demo/demo.ipynb

{
 "cells": [
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "# Metadata for pretty-jupyter HTML rendering\n",
    "# pip install pretty-jupyter\n",
    "# jupyter nbconvert --to html --template pj demo.ipynb\n",
    "\n",
    "title: \"Predicting building age with EUBUCCO footprints\"\n",
    "output:\n",
    "    general:\n",
    "        input: true\n",
    "    html:\n",
    "        toc: true\n",
    "        code_folding: show\n",
    "        theme: paper # https://pretty-jupyter.readthedocs.io/en/latest/styling_settings.html#themes"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Complete notebook available at: https://github.com/ai4up/ufo-prediction/blob/main/demo/demo.ipynb"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Motivation\n",
    "\n",
    "Building attributes such as building height, type, and construction year are not available for all buildings in EUBUCCO. However, for many prospective use cases of the dataset, such as energy modeling, the building attributes are of high importance. This notebook shows how the available building footprints can be used to estimate missing building attributes with supervised machine learning."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# -.-|m { input: false }\n",
    "\n",
    "!pip install -r ../requirements.txt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/Caskroom/miniconda/base/envs/ufo-prediction/lib/python3.9/site-packages/geopandas/_compat.py:106: UserWarning: The Shapely GEOS version (3.9.1-CAPI-1.14.2) is incompatible with the GEOS version PyGEOS was compiled with (3.10.1-CAPI-1.16.0). Conversions between both will be slow.\n",
      "  warnings.warn(\n"
     ]
    }
   ],
   "source": [
    "# -.-|m { input: false }\n",
    "\n",
    "# standard libs\n",
    "import os\n",
    "import sys\n",
    "import logging\n",
    "\n",
    "# project lib\n",
    "PROJECT_SRC_PATH = os.path.join(os.path.abspath(''), '..', 'src')\n",
    "sys.path.append(PROJECT_SRC_PATH)\n",
    "\n",
    "import visualizations\n",
    "from prediction_age import AgePredictor, AgeClassifier, AgePredictorComparison\n",
    "import preprocessing as pp\n",
    "\n",
    "# external libs\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import geopandas as gpd\n",
    "from xgboost import XGBRegressor, XGBClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# -.-|m { input: false }\n",
    "\n",
    "logging.basicConfig(format='%(asctime)s | %(levelname)s : %(message)s', level=logging.INFO)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Demo sample of ~20k buildings for Spain, ~50k for France, and 170k for the Netherlands. All 117 urban form features, lat lon, as well as some auxiliary attributes like city name, neighborhood, building type, etc. are included.\n",
    "\n",
    "The demo samples are stored using Git Large File Storage (LFS). To download them explicitly use:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "!git lfs pull"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "DATA_DIR = '.'\n",
    "\n",
    "path_data_NLD = os.path.join(DATA_DIR, 'df-NLD-exp.pkl')\n",
    "path_data_FRA = os.path.join(DATA_DIR, 'df-FRA-exp.pkl')\n",
    "path_data_ESP = os.path.join(DATA_DIR, 'df-ESP-exp.pkl')\n",
    "\n",
    "df = pd.read_pickle(path_data_NLD)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "xgb_model_params = {'tree_method': 'hist'}\n",
    "xgb_hyperparams = {\n",
    "    'max_depth': 5,\n",
    "    'learning_rate': 0.1,\n",
    "    'n_estimators': 500,\n",
    "    'colsample_bytree': 0.5,\n",
    "    'subsample': 1.0,\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-01-30 11:30:49,926 | INFO : Dataset length: 176319\n",
      "2023-01-30 11:30:49,933 | INFO : Dataset allocated memory: 184 MB\n",
      "2023-01-30 11:30:50,091 | INFO : No abort signal received. Continuing... To abort please create /Users/fnachtigall/workspace/ufo-prediction/metadata/local-def98684.abort.\n",
      "2023-01-30 11:30:50,094 | INFO : Dataset standard deviation: 28.072581085899984\n",
      "2023-01-30 11:30:50,096 | INFO : Dataset mean age: 1974.2587129010487\n",
      "2023-01-30 11:30:50,097 | INFO : Training dataset length: 141055\n",
      "2023-01-30 11:30:50,098 | INFO : Test dataset length: 35264\n",
      "2023-01-30 11:30:50,736 | INFO : Test dataset standard deviation after preprocessing: 25.07194896229578\n",
      "2023-01-30 11:30:50,737 | INFO : Test dataset mean age after preprocessing: 1975.4969008264463\n",
      "2023-01-30 11:30:50,738 | INFO : Training dataset length after preprocessing: 139362\n",
      "2023-01-30 11:30:50,739 | INFO : Test dataset length after preprocessing: 34848\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2023-01-30 11:30:59,430 | ERROR : Failed to calculate energy error: ['residential_type']\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MAE: 10.73 y\n",
      "RMSE: 16.85 y\n",
      "R2: 0.5483\n",
      "R2: nan\n",
      "MAPE: nan\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1008x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x720 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "predictor = AgePredictor(\n",
    "    model=XGBRegressor(**xgb_model_params),\n",
    "    df=df,\n",
    "    test_training_split=pp.split_80_20,\n",
    "    # cross_validation_split=pp.cross_validation,\n",
    "    early_stopping=True,\n",
    "    hyperparameters=xgb_hyperparams,\n",
    "    preprocessing_stages=[pp.remove_outliers]\n",
    ")\n",
    "\n",
    "predictor.evaluate()\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Classification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-01-30 11:31:17,623 | INFO : Generated bins: [1900, 1965, 1975, 1992, 2006, 2015, 2022]\n",
      "2023-01-30 11:31:17,624 | INFO : Generated bins with the following labels: ['1900-1964', '1965-1974', '1975-1991', '1992-2005', '2006-2014', '2015-2021']\n",
      "2023-01-30 11:31:17,807 | INFO : Dataset length: 176319\n",
      "2023-01-30 11:31:17,813 | INFO : Dataset allocated memory: 184 MB\n",
      "2023-01-30 11:31:18,000 | INFO : No abort signal received. Continuing... To abort please create /Users/fnachtigall/workspace/ufo-prediction/metadata/local-fd77a0e5.abort.\n",
      "2023-01-30 11:31:18,000 | INFO : Training dataset length: 141055\n",
      "2023-01-30 11:31:18,001 | INFO : Test dataset length: 35264\n",
      "2023-01-30 11:31:18,004 | INFO : Test cities: ['Apeldoorn' 'Groesbeek' 'Roggel en Neer' 'Dronten' 'Rijnwaarden'\n",
      " 'Jacobswoude' 'IJsselstein' 'Dinxperlo' 'Kapelle' 'Renkum']\n",
      "2023-01-30 11:31:18,362 | INFO : age attribute has been categorized (lowest age included: 1900; highest age included: 2021).\n",
      "2023-01-30 11:31:18,516 | INFO : age attribute has been categorized (lowest age included: 1900; highest age included: 2021).\n",
      "2023-01-30 11:31:18,875 | INFO : Training dataset length after preprocessing: 139362\n",
      "2023-01-30 11:31:18,876 | INFO : Test dataset length after preprocessing: 34848\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification report:\n",
      "               precision    recall  f1-score  support\n",
