2021-03-11 20:57:34 +01:00
########################################################################################################################
# GLOBAL CONFIGURATION #
########################################################################################################################
2019-11-05 21:17:20 +01:00
2020-11-26 01:42:11 +01:00
# See https://www.rapids.science/latest/setup/configuration/#participant-files
2022-02-23 11:10:54 +01:00
PIDS : [ 'p031' ]
2020-02-10 22:45:34 +01:00
2020-11-26 01:42:11 +01:00
# See https://www.rapids.science/latest/setup/configuration/#automatic-creation-of-participant-files
2020-10-27 22:13:16 +01:00
CREATE_PARTICIPANT_FILES :
2021-12-01 18:14:27 +01:00
USERNAMES_CSV : "data/external/main_study_usernames.csv"
CSV_FILE_PATH : "data/external/main_study_participants.csv" # see docs for required format
2020-10-21 01:12:01 +02:00
PHONE_SECTION :
2021-03-11 20:51:16 +01:00
ADD : True
2020-10-21 01:12:01 +02:00
IGNORED_DEVICE_IDS : [ ]
FITBIT_SECTION :
2021-12-01 18:14:27 +01:00
ADD : False
2020-10-27 22:13:16 +01:00
IGNORED_DEVICE_IDS : [ ]
2021-02-21 23:30:30 +01:00
EMPATICA_SECTION :
2021-12-01 18:14:27 +01:00
ADD : False
2021-03-11 20:30:42 +01:00
IGNORED_DEVICE_IDS : [ ]
2020-10-27 22:13:16 +01:00
2020-12-03 00:41:03 +01:00
# See https://www.rapids.science/latest/setup/configuration/#time-segments
TIME_SEGMENTS : &time_segments
2020-11-25 01:12:16 +01:00
TYPE : PERIODIC # FREQUENCY, PERIODIC, EVENT
2021-12-08 18:20:22 +01:00
FILE : "data/external/timesegments_daily.csv"
INCLUDE_PAST_PERIODIC_SEGMENTS : TRUE # Only relevant if TYPE=PERIODIC, see docs
2020-02-10 22:45:34 +01:00
2021-03-11 19:23:56 +01:00
# See https://www.rapids.science/latest/setup/configuration/#timezone-of-your-study
2021-03-05 23:49:37 +01:00
TIMEZONE :
2021-12-08 18:21:29 +01:00
TYPE : MULTIPLE
2021-03-05 23:49:37 +01:00
SINGLE :
2021-11-17 18:53:44 +01:00
TZCODE : Europe/Ljubljana
2021-03-05 23:49:37 +01:00
MULTIPLE :
2021-12-15 18:09:30 +01:00
TZ_FILE : data/external/timezone.csv
2021-12-08 18:21:29 +01:00
TZCODES_FILE : data/external/multiple_timezones.csv
IF_MISSING_TZCODE : USE_DEFAULT
2021-11-17 18:53:44 +01:00
DEFAULT_TZCODE : Europe/Ljubljana
2021-12-08 18:59:10 +01:00
FITBIT :
ALLOW_MULTIPLE_TZ_PER_DEVICE : False
INFER_FROM_SMARTPHONE_TZ : False
2020-11-26 01:42:11 +01:00
########################################################################################################################
# PHONE #
########################################################################################################################
2021-03-14 16:40:04 +01:00
# See https://www.rapids.science/latest/setup/configuration/#data-stream-configuration
2021-03-08 21:58:26 +01:00
PHONE_DATA_STREAMS :
2021-11-17 18:53:44 +01:00
USE : aware_postgresql
2021-03-08 21:58:26 +01:00
# AVAILABLE:
aware_mysql :
DATABASE_GROUP : MY_GROUP
2021-11-17 18:53:44 +01:00
aware_postgresql :
DATABASE_GROUP : PSQL_STRAW
2021-03-12 01:32:11 +01:00
aware_csv :
FOLDER : data/external/aware_csv
2021-03-14 18:33:43 +01:00
aware_influxdb :
DATABASE_GROUP : MY_GROUP
2020-11-26 01:42:11 +01:00
# Sensors ------
# https://www.rapids.science/latest/features/phone-accelerometer/
PHONE_ACCELEROMETER :
2021-03-13 00:14:49 +01:00
CONTAINER : accelerometer
2020-11-25 01:12:16 +01:00
PROVIDERS :
RAPIDS :
2022-01-07 12:20:51 +01:00
COMPUTE : False
2020-11-26 01:42:11 +01:00
FEATURES : [ "maxmagnitude" , "minmagnitude" , "avgmagnitude" , "medianmagnitude" , "stdmagnitude" ]
2021-03-15 00:52:14 +01:00
SRC_SCRIPT : src/features/phone_accelerometer/rapids/main.py
2020-11-26 01:42:11 +01:00
PANDA :
2022-01-07 12:20:51 +01:00
COMPUTE : False
2020-11-26 01:42:11 +01:00
VALID_SENSED_MINUTES : False
FEATURES :
exertional_activity_episode : [ "sumduration" , "maxduration" , "minduration" , "avgduration" , "medianduration" , "stdduration" ]
nonexertional_activity_episode : [ "sumduration" , "maxduration" , "minduration" , "avgduration" , "medianduration" , "stdduration" ]
2021-03-15 00:52:14 +01:00
SRC_SCRIPT : src/features/phone_accelerometer/panda/main.py
2020-07-09 19:01:50 +02:00
2020-11-26 01:42:11 +01:00
# See https://www.rapids.science/latest/features/phone-activity-recognition/
PHONE_ACTIVITY_RECOGNITION :
2021-03-13 00:14:49 +01:00
CONTAINER :
2021-12-08 18:37:34 +01:00
ANDROID : google_ar
2020-11-26 01:42:11 +01:00
IOS : plugin_ios_activity_recognition
2021-02-22 01:34:42 +01:00
EPISODE_THRESHOLD_BETWEEN_ROWS : 5 # minutes. Max time difference for two consecutive rows to be considered within the same AR episode.
