def filter_one_method(self, data_instances, method):
if method == consts.IV_VALUE_THRES:
LOGGER.debug("In host party, sending select_cols")
self._send_select_cols(consts.IV_VALUE_THRES)
self._received_result_cols(filter_name=consts.IV_VALUE_THRES)
LOGGER.info(
"[Result][FeatureSelection][Host]Finish iv value threshold filter. Current left cols are: {}"
.format(self.left_cols))
if method == consts.IV_PERCENTILE:
self._send_select_cols(consts.IV_PERCENTILE)
self._received_result_cols(filter_name=consts.IV_PERCENTILE)
LOGGER.info(
"[Result][FeatureSelection][Host]Finish iv percentile filter. Current left cols are: {}"
.format(self.left_cols))
if method == consts.COEFFICIENT_OF_VARIATION_VALUE_THRES:
variance_coe_param = self.model_param.variance_coe_param
coe_filter = feature_selection.CoeffOfVarValueFilter(
variance_coe_param, self.cols, self.static_obj)
new_left_cols = coe_filter.fit(data_instances)
self._renew_final_left_cols(new_left_cols)
self.static_obj = coe_filter.statics_obj
self.variance_coe_meta = coe_filter.get_meta_obj()
self.results.append(coe_filter.get_param_obj())
LOGGER.debug(
"[Result][FeatureSelection][Host]Finish coeffiecient_of_variation value threshold filter."
" Current left cols are: {}".format(self.left_cols))
if method == consts.UNIQUE_VALUE:
unique_param = self.model_param.unique_param
unique_filter = feature_selection.UniqueValueFilter(
unique_param, self.cols, self.static_obj)
new_left_cols = unique_filter.fit(data_instances)
self._renew_final_left_cols(new_left_cols)
self.static_obj = unique_filter.statics_obj
self.unique_meta = unique_filter.get_meta_obj()
self.results.append(unique_filter.get_param_obj())
# self._renew_left_col_names()
LOGGER.info(
"[Result][FeatureSelection][Host]Finish unique value filter. Current left cols are: {}"
.format(self.left_cols))
if method == consts.OUTLIER_COLS:
outlier_param = self.model_param.outlier_param
outlier_filter = feature_selection.OutlierFilter(
outlier_param, self.cols)
new_left_cols = outlier_filter.fit(data_instances)
self._renew_final_left_cols(new_left_cols)
self.outlier_meta = outlier_filter.get_meta_obj()
self.results.append(outlier_filter.get_param_obj())
# self._renew_left_col_names()
LOGGER.info(
"[Result][FeatureSelection][Host]Finish outlier cols filter. Current left cols are: {}"
.format(self.left_cols))
def filter_one_method(self, data_instances, method):
if method == consts.IV_VALUE_THRES:
iv_param = self.params.iv_value_param
iv_filter = feature_selection.IVValueSelectFilter(
iv_param, self.left_col_names, self.binning_model)
new_left_cols = iv_filter.fit(data_instances)
self.left_cols = new_left_cols
if not self.local_only:
host_select_cols = self._get_host_select_cols(
consts.IV_VALUE_THRES)
host_left_cols = iv_filter.host_cols
# Only one host
left_cols = host_left_cols.get(consts.HOST)
new_result = {}
for host_col_name in host_select_cols:
if host_col_name not in left_cols:
LOGGER.warning(
"Host column {} has not been set in feature binning module"
.format(host_col_name))
continue
is_left = left_cols.get(host_col_name)
new_result[host_col_name] = is_left
self.host_left_cols = new_result
self._send_host_result_cols(consts.IV_VALUE_THRES)
LOGGER.info(
"[Result][FeatureSelection][Guest] Finish iv value threshold filter. Host left cols are: {}"
.format(self.host_left_cols))
iv_filter.display_feature_result(self.party_name)
LOGGER.info(
"[Result][FeatureSelection][Guest] Finish iv value threshold filter. Self left cols are: {}"
.format(self.left_cols))
self.iv_value_meta = iv_filter.get_meta_obj()
self.results.append(iv_filter.get_param_obj())
self._renew_left_col_names()
if method == consts.IV_PERCENTILE:
iv_param = self.params.iv_percentile_param
if self.local_only:
