result_virtualization.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "import os\n",
    "# setup pyspark jars\n",
    "os.environ[\"PYSPARK_SUBMIT_ARGS\"] = \"--jars /home/cloudera/twitter_stream/libs/spark-core_2.11-1.5.2.logging.jar,/usr/lib/hbase/hbase-spark-1.2.0-cdh5.14.0.jar,/usr/lib/hive/lib/guava-14.0.1.jar,/usr/lib/hive/lib/hive-hbase-handler-1.1.0-cdh5.14.0.jar,/usr/lib/hbase/lib/htrace-core.jar,/usr/lib/zookeeper/zookeeper.jar,/usr/lib/hbase/hbase-client.jar,/usr/lib/hbase/hbase-common.jar,/usr/lib/hbase/hbase-protocol.jar,/usr/lib/hbase/hbase-server.jar,/usr/lib/hbase/hbase-hadoop-compat.jar,/usr/lib/hbase/hbase-hadoop2-compat.jar,/usr/lib/hbase/lib/metrics-core-2.2.0.jar pyspark-shell\"\n",
    "\n",
    "from __future__ import unicode_literals\n",
    "\n",
    "from pyspark import SparkContext\n",
    "from pyspark.sql import SparkSession\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "plt.rc('font', family='Arial')\n",
    "import matplotlib.dates as mdates\n",
    "from matplotlib.ticker import IndexFormatter\n",
    "%matplotlib inline\n",
    "#pd.__version__\n",
    "\n",
    "# show col title\n",
    "def show_col_title(p, data):\n",
    "    for rect, total in zip(p.patches, data):\n",
    "        p.text(rect.get_x() + rect.get_width() / 2, rect.get_height(), '{0:,}'.format(total), \n",
    "                    ha='center', va='bottom', fontsize=12, fontweight='bold')\n",
    "\n",
    "\n",
    "sqlContext = SparkSession\\\n",
    "    .builder\\\n",
    "    .appName(\"Python Spark SQL Hive matplotlib\") \\\n",
    "    .config(\"hive.metastore.uris\", \"thrift://127.0.0.1:9083\") \\\n",
    "    .enableHiveSupport() \\\n",
    "    .getOrCreate()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# top 10 hashtags\n",
    "df_hashtag = sqlContext.sql('SELECT hashtag AS hashtag, COUNT(*) AS total\\\n",
    "                        FROM default.tweets LATERAL VIEW explode(SPLIT(lower(general_hashtags),\",\")) aTable AS hashtag\\\n",
    "                        WHERE general_hashtags IS NOT NULL AND LENGTH(general_hashtags) > 0\\\n",
    "                        GROUP BY hashtag\\\n",
    "                        ORDER BY total DESC\\\n",
    "                        LIMIT 10').toPandas()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x504 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#print(df_hashtag)\n",
    "fig1, ax1 = plt.subplots(figsize=(12,7))\n",
    "ax1.tick_params(axis='x', labelsize=15)\n",
    "ax1.tick_params(axis='y', labelsize=10)\n",
    "ax1.set_title('Top 10 Hashtags', fontsize=15, fontweight='bold')\n",
    "chart1 = df_hashtag.plot(ax=ax1, kind='bar',x='hashtag', y='total')\n",
    "show_col_title(chart1, df_hashtag['total'])\n",
    "# save result to file\n",
    "#plt.savefig('tweet_by_country', format='png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# top 10 cities in US\n",
    "#WHERE place_place_type = \"city\" AND place_country_code = \"US\"\\\n",
    "df_cities = sqlContext.sql('SELECT CONCAT(place_name,\" - \",place_country) AS name, COUNT(*) AS total\\\n",
    "                        FROM default.tweets\\\n",
    "                        GROUP BY place_name, place_country\\\n",
    "                        ORDER BY total DESC\\\n",
    "                        LIMIT 10').toPandas()\n",
    "#print(df_cities)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.figure(figsize=(20,10))\n",
    "ax2 = plt.subplot(121, aspect='equal')\n",
    "\n",
    "explode=[0] * len(df_cities)\n",
    "#explode first one\n",
    "explode[0] = 0.1\n",
    "\n",
    "df_cities.plot(kind='pie', y = 'total', ax=ax2, autopct='%1.1f%%', explode=tuple(explode),\n",
    "          startangle=90, shadow=False, labels=df_cities['name'], legend = False, fontsize=14)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# \n",
    "df_lang = sqlContext.sql('SELECT UPPER(general_lang) AS group, COUNT(*) AS total\\\n",
    "                        FROM default.tweets\\\n",
    "                        GROUP BY general_lang\\\n",
    "                        ORDER BY total DESC\\\n",
    "                        LIMIT 10').toPandas()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "fig3, ax3 = plt.subplots(figsize=(12,7))\n",
    "ax3.tick_params(axis='x', labelsize=15)\n",
    "ax3.tick_params(axis='y', labelsize=10)\n",
    "ax3.set_title('Top 10 most popular languages', fontsize=15, fontweight='bold')\n",
    "\n",
    "chart3 = df_lang.plot(ax=ax3, kind='bar',x='group', y='total')\n",
    "show_col_title(chart3, df_lang['total'])\n",
    "\n",
    "#plt.savefig('tweet_by_country', format='png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# total by hour - localtime CST time zone\n",
    "df4 = sqlContext.sql('SELECT CONCAT(MONTH(local_created_ts), \"/\", DAY(local_created_ts), \" \", HOUR(local_created_ts), \"h\") AS group, COUNT(*) total\\\n",
    "                    FROM (SELECT from_utc_timestamp(from_unixtime(general_created_ts), \"CST\") AS local_created_ts \\\n",
    "                            FROM default.tweets) AS tbl\\\n",
    "                    GROUP BY group\\\n",
    "                    ORDER BY group').toPandas()\n",
    "\n",
    "df4.set_index('group',inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#print(df4)\n",
    "fig, ax4 = plt.subplots(figsize=(15,7))\n",
    "ax4.set_title(\"\", fontsize=18, )\n",
    "ax4.set_title('Tweets by hour', fontsize=15, fontweight='bold')\n",
    "chart4 = df4.plot(ax=ax4, table=True)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}