From 08e5768e9a446db8ab95152d2e9403a0e635ec63 Mon Sep 17 00:00:00 2001
From: Christoph Helma <helma@in-silico.ch>
Date: Mon, 8 Mar 2021 17:41:26 +0100
Subject: cdk predictions fixed

---
 bin/preprocessing.R | 7 +++++++
 1 file changed, 7 insertions(+)
 create mode 100644 bin/preprocessing.R

(limited to 'bin/preprocessing.R')

diff --git a/bin/preprocessing.R b/bin/preprocessing.R
new file mode 100644
index 0000000..393bf46
--- /dev/null
+++ b/bin/preprocessing.R
@@ -0,0 +1,7 @@
+#!/usr/bin/env Rscript
+library(caret)
+args = commandArgs(trailingOnly=TRUE)
+variables = read.csv(args[1])
+scaling = preProcess(variables, method = c("nzv","corr","center", "scale"))
+scaled = predict(scaling,variables)
+write.csv(scaled,file=args[2], row.names=F, quote=F)
-- 
cgit v1.2.3