{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# ML Enablement Workshop: サービスの解約率改善シナリオ\n",
    "\n",
    "サービスの解約率改善をテーマに、MLOpsのプロセスのビジネス理解から評価までを体験できるシナリオです。\n",
    "\n",
    "## コンテンツ\n",
    "\n",
    "事前準備\n",
    "\n",
    "* 環境構築\n",
    "\n",
    "シナリオ\n",
    "\n",
    "1. ビジネス理解: Business Understanding\n",
    "2. データ分析: Analyze\n",
    "3. データ充足: Prepare\n",
    "4. 前処理: Preprocess\n",
    "5. 学習: Train\n",
    "6. 評価: Test\n",
    "   1. 性能評価\n",
    "   1. コスト分析\n",
    "   1. 最適な閾値を探す"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "## 環境構築\n",
    "\n",
    "プロジェクトを進めるために、機械学習モデルの開発環境を構築します。\n",
    "\n",
    "今回は Amazon SageMaker Studio Lab を使用し、 Conda による環境構築を行います。 `scenario_churn` フォルダにある `environment.yml` を右クリックして `Build Conda Environment` をクリックすると環境が構築できます。\n",
    "\n",
    "![environment_setup_001.png](./images/001.png)\n",
    "\n",
    "完了したら、Jupyter Notebookの右上にある虫の隣のボタンをクリックしカーネルを切り替えます(作成した仮想環境が反映されるまでに少し時間がかかります)。\n",
    "\n",
    "![environment_setup_002.png](./images/002.png)\n",
    "\n",
    "ターミナルから環境を構築する場合は、[こちら](https://github.com/aws-samples/aws-ml-enablement-workshop/blob/main/notebooks/00_environment_setup.ipynb)をご確認ください。\n",
    "\n",
    "環境構築の詳細を知りたい方は、 [機械学習モデルの開発環境を構築する](https://youtu.be/C8VYnJ-DF3I) をご参照ください。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "## 1.ビジネス理解: Business Understanding\n",
    "\n",
    "<img src=\"images/000.png\" alt=\"000.png\" width=200/>\n",
    "\n",
    "[*The cat tries to decrease the customer churn rate in the office.*](https://huggingface.co/spaces/stabilityai/stable-diffusion)\n",
    "\n",
    "どのようなビジネスであっても、顧客を失うことは大きな損害です。このシナリオでは、サービスに満足していない顧客を機械学習 (Machine Learning, ML) で自動で認識してアクションをとれるようにする方法を説明します。このような顧客離れの分析は Customer Churn Prediction と呼ばれています。\n",
    "\n",
    "今回は携帯電話会社での顧客離れ防止を取り上げます。携帯電話会社が、ある顧客が解約しそうと察知したら、その顧客にタイムリーにインセンティブを与えます。インセンティブとは、電話のアップグレードやクーポン(※)、新機能の紹介などです。インセンティブにより引き続き携帯電話会社を使おうと思うかもしれません。インセンティブは、顧客が解約して再度獲得するまでにかかるコストよりもずっと小さいことが多いです。\n",
    "\n",
    "※解約するふりをするとクーポンがもらえるという仕組みはユーザーの質を下げるため、顧客離反防止のためにクーポンを発行するケースは少ないと思います。\n",
    "\n",
    "素朴なビジネス上の欲求として、最小のコストで解約を防ぎたいものです。今回はビジネス KPI を軸に機械学習で解く課題の設定をします。\n",
    "\n",
    "1. ビジネス課題を定義する\n",
    "2. 課題解決のシナリオを描く\n",
    "3. MLのプロセスを特定する\n",
    "\n",
    "### 1.1 ビジネス課題を定義する\n",
    "\n",
    "課題とは、現状 (As-is) と 理想状態 (To-be) のギャップです。今回は、携帯電話会社で顧客獲得競争のため CAC が LTV の 1 / 2 と割高になっているのが現状で、理想状態として CAC が LTV の 1 / 3 以下に抑えられていることとします。\n",
    "\n",
    "聞きなじみがない単語がばらばらと出てきてうんざりしたかもしれません。ディープラーニングの実装にすぐ取り掛かりたいかもしれませんが、機械学習の活用プロジェクトが失敗するのはビジネス理解のフェーズをきちんと行わなかったことが理由であることがほとんどです(詳細が気になる方は[機械学習の価値を計算する](https://youtu.be/csiMBxUkAEc)をぜひご視聴ください)。プロジェクトを成功させるために、ビジネスの課題がどういうものなのか明らかにしましょう。\n",
    "\n",
    "機械学習活用プロジェクトが失敗する理由のTop5:\n",
    "\n",
    "1. **ビジネスの目標がはっきりしていない**\n",
    "2. データの品質が不十分\n",
    "3. プロジェクトのスポンサーが不在\n",
    "4. **チーム間の連携が不十分**\n",
    "5. データサイエンティストなどの専門職の不在\n",
    "\n",
    "LTV (Life Time Value) は顧客一人当たりの価値を表す値で、利益ベースの場合次の式で計算できます。\n",
    "\n",
    "* LTV = 顧客一人当たりの月次利益 x 平均利用月数\n",
    "  * = **顧客一人当たりの月次利益 / 月の解約率**\n",
    "\n",
    "平均利用月数と 1 / 月の解約率が等価になります。継続するということは解約しないということなので、利用月数の期待値は、初月は 1 - 解約率 、次月は ( 1 - 解約率 )^2 ・・・と継続月数分だけ続け、和を取ることになります。数学 B の等比数列の公式を使うとこの式を導けます。\n",
    "\n",
    "![LTVの説明](images/005.png)\n",
    "\n",
    "CAC (Customer Acquisition Cost) は顧客一人当たりの獲得コストです。LTV が CAC を上回っていないと、いわば原価割れしている状態になります。一般的には、 CAC を LTV の 1 / 3 に抑えることが重要といわれています。\n",
    "\n",
    "現状の LTV は 450,000 (※)、 CAC は　200,000 と仮定します。 CAC が LTV の 2.25 なので、理想状態と乖離があります。これが「ビジネス課題」です。\n",
    "\n",
    "※ LTV の計算について、顧客一人当たりの月次売上を 5,000 円、顧客の解約防止にかけるコストを 10% の 500 円とし、差し引き 4,500 円を顧客一人当たりの月次利益としました。解約率を 1% とし、 4,500 / 0.01 で 450,000 となります。LTV から逆算すると約 100 カ月、8 年使う計算ですが 5 年以上使う方が半数以上との調査があるので、おおむね的を得ている計算結果と思います。\n",
    "\n",
    "### 1.2 課題解決のシナリオを描く\n",
    "\n",
    "理想状態に至るための打ち手は、次のように整理できます。\n",
    "\n",
    "* LTV の向上\n",
    "   * 顧客一人当たりの月次利益を向上する\n",
    "   * 解約率を下げる\n",
    "* CAC の削減\n",
    "\n",
    "携帯電話各社の競争は厳しく、 CAC は下げられない、 LTV (≒利用料金) も上げられない。解約率は現状 1% ですが業界トップの D 社は 0.6% と開きがあり、ここに伸びしろがあると結論付けられたとします。そこで、課題解決のシナリオとして、解約率を 0.6% と業界最低水準にし、LTV を 4,500 / 0.006 = 750,000 と、 CAC の 3.75 倍を目指すというシナリオを描きます。\n",
    "\n",
    "### 1.3 MLのプロセスを特定する\n",
    "\n",
    "課題解決のシナリオにおいて機械学習の出番はあるか特定します。シナリオの重要な点は、解約率を下げつつも LTV 構成要素である利益は維持しなければならない、つまりコストは現状の売上の 10% ( 500 円 ) を超過できないことです。効率的に解約率を下げる必要があり、解約防止のためのインセンティブを最適化するのに機械学習が役立ちそうです。\n",
    "\n",
    "プロダクトマネージャーなどビジネスを検討する人とコミュニケーションを取り、課題解決のシナリオ、そこでの機械学習の役割を確認する必要があります。今回は、次のプロジェクト目標が設定され予算が承認されたとします。\n",
    "\n",
    "**解約率を0.6%にしつつ、顧客一人当たりの解約防止コストを500円(売上の10%)に維持することで、 LTV を CAC の 3 倍以上にする。**\n",
    "\n",
    "この目標の達成は機械学習モデルの精度だけでなくマーケティング部門のオペレーションにも大きく左右されます。なぜなら、解約防止のインセンティブを検討・送付するのはマーケティング部門などになるからです (少なくとも機械学習モデルの開発チームではないでしょう) 。プロジェクト目標を計算式で表現することで、計算に登場する数値に責任を関係者が明確になり、協力を仰ぐことができます。機械学習活用プロジェクトが失敗する理由の多くはビジネス理解のプロセスで防ぐことができます。\n",
    "\n",
    "**Discussion**\n",
    "\n",
    "*皆さんの会社では、誰が解約率防止のアクションを取っているでしょうか? 目標を達成するためにどんなチームの協力が必要でしょうか?*\n",
    "\n",
    "### Key takeaways\n",
    "\n",
    "* ビジネス課題は、 To-be と As-is のギャップで定義される。\n",
    "* 「ビジネスの目標がはっきりしていない」は、機械学習の活用プロジェクトが失敗する理由 1 位である。\n",
    "* プロジェクト目標を計算式で表すことで、目標達成の関係者が明確になる。\n",
    "\n",
    "### 参考文献\n",
    "\n",
    "* [ゴリゴリの実務者が書いた、LTVを正しく理解・計算する3つのステップ](https://note.biz.moneyforward.com/n/n58355972b335)\n",
    "* [スマートフォン利用者の月額利用料金は4,617円](https://www.m2ri.jp/release/detail.html?id=525)\n",
    "* [キャンペーンや新料金プランで市場が流動化し、解約率は3社揃って上昇](https://k-tai.watch.impress.co.jp/docs/column/mca/1345010.html#:~:text=%E5%90%84%E7%A4%BE%E5%85%AC%E8%A1%A8%E6%95%B0%E5%80%A4%E3%81%8B%E3%82%89%E3%80%81%E3%82%B9%E3%83%9E%E3%83%BC%E3%83%88%E3%83%95%E3%82%A9%E3%83%B3,%E3%81%A6%E4%B8%8A%E6%98%87%E3%81%97%E3%81%A6%E3%81%84%E3%82%8B%E3%80%82)\n",
    "* [「携帯電話の違約金1,000円」の認知度は75.4％](https://mmdlabo.jp/investigation/detail_1806.html)\n",
    "* [カスタマーサクセスの費用事情：世界のSaaS企業はお金をどう使っているの？](https://success-lab.jp/20171014-2/)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "## 2.データ分析 : Analyze\n",
    "\n",
    "モデルを構築する前に、データ分析を行います。ビジネス理解で設定した目標を達成するのに今あるデータの質・量が十分か確認するためです。\n",
    "\n",
    "携帯電話会社は、どの顧客が最終的に解約したか、または、サービスを使い続けたかの履歴データをもっています。顧客の履歴データの格納先は企業によって様々です。開発チームではなくマーケティングチームが利用しているサービスに格納されていることもあるでしょう。サービスによっては、最新データしか保持しないためETLツールを使って定期的にデータウェアハウスへロードして履歴データを作る必要があるかもしれません。\n",
    "\n",
    "* Salesforce\n",
    "* HubSpot\n",
    "* Relational Database (MySQLやPostgreSQLなど)\n",
    "* Firestore\n",
    "* Google SpreadSheet / Notion\n",
    "* Google Analytics (BigQuery)\n",
    "\n",
    "今回は、素晴らしいことにデータがすでに手に入っていることとします。ここで利用するデータセットは書籍 [Discovering Knowledge in Data](https://www.amazon.com/dp/0470908742/) で Daniel T. Larose が言及しているもので、一般に利用可能です。データセットは、著者によって University of California Irvine Repository of Machine Learning Datasets へ提供されています。\n",
    "\n",
    "データを集める必要がある場合、関係部門へ依頼する必要があります。この時、ビジネス理解のフェーズでビジネス目標を明確にし、共有しておくと協力が得やすくなります。単にデータ分析で何かやってみたいからデータをくださいと依頼するのと、「業界トップの解約率0.5%を達成し LTV を最大化するために・・・」という目標の元協力を仰ぐのとでは、結果が大きく異なるでしょう。\n",
    "\n",
    "では、データ分析を始めましょう。はじめに、使用するライブラリを読み込みます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from pathlib import Path\n",
    "from IPython.display import display\n",
    "import xgboost"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`data/raw`フォルダにある`churn.txt` を `pandas` を利用して読み込みます。 `pandas` は、表形式のデータを読み込んで、様々な加工ができるライブラリです。"
   ]
  },
  {
   "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>State</th>\n",
       "      <th>Account Length</th>\n",
       "      <th>Area Code</th>\n",
       "      <th>Phone</th>\n",
       "      <th>Int'l Plan</th>\n",
       "      <th>VMail Plan</th>\n",
       "      <th>VMail Message</th>\n",
       "      <th>Day Mins</th>\n",
       "      <th>Day Calls</th>\n",
       "      <th>Day Charge</th>\n",
       "      <th>Eve Mins</th>\n",
       "      <th>Eve Calls</th>\n",
       "      <th>Eve Charge</th>\n",
       "      <th>Night Mins</th>\n",
       "      <th>Night Calls</th>\n",
       "      <th>Night Charge</th>\n",
       "      <th>Intl Mins</th>\n",
       "      <th>Intl Calls</th>\n",
       "      <th>Intl Charge</th>\n",
       "      <th>CustServ Calls</th>\n",
       "      <th>Churn?</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>PA</td>\n",
       "      <td>163</td>\n",
       "      <td>806</td>\n",
       "      <td>403-2562</td>\n",
       "      <td>no</td>\n",
       "      <td>yes</td>\n",
       "      <td>300</td>\n",
       "      <td>8.162204</td>\n",
       "      <td>3</td>\n",
       "      <td>7.579174</td>\n",
       "      <td>3.933035</td>\n",
       "      <td>4</td>\n",
       "      <td>6.508639</td>\n",
       "      <td>4.065759</td>\n",
       "      <td>100</td>\n",
       "      <td>5.111624</td>\n",
       "      <td>4.928160</td>\n",
       "      <td>6</td>\n",
       "      <td>5.673203</td>\n",
       "      <td>3</td>\n",
       "      <td>True.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>SC</td>\n",
       "      <td>15</td>\n",
       "      <td>836</td>\n",
       "      <td>158-8416</td>\n",
       "      <td>yes</td>\n",
       "      <td>no</td>\n",
       "      <td>0</td>\n",
       "      <td>10.018993</td>\n",
       "      <td>4</td>\n",
       "      <td>4.226289</td>\n",
       "      <td>2.325005</td>\n",
       "      <td>0</td>\n",
       "      <td>9.972592</td>\n",
       "      <td>7.141040</td>\n",
       "      <td>200</td>\n",
       "      <td>6.436188</td>\n",
       "      <td>3.221748</td>\n",
       "      <td>6</td>\n",
       "      <td>2.559749</td>\n",
       "      <td>8</td>\n",
       "      <td>False.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>MO</td>\n",
       "      <td>131</td>\n",
       "      <td>777</td>\n",
       "      <td>896-6253</td>\n",
       "      <td>no</td>\n",
       "      <td>yes</td>\n",
       "      <td>300</td>\n",
       "      <td>4.708490</td>\n",
       "      <td>3</td>\n",
       "      <td>4.768160</td>\n",
       "      <td>4.537466</td>\n",
       "      <td>3</td>\n",
       "      <td>4.566715</td>\n",
       "      <td>5.363235</td>\n",
       "      <td>100</td>\n",
       "      <td>5.142451</td>\n",
       "      <td>7.139023</td>\n",
       "      <td>2</td>\n",
       "      <td>6.254157</td>\n",
       "      <td>4</td>\n",
       "      <td>False.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>WY</td>\n",
       "      <td>75</td>\n",
       "      <td>878</td>\n",
       "      <td>817-5729</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>700</td>\n",
       "      <td>1.268734</td>\n",
       "      <td>3</td>\n",
       "      <td>2.567642</td>\n",
       "      <td>2.528748</td>\n",
       "      <td>5</td>\n",
       "      <td>2.333624</td>\n",
       "      <td>3.773586</td>\n",
       "      <td>450</td>\n",
       "      <td>3.814413</td>\n",
       "      <td>2.245779</td>\n",
       "      <td>6</td>\n",
       "      <td>1.080692</td>\n",
       "      <td>6</td>\n",
       "      <td>False.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>WY</td>\n",
       "      <td>146</td>\n",
       "      <td>878</td>\n",
       "      <td>450-4942</td>\n",
       "      <td>yes</td>\n",
       "      <td>no</td>\n",
       "      <td>0</td>\n",
       "      <td>2.696177</td>\n",
       "      <td>3</td>\n",
       "      <td>5.908916</td>\n",
       "      <td>6.015337</td>\n",
       "      <td>3</td>\n",
       "      <td>3.670408</td>\n",
       "      <td>3.751673</td>\n",
       "      <td>250</td>\n",
       "      <td>2.796812</td>\n",
       "      <td>6.905545</td>\n",
       "      <td>4</td>\n",
       "      <td>7.134343</td>\n",
       "      <td>6</td>\n",
       "      <td>True.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4995</th>\n",
       "      <td>NH</td>\n",
       "      <td>4</td>\n",
       "      <td>787</td>\n",
       "      <td>151-3162</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>800</td>\n",
       "      <td>10.862632</td>\n",
       "      <td>5</td>\n",
       "      <td>7.250969</td>\n",
       "      <td>6.936164</td>\n",
       "      <td>1</td>\n",
       "      <td>8.026482</td>\n",
       "      <td>4.921314</td>\n",
       "      <td>350</td>\n",
       "      <td>6.748489</td>\n",
       "      <td>4.872570</td>\n",
       "      <td>8</td>\n",
       "      <td>2.122530</td>\n",
       "      <td>9</td>\n",
       "      <td>False.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4996</th>\n",
       "      <td>SD</td>\n",
       "      <td>140</td>\n",
       "      <td>836</td>\n",
       "      <td>351-5993</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>0</td>\n",
       "      <td>1.581127</td>\n",
       "      <td>8</td>\n",
       "      <td>3.758307</td>\n",
       "      <td>7.377591</td>\n",
       "      <td>7</td>\n",
       "      <td>1.328827</td>\n",
       "      <td>0.939932</td>\n",
       "      <td>300</td>\n",
       "      <td>4.522661</td>\n",
       "      <td>6.938571</td>\n",
       "      <td>2</td>\n",
       "      <td>4.600473</td>\n",
       "      <td>4</td>\n",
       "      <td>False.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4997</th>\n",
       "      <td>SC</td>\n",
       "      <td>32</td>\n",
       "      <td>836</td>\n",
       "      <td>370-3127</td>\n",
       "      <td>no</td>\n",
       "      <td>yes</td>\n",
       "      <td>700</td>\n",
       "      <td>0.163836</td>\n",
       "      <td>5</td>\n",
       "      <td>4.243980</td>\n",
       "      <td>5.841852</td>\n",
       "      <td>3</td>\n",
       "      <td>2.340554</td>\n",
       "      <td>0.939469</td>\n",
       "      <td>450</td>\n",
       "      <td>5.157898</td>\n",
       "      <td>4.388328</td>\n",
       "      <td>7</td>\n",
       "      <td>1.060340</td>\n",
       "      <td>6</td>\n",
       "      <td>False.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4998</th>\n",
       "      <td>MA</td>\n",
       "      <td>142</td>\n",
       "      <td>776</td>\n",
       "      <td>604-2108</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>600</td>\n",
       "      <td>2.034454</td>\n",
       "      <td>5</td>\n",
       "      <td>3.014859</td>\n",
       "      <td>4.140554</td>\n",
       "      <td>3</td>\n",
       "      <td>3.470372</td>\n",
       "      <td>6.076043</td>\n",
       "      <td>150</td>\n",
       "      <td>4.362780</td>\n",
       "      <td>7.173376</td>\n",
       "      <td>3</td>\n",
       "      <td>4.871900</td>\n",
       "      <td>7</td>\n",
       "      <td>True.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4999</th>\n",
       "      <td>AL</td>\n",
       "      <td>141</td>\n",
       "      <td>657</td>\n",
