PySpark_ML.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "0ece5315",
   "metadata": {},
   "source": [
    "# ML task on Used Cars dataset\n",
    "**Target:**\n",
    "Predict how many days `duration_listed` it will take to sell a car on a web platform."
   ]
  },
  {
   "cell_type": "markdown",
   "source": [
    "<img src=\"cars.jpg\" width=1256 height=353 />"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "642cfe57",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyspark.sql import SparkSession"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "74e97b4c",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "21/12/13 09:28:58 WARN Utils: Your hostname, Dmitrijs-MacBook-Pro.local resolves to a loopback address: 127.0.0.1; using 10.16.104.9 instead (on interface en0)\n",
      "21/12/13 09:28:58 WARN Utils: Set SPARK_LOCAL_IP if you need to bind to another address\n",
      "Using Spark's default log4j profile: org/apache/spark/log4j-defaults.properties\n",
      "Setting default log level to \"WARN\".\n",
      "To adjust logging level use sc.setLogLevel(newLevel). For SparkR, use setLogLevel(newLevel).\n",
      "21/12/13 09:29:02 WARN NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable\n"
     ]
    }
   ],
   "source": [
    "appName = \"ML pipline\"\n",
    "master = \"local[4]\"\n",
    "\n",
    "spark = SparkSession.builder.master(master).appName(appName).getOrCreate()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "4ba79ed0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": "<pyspark.sql.session.SparkSession at 0x10d0edc90>",
      "text/html": "\n            <div>\n                <p><b>SparkSession - in-memory</b></p>\n                \n        <div>\n            <p><b>SparkContext</b></p>\n\n            <p><a href=\"http://10.16.104.9:4040\">Spark UI</a></p>\n\n            <dl>\n              <dt>Version</dt>\n                <dd><code>v3.2.0</code></dd>\n              <dt>Master</dt>\n                <dd><code>local[4]</code></dd>\n              <dt>AppName</dt>\n                <dd><code>ML pipline</code></dd>\n            </dl>\n        </div>\n        \n            </div>\n        "
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "spark"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "64c6785a",
   "metadata": {},
   "source": [
    "# 0. Reading Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "6b97906b",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                \r"
     ]
    }
   ],
   "source": [
    "df_cars = spark.read.csv('cars.csv', header='True', inferSchema=True)\n",
    "TARGET = 'duration_listed'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "0ea5e0e6",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_cars = df_cars.dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "2a926d72",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "root\n",
      " |-- manufacturer_name: string (nullable = true)\n",
      " |-- model_name: string (nullable = true)\n",
      " |-- transmission: string (nullable = true)\n",
      " |-- odometer_value: integer (nullable = true)\n",
      " |-- year_produced: integer (nullable = true)\n",
      " |-- engine_capacity: double (nullable = true)\n",
      " |-- body_type: string (nullable = true)\n",
      " |-- price_usd: double (nullable = true)\n",
      " |-- location_region: string (nullable = true)\n",
      " |-- number_of_photos: integer (nullable = true)\n",
      " |-- up_counter: integer (nullable = true)\n",
      " |-- duration_listed: integer (nullable = true)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "df_cars.printSchema()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c9566b28",
   "metadata": {},
   "source": [
    "### Separating different data types"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "e1ccc9b3",
   "metadata": {},
   "outputs": [],
   "source": [
    "str_cols = [item[0] for item in df_cars.dtypes if item[1].startswith('string')]\n",
    "\n",
    "float_cols = [item[0] for item in df_cars.dtypes if item[1].startswith('double')]\n",
    "\n",
    "int_cols = [item[0] for item in df_cars.dtypes if item[1].startswith('int')]\n",
    "\n",
    "bool_cols = [item[0] for item in df_cars.dtypes if item[1].startswith('boolean')]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b3064ae7",
   "metadata": {},
   "source": [
    "# 1. Splitting the data into train and test"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "81d8584f",
