2020-01-20 生成list,并且划分数据集

2020-01-20  本文已影响0人  Joyner2018

生成list,并且划分数据集

#!/bin/sh

CURDIR=$(pwd)

genlist(){

    ls $1/ | while read line;

    do

        echo "${1}/${line} ${2}"

    done

}

genlist ${CURDIR}/non-smoking 0 > temp__0.txt

genlist ${CURDIR}/smoking 1 > temp__1.txt

LEN_0=$(cat temp__0.txt | wc -l)

LEN_1=$(cat temp__1.txt | wc -l)

LEN_0_TEST=$((${LEN_0}/5))

LEN_0_TRAIN=$((${LEN_0}-${LEN_0_TEST}))

LEN_1_TEST=$((${LEN_1}/5))

LEN_1_TRAIN=$((${LEN_1}-${LEN_1_TEST}))

cat temp__0.txt | head -n ${LEN_0_TRAIN} > train_0.txt

cat temp__0.txt | tail -n ${LEN_0_TEST} > test_0.txt

cat temp__1.txt | head -n ${LEN_1_TRAIN} > train_1.txt

cat temp__1.txt | tail -n ${LEN_1_TEST} > test_1.txt

cat train_0.txt train_1.txt > train.txt

cat test_0.txt test_1.txt > test.txt

rm temp__0.txt temp__1.txt

rm train_0.txt test_0.txt

rm train_1.txt test_1.txt

上一篇 下一篇

猜你喜欢

热点阅读