2020-08-31 19:34:15 +02:00
PROVIDERS :
RAPIDS :
2021-12-08 18:37:34 +01:00
COMPUTE : True
2020-11-26 01:42:11 +01:00
FEATURES : [ "count" , "mostcommonactivity" , "countuniqueactivities" , "durationstationary" , "durationmobile" , "durationvehicle" ]
ACTIVITY_CLASSES :
STATIONARY : [ "still" , "tilting" ]
MOBILE : [ "on_foot" , "walking" , "running" , "on_bicycle" ]
VEHICLE : [ "in_vehicle" ]
2021-03-15 00:52:14 +01:00
SRC_SCRIPT : src/features/phone_activity_recognition/rapids/main.py
2020-11-26 01:42:11 +01:00
2021-12-08 18:59:10 +01:00
# See https://www.rapids.science/latest/features/phone-applications-crashes/
PHONE_APPLICATIONS_CRASHES :
CONTAINER : applications_crashes
APPLICATION_CATEGORIES :
CATALOGUE_SOURCE : FILE # FILE (genres are read from CATALOGUE_FILE) or GOOGLE (genres are scrapped from the Play Store)
CATALOGUE_FILE : "data/external/stachl_application_genre_catalogue.csv"
UPDATE_CATALOGUE_FILE : False # if CATALOGUE_SOURCE is equal to FILE, whether or not to update CATALOGUE_FILE, if CATALOGUE_SOURCE is equal to GOOGLE all scraped genres will be saved to CATALOGUE_FILE
SCRAPE_MISSING_CATEGORIES : False # whether or not to scrape missing genres, only effective if CATALOGUE_SOURCE is equal to FILE. If CATALOGUE_SOURCE is equal to GOOGLE, all genres are scraped anyway
PROVIDERS : # None implemented yet but this sensor can be used in PHONE_DATA_YIELD
2020-11-26 01:42:11 +01:00
# See https://www.rapids.science/latest/features/phone-applications-foreground/
PHONE_APPLICATIONS_FOREGROUND :
2021-12-08 18:37:34 +01:00
CONTAINER : applications
2020-11-26 01:42:11 +01:00
APPLICATION_CATEGORIES :
CATALOGUE_SOURCE : FILE # FILE (genres are read from CATALOGUE_FILE) or GOOGLE (genres are scrapped from the Play Store)
CATALOGUE_FILE : "data/external/stachl_application_genre_catalogue.csv"
2021-12-15 17:26:54 +01:00
PACKAGE_NAMES_HASHED : True
2020-11-26 01:42:11 +01:00
UPDATE_CATALOGUE_FILE : False # if CATALOGUE_SOURCE is equal to FILE, whether or not to update CATALOGUE_FILE, if CATALOGUE_SOURCE is equal to GOOGLE all scraped genres will be saved to CATALOGUE_FILE
SCRAPE_MISSING_CATEGORIES : False # whether or not to scrape missing genres, only effective if CATALOGUE_SOURCE is equal to FILE. If CATALOGUE_SOURCE is equal to GOOGLE, all genres are scraped anyway
PROVIDERS :
RAPIDS :
2021-12-08 18:37:34 +01:00
COMPUTE : True
INCLUDE_EPISODE_FEATURES : True
2020-11-26 01:42:11 +01:00
SINGLE_CATEGORIES : [ "all" , "email" ]
MULTIPLE_CATEGORIES :
social : [ "socialnetworks" , "socialmediatools" ]
entertainment : [ "entertainment" , "gamingknowledge" , "gamingcasual" , "gamingadventure" , "gamingstrategy" , "gamingtoolscommunity" , "gamingroleplaying" , "gamingaction" , "gaminglogic" , "gamingsports" , "gamingsimulation" ]
2021-07-02 00:08:33 +02:00
CUSTOM_CATEGORIES :
social_media : [ "com.google.android.youtube" , "com.snapchat.android" , "com.instagram.android" , "com.zhiliaoapp.musically" , "com.facebook.katana" ]
dating : [ "com.tinder" , "com.relance.happycouple" , "com.kiwi.joyride" ]
2020-11-26 01:42:11 +01:00
SINGLE_APPS : [ "top1global" , "com.facebook.moments" , "com.google.android.youtube" , "com.twitter.android" ] # There's no entropy for single apps
EXCLUDED_CATEGORIES : [ ]
2021-12-08 18:37:34 +01:00
EXCLUDED_APPS : [ "com.fitbit.FitbitMobile" , "com.aware.plugin.upmc.cancer" ] # TODO list system apps?
2021-04-13 22:45:32 +02:00
FEATURES :
2021-07-02 00:08:33 +02:00
APP_EVENTS : [ "countevent" , "timeoffirstuse" , "timeoflastuse" , "frequencyentropy" ]
APP_EPISODES : [ "countepisode" , "minduration" , "maxduration" , "meanduration" , "sumduration" ]
2021-04-13 22:45:32 +02:00
IGNORE_EPISODES_SHORTER_THAN : 0 # in minutes, set to 0 to disable
IGNORE_EPISODES_LONGER_THAN : 300 # in minutes, set to 0 to disable
2021-03-15 00:52:14 +01:00
SRC_SCRIPT : src/features/phone_applications_foreground/rapids/main.py
2020-11-26 01:42:11 +01:00
2021-01-13 23:08:19 +01:00
# See https://www.rapids.science/latest/features/phone-applications-notifications/
PHONE_APPLICATIONS_NOTIFICATIONS :
2021-12-08 18:37:34 +01:00
CONTAINER : notifications
2021-01-14 01:05:36 +01:00
APPLICATION_CATEGORIES :
CATALOGUE_SOURCE : FILE # FILE (genres are read from CATALOGUE_FILE) or GOOGLE (genres are scrapped from the Play Store)
CATALOGUE_FILE : "data/external/stachl_application_genre_catalogue.csv"
UPDATE_CATALOGUE_FILE : False # if CATALOGUE_SOURCE is equal to FILE, whether or not to update CATALOGUE_FILE, if CATALOGUE_SOURCE is equal to GOOGLE all scraped genres will be saved to CATALOGUE_FILE
SCRAPE_MISSING_CATEGORIES : False # whether or not to scrape missing genres, only effective if CATALOGUE_SOURCE is equal to FILE. If CATALOGUE_SOURCE is equal to GOOGLE, all genres are scraped anyway
2021-01-13 23:08:19 +01:00
PROVIDERS : # None implemented yet but this sensor can be used in PHONE_DATA_YIELD
2020-11-26 01:42:11 +01:00
# See https://www.rapids.science/latest/features/phone-battery/
PHONE_BATTERY :
2021-03-13 00:14:49 +01:00
CONTAINER : battery
2020-11-26 01:42:11 +01:00
EPISODE_THRESHOLD_BETWEEN_ROWS : 30 # minutes. Max time difference for two consecutive rows to be considered within the same battery episode.
PROVIDERS :
RAPIDS :
2021-12-08 18:37:34 +01:00
COMPUTE : True
2020-11-26 01:42:11 +01:00
FEATURES : [ "countdischarge" , "sumdurationdischarge" , "countcharge" , "sumdurationcharge" , "avgconsumptionrate" , "maxconsumptionrate" ]
2021-03-15 00:52:14 +01:00
SRC_SCRIPT : src/features/phone_battery/rapids/main.py
2020-11-26 01:42:11 +01:00
# See https://www.rapids.science/latest/features/phone-bluetooth/
PHONE_BLUETOOTH :
2021-03-13 00:14:49 +01:00
CONTAINER : bluetooth
2020-11-26 01:42:11 +01:00
PROVIDERS :
RAPIDS :
2021-12-08 18:37:34 +01:00
COMPUTE : True
2020-11-26 01:42:11 +01:00
FEATURES : [ "countscans" , "uniquedevices" , "countscansmostuniquedevice" ]
2021-03-15 00:52:14 +01:00
SRC_SCRIPT : src/features/phone_bluetooth/rapids/main.R
2020-12-11 18:03:22 +01:00
DORYAB :
2021-12-08 18:37:34 +01:00
COMPUTE : True
2020-12-11 18:03:22 +01:00
FEATURES :
2020-12-12 23:01:46 +01:00
ALL :
DEVICES : [ "countscans" , "uniquedevices" , "meanscans" , "stdscans" ]
SCANS_MOST_FREQUENT_DEVICE : [ "withinsegments" , "acrosssegments" , "acrossdataset" ]
SCANS_LEAST_FREQUENT_DEVICE : [ "withinsegments" , "acrosssegments" , "acrossdataset" ]
OWN :
DEVICES : [ "countscans" , "uniquedevices" , "meanscans" , "stdscans" ]
SCANS_MOST_FREQUENT_DEVICE : [ "withinsegments" , "acrosssegments" , "acrossdataset" ]
SCANS_LEAST_FREQUENT_DEVICE : [ "withinsegments" , "acrosssegments" , "acrossdataset" ]
OTHERS :
DEVICES : [ "countscans" , "uniquedevices" , "meanscans" , "stdscans" ]
SCANS_MOST_FREQUENT_DEVICE : [ "withinsegments" , "acrosssegments" , "acrossdataset" ]
SCANS_LEAST_FREQUENT_DEVICE : [ "withinsegments" , "acrosssegments" , "acrossdataset" ]
2021-03-15 00:52:14 +01:00
SRC_SCRIPT : src/features/phone_bluetooth/doryab/main.py
2019-11-06 21:38:08 +01:00
2020-11-26 01:42:11 +01:00
# See https://www.rapids.science/latest/features/phone-calls/
2020-10-19 21:07:12 +02:00
PHONE_CALLS :
2021-11-17 18:53:44 +01:00
CONTAINER : call
2020-08-28 23:50:49 +02:00
PROVIDERS :
RAPIDS :
2021-11-17 18:53:44 +01:00
COMPUTE : True
2021-09-02 00:54:39 +02:00
FEATURES_TYPE : EPISODES # EVENTS or EPISODES
2020-08-28 23:50:49 +02:00
CALL_TYPES : [ missed, incoming, outgoing]
FEATURES :
missed : [ count, distinctcontacts, timefirstcall, timelastcall, countmostfrequentcontact]
incoming : [ count, distinctcontacts, meanduration, sumduration, minduration, maxduration, stdduration, modeduration, entropyduration, timefirstcall, timelastcall, countmostfrequentcontact]
outgoing : [ count, distinctcontacts, meanduration, sumduration, minduration, maxduration, stdduration, modeduration, entropyduration, timefirstcall, timelastcall, countmostfrequentcontact]
2021-03-15 00:52:14 +01:00
SRC_SCRIPT : src/features/phone_calls/rapids/main.R
2019-11-05 18:34:22 +01:00
2020-11-26 01:42:11 +01:00
# See https://www.rapids.science/latest/features/phone-conversation/
2021-12-08 18:37:34 +01:00
PHONE_CONVERSATION : # TODO Adapt for speech
2021-03-13 00:14:49 +01:00
CONTAINER :
2020-11-26 01:42:11 +01:00
ANDROID : plugin_studentlife_audio_android
IOS : plugin_studentlife_audio
PROVIDERS :
RAPIDS :
COMPUTE : False
FEATURES : [ "minutessilence" , "minutesnoise" , "minutesvoice" , "minutesunknown" , "sumconversationduration" , "avgconversationduration" ,
"sdconversationduration" , "minconversationduration" , "maxconversationduration" , "timefirstconversation" , "timelastconversation" , "noisesumenergy" ,
"noiseavgenergy" , "noisesdenergy" , "noiseminenergy" , "noisemaxenergy" , "voicesumenergy" ,
"voiceavgenergy" , "voicesdenergy" , "voiceminenergy" , "voicemaxenergy" , "silencesensedfraction" , "noisesensedfraction" ,
"voicesensedfraction" , "unknownsensedfraction" , "silenceexpectedfraction" , "noiseexpectedfraction" , "voiceexpectedfraction" ,
"unknownexpectedfraction" , "countconversation" ]
RECORDING_MINUTES : 1
PAUSED_MINUTES : 3
2021-03-15 00:52:14 +01:00
SRC_SCRIPT : src/features/phone_conversation/rapids/main.py
2020-11-26 01:42:11 +01:00
# See https://www.rapids.science/latest/features/phone-data-yield/
PHONE_DATA_YIELD :
2022-01-07 12:20:51 +01:00
SENSORS : [ #PHONE_ACCELEROMETER,
2021-12-08 19:04:19 +01:00
PHONE_ACTIVITY_RECOGNITION,
PHONE_APPLICATIONS_FOREGROUND,
PHONE_APPLICATIONS_NOTIFICATIONS,
PHONE_BATTERY,
PHONE_BLUETOOTH,
PHONE_CALLS,
PHONE_LIGHT,
PHONE_LOCATIONS,
PHONE_MESSAGES,
PHONE_SCREEN,
PHONE_WIFI_VISIBLE]
2020-11-26 01:42:11 +01:00
PROVIDERS :
RAPIDS :
2021-12-08 18:37:34 +01:00
COMPUTE : True
2020-11-26 01:42:11 +01:00
FEATURES : [ ratiovalidyieldedminutes, ratiovalidyieldedhours]
2021-02-22 01:34:42 +01:00
MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS : 0.5 # 0 to 1, minimum percentage of valid minutes in an hour to be considered valid.
2021-03-15 00:52:14 +01:00
SRC_SCRIPT : src/features/phone_data_yield/rapids/main.R
2020-11-26 01:42:11 +01:00
2021-12-08 18:59:10 +01:00
# See https://www.rapids.science/latest/features/phone-keyboard/
PHONE_KEYBOARD :
CONTAINER : keyboard
PROVIDERS :
RAPIDS :
COMPUTE : False
FEATURES : [ "sessioncount" , "averageinterkeydelay" , "averagesessionlength" , "changeintextlengthlessthanminusone" , "changeintextlengthequaltominusone" , "changeintextlengthequaltoone" , "changeintextlengthmorethanone" , "maxtextlength" , "lastmessagelength" , "totalkeyboardtouches" ]
SRC_SCRIPT : src/features/phone_keyboard/rapids/main.py
2020-11-26 01:42:11 +01:00
# See https://www.rapids.science/latest/features/phone-light/
PHONE_LIGHT :
2021-12-08 20:02:47 +01:00
CONTAINER : light_sensor
2020-11-26 01:42:11 +01:00
PROVIDERS :
RAPIDS :
2021-12-08 18:37:34 +01:00
COMPUTE : True
2020-11-26 01:42:11 +01:00
FEATURES : [ "count" , "maxlux" , "minlux" , "avglux" , "medianlux" , "stdlux" ]
2021-03-15 00:52:14 +01:00
SRC_SCRIPT : src/features/phone_light/rapids/main.py
2020-11-26 01:42:11 +01:00
# See https://www.rapids.science/latest/features/phone-locations/
2020-10-19 21:07:12 +02:00
PHONE_LOCATIONS :
2021-03-13 00:14:49 +01:00
CONTAINER : locations
2021-01-14 20:27:35 +01:00
LOCATIONS_TO_USE : ALL_RESAMPLED # ALL, GPS, ALL_RESAMPLED, OR FUSED_RESAMPLED
2020-08-28 19:53:00 +02:00
FUSED_RESAMPLED_CONSECUTIVE_THRESHOLD : 30 # minutes, only replicate location samples to the next sensed bin if the phone did not stop collecting data for more than this threshold
FUSED_RESAMPLED_TIME_SINCE_VALID_LOCATION : 720 # minutes, only replicate location samples to consecutive sensed bins if they were logged within this threshold after a valid location row
2021-09-15 20:28:09 +02:00
ACCURACY_LIMIT : 100 # meters, drops location coordinates with an accuracy equal or higher than this. This number means there's a 68% probability the true location is within this radius
2020-08-28 19:53:00 +02:00
PROVIDERS :
DORYAB :
2021-12-08 18:37:34 +01:00
COMPUTE : True
2021-04-09 18:05:25 +02:00
FEATURES : [ "locationvariance" , "loglocationvariance" , "totaldistance" , "avgspeed" , "varspeed" , "numberofsignificantplaces" , "numberlocationtransitions" , "radiusgyration" , "timeattop1location" , "timeattop2location" , "timeattop3location" , "movingtostaticratio" , "outlierstimepercent" , "maxlengthstayatclusters" , "minlengthstayatclusters" , "avglengthstayatclusters" , "stdlengthstayatclusters" , "locationentropy" , "normalizedlocationentropy" , "timeathome" , "homelabel" ]
2021-03-19 17:04:12 +01:00
DBSCAN_EPS : 100 # meters
2020-08-28 19:53:00 +02:00
DBSCAN_MINSAMPLES : 5
THRESHOLD_STATIC : 1 # km/h
2021-04-09 18:05:25 +02:00
MAXIMUM_ROW_GAP : 300 # seconds
2020-08-28 23:50:49 +02:00
MINUTES_DATA_USED : False
2021-04-09 18:05:25 +02:00
CLUSTER_ON : PARTICIPANT_DATASET # PARTICIPANT_DATASET, TIME_SEGMENT, TIME_SEGMENT_INSTANCE
INFER_HOME_LOCATION_STRATEGY : DORYAB_STRATEGY # DORYAB_STRATEGY, SUN_LI_VEGA_STRATEGY
MINIMUM_DAYS_TO_DETECT_HOME_CHANGES : 3
CLUSTERING_ALGORITHM : DBSCAN # DBSCAN, OPTICS
RADIUS_FOR_HOME : 100
2021-03-15 00:52:14 +01:00
SRC_SCRIPT : src/features/phone_locations/doryab/main.py
2020-08-28 19:53:00 +02:00
BARNETT :
2021-12-08 18:37:34 +01:00
COMPUTE : True
2020-08-28 19:53:00 +02:00
FEATURES : [ "hometime" , "disttravelled" , "rog" , "maxdiam" , "maxhomedist" , "siglocsvisited" , "avgflightlen" , "stdflightlen" , "avgflightdur" , "stdflightdur" , "probpause" , "siglocentropy" , "circdnrtn" , "wkenddayrtn" ]
2021-03-05 23:49:37 +01:00
IF_MULTIPLE_TIMEZONES : USE_MOST_COMMON
2020-08-28 23:50:49 +02:00
MINUTES_DATA_USED : False # Use this for quality control purposes, how many minutes of data (location coordinates gruped by minute) were used to compute features
2021-03-15 00:52:14 +01:00
SRC_SCRIPT : src/features/phone_locations/barnett/main.R
2020-07-16 20:26:43 +02:00
2021-03-10 22:26:50 +01:00
# See https://www.rapids.science/latest/features/phone-log/
PHONE_LOG :
2021-03-13 00:14:49 +01:00
CONTAINER :
2021-03-10 22:26:50 +01:00
ANDROID : aware_log
IOS : ios_aware_log
PROVIDERS : # None implemented yet but this sensor can be used in PHONE_DATA_YIELD
2020-11-26 01:42:11 +01:00
# See https://www.rapids.science/latest/features/phone-messages/
PHONE_MESSAGES :
2021-12-08 18:37:34 +01:00
CONTAINER : sms
2020-08-31 23:08:26 +02:00
PROVIDERS :
RAPIDS :
2021-12-08 18:37:34 +01:00
COMPUTE : True
2020-11-26 01:42:11 +01:00
MESSAGES_TYPES : [ received, sent]
FEATURES :