   "metadata": {},
   "source": [
    "### 1.1 Simple splitting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "eb8f0d8a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split the data into train and test sets\n",
    "train, test = df_cars.randomSplit([.75, .25], seed=1234)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "bf51b069",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Stage 5:>                                                          (0 + 1) / 1]\r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train size: 28854\n",
      "Test size:  9667\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                \r"
     ]
    }
   ],
   "source": [
    "print(f'Train size: {train.count()}\\nTest size:  {test.count()}')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c6325f47",
   "metadata": {},
   "source": [
    "**Note**\n",
    "\n",
    "Stratified split need to be done by hands"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "08ce32da",
   "metadata": {},
   "source": [
    "### 1.2 Stratified splitting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "e25b6d5c",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Stage 8:>                                                          (0 + 1) / 1]\r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+------------+-----+\n",
      "|transmission|count|\n",
      "+------------+-----+\n",
      "|   automatic|12888|\n",
      "|  mechanical|25633|\n",
      "+------------+-----+\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                \r"
     ]
    }
   ],
   "source": [
    "df_cars.groupby('transmission').count().show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "0dafd89c",
   "metadata": {},
   "outputs": [],
   "source": [
    "zeros = df_cars.filter(\"transmission = 'automatic'\")\n",
    "ones = df_cars.filter(\"transmission = 'mechanical'\")\n",
    "# split datasets into training and testing\n",
    "train0, test0 = zeros.randomSplit([.75, .25], seed=1234)\n",
    "train1, test1 = ones.randomSplit([.75, .25], seed=1234)\n",
    "# stack datasets back together\n",
    "train = train0.union(train1)\n",
    "test = test0.union(test1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "9908ecda",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                \r"
     ]
    },
    {
     "data": {
      "text/plain": "(28900, 9621)"
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.count(), test.count()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "54c84b60",
   "metadata": {},
   "source": [
    "# 2. Encoding Categorical variables"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3fd2a8b6",
   "metadata": {},
   "source": [
    "### 2.1 Label encoding (string indexing)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "5f712692",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyspark.ml.feature import StringIndexer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "ec4106aa",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                \r"
     ]
    }
   ],
   "source": [
    "str_cols_ind =[col +'_ind' for col in str_cols]\n",
    "\n",
    "str_indexer = StringIndexer(inputCols=str_cols, outputCols=str_cols_ind)\n",
    "\n",
    "train = str_indexer.fit(train).transform(train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "e7a40f1b",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Stage 20:=============================>                            (1 + 1) / 2]\r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+-----------------+---------------------+\n",
      "|manufacturer_name|manufacturer_name_ind|\n",
      "+-----------------+---------------------+\n",
      "|       Volkswagen|                  0.0|\n",
      "|             Opel|                  1.0|\n",
      "|              BMW|                  2.0|\n",
      "|             Ford|                  3.0|\n",
      "|             Audi|                  4.0|\n",
      "+-----------------+---------------------+\n",
      "only showing top 5 rows\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                \r"
     ]
    }
   ],
   "source": [
    "train.select('manufacturer_name', 'manufacturer_name_ind').distinct().sort('manufacturer_name_ind').show(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fa801085",
   "metadata": {},
   "source": [