       "      <td>294-2849</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>500</td>\n",
       "      <td>1.803907</td>\n",
       "      <td>0</td>\n",
       "      <td>5.125716</td>\n",
       "      <td>8.357508</td>\n",
       "      <td>0</td>\n",
       "      <td>2.109823</td>\n",
       "      <td>2.624299</td>\n",
       "      <td>400</td>\n",
       "      <td>3.713631</td>\n",
       "      <td>5.798783</td>\n",
       "      <td>6</td>\n",
       "      <td>5.485345</td>\n",
       "      <td>7</td>\n",
       "      <td>False.</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5000 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     State  Account Length  Area Code     Phone Int'l Plan VMail Plan  \\\n",
       "0       PA             163        806  403-2562         no        yes   \n",
       "1       SC              15        836  158-8416        yes         no   \n",
       "2       MO             131        777  896-6253         no        yes   \n",
       "3       WY              75        878  817-5729        yes        yes   \n",
       "4       WY             146        878  450-4942        yes         no   \n",
       "...    ...             ...        ...       ...        ...        ...   \n",
       "4995    NH               4        787  151-3162        yes        yes   \n",
       "4996    SD             140        836  351-5993         no         no   \n",
       "4997    SC              32        836  370-3127         no        yes   \n",
       "4998    MA             142        776  604-2108        yes        yes   \n",
       "4999    AL             141        657  294-2849        yes        yes   \n",
       "\n",
       "      VMail Message   Day Mins  Day Calls  Day Charge  Eve Mins  Eve Calls  \\\n",
       "0               300   8.162204          3    7.579174  3.933035          4   \n",
       "1                 0  10.018993          4    4.226289  2.325005          0   \n",
       "2               300   4.708490          3    4.768160  4.537466          3   \n",
       "3               700   1.268734          3    2.567642  2.528748          5   \n",
       "4                 0   2.696177          3    5.908916  6.015337          3   \n",
       "...             ...        ...        ...         ...       ...        ...   \n",
       "4995            800  10.862632          5    7.250969  6.936164          1   \n",
       "4996              0   1.581127          8    3.758307  7.377591          7   \n",
       "4997            700   0.163836          5    4.243980  5.841852          3   \n",
       "4998            600   2.034454          5    3.014859  4.140554          3   \n",
       "4999            500   1.803907          0    5.125716  8.357508          0   \n",
       "\n",
       "      Eve Charge  Night Mins  Night Calls  Night Charge  Intl Mins  \\\n",
       "0       6.508639    4.065759          100      5.111624   4.928160   \n",
       "1       9.972592    7.141040          200      6.436188   3.221748   \n",
       "2       4.566715    5.363235          100      5.142451   7.139023   \n",
       "3       2.333624    3.773586          450      3.814413   2.245779   \n",
       "4       3.670408    3.751673          250      2.796812   6.905545   \n",
       "...          ...         ...          ...           ...        ...   \n",
       "4995    8.026482    4.921314          350      6.748489   4.872570   \n",
       "4996    1.328827    0.939932          300      4.522661   6.938571   \n",
       "4997    2.340554    0.939469          450      5.157898   4.388328   \n",
       "4998    3.470372    6.076043          150      4.362780   7.173376   \n",
       "4999    2.109823    2.624299          400      3.713631   5.798783   \n",
       "\n",
       "      Intl Calls  Intl Charge  CustServ Calls  Churn?  \n",
       "0              6     5.673203               3   True.  \n",
       "1              6     2.559749               8  False.  \n",
       "2              2     6.254157               4  False.  \n",
       "3              6     1.080692               6  False.  \n",
       "4              4     7.134343               6   True.  \n",
       "...          ...          ...             ...     ...  \n",
       "4995           8     2.122530               9  False.  \n",
       "4996           2     4.600473               4  False.  \n",
       "4997           7     1.060340               6  False.  \n",
       "4998           3     4.871900               7   True.  \n",
       "4999           6     5.485345               7  False.  \n",
       "\n",
       "[5000 rows x 21 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_root = Path(\"../../data/\")\n",
    "churn = pd.read_csv(data_root.joinpath(\"raw/churn.txt\"))\n",
    "pd.set_option(\"display.max_columns\", len(churn.columns))\n",
    "churn"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "データは 5,000 行で、現在の機械学習としてはやや小さいデータセットです。各データのレコードは、ある米国の携帯電話会社の顧客プロフィールに関する 21 の属性からなります。\n",
    "\n",
    "- `State` : 顧客が居住している米国州で、2文字の省略形で記載されます (OHとかNJのように)\n",
    "- `Account Length` : アカウントが利用可能になってからの経過日数\n",
    "- `Area Code` : 顧客の電話番号に対応する3桁のエリアコード (市外局番)\n",
    "- `Phone` : 残りの7桁の電話番号\n",
    "- `Int’l Plan` : 国際電話のプランに加入しているかどうか (yes/no)\n",
    "- `VMail Plan` : Voice mail の機能を利用しているかどうか (yes/no)\n",
    "- `VMail Message` : 1ヶ月の Voice mail のメッセージの平均長\n",
    "- `Day Mins` : 1日に通話した時間(分)の総和\n",
    "- `Day Calls` : 1日に通話した回数の総和\n",
    "- `Day Charge` : 日中の通話にかかった料金\n",
    "- `Eve Mins, Eve Calls, Eve Charge` : 夜間通話にかかった料金\n",
    "- `Night Mins`, `Night Calls`, `Night Charge` : 深夜通話にかかった料金\n",
    "- `Intl Mins`, `Intl Calls`, `Intl Charge` : 国際通話にかかった料金\n",
    "- `CustServ Calls` : カスタマーサービスに電話をかけた回数\n",
    "- `Churn?` : そのサービスから解約したかどうか (true/false)\n",
    "\n",
    "**Discussion**\n",
    "\n",
    "*皆さんのサービスで似たような項目はあるでしょうか? `Int’l Plan` や `VMail Plan` は有償プランや特定機能の利用有無、 `Day Mins` や `Day Calls` は 1 日の利用時間、 1 日のログイン回数などに置き換えて考えられるかもしれません。*\n",
    "\n",
    "最後の属性 `Churn?` が機械学習モデルで予測する対象になります。予測の対象は**目的変数**と呼ばれます。今回の目的変数は 2 値 (binary) なので、今回作成するモデルは 2 値の予測を行います。これは 2 値分類といわれます。\n",
    "\n",
    "それではデータを詳しく見てみます。まずは各列のデータ型を確認します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "State              object\n",
       "Account Length      int64\n",
       "Area Code           int64\n",
       "Phone              object\n",
       "Int'l Plan         object\n",
       "VMail Plan         object\n",
       "VMail Message       int64\n",
       "Day Mins          float64\n",
       "Day Calls           int64\n",
       "Day Charge        float64\n",
       "Eve Mins          float64\n",
       "Eve Calls           int64\n",
       "Eve Charge        float64\n",
       "Night Mins        float64\n",
       "Night Calls         int64\n",
       "Night Charge      float64\n",
       "Intl Mins         float64\n",
       "Intl Calls          int64\n",
       "Intl Charge       float64\n",
       "CustServ Calls      int64\n",
       "Churn?             object\n",
       "dtype: object"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "churn.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`object` 形式のデータは、選択式項目や文字列のデータです。こうした変数を **カテゴリ変数** と呼びます。たとえば、ユーザーの所在地 ( 千葉県や東京都 ) サービス利用有無です。\n",
    "\n",
    "カテゴリ変数から分析していきましょう。`describe()` を利用すると各属性の統計量を一度に見ることができます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>State</th>\n",
       "      <th>Phone</th>\n",
       "      <th>Int'l Plan</th>\n",
       "      <th>VMail Plan</th>\n",
       "      <th>Churn?</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>5000</td>\n",
       "      <td>5000</td>\n",
       "      <td>5000</td>\n",
       "      <td>5000</td>\n",
       "      <td>5000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>51</td>\n",
       "      <td>4999</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>RI</td>\n",
       "      <td>614-5668</td>\n",
       "      <td>no</td>\n",
       "      <td>yes</td>\n",
       "      <td>False.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>120</td>\n",
       "      <td>2</td>\n",
       "      <td>2507</td>\n",
       "      <td>2512</td>\n",
       "      <td>2502</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       State     Phone Int'l Plan VMail Plan  Churn?\n",
       "count   5000      5000       5000       5000    5000\n",
       "unique    51      4999          2          2       2\n",
       "top       RI  614-5668         no        yes  False.\n",
       "freq     120         2       2507       2512    2502"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "churn.select_dtypes(include=['object']).describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`count` は件数、 `unique` は項目数です。 `yes` / `no` の 2 値である `Int'l Plan` は 2 になっています。当然ですが電話番号である `Phone` は件数と項目数が一致するはずです ( 1 件ずれているのは、本当なら確認が必要かもしれません ) 。\n",
    "\n",
    "データ型、カテゴリ変数の統計量から次の洞察が得られます。\n",
    "\n",
    "* `Area code` は本来 `object` として扱うべきですが数値データとみなされているようなので、非数値に変換しましょう。\n",
    "* `Phone` については、ユーザーに対しほぼ一意のIDとなっています。 Idから解約するか予測できるわけではないので、使うのは止めるべきでしょう。\n",
    "\n",
    "`Area code` の変換と、 `Phone` の削除を行います。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "churn['Area Code'] = churn['Area Code'].astype(object)\n",
    "churn = churn.drop('Phone', axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "カテゴリ変数について、値の出現頻度に偏りがあるか確認します。例えば一部の地区のデータしかなかったら、他の地区のデータを拡充したほうがよいでしょう。\n",
    "\n",
    "`pandas` では `value_counts` を使い値の出現数を数えられます。 `normalize=True` を指定すると出現数から割合を計算できます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 700x1000 with 5 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "categorical_columns = churn.select_dtypes(include=['object']).columns\n",
    "fig, axes = plt.subplots(nrows=len(categorical_columns), ncols=1, figsize=(7, 10))\n",
    "for i, column in enumerate([c for c in churn.select_dtypes(include=[\"object\"]).columns]):\n",
    "        title = f\"Frequency of {column}\"\n",
    "        churn[column].value_counts(normalize=True).sort_values(ascending=False).plot.bar(title=title, ax=axes[i])\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "データセット中に偏りがあまりないことがわかります。\n",
    "\n",
    "次に、目的変数 `Churn?` が True か False かで、出現頻度に差があるか見てみます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 700x1000 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(nrows=len(categorical_columns) - 1, ncols=1, figsize=(7, 10))\n",
    "for i, column in enumerate([c for c in categorical_columns if c != \"Churn?\"]):\n",
    "        title = f\"Frequency comparison of {column} between churn or not\"\n",
    "        pd.crosstab(index=churn[column], columns=churn[\"Churn?\"], normalize='columns').plot.area(title=title, ax=axes[i], stacked=False)\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "カテゴリ変数のデータ分析から次のことが読み取れます。\n",
    "\n",
    "- 各項目の頻度はだいたい一様に分布しています。解約したか否かでも分布は変わりません。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "続いて、数値である **量的変数** のデータに関して統計量とデータの分布を確認します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>Account Length</th>\n",
       "      <th>VMail Message</th>\n",
       "      <th>Day Mins</th>\n",
       "      <th>Day Calls</th>\n",
       "      <th>Day Charge</th>\n",
       "      <th>Eve Mins</th>\n",
       "      <th>Eve Calls</th>\n",
       "      <th>Eve Charge</th>\n",
       "      <th>Night Mins</th>\n",
       "      <th>Night Calls</th>\n",
       "      <th>Night Charge</th>\n",
       "      <th>Intl Mins</th>\n",
       "      <th>Intl Calls</th>\n",
       "      <th>Intl Charge</th>\n",
       "      <th>CustServ Calls</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>5000.000000</td>\n",
       "      <td>5000.000000</td>\n",
       "      <td>5000.000000</td>\n",
       "      <td>5000.00000</td>\n",
       "      <td>5000.000000</td>\n",
       "      <td>5000.000000</td>\n",
       "      <td>5000.000000</td>\n",
       "      <td>5000.000000</td>\n",
       "      <td>5000.000000</td>\n",
       "      <td>5000.000000</td>\n",
       "      <td>5000.000000</td>\n",
       "      <td>5000.000000</td>\n",
       "      <td>5000.000000</td>\n",
       "      <td>5000.000000</td>\n",
       "      <td>5000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>101.675800</td>\n",
       "      <td>226.680000</td>\n",
       "      <td>5.518757</td>\n",
       "      <td>3.50460</td>\n",
       "      <td>5.018902</td>\n",
       "      <td>5.026199</td>\n",
       "      <td>3.140400</td>\n",
       "      <td>5.017557</td>\n",
       "      <td>4.000917</td>\n",
       "      <td>224.790000</td>\n",
       "      <td>5.023490</td>\n",
       "      <td>5.025876</td>\n",
       "      <td>5.475400</td>\n",
       "      <td>4.328242</td>\n",
       "      <td>5.525800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>57.596762</td>\n",
       "      <td>273.998527</td>\n",
       "      <td>3.433485</td>\n",
       "      <td>1.68812</td>\n",
       "      <td>2.195759</td>\n",
       "      <td>2.135487</td>\n",
       "      <td>2.525621</td>\n",
       "      <td>2.127857</td>\n",
       "      <td>1.631001</td>\n",
       "      <td>97.302875</td>\n",
       "      <td>1.748900</td>\n",
       "      <td>1.019302</td>\n",
       "      <td>1.877045</td>\n",
       "      <td>2.440311</td>\n",
       "      <td>2.041217</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000215</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.004777</td>\n",
       "      <td>0.004659</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.013573</td>\n",
       "      <td>0.008468</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.054863</td>\n",
       "      <td>1.648514</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000769</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>52.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.682384</td>\n",
       "      <td>2.00000</td>\n",
       "      <td>3.470151</td>\n",
       "      <td>3.588466</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.529613</td>\n",
       "      <td>2.921998</td>\n",
       "      <td>150.000000</td>\n",
       "      <td>3.873157</td>\n",
       "      <td>4.349726</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>2.468225</td>\n",
       "      <td>4.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>102.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>5.336245</td>\n",
       "      <td>3.00000</td>\n",
       "      <td>4.988291</td>\n",
       "      <td>5.145656</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>5.006860</td>\n",
       "      <td>3.962089</td>\n",
       "      <td>200.000000</td>\n",
       "      <td>5.169154</td>\n",
       "      <td>5.034905</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>4.214058</td>\n",
