GBM_Tuning.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "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>host_response_rate</th>\n",
       "      <th>host_listings_count</th>\n",
       "      <th>host_has_profile_pic</th>\n",
       "      <th>host_identity_verified</th>\n",
       "      <th>accommodates</th>\n",
       "      <th>bathrooms</th>\n",
       "      <th>bedrooms</th>\n",
       "      <th>beds</th>\n",
       "      <th>price</th>\n",
       "      <th>security_deposit</th>\n",
       "      <th>...</th>\n",
       "      <th>Amenties Include parking</th>\n",
       "      <th>Amenties Include Pets</th>\n",
       "      <th>Amenties Include Dishes and silverware</th>\n",
       "      <th>Amenties Include Coffee maker</th>\n",
       "      <th>Amenties Include Microwave</th>\n",
       "      <th>Amenties Include Refrigerator</th>\n",
       "      <th>offers_weekly_price</th>\n",
       "      <th>offers_monthly_price</th>\n",
       "      <th>charges_security_deposit</th>\n",
       "      <th>charges_cleaning_fee</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>5435.0</th>\n",
       "      <td>100.000000</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>145.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20311.0</th>\n",
       "      <td>100.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>70.0</td>\n",
       "      <td>150.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27024.0</th>\n",
       "      <td>100.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>150.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>62264.0</th>\n",
       "      <td>100.000000</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>65.0</td>\n",
       "      <td>300.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>70381.0</th>\n",
       "      <td>93.247853</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>450.0</td>\n",
       "      <td>227.138887</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 69 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         host_response_rate  host_listings_count  host_has_profile_pic  \\\n",
       "id                                                                       \n",
       "5435.0           100.000000                  2.0                   1.0   \n",
       "20311.0          100.000000                  1.0                   1.0   \n",
       "27024.0          100.000000                  1.0                   1.0   \n",
       "62264.0          100.000000                  2.0                   1.0   \n",
       "70381.0           93.247853                  1.0                   1.0   \n",
       "\n",
       "         host_identity_verified  accommodates  bathrooms  bedrooms  beds  \\\n",
       "id                                                                         \n",
       "5435.0                      0.0           2.0        1.0       1.0   1.0   \n",
       "20311.0                     1.0           2.0        1.0       1.0   1.0   \n",
       "27024.0                     1.0           4.0        1.0       1.0   3.0   \n",
       "62264.0                     1.0           2.0        1.0       1.0   1.0   \n",
       "70381.0                     0.0           5.0        2.0       3.0   3.0   \n",
       "\n",
       "         price  security_deposit  ...  Amenties Include parking  \\\n",
       "id                                ...                             \n",
       "5435.0   145.0          0.000000  ...                         0   \n",
       "20311.0   70.0        150.000000  ...                         0   \n",
       "27024.0  150.0          0.000000  ...                         0   \n",
       "62264.0   65.0        300.000000  ...                         1   \n",
       "70381.0  450.0        227.138887  ...                         0   \n",
       "\n",
       "         Amenties Include Pets  Amenties Include Dishes and silverware  \\\n",
       "id                                                                       \n",
       "5435.0                       1                                       1   \n",
       "20311.0                      0                                       0   \n",
       "27024.0                      0                                       0   \n",
       "62264.0                      0                                       0   \n",
       "70381.0                      1                                       0   \n",
       "\n",
       "         Amenties Include Coffee maker  Amenties Include Microwave  \\\n",
       "id                                                                   \n",
       "5435.0                               1                           1   \n",
       "20311.0                              0                           0   \n",
       "27024.0                              0                           0   \n",
       "62264.0                              0                           0   \n",
       "70381.0                              0                           0   \n",
       "\n",
       "         Amenties Include Refrigerator  offers_weekly_price  \\\n",
       "id                                                            \n",
       "5435.0                               1                    0   \n",
       "20311.0                              0                    1   \n",
       "27024.0                              0                    1   \n",
       "62264.0                              0                    1   \n",
       "70381.0                              0                    0   \n",
       "\n",
       "         offers_monthly_price  charges_security_deposit  charges_cleaning_fee  \n",
       "id                                                                             \n",
       "5435.0                      1                         1                     1  \n",
       "20311.0                     1                         1                     1  \n",
       "27024.0                     1                         1                     1  \n",
       "62264.0                     1                         1                     1  \n",
       "70381.0                     0                         0                     1  \n",
       "\n",
       "[5 rows x 69 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import os\n",
    "import numpy as np\n",
    "controllable = pd.read_csv(os.getcwd()+'/Airbnb_Optimization/datasets/controllable_sampled.csv',index_col='id')\n",
    "controllable.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['host_response_rate', 'host_listings_count', 'host_has_profile_pic',\n",
       "       'host_identity_verified', 'accommodates', 'bathrooms', 'bedrooms',\n",
       "       'beds', 'price', 'security_deposit', 'cleaning_fee', 'guests_included',\n",
       "       'extra_people', 'minimum_nights', 'maximum_nights', 'instant_bookable',\n",
       "       'require_guest_profile_picture', 'require_guest_phone_verification',\n",
       "       'calculated_host_listings_count_entire_homes',\n",
       "       'calculated_host_listings_count_private_rooms',\n",
       "       'calculated_host_listings_count_shared_rooms', 'Label',\n",
       "       'host_response_time_a few days or more',\n",
       "       'host_response_time_within a day',\n",
       "       'host_response_time_within a few hours',\n",
       "       'host_response_time_within an hour', 'host_response_time_nan',\n",
       "       'property_type_Apartment', 'property_type_House', 'property_type_Loft',\n",
       "       'property_type_Other', 'property_type_Townhouse',\n",
       "       'room_type_Entire home/apt', 'room_type_Hotel room',\n",
       "       'room_type_Private room', 'room_type_Shared room', 'bed_type_Airbed',\n",
       "       'bed_type_Couch', 'bed_type_Futon', 'bed_type_Pull-out Sofa',\n",
       "       'bed_type_Real Bed', 'cancellation_policy_flexible',\n",
       "       'cancellation_policy_moderate', 'cancellation_policy_strict',\n",
       "       'cancellation_policy_strict_14_with_grace_period',\n",
       "       'cancellation_policy_super_strict_30',\n",
       "       'cancellation_policy_super_strict_60', 'host_tenure', 'is_new_host',\n",
       "       'number_of_host_verifications', 'Amenties Include TV',\n",
       "       'Amenties Include Air conditioning',\n",
       "       'Amenties Include Wheelchair accessible', 'Amenties Include Elevator',\n",
       "       'Amenties Include Washer', 'Amenties Include Dryer',\n",
       "       'Amenties Include Internet', 'Amenties Include Wifi',\n",
       "       'Amenties Include Gym', 'Amenties Include parking',\n",
       "       'Amenties Include Pets', 'Amenties Include Dishes and silverware',\n",
       "       'Amenties Include Coffee maker', 'Amenties Include Microwave',\n",
       "       'Amenties Include Refrigerator', 'offers_weekly_price',\n",
       "       'offers_monthly_price', 'charges_security_deposit',\n",
       "       'charges_cleaning_fee'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "controllable.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GradientBoostingClassifier(max_depth=8, max_features='sqrt',\n",
       "                           min_samples_leaf=50, min_samples_split=500,\n",
       "                           random_state=10, subsample=0.8)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import GradientBoostingClassifier\n",
    "from sklearn.metrics import confusion_matrix,roc_auc_score\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X = controllable.drop(columns=['Label'])\n",
    "Y = controllable['Label']\n",
    "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, train_size=0.80, random_state=42)\n",
    "\n",
    "\n",
    "n_est_lim = 1000\n",
    "gbc = GradientBoostingClassifier(n_estimators=100, max_depth=8,max_features='sqrt',\n",
    "                                                  min_samples_leaf=50,\n",
    "                                                  min_samples_split=500,\n",
    "                                                  random_state=10,\n",
    "                                                  subsample=0.8)\n",
    "gbc.fit(X_train, Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.640542577241899"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "gbc=GradientBoostingClassifier(learning_rate=0.025, max_depth=10,\n",
    "                            max_features='sqrt',\n",
    "                            min_samples_leaf=0.1489795918367347,\n",
    "                            min_samples_split=0.22244897959183674,\n",
    "                            n_estimators=2500)\n",
