{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"![image.png](attachment:fbf07b94-c804-4851-b358-8011c45fc024.png)","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19"},"attachments":{"fbf07b94-c804-4851-b358-8011c45fc024.png":{"image/png":"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"}}},{"cell_type":"markdown","source":"# Competition\n---","metadata":{"execution":{"iopub.status.busy":"2022-05-26T07:00:47.872288Z","iopub.execute_input":"2022-05-26T07:00:47.874121Z","iopub.status.idle":"2022-05-26T07:00:47.902034Z","shell.execute_reply.started":"2022-05-26T07:00:47.874007Z","shell.execute_reply":"2022-05-26T07:00:47.901076Z"}}},{"cell_type":"markdown","source":"・[American Express - Default Prediction](https://www.kaggle.com/competitions/amex-default-prediction)","metadata":{}},{"cell_type":"markdown","source":"# Description\n---","metadata":{}},{"cell_type":"markdown","source":"レストランでの食事やコンサートのチケット購入など、現代の生活では日々の買い物にクレジットカードの利便性が欠かせません。<br>\nクレジットカードがあれば、多額の現金を持ち歩く必要がなく、また、買い物の全額を前払いして、長期にわたって支払うことができます。<br>\nしかし、カード発行会社は、私たちが請求した金額をきちんと返済してくれることをどうやって確認するのでしょうか？<br>\nこの問題は複雑で、多くの解決策がありますが、このコンペティションでは、さらに多くの改善策が検討されています。\n\n貸し倒れ予測は、消費者金融ビジネスのリスク管理の中心的存在です。<br>\n貸し倒れを予測することで、貸し出しの決定を最適化し、より良い顧客体験と健全なビジネス経済を実現することができます。<br>\n現在のモデルは、リスク管理を支援するために存在しています。<br>\nしかし、現在使用されているモデルを凌駕する、より優れたモデルを作成することは可能です。\n\nアメリカン・エキスプレスは、世界的に統合された決済企業です。<br>\n世界最大の決済カード発行会社である同社は、生活を豊かにし、ビジネスの成功をもたらす商品、洞察、体験へのアクセスを顧客に提供しています。\n\nこのコンペティションでは、機械学習のスキルを応用して、クレジット・デフォルトを予測します。<br>\n具体的には、産業界規模のデータセットを活用し、現在の生産モデルに挑戦する機械学習モデルを構築していただきます。<br>\nトレーニング、検証、テストの各データセットには、時系列行動データおよび匿名化された顧客プロファイル情報が含まれます。<br>\n特徴量の作成から、モデル内でのデータの有機的な利用まで、最も強力なモデルを作るためのあらゆる手法を自由に探求することができます。\n\n成功すれば、クレジットカードの審査が通りやすくなり、カード会員にとってより良い顧客体験の創造に貢献できます。<br>\n優れたソリューションは、世界最大のクレジットカード発行会社が使用しているクレジットデフォルト予測モデルに挑戦し、<br>\n賞金やアメリカン・エキスプレスとの面接の機会、そしてやりがいのある新しいキャリアを獲得する可能性があります。","metadata":{}},{"cell_type":"markdown","source":"# Evaluation\n___","metadata":{"execution":{"iopub.status.busy":"2022-05-27T05:29:29.521227Z","iopub.execute_input":"2022-05-27T05:29:29.521749Z","iopub.status.idle":"2022-05-27T05:29:29.534023Z","shell.execute_reply.started":"2022-05-27T05:29:29.521712Z","shell.execute_reply":"2022-05-27T05:29:29.533136Z"}}},{"cell_type":"markdown","source":"### Submission File\nテスト集合の各 customer_ID に対して、ターゲット変数の確率を予測する必要があります。<br>\nファイルはヘッダを含み、以下のフォーマットである必要があります。","metadata":{}},{"cell_type":"markdown","source":"# Timeline\n---","metadata":{"execution":{"iopub.status.busy":"2022-05-27T04:43:39.118752Z","iopub.execute_input":"2022-05-27T04:43:39.119407Z","iopub.status.idle":"2022-05-27T04:43:39.12436Z","shell.execute_reply.started":"2022-05-27T04:43:39.119372Z","shell.execute_reply":"2022-05-27T04:43:39.123362Z"}}},{"cell_type":"markdown","source":"2022年5月25日 - 開始日。<br>\n2022年8月17日 - エントリー締切日。出場するには、この日までに競技規則に同意する必要があります。<br>\n2022年8月17日 - チーム合併の締切日。この日が、参加者がチームに参加したり合併したりできる最後の日です。<br>\n2022年8月24日 - 最終提出締切日。<br>\n\nすべての締め切りは、特に断りのない限り、該当する日の午後11時59分（UTC）です。<br>\n日本時間では、2022年8月25日 午前8時59分（JTC）となります。<br>\n大会主催者は、必要と判断した場合、コンテストのスケジュールを更新する権利を有します。","metadata":{}},{"cell_type":"markdown","source":"# Data\n---","metadata":{}},{"cell_type":"markdown","source":"このコンペティションの目的は、毎月の顧客プロファイルに基づいて、ある顧客が将来クレジットカードの残高額を返済しない確率を予測することです。<br>\nターゲットのバイナリ変数は、最新のクレジットカード明細書の後、18ヶ月のパフォーマンスウィンドウを観察することによって計算され、<br>\nもし顧客が最新の明細書の日付から120日以内に返済額を支払わない場合は、デフォルトイベントとみなされます。\n\nデータセットには、各顧客の各明細書日付における集約されたプロファイル特徴が含まれています。<br>\n特徴は匿名化、正規化されており、以下の一般的なカテゴリーに分類されます。\n\n`D_*` = 延滞変数<br>\n`S_*` = 支出変数<br>\n`P_*` = 支払い変数<br>\n`B_*` = 残高変数<br>\n`R_*` = リスク変数<br>\n\nであり、以下の特徴はカテゴリ的である。