    "### 2.2 One hot encoding"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "d86230b5",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyspark.ml.feature import OneHotEncoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "3fbc4559",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a OneHotEncoder\n",
    "\n",
    "str_dummys = [col + '_dummy' for col in str_cols]\n",
    "\n",
    "onehot = OneHotEncoder(inputCols=str_cols_ind, outputCols=str_dummys)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "d14cf93e",
   "metadata": {},
   "outputs": [],
   "source": [
    "train = onehot.fit(train).transform(train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "ad808d94",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Stage 23:=============================>                            (1 + 1) / 2]\r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+-----------------+---------------------+-----------------------+\n",
      "|manufacturer_name|manufacturer_name_ind|manufacturer_name_dummy|\n",
      "+-----------------+---------------------+-----------------------+\n",
      "|       Volkswagen|                  0.0|         (54,[0],[1.0])|\n",
      "|             Opel|                  1.0|         (54,[1],[1.0])|\n",
      "|              BMW|                  2.0|         (54,[2],[1.0])|\n",
      "|             Ford|                  3.0|         (54,[3],[1.0])|\n",
      "|             Audi|                  4.0|         (54,[4],[1.0])|\n",
      "|          Renault|                  5.0|         (54,[5],[1.0])|\n",
      "|    Mercedes-Benz|                  6.0|         (54,[6],[1.0])|\n",
      "|          Peugeot|                  7.0|         (54,[7],[1.0])|\n",
      "|          Citroen|                  8.0|         (54,[8],[1.0])|\n",
      "|           Nissan|                  9.0|         (54,[9],[1.0])|\n",
      "|            Mazda|                 10.0|        (54,[10],[1.0])|\n",
      "|           Toyota|                 11.0|        (54,[11],[1.0])|\n",
      "|          Hyundai|                 12.0|        (54,[12],[1.0])|\n",
      "|            Skoda|                 13.0|        (54,[13],[1.0])|\n",
      "|              Kia|                 14.0|        (54,[14],[1.0])|\n",
      "|       Mitsubishi|                 15.0|        (54,[15],[1.0])|\n",
      "|             Fiat|                 16.0|        (54,[16],[1.0])|\n",
      "|            Honda|                 17.0|        (54,[17],[1.0])|\n",
      "|            Volvo|                 18.0|        (54,[18],[1.0])|\n",
      "|              ВАЗ|                 19.0|        (54,[19],[1.0])|\n",
      "+-----------------+---------------------+-----------------------+\n",
      "only showing top 20 rows\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                \r"
     ]
    }
   ],
   "source": [
    "train.select('manufacturer_name',\n",
    "             'manufacturer_name_ind',\n",
    "             'manufacturer_name_dummy')\\\n",
    "            .distinct()\\\n",
    "            .sort('manufacturer_name_ind')\\\n",
    "            .show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aea403b1",
   "metadata": {},
   "source": [
    "### `Note` Dense vs Sparse vectors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "1b8d6c0e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": "DenseVector([1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 7.0, 0.0, 0.0])"
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from pyspark.ml.linalg import SparseVector, DenseVector\n",
    "\n",
    "DenseVector([1, 0, 0, 0, 0, 0, 7, 0, 0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "2bd7b2d7",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": "SparseVector(9, {0: 1.0, 6: 7.0})"
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "SparseVector(9, [0, 6], [1, 7])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "86f377a6",
   "metadata": {},
   "source": [
    "# 4. Vectorizing predictors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "abd3a95d",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyspark.ml.feature import VectorAssembler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "4e006276",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Stage 26:>                                                         (0 + 1) / 1]\r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+--------------------+\n",
      "|            features|\n",
      "+--------------------+\n",
      "|(1150,[0,1,2,3,4,...|\n",