iv_filter = feature_selection.IVPercentileFilter(
iv_param, self.left_col_names, {}, self.binning_model)
self.left_cols = iv_filter.fit(data_instances)
else:
host_select_cols = self._get_host_select_cols(
consts.IV_PERCENTILE)
host_cols = {consts.HOST: host_select_cols}
iv_filter = feature_selection.IVPercentileFilter(
iv_param, self.left_col_names, host_cols,
self.binning_model)
self.left_cols = iv_filter.fit(data_instances)
host_left_cols = iv_filter.host_cols
# Only one host
left_cols = host_left_cols.get(consts.HOST)
new_result = {}
for host_col_name in host_select_cols:
if host_col_name not in left_cols:
LOGGER.warning(
"Host column {} has not been set in feature binning module"
.format(host_col_name))
continue
is_left = left_cols.get(host_col_name)
new_result[host_col_name] = is_left
self.host_left_cols = new_result
self._send_host_result_cols(consts.IV_PERCENTILE)
LOGGER.info(
"[Result][FeatureSelection][Host]Finish iv value threshold filter. Host left cols are: {}"
.format(self.host_left_cols))
iv_filter.display_feature_result(self.party_name)
LOGGER.info(
"[Result][FeatureSelection][Guest]Finish iv value threshold filter. Self left cols are: {}"
.format(self.left_cols))
self.iv_percentile_meta = iv_filter.get_meta_obj()
self.results.append(iv_filter.get_param_obj())
self._renew_left_col_names()
if method == consts.COEFFICIENT_OF_VARIATION_VALUE_THRES:
coe_param = self.params.coe_param
coe_filter = feature_selection.CoeffOfVarValueFilter(
coe_param, self.left_col_names, self.static_obj)
self.left_cols = coe_filter.fit(data_instances)
self.static_obj = coe_filter.statics_obj
self.coe_meta = coe_filter.get_meta_obj()
self.results.append(coe_filter.get_param_obj())
self._renew_left_col_names()
coe_filter.display_feature_result(self.party_name)
LOGGER.info(
"[Result][FeatureSelection][Guest] Finish coefficient threshold filter. Self left cols are: {}"
.format(self.left_cols))
if method == consts.UNIQUE_VALUE:
unique_param = self.params.unique_param
unique_filter = feature_selection.UniqueValueFilter(
unique_param, self.left_col_names, self.static_obj)
self.left_cols = unique_filter.fit(data_instances)
self.static_obj = unique_filter.statics_obj
self.unique_meta = unique_filter.get_meta_obj()
self.results.append(unique_filter.get_param_obj())
self._renew_left_col_names()
unique_filter.display_feature_result(self.party_name)
LOGGER.info(
"[Result][FeatureSelection][Guest]Finish unique value filter. Current left cols are: {}"
.format(self.left_cols))
if method == consts.OUTLIER_COLS:
outlier_param = self.params.outlier_param
outlier_filter = feature_selection.OutlierFilter(
outlier_param, self.left_col_names)
self.left_cols = outlier_filter.fit(data_instances)
self.outlier_meta = outlier_filter.get_meta_obj()
self.results.append(outlier_filter.get_param_obj())
self._renew_left_col_names()
outlier_filter.display_feature_result(self.party_name)
LOGGER.info(
"[Result][FeatureSelection][Guest]Finish outlier cols filter. Current left cols are: {}"
.format(self.left_cols))
def filter_one_method(self, data_instances, method):
if method == consts.IV_VALUE_THRES:
self._calculates_iv_attrs(data_instances,
flowid_postfix='iv_value')
iv_param = self.params.iv_param
iv_filter = feature_selection.IVValueSelectFilter(
iv_param, self.left_cols, self.guest_iv_attrs)
new_left_cols = iv_filter.filter()
self.results.append(iv_filter.to_result())
# Renew current left cols and iv_attrs
new_iv_list = self._renew_iv_attrs(new_left_cols, self.left_cols,
self.guest_iv_attrs)
self.guest_iv_attrs = new_iv_list
self.left_cols = new_left_cols
if not self.local_only:
self._filter_host_iv_value()
LOGGER.info(
"Finish iv value threshold filter. Current left cols are: {}".
