DSF_RF.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "710\n",
      "0.6168549087749783\n",
      "735\n",
      "0.6385751520417029\n",
      "708\n",
      "0.6151172893136403\n",
      "838\n",
      "0.7280625543006082\n",
      "850\n",
      "0.738488271068636\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "707\n",
      "0.6142484795829714\n",
      "705\n",
      "0.6125108601216334\n",
      "720\n",
      "0.6255430060816681\n",
      "719\n",
      "0.6246741963509991\n",
      "691\n",
      "0.6003475238922676\n",
      "691\n",
      "0.6003475238922676\n",
      "894\n",
      "0.7767158992180713\n",
      "894\n",
      "0.7767158992180713\n",
      "894\n",
      "0.7767158992180713\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "719\n",
      "0.6246741963509991\n",
      "706\n",
      "0.6133796698523023\n",
      "706\n",
      "0.6133796698523023\n",
      "719\n",
      "0.6246741963509991\n",
      "719\n",
      "0.6246741963509991\n",
      "708\n",
      "0.6151172893136403\n",
      "708\n",
      "0.6151172893136403\n",
      "708\n",
      "0.6151172893136403\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "705\n",
      "0.6125108601216334\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "712\n",
      "0.6185925282363163\n",
      "708\n",
      "0.6151172893136403\n",
      "708\n",
      "0.6151172893136403\n",
      "697\n",
      "0.6055603822762815\n",
      "826\n",
      "0.7176368375325803\n",
      "815\n",
      "0.7080799304952216\n",
      "678\n",
      "0.5890529973935708\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n",
      "702\n",
      "0.6099044309296264\n"
     ]
    }
   ],
   "source": [
    "# set dataset name and run\n",
    "# The output accuracy files for both majority voting and average probabilities\n",
    "# are exported to the folder DSF Random Forest/dataset/\n",
    "# the last line in both files contains the best variant of RF_d_tT where d is in {5,10,15,20}, and T in [2,25]\n",
    "# The outputs of training process are exported to the folder DSF Random Forest/dataset/DetailsFiles/\n",
    "dataset='spambase'\n",
    "\n",
    "import csv,operator,sys,os\n",
    "import numpy as np\n",
    "import sklearn\n",
    "import json\n",
    "import FeatureGenerators.ReadData as ReadData\n",
    "import time\n",
    "from functools import reduce\n",
    "import numpy as np\n",
    "\n",
    "\n",
    "sys.path.append('arch-forest/data/adult/')\n",
    "sys.path.append('arch-forest/data/bank/')\n",
    "sys.path.append('arch-forest/data/wine-quality/')\n",
    "sys.path.append('arch-forest/data/')\n",
    "sys.path.append('arch-forest/code/')\n",
    "import trainForest\n",
    "import Tree\n",
    "\n",
    "\n",
    "filesPath = \"../frequentTreesInRandomForests/arch-forest/data/\"\n",
    "frequentTreesPath = \"../frequentTreesInRandomForests/forests/rootedFrequentTrees/\"\n",
    "resultsPath = \"../frequentTreesInRandomForests/DSF Random Forest/\"\n",
    "\n",
    "\n",
    "#import train data\n",
    "from FeatureGenerators.ReadData import readDataAdult,readWine, readWineTest\n",
    "\n",
    "if (dataset == 'adult'):\n",
    "    X_train,Y_train = ReadData.readDataAdultNormal('train')\n",
    "    X_test,Y_test = ReadData.readDataAdultNormal('test')\n",
    "if (dataset == 'spambase'):\n",
    "    X_train,Y_train = ReadData.readDataSpambase('train')\n",
    "    X_test,Y_test = ReadData.readDataSpambase('test')\n",
    "if (dataset == 'letter'):\n",
    "    X_train,Y_train = ReadData.readDataLetter('train')\n",
    "    X_test,Y_test = ReadData.readDataLetter('test')    \n",
    "if (dataset == 'wine-quality'):\n",
    "        X_train,Y_train = ReadData.readWine()\n",
    "        #X_test,Y_test = ReadData.readWineTest()\n",
    "\n",
    "\n",
    "\n",
    "# Classification\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.utils.estimator_checks import check_estimator\n",
    "from sklearn.metrics import accuracy_score\n",
    "    \n",
    "import FeatureGenerators.DecisionSnippetFeatures\n",
    "\n",
    "results_list = []\n",
    "time_list = []\n",
    "scoring_function = 'accuracy'\n",
    "pattern_max_size=6\n",
    "variant = 'NoLeafEdges'\n",
    "countsList =[]\n",
    "\n",
    "predictions_results_majority_vote = []\n",
    "predictions_results_average_probability = []\n",
    "best_prediction_majority_vote = 0\n",
    "best_prediction_majority_vote_dsf = ''\n",