    "gbc.fit(X_train, Y_train)\n",
    "y_pred = gbc.predict(X_test)\n",
    "accuracy_score(Y_test,y_pred)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "({'subsample': 1.0,\n",
       "  'n_estimators': 2500,\n",
       "  'min_samples_split': 0.22244897959183674,\n",
       "  'min_samples_leaf': 0.1489795918367347,\n",
       "  'max_features': 'sqrt',\n",
       "  'max_depth': 10,\n",
       "  'loss': 'deviance',\n",
       "  'learning_rate': 0.025,\n",
       "  'criterion': 'friedman_mse'},\n",
       " GradientBoostingClassifier(learning_rate=0.025, max_depth=10,\n",
       "                            max_features='sqrt',\n",
       "                            min_samples_leaf=0.1489795918367347,\n",
       "                            min_samples_split=0.22244897959183674,\n",
       "                            n_estimators=2500))"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "parameters = {\n",
    "    \"loss\":[\"deviance\"],\n",
    "    \"learning_rate\": [0.01, 0.025, 0.05, 0.075, 0.1, 0.15, 0.2],\n",
    "    \"min_samples_split\": np.linspace(0.1, 0.5, 50),\n",
    "    \"min_samples_leaf\": np.linspace(0.1, 0.5, 50),\n",
    "    \"max_depth\":[3,5,7,8,9,10],\n",
    "    \"max_features\":[\"log2\",\"sqrt\"],\n",
    "    \"criterion\": [\"friedman_mse\",  \"mae\"],\n",
    "    \"subsample\":[0.5, 0.618, 0.8, 0.85, 0.9, 0.95, 1.0],\n",
    "    \"n_estimators\": [100,200,500,1000,1500,2000,2500]\n",
    "    }\n",
    "\n",
    "\n",
    "clf = RandomizedSearchCV(GradientBoostingClassifier(), parameters, random_state=0)\n",
    "search = clf.fit(X_train, Y_train)\n",
    "search.best_params_,search.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import GradientBoostingClassifier\n",
    "from sklearn.metrics import confusion_matrix,roc_auc_score\n",
    "from sklearn.model_selection import train_test_split\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = gbc.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import accuracy_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6419241396634011"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred = gbc.predict(X_test)\n",
    "accuracy_score(Y_test,y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f3fbe07fbe0>"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from pylab import rcParams\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "rcParams['figure.figsize'] = 12, 5\n",
    "\n",
    "\n",
    "def LogLossP(Pt, Yt):\n",
    "    return -1*((Yt==1)*np.log(Pt)+(Yt==0)*np.log(1-Pt)).mean()\n",
    "\n",
    "p_train = gbc.staged_predict_proba(X_train)\n",
    "p_test = gbc.staged_predict_proba(X_test)\n",
    "\n",
    "ll_train = []\n",
    "ll_test = []\n",
    "\n",
    "for p in p_train:\n",
    "    ll_train.append(LogLossP(p[:, 1], Y_train))\n",
    "    \n",
    "for p in p_test:\n",
    "    ll_test.append(LogLossP(p[:, 1], Y_test))\n",
    "    \n",
    "\n",
    "fig = plt.figure()\n",
    "ax = fig.add_subplot(111)\n",
    "plt.plot(np.arange(1, n_est_lim + 1), ll_train, 'b-', label='Train')\n",
    "plt.plot(np.arange(1, n_est_lim + 1), ll_test, 'r-', label='Test')\n",
    "plt.title('Training and Testing Error as a Function of N-Estimators')\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/dawooood/anaconda3/lib/python3.8/site-packages/sklearn/model_selection/_search.py:847: FutureWarning: The parameter 'iid' is deprecated in 0.22 and will be removed in 0.24.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5,\n",
       "             estimator=GradientBoostingClassifier(max_depth=8,\n",
       "                                                  max_features='sqrt',\n",
       "                                                  min_samples_leaf=50,\n",
       "                                                  min_samples_split=500,\n",
       "                                                  random_state=10,\n",
       "                                                  subsample=0.8),\n",
       "             iid=False, n_jobs=4,\n",
       "             param_grid={'n_estimators': [100, 200, 300, 400, 500, 600, 700,\n",
       "                                          800, 900, 1000, 1100, 1200, 1300,\n",
       "                                          1400, 1500]},\n",
       "             scoring='roc_auc')"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "param_test1 = {'n_estimators':[100,\n",
    " 200,\n",
    " 300,\n",
    " 400,\n",
    " 500,\n",
    " 600,\n",
    " 700,\n",
    " 800,\n",
    " 900,\n",
    " 1000,\n",
    " 1100,\n",
    " 1200,\n",
    " 1300,\n",
    " 1400,\n",
    " 1500]}\n",
    "gsearch1 = GridSearchCV(estimator = GradientBoostingClassifier(learning_rate=0.1, min_samples_split=500,min_samples_leaf=50,max_depth=8,max_features='sqrt',subsample=0.8,random_state=10), \n",
    "param_grid = param_test1, scoring='roc_auc',n_jobs=4,iid=False, cv=5)\n",
    "gsearch1.fit(X_train,Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'cv': 5,\n",
       " 'error_score': nan,\n",
       " 'estimator__ccp_alpha': 0.0,\n",
       " 'estimator__criterion': 'friedman_mse',\n",
       " 'estimator__init': None,\n",
       " 'estimator__learning_rate': 0.1,\n",
       " 'estimator__loss': 'deviance',\n",
       " 'estimator__max_depth': 8,\n",
       " 'estimator__max_features': 'sqrt',\n",
       " 'estimator__max_leaf_nodes': None,\n",
       " 'estimator__min_impurity_decrease': 0.0,\n",
       " 'estimator__min_impurity_split': None,\n",
       " 'estimator__min_samples_leaf': 50,\n",
       " 'estimator__min_samples_split': 500,\n",
       " 'estimator__min_weight_fraction_leaf': 0.0,\n",
       " 'estimator__n_estimators': 100,\n",
       " 'estimator__n_iter_no_change': None,\n",
       " 'estimator__presort': 'deprecated',\n",
       " 'estimator__random_state': 10,\n",
       " 'estimator__subsample': 0.8,\n",
       " 'estimator__tol': 0.0001,\n",
       " 'estimator__validation_fraction': 0.1,\n",
       " 'estimator__verbose': 0,\n",
       " 'estimator__warm_start': False,\n",
       " 'estimator': GradientBoostingClassifier(max_depth=8, max_features='sqrt',\n",
       "                            min_samples_leaf=50, min_samples_split=500,\n",
       "                            random_state=10, subsample=0.8),\n",
       " 'iid': False,\n",
       " 'n_jobs': 4,\n",
       " 'param_grid': {'n_estimators': [100,\n",
       "   200,\n",
       "   300,\n",
       "   400,\n",
       "   500,\n",
       "   600,\n",
       "   700,\n",
       "   800,\n",
       "   900,\n",
       "   1000,\n",
       "   1100,\n",
       "   1200,\n",
       "   1300,\n",
       "   1400,\n",
       "   1500]},\n",
       " 'pre_dispatch': '2*n_jobs',\n",
       " 'refit': True,\n",
       " 'return_train_score': False,\n",
       " 'scoring': 'roc_auc',\n",
       " 'verbose': 0}"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gsearch1.get_params()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "a=[x for x in range(100,1600,100)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[100,\n",
       " 200,\n",
       " 300,\n",
       " 400,\n",
       " 500,\n",
       " 600,\n",
       " 700,\n",
       " 800,\n",
       " 900,\n",
       " 1000,\n",
       " 1100,\n",
       " 1200,\n",
       " 1300,\n",
       " 1400,\n",
       " 1500]"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "({'n_estimators': 100}, 0.7002949759074061)"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gsearch1.best_params_,gsearch1.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/dawooood/anaconda3/lib/python3.8/site-packages/sklearn/model_selection/_search.py:847: FutureWarning: The parameter 'iid' is deprecated in 0.22 and will be removed in 0.24.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "({'max_depth': 5, 'min_samples_split': 1000}, 0.7013061703841229)"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "param_test2 = {'max_depth':range(5,16,2), 'min_samples_split':range(200,1001,200)}\n",
    "gsearch1 = GridSearchCV(estimator = GradientBoostingClassifier(n_estimators=100, learning_rate=0.1, max_features='sqrt',subsample=0.8,random_state=10), \n",
    "param_grid = param_test2, scoring='roc_auc',n_jobs=4,iid=False, cv=5)\n",
    "gsearch1.fit(X_train,Y_train)\n",
    "gsearch1.best_params_,gsearch1.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6372770660638031"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gbc = GradientBoostingClassifier(n_estimators=100, max_depth=5,max_features='sqrt',\n",
    "                                                  min_samples_leaf=50,\n",
    "                                                  min_samples_split=1000,\n",
    "                                                  random_state=10,\n",
    "                                                  subsample=0.8)\n",
    "gbc.fit(X_train, Y_train)\n",
    "y_pred = gbc.predict(X_test)\n",
    "accuracy_score(Y_test,y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/dawooood/anaconda3/lib/python3.8/site-packages/sklearn/model_selection/_search.py:847: FutureWarning: The parameter 'iid' is deprecated in 0.22 and will be removed in 0.24.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "({'max_depth': 5, 'min_samples_split': 1000}, 0.7013061703841229)"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "param_test3 = {'min_samples_split':range(1000,2100,200), 'min_samples_leaf':range(30,71,10)}\n",
    "gsearch3 = GridSearchCV(estimator = GradientBoostingClassifier(learning_rate=0.1, n_estimators=100,max_depth=5,max_features='sqrt', subsample=0.8, random_state=10), \n",
    "param_grid = param_test3, scoring='roc_auc',n_jobs=4,iid=False, cv=5)\n",
    "gsearch3.fit(X_train,Y_train)\n",
    "gsearch3.best_params_,gsearch3.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "({'min_samples_leaf': 70, 'min_samples_split': 1400}, 0.7031256854162494)"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gsearch3.best_params_,gsearch3.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GradientBoostingClassifier(max_depth=5, max_features='sqrt',\n",
       "                           min_samples_leaf=70, min_samples_split=1400,\n",
       "                           random_state=10, subsample=0.8)"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gsearch3.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6347023360964582"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gbc = GradientBoostingClassifier(n_estimators=100, max_depth=5,max_features='sqrt',\n",
    "                                                  min_samples_leaf=70,\n",
