jezrael Feb 09 '20 at 06:52
@jezrael can you be more specific on how to use your solution? – Dmitriy Grankin Feb 09 '20 at 06:54
Can you add your all code? – jezrael Feb 09 '20 at 06:55
@jezrael `df['date'].dtype` `dtype(' – Dmitriy Grankin Feb 09 '20 at 06:59
I not ask for dtype, but for your completely code – jezrael Feb 09 '20 at 07:00
1

Change `df = other_df[['id','date','price']]` to `df = other_df[['id','date','price']].copy()` – jezrael Feb 09 '20 at 07:10
@jezrael the complete code see updated – Dmitriy Grankin Feb 09 '20 at 07:10

Pandas SettingWithCopyError

0 Answers0