Update docs: add visualization section

docs/features/extracted.rst

@@ -49,6 +49,8 @@ Global Parameters
     - ``FIXED_TIMEZONE``. See ``TIMEZONE`` above. This assumes that all data of all participants was collected within one time zone.
     - Support for multiple time zones for each participant coming soon based on the ``timezone`` table collected by Aware.
 
+.. _phone-valid-sensed-bins:
+
 - ``PHONE_VALID_SENSED_BINS``
      Contains three attributes: ``COMPUTE``, ``BIN_SIZE`` and ``TABLES``. See the PHONE_VALID_SENSED_BINS_ section in the ``config.yaml`` file
 

docs/index.rst

@@ -29,6 +29,16 @@ Contents:
 
    features/extracted
 
+.. toctree::
+   :maxdepth: 2
+   :caption: Visualization
+
+   visualization/histogram_of_valid_sensed_hours
+   visualization/heatmap_of_phone_sensed_bins
+   visualization/heatmap_of_days_by_sensors
+   visualization/overall_compliance_heatmap
+   visualization/heatmap_of_correlation_matrix_between_features
+
 .. toctree::
    :maxdepth: 2
    :caption: Developers

docs/visualization/heatmap_of_correlation_matrix_between_features.rst

@@ -0,0 +1,6 @@
+.. _heatmap-of-correlation-matrix-between-features
+
+Heatmap of correlation matrix between features
+==============================================
+
+Columns and rows are features computed in RAPIDS, cells’ color represents the correlation coefficient between all days of data for every pair of feature of all participants. The user can specify a minimum number of observations required to compute the correlation between two features using the ``MIN_ROWS_RATIO`` parameter (0.5 by default). In addition, this plot can be configured to only display correlation coefficients above a threshold using the ``CORR_THRESHOLD`` parameter (0.1 by default).

docs/visualization/heatmap_of_days_by_sensors.rst

@@ -0,0 +1,6 @@
+.. _heatmap-of-days-by-sensors
+
+Heatmap of days by sensors
+==========================
+
+In this heatmap rows are sensors, columns are days and cells’ color shows the normalized (0 to 1) number of valid sensed hours (See valid sensed :ref:`bins<phone-valid-sensed-bins>` and :ref:`days<phone-valid-sensed-days>` sections) collected by a sensor during a day for a participant. The user can decide how many days of data to show starting from the first day of each participant. This plot can used to judge missing data on a per participant, per sensor basis as well as the number of valid sensed hours (usable data) for each day.

docs/visualization/heatmap_of_phone_sensed_bins.rst

@@ -0,0 +1,6 @@
+.. _heatmap-of-phone-sensed-bins:
+
+Heatmap of phone sensed bins
+============================
+
+In this heatmap rows are dates, columns are sensed bins for a participant, and cells’ color shows the number of mobile sensors that logged at least one row of data during that bin. This plot shows the periods of time without data for a participant and can be used as a rough indication of whether time-based sensors were following their sensing schedule (e.g. if location was being sensed every 2 minutes).

docs/visualization/histogram_of_valid_sensed_hours.rst

@@ -0,0 +1,6 @@
+.. _histogram-of-valid-sensed-hours:
+
+Histogram of valid sensed hours
+===============================
+
+This histogram shows the valid sensed hours of all participants processed in RAPIDS (See valid sensed :ref:`bins<phone-valid-sensed-bins>` and :ref:`days<phone-valid-sensed-days>` sections). It can be used as a rough indication of the AWARE client monitoring coverage during a study for all participants.

docs/visualization/overall_compliance_heatmap.rst

@@ -0,0 +1,6 @@
+.. _overall-compliance-heatmap
+
+Overall compliance heatmap
+==========================
+
+In this heatmap rows are participants, columns are days and cells’ color shows the valid sensed hours for a participant during a day (See valid sensed :ref:`bins<phone-valid-sensed-bins>` and :ref:`days<phone-valid-sensed-days>` sections). This plot can be configured to show a certain number of days before today using the ``EXPECTED_NUM_OF_DAYS`` parameter (by default -1 showing all days for every participant). As different participants might join the study on different dates, the x-axis has a day index instead of a date. This plot gives the user a quick overview of the amount of data collected per person and is complementary to the histogram of valid sensed hours as it is broken down per participant and per day.