      "|(1150,[0,1,2,3,4,...|\n",
      "|(1150,[0,1,2,3,4,...|\n",
      "|(1150,[0,1,2,3,4,...|\n",
      "|(1150,[0,1,2,3,4,...|\n",
      "|(1150,[0,1,2,3,4,...|\n",
      "|(1150,[0,1,2,3,4,...|\n",
      "|(1150,[0,1,2,3,4,...|\n",
      "|(1150,[0,1,2,3,4,...|\n",
      "|(1150,[0,1,2,3,4,...|\n",
      "|(1150,[0,1,2,3,4,...|\n",
      "|(1150,[0,1,2,3,4,...|\n",
      "|(1150,[0,1,2,3,4,...|\n",
      "|(1150,[0,1,2,3,4,...|\n",
      "|(1150,[0,1,2,3,4,...|\n",
      "|(1150,[0,1,2,3,4,...|\n",
      "|(1150,[0,1,2,3,4,...|\n",
      "|(1150,[0,1,2,3,4,...|\n",
      "|(1150,[0,1,2,3,4,...|\n",
      "|(1150,[0,1,2,3,4,...|\n",
      "+--------------------+\n",
      "only showing top 20 rows\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                \r"
     ]
    }
   ],
   "source": [
    "train = train.select(int_cols + float_cols + str_dummys)\n",
    "\n",
    "vec_assembler = VectorAssembler(inputCols=train.columns, outputCol='features')\n",
    "\n",
    "train = vec_assembler.transform(train)\n",
    "\n",
    "train.select('features').show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d00f3192",
   "metadata": {},
   "source": [
    "# 5. Baseline model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "39426350",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pyspark.ml"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "outputs": [
    {
     "data": {
      "text/plain": "['ABCMeta',\n 'AFTSurvivalRegression',\n 'AFTSurvivalRegressionModel',\n 'DataFrame',\n 'DecisionTreeRegressionModel',\n 'DecisionTreeRegressor',\n 'FMRegressionModel',\n 'FMRegressor',\n 'GBTRegressionModel',\n 'GBTRegressor',\n 'GeneralJavaMLWritable',\n 'GeneralizedLinearRegression',\n 'GeneralizedLinearRegressionModel',\n 'GeneralizedLinearRegressionSummary',\n 'GeneralizedLinearRegressionTrainingSummary',\n 'HasAggregationDepth',\n 'HasElasticNetParam',\n 'HasFeaturesCol',\n 'HasFitIntercept',\n 'HasLabelCol',\n 'HasLoss',\n 'HasMaxBlockSizeInMB',\n 'HasMaxIter',\n 'HasPredictionCol',\n 'HasRegParam',\n 'HasSeed',\n 'HasSolver',\n 'HasStandardization',\n 'HasStepSize',\n 'HasTol',\n 'HasTrainingSummary',\n 'HasVarianceCol',\n 'HasWeightCol',\n 'IsotonicRegression',\n 'IsotonicRegressionModel',\n 'JavaEstimator',\n 'JavaMLReadable',\n 'JavaMLWritable',\n 'JavaModel',\n 'JavaPredictionModel',\n 'JavaPredictor',\n 'JavaWrapper',\n 'LinearRegression',\n 'LinearRegressionModel',\n 'LinearRegressionSummary',\n 'LinearRegressionTrainingSummary',\n 'Param',\n 'Params',\n 'PredictionModel',\n 'Predictor',\n 'RandomForestRegressionModel',\n 'RandomForestRegressor',\n 'RegressionModel',\n 'Regressor',\n 'TypeConverters',\n '_AFTSurvivalRegressionParams',\n '_DecisionTreeModel',\n '_DecisionTreeParams',\n '_DecisionTreeRegressorParams',\n '_FactorizationMachinesParams',\n '_GBTParams',\n '_GBTRegressorParams',\n '_GeneralizedLinearRegressionParams',\n '_IsotonicRegressionParams',\n '_JavaRegressionModel',\n '_JavaRegressor',\n '_LinearRegressionParams',\n '_PredictorParams',\n '_RandomForestParams',\n '_RandomForestRegressorParams',\n '_TreeEnsembleModel',\n '_TreeRegressorParams',\n '__all__',\n '__builtins__',\n '__cached__',\n '__doc__',\n '__file__',\n '__loader__',\n '__name__',\n '__package__',\n '__spec__',\n 'inherit_doc',\n 'keyword_only',\n 'since',\n 'sys']"
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dir(pyspark.ml.regression)\n",
    "dir(pyspark.ml.regression)"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "703c0404",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyspark.ml.regression import LinearRegression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "53932736",
   "metadata": {},
   "outputs": [],
   "source": [
    "reg = LinearRegression(labelCol=TARGET, featuresCol='features')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "a11fe048",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "21/12/13 09:34:29 WARN Instrumentation: [c33069d2] regParam is zero, which might cause numerical instability and overfitting.\n",
      "21/12/13 09:34:32 WARN InstanceBuilder$NativeBLAS: Failed to load implementation from:dev.ludovic.netlib.blas.JNIBLAS\n",
      "21/12/13 09:34:33 WARN InstanceBuilder$NativeBLAS: Failed to load implementation from:dev.ludovic.netlib.blas.ForeignLinkerBLAS\n",
      "21/12/13 09:34:34 WARN InstanceBuilder$NativeLAPACK: Failed to load implementation from:dev.ludovic.netlib.lapack.JNILAPACK\n",
