76 lines
3.9 KiB
Python
76 lines
3.9 KiB
Python
import pandas as pd
|
|
from scipy.stats import entropy
|
|
|
|
|
|
def statsFeatures(ibi_data, features, ibi_features):
|
|
col_name = "inter_beat_interval"
|
|
if "sumibi" in features:
|
|
ibi_features["sumibi"] = ibi_data[["local_segment", col_name]].groupby(["local_segment"])[
|
|
col_name].sum()
|
|
if "maxibi" in features:
|
|
ibi_features["maxibi"] = ibi_data[["local_segment", col_name]].groupby(["local_segment"])[
|
|
col_name].max()
|
|
if "minibi" in features:
|
|
ibi_features["minibi"] = ibi_data[["local_segment", col_name]].groupby(["local_segment"])[
|
|
col_name].min()
|
|
if "avgibi" in features:
|
|
ibi_features["avgibi"] = ibi_data[["local_segment", col_name]].groupby(["local_segment"])[
|
|
col_name].mean()
|
|
if "medianibi" in features:
|
|
ibi_features["medianibi"] = ibi_data[["local_segment", col_name]].groupby(["local_segment"])[
|
|
col_name].median()
|
|
if "modeibi" in features:
|
|
ibi_features["modeibi"] = ibi_data[["local_segment", col_name]].groupby(["local_segment"])[
|
|
col_name].agg(lambda x: pd.Series.mode(x)[0])
|
|
if "stdibi" in features:
|
|
ibi_features["stdibi"] = ibi_data[["local_segment", col_name]].groupby(["local_segment"])[
|
|
col_name].std()
|
|
if "diffmaxmodeibi" in features:
|
|
ibi_features["diffmaxmodeibi"] = ibi_data[["local_segment", col_name]].groupby(["local_segment"])[
|
|
col_name].max() - \
|
|
ibi_data[["local_segment", col_name]].groupby(["local_segment"])[
|
|
col_name].agg(lambda x: pd.Series.mode(x)[0])
|
|
if "diffminmodeibi" in features:
|
|
ibi_features["diffminmodeibi"] = ibi_data[["local_segment", col_name]].groupby(["local_segment"])[
|
|
col_name].agg(lambda x: pd.Series.mode(x)[0]) - \
|
|
ibi_data[["local_segment", col_name]].groupby(["local_segment"])[
|
|
col_name].min()
|
|
if "entropyibi" in features:
|
|
ibi_features["entropyibi"] = ibi_data[["local_segment", col_name]].groupby(["local_segment"])[
|
|
col_name].agg(entropy)
|
|
|
|
return ibi_features
|
|
|
|
|
|
def extractIBIFeaturesFromIntradayData(ibi_intraday_data, features, time_segment, filter_data_by_segment):
|
|
ibi_intraday_features = pd.DataFrame(columns=["local_segment"] + features)
|
|
if not ibi_intraday_data.empty:
|
|
ibi_intraday_data = filter_data_by_segment(ibi_intraday_data, time_segment)
|
|
|
|
if not ibi_intraday_data.empty:
|
|
ibi_intraday_features = pd.DataFrame()
|
|
|
|
# get stats of ibi
|
|
ibi_intraday_features = statsFeatures(ibi_intraday_data, features, ibi_intraday_features)
|
|
|
|
ibi_intraday_features.reset_index(inplace=True)
|
|
|
|
return ibi_intraday_features
|
|
|
|
|
|
def dbdp_features(sensor_data_files, time_segment, provider, filter_data_by_segment, *args, **kwargs):
|
|
ibi_intraday_data = pd.read_csv(sensor_data_files["sensor_data"])
|
|
|
|
requested_intraday_features = provider["FEATURES"]
|
|
# name of the features this function can compute
|
|
base_intraday_features_names = ["maxibi", "minibi", "avgibi", "medianibi", "modeibi", "stdibi", "diffmaxmodeibi",
|
|
"diffminmodeibi", "entropyibi"]
|
|
# the subset of requested features this function can compute
|
|
intraday_features_to_compute = list(set(requested_intraday_features) & set(base_intraday_features_names))
|
|
|
|
# extract features from intraday data
|
|
ibi_intraday_features = extractIBIFeaturesFromIntradayData(ibi_intraday_data,
|
|
intraday_features_to_compute, time_segment,
|
|
filter_data_by_segment)
|
|
|
|
return ibi_intraday_features |