how converting object to float in dataframe

Question

I have a dataframe in each columns I have numbers but the type is object! how I can convert object to float?

I tried this : pd.to_numeric(df['Close_x'],errors='coerce')

but the error says:

TypeError: arg must be a list, tuple, 1-d array, or Series

I tried concatenated['Close_x'].astype(float)

the error says:

ValueError: could not convert string to float: '1,001.96'

I get this error: ` 'DataFrame' object has no attribute 'str'` — , Aug 15 '20 at 04:35
``Close_x`` might be duplicate, print(concatenated.columns) & see for your self. — sushanth, Aug 15 '20 at 04:38

score 1 · Accepted Answer · edited Aug 15 '20 at 05:17

1

These numbers have , as a separator so first, you need to replace it with an empty string, then convert it to a floating-point number.

df = pd.DataFrame(['1,001.96', '1001.98'], columns=['value'])
pd.to_numeric(df['value'].replace(',', '',regex=True))

or

df.apply(lambda x: x.str.replace(',', '')).astype(float)

Note: df.apply is slower than pd.to_numeric for large dataframes

edited Aug 15 '20 at 05:17

answered Aug 15 '20 at 05:14

jigsaw373

1 Answers1