FIPS Simulation Walkthrough
FIPS-simulation-walkthrough.RmdIntroduction
This vignette demonstrates how to generate a sleep times dataframe and then proceed to process this for modeling with the Three Process Model of Fatigue. The code here demonstrates the usage from a higher-level user perspective.
Note that in future releases, we may seek to create functions that explicitly generate these sleep times for you. Furthermore, it is possible to simply create the sleeptimes in Excel or another program and transform to the format shown here. Therefore, the sleep generation features are presented here for practical purposes.
Data Preparation
Sleep Times Format
The dataframe below shows a prototypical FIPS ‘sleep times’
dataframe. This form of dataframe is intended for individuals who are
manually inputting sleep history (e.g., from pencil forms to a
spreadsheet). Below, we generate this data structure ourselves for
convenience. Specifically, we want to simulate a scenario where an
individual obtains exactly 7.5 hours of sleep per night for 6 nights.
There are likely multiple ways to achieve this, but the generation
function below allows sufficient flexibility to have continuously offset
sleeptimes (e.g., by changing by = '24 hours' to another
value).
The sleeptimes should correspond to only one participant (multiple
people are not currently supported in FIPS without map functions). The
default column names for this dataframe are sleep.id,
sleep.start, and sleep.end. It is recommended
that you explicitly specify timezones in all functions to avoid any
silent errors relating to time zone specifications.
example.sleeptimes <- tibble::tibble(
sleep.start = seq(
from = lubridate::ymd_hms('2018-05-01 23:00:00', tz = "Australia/Perth"),
to = lubridate::ymd_hms('2018-05-07 17:00:00', tz = "Australia/Perth"),
by = '24 hours'),
sleep.end = sleep.start + lubridate::dhours(7.5),
sleep.id = rank(sleep.start))
print(example.sleeptimes)
#> # A tibble: 6 x 3
#> sleep.start sleep.end sleep.id
#> <dttm> <dttm> <dbl>
#> 1 2018-05-01 23:00:00 2018-05-02 06:30:00 1
#> 2 2018-05-02 23:00:00 2018-05-03 06:30:00 2
#> 3 2018-05-03 23:00:00 2018-05-04 06:30:00 3
#> 4 2018-05-04 23:00:00 2018-05-05 06:30:00 4
#> 5 2018-05-05 23:00:00 2018-05-06 06:30:00 5
#> 6 2018-05-06 23:00:00 2018-05-07 06:30:00 6Prior to actually conducting the simulation, you must
convert the sleep times to the continuous “FIPS_df” format. This format
is a continuous time series style dataframe that contains calculated
variables to be interpreted by the FIPS model functions. This dataframe
contains the following headings:
datetime, sleep.id, wake_status, wake_status_int, change_point, switch_direction, status_duration, total_prev, time, day, sim_hours.
Information regarding these will be presented in the section below, but
first let’s quickly run through how to generate the FIPS dataframe from
sleep times.
The parse_sleeptimes function from FIPS will takes in
‘sleep times’ generated previously, and note the arguments below.
-
sleeptimes— The sleep times dataframe generated previously -
series.start— This is the start datetime of the entire simulation series. -
series.end— This will be the datetime of the entire simulation series -
sleep.start.col— This is the name of yoursleep.startcolumn in sleep times if changed from default above -
sleep.end.col— This is the name of yoursleep.endcolumn in sleep times if changed from default above -
sleep.id.col— This is the name of yoursleep.idcolumn in sleep times if changed from default above -
roundvalue— The epoch of the series (i.e., rounding value). Please read further information below!
The setting of roundvalue is critically
important. It determines the epoch or spacing between observations
in your dataset (in minutes). At a value of 1, the
simulation is updated every 1 minute. Consequently, this will increase
the size (in rows) of your dataset by a factor of 5 (relative to
5 minutes). In most cases, 5, 10 or even 15 minutes should
be sufficient.
Moreover, note that all sleep observations will have their datetime
rounded to this value. For example, with a roundvalue = 5
the datetime 2018-05-07 21:02:02 would be rounded to
2018-05-07 21:00:00. For this reason, it is ideal if your
series.start and series.end are rounded to the
same epoch value as you request. Seconds should never be included in
your datetime (biomathematical models are not sensitive to this time
resolution anyway).
# Simulation start date time (i.e., when you want first predictions to begin)
simulation.start = lubridate::ymd_hms('2018-05-01 07:00:00', tz = "Australia/Perth")
# Simulation end date time (i.e., when you want predictions to end)
# In this case it ends
simulation.end = lubridate::ymd_hms('2018-05-07 21:00:00', tz = "Australia/Perth")
# The Continuous FIPS_df dataframe format
# This creates the format ready for simulation
simulated.dataframe = parse_sleeptimes(
sleeptimes = example.sleeptimes,
series.start = simulation.start,
series.end = simulation.end,
sleep.start.col = "sleep.start",
sleep.end.col = "sleep.end",
sleep.id.col = "sleep.id",
roundvalue = 5)
print(simulated.dataframe)
#> # A tibble: 1,897 x 11
#> datetime sleep.id wake_status wake_status_int change_point
#> * <dttm> <dbl> <lgl> <int> <dbl>
#> 1 2018-05-01 07:00:00 NA TRUE 1 0
#> 2 2018-05-01 07:05:00 NA TRUE 1 0
#> 3 2018-05-01 07:10:00 NA TRUE 1 0
#> 4 2018-05-01 07:15:00 NA TRUE 1 0
#> 5 2018-05-01 07:20:00 NA TRUE 1 0
#> 6 2018-05-01 07:25:00 NA TRUE 1 0
#> 7 2018-05-01 07:30:00 NA TRUE 1 0
#> 8 2018-05-01 07:35:00 NA TRUE 1 0
#> 9 2018-05-01 07:40:00 NA TRUE 1 0
#> 10 2018-05-01 07:45:00 NA TRUE 1 0
#> # ... with 1,887 more rows, and 6 more variables: switch_direction <chr>,
#> # status_duration <dbl>, total_prev <dbl>, time <dbl>, day <int>,
#> # sim_hours <dbl>Bitvector Sequence Format
It is common to represent sleep history information as a bitvector (i.e., a sequence of 1’s and 0’s). In bitvectors, sleep and wake times are represented by 1’s or 0’s, with each bit representing an equal time duration (e.g., 1 minute in that status). The bitvector sequence also must be relative to a start datetime. This format is commonly outputted from actigraphy devices (a form of wearable sleep tracker) and corresponding sleep detection algorithms (e.g., Cole-Kripke). Other proprietary software packages (e.g., SAFTE-FAST) require imported data to be in bit vector form (though the exact required formats do vary).
FIPS expects bitvectors to repesent sleep as
0 and wake as 1. The
parse_sleepwake_sequence function can transform a bit
vector sequence to a compliant FIPS_df. Below, an example
of this data and steps to transform are provided, however, note that we
do not use this dataframe again within this vignette.
