def main(config="../../config.yaml", namespace=""):
# obtain config
if isinstance(config, str):
config = load_job_config(config)
parties = config.parties
guest = parties.guest[0]
host = parties.host[0]
arbiter = parties.arbiter[0]
guest_train_data = {"name": "vehicle_scale_hetero_guest", "namespace": f"experiment{namespace}"}
host_train_data = {"name": "vehicle_scale_hetero_host", "namespace": f"experiment{namespace}"}
pipeline = PipeLine().set_initiator(role='guest', party_id=guest).set_roles(guest=guest, host=host, arbiter=arbiter)
reader_0 = Reader(name="reader_0")
reader_0.get_party_instance(role='guest', party_id=guest).component_param(table=guest_train_data)
reader_0.get_party_instance(role='host', party_id=host).component_param(table=host_train_data)
data_transform_0 = DataTransform(name="data_transform_0")
data_transform_0.get_party_instance(role='guest', party_id=guest).component_param(with_label=True, output_format="dense",
label_type="int", label_name="y")
data_transform_0.get_party_instance(role='host', party_id=host).component_param(with_label=False)
intersection_0 = Intersection(name="intersection_0", intersect_method="rsa", sync_intersect_ids=True,
only_output_key=False)
hetero_lr_0 = HeteroLR(name="hetero_lr_0", penalty="L2", optimizer="nesterov_momentum_sgd",
tol=0.0001, alpha=0.0001, max_iter=30, batch_size=-1,
early_stop="diff", learning_rate=0.15, init_param={"init_method": "zeros"})
local_baseline_0 = LocalBaseline(name="local_baseline_0", model_name="LogisticRegression",
model_opts={"penalty": "l2", "tol": 0.0001, "C": 1.0, "fit_intercept": True,
"solver": "lbfgs", "max_iter": 5, "multi_class": "ovr"})
local_baseline_0.get_party_instance(role='guest', party_id=guest).component_param(need_run=True)
local_baseline_0.get_party_instance(role='host', party_id=host).component_param(need_run=False)
evaluation_0 = Evaluation(name="evaluation_0", eval_type="multi", pos_label=1)
evaluation_0.get_party_instance(role='guest', party_id=guest).component_param(need_run=True)
evaluation_0.get_party_instance(role='host', party_id=host).component_param(need_run=False)
pipeline.add_component(reader_0)
pipeline.add_component(data_transform_0, data=Data(data=reader_0.output.data))
pipeline.add_component(intersection_0, data=Data(data=data_transform_0.output.data))
pipeline.add_component(hetero_lr_0, data=Data(train_data=intersection_0.output.data))
pipeline.add_component(local_baseline_0, data=Data(train_data=intersection_0.output.data))
pipeline.add_component(evaluation_0, data=Data(data=[hetero_lr_0.output.data, local_baseline_0.output.data]))
pipeline.compile()
pipeline.fit()
# predict
pipeline.deploy_component([data_transform_0, intersection_0, hetero_lr_0, local_baseline_0])
predict_pipeline = PipeLine()
predict_pipeline.add_component(reader_0)
predict_pipeline.add_component(pipeline,
data=Data(predict_input={pipeline.data_transform_0.input.data: reader_0.output.data}))
predict_pipeline.add_component(evaluation_0, data=Data(data=[hetero_lr_0.output.data, local_baseline_0.output.data]))
predict_pipeline.predict()
def main(config="../../config.yaml", namespace=""):
# obtain config
if isinstance(config, str):
config = load_job_config(config)
parties = config.parties
guest = parties.guest[0]
host = parties.host[0]
arbiter = parties.arbiter[0]
guest_train_data = {"name": "breast_hetero_guest", "namespace": f"experiment{namespace}"}
host_train_data = {"name": "breast_hetero_host", "namespace": f"experiment{namespace}"}
# initialize pipeline
pipeline = PipeLine()
# set job initiator
pipeline.set_initiator(role="guest", party_id=guest).set_roles(guest=guest, host=host, arbiter=arbiter)
# define Reader components to read in data
reader_0 = Reader(name="reader_0")
# configure Reader for guest
reader_0.get_party_instance(role="guest", party_id=guest).component_param(table=guest_train_data)
reader_0.get_party_instance(role="host", party_id=host).component_param(table=host_train_data)
data_transform_0 = DataTransform(name="data_transform_0") # start component numbering at 0
data_transform_0_guest_party_instance = data_transform_0.get_party_instance(role="guest", party_id=guest)
data_transform_0_guest_party_instance.component_param(with_label=True, output_format="dense")
data_transform_0.get_party_instance(role="host", party_id=host).component_param(with_label=False,
output_format="dense")
intersection_0 = Intersection(name="intersection_0")
label_transform_0 = LabelTransform(name="label_transform_0")
label_transform_0.get_party_instance(role="host", party_id=host).component_param(need_run=False)
hetero_lr_0 = HeteroLR(name="hetero_lr_0", penalty="L2", optimizer="sgd", tol=0.001,
alpha=0.01, max_iter=20, early_stop="weight_diff", batch_size=-1,
learning_rate=0.15, decay=0.0, decay_sqrt=False,
init_param={"init_method": "zeros"},
encrypted_mode_calculator_param={"mode": "fast"},
floating_point_precision=23)
label_transform_1 = LabelTransform(name="label_transform_1")
evaluation_0 = Evaluation(name="evaluation_0", eval_type="binary", pos_label=1)
# add components to pipeline, in order of task execution
pipeline.add_component(reader_0)
pipeline.add_component(data_transform_0, data=Data(data=reader_0.output.data))
pipeline.add_component(intersection_0, data=Data(data=data_transform_0.output.data))
pipeline.add_component(label_transform_0, data=Data(data=intersection_0.output.data))
pipeline.add_component(hetero_lr_0, data=Data(train_data=label_transform_0.output.data))
pipeline.add_component(label_transform_1, data=Data(data=hetero_lr_0.output.data), model=Model(label_transform_0.output.model))
pipeline.add_component(evaluation_0, data=Data(data=label_transform_1.output.data))
# compile pipeline once finished adding modules, this step will form conf and dsl files for running job
pipeline.compile()
# fit model
pipeline.fit()
def main(config="../../config.yaml", namespace=""):
# obtain config
if isinstance(config, str):
config = load_job_config(config)
parties = config.parties
guest = parties.guest[0]
host = parties.host[0]
arbiter = parties.arbiter[0]
backend = config.backend
work_mode = config.work_mode
guest_train_data = {"name": "breast_hetero_guest", "namespace": f"experiment{namespace}"}
host_train_data = {"name": "breast_hetero_host", "namespace": f"experiment{namespace}"}
pipeline = PipeLine().set_initiator(role='guest', party_id=guest).set_roles(guest=guest, host=host, arbiter=arbiter)
reader_0 = Reader(name="reader_0")
reader_0.get_party_instance(role='guest', party_id=guest).component_param(table=guest_train_data)
reader_0.get_party_instance(role='host', party_id=host).component_param(table=host_train_data)
dataio_0 = DataIO(name="dataio_0")
dataio_0.get_party_instance(role='guest', party_id=guest).component_param(with_label=True, missing_fill=True,
outlier_replace=True)
dataio_0.get_party_instance(role='host', party_id=host).component_param(with_label=False, missing_fill=True,
outlier_replace=True)
intersection_0 = Intersection(name="intersection_0")
federated_sample_0 = FederatedSample(name="federated_sample_0", mode="stratified", method="upsample",
fractions=[[0, 1.5], [1, 2.0]])
feature_scale_0 = FeatureScale(name="feature_scale_0", method="min_max_scale", mode="normal")
feature_scale_0.get_party_instance(role='guest', party_id=guest).component_param(feat_upper=[1, 2, 1, 1, 0.5, 1, 2, 2, 1, 1])
hetero_feature_binning_0 = HeteroFeatureBinning(name="hetero_feature_binning_0")
hetero_feature_selection_0 = HeteroFeatureSelection(name="hetero_feature_selection_0")
one_hot_0 = OneHotEncoder(name="one_hot_0")
hetero_lr_0 = HeteroLR(name="hetero_lr_0", penalty="L2", optimizer="rmsprop", tol=1e-5,
init_param={"init_method": "random_uniform"},
alpha=0.01, max_iter=10, early_stop="diff", batch_size=320, learning_rate=0.15)
evaluation_0 = Evaluation(name="evaluation_0")
pipeline.add_component(reader_0)
pipeline.add_component(dataio_0, data=Data(data=reader_0.output.data))
pipeline.add_component(intersection_0, data=Data(data=dataio_0.output.data))
pipeline.add_component(federated_sample_0, data=Data(data=intersection_0.output.data))
pipeline.add_component(feature_scale_0, data=Data(data=federated_sample_0.output.data))
pipeline.add_component(hetero_feature_binning_0, data=Data(data=feature_scale_0.output.data))
pipeline.add_component(hetero_feature_selection_0, data=Data(data=hetero_feature_binning_0.output.data))
pipeline.add_component(one_hot_0, data=Data(data=hetero_feature_selection_0.output.data))
pipeline.add_component(hetero_lr_0, data=Data(train_data=one_hot_0.output.data))
pipeline.add_component(evaluation_0, data=Data(data=hetero_lr_0.output.data))
pipeline.compile()
job_parameters = JobParameters(backend=backend, work_mode=work_mode)
pipeline.fit(job_parameters)
print(pipeline.get_component("evaluation_0").get_summary())
def main(config="../../config.yaml", namespace=""):
# obtain config
if isinstance(config, str):
config = load_job_config(config)
parties = config.parties
guest = parties.guest[0]
hosts = parties.host
arbiter = parties.arbiter[0]
backend = config.backend
work_mode = config.work_mode
guest_train_data = {"name": "breast_hetero_guest", "namespace": f"experiment{namespace}"}
host_train_data = {"name": "breast_hetero_host", "namespace": f"experiment{namespace}"}
pipeline = PipeLine().set_initiator(role='guest', party_id=guest).set_roles(guest=guest, host=hosts, arbiter=arbiter)
reader_0 = Reader(name="reader_0")
reader_0.get_party_instance(role='guest', party_id=guest).component_param(table=guest_train_data)
reader_0.get_party_instance(role='host', party_id=hosts).component_param(table=host_train_data)
dataio_0 = DataIO(name="dataio_0")
dataio_0.get_party_instance(role='guest', party_id=guest).component_param(with_label=True, label_name="y",
label_type="int", output_format="dense")
dataio_0.get_party_instance(role='host', party_id=hosts).component_param(with_label=False)
intersection_0 = Intersection(name="intersection_0")
sample_weight_0 = SampleWeight(name="sample_weight_0")
sample_weight_0.get_party_instance(role='guest', party_id=guest).component_param(need_run=True,
class_weight="balanced")
sample_weight_0.get_party_instance(role='host', party_id=hosts).component_param(need_run=False)
hetero_lr_0 = HeteroLR(name="hetero_lr_0", optimizer="nesterov_momentum_sgd", tol=0.001,
alpha=0.01, max_iter=20, early_stop="weight_diff", batch_size=-1,
learning_rate=0.15,
init_param={"init_method": "zeros"})
evaluation_0 = Evaluation(name="evaluation_0", eval_type="binary", pos_label=1)
# evaluation_0.get_party_instance(role='host', party_id=host).component_param(need_run=False)
pipeline.add_component(reader_0)