       "      <td>6.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>151.000000</td>\n",
       "      <td>400.000000</td>\n",
       "      <td>7.936151</td>\n",
       "      <td>5.00000</td>\n",
       "      <td>6.559750</td>\n",
       "      <td>6.552962</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>6.491725</td>\n",
       "      <td>5.100128</td>\n",
       "      <td>300.000000</td>\n",
       "      <td>6.272015</td>\n",
       "      <td>5.716386</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>5.960654</td>\n",
       "      <td>7.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>200.000000</td>\n",
       "      <td>1300.000000</td>\n",
       "      <td>16.897529</td>\n",
       "      <td>10.00000</td>\n",
       "      <td>12.731936</td>\n",
       "      <td>13.622097</td>\n",
       "      <td>14.000000</td>\n",
       "      <td>12.352871</td>\n",
       "      <td>10.183378</td>\n",
       "      <td>550.000000</td>\n",
       "      <td>10.407778</td>\n",
       "      <td>8.405644</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>14.212261</td>\n",
       "      <td>13.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Account Length  VMail Message     Day Mins   Day Calls   Day Charge  \\\n",
       "count     5000.000000    5000.000000  5000.000000  5000.00000  5000.000000   \n",
       "mean       101.675800     226.680000     5.518757     3.50460     5.018902   \n",
       "std         57.596762     273.998527     3.433485     1.68812     2.195759   \n",
       "min          1.000000       0.000000     0.000215     0.00000     0.004777   \n",
       "25%         52.000000       0.000000     2.682384     2.00000     3.470151   \n",
       "50%        102.000000       0.000000     5.336245     3.00000     4.988291   \n",
       "75%        151.000000     400.000000     7.936151     5.00000     6.559750   \n",
       "max        200.000000    1300.000000    16.897529    10.00000    12.731936   \n",
       "\n",
       "          Eve Mins    Eve Calls   Eve Charge   Night Mins  Night Calls  \\\n",
       "count  5000.000000  5000.000000  5000.000000  5000.000000  5000.000000   \n",
       "mean      5.026199     3.140400     5.017557     4.000917   224.790000   \n",
       "std       2.135487     2.525621     2.127857     1.631001    97.302875   \n",
       "min       0.004659     0.000000     0.013573     0.008468     0.000000   \n",
       "25%       3.588466     1.000000     3.529613     2.921998   150.000000   \n",
       "50%       5.145656     3.000000     5.006860     3.962089   200.000000   \n",
       "75%       6.552962     5.000000     6.491725     5.100128   300.000000   \n",
       "max      13.622097    14.000000    12.352871    10.183378   550.000000   \n",
       "\n",
       "       Night Charge    Intl Mins   Intl Calls  Intl Charge  CustServ Calls  \n",
       "count   5000.000000  5000.000000  5000.000000  5000.000000     5000.000000  \n",
       "mean       5.023490     5.025876     5.475400     4.328242        5.525800  \n",
       "std        1.748900     1.019302     1.877045     2.440311        2.041217  \n",
       "min        0.054863     1.648514     0.000000     0.000769        0.000000  \n",
       "25%        3.873157     4.349726     4.000000     2.468225        4.000000  \n",
       "50%        5.169154     5.034905     5.000000     4.214058        6.000000  \n",
       "75%        6.272015     5.716386     7.000000     5.960654        7.000000  \n",
       "max       10.407778     8.405644    12.000000    14.212261       13.000000  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "churn.select_dtypes(exclude=['object']).describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[<AxesSubplot:title={'center':'Account Length'}>,\n",
       "        <AxesSubplot:title={'center':'VMail Message'}>,\n",
       "        <AxesSubplot:title={'center':'Day Mins'}>,\n",
       "        <AxesSubplot:title={'center':'Day Calls'}>],\n",
       "       [<AxesSubplot:title={'center':'Day Charge'}>,\n",
       "        <AxesSubplot:title={'center':'Eve Mins'}>,\n",
       "        <AxesSubplot:title={'center':'Eve Calls'}>,\n",
       "        <AxesSubplot:title={'center':'Eve Charge'}>],\n",
       "       [<AxesSubplot:title={'center':'Night Mins'}>,\n",
       "        <AxesSubplot:title={'center':'Night Calls'}>,\n",
       "        <AxesSubplot:title={'center':'Night Charge'}>,\n",
       "        <AxesSubplot:title={'center':'Intl Mins'}>],\n",
       "       [<AxesSubplot:title={'center':'Intl Calls'}>,\n",
       "        <AxesSubplot:title={'center':'Intl Charge'}>,\n",
       "        <AxesSubplot:title={'center':'CustServ Calls'}>, <AxesSubplot:>]],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 16 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "churn.select_dtypes(exclude=['object']).hist(bins=30, figsize=(10, 10))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "数値項目の多くは正規分布のような釣り鐘の分布になっています。ただ、`VMail Message` は例外です。  \n",
    "数値項目も、目的変数 `Churn?` が True か False かで分布が変わるか見てみます。オレンジが解約、青が解約しないユーザーです。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 16 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(nrows=4, ncols=4, figsize=(10, 10))\n",
    "for i, column in enumerate(churn.select_dtypes(exclude=['object']).columns):\n",
    "    row = i // 4\n",
    "    col = i % 4\n",
    "    hist = churn[[column, \"Churn?\"]].groupby(\"Churn?\").hist(\n",
    "            bins=np.linspace(churn[column].min(),\n",
    "            churn[column].max(), 30), ax=axes[row, col], alpha=0.5)\n",
    "\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "カテゴリ変数、量的変数のデータ分析の結果から、解約する顧客について、以下のような傾向が見られます。\n",
    "\n",
    "- 地理的にほぼ一様に分散している\n",
    "- VoiceMailをあまり利用していない\n",
    "- 支払い料金 (`Charge`)が解約しているユーザーに比べて高い\n",
    "- `Eve Call`、`Night Call` は解約していないユーザーよりも回数が少ないにもかかわらず、`Eve Charge`、`Night Charge` の料金は高い。\n",
    "\n",
    "料金プランの問題かもしれませんが、夜間~深夜の価格体系に問題があるかもしれません。\n",
    "\n",
    "ここで、解約する顧客に関して、`Day Mins` と `Day Charge` は似たような分布を示しています。話せば話すほど、通常課金されるので、驚くことではないです。どの程度似ているか調べるために、`corr()` を使います。 `corr()` では、相関係数を求めることができます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\ttaakkaa\\AppData\\Local\\Temp\\ipykernel_53272\\173422500.py:2: FutureWarning: this method is deprecated in favour of `Styler.format(precision=..)`\n",
      "  corr.style.background_gradient(cmap=\"Greens\").set_precision(2)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_4e9ac_row0_col0, #T_4e9ac_row1_col1, #T_4e9ac_row2_col2, #T_4e9ac_row3_col3, #T_4e9ac_row4_col4, #T_4e9ac_row5_col5, #T_4e9ac_row6_col6, #T_4e9ac_row7_col7, #T_4e9ac_row8_col8, #T_4e9ac_row9_col9, #T_4e9ac_row10_col10, #T_4e9ac_row11_col11, #T_4e9ac_row12_col12, #T_4e9ac_row13_col13, #T_4e9ac_row14_col14 {\n",
       "  background-color: #00441b;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_4e9ac_row0_col1, #T_4e9ac_row12_col3 {\n",
       "  background-color: #e0f3db;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row0_col2, #T_4e9ac_row1_col7, #T_4e9ac_row5_col14 {\n",
       "  background-color: #b8e3b2;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row0_col3, #T_4e9ac_row2_col6 {\n",
       "  background-color: #d5efcf;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row0_col4, #T_4e9ac_row9_col5 {\n",
       "  background-color: #d3eecd;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row0_col5, #T_4e9ac_row10_col8, #T_4e9ac_row11_col1 {\n",
       "  background-color: #e1f3dc;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row0_col6, #T_4e9ac_row9_col12 {\n",
       "  background-color: #acdea6;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row0_col7, #T_4e9ac_row0_col9, #T_4e9ac_row11_col2, #T_4e9ac_row14_col7 {\n",
       "  background-color: #b5e1ae;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row0_col8 {\n",
       "  background-color: #c3e7bc;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row0_col10, #T_4e9ac_row3_col9, #T_4e9ac_row12_col13 {\n",
       "  background-color: #c6e8bf;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row0_col11, #T_4e9ac_row5_col0 {\n",
       "  background-color: #f2faef;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row0_col12, #T_4e9ac_row1_col3, #T_4e9ac_row3_col8, #T_4e9ac_row11_col12 {\n",
       "  background-color: #d7efd1;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row0_col13, #T_4e9ac_row6_col8, #T_4e9ac_row10_col5 {\n",
       "  background-color: #92d28f;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row0_col14 {\n",
       "  background-color: #a0d99b;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row1_col0, #T_4e9ac_row2_col11 {\n",
       "  background-color: #f4fbf1;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row1_col2, #T_4e9ac_row5_col3 {\n",
       "  background-color: #d1edcb;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row1_col4, #T_4e9ac_row6_col0 {\n",
       "  background-color: #eff9eb;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row1_col5, #T_4e9ac_row8_col0, #T_4e9ac_row11_col0 {\n",
       "  background-color: #f0f9ec;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row1_col6, #T_4e9ac_row3_col2, #T_4e9ac_row5_col12 {\n",
       "  background-color: #c8e9c1;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row1_col8, #T_4e9ac_row1_col12 {\n",
       "  background-color: #bae3b3;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row1_col9, #T_4e9ac_row1_col13 {\n",
       "  background-color: #97d492;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row1_col10 {\n",
       "  background-color: #dbf1d5;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row1_col11, #T_4e9ac_row2_col9, #T_4e9ac_row5_col11, #T_4e9ac_row8_col11, #T_4e9ac_row9_col7, #T_4e9ac_row13_col11 {\n",
       "  background-color: #f6fcf4;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row1_col14, #T_4e9ac_row8_col2 {\n",
       "  background-color: #8ace88;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row2_col0, #T_4e9ac_row6_col11, #T_4e9ac_row7_col0, #T_4e9ac_row10_col1, #T_4e9ac_row14_col6 {\n",
       "  background-color: #f4fbf2;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row2_col1, #T_4e9ac_row4_col11, #T_4e9ac_row9_col0, #T_4e9ac_row10_col11, #T_4e9ac_row12_col0 {\n",
       "  background-color: #f2faf0;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row2_col3, #T_4e9ac_row8_col3 {\n",
       "  background-color: #e7f6e3;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row2_col4 {\n",
       "  background-color: #268e47;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_4e9ac_row2_col5 {\n",
       "  background-color: #5db96b;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_4e9ac_row2_col7 {\n",
       "  background-color: #097532;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_4e9ac_row2_col8, #T_4e9ac_row7_col14, #T_4e9ac_row11_col13, #T_4e9ac_row11_col14, #T_4e9ac_row12_col14 {\n",
       "  background-color: #9bd696;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row2_col10, #T_4e9ac_row10_col7 {\n",
       "  background-color: #319a50;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_4e9ac_row2_col12 {\n",
       "  background-color: #9fd899;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row2_col13 {\n",
       "  background-color: #63bc6e;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_4e9ac_row2_col14, #T_4e9ac_row12_col8 {\n",
       "  background-color: #bee5b8;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row3_col0, #T_4e9ac_row10_col9, #T_4e9ac_row13_col10 {\n",
       "  background-color: #f1faee;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row3_col1, #T_4e9ac_row8_col9 {\n",
       "  background-color: #ddf2d8;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row3_col4, #T_4e9ac_row4_col0 {\n",
       "  background-color: #f3faf0;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row3_col5, #T_4e9ac_row14_col2 {\n",
       "  background-color: #daf0d4;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row3_col6 {\n",
       "  background-color: #87cd86;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row3_col7 {\n",
       "  background-color: #bde5b6;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row3_col10, #T_4e9ac_row11_col7 {\n",
       "  background-color: #b4e1ad;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row3_col11, #T_4e9ac_row4_col1, #T_4e9ac_row4_col3, #T_4e9ac_row6_col7, #T_4e9ac_row7_col6, #T_4e9ac_row7_col9, #T_4e9ac_row9_col2, #T_4e9ac_row9_col10, #T_4e9ac_row10_col0, #T_4e9ac_row13_col12, #T_4e9ac_row13_col14, #T_4e9ac_row14_col0, #T_4e9ac_row14_col4, #T_4e9ac_row14_col5, #T_4e9ac_row14_col8, #T_4e9ac_row14_col13 {\n",
       "  background-color: #f7fcf5;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row3_col12 {\n",
       "  background-color: #def2d9;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row3_col13, #T_4e9ac_row8_col4 {\n",
       "  background-color: #b0dfaa;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row3_col14, #T_4e9ac_row5_col9 {\n",
       "  background-color: #a7dba0;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row4_col2 {\n",
       "  background-color: #1d8640;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_4e9ac_row4_col5 {\n",
       "  background-color: #3fa85b;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_4e9ac_row4_col6 {\n",
       "  background-color: #7cc87c;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row4_col7 {\n",
       "  background-color: #56b567;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_4e9ac_row4_col8 {\n",
       "  background-color: #a5db9f;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row4_col9 {\n",
       "  background-color: #cdecc7;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row4_col10, #T_4e9ac_row4_col13 {\n",
       "  background-color: #60ba6c;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_4e9ac_row4_col12 {\n",
       "  background-color: #bce4b5;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row4_col14, #T_4e9ac_row11_col8 {\n",
       "  background-color: #caeac3;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row5_col1, #T_4e9ac_row8_col5, #T_4e9ac_row13_col0 {\n",
       "  background-color: #eef8ea;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row5_col2 {\n",
       "  background-color: #3da65a;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_4e9ac_row5_col4, #T_4e9ac_row10_col14 {\n",
       "  background-color: #38a156;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_4e9ac_row5_col6 {\n",
       "  background-color: #c1e6ba;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row5_col7, #T_4e9ac_row7_col4 {\n",
       "  background-color: #73c476;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row5_col8, #T_4e9ac_row6_col2, #T_4e9ac_row8_col10, #T_4e9ac_row10_col12 {\n",