# Simulation start date time (i.e., when you want first predictions to begin)
simulation.start = lubridate::ymd_hms('2018-05-01 10:00:00', tz = "Australia/Perth")
# Example bitvector sequence. This typically would be imported directly via a textfile.
# Here we generate, though typically this would be returned by a ReadLines/read.delim/read.table
bv.sleep.sequence = rep(rep(c(1,0), 6), sample(20:40, 12))
bv.sim.dataframe = parse_sleepwake_sequence(
seq = bv.sleep.sequence,
series.start = simulation.start,
epoch = 15)
print(bv.sim.dataframe)
#> datetime sleep.id time day sim_hours wake_status_int
#> 1 2018-05-01 10:00:00 NA 10.00 1 0.00 1
#> 2 2018-05-01 10:15:00 NA 10.25 1 0.25 1
#> 3 2018-05-01 10:30:00 NA 10.50 1 0.50 1
#> 4 2018-05-01 10:45:00 NA 10.75 1 0.75 1
#> 5 2018-05-01 11:00:00 NA 11.00 1 1.00 1
#> 6 2018-05-01 11:15:00 NA 11.25 1 1.25 1
#> 7 2018-05-01 11:30:00 NA 11.50 1 1.50 1
#> 8 2018-05-01 11:45:00 NA 11.75 1 1.75 1
#> 9 2018-05-01 12:00:00 NA 12.00 1 2.00 1
#> 10 2018-05-01 12:15:00 NA 12.25 1 2.25 1
#> 11 2018-05-01 12:30:00 NA 12.50 1 2.50 1
#> 12 2018-05-01 12:45:00 NA 12.75 1 2.75 1
#> 13 2018-05-01 13:00:00 NA 13.00 1 3.00 1
#> 14 2018-05-01 13:15:00 NA 13.25 1 3.25 1
#> 15 2018-05-01 13:30:00 NA 13.50 1 3.50 1
#> 16 2018-05-01 13:45:00 NA 13.75 1 3.75 1
#> 17 2018-05-01 14:00:00 NA 14.00 1 4.00 1
#> 18 2018-05-01 14:15:00 NA 14.25 1 4.25 1
#> 19 2018-05-01 14:30:00 NA 14.50 1 4.50 1
#> 20 2018-05-01 14:45:00 NA 14.75 1 4.75 1
#> 21 2018-05-01 15:00:00 NA 15.00 1 5.00 1
#> 22 2018-05-01 15:15:00 NA 15.25 1 5.25 1
#> 23 2018-05-01 15:30:00 NA 15.50 1 5.50 1
#> 24 2018-05-01 15:45:00 NA 15.75 1 5.75 1
#> 25 2018-05-01 16:00:00 NA 16.00 1 6.00 1
#> 26 2018-05-01 16:15:00 NA 16.25 1 6.25 1
#> 27 2018-05-01 16:30:00 1 16.50 1 6.50 0
#> 28 2018-05-01 16:45:00 1 16.75 1 6.75 0
#> 29 2018-05-01 17:00:00 1 17.00 1 7.00 0
#> 30 2018-05-01 17:15:00 1 17.25 1 7.25 0
#> 31 2018-05-01 17:30:00 1 17.50 1 7.50 0
#> 32 2018-05-01 17:45:00 1 17.75 1 7.75 0
#> 33 2018-05-01 18:00:00 1 18.00 1 8.00 0
#> 34 2018-05-01 18:15:00 1 18.25 1 8.25 0
#> 35 2018-05-01 18:30:00 1 18.50 1 8.50 0
#> 36 2018-05-01 18:45:00 1 18.75 1 8.75 0
#> 37 2018-05-01 19:00:00 1 19.00 1 9.00 0
#> 38 2018-05-01 19:15:00 1 19.25 1 9.25 0
#> 39 2018-05-01 19:30:00 1 19.50 1 9.50 0
#> 40 2018-05-01 19:45:00 1 19.75 1 9.75 0
#> 41 2018-05-01 20:00:00 1 20.00 1 10.00 0
#> 42 2018-05-01 20:15:00 1 20.25 1 10.25 0
#> 43 2018-05-01 20:30:00 1 20.50 1 10.50 0
#> 44 2018-05-01 20:45:00 1 20.75 1 10.75 0
#> 45 2018-05-01 21:00:00 1 21.00 1 11.00 0
#> 46 2018-05-01 21:15:00 1 21.25 1 11.25 0
#> 47 2018-05-01 21:30:00 1 21.50 1 11.50 0
#> 48 2018-05-01 21:45:00 1 21.75 1 11.75 0
#> 49 2018-05-01 22:00:00 1 22.00 1 12.00 0
#> 50 2018-05-01 22:15:00 1 22.25 1 12.25 0
#> 51 2018-05-01 22:30:00 1 22.50 1 12.50 0
#> 52 2018-05-01 22:45:00 1 22.75 1 12.75 0
#> 53 2018-05-01 23:00:00 1 23.00 1 13.00 0
#> 54 2018-05-01 23:15:00 NA 23.25 1 13.25 1
#> 55 2018-05-01 23:30:00 NA 23.50 1 13.50 1
#> 56 2018-05-01 23:45:00 NA 23.75 1 13.75 1
#> 57 2018-05-02 00:00:00 NA 0.00 2 14.00 1
#> 58 2018-05-02 00:15:00 NA 0.25 2 14.25 1
#> 59 2018-05-02 00:30:00 NA 0.50 2 14.50 1
#> 60 2018-05-02 00:45:00 NA 0.75 2 14.75 1
#> 61 2018-05-02 01:00:00 NA 1.00 2 15.00 1
#> 62 2018-05-02 01:15:00 NA 1.25 2 15.25 1
#> 63 2018-05-02 01:30:00 NA 1.50 2 15.50 1
#> 64 2018-05-02 01:45:00 NA 1.75 2 15.75 1
#> 65 2018-05-02 02:00:00 NA 2.00 2 16.00 1
#> 66 2018-05-02 02:15:00 NA 2.25 2 16.25 1
#> 67 2018-05-02 02:30:00 NA 2.50 2 16.50 1
#> 68 2018-05-02 02:45:00 NA 2.75 2 16.75 1
#> 69 2018-05-02 03:00:00 NA 3.00 2 17.00 1
#> 70 2018-05-02 03:15:00 NA 3.25 2 17.25 1
#> 71 2018-05-02 03:30:00 NA 3.50 2 17.50 1
#> 72 2018-05-02 03:45:00 NA 3.75 2 17.75 1
#> 73 2018-05-02 04:00:00 NA 4.00 2 18.00 1
#> 74 2018-05-02 04:15:00 NA 4.25 2 18.25 1
#> 75 2018-05-02 04:30:00 NA 4.50 2 18.50 1
#> 76 2018-05-02 04:45:00 NA 4.75 2 18.75 1
#> 77 2018-05-02 05:00:00 NA 5.00 2 19.00 1
#> 78 2018-05-02 05:15:00 NA 5.25 2 19.25 1
#> 79 2018-05-02 05:30:00 NA 5.50 2 19.50 1
#> 80 2018-05-02 05:45:00 NA 5.75 2 19.75 1
#> 81 2018-05-02 06:00:00 NA 6.00 2 20.00 1
#> 82 2018-05-02 06:15:00 NA 6.25 2 20.25 1
#> 83 2018-05-02 06:30:00 NA 6.50 2 20.50 1
#> 84 2018-05-02 06:45:00 NA 6.75 2 20.75 1
#> 85 2018-05-02 07:00:00 NA 7.00 2 21.00 1
#> 86 2018-05-02 07:15:00 NA 7.25 2 21.25 1
#> 87 2018-05-02 07:30:00 NA 7.50 2 21.50 1
