"""
Module description:
"""
__version__ = '0.3.1'
__author__ = 'Felice Antonio Merra, Vito Walter Anelli, Claudio Pomo'
__email__ = 'felice.merra@poliba.it, vitowalter.anelli@poliba.it, claudio.pomo@poliba.it'
from ast import literal_eval as make_tuple
import numpy as np
from tqdm import tqdm
from elliot.dataset.samplers import pointwise_pos_neg_sampler as pws
from elliot.recommender.base_recommender_model import BaseRecommenderModel
from elliot.recommender.base_recommender_model import init_charger
from elliot.recommender.neural.ConvMF.convolutional_matrix_factorization_model import \
ConvMatrixFactorizationModel
from elliot.recommender.recommender_utils_mixin import RecMixin
[docs]class ConvMF(RecMixin, BaseRecommenderModel):
r"""
Convolutional Matrix Factorization for Document Context-Aware Recommendation
For further details, please refer to the `paper <https://dl.acm.org/doi/10.1145/2959100.2959165>`_
Args:
embedding_size: Embedding dimension
lr: Learning rate
l_w: Regularization coefficient
l_b: Regularization coefficient of bias
cnn_channels: List of channels
cnn_kernels: List of kernels
cnn_strides: List of strides
dropout_prob: Dropout probability applied on the convolutional layers
To include the recommendation model, add it to the config file adopting the following pattern:
.. code:: yaml
models:
ConvMF:
meta:
save_recs: True
epochs: 10
batch_size: 512
embedding_size: 100
lr: 0.001
l_w: 0.005
l_b: 0.0005
cnn_channels: (1, 32, 32)
cnn_kernels: (2,2)
cnn_strides: (2,2)
dropout_prob: 0
"""
@init_charger
def __init__(self, data, config, params, *args, **kwargs):
"""
Args:
data:
config:
params:
*args:
**kwargs:
"""
self._sampler = pws.Sampler(self._data.i_train_dict)
self._params_list = [
("_lr", "lr", "lr", 0.001, None, None),
("_embedding_size", "embedding_size", "embedding_size", 100, None, None),
("_cnn_channels", "cnn_channels", "cnn_channels", "(1, 32, 32)", lambda x: list(make_tuple(str(x))),
lambda x: self._batch_remove(str(x), " []").replace(",", "-")),
("_cnn_kernels", "cnn_kernels", "cnn_kernels", "(2,2)", lambda x: list(make_tuple(str(x))),
lambda x: self._batch_remove(str(x), " []").replace(",", "-")),
("_cnn_strides", "cnn_strides", "cnn_strides", "(2,2)", lambda x: list(make_tuple(str(x))),
lambda x: self._batch_remove(str(x), " []").replace(",", "-")),
("_dropout_prob", "dropout_prob", "dropout_prob", 0, None, None),
("_l_w", "l_w", "l_w", 0.005, None, None),
("_l_b", "l_b", "l_b", 0.0005, None, None),
]
self.autoset_params()
if self._batch_size < 1:
self._batch_size = self._data.transactions
self._ratings = self._data.train_dict
self._sp_i_train = self._data.sp_i_train
self._i_items_set = list(range(self._num_items))
self._model = ConvMatrixFactorizationModel(self._num_users, self._num_items, self._embedding_size,
self._lr, self._cnn_channels, self._cnn_kernels,
self._cnn_strides, self._dropout_prob, self._l_w, self._l_b,
self._seed
)
@property
def name(self):
return "ConvMF" \
+ f"_{self.get_base_params_shortcut()}" \
+ f"_{self.get_params_shortcut()}"
[docs] def train(self):
if self._restore:
return self.restore_weights()
for it in self.iterate(self._epochs):
loss = 0
steps = 0
with tqdm(total=int(self._data.transactions // self._batch_size), disable=not self._verbose) as t:
for batch in self._sampler.step(self._data.transactions, self._batch_size):
steps += 1
loss += self._model.train_step(batch)
t.set_postfix({'loss': f'{loss.numpy() / steps:.5f}'})
t.update()
self.evaluate(it, loss.numpy()/(it + 1))
[docs] def get_recommendations(self, k: int = 100):
predictions_top_k_test = {}
predictions_top_k_val = {}
for index, offset in enumerate(range(0, self._num_users, self._batch_size)):
offset_stop = min(offset + self._batch_size, self._num_users)
predictions = self._model.get_recs(
(
np.repeat(np.array(list(range(offset, offset_stop)))[:, None], repeats=self._num_items, axis=1),
np.array([self._i_items_set for _ in range(offset, offset_stop)])
)
)
recs_val, recs_test = self.process_protocol(k, predictions, offset, offset_stop)
predictions_top_k_val.update(recs_val)
predictions_top_k_test.update(recs_test)
return predictions_top_k_val, predictions_top_k_test