    "                                                  min_samples_split=1400,\n",
    "                                                  random_state=10,\n",
    "                                                  subsample=0.8)\n",
    "gbc.fit(X_train, Y_train)\n",
    "y_pred = gbc.predict(X_test)\n",
    "accuracy_score(Y_test,y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-0.09799257, -1.33163724,  0.04562769, ..., -0.5252299 ,\n",
       "       -1.21696234, -0.33378999])"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gsearch3.decision_function(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/dawooood/anaconda3/lib/python3.8/site-packages/sklearn/model_selection/_search.py:847: FutureWarning: The parameter 'iid' is deprecated in 0.22 and will be removed in 0.24.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "({'max_features': 30}, 0.699502205902478)"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "param_test4 = {'max_features':range(10,230,5)}\n",
    "\n",
    "gsearch4 = GridSearchCV(estimator = GradientBoostingClassifier(learning_rate=0.1, n_estimators=100,max_depth=5,max_features='sqrt', subsample=0.8, random_state=10, min_samples_leaf = 70, min_samples_split = 1400), \n",
    "param_grid = param_test4, scoring='roc_auc',n_jobs=4,iid=False, cv=5)\n",
    "gsearch4.fit(X_train,Y_train)\n",
    "gsearch4.best_params_,gsearch4.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6376538558151218"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gbc = GradientBoostingClassifier(n_estimators=200, max_depth=5,max_features= 30,\n",
    "                                                  min_samples_leaf=70,\n",
    "                                                  min_samples_split=1400,\n",
    "                                                  random_state=10,\n",
    "                                                  subsample=0.8,learning_rate=0.05)\n",
    "gbc.fit(X_train, Y_train)\n",
    "y_pred = gbc.predict(X_test)\n",
    "accuracy_score(Y_test,y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/dawooood/anaconda3/lib/python3.8/site-packages/sklearn/model_selection/_search.py:847: FutureWarning: The parameter 'iid' is deprecated in 0.22 and will be removed in 0.24.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "({'subsample': 0.8}, 0.699502205902478)"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "param_test5 = {'subsample':[0.6,0.7,0.75,0.8,0.85,0.9]}\n",
    "\n",
    "gsearch5 = GridSearchCV(estimator = GradientBoostingClassifier(learning_rate=0.1, n_estimators=100,max_depth=5,max_features='sqrt', random_state=10, min_samples_leaf = 70, min_samples_split = 1400), \n",
    "param_grid = param_test5, scoring='roc_auc',n_jobs=4,iid=False, cv=5)\n",
    "gsearch5.fit(X_train,Y_train)\n",
    "gsearch5.best_params_,gsearch4.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "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>idpop_local_sum</th>\n",
       "      <th>idpop_citywide_sum</th>\n",
       "      <th>idpop_local_food_sum</th>\n",
       "      <th>idpop_citywide_food_sum</th>\n",
       "      <th>idpop_local_nightlife_sum</th>\n",
       "      <th>idpop_citywide_nightlife_sum</th>\n",
       "      <th>idpop_local_travel_sum</th>\n",
       "      <th>idpop_citywide_travel_sum</th>\n",
       "      <th>venue_count_500m</th>\n",
       "      <th>sub_cnt_20m</th>\n",
       "      <th>...</th>\n",
       "      <th>neighbourhood_West Village</th>\n",
       "      <th>neighbourhood_Westchester Village</th>\n",
       "      <th>neighbourhood_Westerleigh</th>\n",
       "      <th>neighbourhood_Whitestone</th>\n",
       "      <th>neighbourhood_Williamsbridge</th>\n",
       "      <th>neighbourhood_Williamsburg</th>\n",
       "      <th>neighbourhood_Windsor Terrace</th>\n",
       "      <th>neighbourhood_Woodhaven</th>\n",
       "      <th>neighbourhood_Woodlawn</th>\n",
       "      <th>neighbourhood_Woodside</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>33699563.0</th>\n",
       "      <td>7.992767</td>\n",
       "      <td>0.803436</td>\n",
       "      <td>7.157023</td>\n",
       "      <td>0.805422</td>\n",
       "      <td>13.380945</td>\n",
       "      <td>0.630482</td>\n",
       "      <td>7.921177</td>\n",
       "      <td>0.630585</td>\n",
       "      <td>44</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16585605.0</th>\n",
       "      <td>0.000769</td>\n",
       "      <td>0.739474</td>\n",
       "      <td>0.000769</td>\n",
       "      <td>0.608763</td>\n",
       "      <td>0.000769</td>\n",
       "      <td>0.752475</td>\n",
       "      <td>0.000769</td>\n",
       "      <td>0.892192</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31767505.0</th>\n",
       "      <td>24.897783</td>\n",
       "      <td>2.047967</td>\n",
       "      <td>27.071885</td>\n",
       "      <td>2.982674</td>\n",
       "      <td>11.118003</td>\n",
       "      <td>1.365497</td>\n",
       "      <td>21.060435</td>\n",
       "      <td>2.646382</td>\n",
       "      <td>78</td>\n",
       "      <td>12</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1655390.0</th>\n",
       "      <td>2.110251</td>\n",
       "      <td>0.568449</td>\n",
       "      <td>2.446163</td>\n",
       "      <td>0.568007</td>\n",
       "      <td>1.857215</td>\n",
       "      <td>0.577803</td>\n",
       "      <td>2.622797</td>\n",
       "      <td>0.551869</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10615883.0</th>\n",
       "      <td>10.972633</td>\n",
       "      <td>0.684232</td>\n",
       "      <td>4.148196</td>\n",
       "      <td>0.725314</td>\n",
       "      <td>6.891011</td>\n",
       "      <td>0.619826</td>\n",
       "      <td>3.914252</td>\n",
       "      <td>0.598249</td>\n",
       "      <td>6</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 229 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            idpop_local_sum  idpop_citywide_sum  idpop_local_food_sum  \\\n",
       "id                                                                      \n",
       "33699563.0         7.992767            0.803436              7.157023   \n",
       "16585605.0         0.000769            0.739474              0.000769   \n",
       "31767505.0        24.897783            2.047967             27.071885   \n",
       "1655390.0          2.110251            0.568449              2.446163   \n",
       "10615883.0        10.972633            0.684232              4.148196   \n",
       "\n",
       "            idpop_citywide_food_sum  idpop_local_nightlife_sum  \\\n",
       "id                                                               \n",
       "33699563.0                 0.805422                  13.380945   \n",
       "16585605.0                 0.608763                   0.000769   \n",
       "31767505.0                 2.982674                  11.118003   \n",
       "1655390.0                  0.568007                   1.857215   \n",
       "10615883.0                 0.725314                   6.891011   \n",
       "\n",
       "            idpop_citywide_nightlife_sum  idpop_local_travel_sum  \\\n",
       "id                                                                 \n",
       "33699563.0                      0.630482                7.921177   \n",
       "16585605.0                      0.752475                0.000769   \n",
       "31767505.0                      1.365497               21.060435   \n",
       "1655390.0                       0.577803                2.622797   \n",
       "10615883.0                      0.619826                3.914252   \n",
       "\n",
       "            idpop_citywide_travel_sum  venue_count_500m  sub_cnt_20m  ...  \\\n",
       "id                                                                    ...   \n",
       "33699563.0                   0.630585                44            4  ...   \n",
       "16585605.0                   0.892192                 6            1  ...   \n",
       "31767505.0                   2.646382                78           12  ...   \n",
       "1655390.0                    0.551869                 1            1  ...   \n",
       "10615883.0                   0.598249                 6            2  ...   \n",
       "\n",
       "            neighbourhood_West Village  neighbourhood_Westchester Village  \\\n",
       "id                                                                          \n",
       "33699563.0                           0                                  0   \n",
       "16585605.0                           0                                  0   \n",
       "31767505.0                           0                                  0   \n",
       "1655390.0                            0                                  0   \n",
       "10615883.0                           0                                  0   \n",
       "\n",
       "            neighbourhood_Westerleigh  neighbourhood_Whitestone  \\\n",
       "id                                                                \n",
       "33699563.0                          0                         0   \n",
       "16585605.0                          0                         0   \n",
       "31767505.0                          0                         0   \n",
       "1655390.0                           0                         0   \n",
       "10615883.0                          0                         0   \n",
       "\n",
       "            neighbourhood_Williamsbridge  neighbourhood_Williamsburg  \\\n",
       "id                                                                     \n",
       "33699563.0                             0                           0   \n",
       "16585605.0                             0                           0   \n",
       "31767505.0                             0                           0   \n",
       "1655390.0                              0                           0   \n",
       "10615883.0                             0                           0   \n",
       "\n",
       "            neighbourhood_Windsor Terrace  neighbourhood_Woodhaven  \\\n",
       "id                                                                   \n",
       "33699563.0                              0                        0   \n",
       "16585605.0                              0                        0   \n",
       "31767505.0                              0                        0   \n",
       "1655390.0                               0                        0   \n",
       "10615883.0                              0                        0   \n",
