Multiple_Time_Series_using_Apache_Spark_and_Prophet.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "-aF9dJFVItfM"
   },
   "outputs": [],
   "source": [
    "## Set Python - Spark environment.\n",
    "import os\n",
    "import sys\n",
    "#Setting the path for Spark\n",
    "# os.environ['SPARK_HOME'] = \"/home/Kunal/Downloads/spark-2.4.7-bin-hadoop2.7\"\n",
    "os.environ['PYSPARK_SUBMIT_ARGS'] = '--packages org.apache.spark:spark-sql-kafka-0-10_2.11:2.4.7,org.apache.spark:spark-streaming-kafka-0-10_2.11:2.4.7 pyspark-shell'\n",
    "os.environ['PYSPARK_PYTHON'] = sys.executable\n",
    "os.environ['PYSPARK_DRIVER_PYTHON'] = sys.executable\n",
    "# !pip install pyarrow==0.14"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "NUe7BW37S6VA"
   },
   "outputs": [],
   "source": [
    "# !pip install fbprophet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "CY5G7SE9PLEB"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Importing plotly failed. Interactive plots will not work.\n"
     ]
    }
   ],
   "source": [
    "import matplotlib as mpl\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "\n",
    "from fbprophet import Prophet\n",
    "\n",
    "mpl.rcParams['figure.figsize'] = (10, 8)\n",
    "mpl.rcParams['axes.grid'] = False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Create SparkContext, SparkSession\n",
    "from pyspark.sql import *\n",
    "import pyspark\n",
    "from pyspark.conf import SparkConf\n",
    "from pyspark import SparkContext\n",
    "from pyspark.sql import SparkSession\n",
    "from pyspark.sql.types import *\n",
    "# from os.path import join, abspath\n",
    "# warehouse_location = '/user/hive/warehouse'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "id": "qE2sxBUnXPIP"
   },
   "outputs": [],
   "source": [
    "# #setting configuration\n",
    "# conf = SparkConf().setAppName(\"Python_Spark_PHD_Complete\").setMaster('local')\n",
    "# sc = SparkContext(conf=conf)\n",
    "# # spark = SparkSession(sc)\n",
    "# spark = SparkSession \\\n",
    "#             .builder \\\n",
    "#             .appName(\"Python Spark Multiple Time Series\") \\\n",
    "#             .config(\"spark.sql.catalogImplementation=hive\").enableHiveSupport() \\\n",
    "#             .getOrCreate()\n",
    "#setting configuration\n",
    "conf = SparkConf().setAppName(\"Python_Spark_PHD_Complete\").setMaster('yarn-client').set(\"spark.driver.memory\",\"5g\")\n",
    "sc = SparkContext(conf=conf)\n",
    "# sc=SparkContext.getOrCreate(conf=create_spark_conf().setMaster(\"local[4]\").set(\"spark.driver.memory\",\"8g\").set(\"spark.executor.memory\", '8g').set('spark.executor.cores', 4))\n",
    "sc.setLogLevel(\"ERROR\")\n",
    "# spark = SparkSession(sc)\n",
    "spark = SparkSession \\\n",
    "            .builder \\\n",
    "            .appName(\"Python Spark SQL basic example\") \\\n",
    "            .config(\"spark.sql.catalogImplementation=hive\").enableHiveSupport() \\\n",
    "            .getOrCreate()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "id": "4dmzbVV5rrkJ"
   },
   "outputs": [],
   "source": [
    "#Reading household dataset from hadoop cluster\n",
    "sales_weekly_df=spark.read.csv(\"hdfs://localhost:9000//user/kunal/weekly_sales_data.csv\",header=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 416
    },
    "id": "vwpMqvTdPsEx",
    "outputId": "210d76a0-defa-44a2-b5dc-974f4f684019"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+---------+--------+-----------+\n",
      "|     date|store_id|      sales|\n",
      "+---------+--------+-----------+\n",
      "| 1/6/2019|      25|736657.2267|\n",
      "| 1/6/2019|      26|     233711|\n",
      "| 1/6/2019|      28|692156.3794|\n",
      "| 1/6/2019|      29|   213706.8|\n",
      "| 1/6/2019|      30|     266455|\n",
      "| 1/6/2019|      31|837342.8156|\n",
      "| 1/6/2019|      35|   201013.6|\n",
      "| 1/6/2019|      38|  242397.03|\n",
      "| 1/6/2019|      41|   244189.4|\n",
