astype(int) converts 4 to 3-CodePudding

I have a dataframe column like below

df2['over_ball']

0     6.0
1     5.0
2     4.0
3     4.0
4     3.0
5     2.0
6     1.0
7     6.0
8     5.0
9     4.0
10    3.0
11    3.0
12    2.0
13    1.0

So I thought of removing the digits after decimal and used astype(int). But I see that astype(int) converts the values of 4.0 to 3 and 2.0 to 1.

df2['over_ball'].astype(int)


0     6
1     5
2     3
3     3
4     3
5     1
6     1
7     6
8     5
9     3
10    3
11    3
12    1
13    1

Why is this happening?

CodePudding user response：

Based on your comment, your error is reproducible.

df2 = pd.DataFrame({'over_ball': [3.99999999999, 1.99999999999]})
print(df2)

# Output:
   over_ball
0        4.0  # instead of 3.99999999999
1        2.0  # instead of 1.99999999999

A simple cast doesn't work as expected:

>>> df2['over_ball'].astype(int)
0    3
1    1
Name: over_ball, dtype: int64

You need to use round before astype(int):

>>> df2['over_ball'].round().astype(int)
0    4
1    2
Name: over_ball, dtype: int64

Tip:

pd.options.display.float_format = "{:,.20f}".format
print(df2)

# Output:
               over_ball
0 3.99999999998999999917
1 1.99999999998999999917