       "\n",
       "            neighbourhood_Woodlawn  neighbourhood_Woodside  \n",
       "id                                                          \n",
       "33699563.0                       0                       0  \n",
       "16585605.0                       0                       0  \n",
       "31767505.0                       0                       0  \n",
       "1655390.0                        0                       0  \n",
       "10615883.0                       0                       0  \n",
       "\n",
       "[5 rows x 229 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import os\n",
    "uncontrollable = pd.read_csv(os.getcwd()+'/Airbnb_Optimization/datasets/uncontrollable_sample.csv',index_col='id')\n",
    "uncontrollable.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "scaler = StandardScaler()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = uncontrollable.drop(columns=['Label'])\n",
    "\n",
    "#X = scaler.fit_transform(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import confusion_matrix,roc_auc_score\n",
    "from sklearn.model_selection import train_test_split\n",
    "X = uncontrollable.drop(columns=['Label'])\n",
    "Y = list(uncontrollable['Label'].T)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " ...]"
      ]
     },
     "execution_count": 118,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GradientBoostingClassifier(max_depth=8, max_features='sqrt',\n",
       "                           min_samples_leaf=50, min_samples_split=500,\n",
       "                           random_state=10, subsample=0.8)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix,roc_auc_score\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X = uncontrollable.drop(columns=['Label'])\n",
    "Y = list(uncontrollable['Label'].T)\n",
    "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.80, random_state=42)\n",
    "\n",
    "\n",
    "\n",
    "gbc = GradientBoostingClassifier(n_estimators=100, max_depth=8,max_features='sqrt',\n",
    "                                                  min_samples_leaf=50,\n",
    "                                                  min_samples_split=500,\n",
    "                                                  random_state=10,\n",
    "                                                  subsample=0.8)\n",
    "gbc.fit(X_train, Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5535970102771722"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred = gbc.predict(X_test)\n",
    "accuracy_score(Y_test,y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/dawooood/anaconda3/lib/python3.8/site-packages/sklearn/model_selection/_search.py:847: FutureWarning: The parameter 'iid' is deprecated in 0.22 and will be removed in 0.24.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "({'n_estimators': 100}, 0.5631091060036008)"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "param_test1 = {'n_estimators':[100,\n",
    " 200,\n",
    " 300,\n",
    " 400,\n",
    " 500,\n",
    " 600,\n",
    " 700,\n",
    " 800,\n",
    " 900,\n",
    " 1000,\n",
    " 1100,\n",
    " 1200,\n",
    " 1300,\n",
    " 1400,\n",
    " 1500]}\n",
    "gsearch1 = GridSearchCV(estimator = GradientBoostingClassifier(learning_rate=0.1, min_samples_split=500,min_samples_leaf=50,max_depth=8,max_features='sqrt',subsample=0.8,random_state=10), \n",
    "param_grid = param_test1, scoring='roc_auc',n_jobs=4,iid=False, cv=5)\n",
    "gsearch1.fit(X_train,Y_train)\n",
    "gsearch1.best_params_ , gsearch1.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/dawooood/anaconda3/lib/python3.8/site-packages/sklearn/model_selection/_search.py:847: FutureWarning: The parameter 'iid' is deprecated in 0.22 and will be removed in 0.24.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "({'max_depth': 7, 'min_samples_split': 600}, 0.5629771006894682)"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "param_test2 = {'max_depth':range(5,16,2), 'min_samples_split':range(200,1001,200)}\n",
    "gsearch2 = GridSearchCV(estimator = GradientBoostingClassifier(n_estimators=100, learning_rate=0.1, max_features='sqrt',subsample=0.8,random_state=10), \n",
    "param_grid = param_test2, scoring='roc_auc',n_jobs=4,iid=False, cv=5)\n",
    "gsearch2.fit(X_train,Y_train)\n",
    "gsearch2.best_params_,gsearch2.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "69"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(controllable.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/dawooood/anaconda3/lib/python3.8/site-packages/sklearn/model_selection/_search.py:847: FutureWarning: The parameter 'iid' is deprecated in 0.22 and will be removed in 0.24.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "({'min_samples_leaf': 30, 'min_samples_split': 1800}, 0.5667199514301762)"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "param_test3 = {'min_samples_split':range(600,1900,200), 'min_samples_leaf':range(30,71,10)}\n",
    "gsearch3 = GridSearchCV(estimator = GradientBoostingClassifier(learning_rate=0.1, n_estimators=100,max_depth=7,max_features='sqrt', subsample=0.8, random_state=10), \n",
    "param_grid = param_test3, scoring='roc_auc',n_jobs=4,iid=False, cv=5)\n",
    "gsearch3.fit(X_train,Y_train)\n",
    "gsearch3.best_params_,gsearch3.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/dawooood/anaconda3/lib/python3.8/site-packages/sklearn/model_selection/_search.py:847: FutureWarning: The parameter 'iid' is deprecated in 0.22 and will be removed in 0.24.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "({'max_features': 15}, 0.5667199514301762)"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "param_test4 = {'max_features':range(10,230,5)}\n",
    "\n",
    "gsearch4 = GridSearchCV(estimator = GradientBoostingClassifier(learning_rate=0.1, n_estimators=100,max_depth=7, subsample=0.8, random_state=10, min_samples_leaf = 30, min_samples_split = 1800), \n",
    "param_grid = param_test4, scoring='roc_auc',n_jobs=4,iid=False, cv=5)\n",
    "gsearch4.fit(X_train,Y_train)\n",
    "gsearch4.best_params_,gsearch4.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5538461538461539"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gbc = GradientBoostingClassifier(n_estimators=100, max_depth=8,max_features=15,\n",
    "                                                  min_samples_leaf=30,\n",
    "                                                  min_samples_split=1800,\n",
    "                                                  random_state=10,\n",
    "                                                 subsample=0.8)\n",
    "gbc.fit(X_train,Y_train)\n",
    "y_pred = gbc.predict(X_test)\n",
    "accuracy_score(Y_test,y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5519775770787917"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gbc = GradientBoostingClassifier(n_estimators=400, max_depth=8,max_features=15,\n",
    "                                                  min_samples_leaf=30,\n",
    "                                                  min_samples_split=1800,\n",
    "                                                  random_state=10,\n",
    "                                                 subsample=0.8,learning_rate=0.01)\n",
    "gbc.fit(X_train,Y_train)\n",
    "y_pred = gbc.predict(X_test)\n",
    "accuracy_score(Y_test,y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [],
   "source": [
    "neigh = [col for col in uncontrollable.columns if 'neighbourhood' in col]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [],
   "source": [
    "uncontrollable= uncontrollable.drop(columns=neigh)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GradientBoostingClassifier(max_depth=8, max_features='sqrt',\n",
       "                           min_samples_leaf=50, min_samples_split=500,\n",
       "                           random_state=10, subsample=0.8)"
      ]
     },
     "execution_count": 127,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = uncontrollable.drop(columns=['Label'])\n",
    "Y = uncontrollable['Label']\n",
    "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.80, random_state=42)\n",
    "\n",
    "\n",
    "\n",
    "gbc = GradientBoostingClassifier(n_estimators=100, max_depth=8,max_features='sqrt',\n",
    "                                                  min_samples_leaf=50,\n",
    "                                                  min_samples_split=500,\n",
    "                                                  random_state=10,\n",
    "                                                  subsample=0.8)\n",
    "gbc.fit(X_train, Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5519775770787917"
      ]
     },
     "execution_count": 128,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred = gbc.predict(X_test)\n",
    "accuracy_score(Y_test,y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.0407926 , 0.06534036, 0.05556221, 0.06689587, 0.05756247,\n",
       "       0.05751694, 0.0616179 , 0.07532267, 0.04882817, 0.01078717,\n",
       "       0.02514048, 0.04787747, 0.05508456, 0.03907503, 0.01956987,\n",
       "       0.00764856, 0.01715563, 0.01781164, 0.02469596, 0.0232134 ,\n",
       "       0.01043404, 0.01601894, 0.01708586, 0.02061521, 0.02112663,\n",
       "       0.01298816, 0.02230105, 0.0172284 , 0.01164966, 0.01421802,\n",
       "       0.01883505])"
      ]
     },
     "execution_count": 129,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gbc.feature_importances_\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "#Get feature importances\n",
    "feature_mi = gbc.feature_importances_\n",
    "\n",
    "#Sort indices\n",
    "featureinds = feature_mi.argsort()\n",
    "\n",
    "plt.figure(figsize=(10,5))\n",
    "plt.title(\"Feature Importance\")\n",
    "\n",
    "#Plot features in order of importance\n",
    "plt.bar(X_train.columns.values[featureinds[::-1]],feature_mi[featureinds[::-1]])\n",
    "plt.xticks(X_train.columns.values[featureinds[::-1]],rotation=\"90\")\n",
    "plt.xlabel('Features')\n",
    "plt.ylabel('Importance')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 100.        ,  114.14141414,  128.28282828,  142.42424242,\n",
       "        156.56565657,  170.70707071,  184.84848485,  198.98989899,\n",
       "        213.13131313,  227.27272727,  241.41414141,  255.55555556,\n",
       "        269.6969697 ,  283.83838384,  297.97979798,  312.12121212,\n",
       "        326.26262626,  340.4040404 ,  354.54545455,  368.68686869,\n",
       "        382.82828283,  396.96969697,  411.11111111,  425.25252525,\n",
       "        439.39393939,  453.53535354,  467.67676768,  481.81818182,\n",
