# Copyright 2018 The Cornac Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ============================================================================
import numpy as np
from ..recommender import Recommender
from ...exception import ScoreException
[docs]
class CVAECF(Recommender):
"""Conditional Variational Autoencoder for Collaborative Filtering.
Parameters
----------
z_dim: int, optional, default: 20
The dimension of the stochastic user factors ``z'' representing the preference information.
h_dim: int, optional, default: 20
The dimension of the stochastic user factors ``h'' representing the auxiliary data.
autoencoder_structure: list, default: [20]
The number of neurons of encoder/decoder hidden layer for CVAE.
For example, when autoencoder_structure = [20],
the CVAE encoder structures will be [y_dim, 20, z_dim] and [x_dim, 20, h_dim],
the decoder structure will be [z_dim + h_dim, 20, y_dim], where y and x are respectively the preference and \
auxiliary data.
act_fn: str, default: 'tanh'
Name of the activation function used between hidden layers of the auto-encoder.
Supported functions: ['sigmoid', 'tanh', 'elu', 'relu', 'relu6']
likelihood: str, default: 'mult'
Name of the likelihood function used for modeling user preferences.
Supported choices:
mult: Multinomial likelihood
bern: Bernoulli likelihood
gaus: Gaussian likelihood
pois: Poisson likelihood
n_epochs: int, optional, default: 100
The number of epochs for SGD.
batch_size: int, optional, default: 128
The batch size.
learning_rate: float, optional, default: 0.001
The learning rate for Adam.
beta: float, optional, default: 1.0
The weight of the KL term KL(q(z|y)||p(z)) as in beta-VAE.
alpha_1: float, optional, default: 1.0
The weight of the KL term KL(q(h|x)||p(h|x)).
alpha_2: float, optional, default: 1.0
The weight of the KL term KL(q(h|x)||q(h|y)).
name: string, optional, default: 'CVAECF'
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.
verbose: boolean, optional, default: False
When True, some running logs are displayed.
seed: int, optional, default: None
Random seed for parameters initialization.
use_gpu: boolean, optional, default: False
If True and your system supports CUDA then training is performed on GPUs.
user auxiliary data : See "cornac/examples/cvaecf_filmtrust.py" for an example of how to use \
cornac's graph modality to load and provide the ``user network'' for CVAECF.
References
----------
* Lee, Wonsung, Kyungwoo Song, and Il-Chul Moon. "Augmented variational autoencoders for collaborative filtering \
with auxiliary information." Proceedings of ACM CIKM. 2017.
"""
def __init__(
self,
name="CVAECF",
z_dim=20,
h_dim=20,
autoencoder_structure=[20],
act_fn="tanh",
likelihood="mult",
n_epochs=100,
batch_size=128,
learning_rate=0.001,
beta=1.0,
alpha_1=1.0,
alpha_2=1.0,
trainable=True,
verbose=False,
seed=None,
use_gpu=False,
):
Recommender.__init__(self, name=name, trainable=trainable, verbose=verbose)
self.z_dim = z_dim
self.h_dim = h_dim
self.autoencoder_structure = autoencoder_structure
self.act_fn = act_fn
self.likelihood = likelihood
self.batch_size = batch_size
self.n_epochs = n_epochs
self.learning_rate = learning_rate
self.beta = beta
self.alpha_1 = alpha_1
self.alpha_2 = alpha_2
self.seed = seed
self.use_gpu = use_gpu
[docs]
def fit(self, train_set, val_set=None):
"""Fit the model to observations.
Parameters
----------
train_set: :obj:`cornac.data.Dataset`, required
User-Item preference data as well as additional modalities.
val_set: :obj:`cornac.data.Dataset`, optional, default: None
User-Item preference data for model selection purposes (e.g., early stopping).
Returns
-------
self : object
"""
Recommender.fit(self, train_set, val_set)
import torch
from .cvaecf import CVAE, learn
self.device = (
torch.device("cuda:0")
if (self.use_gpu and torch.cuda.is_available())
else torch.device("cpu")
)
self.r_mat = train_set.matrix
self.u_adj_mat = train_set.user_graph.matrix
if self.trainable:
if self.seed is not None:
torch.manual_seed(self.seed)
torch.cuda.manual_seed(self.seed)
if not hasattr(self, "cvae"):
n_items = self.r_mat.shape[1]
n_users = self.r_mat.shape[0]
self.cvae = CVAE(
self.z_dim,
self.h_dim,
[n_items] + self.autoencoder_structure,
[n_users] + self.autoencoder_structure,
self.act_fn,
self.likelihood,
).to(self.device)
learn(
self.cvae,
train_set,
n_epochs=self.n_epochs,
batch_size=self.batch_size,
learn_rate=self.learning_rate,
beta=self.beta,
alpha_1=self.alpha_1,
alpha_2=self.alpha_2,
verbose=self.verbose,
device=self.device,
)
elif self.verbose:
print("%s is trained already (trainable = False)" % (self.name))
return self
[docs]
def score(self, user_idx, item_idx=None):
"""Predict the scores/ratings of a user for an item.
Parameters
----------
user_idx: int, required
The index of the user for whom to perform score prediction.
item_idx: int, optional, default: None
The index of the item for which to perform score prediction.
If None, scores for all known items will be returned.
Returns
-------
res : A scalar or a Numpy array
Relative scores that the user gives to the item or to all known items
"""
if self.is_unknown_user(user_idx):
raise ScoreException("Can't make score prediction for user %d" % user_idx)
if item_idx is not None and self.is_unknown_item(item_idx):
raise ScoreException("Can't make score prediction for item %d" % item_idx)
import torch
if item_idx is None:
y_u = self.r_mat[user_idx].copy()
y_u.data = np.ones(len(y_u.data))
y_u = torch.tensor(y_u.A, dtype=torch.float32, device=self.device)
z_u, _ = self.cvae.encode_qz(y_u)
x_u = self.u_adj_mat[user_idx].copy()
x_u.data = np.ones(len(x_u.data))
x_u = torch.tensor(x_u.A, dtype=torch.float32, device=self.device)
h_u, _ = self.cvae.encode_qhx(x_u)
known_item_scores = self.cvae.decode(z_u, h_u).data.cpu().numpy().flatten()
return known_item_scores
else:
y_u = self.r_mat[user_idx].copy()
y_u.data = np.ones(len(y_u.data))
y_u = torch.tensor(y_u.A, dtype=torch.float32, device=self.device)
z_u, _ = self.cvae.encode_qz(y_u)
x_u = self.u_adj_mat[user_idx].copy()
x_u.data = np.ones(len(x_u.data))
x_u = torch.tensor(x_u.A, dtype=torch.float32, device=self.device)
h_u, _ = self.cvae.encode_qhx(x_u)
user_pred = (
self.cvae.decode(z_u, h_u).data.cpu().numpy().flatten()[item_idx]
)
return user_pred
