Classification Model.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Classification Models for ifood-data-business-analyst-test-latam-main\n",
    "\n",
    "By: Alejandro Rojas Benítez\n",
    "https://www.linkedin.com/in/arojasb3/\n",
    "\n",
    "this module works with the standard anaconda package except imbalanced-learn\n",
    "\n",
    "It will need the command pip install imbalanced-learn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import numpy as np\n",
    "import warnings\n",
    "from sklearn.manifold import TSNE\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.cluster import KMeans\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from copy import deepcopy\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import roc_curve, auc\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import f1_score\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import recall_score\n",
    "from sklearn.metrics import plot_roc_curve\n",
    "\n",
    "from imblearn.under_sampling import RandomUnderSampler\n",
    "warnings.filterwarnings('ignore')\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "df2 = pd.read_csv('data/not_scaled_data_with_clusters.csv')\n",
    "df_scaled = pd.read_csv('data/scaled_data_with_clusters.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# numerical variables we will take into account\n",
    "numerical_cols = [ 'Income', 'Kidhome',\n",
    "       'Teenhome', 'Recency', 'MntWines', 'MntFruits',\n",
    "       'MntMeatProducts', 'MntFishProducts', 'MntSweetProducts',\n",
    "       'MntGoldProds', 'NumDealsPurchases', 'NumWebPurchases',\n",
    "       'NumCatalogPurchases', 'NumStorePurchases', 'NumWebVisitsMonth',\n",
    "                 'age', 'enrollment_time', 'clusters_kmeans']\n",
    "\n",
    "# categorical variables we will take into account. We will ignore Complain since it is too unbalanced\n",
    "categorical_cols = ['Education', 'Marital_Status']\n",
    "\n",
    "# campaigns for cluster analysis\n",
    "campaings = ['AcceptedCmp1', 'AcceptedCmp2', 'AcceptedCmp3', 'AcceptedCmp4', 'AcceptedCmp5', 'Response']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "random_state = 42"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Adapting the dataset\n",
    "We will transfor all the categorical variables into dummies, creating new binary columns for each of their values. We will use `Education`, `Marital_Status` and `clusters_kmeans`.\n",
    "\n",
    "This will enable an easier analysis for the influence of those variables in the model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>Income</th>\n",
       "      <th>Kidhome</th>\n",
       "      <th>Teenhome</th>\n",
       "      <th>Recency</th>\n",
       "      <th>MntWines</th>\n",
       "      <th>MntFruits</th>\n",
       "      <th>MntMeatProducts</th>\n",
       "      <th>MntFishProducts</th>\n",
       "      <th>MntSweetProducts</th>\n",
       "      <th>MntGoldProds</th>\n",
       "      <th>...</th>\n",
       "      <th>Education_PhD</th>\n",
       "      <th>Marital_Status_Divorced</th>\n",
       "      <th>Marital_Status_Married</th>\n",
       "      <th>Marital_Status_Single</th>\n",
       "      <th>Marital_Status_Together</th>\n",
       "      <th>Marital_Status_Widow</th>\n",
       "      <th>clusters_kmeans_0</th>\n",
       "      <th>clusters_kmeans_1</th>\n",
       "      <th>clusters_kmeans_2</th>\n",
       "      <th>clusters_kmeans_3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.516867</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.585859</td>\n",
       "      <td>0.425318</td>\n",
       "      <td>0.442211</td>\n",
       "      <td>0.554878</td>\n",
       "      <td>0.664093</td>\n",
       "      <td>0.444444</td>\n",
       "      <td>0.353414</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.396485</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.383838</td>\n",
       "      <td>0.007368</td>\n",
       "      <td>0.005025</td>\n",
       "      <td>0.006098</td>\n",
       "      <td>0.007722</td>\n",
       "      <td>0.005051</td>\n",
       "      <td>0.024096</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.654408</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.262626</td>\n",
       "      <td>0.285332</td>\n",
       "      <td>0.246231</td>\n",
       "      <td>0.129065</td>\n",
       "      <td>0.428571</td>\n",
       "      <td>0.106061</td>\n",
       "      <td>0.168675</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.195425</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.262626</td>\n",
       "      <td>0.007368</td>\n",
       "      <td>0.020101</td>\n",
       "      <td>0.020325</td>\n",
       "      <td>0.038610</td>\n",
       "      <td>0.015152</td>\n",
       "      <td>0.020080</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.518449</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.949495</td>\n",
       "      <td>0.115874</td>\n",
       "      <td>0.216080</td>\n",
       "      <td>0.119919</td>\n",
       "      <td>0.177606</td>\n",
       "      <td>0.136364</td>\n",
       "      <td>0.060241</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 36 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     Income  Kidhome  Teenhome   Recency  MntWines  MntFruits  \\\n",
