CODEWITHSUNDEEP
pythonpandasnumpydataframe
Siraj S.
Siraj S.

Reputation: 3751

how to update multiple columns of pandas dataframe inplace and in single instance

consider below pandas DataFrame

df = pd.DataFrame(np.random.randint(1,10,15).reshape(-1,3), columns = list('abc'))

i want to update values of all the three columns as ratios in place (ie columns values divided by sum of all three columns, axis = 1) in a single instance.

something like:

df = df.loc[:,['a','b','c']] / df.loc[:,['a','b','c']].sum(axis = 1)

how can i achieve this?

Upvotes: 0

Views: 456

Answers (1)

akuiper
akuiper

Reputation: 214967

You can use pandas.DataFrame.div method:

df.div(df.sum())

#          a           b           c
#0  0.071429    0.368421    0.173913
#1  0.214286    0.210526    0.391304
#2  0.214286    0.052632    0.043478
#3  0.285714    0.105263    0.130435
#4  0.214286    0.263158    0.260870

To divide by rows, specify axis accordingly:

df1 = df.div(df.sum(axis = 1), axis = 0)

Here is a test that the result data frame has rowsum of one.

df1.sum(axis = 1)

#0    1.0
#1    1.0
#2    1.0
#3    1.0
#4    1.0
#dtype: float64

Upvotes: 3

Related Questions