\n\n[`B_30`, `B_38`, `d_114`, `d_116`, `d_117`, `d_120`, `d_126`, `d_63`, `d_64`, `d_66`, `d_68`]\n\nあなたのタスクは、各 `customer_ID` について、将来の支払い不履行の確率を予測することです。（`target = 1`）<br>\nこのデータセットでは、ネガティブ・クラスは5%でサブサンプルされているので、スコアリング・メトリックでは20倍の重み付けを受けることに注意してください。","metadata":{}},{"cell_type":"markdown","source":"# Library\n---","metadata":{}},{"cell_type":"code","source":"import gc\nimport optuna\nimport numpy as np\nimport pandas as pd\nimport seaborn as sns\nimport lightgbm as lgb\nimport matplotlib.pyplot as plt\nfrom sklearn.model_selection import train_test_split","metadata":{"execution":{"iopub.status.busy":"2022-05-28T03:32:29.939645Z","iopub.execute_input":"2022-05-28T03:32:29.940527Z","iopub.status.idle":"2022-05-28T03:32:33.510437Z","shell.execute_reply.started":"2022-05-28T03:32:29.940386Z","shell.execute_reply":"2022-05-28T03:32:33.509313Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"sns.set()\npd.set_option('display.max_rows', 200)","metadata":{"execution":{"iopub.status.busy":"2022-05-28T03:33:50.120546Z","iopub.execute_input":"2022-05-28T03:33:50.121115Z","iopub.status.idle":"2022-05-28T03:33:50.128866Z","shell.execute_reply.started":"2022-05-28T03:33:50.121075Z","shell.execute_reply":"2022-05-28T03:33:50.126849Z"},"_kg_hide-input":false,"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Dataset\n---","metadata":{}},{"cell_type":"code","source":"train = pd.read_pickle('../input/creating-smaller-train-test-data/amex_train_data.pkl')\nprint(train.shape)\ntrain.head(10)","metadata":{"execution":{"iopub.status.busy":"2022-05-28T03:33:00.704625Z","iopub.execute_input":"2022-05-28T03:33:00.705126Z","iopub.status.idle":"2022-05-28T03:33:22.908682Z","shell.execute_reply.started":"2022-05-28T03:33:00.705085Z","shell.execute_reply":"2022-05-28T03:33:22.907729Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"labels = pd.read_csv('../input/amex-default-prediction/train_labels.csv')\nprint(labels.shape)\nlabels.head(10)","metadata":{"execution":{"iopub.status.busy":"2022-05-27T05:36:31.744324Z","iopub.execute_input":"2022-05-27T05:36:31.744614Z","iopub.status.idle":"2022-05-27T05:36:32.596801Z","shell.execute_reply.started":"2022-05-27T05:36:31.74459Z","shell.execute_reply":"2022-05-27T05:36:32.595818Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"sample_submission = pd.read_csv('../input/amex-default-prediction/sample_submission.csv')\nprint(sample_submission.shape)\nsample_submission.head(10)","metadata":{"execution":{"iopub.status.busy":"2022-05-27T05:36:32.598751Z","iopub.execute_input":"2022-05-27T05:36:32.599565Z","iopub.status.idle":"2022-05-27T05:36:34.350855Z","shell.execute_reply.started":"2022-05-27T05:36:32.599527Z","shell.execute_reply":"2022-05-27T05:36:34.349915Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# EDA\n---","metadata":{}},{"cell_type":"markdown","source":"### Train","metadata":{}},{"cell_type":"code","source":"print(train.shape)\ntrain.head(10)","metadata":{"execution":{"iopub.status.busy":"2022-05-27T05:36:34.35203Z","iopub.execute_input":"2022-05-27T05:36:34.352477Z","iopub.status.idle":"2022-05-27T05:36:34.384561Z","shell.execute_reply.started":"2022-05-27T05:36:34.352436Z","shell.execute_reply":"2022-05-27T05:36:34.383682Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data = pd.DataFrame(['D_*', 'S_*', 'P_*', 'B_*', 'R_*'], columns=['カラム'])\ndata['意味'] = ['延滞変数', '支出変数', '支払い変数', '残高変数', 