    "best_prediction_average_probability = 0\n",
    "best_prediction_average_probability_dsf = ''\n",
    "\n",
    "for rf_depth in (5,10,15,20):\n",
    "    for frequency in range(2,26):\n",
    "\n",
    "    \n",
    "    \n",
    "        rootedFrequentTrees = \"RF_\"+str(rf_depth)+\"_t\"+str(frequency)\n",
    "        f = open(frequentTreesPath+dataset+'/'+variant+'/leq'+str(pattern_max_size)+'/'+rootedFrequentTrees+'.json')\n",
    "        frequentpatterns = json.load(f)\n",
    "        f.close()\n",
    "\n",
    "        dsf = FeatureGenerators.DecisionSnippetFeatures.FrequentSubtreeFeatures(map(lambda x: x['pattern'], frequentpatterns[-200:])) \n",
    "\n",
    "        fts = dsf.fit_transform(X_train,0)\n",
    "\n",
    "\n",
    "\n",
    "        class_labels_list = ''\n",
    "        class_labels_list_P = ''\n",
    "\n",
    "        one_counter=0\n",
    "        zero_counter=0\n",
    "        max_id = 0\n",
    "        max_class_id = 0\n",
    "        trees_count = len(fts[0])\n",
    "\n",
    "        for i in range(0,len(fts)):\n",
    "            if (Y_train[i] > max_class_id):\n",
    "                max_class_id +=1\n",
    "\n",
    "\n",
    "            for j in range(0,len(fts[i])):\n",
    "                if (fts[i][j] > max_id):\n",
    "                    max_id = fts[i][j]\n",
    "                  \n",
    "        id_count= max_id + 1            \n",
    "        class_count = max_class_id + 1  \n",
    "\n",
    "        for i in range(0,class_count):\n",
    "            class_labels_list += 'Y = '+str(i)+','\n",
    "            class_labels_list_P += 'P(Y) = '+str(i)+','\n",
    "    \n",
    "            \n",
    "          \n",
    "        \n",
    "\n",
    "        Matrix = np.zeros((trees_count,id_count,class_count))\n",
    "\n",
    "\n",
    "        for i in range(0,len(fts)):\n",
    "            for j in range(0,trees_count):\n",
    "\n",
    "                     Matrix[j][fts[i][j]][Y_train[i]] +=1\n",
    "\n",
    "                            \n",
    " \n",
    "\n",
    "\n",
    "        \n",
    "        Prefered_Class_Matrix = np.zeros((trees_count,id_count))\n",
    "        Prefered_Class_Matrix_P = np.zeros((trees_count,id_count,class_count))\n",
    "        f= open(resultsPath+'/'+dataset+'/Details_Files/'+'DSF_as_RF_'+str(rf_depth)+'_t'+str(frequency)+'_'+dataset+'_n.csv',\"w\")\n",
    "        f.write('Feature ,Leaf id,'+class_labels_list+' Preferred Class, '+class_labels_list_P+'\\n')\n",
    "        for j in range(0,trees_count):\n",
    "            for k in range(0,id_count):\n",
    "                f.write(str(j)+','+str(k)+',')\n",
    "                for c in range(0,class_count):\n",
    "                        f.write(str(Matrix[j][k][c])+',')\n",
    "                Prefered_Class_Matrix[j][k] = np.argmax(Matrix[j][k])       \n",
    "                f.write(str(np.argmax(Matrix[j][k]))+',')  \n",
    "                for c in range(0,class_count):\n",
    "                    if (np.sum(Matrix[j][k]) != 0):\n",
    "                        f.write(str(round(Matrix[j][k][c]/np.sum(Matrix[j][k]),3))+',')\n",
    "                        Prefered_Class_Matrix_P[j][k][c] = round(Matrix[j][k][c]/np.sum(Matrix[j][k]),3)\n",
    "                    else:\n",
    "                        f.write('0,')\n",
    "                f.write('\\n')    \n",
    "\n",
    "        f.close()            \n",
    "        \n",
    "        \n",
    "        fts_test = dsf.fit_transform(X_test,0)\n",
    "\n",
    "\n",
    "\n",
    "        labels_array = [0] * class_count\n",
    "        labels_array_p = [0] * class_count\n",
    "        predictedClass = 0\n",
    "        correctPredictions = 0\n",
    "        predictedClass_p = 0\n",
    "        correctPredictions_p = 0\n",
    "\n",
    "        for i in range(0,len(fts_test)):\n",
    "            labels_array = [0] * class_count\n",
    "            for j in range(0,trees_count):\n",
    "                for c in range(0,class_count):\n",
    "\n",
    "                    if (Prefered_Class_Matrix[j][fts_test[i][j]] == c):\n",
    "                        labels_array[c] += 1\n",
    "\n",
    "                        labels_array_p[c] += Prefered_Class_Matrix_P[j][fts_test[i][j]][c] \n",
    "\n",
    "            predictedClass =np.argmax(labels_array)\n",
    "            predictedClass_p =np.argmax(labels_array_p)\n",