       "        495.95959596,  510.1010101 ,  524.24242424,  538.38383838,\n",
       "        552.52525253,  566.66666667,  580.80808081,  594.94949495,\n",
       "        609.09090909,  623.23232323,  637.37373737,  651.51515152,\n",
       "        665.65656566,  679.7979798 ,  693.93939394,  708.08080808,\n",
       "        722.22222222,  736.36363636,  750.50505051,  764.64646465,\n",
       "        778.78787879,  792.92929293,  807.07070707,  821.21212121,\n",
       "        835.35353535,  849.49494949,  863.63636364,  877.77777778,\n",
       "        891.91919192,  906.06060606,  920.2020202 ,  934.34343434,\n",
       "        948.48484848,  962.62626263,  976.76767677,  990.90909091,\n",
       "       1005.05050505, 1019.19191919, 1033.33333333, 1047.47474747,\n",
       "       1061.61616162, 1075.75757576, 1089.8989899 , 1104.04040404,\n",
       "       1118.18181818, 1132.32323232, 1146.46464646, 1160.60606061,\n",
       "       1174.74747475, 1188.88888889, 1203.03030303, 1217.17171717,\n",
       "       1231.31313131, 1245.45454545, 1259.5959596 , 1273.73737374,\n",
       "       1287.87878788, 1302.02020202, 1316.16161616, 1330.3030303 ,\n",
       "       1344.44444444, 1358.58585859, 1372.72727273, 1386.86868687,\n",
       "       1401.01010101, 1415.15151515, 1429.29292929, 1443.43434343,\n",
       "       1457.57575758, 1471.71717172, 1485.85858586, 1500.        ])"
      ]
     },
     "execution_count": 133,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.linspace(100,1500,100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = uncontrollable.drop(columns=['Label'])\n",
    "Y = uncontrollable['Label']\n",
    "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.80, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "parameters = {\n",
    "    \"loss\":[\"deviance\"],\n",
    "    \"learning_rate\": [0.01, 0.025, 0.05, 0.075, 0.1, 0.15, 0.2],\n",
    "    \"min_samples_split\": range(100, 2000, 10),\n",
    "    \"min_samples_leaf\": range(5,200,5),\n",
    "    \"max_depth\":[3,5,7,8],\n",
    "    \"max_features\":[\"log2\",\"sqrt\"],\n",
    "    \"criterion\": [\"friedman_mse\",  \"mae\"],\n",
    "    \"subsample\":[0.5, 0.618, 0.8, 0.85, 0.9, 0.95, 1.0],\n",
    "    \"n_estimators\": range(100,2100,100)\n",
    "    }\n",
    "\n",
    "clf = GridSearchCV(GradientBoostingClassifier(), parameters, cv=10, n_jobs=-1)\n",
    "\n",
    "clf.fit(X_train, Y_train)\n",
    "print(clf.score(trainX, trainY))\n",
    "print(clf.best_params_)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'subsample': 0.9,\n",
       " 'n_estimators': 1000,\n",
       " 'min_samples_split': 0.42727272727272736,\n",
       " 'min_samples_leaf': 0.24545454545454548,\n",
       " 'max_features': 'sqrt',\n",
       " 'max_depth': 7,\n",
       " 'loss': 'deviance',\n",
       " 'learning_rate': 0.05,\n",
       " 'criterion': 'friedman_mse'}"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "import numpy as np\n",
    "parameters = {\n",
    "    \"loss\":[\"deviance\"],\n",
    "    \"learning_rate\": [0.01, 0.025, 0.05, 0.075, 0.1, 0.15, 0.2],\n",
    "    \"min_samples_split\": np.linspace(0.1, 0.5, 12),\n",
    "    \"min_samples_leaf\": np.linspace(0.1, 0.5, 12),\n",
    "    \"max_depth\":[3,5,7,8],\n",
    "    \"max_features\":[\"log2\",\"sqrt\"],\n",
    "    \"criterion\": [\"friedman_mse\",  \"mae\"],\n",
    "    \"subsample\":[0.5, 0.618, 0.8, 0.85, 0.9, 0.95, 1.0],\n",
    "    \"n_estimators\": [100,200,500,1000]\n",
    "    }\n",
    "\n",
    "\n",
    "clf = RandomizedSearchCV(GradientBoostingClassifier(), parameters, random_state=0)\n",
    "search = clf.fit(X_train, Y_train)\n",
    "search.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GradientBoostingClassifier(learning_rate=0.05, max_depth=7, max_features='sqrt',\n",
       "                           min_samples_leaf=0.24545454545454548,\n",
       "                           min_samples_split=0.42727272727272736,\n",
       "                           n_estimators=1000, subsample=0.9)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5504827156649019"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gbc = GradientBoostingClassifier(learning_rate=0.05, max_depth=7, max_features='sqrt',\n",
    "                           min_samples_leaf=0.24545454545454548,\n",
    "                           min_samples_split=0.42727272727272736,\n",
    "                           n_estimators=1000, subsample=0.9)\n",
    "\n",
    "gbc.fit(X_train,Y_train)\n",
    "y_pred = gbc.predict(X_test)\n",
    "accuracy_score(Y_test,y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "({'subsample': 0.95,\n",
       "  'n_estimators': 100,\n",
       "  'min_samples_split': 0.3040816326530612,\n",
       "  'min_samples_leaf': 0.1979591836734694,\n",
       "  'max_features': 'log2',\n",
       "  'max_depth': 7,\n",
       "  'loss': 'deviance',\n",
       "  'learning_rate': 0.075,\n",
       "  'criterion': 'friedman_mse'},\n",
       " GradientBoostingClassifier(learning_rate=0.075, max_depth=7,\n",
       "                            max_features='log2',\n",
       "                            min_samples_leaf=0.1979591836734694,\n",
       "                            min_samples_split=0.3040816326530612,\n",
       "                            subsample=0.95))"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "parameters = {\n",
    "    \"loss\":[\"deviance\"],\n",
    "    \"learning_rate\": [0.01, 0.025, 0.05, 0.075, 0.1, 0.15, 0.2],\n",
    "    \"min_samples_split\": np.linspace(0.1, 0.5, 50),\n",
    "    \"min_samples_leaf\": np.linspace(0.1, 0.5, 50),\n",
    "    \"max_depth\":[3,5,7,8,9,10],\n",
    "    \"max_features\":[\"log2\",\"sqrt\"],\n",
    "    \"criterion\": [\"friedman_mse\",  \"mae\"],\n",
    "    \"subsample\":[0.5, 0.618, 0.8, 0.85, 0.9, 0.95, 1.0],\n",
    "    \"n_estimators\": [100,200,500,1000,1500,2000,2500]\n",
    "    }\n",
    "\n",
    "\n",
    "clf = RandomizedSearchCV(GradientBoostingClassifier(), parameters, random_state=0)\n",
    "search = clf.fit(X_train, Y_train)\n",
    "search.best_params_,search.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5591786411927839"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5492369978199938"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gbc = GradientBoostingClassifier(learning_rate=0.075, max_depth=7,\n",
    "                            max_features='log2',\n",
    "                            min_samples_leaf=0.1979591836734694,\n",
    "                            min_samples_split=0.3040816326530612,\n",
    "                            subsample=0.95)\n",
    "\n",
    "gbc.fit(X_train,Y_train)\n",
    "y_pred = gbc.predict(X_test)\n",
    "accuracy_score(Y_test,y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "20068"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import os\n",
    "uncontrollable = pd.read_csv(os.getcwd()+'/Airbnb_Optimization/datasets/uncontrollable_sample.csv',index_col='id')\n",
    "uncontrollable.head()\n",
    "len(uncontrollable)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import confusion_matrix,roc_auc_score\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import GradientBoostingClassifier\n",
    "X = uncontrollable.drop(columns=['Label'])\n",
    "Y = list(uncontrollable['Label'].T)\n",
    "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, train_size=0.80, random_state=42)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(16054, 4014)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(X_train),len(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "({'subsample': 1.0,\n",
       "  'n_estimators': 2500,\n",
       "  'min_samples_split': 0.22244897959183674,\n",
       "  'min_samples_leaf': 0.1489795918367347,\n",
       "  'max_features': 'sqrt',\n",
       "  'max_depth': 10,\n",
       "  'loss': 'deviance',\n",
       "  'learning_rate': 0.025,\n",
       "  'criterion': 'friedman_mse'},\n",
       " GradientBoostingClassifier(learning_rate=0.025, max_depth=10,\n",
       "                            max_features='sqrt',\n",
       "                            min_samples_leaf=0.1489795918367347,\n",
       "                            min_samples_split=0.22244897959183674,\n",
       "                            n_estimators=2500))"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "parameters = {\n",
    "    \"loss\":[\"deviance\"],\n",
    "    \"learning_rate\": [0.01, 0.025, 0.05, 0.075, 0.1, 0.15, 0.2],\n",
    "    \"min_samples_split\": np.linspace(0.1, 0.5, 50),\n",
    "    \"min_samples_leaf\": np.linspace(0.1, 0.5, 50),\n",
    "    \"max_depth\":[3,5,7,8,9,10],\n",
    "    \"max_features\":[\"log2\",\"sqrt\"],\n",
    "    \"criterion\": [\"friedman_mse\",  \"mae\"],\n",
    "    \"subsample\":[0.5, 0.618, 0.8, 0.85, 0.9, 0.95, 1.0],\n",
    "    \"n_estimators\": [100,200,500,1000,1500,2000,2500]\n",
    "    }\n",
    "\n",
    "\n",
    "clf = RandomizedSearchCV(GradientBoostingClassifier(), parameters, random_state=0)\n",
    "search = clf.fit(X_train, Y_train)\n",
    "search.best_params_,search.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5697558545092177"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "gbc = GradientBoostingClassifier(learning_rate=0.025, max_depth=10,\n",
    "                            max_features='sqrt',\n",
    "                            min_samples_leaf=0.1489795918367347,\n",
    "                            min_samples_split=0.22244897959183674,\n",
    "                            n_estimators=2500)\n",
    "gbc.fit(X_train,Y_train)\n",
    "y_pred = gbc.predict(X_test)\n",