       "  background-color: #d8f0d2;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row5_col10 {\n",
       "  background-color: #70c274;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row5_col13 {\n",
       "  background-color: #369f54;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_4e9ac_row6_col1 {\n",
       "  background-color: #edf8ea;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row6_col3 {\n",
       "  background-color: #aedea7;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row6_col4 {\n",
       "  background-color: #99d595;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row6_col5, #T_4e9ac_row9_col8, #T_4e9ac_row13_col3 {\n",
       "  background-color: #ebf7e7;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row6_col9, #T_4e9ac_row9_col14 {\n",
       "  background-color: #81ca81;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row6_col10 {\n",
       "  background-color: #f0f9ed;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row6_col12, #T_4e9ac_row10_col6 {\n",
       "  background-color: #e9f7e5;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row6_col13, #T_4e9ac_row13_col4 {\n",
       "  background-color: #95d391;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row6_col14, #T_4e9ac_row7_col3 {\n",
       "  background-color: #e2f4dd;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row7_col1, #T_4e9ac_row11_col5 {\n",
       "  background-color: #e4f5df;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row7_col2 {\n",
       "  background-color: #0a7633;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_4e9ac_row7_col5 {\n",
       "  background-color: #a2d99c;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row7_col8, #T_4e9ac_row14_col9 {\n",
       "  background-color: #9cd797;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row7_col10 {\n",
       "  background-color: #359e53;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_4e9ac_row7_col11 {\n",
       "  background-color: #f5fbf2;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row7_col12 {\n",
       "  background-color: #b1e0ab;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row7_col13 {\n",
       "  background-color: #62bb6d;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_4e9ac_row8_col1, #T_4e9ac_row12_col5, #T_4e9ac_row13_col9 {\n",
       "  background-color: #d2edcc;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row8_col6, #T_4e9ac_row12_col2 {\n",
       "  background-color: #7fc97f;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row8_col7 {\n",
       "  background-color: #88ce87;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row8_col12, #T_4e9ac_row10_col3 {\n",
       "  background-color: #ceecc8;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row8_col13 {\n",
       "  background-color: #5bb86a;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_4e9ac_row8_col14, #T_4e9ac_row10_col13 {\n",
       "  background-color: #d4eece;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row9_col1, #T_4e9ac_row12_col1 {\n",
       "  background-color: #c2e7bb;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row9_col3 {\n",
       "  background-color: #e7f6e2;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row9_col4 {\n",
       "  background-color: #e8f6e4;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row9_col6 {\n",
       "  background-color: #80ca80;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row9_col11, #T_4e9ac_row12_col11, #T_4e9ac_row14_col11 {\n",
       "  background-color: #f5fbf3;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row9_col13, #T_4e9ac_row12_col4 {\n",
       "  background-color: #b7e2b1;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row10_col2 {\n",
       "  background-color: #2f974e;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_4e9ac_row10_col4 {\n",
       "  background-color: #72c375;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row11_col3, #T_4e9ac_row13_col1 {\n",
       "  background-color: #dcf2d7;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row11_col4 {\n",
       "  background-color: #cfecc9;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row11_col6 {\n",
       "  background-color: #b2e0ac;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row11_col9 {\n",
       "  background-color: #b6e2af;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row11_col10 {\n",
       "  background-color: #bbe4b4;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row12_col6 {\n",
       "  background-color: #c9eac2;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row12_col7 {\n",
       "  background-color: #8dd08a;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row12_col9 {\n",
       "  background-color: #8bcf89;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row12_col10 {\n",
       "  background-color: #c0e6b9;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row13_col2 {\n",
       "  background-color: #7dc87e;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row13_col5 {\n",
       "  background-color: #68be70;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row13_col6 {\n",
       "  background-color: #afdfa8;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row13_col7 {\n",
       "  background-color: #79c67a;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row13_col8 {\n",
       "  background-color: #84cc83;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row14_col1 {\n",
       "  background-color: #d0edca;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row14_col3 {\n",
       "  background-color: #e5f5e0;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_4e9ac_row14_col10 {\n",
       "  background-color: #4db163;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_4e9ac_row14_col12 {\n",
       "  background-color: #d9f0d3;\n",
       "  color: #000000;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_4e9ac\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_4e9ac_level0_col0\" class=\"col_heading level0 col0\" >Account Length</th>\n",
       "      <th id=\"T_4e9ac_level0_col1\" class=\"col_heading level0 col1\" >VMail Message</th>\n",
       "      <th id=\"T_4e9ac_level0_col2\" class=\"col_heading level0 col2\" >Day Mins</th>\n",
       "      <th id=\"T_4e9ac_level0_col3\" class=\"col_heading level0 col3\" >Day Calls</th>\n",
       "      <th id=\"T_4e9ac_level0_col4\" class=\"col_heading level0 col4\" >Day Charge</th>\n",
       "      <th id=\"T_4e9ac_level0_col5\" class=\"col_heading level0 col5\" >Eve Mins</th>\n",
       "      <th id=\"T_4e9ac_level0_col6\" class=\"col_heading level0 col6\" >Eve Calls</th>\n",
       "      <th id=\"T_4e9ac_level0_col7\" class=\"col_heading level0 col7\" >Eve Charge</th>\n",
       "      <th id=\"T_4e9ac_level0_col8\" class=\"col_heading level0 col8\" >Night Mins</th>\n",
       "      <th id=\"T_4e9ac_level0_col9\" class=\"col_heading level0 col9\" >Night Calls</th>\n",
       "      <th id=\"T_4e9ac_level0_col10\" class=\"col_heading level0 col10\" >Night Charge</th>\n",
       "      <th id=\"T_4e9ac_level0_col11\" class=\"col_heading level0 col11\" >Intl Mins</th>\n",
       "      <th id=\"T_4e9ac_level0_col12\" class=\"col_heading level0 col12\" >Intl Calls</th>\n",
       "      <th id=\"T_4e9ac_level0_col13\" class=\"col_heading level0 col13\" >Intl Charge</th>\n",
       "      <th id=\"T_4e9ac_level0_col14\" class=\"col_heading level0 col14\" >CustServ Calls</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_4e9ac_level0_row0\" class=\"row_heading level0 row0\" >Account Length</th>\n",
       "      <td id=\"T_4e9ac_row0_col0\" class=\"data row0 col0\" >1.00</td>\n",
       "      <td id=\"T_4e9ac_row0_col1\" class=\"data row0 col1\" >-0.01</td>\n",
       "      <td id=\"T_4e9ac_row0_col2\" class=\"data row0 col2\" >-0.02</td>\n",
       "      <td id=\"T_4e9ac_row0_col3\" class=\"data row0 col3\" >0.01</td>\n",
       "      <td id=\"T_4e9ac_row0_col4\" class=\"data row0 col4\" >-0.01</td>\n",
       "      <td id=\"T_4e9ac_row0_col5\" class=\"data row0 col5\" >0.00</td>\n",
       "      <td id=\"T_4e9ac_row0_col6\" class=\"data row0 col6\" >0.03</td>\n",
       "      <td id=\"T_4e9ac_row0_col7\" class=\"data row0 col7\" >-0.01</td>\n",
       "      <td id=\"T_4e9ac_row0_col8\" class=\"data row0 col8\" >0.02</td>\n",
       "      <td id=\"T_4e9ac_row0_col9\" class=\"data row0 col9\" >-0.00</td>\n",
       "      <td id=\"T_4e9ac_row0_col10\" class=\"data row0 col10\" >-0.03</td>\n",
       "      <td id=\"T_4e9ac_row0_col11\" class=\"data row0 col11\" >0.02</td>\n",
       "      <td id=\"T_4e9ac_row0_col12\" class=\"data row0 col12\" >-0.00</td>\n",
       "      <td id=\"T_4e9ac_row0_col13\" class=\"data row0 col13\" >0.03</td>\n",
       "      <td id=\"T_4e9ac_row0_col14\" class=\"data row0 col14\" >-0.04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4e9ac_level0_row1\" class=\"row_heading level0 row1\" >VMail Message</th>\n",
       "      <td id=\"T_4e9ac_row1_col0\" class=\"data row1 col0\" >-0.01</td>\n",
       "      <td id=\"T_4e9ac_row1_col1\" class=\"data row1 col1\" >1.00</td>\n",
       "      <td id=\"T_4e9ac_row1_col2\" class=\"data row1 col2\" >-0.14</td>\n",
       "      <td id=\"T_4e9ac_row1_col3\" class=\"data row1 col3\" >0.00</td>\n",
       "      <td id=\"T_4e9ac_row1_col4\" class=\"data row1 col4\" >-0.18</td>\n",
       "      <td id=\"T_4e9ac_row1_col5\" class=\"data row1 col5\" >-0.10</td>\n",
       "      <td id=\"T_4e9ac_row1_col6\" class=\"data row1 col6\" >-0.10</td>\n",
       "      <td id=\"T_4e9ac_row1_col7\" class=\"data row1 col7\" >-0.03</td>\n",
       "      <td id=\"T_4e9ac_row1_col8\" class=\"data row1 col8\" >0.06</td>\n",
       "      <td id=\"T_4e9ac_row1_col9\" class=\"data row1 col9\" >0.14</td>\n",
       "      <td id=\"T_4e9ac_row1_col10\" class=\"data row1 col10\" >-0.16</td>\n",
       "      <td id=\"T_4e9ac_row1_col11\" class=\"data row1 col11\" >-0.02</td>\n",
       "      <td id=\"T_4e9ac_row1_col12\" class=\"data row1 col12\" >0.13</td>\n",
       "      <td id=\"T_4e9ac_row1_col13\" class=\"data row1 col13\" >0.01</td>\n",
       "      <td id=\"T_4e9ac_row1_col14\" class=\"data row1 col14\" >0.07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4e9ac_level0_row2\" class=\"row_heading level0 row2\" >Day Mins</th>\n",
       "      <td id=\"T_4e9ac_row2_col0\" class=\"data row2 col0\" >-0.02</td>\n",
       "      <td id=\"T_4e9ac_row2_col1\" class=\"data row2 col1\" >-0.14</td>\n",
       "      <td id=\"T_4e9ac_row2_col2\" class=\"data row2 col2\" >1.00</td>\n",
       "      <td id=\"T_4e9ac_row2_col3\" class=\"data row2 col3\" >-0.09</td>\n",
       "      <td id=\"T_4e9ac_row2_col4\" class=\"data row2 col4\" >0.67</td>\n",
       "      <td id=\"T_4e9ac_row2_col5\" class=\"data row2 col5\" >0.48</td>\n",
       "      <td id=\"T_4e9ac_row2_col6\" class=\"data row2 col6\" >-0.18</td>\n",
       "      <td id=\"T_4e9ac_row2_col7\" class=\"data row2 col7\" >0.77</td>\n",
       "      <td id=\"T_4e9ac_row2_col8\" class=\"data row2 col8\" >0.19</td>\n",
       "      <td id=\"T_4e9ac_row2_col9\" class=\"data row2 col9\" >-0.45</td>\n",
       "      <td id=\"T_4e9ac_row2_col10\" class=\"data row2 col10\" >0.57</td>\n",
       "      <td id=\"T_4e9ac_row2_col11\" class=\"data row2 col11\" >0.00</td>\n",
       "      <td id=\"T_4e9ac_row2_col12\" class=\"data row2 col12\" >0.24</td>\n",
       "      <td id=\"T_4e9ac_row2_col13\" class=\"data row2 col13\" >0.24</td>\n",
       "      <td id=\"T_4e9ac_row2_col14\" class=\"data row2 col14\" >-0.20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4e9ac_level0_row3\" class=\"row_heading level0 row3\" >Day Calls</th>\n",
       "      <td id=\"T_4e9ac_row3_col0\" class=\"data row3 col0\" >0.01</td>\n",
       "      <td id=\"T_4e9ac_row3_col1\" class=\"data row3 col1\" >0.00</td>\n",
       "      <td id=\"T_4e9ac_row3_col2\" class=\"data row3 col2\" >-0.09</td>\n",
       "      <td id=\"T_4e9ac_row3_col3\" class=\"data row3 col3\" >1.00</td>\n",
       "      <td id=\"T_4e9ac_row3_col4\" class=\"data row3 col4\" >-0.22</td>\n",
       "      <td id=\"T_4e9ac_row3_col5\" class=\"data row3 col5\" >0.03</td>\n",
       "      <td id=\"T_4e9ac_row3_col6\" class=\"data row3 col6\" >0.19</td>\n",
       "      <td id=\"T_4e9ac_row3_col7\" class=\"data row3 col7\" >-0.05</td>\n",
       "      <td id=\"T_4e9ac_row3_col8\" class=\"data row3 col8\" >-0.09</td>\n",
       "      <td id=\"T_4e9ac_row3_col9\" class=\"data row3 col9\" >-0.08</td>\n",
       "      <td id=\"T_4e9ac_row3_col10\" class=\"data row3 col10\" >0.05</td>\n",
       "      <td id=\"T_4e9ac_row3_col11\" class=\"data row3 col11\" >-0.02</td>\n",
       "      <td id=\"T_4e9ac_row3_col12\" class=\"data row3 col12\" >-0.05</td>\n",
       "      <td id=\"T_4e9ac_row3_col13\" class=\"data row3 col13\" >-0.12</td>\n",
       "      <td id=\"T_4e9ac_row3_col14\" class=\"data row3 col14\" >-0.07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4e9ac_level0_row4\" class=\"row_heading level0 row4\" >Day Charge</th>\n",
       "      <td id=\"T_4e9ac_row4_col0\" class=\"data row4 col0\" >-0.01</td>\n",
       "      <td id=\"T_4e9ac_row4_col1\" class=\"data row4 col1\" >-0.18</td>\n",
       "      <td id=\"T_4e9ac_row4_col2\" class=\"data row4 col2\" >0.67</td>\n",
       "      <td id=\"T_4e9ac_row4_col3\" class=\"data row4 col3\" >-0.22</td>\n",
       "      <td id=\"T_4e9ac_row4_col4\" class=\"data row4 col4\" >1.00</td>\n",
       "      <td id=\"T_4e9ac_row4_col5\" class=\"data row4 col5\" >0.57</td>\n",
       "      <td id=\"T_4e9ac_row4_col6\" class=\"data row4 col6\" >0.24</td>\n",
       "      <td id=\"T_4e9ac_row4_col7\" class=\"data row4 col7\" >0.37</td>\n",
       "      <td id=\"T_4e9ac_row4_col8\" class=\"data row4 col8\" >0.15</td>\n",
       "      <td id=\"T_4e9ac_row4_col9\" class=\"data row4 col9\" >-0.13</td>\n",
       "      <td id=\"T_4e9ac_row4_col10\" class=\"data row4 col10\" >0.37</td>\n",
       "      <td id=\"T_4e9ac_row4_col11\" class=\"data row4 col11\" >0.01</td>\n",
       "      <td id=\"T_4e9ac_row4_col12\" class=\"data row4 col12\" >0.12</td>\n",
       "      <td id=\"T_4e9ac_row4_col13\" class=\"data row4 col13\" >0.25</td>\n",
       "      <td id=\"T_4e9ac_row4_col14\" class=\"data row4 col14\" >-0.26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4e9ac_level0_row5\" class=\"row_heading level0 row5\" >Eve Mins</th>\n",
       "      <td id=\"T_4e9ac_row5_col0\" class=\"data row5 col0\" >0.00</td>\n",
       "      <td id=\"T_4e9ac_row5_col1\" class=\"data row5 col1\" >-0.10</td>\n",