    "\n",
    "            if (predictedClass == Y_test[i]):\n",
    "                correctPredictions += 1\n",
    "            labels_array = [0] * class_count\n",
    "            if (predictedClass_p == Y_test[i]):\n",
    "                correctPredictions_p += 1\n",
    "            labels_array_p = [0] * class_count\n",
    "\n",
    "        print(correctPredictions)\n",
    "        \n",
    "        print(correctPredictions/len(fts_test))\n",
    "        predictions_results_majority_vote.append(correctPredictions/len(fts_test))\n",
    "        predictions_results_average_probability.append(correctPredictions_p/len(fts_test))\n",
    "        \n",
    "        if (correctPredictions/len(fts_test) > best_prediction_majority_vote):\n",
    "            best_prediction_majority_vote = correctPredictions/len(fts_test)\n",
    "            best_prediction_majority_vote_dsf = 'RF_'+str(rf_depth)+'_t'+str(frequency)\n",
    "            \n",
    "        if (correctPredictions_p/len(fts_test) > best_prediction_average_probability):\n",
    "            best_prediction_average_probability = correctPredictions_p/len(fts_test)\n",
    "            best_prediction_average_probability_dsf = 'RF_'+str(rf_depth)+'_t'+str(frequency)    \n",
    "        \n",
    "        \n",
    "f= open(resultsPath+'/'+dataset+'/'+'DSF_as_RF_accuracy_'+dataset+'_majority_vote'+'.csv',\"w\")\n",
    "f.write('DSF,accuracy,\\n')\n",
    "counter = 0\n",
    "for rf_depth in (5,10,15,20):\n",
    "    for frequency in range(2,26):\n",
    "        f.write('RF_'+str(rf_depth)+'_t'+str(frequency)+','+str(predictions_results_majority_vote[counter])+','+'\\n')\n",
    "        counter += 1\n",
    "f.write('Best ,'+best_prediction_majority_vote_dsf+','+str(best_prediction_majority_vote)+',\\n')\n",
    "f.close()\n",
    "\n",
    "\n",
    "f= open(resultsPath+'/'+dataset+'/'+'DSF_as_RF_accuracy_'+dataset+'_average_probability'+'.csv',\"w\")\n",
    "f.write('DSF,accuracy,\\n')\n",
    "counter = 0\n",
    "for rf_depth in (5,10,15,20):\n",
    "    for frequency in range(2,26):\n",
    "        f.write('RF_'+str(rf_depth)+'_t'+str(frequency)+','+str(predictions_results_average_probability[counter])+','+'\\n')\n",
    "        counter += 1\n",
    "f.write('Best ,'+best_prediction_average_probability_dsf+','+str(best_prediction_average_probability)+',\\n')\n",
    "f.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2\n",
      "3\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "class_labels_list = ''\n",
    "class_labels_list_P = ''\n",
    "\n",
    "one_counter=0\n",
    "zero_counter=0\n",
    "max_id = 0\n",
    "max_class_id = 0\n",
    "trees_count = len(fts[0])\n",
    "\n",
    "for i in range(0,len(fts)):\n",
    "    if (Y_train[i] > max_class_id):\n",
    "        max_class_id +=1\n",
    "\n",
    "        \n",
    "    for j in range(0,len(fts[i])):\n",
    "        if (fts[i][j] > max_id):\n",
    "            max_id = fts[i][j]\n",
    "                  \n",
    "id_count= max_id + 1            \n",
    "class_count = max_class_id + 1  \n",
    "\n",
    "for i in range(0,class_count):\n",
    "    class_labels_list += 'Y = '+str(i)+','\n",
    "    class_labels_list_P += 'P(Y) = '+str(i)+','\n",
    "    \n",
    "            \n",
    "          \n",
    "print(class_count)  \n",
    "print(id_count)\n",
    "\n",
    "Matrix = np.zeros((trees_count,id_count,class_count))\n",
    "\n",
    "\n",
    "for i in range(0,len(fts)):\n",
    "    for j in range(0,trees_count):\n",
    "       \n",
    "             Matrix[j][fts[i][j]][Y_train[i]] +=1\n",
    "\n",
    "                            "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1 1 2 1 1 2 1 1 2 2 1 1 2 2 1 1 1 1 0 1 1]\n"
     ]
    }
   ],
   "source": [
    "print(fts[0]) \n",
    "Prefered_Class_Matrix = np.zeros((trees_count,id_count))\n",
    "Prefered_Class_Matrix_P = np.zeros((trees_count,id_count,class_count))\n",
    "f= open(resultsPath+'/'+dataset+'/'+'DSF_as_RF_'+str(rf_depth)+'_t'+str(frequency)+'_'+dataset+'_n.csv',\"w\")\n",
    "f.write('Feature ,Leaf id,'+class_labels_list+' Preferred Class, '+class_labels_list_P+'\\n')\n",
    "for j in range(0,trees_count):\n",
    "    for k in range(0,id_count):\n",
    "        f.write(str(j)+','+str(k)+',')\n",
    "        for c in range(0,class_count):\n",
    "                f.write(str(Matrix[j][k][c])+',')\n",
