diff --git a/404.html b/404.html index 3492b564..b9f613b4 100644 --- a/404.html +++ b/404.html @@ -42,7 +42,7 @@
diff --git a/LICENSE-text.html b/LICENSE-text.html index a968a8ee..99590c48 100644 --- a/LICENSE-text.html +++ b/LICENSE-text.html @@ -25,7 +25,7 @@ diff --git a/articles/custom-explainers.html b/articles/custom-explainers.html index c715e4a4..383876f3 100644 --- a/articles/custom-explainers.html +++ b/articles/custom-explainers.html @@ -43,7 +43,7 @@ diff --git a/articles/global-survshap.html b/articles/global-survshap.html index e5fc7c8c..4eb15a74 100644 --- a/articles/global-survshap.html +++ b/articles/global-survshap.html @@ -43,7 +43,7 @@ diff --git a/articles/global-survshap_files/figure-html/unnamed-chunk-5-1.png b/articles/global-survshap_files/figure-html/unnamed-chunk-5-1.png index c5f6517b..471c679e 100644 Binary files a/articles/global-survshap_files/figure-html/unnamed-chunk-5-1.png and b/articles/global-survshap_files/figure-html/unnamed-chunk-5-1.png differ diff --git a/articles/index.html b/articles/index.html index 7f36eae2..fbe8b8cd 100644 --- a/articles/index.html +++ b/articles/index.html @@ -25,7 +25,7 @@ diff --git a/articles/mlr3proba-usage.html b/articles/mlr3proba-usage.html index cce2e5ec..05996566 100644 --- a/articles/mlr3proba-usage.html +++ b/articles/mlr3proba-usage.html @@ -43,7 +43,7 @@ @@ -190,7 +190,7 @@"accumulated"
.
pdp_2d <- model_profile_2d(exp, variables = list(c("karno", "age")))
-#> Aggregating predictions.. Progress: 41%. Estimated remaining time: 45 seconds.
-#> Aggregating predictions.. Progress: 81%. Estimated remaining time: 14 seconds.
+#> Aggregating predictions.. Progress: 67%. Estimated remaining time: 15 seconds.
pdp_2d_num_cat <- model_profile_2d(exp, variables = list(c("karno", "celltype")))
These explanations can be plotted using the plot function.
diff --git a/articles/pdp_files/figure-html/unnamed-chunk-6-1.png b/articles/pdp_files/figure-html/unnamed-chunk-6-1.png index 4dddbcdf..4b891215 100644 Binary files a/articles/pdp_files/figure-html/unnamed-chunk-6-1.png and b/articles/pdp_files/figure-html/unnamed-chunk-6-1.png differ diff --git a/articles/pdp_files/figure-html/unnamed-chunk-7-1.png b/articles/pdp_files/figure-html/unnamed-chunk-7-1.png index 5aab7793..ee8dd903 100644 Binary files a/articles/pdp_files/figure-html/unnamed-chunk-7-1.png and b/articles/pdp_files/figure-html/unnamed-chunk-7-1.png differ diff --git a/articles/survex-usage.html b/articles/survex-usage.html index bb9cbeed..ed5438c2 100644 --- a/articles/survex-usage.html +++ b/articles/survex-usage.html @@ -43,7 +43,7 @@
calculation_method
for surv_shap()
called "treeshap"
that uses the treeshap
package (#75)categorical_variables
were provided[1] Brier, Glenn W. "Verification of forecasts expressed in terms of probability." Monthly Weather Review 78.1 (1950): 1-3.
[2] Graf, Erika, et al. "Assessment and comparison of prognostic classification schemes for survival data." Statistics in Medicine 18.17‐18 (1999): 2529-2545.
[1] Brier, Glenn W. "Verification of forecasts expressed in terms of probability." Monthly Weather Review 78.1 (1950): 1-3.
[2] Graf, Erika, et al. "Assessment and comparison of prognostic classification schemes for survival data." Statistics in Medicine 18.17‐18 (1999): 2529-2545.
[1] Harrell, F.E., Jr., et al. "Regression modelling strategies for improved prognostic prediction." Statistics in Medicine 3.2 (1984): 143-152.
