Round decimals row-wise instead of column-wise in pandas

Question

According to the pandas.DataFrame.round documentation I can decide the rounding also column-wise. However, there is nothing written about a row-wise rounding. For instance, I have

               A
count       1010.00009
measure     54.45678
average     0.50483  

How could I make it that

               A
count       1010
measure     54.46
average     0.5048

Answer 1

When something supports columns but not the index, your only option is to transpose... twice.

df.T.round(decimals=dict(zip(df.index, [0, 2, 4]))).T

Or,

df.T.round(decimals=pd.Series([0, 2, 4], index=df.index)).T

Adding a .astype(object) call at the end, we have—

              A
count      1010
measure   54.46
average  0.5048

Answer 2

With zip and round (from python), and for print I am adding astype(object)

df.A=[round(x,y) for x , y in zip(df.A,[0,2,4])]
df.astype(object)

              A
count      1010
measure   54.46
average  0.5048

Answer 3

我们转置 DataFrame，然后将列舍入并再次转置。

df = df.T.round( { "row_name": 0, "row2" : 2, "row3" : 4 } ).T