    "        Prefered_Class_Matrix[j][k] = np.argmax(Matrix[j][k])       \n",
    "        f.write(str(np.argmax(Matrix[j][k]))+',')  \n",
    "        for c in range(0,class_count):\n",
    "            if (np.sum(Matrix[j][k]) != 0):\n",
    "                f.write(str(round(Matrix[j][k][c]/np.sum(Matrix[j][k]),3))+',')\n",
    "                Prefered_Class_Matrix_P[j][k][c] = round(Matrix[j][k][c]/np.sum(Matrix[j][k]),3)\n",
    "            else:\n",
    "                f.write('0,')\n",
    "        f.write('\\n')    \n",
    "                \n",
    "f.close()            \n",
    "        \n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[  0.   0.   0.   0.   0.   0.   0.   0.   0.   0.   0.   0.   0.   0.\n",
      "    0.   0.   0.   0.   0.   0.   0.   0.   0.   0.   0.   0.]\n",
      " [789. 766. 658. 805. 754. 581. 773. 722. 755. 747. 673. 761. 749. 745.\n",
      "  753. 802. 771. 758. 744. 410. 427. 117. 635. 731. 191. 733.]\n",
      " [  0.   0.  78.   0.  14. 194.   0.  12.   0.   0.  66.   0.  43.  38.\n",
      "    0.   1.  12.   0.   4. 386. 386. 647. 117.  56. 595.   1.]\n",
      " [  0.   0.   0.   0.   0.   0.   0.   0.   0.   0.   0.   0.   0.   0.\n",
      "    0.   0.   0.   0.   0.   0.   0.   0.   0.   0.   0.   0.]\n",
      " [  0.   0.   0.   0.   0.   0.   0.   0.   0.   0.   0.   0.   0.   0.\n",
      "    0.   0.   0.   0.   0.   0.   0.   0.   0.   0.   0.   0.]]\n",
      "0\n",
      "17350\n",
      "2650\n",
      "0\n",
      "0\n",
      "20000\n"
     ]
    }
   ],
   "source": [
    "j = 69\n",
    "print(Matrix[j])\n",
    "\n",
    "zero_counter = 0\n",
    "one_counter = 0\n",
    "two_counter = 0\n",
    "three_counter = 0\n",
    "four_counter = 0\n",
    "for i in range(0,len(fts)):\n",
    "    if (fts[i][j] == 0):\n",
    "        zero_counter += 1\n",
    "    if (fts[i][j] == 1):\n",
    "        one_counter += 1\n",
    "    if (fts[i][j] == 2):\n",
    "        two_counter += 1\n",
    "    if (fts[i][j] == 3):\n",
    "        three_counter += 1\n",
    "    if (fts[i][j] == 4):\n",
    "        four_counter += 1    \n",
    "print(zero_counter)\n",
    "print(one_counter)\n",
    "print(two_counter)\n",
    "print(three_counter)\n",
    "print(four_counter)\n",
    "print(zero_counter + one_counter + two_counter + three_counter + four_counter)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "fts_test = dsf.fit_transform(X_test,0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3160431\n"
     ]
    }
   ],
   "source": [
    "\n",
    "correct_predictions = 0\n",
    "for i in range(0,len(Y_test)): \n",
    "    for j in range(0,trees_count):\n",
    "        for k in range(0,id_count):\n",
    "            if (Prefered_Class_Matrix[j][k] == Y_test[i]):\n",
    "                correct_predictions += 1\n",
    "\n",
    "print(correct_predictions)\n",
    "                "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "782\n",
      "0.6794092093831451\n"
     ]
    }
   ],
   "source": [
    "\n",
    "labels_array = [0] * class_count\n",
    "predictedClass = 0\n",
    "correctPredictions = 0\n",
    "\n",
    "for i in range(0,len(fts_test)):\n",
    "    labels_array = [0] * class_count\n",
    "    for j in range(0,trees_count):\n",
    "        for c in range(0,class_count):\n",
    "        \n",
    "            if (Prefered_Class_Matrix[j][fts_test[i][j]] == c):\n",
    "                #labels_array[c] += 1\n",
    "                \n",
    "                labels_array[c] += Prefered_Class_Matrix_P[j][fts_test[i][j]][c] \n",
    "    \n",
    "    predictedClass =np.argmax(labels_array)\n",
    "    \n",
    "    if (predictedClass == Y_test[i]):\n",
    "        correctPredictions += 1\n",
    "    labels_array = [0] * class_count\n",
    "\n",
    "print(correctPredictions)\n",
    "print(correctPredictions/len(fts_test))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1813\n",
      "4601\n"
     ]
    }
   ],
   "source": [
    "counter = 0\n",
    "for i in range(0,len(Y_train)):\n",
    "    if (Y_train[i] == 1):\n",
    "        counter += 1\n",
    "   \n",
    "print(counter)\n",
    "print(len(Y_train))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 193,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0. 0.]\n",
      "0.0\n",
      "[24550.  6315.]\n",
      "0.0\n",
      "[ 170. 1526.]\n",
      "1.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[24440.  7128.]\n",