    "accuracy_score(Y_test,y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-25-2a6260a9e29a>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m     15\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     16\u001b[0m \u001b[0mclf\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mRandomizedSearchCV\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mGradientBoostingClassifier\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mparameters\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mrandom_state\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 17\u001b[0;31m \u001b[0msearch\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mclf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_train\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mY_train\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     18\u001b[0m \u001b[0msearch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbest_params_\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/sklearn/utils/validation.py\u001b[0m in \u001b[0;36minner_f\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m     71\u001b[0m                           FutureWarning)\n\u001b[1;32m     72\u001b[0m         \u001b[0mkwargs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mupdate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m{\u001b[0m\u001b[0mk\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0marg\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mk\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0marg\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mzip\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msig\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mparameters\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 73\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     74\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0minner_f\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     75\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/sklearn/model_selection/_search.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, X, y, groups, **fit_params)\u001b[0m\n\u001b[1;32m    734\u001b[0m                 \u001b[0;32mreturn\u001b[0m \u001b[0mresults\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    735\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 736\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_run_search\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mevaluate_candidates\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    737\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    738\u001b[0m         \u001b[0;31m# For multi-metric evaluation, store the best_index_, best_params_ and\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/sklearn/model_selection/_search.py\u001b[0m in \u001b[0;36m_run_search\u001b[0;34m(self, evaluate_candidates)\u001b[0m\n\u001b[1;32m   1527\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_run_search\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevaluate_candidates\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1528\u001b[0m         \u001b[0;34m\"\"\"Search n_iter candidates from param_distributions\"\"\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1529\u001b[0;31m         evaluate_candidates(ParameterSampler(\n\u001b[0m\u001b[1;32m   1530\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mparam_distributions\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_iter\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1531\u001b[0m             random_state=self.random_state))\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/sklearn/model_selection/_search.py\u001b[0m in \u001b[0;36mevaluate_candidates\u001b[0;34m(candidate_params)\u001b[0m\n\u001b[1;32m    706\u001b[0m                               n_splits, n_candidates, n_candidates * n_splits))\n\u001b[1;32m    707\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 708\u001b[0;31m                 out = parallel(delayed(_fit_and_score)(clone(base_estimator),\n\u001b[0m\u001b[1;32m    709\u001b[0m                                                        \u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    710\u001b[0m                                                        \u001b[0mtrain\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtrain\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtest\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtest\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/joblib/parallel.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, iterable)\u001b[0m\n\u001b[1;32m   1030\u001b[0m                 \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_iterating\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_original_iterator\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1031\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1032\u001b[0;31m             \u001b[0;32mwhile\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdispatch_one_batch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0miterator\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1033\u001b[0m                 \u001b[0;32mpass\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1034\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/joblib/parallel.py\u001b[0m in \u001b[0;36mdispatch_one_batch\u001b[0;34m(self, iterator)\u001b[0m\n\u001b[1;32m    845\u001b[0m                 \u001b[0;32mreturn\u001b[0m \u001b[0;32mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    846\u001b[0m             \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 847\u001b[0;31m                 \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_dispatch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtasks\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    848\u001b[0m                 \u001b[0;32mreturn\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    849\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/joblib/parallel.py\u001b[0m in \u001b[0;36m_dispatch\u001b[0;34m(self, batch)\u001b[0m\n\u001b[1;32m    763\u001b[0m         \u001b[0;32mwith\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_lock\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    764\u001b[0m             \u001b[0mjob_idx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_jobs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 765\u001b[0;31m             \u001b[0mjob\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_backend\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mapply_async\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbatch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcallback\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcb\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    766\u001b[0m             \u001b[0;31m# A job can complete so quickly than its callback is\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    767\u001b[0m             \u001b[0;31m# called before we get here, causing self._jobs to\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/joblib/_parallel_backends.py\u001b[0m in \u001b[0;36mapply_async\u001b[0;34m(self, func, callback)\u001b[0m\n\u001b[1;32m    206\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mapply_async\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcallback\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    207\u001b[0m         \u001b[0;34m\"\"\"Schedule a func to be run\"\"\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 208\u001b[0;31m         \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mImmediateResult\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfunc\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    209\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mcallback\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    210\u001b[0m             \u001b[0mcallback\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mresult\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/joblib/_parallel_backends.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, batch)\u001b[0m\n\u001b[1;32m    570\u001b[0m         \u001b[0;31m# Don't delay the application, to avoid keeping the input\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    571\u001b[0m         \u001b[0;31m# arguments in memory\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 572\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mresults\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mbatch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    573\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    574\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mget\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/joblib/parallel.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    250\u001b[0m         \u001b[0;31m# change the default number of processes to -1\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    251\u001b[0m         \u001b[0;32mwith\u001b[0m \u001b[0mparallel_backend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_backend\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn_jobs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_n_jobs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 252\u001b[0;31m             return [func(*args, **kwargs)\n\u001b[0m\u001b[1;32m    253\u001b[0m                     for func, args, kwargs in self.items]\n\u001b[1;32m    254\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/joblib/parallel.py\u001b[0m in \u001b[0;36m<listcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m    250\u001b[0m         \u001b[0;31m# change the default number of processes to -1\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    251\u001b[0m         \u001b[0;32mwith\u001b[0m \u001b[0mparallel_backend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_backend\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn_jobs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_n_jobs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 252\u001b[0;31m             return [func(*args, **kwargs)\n\u001b[0m\u001b[1;32m    253\u001b[0m                     for func, args, kwargs in self.items]\n\u001b[1;32m    254\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/sklearn/model_selection/_validation.py\u001b[0m in \u001b[0;36m_fit_and_score\u001b[0;34m(estimator, X, y, scorer, train, test, verbose, parameters, fit_params, return_train_score, return_parameters, return_n_test_samples, return_times, return_estimator, error_score)\u001b[0m\n\u001b[1;32m    529\u001b[0m             \u001b[0mestimator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_train\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mfit_params\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    530\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 531\u001b[0;31m             \u001b[0mestimator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_train\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_train\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mfit_params\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    532\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    533\u001b[0m     \u001b[0;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/sklearn/ensemble/_gb.