[1] Harrell, F.E., Jr., et al. "Regression modelling strategies for improved prognostic prediction." Statistics in Medicine 3.2 (1984): 143-152.
This function calculates the Cumulative/Dynamic AUC metric for a survival model. It is done using the -estimator proposed proposed by Uno et al. [1], -and Hung and Chang [2].
+estimator proposed proposed by Uno et al. [1], +and Hung and Chang [2].[1] Uno, Hajime, et al. "Evaluating prediction rules for t-year survivors with censored regression models." Journal of the American Statistical Association 102.478 (2007): 527-537.
[2] Hung, Hung, and Chin‐Tsang Chiang. "Optimal composite markers for time dependent receiver operating characteristic curves with censored survival data." Scandinavian Journal of Statistics 37.4 (2010): 664-679.
[1] Uno, Hajime, et al. "Evaluating prediction rules for t-year survivors with censored regression models." Journal of the American Statistical Association 102.478 (2007): 527-537.
[2] Hung, Hung, and Chin‐Tsang Chiang. "Optimal composite markers for time dependent receiver operating characteristic curves with censored survival data." Scandinavian Journal of Statistics 37.4 (2010): 664-679.
[1] Brier, Glenn W. "Verification of forecasts expressed in terms of probability." Monthly Weather Review 78.1 (1950): 1-3.
[2] Graf, Erika, et al. "Assessment and comparison of prognostic classification schemes for survival data." Statistics in Medicine 18.17‐18 (1999): 2529-2545.
[1] Brier, Glenn W. "Verification of forecasts expressed in terms of probability." Monthly Weather Review 78.1 (1950): 1-3.
[2] Graf, Erika, et al. "Assessment and comparison of prognostic classification schemes for survival data." Statistics in Medicine 18.17‐18 (1999): 2529-2545.
numeric from 0 to 1, higher values indicate better performance
-#' @section References:
[1] Uno, Hajime, et al. "Evaluating prediction rules for t-year survivors with censored regression models." Journal of the American Statistical Association 102.478 (2007): 527-537.
[2] Hung, Hung, and Chin‐Tsang Chiang. "Optimal composite markers for time‐dependent receiver operating characteristic curves with censored survival data." Scandinavian Journal of Statistics 37.4 (2010): 664-679.
#' @section References:
[1] Uno, Hajime, et al. "Evaluating prediction rules for t-year survivors with censored regression models." Journal of the American Statistical Association 102.478 (2007): 527-537.
[2] Hung, Hung, and Chin‐Tsang Chiang. "Optimal composite markers for time‐dependent receiver operating characteristic curves with censored survival data." Scandinavian Journal of Statistics 37.4 (2010): 664-679.
a function with standardized parameters (y_true
, risk
, surv
, times
) that can be used to calculate loss
if(FALSE) -measure <- msr("surv.calib_beta") -mlr_measure <- loss_adapt_mlr3proba(measure)
- - +if(FALSE){
+ measure <- msr("surv.calib_beta")
+ mlr_measure <- loss_adapt_mlr3proba(measure)
+}
+
+
[1] Graf, Erika, et al. "Assessment and comparison of prognostic classification schemes for survival data." Statistics in Medicine 18.17‐18 (1999): 2529-2545.
[1] Graf, Erika, et al. "Assessment and comparison of prognostic classification schemes for survival data." Statistics in Medicine 18.17‐18 (1999): 2529-2545.
[1] Harrell, F.E., Jr., et al. "Regression modelling strategies for improved prognostic prediction." Statistics in Medicine 3.2 (1984): 143-152.
[1] Harrell, F.E., Jr., et al. "Regression modelling strategies for improved prognostic prediction." Statistics in Medicine 3.2 (1984): 143-152.
a numeric vector of length equal to the length of the times vector, each value (from the range from 0 to 1) represents 1 - AUC metric at a specific time point, with lower values indicating better performance.
-#' @section References:
[1] Uno, Hajime, et al. "Evaluating prediction rules for t-year survivors with censored regression models." Journal of the American Statistical Association 102.478 (2007): 527-537.
[2] Hung, Hung, and Chin‐Tsang Chiang. "Optimal composite markers for time‐dependent receiver operating characteristic curves with censored survival data." Scandinavian Journal of Statistics 37.4 (2010): 664-679.