      "1900-1964      0.751537  0.842825  0.794567     8850\n",
      "1965-1974      0.875129  0.834151  0.854149     7133\n",
      "1975-1991      0.904658  0.799159  0.848642     8798\n",
      "1992-2005      0.852081  0.774682  0.811540     6209\n",
      "2006-2014      0.595462  0.695315  0.641526     3095\n",
      "2015-2021      0.496798  0.711664  0.585129      763\n",
      "accuracy       0.801911  0.801911  0.801911        0\n",
      "macro avg      0.745944  0.776299  0.755926    34848\n",
      "weighted avg   0.813968  0.801911  0.805261    34848\n",
      "Cohen’s kappa: 0.7501\n",
      "Matthews correlation coefficient (MCC): 0.7513\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "FixedFormatter should only be used together with FixedLocator\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1008x720 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "tabula_nl_bins = [1900, 1965, 1975, 1992, 2006, 2015, 2022]\n",
    "equally_sized_bins = (1900, 2020, 10)\n",
    "\n",
    "classifier = AgeClassifier(\n",
    "    model=XGBClassifier(**xgb_model_params),\n",
    "    df=df,\n",
    "    test_training_split=pp.split_80_20,\n",
    "    # cross_validation_split=pp.cross_validation,\n",
    "    preprocessing_stages=[pp.remove_outliers],\n",
    "    hyperparameters=xgb_hyperparams,\n",
    "    mitigate_class_imbalance=True,\n",
    "    # bin_config=equally_sized_bins,\n",
    "    bins=tabula_nl_bins,\n",
    ")\n",
    "classifier.evaluate()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Country and generalization comparison"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The `AgePredictorComparison` faciliates comparisons between differently configured training runs, for example to compare the prediction performance between countries, cross-validation strategies, oversampling strategies or any other preprocessing steps."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-10-21 16:42:52,510 | INFO : Starting experiment Spain_random-cv...\n",
      "2022-10-21 16:42:53,720 | INFO : Dataset length: 974\n",
      "2022-10-21 16:42:53,726 | INFO : Dataset allocated memory: 1 MB\n",
      "2022-10-21 16:42:53,731 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-1eb3cae2.abort.\n",
      "2022-10-21 16:42:53,731 | INFO : Dataset standard deviation: 45.78138305502198\n",
      "2022-10-21 16:42:53,732 | INFO : Dataset mean age: 1950.3613963039015\n",
      "2022-10-21 16:42:53,733 | INFO : Training dataset length: 779\n",
      "2022-10-21 16:42:53,733 | INFO : Test dataset length: 195\n",
      "2022-10-21 16:42:53,748 | INFO : Test dataset standard deviation after preprocessing: 34.49407581359104\n",
      "2022-10-21 16:42:53,749 | INFO : Test dataset mean age after preprocessing: 1958.8352272727273\n",
      "2022-10-21 16:42:53,749 | INFO : Training dataset length after preprocessing: 710\n",
      "2022-10-21 16:42:53,749 | INFO : Test dataset length after preprocessing: 176\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:42:54,115 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-1eb3cae2.abort.\n",
      "2022-10-21 16:42:54,116 | INFO : Dataset standard deviation: 45.78138305502198\n",
      "2022-10-21 16:42:54,117 | INFO : Dataset mean age: 1950.3613963039015\n",
      "2022-10-21 16:42:54,117 | INFO : Training dataset length: 779\n",
      "2022-10-21 16:42:54,118 | INFO : Test dataset length: 195\n",
      "2022-10-21 16:42:54,136 | INFO : Test dataset standard deviation after preprocessing: 32.66656219614004\n",
      "2022-10-21 16:42:54,137 | INFO : Test dataset mean age after preprocessing: 1963.875\n",
      "2022-10-21 16:42:54,138 | INFO : Training dataset length after preprocessing: 710\n",
      "2022-10-21 16:42:54,138 | INFO : Test dataset length after preprocessing: 176\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:42:54,590 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-1eb3cae2.abort.\n",
      "2022-10-21 16:42:54,591 | INFO : Dataset standard deviation: 45.78138305502198\n",
      "2022-10-21 16:42:54,592 | INFO : Dataset mean age: 1950.3613963039015\n",
      "2022-10-21 16:42:54,593 | INFO : Training dataset length: 779\n",
      "2022-10-21 16:42:54,594 | INFO : Test dataset length: 195\n",
      "2022-10-21 16:42:54,614 | INFO : Test dataset standard deviation after preprocessing: 32.182346310183064\n",
      "2022-10-21 16:42:54,615 | INFO : Test dataset mean age after preprocessing: 1961.0777777777778\n",
      "2022-10-21 16:42:54,616 | INFO : Training dataset length after preprocessing: 706\n",
      "2022-10-21 16:42:54,616 | INFO : Test dataset length after preprocessing: 180\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:42:55,152 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-1eb3cae2.abort.\n",
      "2022-10-21 16:42:55,153 | INFO : Dataset standard deviation: 45.78138305502198\n",
      "2022-10-21 16:42:55,154 | INFO : Dataset mean age: 1950.3613963039015\n",
      "2022-10-21 16:42:55,155 | INFO : Training dataset length: 779\n",
      "2022-10-21 16:42:55,155 | INFO : Test dataset length: 195\n",
      "2022-10-21 16:42:55,175 | INFO : Test dataset standard deviation after preprocessing: 32.5996906018213\n",
      "2022-10-21 16:42:55,176 | INFO : Test dataset mean age after preprocessing: 1956.0168539325844\n",
      "2022-10-21 16:42:55,177 | INFO : Training dataset length after preprocessing: 708\n",
      "2022-10-21 16:42:55,177 | INFO : Test dataset length after preprocessing: 178\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:42:55,545 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-1eb3cae2.abort.\n",
      "2022-10-21 16:42:55,546 | INFO : Dataset standard deviation: 45.78138305502198\n",
      "2022-10-21 16:42:55,547 | INFO : Dataset mean age: 1950.3613963039015\n",
      "2022-10-21 16:42:55,548 | INFO : Training dataset length: 780\n",
      "2022-10-21 16:42:55,548 | INFO : Test dataset length: 194\n",
      "2022-10-21 16:42:55,568 | INFO : Test dataset standard deviation after preprocessing: 32.935031974613764\n",
      "2022-10-21 16:42:55,569 | INFO : Test dataset mean age after preprocessing: 1958.721590909091\n",
      "2022-10-21 16:42:55,569 | INFO : Training dataset length after preprocessing: 710\n",
      "2022-10-21 16:42:55,570 | INFO : Test dataset length after preprocessing: 176\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:42:56,318 | INFO : Starting experiment France_random-cv...\n",
      "2022-10-21 16:42:56,443 | INFO : Dataset length: 19625\n",
      "2022-10-21 16:42:56,448 | INFO : Dataset allocated memory: 20 MB\n",
      "2022-10-21 16:42:56,463 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-6c1df9b3.abort.\n",
      "2022-10-21 16:42:56,464 | INFO : Dataset standard deviation: 69.97507071401162\n",
      "2022-10-21 16:42:56,465 | INFO : Dataset mean age: 1944.1809426751593\n",
      "2022-10-21 16:42:56,466 | INFO : Training dataset length: 15700\n",
      "2022-10-21 16:42:56,467 | INFO : Test dataset length: 3925\n",
      "2022-10-21 16:42:56,516 | INFO : Test dataset standard deviation after preprocessing: 26.351582324305316\n",
      "2022-10-21 16:42:56,516 | INFO : Test dataset mean age after preprocessing: 1979.3209530483532\n",
