transpose multiple rows to columns with pandas

Question

I have this excel table read in jupyter notebook with pandas. I want to melt the upper row side of the table into column. The table looks like as follow:

                              ori code   cgk       cgk    clg       clg
                              ori city   jakarta   NaN    cilegon   NaN
                              ori prop   jakarta   NaN    banten    NaN
                              ori area   jawa      NaN    jawa      NaN

code    city       district   island    type a     days    type b    days
001     jakarta    jakarta    jawa      12000       2      13000      3
002     surabaya   surabaya   jawa      13000       3      14000      4

I realized that df.melt should be worked to transpose the upper rows, but the type & dayscolumns, and also the 4 rows and the NaN value on it get me confuse on how to do that correctly.

The desire clean dataframe I need is as follow:

code   city        district    island    type       price_type   days    ori_code   ori_city     ori_prop   ori_area
001    jakarta     jakarta     jawa      type a     12000         2         cgk     jakarta    jakarta      jawa
001    jakarta     jakarta     jawa      type b     13000         3         clg     cilegon    banten       jawa 
002    surabaya    surabaya    jawa      type a     13000         3         cgk     jakarta    jakarta      jawa
002    surabaya    surabaya    jawa      type b     14000         4         clg     cilegon    banten       jawa

The ori_code, ori_city, ori_prop, ori_area would become column names.

So far what I have done is set fix index name which are code, city, district and also island.

df = df.set_index(['code','city','district','island'])

can anyone help me to solve this problem? Any helps would be much appreciated. Thank you in advance.

kindly share a reproducible form : df.to_dict()

sammywemmy
– sammywemmy

2022-12-21 05:59:25 +00:00
Commented Dec 21, 2022 at 5:59 — sammywemmy
– sammywemmy, Commented Dec 21, 2022 at 5:59

Hiren Namera · Accepted Answer · 2022-12-21 04:22:19Z

1

For this you can use pandas melt function like this:

import pandas as pd

# Set the index for the DataFrame
df = df.set_index(['code', 'city', 'district', 'island'])

# Use pd.melt to reshape the data
df = pd.melt(df, id_vars=['code', 'city', 'district', 'island'], var_name='type', value_name='price_type')

# Split the 'type' column into two columns: 'type' and 'days'
df[['type', 'days']] = df['type'].str.split(' ', expand=True)

# Drop the 'ori code', 'ori city', 'ori prop', and 'ori area' columns
df = df.drop(columns=['ori code', 'ori city', 'ori prop', 'ori area'])

# Reorder the columns
df = df[['code', 'city', 'district', 'island', 'type', 'price_type', 'days', 'ori_code', 'ori_city', 'ori_prop', 'ori_area']]

# Display the resulting DataFrame
print(df)

answered Dec 21, 2022 at 4:22

Hiren Namera

4343 silver badges12 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

transpose multiple rows to columns with pandas

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related