#' @section References:
[1] Uno, Hajime, et al. "Evaluating prediction rules for t-year survivors with censored regression models." Journal of the American Statistical Association 102.478 (2007): 527-537.
[2] Hung, Hung, and Chin‐Tsang Chiang. "Optimal composite markers for time‐dependent receiver operating characteristic curves with censored survival data." Scandinavian Journal of Statistics 37.4 (2010): 664-679.
numeric from 0 to 1, lower values indicate better performance
-#' @section References:
[1] Uno, Hajime, et al. "Evaluating prediction rules for t-year survivors with censored regression models." Journal of the American Statistical Association 102.478 (2007): 527-537.
[2] Hung, Hung, and Chin‐Tsang Chiang. "Optimal composite markers for time‐dependent receiver operating characteristic curves with censored survival data." Scandinavian Journal of Statistics 37.4 (2010): 664-679.
#' @section References:
[1] Uno, Hajime, et al. "Evaluating prediction rules for t-year survivors with censored regression models." Journal of the American Statistical Association 102.478 (2007): 527-537.
[2] Hung, Hung, and Chin‐Tsang Chiang. "Optimal composite markers for time‐dependent receiver operating characteristic curves with censored survival data." Scandinavian Journal of Statistics 37.4 (2010): 664-679.
An object of class "model_performance_survival"
. It's a list of metric values calculated for the model. It contains:
Harrell's concordance index [1]
C/D AUC using the estimator proposed by Uno et. al [4]
An object of class "model_performance_survival"
. It's a list of metric values calculated for the model. It contains:
Harrell's concordance index [1]
Brier score [2, 3]
C/D AUC using the estimator proposed by Uno et. al [4]
integral of the Brier score
integral of the C/D AUC
[1] Harrell, F.E., Jr., et al. "Regression modelling strategies for improved prognostic prediction." Statistics in Medicine 3.2 (1984): 143-152.
[2] Brier, Glenn W. "Verification of forecasts expressed in terms of probability." Monthly Weather Review 78.1 (1950): 1-3.
[3] Graf, Erika, et al. "Assessment and comparison of prognostic classification schemes for survival data." Statistics in Medicine 18.17‐18 (1999): 2529-2545.
[4] Uno, Hajime, et al. "Evaluating prediction rules for t-year survivors with censored regression models." Journal of the American Statistical Association 102.478 (2007): 527-537.
[1] Harrell, F.E., Jr., et al. "Regression modelling strategies for improved prognostic prediction." Statistics in Medicine 3.2 (1984): 143-152.
[2] Brier, Glenn W. "Verification of forecasts expressed in terms of probability." Monthly Weather Review 78.1 (1950): 1-3.
[3] Graf, Erika, et al. "Assessment and comparison of prognostic classification schemes for survival data." Statistics in Medicine 18.17‐18 (1999): 2529-2545.
[4] Uno, Hajime, et al. "Evaluating prediction rules for t-year survivors with censored regression models." Journal of the American Statistical Association 102.478 (2007): 527-537.
a two element numeric vector or matrix of one row and two columns, the first element being the true observed time and the second the status of the observation, used for plotting
a positive integer, number of observations used as the background data
a character, either "kernelshap"
for use of kernelshap
library (providing faster Kernel SHAP with refinements) or "exact_kernel"
for exact Kernel SHAP estimation
a character, either "kernelshap"
for use of kernelshap
library (providing faster Kernel SHAP with refinements), "exact_kernel"
for exact Kernel SHAP estimation, or "treeshap"
for use of treeshap
library (efficient implementation to compute SHAP values for tree-based models).
