I will need a little help on this small matters. Please let me know your thoughts on that.
In this test, we will see that '1.731' in a text file transform to '1.7309999999999999' in a dataframe.
This process is included in a larger process that compare 2 dataframe cell by cell to report any difference. So i dont want to round it, that will NOT be a solution for me
Some piece of code you can test:
# This will Work correctly
df1 = pd.DataFrame([['B',0.05,1428.6443,674,
1.731]], columns = range(5))
# This wont Work correctly
str_path = 'file_test.txt'
df1 = pd.read_csv(str_path, sep = '\t', header = None)
print(df1)
for i_row, t_row in df1.iterrows():
i_col = 0
for value in t_row:
try:
flt_1 = float(df1.iloc[i_row, i_col])
if abs(flt_1 - 1.731) < 0.001:
print('-----------------')
print('i_row: ', i_row)
print('value: \t', value)
print('flt_1: \t', flt_1)
print('iat: \t', df1.iat[i_row, i_col])
print('iloc: \t', df1.iloc[i_row, i_col])
print('loc: \t', df1.loc[i_row, i_col])
print('value: \t', value)
print('t_row: \t', t_row[i_col])
print('t_row: \n', t_row, '\n\n')
except: pass
# End LOOP
i_col += 1
The result I have is:
0 1 2 3 4 ... 7 8 9 10 11
0 B LU129109 HK008 0966.HK - ... 1428.6443 - - 674 1.731
[1 rows x 12 columns]
-----------------
i_row: 0
value: 1.7309999999999999
flt_1: 1.7309999999999999
iat: 1.7309999999999999
iloc: 1.7309999999999999
loc: 1.7309999999999999
value: 1.7309999999999999
t_row: 1.7309999999999999
t_row:
0 B
1 LU129109
2 HK008
3 0966.HK
4 -
5 XHKG
6 surance Holdings Company ORD HKD 0.05
7 1428.64
8 -
9 -
10 674
11 1.731
