This object stores the predictions returned by a learner of class LearnerCompRisks.
The task_type is set to "cmprsk".
For accessing survival and hazard functions, as well as other complex methods from a LearnerCompRisks object is not possible atm.
See also
Other Prediction:
PredictionDens,
PredictionSurv
Super class
mlr3::Prediction -> PredictionCompRisks
Active bindings
truth(
Surv)
True (observed) outcome.cif(
list())
Access the stored CIFs.
Methods
Method new()
Creates a new instance of this R6 class.
Usage
PredictionCompRisks$new(
task = NULL,
row_ids = task$row_ids,
truth = task$truth(),
cif = NULL,
check = TRUE
)Arguments
task(TaskSurv)
Task, used to extract defaults forrow_idsandtruth.row_ids(
integer())
Row ids of the predicted observations, i.e. the row ids of the test set.truth(
survival::Surv())
True (observed) response.cif(
list())
Alistof two or morematrixobjects. Each matrix represents a different competing event and it stores the Cumulative Incidence function for each test observation. In each matrix, rows represent observations and columns time points. The names of thelistmust correspond to the competing event names (task$cmp_events).check(
logical(1))
IfTRUE, performs argument checks and predict type conversions.
Examples
library(mlr3)
task = tsk("pbc")
learner = lrn("cmprsk.aalen")
part = partition(task)
p = learner$train(task, part$train)$predict(task, part$test)
p
#>
#> ── <PredictionCompRisks> for 91 observations: ──────────────────────────────────
#> row_ids time event CIF
#> 1 13 2 <list[2]>
#> 6 60 0 <list[2]>
#> 18 44 2 <list[2]>
#> --- --- --- ---
#> 264 42 0 <list[2]>
#> 269 39 0 <list[2]>
#> 275 27 0 <list[2]>
# CIF list: 1 matrix (obs x times) per competing event
names(p$cif) # competing events
#> [1] "1" "2"
# CIF matrix for competing event 1 (first 5 test observations and 20 time points)
p$cif[["1"]][1:5, 1:20]
#> 1 2 4 6 7 9 10 11 12 16 17 18 19 20
#> [1,] 0 0 0 0 0 0 0 0 0 0 0.005405405 0.005405405 0.005405405 0.005405405
#> [2,] 0 0 0 0 0 0 0 0 0 0 0.005405405 0.005405405 0.005405405 0.005405405
#> [3,] 0 0 0 0 0 0 0 0 0 0 0.005405405 0.005405405 0.005405405 0.005405405
#> [4,] 0 0 0 0 0 0 0 0 0 0 0.005405405 0.005405405 0.005405405 0.005405405
#> [5,] 0 0 0 0 0 0 0 0 0 0 0.005405405 0.005405405 0.005405405 0.005405405
#> 22 24 25 26 27 28
#> [1,] 0.005405405 0.01081081 0.01081081 0.01081081 0.01624979 0.02168877
#> [2,] 0.005405405 0.01081081 0.01081081 0.01081081 0.01624979 0.02168877
#> [3,] 0.005405405 0.01081081 0.01081081 0.01081081 0.01624979 0.02168877
#> [4,] 0.005405405 0.01081081 0.01081081 0.01081081 0.01624979 0.02168877
#> [5,] 0.005405405 0.01081081 0.01081081 0.01081081 0.01624979 0.02168877
# CIF matrix for competing event 2 (first 5 test observations and 20 time points)
p$cif[["2"]][1:5, 1:20]
#> 1 2 4 6 7 9
#> [1,] 0.01081081 0.02162162 0.03243243 0.04324324 0.04864865 0.05405405
#> [2,] 0.01081081 0.02162162 0.03243243 0.04324324 0.04864865 0.05405405
#> [3,] 0.01081081 0.02162162 0.03243243 0.04324324 0.04864865 0.05405405
#> [4,] 0.01081081 0.02162162 0.03243243 0.04324324 0.04864865 0.05405405
#> [5,] 0.01081081 0.02162162 0.03243243 0.04324324 0.04864865 0.05405405
#> 10 11 12 16 17 18
#> [1,] 0.05945946 0.06486486 0.07027027 0.07567568 0.07567568 0.08108108
#> [2,] 0.05945946 0.06486486 0.07027027 0.07567568 0.07567568 0.08108108
#> [3,] 0.05945946 0.06486486 0.07027027 0.07567568 0.07567568 0.08108108
#> [4,] 0.05945946 0.06486486 0.07027027 0.07567568 0.07567568 0.08108108
#> [5,] 0.05945946 0.06486486 0.07027027 0.07567568 0.07567568 0.08108108
#> 19 20 22 24 25 26 27
#> [1,] 0.08648649 0.09189189 0.0972973 0.1027027 0.1135135 0.1189525 0.1189525
#> [2,] 0.08648649 0.09189189 0.0972973 0.1027027 0.1135135 0.1189525 0.1189525
#> [3,] 0.08648649 0.09189189 0.0972973 0.1027027 0.1135135 0.1189525 0.1189525