'リスク変数']\ndata","metadata":{"execution":{"iopub.status.busy":"2022-05-28T03:32:33.512668Z","iopub.execute_input":"2022-05-28T03:32:33.513269Z","iopub.status.idle":"2022-05-28T03:32:33.542597Z","shell.execute_reply.started":"2022-05-28T03:32:33.513229Z","shell.execute_reply":"2022-05-28T03:32:33.54164Z"},"_kg_hide-input":true,"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"pd.DataFrame(train.isnull().sum(), columns=['isnull'])","metadata":{"execution":{"iopub.status.busy":"2022-05-28T03:35:54.513968Z","iopub.execute_input":"2022-05-28T03:35:54.514452Z","iopub.status.idle":"2022-05-28T03:36:00.306969Z","shell.execute_reply.started":"2022-05-28T03:35:54.514408Z","shell.execute_reply":"2022-05-28T03:36:00.305726Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"pd.DataFrame(train.nunique(), columns=['nunique'])","metadata":{"execution":{"iopub.status.busy":"2022-05-28T03:37:05.519829Z","iopub.execute_input":"2022-05-28T03:37:05.520304Z","iopub.status.idle":"2022-05-28T03:37:24.374651Z","shell.execute_reply.started":"2022-05-28T03:37:05.520266Z","shell.execute_reply":"2022-05-28T03:37:24.373636Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"figure ,ax = plt.subplots(1, 2, figsize=(12,5))\n\ntrain['target'].value_counts().plot.pie(explode=[0,0.1],autopct='%1.1f%%', ax=ax[0], shadow=True)\nax[0].set_title('target')\nax[0].set_ylabel('')\n\nsns.countplot(x='target', data=train, ax=ax[1])\nax[1].set_title('target')\n\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2022-05-27T05:36:57.194656Z","iopub.execute_input":"2022-05-27T05:36:57.195125Z","iopub.status.idle":"2022-05-27T05:36:58.007765Z","shell.execute_reply.started":"2022-05-27T05:36:57.195083Z","shell.execute_reply":"2022-05-27T05:36:58.006648Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Preprocess\n---","metadata":{}},{"cell_type":"code","source":"train = train.groupby('customer_ID').tail(1).set_index('customer_ID', drop=True).sort_index().drop(['S_2'], axis='columns')\nprint(train.shape)\ntrain.head()","metadata":{"execution":{"iopub.status.busy":"2022-05-27T05:36:58.009535Z","iopub.execute_input":"2022-05-27T05:36:58.010001Z","iopub.status.idle":"2022-05-27T05:37:03.617407Z","shell.execute_reply.started":"2022-05-27T05:36:58.009955Z","shell.execute_reply":"2022-05-27T05:37:03.616448Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_train = train[[col for col in train.columns if col != 'target']]\ny_train = pd.DataFrame(train['target'])\n\nprint(X_train.shape)\nprint(y_train.shape)","metadata":{"execution":{"iopub.status.busy":"2022-05-27T05:37:03.618625Z","iopub.execute_input":"2022-05-27T05:37:03.618934Z","iopub.status.idle":"2022-05-27T05:37:03.916735Z","shell.execute_reply.started":"2022-05-27T05:37:03.618909Z","shell.execute_reply":"2022-05-27T05:37:03.916041Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"cat_cols = [col for col in X_train.columns if X_train[col].dtype == 'object']","metadata":{"execution":{"iopub.status.busy":"2022-05-27T05:37:03.917843Z","iopub.execute_input":"2022-05-27T05:37:03.918328Z","iopub.status.idle":"2022-05-27T05:37:03.928904Z","shell.execute_reply.started":"2022-05-27T05:37:03.918299Z","shell.execute_reply":"2022-05-27T05:37:03.928211Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"for col in cat_cols:\n    X_train[col] = X_train[col].astype('category')","metadata":{"execution":{"iopub.status.busy":"2022-05-27T05:37:03.93145Z","iopub.execute_input":"2022-05-27T05:37:03.931817Z","iopub.status.idle":"2022-05-27T05:37:03.942027Z","shell.execute_reply.started":"2022-05-27T05:37:03.931784Z","shell.execute_reply":"2022-05-27T05:37:03.941067Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"del