熊猫read_csv-千位分隔符工作不一致

Question

I am trying to read a csv file with commas as thousands separators into a pandas dataframe. The one time I am running the code it runs fine, while the other time it does not parse one of the columns correctly. Please find an example of my code below:

df_uv = pd.read_csv(file, sep=',', parse_dates=[0, 1, 2], usecols=columns_to_use,
                    thousands=',').fillna(0)

Please find below an example of the way the data is formatted in the csv file

03 Jan 2017,29 Dec 2016,30 Dec 2016,XXX,XXXX,XXXX,,,,,"14,214.08","-9,095.36","23,309.44","7,530.48","-15,962.64","23,493.12",

Any tips on how to tackle this issue?

Answer 1

I decided to change my code into the following:

df_uv = pd.read_csv(file, sep=',', parse_dates=[0, 1, 2], thousands=',').fillna(0)
df_uv = df_uv[columns_to_use]

Which is working completely fine.