pipeline.add_component(dataio_0, data=Data(data=reader_0.output.data))
pipeline.add_component(intersection_0, data=Data(data=dataio_0.output.data))
pipeline.add_component(sample_weight_0, data=Data(data=intersection_0.output.data))
pipeline.add_component(hetero_lr_0, data=Data(train_data=sample_weight_0.output.data))
pipeline.add_component(evaluation_0, data=Data(data=hetero_lr_0.output.data))
pipeline.compile()
job_parameters = JobParameters(backend=backend, work_mode=work_mode)
pipeline.fit(job_parameters)
def main(config="../../config.yaml", namespace=""):
# obtain config
if isinstance(config, str):
config = load_job_config(config)
parties = config.parties
guest = parties.guest[0]
host = parties.host[0]
arbiter = parties.arbiter[0]
backend = config.backend
work_mode = config.work_mode
guest_train_data = {"name": "breast_hetero_mini_guest", "namespace": f"experiment{namespace}"}
host_train_data = {"name": "breast_hetero_mini_host", "namespace": f"experiment{namespace}"}
pipeline = PipeLine().set_initiator(role='guest', party_id=guest).set_roles(guest=guest, host=host, arbiter=arbiter)
reader_0 = Reader(name="reader_0")
reader_0.get_party_instance(role='guest', party_id=guest).algorithm_param(table=guest_train_data)
reader_0.get_party_instance(role='host', party_id=host).algorithm_param(table=host_train_data)
dataio_0 = DataIO(name="dataio_0")
dataio_0.get_party_instance(role='guest', party_id=guest).algorithm_param(with_label=True, output_format="dense")
dataio_0.get_party_instance(role='host', party_id=host).algorithm_param(with_label=False)
intersection_0 = Intersection(name="intersection_0")
hetero_lr_0 = HeteroLR(name="hetero_lr_0", early_stop="diff", max_iter=5,
penalty="None", optimizer="sgd", tol=0.001,
batch_size=-1, learning_rate=0.15, decay=0.0,
decay_sqrt=False,
init_param={"init_method": "zeros"},
encrypted_mode_calculator_param={"mode": "fast"},
stepwise_param={"score_name": "AIC", "direction": "backward",
"need_stepwise": True, "max_step": 2, "nvmin": 2
})
pipeline.add_component(reader_0)
pipeline.add_component(dataio_0, data=Data(data=reader_0.output.data))
pipeline.add_component(intersection_0, data=Data(data=dataio_0.output.data))
pipeline.add_component(hetero_lr_0, data=Data(train_data=intersection_0.output.data))
pipeline.compile()
pipeline.fit(backend=backend, work_mode=work_mode)
def main(config="../../config.yaml", namespace=""):
# obtain config
if isinstance(config, str):
config = load_job_config(config)
parties = config.parties
guest = parties.guest[0]
host = parties.host[0]
arbiter = parties.arbiter[0]
guest_train_data = {"name": "breast_hetero_guest", "namespace": f"experiment{namespace}"}
host_train_data = {"name": "breast_hetero_host", "namespace": f"experiment{namespace}"}
# initialize pipeline
pipeline = PipeLine()
# set job initiator
pipeline.set_initiator(role='guest', party_id=guest)
# set participants information
pipeline.set_roles(guest=guest, host=host, arbiter=arbiter)
# define Reader components to read in data
reader_0 = Reader(name="reader_0")
# configure Reader for guest
reader_0.get_party_instance(role='guest', party_id=guest).component_param(table=guest_train_data)
# configure Reader for host
reader_0.get_party_instance(role='host', party_id=host).component_param(table=host_train_data)
# define DataTransform components
data_transform_0 = DataTransform(name="data_transform_0", with_label=True, output_format="dense") # start component numbering at 0
data_transform_0.get_party_instance(role="host", party_id=host).component_param(with_label=False)
intersect_0 = Intersection(name='intersect_0')
scale_0 = FeatureScale(name='scale_0', need_run=False)
sample_weight_0 = SampleWeight(name="sample_weight_0", class_weight={"0": 1, "1": 2})
sample_weight_0.get_party_instance(role="host", party_id=host).component_param(need_run=False)
param = {
"penalty": None,
"optimizer": "sgd",
"tol": 1e-05,
"alpha": 0.01,
"max_iter": 3,
"early_stop": "diff",
"batch_size": 320,
"learning_rate": 0.15,
"decay": 0,
"decay_sqrt": True,
"init_param": {
"init_method": "ones"
},
"cv_param": {
"n_splits": 5,
"shuffle": True,
"random_seed": 33,
"need_cv": False
}
}
hetero_lr_0 = HeteroLR(name='hetero_lr_0', **param)
evaluation_0 = Evaluation(name='evaluation_0')
# add components to pipeline, in order of task execution
pipeline.add_component(reader_0)
pipeline.add_component(data_transform_0, data=Data(data=reader_0.output.data))
pipeline.add_component(intersect_0, data=Data(data=data_transform_0.output.data))
# set data input sources of intersection components
pipeline.add_component(scale_0, data=Data(data=intersect_0.output.data))
pipeline.add_component(sample_weight_0, data=Data(data=scale_0.output.data))
pipeline.add_component(hetero_lr_0, data=Data(train_data=sample_weight_0.output.data))
pipeline.add_component(evaluation_0, data=Data(data=hetero_lr_0.output.data))
# compile pipeline once finished adding modules, this step will form conf and dsl files for running job
pipeline.compile()
# fit model
pipeline.fit()
# query component summary
print(json.dumps(pipeline.get_component("evaluation_0").get_summary(), indent=4, ensure_ascii=False))
def make_normal_dsl(config, namespace, lr_param, is_multi_host=False, has_validate=False,
is_cv=False, is_ovr=False, is_dense=True, need_evaluation=True):
parties = config.parties
guest = parties.guest[0]
if is_multi_host:
hosts = parties.host
else:
hosts = parties.host[0]
arbiter = parties.arbiter[0]
if is_ovr:
guest_train_data = {"name": "vehicle_scale_hetero_guest", "namespace": f"experiment{namespace}"}
host_train_data = {"name": "vehicle_scale_hetero_host", "namespace": f"experiment{namespace}"}
guest_eval_data = {"name": "vehicle_scale_hetero_guest", "namespace": f"experiment{namespace}"}
host_eval_data = {"name": "vehicle_scale_hetero_host", "namespace": f"experiment{namespace}"}
else:
guest_train_data = {"name": "breast_hetero_guest", "namespace": f"experiment{namespace}"}
host_train_data = {"name": "breast_hetero_host", "namespace": f"experiment{namespace}"}
guest_eval_data = {"name": "breast_hetero_guest", "namespace": f"experiment{namespace}"}
host_eval_data = {"name": "breast_hetero_host", "namespace": f"experiment{namespace}"}
train_line = []
# initialize pipeline
pipeline = PipeLine()
# set job initiator
pipeline.set_initiator(role='guest', party_id=guest)
# set participants information
pipeline.set_roles(guest=guest, host=hosts, arbiter=arbiter)
# define Reader components to read in data
reader_0 = Reader(name="reader_0")
# configure Reader for guest
reader_0.get_party_instance(role='guest', party_id=guest).component_param(table=guest_train_data)
# configure Reader for host
reader_0.get_party_instance(role='host', party_id=hosts).component_param(table=host_train_data)
# define DataTransform components
if is_dense:
data_transform_0 = DataTransform(name="data_transform_0", output_format='dense')
else:
data_transform_0 = DataTransform(name="data_transform_0", output_format='sparse')
# get DataTransform party instance of guest
data_transform_0_guest_party_instance = data_transform_0.get_party_instance(role='guest', party_id=guest)
# configure DataTransform for guest
data_transform_0_guest_party_instance.component_param(with_label=True)
# get and configure DataTransform party instance of host
data_transform_0.get_party_instance(role='host', party_id=hosts).component_param(with_label=False)
train_line.append(data_transform_0)
# define Intersection components
intersection_0 = Intersection(name="intersection_0")
pipeline.add_component(reader_0)
pipeline.add_component(data_transform_0, data=Data(data=reader_0.output.data))
pipeline.add_component(intersection_0, data=Data(data=data_transform_0.output.data))
train_line.append(intersection_0)
last_cpn = None
if has_validate:
reader_1 = Reader(name="reader_1")
reader_1.get_party_instance(role='guest', party_id=guest).component_param(table=guest_eval_data)
reader_1.get_party_instance(role='host', party_id=hosts).component_param(table=host_eval_data)
pipeline.add_component(reader_1)
last_cpn = reader_1
for cpn in train_line:
cpn_name = cpn.name
new_name = "_".join(cpn_name.split('_')[:-1] + ['1'])
validate_cpn = type(cpn)(name=new_name)
if hasattr(cpn.output, "model"):
pipeline.add_component(validate_cpn, data=Data(data=last_cpn.output.data),
model=Model(cpn.output.model))
else:
pipeline.add_component(validate_cpn, data=Data(data=last_cpn.output.data))
last_cpn = validate_cpn
hetero_lr_0 = HeteroLR(**lr_param)
if has_validate:
pipeline.add_component(hetero_lr_0, data=Data(train_data=intersection_0.output.data,
validate_data=last_cpn.output.data))
else:
pipeline.add_component(hetero_lr_0, data=Data(train_data=intersection_0.output.data))
if is_cv:
pipeline.compile()
return pipeline
evaluation_data = [hetero_lr_0.output.data]
if has_validate:
hetero_lr_1 = HeteroLR(name='hetero_lr_1')
pipeline.add_component(hetero_lr_1, data=Data(test_data=last_cpn.output.data),
model=Model(hetero_lr_0.output.model))
evaluation_data.append(hetero_lr_1.output.data)
if need_evaluation:
evaluation_0 = Evaluation(name="evaluation_0", eval_type="binary")
pipeline.add_component(evaluation_0, data=Data(data=evaluation_data))
pipeline.compile()
return pipeline
def make_feature_engineering_dsl(config, namespace, lr_param, is_multi_host=False, has_validate=False,
is_cv=False, is_ovr=False):
parties = config.parties
guest = parties.guest[0]
if is_multi_host:
hosts = parties.host
else:
hosts = parties.host[0]
arbiter = parties.arbiter[0]
if is_ovr:
guest_train_data = {"name": "vehicle_scale_hetero_guest", "namespace": f"experiment{namespace}"}
host_train_data = {"name": "vehicle_scale_hetero_host", "namespace": f"experiment{namespace}"}
guest_eval_data = {"name": "vehicle_scale_hetero_guest", "namespace": f"experiment{namespace}"}
host_eval_data = {"name": "vehicle_scale_hetero_host", "namespace": f"experiment{namespace}"}
else:
guest_train_data = {"name": "breast_hetero_guest", "namespace": f"experiment{namespace}"}
host_train_data = {"name": "breast_hetero_host", "namespace": f"experiment{namespace}"}
guest_eval_data = {"name": "breast_hetero_guest", "namespace": f"experiment{namespace}"}
host_eval_data = {"name": "breast_hetero_host", "namespace": f"experiment{namespace}"}
train_line = []
# initialize pipeline
pipeline = PipeLine()
# set job initiator
pipeline.set_initiator(role='guest', party_id=guest)
# set participants information
pipeline.set_roles(guest=guest, host=hosts, arbiter=arbiter)
# define Reader components to read in data
reader_0 = Reader(name="reader_0")
# configure Reader for guest
reader_0.get_party_instance(role='guest', party_id=guest).component_param(table=guest_train_data)
