get_posneg.py

def get_posneg(df, anchor):
    """
        Given a user id anchor, it gives back the max number of triplets [anchor, positive, negative]
        available.
        Triplets are randomly shuffled to better feed the training network.
        
        Parameters
        ----------
            df : Pandas DataFrame
                    Dataframe containing ratings, having user id as rows, movie id as columns
                    
            anchor : int
                        user id to be serving as anchor.
        
        Return
        ------
            pn_list : list of int
                        list of P,N elements
                            values : Positive (Movie id evaluated at least 4.0),
                                     Negative (Movie id evaluates at most 3.0)
    """
    POS_THR = 4.0
    NEG_THR = 3.0
    
    ps = df.loc[anchor] >= POS_THR 
    pos_ids = ps[ps].index.values
    
    ns = df.loc[anchor] <= NEG_THR 
    neg_ids = ns[ns].index.values
    
    n, m = len(pos_ids), len(neg_ids)
    pn_list = []
    
    for i in range(n):
        for j in range(m):
            pn_elem = [pos_ids[i], neg_ids[j]]
            pn_list.append(pn_elem)
            
    import random
    random.shuffle(pn_list)
    
    return pn_list