class LR():
def __init__(self,
features_sizes,
loss_type='rmse',
hash_size=None,
metric_type=None):
self.model = Alita_DeepFM(features_sizes=features_sizes,
loss_type=loss_type,
use_LR=True,
use_FM=False,
use_MLP=False,
hash_size=hash_size,
metric_type=metric_type)
def fit(self,
ids_train,
ids_test,
y_train,
y_test,
lr=0.001,
N_EPOCH=50,
batch_size=200,
early_stopping_rounds=20):
return self.model.fit(ids_train,
ids_test,
y_train,
y_test,
lr=lr,
N_EPOCH=N_EPOCH,
batch_size=batch_size,
early_stopping_rounds=early_stopping_rounds)
def predict(self, ids_pred):
return self.model.predict(ids_pred)
def __init__(self,
features_sizes,
dense_features_size=0,
loss_type='rmse',
k=10,
deep_layers=(256, 256),
activation=tf.nn.relu,
FM_ignore_interaction=None,
dropout_keeprate=1.0,
autoint_params=None,
metric_type=None):
self.model = Alita_DeepFM(features_sizes=features_sizes,
dense_features_size=dense_features_size,
loss_type=loss_type,
k=k,
deep_layers=deep_layers,
activation=activation,
use_LR=True,
use_FM=False,
use_MLP=True,
use_AutoInt=True,
FM_ignore_interaction=FM_ignore_interaction,
autoint_params=autoint_params,
dropout_keeprate=dropout_keeprate,
metric_type=metric_type)
class DeepAutoInt():
def __init__(self,
features_sizes,
dense_features_size=0,
loss_type='rmse',
k=10,
deep_layers=(256, 256),
activation=tf.nn.relu,
FM_ignore_interaction=None,
dropout_keeprate=1.0,
autoint_params=None,
metric_type=None):
self.model = Alita_DeepFM(features_sizes=features_sizes,
dense_features_size=dense_features_size,
loss_type=loss_type,
k=k,
deep_layers=deep_layers,
activation=activation,
use_LR=True,
use_FM=False,
use_MLP=True,
use_AutoInt=True,
FM_ignore_interaction=FM_ignore_interaction,
autoint_params=autoint_params,
dropout_keeprate=dropout_keeprate,
metric_type=metric_type)
def fit(self,
ids_train,
ids_test,
y_train,
y_test,
dense_train=None,
dense_test=None,
lr=0.001,
N_EPOCH=50,
batch_size=200,
early_stopping_rounds=20):
return self.model.fit(ids_train,
ids_test,
y_train,
y_test,
dense_train=dense_train,
dense_test=dense_test,
lr=lr,
N_EPOCH=N_EPOCH,
batch_size=batch_size,
early_stopping_rounds=early_stopping_rounds)
def predict(
self,
ids_pred,
dense_pred=None,
):
return self.model.predict(ids_pred, dense_pred=dense_pred)
def __init__(self,
features_sizes,
loss_type='rmse',
hash_size=None,
metric_type=None):
self.model = Alita_DeepFM(features_sizes=features_sizes,
loss_type=loss_type,
use_LR=True,
use_FM=False,
use_MLP=False,
hash_size=hash_size,
metric_type=metric_type)
def __init__(self,
features_sizes,
loss_type='rmse',
k=10,
dropout_keeprate=1.0,
metric_type=None):
self.model = Alita_DeepFM(features_sizes=features_sizes,
loss_type=loss_type,
k=k,
use_LR=True,
use_FM=True,
use_MLP=False,
use_NFM=True,
dropout_keeprate=dropout_keeprate,
metric_type=metric_type)
def __init__(self,
features_sizes,
loss_type='rmse',
k=10,
FM_ignore_interaction=None,
dropout_keeprate=1.0,
hash_size=None,
metric_type=None):
self.model = Alita_DeepFM(features_sizes=features_sizes,
loss_type=loss_type,
k=k,
use_LR=True,
use_FM=True,
use_MLP=False,
FM_ignore_interaction=FM_ignore_interaction,
dropout_keeprate=dropout_keeprate,
hash_size=hash_size,
metric_type=metric_type)
def __init__(self,
features_sizes,
loss_type='rmse',
deep_layers=(256, 256),
activation=tf.nn.relu,
k=10,
dropout_keeprate=1.0,
metric_type=None):
self.model = Alita_DeepFM(features_sizes=features_sizes,
loss_type=loss_type,
k=k,
deep_layers=deep_layers,
activation=activation,
use_LR=True,
use_FM=False,
use_MLP=True,
dropout_keeprate=dropout_keeprate,
metric_type=metric_type)