       "0  0.516867      0.0       0.0  0.585859  0.425318   0.442211   \n",
       "1  0.396485      0.5       0.5  0.383838  0.007368   0.005025   \n",
       "2  0.654408      0.0       0.0  0.262626  0.285332   0.246231   \n",
       "3  0.195425      0.5       0.0  0.262626  0.007368   0.020101   \n",
       "4  0.518449      0.5       0.0  0.949495  0.115874   0.216080   \n",
       "\n",
       "   MntMeatProducts  MntFishProducts  MntSweetProducts  MntGoldProds  ...  \\\n",
       "0         0.554878         0.664093          0.444444      0.353414  ...   \n",
       "1         0.006098         0.007722          0.005051      0.024096  ...   \n",
       "2         0.129065         0.428571          0.106061      0.168675  ...   \n",
       "3         0.020325         0.038610          0.015152      0.020080  ...   \n",
       "4         0.119919         0.177606          0.136364      0.060241  ...   \n",
       "\n",
       "   Education_PhD  Marital_Status_Divorced  Marital_Status_Married  \\\n",
       "0              0                        0                       0   \n",
       "1              0                        0                       0   \n",
       "2              0                        0                       0   \n",
       "3              0                        0                       0   \n",
       "4              1                        0                       1   \n",
       "\n",
       "   Marital_Status_Single  Marital_Status_Together  Marital_Status_Widow  \\\n",
       "0                      1                        0                     0   \n",
       "1                      1                        0                     0   \n",
       "2                      0                        1                     0   \n",
       "3                      0                        1                     0   \n",
       "4                      0                        0                     0   \n",
       "\n",
       "   clusters_kmeans_0  clusters_kmeans_1  clusters_kmeans_2  clusters_kmeans_3  \n",
       "0                  0                  0                  1                  0  \n",
       "1                  0                  0                  0                  1  \n",
       "2                  0                  0                  1                  0  \n",
       "3                  1                  0                  0                  0  \n",
       "4                  1                  0                  0                  0  \n",
       "\n",
       "[5 rows x 36 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create a new dataset where categorical variables are dummies\n",
    "X = pd.get_dummies(df_scaled[numerical_cols + categorical_cols + campaings[:-1]])\n",
    "\n",
    "X = pd.get_dummies(X, columns=[ 'clusters_kmeans'])\n",
    "X.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The dataset looks good, now lets prepare our the classification, which will be the success of the last campaign `Response`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = df_scaled['Response']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Wait!\n",
    "Our data is unbalanced, only 15% of the customers accepted the campaign, that makes the creation of the model a lot harder because we have a lot more samples of one class. \n",
    "\n",
    "To fix that issue we will undersample the dataset, basically for every 2 customers that rejected the campaign, we will have one that accepted it. Leaving a proportion of 33% instead."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# undersampling thanks to https://machinelearningmastery.com/random-oversampling-and-undersampling-for-imbalanced-classification/\n",
    "undersample = RandomUnderSampler(sampling_strategy=0.5, random_state = random_state)\n",
    "\n",
    "X_under, y_under= undersample.fit_resample(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((990, 36), (2185, 36))"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_under.shape, X.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    660\n",
       "1    330\n",
       "Name: Response, dtype: int64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_under.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "That looks a lot better, we went down from 2185 customers to only 990 but with that proportion. This will make the fitting process a lot easier.\n",
    "\n",
    "Now lets start splitting our data, we will use the usual 60% training data, 20% validation and 20% test.\n",
    "\n",
    "We will train our model with the training set, tune a bit some parameters with the validation and at the end check its performance with the test set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# split dataset into 60% 20% 20%\n",
    "X_train, X_test, y_train, y_test = train_test_split(X_under, y_under, test_size=0.2, random_state=random_state)\n",
    "\n",
    "X_train, X_val, y_train, y_val = train_test_split(X_train, y_train, test_size=0.25, random_state=random_state)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Logistic Regression\n",
    "\n",
    "As an initial approach we will use a logistic regression, it is a fairly simple model that enables us to see an importance of an influence of each of the variables: KISS principle.\n",
    "\n",
