rapids/config.yaml

# Valid database table names
SENSORS: [applications_crashes, applications_foreground, applications_notifications, battery, bluetooth, calls, locations, messages, plugin_ambient_noise, plugin_device_usage, plugin_google_activity_recognition, plugin_ios_activity_recognition, screen]

FITBIT_TABLE: [fitbit_data]
FITBIT_SENSORS: [heartrate, steps, sleep, calories]

# Participants to include in the analysis
# You must create a file for each participant
# named pXXX containing their device_id
PIDS: [p01, p02]

# Global var with common day segments
DAY_SEGMENTS: &day_segments
  [daily, morning, afternoon, evening, night]

# Global timezone
# Use codes from https://en.wikipedia.org/wiki/List_of_tz_database_time_zones
# Double check your code, for example EST is not US Eastern Time.
TIMEZONE: &timezone
  America/New_York

DATABASE_GROUP: &database_group
  MY_GROUP

DOWNLOAD_PARTICIPANTS:
  IGNORED_DEVICE_IDS: [] # for example "5a1dd68c-6cd1-48fe-ae1e-14344ac5215f"
  GROUP: *database_group

# Download data config
DOWNLOAD_DATASET:
  GROUP: *database_group

# Readable datetime config
READABLE_DATETIME:
  FIXED_TIMEZONE: *timezone

# Communication SMS features config, TYPES and FEATURES keys need to match
SMS:
  TYPES : [received, sent]
  FEATURES: 
    received: [count, distinctcontacts, timefirstsms, timelastsms, countmostfrequentcontact]
    sent: [count, distinctcontacts, timefirstsms, timelastsms, countmostfrequentcontact]
  DAY_SEGMENTS: *day_segments  

# Communication call features config, TYPES and FEATURES keys need to match
CALLS:
  TYPES: [missed, incoming, outgoing]
  FEATURES:
    missed:  [count, distinctcontacts, timefirstcall, timelastcall, countmostfrequentcontact]
    incoming: [count, distinctcontacts, meanduration, sumduration, minduration, maxduration, stdduration, modeduration, hubermduration, varqnduration, entropyduration, timefirstcall, timelastcall, countmostfrequentcontact]
    outgoing: [count, distinctcontacts, meanduration, sumduration, minduration, maxduration, stdduration, modeduration, hubermduration, varqnduration, entropyduration, timefirstcall, timelastcall, countmostfrequentcontact]
  DAY_SEGMENTS: *day_segments

APPLICATION_GENRES:
  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_GENRES: 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

PHONE_VALID_SENSED_DAYS:
  BIN_SIZE: 5 # (in minutes)
  MIN_VALID_HOURS: 20 # (out of 24)
  MIN_BINS_PER_HOUR: 8 # (out of 60min/BIN_SIZE bins)

RESAMPLE_FUSED_LOCATION:
  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
  TIME_SINCE_VALID_LOCATION: 12 # hours, only replicate location samples to consecutive sensed bins if they were logged within this threshold after a valid location row
  TIMEZONE: *timezone

BARNETT_LOCATION:
  DAY_SEGMENTS: [daily] # These features are only available on a daily basis
  FEATURES: ["hometime","disttravelled","rog","maxdiam","maxhomedist","siglocsvisited","avgflightlen","stdflightlen","avgflightdur","stdflightdur","probpause","siglocentropy","circdnrtn","wkenddayrtn"]
  LOCATIONS_TO_USE: ALL # ALL, ALL_EXCEPT_FUSED OR RESAMPLE_FUSED
  ACCURACY_LIMIT: 51 # meters, drops location coordinates with an accuracy higher than this. This number means there's a 68% probability the true location is within this radius
  TIMEZONE: *timezone

BLUETOOTH:
  DAY_SEGMENTS: *day_segments
  FEATURES: ["countscans", "uniquedevices", "countscansmostuniquedevice"]

ACTIVITY_RECOGNITION:
  DAY_SEGMENTS: *day_segments
  FEATURES: ['count','mostcommonactivity','countuniqueactivities','activitychangecount','sumstationary','summobile','sumvehicle']

BATTERY:
  DAY_SEGMENTS: *day_segments
  FEATURES: ["countdischarge", "sumdurationdischarge", "countcharge", "sumdurationcharge", "avgconsumptionrate", "maxconsumptionrate"]

SCREEN:
  DAY_SEGMENTS: *day_segments
  REFERENCE_HOUR_FIRST_USE: 0
  FEATURES_DELTAS: ["countepisode", "episodepersensedminutes", "sumduration", "maxduration", "minduration", "avgduration", "stdduration", "firstuseafter"]
  EPISODE_TYPES: ["unlock"]

LIGHT:
  DAY_SEGMENTS: *day_segments
  FEATURES: ["count", "maxlux", "minlux", "avglux", "medianlux", "stdlux"]

ACCELEROMETER:
  DAY_SEGMENTS: *day_segments
  FEATURES: ["maxmagnitude", "minmagnitude", "avgmagnitude", "medianmagnitude", "stdmagnitude", "ratioexertionalactivityepisodes", "sumexertionalactivityepisodes", "longestexertionalactivityepisode", "longestnonexertionalactivityepisode", "countexertionalactivityepisodes", "countnonexertionalactivityepisodes"]

APPLICATIONS_FOREGROUND:
  DAY_SEGMENTS: *day_segments
  SINGLE_CATEGORIES: ["all", "video"]
  MULTIPLE_CATEGORIES:
    social: ["socialnetworks", "socialmediatools"]
    entertainment: ["entertainment", "gamingknowledge", "gamingcasual", "gamingadventure", "gamingstrategy", "gamingtoolscommunity", "gamingroleplaying", "gamingaction", "gaminglogic", "gamingsports", "gamingsimulation"]
  SINGLE_APPS: ["top1global", "com.facebook.moments", "com.google.android.youtube", "com.twitter.android"] # There's no entropy for single apps
  EXCLUDED_CATEGORIES: ["system_apps", "video"]
  EXCLUDED_APPS: ["com.fitbit.FitbitMobile", "com.aware.plugin.upmc.cancer"]
  FEATURES: ["count", "timeoffirstuse", "timeoflastuse", "frequencyentropy"]

HEARTRATE:
  DAY_SEGMENTS: *day_segments
  FEATURES: ["maxhr", "minhr", "avghr", "medianhr", "modehr", "stdhr", "diffmaxmodehr", "diffminmodehr", "entropyhr", "lengthoutofrange", "lengthfatburn", "lengthcardio", "lengthpeak"]

STEP:
  DAY_SEGMENTS: *day_segments
  FEATURES:
    ALL_STEPS: ["sumallsteps", "maxallsteps", "minallsteps", "avgallsteps", "stdallsteps"]
    SEDENTARY_BOUT: ["countsedentarybout", "maxdurationsedentarybout", "mindurationsedentarybout", "avgdurationsedentarybout", "stddurationsedentarybout", "sumdurationsedentarybout"]
    ACTIVE_BOUT: ["countactivebout", "maxdurationactivebout", "mindurationactivebout", "avgdurationactivebout", "stddurationactivebout"]
  THRESHOLD_ACTIVE_BOUT: 10 # steps
  INCLUDE_ZERO_STEP_ROWS: True

WIFI:
  DAY_SEGMENTS: *day_segments
  FEATURES: ["countscans", "uniquedevices", "countscansmostuniquedevice"]

PARAMS_FOR_ANALYSIS:
  GROUNDTRUTH_TABLE: participant_info
  SOURCES: &sources ["phone_features", "fitbit_features", "phone_fitbit_features"]
  DAY_SEGMENTS: *day_segments
  PHONE_FEATURES: [accelerometer, applications_foreground, battery, call_incoming, call_missed, call_outgoing, activity_recognition, light, location_barnett, screen, sms_received, sms_sent]
  FITBIT_FEATURES: [fitbit_heartrate, fitbit_step]
  PHONE_FITBIT_FEATURES: "" # This array is merged in the input_merge_features_of_single_participant function in models.snakefile
  DEMOGRAPHIC_FEATURES: [age, gender, inpatientdays]
  CATEGORICAL_DEMOGRAPHIC_FEATURES: ["gender"]
  