       "      <td id=\"T_4e9ac_row5_col2\" class=\"data row5 col2\" >0.48</td>\n",
       "      <td id=\"T_4e9ac_row5_col3\" class=\"data row5 col3\" >0.03</td>\n",
       "      <td id=\"T_4e9ac_row5_col4\" class=\"data row5 col4\" >0.57</td>\n",
       "      <td id=\"T_4e9ac_row5_col5\" class=\"data row5 col5\" >1.00</td>\n",
       "      <td id=\"T_4e9ac_row5_col6\" class=\"data row5 col6\" >-0.07</td>\n",
       "      <td id=\"T_4e9ac_row5_col7\" class=\"data row5 col7\" >0.27</td>\n",
       "      <td id=\"T_4e9ac_row5_col8\" class=\"data row5 col8\" >-0.09</td>\n",
       "      <td id=\"T_4e9ac_row5_col9\" class=\"data row5 col9\" >0.07</td>\n",
       "      <td id=\"T_4e9ac_row5_col10\" class=\"data row5 col10\" >0.32</td>\n",
       "      <td id=\"T_4e9ac_row5_col11\" class=\"data row5 col11\" >-0.02</td>\n",
       "      <td id=\"T_4e9ac_row5_col12\" class=\"data row5 col12\" >0.07</td>\n",
       "      <td id=\"T_4e9ac_row5_col13\" class=\"data row5 col13\" >0.45</td>\n",
       "      <td id=\"T_4e9ac_row5_col14\" class=\"data row5 col14\" >-0.17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4e9ac_level0_row6\" class=\"row_heading level0 row6\" >Eve Calls</th>\n",
       "      <td id=\"T_4e9ac_row6_col0\" class=\"data row6 col0\" >0.03</td>\n",
       "      <td id=\"T_4e9ac_row6_col1\" class=\"data row6 col1\" >-0.10</td>\n",
       "      <td id=\"T_4e9ac_row6_col2\" class=\"data row6 col2\" >-0.18</td>\n",
       "      <td id=\"T_4e9ac_row6_col3\" class=\"data row6 col3\" >0.19</td>\n",
       "      <td id=\"T_4e9ac_row6_col4\" class=\"data row6 col4\" >0.24</td>\n",
       "      <td id=\"T_4e9ac_row6_col5\" class=\"data row6 col5\" >-0.07</td>\n",
       "      <td id=\"T_4e9ac_row6_col6\" class=\"data row6 col6\" >1.00</td>\n",
       "      <td id=\"T_4e9ac_row6_col7\" class=\"data row6 col7\" >-0.47</td>\n",
       "      <td id=\"T_4e9ac_row6_col8\" class=\"data row6 col8\" >0.22</td>\n",
       "      <td id=\"T_4e9ac_row6_col9\" class=\"data row6 col9\" >0.22</td>\n",
       "      <td id=\"T_4e9ac_row6_col10\" class=\"data row6 col10\" >-0.32</td>\n",
       "      <td id=\"T_4e9ac_row6_col11\" class=\"data row6 col11\" >-0.00</td>\n",
       "      <td id=\"T_4e9ac_row6_col12\" class=\"data row6 col12\" >-0.11</td>\n",
       "      <td id=\"T_4e9ac_row6_col13\" class=\"data row6 col13\" >0.02</td>\n",
       "      <td id=\"T_4e9ac_row6_col14\" class=\"data row6 col14\" >-0.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4e9ac_level0_row7\" class=\"row_heading level0 row7\" >Eve Charge</th>\n",
       "      <td id=\"T_4e9ac_row7_col0\" class=\"data row7 col0\" >-0.01</td>\n",
       "      <td id=\"T_4e9ac_row7_col1\" class=\"data row7 col1\" >-0.03</td>\n",
       "      <td id=\"T_4e9ac_row7_col2\" class=\"data row7 col2\" >0.77</td>\n",
       "      <td id=\"T_4e9ac_row7_col3\" class=\"data row7 col3\" >-0.05</td>\n",
       "      <td id=\"T_4e9ac_row7_col4\" class=\"data row7 col4\" >0.37</td>\n",
       "      <td id=\"T_4e9ac_row7_col5\" class=\"data row7 col5\" >0.27</td>\n",
       "      <td id=\"T_4e9ac_row7_col6\" class=\"data row7 col6\" >-0.47</td>\n",
       "      <td id=\"T_4e9ac_row7_col7\" class=\"data row7 col7\" >1.00</td>\n",
       "      <td id=\"T_4e9ac_row7_col8\" class=\"data row7 col8\" >0.18</td>\n",
       "      <td id=\"T_4e9ac_row7_col9\" class=\"data row7 col9\" >-0.45</td>\n",
       "      <td id=\"T_4e9ac_row7_col10\" class=\"data row7 col10\" >0.55</td>\n",
       "      <td id=\"T_4e9ac_row7_col11\" class=\"data row7 col11\" >-0.00</td>\n",
       "      <td id=\"T_4e9ac_row7_col12\" class=\"data row7 col12\" >0.16</td>\n",
       "      <td id=\"T_4e9ac_row7_col13\" class=\"data row7 col13\" >0.24</td>\n",
       "      <td id=\"T_4e9ac_row7_col14\" class=\"data row7 col14\" >-0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4e9ac_level0_row8\" class=\"row_heading level0 row8\" >Night Mins</th>\n",
       "      <td id=\"T_4e9ac_row8_col0\" class=\"data row8 col0\" >0.02</td>\n",
       "      <td id=\"T_4e9ac_row8_col1\" class=\"data row8 col1\" >0.06</td>\n",
       "      <td id=\"T_4e9ac_row8_col2\" class=\"data row8 col2\" >0.19</td>\n",
       "      <td id=\"T_4e9ac_row8_col3\" class=\"data row8 col3\" >-0.09</td>\n",
       "      <td id=\"T_4e9ac_row8_col4\" class=\"data row8 col4\" >0.15</td>\n",
       "      <td id=\"T_4e9ac_row8_col5\" class=\"data row8 col5\" >-0.09</td>\n",
       "      <td id=\"T_4e9ac_row8_col6\" class=\"data row8 col6\" >0.22</td>\n",
       "      <td id=\"T_4e9ac_row8_col7\" class=\"data row8 col7\" >0.18</td>\n",
       "      <td id=\"T_4e9ac_row8_col8\" class=\"data row8 col8\" >1.00</td>\n",
       "      <td id=\"T_4e9ac_row8_col9\" class=\"data row8 col9\" >-0.22</td>\n",
       "      <td id=\"T_4e9ac_row8_col10\" class=\"data row8 col10\" >-0.14</td>\n",
       "      <td id=\"T_4e9ac_row8_col11\" class=\"data row8 col11\" >-0.01</td>\n",
       "      <td id=\"T_4e9ac_row8_col12\" class=\"data row8 col12\" >0.04</td>\n",
       "      <td id=\"T_4e9ac_row8_col13\" class=\"data row8 col13\" >0.27</td>\n",
       "      <td id=\"T_4e9ac_row8_col14\" class=\"data row8 col14\" >-0.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4e9ac_level0_row9\" class=\"row_heading level0 row9\" >Night Calls</th>\n",
       "      <td id=\"T_4e9ac_row9_col0\" class=\"data row9 col0\" >-0.00</td>\n",
       "      <td id=\"T_4e9ac_row9_col1\" class=\"data row9 col1\" >0.14</td>\n",
       "      <td id=\"T_4e9ac_row9_col2\" class=\"data row9 col2\" >-0.45</td>\n",
       "      <td id=\"T_4e9ac_row9_col3\" class=\"data row9 col3\" >-0.08</td>\n",
       "      <td id=\"T_4e9ac_row9_col4\" class=\"data row9 col4\" >-0.13</td>\n",
       "      <td id=\"T_4e9ac_row9_col5\" class=\"data row9 col5\" >0.07</td>\n",
       "      <td id=\"T_4e9ac_row9_col6\" class=\"data row9 col6\" >0.22</td>\n",
       "      <td id=\"T_4e9ac_row9_col7\" class=\"data row9 col7\" >-0.45</td>\n",
       "      <td id=\"T_4e9ac_row9_col8\" class=\"data row9 col8\" >-0.22</td>\n",
       "      <td id=\"T_4e9ac_row9_col9\" class=\"data row9 col9\" >1.00</td>\n",
       "      <td id=\"T_4e9ac_row9_col10\" class=\"data row9 col10\" >-0.39</td>\n",
       "      <td id=\"T_4e9ac_row9_col11\" class=\"data row9 col11\" >-0.01</td>\n",
       "      <td id=\"T_4e9ac_row9_col12\" class=\"data row9 col12\" >0.18</td>\n",
       "      <td id=\"T_4e9ac_row9_col13\" class=\"data row9 col13\" >-0.16</td>\n",
       "      <td id=\"T_4e9ac_row9_col14\" class=\"data row9 col14\" >0.11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4e9ac_level0_row10\" class=\"row_heading level0 row10\" >Night Charge</th>\n",
       "      <td id=\"T_4e9ac_row10_col0\" class=\"data row10 col0\" >-0.03</td>\n",
       "      <td id=\"T_4e9ac_row10_col1\" class=\"data row10 col1\" >-0.16</td>\n",
       "      <td id=\"T_4e9ac_row10_col2\" class=\"data row10 col2\" >0.57</td>\n",
       "      <td id=\"T_4e9ac_row10_col3\" class=\"data row10 col3\" >0.05</td>\n",
       "      <td id=\"T_4e9ac_row10_col4\" class=\"data row10 col4\" >0.37</td>\n",
       "      <td id=\"T_4e9ac_row10_col5\" class=\"data row10 col5\" >0.32</td>\n",
       "      <td id=\"T_4e9ac_row10_col6\" class=\"data row10 col6\" >-0.32</td>\n",
       "      <td id=\"T_4e9ac_row10_col7\" class=\"data row10 col7\" >0.55</td>\n",
       "      <td id=\"T_4e9ac_row10_col8\" class=\"data row10 col8\" >-0.14</td>\n",
       "      <td id=\"T_4e9ac_row10_col9\" class=\"data row10 col9\" >-0.39</td>\n",
       "      <td id=\"T_4e9ac_row10_col10\" class=\"data row10 col10\" >1.00</td>\n",
       "      <td id=\"T_4e9ac_row10_col11\" class=\"data row10 col11\" >0.01</td>\n",
       "      <td id=\"T_4e9ac_row10_col12\" class=\"data row10 col12\" >-0.01</td>\n",
       "      <td id=\"T_4e9ac_row10_col13\" class=\"data row10 col13\" >-0.33</td>\n",
       "      <td id=\"T_4e9ac_row10_col14\" class=\"data row10 col14\" >0.44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4e9ac_level0_row11\" class=\"row_heading level0 row11\" >Intl Mins</th>\n",
       "      <td id=\"T_4e9ac_row11_col0\" class=\"data row11 col0\" >0.02</td>\n",
       "      <td id=\"T_4e9ac_row11_col1\" class=\"data row11 col1\" >-0.02</td>\n",
       "      <td id=\"T_4e9ac_row11_col2\" class=\"data row11 col2\" >0.00</td>\n",
       "      <td id=\"T_4e9ac_row11_col3\" class=\"data row11 col3\" >-0.02</td>\n",
       "      <td id=\"T_4e9ac_row11_col4\" class=\"data row11 col4\" >0.01</td>\n",
       "      <td id=\"T_4e9ac_row11_col5\" class=\"data row11 col5\" >-0.02</td>\n",
       "      <td id=\"T_4e9ac_row11_col6\" class=\"data row11 col6\" >-0.00</td>\n",
       "      <td id=\"T_4e9ac_row11_col7\" class=\"data row11 col7\" >-0.00</td>\n",
       "      <td id=\"T_4e9ac_row11_col8\" class=\"data row11 col8\" >-0.01</td>\n",
       "      <td id=\"T_4e9ac_row11_col9\" class=\"data row11 col9\" >-0.01</td>\n",
       "      <td id=\"T_4e9ac_row11_col10\" class=\"data row11 col10\" >0.01</td>\n",
       "      <td id=\"T_4e9ac_row11_col11\" class=\"data row11 col11\" >1.00</td>\n",
       "      <td id=\"T_4e9ac_row11_col12\" class=\"data row11 col12\" >-0.01</td>\n",
       "      <td id=\"T_4e9ac_row11_col13\" class=\"data row11 col13\" >-0.01</td>\n",
       "      <td id=\"T_4e9ac_row11_col14\" class=\"data row11 col14\" >-0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4e9ac_level0_row12\" class=\"row_heading level0 row12\" >Intl Calls</th>\n",
       "      <td id=\"T_4e9ac_row12_col0\" class=\"data row12 col0\" >-0.00</td>\n",
       "      <td id=\"T_4e9ac_row12_col1\" class=\"data row12 col1\" >0.13</td>\n",
       "      <td id=\"T_4e9ac_row12_col2\" class=\"data row12 col2\" >0.24</td>\n",
       "      <td id=\"T_4e9ac_row12_col3\" class=\"data row12 col3\" >-0.05</td>\n",
       "      <td id=\"T_4e9ac_row12_col4\" class=\"data row12 col4\" >0.12</td>\n",
       "      <td id=\"T_4e9ac_row12_col5\" class=\"data row12 col5\" >0.07</td>\n",
       "      <td id=\"T_4e9ac_row12_col6\" class=\"data row12 col6\" >-0.11</td>\n",
       "      <td id=\"T_4e9ac_row12_col7\" class=\"data row12 col7\" >0.16</td>\n",
       "      <td id=\"T_4e9ac_row12_col8\" class=\"data row12 col8\" >0.04</td>\n",
       "      <td id=\"T_4e9ac_row12_col9\" class=\"data row12 col9\" >0.18</td>\n",
       "      <td id=\"T_4e9ac_row12_col10\" class=\"data row12 col10\" >-0.01</td>\n",
       "      <td id=\"T_4e9ac_row12_col11\" class=\"data row12 col11\" >-0.01</td>\n",
       "      <td id=\"T_4e9ac_row12_col12\" class=\"data row12 col12\" >1.00</td>\n",
       "      <td id=\"T_4e9ac_row12_col13\" class=\"data row12 col13\" >-0.23</td>\n",
       "      <td id=\"T_4e9ac_row12_col14\" class=\"data row12 col14\" >-0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4e9ac_level0_row13\" class=\"row_heading level0 row13\" >Intl Charge</th>\n",
       "      <td id=\"T_4e9ac_row13_col0\" class=\"data row13 col0\" >0.03</td>\n",
       "      <td id=\"T_4e9ac_row13_col1\" class=\"data row13 col1\" >0.01</td>\n",
       "      <td id=\"T_4e9ac_row13_col2\" class=\"data row13 col2\" >0.24</td>\n",
       "      <td id=\"T_4e9ac_row13_col3\" class=\"data row13 col3\" >-0.12</td>\n",
       "      <td id=\"T_4e9ac_row13_col4\" class=\"data row13 col4\" >0.25</td>\n",
       "      <td id=\"T_4e9ac_row13_col5\" class=\"data row13 col5\" >0.45</td>\n",
       "      <td id=\"T_4e9ac_row13_col6\" class=\"data row13 col6\" >0.02</td>\n",
       "      <td id=\"T_4e9ac_row13_col7\" class=\"data row13 col7\" >0.24</td>\n",
       "      <td id=\"T_4e9ac_row13_col8\" class=\"data row13 col8\" >0.27</td>\n",
       "      <td id=\"T_4e9ac_row13_col9\" class=\"data row13 col9\" >-0.16</td>\n",
       "      <td id=\"T_4e9ac_row13_col10\" class=\"data row13 col10\" >-0.33</td>\n",
       "      <td id=\"T_4e9ac_row13_col11\" class=\"data row13 col11\" >-0.01</td>\n",
       "      <td id=\"T_4e9ac_row13_col12\" class=\"data row13 col12\" >-0.23</td>\n",
       "      <td id=\"T_4e9ac_row13_col13\" class=\"data row13 col13\" >1.00</td>\n",
       "      <td id=\"T_4e9ac_row13_col14\" class=\"data row13 col14\" >-0.66</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4e9ac_level0_row14\" class=\"row_heading level0 row14\" >CustServ Calls</th>\n",
       "      <td id=\"T_4e9ac_row14_col0\" class=\"data row14 col0\" >-0.04</td>\n",
       "      <td id=\"T_4e9ac_row14_col1\" class=\"data row14 col1\" >0.07</td>\n",
       "      <td id=\"T_4e9ac_row14_col2\" class=\"data row14 col2\" >-0.20</td>\n",
       "      <td id=\"T_4e9ac_row14_col3\" class=\"data row14 col3\" >-0.07</td>\n",
       "      <td id=\"T_4e9ac_row14_col4\" class=\"data row14 col4\" >-0.26</td>\n",
       "      <td id=\"T_4e9ac_row14_col5\" class=\"data row14 col5\" >-0.17</td>\n",
       "      <td id=\"T_4e9ac_row14_col6\" class=\"data row14 col6\" >-0.43</td>\n",
       "      <td id=\"T_4e9ac_row14_col7\" class=\"data row14 col7\" >-0.01</td>\n",
       "      <td id=\"T_4e9ac_row14_col8\" class=\"data row14 col8\" >-0.33</td>\n",
       "      <td id=\"T_4e9ac_row14_col9\" class=\"data row14 col9\" >0.11</td>\n",
       "      <td id=\"T_4e9ac_row14_col10\" class=\"data row14 col10\" >0.44</td>\n",
       "      <td id=\"T_4e9ac_row14_col11\" class=\"data row14 col11\" >-0.01</td>\n",
       "      <td id=\"T_4e9ac_row14_col12\" class=\"data row14 col12\" >-0.01</td>\n",
       "      <td id=\"T_4e9ac_row14_col13\" class=\"data row14 col13\" >-0.66</td>\n",
       "      <td id=\"T_4e9ac_row14_col14\" class=\"data row14 col14\" >1.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x246d58cdc60>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corr = churn.corr()\n",
    "corr.style.background_gradient(cmap=\"Greens\").set_precision(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "いくつかの変数は互いに100%の相関をもっています。このような同じ傾向を示す変数があるとき、機械学習のアルゴリズムによっては全くうまくいかないことがあり、そうでなくても結果が偏ったりしてしまうことがあります。これらの相関の強いペアは削除することが好ましいです。\n",
    "\n",
    "`Day Charge` に対する `Day Mins`、`Night Charge` に対する `Night Mins`、`Intl Charge` に対する `Intl Mins` を削除します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "churn = churn.drop(['Day Mins', 'Eve Mins', 'Night Mins', 'Intl Mins'], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "データ分析の過程で変数の削除や修正を行いました。この作業が妥当かどうかについては、データを扱っているアナリストや業務に詳しドメインエキスパートにヒアリングすべきでしょう。より詳しいデータ分析プロセスの進め方は [データから価値を創出できるか診断する](https://youtu.be/iYV4slOPoYE) をご参照ください。\n",
    "\n",
    "ここまででデータの分析は完了です。\n",
    "\n",
    "### Key takeaways\n",
    "\n",
    "* データはカテゴリ変数、量的変数に分けて分析を行う。\n",
    "* データの頻度、目的変数との相関を可視化することはデータから予測ができるか知ることに役立つ。\n",
    "* 相関の高い項目があると学習が困難になるため、確認の上削除する。\n",
    "\n",
    "---\n",
    "## 3.データ充足 : Prepare\n",