#> [4,] 0.08648649 0.09189189 0.0972973 0.1027027 0.1135135 0.1189525 0.1189525
#> [5,] 0.08648649 0.09189189 0.0972973 0.1027027 0.1135135 0.1189525 0.1189525
#> 28
#> [1,] 0.1298305
#> [2,] 0.1298305
#> [3,] 0.1298305
#> [4,] 0.1298305
#> [5,] 0.1298305
# data.table conversion
tab = as.data.table(p)
tab$CIF[[1]] # for first test observation, list of CIF vectors
#> $`1`
#> 1 2 4 6 7 9
#> 0.000000000 0.000000000 0.000000000 0.000000000 0.000000000 0.000000000
#> 10 11 12 16 17 18
#> 0.000000000 0.000000000 0.000000000 0.000000000 0.005405405 0.005405405
#> 19 20 22 24 25 26
#> 0.005405405 0.005405405 0.005405405 0.010810811 0.010810811 0.010810811
#> 27 28 29 30 31 32
#> 0.016249790 0.021688770 0.027127749 0.027127749 0.027127749 0.027127749
#> 33 35 37 38 39 40
#> 0.027127749 0.038224746 0.038224746 0.038224746 0.038224746 0.038224746
#> 41 42 43 44 46 47
#> 0.038224746 0.038224746 0.038224746 0.038224746 0.038224746 0.044505371
#> 48 49 50 51 53 54
#> 0.044505371 0.050895708 0.057286045 0.057286045 0.057286045 0.057286045
#> 55 57 58 60 61 62
#> 0.057286045 0.057286045 0.057286045 0.057286045 0.057286045 0.057286045
#> 63 64 66 67 69 70
#> 0.057286045 0.057286045 0.057286045 0.057286045 0.057286045 0.057286045
#> 71 72 73 74 75 76
#> 0.057286045 0.057286045 0.057286045 0.057286045 0.057286045 0.057286045
#> 77 78 79 80 81 83
#> 0.057286045 0.057286045 0.057286045 0.057286045 0.075731410 0.075731410
#> 84 85 86 87 89 90
#> 0.075731410 0.075731410 0.075731410 0.075731410 0.075731410 0.075731410
#> 91 93 94 96 97 98
#> 0.075731410 0.075731410 0.075731410 0.075731410 0.075731410 0.075731410
#> 100 101 103 104 106 107
#> 0.075731410 0.075731410 0.075731410 0.075731410 0.075731410 0.075731410
#> 108 109 110 111 112 117
#> 0.075731410 0.075731410 0.075731410 0.075731410 0.075731410 0.075731410
#> 118 120 123 125 126 128
#> 0.075731410 0.075731410 0.075731410 0.075731410 0.075731410 0.075731410
#> 129 131 132 133 134 135
#> 0.075731410 0.075731410 0.075731410 0.075731410 0.075731410 0.075731410
#> 137 139 143 145 147 149
#> 0.075731410 0.075731410 0.075731410 0.075731410 0.075731410 0.075731410
#>
#> $`2`
#> 1 2 4 6 7 9 10
#> 0.01081081 0.02162162 0.03243243 0.04324324 0.04864865 0.05405405 0.05945946
#> 11 12 16 17 18 19 20
#> 0.06486486 0.07027027 0.07567568 0.07567568 0.08108108 0.08648649 0.09189189
#> 22 24 25 26 27 28 29
#> 0.09729730 0.10270270 0.11351351 0.11895249 0.11895249 0.12983045 0.14070841
#> 30 31 32 33 35 37 38
#> 0.14070841 0.14618317 0.14618317 0.15169443 0.16833993 0.17388843 0.18514395
#> 39 40 41 42 43 44 46
#> 0.20202724 0.20202724 0.20202724 0.20202724 0.20202724 0.20202724 0.21438087
#> 47 48 49 50 51 53 54
#> 0.22066149 0.22699626 0.22699626 0.22699626 0.23344417 0.23344417 0.23344417
#> 55 57 58 60 61 62 63
#> 0.24019912 0.24701965 0.25384019 0.26804377 0.26804377 0.26804377 0.27545773
#> 64 66 67 69 70 71 72
#> 0.27545773 0.27545773 0.28340125 0.29134478 0.29134478 0.29134478 0.29134478
#> 73 74 75 76 77 78 79
#> 0.29134478 0.29969567 0.31639744 0.31639744 0.31639744 0.33429220 0.34323958
#> 80 81 83 84 85 86 87
#> 0.34323958 0.35246226 0.36168494 0.37122026 0.38109612 0.38109612 0.38109612
#> 89 90 91 93 94 96 97
#> 0.38109612 0.39154175 0.40198737 0.41264617 0.41264617 0.41264617 0.41264617
#> 98 100 101 103 104 106 107
#> 0.41264617 0.41264617 0.43700914 0.43700914 0.44983176 0.46265438 0.47584335
#> 108 109 110 111 112 117 118
#> 0.47584335 0.47584335 0.48985664 0.50386993 0.51836644 0.54843326 0.54843326
#> 120 123 125 126 128 129 131
#> 0.54843326 0.56633018 0.56633018 0.60400791 0.60400791 0.60400791 0.60400791
#> 132 133 134 135 137 139 143
#> 0.60400791 0.60400791 0.63312252 0.63312252 0.66547208 0.66547208 0.66547208
#> 145 147 149
#> 0.66547208 0.66547208 0.66547208
#>