    "As a parameter we will play around the C parameter from sci kit learn, which refers to the inverse regularization strength, where lower values specify stronger regularization. Our optimization criteria will be maximizing the f1 score, this choice because we want to be able to share the accuracy and recall values with the marketing deparment team, 2 easier metrics to take into a real approach compared to the AUC."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# logistic regression parameters set\n",
    "regularization = np.linspace(0.01, 1, 100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# finding best regularization for our logistic regression\n",
    "prev_f1 = 0\n",
    "logistic = None\n",
    "best_c = 0\n",
    "for C in regularization:\n",
    "    clf = LogisticRegression(random_state=random_state, C = C)\n",
    "    clf.fit(X_train, y_train)\n",
    "    curr_f1 = f1_score(clf.predict(X_val), y_val, average='binary')\n",
    "    \n",
    "    if prev_f1 < curr_f1:\n",
    "        prev_f1 = curr_f1\n",
    "        logistic = clf\n",
    "        best_c = C\n",
    "C"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Well, we ended up with the default C value but anyway it was a good try!\n",
    "\n",
    "\n",
    "Now lets check our performance metrics for the best model we found"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.797979797979798"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy_score(logistic.predict(X_test), y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7735849056603774"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "recall_score(logistic.predict(X_test), y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6721311475409837"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f1_score(logistic.predict(X_test), y_test, average='binary')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_roc_curve(logistic, X_test, y_test)\n",
    "plt.plot([0, 1], [0, 1], color='navy', linestyle='--');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "That could be decent, for the moment we can be happy about our results. Now lets try to understand a bit on how the varibales affect our logistic regression."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "models_coefficients = pd.DataFrame(logistic.coef_[0], X_train.columns, columns = ['log_coefficients'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = models_coefficients.sort_values(by='log_coefficients',ascending=False).plot(kind='barh', figsize=(16,8), grid=True, title='Coefficients for Logistic Regression Model')\n",
    "\n",
    "ax.set_xlabel('Coefficient Value');\n",
    "\n",
    "plt.savefig('figures/log_coefficients.pdf')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Thats interesting, so recency affects the campaign in a negative way. Customers with higher recency have a much higher chance to reject the campaign. Other 2 variables that heavy influence the rejection are the number of store purchases and the number of teens at home. \n",
    "\n",
    "On the other hand, high enrollment times, having accepted the third campaign and the number of web purchases heavily influence the acceptance of the last campaign.\n",
    "\n",
    "Now, looking at how our clusters obtained from the kmeans affect the model, clusters 1 and 3 tend to reject but their influence is fairly low while clusters 0 and 2 help the acceptance also with a low coefficient. It seems that the clusters we obtained before can influence the model, remembering that cluster 0 was the group that had the highest acceptance of them all.\n",
    "\n",
    "Another interesting thing is that the customers from cluster 0 are middle of the pack on web purchases but the highest on catalog purchases which makes sense with the logistic model.\n",
    "\n",
    "### Random Forest\n",
    "\n",
    "Now that we got some results with the logistic regression, lets focus on another model. We will try now the random forest and then plot the importance of the variables for the best model.\n",
    "\n",
    "In this approach we will try different number of trees and different max depths for the random forest. Again, optimizing around the f1 score. We defined the upper bound of the depth as the number of features we have in our dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Random Forest parameters\n",
    "number_of_trees = range(100, 151)\n",
    "\n",
    "depths = range(2, len(X.columns))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(107, 9)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# finding best regularization for our logistic regression\n",
    "prev_f1_rf = 0\n",
    "rf = None\n",
    "best_trees = 0\n",
    "best_depth = 0\n",
    "for trees, depth in zip(number_of_trees, depths):\n",
    "    clf = RandomForestClassifier(n_estimators=trees, max_depth=depth, random_state=random_state)\n",
    "    clf.fit(X_train, y_train)\n",
    "    curr_f1 = f1_score(clf.predict(X_val), y_val, average='binary')\n",
    "    \n",
    "    if prev_f1_rf < curr_f1:\n",
    "        prev_f1_rf = curr_f1\n",
    "        rf = clf\n",
    "        best_trees = trees\n",
    "        best_depth = depth\n",
    "\n",
    "best_trees, best_depth"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ok, our best random forest model was made with 108 trees and a maximum depth of 10.\n",
    "\n",
