Doubt at Implementation

[Marcel Caraciolo]

Hi Folks,

I've been developing the MatrixDataModel using Scipy and numpy arrays.

As soon as I got the implementation of preferences_from_user, one issue came to me:

when i get an user_id which does not have any preferences in the dataset, which result would be

the correct for this user_id:

For instance, consider the user_id Maria Gabriela: 

a)  It must not return anything, just an empty list since it does not have any preferences for that user.  []

b)  It must return all preferences with the value np.NaN, which represents that the user_id has not have evaluated yet that item_id. In our case it would return:

 ('Maria Gabriela', [('Just My Luck', nan), ('Lady in the Water', nan), ('Snakes on a Plane', nan), ('Superman Returns', nan), ('The Night Listener', nan), ('You, Me and Dupree', nan)]), 

This question also applies in cases of user_ids who have evaluated some item_ids and didn't preferred others. Does it return all the ids combined with NaNs and the preferences or only the item_ids wich the user preferred ?

What do you think ?

Regards,

--
Marcel Pinheiro Caraciolo
M.S.C. Candidate at CIN/UFPE

   http://www.mobideia.com
   http://aimotion.blogspot.com/

[Ricardo Caspirro]

In the case return just an empty list.

But could have other methods for return 1..n values NaN.

Ricardo Caspirro

[Marcel Caraciolo]

I decided to remove all NaN.

If the developer wants to get the NaN, he just need to implement their custom methods for fetching them.

Thanks Ricardo!

Marcel