Now, we’ll see how we can get the substring for all the values of a column in a Pandas dataframe. This extraction can be very useful when working with data. For example, we have the first name and last name of different people in a column and we need to extract the first 3 letters of their name to create their username.
Example 1:
We can loop through the range of the column and calculate the substring for each value in the column.
# importing pandas as pdimport pandas as pd # creating a dictionarydict = {'Name':["John Smith", "Mark Wellington", "Rosie Bates", "Emily Edward"]} # converting the dictionary to a# dataframedf = pd.DataFrame.from_dict(dict) # storing first 3 letters of namefor i in range(0, len(df)): df.iloc[i].Name = df.iloc[i].Name[:3] df |
Output:
Note: For more information, refer Python Extracting Rows Using Pandas
Example 2: In this example we’ll use str.slice().
# importing pandas as pdimport pandas as pd # creating a dictionarydict = {'Name':["John Smith", "Mark Wellington", "Rosie Bates", "Emily Edward"]} # converting the dictionary to a # dataframedf = pd.DataFrame.from_dict(dict) # storing first 3 letters of name as usernamedf['UserName'] = df['Name'].str.slice(0, 3) df |
Output:
Example 3: We can also use the str accessor in a different way by using square brackets.
# importing pandas as pdimport pandas as pd # creating a dictionarydict = {'Name':["John Smith", "Mark Wellington", "Rosie Bates", "Emily Edward"]} # converting the dictionary to a dataframedf = pd.DataFrame.from_dict(dict) # storing first 3 letters of name as usernamedf['UserName'] = df['Name'].str[:3] df |
Output:
Example 4: We can also use str.extract for this task. In this example we’ll store last name of each person in “LastName” column.
# importing pandas as pdimport pandas as pd # creating a dictionarydict = {'Name':["John Smith", "Mark Wellington", "Rosie Bates", "Emily Edward"]} # converting the dictionary to a dataframedf = pd.DataFrame.from_dict(dict) # storing lastname of each persondf['LastName'] = df.Name.str.extract(r'\b(\w+)$', expand = True) df |
Output:
