Added the Spherical k-means based recommender

cornac/models/__init__.py

@@ -4,3 +4,4 @@
 from .pcrl import *
 from .pmf import *
 from .ibpr import *
+from .skm import *

cornac/models/hpf/recom_hpf.py

@@ -24,7 +24,7 @@ class Hpf(Recommender):
         The name of the recommender model.
 
     trainable: boolean, optional, default: True
-        When False, the model is not trained and Cornac assumes that the model already \
+        When False, the model is not trained and Cornac assumes that the model is already \
         pre-trained (Theta and Beta are not None). 
 
     init_params: dictionary, optional, default: None

cornac/models/skm/__init__.py

@@ -0,0 +1,5 @@
+
+from .recom_skmeans import Skmeans
+
+
+__all__ = ['Skmeans']

cornac/models/skm/recom_skmeans.py

@@ -0,0 +1,106 @@
+"""
+@author: Aghiles Salah <asalah@smu.edu.sg>
+"""
+
+import numpy as np
+from ..recommender import Recommender
+from .skmeans import *
+
+
+class Skmeans(Recommender):
+    """Spherical k-means based recommender.
+
+    Parameters
+    ----------
+    k: int, optional, default: 5
+        The number of clusters.
+
+    max_iter: int, optional, default: 100
+        Maximum number of iterations.
+
+    name: string, optional, default: 'Skmeans'
+        The name of the recommender model.
+
+    trainable: boolean, optional, default: True
+        When False, the model is not trained and Cornac assumes that the model is already \
+        pre-trained (Theta and Beta are not None). 
+        
+    tol : float, optional, default: 1e-6
+        Relative tolerance with regards to skmeans' criterion to declare convergence.
+        
+    verbose: boolean, optional, default: True
+        When True, the skmeans criterion (likelihood) is displayed after each iteration.
+
+    init_par: numpy 1d array, optional, default: None
+        The initial object parition, 1d array contaning the cluster label (int type starting from 0) \
+        of each object (user). If par = None, then skmeans is initialized randomly.
+      
+    centroids: csc_matrix, shape (k,n_users)
+        The maxtrix of cluster centroids.
+
+    References
+    ----------
+    * Salah, Aghiles, Nicoleta Rogovschi, and Mohamed Nadif. "A dynamic collaborative filtering system \
+    via a weighted clustering approach." Neurocomputing 175 (2016): 206-215.
+    """
+
+    def __init__(self, k=5, max_iter=100, name = "Skmeans",trainable = True, tol = 1e-6, verbose = True, init_par = None):
+        Recommender.__init__(self,name=name, trainable = trainable)
+        self.k = k
+        self.par = init_par
+        self.max_iter = max_iter
+        self.tol = tol
+        self.verbose = verbose
+
+        self.centroids = None #matrix of cluster centroids
+
+
+    #fit the recommender model to the traning data    
+    def fit(self,X):
+        """Fit the model to observations.
+
+        Parameters
+        ----------
+        X: scipy sparse matrix, required
+            the user-item preference matrix (traning data), in a scipy sparse format\
+            (e.g., csc_matrix).
+        """
+        if self.trainable:
+            #Skmeans requires rows of X to have a unit L2 norm. We therefore need to make a copy of X as we should not modify the latter.
+            X1 = X.copy()
+            X1 = X1.multiply(sp.csc_matrix(1./(np.sqrt(X1.multiply(X1).sum(1).A1)+1e-20)).T)
+            res = skmeans(X1,k = self.k, max_iter = self.max_iter,tol = self.tol,verbose = self.verbose,init_par = self.par)
+            self.centroids = res['centroids']
+            self.par = res['partition']
+            self.user_center_sim = X1*self.centroids.T   #user-centroid cosine similarity matrix
+            del(X1)
+        else:
+            print('%s is trained already (trainable = False)' % (self.name))
+
+
+
+
+    #get prefiction for a single user (predictions for one user at a time for efficiency purposes)
+    #predictions are not stored for the same efficiency reasons"""         
+    def predict(self,index_user):
+        """Predic the scores (ratings) of a user for all items.
+
+        Parameters
+        ----------
+        index_user: int, required
+            The index of the user for whom to perform predictions.
+
+        Returns
+        -------
+        Numpy 1d array 
+            Array containing the predicted values for all items
+        """
+        user_pred = self.centroids.multiply(self.user_center_sim[index_user,:].T)
+        #transform user_pred to a flatten array
+        user_pred = user_pred.sum(0).A1/(self.user_center_sim[index_user,:].sum()+1e-20)    #weighted average of cluster centroids
+
+        return user_pred
+
+
+
+

cornac/models/skm/skmeans.py

@@ -0,0 +1,63 @@
+# -*- coding: utf-8 -*-
+"""
+@author: Aghiles Salah
+"""
+
+import numpy as np
+import scipy.sparse as sp
+
+
+def skmeans(X,k=5, max_iter = 100,tol = 1e-6, verbose = True,init_par = None):
+    #The Spherical k-means clustering algorithm
+
+    n = X.shape[0]
+    #normalize rows of X so as they lie on a unit hypersphere
+    X = X.multiply(sp.csc_matrix(1./(np.sqrt(X.multiply(X).sum(1).A1)+1e-20)).T)
+    if init_par is None:
+        par= np.random.randint(k,size=n)
+    else:
+        par = init_par
+
+    #Initialisation of the classification matrix Z
+    Z=sp.csc_matrix((n,k))
+    Z[np.arange(n),par]=1    
+
+
+    change = True
+    l_init= -1e1000
+    l=[]
+    iter_ = 0
+    while change :
+        change=False
+        #Update centroids
+        MU=Z.T*X
+        #project centroids to the unit hypersphere
+        MU = MU.multiply(sp.csc_matrix(1./np.sqrt(MU.multiply(MU).sum(1).A1)).T)
+        #MU = sp.csc_matrix(MU)
+
+        #Object Assignements
+        Z1=X*MU.T
+        par = Z1.argmax(1).A1     #The object partition in k clusters
+        #update the classification matrix
+        Z=sp.csc_matrix((n,k))
+        Z[np.arange(len(par)), par] = 1	
+
+
+        #Skmeans criteria (likelihood) 
+        l_t = Z1.multiply(Z).sum()
+
+        if np.abs(l_t - l_init) > tol:
+            if verbose:
+                print('Iter %i, likelihood: %f' % (iter_, l_t))
+            l_init=l_t
+            change=True
+            l.append(l_t)
+            iter_+=1
+
+    return {"centroids": MU, "partition": par}
+
+
+
+
+
+

docs/build/html/_modules/cornac/models/hpf/recom_hpf.html

@@ -180,7 +180,7 @@ <h1>Source code for cornac.models.hpf.recom_hpf</h1><div class="highlight"><pre>
 <span class="sd">        The name of the recommender model.</span>
 
 <span class="sd">    trainable: boolean, optional, default: True</span>
-<span class="sd">        When False, the model is not trained and Cornac assumes that the model already \</span>
+<span class="sd">        When False, the model is not trained and Cornac assumes that the model is already \</span>
 <span class="sd">        pre-trained (Theta and Beta are not None). </span>
 
 <span class="sd">    init_params: dictionary, optional, default: None</span>