Hey guys! Hoping someone can help me with a pandas issue!

So I have a dataframe with a column that contains lists, which look like this:```
0 [Paid Search]
1 [Paid Search, unavailable]
2 [Paid Search, unavailable, Paid Search]
3 [Paid Search, unavailable, Paid Search, unavai...
4 [Paid Search, unavailable, Paid Search, unavai...
5 [Paid Search, unavailable, Paid Search, unavai...
6 [Paid Search, unavailable, Paid Search, unavai...
7 [Paid Search, unavailable, Paid Search, unavai...
8 [Paid Search, unavailable, Paid Search, unavai...
9 [Paid Search, unavailable, Paid Search, unavai...
10 [Paid Search, unavailable, Paid Search, Direct]
11 [Paid Search, unavailable, Paid Search, Direct...
12 [Paid Search, unavailable, Paid Search, Direct...
13 [Paid Search, unavailable, Paid Search, Direct...
14 [Paid Search, unavailable, Paid Search, Direct...
15 [Paid Search, unavailable, Paid Search, Direct...
16 [Paid Search, unavailable, Paid Search, Direct...
17 [Paid Search, unavailable, Paid Search, Direct...
18 [Paid Search, unavailable, Paid Search, Direct...
19 [Paid Search, unavailable, Paid Search, Direct...
```

etc.

And I'd like to find out all the unique values that are present in the lists in that column... any thoughts?

oops, it didn't print very nicely here :( sorry about that, but each number is a new row anyway

no worries! I got that down, the data I posted is already cleaned in a previous function so in that function I just tried adding each channel to a set. :)

Second problem – getting average length of each list in a row? feels unnecessary to iterate through the whole thing?

Second problem – getting average length of each list in a row? feels unnecessary to iterate through the whole thing?

:+1: for the first problem

Second problem: I am afraid you have to iterate, specially if the lengths are different between rows. Suggestion? First thing that comes to my mind is the

Second problem: I am afraid you have to iterate, specially if the lengths are different between rows. Suggestion? First thing that comes to my mind is the

`apply`

method in pandas? It will iterate inefficiently. I think you can get a better result by considering it as a numpy problem and vectorise but I am not sure...
@webel ^^

I'm thinking some combo of apply and reduce but having some syntax issues, thanks @evaristoc for the guidance!

webel sends brownie points to @evaristoc :sparkles: :thumbsup: :sparkles:

:cookie: 301 | @evaristoc |http://www.freecodecamp.com/evaristoc