format(self.left_cols))
if method == consts.IV_PERCENTILE:
self._calculates_iv_attrs(data_instances,
flowid_postfix='iv_percentile')
iv_param = self.params.iv_param
iv_filter = feature_selection.IVPercentileFilter(iv_param)
iv_filter.add_attrs(self.guest_iv_attrs, self.left_cols)
if not self.local_only:
iv_filter.add_attrs(self.host_iv_attrs, self.host_left_cols)
left_cols = iv_filter.filter_multiple_parties()
new_left_cols = left_cols[0]
self.results.append(iv_filter.to_result())
# Renew current left cols and iv_attrs
new_iv_list = self._renew_iv_attrs(new_left_cols, self.left_cols,
self.guest_iv_attrs)
self.guest_iv_attrs = new_iv_list
self.left_cols = new_left_cols
# If host has participated, send result to host
if len(left_cols) > 1:
new_host_left_cols = left_cols[1]
new_host_iv_list = self._renew_iv_attrs(
new_host_left_cols, self.host_left_cols,
self.host_iv_attrs)
self.host_iv_attrs = new_host_iv_list
self.host_left_cols = new_host_left_cols
self._send_host_result_cols()
LOGGER.info(
"Finish iv percentile filter. Current left cols are: {}".
format(self.left_cols))
if method == consts.COEFFICIENT_OF_VARIATION_VALUE_THRES:
coe_param = self.params.coe_param
coe_filter = feature_selection.CoeffOfVarValueFilter(
coe_param, self.left_cols, self.static_obj)
self.left_cols = coe_filter.filter(data_instances)
self.static_obj = coe_filter.statics_obj
self.results.append(coe_filter.to_result())
LOGGER.info(
"Finish coeffiecient_of_variation value threshold filter. Current left cols are: {}"
.format(self.left_cols))
if method == consts.UNIQUE_VALUE:
unique_param = self.params.unique_param
unique_filter = feature_selection.UniqueValueFilter(
unique_param, self.left_cols, self.static_obj)
self.left_cols = unique_filter.filter(data_instances)
self.static_obj = unique_filter.statics_obj
self.results.append(unique_filter.to_result())
LOGGER.info(
"Finish unique value filter. Current left cols are: {}".format(
self.left_cols))
if method == consts.OUTLIER_COLS:
outlier_param = self.params.outlier_param
outlier_filter = feature_selection.OutlierFilter(
outlier_param, self.left_cols)
self.left_cols = outlier_filter.filter(data_instances)
self.results.append(outlier_filter.to_result())
LOGGER.info(
"Finish outlier cols filter. Current left cols are: {}".format(
self.left_cols))
def filter_one_method(self, data_instances, method):
if method == consts.IV_VALUE_THRES:
iv_param = self.model_param.iv_value_param
if not self.local_only:
host_select_cols = self._get_host_select_cols(
consts.IV_VALUE_THRES)
LOGGER.debug("In iv value filter, host_select_cols: {}".format(
host_select_cols))
iv_filter = feature_selection.IVValueSelectFilter(
iv_param,
self.filter_result.this_to_select_cols_index,
self.binning_model,
host_select_cols=host_select_cols)
new_left_cols = iv_filter.fit(data_instances, fit_host=True)
# self._renew_final_left_cols(new_left_cols)
self.filter_result.add_left_col_index(new_left_cols)
host_left_cols = iv_filter.host_cols
left_cols = host_left_cols.get(consts.HOST)
left_cols = {int(k): v for k, v in left_cols.items()}
self.host_filter_result.add_left_cols(left_cols)
LOGGER.debug(
"In Guest IV filter, host_select_cols: {}, host_left_cols: {}"
.format(host_select_cols, host_left_cols))
# new_result = {}
# for host_col_idx, _ in host_select_cols.items():
# host_col_idx = int(host_col_idx)
# is_left = left_cols.get(host_col_idx)
# new_result[host_col_idx] = is_left
# self.host_left_cols = new_result
# self._add_host_left_cols(self.host_left_cols)
self._send_host_result_cols(consts.IV_VALUE_THRES)
LOGGER.debug(
"[Result][FeatureSelection][Guest] Finish iv value threshold filter. Host left cols are: {}"
.format(self.host_filter_result.get_left_cols()))
else:
iv_filter = feature_selection.IVValueSelectFilter(
iv_param, self.filter_result.this_to_select_cols_index,
