{"metadata": {"kernelspec": {"name": "python3", "display_name": "Python 3", "language": "python"}, "language_info": {"file_extension": ".py", "mimetype": "text/x-python", "pygments_lexer": "ipython3", "name": "python", "codemirror_mode": {"name": "ipython", "version": 3}, "version": "3.6.3", "nbconvert_exporter": "python"}}, "nbformat_minor": 1, "nbformat": 4, "cells": [{"metadata": {"_cell_guid": "cad50af4-474c-4b92-8ef5-e48a1f8aee8c", "collapsed": true, "_uuid": "953022e4dea927e29a92a4bca75fe634a77dc54b"}, "source": ["##### Global Imports\n", "import pandas as pd\n", "import random\n"], "execution_count": null, "cell_type": "code", "outputs": []}, {"metadata": {}, "source": ["# Filter Row Numbers\n", "\n", "n = 125497040 #number of records in file (excludes header)\n", "s = 10000 #desired sample size\n", "select = sorted(random.sample(range(1,n+1),s))\n", "skip  = tuple(set(range(1,n+1)) - set(select))\n"], "execution_count": null, "cell_type": "code", "outputs": []}, {"metadata": {}, "source": ["df_train = pd.read_csv(\"../input/train.csv\",skiprows=skip)"], "execution_count": null, "cell_type": "code", "outputs": []}, {"metadata": {}, "source": ["df_train.shape"], "execution_count": null, "cell_type": "code", "outputs": []}, {"metadata": {}, "source": ["**Kindly Upvote If You Find It Useful !!**"], "cell_type": "markdown"}, {"metadata": {"collapsed": true}, "source": [], "execution_count": null, "cell_type": "code", "outputs": []}]}