plot.aggregated_surv_shap
variable
- variable for which the profile is to be plotted, by default first from result data
color_variable
- variable used to denote the color, by default equal to variable
-
#' ## plot.aggregated_surv_shap(geom = "curves")
variable
- variable for which SurvSHAP(t) curves are to be plotted, by default first from result data
+
+
+
+plot.aggregated_surv_shap(geom = "curves")
+
+variable
- variable for which SurvSHAP(t) curves are to be plotted, by default first from result data
boxplot
- whether to plot functional boxplot with marked outliers or all curves colored by variable value
+coef
- length of the functional boxplot's whiskers as multiple of IQR, by default 1.5
@@ -209,8 +215,11 @@ Examples
#> Observations with outlying SurvSHAP(t) values:
#> trt celltype karno diagtime age prior
#> 1 1 squamous 60 7 69 0
+#> 2 1 squamous 70 5 64 10
#> 3 1 squamous 60 3 38 0
+#> 4 1 squamous 60 9 63 10
#> 18 1 smallcell 40 2 35 0
+#> 19 1 smallcell 80 4 63 10
#> 127 2 large 70 15 68 10
#> 128 2 large 30 4 39 10
diff --git a/reference/plot.model_diagnostics_survival.html b/reference/plot.model_diagnostics_survival.html
index 5f6bd64b..e206a2c6 100644
--- a/reference/plot.model_diagnostics_survival.html
+++ b/reference/plot.model_diagnostics_survival.html
@@ -26,7 +26,7 @@
diff --git a/reference/plot.model_parts_survival-1.png b/reference/plot.model_parts_survival-1.png
index 6caaacf1..39ae935b 100644
Binary files a/reference/plot.model_parts_survival-1.png and b/reference/plot.model_parts_survival-1.png differ
diff --git a/reference/plot.model_parts_survival.html b/reference/plot.model_parts_survival.html
index bab75b94..d97a67b3 100644
--- a/reference/plot.model_parts_survival.html
+++ b/reference/plot.model_parts_survival.html
@@ -26,7 +26,7 @@
diff --git a/reference/plot.model_performance_survival-1.png b/reference/plot.model_performance_survival-1.png
index 6f1a76b0..68d779ce 100644
Binary files a/reference/plot.model_performance_survival-1.png and b/reference/plot.model_performance_survival-1.png differ
diff --git a/reference/plot.model_performance_survival-2.png b/reference/plot.model_performance_survival-2.png
index 9f7520b4..24573fbf 100644
Binary files a/reference/plot.model_performance_survival-2.png and b/reference/plot.model_performance_survival-2.png differ
diff --git a/reference/plot.model_performance_survival.html b/reference/plot.model_performance_survival.html
index 394e7629..960f2f88 100644
--- a/reference/plot.model_performance_survival.html
+++ b/reference/plot.model_performance_survival.html
@@ -26,7 +26,7 @@
diff --git a/reference/plot.model_profile_2d_survival-1.png b/reference/plot.model_profile_2d_survival-1.png
index 35c3cca5..948c89f7 100644
Binary files a/reference/plot.model_profile_2d_survival-1.png and b/reference/plot.model_profile_2d_survival-1.png differ
diff --git a/reference/plot.model_profile_2d_survival-2.png b/reference/plot.model_profile_2d_survival-2.png
index 8c94d410..c8da4710 100644
Binary files a/reference/plot.model_profile_2d_survival-2.png and b/reference/plot.model_profile_2d_survival-2.png differ
diff --git a/reference/plot.model_profile_2d_survival.html b/reference/plot.model_profile_2d_survival.html
index e79b0b52..0d29c4e1 100644
--- a/reference/plot.model_profile_2d_survival.html
+++ b/reference/plot.model_profile_2d_survival.html
@@ -26,7 +26,7 @@
@@ -172,19 +172,19 @@ Examples
)
head(cph_model_profile_2d$result)
#> _v1name_ _v2name_ _v1type_ _v2type_ _v1value_ _v2value_ _times_
-#> 1 age celltype numerical categorical 34 adeno 1.5
-#> 2 age celltype numerical categorical 35.9583333333333 adeno 1.5
-#> 3 age celltype numerical categorical 37.9166666666667 adeno 1.5
-#> 4 age celltype numerical categorical 39.875 adeno 1.5
-#> 5 age celltype numerical categorical 41.8333333333333 adeno 1.5