      "2022-10-21 16:42:56,517 | INFO : Training dataset length after preprocessing: 11563\n",
      "2022-10-21 16:42:56,518 | INFO : Test dataset length after preprocessing: 2854\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:42:59,261 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-6c1df9b3.abort.\n",
      "2022-10-21 16:42:59,263 | INFO : Dataset standard deviation: 69.97507071401162\n",
      "2022-10-21 16:42:59,264 | INFO : Dataset mean age: 1944.1809426751593\n",
      "2022-10-21 16:42:59,265 | INFO : Training dataset length: 15700\n",
      "2022-10-21 16:42:59,265 | INFO : Test dataset length: 3925\n",
      "2022-10-21 16:42:59,321 | INFO : Test dataset standard deviation after preprocessing: 26.22312882519785\n",
      "2022-10-21 16:42:59,322 | INFO : Test dataset mean age after preprocessing: 1979.7270861833106\n",
      "2022-10-21 16:42:59,323 | INFO : Training dataset length after preprocessing: 11493\n",
      "2022-10-21 16:42:59,323 | INFO : Test dataset length after preprocessing: 2924\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:43:02,304 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-6c1df9b3.abort.\n",
      "2022-10-21 16:43:02,306 | INFO : Dataset standard deviation: 69.97507071401162\n",
      "2022-10-21 16:43:02,307 | INFO : Dataset mean age: 1944.1809426751593\n",
      "2022-10-21 16:43:02,308 | INFO : Training dataset length: 15700\n",
      "2022-10-21 16:43:02,309 | INFO : Test dataset length: 3925\n",
      "2022-10-21 16:43:02,359 | INFO : Test dataset standard deviation after preprocessing: 26.47775043789445\n",
      "2022-10-21 16:43:02,360 | INFO : Test dataset mean age after preprocessing: 1979.8195750609543\n",
      "2022-10-21 16:43:02,360 | INFO : Training dataset length after preprocessing: 11546\n",
      "2022-10-21 16:43:02,361 | INFO : Test dataset length after preprocessing: 2871\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:43:05,347 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-6c1df9b3.abort.\n",
      "2022-10-21 16:43:05,349 | INFO : Dataset standard deviation: 69.97507071401162\n",
      "2022-10-21 16:43:05,351 | INFO : Dataset mean age: 1944.1809426751593\n",
      "2022-10-21 16:43:05,352 | INFO : Training dataset length: 15700\n",
      "2022-10-21 16:43:05,352 | INFO : Test dataset length: 3925\n",
      "2022-10-21 16:43:05,416 | INFO : Test dataset standard deviation after preprocessing: 25.766233816924423\n",
      "2022-10-21 16:43:05,417 | INFO : Test dataset mean age after preprocessing: 1981.037023977433\n",
      "2022-10-21 16:43:05,417 | INFO : Training dataset length after preprocessing: 11581\n",
      "2022-10-21 16:43:05,418 | INFO : Test dataset length after preprocessing: 2836\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:43:08,584 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-6c1df9b3.abort.\n",
      "2022-10-21 16:43:08,586 | INFO : Dataset standard deviation: 69.97507071401162\n",
      "2022-10-21 16:43:08,587 | INFO : Dataset mean age: 1944.1809426751593\n",
      "2022-10-21 16:43:08,588 | INFO : Training dataset length: 15700\n",
      "2022-10-21 16:43:08,589 | INFO : Test dataset length: 3925\n",
      "2022-10-21 16:43:08,639 | INFO : Test dataset standard deviation after preprocessing: 27.01922630583458\n",
      "2022-10-21 16:43:08,641 | INFO : Test dataset mean age after preprocessing: 1979.3833560709413\n",
      "2022-10-21 16:43:08,642 | INFO : Training dataset length after preprocessing: 11485\n",
      "2022-10-21 16:43:08,642 | INFO : Test dataset length after preprocessing: 2932\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:43:12,079 | INFO : Starting experiment Netherlands_random-cv...\n",
      "2022-10-21 16:43:13,168 | INFO : Dataset length: 147706\n",
      "2022-10-21 16:43:13,173 | INFO : Dataset allocated memory: 154 MB\n",
      "2022-10-21 16:43:13,244 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-65ae921e.abort.\n",
      "2022-10-21 16:43:13,246 | INFO : Dataset standard deviation: 28.04851594248268\n",
      "2022-10-21 16:43:13,248 | INFO : Dataset mean age: 1974.6345036762216\n",
      "2022-10-21 16:43:13,248 | INFO : Training dataset length: 118164\n",
      "2022-10-21 16:43:13,249 | INFO : Test dataset length: 29542\n",
      "2022-10-21 16:43:13,633 | INFO : Test dataset standard deviation after preprocessing: 25.39573319971459\n",
      "2022-10-21 16:43:13,634 | INFO : Test dataset mean age after preprocessing: 1975.805232159978\n",
      "2022-10-21 16:43:13,635 | INFO : Training dataset length after preprocessing: 116840\n",
      "2022-10-21 16:43:13,635 | INFO : Test dataset length after preprocessing: 29204\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:43:22,576 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-65ae921e.abort.\n",
      "2022-10-21 16:43:22,578 | INFO : Dataset standard deviation: 28.04851594248268\n",
      "2022-10-21 16:43:22,580 | INFO : Dataset mean age: 1974.6345036762216\n",
      "2022-10-21 16:43:22,580 | INFO : Training dataset length: 118165\n",
      "2022-10-21 16:43:22,581 | INFO : Test dataset length: 29541\n",
      "2022-10-21 16:43:22,996 | INFO : Test dataset standard deviation after preprocessing: 25.223689972656587\n",
      "2022-10-21 16:43:22,997 | INFO : Test dataset mean age after preprocessing: 1975.7403730960123\n",
      "2022-10-21 16:43:22,997 | INFO : Training dataset length after preprocessing: 116829\n",
      "2022-10-21 16:43:22,998 | INFO : Test dataset length after preprocessing: 29215\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:43:32,525 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-65ae921e.abort.\n",
      "2022-10-21 16:43:32,528 | INFO : Dataset standard deviation: 28.04851594248268\n",
      "2022-10-21 16:43:32,529 | INFO : Dataset mean age: 1974.6345036762216\n",
      "2022-10-21 16:43:32,530 | INFO : Training dataset length: 118165\n",
      "2022-10-21 16:43:32,531 | INFO : Test dataset length: 29541\n",
      "2022-10-21 16:43:32,937 | INFO : Test dataset standard deviation after preprocessing: 25.49599785716712\n",
      "2022-10-21 16:43:32,938 | INFO : Test dataset mean age after preprocessing: 1975.7676598364862\n",
      "2022-10-21 16:43:32,939 | INFO : Training dataset length after preprocessing: 116811\n",
      "2022-10-21 16:43:32,940 | INFO : Test dataset length after preprocessing: 29233\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:43:42,334 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-65ae921e.abort.\n",
      "2022-10-21 16:43:42,336 | INFO : Dataset standard deviation: 28.04851594248268\n",
      "2022-10-21 16:43:42,338 | INFO : Dataset mean age: 1974.6345036762216\n",
      "2022-10-21 16:43:42,339 | INFO : Training dataset length: 118165\n",
      "2022-10-21 16:43:42,340 | INFO : Test dataset length: 29541\n",
      "2022-10-21 16:43:42,740 | INFO : Test dataset standard deviation after preprocessing: 25.47460972363702\n",
      "2022-10-21 16:43:42,741 | INFO : Test dataset mean age after preprocessing: 1975.7596710639027\n",
      "2022-10-21 16:43:42,742 | INFO : Training dataset length after preprocessing: 116859\n",
