{"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":"code","source":"from IPython.core.display import display, HTML, Javascript\n\n# ----- Notebook Theme -----\ncolor_map = ['#f4a261', '#e8f6f3', '#d0ece7', '#a2d9ce', '#73c6b6', '#45b39d', \n                        '#16a085', '#138d75', '#117a65', '#0e6655', '#e76f51']\n\nprompt = color_map[-1]\nmain_color = color_map[0]\nstrong_main_color = color_map[1]\ncustom_colors = [strong_main_color, main_color]\n\ncss_file = ''' \n\ndiv #notebook {\nbackground-color: white;\nline-height: 20px;\n}\n\n#notebook-container {\n%s\nmargin-top: 2em;\npadding-top: 2em;\nborder-top: 4px solid %s; /* light orange */\n-webkit-box-shadow: 0px 0px 8px 2px rgba(224, 212, 226, 0.5); /* pink */\n    box-shadow: 0px 0px 8px 2px rgba(224, 212, 226, 0.5); /* pink */\n}\n\ndiv .input {\nmargin-bottom: 1em;\n}\n\n.rendered_html h1, .rendered_html h2, .rendered_html h3, .rendered_html h4, .rendered_html h5, .rendered_html h6 {\ncolor: %s; /* light orange */\nfont-weight: 600;\n}\n\ndiv.input_area {\nborder: none;\n    background-color: %s; /* rgba(229, 143, 101, 0.1); light orange [exactly #E58F65] */\n    border-top: 2px solid %s; /* light orange */\n}\n\ndiv.input_prompt {\ncolor: %s; /* light blue */\n}\n\ndiv.output_prompt {\ncolor: %s; /* strong orange */\n}\n\ndiv.cell.selected:before, div.cell.selected.jupyter-soft-selected:before {\nbackground: %s; /* light orange */\n}\n\ndiv.cell.selected, div.cell.selected.jupyter-soft-selected {\n    border-color: %s; /* light orange */\n}\n\n.edit_mode div.cell.selected:before {\nbackground: %s; /* light orange */\n}\n\n.edit_mode div.cell.selected {\nborder-color: %s; /* light orange */\n\n}\n'''\ndef to_rgb(h): \n    return tuple(int(h[i:i+2], 16) for i in [0, 2, 4])\n\nmain_color_rgba = 'rgba(%s, %s, %s, 0.1)' % (to_rgb(main_color[1:]))\nopen('notebook.css', 'w').write(css_file % ('width: 95%;', main_color, main_color, main_color_rgba, main_color,  main_color, prompt, main_color, main_color, main_color, main_color))\n\ndef nb(): \n    return HTML(\"<style>\" + open(\"notebook.css\", \"r\").read() + \"</style>\")\nnb()","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2022-07-05T19:54:31.186156Z","iopub.execute_input":"2022-07-05T19:54:31.186517Z","iopub.status.idle":"2022-07-05T19:54:31.201969Z","shell.execute_reply.started":"2022-07-05T19:54:31.186488Z","shell.execute_reply":"2022-07-05T19:54:31.200926Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"c1,c2,f1,f2,fs1,fs2=\\\n'#e76f51','#eb3446','Montserrat','Smokum',55,10\ndef dhtml(string:str='sometext', fontcolor='#f4a261', font='Montserrat',fontsize=65):\n    display(HTML(f\"\"\"<style>@import 'https://fonts.googleapis.com/css2?family=Montserrat:wght@900&display=swap'</style><h4 style='font-family: {font}; color: {fontcolor}; font-size: {fontsize}px; text-shadow: 3px 3px #e76f51;'>{string}</h4>\"\"\"))\n    \n    \ndhtml('AM-EX CatBoost + WandB')","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2022-07-05T19:55:21.539074Z","iopub.execute_input":"2022-07-05T19:55:21.539495Z","iopub.status.idle":"2022-07-05T19:55:21.548102Z","shell.execute_reply.started":"2022-07-05T19:55:21.539465Z","shell.execute_reply":"2022-07-05T19:55:21.546852Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"dhtml('Pre-Settings', fontsize=45)","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2022-07-05T19:55:26.823115Z","iopub.execute_input":"2022-07-05T19:55:26.823606Z","iopub.status.idle":"2022-07-05T19:55:26.831717Z","shell.execute_reply.started":"2022-07-05T19:55:26.823562Z","shell.execute_reply":"2022-07-05T19:55:26.830729Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def dhtml2(string:str='sometext', fontcolor='#f4a261', font='Montserrat',fontsize=20):\n    display(HTML(\"\"\"\n    <style>\n    @import 'https://fonts.googleapis.com/css2?family=Montserrat:wght@900&display=swap'\n    </style>\n    \n    <style>\n    .someclass{\n        width: 45em;\n        box-shadow: 4px 4px #e76f51;\n        padding: 20px;\n        