s444465/ium_444465
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

script fixes

Browse Source
This commit is contained in:
Andrzej Preibisz 2022-03-27 17:23:55 +02:00
parent cc4a903383
commit c1b7825305
1 changed files with 7 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
dataset_script.sh
View File

@ -1,6 +1,9 @@
#!/bin/bash #!/bin/bash
CUTOFF=$1 CUTOFF=$1
! rm -rf heart_2020_*
! rm -rf personal-key-indicators-of-heart-disease.zip
TEST_SIZE=$(($CUTOFF * 1/5)) TEST_SIZE=$(($CUTOFF * 1/5))
TRAIN_START=$(($TEST_SIZE+1)) TRAIN_START=$(($TEST_SIZE+1))
echo $TEST_SIZE echo $TEST_SIZE
@ -8,12 +11,11 @@ echo $TRAIN_START
echo $KAGGLE_USERNAME echo $KAGGLE_USERNAME
kaggle datasets download -d kamilpytlak/personal-key-indicators-of-heart-disease --force kaggle datasets download -d kamilpytlak/personal-key-indicators-of-heart-disease --force
! unzip personal-key-indicators-of-heart-disease -d ./ unzip personal-key-indicators-of-heart-disease -d ./
! wc -l heart_2020_cleaned.csv head -n $CUTOFF heart_2020_cleaned.csv | shuf > heart_2020_sorted.csv
! head -n $CUTOFF heart_2020_cleaned.csv | shuf > heart_2020_sorted.csv
! head -n $TEST_SIZE heart_2020_sorted.csv > heart_2020_test.csv head -n $TEST_SIZE heart_2020_sorted.csv > heart_2020_test.csv
! tail -n +$TRAIN_START heart_2020_sorted.csv > heart_2020_train.csv tail -n +$TRAIN_START heart_2020_sorted.csv > heart_2020_train.csv