      "2022-10-21 16:43:42,742 | INFO : Test dataset length after preprocessing: 29185\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:43:52,274 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-65ae921e.abort.\n",
      "2022-10-21 16:43:52,277 | INFO : Dataset standard deviation: 28.04851594248268\n",
      "2022-10-21 16:43:52,278 | INFO : Dataset mean age: 1974.6345036762216\n",
      "2022-10-21 16:43:52,279 | INFO : Training dataset length: 118165\n",
      "2022-10-21 16:43:52,280 | INFO : Test dataset length: 29541\n",
      "2022-10-21 16:43:52,679 | INFO : Test dataset standard deviation after preprocessing: 25.307730713106885\n",
      "2022-10-21 16:43:52,680 | INFO : Test dataset mean age after preprocessing: 1975.7559146779881\n",
      "2022-10-21 16:43:52,681 | INFO : Training dataset length after preprocessing: 116837\n",
      "2022-10-21 16:43:52,681 | INFO : Test dataset length after preprocessing: 29207\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:03,043 | INFO : Starting experiment Spain_neighborhood-cv...\n",
      "2022-10-21 16:44:04,252 | INFO : Dataset length: 974\n",
      "2022-10-21 16:44:04,257 | INFO : Dataset allocated memory: 1 MB\n",
      "2022-10-21 16:44:04,258 | INFO : Reusing neighborhood column existing in data.\n",
      "2022-10-21 16:44:04,262 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-7d57b122.abort.\n",
      "2022-10-21 16:44:04,263 | INFO : Dataset standard deviation: 45.78138305502198\n",
      "2022-10-21 16:44:04,263 | INFO : Dataset mean age: 1950.3613963039015\n",
      "2022-10-21 16:44:04,264 | INFO : Training dataset length: 675\n",
      "2022-10-21 16:44:04,264 | INFO : Test dataset length: 299\n",
      "2022-10-21 16:44:04,279 | INFO : Test dataset standard deviation after preprocessing: 31.956217364704337\n",
      "2022-10-21 16:44:04,280 | INFO : Test dataset mean age after preprocessing: 1960.3154362416108\n",
      "2022-10-21 16:44:04,280 | INFO : Training dataset length after preprocessing: 588\n",
      "2022-10-21 16:44:04,281 | INFO : Test dataset length after preprocessing: 298\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:04,757 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-7d57b122.abort.\n",
      "2022-10-21 16:44:04,758 | INFO : Dataset standard deviation: 45.78138305502198\n",
      "2022-10-21 16:44:04,759 | INFO : Dataset mean age: 1950.3613963039015\n",
      "2022-10-21 16:44:04,760 | INFO : Training dataset length: 740\n",
      "2022-10-21 16:44:04,761 | INFO : Test dataset length: 234\n",
      "2022-10-21 16:44:04,783 | INFO : Test dataset standard deviation after preprocessing: 35.199759365096945\n",
      "2022-10-21 16:44:04,784 | INFO : Test dataset mean age after preprocessing: 1941.65\n",
      "2022-10-21 16:44:04,785 | INFO : Training dataset length after preprocessing: 666\n",
      "2022-10-21 16:44:04,785 | INFO : Test dataset length after preprocessing: 220\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:05,160 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-7d57b122.abort.\n",
      "2022-10-21 16:44:05,161 | INFO : Dataset standard deviation: 45.78138305502198\n",
      "2022-10-21 16:44:05,162 | INFO : Dataset mean age: 1950.3613963039015\n",
      "2022-10-21 16:44:05,163 | INFO : Training dataset length: 779\n",
      "2022-10-21 16:44:05,163 | INFO : Test dataset length: 195\n",
      "2022-10-21 16:44:05,184 | INFO : Test dataset standard deviation after preprocessing: 28.33530596863871\n",
      "2022-10-21 16:44:05,185 | INFO : Test dataset mean age after preprocessing: 1957.936507936508\n",
      "2022-10-21 16:44:05,186 | INFO : Training dataset length after preprocessing: 697\n",
      "2022-10-21 16:44:05,186 | INFO : Test dataset length after preprocessing: 189\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:05,655 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-7d57b122.abort.\n",
      "2022-10-21 16:44:05,656 | INFO : Dataset standard deviation: 45.78138305502198\n",
      "2022-10-21 16:44:05,657 | INFO : Dataset mean age: 1950.3613963039015\n",
      "2022-10-21 16:44:05,658 | INFO : Training dataset length: 863\n",
      "2022-10-21 16:44:05,659 | INFO : Test dataset length: 111\n",
      "2022-10-21 16:44:05,679 | INFO : Test dataset standard deviation after preprocessing: 20.35446077489167\n",
      "2022-10-21 16:44:05,680 | INFO : Test dataset mean age after preprocessing: 1982.904761904762\n",
      "2022-10-21 16:44:05,681 | INFO : Training dataset length after preprocessing: 802\n",
      "2022-10-21 16:44:05,682 | INFO : Test dataset length after preprocessing: 84\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:06,104 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-7d57b122.abort.\n",
      "2022-10-21 16:44:06,105 | INFO : Dataset standard deviation: 45.78138305502198\n",
      "2022-10-21 16:44:06,106 | INFO : Dataset mean age: 1950.3613963039015\n",
      "2022-10-21 16:44:06,107 | INFO : Training dataset length: 839\n",
      "2022-10-21 16:44:06,108 | INFO : Test dataset length: 135\n",
      "2022-10-21 16:44:06,128 | INFO : Test dataset standard deviation after preprocessing: 19.855057095693187\n",
      "2022-10-21 16:44:06,129 | INFO : Test dataset mean age after preprocessing: 1982.5894736842106\n",
      "2022-10-21 16:44:06,130 | INFO : Training dataset length after preprocessing: 791\n",
      "2022-10-21 16:44:06,131 | INFO : Test dataset length after preprocessing: 95\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:07,110 | INFO : Starting experiment France_neighborhood-cv...\n",
      "2022-10-21 16:44:07,218 | INFO : Dataset length: 19625\n",
      "2022-10-21 16:44:07,224 | INFO : Dataset allocated memory: 20 MB\n",
      "2022-10-21 16:44:07,225 | INFO : Reusing neighborhood column existing in data.\n",
      "2022-10-21 16:44:07,245 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-d0357291.abort.\n",
      "2022-10-21 16:44:07,247 | INFO : Dataset standard deviation: 69.97507071401162\n",
      "2022-10-21 16:44:07,248 | INFO : Dataset mean age: 1944.1809426751593\n",
      "2022-10-21 16:44:07,249 | INFO : Training dataset length: 15699\n",
      "2022-10-21 16:44:07,249 | INFO : Test dataset length: 3926\n",
      "2022-10-21 16:44:07,292 | INFO : Test dataset standard deviation after preprocessing: 26.147640834969383\n",
      "2022-10-21 16:44:07,293 | INFO : Test dataset mean age after preprocessing: 1980.0678851174935\n",
      "2022-10-21 16:44:07,293 | INFO : Training dataset length after preprocessing: 11736\n",
      "2022-10-21 16:44:07,294 | INFO : Test dataset length after preprocessing: 2681\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:08,574 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-d0357291.abort.\n",
      "2022-10-21 16:44:08,576 | INFO : Dataset standard deviation: 69.97507071401162\n",
      "2022-10-21 16:44:08,577 | INFO : Dataset mean age: 1944.1809426751593\n",
      "2022-10-21 16:44:08,578 | INFO : Training dataset length: 15700\n",
      "2022-10-21 16:44:08,579 | INFO : Test dataset length: 3925\n",
      "2022-10-21 16:44:08,638 | INFO : Test dataset standard deviation after preprocessing: 22.059129423039277\n",