      "| 1/6/2019|      44|   164553.5|\n",
      "|1/13/2019|      25|696457.1717|\n",
      "|1/13/2019|      26|  261136.45|\n",
      "|1/13/2019|      28|  787359.03|\n",
      "|1/13/2019|      29|     279218|\n",
      "|1/13/2019|      30|  252519.59|\n",
      "|1/13/2019|      31|836604.5717|\n",
      "|1/13/2019|      35| 277550.185|\n",
      "|1/13/2019|      38| 215895.275|\n",
      "|1/13/2019|      41|   231992.5|\n",
      "|1/13/2019|      44|     134188|\n",
      "+---------+--------+-----------+\n",
      "only showing top 20 rows\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#show rows\n",
    "sales_weekly_df.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Convert Date column into time stamp format\n",
    "from datetime import datetime\n",
    "from pyspark.sql.functions import col, udf\n",
    "from pyspark.sql.types import DateType,TimestampType\n",
    "func =  udf (lambda x: datetime.strptime(x, '%m/%d/%Y'), TimestampType())\n",
    "spark_df = sales_weekly_df.withColumn('Date', func(col('date')))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+-------------------+--------+-----------+\n",
      "|               Date|store_id|      sales|\n",
      "+-------------------+--------+-----------+\n",
      "|2019-01-06 00:00:00|      25|736657.2267|\n",
      "|2019-01-06 00:00:00|      26|     233711|\n",
      "|2019-01-06 00:00:00|      28|692156.3794|\n",
      "|2019-01-06 00:00:00|      29|   213706.8|\n",
      "|2019-01-06 00:00:00|      30|     266455|\n",
      "|2019-01-06 00:00:00|      31|837342.8156|\n",
      "|2019-01-06 00:00:00|      35|   201013.6|\n",
      "|2019-01-06 00:00:00|      38|  242397.03|\n",
      "|2019-01-06 00:00:00|      41|   244189.4|\n",
      "|2019-01-06 00:00:00|      44|   164553.5|\n",
      "|2019-01-13 00:00:00|      25|696457.1717|\n",
      "|2019-01-13 00:00:00|      26|  261136.45|\n",
      "|2019-01-13 00:00:00|      28|  787359.03|\n",
      "|2019-01-13 00:00:00|      29|     279218|\n",
      "|2019-01-13 00:00:00|      30|  252519.59|\n",
      "|2019-01-13 00:00:00|      31|836604.5717|\n",
      "|2019-01-13 00:00:00|      35| 277550.185|\n",
      "|2019-01-13 00:00:00|      38| 215895.275|\n",
      "|2019-01-13 00:00:00|      41|   231992.5|\n",
      "|2019-01-13 00:00:00|      44|     134188|\n",
      "+-------------------+--------+-----------+\n",
      "only showing top 20 rows\n",
      "\n"
     ]
    }
   ],
   "source": [
    "spark_df.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 106
    },
    "id": "If4w2bE7P9Ij",
    "outputId": "a5db4483-b637-423b-c0bd-43f8c877f396"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "root\n",
      " |-- Date: timestamp (nullable = true)\n",
      " |-- store_id: string (nullable = true)\n",
      " |-- sales: string (nullable = true)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "sdf=spark_df.select(\"Date\",\"store_id\",\"sales\")\n",
    "sdf.printSchema()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "sdf=sdf.withColumn('sales',sdf['sales'].cast('double'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "sdf=sdf.withColumn('store_id',sdf['store_id'].cast('double'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "root\n",
      " |-- Date: timestamp (nullable = true)\n",
      " |-- store_id: double (nullable = true)\n",
      " |-- sales: double (nullable = true)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "sdf.printSchema()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "id": "p6GuZ8PyP9GP",
    "outputId": "4f6bdfc5-0fc7-4186-d0b5-d73b061efaf6"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "500"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sdf.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 285
    },
    "id": "Bk4MebeQP9DM",
    "outputId": "a1c22b56-8edb-410a-ef71-534580661764"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+--------+---------------+\n",
      "|store_id|count(store_id)|\n",
      "+--------+---------------+\n",