#> 88 2018-05-02 07:45:00 NA 7.75 2 21.75 1
#> 89 2018-05-02 08:00:00 NA 8.00 2 22.00 1
#> 90 2018-05-02 08:15:00 NA 8.25 2 22.25 1
#> 91 2018-05-02 08:30:00 NA 8.50 2 22.50 1
#> 92 2018-05-02 08:45:00 2 8.75 2 22.75 0
#> 93 2018-05-02 09:00:00 2 9.00 2 23.00 0
#> 94 2018-05-02 09:15:00 2 9.25 2 23.25 0
#> 95 2018-05-02 09:30:00 2 9.50 2 23.50 0
#> 96 2018-05-02 09:45:00 2 9.75 2 23.75 0
#> 97 2018-05-02 10:00:00 2 10.00 2 24.00 0
#> 98 2018-05-02 10:15:00 2 10.25 2 24.25 0
#> 99 2018-05-02 10:30:00 2 10.50 2 24.50 0
#> 100 2018-05-02 10:45:00 2 10.75 2 24.75 0
#> 101 2018-05-02 11:00:00 2 11.00 2 25.00 0
#> 102 2018-05-02 11:15:00 2 11.25 2 25.25 0
#> 103 2018-05-02 11:30:00 2 11.50 2 25.50 0
#> 104 2018-05-02 11:45:00 2 11.75 2 25.75 0
#> 105 2018-05-02 12:00:00 2 12.00 2 26.00 0
#> 106 2018-05-02 12:15:00 2 12.25 2 26.25 0
#> 107 2018-05-02 12:30:00 2 12.50 2 26.50 0
#> 108 2018-05-02 12:45:00 2 12.75 2 26.75 0
#> 109 2018-05-02 13:00:00 2 13.00 2 27.00 0
#> 110 2018-05-02 13:15:00 2 13.25 2 27.25 0
#> 111 2018-05-02 13:30:00 2 13.50 2 27.50 0
#> 112 2018-05-02 13:45:00 2 13.75 2 27.75 0
#> 113 2018-05-02 14:00:00 2 14.00 2 28.00 0
#> 114 2018-05-02 14:15:00 2 14.25 2 28.25 0
#> 115 2018-05-02 14:30:00 2 14.50 2 28.50 0
#> 116 2018-05-02 14:45:00 2 14.75 2 28.75 0
#> 117 2018-05-02 15:00:00 NA 15.00 2 29.00 1
#> 118 2018-05-02 15:15:00 NA 15.25 2 29.25 1
#> 119 2018-05-02 15:30:00 NA 15.50 2 29.50 1
#> 120 2018-05-02 15:45:00 NA 15.75 2 29.75 1
#> 121 2018-05-02 16:00:00 NA 16.00 2 30.00 1
#> 122 2018-05-02 16:15:00 NA 16.25 2 30.25 1
#> 123 2018-05-02 16:30:00 NA 16.50 2 30.50 1
#> 124 2018-05-02 16:45:00 NA 16.75 2 30.75 1
#> 125 2018-05-02 17:00:00 NA 17.00 2 31.00 1
#> 126 2018-05-02 17:15:00 NA 17.25 2 31.25 1
#> 127 2018-05-02 17:30:00 NA 17.50 2 31.50 1
#> 128 2018-05-02 17:45:00 NA 17.75 2 31.75 1
#> 129 2018-05-02 18:00:00 NA 18.00 2 32.00 1
#> 130 2018-05-02 18:15:00 NA 18.25 2 32.25 1
#> 131 2018-05-02 18:30:00 NA 18.50 2 32.50 1
#> 132 2018-05-02 18:45:00 NA 18.75 2 32.75 1
#> 133 2018-05-02 19:00:00 NA 19.00 2 33.00 1
#> 134 2018-05-02 19:15:00 NA 19.25 2 33.25 1
#> 135 2018-05-02 19:30:00 NA 19.50 2 33.50 1
#> 136 2018-05-02 19:45:00 NA 19.75 2 33.75 1
#> 137 2018-05-02 20:00:00 NA 20.00 2 34.00 1
#> 138 2018-05-02 20:15:00 NA 20.25 2 34.25 1
#> 139 2018-05-02 20:30:00 NA 20.50 2 34.50 1
#> 140 2018-05-02 20:45:00 NA 20.75 2 34.75 1
#> 141 2018-05-02 21:00:00 3 21.00 2 35.00 0
#> 142 2018-05-02 21:15:00 3 21.25 2 35.25 0
#> 143 2018-05-02 21:30:00 3 21.50 2 35.50 0
#> 144 2018-05-02 21:45:00 3 21.75 2 35.75 0
#> 145 2018-05-02 22:00:00 3 22.00 2 36.00 0
#> 146 2018-05-02 22:15:00 3 22.25 2 36.25 0
#> 147 2018-05-02 22:30:00 3 22.50 2 36.50 0
#> 148 2018-05-02 22:45:00 3 22.75 2 36.75 0
#> 149 2018-05-02 23:00:00 3 23.00 2 37.00 0
#> 150 2018-05-02 23:15:00 3 23.25 2 37.25 0
#> 151 2018-05-02 23:30:00 3 23.50 2 37.50 0
#> 152 2018-05-02 23:45:00 3 23.75 2 37.75 0
#> 153 2018-05-03 00:00:00 3 0.00 3 38.00 0
#> 154 2018-05-03 00:15:00 3 0.25 3 38.25 0
#> 155 2018-05-03 00:30:00 3 0.50 3 38.50 0
#> 156 2018-05-03 00:45:00 3 0.75 3 38.75 0
#> 157 2018-05-03 01:00:00 3 1.00 3 39.00 0
#> 158 2018-05-03 01:15:00 3 1.25 3 39.25 0
#> 159 2018-05-03 01:30:00 3 1.50 3 39.50 0
#> 160 2018-05-03 01:45:00 3 1.75 3 39.75 0
#> 161 2018-05-03 02:00:00 3 2.00 3 40.00 0
#> 162 2018-05-03 02:15:00 3 2.25 3 40.25 0
#> 163 2018-05-03 02:30:00 3 2.50 3 40.50 0
#> 164 2018-05-03 02:45:00 3 2.75 3 40.75 0
#> 165 2018-05-03 03:00:00 3 3.00 3 41.00 0
#> 166 2018-05-03 03:15:00 3 3.25 3 41.25 0
#> 167 2018-05-03 03:30:00 3 3.50 3 41.50 0
#> 168 2018-05-03 03:45:00 3 3.75 3 41.75 0
#> 169 2018-05-03 04:00:00 3 4.00 3 42.00 0
#> 170 2018-05-03 04:15:00 3 4.25 3 42.25 0
#> 171 2018-05-03 04:30:00 3 4.50 3 42.50 0
#> 172 2018-05-03 04:45:00 3 4.75 3 42.75 0
#> 173 2018-05-03 05:00:00 3 5.00 3 43.00 0
#> 174 2018-05-03 05:15:00 3 5.25 3 43.25 0
#> 175 2018-05-03 05:30:00 3 5.50 3 43.50 0
#> 176 2018-05-03 05:45:00 3 5.75 3 43.75 0
#> 177 2018-05-03 06:00:00 3 6.00 3 44.00 0
#> 178 2018-05-03 06:15:00 NA 6.25 3 44.25 1
#> 179 2018-05-03 06:30:00 NA 6.50 3 44.50 1
#> 180 2018-05-03 06:45:00 NA 6.75 3 44.75 1
#> 181 2018-05-03 07:00:00 NA 7.00 3 45.00 1
#> 182 2018-05-03 07:15:00 NA 7.25 3 45.25 1
#> 183 2018-05-03 07:30:00 NA 7.50 3 45.50 1
#> 184 2018-05-03 07:45:00 NA 7.75 3 45.75 1
#> 185 2018-05-03 08:00:00 NA 8.00 3 46.00 1
#> 186 2018-05-03 08:15:00 NA 8.25 3 46.25 1