      "2022-10-21 16:44:08,639 | INFO : Test dataset mean age after preprocessing: 1984.1629941672068\n",
      "2022-10-21 16:44:08,640 | INFO : Training dataset length after preprocessing: 11331\n",
      "2022-10-21 16:44:08,640 | INFO : Test dataset length after preprocessing: 3086\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:09,719 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-d0357291.abort.\n",
      "2022-10-21 16:44:09,721 | INFO : Dataset standard deviation: 69.97507071401162\n",
      "2022-10-21 16:44:09,722 | INFO : Dataset mean age: 1944.1809426751593\n",
      "2022-10-21 16:44:09,723 | INFO : Training dataset length: 15700\n",
      "2022-10-21 16:44:09,724 | INFO : Test dataset length: 3925\n",
      "2022-10-21 16:44:09,786 | INFO : Test dataset standard deviation after preprocessing: 29.74950952572091\n",
      "2022-10-21 16:44:09,787 | INFO : Test dataset mean age after preprocessing: 1972.266184448463\n",
      "2022-10-21 16:44:09,787 | INFO : Training dataset length after preprocessing: 11652\n",
      "2022-10-21 16:44:09,788 | INFO : Test dataset length after preprocessing: 2765\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:10,614 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-d0357291.abort.\n",
      "2022-10-21 16:44:10,616 | INFO : Dataset standard deviation: 69.97507071401162\n",
      "2022-10-21 16:44:10,618 | INFO : Dataset mean age: 1944.1809426751593\n",
      "2022-10-21 16:44:10,619 | INFO : Training dataset length: 15700\n",
      "2022-10-21 16:44:10,619 | INFO : Test dataset length: 3925\n",
      "2022-10-21 16:44:10,674 | INFO : Test dataset standard deviation after preprocessing: 28.808374396670327\n",
      "2022-10-21 16:44:10,675 | INFO : Test dataset mean age after preprocessing: 1978.5458290422246\n",
      "2022-10-21 16:44:10,676 | INFO : Training dataset length after preprocessing: 11504\n",
      "2022-10-21 16:44:10,677 | INFO : Test dataset length after preprocessing: 2913\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:11,772 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-d0357291.abort.\n",
      "2022-10-21 16:44:11,775 | INFO : Dataset standard deviation: 69.97507071401162\n",
      "2022-10-21 16:44:11,777 | INFO : Dataset mean age: 1944.1809426751593\n",
      "2022-10-21 16:44:11,778 | INFO : Training dataset length: 15701\n",
      "2022-10-21 16:44:11,778 | INFO : Test dataset length: 3924\n",
      "2022-10-21 16:44:11,838 | INFO : Test dataset standard deviation after preprocessing: 23.00641214340772\n",
      "2022-10-21 16:44:11,839 | INFO : Test dataset mean age after preprocessing: 1983.522880215343\n",
      "2022-10-21 16:44:11,840 | INFO : Training dataset length after preprocessing: 11445\n",
      "2022-10-21 16:44:11,841 | INFO : Test dataset length after preprocessing: 2972\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:13,869 | INFO : Starting experiment Netherlands_neighborhood-cv...\n",
      "2022-10-21 16:44:14,927 | INFO : Dataset length: 147706\n",
      "2022-10-21 16:44:14,932 | INFO : Dataset allocated memory: 154 MB\n",
      "2022-10-21 16:44:14,934 | INFO : Reusing neighborhood column existing in data.\n",
      "2022-10-21 16:44:15,087 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-4ca9cd8d.abort.\n",
      "2022-10-21 16:44:15,089 | INFO : Dataset standard deviation: 28.04851594248268\n",
      "2022-10-21 16:44:15,091 | INFO : Dataset mean age: 1974.6345036762216\n",
      "2022-10-21 16:44:15,091 | INFO : Training dataset length: 118164\n",
      "2022-10-21 16:44:15,091 | INFO : Test dataset length: 29542\n",
      "2022-10-21 16:44:15,468 | INFO : Test dataset standard deviation after preprocessing: 28.935082154644657\n",
      "2022-10-21 16:44:15,469 | INFO : Test dataset mean age after preprocessing: 1970.0846606350408\n",
      "2022-10-21 16:44:15,469 | INFO : Training dataset length after preprocessing: 116975\n",
      "2022-10-21 16:44:15,470 | INFO : Test dataset length after preprocessing: 29069\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:22,283 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-4ca9cd8d.abort.\n",
      "2022-10-21 16:44:22,286 | INFO : Dataset standard deviation: 28.04851594248268\n",
      "2022-10-21 16:44:22,288 | INFO : Dataset mean age: 1974.6345036762216\n",
      "2022-10-21 16:44:22,289 | INFO : Training dataset length: 118165\n",
      "2022-10-21 16:44:22,289 | INFO : Test dataset length: 29541\n",
      "2022-10-21 16:44:22,703 | INFO : Test dataset standard deviation after preprocessing: 24.95239689089501\n",
      "2022-10-21 16:44:22,704 | INFO : Test dataset mean age after preprocessing: 1973.4328429695518\n",
      "2022-10-21 16:44:22,705 | INFO : Training dataset length after preprocessing: 116814\n",
      "2022-10-21 16:44:22,706 | INFO : Test dataset length after preprocessing: 29230\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:28,201 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-4ca9cd8d.abort.\n",
      "2022-10-21 16:44:28,204 | INFO : Dataset standard deviation: 28.04851594248268\n",
      "2022-10-21 16:44:28,206 | INFO : Dataset mean age: 1974.6345036762216\n",
      "2022-10-21 16:44:28,207 | INFO : Training dataset length: 118165\n",
      "2022-10-21 16:44:28,207 | INFO : Test dataset length: 29541\n",
      "2022-10-21 16:44:28,621 | INFO : Test dataset standard deviation after preprocessing: 26.086501899919966\n",
      "2022-10-21 16:44:28,622 | INFO : Test dataset mean age after preprocessing: 1977.429920561863\n",
      "2022-10-21 16:44:28,623 | INFO : Training dataset length after preprocessing: 116713\n",
      "2022-10-21 16:44:28,623 | INFO : Test dataset length after preprocessing: 29331\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:32,777 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-4ca9cd8d.abort.\n",
      "2022-10-21 16:44:32,780 | INFO : Dataset standard deviation: 28.04851594248268\n",
      "2022-10-21 16:44:32,782 | INFO : Dataset mean age: 1974.6345036762216\n",
      "2022-10-21 16:44:32,782 | INFO : Training dataset length: 118165\n",
      "2022-10-21 16:44:32,783 | INFO : Test dataset length: 29541\n",
      "2022-10-21 16:44:33,182 | INFO : Test dataset standard deviation after preprocessing: 21.945850557420872\n",
      "2022-10-21 16:44:33,183 | INFO : Test dataset mean age after preprocessing: 1979.4419145299146\n",
      "2022-10-21 16:44:33,183 | INFO : Training dataset length after preprocessing: 116794\n",
      "2022-10-21 16:44:33,184 | INFO : Test dataset length after preprocessing: 29250\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:37,025 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-4ca9cd8d.abort.\n",
      "2022-10-21 16:44:37,027 | INFO : Dataset standard deviation: 28.04851594248268\n",
      "2022-10-21 16:44:37,029 | INFO : Dataset mean age: 1974.6345036762216\n",
      "2022-10-21 16:44:37,030 | INFO : Training dataset length: 118165\n",
      "2022-10-21 16:44:37,031 | INFO : Test dataset length: 29541\n",
      "2022-10-21 16:44:37,494 | INFO : Test dataset standard deviation after preprocessing: 23.20964775454083\n",
