CODEWITHSUNDEEP
python-3.xpandasdataframecategorical-data
Jayit Ghosh
Jayit Ghosh

Reputation: 141

pandas.dataframe.astype is not converting dtype

I am trying to convert some columns from object to Categorical columns.

    # dtyp_cat = 'category'
        # mapper = {'Segment':dtyp_cat,
        #           "Sub-Category":dtyp_cat,
        #           "Postal Code":dtyp_cat,
        #           "Region":dtyp_cat,
        #          }
    
    df.astype({'Segment':'category'})
    df.dtypes

But the output is still object type.
enter image description here

Dataset is hosted at:

url = r"https://raw.githubusercontent.com/jaegarbomb/TSF_GRIP/main/Retail_EDA/Superstore.csv"
df = pd.read_csv(url)

Upvotes: 0

Views: 2368

Answers (1)

Serge de Gosson de Varennes
Serge de Gosson de Varennes

Reputation: 11474

Do this:

df['Segment'] = df.Segment.astype('category')

Which returns

RangeIndex: 9994 entries, 0 to 9993
Data columns (total 13 columns):
 #   Column        Non-Null Count  Dtype   
---  ------        --------------  -----   
 0   Ship Mode     9994 non-null   object  
 1   Segment       9994 non-null   category
 2   Country       9994 non-null   object  
 3   City          9994 non-null   object  
 4   State         9994 non-null   object  
 5   Postal Code   9994 non-null   int64   
 6   Region        9994 non-null   object  
 7   Category      9994 non-null   object  
 8   Sub-Category  9994 non-null   object  
 9   Sales         9994 non-null   float64 
 10  Quantity      9994 non-null   int64   
 11  Discount      9994 non-null   float64 
 12  Profit        9994 non-null   float64 
dtypes: category(1), float64(3), int64(2), object(7)
memory usage: 946.9+ KB

EDIT

If you want to convert several columns (In your case, I suppose it is all that are objects, you need to drop those that aren't, convert what's left and then reattach the other columns.

df2 = df.drop([ 'Postal Code', 'Sales', 'Quantity', 'Discount', 'Profit'], axis=1)
df3 = df2.apply(lambda x: x.astype('category'))

which gives

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 9994 entries, 0 to 9993
Data columns (total 8 columns):
 #   Column        Non-Null Count  Dtype   
---  ------        --------------  -----   
 0   Ship Mode     9994 non-null   category
 1   Segment       9994 non-null   category
 2   Country       9994 non-null   category
 3   City          9994 non-null   category
 4   State         9994 non-null   category
 5   Region        9994 non-null   category
 6   Category      9994 non-null   category
 7   Sub-Category  9994 non-null   category
dtypes: category(8)
memory usage: 115.2 KB

I'll leave the appending the other columns to you. A hint would be:

df4 = pd.concat([df3, df], axis=1, sort=False)
df_final = df4.loc[:,~df4.columns.duplicated()]

Upvotes: 1

Related Questions