position: relative;\n        background: #ffebdb;\n        display: flex;\n        justify-content: center;\n        \n    }\n    \n    .textclass{\n        line-height: 25px;\n        font-family: Montserrat;\n        color: #e76f51;\n        font-size: 16px;\n    }\n    \n    .fire{\n        position: relative;\n        left: -50%;\n        top: +160px;\n    }\n    \n    </style>\n    \n    <div class=\"someclass\">\n        \n        \n        <p class=\"textclass\">CatBoost is an algorithm for gradient boosting on decision trees. It is developed by Yandex researchers and engineers, and is used for search, recommendation systems, personal assistant, self-driving cars, weather prediction and many other tasks at Yandex and in other companies, including CERN, Cloudflare, Careem taxi. It is in open-source and can be used by anyone.</p>\n        <p class=\"fire\">🔥</p>\n        \n    </div>\n    \"\"\"))\n    \ndhtml2()","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2022-07-05T20:11:10.485034Z","iopub.execute_input":"2022-07-05T20:11:10.485378Z","iopub.status.idle":"2022-07-05T20:11:10.494836Z","shell.execute_reply.started":"2022-07-05T20:11:10.485349Z","shell.execute_reply":"2022-07-05T20:11:10.493883Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# ====================================================\n# Library\n# ====================================================\nimport os\nimport gc\nimport warnings\nwarnings.filterwarnings('ignore')\nimport random\n\nfrom tqdm.notebook import tqdm\n\nimport scipy as sp\nimport numpy as np\nimport pandas as pd\n\nimport matplotlib.pyplot as plt\nimport itertools\n\nfrom sklearn.metrics import confusion_matrix\nfrom sklearn.preprocessing import LabelEncoder\n\nfrom sklearn.model_selection import StratifiedKFold\n\nfrom catboost import CatBoostClassifier\n\npd.set_option('display.max_rows', 500)\npd.set_option('display.max_columns', 500)\npd.set_option('display.width', 1000)\n\n#!pip install --upgrade wandb\n\n\nimport wandb\n\ntry:\n    from kaggle_secrets import UserSecretsClient\n    user_secrets = UserSecretsClient()\n    api_key = user_secrets.get_secret(\"wandb_api_key\")\n    wandb.login(key=api_key)\n    anony = None\nexcept:\n    anony = \"must\"\n    print('If you want to use your W&B account, go to Add-ons -> Secrets and provide your W&B access token. Use the Label name as wandb_api. \\nGet your W&B access token from here: https://wandb.ai/authorize')","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","_kg_hide-input":false,"_kg_hide-output":true,"execution":{"iopub.status.busy":"2022-07-04T18:38:40.994729Z","iopub.execute_input":"2022-07-04T18:38:40.995407Z","iopub.status.idle":"2022-07-04T18:38:44.20316Z","shell.execute_reply.started":"2022-07-04T18:38:40.995371Z","shell.execute_reply":"2022-07-04T18:38:44.201791Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"class CONFIG:\n    SEED = 42\n    VER = 1\n    FOLDS = 5\n    \ndef seed_everything(seed: int = 42) -> None:\n    random.seed(seed)\n    np.random.seed(seed)\n    os.environ[\"PYTHONHASHSEED\"] = str(seed)\n    \nseed_everything(CONFIG.SEED)\n\ndef pre_proccesed_data(path:str)->pd.DataFrame:\n    \"\"\"\n    :path: str - path to data with format pickle\n        ../input/amex-agg-data-pickle/train_agg.pkl\n    \"\"\"\n    data = pd.read_pickle(path, compression=\"gzip\")\n    for col in data.columns:\n        if data[col].dtype=='float16':\n            data[col] = data[col].astype('float32').round(decimals=2).astype('float16')\n            \n    cat_features = [\n    \"B_30\",\n    \"B_38\",\n    \"D_114\",\n    \"D_116\",\n    \"D_117\",\n    \"D_120\",\n    \"D_126\",\n    \"D_63\",\n    \"D_64\",\n    \"D_66\",\n    \"D_68\"\n    ]\n    \n    cat_features = [f\"{cf}_last\" for cf in cat_features]\n    le_encoder = LabelEncoder()\n    \n    for categorical_feature in cat_features:\n        data[categorical_feature] = le_encoder.fit_transform(data[categorical_feature])\n        #data[categorical_feature] = le_encoder.transform(data[categorical_feature])\n        \n    return data\n\ndef amex_metric_mod(y_true, y_pred):\n\n    labels     = np.transpose(np.array([y_true, y_pred]))\n    labels     = labels[labels[:, 1].argsort()[::-1]]\n    weights    = np.where(labels[:,0]==0, 20, 1)\n    cut_vals   = labels[np.cumsum(weights) <= int(0.04 * np.sum(weights))]\n    top_four   = np.sum(cut_vals[:,0]) / np.sum(labels[:,0])\n\n    gini = [0,0]\n    for i in [1,0]:\n        labels         = np.transpose(np.array([y_true, y_pred]))\n        labels         = labels[labels[:, i].argsort()[::-1]]\n        weight         = np.where(labels[:,0]==0, 20, 1)\n        weight_random  = np.cumsum(weight / np.sum(weight))\n        total_pos      = np.sum(labels[:, 0] *  weight)\n        cum_pos_found  = np.cumsum(labels[:, 0] * weight)\n        lorentz        = cum_pos_found / total_pos\n        gini[i]        = np.sum((lorentz - weight_random) * weight)\n\n    return 0.5 * (gini[1]/gini[0] + top_four)\n\ntrain_data = pre_proccesed_data('../input/amex-agg-data-pickle/train_agg.pkl')\ntest_data = pre_proccesed_data('../input/amex-agg-data-pickle/test_agg.pkl')\n\nFEATURES = [col for col in train_data.columns if col not in ['target']]\n","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2022-07-04T20:45:56.252037Z","iopub.execute_input":"2022-07-04T20:45:56.252415Z","iopub.status.idle":"2022-07-04T20:45:58.074255Z","shell.execute_reply.started":"2022-07-04T20:45:56.252385Z","shell.execute_reply":"2022-07-04T20:45:58.072236Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"dhtml('CATBOOST', fontsize=45)","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2022-07-05T20:11:55.218979Z","iopub.execute_input":"2022-07-05T20:11:55.219327Z","iopub.status.idle":"2022-07-05T20:11:55.22563Z","shell.execute_reply.started":"2022-07-05T20:11:55.219298Z","shell.execute_reply":"2022-07-05T20:11:55.224629Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"!mkdir models\n!mkdir data\n\n#train_data.to_pandas()\noof = []\nimportance = []\ny_test = pd.DataFrame()\ngc.collect()\nnum_fold = 0\nskf = StratifiedKFold(n_splits=CONFIG.FOLDS, shuffle=True, random_state=CONFIG.SEED)\n\nfor train_idx, valid_idx in tqdm(skf.split(train_data[FEATURES], train_data.target), total=skf.get_n_splits(), desc=\"k-fold\"):\n    run = wandb.init(project=\"AMEX\", entity=\"torchme\")\n    wandb.run.name = f'AMEX_CATBOOST_{num_fold}'\n    \n    num_fold += 1\n    print('#'*50)\n    print(f'Fold: {num_fold}')\n    \n    #Split data to folds\n    tr_x, tr_y = train_data.iloc[train_idx].reset_index(drop=True)[FEATURES], train_data.iloc[train_idx].reset_index(drop=True)['target']\n    print(f'train X shape: {tr_x.shape}, train Y shape: {tr_y.shape}')\n    val_x, val_y = train_data.iloc[valid_idx].reset_index(drop=True)[FEATURES], train_data.iloc[valid_idx].reset_index(drop=True)['target']\n    print(f'valid X shape: {tr_x.shape}, valid Y shape: {tr_y.shape}')\n    print('#'*50)\n    print(' ')\n    #Save Model\n    try:\n        clf = CatBoostClassifier(iterations=5000, random_state=CONFIG.SEED, task_type='GPU')\n        clf.fit(tr_x, tr_y, eval_set=[(val_x, val_y)], verbose=1000)\n        \n        importance.append(clf.get_feature_importance())\n    finally:\n        clf.save_model(f'models/catboost_amex_model_{num_fold}.cbm')\n        artifact = wandb.Artifact(name=f'catboost_{num_fold}', type='model')\n        artifact.add_file(f'models/catboost_amex_model_{num_fold}.cbm')\n        run.log_artifact(artifact)\n        \n        print(' ')\n        print('Model has been saved')\n        print(' ')\n    \n    print(' ')\n    preds = clf.predict_proba(val_x)[:, 1]\n    print(f'KAGGLE METRICS: {amex_metric_mod(val_y, preds):.6f}')\n    preds_test = clf.predict_proba(test_data)[:, 1]\n    y_test[f'fold_{num_fold}'] = preds_test\n    \n    oof.extend(preds)\n    \n    wandb.log({'kaggle_metric': amex_metric_mod(val_y, preds)})\n    \n    #CLEAR RAM\n    del tr_x, tr_y, val_x, val_y, preds, preds_test, clf\n    