received : [ count, distinctcontacts, timefirstmessage, timelastmessage, countmostfrequentcontact]
sent : [ count, distinctcontacts, timefirstmessage, timelastmessage, countmostfrequentcontact]
2021-03-15 00:52:14 +01:00
SRC_SCRIPT : src/features/phone_messages/rapids/main.R
2020-08-31 23:08:26 +02:00
2020-11-26 01:42:11 +01:00
# See https://www.rapids.science/latest/features/phone-screen/
2020-10-19 21:07:12 +02:00
PHONE_SCREEN :
2021-03-13 00:14:49 +01:00
CONTAINER : screen
2020-09-19 02:25:29 +02:00
PROVIDERS :
RAPIDS :
2021-12-08 18:37:34 +01:00
COMPUTE : True
2020-09-19 02:25:29 +02:00
REFERENCE_HOUR_FIRST_USE : 0
IGNORE_EPISODES_SHORTER_THAN : 0 # in minutes, set to 0 to disable
2021-10-14 15:26:06 +02:00
IGNORE_EPISODES_LONGER_THAN : 360 # in minutes, set to 0 to disable
2020-09-19 02:25:29 +02:00
FEATURES : [ "countepisode" , "sumduration" , "maxduration" , "minduration" , "avgduration" , "stdduration" , "firstuseafter" ] # "episodepersensedminutes" needs to be added later
EPISODE_TYPES : [ "unlock" ]
2021-03-15 00:52:14 +01:00
SRC_SCRIPT : src/features/phone_screen/rapids/main.py
2020-01-14 15:51:39 +01:00
2020-11-26 01:42:11 +01:00
# See https://www.rapids.science/latest/features/phone-wifi-connected/
PHONE_WIFI_CONNECTED :
2021-03-15 00:52:14 +01:00
CONTAINER : sensor_wifi
2020-09-01 21:25:35 +02:00
PROVIDERS :
RAPIDS :
2020-09-03 20:36:42 +02:00
COMPUTE : False
2020-11-26 01:42:11 +01:00
FEATURES : [ "countscans" , "uniquedevices" , "countscansmostuniquedevice" ]
2021-03-15 00:52:14 +01:00
SRC_SCRIPT : src/features/phone_wifi_connected/rapids/main.R
2020-02-07 17:52:55 +01:00
2020-11-26 01:42:11 +01:00
# See https://www.rapids.science/latest/features/phone-wifi-visible/
2020-10-19 21:07:12 +02:00
PHONE_WIFI_VISIBLE :
2021-03-15 00:52:14 +01:00
CONTAINER : wifi
2020-10-19 21:07:12 +02:00
PROVIDERS :
RAPIDS :
2021-12-08 18:37:34 +01:00
COMPUTE : True
2020-10-19 21:07:12 +02:00
FEATURES : [ "countscans" , "uniquedevices" , "countscansmostuniquedevice" ]
2021-03-15 00:52:14 +01:00
SRC_SCRIPT : src/features/phone_wifi_visible/rapids/main.R
2020-10-19 21:07:12 +02:00
2021-12-08 18:59:10 +01:00
########################################################################################################################
# FITBIT #
########################################################################################################################
# See https://www.rapids.science/latest/setup/configuration/#data-stream-configuration
FITBIT_DATA_STREAMS :
USE : fitbitjson_mysql
# AVAILABLE:
fitbitjson_mysql :
DATABASE_GROUP : MY_GROUP
SLEEP_SUMMARY_LAST_NIGHT_END : 660 # a number ranged from 0 (midnight) to 1439 (23:59) which denotes number of minutes after midnight. By default, 660 (11:00).
fitbitparsed_mysql :
DATABASE_GROUP : MY_GROUP
SLEEP_SUMMARY_LAST_NIGHT_END : 660 # a number ranged from 0 (midnight) to 1439 (23:59) which denotes number of minutes after midnight. By default, 660 (11:00).
fitbitjson_csv :
FOLDER : data/external/fitbit_csv
SLEEP_SUMMARY_LAST_NIGHT_END : 660 # a number ranged from 0 (midnight) to 1439 (23:59) which denotes number of minutes after midnight. By default, 660 (11:00).
fitbitparsed_csv :
FOLDER : data/external/fitbit_csv
SLEEP_SUMMARY_LAST_NIGHT_END : 660 # a number ranged from 0 (midnight) to 1439 (23:59) which denotes number of minutes after midnight. By default, 660 (11:00).
# Sensors ------
# See https://www.rapids.science/latest/features/fitbit-calories-intraday/
FITBIT_CALORIES_INTRADAY :
CONTAINER : fitbit_data
PROVIDERS :
RAPIDS :
COMPUTE : False
EPISODE_TYPE : [ sedentary, lightlyactive, fairlyactive, veryactive, mvpa, lowmet, highmet]
EPISODE_TIME_THRESHOLD : 5 # minutes
EPISODE_MET_THRESHOLD : 3
EPISODE_MVPA_CATEGORIES : [ fairlyactive, veryactive]
EPISODE_REFERENCE_TIME : MIDNIGHT # or START_OF_THE_SEGMENT
FEATURES : [ count, sumduration, avgduration, minduration, maxduration, stdduration, starttimefirst, endtimefirst, starttimelast, endtimelast, starttimelongest, endtimelongest, summet, avgmet, maxmet, minmet, stdmet, sumcalories, avgcalories, maxcalories, mincalories, stdcalories]
SRC_SCRIPT : src/features/fitbit_calories_intraday/rapids/main.R
# See https://www.rapids.science/latest/features/fitbit-data-yield/
FITBIT_DATA_YIELD :
SENSOR : FITBIT_HEARTRATE_INTRADAY
PROVIDERS :
RAPIDS :
COMPUTE : False
FEATURES : [ ratiovalidyieldedminutes, ratiovalidyieldedhours]
MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS : 0.5 # 0 to 1, minimum percentage of valid minutes in an hour to be considered valid.
