Python Pandas: Get index of rows which column matches certain value
df.iloc[i]
returns the ith
row of df
. i
does not refer to the index label, i
is a 0-based index.
In contrast, the attribute index
returns actual index labels, not numeric row-indices:
df.index[df['BoolCol'] == True].tolist()
or equivalently,
df.index[df['BoolCol']].tolist()
You can see the difference quite clearly by playing with a DataFrame witha non-default index that does not equal to the row's numerical position:
df = pd.DataFrame({'BoolCol': [True, False, False, True, True]}, index=[10,20,30,40,50])In [53]: dfOut[53]: BoolCol10 True20 False30 False40 True50 True[5 rows x 1 columns]In [54]: df.index[df['BoolCol']].tolist()Out[54]: [10, 40, 50]
If you want to use the index,
In [56]: idx = df.index[df['BoolCol']]In [57]: idxOut[57]: Int64Index([10, 40, 50], dtype='int64')
then you can select the rows using loc
instead of iloc
:
In [58]: df.loc[idx]Out[58]: BoolCol10 True40 True50 True[3 rows x 1 columns]
Note that loc
can also accept boolean arrays:
In [55]: df.loc[df['BoolCol']]Out[55]: BoolCol10 True40 True50 True[3 rows x 1 columns]
If you have a boolean array, mask
, and need ordinal index values, you can compute them using np.flatnonzero
:
In [110]: np.flatnonzero(df['BoolCol'])Out[112]: array([0, 3, 4])
Use df.iloc
to select rows by ordinal index:
In [113]: df.iloc[np.flatnonzero(df['BoolCol'])]Out[113]: BoolCol10 True40 True50 True
Can be done using numpy where() function:
import pandas as pdimport numpy as npIn [716]: df = pd.DataFrame({"gene_name": ['SLC45A1', 'NECAP2', 'CLIC4', 'ADC', 'AGBL4'] , "BoolCol": [False, True, False, True, True] }, index=list("abcde"))In [717]: dfOut[717]: BoolCol gene_namea False SLC45A1b True NECAP2c False CLIC4d True ADCe True AGBL4In [718]: np.where(df["BoolCol"] == True)Out[718]: (array([1, 3, 4]),)In [719]: select_indices = list(np.where(df["BoolCol"] == True)[0])In [720]: df.iloc[select_indices]Out[720]: BoolCol gene_nameb True NECAP2d True ADCe True AGBL4
Though you don't always need index for a match, but incase if you need:
In [796]: df.iloc[select_indices].indexOut[796]: Index([u'b', u'd', u'e'], dtype='object')In [797]: df.iloc[select_indices].index.tolist()Out[797]: ['b', 'd', 'e']