    "\n",
    "分析した結果、データの質・量が足りないことがわかった場合、データ充足のプロセスでデータの収集と作成を行います。不十分なデータでモデルの構築をしても、不十分な結果しか得られないからです。データの作成は画像に対するラベルづけなどの作業、データの収集はラベル付するためのデータを集める作業です。データの収集を行うのであればデータアーキテクト、データの作成をするなら業務に詳しいドメインエキスパートと協力する必要があるでしょう。\n",
    "\n",
    "**Discussion**\n",
    "\n",
    "*解約率の予測に関わるデータを収集、作成するのにどんなチームの協力が必要でしょうか?*\n",
    "\n",
    "今回はデータはすでに整備されているとして前処理に進みます。\n",
    "\n",
    "---\n",
    "## 4.前処理 : Preprocess\n",
    "\n",
    "データ充足で用意したデータを、機械学習モデルが認識しやすいデータに変換します。機械学習モデルは数値しか扱えないため、カテゴリ変数は量的変数に変換する必要があります。 `get_dummies()` はカテゴリ変数を量的変数へ変換する処理の一つです。 `Int'l Plan` は `yes` と `no` 2 つの値を持っていますが、その値ごとに該当する場合は 1 、そうでない場合 0 とする量的変数を作ります。これを One-hot エンコーディングと呼びます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>no</th>\n",
       "      <th>yes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4995</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4996</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4997</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4998</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4999</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5000 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      no  yes\n",
       "0      1    0\n",
       "1      0    1\n",
       "2      1    0\n",
       "3      0    1\n",
       "4      0    1\n",
       "...   ..  ...\n",
       "4995   0    1\n",
       "4996   1    0\n",
       "4997   1    0\n",
       "4998   0    1\n",
       "4999   0    1\n",
       "\n",
       "[5000 rows x 2 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.get_dummies(churn[\"Int'l Plan\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`get_dummies()` を使い、目的変数である `Churn?` 以外のすべての項目に対し One-hot エンコーディングを行います。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\tools\\miniconda3\\envs\\ml-handson-churn\\lib\\site-packages\\pandas\\core\\algorithms.py:798: FutureWarning: In a future version, the Index constructor will not infer numeric dtypes when passed object-dtype sequences (matching Series behavior)\n",
      "  uniques = Index(uniques)\n"
     ]
    }
   ],
   "source": [
    "model_data = pd.get_dummies(churn.loc[:, churn.columns != \"Churn?\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>Churn?</th>\n",
       "      <th>Account Length</th>\n",
       "      <th>VMail Message</th>\n",
       "      <th>Day Calls</th>\n",
       "      <th>Day Charge</th>\n",
       "      <th>Eve Calls</th>\n",
       "      <th>Eve Charge</th>\n",
       "      <th>Night Calls</th>\n",
       "      <th>Night Charge</th>\n",
       "      <th>Intl Calls</th>\n",
       "      <th>...</th>\n",
       "      <th>Area Code_858</th>\n",
       "      <th>Area Code_866</th>\n",
       "      <th>Area Code_868</th>\n",
       "      <th>Area Code_876</th>\n",
       "      <th>Area Code_877</th>\n",
       "      <th>Area Code_878</th>\n",
       "      <th>Int'l Plan_no</th>\n",
       "      <th>Int'l Plan_yes</th>\n",
       "      <th>VMail Plan_no</th>\n",
       "      <th>VMail Plan_yes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>163</td>\n",
       "      <td>300</td>\n",
       "      <td>3</td>\n",
       "      <td>7.579174</td>\n",
       "      <td>4</td>\n",
       "      <td>6.508639</td>\n",
       "      <td>100</td>\n",
       "      <td>5.111624</td>\n",
       "      <td>6</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>15</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>4.226289</td>\n",
       "      <td>0</td>\n",
       "      <td>9.972592</td>\n",
       "      <td>200</td>\n",
       "      <td>6.436188</td>\n",
       "      <td>6</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>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>131</td>\n",
       "      <td>300</td>\n",
       "      <td>3</td>\n",
       "      <td>4.768160</td>\n",
       "      <td>3</td>\n",
       "      <td>4.566715</td>\n",
       "      <td>100</td>\n",
       "      <td>5.142451</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>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>75</td>\n",
       "      <td>700</td>\n",
       "      <td>3</td>\n",
       "      <td>2.567642</td>\n",
       "      <td>5</td>\n",
       "      <td>2.333624</td>\n",
       "      <td>450</td>\n",
       "      <td>3.814413</td>\n",
       "      <td>6</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>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>146</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>5.908916</td>\n",
       "      <td>3</td>\n",
       "      <td>3.670408</td>\n",
       "      <td>250</td>\n",
       "      <td>2.796812</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>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4995</th>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>800</td>\n",
       "      <td>5</td>\n",
       "      <td>7.250969</td>\n",
       "      <td>1</td>\n",
       "      <td>8.026482</td>\n",
       "      <td>350</td>\n",
       "      <td>6.748489</td>\n",
       "      <td>8</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>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4996</th>\n",
       "      <td>0</td>\n",
       "      <td>140</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>3.758307</td>\n",
       "      <td>7</td>\n",
       "      <td>1.328827</td>\n",
       "      <td>300</td>\n",
       "      <td>4.522661</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>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4997</th>\n",
       "      <td>0</td>\n",
       "      <td>32</td>\n",
       "      <td>700</td>\n",
       "      <td>5</td>\n",
       "      <td>4.243980</td>\n",
       "      <td>3</td>\n",
       "      <td>2.340554</td>\n",
       "      <td>450</td>\n",
       "      <td>5.157898</td>\n",
       "      <td>7</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4998</th>\n",
       "      <td>1</td>\n",
       "      <td>142</td>\n",
       "      <td>600</td>\n",
       "      <td>5</td>\n",
       "      <td>3.014859</td>\n",
       "      <td>3</td>\n",
       "      <td>3.470372</td>\n",
       "      <td>150</td>\n",
       "      <td>4.362780</td>\n",
       "      <td>3</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>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4999</th>\n",
       "      <td>0</td>\n",
       "      <td>141</td>\n",
       "      <td>500</td>\n",
       "      <td>0</td>\n",
       "      <td>5.125716</td>\n",
       "      <td>0</td>\n",
       "      <td>2.109823</td>\n",
       "      <td>400</td>\n",
       "      <td>3.713631</td>\n",
       "      <td>6</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>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5000 rows × 100 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Churn?  Account Length  VMail Message  Day Calls  Day Charge  Eve Calls  \\\n",
       "0          1             163            300          3    7.579174          4   \n",
       "1          0              15              0          4    4.226289          0   \n",
       "2          0             131            300          3    4.768160          3   \n",
       "3          0              75            700          3    2.567642          5   \n",
       "4          1             146              0          3    5.908916          3   \n",
       "...      ...             ...            ...        ...         ...        ...   \n",
       "4995       0               4            800          5    7.250969          1   \n",
       "4996       0             140              0          8    3.758307          7   \n",
       "4997       0              32            700          5    4.243980          3   \n",
       "4998       1             142            600          5    3.014859          3   \n",
       "4999       0             141            500          0    5.125716          0   \n",
       "\n",
       "      Eve Charge  Night Calls  Night Charge  Intl Calls  ...  Area Code_858  \\\n",
       "0       6.508639          100      5.111624           6  ...              0   \n",
       "1       9.972592          200      6.436188           6  ...              0   \n",
       "2       4.566715          100      5.142451           2  ...              0   \n",
       "3       2.333624          450      3.814413           6  ...              0   \n",
       "4       3.670408          250      2.796812           4  ...              0   \n",
       "...          ...          ...           ...         ...  ...            ...   \n",
       "4995    8.026482          350      6.748489           8  ...              0   \n",
       "4996    1.328827          300      4.522661           2  ...              0   \n",
       "4997    2.340554          450      5.157898           7  ...              0   \n",
       "4998    3.470372          150      4.362780           3  ...              0   \n",
       "4999    2.109823          400      3.713631           6  ...              0   \n",
       "\n",
       "      Area Code_866  Area Code_868  Area Code_876  Area Code_877  \\\n",
       "0                 0              0              0              0   \n",
       "1                 0              0              0              0   \n",
       "2                 0              0              0              0   \n",
       "3                 0              0              0              0   \n",
       "4                 0              0              0              0   \n",
       "...             ...            ...            ...            ...   \n",
       "4995              0              0              0              0   \n",
       "4996              0              0              0              0   \n",
       "4997              0              0              0              0   \n",
       "4998              0              0              0              0   \n",
       "4999              0              0              0              0   \n",
       "\n",
       "      Area Code_878  Int'l Plan_no  Int'l Plan_yes  VMail Plan_no  \\\n",
       "0                 0              1               0              0   \n",
       "1                 0              0               1              1   \n",
       "2                 0              1               0              0   \n",
       "3                 1              0               1              0   \n",
       "4                 1              0               1              1   \n",
       "...             ...            ...             ...            ...   \n",
       "4995              0              0               1              0   \n",
       "4996              0              1               0              1   \n",
       "4997              0              1               0              0   \n",
       "4998              0              0               1              0   \n",
       "4999              0              0               1              0   \n",
       "\n",
       "      VMail Plan_yes  \n",
       "0                  1  \n",
       "1                  0  \n",
       "2                  1  \n",
       "3                  1  \n",
       "4                  0  \n",
       "...              ...  \n",
       "4995               1  \n",
       "4996               0  \n",
       "4997               1  \n",
       "4998               1  \n",
       "4999               1  \n",
       "\n",
       "[5000 rows x 100 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_data = pd.concat([churn[\"Churn?\"].apply(lambda x: 1 if x == \"True.\" else 0), model_data], axis=1)\n",
    "model_data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "データを学習用、バリデーション用、テスト用にわけておきます。これによって Overfitting (学習用データには精度が高いが、実際に利用すると精度が低い、といった状況) を検出しやすくなり、未知のテストデータに対する精度を確認することができます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_data, validation_data, test_data = np.split(model_data.sample(frac=1, random_state=1729), [int(0.7 * len(model_data)), int(0.9 * len(model_data))])\n",
    "train_data.to_csv(data_root.joinpath(\"interim/churn_train.csv\"), header=False, index=False)\n",
    "validation_data.to_csv(data_root.joinpath(\"interim/churn_validation.csv\"), header=False, index=False)\n",
    "test_data.to_csv(data_root.joinpath(\"interim/churn_test.csv\"), header=False, index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "学習には学習用データとバリデーション用データのみを使用し、テスト用データは評価に使用します。\n",
    "\n",
    "\n",
    "### Key takeaways\n",
    "\n",
    "* 機械学習モデルで扱える変数は数値のみのため、カテゴリ変数は量的変数に変換する。\n",
    "* データを学習用、検証用、テスト用に分けることで学習用データへの過剰な適合 ( Overfitting ) を検出できる。\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "## 5.学習 : Train\n",
    "\n",
    "ビジネス目標の達成に適したモデルを選択し、前処理したデータで学習します。解約する度合いを数値の大小で表すモデルが良さそうです。線形回帰のようなアルゴリズムでも予測が行えますが、今回は勾配ブースティング木 (Gradient Boosted Tree)を利用します。 XGBoost は、特徴間の非線形な関係を考慮した勾配ブースティング木を利用しており、特徴間の複雑な関連性を扱うことができます。\n",
    "\n",
    "それでは学習を始めましょう。まず、目的変数と目的変数を予測するための変数 ( **説明変数**と呼びます。実体は量的変数やカテゴリです )にデータを分けます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_x, train_y = train_data.drop(['Churn?'], axis=1), train_data['Churn?']\n",
    "test_x, test_y = test_data.drop(['Churn?'], axis=1), test_data['Churn?']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "XGBoost のモデルを作成し、学習を開始します"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;background-color: white;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>XGBClassifier(base_score=0.5, booster=&#x27;gbtree&#x27;, callbacks=None,\n",
       "              colsample_bylevel=1, colsample_bynode=1, colsample_bytree=1,\n",
       "              early_stopping_rounds=None, enable_categorical=False,\n",
       "              eval_metric=None, gamma=0, gpu_id=-1, grow_policy=&#x27;depthwise&#x27;,\n",
       "              importance_type=None, interaction_constraints=&#x27;&#x27;,\n",
       "              learning_rate=0.300000012, max_bin=256, max_cat_to_onehot=4,\n",
       "              max_delta_step=0, max_depth=6, max_leaves=0, min_child_weight=1,\n",
       "              missing=nan, monotone_constraints=&#x27;()&#x27;, n_estimators=100,\n",
       "              n_jobs=0, num_parallel_tree=1, predictor=&#x27;auto&#x27;, random_state=0,\n",
       "              reg_alpha=0, reg_lambda=1, ...)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">XGBClassifier</label><div class=\"sk-toggleable__content\"><pre>XGBClassifier(base_score=0.5, booster=&#x27;gbtree&#x27;, callbacks=None,\n",
       "              colsample_bylevel=1, colsample_bynode=1, colsample_bytree=1,\n",
       "              early_stopping_rounds=None, enable_categorical=False,\n",
       "              eval_metric=None, gamma=0, gpu_id=-1, grow_policy=&#x27;depthwise&#x27;,\n",
       "              importance_type=None, interaction_constraints=&#x27;&#x27;,\n",
