pandas how to create simple cross-tab without aggregation?

Question

I have a pandas table with 3 columns: parent_male, parent_female, offsprings - all strings. I want to create a simple sparse crosstab table of male vs female and the offsprings as the values - how can I write an aggfunc that do so. (no real aggregation is needed) - just put an empty string in the blanks.

score 6 · Accepted Answer · edited Sep 26 '17 at 22:04

6

IIUC you need pivot:

df = df.pivot(index='parent_male', columns='parent_female', values='offsprings')

If get error:

ValueError: Index contains duplicate entries, cannot reshape

use pivot_table

So final solution is:

ct = pd.pivot_table(d['male'], d['female'], d['offsprings'], aggfunc=','.join)

edited Sep 26 '17 at 22:04

LW001

2,452
6
27
36

answered Oct 30 '16 at 09:07

jezrael

822,522
95
1,334
1,252

thanks - I had a problem with the values as strings, since the index contains duplicate entries. I found the solution below... – Dror Hilman Oct 30 '16 at 09:09
1

if have there problem you can use `aggfunc=','.join` ;) – jezrael Oct 30 '16 at 09:10
final solution, that works well thanks to you guys: ct = pd.crosstab(hyb['male'], hyb['female'], hyb['line'], aggfunc=','.join) – Dror Hilman Oct 30 '16 at 09:14
There is interesting [question](http://stackoverflow.com/q/36267745/2901002) about `crosstab` vs `pivot_table`. – jezrael Oct 30 '16 at 09:21

score 2 · Answer 2 · edited May 23 '17 at 12:18

2

I found the answer here... Pandas Groupby Agg Function Does Not Reduce and I used the info. from the comments above...

ct = pd.crosstab(d['male'], d['female'], d['offsprings'], aggfunc=','.join)

edited May 23 '17 at 12:18

Community

1
1

answered Oct 30 '16 at 09:08

Dror Hilman

6,837
9
39
56

pandas how to create simple cross-tab without aggregation?

2 Answers2