Get a list from Pandas DataFrame column headers
Ring the bell! You just need to dance with df.columns.tolist() to get a list of column headers from a Pandas DataFrame. Here's a quick run:
Get your groove on with a list like: ['Column1', 'Column2', ...]. It's the most direct and effective route to stay in sync with your DataFrame's structure.
More tricks, less hocus-pocus
Let's kick the elegance up a notch. Beyond df.columns.tolist(), there are a few other cool cats in town for accessing DataFrame headers.
Deconstructing the DataFrame
Unpack your DataFrame like a boss in Python 3.5+:
Feels like opening a Christmas present, right?
Or store the headers straight into variables:
Like a surgeon doing a precision operation. No mess with intermediate variables or function calls.
Strive for speed
Are you a speed junkie? Step on the gas with:
my_dataframe.columns.values.tolist()df.columns.to_numpy().tolist()
Use %timeit to measure your speed because remember: fast cars need speedometers.
Mastering the lego house
Add a twist, get a set instead of a list for quick lookups with {*df}. Because speed sometimes matters more than order, right?
The world is not always fast and furious. Take control knowing your environment. Updates in Python (PEP 448) and Pandas (0.16.0 and onwards) can save your day.
Rock the comprehension list
If loops rock your boat:
But beware! You might get the fame but not the efficiency.
Dig deeper, it's not just a list
Let's not pigeonhole ourselves with a simple list. Here's a set of muscle flexing approaches in accessing and transforming DataFrame column headers.
In the form of sets
Because sometimes, order does not matter - {*df}. Fast and Furious.
Immutable sequencing
Like the law of gravity, certain things just don't change - tuple(df).
Printing with a style
Getting a list of headers is cool, but what about printing them neat?
Bring in the print party π
Maintain order
Remember: list(df) preserves the original order. Just like good old times.
Was this article helpful?