SRC_SCRIPT : src/features/fitbit_data_yield/rapids/main.R
# See https://www.rapids.science/latest/features/fitbit-heartrate-summary/
FITBIT_HEARTRATE_SUMMARY :
CONTAINER : heartrate_summary
PROVIDERS :
RAPIDS :
COMPUTE : False
FEATURES : [ "maxrestinghr" , "minrestinghr" , "avgrestinghr" , "medianrestinghr" , "moderestinghr" , "stdrestinghr" , "diffmaxmoderestinghr" , "diffminmoderestinghr" , "entropyrestinghr" ] # calories features' accuracy depend on the accuracy of the participants fitbit profile (e.g. height, weight) use these with care : [ "sumcaloriesoutofrange" , "maxcaloriesoutofrange" , "mincaloriesoutofrange" , "avgcaloriesoutofrange" , "mediancaloriesoutofrange" , "stdcaloriesoutofrange" , "entropycaloriesoutofrange" , "sumcaloriesfatburn" , "maxcaloriesfatburn" , "mincaloriesfatburn" , "avgcaloriesfatburn" , "mediancaloriesfatburn" , "stdcaloriesfatburn" , "entropycaloriesfatburn" , "sumcaloriescardio" , "maxcaloriescardio" , "mincaloriescardio" , "avgcaloriescardio" , "mediancaloriescardio" , "stdcaloriescardio" , "entropycaloriescardio" , "sumcaloriespeak" , "maxcaloriespeak" , "mincaloriespeak" , "avgcaloriespeak" , "mediancaloriespeak" , "stdcaloriespeak" , "entropycaloriespeak" ]
SRC_SCRIPT : src/features/fitbit_heartrate_summary/rapids/main.py
# See https://www.rapids.science/latest/features/fitbit-heartrate-intraday/
FITBIT_HEARTRATE_INTRADAY :
CONTAINER : heartrate_intraday
PROVIDERS :
RAPIDS :
COMPUTE : False
FEATURES : [ "maxhr" , "minhr" , "avghr" , "medianhr" , "modehr" , "stdhr" , "diffmaxmodehr" , "diffminmodehr" , "entropyhr" , "minutesonoutofrangezone" , "minutesonfatburnzone" , "minutesoncardiozone" , "minutesonpeakzone" ]
SRC_SCRIPT : src/features/fitbit_heartrate_intraday/rapids/main.py
# See https://www.rapids.science/latest/features/fitbit-sleep-summary/
FITBIT_SLEEP_SUMMARY :
CONTAINER : sleep_summary
PROVIDERS :
RAPIDS :
COMPUTE : False
FEATURES : [ "firstwaketime" , "lastwaketime" , "firstbedtime" , "lastbedtime" , "countepisode" , "avgefficiency" , "sumdurationafterwakeup" , "sumdurationasleep" , "sumdurationawake" , "sumdurationtofallasleep" , "sumdurationinbed" , "avgdurationafterwakeup" , "avgdurationasleep" , "avgdurationawake" , "avgdurationtofallasleep" , "avgdurationinbed" ]
SLEEP_TYPES : [ "main" , "nap" , "all" ]
SRC_SCRIPT : src/features/fitbit_sleep_summary/rapids/main.py
# See https://www.rapids.science/latest/features/fitbit-sleep-intraday/
FITBIT_SLEEP_INTRADAY :
CONTAINER : sleep_intraday
PROVIDERS :
RAPIDS :
COMPUTE : False
FEATURES :
LEVELS_AND_TYPES : [ countepisode, sumduration, maxduration, minduration, avgduration, medianduration, stdduration]
RATIOS_TYPE : [ count, duration]
RATIOS_SCOPE : [ ACROSS_LEVELS, ACROSS_TYPES, WITHIN_LEVELS, WITHIN_TYPES]
SLEEP_LEVELS :
INCLUDE_ALL_GROUPS : True
CLASSIC : [ awake, restless, asleep]
STAGES : [ wake, deep, light, rem]
UNIFIED : [ awake, asleep]
SLEEP_TYPES : [ main, nap, all]
SRC_SCRIPT : src/features/fitbit_sleep_intraday/rapids/main.py
PRICE :
COMPUTE : False
FEATURES : [ avgduration, avgratioduration, avgstarttimeofepisodemain, avgendtimeofepisodemain, avgmidpointofepisodemain, stdstarttimeofepisodemain, stdendtimeofepisodemain, stdmidpointofepisodemain, socialjetlag, rmssdmeanstarttimeofepisodemain, rmssdmeanendtimeofepisodemain, rmssdmeanmidpointofepisodemain, rmssdmedianstarttimeofepisodemain, rmssdmedianendtimeofepisodemain, rmssdmedianmidpointofepisodemain]
SLEEP_LEVELS :
INCLUDE_ALL_GROUPS : True
CLASSIC : [ awake, restless, asleep]
STAGES : [ wake, deep, light, rem]
UNIFIED : [ awake, asleep]
DAY_TYPES : [ WEEKEND, WEEK, ALL]
LAST_NIGHT_END : 660 # number of minutes after midnight (11:00) 11*60
SRC_SCRIPT : src/features/fitbit_sleep_intraday/price/main.py
# See https://www.rapids.science/latest/features/fitbit-steps-summary/
FITBIT_STEPS_SUMMARY :
CONTAINER : steps_summary
PROVIDERS :
RAPIDS :
COMPUTE : False
FEATURES : [ "maxsumsteps" , "minsumsteps" , "avgsumsteps" , "mediansumsteps" , "stdsumsteps" ]
SRC_SCRIPT : src/features/fitbit_steps_summary/rapids/main.py
# See https://www.rapids.science/latest/features/fitbit-steps-intraday/
FITBIT_STEPS_INTRADAY :
CONTAINER : steps_intraday
EXCLUDE_SLEEP : # you can exclude step data that was logged during sleep periods
TIME_BASED :
EXCLUDE : False
START_TIME : "23:00"
END_TIME : "07:00"
FITBIT_BASED :
EXCLUDE : False
PROVIDERS :
RAPIDS :
COMPUTE : False
FEATURES :
2021-11-19 00:35:27 +01:00
STEPS : [ "sum" , "max" , "min" , "avg" , "std" , "firststeptime" , "laststeptime" ]
2021-12-08 18:59:10 +01:00
SEDENTARY_BOUT : [ "countepisode" , "sumduration" , "maxduration" , "minduration" , "avgduration" , "stdduration" ]