-#> 6 age celltype numerical categorical 43.7916666666667 adeno 1.5
+#> 1 age celltype numerical categorical 35 adeno 1.5
+#> 2 age celltype numerical categorical 36.5416666666667 adeno 1.5
+#> 3 age celltype numerical categorical 38.0833333333333 adeno 1.5
+#> 4 age celltype numerical categorical 39.625 adeno 1.5
+#> 5 age celltype numerical categorical 41.1666666666667 adeno 1.5
+#> 6 age celltype numerical categorical 42.7083333333333 adeno 1.5
#> _label_ _yhat_
-#> 1 coxph 0.9721413
-#> 2 coxph 0.9726016
-#> 3 coxph 0.9730544
-#> 4 coxph 0.9734999
-#> 5 coxph 0.9739382
-#> 6 coxph 0.9743694
+#> 1 coxph 0.9698524
+#> 2 coxph 0.9702453
+#> 3 coxph 0.9706333
+#> 4 coxph 0.9710162
+#> 5 coxph 0.9713943
+#> 6 coxph 0.9717676
plot(cph_model_profile_2d, variables = list(c("age", "celltype")), times = cph_exp$times[20])
@@ -194,19 +194,19 @@ Examples
)
head(cph_model_profile_2d_ale$result)
#> _v1name_ _v2name_ _v1type_ _v2type_ _v1value_ _v2value_ _times_ _yhat_
-#> 1 age karno numerical numerical 34 10 1.5 0.9808874
-#> 2 age karno numerical numerical 34 10 4.0 0.9534604
-#> 3 age karno numerical numerical 34 10 7.0 0.9270583
-#> 4 age karno numerical numerical 34 10 8.0 0.8934681
-#> 5 age karno numerical numerical 34 10 10.0 0.8774408
-#> 6 age karno numerical numerical 34 10 12.0 0.8545444
+#> 1 age karno numerical numerical 34 10 1.5 0.9808187
+#> 2 age karno numerical numerical 34 10 4.0 0.9529722
+#> 3 age karno numerical numerical 34 10 7.0 0.9258121
+#> 4 age karno numerical numerical 34 10 8.0 0.8907776
+#> 5 age karno numerical numerical 34 10 10.0 0.8738869
+#> 6 age karno numerical numerical 34 10 12.0 0.8495871
#> _right_ _left_ _top_ _bottom_ _count_ _label_
-#> 1 36 34 10 20 0 coxph
-#> 2 36 34 10 20 0 coxph
-#> 3 36 34 10 20 0 coxph
-#> 4 36 34 10 20 0 coxph
-#> 5 36 34 10 20 0 coxph
-#> 6 36 34 10 20 0 coxph
+#> 1 36 34 10 15 0 coxph
+#> 2 36 34 10 15 0 coxph
+#> 3 36 34 10 15 0 coxph
+#> 4 36 34 10 15 0 coxph
+#> 5 36 34 10 15 0 coxph
+#> 6 36 34 10 15 0 coxph
plot(cph_model_profile_2d_ale, times = cph_exp$times[c(10, 20)], marginalize_over_time = TRUE)
# }
diff --git a/reference/plot.model_profile_survival-1.png b/reference/plot.model_profile_survival-1.png
index 8e5d8fc5..471d4955 100644
Binary files a/reference/plot.model_profile_survival-1.png and b/reference/plot.model_profile_survival-1.png differ
diff --git a/reference/plot.model_profile_survival-2.png b/reference/plot.model_profile_survival-2.png
index 8de544af..119513ea 100644
Binary files a/reference/plot.model_profile_survival-2.png and b/reference/plot.model_profile_survival-2.png differ
diff --git a/reference/plot.model_profile_survival-3.png b/reference/plot.model_profile_survival-3.png
index e0d15962..121d3671 100644
Binary files a/reference/plot.model_profile_survival-3.png and b/reference/plot.model_profile_survival-3.png differ
diff --git a/reference/plot.model_profile_survival-4.png b/reference/plot.model_profile_survival-4.png
index 4b11af74..c8f8e900 100644
Binary files a/reference/plot.model_profile_survival-4.png and b/reference/plot.model_profile_survival-4.png differ
diff --git a/reference/plot.model_profile_survival-5.png b/reference/plot.model_profile_survival-5.png
index 0ef3582c..b33d28ec 100644
Binary files a/reference/plot.model_profile_survival-5.png and b/reference/plot.model_profile_survival-5.png differ
diff --git a/reference/plot.model_profile_survival-6.png b/reference/plot.model_profile_survival-6.png
index d95e2ae0..1f8b2415 100644
Binary files a/reference/plot.model_profile_survival-6.png and b/reference/plot.model_profile_survival-6.png differ
diff --git a/reference/plot.model_profile_survival.html b/reference/plot.model_profile_survival.html
index 3b4a1ab8..3807fede 100644
--- a/reference/plot.model_profile_survival.html
+++ b/reference/plot.model_profile_survival.html
