CODEWITHSUNDEEP
python-3.xpandasdictionary
haben
haben

Reputation: 65

Changing abbreviated state names with full name

I have a dataframe with a column in it containing state names. The names are a mix of US states abbreviations and other countries state name. And, I only want to change the us state names and leave others as it is.

Data

   Country        State          
1  United States  MI
2  United States  PA
3  New Zealand    Auckland
4  France         Île-de-France
5  United States  FL

I have tried this code:

states = {
        'AK': 'Alaska',
        'AL': 'Alabama',
        'AR': 'Arkansas',
        'AS': 'American Samoa',
        'AZ': 'Arizona',
        'CA': 'California',
        'CO': 'Colorado',
        'CT': 'Connecticut',
        'DC': 'District of Columbia',
        'DE': 'Delaware',
         .
         .
         .
      }
states = {state: abbrev for state, abbrev in states.items()}
Data['State_full'] = Data['State'].map(states)

It replaces the US states as it should but it also replaces the other countries state name with None. What am i missing? Thanks in advance.

Upvotes: 0

Views: 3641

Answers (2)

mozway
mozway

Reputation: 262124

If efficiency is important, you can use states.get with the same parameter as default value:

df['State'] = df['State'].map(lambda x: states.get(x, x))

map is faster than replace

Example on 100k rows, map is twice faster:

%%timeit
df['State'].map(lambda x: states.get(x, x))
# 13.3 ms ± 501 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)

%%timeit
df['State'].replace(states)
# 30.7 ms ± 1.57 ms per loop (mean ± std. dev. of 7 runs, 10 loops each)

Upvotes: 1

user17242583
user17242583

Reputation:

You can use Series.replace. It can take a dict, where the keys of the dict are values to find, and the values of the dict are the replacements. If a value isn't found in the dict, it will be left as-is.

df['State'] = df['State'].replace(states)

Output:

>>> df
         Country          State
1  United States      Minnesota
2  United States   Pennsylvania
3    New Zealand       Auckland
4         France  Île-de-France
5  United States        Florida

Upvotes: 1

Related Questions