# configure Reader for host
reader_0.get_party_instance(role='host', party_id=hosts).component_param(table=host_train_data)
# define DataTransform components
data_transform_0 = DataTransform(name="data_transform_0") # start component numbering at 0
# get DataTransform party instance of guest
data_transform_0_guest_party_instance = data_transform_0.get_party_instance(role='guest', party_id=guest)
# configure DataTransform for guest
data_transform_0_guest_party_instance.component_param(with_label=True, output_format="dense")
# get and configure DataTransform party instance of host
data_transform_0.get_party_instance(role='host', party_id=hosts).component_param(with_label=False)
train_line.append(data_transform_0)
# define Intersection components
intersection_0 = Intersection(name="intersection_0")
pipeline.add_component(reader_0)
pipeline.add_component(data_transform_0, data=Data(data=reader_0.output.data))
pipeline.add_component(intersection_0, data=Data(data=data_transform_0.output.data))
train_line.append(intersection_0)
feature_scale_0 = FeatureScale(name='feature_scale_0', method="standard_scale",
need_run=True)
pipeline.add_component(feature_scale_0, data=Data(data=intersection_0.output.data))
train_line.append(feature_scale_0)
binning_param = {
"method": "quantile",
"compress_thres": 10000,
"head_size": 10000,
"error": 0.001,
"bin_num": 10,
"bin_indexes": -1,
"adjustment_factor": 0.5,
"local_only": False,
"need_run": True,
"transform_param": {
"transform_cols": -1,
"transform_type": "bin_num"
}
}
hetero_feature_binning_0 = HeteroFeatureBinning(name='hetero_feature_binning_0',
**binning_param)
pipeline.add_component(hetero_feature_binning_0, data=Data(data=feature_scale_0.output.data))
train_line.append(hetero_feature_binning_0)
selection_param = {
"select_col_indexes": -1,
"filter_methods": [
"manually",
"iv_value_thres",
"iv_percentile"
],
"manually_param": {
"filter_out_indexes": None
},
"iv_value_param": {
"value_threshold": 1.0
},
"iv_percentile_param": {
"percentile_threshold": 0.9
},
"need_run": True
}
hetero_feature_selection_0 = HeteroFeatureSelection(name='hetero_feature_selection_0',
**selection_param)
pipeline.add_component(hetero_feature_selection_0, data=Data(data=hetero_feature_binning_0.output.data),
model=Model(isometric_model=[hetero_feature_binning_0.output.model]))
train_line.append(hetero_feature_selection_0)
onehot_param = {
"transform_col_indexes": -1,
"transform_col_names": None,
"need_run": True
}
one_hot_encoder_0 = OneHotEncoder(name='one_hot_encoder_0', **onehot_param)
pipeline.add_component(one_hot_encoder_0, data=Data(data=hetero_feature_selection_0.output.data))
train_line.append(one_hot_encoder_0)
last_cpn = None
if has_validate:
reader_1 = Reader(name="reader_1")
reader_1.get_party_instance(role='guest', party_id=guest).component_param(table=guest_eval_data)
reader_1.get_party_instance(role='host', party_id=hosts).component_param(table=host_eval_data)
pipeline.add_component(reader_1)
last_cpn = reader_1
for cpn in train_line:
cpn_name = cpn.name
new_name = "_".join(cpn_name.split('_')[:-1] + ['1'])
validate_cpn = type(cpn)(name=new_name)
if hasattr(cpn.output, "model"):
pipeline.add_component(validate_cpn, data=Data(data=last_cpn.output.data),
model=Model(cpn.output.model))
else:
pipeline.add_component(validate_cpn, data=Data(data=last_cpn.output.data))
last_cpn = validate_cpn
hetero_lr_0 = HeteroLR(**lr_param)
if has_validate:
pipeline.add_component(hetero_lr_0, data=Data(train_data=one_hot_encoder_0.output.data,
validate_data=last_cpn.output.data))
else:
pipeline.add_component(hetero_lr_0, data=Data(train_data=one_hot_encoder_0.output.data))
if is_cv:
pipeline.compile()
return pipeline
evaluation_data = [hetero_lr_0.output.data]
if has_validate:
hetero_lr_1 = HeteroLR(name='hetero_lr_1')
pipeline.add_component(hetero_lr_1, data=Data(test_data=last_cpn.output.data),
model=Model(hetero_lr_0.output.model))
evaluation_data.append(hetero_lr_1.output.data)
evaluation_0 = Evaluation(name="evaluation_0", eval_type="binary")
pipeline.add_component(evaluation_0, data=Data(data=evaluation_data))
pipeline.compile()
return pipeline
def main(config="../../config.yaml", namespace=""):
if isinstance(config, str):
config = load_job_config(config)
parties = config.parties
guest = parties.guest[0]
hosts = parties.host[0]
arbiter = parties.arbiter[0]
guest_train_data = {
"name": "breast_hetero_guest",
"namespace": f"experiment{namespace}"
}
host_train_data = {
"name": "breast_hetero_host",
"namespace": f"experiment{namespace}"
}
# guest_train_data = {"name": "default_credit_hetero_guest", "namespace": f"experiment{namespace}"}
# host_train_data = {"name": "default_credit_hetero_host", "namespace": f"experiment{namespace}"}
# initialize pipeline
pipeline = PipeLine()
# set job initiator
pipeline.set_initiator(role='guest', party_id=guest)
# set participants information
pipeline.set_roles(guest=guest, host=hosts, arbiter=arbiter)
# define Reader components to read in data
reader_0 = Reader(name="reader_0")
# configure Reader for guest
reader_0.get_party_instance(
role='guest', party_id=guest).component_param(table=guest_train_data)
# configure Reader for host
reader_0.get_party_instance(
role='host', party_id=hosts).component_param(table=host_train_data)
data_transform_0 = DataTransform(name="data_transform_0",
output_format='dense')
# get DataTransform party instance of guest
data_transform_0_guest_party_instance = data_transform_0.get_party_instance(
role='guest', party_id=guest)
# configure DataTransform for guest
data_transform_0_guest_party_instance.component_param(with_label=True)
# get and configure DataTransform party instance of host
data_transform_0.get_party_instance(
role='host', party_id=hosts).component_param(with_label=False)
# define Intersection components
intersection_0 = Intersection(name="intersection_0")
pipeline.add_component(reader_0)
pipeline.add_component(data_transform_0,
data=Data(data=reader_0.output.data))
pipeline.add_component(intersection_0,
data=Data(data=data_transform_0.output.data))
param = {
"model_id": "arbiter-9999#guest-10000#host-9999#model",
"model_version": "202108311438379703480",
"component_name": "hetero_lr_0",
"step_index": 2
}
model_loader_0 = ModelLoader(name="model_loader_0", **param)
lr_param = {
"penalty": "L2",
"optimizer": "rmsprop",
"tol": 0.0001,
"alpha": 0.01,
"early_stop": "diff",
"batch_size": -1,
"learning_rate": 0.15,
"init_param": {
"init_method": "zeros",
"fit_intercept": True
},
"encrypt_param": {
"key_length": 1024
},
"callback_param": {
"callbacks": ["ModelCheckpoint"],
"validation_freqs": 1,
"early_stopping_rounds": 1,
"metrics": None,
"use_first_metric_only": False,
"save_freq": 1
}
}
hetero_lr_0 = HeteroLR(name="hetero_lr_0", max_iter=30, **lr_param)
pipeline.add_component(model_loader_0)
pipeline.add_component(hetero_lr_0,
data=Data(train_data=intersection_0.output.data),
model=Model(model=model_loader_0.output.model))
evaluation_0 = Evaluation(name="evaluation_0", eval_type="binary")
pipeline.add_component(evaluation_0,
data=Data(data=hetero_lr_0.output.data))
pipeline.compile()
# fit model
pipeline.fit()
# query component summary
prettify(pipeline.get_component("hetero_lr_0").get_summary())
prettify(pipeline.get_component("evaluation_0").get_summary())
return pipeline
def main(config="../../config.yaml", param="./lr_config.yaml", namespace=""):
# obtain config
if isinstance(config, str):
config = load_job_config(config)
parties = config.parties
guest = parties.guest[0]
host = parties.host[0]
arbiter = parties.arbiter[0]
if isinstance(param, str):
param = JobConfig.load_from_file(param)
assert isinstance(param, dict)
data_set = param.get("data_guest").split('/')[-1]
if data_set == "default_credit_hetero_guest.csv":
guest_data_table = 'default_credit_hetero_guest'
host_data_table = 'default_credit_hetero_host'
elif data_set == 'breast_hetero_guest.csv':
guest_data_table = 'breast_hetero_guest'
host_data_table = 'breast_hetero_host'
elif data_set == 'give_credit_hetero_guest.csv':
guest_data_table = 'give_credit_hetero_guest'
host_data_table = 'give_credit_hetero_host'
elif data_set == 'epsilon_5k_hetero_guest.csv':
guest_data_table = 'epsilon_5k_hetero_guest'
host_data_table = 'epsilon_5k_hetero_host'
else:
raise ValueError(f"Cannot recognized data_set: {data_set}")
guest_train_data = {"name": guest_data_table, "namespace": f"experiment{namespace}"}
host_train_data = {"name": host_data_table, "namespace": f"experiment{namespace}"}
# initialize pipeline
pipeline = PipeLine()
# set job initiator
pipeline.set_initiator(role='guest', party_id=guest)
# set participants information
pipeline.set_roles(guest=guest, host=host, arbiter=arbiter)
# define Reader components to read in data
reader_0 = Reader(name="reader_0")
# configure Reader for guest
reader_0.get_party_instance(role='guest', party_id=guest).component_param(table=guest_train_data)
# configure Reader for host
reader_0.get_party_instance(role='host', party_id=host).component_param(table=host_train_data)
# define DataTransform components
data_transform_0 = DataTransform(name="data_transform_0") # start component numbering at 0
# get DataTransform party instance of guest
data_transform_0_guest_party_instance = data_transform_0.get_party_instance(role='guest', party_id=guest)
# configure DataTransform for guest
data_transform_0_guest_party_instance.component_param(with_label=True, output_format="dense")
# get and configure DataTransform party instance of host
data_transform_0.get_party_instance(role='host', party_id=host).component_param(with_label=False)
# define Intersection component
intersection_0 = Intersection(name="intersection_0")
lr_param = {
}
config_param = {
"penalty": param["penalty"],
"max_iter": param["max_iter"],
"alpha": param["alpha"],
"learning_rate": param["learning_rate"],
"optimizer": param["optimizer"],
"batch_size": param["batch_size"],
"shuffle": False,
"masked_rate": 0,
"early_stop": "diff",
"tol": 1e-5,
"floating_point_precision": param.get("floating_point_precision"),
"init_param": {
"init_method": param.get("init_method", 'random_uniform'),
"random_seed": param.get("random_seed", 103)
}
}
lr_param.update(config_param)
print(f"lr_param: {lr_param}, data_set: {data_set}")
hetero_lr_0 = HeteroLR(name='hetero_lr_0', **lr_param)
hetero_lr_1 = HeteroLR(name='hetero_lr_1')
evaluation_0 = Evaluation(name='evaluation_0', eval_type="binary")
# add components to pipeline, in order of task execution
pipeline.add_component(reader_0)
pipeline.add_component(data_transform_0, data=Data(data=reader_0.output.data))