      "21/12/13 09:34:35 WARN Instrumentation: [c33069d2] Cholesky solver failed due to singular covariance matrix. Retrying with Quasi-Newton solver.\n",
      "21/12/13 09:34:36 WARN BLAS: Failed to load implementation from: com.github.fommil.netlib.NativeSystemBLAS\n",
      "21/12/13 09:34:36 WARN BLAS: Failed to load implementation from: com.github.fommil.netlib.NativeRefBLAS\n",
      "                                                                                \r"
     ]
    }
   ],
   "source": [
    "reg = reg.fit(train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "8dcf42b8",
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions = reg.transform(train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "e6cf3fd8",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Stage 29:>                                                         (0 + 1) / 1]\r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+------------------+---------------+\n",
      "|        prediction|duration_listed|\n",
      "+------------------+---------------+\n",
      "|171.99966937750236|            172|\n",
      "|152.00043232862572|            152|\n",
      "| 61.00039985877073|             61|\n",
      "|228.00881627833198|            228|\n",
      "|21.004959882561277|             21|\n",
      "| 124.0051284774795|            124|\n",
      "|1181.9997296909503|           1182|\n",
      "|186.00255264048337|            186|\n",
      "|  118.998348281514|            119|\n",
      "|40.000285416312664|             40|\n",
      "|2078.9957483423136|           2079|\n",
      "| 4.998532476567573|              5|\n",
      "|11.005056153951378|             11|\n",
      "|265.00008605794403|            265|\n",
      "|154.00028038763838|            154|\n",
      "|  789.997760545205|            790|\n",
      "|  90.9983982943395|             91|\n",
      "| 7.998422849330582|              8|\n",
      "| 66.99337051761036|             67|\n",
      "|251.99684822765752|            252|\n",
      "+------------------+---------------+\n",
      "only showing top 20 rows\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                \r"
     ]
    }
   ],
   "source": [
    "predictions.select('prediction', TARGET).show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "e3a825db",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyspark.ml.evaluation import RegressionEvaluator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "76a5115a",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Stage 30:=============================>                            (1 + 1) / 2]\r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R2 on train data = 0.9999999943022991\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                \r"
     ]
    }
   ],
   "source": [
    "lr_evaluator = RegressionEvaluator(\n",
    "    labelCol=TARGET, predictionCol=\"prediction\", metricName=\"r2\")\n",
    "r2 = lr_evaluator.evaluate(predictions)\n",
    "print(f'R2 on train data = {r2}')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4ad3ffe8",
   "metadata": {},
   "source": [
    "# 6. Pipline"
   ]
  },
  {
   "attachments": {
    "Screenshot%202021-12-12%20at%202.24.36%20AM.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "id": "4cf6eca8",
   "metadata": {},
   "source": [
    "![Screenshot%202021-12-12%20at%202.24.36%20AM.png](attachment:Screenshot%202021-12-12%20at%202.24.36%20AM.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "e0ae3a87",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyspark.ml import Pipeline\n",
    "\n",
    "from pyspark.ml.regression import GBTRegressor\n",
    "\n",
    "stages = []\n",
    "\n",
    "#depend on categorical columns: country and types of emission\n",
    "\n",
    "for categoricalCol in str_cols:\n",
    "    stringIndexer = StringIndexer(inputCol = categoricalCol,\n",
    "                                  outputCol = categoricalCol + '_Index', handleInvalid = 'keep')\n",
    "    encoder = OneHotEncoder(inputCol=stringIndexer.getOutputCol(),\n",
    "                            outputCol=categoricalCol + \"_classVec\")\n",
    "    stages.extend([stringIndexer, encoder])\n",
    "\n",
    "assemblerInputs = [c + \"_classVec\" for c in str_cols] + float_cols + int_cols\n",
    "\n",
    "# transfrom columns\n",
    "assembler = VectorAssembler(inputCols=assemblerInputs, outputCol=\"features\")\n",