      "0.0\n",
      "[280. 713.]\n",
      "1.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[8521.  377.]\n",
      "0.0\n",
      "[16199.  7464.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[6987. 2878.]\n",
      "0.0\n",
      "[17733.  4963.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[24707.  7831.]\n",
      "0.0\n",
      "[13. 10.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[23436.  7755.]\n",
      "0.0\n",
      "[1284.   86.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[5196.  387.]\n",
      "0.0\n",
      "[19524.  7454.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[23776.  7804.]\n",
      "0.0\n",
      "[944.  37.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[24613.  7535.]\n",
      "0.0\n",
      "[107. 306.]\n",
      "1.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[11918.  1245.]\n",
      "0.0\n",
      "[12802.  6596.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[20740.  7378.]\n",
      "0.0\n",
      "[3980.  463.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[23068.  7650.]\n",
      "0.0\n",
      "[1652.  191.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[23443.  7521.]\n",
      "0.0\n",
      "[1277.  320.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[23761.  7775.]\n",
      "0.0\n",
      "[959.  66.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[21492.  7623.]\n",
      "0.0\n",
      "[3228.  218.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[23918.  7576.]\n",
      "0.0\n",
      "[802. 265.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[12835.  1919.]\n",
      "0.0\n",
      "[11885.  5922.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[19423.  4175.]\n",
      "0.0\n",
      "[5297. 3666.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[19760.  3667.]\n",
      "0.0\n",
      "[4960. 4174.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[15128.  6662.]\n",
      "0.0\n",
      "[9592. 1179.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[21457.  7334.]\n",
      "0.0\n",
      "[3263.  507.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[17566.  3910.]\n",
      "0.0\n",
      "[7154. 3931.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[24075.  7558.]\n",
      "0.0\n",
      "[645. 283.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[23696.  6153.]\n",
      "0.0\n",
      "[1024. 1688.]\n",
      "1.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[19173.  4062.]\n",
      "0.0\n",
      "[5547. 3779.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[24700.  7741.]\n",
      "0.0\n",
      "[ 20. 100.]\n",
      "1.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[24700.  6462.]\n",
      "0.0\n",
      "[  20. 1379.]\n",
      "1.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[24440.  7128.]\n",
      "0.0\n",
      "[280. 713.]\n",
      "1.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[9200.  511.]\n",
      "0.0\n",
      "[15520.  7330.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[21983.  7454.]\n",
      "0.0\n",
      "[2737.  387.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[9890.  682.]\n",
      "0.0\n",
      "[14830.  7159.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[20562.  3932.]\n",
      "0.0\n",
      "[4158. 3909.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[24445.  7805.]\n",
      "0.0\n",
      "[275.  36.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[23060.  7649.]\n",
      "0.0\n",
      "[1660.  192.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[16436.  1149.]\n",
      "0.0\n",
      "[8284. 6692.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[15894.  6166.]\n",
      "0.0\n",
      "[8826. 1675.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[23775.  7488.]\n",
      "0.0\n",
      "[945. 353.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[23841.  7726.]\n",
      "0.0\n",
      "[879. 115.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[23699.  7480.]\n",
      "0.0\n",
      "[1021.  361.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[17445.  1923.]\n",
      "0.0\n",
      "[7275. 5918.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[23957.  7565.]\n",
      "0.0\n",
      "[763. 276.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[19285.  4101.]\n",
      "0.0\n",
      "[5435. 3740.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[5135.  381.]\n",
