CODEWITHSUNDEEP
pythonpandas
Alan
Alan

Reputation: 2498

How to Check if a Column Contains a List in Pandas (Python)?

When I run a drop_duplicates on my DataFrame I am getting an unhashable type: list error. It's because my df contains lists. However, it's non-obvious to me that it does. When I run df.dtypes the columns return as object.

I could do df.apply(lambda x: str(x) if isinstance(x, list) else x), which stringifies everything. However there are other unhashable types I would have to try. Like, for all I know there are dictionaries or tuples holding mutables, and these would give the same error

In order to overcome this, I have had to manually inspect my df to figure which columns are list, and handle them accordingly (dropping them worked).I don't like this. Is there an easier pandas command to return columns containing lists? For example, might there be something like, if my dataframe is:

df = pd.DataFrame([[1, 'a', [1,2,3]], [2, "b", [4,5,6]]], columns = ["number", "letter", "unhashable"])

Which looks like:

    number  letter  unhashable
0   1   a   [1, 2, 3]
1   2   b   [4, 5, 6]

Then there's a method I can run on df, something like df.detailed_dtypes, to give me back

number: int,
letter: str,
unhashable: list

Upvotes: 1

Views: 424

Answers (1)

BENY
BENY

Reputation: 323276

Try with

df.iloc[0].map(lambda x : type(x).__name__)
Out[136]: 
number        int64
letter          str
unhashable     list
Name: 0, dtype: object

Upvotes: 3

Related Questions