      "2022-10-21 16:44:37,495 | INFO : Test dataset mean age after preprocessing: 1978.4059113976134\n",
      "2022-10-21 16:44:37,496 | INFO : Training dataset length after preprocessing: 116880\n",
      "2022-10-21 16:44:37,496 | INFO : Test dataset length after preprocessing: 29164\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:43,079 | INFO : Starting experiment Spain_city-cv...\n",
      "2022-10-21 16:44:44,222 | INFO : Dataset length: 974\n",
      "2022-10-21 16:44:44,227 | INFO : Dataset allocated memory: 1 MB\n",
      "2022-10-21 16:44:44,230 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-663bb9a8.abort.\n",
      "2022-10-21 16:44:44,231 | INFO : Dataset standard deviation: 45.78138305502198\n",
      "2022-10-21 16:44:44,232 | INFO : Dataset mean age: 1950.3613963039015\n",
      "2022-10-21 16:44:44,232 | INFO : Training dataset length: 675\n",
      "2022-10-21 16:44:44,232 | INFO : Test dataset length: 299\n",
      "2022-10-21 16:44:44,247 | INFO : Test dataset standard deviation after preprocessing: 31.956217364704337\n",
      "2022-10-21 16:44:44,248 | INFO : Test dataset mean age after preprocessing: 1960.3154362416108\n",
      "2022-10-21 16:44:44,249 | INFO : Training dataset length after preprocessing: 588\n",
      "2022-10-21 16:44:44,249 | INFO : Test dataset length after preprocessing: 298\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:44,661 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-663bb9a8.abort.\n",
      "2022-10-21 16:44:44,662 | INFO : Dataset standard deviation: 45.78138305502198\n",
      "2022-10-21 16:44:44,663 | INFO : Dataset mean age: 1950.3613963039015\n",
      "2022-10-21 16:44:44,664 | INFO : Training dataset length: 739\n",
      "2022-10-21 16:44:44,664 | INFO : Test dataset length: 235\n",
      "2022-10-21 16:44:44,684 | INFO : Test dataset standard deviation after preprocessing: 35.33091551481178\n",
      "2022-10-21 16:44:44,685 | INFO : Test dataset mean age after preprocessing: 1941.9095022624435\n",
      "2022-10-21 16:44:44,685 | INFO : Training dataset length after preprocessing: 665\n",
      "2022-10-21 16:44:44,686 | INFO : Test dataset length after preprocessing: 221\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:45,020 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-663bb9a8.abort.\n",
      "2022-10-21 16:44:45,021 | INFO : Dataset standard deviation: 45.78138305502198\n",
      "2022-10-21 16:44:45,022 | INFO : Dataset mean age: 1950.3613963039015\n",
      "2022-10-21 16:44:45,023 | INFO : Training dataset length: 779\n",
      "2022-10-21 16:44:45,024 | INFO : Test dataset length: 195\n",
      "2022-10-21 16:44:45,046 | INFO : Test dataset standard deviation after preprocessing: 28.33530596863871\n",
      "2022-10-21 16:44:45,047 | INFO : Test dataset mean age after preprocessing: 1957.936507936508\n",
      "2022-10-21 16:44:45,047 | INFO : Training dataset length after preprocessing: 697\n",
      "2022-10-21 16:44:45,048 | INFO : Test dataset length after preprocessing: 189\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:45,483 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-663bb9a8.abort.\n",
      "2022-10-21 16:44:45,484 | INFO : Dataset standard deviation: 45.78138305502198\n",
      "2022-10-21 16:44:45,485 | INFO : Dataset mean age: 1950.3613963039015\n",
      "2022-10-21 16:44:45,486 | INFO : Training dataset length: 807\n",
      "2022-10-21 16:44:45,486 | INFO : Test dataset length: 167\n",
      "2022-10-21 16:44:45,507 | INFO : Test dataset standard deviation after preprocessing: 19.367780449505045\n",
      "2022-10-21 16:44:45,508 | INFO : Test dataset mean age after preprocessing: 1982.9862068965517\n",
      "2022-10-21 16:44:45,509 | INFO : Training dataset length after preprocessing: 741\n",
      "2022-10-21 16:44:45,510 | INFO : Test dataset length after preprocessing: 145\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:46,105 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-663bb9a8.abort.\n",
      "2022-10-21 16:44:46,106 | INFO : Dataset standard deviation: 45.78138305502198\n",
      "2022-10-21 16:44:46,107 | INFO : Dataset mean age: 1950.3613963039015\n",
      "2022-10-21 16:44:46,107 | INFO : Training dataset length: 896\n",
      "2022-10-21 16:44:46,108 | INFO : Test dataset length: 78\n",
      "2022-10-21 16:44:46,127 | INFO : Test dataset standard deviation after preprocessing: 23.100488647593057\n",
      "2022-10-21 16:44:46,128 | INFO : Test dataset mean age after preprocessing: 1981.1515151515152\n",
      "2022-10-21 16:44:46,129 | INFO : Training dataset length after preprocessing: 853\n",
      "2022-10-21 16:44:46,130 | INFO : Test dataset length after preprocessing: 33\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:47,208 | INFO : Starting experiment France_city-cv...\n",
      "2022-10-21 16:44:47,311 | INFO : Dataset length: 19625\n",
      "2022-10-21 16:44:47,316 | INFO : Dataset allocated memory: 20 MB\n",
      "2022-10-21 16:44:47,333 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-399966f3.abort.\n",
      "2022-10-21 16:44:47,335 | INFO : Dataset standard deviation: 69.97507071401162\n",
      "2022-10-21 16:44:47,336 | INFO : Dataset mean age: 1944.1809426751593\n",
      "2022-10-21 16:44:47,336 | INFO : Training dataset length: 12937\n",
      "2022-10-21 16:44:47,337 | INFO : Test dataset length: 6688\n",
      "2022-10-21 16:44:47,376 | INFO : Test dataset standard deviation after preprocessing: 27.74388307712151\n",
      "2022-10-21 16:44:47,377 | INFO : Test dataset mean age after preprocessing: 1980.5720154043645\n",
      "2022-10-21 16:44:47,377 | INFO : Training dataset length after preprocessing: 10522\n",
      "2022-10-21 16:44:47,377 | INFO : Test dataset length after preprocessing: 3895\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:48,071 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-399966f3.abort.\n",
      "2022-10-21 16:44:48,073 | INFO : Dataset standard deviation: 69.97507071401162\n",
      "2022-10-21 16:44:48,074 | INFO : Dataset mean age: 1944.1809426751593\n",
      "2022-10-21 16:44:48,075 | INFO : Training dataset length: 12948\n",
      "2022-10-21 16:44:48,076 | INFO : Test dataset length: 6677\n",
      "2022-10-21 16:44:48,135 | INFO : Test dataset standard deviation after preprocessing: 17.789735118393526\n",
      "2022-10-21 16:44:48,137 | INFO : Test dataset mean age after preprocessing: 1987.3874465987512\n",
      "2022-10-21 16:44:48,138 | INFO : Training dataset length after preprocessing: 8331\n",
      "2022-10-21 16:44:48,138 | INFO : Test dataset length after preprocessing: 6086\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:49,052 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-399966f3.abort.\n",
      "2022-10-21 16:44:49,054 | INFO : Dataset standard deviation: 69.97507071401162\n",
      "2022-10-21 16:44:49,055 | INFO : Dataset mean age: 1944.1809426751593\n",
      "2022-10-21 16:44:49,056 | INFO : Training dataset length: 15536\n",
      "2022-10-21 16:44:49,056 | INFO : Test dataset length: 4089\n",
      "2022-10-21 16:44:49,122 | INFO : Test dataset standard deviation after preprocessing: 31.76171429473107\n",