      "0.0\n",
      "[19585.  7460.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[24655.  7707.]\n",
      "0.0\n",
      "[ 65. 134.]\n",
      "1.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[23956.  6882.]\n",
      "0.0\n",
      "[764. 959.]\n",
      "1.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[21562.  7704.]\n",
      "0.0\n",
      "[3158.  137.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[24638.  6345.]\n",
      "0.0\n",
      "[  82. 1496.]\n",
      "1.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[4021.  724.]\n",
      "0.0\n",
      "[20699.  7117.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[14528.  7350.]\n",
      "0.0\n",
      "[10192.   491.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[9592. 1179.]\n",
      "0.0\n",
      "[15128.  6662.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[24700.  6462.]\n",
      "0.0\n",
      "[  20. 1379.]\n",
      "1.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[23897.  7096.]\n",
      "0.0\n",
      "[823. 745.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[24720.  7841.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[18739.  3306.]\n",
      "0.0\n",
      "[5981. 4535.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[24567.  7418.]\n",
      "0.0\n",
      "[153. 423.]\n",
      "1.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[21586.  5620.]\n",
      "0.0\n",
      "[3134. 2221.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[24226.  7219.]\n",
      "0.0\n",
      "[494. 622.]\n",
      "1.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[24440.  7128.]\n",
      "0.0\n",
      "[280. 713.]\n",
      "1.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[22968.  7591.]\n",
      "0.0\n",
      "[1752.  250.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[18816.  6454.]\n",
      "0.0\n",
      "[5904. 1387.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[22439.  5982.]\n",
      "0.0\n",
      "[2281. 1859.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[19719.  7774.]\n",
      "0.0\n",
      "[5001.   67.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[22903.  7117.]\n",
      "0.0\n",
      "[1817.  724.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[21550.  6912.]\n",
      "0.0\n",
      "[3170.  929.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[17271.  6985.]\n",
      "0.0\n",
      "[7449.  856.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[22622.  5873.]\n",
      "0.0\n",
      "[2098. 1968.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[15754.   820.]\n",
      "0.0\n",
      "[682. 329.]\n",
      "0.0\n",
      "[8284. 6692.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[16436.  1149.]\n",
      "0.0\n",
      "[7810. 6591.]\n",
      "0.0\n",
      "[474. 101.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[13786.  4367.]\n",
      "0.0\n",
      "[2108. 1799.]\n",
      "0.0\n",
      "[8826. 1675.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[14528.  7350.]\n",
      "0.0\n",
      "[9135.  314.]\n",
      "0.0\n",
      "[1057.  177.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[0. 0.]\n",
      "0.0\n",
      "[22100.  5442.]\n",
      "0.0\n",
      "[339. 540.]\n",
      "1.0\n",
      "[2281. 1859.]\n",
      "0.0\n"
     ]
    }
   ],
   "source": [
    "for j in range(0,trees_count):\n",
    "    for k in range(0,id_count):\n",
    "        print(Matrix[j][k])\n",
    "        print(Prefered_Class_Matrix[j][k])\n",
    "       \n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "def getFeatureVectorAdult(entries):\n",
    "\tx = []\n",
    "\tx.append(float(entries[0])) # age = continous\n",
    "\n",
    "\tworkclass = [0 for i in range(8)]\n",
    "\tif entries[1] == \"Private\":\n",
    "\t\tworkclass[0] = 1\n",
    "\telif entries[1] == \"Self-emp-not-inc\":\n",
    "\t\tworkclass[1] = 1\n",
    "\telif entries[1] == \"Self-emp-inc\":\n",
    "\t\tworkclass[2] = 1\n",
    "\telif entries[1] == \"Federal-gov\":\n",
    "\t\tworkclass[3] = 1\n",
    "\telif entries[1] == \"Local-gov\":\n",
    "\t\tworkclass[4] = 1\n",
    "\telif entries[1] == \"State-gov\":\n",
    "\t\tworkclass[5] = 1\n",
    "\telif entries[1] == \"Without-pay\":\n",
    "\t\tworkclass[6] = 1\n",
    "\telse: #Never-worked\n",
    "\t\tworkclass[7] = 1\n",
    "\tx.extend(workclass)\n",