#> 187 2018-05-03 08:30:00 NA 8.50 3 46.50 1
#> 188 2018-05-03 08:45:00 NA 8.75 3 46.75 1
#> 189 2018-05-03 09:00:00 NA 9.00 3 47.00 1
#> 190 2018-05-03 09:15:00 NA 9.25 3 47.25 1
#> 191 2018-05-03 09:30:00 NA 9.50 3 47.50 1
#> 192 2018-05-03 09:45:00 NA 9.75 3 47.75 1
#> 193 2018-05-03 10:00:00 NA 10.00 3 48.00 1
#> 194 2018-05-03 10:15:00 NA 10.25 3 48.25 1
#> 195 2018-05-03 10:30:00 NA 10.50 3 48.50 1
#> 196 2018-05-03 10:45:00 NA 10.75 3 48.75 1
#> 197 2018-05-03 11:00:00 NA 11.00 3 49.00 1
#> 198 2018-05-03 11:15:00 NA 11.25 3 49.25 1
#> 199 2018-05-03 11:30:00 NA 11.50 3 49.50 1
#> 200 2018-05-03 11:45:00 NA 11.75 3 49.75 1
#> 201 2018-05-03 12:00:00 NA 12.00 3 50.00 1
#> 202 2018-05-03 12:15:00 NA 12.25 3 50.25 1
#> 203 2018-05-03 12:30:00 NA 12.50 3 50.50 1
#> 204 2018-05-03 12:45:00 NA 12.75 3 50.75 1
#> 205 2018-05-03 13:00:00 NA 13.00 3 51.00 1
#> 206 2018-05-03 13:15:00 NA 13.25 3 51.25 1
#> 207 2018-05-03 13:30:00 NA 13.50 3 51.50 1
#> 208 2018-05-03 13:45:00 NA 13.75 3 51.75 1
#> 209 2018-05-03 14:00:00 NA 14.00 3 52.00 1
#> 210 2018-05-03 14:15:00 NA 14.25 3 52.25 1
#> 211 2018-05-03 14:30:00 NA 14.50 3 52.50 1
#> 212 2018-05-03 14:45:00 NA 14.75 3 52.75 1
#> 213 2018-05-03 15:00:00 NA 15.00 3 53.00 1
#> 214 2018-05-03 15:15:00 NA 15.25 3 53.25 1
#> 215 2018-05-03 15:30:00 NA 15.50 3 53.50 1
#> 216 2018-05-03 15:45:00 NA 15.75 3 53.75 1
#> 217 2018-05-03 16:00:00 4 16.00 3 54.00 0
#> 218 2018-05-03 16:15:00 4 16.25 3 54.25 0
#> 219 2018-05-03 16:30:00 4 16.50 3 54.50 0
#> 220 2018-05-03 16:45:00 4 16.75 3 54.75 0
#> 221 2018-05-03 17:00:00 4 17.00 3 55.00 0
#> 222 2018-05-03 17:15:00 4 17.25 3 55.25 0
#> 223 2018-05-03 17:30:00 4 17.50 3 55.50 0
#> 224 2018-05-03 17:45:00 4 17.75 3 55.75 0
#> 225 2018-05-03 18:00:00 4 18.00 3 56.00 0
#> 226 2018-05-03 18:15:00 4 18.25 3 56.25 0
#> 227 2018-05-03 18:30:00 4 18.50 3 56.50 0
#> 228 2018-05-03 18:45:00 4 18.75 3 56.75 0
#> 229 2018-05-03 19:00:00 4 19.00 3 57.00 0
#> 230 2018-05-03 19:15:00 4 19.25 3 57.25 0
#> 231 2018-05-03 19:30:00 4 19.50 3 57.50 0
#> 232 2018-05-03 19:45:00 4 19.75 3 57.75 0
#> 233 2018-05-03 20:00:00 4 20.00 3 58.00 0
#> 234 2018-05-03 20:15:00 4 20.25 3 58.25 0
#> 235 2018-05-03 20:30:00 4 20.50 3 58.50 0
#> 236 2018-05-03 20:45:00 4 20.75 3 58.75 0
#> 237 2018-05-03 21:00:00 4 21.00 3 59.00 0
#> 238 2018-05-03 21:15:00 NA 21.25 3 59.25 1
#> 239 2018-05-03 21:30:00 NA 21.50 3 59.50 1
#> 240 2018-05-03 21:45:00 NA 21.75 3 59.75 1
#> 241 2018-05-03 22:00:00 NA 22.00 3 60.00 1
#> 242 2018-05-03 22:15:00 NA 22.25 3 60.25 1
#> 243 2018-05-03 22:30:00 NA 22.50 3 60.50 1
#> 244 2018-05-03 22:45:00 NA 22.75 3 60.75 1
#> 245 2018-05-03 23:00:00 NA 23.00 3 61.00 1
#> 246 2018-05-03 23:15:00 NA 23.25 3 61.25 1
#> 247 2018-05-03 23:30:00 NA 23.50 3 61.50 1
#> 248 2018-05-03 23:45:00 NA 23.75 3 61.75 1
#> 249 2018-05-04 00:00:00 NA 0.00 4 62.00 1
#> 250 2018-05-04 00:15:00 NA 0.25 4 62.25 1
#> 251 2018-05-04 00:30:00 NA 0.50 4 62.50 1
#> 252 2018-05-04 00:45:00 NA 0.75 4 62.75 1
#> 253 2018-05-04 01:00:00 NA 1.00 4 63.00 1
#> 254 2018-05-04 01:15:00 NA 1.25 4 63.25 1
#> 255 2018-05-04 01:30:00 NA 1.50 4 63.50 1
#> 256 2018-05-04 01:45:00 NA 1.75 4 63.75 1
#> 257 2018-05-04 02:00:00 NA 2.00 4 64.00 1
#> 258 2018-05-04 02:15:00 NA 2.25 4 64.25 1
#> 259 2018-05-04 02:30:00 NA 2.50 4 64.50 1
#> 260 2018-05-04 02:45:00 NA 2.75 4 64.75 1
#> 261 2018-05-04 03:00:00 NA 3.00 4 65.00 1
#> 262 2018-05-04 03:15:00 NA 3.25 4 65.25 1
#> 263 2018-05-04 03:30:00 NA 3.50 4 65.50 1
#> 264 2018-05-04 03:45:00 NA 3.75 4 65.75 1
#> 265 2018-05-04 04:00:00 NA 4.00 4 66.00 1
#> 266 2018-05-04 04:15:00 NA 4.25 4 66.25 1
#> 267 2018-05-04 04:30:00 NA 4.50 4 66.50 1
#> 268 2018-05-04 04:45:00 NA 4.75 4 66.75 1
#> 269 2018-05-04 05:00:00 5 5.00 4 67.00 0
#> 270 2018-05-04 05:15:00 5 5.25 4 67.25 0
#> 271 2018-05-04 05:30:00 5 5.50 4 67.50 0
#> 272 2018-05-04 05:45:00 5 5.75 4 67.75 0
#> 273 2018-05-04 06:00:00 5 6.00 4 68.00 0
#> 274 2018-05-04 06:15:00 5 6.25 4 68.25 0
#> 275 2018-05-04 06:30:00 5 6.50 4 68.50 0
#> 276 2018-05-04 06:45:00 5 6.75 4 68.75 0
#> 277 2018-05-04 07:00:00 5 7.00 4 69.00 0
#> 278 2018-05-04 07:15:00 5 7.25 4 69.25 0
#> 279 2018-05-04 07:30:00 5 7.50 4 69.50 0
#> 280 2018-05-04 07:45:00 5 7.75 4 69.75 0
#> 281 2018-05-04 08:00:00 5 8.00 4 70.00 0
#> 282 2018-05-04 08:15:00 5 8.25 4 70.25 0
#> 283 2018-05-04 08:30:00 5 8.50 4 70.50 0
