Added sh file
This commit is contained in:
parent
762bb762bc
commit
84eea01ea3
18
dataset_script.sh
Normal file
18
dataset_script.sh
Normal file
@ -0,0 +1,18 @@
|
||||
! kaggle datasets download kamilpytlak/personal-key-indicators-of-heart-disease
|
||||
! unzip personal-key-indicators-of-heart-disease -d ./
|
||||
! wc -l heart_2020_cleaned.csv
|
||||
|
||||
! head -n 10 heart_2020_cleaned.csv
|
||||
echo "Categories: \n"
|
||||
! cut -f 10 -d "," heart_2020_cleaned.csv | sort | uniq -c
|
||||
! cut -f 12 -d "," heart_2020_cleaned.csv | sort | uniq -c
|
||||
|
||||
|
||||
! grep -P "^$" -n heart_2020_cleaned.csv
|
||||
! head -n -1 heart_2020_cleaned.csv | shuf > heart_2020_sorted.csv
|
||||
|
||||
! head -n 4000 heart_2020_sorted.csv > heart_2020_test.csv
|
||||
! tail -n +4001 heart_2020_sorted.csv > heart_2020_train.csv
|
||||
|
||||
! wc -l heart_2020_t*
|
||||
|
319796
heart_2020_cleaned.csv
319796
heart_2020_cleaned.csv
File diff suppressed because it is too large
Load Diff
Loading…
Reference in New Issue
Block a user