# -*- coding: utf-8 -*-
"""
@author: Guo Jingyao
"""
from .vbpr import *
from ..recommender import Recommender
[docs]class VBPR(Recommender):
"""Visual Bayesian Personalized Ranking.
Parameters
----------
k: int, optional, default: 5
The dimension of the latent factors.
d: int, optional, default: 5
The dimension of the latent factors.
max_iter: int, optional, default: 100
Maximum number of iterations or the number of epochs for SGD.
aux_info:ndarray, shape (n_items, feature dimension), optional, default:None
Image features of items
learning_rate: float, optional, default: 0.001
The learning rate for SGD.
lamda: float, optional, default: 0.01
The regularization parameter.
batch_size: int, optional, default: 100
The batch size for SGD.
name: string, optional, default: 'BRP'
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 \
pre-trained (U and V are not None).
init_params: dictionary, optional, default: None
List of initial parameters, e.g., init_params = {'U':U, 'V':V} \
please see below the definition of U and V.
U: ndarray, shape (n_users,k)
The user latent factors, optional initialization via init_params.
V: ndarray, shape (n_items,k)
The item latent factors, optional initialization via init_params.
E: ndarray, shape (d, feature dimension)
The matrix embedding deep CNN feature, optional initialization via init_params.
Ue: ndarray, shape (n_users, d)
The visual factors of users, optional initialization via init_params.
References
----------
* HE, Ruining et MCAULEY, Julian. VBPR: Visual Bayesian Personalized Ranking from Implicit Feedback. In : AAAI. 2016. p. 144-150.
"""
def __init__(self, k=10, d=10, max_iter=100, aux_info=None, learning_rate=0.001, lamda=0.01, batch_size=100,
name="vbpr", trainable=True,
init_params=None):
Recommender.__init__(self, name=name, trainable=trainable)
self.k = k
self.d = d
self.init_params = init_params
self.aux_info = aux_info
self.max_iter = max_iter
self.name = name
self.learning_rate = learning_rate
self.lamda = lamda
self.batch_size = batch_size
self.U = init_params['U'] # matrix of user factors
self.V = init_params['V'] # matrix of item factors
self.E = init_params['E'] # matrix embedding deep CNN feature
self.Ue = init_params['Ue'] # visual factors of users
# fit the recommender model to the traning data
[docs] 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:
# change the data to original user Id item Id and rating format
cooX = X.tocoo()
data = np.ndarray(shape=(len(cooX.data), 3), dtype=float)
data[:, 0] = cooX.row
data[:, 1] = cooX.col
data[:, 2] = cooX.data
print('Learning...')
res = vbpr(X, data, k=self.k, d=self.d, aux_info=self.aux_info, n_epochs=self.max_iter, lamda=self.lamda,
learning_rate=self.learning_rate,
batch_size=self.batch_size, init_params=self.init_params)
self.U = res['U']
self.V = res['V']
self.Ue = res['Ue']
self.E = res['E']
print('Learning completed')
else:
print('%s is trained already (trainable = False)' % (self.name))
[docs] def score(self, user_index, item_indexes = None):
"""Predict the scores/ratings of a user for a list of items.
Parameters
----------
user_index: int, required
The index of the user for whom to perform score predictions.
item_indexes: 1d array, optional, default: None
A list of item indexes for which to predict the rating score.\
When "None", score prediction is performed for all test items of the given user.
Returns
-------
Numpy 1d array
Array containing the predicted values for the items of interest
"""
if item_indexes is None:
user_pred = self.U[user_index, :].dot(self.V.T) + self.Ue[user_index, :].dot(self.E).dot(self.aux_info.T)
# user_pred = self.U[index_user, :].dot(self.V.T) + self.Ue[index_user, :]*self.E.dot(self.aux_info.T)
else:
user_pred = self.U[user_index, :].dot(self.V[item_indexes,:].T) + self.Ue[user_index, :].dot(self.E).dot(self.aux_info[item_indexes,:].T)
# transform user_pred to a flatten array, but keep thinking about another possible format
user_pred = np.array(user_pred, dtype='float64').flatten()
return user_pred
[docs] def rank(self, user_index, known_items = None):
"""Rank all test items for a given user.
Parameters
----------
user_index: int, required
The index of the user for whom to perform item raking.
known_items: 1d array, optional, default: None
A list of item indices already known by the user
Returns
-------
Numpy 1d array
Array of item indices sorted (in decreasing order) relative to some user preference scores.
"""
u_pref_score = np.array(self.score(user_index))
if known_items is not None:
u_pref_score[known_items] = None
rank_item_list = (-u_pref_score).argsort() # ordering the items (in decreasing order) according to the preference score
return rank_item_list