ACTIVE_BOUT : [ "countepisode" , "sumduration" , "maxduration" , "minduration" , "avgduration" , "stdduration" ]
2021-11-19 00:35:27 +01:00
REFERENCE_HOUR : 0
2021-12-08 18:59:10 +01:00
THRESHOLD_ACTIVE_BOUT : 10 # steps
INCLUDE_ZERO_STEP_ROWS : False
SRC_SCRIPT : src/features/fitbit_steps_intraday/rapids/main.py
2020-12-15 02:30:34 +01:00
########################################################################################################################
# EMPATICA #
########################################################################################################################
2021-03-07 05:16:59 +01:00
EMPATICA_DATA_STREAMS :
2021-03-11 20:30:42 +01:00
USE : empatica_zip
2021-03-07 05:16:59 +01:00
# AVAILABLE:
2021-03-11 20:30:42 +01:00
empatica_zip :
2020-12-16 00:19:11 +01:00
FOLDER : data/external/empatica
2020-12-15 02:30:34 +01:00
# Sensors ------
2021-03-11 00:51:55 +01:00
# See https://www.rapids.science/latest/features/empatica-accelerometer/
2020-12-15 02:30:34 +01:00
EMPATICA_ACCELEROMETER :
2021-03-13 00:14:49 +01:00
CONTAINER : ACC
2020-12-15 02:30:34 +01:00
PROVIDERS :
DBDP :
2021-02-02 23:19:10 +01:00
COMPUTE : False
FEATURES : [ "maxmagnitude" , "minmagnitude" , "avgmagnitude" , "medianmagnitude" , "stdmagnitude" ]
2021-03-15 00:52:14 +01:00
SRC_SCRIPT : src/features/empatica_accelerometer/dbdp/main.py
2020-12-15 02:30:34 +01:00
2021-03-11 00:51:55 +01:00
# See https://www.rapids.science/latest/features/empatica-heartrate/
2020-12-15 02:30:34 +01:00
EMPATICA_HEARTRATE :
2021-03-13 00:14:49 +01:00
CONTAINER : HR
2020-12-15 02:30:34 +01:00
PROVIDERS :
DBDP :
2021-02-02 23:34:56 +01:00
COMPUTE : False
FEATURES : [ "maxhr" , "minhr" , "avghr" , "medianhr" , "modehr" , "stdhr" , "diffmaxmodehr" , "diffminmodehr" , "entropyhr" ]
2021-03-15 00:52:14 +01:00
SRC_SCRIPT : src/features/empatica_heartrate/dbdp/main.py
2020-12-15 02:30:34 +01:00
2021-03-11 00:51:55 +01:00
# See https://www.rapids.science/latest/features/empatica-temperature/
2020-12-15 02:30:34 +01:00
EMPATICA_TEMPERATURE :
2021-03-13 00:14:49 +01:00
CONTAINER : TEMP
2020-12-15 02:30:34 +01:00
PROVIDERS :
DBDP :
2021-02-02 23:34:56 +01:00
COMPUTE : False
2021-02-12 02:56:27 +01:00
FEATURES : [ "maxtemp" , "mintemp" , "avgtemp" , "mediantemp" , "modetemp" , "stdtemp" , "diffmaxmodetemp" , "diffminmodetemp" , "entropytemp" ]
2021-03-15 00:52:14 +01:00
SRC_SCRIPT : src/features/empatica_temperature/dbdp/main.py
2020-12-15 02:30:34 +01:00
2021-03-11 00:51:55 +01:00
# See https://www.rapids.science/latest/features/empatica-electrodermal-activity/
2020-12-15 02:30:34 +01:00
EMPATICA_ELECTRODERMAL_ACTIVITY :
2021-03-13 00:14:49 +01:00
CONTAINER : EDA
2020-12-15 02:30:34 +01:00
PROVIDERS :
DBDP :
2021-02-02 23:34:56 +01:00
COMPUTE : False
2021-02-12 02:56:27 +01:00
FEATURES : [ "maxeda" , "mineda" , "avgeda" , "medianeda" , "modeeda" , "stdeda" , "diffmaxmodeeda" , "diffminmodeeda" , "entropyeda" ]
2021-03-15 00:52:14 +01:00
SRC_SCRIPT : src/features/empatica_electrodermal_activity/dbdp/main.py
2020-12-15 02:30:34 +01:00
2021-03-11 00:51:55 +01:00
# See https://www.rapids.science/latest/features/empatica-blood-volume-pulse/
2020-12-15 02:30:34 +01:00
EMPATICA_BLOOD_VOLUME_PULSE :
2021-03-13 00:14:49 +01:00
CONTAINER : BVP
2020-12-15 02:30:34 +01:00
PROVIDERS :
DBDP :
2021-02-02 23:34:56 +01:00
COMPUTE : False
2021-02-12 02:56:27 +01:00
FEATURES : [ "maxbvp" , "minbvp" , "avgbvp" , "medianbvp" , "modebvp" , "stdbvp" , "diffmaxmodebvp" , "diffminmodebvp" , "entropybvp" ]
2021-03-15 00:52:14 +01:00
SRC_SCRIPT : src/features/empatica_blood_volume_pulse/dbdp/main.py
2020-12-15 02:30:34 +01:00
2021-03-11 00:51:55 +01:00
# See https://www.rapids.science/latest/features/empatica-inter-beat-interval/
2020-12-15 02:30:34 +01:00
EMPATICA_INTER_BEAT_INTERVAL :
2021-03-13 00:14:49 +01:00
CONTAINER : IBI
2020-12-15 02:30:34 +01:00
PROVIDERS :
DBDP :
2021-02-02 23:34:56 +01:00
COMPUTE : False
2021-02-12 02:56:27 +01:00
FEATURES : [ "maxibi" , "minibi" , "avgibi" , "medianibi" , "modeibi" , "stdibi" , "diffmaxmodeibi" , "diffminmodeibi" , "entropyibi" ]
2021-03-15 00:52:14 +01:00
SRC_SCRIPT : src/features/empatica_inter_beat_interval/dbdp/main.py
2020-12-15 02:30:34 +01:00
2021-03-11 00:51:55 +01:00
# See https://www.rapids.science/latest/features/empatica-tags/
2020-12-15 02:30:34 +01:00
EMPATICA_TAGS :
2021-03-13 00:14:49 +01:00
CONTAINER : TAGS
2021-02-16 00:29:05 +01:00
PROVIDERS : # None implemented yet
2020-05-15 23:51:00 +02:00
2020-11-26 01:42:11 +01:00
########################################################################################################################