pipeline.add_component(intersection_0, data=Data(data=data_transform_0.output.data))
pipeline.add_component(hetero_lr_0, data=Data(train_data=intersection_0.output.data))
pipeline.add_component(hetero_lr_1, data=Data(test_data=intersection_0.output.data),
model=Model(hetero_lr_0.output.model))
pipeline.add_component(evaluation_0, data=Data(data=hetero_lr_0.output.data))
# compile pipeline once finished adding modules, this step will form conf and dsl files for running job
pipeline.compile()
# fit model
job_parameters = JobParameters()
pipeline.fit(job_parameters)
lr_0_data = pipeline.get_component("hetero_lr_0").get_output_data().get("data")
lr_1_data = pipeline.get_component("hetero_lr_1").get_output_data().get("data")
lr_0_score = extract_data(lr_0_data, "predict_result")
lr_0_label = extract_data(lr_0_data, "label")
lr_1_score = extract_data(lr_1_data, "predict_result")
lr_1_label = extract_data(lr_1_data, "label")
lr_0_score_label = extract_data(lr_0_data, "predict_result", keep_id=True)
lr_1_score_label = extract_data(lr_1_data, "predict_result", keep_id=True)
result_summary = parse_summary_result(pipeline.get_component("evaluation_0").get_summary())
metric_lr = {
"score_diversity_ratio": classification_metric.Distribution.compute(lr_0_score_label, lr_1_score_label),
"ks_2samp": classification_metric.KSTest.compute(lr_0_score, lr_1_score),
"mAP_D_value": classification_metric.AveragePrecisionScore().compute(lr_0_score, lr_1_score, lr_0_label,
lr_1_label)}
result_summary["distribution_metrics"] = {"hetero_lr": metric_lr}
data_summary = {"train": {"guest": guest_train_data["name"], "host": host_train_data["name"]},
"test": {"guest": guest_train_data["name"], "host": host_train_data["name"]}
}
return data_summary, result_summary
def main(config="../../config.yaml", namespace=""):
# obtain config
if isinstance(config, str):
config = load_job_config(config)
parties = config.parties
guest = parties.guest[0]
host = parties.host[0]
arbiter = parties.arbiter[0]
guest_train_data = {"name": "breast_hetero_guest", "namespace": f"experiment{namespace}"}
host_train_data = {"name": "breast_hetero_host", "namespace": f"experiment{namespace}"}
pipeline = PipeLine().set_initiator(role='guest', party_id=guest).set_roles(guest=guest, host=host, arbiter=arbiter)
reader_0 = Reader(name="reader_0")
reader_0.get_party_instance(role='guest', party_id=guest).component_param(table=guest_train_data)
reader_0.get_party_instance(role='host', party_id=host).component_param(table=host_train_data)
reader_1 = Reader(name="reader_1")
reader_1.get_party_instance(role='guest', party_id=guest).component_param(table=guest_train_data)
reader_1.get_party_instance(role='host', party_id=host).component_param(table=host_train_data)
data_transform_0 = DataTransform(name="data_transform_0")
data_transform_1 = DataTransform(name="data_transform_1")
data_transform_0.get_party_instance(
role='guest', party_id=guest).component_param(
with_label=True, output_format="dense")
data_transform_0.get_party_instance(role='host', party_id=host).component_param(with_label=False)
data_transform_1.get_party_instance(
role='guest', party_id=guest).component_param(
with_label=True, output_format="dense")
data_transform_1.get_party_instance(role='host', party_id=host).component_param(with_label=False)
intersect_0 = Intersection(name="intersection_0")
intersect_1 = Intersection(name="intersection_1")
union_0 = Union(name="union_0")
hetero_lr_0 = HeteroLR(name="hetero_lr_0", max_iter=3, early_stop="weight_diff",
optimizer="nesterov_momentum_sgd", tol=1E-4, alpha=0.01,
learning_rate=0.15, init_param={"init_method": "random_uniform"})
evaluation_0 = Evaluation(name="evaluation_0", eval_type="binary", pos_label=1)
evaluation_0.get_party_instance(role='host', party_id=host).component_param(need_run=False)
pipeline.add_component(reader_0)
pipeline.add_component(reader_1)
pipeline.add_component(data_transform_0, data=Data(data=reader_0.output.data))
pipeline.add_component(
data_transform_1, data=Data(
data=reader_1.output.data), model=Model(
data_transform_0.output.model))
pipeline.add_component(intersect_0, data=Data(data=data_transform_0.output.data))
pipeline.add_component(intersect_1, data=Data(data=data_transform_1.output.data))
pipeline.add_component(union_0, data=Data(data=[intersect_0.output.data, intersect_1.output.data]))
pipeline.add_component(hetero_lr_0, data=Data(train_data=union_0.output.data))
pipeline.add_component(evaluation_0, data=Data(data=hetero_lr_0.output.data))
pipeline.compile()
pipeline.fit()
def main(config="../../config.yaml", namespace=""):
# obtain config
if isinstance(config, str):
config = load_job_config(config)
parties = config.parties
guest = parties.guest[0]
host = parties.host[0]
arbiter = parties.arbiter[0]
backend = config.backend
work_mode = config.work_mode
guest_train_data = {
"name": "breast_hetero_guest",
"namespace": f"experiment{namespace}"
}
host_train_data = {
"name": "breast_hetero_host",
"namespace": f"experiment{namespace}"
}
pipeline = PipeLine().set_initiator(
role='guest', party_id=guest).set_roles(guest=guest,
host=host,
arbiter=arbiter)
reader_0 = Reader(name="reader_0")
reader_0.get_party_instance(
role='guest', party_id=guest).component_param(table=guest_train_data)
reader_0.get_party_instance(
role='host', party_id=host).component_param(table=host_train_data)
dataio_0 = DataIO(name="dataio_0")
dataio_0.get_party_instance(role='guest', party_id=guest).component_param(
with_label=True,
label_name="y",
label_type="int",
output_format="dense")
dataio_0.get_party_instance(
role='host', party_id=host).component_param(with_label=False)
intersection_0 = Intersection(name="intersection_0")
binning_param = {
"name": 'hetero_feature_binning_0',
"method": "quantile",
"compress_thres": 10000,
"head_size": 10000,
"error": 0.001,
"bin_num": 10,
"bin_indexes": -1,
"bin_names": None,
"category_indexes": None,
"category_names": None,
"adjustment_factor": 0.5,
"local_only": False,
"transform_param": {
"transform_cols": -1,
"transform_names": None,
"transform_type": "bin_num"
}
}
selection_param = {
"name": "hetero_feature_selection_0",
"select_col_indexes": -1,
"select_names": [],
"filter_methods": ["iv_value_thres"],
"iv_value_param": {
"value_threshold": 0.1
}
}
hetero_feature_binning_0 = HeteroFeatureBinning(**binning_param)
hetero_feature_selection_0 = HeteroFeatureSelection(**selection_param)
sample_weight_0 = SampleWeight(name="sample_weight_0")
sample_weight_0.get_party_instance(
role='guest', party_id=guest).component_param(need_run=True,
class_weight={
"0": 1,
"1": 2
})
sample_weight_0.get_party_instance(
role='host', party_id=host).component_param(need_run=False)
feature_scale_0 = FeatureScale(name="feature_scale_0",
method="standard_scale",
need_run=True)
hetero_lr_0 = HeteroLR(name="hetero_lr_0",
optimizer="nesterov_momentum_sgd",
tol=0.001,
alpha=0.01,
max_iter=20,
early_stop="weight_diff",
batch_size=-1,
learning_rate=0.15,
init_param={"init_method": "zeros"})
evaluation_0 = Evaluation(name="evaluation_0",
eval_type="binary",
pos_label=1)
# evaluation_0.get_party_instance(role='host', party_id=host).component_param(need_run=False)
pipeline.add_component(reader_0)
pipeline.add_component(dataio_0, data=Data(data=reader_0.output.data))
pipeline.add_component(intersection_0,
data=Data(data=dataio_0.output.data))
pipeline.add_component(sample_weight_0,
data=Data(data=intersection_0.output.data))
pipeline.add_component(hetero_feature_binning_0,
data=Data(data=sample_weight_0.output.data))
pipeline.add_component(
hetero_feature_selection_0,
data=Data(data=hetero_feature_binning_0.output.data),
model=Model(isometric_model=[hetero_feature_binning_0.output.model]))
pipeline.add_component(feature_scale_0,
data=Data(hetero_feature_selection_0.output.data))
pipeline.add_component(hetero_lr_0,
data=Data(train_data=feature_scale_0.output.data))
pipeline.add_component(evaluation_0,
data=Data(data=hetero_lr_0.output.data))
pipeline.compile()
job_parameters = JobParameters(backend=backend, work_mode=work_mode)
pipeline.fit(job_parameters)
def main(config="../../config.yaml", namespace=""):
# obtain config
if isinstance(config, str):
config = load_job_config(config)
parties = config.parties
guest = parties.guest[0]
host = parties.host[0]
arbiter = parties.arbiter[0]
guest_train_data = {
"name": "breast_hetero_guest",
"namespace": f"experiment{namespace}"
}
host_train_data = {
"name": "breast_hetero_host",
"namespace": f"experiment{namespace}"
}
pipeline = PipeLine().set_initiator(
role='guest', party_id=guest).set_roles(guest=guest,
host=host,
arbiter=arbiter)
reader_0 = Reader(name="reader_0")
reader_0.get_party_instance(
role='guest', party_id=guest).component_param(table=guest_train_data)
reader_0.get_party_instance(
role='host', party_id=host).component_param(table=host_train_data)
reader_1 = Reader(name="reader_1")
reader_1.get_party_instance(
role='guest', party_id=guest).component_param(table=guest_train_data)
reader_1.get_party_instance(
role='host', party_id=host).component_param(table=host_train_data)
reader_2 = Reader(name="reader_2")
reader_2.get_party_instance(
role='guest', party_id=guest).component_param(table=guest_train_data)
reader_2.get_party_instance(
role='host', party_id=host).component_param(table=host_train_data)
data_transform_0 = DataTransform(name="data_transform_0")
data_transform_0.get_party_instance(
role='guest', party_id=guest).component_param(with_label=True,
missing_fill=True,
outlier_replace=True)
data_transform_0.get_party_instance(
role='host', party_id=host).component_param(with_label=False,
missing_fill=True,
outlier_replace=True)
data_transform_1 = DataTransform(name="data_transform_1")
data_transform_2 = DataTransform(name="data_transform_2")
intersection_0 = Intersection(name="intersection_0")
intersection_1 = Intersection(name="intersection_1")
intersection_2 = Intersection(name="intersection_2")
union_0 = Union(name="union_0")
federated_sample_0 = FederatedSample(name="federated_sample_0",
mode="stratified",
method="downsample",
fractions=[[0, 1.0], [1, 1.0]])
feature_scale_0 = FeatureScale(name="feature_scale_0")
feature_scale_1 = FeatureScale(name="feature_scale_1")
hetero_feature_binning_0 = HeteroFeatureBinning(
name="hetero_feature_binning_0")
hetero_feature_binning_1 = HeteroFeatureBinning(
name="hetero_feature_binning_1")
hetero_feature_selection_0 = HeteroFeatureSelection(
name="hetero_feature_selection_0")
hetero_feature_selection_1 = HeteroFeatureSelection(
name="hetero_feature_selection_1")
one_hot_0 = OneHotEncoder(name="one_hot_0")
one_hot_1 = OneHotEncoder(name="one_hot_1")
hetero_lr_0 = HeteroLR(name="hetero_lr_0",
penalty="L2",