#> 284 2018-05-04 08:45:00 5 8.75 4 70.75 0
#> 285 2018-05-04 09:00:00 5 9.00 4 71.00 0
#> 286 2018-05-04 09:15:00 5 9.25 4 71.25 0
#> 287 2018-05-04 09:30:00 5 9.50 4 71.50 0
#> 288 2018-05-04 09:45:00 5 9.75 4 71.75 0
#> 289 2018-05-04 10:00:00 5 10.00 4 72.00 0
#> 290 2018-05-04 10:15:00 5 10.25 4 72.25 0
#> 291 2018-05-04 10:30:00 5 10.50 4 72.50 0
#> 292 2018-05-04 10:45:00 5 10.75 4 72.75 0
#> 293 2018-05-04 11:00:00 5 11.00 4 73.00 0
#> 294 2018-05-04 11:15:00 5 11.25 4 73.25 0
#> 295 2018-05-04 11:30:00 5 11.50 4 73.50 0
#> 296 2018-05-04 11:45:00 5 11.75 4 73.75 0
#> 297 2018-05-04 12:00:00 5 12.00 4 74.00 0
#> 298 2018-05-04 12:15:00 5 12.25 4 74.25 0
#> 299 2018-05-04 12:30:00 5 12.50 4 74.50 0
#> 300 2018-05-04 12:45:00 5 12.75 4 74.75 0
#> 301 2018-05-04 13:00:00 5 13.00 4 75.00 0
#> 302 2018-05-04 13:15:00 5 13.25 4 75.25 0
#> 303 2018-05-04 13:30:00 5 13.50 4 75.50 0
#> 304 2018-05-04 13:45:00 NA 13.75 4 75.75 1
#> 305 2018-05-04 14:00:00 NA 14.00 4 76.00 1
#> 306 2018-05-04 14:15:00 NA 14.25 4 76.25 1
#> 307 2018-05-04 14:30:00 NA 14.50 4 76.50 1
#> 308 2018-05-04 14:45:00 NA 14.75 4 76.75 1
#> 309 2018-05-04 15:00:00 NA 15.00 4 77.00 1
#> 310 2018-05-04 15:15:00 NA 15.25 4 77.25 1
#> 311 2018-05-04 15:30:00 NA 15.50 4 77.50 1
#> 312 2018-05-04 15:45:00 NA 15.75 4 77.75 1
#> 313 2018-05-04 16:00:00 NA 16.00 4 78.00 1
#> 314 2018-05-04 16:15:00 NA 16.25 4 78.25 1
#> 315 2018-05-04 16:30:00 NA 16.50 4 78.50 1
#> 316 2018-05-04 16:45:00 NA 16.75 4 78.75 1
#> 317 2018-05-04 17:00:00 NA 17.00 4 79.00 1
#> 318 2018-05-04 17:15:00 NA 17.25 4 79.25 1
#> 319 2018-05-04 17:30:00 NA 17.50 4 79.50 1
#> 320 2018-05-04 17:45:00 NA 17.75 4 79.75 1
#> 321 2018-05-04 18:00:00 NA 18.00 4 80.00 1
#> 322 2018-05-04 18:15:00 NA 18.25 4 80.25 1
#> 323 2018-05-04 18:30:00 NA 18.50 4 80.50 1
#> 324 2018-05-04 18:45:00 NA 18.75 4 80.75 1
#> 325 2018-05-04 19:00:00 NA 19.00 4 81.00 1
#> 326 2018-05-04 19:15:00 NA 19.25 4 81.25 1
#> 327 2018-05-04 19:30:00 6 19.50 4 81.50 0
#> 328 2018-05-04 19:45:00 6 19.75 4 81.75 0
#> 329 2018-05-04 20:00:00 6 20.00 4 82.00 0
#> 330 2018-05-04 20:15:00 6 20.25 4 82.25 0
#> 331 2018-05-04 20:30:00 6 20.50 4 82.50 0
#> 332 2018-05-04 20:45:00 6 20.75 4 82.75 0
#> 333 2018-05-04 21:00:00 6 21.00 4 83.00 0
#> 334 2018-05-04 21:15:00 6 21.25 4 83.25 0
#> 335 2018-05-04 21:30:00 6 21.50 4 83.50 0
#> 336 2018-05-04 21:45:00 6 21.75 4 83.75 0
#> 337 2018-05-04 22:00:00 6 22.00 4 84.00 0
#> 338 2018-05-04 22:15:00 6 22.25 4 84.25 0
#> 339 2018-05-04 22:30:00 6 22.50 4 84.50 0
#> 340 2018-05-04 22:45:00 6 22.75 4 84.75 0
#> 341 2018-05-04 23:00:00 6 23.00 4 85.00 0
#> 342 2018-05-04 23:15:00 6 23.25 4 85.25 0
#> 343 2018-05-04 23:30:00 6 23.50 4 85.50 0
#> 344 2018-05-04 23:45:00 6 23.75 4 85.75 0
#> 345 2018-05-05 00:00:00 6 0.00 5 86.00 0
#> 346 2018-05-05 00:15:00 6 0.25 5 86.25 0
#> 347 2018-05-05 00:30:00 6 0.50 5 86.50 0
#> 348 2018-05-05 00:45:00 6 0.75 5 86.75 0
#> 349 2018-05-05 01:00:00 6 1.00 5 87.00 0
#> 350 2018-05-05 01:15:00 6 1.25 5 87.25 0
#> 351 2018-05-05 01:30:00 6 1.50 5 87.50 0
#> 352 2018-05-05 01:45:00 6 1.75 5 87.75 0
#> 353 2018-05-05 02:00:00 6 2.00 5 88.00 0
#> 354 2018-05-05 02:15:00 6 2.25 5 88.25 0
#> 355 2018-05-05 02:30:00 6 2.50 5 88.50 0
#> 356 2018-05-05 02:45:00 6 2.75 5 88.75 0
#> wake_status change_point switch_direction status_duration total_prev
#> 1 TRUE 0 0 0.00 0.00
#> 2 TRUE 0 0 0.25 0.00
#> 3 TRUE 0 0 0.50 0.00
#> 4 TRUE 0 0 0.75 0.00
#> 5 TRUE 0 0 1.00 0.00
#> 6 TRUE 0 0 1.25 0.00
#> 7 TRUE 0 0 1.50 0.00
#> 8 TRUE 0 0 1.75 0.00
#> 9 TRUE 0 0 2.00 0.00
#> 10 TRUE 0 0 2.25 0.00
#> 11 TRUE 0 0 2.50 0.00
#> 12 TRUE 0 0 2.75 0.00
#> 13 TRUE 0 0 3.00 0.00
#> 14 TRUE 0 0 3.25 0.00
#> 15 TRUE 0 0 3.50 0.00
#> 16 TRUE 0 0 3.75 0.00
#> 17 TRUE 0 0 4.00 0.00
#> 18 TRUE 0 0 4.25 0.00
#> 19 TRUE 0 0 4.50 0.00
#> 20 TRUE 0 0 4.75 0.00
#> 21 TRUE 0 0 5.00 0.00
#> 22 TRUE 0 0 5.25 0.00
#> 23 TRUE 0 0 5.50 0.00
#> 24 TRUE 0 0 5.75 0.00
#> 25 TRUE 0 0 6.00 0.00
#> 26 TRUE 0 0 6.25 0.00
#> 27 FALSE 1 Sleep 0.00 6.50
#> 28 FALSE 0 0 0.25 0.00
#> 29 FALSE 0 0 0.50 0.00
#> 30 FALSE 0 0 0.75 0.00
#> 31 FALSE 0 0 1.00 0.00
#> 32 FALSE 0 0 1.25 0.00
#> 33 FALSE 0 0 1.50 0.00
#> 34 FALSE 0 0 1.75 0.00
#> 35 FALSE 0 0 2.00 0.00
#> 36 FALSE 0 0 2.25 0.00
#> 37 FALSE 0 0 2.50 0.00
#> 38 FALSE 0 0 2.75 0.00
#> 39 FALSE 0 0 3.00 0.00
#> 40 FALSE 0 0 3.25 0.00
#> 41 FALSE 0 0 3.50 0.00
#> 42 FALSE 0 0 3.75 0.00
