This commit is contained in:
patrycjalazna 2021-03-28 17:11:30 +02:00
parent 1495840296
commit 0d1787fe23

View File

@ -16,14 +16,14 @@ head -n -1 avocado-updated-2020.csv | shuf > avocado-updated-2020-shuf.csv
head -n "$1" avocado-updated-2020-shuf.csv > avocado-2020.csv head -n "$1" avocado-updated-2020-shuf.csv > avocado-2020.csv
TRUNCATED_SIZE=$(wc -l avocado-2020.csv | awk '{print $1}') TRUNCATED_SIZE=$(wc -l avocado-2020.csv | awk '{print $1}')
TEST_SIZE=0,2*$TRUNCATED_SIZE declare -i test_size=0,2*$TRUNCATED_SIZE
DEV_SIZE=2*0,2*$TRUNCATED_SIZE declare -i dev_size=2*0,2*$TRUNCATED_SIZE
TRAIN_SIZE=2*0,2*$TRUNCATED_SIZE+1 declare -i train_size=2*0,2*$TRUNCATED_SIZE+1
# podzial na train/dev/test 6/2/2 # podzial na train/dev/test 6/2/2
head -n $TEST_SIZE avocado-2020.csv > avocado-updated-2020-test.csv head -n $test_size avocado-2020.csv > avocado-updated-2020-test.csv
head -n $DEV_SIZE avocado-2020.csv | tail -n $TEST_SIZE > avocado-updated-2020-dev.csv head -n $dev_size avocado-2020.csv | tail -n $TEST_SIZE > avocado-updated-2020-dev.csv
tail -n +$TRAIN_SIZE avocado-2020.csv > avocado-updated-2020-train.csv tail -n +$train_size avocado-2020.csv > avocado-updated-2020-train.csv
wc -l avocado-updated-2020-*.csv wc -l avocado-updated-2020-*.csv