    "stages.append(assembler)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "97e2746c",
   "metadata": {},
   "outputs": [],
   "source": [
    "# train/test split (25% test)\n",
    "trainingData, testData = df_cars.randomSplit([0.75, 0.25])\n",
    "\n",
    "# Train Gradient Boosting on trees\n",
    "gbt = GBTRegressor(labelCol=TARGET, featuresCol=\"features\", maxIter=100)\n",
    "stages.append(gbt)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "e3d6dd15",
   "metadata": {},
   "outputs": [],
   "source": [
    "# build pipline\n",
    "pipeline = Pipeline(stages=stages)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "4e5c65a6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": "{Param(parent='Pipeline_e48a9f83fae4', name='stages', doc='a list of pipeline stages'): [StringIndexer_75028316c10a,\n  OneHotEncoder_e8eb8d7d0208,\n  StringIndexer_a57dfa328cdb,\n  OneHotEncoder_b9bd11cf5b74,\n  StringIndexer_55c1d65eef99,\n  OneHotEncoder_daed2730b553,\n  StringIndexer_3a48b31a8129,\n  OneHotEncoder_0c71e099b4ae,\n  StringIndexer_bc49f9767605,\n  OneHotEncoder_508e6c2109de,\n  VectorAssembler_a92153f04556,\n  GBTRegressor_82d01005af6f]}"
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pipeline.extractParamMap()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "0c1cfc67",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                \r"
     ]
    }
   ],
   "source": [
    "# тренируем модель\n",
    "model = pipeline.fit(trainingData)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "9e9d4816",
   "metadata": {},
   "outputs": [],
   "source": [
    "# делаем предсказания на тестовой выборке\n",
    "predictions = model.transform(testData)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d0ac6cd4",
   "metadata": {},
   "source": [
    "# 7. Validating the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "fc1a4723",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+---------------+------------------+\n",
      "|duration_listed|        prediction|\n",
      "+---------------+------------------+\n",
      "|             61|56.655730313094544|\n",
      "|              4| 7.842711253774789|\n",
      "|            228| 315.5032535080454|\n",
      "|             21|32.275357676082386|\n",
      "|              5| 8.866510680077619|\n",
      "|            265| 802.4130740066757|\n",
      "|              8| 8.866510680077619|\n",
      "|             63| 71.09484842356053|\n",
      "|            252| 322.3089861038882|\n",
      "|             90| 89.70395596333664|\n",
      "|             43| 40.32774543866011|\n",
      "|             81| 79.01846394233387|\n",
      "|             92| 89.68249438734743|\n",
      "|             36| 39.58584747699532|\n",
      "|            121|122.28051140789209|\n",
      "|            188|186.34825542857732|\n",
      "|             25| 26.46774537831986|\n",
      "|             58| 56.63426873710533|\n",
      "|            100|107.07030157462191|\n",
      "|            208|231.91639673468654|\n",
      "+---------------+------------------+\n",
      "only showing top 20 rows\n",
      "\n"
     ]
    }
   ],
   "source": [
    "predictions.select(TARGET, \"prediction\").show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "6983a761",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "[Stage 1057:>                                                       (0 + 1) / 1]\r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R2 on test data = 0.74\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "                                                                                \r"
     ]
    }
   ],
   "source": [
    "gbt_evaluator = RegressionEvaluator(\n",
    "    labelCol=TARGET, predictionCol=\"prediction\", metricName=\"r2\")\n",
    "r2 = gbt_evaluator.evaluate(predictions)\n",
    "print(f\"R2 on test data = {r2:.2f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c189c33e",
   "metadata": {},
   "source": [
    "# 8. Save model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "6690f38a",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.save('model_GBT')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "386f0b8d",
   "metadata": {},
   "outputs": [],
   "source": [
    "spark.stop()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "77eb5d75",
   "metadata": {},
   "source": [
    "---"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.10.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}