    "Now lets check the performance metrics."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7474747474747475"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy_score(rf.predict(X_test), y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6557377049180327"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "recall_score(rf.predict(X_test), y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6153846153846154"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f1_score(rf.predict(X_test), y_test, average='binary')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_roc_curve(rf, X_test, y_test)\n",
    "plt.plot([0, 1], [0, 1], color='navy', linestyle='--');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Well, all metrics for our test set are worse on the Random Forest than on the logistic regression, anyway they are a bit close. \n",
    "Lets check the most important variables for the campaign, maybe we can find some similarities between them both."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "models_coefficients_rf = pd.DataFrame(rf.feature_importances_, X_train.columns, columns = ['rf_coefficients'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = models_coefficients_rf.sort_values(by='rf_coefficients',ascending=False).plot(kind='barh', figsize=(16,8), grid=True, title='Feature Importance for Random Forest Model')\n",
    "\n",
    "ax.set_xlabel('Feature Importance');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Whoa! thats a lot of variables, lets plot the top 5 below"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA+oAAAHwCAYAAAAxTJUGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdebhdZXk3/u9NgkIJggxFEREUB2QWVJxKqAoKCtXKizZVU6uYKqL+0FdsfXFuURxwAotTHFCRKE6I4sCBOqJBICIgKmlFHBAUCQ2Q4PP7Y68Dm8NJcnISctYhn8915crea3iee6/1nMB3P2utU621AAAAAP2wwVQXAAAAANxGUAcAAIAeEdQBAACgRwR1AAAA6BFBHQAAAHpEUAcAAIAeEdQBgN6pqqdV1a+qaklV7TXV9ayOqppbVd+e6jruyqpqflW9aYLbLq6qJ9zZNQGsTYI6ACvUhaTRP3+pqqVD7+espT7mV9XNY/qasYJtZ3d1DG/7pTXsf3ZVXbkmbUyiz94Eh6oaqarnT3Ud43hbkiNba7Naaz9e08a6z3ljN2b+UFWfq6p7r4U6p0xV7VBVbczPw4XruIZWVTutZP3cbpt3jFn+d93y+Xd6kQDTkKAOwAp1IWlWa21Wkv9J8tShZaesxa7eOtxXa+2WlWx71Zhtn7oW61htVTVzKvufrBro8/8H3C/JxZPZcUVf9KQL/kl2SjIrgy8D7go2H/p52GN1d14HY/gXSQ4f089zkvzsTu4XYNrq83+gAeipqrp7VZ1QVVd1f06oqrt362ZX1ZVV9a/dzOXitTX7voqa9q2q71bVn6rqwqqaPbTun6rqkqq6vqp+WVUv7JZvkuTMJNsOzUhuO/ay2rGz7t1nelVVXZTkhqqaubL+V1H33Kr6TlW9s9v3l1X16G75r6rq91X13KHt51fV+6vq693nOaeq7je0/tFV9cOquq77+9FD60aq6s1V9Z0k/5vk40kel+S93Wd/b7fdu7q+/1xVC6vqcUNtvK6qPlNVH+v6v7iq9hlaf99utvrqqrpmtM1u3fO68/DHqvracN1D29y9qpYkmZHkwqr6Rbd8567+P3V9HjLmmJxUVV+pqhuS7L+yY95a+1OSzyfZc6iNccdIt250TB/dnY/fVNU/Da3fsqq+2B2v85I8YMxnWtU5eVM3dpZU1Ze69k7p2vthVe2wss8znm4cf7Gqrq2qn1fVC4bWva6qFlTVJ6rqz0nmVtVmVfWh7rP9uqtpRrf9Tt04u64GP9OndsvP7Zq8sKv98BWU89ski5Ic2O23RZJHJ/nimJoP6c7tn7rjsvPQur2q6vzu/JyaZKMx+z6lqi7o9v1uVe2+uscMoE8EdQAm49+S7JtB0NkjySOSvGZo/b2SbJXkPkmem+TkqnrwStp7URcoFlbV369uMVV1nyRnJHlTki2SvCLJZ6tq626T3yd5SpJ7JPmnJO+sqoe11m5I8uTcfpb+qgl2+6wkByfZPMk2q+h/VR6Z5KIkWyb5ZJJPJ3l4BjO//5hBkJ41tP2cJG/M4BhfkOSU7jhs0dXx7q6tdyQ5o6q2HNr32UmOSLJpkrlJ/iu3XWJ+ZLfNDzM4t1t09ZxWVcPB6JCuxs0zCFujAX9Gki8n+e8kO2Rw/j/drfu7JP+a5OlJtu76/dTYA9Fau6mb9U6SPVprD6iqDZN8KclZSf46yUuSnDJmTP1Dkjd3n2ul94d3x+PpSX4+tHjcMTK0/l5JNus+0z8neV9V3bNb974kNya5d5LndX9G+5rIOXlmBuflPhmE/O8l+UgGx/+SJK9d2edZgU8luTLJtkmekeTfq+rxQ+sPTbIgg3N4SpKPJlmewZjbK8kBSUZviXhjBsf+nkm2S/KeJGmt/U23fo9u/Jy6kno+lsEs+ujn/UKSm0ZXVtWDuppflsH4+EqSL1XV3arqbhl8sfLxDI7JaUn+fmjfhyX5cJIXZnCM/zPJF6v78hBgOhLUAZiMOUne0Fr7fWvt6iSvzyBoDPt/Xeg6J4Og8n9W0Na7kzwwgwD2/5LMr6rHrKTvbbtZs9E//yeDMPuV1tpXWmt/aa19PcmPkhyUJK21M1prv2gD52QQOh634i4m5N2ttV+11pauqv8JuKK19pHukv9Tk9w3g+N7U2vtrCQ3ZxCgRp3RWju3tXZTBl+aPKqq7pvBFweXt9Y+3lpb3lr7VJJLkwzfHjC/tXZxt37ZeMW01j7RWrum2+btSe6eZDgUf7v7rLdkEJ5GL7d+RAbB8JWttRtaaze21kZD8wuT/Edr7ZLW2vIk/55kz/Fm1cexbwaXqh/XWru5tfatDL4QeNbQNl9orX2nO/43rqCdd1fVdUn+kMGXHC8Z+syrGiPLMjgny1prX0myJMmDuy8n/j7Jsd1n/kkGoXfURM7JR7q+r8vgCo9ftNa+0R2n0zIIzivzh6Gfh1d0Y+GxSV7VnYMLknwwt/8Z/V5r7fOttb9k8OXEk5O8rPsMv0/yzgwC9ehnv1+Sbcec09VxepLZVbVZBoH9Y2PWH57BuP56Ny7flmTjDGbe902yYZITuuO/IIMvk0a9IMl/ttZ+0Fq7pbX20Qy+BNh3EnUC9IKgDsBkbJvBrOmo/+6WjfpjN1u9ovW3aq2dPxQKv5LB7N7TV9L3Va21zYf+fCaDEHHYcIDPIKjcO0mq6slV9f1u1v5PGQTorVbzM4/1q6HXK+1/An439HppkrTWxi4bnlG/te/W2pIk12ZwfMeel3Tv77OCusfVXeJ9SXep858ymEkePl6/HXr9v0k2qsH9x/dN8t9dwBzrfkneNXR8rk1SY2pbkW2T/KoLlZP+XEmOaq1tlmT33DY7nGRCY+SaMZ/rfzM4J1snmTmm/+FzMJFzMvZcr+zcj2eroZ+Ht3V9Xttau34lfY4dvxsm+c3Q+fnPDL48S5L/m8G5Oq+7NP15WU3dF1pnZHDlzVatte+M2eR2x6k717/qat42ya9ba23M5xmu/+gxP3/3zQr+zQGYDqblA3AAmHJX5fYP+9q+WzbqnlW1yVBY3z7JTybYdssgFKyOXyX5eGvtBWNXdJe/fjaDWbwvtNaWVdXnh/poY/dJckOSvxp6f68V1LnK/u8k9x190V0Sv0UGx3/0vAzbPslXh96P/by3e1+D+9FfleTxSS5urf2lqv6YiZ2TXyXZvqpmjhPWf5XkzW1yDyG8Ksl9q2qDobC+fW7/MLLxzuO4WmuLavAMgvd1l03fLSsfIytzdQaXjN83g5ny0dqGa1/VOVnbrkqyRVVtOhTWt0/y66Ftxo7fmzII0Hf4kqW19tsMZq1TVY9N8o2qOre19vOx267Cx5J8K4MrcMarebfRN1VVGRzTX3e13qeqaiisb5/BQ+pG639za+3Nq1kPQG+ZUQdgMj6V5DVVtXVVbZXk2CSfGLPN67v7Sx+Xwb2/p43XUFU9o6pmVdUGVXVABpeRf3G8bVfiE0meWlUHVtWMqtqoBg8A2y6DEHb3dIGqqp6cwf23o36XZMvuktxRFyQ5qKq2qKp7ZXDf7GT7vzMcVFWP7e7dfWOSH7TWfpXBfb0Pqqp/qMED7g5P8tAMLhNfkd8luf/Q+00zCJ5XJ5lZVcdmcGn0RJyX5DdJjquqTbrjMHobw/uTvLqqdkmSGjy87LAJtvuDDL48+b9VtWENHtT31HT3v0/SRzOYMT4kqx4jK9Rd/v+5JK+rqr+qqodm8FyGUZM5J2ukGwvfTfIf3TnYPYP76sf9kqS19psMLvV/e1Xdo/tZfEBV7ZckVXXY0Fj+YwbBefQ3M4wdPytzTpInprvHfYzPJDm4qh7fPZPg6Ay+PPhuBvfsL09yVHcMn57BbRajPpBkXlU9sgY2qaqDq2rTCdYF0DuCOgCT8aYM7sG+KIOnOZ/fLRv12wz+h/6qDMLBvNbapWMb6bw0g1mzPyU5PskLWmsjq1NMF0wOzeBhZVdnMMP2yiQbdDOKR2UQBP6YwUPHvji076UZfPHwy+6y2W0zuO/6wiSLMwgwK3tI1kr7X53PsRo+mcEDxq5NsncGzwxIa+2aDL4UOTrJNRlcsvyU1tofVtLWu5I8owZPYn93kq9lcJ/0zzK4vPjGTOyy8tHQ+tQM7qf/nwweZnZ4t+70JG9J8ukaPGn8JxncFz2Rdm/OIFA/OYP7y09M8pyVjKmJtvnuDJ6lsNIxMgFHZnB5+m+TzM/gQXCj/UzmnKwNz8rggX5XZXB/+Gu7ZyesyHMy+MLipxkcgwW57daNhyf5QQ2exv/FJC9trV3RrXtdko8OPS9ihbr7/7/ZWrt2nHWXZfAl3XsyOMdPzeDXQd7cnaunZ/Dwwz9mMKY+N7TvjzKY8X9vt/7n3bYA01bd/nYfAFgz3WznJ1prd9Zs8nqtquYnubK19ppVbQsATE9m1AEAAKBHBHUAAADoEZe+AwAAQI+YUQcAAIAeEdQBAACgR2ZOdQHc0eabb9522mmnqS4DJu2GG27IJptsMtVlwKQYv0x3xjDTmfHLdDfRMbxw4cI/tNa2XtF6Qb2Httlmm/zoRz+a6jJg0kZGRjJ79uypLgMmxfhlujOGmc6MX6a7iY7hqvrvla136TsAAAD0iKAOAAAAPSKoAwAAQI+4Rx0AAGCaWbZsWa688srceOONU10KQzbbbLNccsklt77faKONst1222XDDTdcrXYEdQAAgGnmyiuvzKabbpoddtghVTXV5dC5/vrrs+mmmyZJWmu55pprcuWVV2bHHXdcrXZc+g4AADDN3Hjjjdlyyy2F9B6rqmy55ZaTuupBUAcAAJiGhPT+m+w5EtQBAACgR9yjDgAAMM3tcMwZa7W9xccdvNr7nHbaaTn22GNzr3vdK2efffZarWfYpZdemmc+85mpqixYsCBnnHFGTjrppDzsYQ/L4Ycfnp/+9Kc55phjVrj/ox/96Hz3u9+dVN/z58/PAQcckG233Xay5U+IoA4AAMAaaa3lAx/4QE488cTsv//+d2pfn//853PooYfm9a9/fZLkxBNPzJlnnnnrA9sOOeSQle4/2ZCeDIL6rrvueqcHdZe+AwAAsNoWL16cnXfeOS960YuywQYb5Otf/3rmzZuXV77yleNuf8stt+QVr3hFdtttt+y+++55z3vekyT55je/mb322iu77bZbnve85+Wmm25KkixcuDD77bdf9t577xx44IH5zW9+k6985Ss54YQT8sEPfjD7779/5s2bl1/+8pc55JBD8s53vjPz58/PkUcemST53e9+l6c97WnZY489sscee9wa0GfNmnVrTccff3we/vCHZ/fdd89rX/va232uF7zgBdlll11ywAEHZOnSpVmwYEF+9KMfZc6cOdlzzz2zdOnSHHPMMXnoQx+a3XffPa94xSvW2rEV1AEAAJiUyy67LM95znPSWst+++2XU045Jccff/y425588sm54oor8uMf/zgXXXRR5syZkxtvvDFz587NqaeemkWLFmX58uU56aSTsmzZsrzkJS/JggULsnDhwjzvec/Lv/3bv+Wggw7KvHnz8vKXvzxnn3123v/+92fbbbfN2WefnZe//OW36++oo47KfvvtlwsvvDDnn39+dtlll9utP+uss3L55ZfnvPPOywUXXJCFCxfm3HPPTZJcfvnlefGLX5yLL744m2++eT772c/mGc94RvbZZ5+ccsopueCCC7J06dKcfvrpufjii3PRRRflNa95zVo7ri59BwAAYFLud7/7Zd99953Qtt/4xjcyb968zJw5iKFbbLFFLrzwwuy444550IMelCR57nOfm/e97315whOekJ/85Cd54hOfmGQwG3/ve997tWr71re+lY997GNJkhkzZmSzzTa73fqzzjorZ511Vvbaa68kyZIlS3L55Zdn++23z4477pg999wzSbL33ntn8eLFd2j/Hve4RzbaaKM8//nPz8EHH5ynPOUpt14NsKYEdQAAACZlk002mfC2rbU7/Lqy1toKt91ll13yve99b43qW1U9r371q/PCF77wdssXL16cu9/97re+nzFjRpYuXXqH/WfOnJnzzjsv3/zmN/PpT386733ve/OFL3xhrdTm0ncAAADudAcccEDe//73Z/ny5UmSa6+9Ng95yEOyePHi/PznP0+SfPzjH89+++2XBz/4wbn66qtvDerLli3LxRdfvFr9Pf7xj89JJ52UZDAj/+c///l26w888MB8+MMfzpIlS5Ikv/71r/P73/9+pW1uuummuf7665MMZuCvu+66HHTQQTnhhBNywQUXrFZ9K2NGHQAAYJqbzK9TW9ee//zn52c/+1l23333bLjhhnnBC16QI488Mh/5yEdy2GGHZfny5Xn4wx+eefPm5W53u1sWLFiQo446Ktddd12WL1+el73sZXe4z3xl3vWud+WII47Ihz70ocyYMSMnnXRSHvWoR926/oADDsgll1xy67JZs2blE5/4RGbMmLHCNufOnZt58+Zl4403zplnnplDDz00N954Y1preec73zn5gzNGrehSA6bOgx/84HbZZZdNdRkwaSMjI5k9e/ZUlwGTYvwy3RnDTGfG78Rdcskl2Xnnnae6DMa4/vrrs+mmm95u2XjnqqoWttb2WVE7Ln0HAACAHnHpOwAAAGvN1772tbzqVa+63bIdd9wxp59++hRVNP0I6gAAAKw1Bx54YA488MCpLmNaE9R7aOmyW7LDMWdMdRkwaUfvtjxzjWGmKeOX6c4Yvs10eLgWrInxft0Z/TLZZ8K5Rx0AAGCa2WijjXLNNddMOghy52ut5ZprrslGG2202vuaUQcAAJhmtttuu1x55ZW5+uqrp7oUhtx44423C+YbbbRRtttuu9VuR1AHAACYZjbccMPsuOOOU10GY4yMjGSvvfZa43Zc+g4AAAA9IqgDAABAjwjqAAAA0COCOgAAAPSIoA4AAAA9IqgDAABAjwjqAAAA0COCOgAAAPSIoA4AAAA9IqgDAABAjwjqAAAA0COCOgAAAPSIoA4AAAA9IqgDAABAj6wyqFdVq6qPD72fWVVXV9WXV7Hf5lX1oqH3O3RtvXFo2VZVtayq3juZ4qtqz6o6aOj93K62C6rqp1X1gsm0O1TvTya57+yqevRk+wYAAGD9NZEZ9RuS7FpVG3fvn5jk1xPYb/MkLxqz7JdJnjL0/rAkF0+grRXZM8lBY5ad2lrbM8nsJP9eVdsMr6yqmWvQ30TNTiKoAwAAsNomeun7mUkO7l4/K8mnRldU1euq6sNVNVJVv6yqo7pVxyV5QDe7fXy3bGmSS6pqn+794Uk+M9TW1lX12ar6YffnMd3yR1TVd6vqx93fD66quyV5Q5LDuz4OHy64tfb7JL9Icr+qml9V76iqs5O8pZuJ/35VXVRVp1fVPbt+9q6qC6vqe0lePFTX3OFZ/6r6clXN7l4/qarO7/b7ZlXtkGRekpd3dT2uqg6rqp9025w7wWMOAADAemiis8ufTnJsd7n77kk+nORxQ+sfkmT/JJsmuayqTkpyTJJdu9ntdAF2tK1nVtVvk9yS5Kok23br3pXkna21b1fV9km+lmTnJJcm+ZvW2vKqekKSf2+t/X1VHZtkn9bakV0fc0cLqqr7J7l/kp93ix6U5AmttVuq6qIkL2mtnVNVb0jy2iQvS/KRoeWjXy6sUFVtneQDXW1XVNUWrbVrq+r9SZa01t7WbbcoyYGttV9X1eYraOuIJEckyVZbbZ1jd1u+qu6ht7bZODnaGGaaMn6Z7ozh24yMjEx1CaymJUuWOG9Ma2trDE8oqLfWLuqC9rOSfGWcTc5ord2U5Kaq+n2SbcbZZtRXk7wxye+SnDpm3ROSPLSqRt/fo6o2TbJZko9W1QOTtCQbrqT9w6vqsUluSvLCLjgnyWldSN8syeattXO67T+a5LRxln88yZNX0k+S7Jvk3NbaFUnSWrt2Bdt9J8n8qvpMks+Nt0Fr7eQkJyfJ9vffqb190bq4Qh/uHEfvtjzGMNOV8ct0ZwzfZvGc2VNdAqtpZGQks2fPnuoyYNLW1hhenX/Fv5jkbRncf73lmHU3Db2+ZWXtttZurqqFSY5OskuSpw6t3iDJo1prS4f3qar3JDm7tfa07guDkZXUeeroDPsYN6xknySpDL4EGM/y3P42gY0msM+tWmvzquqRGdw+cEFV7dlau2ZV+wEAALD+WZ1fz/bhJG9orS2a4PbXZ3Ap/HjenuRV44TVs5LcGrKras/u5Wa57QF2cyfYx7haa9cl+WNVjV66/+wk57TW/pTkum42PknmDO22OMmeVbVBVd03ySO65d9Lsl9V7djVu8V4dVXVA1prP2itHZvkD0nuuzo1AwAAsP6YcFBvrV3ZWnvXamx/TZLvdA9RO37Muotbax8dZ7ejkuzTPeTtpxk8lC1J3prkP6rqO0lmDG1/dgaXyt/hYXKr8Nwkx3f3qu+ZwUPpkuSfkryve5jc8Kz+d5JckWRRBlcVnN99jqszuK/8c1V1YW67lP9LSZ42+jC5rq9F3a97OzfJhatRKwAAAOuRVV763lqbNc6ykXSXn7fWXjdm3a5Dr/9hzK67jnmf1tr8JPO713/I4EnwY7f5XgYPgxv1/7rl1yZ5+JjN54+z/9wx7y/I4P7ysdstTLLH0KLXdctbbj/DPrzPmRk8FX942c8yeOjeqP8ab18AAAAYa3UufQcAAADuZII6AAAA9IigDgAAAD0iqAMAAECPCOoAAADQI4I6AAAA9IigDgAAAD0iqAMAAECPCOoAAADQI4I6AAAA9IigDgAAAD0iqAMAAECPCOoAAADQI4I6AAAA9IigDgAAAD0iqAMAAECPzJzqArijjTeckcuOO3iqy4BJGxkZyeI5s6e6DJgU45fpzhgGmP7MqAMAAECPCOoAAADQI4I6AAAA9IigDgAAAD0iqAMAAECPCOoAAADQI4I6AAAA9IigDgAAAD0iqAMAAECPCOoAAADQI4I6AAAA9IigDgAAAD0iqAMAAECPCOoAAADQI4I6AAAA9IigDgAAAD0iqAMAAECPCOoAAADQI4I6AAAA9IigDgAAAD0iqAMAAECPCOoAAADQI4I6AAAA9IigDgAAAD0iqAMAAECPCOoAAADQI4I6AAAA9IigDgAAAD0iqAMAAECPCOoAAADQI4I6AAAA9IigDgAAAD0iqAMAAECPCOoAAADQI4I6AAAA9IigDgAAAD0iqAMAAECPCOoAAADQI4I6AAAA9IigDgAAAD0iqAMAAECPCOoAAADQI4I6AAAA9MjMqS6AO1q67JbscMwZU10GTNrRuy3PXGOYacr4Zbozhlds8XEHT3UJABNiRh0