def _init_model(self, params):
self.model_param = params
self.alpha = params.alpha
self.init_param_obj = params.init_param
self.fit_intercept = self.init_param_obj.fit_intercept
self.batch_size = params.batch_size
self.max_iter = params.max_iter
self.optimizer = optimizer_factory(params)
self.converge_func = converge_func_factory(params.early_stop, params.tol)
self.encrypted_calculator = None
self.validation_freqs = params.validation_freqs
def _init_model(self, params):
super(HomoFMBase, self)._init_model(params)
# if params.encrypt_param.method == consts.PAILLIER:
# self.cipher_operator = PaillierEncrypt()
# else:
# self.cipher_operator = FakeEncrypt()
self.transfer_variable = HomoFMTransferVariable()
self.aggregator.register_aggregator(self.transfer_variable)
self.optimizer = optimizer_factory(params)
self.aggregate_iters = params.aggregate_iters
def _init_model(self, params):
super(HomoLRBase, self)._init_model(params)
self.re_encrypt_batches = params.re_encrypt_batches
if params.encrypt_param.method == consts.PAILLIER:
self.cipher_operator = PaillierEncrypt()
else:
self.cipher_operator = FakeEncrypt()
self.transfer_variable = HomoZclAlexTransferVariable()
self.aggregator.register_aggregator(self.transfer_variable)
self.optimizer = optimizer_factory(params)
self.aggregate_iters = params.aggregate_iters
self.max_iter = 50
def _init_model(self, params):
self.model_param = params
self.alpha = params.alpha
self.init_param_obj = params.init_param
# self.fit_intercept = self.init_param_obj.fit_intercept
self.batch_size = params.batch_size
self.max_iter = params.max_iter
self.optimizer = optimizer_factory(params)
self.converge_func = converge_func_factory(params.early_stop, params.tol)
self.encrypted_calculator = None
self.validation_freqs = params.validation_freqs
self.validation_strategy = None
self.early_stopping_rounds = params.early_stopping_rounds
self.metrics = params.metrics
self.use_first_metric_only = params.use_first_metric_only
def _init_model(self, params):
self.model_param = params
self.alpha = params.alpha
self.init_param_obj = params.init_param
# self.fit_intercept = self.init_param_obj.fit_intercept
self.batch_size = params.batch_size
if hasattr(params, "shuffle"):
self.shuffle = params.shuffle
if hasattr(params, "masked_rate"):
self.masked_rate = params.masked_rate
if hasattr(params, "batch_strategy"):
self.batch_strategy = params.batch_strategy
self.max_iter = params.max_iter
self.optimizer = optimizer_factory(params)
self.converge_func = converge_func_factory(params.early_stop, params.tol)
self.validation_freqs = params.callback_param.validation_freqs
self.validation_strategy = None
self.early_stopping_rounds = params.callback_param.early_stopping_rounds
self.metrics = params.callback_param.metrics
self.use_first_metric_only = params.callback_param.use_first_metric_only