ium_434742/avocado-preprocessing.sh
patrycjalazna b6c1bb0227 cutoff
2021-03-28 16:47:11 +02:00

39 lines
1.3 KiB
Bash
Executable File

# zadanie 5
# pobranie zbioru danych
kaggle datasets download -d timmate/avocado-prices-2020
unzip -o avocado-prices-2020.zip
TOTAL_SIZE=$(wc -l avocado-updated-2020.csv)
echo "size $TOTAL_SIZE"
# przetwarzanie pliku
grep -P "^$" -n avocado-updated-2020.csv
# usuniecie zbednej kolumny year (redundantne dane)
cut -d ',' -f12 --complement avocado-updated-2020.csv
head -n 1 avocado-updated-2020.csv > header.csv
head -n -1 avocado-updated-2020.csv | shuf > avocado-updated-2020-shuf.csv
head -n "$1" avocado-updated-2020-shuf.csv > avocado-2020.csv
TRUNCATED_SIZE=$(wc -l avocado-2020.csv)
echo "truncated size $TRUNCATED_SIZE"
# podzial na train/dev/test 6/2/2
head -n 6609 avocado-2020.csv > avocado-updated-2020-test.csv
head -n 13218 avocado-2020.csv | tail -n 6609 > avocado-updated-2020-dev.csv
tail -n +13219 avocado-2020.csv > avocado-updated-2020-train.csv
wc -l avocado-updated-2020-*.csv
cat header.csv avocado-updated-2020-test.csv > test.csv
cat header.csv avocado-updated-2020-dev.csv > dev.csv
cat header.csv avocado-updated-2020-train.csv > train.csv
# usuniecie pomocniczych plikow
rm avocado-updated-2020-test.csv
rm avocado-updated-2020-dev.csv
rm avocado-updated-2020-train.csv
rm avocado-updated-2020-shuf.csv
rm avocado-prices-2020.zip
rm header.csv