train\ngc.collect()","metadata":{"_kg_hide-output":true,"execution":{"iopub.status.busy":"2022-05-27T05:37:03.943362Z","iopub.execute_input":"2022-05-27T05:37:03.944434Z","iopub.status.idle":"2022-05-27T05:37:04.08986Z","shell.execute_reply.started":"2022-05-27T05:37:03.94439Z","shell.execute_reply":"2022-05-27T05:37:04.089103Z"},"_kg_hide-input":true,"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Metric\n---","metadata":{}},{"cell_type":"code","source":"def amex_metric(y_true: pd.DataFrame, y_pred: pd.DataFrame) -> float:\n\n    def top_four_percent_captured(y_true: pd.DataFrame, y_pred: pd.DataFrame) -> float:\n        df = (pd.concat([y_true, y_pred], axis='columns')\n              .sort_values('prediction', ascending=False))\n        df['weight'] = df['target'].apply(lambda x: 20 if x==0 else 1)\n        four_pct_cutoff = int(0.04 * df['weight'].sum())\n        df['weight_cumsum'] = df['weight'].cumsum()\n        df_cutoff = df.loc[df['weight_cumsum'] <= four_pct_cutoff]\n        return (df_cutoff['target'] == 1).sum() / (df['target'] == 1).sum()\n        \n    def weighted_gini(y_true: pd.DataFrame, y_pred: pd.DataFrame) -> float:\n        df = (pd.concat([y_true, y_pred], axis='columns')\n              .sort_values('prediction', ascending=False))\n        df['weight'] = df['target'].apply(lambda x: 20 if x==0 else 1)\n        df['random'] = (df['weight'] / df['weight'].sum()).cumsum()\n        total_pos = (df['target'] * df['weight']).sum()\n        df['cum_pos_found'] = (df['target'] * df['weight']).cumsum()\n        df['lorentz'] = df['cum_pos_found'] / total_pos\n        df['gini'] = (df['lorentz'] - df['random']) * df['weight']\n        return df['gini'].sum()\n\n    def normalized_weighted_gini(y_true: pd.DataFrame, y_pred: pd.DataFrame) -> float:\n        y_true_pred = y_true.rename(columns={'target': 'prediction'})\n        return weighted_gini(y_true, y_pred) / weighted_gini(y_true, y_true_pred)\n\n    g = normalized_weighted_gini(y_true, y_pred)\n    d = top_four_percent_captured(y_true, y_pred)\n\n    return 0.5 * (g + d)","metadata":{"execution":{"iopub.status.busy":"2022-05-27T05:37:04.091245Z","iopub.execute_input":"2022-05-27T05:37:04.091607Z","iopub.status.idle":"2022-05-27T05:37:04.104434Z","shell.execute_reply.started":"2022-05-27T05:37:04.091578Z","shell.execute_reply":"2022-05-27T05:37:04.103477Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Model\n---","metadata":{}},{"cell_type":"code","source":"X_tr, X_val, y_tr, y_val = train_test_split(X_train, y_train, test_size=0.2, random_state=22)","metadata":{"execution":{"iopub.status.busy":"2022-05-27T05:37:04.106006Z","iopub.execute_input":"2022-05-27T05:37:04.106638Z","iopub.status.idle":"2022-05-27T05:37:05.268537Z","shell.execute_reply.started":"2022-05-27T05:37:04.106606Z","shell.execute_reply":"2022-05-27T05:37:05.267514Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"y_val = pd.DataFrame(y_val)","metadata":{"execution":{"iopub.status.busy":"2022-05-27T05:37:05.269895Z","iopub.execute_input":"2022-05-27T05:37:05.270213Z","iopub.status.idle":"2022-05-27T05:37:05.273826Z","shell.execute_reply.started":"2022-05-27T05:37:05.270184Z","shell.execute_reply":"2022-05-27T05:37:05.272952Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"'''\ndef create_model(trial):\n    num_leaves = trial.suggest_int('num_leaves', 2, 31)\n    n_estimators = trial.suggest_int('n_estimators', 50, 300)\n    learning_rate = trial.suggest_uniform('learning_rate', 0.0001, 0.99)\n    max_depth = trial.suggest_int('max_depth', 