def __init__(self, neg_emb, neg_layer, hidden_size, num_layers, p_layers, *args, **kwargs):
self.neg_emb = neg_emb
self.hidden_size = hidden_size
self.num_layers = num_layers
self.p_layers = p_layers if type(p_layers) == list else eval(p_layers)
self.neg_layer = neg_layer if type(neg_layer) == list else eval(neg_layer)
RecModel.__init__(self, *args, **kwargs)
def __init__(self, user_feature_num, item_feature_num, feature_dims,
f_vector_size, cb_hidden_layers, attention_size, cs_ratio,
*args, **kwargs):
self.user_feature_num = user_feature_num
self.item_feature_num = item_feature_num
self.feature_dims = feature_dims
self.f_vector_size = f_vector_size
self.cb_hidden_layers = cb_hidden_layers if type(
cb_hidden_layers) == list else eval(cb_hidden_layers)
self.attention_size = attention_size
self.cs_ratio = cs_ratio
RecModel.__init__(self, *args, **kwargs)
def __init__(self, hash_u_num, hash_layers, tree_layers, transfer_att_size,
cs_ratio, sample_max_n, sample_r_n, *args, **kwargs):
self.hash_u_num = hash_u_num
self.hash_layers = hash_layers if type(hash_layers) == list else eval(
hash_layers)
self.tree_layers = tree_layers if type(tree_layers) == list else eval(
tree_layers)
self.sample_max_n, self.sample_r_n = sample_max_n, sample_r_n
self.transfer_att_size = transfer_att_size
self.cs_ratio = cs_ratio
self.rec_paras_record, self.hash_paras_record = {}, {}
RecModel.__init__(self, *args, **kwargs)
def __init__(self, layers, p_layers, hash_u_num, hash_layers, tree_layers,
transfer_att_size, cs_ratio, sample_max_n, sample_r_n, *args,
**kwargs):
self.layers = layers if type(layers) == list else eval(layers)
self.p_layers = p_layers if type(p_layers) == list else eval(p_layers)
self.hash_u_num = hash_u_num
self.hash_layers = hash_layers if type(hash_layers) == list else eval(
hash_layers)
self.tree_layers = tree_layers if type(tree_layers) == list else eval(
tree_layers)
self.transfer_att_size = transfer_att_size
self.sample_max_n, self.sample_r_n = sample_max_n, sample_r_n
self.cs_ratio = cs_ratio
RecModel.__init__(self, *args, **kwargs)
def parse_model_args(parser, model_name='PreHash'):
parser.add_argument('--hash_u_num',
type=int,
default=128,
help='Size of user hash.')
parser.add_argument('--sample_max_n',
type=int,
default=128,
help='Sample top-n when learn hash.')
parser.add_argument('--sample_r_n',
type=int,
default=128,
help='Sample random-n when learn hash.')
parser.add_argument('--hash_layers',
type=str,
default='[32]',
help='MLP layer sizes of hash')
parser.add_argument(
'--tree_layers',
type=str,
default='[64]',
help='Number of branches in each level of the hash tree')
parser.add_argument(
'--transfer_att_size',
type=int,
default=16,
help=
'Size of attention layer of transfer layer (combine the hash and cf vector)'
)
parser.add_argument('--cs_ratio',
type=float,
default=0.1,
help='Cold-Sampling ratio of each batch.')
return RecModel.parse_model_args(parser, model_name)
def parse_model_args(parser, model_name='GRU4Rec'):
parser.add_argument('--hidden_size',
type=int,
default=64,
help='Size of hidden vectors in GRU.')
parser.add_argument('--num_layers',
type=int,
default=1,
help='Number of GRU layers.')
parser.add_argument('--p_layers',
type=str,
default='[64]',
help="Size of each layer.")
parser.add_argument(
'--neg_emb',
type=int,
default=1,
help="Whether use negative interaction embeddings.")
parser.add_argument(
'--neg_layer',
type=str,
default='[]',
help=
"Whether use a neg_layer to transfer negative interaction embeddings. "
"[] means using -v. It is ignored when neg_emb=1")
return RecModel.parse_model_args(parser, model_name)
def __init__(self, user_feature_num, item_feature_num, feature_dims,
f_vector_size, cb_hidden_layers, attention_size, cs_ratio,
hash_u_num, hash_layers, tree_layers, transfer_att_size,
sample_max_n, sample_r_n,
*args, **kwargs):
self.user_feature_num = user_feature_num
self.item_feature_num = item_feature_num
self.feature_dims = feature_dims
self.f_vector_size = f_vector_size
self.cb_hidden_layers = cb_hidden_layers if type(cb_hidden_layers) == list else eval(cb_hidden_layers)
self.attention_size = attention_size
self.cs_ratio = cs_ratio
self.hash_u_num = hash_u_num
self.hash_layers = hash_layers if type(hash_layers) == list else eval(hash_layers)
self.tree_layers = tree_layers if type(tree_layers) == list else eval(tree_layers)
self.transfer_att_size = transfer_att_size
self.sample_max_n, self.sample_r_n = sample_max_n, sample_r_n
RecModel.__init__(self, *args, **kwargs)
def parse_model_args(parser, model_name='NeuMF'):
parser.add_argument('--layers',
type=str,
default='[64]',
help="Size of mlp layers.")
parser.add_argument('--p_layers',
type=str,
default='[]',
help="Size of prediction mlp layers.")
return RecModel.parse_model_args(parser, model_name)
def __init__(self, class_num, feature_num, user_num, item_num,
u_vector_size, i_vector_size, user_feature_num,
item_feature_num, feature_dims, f_vector_size,
cb_hidden_layers, attention_size, cs_ratio, random_seed,
model_path):
self.user_feature_num = user_feature_num
self.item_feature_num = item_feature_num
self.feature_dims = feature_dims
self.f_vector_size = f_vector_size
self.cb_hidden_layers = cb_hidden_layers
self.attention_size = attention_size
self.cs_ratio = cs_ratio
RecModel.__init__(self,
class_num=class_num,
feature_num=feature_num,
user_num=user_num,
item_num=item_num,
u_vector_size=u_vector_size,
i_vector_size=i_vector_size,
random_seed=random_seed,
model_path=model_path)
def parse_model_args(parser, model_name='ACCM'):
parser.add_argument('--f_vector_size',
type=int,
default=64,
help='Size of feature vectors.')
parser.add_argument('--cb_hidden_layers',
type=str,
default='[]',
help="Number of CB part's hidden layer.")
parser.add_argument('--attention_size',
type=int,
default=16,
help='Size of attention layer.')
parser.add_argument('--cs_ratio',
type=float,
default=0.1,
help='Cold-Sampling ratio of each batch.')
return RecModel.parse_model_args(parser, model_name)
def __init__(self, layers, p_layers, *args, **kwargs):
self.layers = layers if type(layers) == list else eval(layers)
self.p_layers = p_layers if type(p_layers) == list else eval(p_layers)
RecModel.__init__(self, *args, **kwargs)