      "2022-10-21 16:44:49,123 | INFO : Test dataset mean age after preprocessing: 1965.4946396664682\n",
      "2022-10-21 16:44:49,123 | INFO : Training dataset length after preprocessing: 11059\n",
      "2022-10-21 16:44:49,124 | INFO : Test dataset length after preprocessing: 3358\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:49,759 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-399966f3.abort.\n",
      "2022-10-21 16:44:49,761 | INFO : Dataset standard deviation: 69.97507071401162\n",
      "2022-10-21 16:44:49,762 | INFO : Dataset mean age: 1944.1809426751593\n",
      "2022-10-21 16:44:49,763 | INFO : Training dataset length: 17857\n",
      "2022-10-21 16:44:49,764 | INFO : Test dataset length: 1768\n",
      "2022-10-21 16:44:49,815 | INFO : Test dataset standard deviation after preprocessing: 23.41372357099903\n",
      "2022-10-21 16:44:49,816 | INFO : Test dataset mean age after preprocessing: 1985.858938547486\n",
      "2022-10-21 16:44:49,817 | INFO : Training dataset length after preprocessing: 13701\n",
      "2022-10-21 16:44:49,817 | INFO : Test dataset length after preprocessing: 716\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:50,794 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-399966f3.abort.\n",
      "2022-10-21 16:44:50,796 | INFO : Dataset standard deviation: 69.97507071401162\n",
      "2022-10-21 16:44:50,797 | INFO : Dataset mean age: 1944.1809426751593\n",
      "2022-10-21 16:44:50,798 | INFO : Training dataset length: 19222\n",
      "2022-10-21 16:44:50,798 | INFO : Test dataset length: 403\n",
      "2022-10-21 16:44:50,857 | INFO : Test dataset standard deviation after preprocessing: 23.613057962244934\n",
      "2022-10-21 16:44:50,858 | INFO : Test dataset mean age after preprocessing: 1966.754143646409\n",
      "2022-10-21 16:44:50,858 | INFO : Training dataset length after preprocessing: 14055\n",
      "2022-10-21 16:44:50,859 | INFO : Test dataset length after preprocessing: 362\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:51,922 | INFO : Starting experiment Netherlands_city-cv...\n",
      "2022-10-21 16:44:52,923 | INFO : Dataset length: 147706\n",
      "2022-10-21 16:44:52,929 | INFO : Dataset allocated memory: 154 MB\n",
      "2022-10-21 16:44:53,048 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-c29b793d.abort.\n",
      "2022-10-21 16:44:53,050 | INFO : Dataset standard deviation: 28.04851594248268\n",
      "2022-10-21 16:44:53,051 | INFO : Dataset mean age: 1974.6345036762216\n",
      "2022-10-21 16:44:53,052 | INFO : Training dataset length: 45572\n",
      "2022-10-21 16:44:53,052 | INFO : Test dataset length: 102134\n",
      "2022-10-21 16:44:53,413 | INFO : Test dataset standard deviation after preprocessing: 27.059995121352962\n",
      "2022-10-21 16:44:53,414 | INFO : Test dataset mean age after preprocessing: 1972.3015981599349\n",
      "2022-10-21 16:44:53,414 | INFO : Training dataset length after preprocessing: 45178\n",
      "2022-10-21 16:44:53,415 | INFO : Test dataset length after preprocessing: 100866\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:55,470 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-c29b793d.abort.\n",
      "2022-10-21 16:44:55,472 | INFO : Dataset standard deviation: 28.04851594248268\n",
      "2022-10-21 16:44:55,474 | INFO : Dataset mean age: 1974.6345036762216\n",
      "2022-10-21 16:44:55,474 | INFO : Training dataset length: 125620\n",
      "2022-10-21 16:44:55,475 | INFO : Test dataset length: 22086\n",
      "2022-10-21 16:44:55,859 | INFO : Test dataset standard deviation after preprocessing: 15.11088725363819\n",
      "2022-10-21 16:44:55,860 | INFO : Test dataset mean age after preprocessing: 1986.4627818527574\n",
      "2022-10-21 16:44:55,861 | INFO : Training dataset length after preprocessing: 123958\n",
      "2022-10-21 16:44:55,861 | INFO : Test dataset length after preprocessing: 22086\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:44:58,316 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-c29b793d.abort.\n",
      "2022-10-21 16:44:58,318 | INFO : Dataset standard deviation: 28.04851594248268\n",
      "2022-10-21 16:44:58,319 | INFO : Dataset mean age: 1974.6345036762216\n",
      "2022-10-21 16:44:58,320 | INFO : Training dataset length: 132211\n",
      "2022-10-21 16:44:58,321 | INFO : Test dataset length: 15495\n",
      "2022-10-21 16:44:58,699 | INFO : Test dataset standard deviation after preprocessing: 19.21771965117587\n",
      "2022-10-21 16:44:58,700 | INFO : Test dataset mean age after preprocessing: 1983.2755998426642\n",
      "2022-10-21 16:44:58,701 | INFO : Training dataset length after preprocessing: 130790\n",
      "2022-10-21 16:44:58,701 | INFO : Test dataset length after preprocessing: 15254\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:45:03,760 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-c29b793d.abort.\n",
      "2022-10-21 16:45:03,763 | INFO : Dataset standard deviation: 28.04851594248268\n",
      "2022-10-21 16:45:03,765 | INFO : Dataset mean age: 1974.6345036762216\n",
      "2022-10-21 16:45:03,765 | INFO : Training dataset length: 142164\n",
      "2022-10-21 16:45:03,766 | INFO : Test dataset length: 5542\n",
      "2022-10-21 16:45:04,256 | INFO : Test dataset standard deviation after preprocessing: 25.221979598040583\n",
      "2022-10-21 16:45:04,257 | INFO : Test dataset mean age after preprocessing: 1974.2217102838185\n",
      "2022-10-21 16:45:04,257 | INFO : Training dataset length after preprocessing: 140618\n",
      "2022-10-21 16:45:04,258 | INFO : Test dataset length after preprocessing: 5426\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n",
      "2022-10-21 16:45:08,058 | INFO : No abort signal received. Continuing... To abort please create /p/projects/eubucco/data/3-ml-inputs/local-c29b793d.abort.\n",
      "2022-10-21 16:45:08,060 | INFO : Dataset standard deviation: 28.04851594248268\n",
      "2022-10-21 16:45:08,061 | INFO : Dataset mean age: 1974.6345036762216\n",
      "2022-10-21 16:45:08,061 | INFO : Training dataset length: 145257\n",
      "2022-10-21 16:45:08,062 | INFO : Test dataset length: 2449\n",
      "2022-10-21 16:45:08,512 | INFO : Test dataset standard deviation after preprocessing: 24.54401759850889\n",
      "2022-10-21 16:45:08,513 | INFO : Test dataset mean age after preprocessing: 1978.6621061359867\n",
      "2022-10-21 16:45:08,513 | INFO : Training dataset length after preprocessing: 143632\n",
      "2022-10-21 16:45:08,514 | INFO : Test dataset length after preprocessing: 2412\n",
      "`early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.\n"
     ]
    }
   ],
   "source": [
    "comparison_config = {\n",
    "    'Spain': {'df': path_data_ESP},\n",
    "    'France': {'df': path_data_FRA},\n",
    "    'Netherlands': {'df': path_data_NLD},\n",
    "}\n",
    "\n",
    "grid_comparison_config = {\n",
    "    'random-cv': {'cross_validation_split': pp.cross_validation},\n",
    "    'neighborhood-cv': {'cross_validation_split': pp.neighborhood_cross_validation},\n",
    "    'city-cv': {'cross_validation_split': pp.city_cross_validation},\n",
    "}\n",
    "\n",
    "comparison = AgePredictorComparison(\n",