@@ -26,7 +26,7 @@
diff --git a/reference/plot.predict_parts_survival-1.png b/reference/plot.predict_parts_survival-1.png
index a44e16a6..cabd5d58 100644
Binary files a/reference/plot.predict_parts_survival-1.png and b/reference/plot.predict_parts_survival-1.png differ
diff --git a/reference/plot.predict_parts_survival-2.png b/reference/plot.predict_parts_survival-2.png
index 0d536f3e..fbd2f833 100644
Binary files a/reference/plot.predict_parts_survival-2.png and b/reference/plot.predict_parts_survival-2.png differ
diff --git a/reference/plot.predict_parts_survival.html b/reference/plot.predict_parts_survival.html
index 860ca9b9..518eeb28 100644
--- a/reference/plot.predict_parts_survival.html
+++ b/reference/plot.predict_parts_survival.html
@@ -26,7 +26,7 @@
diff --git a/reference/plot.predict_profile_survival-1.png b/reference/plot.predict_profile_survival-1.png
index 78d79fcd..ce2d96ca 100644
Binary files a/reference/plot.predict_profile_survival-1.png and b/reference/plot.predict_profile_survival-1.png differ
diff --git a/reference/plot.predict_profile_survival-2.png b/reference/plot.predict_profile_survival-2.png
index 1dd37e97..4f23c342 100644
Binary files a/reference/plot.predict_profile_survival-2.png and b/reference/plot.predict_profile_survival-2.png differ
diff --git a/reference/plot.predict_profile_survival.html b/reference/plot.predict_profile_survival.html
index 930b986c..1b4963e0 100644
--- a/reference/plot.predict_profile_survival.html
+++ b/reference/plot.predict_profile_survival.html
@@ -26,7 +26,7 @@
diff --git a/reference/plot.surv_feature_importance-1.png b/reference/plot.surv_feature_importance-1.png
index 30b6ce06..9a6761f9 100644
Binary files a/reference/plot.surv_feature_importance-1.png and b/reference/plot.surv_feature_importance-1.png differ
diff --git a/reference/plot.surv_feature_importance.html b/reference/plot.surv_feature_importance.html
index 1fe7f405..2d521a1d 100644
--- a/reference/plot.surv_feature_importance.html
+++ b/reference/plot.surv_feature_importance.html
@@ -27,7 +27,7 @@
diff --git a/reference/plot.surv_lime-1.png b/reference/plot.surv_lime-1.png
index e7c2b314..f400df16 100644
Binary files a/reference/plot.surv_lime-1.png and b/reference/plot.surv_lime-1.png differ
diff --git a/reference/plot.surv_lime.html b/reference/plot.surv_lime.html
index d657f812..f1b07ef1 100644
--- a/reference/plot.surv_lime.html
+++ b/reference/plot.surv_lime.html
@@ -26,7 +26,7 @@
diff --git a/reference/plot.surv_model_performance-1.png b/reference/plot.surv_model_performance-1.png
index e7a2c450..fe4c5b4e 100644
Binary files a/reference/plot.surv_model_performance-1.png and b/reference/plot.surv_model_performance-1.png differ
diff --git a/reference/plot.surv_model_performance.html b/reference/plot.surv_model_performance.html
index 355eabf4..d1972bad 100644
--- a/reference/plot.surv_model_performance.html
+++ b/reference/plot.surv_model_performance.html
@@ -26,7 +26,7 @@
diff --git a/reference/plot.surv_model_performance_rocs-1.png b/reference/plot.surv_model_performance_rocs-1.png
index 0b3539cf..2f1b66d8 100644
Binary files a/reference/plot.surv_model_performance_rocs-1.png and b/reference/plot.surv_model_performance_rocs-1.png differ
diff --git a/reference/plot.surv_model_performance_rocs.html b/reference/plot.surv_model_performance_rocs.html
index 5e294edb..e1220d2f 100644
--- a/reference/plot.surv_model_performance_rocs.html
+++ b/reference/plot.surv_model_performance_rocs.html
@@ -26,7 +26,7 @@
diff --git a/reference/plot.surv_shap-1.png b/reference/plot.surv_shap-1.png
index 0b9ca06f..19e7f50c 100644
Binary files a/reference/plot.surv_shap-1.png and b/reference/plot.surv_shap-1.png differ
diff --git a/reference/plot.surv_shap.html b/reference/plot.surv_shap.html
index e13082c3..2de0711d 100644
--- a/reference/plot.surv_shap.html