# PLOTS #
########################################################################################################################
2020-10-22 19:08:52 +02:00
2021-03-11 00:51:55 +01:00
# Data quality ------
# See https://www.rapids.science/latest/visualizations/data-quality-visualizations/#1-histograms-of-phone-data-yield
2020-12-01 20:21:33 +01:00
HISTOGRAM_PHONE_DATA_YIELD :
2021-12-08 18:37:34 +01:00
PLOT : True
2020-12-01 20:21:33 +01:00
2021-03-11 00:51:55 +01:00
# See https://www.rapids.science/latest/visualizations/data-quality-visualizations/#2-heatmaps-of-overall-data-yield
2020-12-04 22:49:29 +01:00
HEATMAP_PHONE_DATA_YIELD_PER_PARTICIPANT_PER_TIME_SEGMENT :
2022-02-23 11:10:54 +01:00
PLOT : False
2021-03-24 02:40:55 +01:00
TIME : RELATIVE_TIME # ABSOLUTE_TIME or RELATIVE_TIME
2020-12-04 22:49:29 +01:00
2021-03-11 00:51:55 +01:00
# See https://www.rapids.science/latest/visualizations/data-quality-visualizations/#3-heatmap-of-recorded-phone-sensors
2020-12-04 03:00:32 +01:00
HEATMAP_SENSORS_PER_MINUTE_PER_TIME_SEGMENT :
2021-12-08 18:37:34 +01:00
PLOT : True
2020-07-22 22:47:32 +02:00
2021-03-11 00:51:55 +01:00
# See https://www.rapids.science/latest/visualizations/data-quality-visualizations/#4-heatmap-of-sensor-row-count
2020-12-04 03:00:32 +01:00
HEATMAP_SENSOR_ROW_COUNT_PER_TIME_SEGMENT :
2020-07-22 22:47:32 +02:00
PLOT : False
2021-03-24 02:40:55 +01:00
SENSORS : [ ]
2020-07-22 22:47:32 +02:00
2021-03-11 00:51:55 +01:00
# Features ------
# See https://www.rapids.science/latest/visualizations/feature-visualizations/#1-heatmap-correlation-matrix
2020-12-04 03:00:32 +01:00
HEATMAP_FEATURE_CORRELATION_MATRIX :
2021-12-08 18:37:34 +01:00
PLOT : True
2020-12-04 03:00:32 +01:00
MIN_ROWS_RATIO : 0.5
CORR_THRESHOLD : 0.1
CORR_METHOD : "pearson" # choose from {"pearson", "kendall", "spearman"}
2020-07-22 22:47:32 +02:00
2021-11-19 16:34:36 +01:00
########################################################################################################################
# Data Cleaning #
########################################################################################################################
ALL_CLEANING_INDIVIDUAL :
PROVIDERS :
RAPIDS :
2022-01-07 12:20:51 +01:00
COMPUTE : True
2021-11-19 16:34:36 +01:00
IMPUTE_SELECTED_EVENT_FEATURES :
COMPUTE : True
MIN_DATA_YIELDED_MINUTES_TO_IMPUTE : 0.33
COLS_NAN_THRESHOLD : 0.3 # set to 1 to disable
COLS_VAR_THRESHOLD : True
2022-01-07 12:20:51 +01:00
ROWS_NAN_THRESHOLD : 1 # set to 1 to disable
2021-11-19 16:34:36 +01:00
DATA_YIELD_FEATURE : RATIO_VALID_YIELDED_HOURS # RATIO_VALID_YIELDED_HOURS or RATIO_VALID_YIELDED_MINUTES
2022-01-07 12:20:51 +01:00
DATA_YIELD_RATIO_THRESHOLD : 0.3 # set to 0 to disable
2021-11-19 16:34:36 +01:00
DROP_HIGHLY_CORRELATED_FEATURES :
2022-01-07 12:20:51 +01:00
COMPUTE : False
2021-11-19 16:34:36 +01:00
MIN_OVERLAP_FOR_CORR_THRESHOLD : 0.5
CORR_THRESHOLD : 0.95
SRC_SCRIPT : src/features/all_cleaning_individual/rapids/main.R
ALL_CLEANING_OVERALL :
PROVIDERS :
RAPIDS :
2022-01-19 13:41:09 +01:00
COMPUTE : True
2021-11-19 16:34:36 +01:00
IMPUTE_SELECTED_EVENT_FEATURES :
COMPUTE : True
MIN_DATA_YIELDED_MINUTES_TO_IMPUTE : 0.33
COLS_NAN_THRESHOLD : 0.3 # set to 1 to disable
COLS_VAR_THRESHOLD : True
2022-01-19 13:41:09 +01:00
ROWS_NAN_THRESHOLD : 1 # set to 1 to disable
2021-11-19 16:34:36 +01:00
DATA_YIELD_FEATURE : RATIO_VALID_YIELDED_HOURS # RATIO_VALID_YIELDED_HOURS or RATIO_VALID_YIELDED_MINUTES
2022-01-19 13:41:09 +01:00
DATA_YIELD_RATIO_THRESHOLD : 0.3 # set to 0 to disable
2021-11-19 16:34:36 +01:00
DROP_HIGHLY_CORRELATED_FEATURES :
2022-01-19 13:41:09 +01:00
COMPUTE : False
2021-11-19 16:34:36 +01:00
MIN_OVERLAP_FOR_CORR_THRESHOLD : 0.5
CORR_THRESHOLD : 0.95
SRC_SCRIPT : src/features/all_cleaning_overall/rapids/main.R
2022-02-04 17:37:00 +01:00
########################################################################################################################
# Analysis Workflow Example #
########################################################################################################################
PARAMS_FOR_ANALYSIS :
2022-02-04 18:21:42 +01:00
BASELINE :
2022-02-04 18:06:02 +01:00
FOLDER : data/external/baseline
2022-02-04 17:37:00 +01:00
CONTAINER : [ results-survey637813_final.csv, # Slovenia
results-survey358134_final.csv, # Belgium 1
results-survey413767_final.csv # Belgium 2
]
2022-02-23 19:08:10 +01:00
QUESTION_LIST : survey637813+question_text.csv
2022-02-23 18:05:23 +01:00
FEATURES : [ age, gender, startlanguage]
2022-02-04 17:37:00 +01:00
CATEGORICAL_FEATURES : [ gender]