      "|    29.0|             50|\n",
      "|    44.0|             50|\n",
      "|    35.0|             50|\n",
      "|    25.0|             50|\n",
      "|    41.0|             50|\n",
      "|    31.0|             50|\n",
      "|    28.0|             50|\n",
      "|    38.0|             50|\n",
      "|    30.0|             50|\n",
      "|    26.0|             50|\n",
      "+--------+---------------+\n",
      "\n"
     ]
    }
   ],
   "source": [
    "sdf.select(['store_id']).groupBy('store_id').agg({'store_id':'count'}).show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "id": "LkO5c9JHP9BM"
   },
   "outputs": [],
   "source": [
    "sdf.createOrReplaceTempView(\"sales\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 285
    },
    "id": "vf5pea8aP8-l",
    "outputId": "7651bdf9-dc49-43b8-b29e-8f1d4970a8b6"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+--------+--------+\n",
      "|store_id|count(1)|\n",
      "+--------+--------+\n",
      "|    25.0|      50|\n",
      "|    26.0|      50|\n",
      "|    28.0|      50|\n",
      "|    29.0|      50|\n",
      "|    30.0|      50|\n",
      "|    31.0|      50|\n",
      "|    35.0|      50|\n",
      "|    38.0|      50|\n",
      "|    41.0|      50|\n",
      "|    44.0|      50|\n",
      "+--------+--------+\n",
      "\n"
     ]
    }
   ],
   "source": [
    "spark.sql(\"select store_id, count(*) from sales group by store_id order by store_id\").show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 481
    },
    "id": "0Yz4zI0qP84H",
    "outputId": "97b23751-eeed-4087-dc23-46be1cfaabc6"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+--------+-------------------+-----------+\n",
      "|store_id|                 ds|          y|\n",
      "+--------+-------------------+-----------+\n",
      "|    25.0|2019-01-06 00:00:00|736657.2267|\n",
      "|    25.0|2019-01-13 00:00:00|696457.1717|\n",
      "|    25.0|2019-01-20 00:00:00| 996876.875|\n",
      "|    25.0|2019-01-27 00:00:00| 1029737.05|\n",
      "|    25.0|2019-02-03 00:00:00|1072873.705|\n",
      "|    25.0|2019-02-10 00:00:00|880462.4681|\n",
      "|    25.0|2019-02-17 00:00:00| 1088756.78|\n",
      "|    25.0|2019-02-24 00:00:00|1266928.313|\n",
      "|    25.0|2019-03-03 00:00:00|907510.1217|\n",
      "|    25.0|2019-03-10 00:00:00|747920.9824|\n",
      "|    25.0|2019-03-17 00:00:00|904750.3158|\n",
      "|    25.0|2019-03-24 00:00:00|1069857.589|\n",
      "|    25.0|2019-03-31 00:00:00|1195574.314|\n",
      "|    25.0|2019-04-07 00:00:00|834911.3333|\n",
      "|    25.0|2019-04-14 00:00:00|788777.4167|\n",
      "|    25.0|2019-04-21 00:00:00|849603.4333|\n",
      "|    25.0|2019-04-28 00:00:00| 965521.165|\n",
      "|    25.0|2019-05-05 00:00:00|1045632.688|\n",
      "|    25.0|2019-05-12 00:00:00| 587768.245|\n",
      "|    25.0|2019-05-19 00:00:00|806767.5946|\n",
      "+--------+-------------------+-----------+\n",
      "only showing top 20 rows\n",
      "\n"
     ]
    }
   ],
   "source": [
    "spark.sql('SELECT store_id, date as ds, sum(sales) as y FROM sales GROUP BY store_id, ds ORDER BY store_id, ds').show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "id": "4CkhDN9CPGU3",
    "outputId": "1599e736-e0f2-4c93-948f-e766c297184a"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sdf.rdd.getNumPartitions()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "sql='SELECT store_id, date as ds, sum(sales) as y FROM sales GROUP BY store_id, ds ORDER BY store_id, ds'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "id": "DskWUcVlP81s"
   },
   "outputs": [],
   "source": [
    "store_part = (spark.sql( sql ).repartition(spark.sparkContext.defaultParallelism, ['store_id'])).cache()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "id": "r-mWsPutuxS6",
    "outputId": "db5006a0-7868-49a7-9226-23fda2b1fc09"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DataFrame[store_id: double, ds: timestamp, y: double]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "store_part"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 231