    "\tx.append(float(entries[2]))\n",
    "\n",
    "\teducation = [0 for i in range(16)]\n",
    "\tif entries[3] == \"Bachelors\":\n",
    "\t\teducation[0] = 1\n",
    "\telif entries[3] == \"Some-college\":\n",
    "\t\teducation[1] = 1\n",
    "\telif entries[3] == \"11th\":\n",
    "\t\teducation[2] = 1\n",
    "\telif entries[3] == \"HS-grad\":\n",
    "\t\teducation[3] = 1\n",
    "\telif entries[3] == \"Prof-school\":\n",
    "\t\teducation[4] = 1\n",
    "\telif entries[3] == \"Assoc-acdm\":\n",
    "\t\teducation[5] = 1\n",
    "\telif entries[3] == \"Assoc-voc\":\n",
    "\t\teducation[6] = 1\n",
    "\telif entries[3] == \"9th\":\n",
    "\t\teducation[7] = 1\n",
    "\telif entries[3] == \"7th-8th\":\n",
    "\t\teducation[8] = 1\n",
    "\telif entries[3] == \"12th\":\n",
    "\t\teducation[9] = 1\n",
    "\telif entries[3] == \"Masters\":\n",
    "\t\teducation[10] = 1\n",
    "\telif entries[3] == \"1st-4th\":\n",
    "\t\teducation[11] = 1\n",
    "\telif entries[3] == \"10th\":\n",
    "\t\teducation[12] = 1\n",
    "\telif entries[3] == \"Doctorate\":\n",
    "\t\teducation[13] = 1\n",
    "\telif entries[3] == \"5th-6th\":\n",
    "\t\teducation[14] = 1\n",
    "\tif entries[3] == \"Preschool\":\n",
    "\t\teducation[15] = 1\n",
    "\n",
    "\tx.extend(education)\n",
    "\tx.append(float(entries[4]))\n",
    "\n",
    "\tmarital = [0 for i in range(7)]\n",
    "\tif entries[5] == \"Married-civ-spouse\":\n",
    "\t\tmarital[0] = 1\n",
    "\telif entries[5] == \"Divorced\":\n",
    "\t\tmarital[1] = 1\n",
    "\telif entries[5] == \"Never-married\":\n",
    "\t\tmarital[2] = 1\n",
    "\telif entries[5] == \"Separated\":\n",
    "\t\tmarital[3] = 1\n",
    "\telif entries[5] == \"Widowed\":\n",
    "\t\tmarital[4] = 1\n",
    "\telif entries[5] == \"Married-spouse-absent\":\n",
    "\t\tmarital[5] = 1\n",
    "\telse:\n",
    "\t\tmarital[6] = 1\n",
    "\tx.extend(marital)\n",
    "\n",
    "\toccupation = [0 for i in range(14)]\n",
    "\tif entries[6] == \"Tech-support\":\n",
    "\t\toccupation[0] = 1\n",
    "\telif entries[6] == \"Craft-repair\":\n",
    "\t\toccupation[1] = 1\n",
    "\telif entries[6] == \"Other-service\":\n",
    "\t\toccupation[2] = 1\n",
    "\telif entries[6] == \"Sales\":\n",
    "\t\toccupation[3] = 1\n",
    "\telif entries[6] == \"Exec-managerial\":\n",
    "\t\toccupation[4] = 1\n",
    "\telif entries[6] == \"Prof-specialty\":\n",
    "\t\toccupation[5] = 1\n",
    "\telif entries[6] == \"Handlers-cleaners\":\n",
    "\t\toccupation[6] = 1\n",
    "\telif entries[6] == \"Machine-op-inspct\":\n",
    "\t\toccupation[7] = 1\n",
    "\telif entries[6] == \"Adm-clerical\":\n",
    "\t\toccupation[8] = 1\n",
    "\telif entries[6] == \"Farming-fishing\":\n",
    "\t\toccupation[9] = 1\n",
    "\telif entries[6] == \"Transport-moving\":\n",
    "\t\toccupation[10] = 1\n",
    "\telif entries[6] == \"Priv-house-serv\":\n",
    "\t\toccupation[11] = 1\n",
    "\telif entries[6] == \"Protective-serv\":\n",
    "\t\toccupation[12] = 1\n",
    "\telse:\n",
    "\t\toccupation[13] = 1\n",
    "\tx.extend(occupation)\n",
    "\n",
    "\trelationship = [0 for i in range(6)]\n",
    "\tif entries[7] == \"Wife\":\n",
    "\t\trelationship[0] = 1\n",
    "\telif entries[7] == \"Own-child\":\n",
    "\t\trelationship[1] = 1\n",
    "\telif entries[7] == \"Husband\":\n",
    "\t\trelationship[2] = 1\n",
    "\telif entries[7] == \"Not-in-family\":\n",
    "\t\trelationship[3] = 1\n",
    "\telif entries[7] == \"Other-relative\":\n",
    "\t\trelationship[4] = 1\n",
    "\telse:\n",
    "\t\trelationship[5] = 1\n",
    "\tx.extend(relationship)\n",
    "\n",
    "\trace = [0 for i in range(5)]\n",
    "\tif entries[8] == \"White\":\n",
    "\t\trace[0] = 1\n",
    "\telif entries[8] == \"Asian-Pac-Islander\":\n",
    "\t\trace[1] = 1\n",
    "\telif entries[8] == \"Amer-Indian-Eskimo\":\n",
    "\t\trace[2] = 1\n",
    "\telif entries[8] == \"Other\":\n",
    "\t\trace[3] = 1\n",
    "\telse:\n",
    "\t\trace[4] = 1\n",
    "\tx.extend(race)\n",
    "\n",
    "\tif (entries[9] == \"Male\"):\n",
    "\t\tx.extend([1,0])\n",
    "\telse:\n",
    "\t\tx.extend([0,1])\n",
    "\n",
    "\tx.append(float(entries[10]))\n",
    "\tx.append(float(entries[11]))\n",
    "\tx.append(float(entries[12]))\n",
    "\n",
    "\tnative = [0 for i in range(42)]\n",
    "\tif entries[14] == \"United-States\":\n",
    "\t\tnative[1] = 1\n",