    "    exp_name='demo',\n",
    "    model=XGBRegressor(**xgb_model_params),\n",
    "    df=None,\n",
    "    frac=0.5,\n",
    "    cross_validation_split=None,\n",
    "    preprocessing_stages=[pp.remove_outliers],\n",
    "    hyperparameters=xgb_hyperparams,\n",
    "    compare_feature_importance=False,\n",
    "    compare_classification_error=False,\n",
    "    include_baseline=False,\n",
    "    save_results=False,\n",
    "    garbage_collect_after_training=True,\n",
    "    comparison_config=comparison_config,\n",
    "    grid_comparison_config=grid_comparison_config,\n",
    ")\n",
    "\n",
    "results = comparison.evaluate()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "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>name</th>\n",
       "      <th>R2</th>\n",
       "      <th>R2_std</th>\n",
       "      <th>MAE</th>\n",
       "      <th>MAE_std</th>\n",
       "      <th>RMSE</th>\n",
       "      <th>RMSE_std</th>\n",
       "      <th>within_5_years</th>\n",
       "      <th>within_10_years</th>\n",
       "      <th>within_20_years</th>\n",
       "      <th>R2_seed_0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Netherlands_city-cv</td>\n",
       "      <td>0.135401</td>\n",
       "      <td>0.0</td>\n",
       "      <td>18.030643</td>\n",
       "      <td>0.0</td>\n",
       "      <td>23.598668</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.221385</td>\n",
       "      <td>0.392977</td>\n",
       "      <td>0.638903</td>\n",
       "      <td>0.135401</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>France_city-cv</td>\n",
       "      <td>0.187767</td>\n",
       "      <td>0.0</td>\n",
       "      <td>18.645831</td>\n",
       "      <td>0.0</td>\n",
       "      <td>23.772030</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.176875</td>\n",
       "      <td>0.345911</td>\n",
       "      <td>0.615315</td>\n",
       "      <td>0.187767</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Spain_city-cv</td>\n",
       "      <td>0.197072</td>\n",
       "      <td>0.0</td>\n",
       "      <td>23.840955</td>\n",
       "      <td>0.0</td>\n",
       "      <td>29.563272</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.126411</td>\n",
       "      <td>0.247178</td>\n",
       "      <td>0.494357</td>\n",
       "      <td>0.197072</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Spain_neighborhood-cv</td>\n",
       "      <td>0.198503</td>\n",
       "      <td>0.0</td>\n",
       "      <td>23.779078</td>\n",
       "      <td>0.0</td>\n",
       "      <td>29.536916</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.129797</td>\n",
       "      <td>0.247178</td>\n",
       "      <td>0.506772</td>\n",
       "      <td>0.198503</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Netherlands_neighborhood-cv</td>\n",
       "      <td>0.304538</td>\n",
       "      <td>0.0</td>\n",
       "      <td>15.884060</td>\n",
       "      <td>0.0</td>\n",
       "      <td>21.164937</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.241489</td>\n",
       "      <td>0.444702</td>\n",
       "      <td>0.699700</td>\n",
       "      <td>0.304538</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>France_neighborhood-cv</td>\n",
       "      <td>0.330228</td>\n",
       "      <td>0.0</td>\n",
       "      <td>16.306574</td>\n",
       "      <td>0.0</td>\n",
       "      <td>21.586864</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.211348</td>\n",
       "      <td>0.408337</td>\n",
       "      <td>0.705209</td>\n",
       "      <td>0.330228</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Spain_random-cv</td>\n",
       "      <td>0.363164</td>\n",
       "      <td>0.0</td>\n",
       "      <td>20.108252</td>\n",
       "      <td>0.0</td>\n",
       "      <td>26.328608</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.180587</td>\n",
       "      <td>0.355530</td>\n",
       "      <td>0.592551</td>\n",
       "      <td>0.363164</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>France_random-cv</td>\n",
       "      <td>0.511105</td>\n",
       "      <td>0.0</td>\n",
       "      <td>12.372172</td>\n",
       "      <td>0.0</td>\n",
       "      <td>18.443089</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.369564</td>\n",
       "      <td>0.593466</td>\n",
       "      <td>0.806340</td>\n",
       "      <td>0.511105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Netherlands_random-cv</td>\n",
       "      <td>0.575725</td>\n",
       "      <td>0.0</td>\n",
       "      <td>10.203823</td>\n",
       "      <td>0.0</td>\n",
       "      <td>16.531180</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.525335</td>\n",
       "      <td>0.695626</td>\n",
       "      <td>0.827052</td>\n",
       "      <td>0.575725</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          name        R2  R2_std        MAE  MAE_std  \\\n",
       "8          Netherlands_city-cv  0.135401     0.0  18.030643      0.0   \n",
       "7               France_city-cv  0.187767     0.0  18.645831      0.0   \n",
       "6                Spain_city-cv  0.197072     0.0  23.840955      0.0   \n",
       "3        Spain_neighborhood-cv  0.198503     0.0  23.779078      0.0   \n",
       "5  Netherlands_neighborhood-cv  0.304538     0.0  15.884060      0.0   \n",
       "4       France_neighborhood-cv  0.330228     0.0  16.306574      0.0   \n",
       "0              Spain_random-cv  0.363164     0.0  20.108252      0.0   \n",
       "1             France_random-cv  0.511105     0.0  12.372172      0.0   \n",
       "2        Netherlands_random-cv  0.575725     0.0  10.203823      0.0   \n",
       "\n",
       "        RMSE  RMSE_std  within_5_years  within_10_years  within_20_years  \\\n",
       "8  23.598668       0.0        0.221385         0.392977         0.638903   \n",
       "7  23.772030       0.0        0.176875         0.345911         0.615315   \n",
       "6  29.563272       0.0        0.126411         0.247178         0.494357   \n",
       "3  29.536916       0.0        0.129797         0.247178         0.506772   \n",
       "5  21.164937       0.0        0.241489         0.444702         0.699700   \n",
       "4  21.586864       0.0        0.211348         0.408337         0.705209   \n",
       "0  26.328608       0.0        0.180587         0.355530         0.592551   \n",
       "1  18.443089       0.0        0.369564         0.593466         0.806340   \n",
       "2  16.531180       0.0        0.525335         0.695626         0.827052   \n",
       "\n",
       "   R2_seed_0  \n",
       "8   0.135401  \n",
       "7   0.187767  \n",
       "6   0.197072  \n",
       "3   0.198503  \n",
       "5   0.304538  \n",
       "4   0.330228  \n",
       "0   0.363164  \n",
       "1   0.511105  \n",
       "2   0.575725  "
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "test-ufo-prediction",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.12"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "d27ed2b28687e829021d274c32d699f3465bdf9c241c45e1c54100cf9589ad2f"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}