Merge pull request #112 from ModelOriented/speedup-permshap

Slight speed-up of permshap()

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: kernelshap
 Title: Kernel SHAP
-Version: 0.4.0
+Version: 0.4.1
 Authors@R: c(
     person("Michael", "Mayer", , "[email protected]", role = c("aut", "cre")),
     person("David", "Watson", , "[email protected]", role = "aut"),

NEWS.md

@@ -1,3 +1,10 @@
+# kernelshap 0.4.1
+
+## Other changes
+
+- Slight speed-up of `permshap()` by saving calculations for the two special permutations of all 0 and all 1.
+- Consequently, the `m_exact` component in the output is reduced by 2.
+
 # kernelshap 0.4.0
 
 ## Major changes

R/permshap.R

@@ -68,7 +68,7 @@ permshap.default <- function(object, X, bg_X, pred_fun = stats::predict,
     message(txt)
   }
 
-  # Baseline and predictions on explanation data (latter not required in algo)
+  # Baseline and predictions on explanation data
   bg_preds <- align_pred(pred_fun(object, bg_X[, colnames(X), drop = FALSE], ...))
   v0 <- wcolMeans(bg_preds, bg_w)            # Average pred of bg data: 1 x K
   v1 <- align_pred(pred_fun(object, X, ...)) # Predictions on X:        n x K
@@ -81,10 +81,10 @@ permshap.default <- function(object, X, bg_X, pred_fun = stats::predict,
 
   # Precalculations that are identical for each row to be explained
   Z <- exact_Z(p, feature_names = feature_names, keep_extremes = TRUE)
-  m_exact <- nrow(Z)
+  m_exact <- nrow(Z) - 2L  # We won't evaluate vz for first and last row
   precalc <- list(
     Z = Z,
-    Z_code = rowpaste(Z),
+    Z_code = rowpaste(Z), 
     bg_X_rep = bg_X[rep(seq_len(bg_n), times = m_exact), , drop = FALSE]
   )
 
@@ -97,9 +97,11 @@ permshap.default <- function(object, X, bg_X, pred_fun = stats::predict,
     parallel_args <- c(list(i = seq_len(n)), parallel_args)
     res <- do.call(foreach::foreach, parallel_args) %dopar% permshap_one(
       x = X[i, , drop = FALSE],
+      v1 = v1[i, , drop = FALSE], 
       object = object,
       pred_fun = pred_fun,
       bg_w = bg_w,
+      v0 = v0,
       precalc = precalc,
       ...
     )
@@ -111,9 +113,11 @@ permshap.default <- function(object, X, bg_X, pred_fun = stats::predict,
     for (i in seq_len(n)) {
       res[[i]] <- permshap_one(
         x = X[i, , drop = FALSE],
+        v1 = v1[i, , drop = FALSE], 
         object = object,
         pred_fun = pred_fun,
         bg_w = bg_w,
+        v0 = v0,
         precalc = precalc,
         ...
       )

R/utils_permshap.R

@@ -20,21 +20,25 @@ shapley_weights <- function(p, ell) {
 #' @keywords internal
 #'
 #' @inheritParams permshap
+#' @param v1 Prediction of `x`.
+#' @param v0 Average prediction on background data.
 #' @param x A single row to be explained.
 #' @param precalc A list with precalculated values that are identical for all rows.
 #' @return A (p x K) matrix of SHAP values.
-permshap_one <- function(x, object, pred_fun, bg_w, precalc, ...) {
-  vz <- get_vz(                                                          #  (m_ex x K)
-    X = x[rep(1L, times = nrow(precalc[["bg_X_rep"]])), , drop = FALSE], #  (m_ex*n_bg x p)
-    bg = precalc[["bg_X_rep"]],                                          #  (m_ex*n_bg x p)
-    Z = precalc[["Z"]],                                                  #  (m_ex x p)
+permshap_one <- function(x, v1, object, pred_fun, bg_w, v0, precalc, ...) {
+  Z <- precalc[["Z"]]                                                    # ((m_ex+2) x K)
+  vz <- get_vz(                                                          # (m_ex x K)
+    X = x[rep(1L, times = nrow(precalc[["bg_X_rep"]])), , drop = FALSE], # (m_ex*n_bg x p)
+    bg = precalc[["bg_X_rep"]],                                          # (m_ex*n_bg x p)
+    Z = Z[2:(nrow(Z) - 1L), , drop = FALSE],                             # (m_ex x p)
     object = object,
     pred_fun = pred_fun,
     w = bg_w,
     ...
   )
+  vz <- rbind(v0, vz, v1)  # we add the cheaply calculated v0 and v1
   rownames(vz) <- precalc[["Z_code"]]
-  shapley_formula(precalc[["Z"]], vz)
+  shapley_formula(Z, vz = vz)
 }
 
 #' Shapley's formula

packaging.R

@@ -15,7 +15,7 @@ library(usethis)
 use_description(
   fields = list(
     Title = "Kernel SHAP",
-    Version = "0.4.0",
+    Version = "0.4.1",
     Description = "Efficient implementation of Kernel SHAP, see Lundberg and Lee (2017),
     and Covert and Lee (2021) <http://proceedings.mlr.press/v130/covert21a>.
     Furthermore, for up to 14 features, exact permutation SHAP values can be calculated.