gc.collect()\n    print(' ')\n    print(' ')\n\nrun.finish()\n    \n        ","metadata":{"execution":{"iopub.status.busy":"2022-07-04T18:40:15.046796Z","iopub.execute_input":"2022-07-04T18:40:15.047448Z","iopub.status.idle":"2022-07-04T19:29:02.767303Z","shell.execute_reply.started":"2022-07-04T18:40:15.047414Z","shell.execute_reply":"2022-07-04T19:29:02.76495Z"},"_kg_hide-input":false,"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"dhtml('WandB Artifact', fontsize=45)\n\ndef dhtml3(string:str='sometext', fontcolor='#f4a261', font='Montserrat',fontsize=20):\n    display(HTML(\"\"\"\n    <style>\n    @import 'https://fonts.googleapis.com/css2?family=Montserrat:wght@900&display=swap'\n    </style>\n    \n    <style>\n    .someclass2{\n        width: 45em;\n        box-shadow: 4px 4px #e76f51;\n        padding: 20px;\n        position: relative;\n        background: #ffebdb;\n        display: flex;\n        justify-content: center;\n        \n    }\n    \n    .textclass2{\n        line-height: 25px;\n        font-family: Montserrat;\n        color: #e76f51;\n        font-size: 16px;\n    }\n    \n    .fire2{\n        position: relative;\n        top: +160px;\n        left: -50%;\n    }\n    \n    </style>\n    \n    <div class=\"someclass2\">\n        \n        \n        <p class=\"textclass2\">An artifact as a versioned folder of data.Entire datasets can be directly stored as artifacts .\n\nW&B Artifacts are used for dataset versioning, model versioning . They are also used for tracking dependencies and results across machine learning pipelines.Artifact references can be used to point to data in other systems like S3, GCP, or your own system.</p>\n        <p class=\"fire2\">🔥</p>\n        \n    </div>\n    \"\"\"))\n    \ndhtml3()","metadata":{"execution":{"iopub.status.busy":"2022-07-05T20:10:48.860582Z","iopub.execute_input":"2022-07-05T20:10:48.861238Z","iopub.status.idle":"2022-07-05T20:10:48.871468Z","shell.execute_reply.started":"2022-07-05T20:10:48.8612Z","shell.execute_reply":"2022-07-05T20:10:48.870571Z"},"_kg_hide-input":true,"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"\nYou can learn more about W&B artifacts [here](https://docs.wandb.ai/guides/artifacts)\n\n![image.png](attachment:e00e83df-6d77-4a7b-a469-cb5623081372.png)\n\n![image.png](attachment:a2d3280c-2cfb-4353-a98e-bf9ecd5d77cc.png)","metadata":{},"attachments":{"e00e83df-6d77-4a7b-a469-cb5623081372.png":{"image/png":"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"},"a2d3280c-2cfb-4353-a98e-bf9ecd5d77cc.png":{"image/png":"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"}}},{"cell_type":"code","source":"dhtml('Catboost predict analysis', fontsize=45)","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2022-07-05T20:13:05.119078Z","iopub.execute_input":"2022-07-05T20:13:05.119506Z","iopub.status.idle":"2022-07-05T20:13:05.128669Z","shell.execute_reply.started":"2022-07-05T20:13:05.119466Z","shell.execute_reply":"2022-07-05T20:13:05.1274Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"del test_data\ngc.collect()\n\n#clf = CatBoostClassifier(iterations=5000, random_state=CONFIG.SEED, task_type='GPU')\n#model.load_model('model_name')\n\ny_test.to_csv('catboost_test_df.csv')","metadata":{"execution":{"iopub.status.busy":"2022-07-04T19:30:06.308752Z","iopub.execute_input":"2022-07-04T19:30:06.309598Z","iopub.status.idle":"2022-07-04T19:30:17.335395Z","shell.execute_reply.started":"2022-07-04T19:30:06.30956Z","shell.execute_reply":"2022-07-04T19:30:17.334377Z"},"_kg_hide-input":true,"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"y_oof_binary = (oof >= np.percentile(oof, 75)).astype(int)\nplt.hist(train_data.target, alpha=0.7);\nplt.hist(y_oof_binary, alpha=0.7);\n#plt.hist(oof)","metadata":{"execution":{"iopub.status.busy":"2022-07-04T20:40:29.238748Z","iopub.execute_input":"2022-07-04T20:40:29.239725Z","iopub.status.idle":"2022-07-04T20:40:29.534697Z","shell.execute_reply.started":"2022-07-04T20:40:29.239686Z","shell.execute_reply":"2022-07-04T20:40:29.533715Z"},"_kg_hide-input":true,"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def plot_confusion_matrix(cm, classes,\n                          normalize = False,\n                          title = 'Confusion matrix\"',\n                          cmap = plt.cm.Blues) :\n    plt.imshow(cm, interpolation = 'nearest', cmap = cmap)\n    plt.title(title)\n    plt.colorbar()\n    tick_marks = np.arange(len(classes))\n    plt.xticks(tick_marks, classes, rotation = 0)\n    plt.yticks(tick_marks, classes)\n\n    thresh = cm.max() / 2.