Unique values from dataframe columns via loop

Question

Is it possible to get unique values from multiple columns? I would like each column to have it's own unique values output via a loop.

df["col A"].unique()

Output

Col A: 3, 4, 1

Col B: 6456, 312, 4

Based of the example data you gave us, could you tell us what is the expected output? This way we can understand your specific problem, since it seems that .unique() does exactly what you want. — Ralubrusto
– Ralubrusto, Commented Dec 22, 2020 at 17:00
If you want some kind of vectorized implementation of unique to all the columns of the df at once you could use df.apply(lambda col: col.unique()) but I'm not sure if it will work because each columns probably will have an output of different length — Yamil Essus
– Yamil Essus, Commented Dec 22, 2020 at 17:04

Tiroha · Accepted Answer · 2020-12-22 17:04:53Z

2

It kinda depends on what you want as output. You can do it like this to get a dict with each column:

unique_vals = {col:df[col].unique() for col in df}

But you probably don't want it as a dataframe like this, because there is no guarantee the amount of unique values is the same for each column.

If you only want to print it it's as simple as:

for col in df:
    print(f'{col}: {df[col].unique()}')

answered Dec 22, 2020 at 17:04

Tiroha

634 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Pygirl · Accepted Answer · 2020-12-22 17:09:34Z

2

Try using list comprehension:

cola, colb = [df[i].unique() for i in df]

or

df.apply(lambda x: x.unique())

answered Dec 22, 2020 at 17:09

Pygirl

13.4k6 gold badges36 silver badges48 bronze badges

BENY · Accepted Answer · 2020-12-22 17:09:04Z

1

Let us try

out = pd.DataFrame(map(set,df.values.T.tolist())).T
Out[161]: 
   0     1
0  1   312
1  3  6456
2  4     4

answered Dec 22, 2020 at 17:09

BENY

324k22 gold badges176 silver badges250 bronze badges