    },
    "id": "9BHeebRdNkap",
    "outputId": "43e21c71-f434-4db5-995b-1b2d99aaa307"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "== Physical Plan ==\n",
      "InMemoryTableScan [store_id#55, ds#126, y#127]\n",
      "   +- InMemoryRelation [store_id#55, ds#126, y#127], StorageLevel(disk, memory, deserialized, 1 replicas)\n",
      "         +- Exchange hashpartitioning(store_id#55, 2)\n",
      "            +- *(4) Sort [store_id#55 ASC NULLS FIRST, ds#126 ASC NULLS FIRST], true, 0\n",
      "               +- Exchange rangepartitioning(store_id#55 ASC NULLS FIRST, ds#126 ASC NULLS FIRST, 200)\n",
      "                  +- *(3) HashAggregate(keys=[store_id#55, date#30], functions=[sum(sales#51)])\n",
      "                     +- Exchange hashpartitioning(store_id#55, date#30, 200)\n",
      "                        +- *(2) HashAggregate(keys=[store_id#55, date#30], functions=[partial_sum(sales#51)])\n",
      "                           +- *(2) Project [pythonUDF0#134 AS Date#30, cast(store_id#11 as double) AS store_id#55, cast(sales#12 as double) AS sales#51]\n",
      "                              +- BatchEvalPython [<lambda>(date#10)], [date#10, store_id#11, sales#12, pythonUDF0#134]\n",
      "                                 +- *(1) FileScan csv [date#10,store_id#11,sales#12] Batched: false, Format: CSV, Location: InMemoryFileIndex[hdfs://localhost:9000/user/kunal/weekly_sales_data.csv], PartitionFilters: [], PushedFilters: [], ReadSchema: struct<date:string,store_id:string,sales:string>\n"
     ]
    }
   ],
   "source": [
    "store_part.explain()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "id": "6E7MgluCuwrL"
   },
   "outputs": [],
   "source": [
    "result_schema =StructType([\n",
    "  StructField('ds',TimestampType()),\n",
    "  StructField('store_id',IntegerType()),\n",
    "  StructField('y',DoubleType()),\n",
    "  StructField('yhat',DoubleType()),\n",
    "  StructField('yhat_upper',DoubleType()),\n",
    "  StructField('yhat_lower',DoubleType())\n",
    "  ])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "id": "4ry6ExeHuwav"
   },
   "outputs": [],
   "source": [
    "from pyspark.sql.functions import pandas_udf, PandasUDFType\n",
    "\n",
    "@pandas_udf( result_schema, PandasUDFType.GROUPED_MAP )\n",
    "def forecast_sales( store_pd ):\n",
    "\n",
    "  model = Prophet(interval_width=0.95,seasonality_mode = 'multiplicative', weekly_seasonality=True, yearly_seasonality=True)\n",
    "\n",
    "  model.fit( store_pd )\n",
    "\n",
    "  future_pd = model.make_future_dataframe(\n",
    "    periods=5, \n",
    "    freq='w'\n",
    "    )\n",
    "  \n",
    "  forecast_pd = model.predict( future_pd )  \n",
    "  \n",
    "  f_pd = forecast_pd[ ['ds','yhat', 'yhat_upper', 'yhat_lower'] ].set_index('ds')\n",
    "  \n",
    "  st_pd = store_pd[['ds','store_id','y']].set_index('ds')\n",
    "  \n",
    "  results_pd = f_pd.join( st_pd, how='left' )\n",
    "  results_pd.reset_index(level=0, inplace=True)\n",
    "  \n",
    "  results_pd['store_id'] = store_pd['store_id'].iloc[0]\n",
    "\n",
    "  return results_pd[ ['ds', 'store_id','y', 'yhat', 'yhat_upper', 'yhat_lower'] ]  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 108
    },
    "id": "cpie_8DUPdIT",
    "outputId": "28d78538-8ef7-4486-a693-90fc1cda328d"
   },
   "outputs": [],
   "source": [
    "from pyspark.sql.functions import current_date\n",
    "\n",
    "results = (\n",
    "  store_part\n",
    "    .groupBy('store_id')\n",
    "    .apply(forecast_sales)\n",
    "    .withColumn('training_date', current_date() )\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "id": "6pAQ0nYnS_Sz",
    "outputId": "2ec6be2f-f7ee-491a-a087-84824a5dcb42"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DataFrame[ds: timestamp, store_id: int, y: double, yhat: double, yhat_upper: double, yhat_lower: double, training_date: date]"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results.cache()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 481