\n    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])) :\n        plt.text(j, i, cm[i, j],\n                 horizontalalignment = 'center',\n                 color = 'white' if cm[i, j] > thresh else 'black')\n\n    plt.tight_layout()\n    plt.ylabel('True label')\n    plt.xlabel('Predicted label')\n    \ncm = confusion_matrix(train_data.target, y_oof_binary)\nclass_names = [0,1]\nplt.figure()\nplot_confusion_matrix(cm,\n                      classes = class_names,\n                      title = f'Confusion matrix at 4%')\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2022-07-04T19:36:09.092303Z","iopub.execute_input":"2022-07-04T19:36:09.093181Z","iopub.status.idle":"2022-07-04T19:36:09.428911Z","shell.execute_reply.started":"2022-07-04T19:36:09.093136Z","shell.execute_reply":"2022-07-04T19:36:09.427756Z"},"_kg_hide-input":false,"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"y_oof_binary = (oof >= np.percentile(oof, 95)).astype(int)\nplt.hist(train_data.target, alpha=0.7)\nplt.hist(y_oof_binary, alpha=0.7)\n#plt.hist(oof)","metadata":{"execution":{"iopub.status.busy":"2022-07-04T19:36:45.477273Z","iopub.execute_input":"2022-07-04T19:36:45.477636Z","iopub.status.idle":"2022-07-04T19:36:45.759078Z","shell.execute_reply.started":"2022-07-04T19:36:45.477608Z","shell.execute_reply":"2022-07-04T19:36:45.75804Z"},"_kg_hide-input":false,"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"cm = confusion_matrix(train_data.target, y_oof_binary)\nclass_names = [0,1]\nplt.figure()\nplot_confusion_matrix(cm,\n                      classes = class_names,\n                      title = f'Confusion matrix at 4%')\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2022-07-04T19:36:45.761032Z","iopub.execute_input":"2022-07-04T19:36:45.761367Z","iopub.status.idle":"2022-07-04T19:36:46.075155Z","shell.execute_reply.started":"2022-07-04T19:36:45.761328Z","shell.execute_reply":"2022-07-04T19:36:46.074134Z"},"_kg_hide-input":false,"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"ssub = pd.read_csv('../input/amex-default-prediction/sample_submission.csv', index_col='customer_ID')\nssub['prediction'] = y_test.T.mean().values * 0.99\n\nssub.to_csv('submission.csv')\n\nssub","metadata":{"execution":{"iopub.status.busy":"2022-07-04T19:42:21.500475Z","iopub.execute_input":"2022-07-04T19:42:21.500832Z","iopub.status.idle":"2022-07-04T19:42:29.676866Z","shell.execute_reply.started":"2022-07-04T19:42:21.500788Z","shell.execute_reply":"2022-07-04T19:42:29.675868Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"dhtml('Reference', fontsize=45)","metadata":{"execution":{"iopub.status.busy":"2022-07-04T20:40:05.747451Z","iopub.execute_input":"2022-07-04T20:40:05.747827Z","iopub.status.idle":"2022-07-04T20:40:05.755077Z","shell.execute_reply.started":"2022-07-04T20:40:05.747781Z","shell.execute_reply":"2022-07-04T20:40:05.753876Z"},"_kg_hide-input":true,"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#https://www.kaggle.com/code/carlolepelaars/ensembling-with-stacknet/notebook\n#https://www.kaggle.com/competitions/amex-default-prediction/discussion/328606","metadata":{"execution":{"iopub.status.busy":"2022-07-04T20:22:57.845062Z","iopub.execute_input":"2022-07-04T20:22:57.845593Z","iopub.status.idle":"2022-07-04T20:22:57.850182Z","shell.execute_reply.started":"2022-07-04T20:22:57.845555Z","shell.execute_reply":"2022-07-04T20:22:57.849173Z"},"trusted":true},"execution_count":null,"outputs":[]}]}