How to convert an asymmetric pairwise distance matrix to dictionary?

user_15887317 0
Viewed 112

I've distance matrix which I want to convert to a dict with two keys and one value. My csv file looks something like this:

City1 City2 City3
City1 0 2.2 3.1
City2 2.1 0 4.0
City3 3.2 4.3 0

And I imported my csv file as a pandas data frame with pd.read_csv and now I want to convert this data frame to a dictionary which should use the first column and the first row as keys and the rest as values. So like:

{('City1','City1') : 0, ('City1','City2') : 2.1, ('City1','City3') : 3.2, ...}

I tried to use pandas .to_dict function but I wasn't able to figure out how to tell this function to not only use the column names as the keys. Thank you in advance.

pandas dataframe dictionary python
1 Answers

Let's try stack then to_dict:

import pandas as pd

df = pd.DataFrame({
    'City1': {'City1': 0, 'City2': 2.1, 'City3': 3.2},
    'City2': {'City1': 2.2, 'City2': 0.0, 'City3': 4.2},
    'City3': {'City1': 3.1, 'City2': 4.0, 'City3': 0},
})

d = df.stack().to_dict()
print(d)

df:

       City1  City2  City3
City1    0.0    2.2    3.1
City2    2.1    0.0    4.0
City3    3.2    4.2    0.0

d:

{('City1', 'City1'): 0.0, ('City1', 'City2'): 2.2, ('City1', 'City3'): 3.1,
 ('City2', 'City1'): 2.1, ('City2', 'City2'): 0.0, ('City2', 'City3'): 4.0,
 ('City3', 'City1'): 3.2, ('City3', 'City2'): 4.2, ('City3', 'City3'): 0.0}
user_154978880
Related