    "\telif entries[14] == \"Cambodia\":\n",
    "\t\tnative[2] = 1\n",
    "\telif entries[14] == \"England\":\n",
    "\t\tnative[3] = 1\n",
    "\telif entries[14] == \"Puerto-Rico\":\n",
    "\t\tnative[4] = 1\n",
    "\telif entries[14] == \"Canada\":\n",
    "\t\tnative[5] = 1\n",
    "\telif entries[14] == \"Germany\":\n",
    "\t\tnative[6] = 1\n",
    "\telif entries[14] == \"Outlying-US(Guam-USVI-etc)\":\n",
    "\t\tnative[7] = 1\n",
    "\telif entries[14] == \"India\":\n",
    "\t\tnative[8] = 1\n",
    "\telif entries[14] == \"Japan\":\n",
    "\t\tnative[9] = 1\n",
    "\telif entries[14] == \"Greece\":\n",
    "\t\tnative[10] = 1\n",
    "\telif entries[14] == \"South\":\n",
    "\t\tnative[11] = 1\n",
    "\telif entries[14] == \"China\":\n",
    "\t\tnative[12] = 1\n",
    "\telif entries[14] == \"Cuba\":\n",
    "\t\tnative[13] = 1\n",
    "\telif entries[14] == \"Iran\":\n",
    "\t\tnative[14] = 1\n",
    "\telif entries[14] == \"Honduras\":\n",
    "\t\tnative[15] = 1\n",
    "\telif entries[14] == \"Philippines\":\n",
    "\t\tnative[16] = 1\n",
    "\telif entries[14] == \"Italy\":\n",
    "\t\tnative[17] = 1\n",
    "\telif entries[14] == \"Poland\":\n",
    "\t\tnative[18] = 1\n",
    "\telif entries[14] == \"Jamaica\":\n",
    "\t\tnative[19] = 1\n",
    "\telif entries[14] == \"Vietnam\":\n",
    "\t\tnative[20] = 1\n",
    "\telif entries[14] == \"Mexico\":\n",
    "\t\tnative[21] = 1\n",
    "\telif entries[14] == \"Portugal\":\n",
    "\t\tnative[22] = 1\n",
    "\telif entries[14] == \"Ireland\":\n",
    "\t\tnative[23] = 1\n",
    "\telif entries[14] == \"France\":\n",
    "\t\tnative[24] = 1\n",
    "\telif entries[14] == \"Dominican-Republic\":\n",
    "\t\tnative[25] = 1\n",
    "\telif entries[14] == \"Laos\":\n",
    "\t\tnative[26] = 1\n",
    "\telif entries[14] == \"Ecuador\":\n",
    "\t\tnative[27] = 1\n",
    "\telif entries[14] == \"Taiwan\":\n",
    "\t\tnative[28] = 1\n",
    "\telif entries[14] == \"Haiti\":\n",
    "\t\tnative[29] = 1\n",
    "\telif entries[14] == \"Columbia\":\n",
    "\t\tnative[30] = 1\n",
    "\telif entries[14] == \"Hungary\":\n",
    "\t\tnative[31] = 1\n",
    "\telif entries[14] == \"Guatemala\":\n",
    "\t\tnative[32] = 1\n",
    "\telif entries[14] == \"Nicaragua\":\n",
    "\t\tnative[33] = 1\n",
    "\telif entries[14] == \"Scotland\":\n",
    "\t\tnative[34] = 1\n",
    "\telif entries[14] == \"Thailand\":\n",
    "\t\tnative[35] = 1\n",
    "\telif entries[14] == \"Yugoslavia\":\n",
    "\t\tnative[36] = 1\n",
    "\telif entries[14] == \"El-Salvador\":\n",
    "\t\tnative[37] = 1\n",
    "\telif entries[14] == \"Trinadad&Tobago\":\n",
    "\t\tnative[38] = 1\n",
    "\telif entries[14] == \"Peru\":\n",
    "\t\tnative[39] = 1\n",
    "\telif entries[14] == \"Hong\":\n",
    "\t\tnative[40] = 1\n",
    "\telse:\n",
    "\t\tnative[41] = 1\n",
    "\n",
    "\treturn x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5598\n",
      "5374\n"
     ]
    }
   ],
   "source": [
    "\n",
    "path = \"/home/falkhoury/Study/Lab/Project/frequentTreesInRandomForests/arch-forest/data/\"\n",
    "X = []\n",
    "Y = []       \n",
    "\n",
    "f = open(path + \"adult/adult.train\")\n",
    "counter = 0\n",
    "for row in f:\n",
    "    \n",
    "    if (len(row) > 1):\n",
    "        entries = row.replace(\"\\n\", \"\").replace(\" \", \"\").split(\",\")\n",
    "        x = getFeatureVectorAdult(entries)\n",
    "\n",
    "        if (entries[-1] == \"<=50K.\" or entries[-1] == \"<=50K\"):\n",
    "            counter += 1\n",
    "            if (counter %3 == 0):\n",
    "                y = 0\n",
    "                Y.append(y)\n",
    "                X.append(x) \n",
    "                \n",
    "                \n",
    "        else:\n",
    "                y = 1\n",
    "                Y.append(y)\n",
    "                X.append(x) \n",
    "\n",
    "        \n",
    "X = np.array(X).astype(dtype=np.int32)\n",
    "Y = np.array(Y)\n",
    "f.close()\n",
    "counter_zero = 0\n",
    "counter_one = 0\n",
    "for i in range(0,len(X)):\n",
    "    if (Y[i] == 0):\n",
    "        counter_zero += 1\n",
    "    if (Y[i] == 1):\n",
    "        counter_one += 1\n",
    "print(counter_zero)\n",
    "print(counter_one)    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "449\n",
      "1151\n"
     ]
    }
   ],
   "source": [
    " "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}