Create Grid Data for single subject from raw accelerometry

Usage

create_subject_predictor(
  data,
  max_signal = 3,
  bin_size = 0.25,
  lags = c(15, 30, 45)
)

Arguments

data: a data.frame with the columns time and vm
max_signal: maximum signal of vector magnitude to create grid over, grid will be 0 to max_signal by bin_size
bin_size: Size of the bins. max_signal must be evenly divisible by bin_size.
lags: the lag (in samples, not seconds) to compute bin inclusion.

Value

A data.frame of subject level lags and cut-values

Examples

df = example_walking %>%
   dplyr::mutate(time = lubridate::floor_date(Sys.time(), "seconds") + time)
create_subject_predictor(df)
#> # A tibble: 207 × 433
#>    second              `lag_15_[0,0.25]_[0,0.25]` `lag_15_[0,0.25]_(0.25,0.5]`
#>    <dttm>                                   <int>                        <int>
#>  1 2023-10-13 14:56:03                          0                            0
#>  2 2023-10-13 14:56:04                          0                            0
#>  3 2023-10-13 14:56:05                          0                            0
#>  4 2023-10-13 14:56:06                          0                            0
#>  5 2023-10-13 14:56:07                          0                            0
#>  6 2023-10-13 14:56:08                          0                            0
#>  7 2023-10-13 14:56:09                          0                            0
#>  8 2023-10-13 14:56:10                          0                            0
#>  9 2023-10-13 14:56:11                          0                            0
#> 10 2023-10-13 14:56:12                          0                            0
#> # ℹ 197 more rows
#> # ℹ 430 more variables: `lag_15_[0,0.25]_(0.5,0.75]` <int>,
#> #   `lag_15_[0,0.25]_(0.75,1]` <int>, `lag_15_[0,0.25]_(1,1.25]` <int>,
#> #   `lag_15_[0,0.25]_(1.25,1.5]` <int>, `lag_15_[0,0.25]_(1.5,1.75]` <int>,
#> #   `lag_15_[0,0.25]_(1.75,2]` <int>, `lag_15_[0,0.25]_(2,2.25]` <int>,
#> #   `lag_15_[0,0.25]_(2.25,2.5]` <int>, `lag_15_[0,0.25]_(2.5,2.75]` <int>,
#> #   `lag_15_[0,0.25]_(2.75,3]` <int>, `lag_15_(0.25,0.5]_[0,0.25]` <int>, …
create_subject_predictor(df, lags = c(1, 15))
#> # A tibble: 207 × 289
#>    second              `lag_1_[0,0.25]_[0,0.25]` `lag_1_[0,0.25]_(0.25,0.5]`
#>    <dttm>                                  <int>                       <int>
#>  1 2023-10-13 14:56:03                         0                           0
#>  2 2023-10-13 14:56:04                         0                           0
#>  3 2023-10-13 14:56:05                         0                           0
#>  4 2023-10-13 14:56:06                         0                           0
#>  5 2023-10-13 14:56:07                         0                           0
#>  6 2023-10-13 14:56:08                         0                           0
#>  7 2023-10-13 14:56:09                         0                           0
#>  8 2023-10-13 14:56:10                         0                           0
#>  9 2023-10-13 14:56:11                         0                           0
#> 10 2023-10-13 14:56:12                         0                           0
#> # ℹ 197 more rows
#> # ℹ 286 more variables: `lag_1_[0,0.25]_(0.5,0.75]` <int>,
#> #   `lag_1_[0,0.25]_(0.75,1]` <int>, `lag_1_[0,0.25]_(1,1.25]` <int>,
#> #   `lag_1_[0,0.25]_(1.25,1.5]` <int>, `lag_1_[0,0.25]_(1.5,1.75]` <int>,
#> #   `lag_1_[0,0.25]_(1.75,2]` <int>, `lag_1_[0,0.25]_(2,2.25]` <int>,
#> #   `lag_1_[0,0.25]_(2.25,2.5]` <int>, `lag_1_[0,0.25]_(2.5,2.75]` <int>,
#> #   `lag_1_[0,0.25]_(2.75,3]` <int>, `lag_1_(0.25,0.5]_[0,0.25]` <int>, …