Refactor light features

pull/95/head
Meng Li 2020-04-01 18:29:53 -04:00
parent ac9fb487a6
commit 1ce76a5380
2 changed files with 39 additions and 24 deletions

View File

@ -0,0 +1,34 @@
import pandas as pd
def base_light_features(light_data, day_segment, requested_features):
# name of the features this function can compute
base_features_names = ["count", "maxlux", "minlux", "avglux", "medianlux", "stdlux"]
# the subset of requested features this function can compute
features_to_compute = list(set(requested_features) & set(base_features_names))
if light_data.empty:
light_features = pd.DataFrame(columns=["local_date"] + ["light_" + day_segment + "_" + x for x in features_to_compute])
else:
if day_segment != "daily":
light_data =light_data[light_data["local_day_segment"] == day_segment]
if not light_data.empty:
light_features = pd.DataFrame()
if "count" in features_to_compute:
light_features["light_" + day_segment + "_count"] = light_data.groupby(["local_date"]).count()["timestamp"]
# get light ambient luminance related features
if "maxlux" in features_to_compute:
light_features["light_" + day_segment + "_maxlux"] = light_data.groupby(["local_date"])["double_light_lux"].max()
if "minlux" in features_to_compute:
light_features["light_" + day_segment + "_minlux"] = light_data.groupby(["local_date"])["double_light_lux"].min()
if "avglux" in features_to_compute:
light_features["light_" + day_segment + "_avglux"] = light_data.groupby(["local_date"])["double_light_lux"].mean()
if "medianlux" in features_to_compute:
light_features["light_" + day_segment + "_medianlux"] = light_data.groupby(["local_date"])["double_light_lux"].median()
if "stdlux" in features_to_compute:
light_features["light_" + day_segment + "_stdlux"] = light_data.groupby(["local_date"])["double_light_lux"].std()
light_features = light_features.reset_index()
return light_features

View File

@ -1,32 +1,13 @@
import pandas as pd
import numpy as np
from light.light_base import base_light_features
light_data = pd.read_csv(snakemake.input[0], parse_dates=["local_date_time", "local_date"])
day_segment = snakemake.params["day_segment"]
metrics = snakemake.params["metrics"]
light_features = pd.DataFrame(columns=["local_date"])
light_features = pd.DataFrame(columns=["local_date"] + ["light_" + day_segment + "_" + x for x in metrics])
if not light_data.empty:
if day_segment != "daily":
light_data =light_data[light_data["local_day_segment"] == day_segment]
light_features = light_features.merge(base_light_features(light_data, day_segment, metrics), on="local_date", how="outer")
if not light_data.empty:
light_features = pd.DataFrame()
if "count" in metrics:
light_features["light_" + day_segment + "_count"] = light_data.groupby(["local_date"]).count()["timestamp"]
# get light ambient luminance related features
if "maxlux" in metrics:
light_features["light_" + day_segment + "_maxlux"] = light_data.groupby(["local_date"])["double_light_lux"].max()
if "minlux" in metrics:
light_features["light_" + day_segment + "_minlux"] = light_data.groupby(["local_date"])["double_light_lux"].min()
if "avglux" in metrics:
light_features["light_" + day_segment + "_avglux"] = light_data.groupby(["local_date"])["double_light_lux"].mean()
if "medianlux" in metrics:
light_features["light_" + day_segment + "_medianlux"] = light_data.groupby(["local_date"])["double_light_lux"].median()
if "stdlux" in metrics:
light_features["light_" + day_segment + "_stdlux"] = light_data.groupby(["local_date"])["double_light_lux"].std()
light_features = light_features.reset_index()
assert len(metrics) + 1 == light_features.shape[1], "The number of features in the output dataframe (=" + str(light_features.shape[1]) + ") does not match the expected value (=" + str(len(metrics)) + " + 1). Verify your light feature extraction functions"
light_features.to_csv(snakemake.output[0], index=False)