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, X, y, sample_weight, monitor)\u001b[0m\n\u001b[1;32m    496\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    497\u001b[0m         \u001b[0;31m# fit the boosting stages\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 498\u001b[0;31m         n_stages = self._fit_stages(\n\u001b[0m\u001b[1;32m    499\u001b[0m             \u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mraw_predictions\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msample_weight\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_rng\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mX_val\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_val\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    500\u001b[0m             sample_weight_val, begin_at_stage, monitor, X_idx_sorted)\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/sklearn/ensemble/_gb.py\u001b[0m in \u001b[0;36m_fit_stages\u001b[0;34m(self, X, y, raw_predictions, sample_weight, random_state, X_val, y_val, sample_weight_val, begin_at_stage, monitor, X_idx_sorted)\u001b[0m\n\u001b[1;32m    553\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    554\u001b[0m             \u001b[0;31m# fit next stage of trees\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 555\u001b[0;31m             raw_predictions = self._fit_stage(\n\u001b[0m\u001b[1;32m    556\u001b[0m                 \u001b[0mi\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mraw_predictions\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msample_weight\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msample_mask\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    557\u001b[0m                 random_state, X_idx_sorted, X_csc, X_csr)\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/sklearn/ensemble/_gb.py\u001b[0m in \u001b[0;36m_fit_stage\u001b[0;34m(self, i, X, y, raw_predictions, sample_weight, sample_mask, random_state, X_idx_sorted, X_csc, X_csr)\u001b[0m\n\u001b[1;32m    209\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    210\u001b[0m             \u001b[0mX\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mX_csr\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mX_csr\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0mX\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 211\u001b[0;31m             tree.fit(X, residual, sample_weight=sample_weight,\n\u001b[0m\u001b[1;32m    212\u001b[0m                      check_input=False, X_idx_sorted=X_idx_sorted)\n\u001b[1;32m    213\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/sklearn/tree/_classes.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, X, y, sample_weight, check_input, X_idx_sorted)\u001b[0m\n\u001b[1;32m   1240\u001b[0m         \"\"\"\n\u001b[1;32m   1241\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1242\u001b[0;31m         super().fit(\n\u001b[0m\u001b[1;32m   1243\u001b[0m             \u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1244\u001b[0m             \u001b[0msample_weight\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0msample_weight\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/sklearn/tree/_classes.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, X, y, sample_weight, check_input, X_idx_sorted)\u001b[0m\n\u001b[1;32m    373\u001b[0m                                            min_impurity_split)\n\u001b[1;32m    374\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 375\u001b[0;31m         \u001b[0mbuilder\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbuild\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtree_\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msample_weight\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mX_idx_sorted\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    376\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    377\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_outputs_\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m1\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mis_classifier\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "import numpy as np\n",
    "parameters = {\n",
    "    #\"loss\":[\"deviance\"],\n",
    "    \"learning_rate\": [0.025] #[0.01, 0.025, 0.05, 0.075, 0.1, 0.15, 0.2],\n",
    "    ,\"min_samples_split\": [0.22244897959183674]#np.linspace(0.1, 0.5, 12),\n",
    "    ,\"min_samples_leaf\": [0.1489795918367347]# np.linspace(0.1, 0.5, 12),\n",
    "    ,\"max_depth\":[10]#3,5,7,8],\n",
    "     ,\"max_features\":np.linspace(0.1,0.5,12),\n",
    "#     \"criterion\": [\"friedman_mse\",  \"mae\"],\n",
    "    #\"subsample\":[0.5, 0.618, 0.8, 0.85, 0.9, 0.95, 1.0],\n",
    "    \"n_estimators\": [2500]#100,200,500,1000]\n",
    "    }\n",
    "\n",
    "\n",
    "clf = RandomizedSearchCV(GradientBoostingClassifier(), parameters, random_state=0)\n",
    "search = clf.fit(X_train, Y_train)\n",
    "search.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'subsample': 1.0,\n",
       " 'n_estimators': 2500,\n",
       " 'min_samples_split': 0.22244897959183674,\n",
       " 'min_samples_leaf': 0.1489795918367347,\n",
       " 'max_features': 'sqrt',\n",
       " 'max_depth': 10,\n",
       " 'loss': 'deviance',\n",
       " 'learning_rate': 0.025,\n",
       " 'criterion': 'friedman_mse'}"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-27-d1fc15a21e3b>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m     13\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     14\u001b[0m \u001b[0mclf\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mRandomizedSearchCV\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mGradientBoostingClassifier\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mparameters\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mrandom_state\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 15\u001b[0;31m \u001b[0msearch\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mclf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_train\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mY_train\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     16\u001b[0m \u001b[0msearch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbest_params_\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0msearch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbest_estimator_\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/sklearn/utils/validation.py\u001b[0m in \u001b[0;36minner_f\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m     71\u001b[0m                           FutureWarning)\n\u001b[1;32m     72\u001b[0m         \u001b[0mkwargs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mupdate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m{\u001b[0m\u001b[0mk\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0marg\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mk\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0marg\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mzip\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msig\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mparameters\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 73\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     74\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0minner_f\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     75\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/sklearn/model_selection/_search.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, X, y, groups, **fit_params)\u001b[0m\n\u001b[1;32m    734\u001b[0m                 \u001b[0;32mreturn\u001b[0m \u001b[0mresults\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    735\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 736\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_run_search\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mevaluate_candidates\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    737\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    738\u001b[0m         \u001b[0;31m# For multi-metric evaluation, store the best_index_, best_params_ and\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/sklearn/model_selection/_search.py\u001b[0m in \u001b[0;36m_run_search\u001b[0;34m(self, evaluate_candidates)\u001b[0m\n\u001b[1;32m   1527\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_run_search\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevaluate_candidates\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1528\u001b[0m         \u001b[0;34m\"\"\"Search n_iter candidates from param_distributions\"\"\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1529\u001b[0;31m         evaluate_candidates(ParameterSampler(\n\u001b[0m\u001b[1;32m   1530\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mparam_distributions\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_iter\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1531\u001b[0m             random_state=self.random_state))\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/sklearn/model_selection/_search.py\u001b[0m in \u001b[0;36mevaluate_candidates\u001b[0;34m(candidate_params)\u001b[0m\n\u001b[1;32m    706\u001b[0m                               n_splits, n_candidates, n_candidates * n_splits))\n\u001b[1;32m    707\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 708\u001b[0;31m                 out = parallel(delayed(_fit_and_score)(clone(base_estimator),\n\u001b[0m\u001b[1;32m    709\u001b[0m                                                        \u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    710\u001b[0m                                                        \u001b[0mtrain\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtrain\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtest\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtest\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/joblib/parallel.