def extract_features_with_param(time_series, window):
## type: (object, object) -> object
## type: (object, object) -> object
"""
Extracts three types of features from the time series.
:param time_series: the time series to extract the feature of
:type time_series: pandas.Series
:param window: the length of window
:type window: int
:return: the value of features
:return type: list with float
"""
# if not tsd_common.is_standard_time_series(time_series, window):
# # add your report of this error here...
#
# return []
# spilt time_series
split_time_series = tsd_common.split_time_series(time_series, window)
normalized_split_time_series = tsd_common.normalize_time_series(
split_time_series)
max_min_normalized_time_series = tsd_common.normalize_time_series_by_max_min(
split_time_series)
s_features_with_parameter1 = statistical_features.get_parameters_features(
max_min_normalized_time_series)
# s_features_with_parameter2 = statistical_features.get_parameters_features(normalized_split_time_series)
features = s_features_with_parameter1
return features
def extract_features(time_series, window):
"""
Extracts three types of features from the time series.
:param time_series: the time series to extract the feature of
:type time_series: pandas.Series
:param window: the length of window
:type window: int
:return: the value of features
:return type: list with float
"""
if not tsd_common.is_standard_time_series(time_series, window):
# add your report of this error here...
return []
# spilt time_series
split_time_series = tsd_common.split_time_series(time_series, window)
# nomalize time_series
normalized_split_time_series = tsd_common.normalize_time_series(split_time_series)
max_min_normalized_time_series = tsd_common.normalize_time_series_by_max_min(split_time_series)
s_features = statistical_features.get_statistical_features(normalized_split_time_series[4])
f_features = fitting_features.get_fitting_features(normalized_split_time_series)
c_features = classification_features.get_classification_features(max_min_normalized_time_series)
# combine features with types
features = s_features + f_features + c_features
return features
def extract_features(time_series, window):
"""
Extracts three types of features from the time series.
:param time_series: the time series to extract the feature of
:type time_series: pandas.Series
:param window: the length of window
:type window: int
:return: the value of features
:return type: list with float
"""
if not tsd_common.is_standard_time_series(time_series, window):
# add your report of this error here...
return []
# spilt time_series
split_time_series = tsd_common.split_time_series(time_series, window)
# nomalize time_series
normalized_split_time_series = tsd_common.normalize_time_series(
split_time_series)
max_min_normalized_time_series = tsd_common.normalize_time_series_by_max_min(
split_time_series)
s_features = statistical_features.get_statistical_features(
normalized_split_time_series[4])
f_features = fitting_features.get_fitting_features(
normalized_split_time_series)
c_features = classification_features.get_classification_features(
max_min_normalized_time_series)
# combine features with types
features = s_features + f_features + c_features
return features
def calculate_all_features(time_series, window):
"""
Extracts three types of features from the time series.
:param time_series: the time series to extract the feature of
:type time_series: pandas.Series
:param window: the length of window
:type window: int
:return: the value of features
:return type: list with float
"""
split_time_series = tsd_common.split_time_series(time_series, window)
normalized_split_time_series = tsd_common.normalize_time_series(
split_time_series)
max_min_normalized_time_series = tsd_common.normalize_time_series_by_max_min(
split_time_series)
# s_features = statistical_features.get_statistical_features(normalized_split_time_series[4])
# c_features = classification_features.get_classification_features(max_min_normalized_time_series)
# f_features = fitting_features.get_fitting_features(normalized_split_time_series)
# s_features_with_parameter1 = feature_calculate.get_parameters_features(max_min_normalized_time_series)
# features = s_features + c_features + f_features + s_features_with_parameter1
anom_feature = feature_calculate.get_classification_features_test(
normalized_split_time_series)
pattern_feature = feature_calculate.get_classification_feature_pattern(
max_min_normalized_time_series)
stat_feature = feature_calculate.get_classification_feature_stat(
max_min_normalized_time_series)
features = stat_feature + anom_feature + pattern_feature
return features
def extract_features_without_param(time_series, window):
## type: (object, object) -> object
## type: (object, object) -> object
"""
Extracts three types of features from the time series.
:param time_series: the time series to extract the feature of
:type time_series: pandas.Series
:param window: the length of window
:type window: int
:return: the value of features
:return type: list with float
"""
# if not tsd_common.is_standard_time_series(time_series, window):
# # add your report of this error here...
#
# return []
# spilt time_series
split_time_series = tsd_common.split_time_series(time_series, window)
split_time_series2 = tsd_common.split_time_series2(time_series, window)
# nomalize time_series
normalized_split_time_series = tsd_common.normalize_time_series(
split_time_series)
max_min_normalized_time_series = tsd_common.normalize_time_series_by_max_min(
split_time_series)
s_features = statistical_features.get_statistical_features(
normalized_split_time_series[4])
f_features = fitting_features.get_fitting_features(
normalized_split_time_series)
c_features = classification_features.get_classification_features(
max_min_normalized_time_series)
# combine features with types
# s_features_without_parameter = statistical_features.calculate_nonparameters_features(normalized_split_time_series[4])
# s_features_with_parameter = statistical_features.get_parameters_features(normalized_split_time_series[4])
# s_features_with_parameter = statistical_features.get_parameters_features(time_series)
# features = c_features
# return s_features_with_parameter
features = s_features + c_features + f_features
# features = c_features
return features