#> 43 FALSE 0 0 4.00 0.00
#> 44 FALSE 0 0 4.25 0.00
#> 45 FALSE 0 0 4.50 0.00
#> 46 FALSE 0 0 4.75 0.00
#> 47 FALSE 0 0 5.00 0.00
#> 48 FALSE 0 0 5.25 0.00
#> 49 FALSE 0 0 5.50 0.00
#> 50 FALSE 0 0 5.75 0.00
#> 51 FALSE 0 0 6.00 0.00
#> 52 FALSE 0 0 6.25 0.00
#> 53 FALSE 0 0 6.50 0.00
#> 54 TRUE 1 Wake 0.00 6.75
#> 55 TRUE 0 0 0.25 0.00
#> 56 TRUE 0 0 0.50 0.00
#> 57 TRUE 0 0 0.75 0.00
#> 58 TRUE 0 0 1.00 0.00
#> 59 TRUE 0 0 1.25 0.00
#> 60 TRUE 0 0 1.50 0.00
#> 61 TRUE 0 0 1.75 0.00
#> 62 TRUE 0 0 2.00 0.00
#> 63 TRUE 0 0 2.25 0.00
#> 64 TRUE 0 0 2.50 0.00
#> 65 TRUE 0 0 2.75 0.00
#> 66 TRUE 0 0 3.00 0.00
#> 67 TRUE 0 0 3.25 0.00
#> 68 TRUE 0 0 3.50 0.00
#> 69 TRUE 0 0 3.75 0.00
#> 70 TRUE 0 0 4.00 0.00
#> 71 TRUE 0 0 4.25 0.00
#> 72 TRUE 0 0 4.50 0.00
#> 73 TRUE 0 0 4.75 0.00
#> 74 TRUE 0 0 5.00 0.00
#> 75 TRUE 0 0 5.25 0.00
#> 76 TRUE 0 0 5.50 0.00
#> 77 TRUE 0 0 5.75 0.00
#> 78 TRUE 0 0 6.00 0.00
#> 79 TRUE 0 0 6.25 0.00
#> 80 TRUE 0 0 6.50 0.00
#> 81 TRUE 0 0 6.75 0.00
#> 82 TRUE 0 0 7.00 0.00
#> 83 TRUE 0 0 7.25 0.00
#> 84 TRUE 0 0 7.50 0.00
#> 85 TRUE 0 0 7.75 0.00
#> 86 TRUE 0 0 8.00 0.00
#> 87 TRUE 0 0 8.25 0.00
#> 88 TRUE 0 0 8.50 0.00
#> 89 TRUE 0 0 8.75 0.00
#> 90 TRUE 0 0 9.00 0.00
#> 91 TRUE 0 0 9.25 0.00
#> 92 FALSE 1 Sleep 0.00 9.50
#> 93 FALSE 0 0 0.25 0.00
#> 94 FALSE 0 0 0.50 0.00
#> 95 FALSE 0 0 0.75 0.00
#> 96 FALSE 0 0 1.00 0.00
#> 97 FALSE 0 0 1.25 0.00
#> 98 FALSE 0 0 1.50 0.00
#> 99 FALSE 0 0 1.75 0.00
#> 100 FALSE 0 0 2.00 0.00
#> 101 FALSE 0 0 2.25 0.00
#> 102 FALSE 0 0 2.50 0.00
#> 103 FALSE 0 0 2.75 0.00
#> 104 FALSE 0 0 3.00 0.00
#> 105 FALSE 0 0 3.25 0.00
#> 106 FALSE 0 0 3.50 0.00
#> 107 FALSE 0 0 3.75 0.00
#> 108 FALSE 0 0 4.00 0.00
#> 109 FALSE 0 0 4.25 0.00
#> 110 FALSE 0 0 4.50 0.00
#> 111 FALSE 0 0 4.75 0.00
#> 112 FALSE 0 0 5.00 0.00
#> 113 FALSE 0 0 5.25 0.00
#> 114 FALSE 0 0 5.50 0.00
#> 115 FALSE 0 0 5.75 0.00
#> 116 FALSE 0 0 6.00 0.00
#> 117 TRUE 1 Wake 0.00 6.25
#> 118 TRUE 0 0 0.25 0.00
#> 119 TRUE 0 0 0.50 0.00
#> 120 TRUE 0 0 0.75 0.00
#> 121 TRUE 0 0 1.00 0.00
#> 122 TRUE 0 0 1.25 0.00
#> 123 TRUE 0 0 1.50 0.00
#> 124 TRUE 0 0 1.75 0.00
#> 125 TRUE 0 0 2.00 0.00
#> 126 TRUE 0 0 2.25 0.00
#> 127 TRUE 0 0 2.50 0.00
#> 128 TRUE 0 0 2.75 0.00
#> 129 TRUE 0 0 3.00 0.00
#> 130 TRUE 0 0 3.25 0.00
#> 131 TRUE 0 0 3.50 0.00
#> 132 TRUE 0 0 3.75 0.00
#> 133 TRUE 0 0 4.00 0.00
#> 134 TRUE 0 0 4.25 0.00
#> 135 TRUE 0 0 4.50 0.00
#> 136 TRUE 0 0 4.75 0.00
#> 137 TRUE 0 0 5.00 0.00
#> 138 TRUE 0 0 5.25 0.00
#> 139 TRUE 0 0 5.50 0.00
#> 140 TRUE 0 0 5.75 0.00
#> 141 FALSE 1 Sleep 0.00 6.00
#> 142 FALSE 0 0 0.25 0.00
#> 143 FALSE 0 0 0.50 0.00
#> 144 FALSE 0 0 0.75 0.00
#> 145 FALSE 0 0 1.00 0.00
#> 146 FALSE 0 0 1.25 0.00
#> 147 FALSE 0 0 1.50 0.00
#> 148 FALSE 0 0 1.75 0.00
#> 149 FALSE 0 0 2.00 0.00
#> 150 FALSE 0 0 2.25 0.00
#> 151 FALSE 0 0 2.50 0.00
#> 152 FALSE 0 0 2.75 0.00
#> 153 FALSE 0 0 3.00 0.00
#> 154 FALSE 0 0 3.25 0.00
#> 155 FALSE 0 0 3.50 0.00
#> 156 FALSE 0 0 3.75 0.00
#> 157 FALSE 0 0 4.00 0.00
#> 158 FALSE 0 0 4.25 0.00
#> 159 FALSE 0 0 4.50 0.00
#> 160 FALSE 0 0 4.75 0.00
#> 161 FALSE 0 0 5.00 0.00
#> 162 FALSE 0 0 5.25 0.00
#> 163 FALSE 0 0 5.50 0.00
#> 164 FALSE 0 0 5.75 0.00
#> 165 FALSE 0 0 6.00 0.00
#> 166 FALSE 0 0 6.25 0.00
#> 167 FALSE 0 0 6.50 0.00
#> 168 FALSE 0 0 6.75 0.00
#> 169 FALSE 0 0 7.00 0.00
#> 170 FALSE 0 0 7.25 0.00
#> 171 FALSE 0 0 7.50 0.00
#> 172 FALSE 0 0 7.75 0.00
#> 173 FALSE 0 0 8.00 0.00
#> 174 FALSE 0 0 8.25 0.00
#> 175 FALSE 0 0 8.50 0.00
#> 176 FALSE 0 0 8.75 0.00
#> 177 FALSE 0 0 9.00 0.00
#> 178 TRUE 1 Wake 0.00 9.25
#> 179 TRUE 0 0 0.25 0.00
#> 180 TRUE 0 0 0.50 0.00
#> 181 TRUE 0 0 0.75 0.00
#> 182 TRUE 0 0 1.00 0.00
#> 183 TRUE 0 0 1.25 0.00
#> 184 TRUE 0 0 1.50 0.00
#> 185 TRUE 0 0 1.75 0.00
#> 186 TRUE 0 0 2.00 0.00
#> 187 TRUE 0 0 2.25 0.00
#> 188 TRUE 0 0 2.50 0.00
#> 189 TRUE 0 0 2.75 0.00
#> 190 TRUE 0 0 3.00 0.00
#> 191 TRUE 0 0 3.25 0.00
#> 192 TRUE 0 0 3.50 0.00
#> 193 TRUE 0 0 3.75 0.00
#> 194 TRUE 0 0 4.00 0.00
#> 195 TRUE 0 0 4.25 0.00
#> 196 TRUE 0 0 4.50 0.00
#> 197 TRUE 0 0 4.75 0.00
#> 198 TRUE 0 0 5.00 0.00
#> 199 TRUE 0 0 5.25 0.00
#> 200 TRUE 0 0 5.50 0.00
#> 201 TRUE 0 0 5.75 0.00
#> 202 TRUE 0 0 6.00 0.00
#> 203 TRUE 0 0 6.25 0.00
#> 204 TRUE 0 0 6.50 0.00
#> 205 TRUE 0 0 6.75 0.00
#> 206 TRUE 0 0 7.00 0.00
#> 207 TRUE 0 0 7.25 0.00
#> 208 TRUE 0 0 7.50 0.00
#> 209 TRUE 0 0 7.75 0.00
#> 210 TRUE 0 0 8.00 0.00
#> 211 TRUE 0 0 8.25 0.00
