This lesson's interactive features are locked, please to keep using them
String Operations
Text data is especially messy. Restaurant names don't follow a consistent format, cuisines are typed by hand, menu items drift... it never ends.
Fortunately, Pandas .str methods let you transform text columns efficiently. You'll often need to:
- Standardize case –
"TACOS","Tacos", and"tacos"should all match - Remove whitespace –
" Los Pollos Hermanos "should be"Los Pollos Hermanos" - Replace text – swap spaces for hyphens in every restaurant name
- Extract patterns – pull the domain out of a reservation email
Leaving text data in a dirty state causes incorrect groupings, failed joins, and misleading reports!
Common String Methods
- Applying uppercase, lowercase, or title case:
df["restaurant"].str.upper() # ALL CAPS df["restaurant"].str.lower() # all lowercase df["city"].str.title() # Title Case - Removing leading/trailing whitespace:
# " Los Pollos Hermanos " becomes "Los Pollos Hermanos" df["restaurant"].str.strip() - Replacing text:
# Replace spaces with hyphens df["restaurant"].str.replace(" ", "-", regex=False) - Checking if a string contains a given substring:
# Returns a boolean Series df["cuisine"].str.contains("taco") - Splitting strings into parts:
# Split a reservation email at the @ symbol df["email"].str.split("@") # Returns a Series of lists df["email"].str.split("@").str[1] # Returns a Series of just the domain parts
Chaining
Because each .str method returns a new Series, you can chain them to apply several transformations in a row:
df["restaurant"].str.strip().str.lower().str.replace(" ", "-", regex=False)
Assignment
Complete the clean_device_registry function. It accepts a DataFrame and returns a copy with two cleaned-up columns.