I have a column that contains Friday-Friday dates ex. Fri March 4 to Fri March 11. I only want to filter the earliest Friday date. Any suggestions. I figured a way to sort out the min value, but I feel like there's a better method
df['Submitted On'] = pd.to_datetime(df['Submitted On'])
early = df['Submitted On'].min()
df = df.loc[df['Submitted On'] != early]
Although I don't know the use case for your data, your method is a little brittle. If for some reason the range of dates in your column changes, then you're filtering out the earliest date regardless of whether it's a Friday or not.
You can use the .dt.dayofweek method for Series which will return integers 0 through 6 for the day of the week meaning Friday is 4, and filter based on the first occurrence of a Friday. For example:
df = pd.DataFrame({'Submitted On': pd.date_range('2022-03-04','2022-03-11'), 'value':range(8)})
df['Submitted On'] = pd.to_datetime(df['Submitted On'])
filtered_df = df.drop(labels=df[df['Submitted On'].dt.dayofweek == 4].index.values[0])
Result:
Submitted On value
1 2022-03-05 1
2 2022-03-06 2
3 2022-03-07 3
4 2022-03-08 4
5 2022-03-09 5
6 2022-03-10 6
7 2022-03-11 7
And note that if I change the date range slightly, it still drops the first Friday:
df = pd.DataFrame({'Submitted On': pd.date_range('2022-03-03','2022-03-12'), 'value':range(10)})
filtered_df = df.drop(labels=df[df['Submitted On'].dt.dayofweek == 4].index.values[0])
Result:
Submitted On value
0 2022-03-03 0
2 2022-03-05 2
3 2022-03-06 3
4 2022-03-07 4
5 2022-03-08 5
6 2022-03-09 6
7 2022-03-10 7
8 2022-03-11 8
9 2022-03-12 9