#!/bin/bash
if kaggle datasets download -d sgonkaggle/youtube-trend-with-subscriber && unzip youtube-trend-with-subscriber.zip; then
    head -n 2 USvideos_modified.csv
    grep -v -e "^$" - USvideos_modified.csv
    COUNT=$(wc -l "USvideos_modified.csv")
    echo "${COUNT}"
    head -n -1 "USvideos_modified.csv" | shuf > "data_shuf"
    head -n "$1" "data_shuf" > "data_train"
    echo "Shuffled dataset"
    wc -l "data_shuf"
    echo "Train dataset"
    wc -l "data_train"
fi