optimizer="rmsprop",
tol=1e-5,
init_param={"init_method": "random_uniform"},
alpha=0.01,
max_iter=3,
early_stop="diff",
batch_size=320,
learning_rate=0.15)
hetero_lr_1 = HeteroLR(name="hetero_lr_1")
hetero_lr_2 = HeteroLR(name="hetero_lr_2",
penalty="L2",
optimizer="rmsprop",
tol=1e-5,
init_param={"init_method": "random_uniform"},
alpha=0.01,
max_iter=3,
early_stop="diff",
batch_size=320,
learning_rate=0.15,
cv_param={
"n_splits": 5,
"shuffle": True,
"random_seed": 103,
"need_cv": True
})
hetero_sshe_lr_0 = HeteroSSHELR(
name="hetero_sshe_lr_0",
reveal_every_iter=True,
reveal_strategy="respectively",
penalty="L2",
optimizer="rmsprop",
tol=1e-5,
batch_size=320,
learning_rate=0.15,
init_param={"init_method": "random_uniform"},
alpha=0.01,
max_iter=3)
hetero_sshe_lr_1 = HeteroSSHELR(name="hetero_sshe_lr_1")
local_baseline_0 = LocalBaseline(name="local_baseline_0",
model_name="LogisticRegression",
model_opts={
"penalty": "l2",
"tol": 0.0001,
"C": 1.0,
"fit_intercept": True,
"solver": "lbfgs",
"max_iter": 5,
"multi_class": "ovr"
})
local_baseline_0.get_party_instance(
role='guest', party_id=guest).component_param(need_run=True)
local_baseline_0.get_party_instance(
role='host', party_id=host).component_param(need_run=False)
local_baseline_1 = LocalBaseline(name="local_baseline_1")
hetero_secureboost_0 = HeteroSecureBoost(name="hetero_secureboost_0",
num_trees=3)
hetero_secureboost_1 = HeteroSecureBoost(name="hetero_secureboost_1")
hetero_secureboost_2 = HeteroSecureBoost(name="hetero_secureboost_2",
num_trees=3,
cv_param={
"shuffle": False,
"need_cv": True
})
hetero_linr_0 = HeteroLinR(name="hetero_linr_0",
penalty="L2",
optimizer="sgd",
tol=0.001,
alpha=0.01,
max_iter=3,
early_stop="weight_diff",
batch_size=-1,
learning_rate=0.15,
decay=0.0,
decay_sqrt=False,
init_param={"init_method": "zeros"},
floating_point_precision=23)
hetero_linr_1 = HeteroLinR(name="hetero_linr_1")
hetero_sshe_linr_0 = HeteroSSHELinR(name="hetero_sshe_linr_0",
max_iter=5,
early_stop="weight_diff",
batch_size=-1)
hetero_sshe_linr_1 = HeteroSSHELinR(name="hetero_sshe_linr_1")
hetero_poisson_0 = HeteroPoisson(name="hetero_poisson_0",
early_stop="weight_diff",
max_iter=10,
alpha=100.0,
batch_size=-1,
learning_rate=0.01,
optimizer="rmsprop",
exposure_colname="exposure",
decay_sqrt=False,
tol=0.001,
init_param={"init_method": "zeros"},
penalty="L2")
hetero_poisson_1 = HeteroPoisson(name="hetero_poisson_1")
hetero_sshe_poisson_0 = HeteroSSHEPoisson(name="hetero_sshe_poisson_0",
max_iter=5)
hetero_sshe_poisson_1 = HeteroSSHEPoisson(name="hetero_sshe_poisson_1")
evaluation_0 = Evaluation(name="evaluation_0")
evaluation_1 = Evaluation(name="evaluation_1")
evaluation_2 = Evaluation(name="evaluation_2")
pipeline.add_component(reader_0)
pipeline.add_component(reader_1)
pipeline.add_component(reader_2)
pipeline.add_component(data_transform_0,
data=Data(data=reader_0.output.data))
pipeline.add_component(data_transform_1,
data=Data(data=reader_1.output.data),
model=Model(model=data_transform_0.output.model))
pipeline.add_component(data_transform_2,
data=Data(data=reader_2.output.data),
model=Model(model=data_transform_0.output.model))
pipeline.add_component(intersection_0,
data=Data(data=data_transform_0.output.data))
pipeline.add_component(intersection_1,
data=Data(data=data_transform_1.output.data))
pipeline.add_component(intersection_2,
data=Data(data=data_transform_2.output.data))
pipeline.add_component(
union_0,
data=Data(
data=[intersection_0.output.data, intersection_2.output.data]))
pipeline.add_component(federated_sample_0,
data=Data(data=intersection_1.output.data))
pipeline.add_component(feature_scale_0,
data=Data(data=union_0.output.data))
pipeline.add_component(feature_scale_1,
data=Data(data=federated_sample_0.output.data),
model=Model(model=feature_scale_0.output.model))
pipeline.add_component(hetero_feature_binning_0,
data=Data(data=feature_scale_0.output.data))
pipeline.add_component(
hetero_feature_binning_1,
data=Data(data=feature_scale_1.output.data),
model=Model(model=hetero_feature_binning_0.output.model))
pipeline.add_component(
hetero_feature_selection_0,
data=Data(data=hetero_feature_binning_0.output.data))
pipeline.add_component(
hetero_feature_selection_1,
data=Data(data=hetero_feature_binning_1.output.data),
model=Model(model=hetero_feature_selection_0.output.model))
pipeline.add_component(
one_hot_0, data=Data(data=hetero_feature_selection_0.output.data))
pipeline.add_component(
one_hot_1,
data=Data(data=hetero_feature_selection_1.output.data),
model=Model(model=one_hot_0.output.model))
pipeline.add_component(hetero_lr_0,
data=Data(train_data=one_hot_0.output.data))
pipeline.add_component(hetero_lr_1,
data=Data(test_data=one_hot_1.output.data),
model=Model(model=hetero_lr_0.output.model))
pipeline.add_component(hetero_lr_2,
data=Data(train_data=one_hot_0.output.data))
pipeline.add_component(local_baseline_0,
data=Data(train_data=one_hot_0.output.data))
pipeline.add_component(local_baseline_1,
data=Data(test_data=one_hot_1.output.data),
model=Model(model=local_baseline_0.output.model))
pipeline.add_component(hetero_sshe_lr_0,
data=Data(train_data=one_hot_0.output.data))
pipeline.add_component(hetero_sshe_lr_1,
data=Data(test_data=one_hot_1.output.data),
model=Model(model=hetero_sshe_lr_0.output.model))
pipeline.add_component(hetero_secureboost_0,
data=Data(train_data=one_hot_0.output.data))
pipeline.add_component(
hetero_secureboost_1,
data=Data(test_data=one_hot_1.output.data),
model=Model(model=hetero_secureboost_0.output.model))
pipeline.add_component(hetero_secureboost_2,
data=Data(train_data=one_hot_0.output.data))
pipeline.add_component(hetero_linr_0,
data=Data(train_data=one_hot_0.output.data))
pipeline.add_component(hetero_linr_1,
data=Data(test_data=one_hot_1.output.data),
model=Model(model=hetero_linr_0.output.model))
pipeline.add_component(hetero_sshe_linr_0,
data=Data(train_data=one_hot_0.output.data))
pipeline.add_component(hetero_sshe_linr_1,
data=Data(test_data=one_hot_1.output.data),
model=Model(model=hetero_sshe_linr_0.output.model))
pipeline.add_component(hetero_poisson_0,
data=Data(train_data=one_hot_0.output.data))
pipeline.add_component(hetero_poisson_1,
data=Data(test_data=one_hot_1.output.data),
model=Model(model=hetero_poisson_0.output.model))
pipeline.add_component(
evaluation_0,
data=Data(data=[
hetero_lr_0.output.data, hetero_lr_1.output.data,
hetero_sshe_lr_0.output.data, hetero_sshe_lr_1.output.data,
local_baseline_0.output.data, local_baseline_1.output.data
]))
pipeline.add_component(hetero_sshe_poisson_0,
data=Data(train_data=one_hot_0.output.data))
pipeline.add_component(
hetero_sshe_poisson_1,
data=Data(test_data=one_hot_1.output.data),
model=Model(model=hetero_sshe_poisson_0.output.model))
pipeline.add_component(
evaluation_1,
data=Data(data=[
hetero_linr_0.output.data, hetero_linr_1.output.data,
hetero_sshe_linr_0.output.data, hetero_linr_1.output.data
]))
pipeline.add_component(
evaluation_2,
data=Data(data=[
hetero_poisson_0.output.data, hetero_poisson_1.output.data,
hetero_sshe_poisson_0.output.data,
hetero_sshe_poisson_1.output.data
]))
pipeline.compile()
pipeline.fit()
print(pipeline.get_component("evaluation_0").get_summary())
print(pipeline.get_component("evaluation_1").get_summary())
print(pipeline.get_component("evaluation_2").get_summary())
def main(config="../../config.yaml", namespace=""):
# obtain config
if isinstance(config, str):
config = load_job_config(config)
parties = config.parties
guest = parties.guest[0]
host = parties.host[0]
arbiter = parties.arbiter[0]
guest_train_data = {
"name": "breast_hetero_guest",
"namespace": f"experiment_sid{namespace}"
}
host_train_data = {
"name": "breast_hetero_host",
"namespace": f"experiment_sid{namespace}"
}
pipeline = PipeLine().set_initiator(role='guest', party_id=guest).\
set_roles(guest=guest, host=host, arbiter=arbiter)
reader_0 = Reader(name="reader_0")
reader_0.get_party_instance(
role='guest', party_id=guest).component_param(table=guest_train_data)
reader_0.get_party_instance(
role='host', party_id=host).component_param(table=host_train_data)
data_transform_0 = DataTransform(name="data_transform_0",
with_match_id=True)
data_transform_0.get_party_instance(
role='guest', party_id=guest).component_param(with_label=True)
data_transform_0.get_party_instance(
role='host', party_id=host).component_param(with_label=False)
intersection_0 = Intersection(name="intersection_0")
lr_param = {
"name": "hetero_lr_0",
"penalty": "L2",
"optimizer": "rmsprop",
"tol": 0.0001,
"alpha": 0.01,
"max_iter": 30,
"early_stop": "diff",
"batch_size": 320,
"learning_rate": 0.15,
"init_param": {
"init_method": "zeros"
},
"sqn_param": {
"update_interval_L": 3,
"memory_M": 5,
"sample_size": 5000,
"random_seed": None
},
"cv_param": {
"n_splits": 5,
"shuffle": False,
"random_seed": 103,
"need_cv": False
}
}
hetero_lr_0 = HeteroLR(**lr_param)
# evaluation_0 = Evaluation(name='evaluation_0')
pipeline.add_component(reader_0)
pipeline.add_component(data_transform_0,
data=Data(data=reader_0.output.data))
pipeline.add_component(intersection_0,
data=Data(data=data_transform_0.output.data))
pipeline.add_component(hetero_lr_0,
data=Data(train_data=intersection_0.output.data))
# pipeline.add_component(evaluation_0, data=Data(data=hetero_lr_0.output.data))
pipeline.compile()
pipeline.fit()
def main(config="../../config.yaml", namespace=""):
# obtain config
if isinstance(config, str):
config = load_job_config(config)
parties = config.parties
guest = parties.guest[0]
host = parties.host[0]
arbiter = parties.arbiter[0]
backend = config.backend
work_mode = config.work_mode
guest_train_data = {
"name": "breast_hetero_guest",
"namespace": f"experiment{namespace}"
}
host_train_data = {
"name": "breast_hetero_host",
"namespace": f"experiment{namespace}"
}
# initialize pipeline
pipeline = PipeLine()
# set job initiator
pipeline.set_initiator(role="guest", party_id=guest)
# set participants information
pipeline.set_roles(guest=guest, host=host, arbiter=arbiter)
# define Reader components to read in data
reader_0 = Reader(name="reader_0")
# configure Reader for guest
reader_0.get_party_instance(
role="guest", party_id=guest).algorithm_param(table=guest_train_data)
# configure Reader for host
reader_0.get_party_instance(
role="host", party_id=host).algorithm_param(table=host_train_data)
# define ColumnExpand components
column_expand_0 = ColumnExpand(name="column_expand_0")
column_expand_0.get_party_instance(
role="guest", party_id=guest).algorithm_param(
need_run=True,
method="manual",
append_header=["x_0", "x_1", "x_2", "x_3"],
fill_value=[0, 0.2, 0.5, 1])
column_expand_0.get_party_instance(