    },
    "id": "S974tPXcPc-v",
    "outputId": "46a958b0-e846-4fc2-c7cb-60fadd0bd043"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "root\n",
      " |-- ds: timestamp (nullable = true)\n",
      " |-- store_id: integer (nullable = true)\n",
      " |-- y: double (nullable = true)\n",
      " |-- yhat: double (nullable = true)\n",
      " |-- yhat_upper: double (nullable = true)\n",
      " |-- yhat_lower: double (nullable = true)\n",
      " |-- training_date: date (nullable = false)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "results.printSchema()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 340
    },
    "id": "pm9V9lRFM_Ef",
    "outputId": "4f318945-6b7a-4529-fe8b-bbc84dbb2638"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "== Physical Plan ==\n",
      "InMemoryTableScan [ds#173, store_id#174, y#175, yhat#176, yhat_upper#177, yhat_lower#178, training_date#185]\n",
      "   +- InMemoryRelation [ds#173, store_id#174, y#175, yhat#176, yhat_upper#177, yhat_lower#178, training_date#185], StorageLevel(disk, memory, deserialized, 1 replicas)\n",
      "         +- *(2) Project [ds#173, store_id#174, y#175, yhat#176, yhat_upper#177, yhat_lower#178, 18826 AS training_date#185]\n",
      "            +- FlatMapGroupsInPandas [store_id#55], forecast_sales(store_id#55, ds#126, y#127), [ds#173, store_id#174, y#175, yhat#176, yhat_upper#177, yhat_lower#178]\n",
      "               +- *(1) Sort [store_id#55 ASC NULLS FIRST], false, 0\n",
      "                  +- InMemoryTableScan [store_id#55, store_id#55, ds#126, y#127]\n",
      "                        +- InMemoryRelation [store_id#55, ds#126, y#127], StorageLevel(disk, memory, deserialized, 1 replicas)\n",
      "                              +- Exchange hashpartitioning(store_id#55, 2)\n",
      "                                 +- *(4) Sort [store_id#55 ASC NULLS FIRST, ds#126 ASC NULLS FIRST], true, 0\n",
      "                                    +- Exchange rangepartitioning(store_id#55 ASC NULLS FIRST, ds#126 ASC NULLS FIRST, 200)\n",
      "                                       +- *(3) HashAggregate(keys=[store_id#55, date#30], functions=[sum(sales#51)])\n",
      "                                          +- Exchange hashpartitioning(store_id#55, date#30, 200)\n",
      "                                             +- *(2) HashAggregate(keys=[store_id#55, date#30], functions=[partial_sum(sales#51)])\n",
      "                                                +- *(2) Project [pythonUDF0#134 AS Date#30, cast(store_id#11 as double) AS store_id#55, cast(sales#12 as double) AS sales#51]\n",
      "                                                   +- BatchEvalPython [<lambda>(date#10)], [date#10, store_id#11, sales#12, pythonUDF0#134]\n",
      "                                                      +- *(1) FileScan csv [date#10,store_id#11,sales#12] Batched: false, Format: CSV, Location: InMemoryFileIndex[hdfs://localhost:9000/user/kunal/weekly_sales_data.csv], PartitionFilters: [], PushedFilters: [], ReadSchema: struct<date:string,store_id:string,sales:string>\n"
     ]
    }
   ],
   "source": [
    "results.explain()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "id": "sXL4kmexSH0D",
    "outputId": "135e076d-a6f0-41ca-b9b3-07ccf407d5d7"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DataFrame[ds: timestamp, store_id: int, y: double, yhat: double, yhat_upper: double, yhat_lower: double, training_date: date]"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results.coalesce(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "id": "9nj260kgSMa7",
    "outputId": "f449b09f-3fc1-414d-c812-c3d0ed60c86e"
   },
   "outputs": [],
   "source": [
    "# results.count()\n",
    "results=results.filter(\"store_id is not null\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "root\n",
      " |-- ds: timestamp (nullable = true)\n",
      " |-- store_id: integer (nullable = true)\n",
