I want to read a csv but it culls the number of decimals:

fname = './sol/Pret-SB_A00DLR0_202205240635.pos'
skiprow = 0
with open(fname) as search:
    for i, line in enumerate(search):
        if "%  GPST" in line:
            skiprow = i
            break
df = pd.read_csv(fname, skiprows=skiprow, delim_whitespace=True, parse_dates=[[0, 1]])
df.head(2)

gives (first 2 rows, first five columns):

enter image description here

the original data (here) has 8 decimal places in the 3rd and 4th columns. I need those.

2211 196568.000 -25.732036008 28.282629130 1387.8994
2211 196569.000 -25.732032386 28.282633712 1389.4025

How do I read a csv and retain the precision of the original data?

CodePudding user response：

How do I read a csv and retain the precision of the original data?

You do have it, pandas simply limit number of digits for presentation purposes, consider following example

import pandas as pd
df = pd.DataFrame({'x':[28.282633712]})
print(df)
print(df.x[0])
print(df.x[0] == 28.282633712)

gives output

           x
0  28.282634
28.282633712
True

CodePudding user response：

You can set the number of displayed digits to 8 like this:

pd.options.display.float_format = "{:,.8f}".format