3, 8)\n    min_child_samples = trial.suggest_int('min_child_samples', 100, 1200)\n    min_data_in_leaf = trial.suggest_int('min_data_in_leaf', 5, 90)\n    bagging_freq = trial.suggest_int('bagging_freq', 1, 7)\n    bagging_fraction = trial.suggest_uniform('bagging_fraction', 0.0001, 1.0)\n    feature_fraction = trial.suggest_uniform('feature_fraction', 0.0001, 1.0)\n    subsample = trial.suggest_uniform('subsample', 0.1, 1.0)\n    colsample_bytree = trial.suggest_uniform('colsample_bytree', 0.1, 1.0)\n    \n    model = lgb.LGBMClassifier(\n        num_leaves=num_leaves,\n        n_estimators=n_estimators,\n        learning_rate=learning_rate,\n        max_depth=max_depth, \n        min_child_samples=min_child_samples, \n        min_data_in_leaf=min_data_in_leaf,\n        bagging_freq=bagging_freq,\n        bagging_fraction=bagging_fraction,\n        feature_fraction=feature_fraction,\n        subsample=subsample,\n        colsample_bytree=colsample_bytree,\n        random_state=666)\n        \n    return model\n\ndef objective(trial):\n    model = create_model(trial)\n    model.fit(X_tr, y_tr, categorical_feature = cat_cols)\n    y_pred = y_val.copy(deep=True)\n    y_pred = y_pred.rename(columns={'target':'prediction'})\n    y_pred['prediction'] = model.predict_proba(X_val)[:,1]\n    score = amex_metric(y_val, y_pred)\n    return score\n\nstudy = optuna.create_study(direction='maximize')\nstudy.optimize(objective, n_trials=70)\nparams = study.best_params\n\nprint(params)\n'''","metadata":{"_kg_hide-output":true,"_kg_hide-input":false,"execution":{"iopub.status.busy":"2022-05-27T05:37:05.27494Z","iopub.execute_input":"2022-05-27T05:37:05.275347Z","iopub.status.idle":"2022-05-27T05:37:05.290751Z","shell.execute_reply.started":"2022-05-27T05:37:05.275321Z","shell.execute_reply":"2022-05-27T05:37:05.289765Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"params = {'num_leaves': 22,\n          'n_estimators': 264,\n          'learning_rate': 0.06612339248781801,\n          'max_depth': 8,\n          'min_child_samples': 548,\n          'min_data_in_leaf': 27,\n          'bagging_freq': 7,\n          'bagging_fraction': 0.7938573920363178,\n          'feature_fraction': 0.36695559140359413,\n          'subsample': 0.8223893456213397,\n          'colsample_bytree': 0.6109251160779336,\n          'random_state': 22}","metadata":{"execution":{"iopub.status.busy":"2022-05-27T05:37:05.292032Z","iopub.execute_input":"2022-05-27T05:37:05.292489Z","iopub.status.idle":"2022-05-27T05:37:05.306955Z","shell.execute_reply.started":"2022-05-27T05:37:05.292434Z","shell.execute_reply":"2022-05-27T05:37:05.305671Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"cls = lgb.LGBMClassifier(**params)\ncls.fit(X_train, y_train)","metadata":{"_kg_hide-output":true,"execution":{"iopub.status.busy":"2022-05-27T05:37:05.308096Z","iopub.execute_input":"2022-05-27T05:37:05.308788Z","iopub.status.idle":"2022-05-27T05:37:35.893166Z","shell.execute_reply.started":"2022-05-27T05:37:05.308757Z","shell.execute_reply":"2022-05-27T05:37:35.892135Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"fig,ax = plt.subplots(figsize=(30, 30))\nlgb.plot_importance(cls, ax=ax,importance_type='gain',max_num_features=190)\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2022-05-27T05:37:35.894461Z","iopub.execute_input":"2022-05-27T05:37:35.895309Z","iopub.status.idle":"2022-05-27T05:37:38.21201Z","shell.execute_reply.started":"2022-05-27T05:37:35.895272Z","shell.execute_reply":"2022-05-27T05:37:38.211244Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"del