#> 212 TRUE 0 0 8.50 0.00
#> 213 TRUE 0 0 8.75 0.00
#> 214 TRUE 0 0 9.00 0.00
#> 215 TRUE 0 0 9.25 0.00
#> 216 TRUE 0 0 9.50 0.00
#> 217 FALSE 1 Sleep 0.00 9.75
#> 218 FALSE 0 0 0.25 0.00
#> 219 FALSE 0 0 0.50 0.00
#> 220 FALSE 0 0 0.75 0.00
#> 221 FALSE 0 0 1.00 0.00
#> 222 FALSE 0 0 1.25 0.00
#> 223 FALSE 0 0 1.50 0.00
#> 224 FALSE 0 0 1.75 0.00
#> 225 FALSE 0 0 2.00 0.00
#> 226 FALSE 0 0 2.25 0.00
#> 227 FALSE 0 0 2.50 0.00
#> 228 FALSE 0 0 2.75 0.00
#> 229 FALSE 0 0 3.00 0.00
#> 230 FALSE 0 0 3.25 0.00
#> 231 FALSE 0 0 3.50 0.00
#> 232 FALSE 0 0 3.75 0.00
#> 233 FALSE 0 0 4.00 0.00
#> 234 FALSE 0 0 4.25 0.00
#> 235 FALSE 0 0 4.50 0.00
#> 236 FALSE 0 0 4.75 0.00
#> 237 FALSE 0 0 5.00 0.00
#> 238 TRUE 1 Wake 0.00 5.25
#> 239 TRUE 0 0 0.25 0.00
#> 240 TRUE 0 0 0.50 0.00
#> 241 TRUE 0 0 0.75 0.00
#> 242 TRUE 0 0 1.00 0.00
#> 243 TRUE 0 0 1.25 0.00
#> 244 TRUE 0 0 1.50 0.00
#> 245 TRUE 0 0 1.75 0.00
#> 246 TRUE 0 0 2.00 0.00
#> 247 TRUE 0 0 2.25 0.00
#> 248 TRUE 0 0 2.50 0.00
#> 249 TRUE 0 0 2.75 0.00
#> 250 TRUE 0 0 3.00 0.00
#> 251 TRUE 0 0 3.25 0.00
#> 252 TRUE 0 0 3.50 0.00
#> 253 TRUE 0 0 3.75 0.00
#> 254 TRUE 0 0 4.00 0.00
#> 255 TRUE 0 0 4.25 0.00
#> 256 TRUE 0 0 4.50 0.00
#> 257 TRUE 0 0 4.75 0.00
#> 258 TRUE 0 0 5.00 0.00
#> 259 TRUE 0 0 5.25 0.00
#> 260 TRUE 0 0 5.50 0.00
#> 261 TRUE 0 0 5.75 0.00
#> 262 TRUE 0 0 6.00 0.00
#> 263 TRUE 0 0 6.25 0.00
#> 264 TRUE 0 0 6.50 0.00
#> 265 TRUE 0 0 6.75 0.00
#> 266 TRUE 0 0 7.00 0.00
#> 267 TRUE 0 0 7.25 0.00
#> 268 TRUE 0 0 7.50 0.00
#> 269 FALSE 1 Sleep 0.00 7.75
#> 270 FALSE 0 0 0.25 0.00
#> 271 FALSE 0 0 0.50 0.00
#> 272 FALSE 0 0 0.75 0.00
#> 273 FALSE 0 0 1.00 0.00
#> 274 FALSE 0 0 1.25 0.00
#> 275 FALSE 0 0 1.50 0.00
#> 276 FALSE 0 0 1.75 0.00
#> 277 FALSE 0 0 2.00 0.00
#> 278 FALSE 0 0 2.25 0.00
#> 279 FALSE 0 0 2.50 0.00
#> 280 FALSE 0 0 2.75 0.00
#> 281 FALSE 0 0 3.00 0.00
#> 282 FALSE 0 0 3.25 0.00
#> 283 FALSE 0 0 3.50 0.00
#> 284 FALSE 0 0 3.75 0.00
#> 285 FALSE 0 0 4.00 0.00
#> 286 FALSE 0 0 4.25 0.00
#> 287 FALSE 0 0 4.50 0.00
#> 288 FALSE 0 0 4.75 0.00
#> 289 FALSE 0 0 5.00 0.00
#> 290 FALSE 0 0 5.25 0.00
#> 291 FALSE 0 0 5.50 0.00
#> 292 FALSE 0 0 5.75 0.00
#> 293 FALSE 0 0 6.00 0.00
#> 294 FALSE 0 0 6.25 0.00
#> 295 FALSE 0 0 6.50 0.00
#> 296 FALSE 0 0 6.75 0.00
#> 297 FALSE 0 0 7.00 0.00
#> 298 FALSE 0 0 7.25 0.00
#> 299 FALSE 0 0 7.50 0.00
#> 300 FALSE 0 0 7.75 0.00
#> 301 FALSE 0 0 8.00 0.00
#> 302 FALSE 0 0 8.25 0.00
#> 303 FALSE 0 0 8.50 0.00
#> 304 TRUE 1 Wake 0.00 8.75
#> 305 TRUE 0 0 0.25 0.00
#> 306 TRUE 0 0 0.50 0.00
#> 307 TRUE 0 0 0.75 0.00
#> 308 TRUE 0 0 1.00 0.00
#> 309 TRUE 0 0 1.25 0.00
#> 310 TRUE 0 0 1.50 0.00
#> 311 TRUE 0 0 1.75 0.00
#> 312 TRUE 0 0 2.00 0.00
#> 313 TRUE 0 0 2.25 0.00
#> 314 TRUE 0 0 2.50 0.00
#> 315 TRUE 0 0 2.75 0.00
#> 316 TRUE 0 0 3.00 0.00
#> 317 TRUE 0 0 3.25 0.00
#> 318 TRUE 0 0 3.50 0.00
#> 319 TRUE 0 0 3.75 0.00
#> 320 TRUE 0 0 4.00 0.00
#> 321 TRUE 0 0 4.25 0.00
#> 322 TRUE 0 0 4.50 0.00
#> 323 TRUE 0 0 4.75 0.00
#> 324 TRUE 0 0 5.00 0.00
#> 325 TRUE 0 0 5.25 0.00
#> 326 TRUE 0 0 5.50 0.00
#> 327 FALSE 1 Sleep 0.00 5.75
#> 328 FALSE 0 0 0.25 0.00
#> 329 FALSE 0 0 0.50 0.00
#> 330 FALSE 0 0 0.75 0.00
#> 331 FALSE 0 0 1.00 0.00
#> 332 FALSE 0 0 1.25 0.00
#> 333 FALSE 0 0 1.50 0.00
#> 334 FALSE 0 0 1.75 0.00
#> 335 FALSE 0 0 2.00 0.00
#> 336 FALSE 0 0 2.25 0.00
#> 337 FALSE 0 0 2.50 0.00
#> 338 FALSE 0 0 2.75 0.00
#> 339 FALSE 0 0 3.00 0.00
#> 340 FALSE 0 0 3.25 0.00
#> 341 FALSE 0 0 3.50 0.00
#> 342 FALSE 0 0 3.75 0.00
#> 343 FALSE 0 0 4.00 0.00
#> 344 FALSE 0 0 4.25 0.00
#> 345 FALSE 0 0 4.50 0.00
#> 346 FALSE 0 0 4.75 0.00
#> 347 FALSE 0 0 5.00 0.00
#> 348 FALSE 0 0 5.25 0.00
#> 349 FALSE 0 0 5.50 0.00
#> 350 FALSE 0 0 5.75 0.00
#> 351 FALSE 0 0 6.00 0.00
#> 352 FALSE 0 0 6.25 0.00
#> 353 FALSE 0 0 6.50 0.00
#> 354 FALSE 0 0 6.75 0.00
#> 355 FALSE 0 0 7.00 0.00
#> 356 FALSE 0 0 7.25 0.00Modelling and the Simulation Dataframe
Now that you have generated the FIPS format, you can now apply the
FIPS simulation functions to this to actually run BMM predictions on the
series. In the example below, we will run a Three Process Model
simulation over the FIPS_df series we just created. To do
this, we will use the FIPS_simulation function, which takes
in three arguments: a FIPS_df object, a specification of a
modeltype (see help for model types currently implemented),
and a pvec which is a vector of parameters for the
model.