      " |-- y: double (nullable = true)\n",
      " |-- yhat: double (nullable = true)\n",
      " |-- yhat_upper: double (nullable = true)\n",
      " |-- yhat_lower: double (nullable = true)\n",
      " |-- training_date: date (nullable = false)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "results.printSchema()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_results=results.withColumn(\"training_date\",results[\"training_date\"].cast(TimestampType()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "root\n",
      " |-- ds: timestamp (nullable = true)\n",
      " |-- store_id: integer (nullable = true)\n",
      " |-- y: double (nullable = true)\n",
      " |-- yhat: double (nullable = true)\n",
      " |-- yhat_upper: double (nullable = true)\n",
      " |-- yhat_lower: double (nullable = true)\n",
      " |-- training_date: timestamp (nullable = false)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "df_results.printSchema()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+-------------------+--------+-----------+------------------+------------------+------------------+-------------------+\n",
      "|                 ds|store_id|          y|              yhat|        yhat_upper|        yhat_lower|      training_date|\n",
      "+-------------------+--------+-----------+------------------+------------------+------------------+-------------------+\n",
      "|2019-01-06 00:00:00|      25|736657.2267| 712575.0694034342| 951145.1714429691|481264.82516048197|2021-07-18 00:00:00|\n",
      "|2019-01-13 00:00:00|      25|696457.1717| 743751.7568928569| 974123.3020284316| 504054.6607367446|2021-07-18 00:00:00|\n",
      "|2019-01-20 00:00:00|      25| 996876.875| 972909.8778010788|1215014.4092873293|  759465.238607354|2021-07-18 00:00:00|\n",
      "|2019-01-27 00:00:00|      25| 1029737.05|1034447.5557582717| 1264969.198907835| 811822.7152067492|2021-07-18 00:00:00|\n",
      "|2019-02-03 00:00:00|      25|1072873.705| 986365.7266742919|1208138.3833476666| 754330.1094141544|2021-07-18 00:00:00|\n",
      "|2019-02-10 00:00:00|      25|880462.4681|1016291.0609313075| 1236954.596423364| 797825.5795935837|2021-07-18 00:00:00|\n",
      "|2019-02-17 00:00:00|      25| 1088756.78|1108793.2648313895|1342140.1629919412| 876096.1986851295|2021-07-18 00:00:00|\n",
      "|2019-02-24 00:00:00|      25|1266928.313|1099825.7781849583|1325639.8058518483| 868227.4548047854|2021-07-18 00:00:00|\n",
      "|2019-03-03 00:00:00|      25|907510.1217| 954930.7385013087|1179312.7625411935|  726515.444990953|2021-07-18 00:00:00|\n",
      "|2019-03-10 00:00:00|      25|747920.9824| 840307.5564991487|1085170.6143633244| 613741.8753766676|2021-07-18 00:00:00|\n",
      "|2019-03-17 00:00:00|      25|904750.3158| 895698.5219441098|1108553.5172415026| 673369.3201938033|2021-07-18 00:00:00|\n",
      "|2019-03-24 00:00:00|      25|1069857.589|1040022.5385686738|1271414.8577645312| 807519.3727823317|2021-07-18 00:00:00|\n",
      "|2019-03-31 00:00:00|      25|1195574.314| 1078329.032876978| 1311366.214073445| 836995.0597982827|2021-07-18 00:00:00|\n",
      "|2019-04-07 00:00:00|      25|834911.3333| 957531.8584494952|1194887.9738064846| 724560.6112828285|2021-07-18 00:00:00|\n",
      "|2019-04-14 00:00:00|      25|788777.4167| 825594.4716836138|1062225.5730295067| 593122.5128832641|2021-07-18 00:00:00|\n",
      "|2019-04-21 00:00:00|      25|849603.4333| 826255.1077112334| 1058406.666442855| 603494.1254724397|2021-07-18 00:00:00|\n",
      "|2019-04-28 00:00:00|      25| 965521.165| 913980.6218878273|1145845.2922387242|  683978.871117027|2021-07-18 00:00:00|\n",
      "|2019-05-05 00:00:00|      25|1045632.688| 931383.5223415782|1154647.2980189256| 686954.9618851955|2021-07-18 00:00:00|\n",
      "|2019-05-12 00:00:00|      25| 587768.245| 831326.9849932519|1057515.2945626732| 597392.3478105812|2021-07-18 00:00:00|\n",
      "|2019-05-19 00:00:00|      25|806767.5946| 740232.6751845297| 982486.2132086994|511095.03032678046|2021-07-18 00:00:00|\n",