AAAB6RFAHAACAHhHUAQAAoEcEdQAAAOgRQR0AAAB6RFAHAACAHhHUAQAAoEcEdQAAAOgRQR0AAAB6RFAHAACAHhHUAQAAoEcEdQAAAOgRQR0AAAB6RFAHAACAHhHUAQAAoEcEdQAAAOgRQR0AAAB6ZJ0F9apqVfXxofczq+rqqvryKvbbvKpeNGbZA6vqy1X1i6paWFVnV9XfrKKduVX13hWsW9L9vUNVLa2qC6rqp1X1/qqa9DGqqsVVtdVk9wcAAGD9sy5n1G9IsmtVbdy9f2KSX09gv82T3BrUq2qjJGckObm19oDW2t5JXpLk/mupzl+01vZMsnuShyb5u+GVVTVzLfUDAAAAd7CuQ+eZSQ5OsiDJs5J8KsnjkqSqXpdk+wwC9/ZJTmitvTvJcUkeUFUXJPl6kkuTfK+19sXRRltrP0nyk66dLZJ8uGvnf5Mc0Vq7aLiIqtoxyScz+PxfHa/Q1tryqvpukp2qam5X90ZJNqmqZ4zXR1Vt2X2mrZOcl6S6/jZJ8pkk2yWZkeSNrbVTV//wAQAAcFe3roP6p5Mc213uvnsGYfdxQ+sfkmT/JJsmuayqTkpyTJJdu1nuVNU7kpy/kj5en+THrbW/q6q/TfKxJHuO2eZdSU5qrX2sql48XiNV9VdJHp/k2CTbJHlUkt1ba9dW1XtW0Mdrk3y7tfaGqjo4yRFdc09KclVr7eCu7c3G6e+I0e232mrrHLvb8pV8ROi3bTZOjjaGmaaMX6Y7Y3jFRkZGproEVmHJkiXOE9Pa2hrD6zSod7POO2Qwm/6VcTY5o7V2U5Kbqur3GQTklaqq05M8MMnPWmtPT/LYJH/f9fetqtpynGD8mNFtknw8yVuG1o3O3rckX2itndnNqH+9tXZtt82K+vibJE/vlp9RVX/stl+U5G1V9ZYkX26t/dc4x+bkJCcnyfb336m9fZEr7Jm+jt5teYxhpivjl+nOGF6xxXNmT3UJrMLIyEhmz5491WXApK2tMTwV/4p/McnbksxOsuWYdTcNvb4l49d3cQaBOEnSWntaVe3TtZl0l5uP0Sa4LLntHvWxbhh6vbI+7tBua+1nVbV3koOS/EdVndVae8MK+gcAAGA9NhW/nu3DSd7QWls0we2vz+BS+FGfTPKYqjpkaNlfDb0+N8mcJKmq2Un+0Fr785g2v5Pkmd3rOROsY9iK+hhe/uQk9+xeb5vkf1trn8jgC4WHTaJPAAAA1gPrfEa9tXZlBveIT3T7a6rqO1X1kyRnttZeWVVPSfKOqjohye8yCPNv6nZ5XZKPVNVFGTzo7bnjNPvSJJ+sqpcm+ewkPsaK+nh9kk9V1flJzknyP93y3ZIcX1V/SbIsyb9Mok8AAADWA+ssqLfWZo2zbCTJSPf6dWPW7Tr0+h/GrLs0g8vIx+vn2iSHjrN8fpL53esrMng43KjjuuWLk+w6Ztfb7buKPq5JcsDQopd3f3+t+wMAAAArNRWXvgMAAAArIKgDAABAjwjqAAAA0COCOgAAAPSIoA4AAAA9IqgDAABAjwjqAAAA0COCOgAAAPSIoA4AAAA9IqgDAABAjwjqAAAA0COCOgAAAPSIoA4AAAA9IqgDAABAjwjqAAAA0COCOgAAAPTIzKkugDvaeMMZuey4g6e6DJi0kZGRLJ4ze6rLgEkxfpnujGGA6c+MOgAAAPSIoA4AAAA9IqgDAABAjwjqAAAA0COCOgAAAPSIoA4AAAA9IqgDAABAjwjqAAAA0COCOgAAAPSIoA4AAAA9IqgDAABAjwjqAAAA0COCOo9U7HoAABkqSURBVAAAAPSIoA4AAAA9IqgDAABAjwjqAAAA0COCOgAAAPSIoA4AAAA9IqgDAABAjwjqAAAA0COCOgAAAPSIoA4AAAA9IqgDAABAjwjqAAAA0COCOgAAAPSIoA4AAAA9IqgDAABAjwjqAAAA0COCOgAAAPSIoA4AAAA9IqgDAABAjwjqAAAA0COCOgAAAPSIoA4AAAA9IqgDAABAjwjqAAAA0COCOgAAAPSIoA4AAAA9IqgDAABAjwjqAAAA0COCOgAAAPSIoA4AAAA9IqgDAABAj8yc6gK4o6XLbskOx5wx1WXApB292/LMNYaZpoxfpjtjeO1afNzBU10CsB4yow4AAAA9IqgDAABAjwjqAAAA0COCOgAAAPSIoA4AAAA9IqgDAABAjwjqAAAA0COCOgAAAPSIoA4AAAA9IqgDAABAjwjqAAAA0COCOgAAAPSIoA4AAAA9IqgDAABAjwjqAAAA0COCOgAAAPSIoA4AAAA9st4E9apaMtU1AAAAwKqsN0EdAAAApoP1LqhX1eyqGqmqBVV1aVWdUlXVrXt4VX23qi6sqvOqatOq2qiqPlJVi6rqx1W1f7ft3Kr6fFV9qaquqKojq+r/67b5flVt0W33gKr6alUtrKr/qqqHTOXnBwAAoN9mTnUBU2SvJLskuSrJd5I8pqrOS3JqksNbaz+sqnskWZrkpUnSWtutC9lnVdWDunZ27draKMnPk7yqtbZXVb0zyXOSnJDk5CTzWmuXV9Ujk5yY5G/HFlRVRyQ5Ikm22mrrHLvb8jvpo8Odb5uNk6ONYaYp45fpzhheu0ZGRqa6hPXKkiVLHHOmtbU1htfXoH5ea+3KJKmqC5LskOS6JL9prf0wSVprf+7WPzbJe7pll1bVfycZDepnt9auT3J9VV2X5Evd8kVJdq+qWUkeneS0btI+Se4+XkGttZMzCPXZ/v47tbcvWl9PDXcFR++2PMYw05Xxy3RnDK9di+fMnuoS1isjIyOZPXv2VJcBk7a2xvD6+q/4TUOvb8ngOFSSNs62Nc6y8dr5y9D7v3RtbpDkT621PSdfKgAAAOuT9e4e9ZW4NMm2VfXwJOnuT5+Z5Nwkc7plD0qyfZLLJtJgNyt/RVUd1u1fVbXHnVE8AAAAdw2Ceqe1dnOSw5O8p6ouTPL1DO49PzHJjKpalME97HNbazetuKU7mJPkn7s2L05y6NqtHAAAgLuS9ebS99barO7vkSQjQ8uPHHr9wyT7jrP73HHam59k/tD7HcZb11q7IsmTJl85AAAA6xMz6gAAANAjgjoAAAD0iKAOAAAAPSKoAwAAQI8I6gAAANAjgjoAAAD0iKAOAAAAPSKoAwAAQI8I6gAAANAjgjoAAAD0iKAOAAAAPSKoAwAAQI8I6gAAANAjgjoAAAD0iKAOAAAAPSKoAwAAQI/MnOoCuKONN5yRy447eKrLgEkbGRnJ4jmzp7oMmBTjl+nOGAaY/syoAwAAQI8I6gAAANAjgjoAAAD0iKAOAAAAPSKoAwAAQI8I6gAAANAjgjoAAAD0iKAOAAAAPSKoAwAAQI8I6gAAANAjgjoAAAD0iKAOAAAAPSKoAwAAQI8I6gAAANAjgjoAAAD0iKAOAAAAPSKoAwAAQI8I6gAAANAjgjoAAAD0iKAOAAAAPSKoAwAAQI8I6gAAANAjgjoAAAD0iKAOAAAAPSKoAwAAQI8I6gAAANAjgjoAAAD0iKAOAAAAPSKoAwAAQI8I6gAAANAjgjoAAAD0iKAOAAAAPSKoAwAAQI8I6gAAANAjgjoAAAD0iKAOAAAAPSKoAwAAQI8I6gAAANAjgjoAAAD0iKAOAAAAPSKoAwAAQI8I6gAAANAjgjoAAAD0yMypLoA7WrrsluxwzBlTXQZM2tG7Lc9cY5hpyvhlujOGmc6m4/hdfNzBU10Cd0Fm1AEAAKBHBHUAAADoEUEdAAAAekRQBwAAgB4R1AEAAKBHBHUAAADoEUEdAAAAekRQBwAAgB4R1AEAAKBHBHUAAADoEUEdAAAAekRQBwAAgB4R1AEAAKBHBHUAAADoEUEdAAAAekRQBwAAgB4R1AEAAKBHZk51AXeGqrolyaIMPt8VSZ7dWvvT1FYFAAAAq3ZXnVFf2lrbs7W2a5Jrk7x4qgsCAACAibirBvVh30tynySpqgdU1VeramFV/VdVPaRbvk1VnV5VF3Z/Ht0t/8eqOq+qLqiq/6yqGd3yJVX15m7b71fVNitqp6reWFUvHS2m2++odX4UAAAAmBaqtTbVNax1VbWktTarC9afTvKh1tpXq+qbSea11i6vqkcm+Y/W2t9W1alJvtdaO6HbZ1aSbZO8NcnTW2vLqurEJN9vrX2sqlqSQ1prX6qqtyb5c2vtTSto555JPtdae1hVbZDk8iSPaK1dM6bmI5IckSRbbbX13see8IF1cKTgzrHNxsnvlk51FTA5xi/TnTHMdDYdx+9u99lsqkugR5YsWZJZs2atcrv9999/YWttnxWtv0veo55k46q6IMkOSRYm+XpVzUry6CSnVdXodnfv/v7bJM9JktbaLUmuq6pnJ9k7yQ+77TdO8vtu+5uTfLl7vTDJE1fUTtfWNVW1V5Jtkvx4bEjvtj85yclJsv39d2pvX3RXPTWsD47ebXmMYaYr45fpzhhmOpuO43fxnNlTXQI9MjIyktmzZ69xO9Prp2DilrbW9qyqzTII1C9OMj/Jn1pre06wjUry0dbaq8dZt6zddinCLVn1cfxgkrlJ7pXkwxPsHwAAgPXQXfoe9dbadUmOSvKKJEuTXFFVhyVJDezRbfrNJP/SLZ9RVffolj2jqv66W75FVd1vFV2O106SnJ7kSUkenuRra+vzAQAAcNdzlw7qSdJa+3GSC5M8M8mcJP9cVRcmuTjJod1mL02yf1UtyuBS9l1aaz9N8pokZ1XVRUm+nuTeq+juDu10Ndyc5Owkn+kuiQcAAIBx3SUvfW+tzRrz/qlDb580zva/y22hfXj5qUlOXVn7rbUFSRasrJ3uIXL7Jjlswh8CAACA9dJdfkZ9qlXVQ5P8PMk3W2uXT3U9AAAA9Ntdcka9T7pL6O8/1XUAAAAwPZhRBwAAgB4R1AEAAKBHBHUAAADoEUEdAAAAekRQBwAAgB4R1AEAAKBHBHUAAADoEUEdAAAAekRQBwAAgB4R1AEAAKBHBHUAAADoEUEdAAAAekRQBwAAgB4R1AEAAKBHZk51AdzRxhvOyGXHHTzVZcCkjYyMZPGc2VNdBkyK8ct0ZwwznRm/MGBGHQAAAHpEUAcAAIAeEdQBAACgRwR1AAAA6BFBHQAAAHpEUAcAAIAeEdQBAACgRwR1AAAA6BFBHQAAAHpEUAcAAIAeEdQBAACgRwR1AAAA6BFBHQAAAHpEUAcAAIAeEdQBAACgRwR1AAAA6BFBHQAAAHpEUAcAAIAeEdQBAACgRwR1AAAA6BFBHQAAAHpEUAcAAIAeEdQBAACgRwR1AAAA6BFBHQAAAHpEUAcAAIAeEdQBAACgRwR1AAAA6BFBHQAAAHpEUAcAAIAeEdQBAACgRwR1AAAA6BFBHQAAAHpEUAcAAIAeEdQBAACgRwR1AAAA6BFBHQAAAHpEUAcAAIAeEdQBAACgRwR1AAAA6BFBHQAAAHpEUAcAAIAeEdQBAACgR2ZOdQHc0dJlt2SHY86Y6jJg0o7ebXnmGsNMU8Yv050xzHRm/LImFh938FSXsNaYUQcAAIAeEdQBAACgRwR1AAAA6BFBHQAAAHpEUAcAAIAeEdQBAACgRwR1AAAA6BFBHQAAAHpEUAcAAIAeEdQBAACgRwR1AAAA6BFBHQAAAHpEUAcAAIAeEdQBAACgRwR1AAAA6BFBHQAAAHpEUAcAAIAe6VVQr6r5VfWM7vVIVe1zJ/f3sqr6q1Vs869j3n/3zqwJAACA9ds6D+pVNXNd97kSL0uy0qCe5HZBvbX26DuvHAAAANZ3kw7qVfWPVXVeVV1QVf9ZVTOqaklVvbmqLqyq71fVNt2286vqHVV1dpK3VNWe3fqLqur0qrrnKvpaUlVvqaqFVfWNqnpEN+P+y6o6pNtmRlUdX1U/7Np9Ybd8drftgqq6tKpOqYGjkmyb5OyurvH6PS7Jxt1nPGW0lqF2z6mqz1TVz6rquKqa0x2TRVX1gG67ravqs11dP6yqx0z2mAMAAHDXV6211d+pauckb03y9Nbasqo6Mcn3k3w0ySGttS9V1VuT/Lm19qaqmp9kqySHttZuqaqLkryktXZOVb0hyT1aay/rtvtya21BVY0keUVr7UdV1ZIc1Fo7s6pOT7JJkoOTPDTJR1tre1bVEUn+uuvv7km+k+SwJPdL8oUkuyS5qlv+ytbat6tqcZJ9Wmt/WMlnXdJamzX2fVXNTvL5JDsnuTbJL5N8sLX22qp6aZIdu8/0ySQndv1tn+RrrbWdx+nniCRHJMlWW22997EnfGC1zgn0yTYbJ79bOtVVwOQYv0x3xjDTmfHLmtjtPptNdQlZsmRJZs2atcrt9t9//4WttRXe6j3Zy9Afn2TvJD+sqiTZOMnvk9yc5MvdNguTPHFon9O6kL5Zks1ba+d0yz+a5LRV9Hdzkq92rxcluan7gmBRkh265Qck2X30HvckmyV5YLfvea21K5Okqi7o9vn26nzgFfhha+03Xbu/SHLWUI37d6+fkOSh3XFKkntU1aatteuHG2qtnZzk5CTZ/v47tbcv6tMdArB6jt5teYxhpivjl+nOGGY6M35ZE4vnzJ7qEjIyMpLZs9e8jsn+FFQGM9mvvt3Cqle026bobxnT/g2T7CtJlg21+5ckNyVJa+0vQ/e8Vwaz9F8bU9Ps0e1XUNeaGG73L0Pv/zLUxwZJHtVa890gAAAAqzTZe9S/meQZVfXXSVJVW1TV/SayY2vtuiR/rKrHdYueneSclewyUV9L8i9VtWFX04OqapNV7HN9kk1Xsc2y0TYn6awkR46+qao916AtAAAA7uImNbPcWvtpVb0myVlVtUGSZUlevBpNPDfJ+7tfjfbLJP80mTrG+GAGl7SfX4PrzK9O8ner2OfkJGdW1W9aa/uvZJuLqur81tqcSdR1VJL3dfflz0xybpJ5k2gHAACA9cCkLwFvrZ2a5NQxi2cNrV+QZEH3eu6YfS9Isu84bc4dej176PVwu68bs8+s7u+/ZPCr1G7369SSjHR/Rrc/cuj1e5K85w4f7vbtvyrJq8bpb2y7w/Xeuq57UN3hK+sDAAAARq3z36MOAAAArJhHKnaq6gdJ7j5m8bNba4umoh4AAADWT4J6p7X2yKmuAQAAAFz6DgAAAD0iqAMAAECPCOoAAADQI4I6AAAA9IigDgAAAD0iqAMAAECPCOoAAADQI4I6AAAA9IigDgAAAD0iqAMAAECPCOoAAADQI4I6AAAA9IigDgAAAD0yc6oL4I423nBGLjvu4KkuAyZtZGQki+fMnuoyYFKMX6Y7Y5jpzPiFATPqAAAA0COCOgAAAPSIoA4AAAA9IqgDAABAjwjqAAAA0COCOgAAAPSIoA4AAAA9IqgDAABAjwjqAAAA0COCOgAAAPSIoA4AAAA9IqgDAABAjwjqAAAA0COCOgAAAPSIoA4AAAA9IqgDAABAjwjqAAAA0COCOgAAAPSIoA4AAAA9IqgDAABAjwjqAAAA0CPVWpvqGhijqq5PctlU1wFrYKskf5jqImCSjF+mO2OY6cz4Zbqb6Bi+X2tt6xWtnLn26mEtuqy1ts9UFwGTVVU/MoaZroxfpjtjmOnM+GW6W1tj2KXvAAAA0COCOgAAAPSIoN5PJ091AbCGjGGmM+OX6c4YZjozfpnu1soY9jA5AAAA6BEz6gAAANAjgvo6VFVPqqrLqurnVXXMOOurqt7drb+oqh420X1hXZjsGK6q+1bV2VV1SVVdXFUvXffVs75bk3+Du/UzqurHVfXldVc13GYN/z9i86paUFWXdv8WP2rdVg9rPIZf3v0/xE+q6lNVtdG6rZ713QTG70Oq6ntVdVNVvWJ19h2PoL6OVNWMJO9L8uQkD03yrKp66JjNnpzkgd2fI5KctBr7wp1qTcZwkuVJjm6t7Zxk3yQvNoZZl9Zw/I56aZJL7uRSYVxrYQy/K8lXW2sPSbJHjGXWsTX8f+H7JDkqyT6ttV2TzEjyzHVUOkx0/F6bwTh92yT2vQNBfd15RJKft9Z+2Vq7Ocmnkxw6ZptDk3ysDXw/yeZVde8J7gt3tkmP4dbab1pr5ydJa+36DP4H8T7rsnjWe2vyb3CqarskByf54LosGoZMegxX1T2S/E2SDyVJa+3m1v7/9u4/VM+yjuP4+0Nb6Gb+yqAfVjMRYYlsOUSXmKhESOjCaH9UtpSiQEHCiorAsqwsSvCPRMN+kFSkQ6ZI05xLyFpTN3/NJGnijxYurDnMzLlvfzzXsePxnOM5z32eZ8/m+wWH8zz3fX3v63sdLh7u77nu+37qX8NMXqLj5zAwD9g/yTxgAfC3YSUuMYP5W1VPVdVG4IXZxk7GQn143gY8Pu79E7yyUJmqzUxipUHrModfkmQRsBTYMOcZSlPrOn8vB74A7B5UgtKr6DKH3wVsB37cbt/4UZKFg0xWmkTfc7iqnqS3SvkYsA3YUVW3DDBXaaIu9VhfsRbqw5NJtk185P5UbWYSKw1alznc25kcAFwPXFhVz8xhbtKr6Xv+Jvkg8FRV3T33aUkz1uUzeB7wHuCHVbUUeBbweTcati6fw4fQW4E8AngrsDDJx+Y4P2k6XeqxvmIt1IfnCeDt494fzisv2ZmqzUxipUHrModJMp9ekX5tVa0eYJ7SZLrM3/cCZyZ5lN7laqcm+fngUpUm1fU84omqGruS6Tp6hbs0TF3m8OnA1qraXlUvAKuB5QPMVZqoSz3WV6yF+vBsBI5KckSS19N7AMaaCW3WAOe0J16eQO+ynm0zjJUGre85nCT07o18qKq+P9y0JaDD/K2qL1XV4VW1qMWtqypXcjRsXebw34HHkxzd2p0GbBla5lJPl3Phx4ATkixo5xSn4QMRNVxd6rG+Yuf1napmpap2JTkfWEvvSZXXVNWDST7T9l8J3AycATwC/Bv45HSxe2AYeg3rMofprUh+HLg/yea27ctVdfMwx6DXro7zV9rj5mAOXwBc204S/4rzW0PW8Vx4Q5LrgHvofZPMJuCq4Y9Cr1Uzmb9J3gzcBRwI7E5yIbC4qp7pp5ZLlbc6S5IkSZI0Krz0XZIkSZKkEWKhLkmSJEnSCLFQlyRJkiRphFioS5IkSZI0QizUJUmSJEkaIRbqkiTt5ZK8mGTzuJ9FfRxjRZLFc58dJFmU5IFBHHuaPpckOWOYfUqSNFf8HnVJkvZ+z1XVko7HWAHcBGyZaUCSeVW1q2O/cy7JPGAJsIze9zJLkrRXcUVdkqR9UJLjkvwuyd1J1iZ5S9v+qSQbk9yb5PokC5IsB84EvttW5I9Msj7JshZzWJJH2+tVSX6d5EbgliQLk1zTjrkpyVmvkteqJDckuTHJ1iTnJ/lci/1jkkNbu/VJLk9yZ5IHkhzfth/a4u9r7Y9t2y9OclWSW4CfAV8HVrbxrExyfDvWpvb76HH5rE7ymyR/SXLZuFw/kOSe9re6rW2b1XglSeqHK+qSJO399k+yub3eCnwEuAI4q6q2J1kJfBM4F1hdVVcDJPkGcF5VXZFkDXBTVV3X9k3X34nAsVX1dJJLgXVVdW6Sg4E/JfltVT07TfwxwFJgP+AR4ItVtTTJD4BzgMtbu4VVtTzJycA1Le5rwKaqWpHkVHpF+djVBMcBJ1XVc0lWAcuq6vw2ngOBk6tqV5LTgUuBs1vckpbP88DDSa4A/gNc3WK2jv0DAfhKH+OVJGlWLNQlSdr7vezS9yTH0Ctqb20F9+uAbW33Ma1APxg4AFjbR3+3VtXT7fX7gTOTXNTe7we8A3homvjbq2onsDPJDuDGtv1+4Nhx7X4BUFV3JDmwFcYn0QrsqlqX5I1JDmrt11TVc1P0eRDw0yRHAQXMH7fvtqraAZBkC/BO4BDgjqra2vrqMl5JkmbFQl2SpH1PgAer6sRJ9v0EWFFV97ZV51OmOMYu/n+L3H4T9o1fPQ5wdlU9PIv8nh/3eve497t5+blJTYir1t9EY+2mW9W+hN4/CD7UHra3fop8Xmw5ZJL+ob/xSpI0K96jLknSvudh4E1JTgRIMj/Ju9u+NwDbkswHPjouZmfbN+ZRepeSA3x4mr7WAhekLd0nWdo9/ZesbMc8CdjRVr3voOWd5BTgH1X1zCSxE8dzEPBke71qBn3/AXhfkiNaX2OXvg9yvJIkARbqkiTtc6rqv/SK6+8kuRfYDCxvu78KbABuBf48LuyXwOfbA9KOBL4HfDbJncBh03R3Cb3LyO9rX8F2yRwO5Z+t/yuB89q2i4FlSe4Dvg18YorY24HFYw+TAy4DvpXk9/RuBZhWVW0HPg2sbn/DX7VdgxyvJEkApGqyq7okSZL2nCTrgYuq6q49nYskScPmirokSZIkSSPEFXVJkiRJkkaIK+qSJEmSJI0QC3VJkiRJkkaIhbokSZIkSSPEQl2SJEmSpBFioS5JkiRJ0gixUJckSZIkaYT8D2qvZ8XLtmsrAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1152x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = models_coefficients_rf.nlargest(5, 'rf_coefficients').plot(kind='barh', figsize=(16,8), grid=True, title='Top 5 Feature Importance for Random Forest Model')\n",
    "\n",
    "ax.set_xlabel('Feature Importance');\n",
    "\n",