The parameter vectors provided by default in the package are those
reported by Ingre
et al. (2014). Please see the help files for citations and further
information. These defaults are customisable, and if you are using
Rstudio or Emacs ESS, you will get full autocompletion with descriptions
of each parameter (See help("TPM_make_pvec", "FIPS")).
We will use the default parameter vector for the Three process model
created by FIPS::TPM_make_pvec(), which takes on the
following values:
|
|
Calling FIPS_simulate() will the produces a
FIPS_df with model predictions and predicted time-varying
process variables (e.g., s, c). The returned
FIPS_df will also now inherit the
FIPS_simulation class. A FIPS_simulation
object has attributes containing the parameter vector, the
modeltype string, and the pvec used, and
several other values used internally. A custom print function of the
object will reveal all this information. Note that this print function
does mask some of the columns for ease of reading.
# Run a simulation with the three process model
TPM.simulation.results = FIPS_simulate(
FIPS_df = simulated.dataframe, # The FIPS_df
modeltype = "TPM", # three process model
pvec = TPM_make_pvec() # parameter vector
)
TPM.simulation.results
#> ---------
#> Model Type: TPM
#> Epoch Value: 5 minutes
#> Simulation duration: 158 hours
#> Time points: 1897
#> Parameters used (pvec input):
#> la ha d g bl
#> 2.4000000 14.3000000 -0.0353000 -0.3813564 12.2000000
#> Cm Ca p Um Ua
#> 0.0000000 2.5000000 16.8000000 -0.5000000 0.5000000
#> Wc Wd S0 KSS_intercept KSS_beta
#> -5.7200000 -1.5100000 7.9600000 10.6000000 -0.6000000
#> For descriptions of these parameters, inspect: help(FIPS::TPM_make_pvec)
#> ---------
#> # A tibble: 1,897 x 10
#> datetime time wake_status sim_hours s c w u
#> <dttm> <dbl> <lgl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 2018-05-01 07:00:00 7 TRUE 0 7.96 -2.10 -5.72 -0.0432
#> 2 2018-05-01 07:05:00 7.08 TRUE 0.0833 7.94 -2.07 -5.04 -0.0348
#> 3 2018-05-01 07:10:00 7.17 TRUE 0.167 7.93 -2.04 -4.45 -0.0272
#> 4 2018-05-01 07:15:00 7.25 TRUE 0.25 7.91 -2.00 -3.92 -0.0206
#> 5 2018-05-01 07:20:00 7.33 TRUE 0.333 7.89 -1.97 -3.46 -0.0149
#> 6 2018-05-01 07:25:00 7.42 TRUE 0.417 7.88 -1.94 -3.05 -0.0100
#> 7 2018-05-01 07:30:00 7.5 TRUE 0.5 7.86 -1.90 -2.69 -0.00616
#> 8 2018-05-01 07:35:00 7.58 TRUE 0.583 7.85 -1.87 -2.37 -0.00321
#> 9 2018-05-01 07:40:00 7.67 TRUE 0.667 7.83 -1.83 -2.09 -0.00122
#> 10 2018-05-01 07:45:00 7.75 TRUE 0.75 7.81 -1.79 -1.84 -0.000171
#> # ... with 1,887 more rows, and 2 more variables: KSS <dbl>, alertness <dbl>-
datetime= vector of datetime stamps separated at equidistance intervals. -
sleep.id= a supplementary variable indicating sleep episode identifier. -
wake_status= Awake (T) or asleep (F) at that epoch interval/epoch -
wake_status_int= Awake (1) or asleep (0) at that epoch interval/epoch -
change_point= Whether the individual changed wake status at that interval/epoch. -
switch_direction= Whether switch was to sleep or to wake -
status_duration= How long individual has been in status at that current time point -
total_prev= If a switch has occured, how long were that in the previous status. -
time= time of day in decimal hour -
day= days into simulation -
sim_hours= hours simulation has run for total -
s— Estimate of S (homeostatic) process at that time -
l— Estimate of l (lower asmytope) process at that time -
c— Estimate of C (24-hour circadian) process at that time -
w— Estimate of W (sleep intertia) process at that time -
u— Estimate of U (12-hour circadian) process at that time -
alertness— Currently justs + c + u -
KSS— This is equal to10.6 + -0.6 * (alertness)
Plotting
Plot 1: Parameter Plots
FIPS provides a default plot method to visualize model predictions and time-varying process estimates over time, discussed in detail in the plotting vignette. These plots aid with debugging and understanding how the different model parameters contribute to predictions.
#Narrow in on the first 90 hours
plot(TPM.simulation.results, plot_stat=c("alertness", "s", "c"),
from= '2018-05-01 23:00:00',
to= as_datetime('2018-05-01 23:00:00') + hours(90))
Plot 2: Heatmap Prediction Plots
There are many other ways that visualizations could help inform
theory and practice. The below heat plot is an example of how dangerous
times in a mission, according to the TPM predictions, could be
visualized. In the example below, the regions in deep orange indicate a
greatly increased risk of fatigue. However, the fact that the increased
fatigue occurs at the start of the mission indicates the initalisation
parameters (e.g., S0) are to blame, so this isn’t a cause
for significant concern.
if (!requireNamespace("colorspace", quietly = TRUE)) {
stop("Package \"colorspace\" needed for this example to work. Please install it.",
call. = FALSE)
}
plot3_w = TPM.simulation.results %>%
# Change epoch to 30 minutes for this plot (optional)
mutate(time = strftime(datetime, format = "%H:%M")) %>%
separate(time, into = c("plot_hour", "plot_minute"), ":") %>%
mutate(halfhour = if_else(plot_minute < 30, 0, 0.5) + as.numeric(plot_hour)) %>%
group_by(day, plot_hour, halfhour) %>%
summarise(KSS = mean(KSS), .groups = "keep") %>%
filter(day > 1) %>%
# Start Plot
ggplot(aes(day, halfhour, fill = KSS)) +
geom_tile(aes(fill = KSS), color = "white") +
scale_x_continuous(breaks = seq(0,7,1)) +
scale_y_continuous(breaks = seq(0,23,1)) +
labs(y = "Hour of Day", x = "Day of Mission") +
colorspace::scale_fill_continuous_divergingx(name = "KSS", palette = "Zissou 1", limits = c(1,9), mid = 5) +
ggtitle(label = "Three Process Model (KSS)", subtitle = "Heatmap of TPM predicted performance by 0.5hr") +
theme(axis.title = element_text(size = 12))
plot(plot3_w)