      "+-------------------+--------+-----------+------------------+------------------+------------------+-------------------+\n",
      "only showing top 20 rows\n",
      "\n"
     ]
    }
   ],
   "source": [
    "df_results.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "id": "G4Zj9bWYPnhD"
   },
   "outputs": [],
   "source": [
    "df_results.createOrReplaceTempView('forecasted')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 285
    },
    "id": "GLQys5g8CUzz",
    "outputId": "b37ade0d-85fe-4b61-9263-8306fef65d30"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+--------+--------+\n",
      "|store_id|count(1)|\n",
      "+--------+--------+\n",
      "|      31|      55|\n",
      "|      28|      55|\n",
      "|      26|      55|\n",
      "|      44|      55|\n",
      "|      41|      55|\n",
      "|      35|      55|\n",
      "|      38|      55|\n",
      "|      25|      55|\n",
      "|      29|      55|\n",
      "|      30|      55|\n",
      "+--------+--------+\n",
      "\n"
     ]
    }
   ],
   "source": [
    "spark.sql(\"select store_id, count(*) from forecasted group by store_id\").show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "id": "dhCPlXG_PczA"
   },
   "outputs": [],
   "source": [
    "final_df = df_results.toPandas()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 416
    },
    "id": "LPX69_FBKBeh",
    "outputId": "2f997c6a-acd1-4f35-e789-aad2c50ea9f7"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ds</th>\n",
       "      <th>store_id</th>\n",
       "      <th>y</th>\n",
       "      <th>yhat</th>\n",
       "      <th>yhat_upper</th>\n",
       "      <th>yhat_lower</th>\n",
       "      <th>training_date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2019-01-06</td>\n",
       "      <td>25</td>\n",
       "      <td>7.366572e+05</td>\n",
       "      <td>7.125751e+05</td>\n",
       "      <td>9.511452e+05</td>\n",
       "      <td>481264.825160</td>\n",
       "      <td>2021-07-18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2019-01-13</td>\n",
       "      <td>25</td>\n",
       "      <td>6.964572e+05</td>\n",
       "      <td>7.437518e+05</td>\n",
       "      <td>9.741233e+05</td>\n",
       "      <td>504054.660737</td>\n",
       "      <td>2021-07-18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2019-01-20</td>\n",
       "      <td>25</td>\n",
       "      <td>9.968769e+05</td>\n",
       "      <td>9.729099e+05</td>\n",
       "      <td>1.215014e+06</td>\n",
       "      <td>759465.238607</td>\n",
       "      <td>2021-07-18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2019-01-27</td>\n",
       "      <td>25</td>\n",
       "      <td>1.029737e+06</td>\n",
       "      <td>1.034448e+06</td>\n",
       "      <td>1.264969e+06</td>\n",
       "      <td>811822.715207</td>\n",
       "      <td>2021-07-18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2019-02-03</td>\n",
       "      <td>25</td>\n",
       "      <td>1.072874e+06</td>\n",
       "      <td>9.863657e+05</td>\n",
       "      <td>1.208138e+06</td>\n",
       "      <td>754330.109414</td>\n",
       "      <td>2021-07-18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>545</th>\n",
       "      <td>2019-12-22</td>\n",
       "      <td>44</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6.265474e+04</td>\n",
       "      <td>1.241497e+05</td>\n",
       "      <td>-1930.359205</td>\n",
       "      <td>2021-07-18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>546</th>\n",
       "      <td>2019-12-29</td>\n",
       "      <td>44</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.657442e+04</td>\n",
       "      <td>9.034891e+04</td>\n",
       "      <td>-34688.853747</td>\n",
       "      <td>2021-07-18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>547</th>\n",
       "      <td>2020-01-05</td>\n",
       "      <td>44</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8.801264e+03</td>\n",