  # Whether or not to include only days with enough valid sensed hours
  # logic can be found in rule phone_valid_sensed_days of rules/preprocessing.snakefile
  DROP_VALID_SENSED_DAYS: 
    ENABLED: True

  # Whether or not to include certain days in the analysis, logic can be found in rule days_to_analyse of rules/mystudy.snakefile
  # If you want to include all days downloaded for each participant, set ENABLED to False
  DAYS_TO_ANALYSE:
    ENABLED: True
    DAYS_BEFORE_SURGERY: 15
    DAYS_IN_HOSPITAL: F # T or F
    DAYS_AFTER_DISCHARGE: 7

  # Cleaning Parameters
  COLS_NAN_THRESHOLD: 0.5
  COLS_VAR_THRESHOLD: True
  ROWS_NAN_THRESHOLD: 0.5
  PARTICIPANT_DAYS_BEFORE_THRESHOLD: 7
  PARTICIPANT_DAYS_AFTER_THRESHOLD: 4

  # Extract summarised features from daily features with any of the following substrings
  NUMERICAL_OPERATORS: ["count", "sum", "length", "avg"]
  CATEGORICAL_OPERATORS: ["mostcommon"]
  
  MODEL_NAMES: ["LogReg", "kNN", "SVM", "DT", "RF", "GB", "XGBoost", "LightGBM"]
  CV_METHODS: ["LeaveOneOut"]
  SUMMARISED: ["summarised"] # "summarised" or "notsummarised"
  SCALER: ["notnormalized", "minmaxscaler", "standardscaler", "robustscaler"]
  RESULT_COMPONENTS: ["fold_predictions", "fold_metrics", "overall_results", "fold_feature_importances"]

  MODEL_HYPERPARAMS:
    LogReg:
      {"clf__C": [0.01, 0.1, 1, 10, 100], "clf__solver": ["newton-cg", "lbfgs", "liblinear", "saga"], "clf__penalty": ["l2"]}
    kNN:
      {"clf__n_neighbors": range(1, 21, 2), "clf__weights": ["uniform", "distance"], "clf__metric": ["euclidean", "manhattan", "minkowski"]}
    SVM:
      {"clf__C": [0.01, 0.1, 1, 10, 100], "clf__gamma": ["scale", "auto"], "clf__kernel": ["rbf", "poly", "sigmoid"]}
    DT:
      {"clf__criterion": ["gini", "entropy"], "clf__max_depth": [None, 3, 5, 7, 9], "clf__max_features": [None, "auto", "sqrt", "log2"]}
    RF:
      {"clf__n_estimators": [2, 5, 10, 100],"clf__max_depth": [None, 3, 5, 7, 9]}
    GB:
      {"clf__learning_rate": [0.01, 0.1, 1], "clf__n_estimators": [5, 10, 100, 200], "clf__subsample": [0.5, 0.7, 1.0], "clf__max_depth": [3, 5, 7, 9]}
    XGBoost:
      {"clf__learning_rate": [0.01, 0.1, 1], "clf__n_estimators": [5, 10, 100, 200], "clf__num_leaves": [5, 16, 31, 62]}
    LightGBM:
      {"clf__learning_rate": [0.01, 0.1, 1], "clf__n_estimators": [5, 10, 100, 200], "clf__num_leaves": [5, 16, 31, 62]}


  # Target Settings:
  # 1 => TARGETS_RATIO_THRESHOLD (ceiling) or more of available CESD scores were TARGETS_VALUE_THRESHOLD or higher; 0 => otherwise
  TARGETS_RATIO_THRESHOLD: 0.5
  TARGETS_VALUE_THRESHOLD: 16