Pandas dataframe：将列转换为具有默认值的数字

Question

I'm surprised I couldn't find anything about this:

Q: How can I convert values of a column to numbers with a default value if casting fails.

I need all values to be numbers in order to perform arithmetical functions on them. The raw input data contains non-convertible values like "" , "**" , "not available" , ....

Eg

1
2
*
4

Should become:

1
2
0
4

What I've tried:

• to_numberic : Does not allow to set a default value. errors=coerce will set the value to NaN , which can not be handled arithmetically afterwards.
• df[COLUMN].apply (makeNumber) with def makeNumber(value): return 0.0 does apparently nothing on the string values.

Answer 1

If no missing values in original use:

df['col'] = pd.to_numeric(df['col'], errors='coerce').fillna(0)