Converting pandas dataframe into basket type orange table

Question

I have the following pandas dataframe and would like any pointers in converting this to an Orange table with a basket format:

orders_item=pandas.DataFrame({'order_number':['001','002','002','002','003','003','003','003','003'],
'item':['Beef','Beef','Fish','Eggs','Beef','Oil','Salt','Pepper','MSG']})

I'll like to convert it to an orange table with Basket format similar to this:

Beef

Beef, Fish, Eggs

Beef, Oil, Salt, Pepper, MSG

Any help will be much appreciated.

score 2 · Accepted Answer · answered May 16 '16 at 02:38

2

I'd groupby the order_id and print the results:

In [11]: for row in orders_item.groupby("order_number")["item"]:
             print(",".join(row[1]))
Beef
Beef,Fish,Eggs
Beef,Oil,Salt,Pepper,MSG

If you want the string:

In [12]: "\n\n".join([", ".join(row[1]) for row in orders_item.groupby("order_number")["item"]])
Out[12]: 'Beef\n\nBeef, Fish, Eggs\n\nBeef, Oil, Salt, Pepper, MSG'

answered May 16 '16 at 02:38

Andy Hayden

359,921
101
625
535

Thanks. How do I store that into an orange table? – BlackHat May 16 '16 at 02:53
1

You save it as a `*.basket` file and load it with Orange? – K3---rnc May 16 '16 at 17:05
@BlackHat See [this question](http://stackoverflow.com/q/5214578/1240268) to save a string to a file in python. Note that the other option is to export to csv (with `.to_csv`) and then read the csv with orange table. – Andy Hayden May 16 '16 at 22:46
Finally get it. I completely ignored the .basket extension. This makes sense. – BlackHat May 17 '16 at 15:56

Converting pandas dataframe into basket type orange table

1 Answers1