self.binning_model)
new_left_cols = iv_filter.fit(data_instances)
# self._renew_final_left_cols(new_left_cols)
self.filter_result.add_left_col_index(new_left_cols)
LOGGER.debug(
"[Result][FeatureSelection][Guest] Finish iv value threshold filter. Self left cols are: {}"
.format(self.filter_result.get_left_cols()))
self.iv_value_meta = iv_filter.get_meta_obj()
self.results.append(iv_filter.get_param_obj())
# self._renew_left_col_names()
if method == consts.IV_PERCENTILE:
iv_param = self.model_param.iv_percentile_param
if self.local_only:
iv_filter = feature_selection.IVPercentileFilter(
iv_param, self.filter_result.this_to_select_cols_index, {},
self.binning_model)
new_left_cols = iv_filter.fit(data_instances)
# self._renew_final_left_cols(new_left_cols)
self.filter_result.add_left_col_index(new_left_cols)
else:
host_select_cols = self._get_host_select_cols(
consts.IV_PERCENTILE)
host_cols = {consts.HOST: host_select_cols}
iv_filter = feature_selection.IVPercentileFilter(
iv_param, self.filter_result.this_to_select_cols_index,
host_cols, self.binning_model)
new_left_cols = iv_filter.fit(data_instances)
# self._renew_final_left_cols(new_left_cols)
self.filter_result.add_left_col_index(new_left_cols)
host_left_cols = iv_filter.host_cols
# Only one host
left_col_index = host_left_cols.get(consts.HOST)
self.host_filter_result.add_left_cols(left_col_index)
self._send_host_result_cols(consts.IV_PERCENTILE)
LOGGER.info(
"[Result][FeatureSelection][Host]Finish iv percentile threshold filter. "
"Host left cols are: {}".format(
self.host_filter_result.get_left_cols()))
LOGGER.debug(
"[Result][FeatureSelection][Guest]Finish iv percentile threshold filter. Self left cols are: {}"
.format(self.filter_result.get_left_cols()))
self.iv_percentile_meta = iv_filter.get_meta_obj()
self.results.append(iv_filter.get_param_obj())
# self._renew_left_col_names()
if method == consts.COEFFICIENT_OF_VARIATION_VALUE_THRES:
variance_coe_param = self.model_param.variance_coe_param
coe_filter = feature_selection.CoeffOfVarValueFilter(
variance_coe_param,
self.filter_result.this_to_select_cols_index, self.static_obj)
new_left_cols = coe_filter.fit(data_instances)
# self._renew_final_left_cols(new_left_cols)
self.filter_result.add_left_col_index(new_left_cols)
self.static_obj = coe_filter.statics_obj
LOGGER.info(
"[Result][FeatureSelection][Guest]Finish coefficient of variance filter. Self left cols are: {}"
.format(self.filter_result.get_left_cols()))
self.variance_coe_meta = coe_filter.get_meta_obj()
self.results.append(coe_filter.get_param_obj())
# self._renew_left_col_names()
if method == consts.UNIQUE_VALUE:
unique_param = self.model_param.unique_param
unique_filter = feature_selection.UniqueValueFilter(
unique_param, self.filter_result.this_to_select_cols_index,
self.static_obj)
new_left_cols = unique_filter.fit(data_instances)
# self._renew_final_left_cols(new_left_cols)
self.filter_result.add_left_col_index(new_left_cols)
self.static_obj = unique_filter.statics_obj
self.unique_meta = unique_filter.get_meta_obj()
self.results.append(unique_filter.get_param_obj())
LOGGER.info(
"[Result][FeatureSelection][Guest]Finish unique value filter. Current left cols are: {}"
.format(self.filter_result.get_left_cols()))
if method == consts.OUTLIER_COLS:
outlier_param = self.model_param.outlier_param
outlier_filter = feature_selection.OutlierFilter(
outlier_param, self.filter_result.this_to_select_cols_index)
new_left_cols = outlier_filter.fit(data_instances)
# self._renew_final_left_cols(new_left_cols)
self.filter_result.add_left_col_index(new_left_cols)
self.outlier_meta = outlier_filter.get_meta_obj()
self.results.append(outlier_filter.get_param_obj())
LOGGER.info(
"[Result][FeatureSelection][Guest]Finish outlier filter. Self left cols are: {}"
.format(self.filter_result.get_left_cols()))
