Add screen episode filter parameters

config.yaml

@@ -91,7 +91,7 @@ DORYAB_LOCATION:
   DBSCAN_MINSAMPLES: 5
   THRESHOLD_STATIC : 1 # km/h
   MAXIMUM_GAP_ALLOWED: 300
-  MINUTES_DATA_USED: True
+  MINUTES_DATA_USED: False
 
 BLUETOOTH:
   COMPUTE: False
@@ -118,6 +118,8 @@ SCREEN:
   DB_TABLE: screen
   DAY_SEGMENTS: *day_segments
   REFERENCE_HOUR_FIRST_USE: 0
+  IGNORE_EPISODES_SHORTER_THAN: 0 # in minutes, set to 0 to disable
+  IGNORE_EPISODES_LONGER_THAN: 0 # in minutes, set to 0 to disable
   FEATURES_DELTAS: ["countepisode", "episodepersensedminutes", "sumduration", "maxduration", "minduration", "avgduration", "stdduration", "firstuseafter"]
   EPISODE_TYPES: ["unlock"]
 

docs/features/extracted.rst

@@ -740,14 +740,16 @@ See `Screen Config Code`_
 
 **Screen Rule Parameters (screen_features):**
 
-=========================    ===================
+============================    ===================
-Name	                     Description
+Name	                        Description
-=========================    ===================
+============================    ===================
-day_segment                  The particular ``day_segments`` that will be analyzed. The available options are ``daily``, ``morning``, ``afternoon``, ``evening``, ``night``
+day_segment                     The particular ``day_segments`` that will be analyzed. The available options are ``daily``, ``morning``, ``afternoon``, ``evening``, ``night``
-reference_hour_first_use     The reference point from which ``firstuseafter`` is to be computed, default is midnight
+reference_hour_first_use        The reference point from which ``firstuseafter`` is to be computed, default is midnight
-features_deltas              Features to be computed, see table below
+ignore_episodes_shorter_than    Ignore episodes that are shorter than this threshold (minutes). Set to 0 to disable this filter.
-episode_types                Currently we only support unlock episodes (from when the phone is unlocked until the screen is off)
+ignore_episodes_longer_than     Ignore episodes that are longer than this threshold (minutes). Set to 0 to disable this filter.
-=========================    ===================
+features_deltas                 Features to be computed, see table below
+episode_types                   Currently we only support unlock episodes (from when the phone is unlocked until the screen is off)
+============================    ===================
 
 .. _screen-episodes-available-features:
 

rules/features.snakefile

@@ -179,6 +179,8 @@ rule screen_features:
         reference_hour_first_use = config["SCREEN"]["REFERENCE_HOUR_FIRST_USE"],
         features_deltas = config["SCREEN"]["FEATURES_DELTAS"],
         episode_types = config["SCREEN"]["EPISODE_TYPES"],
+        ignore_episodes_shorter_than = config["SCREEN"]["IGNORE_EPISODES_SHORTER_THAN"],
+        ignore_episodes_longer_than = config["SCREEN"]["IGNORE_EPISODES_LONGER_THAN"],
         bin_size = config["PHONE_VALID_SENSED_BINS"]["BIN_SIZE"]
     output:
         "data/processed/{pid}/screen_{day_segment}.csv"

src/features/screen/screen_base.py

@@ -44,8 +44,10 @@ def base_screen_features(screen_data, phone_sensed_bins, day_segment, params):
 
     reference_hour_first_use = params["reference_hour_first_use"]
     bin_size = params["bin_size"]
-    requested_features_deltas = params["requested_features_deltas"]
+    requested_features_deltas = params["features_deltas"]
-    requested_episode_types = params["requested_episode_types"]
+    requested_episode_types = params["episode_types"]
+    ignore_episodes_shorter_than = params["ignore_episodes_shorter_than"]
+    ignore_episodes_longer_than = params["ignore_episodes_longer_than"]
 
     # name of the features this function can compute
     base_features_deltas = ["countepisode", "episodepersensedminutes", "sumduration", "maxduration", "minduration", "avgduration", "stdduration", "firstuseafter"]
@@ -65,6 +67,11 @@ def base_screen_features(screen_data, phone_sensed_bins, day_segment, params):
             screen_data = splitMultiSegmentEpisodes(screen_data, day_segment, [])
         screen_data.set_index(["local_start_date"],inplace=True)
 
+        if ignore_episodes_shorter_than > 0:
+            screen_data = screen_data.query('@ignore_episodes_shorter_than <= time_diff')
+        if ignore_episodes_longer_than > 0:
+            screen_data = screen_data.query('time_diff <= @ignore_episodes_longer_than')
+
         if not screen_data.empty:
             screen_features = pd.DataFrame()
             for episode in episode_type_to_compute:

src/features/screen_features.py

@@ -8,16 +8,10 @@ phone_sensed_bins[phone_sensed_bins > 0] = 1
 day_segment = snakemake.params["day_segment"]
 screen_features = pd.DataFrame(columns=["local_date"])
 
-params = {}
+requested_features_deltas = ["firstuseafter" + "{0:0=2d}".format(snakemake.params["reference_hour_first_use"]) if feature_name == "firstuseafter" else feature_name for feature_name in snakemake.params["features_deltas"]]
-params["reference_hour_first_use"] = snakemake.params["reference_hour_first_use"]
+requested_features = ["".join(feature) for feature in itertools.product(requested_features_deltas, snakemake.params["episode_types"])]
-params["bin_size"] = snakemake.params["bin_size"]
-params["requested_features_deltas"] = snakemake.params["features_deltas"]
-params["requested_episode_types"] = snakemake.params["episode_types"]
 
-requested_features_deltas = ["firstuseafter" + "{0:0=2d}".format(params["reference_hour_first_use"]) if feature_name == "firstuseafter" else feature_name for feature_name in params["requested_features_deltas"]]
+screen_features = screen_features.merge(base_screen_features(screen_data, phone_sensed_bins, day_segment, snakemake.params), on="local_date", how="outer")
-requested_features = ["".join(feature) for feature in itertools.product(requested_features_deltas, params["requested_episode_types"])]
-
-screen_features = screen_features.merge(base_screen_features(screen_data, phone_sensed_bins, day_segment, params), on="local_date", how="outer")
 
 assert len(requested_features) + 1 == screen_features.shape[1], "The number of features in the output dataframe (=" + str(screen_features.shape[1]) + ") does not match the expected value (=" + str(len(requested_features)) + " + 1). Verify your screen feature extraction functions"