CODEWITHSUNDEEP
pythonstring
newleaf
newleaf

Reputation: 2457

python convert float to string

I'm trying to read in a csv file through Pandas.

pd.read_csv('zip_mapping.gz',compression='gzip' ,header=None, sep=',')

But somehow I read in zip as float, like

 0     501.0
 1    1220.0
 2    1509.0
 3    1807.0
 4    2047.0

as I don't know zip is in which column before I read in the data, so I could not set dtype in pd.read_csv.

I want to change zip into int, but due to missing values I got " could not convert NA to int "error.

Tried 

str(zip).rstrip('0').rstrip('.')

But got this 

'0     501.0\n1    1220.0\n2    1509.0\n3    1807.0\n4    2047.0\nName: zip, dtype: float64'

Actually I want convert zip in float into str like 501, 1220, 1509, 1807, 2047 then I could further padding leading zeros.

Any suggestion? Thank you.

Upvotes: 2

Views: 12807

Answers (1)

akuiper
akuiper

Reputation: 214937

You can use Series.astype method to convert float to int then to string, here I am using df to refer to the data frame you read in from csv and df.zip to refer to the zip column (adjust accordingly):

df.zip.astype(int).astype(str).str.zfill(5)

#0    00501
#1    01220
#2    01509
#3    01807
#4    02047
#Name: zip, dtype: object

If there is NA in the column, and you want to keep them as is:

df['zip'] = df.zip.dropna().astype(int).astype(str).str.zfill(5)
df

#     zip
#0    NaN
#1  01220
#2  01509
#3  01807
#4  02047

Another option use string formatter:

df.zip.apply(lambda x: x if pd.isnull(x) else "{:05.0f}".format(x))

#0      NaN
#1    01220
#2    01509
#3    01807
#4    02047
#Name: zip, dtype: object

Upvotes: 3

Related Questions