s434784/ium_434784
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Rewrite to bash script

Browse Source
This commit is contained in:
Maciej 2021-03-29 01:55:42 +02:00
parent b5d383f625
commit 572aaeea07
1 changed files with 24 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

25
preparations.sh
View File

@ -1,2 +1,25 @@
curl -OL https://git.wmi.amu.edu.pl/s434784/ium_434784/who_suicide_statistics.csv
#curl -OL https://git.wmi.amu.edu.pl/s434784/ium_434784/raw/branch/master/who_suicide_statistics.csv
# uciecie nazw kolumn
head -n 1 who_suicide_statistics.csv > names.csv
# Randomizacja zbioru.
shuf who_suicide_statistics.csv > data.shuf
#cp who_suicide_statistics.csv data.shuf
# Podzial na 3 pozbiory
#Ustalenie proporcji 6:2:2
let NUMROWS=$(cat data.shuf | wc -l)
let TEST=$((NUMROWS / 10 * 2 ))
let DEV=$((NUMROWS / 10 * 4 ))
let TRAIN=$((NUMROWS - DEV ))
head -n $TEST data.shuf > data.test
head -n $DEV data.shuf | tail -n $TEST > data.dev
head -n $NUMROWS data.shuf | tail -n $TRAIN > data.train
echo "Test rows ${TEST}"
echo "Dev rows ${DEV}"
echo "Train rows ${TRAIN}"
echo "All number of rows ${NUMROWS}"