# zadanie 5

# pobranie zbioru danych
kaggle datasets download -d timmate/avocado-prices-2020
unzip -o avocado-prices-2020.zip
wc -l avocado-updated-2020.csv

# przetwarzanie pliku
grep -P "^$" -n avocado-updated-2020.csv

# usuniecie zbednej kolumny year (redundantne dane)
cut -d -f12 --complement avocado-updated-2020.csv
ghead -n 1 avocado-updated-2020.csv> header.csv
ghead -n -1 avocado-updated-2020.csv | shuf > avocado-updated-2020-shuf.csv

# podzial na train/dev/test
ghead -n 6609 avocado-updated-2020-shuf.csv > avocado-updated-2020-test.csv
ghead -n 13218 avocado-updated-2020-shuf.csv | tail -n 6609 > avocado-updated-2020-dev.csv
tail -n +13219 avocado-updated-2020-shuf.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