class FiBiFM():
def __init__(self,
features_sizes,
loss_type='rmse',
k=10,
dropout_keeprate=1.0,
metric_type=None):
self.model = Alita_DeepFM(features_sizes=features_sizes,
loss_type=loss_type,
k=k,
use_LR=True,
use_FM=True,
use_MLP=False,
use_NFM=False,
use_BiFM=True,
use_SE=True,
dropout_keeprate=dropout_keeprate,
metric_type=metric_type)
def fit(self,
ids_train,
ids_test,
y_train,
y_test,
dense_train=None,
dense_test=None,
lr=0.001,
N_EPOCH=50,
batch_size=200,
early_stopping_rounds=20):
return self.model.fit(ids_train,
ids_test,
y_train,
y_test,
dense_train=dense_train,
dense_test=dense_test,
lr=lr,
N_EPOCH=N_EPOCH,
batch_size=batch_size,
early_stopping_rounds=early_stopping_rounds)
def predict(self, ids_pred, dense_pred=None):
return self.model.predict(ids_pred, dense_pred=dense_pred)
def __init__(self,
features_sizes,
loss_type='rmse',
k=10,
FM_ignore_interaction=None,
dropout_keeprate=1.0,
lambda_l2=0.0,
metric_type=None): #todo lambda_L2 only added for AFM
self.model = Alita_DeepFM(features_sizes=features_sizes,
loss_type=loss_type,
k=k,
use_LR=True,
use_FM=True,
use_CFM=True,
use_MLP=False,
FM_ignore_interaction=FM_ignore_interaction,
dropout_keeprate=dropout_keeprate,
lambda_l2=lambda_l2,
metric_type=metric_type)
class AFM():
def __init__(self,
features_sizes,
loss_type='rmse',
k=10,
FM_ignore_interaction=None,
attention_FM=10,
dropout_keeprate=1.0,
lambda_l2=0.0,
metric_type=None): #todo lambda_L2 only added for AFM
self.model = Alita_DeepFM(features_sizes=features_sizes,
loss_type=loss_type,
k=k,
use_LR=True,
use_FM=True,
use_MLP=False,
FM_ignore_interaction=FM_ignore_interaction,
attention_FM=attention_FM,
dropout_keeprate=dropout_keeprate,
lambda_l2=lambda_l2,
metric_type=metric_type)
def fit(self,
ids_train,
ids_test,
y_train,
y_test,
lr=0.001,
N_EPOCH=50,
batch_size=200,
early_stopping_rounds=20):
return self.model.fit(ids_train,
ids_test,
y_train,
y_test,
lr=lr,
N_EPOCH=N_EPOCH,
batch_size=batch_size,
early_stopping_rounds=early_stopping_rounds)
def predict(self, ids_pred):
return self.model.predict(ids_pred)
class AutoInt():
def __init__(self,
features_sizes,
loss_type='rmse',
k=10,
FM_ignore_interaction=None,
dropout_keeprate=1.0,
autoint_params=None,
metric_type=None):
self.model = Alita_DeepFM(features_sizes=features_sizes,
loss_type=loss_type,
k=k,
use_LR=False,
use_FM=False,
use_MLP=False,
use_AutoInt=True,
FM_ignore_interaction=FM_ignore_interaction,
autoint_params=autoint_params,
dropout_keeprate=dropout_keeprate,
metric_type=metric_type)
def fit(self,
ids_train,
ids_test,
y_train,
y_test,
lr=0.001,
N_EPOCH=50,
batch_size=200,
early_stopping_rounds=20):
return self.model.fit(ids_train,
ids_test,
y_train,
y_test,
lr=lr,
N_EPOCH=N_EPOCH,
batch_size=batch_size,
early_stopping_rounds=early_stopping_rounds)
def predict(self, ids_pred):
return self.model.predict(ids_pred)
class WideAndDeep():
def __init__(self,
features_sizes,
loss_type='rmse',
deep_layers=(256, 256),
activation=tf.nn.relu,
k=10,
dropout_keeprate=1.0,
metric_type=None):
self.model = Alita_DeepFM(features_sizes=features_sizes,
loss_type=loss_type,
k=k,
deep_layers=deep_layers,
activation=activation,
use_LR=True,
use_FM=False,
use_MLP=True,
dropout_keeprate=dropout_keeprate,
metric_type=metric_type)
def fit(self,
ids_train,
ids_test,
y_train,
y_test,
lr=0.001,
N_EPOCH=50,
batch_size=200,
early_stopping_rounds=20):
return self.model.fit(ids_train,
ids_test,
y_train,
y_test,
lr=lr,
N_EPOCH=N_EPOCH,
batch_size=batch_size,
early_stopping_rounds=early_stopping_rounds)
def predict(self, ids_pred):
return self.model.predict(ids_pred)
