split benchmark file in chunks

scripts/split_in_chunks.sh

@@ -0,0 +1,18 @@
+#!/bin/bash
+
+IN_FILE="chembl_processed_unique.smi.gz"
+DIR=$PWD
+
+# go to data dir
+cd $(dirname "$0")/../data
+
+echo Splitting processed file in chunks of 200,000 lines
+mkdir -p chunks
+zcat ${IN_FILE} | split -d -a 1 --additional-suffix .smi -l 200000 - chunks/part
+
+# counts
+cd chunks
+wc -l *.smi | grep .smi | awk '{print $1 > ".count_"$2}'
+
+cd $DIR
+