How to delete rows for a specific date in the Pandas data framework?

Question

I have a Pandas DataFrame using Date as an index. How can I delete all rows with the date "2000-01-06"?

Code example:

import numpy as np
import pandas as pd

dates = pd.date_range('1/1/2000', periods=8)
df = pd.DataFrame(np.random.randn(8, 3), index=dates, columns=['A', 'B', 'C'])
df.index.name = 'Date'

Example DataFrame:

                   A         B         C
Date                                    
2000-01-01 -0.501547 -0.227375  0.275931
2000-01-02  0.994459  1.266288 -0.178603
2000-01-03 -0.982746 -0.339685  0.157390
2000-01-04 -1.013987 -1.074076 -2.346117
2000-01-05 -0.101157 -0.698663  1.025318
2000-01-06 -1.697615 -0.081638  1.075712
2000-01-07  0.617587 -1.561204 -1.685774
2000-01-08  0.049115  0.579139 -1.036961

Answer 1

You can pass the date and time dropto delete this line:

In [12]:
df.drop(pd.to_datetime('2000-01-06'))

Out[12]:
                   A         B         C
Date                                    
2000-01-01 -0.401531 -1.076727  0.519324
2000-01-02  0.022450  0.655763 -0.592045
2000-01-03  0.579927  1.358475  0.803414
2000-01-04  0.346246 -0.252332 -1.347014
2000-01-05  0.101308  0.912279  0.020754
2000-01-07  0.869264  0.699575  0.385521
2000-01-08  0.098829 -0.237605  1.112033

Answer 2

You can also delete the list of values, for example:

date_list = [datetime(2009, 5, 2),
             datetime(2010, 8, 22),
             datetime(2010, 9, 19),
             datetime(2011, 6, 19),
             datetime(2011, 7, 17),
             datetime(2015, 5, 23),
             datetime(2016, 2, 20)]
df = df.drop(date_list)

Note that by putting inplace = True in the drop argument, you do not need to define a new object, but it runs on the same object

df.drop(date_list, inplace=True)