       "              learning_rate=0.300000012, max_bin=256, max_cat_to_onehot=4,\n",
       "              max_delta_step=0, max_depth=6, max_leaves=0, min_child_weight=1,\n",
       "              missing=nan, monotone_constraints=&#x27;()&#x27;, n_estimators=100,\n",
       "              n_jobs=0, num_parallel_tree=1, predictor=&#x27;auto&#x27;, random_state=0,\n",
       "              reg_alpha=0, reg_lambda=1, ...)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "XGBClassifier(base_score=0.5, booster='gbtree', callbacks=None,\n",
       "              colsample_bylevel=1, colsample_bynode=1, colsample_bytree=1,\n",
       "              early_stopping_rounds=None, enable_categorical=False,\n",
       "              eval_metric=None, gamma=0, gpu_id=-1, grow_policy='depthwise',\n",
       "              importance_type=None, interaction_constraints='',\n",
       "              learning_rate=0.300000012, max_bin=256, max_cat_to_onehot=4,\n",
       "              max_delta_step=0, max_depth=6, max_leaves=0, min_child_weight=1,\n",
       "              missing=nan, monotone_constraints='()', n_estimators=100,\n",
       "              n_jobs=0, num_parallel_tree=1, predictor='auto', random_state=0,\n",
       "              reg_alpha=0, reg_lambda=1, ...)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# xgboostモデルの作成\n",
    "clf = xgboost.XGBClassifier()\n",
    "# xgboostモデルの学習\n",
    "clf.fit(train_x, train_y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "モデルの学習は驚くほどあっという間です。近年は機械学習モデルを自動で構築する Auto ML の技術も発達しているため、ゼロからモデルを作る機会はどんどん減ってくるかもしれません。\n",
    "\n",
    "決定木を使ったモデルは特徴に対する重みを分析することができます。`feature_importances_` で重みを可視化してみましょう。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pd.Series(clf.feature_importances_, index=clf.feature_names_in_)[lambda x: x > 0.01].sort_values(ascending=False).plot.bar(figsize=(10, 5))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "データ分析の結果、解約しているユーザーで顕著であった `Night Calls` / `Eve Call` と、`Night Charge` / `Eve Charge` の差がとらえられていそうなことがわかります。モデルの解釈と人間の解釈があっていることを確認することは重要です。\n",
    "\n",
    "### Key takeaways\n",
    "\n",
    "* 問題設定に合わせたモデルを選択する必要がある。\n",
    "* 機械学習フレームワークの発達により、学習は容易になりデータの準備、分析がより重要になっている。\n",
    "* 変数に対する重み `feature_importance` を取得できるモデルを使用することで、人間の感覚とモデルの判断根拠の一致を検証できる。\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "## 7.評価 : Test\n",
    "\n",
    "機械学習モデルができたので、ビジネス目標を達成できるかどうか評価を行います。\n",
    "\n",
    "機械学習モデルの性能を評価した後、その機械学習モデルを使って解約防止のアクションを取った場合のコストを試算します。\n",
    "\n",
    "### 7-1.性能評価\n",
    "\n",
    "まず機械学習モデルの性能を評価します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions = clf.predict_proba(test_x)\n",
    "predictions = predictions[:, 1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "機械学習の性能を評価する方法はいくつかあります。今回は、顧客が解約する `1` と解約しない `0` を予測しますので、この混同行列を作成します。混合行列は以下のように表されます。行列の各要素には、予測が当たっていれば True 、はずれていれば False で、それに予測対象の種別 Positive / Negative を付与した名前が付けられています。\n",
    "\n",
    "![environment_setup_003.png](./images/003.png)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "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>predictions</th>\n",
       "      <th>0.0</th>\n",
       "      <th>1.0</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>actual</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>236</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>9</td>\n",
       "      <td>238</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "predictions  0.0  1.0\n",
       "actual               \n",
       "0            236   17\n",
       "1              9  238"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "confusion_matrix = pd.crosstab(index=test_data.iloc[:, 0], columns=np.round(predictions), rownames=['actual'], colnames=['predictions'])\n",
    "confusion_matrix"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "※ アルゴリズムにはランダムな要素があるので結果は必ずしも一致しません._\n",
    "\n",
    "247人の解約者がいて、それらの237名 (true positives) を正しく予測できました。そして、16名の顧客は解約すると予測しましたが、解約していません (false positives)。10名の顧客は解約しないと予測したにもかかわらず解約してしまいました (false negatives)。\n",
    "\n",
    "性能を評価する指標として、 主に精度(`Accuracy`)、適合率(`Precision`)、再現率(`Recall`)、という 3 つの指標があります。 3 つの指標はそれぞれ評価の観点が異なります。一般的には、適合率と再現率の調和平均である F1 を評価指標として使うことが多いです ( 以下の図は、[機械学習帳の学習ノート](https://github.com/icoxfog417/mlnote-note)の図を転用しています )。\n",
    "\n",
    "![004.png](images/004.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`sklearn.metrics` の `classification_report` を使うと簡単に計算することができます。 F1 の値は `0.95` なのでよく予測できていそうです。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.96      0.93      0.95       253\n",
      "           1       0.93      0.96      0.95       247\n",
      "\n",
      "    accuracy                           0.95       500\n",
      "   macro avg       0.95      0.95      0.95       500\n",
      "weighted avg       0.95      0.95      0.95       500\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "\n",
    "\n",
    "print(classification_report(test_data.iloc[:, 0], np.round(predictions)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 7-2. コスト分析\n",
    "\n",
    "このモデルを使って解約防止のアクションを取った場合、どれぐらいのコストがかかるでしょうか? 目標は 1 顧客当たり 500円 以内でした。混同行列のそれぞれの要素にコストの割り振りを行います。\n",
    "\n",
    "1. True Negative: 解約しない顧客を正確に予測できた場合、コストは発生しないため 0 円とします。\n",
    "2. False Negative: 実際は解約してしまった場合です。失った顧客を再度獲得するコストとして、CAC の 200,000 円を設定します。\n",
    "3. True Positive: 解約を予測できた場合、インセンティブとして 500 円までのインセンティブで解約を防止します。\n",
    "3. False Positive: 実際は解約しなかった場合です。インセンティブ 500 円のインセンティブは本来不要だったことになります。\n",
    "\n",
    "この設定で、コストの計算をしてみます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[     0,    500],\n",
       "       [200000,    500]])"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cost_matrix = np.array([[0, 500], [200000, 500]])\n",
    "cost_matrix"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "混同行列と掛け合わせてコストを算出します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "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>predictions</th>\n",
       "      <th>0.0</th>\n",
       "      <th>1.0</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>actual</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>8500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1800000</td>\n",
       "      <td>119000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "predictions      0.0     1.0\n",
       "actual                      \n",
       "0                  0    8500\n",
       "1            1800000  119000"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "total_cost = confusion_matrix * cost_matrix\n",
    "total_cost"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "コストの合計を顧客の人数で割って、一人当たりのコストを計算します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3855.0"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.sum(np.sum(total_cost)) / len(test_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "一人当たり4253円なので、約 8 倍の予算オーバーです。どうしたらコストを最適化できるでしょうか?\n",
    "\n",
    "重要な点として、`xgboost` が出力する値は 0 から 1 までの連続値なので、四捨五入を行う `np.round()` という関数で解約する `1` と 解約しない `0` に分類していました。つまり、しきい値 0.5 で判断したということです。\n",
    "\n",
    "> confusion_matrix = pd.crosstab(index=test_data.iloc[:, 0], columns=**np.round(predictions)**, rownames=['actual'], colnames=['predictions'])\n",
    "\n",
    "直感的な理解のため、予測結果の連続値をみてみましょう。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(predictions)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "解約しない 0 、 解約する 1 いずれかに偏っており、0.1 から 0.9 までの間でしきい値を設定するのはよさそうです。 `0.3` にしてみましょう。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "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>col_0</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Churn?</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>230</td>\n",
       "      <td>23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>244</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "col_0     0    1\n",
       "Churn?          \n",
       "0       230   23\n",
       "1         3  244"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "confusion_matrix_changed = pd.crosstab(index=test_data.iloc[:, 0], columns=np.where(predictions > 0.3, 1, 0))\n",
    "confusion_matrix_changed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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>col_0</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Churn?</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-6</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-6</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "col_0   0  1\n",
       "Churn?      \n",
       "0      -6  6\n",
       "1      -6  6"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "confusion_matrix_changed - confusion_matrix"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "しきい値を0.5から0.3まで減らしてみたとき、true positives は 4 つ、false positives は 9 つ増え、false negatives は 4 つ減りました。全体の約 7% の顧客に対する予測結果が代わったことになります。解約を予測できなかった false negative の影響は大きいため、しきい値を 0.5 から下げて、可能性があれば解約するとみなすことはコスト最適化を行う上で重要そうです。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 7-3.最適なしきい値を探す\n",
    "\n",
    "コストの関数は以下のようなものになります。\n",
    "\n",
    "```txt\n",
    "cost(C) = 200,000 * FN(C) + 0 * TN(C) + 500 * FP(C) + 500 * TP(C)\n",
    "```\n",
    "\n",
    "FN(C) は false negative の割合で、しきい値 C(cutoff) の関数です。同様にTN, FP, TP も用意します。この関数の値が最小となるようなしきい値 C を探します。\n",
    "最も単純な方法は、候補となる閾値 C で何度もシミュレーションをすることです。以下では100個の値に対してループで計算を行います。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiMAAAGsCAYAAAAPJKchAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAzk0lEQVR4nO3deXzU1b3/8ffMZA9JMAlkgbAJhFVEUATEDQ0Fi6K25XfrVWrVW6pe1PyoFW2LWr3cupVSRaqC/HDlKkvxikJa2VTUgqBI2AlbSAhhyUa2mTm/P0IikQSYkJnvLK/no/PA+c75Zj5zGp0353vO+dqMMUYAAAAWsVtdAAAACG2EEQAAYCnCCAAAsBRhBAAAWIowAgAALEUYAQAAliKMAAAASxFGAACApQgjAADAUoQRAABgqYAKI6tXr9bYsWOVnp4um82mxYsXe3T+448/LpvNdtojNjbWOwUDAICzCqgwUlFRoQEDBujFF19s0fmTJ09WQUFBo0efPn3005/+tJUrBQAA5yqgwsjo0aP11FNP6ZZbbmny9ZqaGj388MPq0KGDYmNjNWTIEK1cubLh9TZt2ig1NbXhcejQIeXm5uquu+7y0ScAAAA/FGZ1Aa3pzjvv1J49e/Tuu+8qPT1dixYt0o9+9CNt2rRJPXr0OK39a6+9pp49e2rEiBEWVAsAAKQAGxk5k127dumdd97Re++9pxEjRujCCy/U5MmTdcUVV+j1118/rX11dbXeeustRkUAALBY0IyMfP311zLGqGfPno2OV1dXKykp6bT2CxcuVFlZme644w5flQgAAJoQNGHE7XbL4XBo/fr1cjgcjV5r06bNae1fe+01/fjHP1ZqaqqvSgQAAE0ImjAycOBAuVwuFRUVnXUOSF5enlasWKElS5b4qDoAANCcgAoj5eXl2rlzZ8PzvLw8bdy4UYmJierZs6duu+023XHHHXr++ec1cOBAFRcX65NPPlH//v01ZsyYhvPmzJmjtLQ0jR492oqPAQAATmEzxhirizhXK1eu1DXXXHPa8QkTJmju3Lmqra3VU089pXnz5ik/P19JSUkaOnSonnjiCfXv319S3eWczp0764477tDTTz/t648AAAB+IKDCCAAACD5Bs7QXAAAEJsIIAACwVEBMYHW73Tp48KDi4uJks9msLgcAAJwDY4zKysqUnp4uu7358Y+ACCMHDx5URkaG1WUAAIAW2L9/vzp27Njs6wERRuLi4iTVfZj4+HiLqwEAAOeitLRUGRkZDd/jzQmIMFJ/aSY+Pp4wAgBAgDnbFAsmsAIAAEsRRgAAgKUIIwAAwFKEEQAAYCnCCAAAsBRhBAAAWIowAgAALEUYAQAAliKMAAAASxFGAACApQgjAADAUoQRAABgqYC4UR4AAPCO//f5Hu0/ekI3Xpyuizq2taQGRkYAAAhhH31XoNc+zVNecYVlNRBGAAAIYYfLqiVJ7eIiLauBMAIAQAirDyPtCSMAAMDXqmpdKq1ySpLatYmyrA7CCAAAIaq4vG5UJMJhV3y0dWtaCCMAAISoU+eL2Gw2y+ogjAAAEKLqw0iyhfNFJMIIAAAhq7i8RpLUrg1hBAAAWMAflvVKhBEAAELW4fIqSVK7NhGW1kEYAQAgRDEyAgAALEUYAQAAljpcThgBAAAWMcZ8PzJi4e6rEmEEAICQVF7tVFWtW5KUHMcEVgAA4GP1oyJtIsMUE2HdVvASYQQAgJDkL5NXJcIIAAAhyV92X5UIIwAAhKTDZXUbnlk9X0QijAAAEJIalvUyMgIAAKzAnBEAAGApwggAALCUv+y+KhFGAAAISf6y+6pEGAEAIOS43eb7pb2MjAAAAF87dqJGLreRJCW1YWkvAADwsfr5IhfEhCvcYX0UsL4CAADgU8Vl/nOJRiKMAAAQcg6X1+2+ShgBAACW+H4lDWEEAABYwJ82PJMIIwAAhBzCCAAAsJQ/7b4qEUYAAAg5/rT7qkQYAQAg5HCZBgAAWKbW5daxE7WSpGQ/2H1VIowAABBSjpy8J43DbtMFMYQRAADgY/WXaJLbRMhut1lcTR3CCAAAIcTfdl+VCCMAAIQUf9t9VSKMAAAQUvxtJY1EGAEAIKQQRgAAgKUadl/lMg0AALDC9yMj/rH7quRhGJk2bZouvfRSxcXFqX379ho3bpy2bdt21vNWrVqlQYMGKSoqSt26ddOsWbNaXDAAAGi5U5f2+guPwsiqVat033336YsvvlBOTo6cTqeysrJUUVHR7Dl5eXkaM2aMRowYoQ0bNujRRx/VpEmTtGDBgvMuHgAAeMYf54yEedL4448/bvT89ddfV/v27bV+/XpdeeWVTZ4za9YsderUSdOnT5ck9e7dW+vWrdNzzz2nW2+9tWVVAwAAj52ocaqixiXJv8LIec0ZKSkpkSQlJiY222bt2rXKyspqdGzUqFFat26damtrmzynurpapaWljR4AAOD8FJfVbQUfFW5Xm0iPxiO8qsVhxBij7OxsXXHFFerXr1+z7QoLC5WSktLoWEpKipxOp4qLi5s8Z9q0aUpISGh4ZGRktLRMAABw0qm7r9ps/rEVvHQeYeT+++/Xt99+q3feeeesbX/4gY0xTR6vN2XKFJWUlDQ89u/f39IyAQDASUWl9ZNX/ecSjeThnJF6//mf/6klS5Zo9erV6tix4xnbpqamqrCwsNGxoqIihYWFKSkpqclzIiMjFRnpXx0FAECgKyipGxlJjfefZb2ShyMjxhjdf//9WrhwoT755BN17dr1rOcMHTpUOTk5jY4tX75cgwcPVnh4uGfVAgCAFisoqZQkpSVEW1xJYx6Fkfvuu09vvvmm3n77bcXFxamwsFCFhYWqrKxsaDNlyhTdcccdDc8nTpyovXv3Kjs7W1u2bNGcOXM0e/ZsTZ48ufU+BQAAOKuDJ0dG0tsG8MjIyy+/rJKSEl199dVKS0treMyfP7+hTUFBgfbt29fwvGvXrlq6dKlWrlypiy++WH/84x81Y8YMlvUCAOBjBcfrBg/S2/rXyIhHc0bqJ56eydy5c087dtVVV+nrr7/25K0AAEArq58zkpYQwCMjAAAgMDldbh0qrb9M418jI4QRAABCQFFZtdxGCrPb/G5pL2EEAIAQUL+SJiU+Sg67/2x4JhFGAAAICQeP++dKGokwAgBASPDXPUYkwggAACGhfmQkjZERAABghfqRkXRGRgAAgBX8dY8RiTACAEBI+H4CKyMjAADAx6qdLhWXV0tiZAQAAFjgUEldEIkMsysxNsLiak5HGAEAIMgdbFjWGyWbzb82PJMIIwAABD1/3mNEIowAABD0/HmPEYkwAgBA0PPnPUYkwggAAEGvgJERAABgpYMnNzxjZAQAAFiiYQIrIyMAAMDXKmtcOn6iVhKraQAAgAXq9xiJjXAoPirM4mqaRhgBACCIfT95NdovNzyTCCMAAAS1U3df9VeEEQAAglj9yIi/rqSRCCMAAAQ1f19JIxFGAAAIav6+x4hEGAEAIKgVHGdkBAAAWKjg5MiIv+4xIhFGAAAIWqVVtSqvdkqS0hkZAQAAvla/kiYhOlwxEf654ZlEGAEAIGgFwh4jEmEEAICg1bDHSFv/nS8iEUYAAAhaBYyMAAAAKx1kZAQAAFiJkREAAGCpQNhjRCKMAAAQlNxuo4Mnd1/twGUaAADgawWlVap2uhVmt/n1hmcSYQQAgKC0t7hCktQpMUZhDv/+uvfv6gAAQIvkHakLI12SYy2u5OwIIwAABKE9J0dGuiQRRgAAgAXyik9Ikrokx1hcydkRRgAACEJ7jjAyAgAALOJyG+07Ujcy0pU5IwAAwNcOHq9UjcutCIfd77eClwgjAAAEnfpLNBmJ0XLYbRZXc3aEEQAAgkz9SppAuEQjEUYAAAg6DStpAmDyqkQYAQAg6OwJoA3PJMIIAABBh8s0AADAMk6XW/uO1m94RhgBAAA+dvB4lZxuo4gwu9Li/ftuvfUIIwAABJH6G+R1ToyRPQCW9UqEEQAAgkrDDfIC5BKNRBgBACCo5AXY5FWJMAIAQFAJpBvk1SOMAAAQRL6/TBNjcSXnjjACAECQqHW5tf9YpSQu0wAAAAscOFYpl9soKtyulLjAWNYrEUYAAAgaDZdokmIDZlmvRBgBACBo5BUH3uRViTACAEDQqF9J0zmAJq9KhBEAAIJGwx4jwT4ysnr1ao0dO1bp6emy2WxavHjxGduvXLlSNpvttMfWrVtbWjMAAGhCwx4jAbSSRpLCPD2hoqJCAwYM0J133qlbb731nM/btm2b4uPjG563a9fO07cGAADNqHG6lR+Ay3qlFoSR0aNHa/To0R6/Ufv27dW2bVuPzwMAAGe3/9gJuY0UE+FQ+7hIq8vxiM/mjAwcOFBpaWkaOXKkVqxYcca21dXVKi0tbfQAAADNq1/W2zkpVjZb4CzrlXwQRtLS0vTKK69owYIFWrhwoTIzMzVy5EitXr262XOmTZumhISEhkdGRoa3ywQAIKB9f4O8wFpJI7XgMo2nMjMzlZmZ2fB86NCh2r9/v5577jldeeWVTZ4zZcoUZWdnNzwvLS0lkAAA0AxjjD74tkCSlJkSf5bW/seSpb2XX365duzY0ezrkZGRio+Pb/QAAABNW557SN/sP66YCId+PqST1eV4zJIwsmHDBqWlpVnx1gAABBWX2+i5ZdskSb8c3lXtAmzyqtSCyzTl5eXauXNnw/O8vDxt3LhRiYmJ6tSpk6ZMmaL8/HzNmzdPkjR9+nR16dJFffv2VU1Njd58800tWLBACxYsaL1PAQBAiFq0IV87isqVEB2ue67sZnU5LeJxGFm3bp2uueaahuf1czsmTJiguXPnqqCgQPv27Wt4vaamRpMnT1Z+fr6io6PVt29fffjhhxozZkwrlA8AQOiqdrr055ztkqRfX32hEqLDLa6oZWzGGGN1EWdTWlqqhIQElZSUMH8EAICTXv8sT098kKuU+EitnHyNoiMcVpfUyLl+f3NvGgAAAlBFtVMvflI3bWLSyB5+F0Q8QRgBACAAzfk0T0cqatQlKUY/GxzY218QRgAACDBlVbV6ZfVuSdJD1/dUuCOwv84Du3oAAELQJ1uLVFbtVLfkWI29KN3qcs4bYQQAgACzbHOhJGl0/1TZ7YF1H5qmEEYAAAggVbUurdh6WJL0o77BsYEoYQQAgACyZkexKmtd6tA2Wv06BMd2F4QRAAACSP0lmuv7pMhmC/xLNBJhBACAgOF0ufWPLYckSaP6plpcTeshjAAAECC+yjuq4ydqlRgboUu7XGB1Oa2GMAIAQICov0RzXe/2CgvwvUVOFTyfBACAIGaM0fLc4LtEIxFGAAAICN8eKFFBSZViIxwa3j3Z6nJaFWEEAIAAUH+J5upe7RUVHrg3xWsKYQQAgADw8ckwEmyXaCTCCAAAfm9nUZl2H65QhMOuazLbWV1OqyOMAADg55Ztrpu4Oqx7kuKiwi2upvURRgAA8HOrttfdiyarT/BdopEIIwAA+DVjjLYUlEqSBnZqa20xXkIYAQDAjx0sqVJZlVNhdpsubNfG6nK8gjACAIAf21ZYNyrSrV2sIsKC82s7OD8VAABBYmthmSQpMzXe4kq8hzACAIAf23YyjPRKjbO4Eu8hjAAA4McIIwAAwDK1Lrd2HS6XJGUSRgAAgK/tPlyhWpdRXGSYOrSNtrocryGMAADgp7aeXEnTMzVONpvN4mq8hzACAICf2tawkiZ4L9FIhBEAAPxWKExelQgjAAD4rYY9RlIIIwAAwMdKq2qVf7xSktQriDc8kwgjAAD4pe0nR0VS46OUEBNucTXeRRgBAMAPbQ2RyasSYQQAAL8UKpNXJcIIAAB+KVSW9UqEEQAA/I4xpmHDM8IIAADwucLSKpVWOeWw29S9fRury/E6wggAAH6mfvJqt+RYRYY5LK7G+wgjAAD4mVCaLyIRRgAA8DuhtJJGIowAAOB3thTUT14N7p1X6xFGAADwI7Uut3YdLpfEyAgAALBAXnGFal1GsREOdWgbbXU5PkEYAQDAT6zbc1T3vfW1JKlXWrzsdpvFFflGmNUFAAAQ6koqa/XfH23VO1/tkyQlxUbokdG9LK7KdwgjAABYaMW2Ij38/rc6XFYtSRo/OENTxvRS25gIiyvzHcIIAAAWKa2q1b1vfq3KWpe6tYvVf93cX5d3S7K6LJ8jjAAAYJG/b8hXZa1LPdq30f9OuiIkdlttChNYAQCwgDFGb3+1X5L08yGdQjaISIQRAAAs8e2BEm0pKFVEmF03D+xgdTmWIowAAGCB+pUzN/RPC6nJqk0hjAAA4GPl1U4t+eagJOn/XJphcTXWI4wAAOBjSzYe1ImauhU0l3VNtLocyxFGAADwsfpLND+/rJNsttDYZfVMCCMAAPjQd/kl2pRfogiHXbdc0tHqcvwCYQQAAB+qHxUZ1S9VibGhPXG1HmEEAAAfOVHj1N831k1c/TcmrjYgjAAA4CMffHNQ5dVOdUmKCclt35tDGAEAwAdqnG69uGKnpLodV+12Jq7WI4wAAOAD8/+1T/uPViq5TaT+/fLOVpfjVwgjAAB42Ykap2Z8Ujcq8sDI7oqJ4D61p/I4jKxevVpjx45Venq6bDabFi9efNZzVq1apUGDBikqKkrdunXTrFmzWlIrAAAB6fXP9uhwWbUyEqM1/tJOVpfjdzwOIxUVFRowYIBefPHFc2qfl5enMWPGaMSIEdqwYYMeffRRTZo0SQsWLPC4WAAAAk3JiVr9bdUuSVL29T0VEcZFiR/yeJxo9OjRGj169Dm3nzVrljp16qTp06dLknr37q1169bpueee06233urp2wMAEFBeXrVLpVVO9UqN040DQvvuvM3xejxbu3atsrKyGh0bNWqU1q1bp9ra2ibPqa6uVmlpaaMHAACBpqi0SnM/z5MkTc7KlIMVNE3y+gyawsJCpaSkNDqWkpIip9Op4uJipaWlnXbOtGnT9MQTT3i7NAAAWk2106VthWVyuU3DsTe+2KuqWrcu6dRWI3u3t7A6/+aT6bw/vAmQMabJ4/WmTJmi7OzshuelpaXKyGCnOgCAf1q1/bB+t3iT9h+tbPL13/6oFzfEOwOvh5HU1FQVFhY2OlZUVKSwsDAlJTW9+1xkZKQiIyO9XRoAAOflcFm1/vi/uVryTd0W73FRYUqIDm/U5vo+KRrCbqtn5PUwMnToUH3wwQeNji1fvlyDBw9WeHh4M2cBAOC/jDH6n3X79V9Lt6qkslZ2m3Tn8K7Kvr6nYiPZQ8RTHvdYeXm5du7c2fA8Ly9PGzduVGJiojp16qQpU6YoPz9f8+bNkyRNnDhRL774orKzs3XPPfdo7dq1mj17tt55553W+xQAAPjIzqJyPbpok77KOypJ6pser/++5SL175hgcWWBy+Mwsm7dOl1zzTUNz+vndkyYMEFz585VQUGB9u3b1/B6165dtXTpUj300EN66aWXlJ6erhkzZrCsFwAQUKqdLs1csUsvr9ylGpdb0eEOZV/fU3cO76IwB3uHnA+bqZ9N6sdKS0uVkJCgkpISxcfHW10OACDErN97TL95/xvtPlwhSboms52evKmfMhJjLK7Mv53r9zcXtgAAOIMj5dWaMOcrlVc71S4uUlPH9tEN/dNYHdOKCCMAAJzBzJW7VF7tVJ+0eL3zH5eftloG54+LXAAANCP/eKXeWLtXkvTI6F4EES8hjAAA0Iy//GO7alxuXd4tUSN6JFtdTtAijAAA0ISdReV6f/0BSdLD7KDqVYQRAACa8PzybXKbuh1UL+l0gdXlBDXCCAAAP/DtgeP66LtC2Wx1d9uFdxFGAAD4gWeXbZMk3TywgzJT4yyuJvgRRgAAOMXnO4u1Zkexwh02PXRdT6vLCQmEEQAATjLG6E8nR0VuG9KZHVZ9hDACAMBJyzYf0jf7jysmwqH7ruludTkhgzACAIAkl9vo+eV1oyK/HN5V7eIiLa4odBBGAACQtGhDvnYUlSshOlz3XNnN6nJCCmEEABDyqp0u/TlnuyTp3qsvZNt3HyOMAABC3ttf7lP+8UqlxEdqwrAuVpcTcggjAICQVl7t1Iuf7JQkTRrZQ1HhDosrCj2EEQBASJvzaZ6OVNSoS1KMfjY4w+pyQhJhBAAQso5V1OjV1bslSdlZmQp38LVoBXodABCyXl61S2XVTvVOi9eP+6dZXU7IIowAAEJSQUml5n6+R5L08KhM2e02awsKYYQRAEBImvHPHapxunVZl0RdndnO6nJCGmEEABBydh8u1/+sOyBJevhHmbLZGBWxEmEEABByns/ZLpfb6Npe7TW4S6LV5YQ8wggAIKR8l1+iD78tkCRNzsq0uBpIhBEAQIh5ZlndzfBuujhdfdLjLa4GkhRmdQEAAHiDMUZf5R1VSWVtw7GCkiqt3n5YYXabsq/vaWF1OBVhBAAQdHYWlevRRZv0Vd7RJl8ff2mGOifF+rgqNIcwAgAIGtVOl15euUszV+xSjcut6HCHeqXFNWqTFBvBqIifIYwAAPyaMUardxRr/9ETZ2znchvNW7tHuw5XSJKuzmynP97UTxmJMb4oE+eBMAIA8Ft7j1Tod4u/05odxed8TnKbSE0d20c/viiN/UMCBGEEAOB3al1uvbpmt/7yjx2qdroVGWbXlT3byXGWcNE5OUb3XtVdCTHhPqoUrYEwAgDwK7kHS5X9Pxu1tbBMkjS8e5KeHtdfXZKZcBqsCCMAAL9x8Hilbp/9pY5U1OiCmHD97oY+uuWSDlxuCXKEEQCAX6iqdenXb67XkYoa9UmL15t3D1FibITVZcEH2IEVAGA5Y4ym/n2zvjlQorYx4frb7YMIIiGEMAIAsNzbX+3T/HX7ZbdJf/23gSzHDTGEEQCApdbvPabHl2yWJP1mVC+N6NHO4orga4QRAIBlthaW6tdvrlety2hM/1RNvKqb1SXBAkxgBQD4XFWtSzP+uUOvrN4tp9uoR/s2euYnA1g1E6IIIwAAn/psZ7EeXbRJe4/Ube8+qm+K/nhTP7WJ5CspVPH/PADAZ578IFdzPsuTJKXGR+mJm/pqVN9Ui6uC1QgjAACf+HL3Ec35LE82mzRhaBf936yeioti23YQRgAAPmCM0TPLtkmSbhvSSY/f2NfiiuBPWE0DAPC6f24p0vq9xxQVbteka3tYXQ78DGEEAOBVLrfRsydHRe4c3lXt46Msrgj+hjACAPCqJd/ka9uhMsVHhWnilRdaXQ78EGEEAOA1NU63XsjZLkmaePWFSohhwipORxgBAHjNu//ap/1HK9UuLlJ3DutqdTnwU4QRAIBXnKhxasY/d0qSJl3bXdERDosrgr8ijAAAWt3B45W6962vVVxerU6JMRp/aSerS4IfY58RAECrcbmN5q3do+eWbVNFjUvhDpumju2jiDD+7ovmEUYAAK0i92Cppiz8Vt8cKJEkDep8gabd0l89U+Isrgz+jjACADgvJ2qc+ss/dui1T/PkchvFRYXpkdG99G+XdpLdzl14cXaEEQBAi63cVqTfLf5OB45VSpLG9E/V42P7srEZPEIYAQCckTFGa3cdUUllrSTJdnKw48NNhfrgm4OSpA5to/XkTX01sneKVWUigBFGAABn9N66A3p4wbdNvma31W3xnn19T8VG8pWCluE3BwDQrKpal57PqbuvTK/UOMVHhcvISJLaxkRo0rU91L9jgpUlIggQRgAAzZq3do8OlVarQ9toLb5vuKLC2bgMrY+F3wCAJpVW1Wrmyl2SpAeu60EQgdcQRgAATXp19W4dP1GrC9vF6paBHawuB0GMMAIAOM3hsmrN/jRPkvSbUZkKc/B1Ae/htwsAcJqXVuzUiRqXBnRM0Ki+qVaXgyDXojAyc+ZMde3aVVFRURo0aJDWrFnTbNuVK1fKZrOd9ti6dWuLiwYAeM/+oyf09pf7JEm/GdVLNhu7qMK7PA4j8+fP14MPPqjHHntMGzZs0IgRIzR69Gjt27fvjOdt27ZNBQUFDY8ePXq0uGgAgPf8+R/bVeNya3j3JF3RI9nqchACPA4jL7zwgu666y7dfffd6t27t6ZPn66MjAy9/PLLZzyvffv2Sk1NbXg4HMzKBgB/s3RTgRZ+nS+pblQE8AWPwkhNTY3Wr1+vrKysRsezsrL0+eefn/HcgQMHKi0tTSNHjtSKFSvO2La6ulqlpaWNHgAA79p+qEyT3/tGkvQfV3bTxRltrS0IIcOjMFJcXCyXy6WUlMb3HkhJSVFhYWGT56SlpemVV17RggULtHDhQmVmZmrkyJFavXp1s+8zbdo0JSQkNDwyMjI8KRMA4KGSylr96o31OlHj0vDuSXp4VKbVJSGEtGgH1h9OZjLGNDvBKTMzU5mZ3/9SDx06VPv379dzzz2nK6+8sslzpkyZouzs7IbnpaWlBBIA8BK32yh7/kblFVeoQ9to/fXfLmEpL3zKo9+25ORkORyO00ZBioqKThstOZPLL79cO3bsaPb1yMhIxcfHN3oAALxjxic79M+tRYoIs2vWvw9SYmyE1SUhxHg0MhIREaFBgwYpJydHN998c8PxnJwc3XTTTef8czZs2KC0tDRP3hoAcJ7Kqmr1u8XfaWtBWaPj2w7VPf+vm/tz0ztYwuPLNNnZ2br99ts1ePBgDR06VK+88or27duniRMnSqq7xJKfn6958+ZJkqZPn64uXbqob9++qqmp0ZtvvqkFCxZowYIFrftJAADNcruNsv/nG+XkHmry9V8M66KfDOro46qAOh6HkfHjx+vIkSN68sknVVBQoH79+mnp0qXq3LmzJKmgoKDRniM1NTWaPHmy8vPzFR0drb59++rDDz/UmDFjWu9TAADO6KUVO5WTe0gRDrue/elFatcmUubkawnR4eqbzuVwWMdmjDFnb2at0tJSJSQkqKSkhPkjAOChFVuL9Mv/9y8ZI/3p1v4af2knq0tCiDjX72+mSwNAENtTXKFJ726QMdLPh3QiiMAvEUYAIEhVVDv1qzfWq6zKqYGd2mrq2D5WlwQ0iTACAEEor7hCd77+L207VKbkNpGa9e+DFBnGbTjgn1q06RkAwD/VON16ZfUuzfhkp2qcbkWF2/Xyv1+ilPgoq0sDmkUYAYAgsX7vUU1ZuEnbD5VLkkb0SNbT4/qrU1KMxZUBZ0YYAYAAV1pVq2c+3qq3vtwnY6Sk2Aj9/sd9dNPF6c3eqgPwJ4QRAAhQxhh9/F2hpi7ZrKKyaknSTwd11KNjeusCtnRHACGMAICfOXDshA6VVp2xTa3L6LU1efrHlrodVbsmx+rpm/tp2IXJvigRaFWEEQDwE+XVTj23bJvmrd0j9zluRxnusOnXV12oe6/prqhwVssgMBFGAMAPLN9cd7mloKRuRCQjMVqOs8z36Jocq0fH9FaPlDhflAh4DWEEACxUVFql3//9Oy3bXHe5pVNijJ4a109X9mxncWWA7xBGAMAipVW1Gv/KF8orrlCY3aZ7ruymSdf2UHQEl1sQWggjAGABt9soe/5G5RVXKD0hSnPuvFS9UrkRKEIT28EDgAX++slO/WNLkSLC7Prb7YMJIghphBEA8LFPth7S9H9ulyQ9Pa6f+ndMsLgiwFqEEQDwobziCj3w7kYZI91+eWf9dHCG1SUBliOMAICPHK2o0cQ31qusyqnBnS/Q73/cx+qSAL/ABFYA8DJjjBZtyNdTH27R0YoatY+L1MzbLlFEGH8fBCTCCAB41Z7iCj22eJM+23lEkpSZEqc/j79Y7eOjLK4M8B+EEQBoBev3HtN76/arxuluOFbrNlq+uVDVTrciw+yaNLKH/uPKbgp3MCICnIowAgDnoaSyVs98vFVvfbmv2TbDuyfp6XH91SU51oeVAYGDMAIALWCM0UffFerxJZtVVFYtSbp5YAf1Tmt8n5huyW00snd72c5ynxkglBFGAOAUG/Yd099W7VZpVW2zbWw2qbTSqU35JZKkbsmxevrm/hp6YZKvygSCCmEEACSVVdXq2WXb9MYXe2XMuZ0T7rDp11ddqHuv6a6ocO4nA7QUYQRAyPv4u0JNXfKdDpV+f7nl6syz3zX34oy26pzEPBDgfBFGAISsg8crNXXJZuXkHpIkdU6K0dPj+uuKHskWVwaEFsIIgJDjchvNW7tHzy3bpooal8LsNv3qqm76z2t7cLkFsABhBEBI2XywRI8u3KRvDtRNPh3U+QL91839lZkad5YzAXgLYQRAUPpk6yFNW7pVB49XNjp+otYlY6S4qDD99ke99PPLOsluZ9ktYCXCCICgUlRWpSc+yNWH3xY02+aG/mmaOrYPW7IDfoIwAiAouN1G7/5rv6Z9tEVlVU457DbddUXXupGPUzYci4qwq30cIQTwJ4QRAAFvx6EyPbpok/6155gkqX+HBE27pb/6dUiwuDIA54IwAiBgVdW69NKKnZq1apdqXUYxEQ7936xM/WJYFzmYBwIEDMIIAL9XVetS/g8mou47ckJP/m+u8oorJEkje7XXk+P6qUPbaCtKBHAeCCMA/JYxRgu+ztfTH+bq2Imm7xXTPi5Sj9/YV6P7pXIzOiBAEUYA+KW84go9tmiTPt91RJIUE+FQuMPe8HqY3aYbLkrT5FGZio8Kt6pMAK2AMAIEoJITtdp44LjMud7RrRn1IwnnM55gs0k22U7+eZ4/7KT1e47pryt2qsbpVmSYXQ9e11N3j+jaKIwACB6EESCAGGP03roDenrpFpVUNn+L+2AxokeynhrXj5vRAUGOMAIEiN2Hy/Xook36YvdRSVKHttG6ILbllyfqB1XOZ3DFqC4gGSMZmfP6WaeKiXDoF8O7aNzFHZgHAoQAwgjg55wut2au3KUXT162iA536KHre+iXw7sqjMsWAIIAYQTwc099uEVzP98jSbqqZzs9Na6fMhJjrC0KAFoRYQTwYwu/PtAQRP50a3/9bHAGly0ABB3CCOCnvssv0ZSFmyRJk0b20PhLO1lcEQB4BxecAT90tKJGv3pjvaqdbl3bq70eHNnD6pIAwGsII4CfcbrcmvTOBuUfr1SXpBj9efzFsnOfFQBBjMs0gI+UVzs1b+0eFZZUnbHd/qMn9OnOYsVEOPS32wcrIZrdRQEEN8II4AM5uYf0h79/p4KzBJFTPfOTi5SZGufFqgDAPxBGAC86VFqlx5ds1kffFUqSMhKjz2kjr8GdL9CVPdv5okQAsBxhBPDQiRqnXluTp9yDpWdsZ2T0+c4jKqt2ymG36Z4R3fTAyB6KjnD4qFIACAyEEcADK7cV6XeLv9OBY5XnfM6AjLaadnN/9UmP92JlABC4CCPAOSgqq9If/3eLPvjmoCQpPSFKv7yiq6LCzzzKkdwmUtf3SZGD1TAA0CzCCHAGbrfR/HX7NW3pFpVWOWW3SXcO76rs63sqNpJ/fQCgNfBfU6AZOw6V6dFFm/SvPcckSf06xGvazRepf8cEiysDgOBCGAF+oKrWpZdW7NSsVbtU6zKKiXAo+/qe+sWwLtwlFwC8gDASIvYeqdDUJZv16Y5imTO0q5/Z0NzKU5tsOvm/Zts0d159+7pzbXXvZTvluU2ynzx+/veC+/4HNP9ZdEpN3zc6UeNUaZVTkjSyV3s9cVNfdbyAu+QCgLcQRoJcrcutV1bv1ox/7lC1090KP/FMUSZ4tI+L1OM39tXofqncJRcAvIwwEsS+3ndMjy7cpK2FZZKkK7on67EbeispNqLJ9vUxwzSTN4yMjKlrZ5pr9MNzTON/rv8Zp76f29Qfq/vTfZ55x5wSmJoqs76O5l6XpO7t25x1pQwAoHUQRoJQWVWtnl22TW98sVfGSImxEfr9j3uf086fAAD4GmEkiBhjtGxzoaYu2axDpdWSpFsv6ajHbuitxGZGQwAAsBph5CxO1Di1fPMhVdQ4ZbfZZLfVTbas/2eH3SabzSbHyQmYjSeAfn/MdsrEzLpjtkaTN+227yd51k/sPJVNNoU56t7XYf/+/erVuNyauWKX/rHlkCSpa3Ksnh7XT8O6J3u5hwAAOD+EkTP455ZD+sPfNyv/+Llv/W21cIdNE6+6UPdd0505DwCAgEAYacLB45V64oPNWra5bpQhPSFK/ToknJxYaeQ2kstt5DYnH27JdXJm5w8naBpjTv55csJmo+d15546IfTUyZWncpu63UCdbtPw3vXt6/VMjdPvb+itHincdh4AEDhaFEZmzpypZ599VgUFBerbt6+mT5+uESNGNNt+1apVys7O1ubNm5Wenq6HH35YEydObHHRreWNL/Zqw75jctjqLoE47Da53EZ/33hQJ2pcCrPbdPeIbpo0srtiIshtAAB4g8ffsPPnz9eDDz6omTNnavjw4frb3/6m0aNHKzc3V506dTqtfV5ensaMGaN77rlHb775pj777DPde++9ateunW699dZW+RAt9VXe0YYbn/3Q4M4X6Kmb+6lXKndaBQDAm2zmXDeMOGnIkCG65JJL9PLLLzcc6927t8aNG6dp06ad1v63v/2tlixZoi1btjQcmzhxor755hutXbv2nN6ztLRUCQkJKikpUXx864WDT7Ye0s6icjndpuESiNNllJkapxv6p8nOnVYBAGixc/3+9mhkpKamRuvXr9cjjzzS6HhWVpY+//zzJs9Zu3atsrKyGh0bNWqUZs+erdraWoWHh592TnV1taqrqxt9GG+4tleKru2V4pWfDQAAzo1Hd/0qLi6Wy+VSSkrjL/CUlBQVFhY2eU5hYWGT7Z1Op4qLi5s8Z9q0aUpISGh4ZGRkeFImAAAIIC26BekPd/E0xpxxZ8+m2jd1vN6UKVNUUlLS8Ni/f39LygQAAAHAo8s0ycnJcjgcp42CFBUVnTb6US81NbXJ9mFhYUpKSmrynMjISEVGRnpSGgAACFAejYxERERo0KBBysnJaXQ8JydHw4YNa/KcoUOHntZ++fLlGjx4cJPzRQAAQGjx+DJNdna2XnvtNc2ZM0dbtmzRQw89pH379jXsGzJlyhTdcccdDe0nTpyovXv3Kjs7W1u2bNGcOXM0e/ZsTZ48ufU+BQAACFge7zMyfvx4HTlyRE8++aQKCgrUr18/LV26VJ07d5YkFRQUaN++fQ3tu3btqqVLl+qhhx7SSy+9pPT0dM2YMcPyPUYAAIB/8HifESt4a58RAADgPef6/d2i1TQAAACthTACAAAsRRgBAACWIowAAABLEUYAAIClCCMAAMBSHu8zYoX61cfeunsvAABoffXf22fbRSQgwkhZWZkkcfdeAAACUFlZmRISEpp9PSA2PXO73Tp48KDi4uLOeHfgsyktLVVGRob279/P5mk+QH/7Fv3tW/S3b9HfvtVa/W2MUVlZmdLT02W3Nz8zJCBGRux2uzp27NhqPy8+Pp5fZh+iv32L/vYt+tu36G/fao3+PtOISD0msAIAAEsRRgAAgKVCKoxERkZq6tSpioyMtLqUkEB/+xb97Vv0t2/R377l6/4OiAmsAAAgeIXUyAgAAPA/hBEAAGApwggAALAUYQQAAFgq6MLIzJkz1bVrV0VFRWnQoEFas2bNGduvWrVKgwYNUlRUlLp166ZZs2b5qNLg4El/L1y4UNdff73atWun+Ph4DR06VMuWLfNhtYHP09/vep999pnCwsJ08cUXe7fAIONpf1dXV+uxxx5T586dFRkZqQsvvFBz5szxUbWBz9P+fuuttzRgwADFxMQoLS1Nd955p44cOeKjagPX6tWrNXbsWKWnp8tms2nx4sVnPcfr35UmiLz77rsmPDzcvPrqqyY3N9c88MADJjY21uzdu7fJ9rt37zYxMTHmgQceMLm5uebVV1814eHh5v333/dx5YHJ0/5+4IEHzJ/+9Cfz1Vdfme3bt5spU6aY8PBw8/XXX/u48sDkaX/XO378uOnWrZvJysoyAwYM8E2xQaAl/X3jjTeaIUOGmJycHJOXl2e+/PJL89lnn/mw6sDlaX+vWbPG2O1285e//MXs3r3brFmzxvTt29eMGzfOx5UHnqVLl5rHHnvMLFiwwEgyixYtOmN7X3xXBlUYueyyy8zEiRMbHevVq5d55JFHmmz/8MMPm169ejU69qtf/cpcfvnlXqsxmHja303p06ePeeKJJ1q7tKDU0v4eP368+d3vfmemTp1KGPGAp/390UcfmYSEBHPkyBFflBd0PO3vZ5991nTr1q3RsRkzZpiOHTt6rcZgdC5hxBfflUFzmaampkbr169XVlZWo+NZWVn6/PPPmzxn7dq1p7UfNWqU1q1bp9raWq/VGgxa0t8/5Ha7VVZWpsTERG+UGFRa2t+vv/66du3apalTp3q7xKDSkv5esmSJBg8erGeeeUYdOnRQz549NXnyZFVWVvqi5IDWkv4eNmyYDhw4oKVLl8oYo0OHDun999/XDTfc4IuSQ4ovvisD4kZ556K4uFgul0spKSmNjqekpKiwsLDJcwoLC5ts73Q6VVxcrLS0NK/VG+ha0t8/9Pzzz6uiokI/+9nPvFFiUGlJf+/YsUOPPPKI1qxZo7CwoPlX3Sda0t+7d+/Wp59+qqioKC1atEjFxcW69957dfToUeaNnEVL+nvYsGF66623NH78eFVVVcnpdOrGG2/UX//6V1+UHFJ88V0ZNCMj9Ww2W6PnxpjTjp2tfVPH0TRP+7veO++8o8cff1zz589X+/btvVVe0DnX/na5XPr5z3+uJ554Qj179vRVeUHHk99vt9stm82mt956S5dddpnGjBmjF154QXPnzmV05Bx50t+5ubmaNGmS/vCHP2j9+vX6+OOPlZeXp4kTJ/qi1JDj7e/KoPnrUnJyshwOx2kpuqio6LREVy81NbXJ9mFhYUpKSvJarcGgJf1db/78+brrrrv03nvv6brrrvNmmUHD0/4uKyvTunXrtGHDBt1///2S6r4sjTEKCwvT8uXLde211/qk9kDUkt/vtLQ0dejQodHt0nv37i1jjA4cOKAePXp4teZA1pL+njZtmoYPH67f/OY3kqSLLrpIsbGxGjFihJ566ilGtluRL74rg2ZkJCIiQoMGDVJOTk6j4zk5ORo2bFiT5wwdOvS09suXL9fgwYMVHh7utVqDQUv6W6obEfnFL36ht99+m2u7HvC0v+Pj47Vp0yZt3Lix4TFx4kRlZmZq48aNGjJkiK9KD0gt+f0ePny4Dh48qPLy8oZj27dvl91uV8eOHb1ab6BrSX+fOHFCdnvjrzCHwyHp+7+1o3X45Luy1abC+oH6pWGzZ882ubm55sEHHzSxsbFmz549xhhjHnnkEXP77bc3tK9frvTQQw+Z3NxcM3v2bJb2esDT/n777bdNWFiYeemll0xBQUHD4/jx41Z9hIDiaX//EKtpPONpf5eVlZmOHTuan/zkJ2bz5s1m1apVpkePHubuu++26iMEFE/7+/XXXzdhYWFm5syZZteuXebTTz81gwcPNpdddplVHyFglJWVmQ0bNpgNGzYYSeaFF14wGzZsaFhGbcV3ZVCFEWOMeemll0znzp1NRESEueSSS8yqVasaXpswYYK56qqrGrVfuXKlGThwoImIiDBdunQxL7/8so8rDmye9PdVV11lJJ32mDBhgu8LD1Ce/n6fijDiOU/7e8uWLea6664z0dHRpmPHjiY7O9ucOHHCx1UHLk/7e8aMGaZPnz4mOjrapKWlmdtuu80cOHDAx1UHnhUrVpzxv8VWfFfajGE8CwAAWCdo5owAAIDARBgBAACWIowAAABLEUYAAIClCCMAAMBShBEAAGApwggAALAUYQQAAFiKMAIAACxFGAEAAJYijAAAAEsRRgAAgKX+P02/l0+1M8lMAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "コストが最小となるしきい値は: 0.02 であり、その際のコストは: 161000 です。\n"
     ]
    }
   ],
   "source": [
    "cutoffs = np.arange(0.01, 1, 0.01)\n",
    "costs = []\n",
    "\n",
    "for c in cutoffs:\n",
    "    _predictions = pd.Categorical(np.where(predictions > c, 1, 0), categories=[0, 1])\n",
    "    _confusion_matrix = pd.crosstab(index=test_data.iloc[:, 0], columns=_predictions, dropna=False)\n",
    "    costs.append(np.sum(np.sum(_confusion_matrix * cost_matrix)))\n",
    "\n",
    "costs = np.array(costs)\n",
    "plt.plot(cutoffs, costs)\n",
    "plt.show()\n",
    "print('コストが最小となるしきい値は:', cutoffs[np.argmin(costs)], 'であり、その際のコストは:', np.min(costs),'です。')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "最適なしきい値の時のコストが得られました。顧客一人当たりのコストを計算してみましょう。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "322.0"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.min(costs) / len(test_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "目標の500円より低い値とできました。コスト目標は達成できましたが、解約率が目標の 0.6% になるかは機械学習モデルの検証だけではわかりません。なぜなら、解約しそうなことが分かった後に行われる解約防止のアクションにも依存するからです。 その意味では、まず解約の予測ができるかどうかと、予測した対象にアプローチすることで解約率を改善できるか、を別々のフェーズに分けて検証したほうが良いでしょう。\n",
    "\n",
    "### Key takeaways\n",
    "\n",
    "* モデルの性能は、精度、適合率、再現率、の 3 つの観点から評価を行う。適合率と再現率のバランスをとった F1 が評価指標としてよく使用される。\n",
    "* モデルの予測は、モデルの出力値だけでなくしきい値の影響も受ける。\n",
    "* モデルの予測に対するコストを定義することで、最適なコストとなるしきい値を求めることができる。\n",
    "\n",
    "機械学習プロジェクトの成果は機械学習モデルのみから得られるものではないことは覚えておくべきことです。ビジネス理解で作成した価値の計算式に関わる関係者と協力することで、はじめて目標が達成できます。\n",
    "\n",
    "<img src=\"images/006.png\" alt=\"000.png\" width=200/>\n",
    "\n",
    "[*The cat approaches to customers to avoid churn.*](https://huggingface.co/spaces/stabilityai/stable-diffusion)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "99f607f2c79ffd9814a22b00079e08a0d69a070c1d0da094d6f294d5d7821e75"
  },
  "kernelspec": {
   "display_name": "ml-handson-churn:Python",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.6"
  },
  "notice": "Copyright 2017 Amazon.com, Inc. or its affiliates. All Rights Reserved.  Licensed under the Apache License, Version 2.0 (the \"License\"). You may not use this file except in compliance with the License. A copy of the License is located at http://aws.amazon.com/apache2.0/ or in the \"license\" file accompanying this file. This file is distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License."
 },
 "nbformat": 4,
 "nbformat_minor": 4
}