    "plt.savefig('figures/top5_features_rf.pdf')\n",
    "plt.savefig('figures/top5_features_rf.png')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A lot better, so enrollment time and recency are 2 common variables with the logistic regression. This is cool, because those 2 doesn't help to generate the new clusters, still are super important for both classifiers. Imagine picking a customer from cluster 0 with high enrollment time and low recency... that sounds like a potential one to accept the last campaign!\n",
    "\n",
    "\n",
    "This created a new question about campaign 3, what about creating the same pie chart we created before for each cluster but now for campaign 3? Lets see how they look"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_campaign_frequency_cluster(df, group, campaings, k):\n",
    "    \"\"\"function that returns a Pandas Data Frame of the \n",
    "    percentage of accepted campaigns per cluster\n",
    "    \"\"\"\n",
    "    kmeans_list = {}\n",
    "    for camp in campaings: kmeans_list[camp] = []\n",
    "    for clust in range(0, k):   \n",
    "        for camp in campaings:\n",
    "            kmeans_list[camp].append(1 - list(df[df[group] == clust][camp].value_counts())[0] / df[df[group] == clust].shape[0])\n",
    "            \n",
    "    return pd.DataFrame(kmeans_list, index=range(0, k))\n",
    "\n",
    "def create_campaign_pie_charts(df, campaign, title):\n",
    "    # thanks to help from https://stackoverflow.com/questions/62169907/seaborn-catplot-kind-count-change-bar-chart-to-pie-chart\n",
    "    \"\"\"function that creates a cool pie chart plot for each cluster for a campaign acceptance\n",
    "    \"\"\"\n",
    "    fig, axes = plt.subplots(nrows=2, ncols=2)\n",
    "    labels = 'Accepted', 'Rejected'\n",
    "    colors = ['#008fd5', '#fc4f30']\n",
    "    explode = (0.1, 0.1)\n",
    "\n",
    "    plt.tight_layout()\n",
    "\n",
    "    for i, col in enumerate(df.index):\n",
    "        numb = df.iloc[i][campaign]\n",
    "        curr = list(axes.flat)[i]\n",
    "        curr.pie([numb, 1-numb], startangle=90, colors=colors, wedgeprops={'edgecolor': 'black'}, autopct='%1.f%%', explode=explode, shadow=True)\n",
    "        curr.set_title(f'Cluster {i}', fontweight='bold')\n",
    "\n",
    "    fig.suptitle(title, fontsize=20, y=1.07, fontweight='bold', x=0.37)\n",
    "    fig.set_figheight(10)\n",
    "    fig.set_figwidth(7)\n",
    "    fig.legend(loc='best', labels=labels, fontsize='medium')\n",
    "    fig.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0736842105263158"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# sucess of campaign 3 on the dataset\n",
    "1 - df2['AcceptedCmp3'].value_counts()[0]/df2['AcceptedCmp3'].value_counts().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x720 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "create_campaign_pie_charts(get_campaign_frequency_cluster(df2, 'clusters_kmeans', campaings, 4), 'AcceptedCmp3', 'Campaign 3 Results per cluster')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Oh, well not really that big of a difference compared to the last campaign. Still we know that customers that approved campaing 3 also have a higher chance of accepting the last one. This are some of the cool conclusions we can find, and why trying different techniques can be super useful on finding trends for our customers."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Conclusions\n",
    "\n",
    "After training this 2 classification models and already knowing some features of the cluster with the highest amount of acceptance:\n",
    "\n",
    "- Customers with high enrollment time and low recency should be aimed for the campaign. We could call them loyal and old customers, which have been subscribed a long time and buy frequently in the store.\n",
    "\n",
    "- Amount spent on both meat and gold seem important, but more importantly as we saw on cluster 0, we want to aim at customers with the higest income. As one can guess, the ones with the most income are also customers that can spend a lot more at the store. We would be combining here loyal customers with high income.\n",
    "\n",
    "- Age can be somewhat important but shouldn't be the main choosing factor at the moment of deploying the campaign.\n",
    "\n",
    "- The customers that approved campaign 3 are somewhat more likely to accept the last campaign. So checking at those old customers that bought stuff in there can be also for the marketing department to review.\n",
    "\n",
    "- We found a group of customers (and a way to classify new ones) where the success of the last campaign was almost double compared to the overall dataset, that can be also huge for the marketing department.\n",
    "\n",
    "- Both classifications models have an accuracy higher than 74% most probably enabling a more smart way to offer the campaign for new customers. Yes acurracy is a bit biased since we have more negative answers and thats why we also calculated the recall.\n",
    "\n",
    "- Recall for both models is over 65%, which is not amazing but the logistic regression achieved recall over 75%. Which tells us that the model guessed correctly for 3 of every 4 customers that actually accepted the campaign."
   ]
  }
 ],
 "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
}