From 8b45ef07ca9b2f1b593b8f93279fe20f15f283d8 Mon Sep 17 00:00:00 2001
From: El Potaeto <pommedeterresautee@msn.com>
Date: Sun, 4 Jan 2015 11:21:39 +0100
Subject: [PATCH] build data.table from raw model data

---
 R-package/R/xgb.plot.tree.R | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)
 create mode 100644 R-package/R/xgb.plot.tree.R

diff --git a/R-package/R/xgb.plot.tree.R b/R-package/R/xgb.plot.tree.R
new file mode 100644
index 000000000..3f60d598a
--- /dev/null
+++ b/R-package/R/xgb.plot.tree.R
@@ -0,0 +1,24 @@
+require(DiagrammeR)
+require(stringr)
+require(data.table)
+require(magrittr)
+text <- readLines('xgb.model.dump') %>% str_trim(side = "both")
+position <- str_match(text, "booster") %>% is.na %>% not %>% which %>% c(length(text)+1)
+
+extract <- function(x, pattern)  str_extract(x, pattern) %>% str_split("=") %>% lapply(function(x) x[2]) %>% unlist %>% as.numeric
+
+#for(i in 1:(length(position)-1)){
+i=1
+  cat(paste("\n",i,"\n"))
+  tree <- text[(position[i]+1):(position[i+1]-1)]
+  paste(tree, collapse = "\n") %>% cat
+branch <- str_match(tree, "leaf") %>% is.na %>% tree[.]
+id <- str_extract(branch, "\\d*:") %>% str_replace(":", "") %>% as.numeric
+feature <- str_extract(branch, "\\[.*\\]")
+yes <- extract(branch, "yes=\\d*") 
+no <- extract(branch, "no=\\d*")
+missing <- extract(branch, "missing=\\d+")
+gain <- extract(branch, "gain=\\d*\\.*\\d*")
+cover <- extract(branch, "cover=\\d*\\.*\\d*")
+dt <- data.table(ID = id, Feature = feature, Yes = yes, No = no, Missing = missing, Gain = gain, Cover = cover)
+#}