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, iterable)\u001b[0m\n\u001b[1;32m   1030\u001b[0m                 \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_iterating\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_original_iterator\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1031\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1032\u001b[0;31m             \u001b[0;32mwhile\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdispatch_one_batch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0miterator\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1033\u001b[0m                 \u001b[0;32mpass\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1034\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/joblib/parallel.py\u001b[0m in \u001b[0;36mdispatch_one_batch\u001b[0;34m(self, iterator)\u001b[0m\n\u001b[1;32m    845\u001b[0m                 \u001b[0;32mreturn\u001b[0m \u001b[0;32mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    846\u001b[0m             \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 847\u001b[0;31m                 \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_dispatch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtasks\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    848\u001b[0m                 \u001b[0;32mreturn\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    849\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/joblib/parallel.py\u001b[0m in \u001b[0;36m_dispatch\u001b[0;34m(self, batch)\u001b[0m\n\u001b[1;32m    763\u001b[0m         \u001b[0;32mwith\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_lock\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    764\u001b[0m             \u001b[0mjob_idx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_jobs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 765\u001b[0;31m             \u001b[0mjob\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_backend\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mapply_async\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbatch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcallback\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcb\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    766\u001b[0m             \u001b[0;31m# A job can complete so quickly than its callback is\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    767\u001b[0m             \u001b[0;31m# called before we get here, causing self._jobs to\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/joblib/_parallel_backends.py\u001b[0m in \u001b[0;36mapply_async\u001b[0;34m(self, func, callback)\u001b[0m\n\u001b[1;32m    206\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mapply_async\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcallback\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    207\u001b[0m         \u001b[0;34m\"\"\"Schedule a func to be run\"\"\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 208\u001b[0;31m         \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mImmediateResult\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfunc\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    209\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mcallback\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    210\u001b[0m             \u001b[0mcallback\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mresult\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/joblib/_parallel_backends.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, batch)\u001b[0m\n\u001b[1;32m    570\u001b[0m         \u001b[0;31m# Don't delay the application, to avoid keeping the input\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    571\u001b[0m         \u001b[0;31m# arguments in memory\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 572\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mresults\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mbatch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    573\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    574\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mget\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/joblib/parallel.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    250\u001b[0m         \u001b[0;31m# change the default number of processes to -1\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    251\u001b[0m         \u001b[0;32mwith\u001b[0m \u001b[0mparallel_backend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_backend\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn_jobs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_n_jobs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 252\u001b[0;31m             return [func(*args, **kwargs)\n\u001b[0m\u001b[1;32m    253\u001b[0m                     for func, args, kwargs in self.items]\n\u001b[1;32m    254\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/joblib/parallel.py\u001b[0m in \u001b[0;36m<listcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m    250\u001b[0m         \u001b[0;31m# change the default number of processes to -1\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    251\u001b[0m         \u001b[0;32mwith\u001b[0m \u001b[0mparallel_backend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_backend\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn_jobs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_n_jobs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 252\u001b[0;31m             return [func(*args, **kwargs)\n\u001b[0m\u001b[1;32m    253\u001b[0m                     for func, args, kwargs in self.items]\n\u001b[1;32m    254\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/sklearn/model_selection/_validation.py\u001b[0m in \u001b[0;36m_fit_and_score\u001b[0;34m(estimator, X, y, scorer, train, test, verbose, parameters, fit_params, return_train_score, return_parameters, return_n_test_samples, return_times, return_estimator, error_score)\u001b[0m\n\u001b[1;32m    529\u001b[0m             \u001b[0mestimator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_train\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mfit_params\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    530\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 531\u001b[0;31m             \u001b[0mestimator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_train\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_train\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mfit_params\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    532\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    533\u001b[0m     \u001b[0;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/sklearn/ensemble/_gb.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, X, y, sample_weight, monitor)\u001b[0m\n\u001b[1;32m    496\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    497\u001b[0m         \u001b[0;31m# fit the boosting stages\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 498\u001b[0;31m         n_stages = self._fit_stages(\n\u001b[0m\u001b[1;32m    499\u001b[0m             \u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mraw_predictions\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msample_weight\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_rng\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mX_val\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_val\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    500\u001b[0m             sample_weight_val, begin_at_stage, monitor, X_idx_sorted)\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/sklearn/ensemble/_gb.py\u001b[0m in \u001b[0;36m_fit_stages\u001b[0;34m(self, X, y, raw_predictions, sample_weight, random_state, X_val, y_val, sample_weight_val, begin_at_stage, monitor, X_idx_sorted)\u001b[0m\n\u001b[1;32m    553\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    554\u001b[0m             \u001b[0;31m# fit next stage of trees\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 555\u001b[0;31m             raw_predictions = self._fit_stage(\n\u001b[0m\u001b[1;32m    556\u001b[0m                 \u001b[0mi\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mraw_predictions\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msample_weight\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msample_mask\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    557\u001b[0m                 random_state, X_idx_sorted, X_csc, X_csr)\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/sklearn/ensemble/_gb.py\u001b[0m in \u001b[0;36m_fit_stage\u001b[0;34m(self, i, X, y, raw_predictions, sample_weight, sample_mask, random_state, X_idx_sorted, X_csc, X_csr)\u001b[0m\n\u001b[1;32m    209\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    210\u001b[0m             \u001b[0mX\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mX_csr\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mX_csr\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0mX\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 211\u001b[0;31m             tree.fit(X, residual, sample_weight=sample_weight,\n\u001b[0m\u001b[1;32m    212\u001b[0m                      check_input=False, X_idx_sorted=X_idx_sorted)\n\u001b[1;32m    213\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/sklearn/tree/_classes.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, X, y, sample_weight, check_input, X_idx_sorted)\u001b[0m\n\u001b[1;32m   1240\u001b[0m         \"\"\"\n\u001b[1;32m   1241\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1242\u001b[0;31m         super().fit(\n\u001b[0m\u001b[1;32m   1243\u001b[0m             \u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1244\u001b[0m             \u001b[0msample_weight\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0msample_weight\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.8/site-packages/sklearn/tree/_classes.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, X, y, sample_weight, check_input, X_idx_sorted)\u001b[0m\n\u001b[1;32m    373\u001b[0m                                            min_impurity_split)\n\u001b[1;32m    374\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 375\u001b[0;31m         \u001b[0mbuilder\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbuild\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtree_\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msample_weight\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mX_idx_sorted\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    376\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    377\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_outputs_\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m1\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mis_classifier\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "parameters = {\n",
    "    \"loss\":[\"deviance\"],\n",
    "    \"learning_rate\": [0.01, 0.025, 0.05, 0.075, 0.1, 0.15, 0.2],\n",
    "    \"min_samples_split\": np.linspace(0.1, 0.5, 12),\n",
    "    \"min_samples_leaf\": np.linspace(0.1, 0.5, 12),\n",
    "    \"max_depth\":[3,5,7,8,9,10],\n",
    "    \"max_features\":np.linspace(0.1,1,24),\n",
    "    \"criterion\": [\"friedman_mse\",  \"mae\"],\n",
    "    \"subsample\":[0.5, 0.618, 0.8, 0.85, 0.9, 0.95, 1.0],\n",
    "    \"n_estimators\": [100,200,500,1000,1500,2000,2500]fd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "a"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "a"
   ]
  },
  {
   "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": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}