       "      <td>7.213202e+04</td>\n",
       "      <td>-53411.941180</td>\n",
       "      <td>2021-07-18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>548</th>\n",
       "      <td>2020-01-12</td>\n",
       "      <td>44</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.665900e+03</td>\n",
       "      <td>6.847400e+04</td>\n",
       "      <td>-56652.550310</td>\n",
       "      <td>2021-07-18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>549</th>\n",
       "      <td>2020-01-19</td>\n",
       "      <td>44</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.708830e+03</td>\n",
       "      <td>6.266114e+04</td>\n",
       "      <td>-60661.669656</td>\n",
       "      <td>2021-07-18</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>550 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            ds  store_id             y          yhat    yhat_upper  \\\n",
       "0   2019-01-06        25  7.366572e+05  7.125751e+05  9.511452e+05   \n",
       "1   2019-01-13        25  6.964572e+05  7.437518e+05  9.741233e+05   \n",
       "2   2019-01-20        25  9.968769e+05  9.729099e+05  1.215014e+06   \n",
       "3   2019-01-27        25  1.029737e+06  1.034448e+06  1.264969e+06   \n",
       "4   2019-02-03        25  1.072874e+06  9.863657e+05  1.208138e+06   \n",
       "..         ...       ...           ...           ...           ...   \n",
       "545 2019-12-22        44           NaN  6.265474e+04  1.241497e+05   \n",
       "546 2019-12-29        44           NaN  2.657442e+04  9.034891e+04   \n",
       "547 2020-01-05        44           NaN  8.801264e+03  7.213202e+04   \n",
       "548 2020-01-12        44           NaN  4.665900e+03  6.847400e+04   \n",
       "549 2020-01-19        44           NaN  2.708830e+03  6.266114e+04   \n",
       "\n",
       "        yhat_lower training_date  \n",
       "0    481264.825160    2021-07-18  \n",
       "1    504054.660737    2021-07-18  \n",
       "2    759465.238607    2021-07-18  \n",
       "3    811822.715207    2021-07-18  \n",
       "4    754330.109414    2021-07-18  \n",
       "..             ...           ...  \n",
       "545   -1930.359205    2021-07-18  \n",
       "546  -34688.853747    2021-07-18  \n",
       "547  -53411.941180    2021-07-18  \n",
       "548  -56652.550310    2021-07-18  \n",
       "549  -60661.669656    2021-07-18  \n",
       "\n",
       "[550 rows x 7 columns]"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "id": "jtc2_p12KBbO"
   },
   "outputs": [],
   "source": [
    "final_df=final_df.set_index('ds')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 526
    },
    "id": "7CKgGuh_KBYX",
    "outputId": "f860bd31-7dd6-4d2c-c5d9-76b98f0a96ad"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f78503b9fd0>"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "final_df.query('store_id==44')[['y','yhat']].plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 526
    },
    "id": "r11ePm2TvAfz",
    "outputId": "00b65550-4106-4290-f4b2-247dbf691cf5"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f784f270128>"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "final_df.query('store_id==26')[['y','yhat']].plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 536
    },
    "id": "wnvd3OdXKBU1",
    "outputId": "ad34e844-9517-4e45-a56f-1bbb129d32c9"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f784f14eba8>"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "final_df.query('store_id==31')[['y','yhat']].plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "id": "J3hfLt3ZP8yv"
   },
   "outputs": [],
   "source": [
    "#END"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "authorship_tag": "ABX9TyP5E/vB/lfYMLc+0oqcRxB1",
   "collapsed_sections": [],
   "include_colab_link": true,
   "name": "Multiple Time Series using Apache Spark and Prophet.ipynb",
   "provenance": []
  },
  "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": "3.6.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}