How to replace values in a pandas dataframe based on conditions?

Question

I attempt to apply a different formula to a new column "result" based on the column 'C' containing the condition. If C is 'add' then I would like to add X and Y. When C is 'mult' the result should be X * Y.

df = pd.DataFrame({'X': [0, 1, 2, 3, 4],
                   'Y': [5, 6, 7, 8, 9],
                   'C': ['add', 'add', 'mult', 'mult', 'mult']})
df['result'] = df['X'] * df['Y']

df.loc[df.C =='add', 'result'] = df.loc[df['C'] =='add', 'X'] \
                                 + df.loc[df['C'] =='add', 'Y']
df

The result I get is:

      C  X  Y  result
0   add  0  5       5
1   add  1  6       5
2  mult  2  7      14
3  mult  3  8      24
4  mult  4  9      36

What I need is 'result' in row 1 being 7

      C  X  Y  result
0   add  0  5       5
1   add  1  6       7
2  mult  2  7      14
3  mult  3  8      24
4  mult  4  9      36

Using your code I have the right result on my machine using pandas 0.24.2 — Benjamin Breton
– Benjamin Breton, Commented May 17, 2019 at 9:22

nassim · Accepted Answer · 2019-05-17 09:29:47Z

3

your code gives right results, but if you want a direct way

df['result'] = df.apply(lambda x :  x['X']+x['Y'] if x['C'] == 'add' else x['X']*x['Y'], axis=1 )

output :

   X  Y     C  result
0  0  5   add       5
1  1  6   add       7
2  2  7  mult      14
3  3  8  mult      24
4  4  9  mult      36

answered May 17, 2019 at 9:29

nassim

1,5811 gold badge16 silver badges28 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

nassim Over a year ago

like every other function it has its downsides and its upsides, if you are talking about performance, you should know that performance depends on your dataset size. So if you have a small dataset just use it, thank you anyway

jezrael · Accepted Answer · 2019-05-17 09:34:55Z

Your solution working nice, also is posible use this alternative with numpy.where:

mask = df.C =='add'
df['result'] = np.where(mask, df['X'] + df['Y'], df['X'] * df['Y'])        
print (df)

   X  Y     C  result
0  0  5   add       5
1  1  6   add       7
2  2  7  mult      14
3  3  8  mult      24
4  4  9  mult      36

If more conditions is possible use numpy.select:

m1 = df.C =='add'
m2 = df.C =='mult'
m3 = df.C =='div'
v1 = df['X'] + df['Y']
v2 = df['X'] * df['Y']
v3 = df['X'] / df['Y']

df['result'] = np.select([m1, m2, m3], [v1, v2, v3])        
print (df)

   X  Y     C     result
0  0  5   add   5.000000
1  1  6   add   7.000000
2  2  7  mult  14.000000
3  3  8  mult  24.000000
4  4  9   div   0.444444

Collectives™ on Stack Overflow

How to replace values in a pandas dataframe based on conditions?

2 Answers 2

1 Comment

Comments

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

Comments

Related