role="host", party_id=host).algorithm_param(need_run=False)
# define DataIO components
dataio_0 = DataIO(name="dataio_0") # start component numbering at 0
# get DataIO party instance of guest
dataio_0_guest_party_instance = dataio_0.get_party_instance(role="guest",
party_id=guest)
# configure DataIO for guest
dataio_0_guest_party_instance.algorithm_param(with_label=True,
output_format="dense")
# get and configure DataIO party instance of host
dataio_0.get_party_instance(
role="host", party_id=host).algorithm_param(with_label=False)
# define Intersection components
intersection_0 = Intersection(name="intersection_0",
intersect_method="rsa",
sync_intersect_ids=True,
only_output_key=False)
param = {
"penalty": "L2",
"optimizer": "nesterov_momentum_sgd",
"tol": 0.0001,
"alpha": 0.01,
"max_iter": 20,
"early_stop": "weight_diff",
"batch_size": -1,
"learning_rate": 0.15,
"init_param": {
"init_method": "random_uniform"
},
"sqn_param": {
"update_interval_L": 3,
"memory_M": 5,
"sample_size": 5000,
"random_seed": None
}
}
hetero_lr_0 = HeteroLR(name="hetero_lr_0", **param)
# add components to pipeline, in order of task execution
pipeline.add_component(reader_0)
pipeline.add_component(column_expand_0,
data=Data(data=reader_0.output.data))
pipeline.add_component(dataio_0,
data=Data(data=column_expand_0.output.data))
# set data input sources of intersection components
pipeline.add_component(intersection_0,
data=Data(data=dataio_0.output.data))
pipeline.add_component(hetero_lr_0,
data=Data(train_data=intersection_0.output.data))
# compile pipeline once finished adding modules, this step will form conf and dsl files for running job
pipeline.compile()
# fit model
pipeline.fit(backend=backend, work_mode=work_mode)
# query component summary
print(pipeline.get_component("hetero_lr_0").get_summary())
# predict
# deploy required components
pipeline.deploy_component(
[column_expand_0, dataio_0, intersection_0, hetero_lr_0])
predict_pipeline = PipeLine()
# add data reader onto predict pipeline
predict_pipeline.add_component(reader_0)
# add selected components from train pipeline onto predict pipeline
# specify data source
predict_pipeline.add_component(
pipeline,
data=Data(predict_input={
pipeline.column_expand_0.input.data: reader_0.output.data
}))
# run predict model
predict_pipeline.predict(backend=backend, work_mode=work_mode)
def main(config="../../config.yaml", namespace=""):
# obtain config
if isinstance(config, str):
config = load_job_config(config)
parties = config.parties
guest = parties.guest[0]
host = parties.host[0]
arbiter = parties.arbiter[0]
guest_train_data = {
"name": "breast_hetero_guest",
"namespace": f"experiment_sid{namespace}"
}
host_train_data = {
"name": "breast_hetero_host",
"namespace": f"experiment_sid{namespace}"
}
pipeline = PipeLine().set_initiator(role='guest', party_id=guest).\
set_roles(guest=guest, host=host, arbiter=arbiter)
reader_0 = Reader(name="reader_0")
reader_0.get_party_instance(
role='guest', party_id=guest).component_param(table=guest_train_data)
reader_0.get_party_instance(
role='host', party_id=host).component_param(table=host_train_data)
data_transform_0 = DataTransform(name="data_transform_0",
with_match_id=True)
data_transform_0.get_party_instance(
role='guest', party_id=guest).component_param(with_label=True)
data_transform_0.get_party_instance(
role='host', party_id=host).component_param(with_label=False)
intersection_0 = Intersection(name="intersection_0")
feature_scale_0 = FeatureScale(name='feature_scale_0',
method="standard_scale",
need_run=True)
binning_param = {
"method": "quantile",
"compress_thres": 10000,
"head_size": 10000,
"error": 0.001,
"bin_num": 10,
"bin_indexes": -1,
"adjustment_factor": 0.5,
"local_only": False,
"need_run": True,
"transform_param": {
"transform_cols": -1,
"transform_type": "bin_num"
}
}
hetero_feature_binning_0 = HeteroFeatureBinning(
name='hetero_feature_binning_0', **binning_param)
statistic_0 = DataStatistics(name='statistic_0', statistics=["95%"])
pearson_0 = HeteroPearson(name='pearson_0', column_indexes=-1)
onehot_0 = OneHotEncoder(name='onehot_0')
selection_param = {
"name":
"hetero_feature_selection_0",
"select_col_indexes":
-1,
"select_names": [],
"filter_methods": [
"manually", "unique_value", "iv_filter",
"coefficient_of_variation_value_thres", "outlier_cols"
],
"manually_param": {
"filter_out_indexes": [0, 1, 2],
"filter_out_names": ["x3"]
},
"unique_param": {
"eps": 1e-06
},
"iv_param": {
"metrics": ["iv", "iv", "iv"],
"filter_type": ["threshold", "top_k", "top_percentile"],
"threshold": [0.001, 100, 0.99]
},
"variance_coe_param": {
"value_threshold": 0.3
},
"outlier_param": {
"percentile": 0.95,
"upper_threshold": 2.0
}
}
hetero_feature_selection_0 = HeteroFeatureSelection(**selection_param)
lr_param = {
"name": "hetero_lr_0",
"penalty": "L2",
"optimizer": "rmsprop",
"tol": 0.0001,
"alpha": 0.01,
"max_iter": 30,
"early_stop": "diff",
"batch_size": 320,
"learning_rate": 0.15,
"init_param": {
"init_method": "zeros"
},
"sqn_param": {
"update_interval_L": 3,
"memory_M": 5,
"sample_size": 5000,
"random_seed": None
},
"cv_param": {
"n_splits": 5,
"shuffle": False,
"random_seed": 103,
"need_cv": False
}
}
hetero_lr_0 = HeteroLR(**lr_param)
evaluation_0 = Evaluation(name='evaluation_0')
pipeline.add_component(reader_0)
pipeline.add_component(data_transform_0,
data=Data(data=reader_0.output.data))
pipeline.add_component(intersection_0,
data=Data(data=data_transform_0.output.data))
pipeline.add_component(feature_scale_0,
data=Data(data=intersection_0.output.data))
pipeline.add_component(hetero_feature_binning_0,
data=Data(data=feature_scale_0.output.data))
pipeline.add_component(statistic_0,
data=Data(data=feature_scale_0.output.data))
pipeline.add_component(pearson_0,
data=Data(data=feature_scale_0.output.data))
pipeline.add_component(
hetero_feature_selection_0,
data=Data(data=hetero_feature_binning_0.output.data),
model=Model(isometric_model=[
hetero_feature_binning_0.output.model, statistic_0.output.model
]))
pipeline.add_component(
onehot_0, data=Data(data=hetero_feature_selection_0.output.data))
pipeline.add_component(hetero_lr_0,
data=Data(train_data=onehot_0.output.data))
pipeline.add_component(evaluation_0,
data=Data(data=hetero_lr_0.output.data))
pipeline.compile()
pipeline.fit()
def main(config="../../config.yaml", namespace=""):
# obtain config
if isinstance(config, str):
config = load_job_config(config)
parties = config.parties
guest = parties.guest[0]
host = parties.host[0]
arbiter = parties.arbiter[0]
backend = config.backend
work_mode = config.work_mode
guest_train_data = {
"name": "default_credit_hetero_guest",
"namespace": f"experiment{namespace}"
}
host_train_data = {
"name": "default_credit_hetero_host",
"namespace": f"experiment{namespace}"
}
# initialize pipeline
pipeline = PipeLine()
# set job initiator
pipeline.set_initiator(role="guest", party_id=guest)
# set participants information
pipeline.set_roles(guest=guest, host=host, arbiter=arbiter)
# define Reader components to read in data
reader_0 = Reader(name="reader_0")
# configure Reader for guest
reader_0.get_party_instance(
role="guest", party_id=guest).component_param(table=guest_train_data)
# configure Reader for host
reader_0.get_party_instance(
role="host", party_id=host).component_param(table=host_train_data)
# define DataIO components
dataio_0 = DataIO(name="dataio_0") # start component numbering at 0
# get DataIO party instance of guest
dataio_0_guest_party_instance = dataio_0.get_party_instance(role="guest",
party_id=guest)
# configure DataIO for guest
dataio_0_guest_party_instance.component_param(with_label=True,
output_format="dense")
# get and configure DataIO party instance of host
dataio_0.get_party_instance(
role="host", party_id=host).component_param(with_label=False)
# define Intersection components
intersection_0 = Intersection(name="intersection_0",
intersect_method="rsa",
sync_intersect_ids=True,
only_output_key=False)
param = {
"penalty": "L2",
"optimizer": "nesterov_momentum_sgd",
"tol": 0.0001,
"alpha": 0.01,
"max_iter": 5,
"early_stop": "weight_diff",
"batch_size": -1,
"learning_rate": 0.15,
"init_param": {
"init_method": "random_uniform"
},
"sqn_param": {
"update_interval_L": 3,
"memory_M": 5,
"sample_size": 5000,
"random_seed": None
}
}
hetero_lr_0 = HeteroLR(name="hetero_lr_0", **param)
# define Scorecard component
scorecard_0 = Scorecard(name="scorecard_0")
scorecard_0.get_party_instance(
role="guest", party_id=guest).component_param(need_run=True,
method="credit",
offset=500,
factor=20,
factor_base=2,
upper_limit_ratio=3,
lower_limit_value=0)
scorecard_0.get_party_instance(
role="host", party_id=host).component_param(need_run=False)
# add components to pipeline, in order of task execution
pipeline.add_component(reader_0)
pipeline.add_component(dataio_0, data=Data(data=reader_0.output.data))
# set data input sources of intersection components
pipeline.add_component(intersection_0,
data=Data(data=dataio_0.output.data))
pipeline.add_component(hetero_lr_0,
data=Data(train_data=intersection_0.output.data))
pipeline.add_component(scorecard_0,
data=Data(data=hetero_lr_0.output.data))
# compile pipeline once finished adding modules, this step will form conf and dsl files for running job
pipeline.compile()
# fit model
job_parameters = JobParameters(backend=backend, work_mode=work_mode)
pipeline.fit(job_parameters)
def main(config="../../config.yaml", param="./lr_config.yaml", namespace=""):
# obtain config
if isinstance(config, str):
config = load_job_config(config)
parties = config.parties
guest = parties.guest[0]
host = parties.host[0]
arbiter = parties.arbiter[0]
backend = config.backend
work_mode = config.work_mode
if isinstance(param, str):
param = JobConfig.load_from_file(param)
assert isinstance(param, dict)
data_set = param.get("data_guest").split('/')[-1]
if data_set == "default_credit_hetero_guest.csv":
guest_data_table = 'default_credit_hetero_guest'
host_data_table = 'default_credit_hetero_host'
elif data_set == 'breast_hetero_guest.csv':
guest_data_table = 'breast_hetero_guest'
host_data_table = 'breast_hetero_host'
elif data_set == 'give_credit_hetero_guest.csv':
guest_data_table = 'give_credit_hetero_guest'
host_data_table = 'give_credit_hetero_host'
elif data_set == 'epsilon_5k_hetero_guest.csv':
guest_data_table = 'epsilon_5k_hetero_guest'
host_data_table = 'epsilon_5k_hetero_host'
else:
raise ValueError(f"Cannot recognized data_set: {data_set}")
guest_train_data = {
"name": guest_data_table,
"namespace": f"experiment{namespace}"
}
host_train_data = {
"name": host_data_table,
"namespace": f"experiment{namespace}"
}
# initialize pipeline
pipeline = PipeLine()
