I have a DF with column containing array of values as per below

I would like to map the values within the array in column "urls" to a dictionary and replace the values to those from the dictionary. Example dictionary below

mapping_dictionary = {'https:/1': 'google',
 'https:/2': 'yahoo',
 'https:/3': 'bing'}

So end result would look something like this:

Is it possible to do this using pandas.replace or pandas mapping?

Thanks in advance

CodePudding user response：

You can explode "urls" columns, map "mapping_dictionary", aggregate by the index to lists:

df['urls'] = df['urls'].explode().map(mapping_dictionary).groupby(level=0).agg(list)

Output:

   name             urls
0   bob  [google, yahoo]
1  mary   [google, bing]

CodePudding user response：

Let us try

df['value' = df.urls.map(lambda x : [*map(mapping_dictionary.get, x)])
0    [google, yahoo]
1     [google, bing]
Name: urls, dtype: object

CodePudding user response：

You can change the urls using apply and lambda, as follows:

import pandas as pd

df = pd.DataFrame({
    'name': ['bob', 'mary'],
    'urls': [['https:/1', 'https:/2'], ['https:/1', 'https:/3']]
})

m_dict = {
    'https:/1': 'google',
    'https:/2': 'yahoo',
    'https:/3': 'bing'
}


df['urls'] = df['urls'].apply(lambda urls: [m_dict[url] for url in urls])

print(df)
#   name             urls
#0   bob  [google, yahoo]
#1  mary   [google, bing]