X_train, y_train, X_tr, X_val, y_tr, y_val\ngc.collect()","metadata":{"_kg_hide-output":true,"execution":{"iopub.status.busy":"2022-05-27T05:37:38.212983Z","iopub.execute_input":"2022-05-27T05:37:38.213718Z","iopub.status.idle":"2022-05-27T05:37:38.390282Z","shell.execute_reply.started":"2022-05-27T05:37:38.213687Z","shell.execute_reply":"2022-05-27T05:37:38.389229Z"},"_kg_hide-input":true,"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Prediction\n---","metadata":{}},{"cell_type":"code","source":"test = pd.read_pickle('../input/creating-smaller-train-test-data/amex_test_data.pkl')\ntest = test.groupby('customer_ID').tail(1).set_index('customer_ID', drop=True).sort_index().drop(['S_2'], axis='columns')","metadata":{"execution":{"iopub.status.busy":"2022-05-27T05:37:38.391501Z","iopub.execute_input":"2022-05-27T05:37:38.391841Z","iopub.status.idle":"2022-05-27T05:38:37.318639Z","shell.execute_reply.started":"2022-05-27T05:37:38.391814Z","shell.execute_reply":"2022-05-27T05:38:37.317637Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_test = test.copy()","metadata":{"execution":{"iopub.status.busy":"2022-05-27T05:38:37.32012Z","iopub.execute_input":"2022-05-27T05:38:37.320648Z","iopub.status.idle":"2022-05-27T05:38:37.446727Z","shell.execute_reply.started":"2022-05-27T05:38:37.320618Z","shell.execute_reply":"2022-05-27T05:38:37.445564Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"cat_cols = [col for col in X_test.columns if X_test[col].dtype == 'object']","metadata":{"execution":{"iopub.status.busy":"2022-05-27T05:38:37.447925Z","iopub.execute_input":"2022-05-27T05:38:37.448285Z","iopub.status.idle":"2022-05-27T05:38:37.461153Z","shell.execute_reply.started":"2022-05-27T05:38:37.448254Z","shell.execute_reply":"2022-05-27T05:38:37.460227Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"for col in cat_cols:\n    X_test[col] = X_test[col].astype('category')","metadata":{"execution":{"iopub.status.busy":"2022-05-27T05:38:37.46251Z","iopub.execute_input":"2022-05-27T05:38:37.462866Z","iopub.status.idle":"2022-05-27T05:38:37.470597Z","shell.execute_reply.started":"2022-05-27T05:38:37.462837Z","shell.execute_reply":"2022-05-27T05:38:37.469806Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"y_pred = cls.predict_proba(X_test)[:, 1]","metadata":{"execution":{"iopub.status.busy":"2022-05-27T05:38:37.471807Z","iopub.execute_input":"2022-05-27T05:38:37.472274Z","iopub.status.idle":"2022-05-27T05:38:43.331281Z","shell.execute_reply.started":"2022-05-27T05:38:37.472246Z","shell.execute_reply":"2022-05-27T05:38:43.330345Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Submit","metadata":{}},{"cell_type":"code","source":"submission = sample_submission.copy()\nsubmission['prediction'] = y_pred","metadata":{"execution":{"iopub.status.busy":"2022-05-27T05:38:43.334246Z","iopub.execute_input":"2022-05-27T05:38:43.334748Z","iopub.status.idle":"2022-05-27T05:38:43.355876Z","shell.execute_reply.started":"2022-05-27T05:38:43.334702Z","shell.execute_reply":"2022-05-27T05:38:43.354843Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"submission.to_csv('submission.csv', index=False)\npd.read_csv('submission.csv').head(10)","metadata":{"execution":{"iopub.status.busy":"2022-05-27T05:38:43.356974Z","iopub.execute_input":"2022-05-27T05:38:43.357288Z","iopub.status.idle":"2022-05-27T05:38:47.696239Z","shell.execute_reply.started":"2022-05-27T05:38:43.357262Z","shell.execute_reply":"2022-05-27T05:38:47.69537Z"},"trusted":true},"execution_count":null,"outputs":[]}]}