# set job initiator
pipeline.set_initiator(role='guest', party_id=guest)
# set participants information
pipeline.set_roles(guest=guest, host=host, arbiter=arbiter)
# define Reader components to read in data
reader_0 = Reader(name="reader_0")
# configure Reader for guest
reader_0.get_party_instance(
role='guest', party_id=guest).component_param(table=guest_train_data)
# configure Reader for host
reader_0.get_party_instance(
role='host', party_id=host).component_param(table=host_train_data)
# define DataIO components
dataio_0 = DataIO(name="dataio_0") # start component numbering at 0
# get DataIO party instance of guest
dataio_0_guest_party_instance = dataio_0.get_party_instance(role='guest',
party_id=guest)
# configure DataIO for guest
dataio_0_guest_party_instance.component_param(with_label=True,
output_format="dense")
# get and configure DataIO party instance of host
dataio_0.get_party_instance(
role='host', party_id=host).component_param(with_label=False)
# define Intersection component
intersection_0 = Intersection(name="intersection_0")
lr_param = {
"validation_freqs": None,
"early_stopping_rounds": None,
}
config_param = {
"penalty": param["penalty"],
"max_iter": param["max_iter"],
"alpha": param["alpha"],
"learning_rate": param["learning_rate"],
"optimizer": param["optimizer"],
"batch_size": param["batch_size"],
"early_stop": "diff",
"tol": 1e-5,
"init_param": {
"init_method": param.get("init_method", 'random_uniform')
}
}
lr_param.update(config_param)
print(f"lr_param: {lr_param}, data_set: {data_set}")
hetero_lr_0 = HeteroLR(name='hetero_lr_0', **lr_param)
evaluation_0 = Evaluation(name='evaluation_0', eval_type="binary")
# add components to pipeline, in order of task execution
pipeline.add_component(reader_0)
pipeline.add_component(dataio_0, data=Data(data=reader_0.output.data))
pipeline.add_component(intersection_0,
data=Data(data=dataio_0.output.data))
pipeline.add_component(hetero_lr_0,
data=Data(train_data=intersection_0.output.data))
pipeline.add_component(evaluation_0,
data=Data(data=hetero_lr_0.output.data))
# compile pipeline once finished adding modules, this step will form conf and dsl files for running job
pipeline.compile()
# fit model
job_parameters = JobParameters(backend=backend, work_mode=work_mode)
pipeline.fit(job_parameters)
# query component summary
print(pipeline.get_component("evaluation_0").get_summary())
data_summary = {
"train": {
"guest": guest_train_data["name"],
"host": host_train_data["name"]
},
"test": {
"guest": guest_train_data["name"],
"host": host_train_data["name"]
}
}
result_summary = pipeline.get_component("evaluation_0").get_summary()
return data_summary, result_summary
def main(config="../../config.yaml", namespace=""):
# obtain config
if isinstance(config, str):
config = load_job_config(config)
parties = config.parties
guest = parties.guest[0]
host = parties.host[0]
arbiter = parties.arbiter[0]
guest_train_data = {
"name": "breast_hetero_guest",
"namespace": f"experiment{namespace}"
}
host_train_data = {
"name": "breast_hetero_host",
"namespace": f"experiment{namespace}"
}
pipeline = PipeLine().set_initiator(
role='guest', party_id=guest).set_roles(guest=guest,
host=host,
arbiter=arbiter)
reader_0 = Reader(name="reader_0")
reader_0.get_party_instance(
role='guest', party_id=guest).component_param(table=guest_train_data)
reader_0.get_party_instance(
role='host', party_id=host).component_param(table=host_train_data)
data_transform_0 = DataTransform(name="data_transform_0")
data_transform_0.get_party_instance(
role='guest', party_id=guest).component_param(with_label=True,
label_name="y",
label_type="int",
output_format="dense")
data_transform_0.get_party_instance(
role='host', party_id=host).component_param(with_label=False)
intersection_0 = Intersection(name="intersection_0")
sample_weight_0 = SampleWeight(name="sample_weight_0")
sample_weight_0.get_party_instance(
role='guest', party_id=guest).component_param(need_run=True,
class_weight="balanced")
sample_weight_0.get_party_instance(
role='host', party_id=host).component_param(need_run=False)
sample_weight_1 = SampleWeight(name="sample_weight_1")
hetero_lr_0 = HeteroLR(name="hetero_lr_0",
optimizer="nesterov_momentum_sgd",
tol=0.001,
alpha=0.01,
max_iter=20,
early_stop="weight_diff",
batch_size=-1,
learning_rate=0.15,
init_param={"init_method": "zeros"})
evaluation_0 = Evaluation(name="evaluation_0",
eval_type="binary",
pos_label=1)
# evaluation_0.get_party_instance(role='host', party_id=host).component_param(need_run=False)
pipeline.add_component(reader_0)
pipeline.add_component(data_transform_0,
data=Data(data=reader_0.output.data))
pipeline.add_component(intersection_0,
data=Data(data=data_transform_0.output.data))
pipeline.add_component(sample_weight_0,
data=Data(data=intersection_0.output.data))
pipeline.add_component(sample_weight_1,
data=Data(data=intersection_0.output.data),
model=Model(model=sample_weight_0.output.model))
pipeline.add_component(hetero_lr_0,
data=Data(train_data=sample_weight_1.output.data))
pipeline.add_component(evaluation_0,
data=Data(data=hetero_lr_0.output.data))
pipeline.compile()
pipeline.fit()
# predict
# deploy required components
pipeline.deploy_component(
[data_transform_0, intersection_0, sample_weight_0, hetero_lr_0])
predict_pipeline = PipeLine()
# add data reader onto predict pipeline
predict_pipeline.add_component(reader_0)
# add selected components from train pipeline onto predict pipeline
# specify data source
predict_pipeline.add_component(
pipeline,
data=Data(predict_input={
pipeline.data_transform_0.input.data: reader_0.output.data
}))
# run predict model
predict_pipeline.predict()
def main(config="../../config.yaml",
param="./vechile_config.yaml",
namespace=""):
# obtain config
if isinstance(config, str):
config = load_job_config(config)
parties = config.parties
guest = parties.guest[0]
host = parties.host[0]
arbiter = parties.arbiter[0]
backend = config.backend
work_mode = config.work_mode
if isinstance(param, str):
param = JobConfig.load_from_file(param)
assert isinstance(param, dict)
"""
guest = 9999
host = 10000
arbiter = 9999
backend = 0
work_mode = 1
param = {"penalty": "L2", "max_iter": 5}
"""
data_set = param.get("data_guest").split('/')[-1]
if data_set == "vehicle_scale_hetero_guest.csv":
guest_data_table = 'vehicle_scale_hetero_guest'
host_data_table = 'vehicle_scale_hetero_host'
else:
raise ValueError(f"Cannot recognized data_set: {data_set}")
guest_train_data = {
"name": guest_data_table,
"namespace": f"experiment{namespace}"
}
host_train_data = {
"name": host_data_table,
"namespace": f"experiment{namespace}"
}
# initialize pipeline
pipeline = PipeLine()
# set job initiator
pipeline.set_initiator(role='guest', party_id=guest)
# set participants information
pipeline.set_roles(guest=guest, host=host, arbiter=arbiter)
# define Reader components to read in data
reader_0 = Reader(name="reader_0")
# configure Reader for guest
reader_0.get_party_instance(
role='guest', party_id=guest).component_param(table=guest_train_data)
# configure Reader for host
reader_0.get_party_instance(
role='host', party_id=host).component_param(table=host_train_data)
# define DataIO components
dataio_0 = DataIO(name="dataio_0") # start component numbering at 0
# get DataIO party instance of guest
dataio_0_guest_party_instance = dataio_0.get_party_instance(role='guest',
party_id=guest)
# configure DataIO for guest
dataio_0_guest_party_instance.component_param(with_label=True,
output_format="dense")
# get and configure DataIO party instance of host
dataio_0.get_party_instance(
role='host', party_id=host).component_param(with_label=False)
# define Intersection component
intersection_0 = Intersection(name="intersection_0")
lr_param = {
"validation_freqs": None,
"early_stopping_rounds": None,
}
config_param = {
"penalty": param["penalty"],
"max_iter": param["max_iter"],
"alpha": param["alpha"],
"learning_rate": param["learning_rate"],
"optimizer": param["optimizer"],
"batch_size": param["batch_size"],
"early_stop": "diff",
"init_param": {
"init_method": param.get("init_method", 'random_uniform'),
"random_seed": param.get("random_seed", 103)
}
}
lr_param.update(config_param)
print(f"lr_param: {lr_param}, data_set: {data_set}")
hetero_lr_0 = HeteroLR(name='hetero_lr_0', **lr_param)
hetero_lr_1 = HeteroLR(name='hetero_lr_1')
evaluation_0 = Evaluation(name='evaluation_0', eval_type="multi")
# add components to pipeline, in order of task execution
pipeline.add_component(reader_0)
pipeline.add_component(dataio_0, data=Data(data=reader_0.output.data))
pipeline.add_component(intersection_0,
data=Data(data=dataio_0.output.data))
pipeline.add_component(hetero_lr_0,
data=Data(train_data=intersection_0.output.data))
pipeline.add_component(hetero_lr_1,
data=Data(test_data=intersection_0.output.data),
model=Model(hetero_lr_0.output.model))
pipeline.add_component(evaluation_0,
data=Data(data=hetero_lr_0.output.data))
# compile pipeline once finished adding modules, this step will form conf and dsl files for running job
pipeline.compile()
# fit model
job_parameters = JobParameters(backend=backend, work_mode=work_mode)
pipeline.fit(job_parameters)
# query component summary
result_summary = parse_summary_result(
pipeline.get_component("evaluation_0").get_summary())
lr_0_data = pipeline.get_component("hetero_lr_0").get_output_data().get(
"data")
lr_1_data = pipeline.get_component("hetero_lr_1").get_output_data().get(
"data")
lr_0_score_label = extract_data(lr_0_data, "predict_result", keep_id=True)
lr_1_score_label = extract_data(lr_1_data, "predict_result", keep_id=True)
metric_lr = {
"score_diversity_ratio":
classification_metric.Distribution.compute(lr_0_score_label,
lr_1_score_label)
}
result_summary["distribution_metrics"] = {"hetero_lr": metric_lr}
data_summary = {
"train": {
"guest": guest_train_data["name"],
"host": host_train_data["name"]
},
"test": {
"guest": guest_train_data["name"],
"host": host_train_data["name"]
}
}
return data_summary, result_summary
def main(config="../../config.yaml", namespace=""):
# obtain config
if isinstance(config, str):
config = load_job_config(config)
parties = config.parties
guest = parties.guest[0]
host = parties.host[0]
arbiter = parties.arbiter[0]
guest_train_data = {
"name": "breast_hetero_guest",
"namespace": f"experiment{namespace}"
}
host_train_data = {
"name": "breast_hetero_host",
"namespace": f"experiment{namespace}"
}
pipeline = PipeLine().set_initiator(
role='guest', party_id=guest).set_roles(guest=guest,
host=host,
arbiter=arbiter)
reader_0 = Reader(name="reader_0")
reader_0.get_party_instance(
role='guest', party_id=guest).component_param(table=guest_train_data)