+++ b/reference/plot.surv_shap.html
@@ -26,7 +26,7 @@
diff --git a/reference/predict.surv_explainer.html b/reference/predict.surv_explainer.html
index 34b7e499..42254687 100644
--- a/reference/predict.surv_explainer.html
+++ b/reference/predict.surv_explainer.html
@@ -25,7 +25,7 @@
@@ -170,8 +170,8 @@ Examples
#> 0.7354128
predict(rsf_ranger_exp, veteran[1, ], output_type = "chf")[, 1:10]
-#> [1] 0.005709320 0.005931542 0.011686444 0.012941830 0.012941830 0.027183645
-#> [7] 0.031544223 0.035825326 0.039158659 0.039158659
+#> [1] 0.009623403 0.009623403 0.027945423 0.033625319 0.035069764 0.061802719
+#> [7] 0.067580103 0.073109355 0.073219245 0.075695435
diff --git a/reference/predict_parts.surv_explainer-2.png b/reference/predict_parts.surv_explainer-2.png
index 92d8eb20..9cc602fb 100644
Binary files a/reference/predict_parts.surv_explainer-2.png and b/reference/predict_parts.surv_explainer-2.png differ
diff --git a/reference/predict_parts.surv_explainer.html b/reference/predict_parts.surv_explainer.html
index bb097276..b598aa4b 100644
--- a/reference/predict_parts.surv_explainer.html
+++ b/reference/predict_parts.surv_explainer.html
@@ -25,7 +25,7 @@
@@ -112,7 +112,7 @@ the maximum number of observations used for calculation of attributions. If NULL
(default) all observations will be used.
the number of observations used for calculation of attributions. If NULL
(default) all explainer data will be used for SurvSHAP(t) and 100 neigbours for SurvLIME.
either "survival"
or "risk"
the type of survival model output that should be considered for explanations. If "survival"
the explanations are based on the survival function. Otherwise the scalar risk predictions are used by the DALEX::predict_parts
function.
either "survival"
, "chf"
or "risk"
the type of survival model output that should be considered for explanations. If "survival"
the explanations are based on the survival function. If "chf"
the explanations are based on the cumulative hazard function. Otherwise the scalar risk predictions are used by the DALEX::predict_parts
function.
categorical_variables
- character vector, names of variables that should be treated as categories (factors are included by default)
k
- a small positive number > 1, added to chf before taking log, so that weigths aren't negative
for survshap
timestamps
- a numeric vector, time points at which the survival function will be evaluated
y_true
- a two element numeric vector or matrix of one row and two columns, the first element being the true observed time and the second the status of the observation, used for plotting
for survshap
y_true
- a two element numeric vector or matrix of one row and two columns, the first element being the true observed time and the second the status of the observation, used for plotting
calculation_method
- a character, either "kernelshap"
for use of kernelshap
library (providing faster Kernel SHAP with refinements) or "exact_kernel"
for exact Kernel SHAP estimation
aggregation_method
- a character, either "mean_absolute"
or "integral"
, "max_absolute"
, "sum_of_squares"
predict_parts.R
additional parameters, passed to internal functions
a positive integer, number of observations used as the background data
a two element numeric vector or matrix of one row and two columns, the first element being the true observed time and the second the status of the observation, used for plotting
a character, either "kernelshap"
for use of kernelshap
library (providing faster Kernel SHAP with refinements) or "exact_kernel"
for exact Kernel SHAP estimation
a character, either "kernelshap"
for use of kernelshap
library (providing faster Kernel SHAP with refinements), "exact_kernel"
for exact Kernel SHAP estimation, or "treeshap"
for use of treeshap
library (efficient implementation to compute SHAP values for tree-based models).