reader_0.get_party_instance(
role='host', party_id=host).component_param(table=host_train_data)
data_transform_0 = DataTransform(name="data_transform_0")
data_transform_0.get_party_instance(
role='guest', party_id=guest).component_param(with_label=True,
missing_fill=True,
outlier_replace=True)
data_transform_0.get_party_instance(
role='host', party_id=host).component_param(with_label=False,
missing_fill=True,
outlier_replace=True)
intersection_0 = Intersection(name="intersection_0")
federated_sample_0 = FederatedSample(name="federated_sample_0",
mode="stratified",
method="upsample",
fractions=[[0, 1.5], [1, 2.0]])
feature_scale_0 = FeatureScale(name="feature_scale_0")
hetero_feature_binning_0 = HeteroFeatureBinning(
name="hetero_feature_binning_0")
hetero_feature_selection_0 = HeteroFeatureSelection(
name="hetero_feature_selection_0")
one_hot_0 = OneHotEncoder(name="one_hot_0")
hetero_lr_0 = HeteroLR(name="hetero_lr_0",
penalty="L2",
optimizer="rmsprop",
tol=1e-5,
init_param={"init_method": "random_uniform"},
alpha=0.01,
max_iter=10,
early_stop="diff",
batch_size=320,
learning_rate=0.15)
hetero_lr_1 = HeteroLR(name="hetero_lr_1",
penalty="L2",
optimizer="rmsprop",
tol=1e-5,
init_param={"init_method": "random_uniform"},
alpha=0.01,
max_iter=10,
early_stop="diff",
batch_size=320,
learning_rate=0.15,
cv_param={
"n_splits": 5,
"shuffle": True,
"random_seed": 103,
"need_cv": True
})
hetero_secureboost_0 = HeteroSecureBoost(name="hetero_secureboost_0",
num_trees=5,
cv_param={
"shuffle": False,
"need_cv": True
})
hetero_secureboost_1 = HeteroSecureBoost(name="hetero_secureboost_1",
num_trees=5)
evaluation_0 = Evaluation(name="evaluation_0")
evaluation_1 = Evaluation(name="evaluation_1")
pipeline.add_component(reader_0)
pipeline.add_component(data_transform_0,
data=Data(data=reader_0.output.data))
pipeline.add_component(intersection_0,
data=Data(data=data_transform_0.output.data))
pipeline.add_component(federated_sample_0,
data=Data(data=intersection_0.output.data))
pipeline.add_component(feature_scale_0,
data=Data(data=federated_sample_0.output.data))
pipeline.add_component(hetero_feature_binning_0,
data=Data(data=feature_scale_0.output.data))
pipeline.add_component(
hetero_feature_selection_0,
data=Data(data=hetero_feature_binning_0.output.data))
pipeline.add_component(
one_hot_0, data=Data(data=hetero_feature_selection_0.output.data))
pipeline.add_component(hetero_lr_0,
data=Data(train_data=one_hot_0.output.data))
pipeline.add_component(hetero_lr_1,
data=Data(train_data=one_hot_0.output.data))
pipeline.add_component(hetero_secureboost_0,
data=Data(train_data=one_hot_0.output.data))
pipeline.add_component(hetero_secureboost_1,
data=Data(train_data=one_hot_0.output.data))
pipeline.add_component(evaluation_0,
data=Data(data=hetero_lr_0.output.data))
pipeline.add_component(evaluation_1,
data=Data(data=hetero_secureboost_1.output.data))
pipeline.compile()
pipeline.fit()
print(pipeline.get_component("evaluation_0").get_summary())
def main():
# parties config
guest = 9999
host = 10000
arbiter = 10000
# specify input data name & namespace in database
guest_train_data = {
"name": "breast_hetero_guest",
"namespace": "experiment"
}
host_train_data = {"name": "breast_hetero_host", "namespace": "experiment"}
guest_eval_data = {
"name": "breast_hetero_guest",
"namespace": "experiment"
}
host_eval_data = {"name": "breast_hetero_host", "namespace": "experiment"}
# initialize pipeline
pipeline = PipeLine()
# set job initiator
pipeline.set_initiator(role="guest", party_id=guest)
# set participants information
pipeline.set_roles(guest=guest, host=host, arbiter=arbiter)
# define Reader components to read in data
reader_0 = Reader(name="reader_0")
# configure Reader for guest
reader_0.get_party_instance(
role="guest", party_id=guest).component_param(table=guest_train_data)
# configure Reader for host
reader_0.get_party_instance(
role="host", party_id=host).component_param(table=host_train_data)
# define DataTransform component
data_transform_0 = DataTransform(name="data_transform_0")
# get DataTransform party instance of guest
data_transform_0_guest_party_instance = data_transform_0.get_party_instance(
role="guest", party_id=guest)
# configure DataTransform for guest
data_transform_0_guest_party_instance.component_param(
with_label=True, output_format="dense")
# get and configure DataTransform party instance of host
data_transform_0.get_party_instance(
role="host", party_id=host).component_param(with_label=False)
# define Intersection components
intersection_0 = Intersection(name="intersection_0")
# define HeteroLR component
hetero_lr_0 = HeteroLR(name="hetero_lr_0",
early_stop="diff",
learning_rate=0.15,
optimizer="rmsprop",
max_iter=10)
# add components to pipeline, in order of task execution
pipeline.add_component(reader_0)
pipeline.add_component(data_transform_0,
data=Data(data=reader_0.output.data))
# set data input sources of intersection components
pipeline.add_component(intersection_0,
data=Data(data=data_transform_0.output.data))
# set train data of hetero_lr_0 component
pipeline.add_component(hetero_lr_0,
data=Data(train_data=intersection_0.output.data))
# compile pipeline once finished adding modules, this step will form conf and dsl files for running job
pipeline.compile()
# fit model
pipeline.fit()
# query component summary
import json
print(
json.dumps(pipeline.get_component("hetero_lr_0").get_summary(),
indent=4))
# predict
# deploy required components
pipeline.deploy_component([data_transform_0, intersection_0, hetero_lr_0])
# initiate predict pipeline
predict_pipeline = PipeLine()
# define new data reader
reader_1 = Reader(name="reader_1")
reader_1.get_party_instance(
role="guest", party_id=guest).component_param(table=guest_eval_data)
reader_1.get_party_instance(
role="host", party_id=host).component_param(table=host_eval_data)
# define evaluation component
evaluation_0 = Evaluation(name="evaluation_0")
evaluation_0.get_party_instance(
role="guest", party_id=guest).component_param(need_run=True,
eval_type="binary")
evaluation_0.get_party_instance(
role="host", party_id=host).component_param(need_run=False)
# add data reader onto predict pipeline
predict_pipeline.add_component(reader_1)
# add selected components from train pipeline onto predict pipeline
# specify data source
predict_pipeline.add_component(
pipeline,
data=Data(predict_input={
pipeline.data_transform_0.input.data: reader_1.output.data
}))
# add evaluation component to predict pipeline
predict_pipeline.add_component(
evaluation_0, data=Data(data=pipeline.hetero_lr_0.output.data))
# run predict model
predict_pipeline.predict()
def main():
# parties config
guest = 9999
host = 10000
arbiter = 10000
# 0 for eggroll, 1 for spark
backend = Backend.EGGROLL
# 0 for standalone, 1 for cluster
work_mode = WorkMode.STANDALONE
# use the work mode below for cluster deployment
# work_mode = WorkMode.CLUSTER
# specify input data name & namespace in database
guest_train_data = {
"name": "breast_hetero_guest",
"namespace": "experiment"
}
host_train_data = {"name": "breast_hetero_host", "namespace": "experiment"}
guest_eval_data = {
"name": "breast_hetero_guest",
"namespace": "experiment"
}
host_eval_data = {"name": "breast_hetero_host", "namespace": "experiment"}
# initialize pipeline
pipeline = PipeLine()
# set job initiator
pipeline.set_initiator(role="guest", party_id=guest)
# set participants information
pipeline.set_roles(guest=guest, host=host, arbiter=arbiter)
# define Reader components to read in data
reader_0 = Reader(name="reader_0")
# configure Reader for guest
reader_0.get_party_instance(
role="guest", party_id=guest).algorithm_param(table=guest_train_data)
# configure Reader for host
reader_0.get_party_instance(
role="host", party_id=host).algorithm_param(table=host_train_data)
reader_1 = Reader(name="reader_1")
reader_1.get_party_instance(
role="guest", party_id=guest).algorithm_param(table=guest_eval_data)
reader_1.get_party_instance(
role="host", party_id=host).algorithm_param(table=host_eval_data)
# define DataIO components
dataio_0 = DataIO(name="dataio_0")
dataio_1 = DataIO(name="dataio_1")
# get DataIO party instance of guest
dataio_0_guest_party_instance = dataio_0.get_party_instance(role="guest",
party_id=guest)
# configure DataIO for guest
dataio_0_guest_party_instance.algorithm_param(with_label=True,
output_format="dense")
# get and configure DataIO party instance of host
dataio_0.get_party_instance(
role="host", party_id=host).algorithm_param(with_label=False)
# define Intersection components
intersection_0 = Intersection(name="intersection_0")
intersection_1 = Intersection(name="intersection_1")
# define HeteroLR component
hetero_lr_0 = HeteroLR(name="hetero_lr_0",
early_stop="weight_diff",
learning_rate=0.15,
optimizer="rmsprop",
max_iter=10,
early_stopping_rounds=2,
validation_freqs=1)
# add components to pipeline, in order of task execution
pipeline.add_component(reader_0)
pipeline.add_component(reader_1)
pipeline.add_component(dataio_0, data=Data(data=reader_0.output.data))
# set dataio_1 to replicate model from dataio_0
pipeline.add_component(dataio_1,
data=Data(data=reader_1.output.data),
model=Model(dataio_0.output.model))
# set data input sources of intersection components
pipeline.add_component(intersection_0,
data=Data(data=dataio_0.output.data))
pipeline.add_component(intersection_1,
data=Data(data=dataio_1.output.data))
# set train & validate data of hetero_lr_0 component
pipeline.add_component(hetero_lr_0,
data=Data(train_data=intersection_0.output.data,
validate_data=intersection_1.output.data))
# compile pipeline once finished adding modules, this step will form conf and dsl files for running job
pipeline.compile()
# fit model
pipeline.fit(backend=backend, work_mode=work_mode)
# query component summary
import json
print(
json.dumps(pipeline.get_component("hetero_lr_0").get_summary(),
indent=4))
# predict
# deploy required components
pipeline.deploy_component([dataio_0, intersection_0, hetero_lr_0])
# initiate predict pipeline
predict_pipeline = PipeLine()
reader_2 = Reader(name="reader_2")
reader_2.get_party_instance(
role="guest", party_id=guest).algorithm_param(table=guest_eval_data)
reader_2.get_party_instance(
role="host", party_id=host).algorithm_param(table=host_eval_data)
# add data reader onto predict pipeline
predict_pipeline.add_component(reader_2)
# add selected components from train pipeline onto predict pipeline
# specify data source
predict_pipeline.add_component(
pipeline,
data=Data(
predict_input={pipeline.dataio_0.input.data: reader_2.output.data
}))
# run predict model
predict_pipeline.predict(backend=backend, work_mode=work_mode)
