04_Correlated_features_and_Outliers.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "bcf5d27a",
   "metadata": {},
   "source": [
    "# Data preparation and transformation exercise\n",
    "\n",
    "## Part IV - Correlated features and outliers\n",
    "\n",
    "The objective of this exercise is to practice various steps of data preprocessing and feature engineering.\n",
    "\n",
    "The scenario is the preparation of data for a ML multilinear regressions.\n",
    "\n",
    "The dataset used is the \"Climate Weather Surface of Brazil - Hourly\", wich is available at <a href=\"https://www.kaggle.com/PROPPG-PPG/hourly-weather-surface-brazil-southeast-region?select=make_dataset.py\">Kaggle</a>.\n",
    "\n",
    "It contains hourly climate data taken from 122 weather stations in Brasil between 2000 and 2021.\n",
    "\n",
    "**Steps:**\n",
    "\n",
    "1. Load data\n",
    "2. Inspect data\n",
    "3. Format features\n",
    "4. Clean messy data\n",
    "5. Remove duplicate values\n",
    "6. Treat missing values\n",
    "7. Imputation\n",
    "8. <a href=\"#Remove-strongly-correlated-features\">Remove strongly correlated features</a>\n",
    "9. <a href=\"#Remove-outliers\">Remove outliers</a>\n",
    "10. Aggregate features\n",
    "11. Encode categorical features\n",
    "12. Feature scaling\n",
    "13. Dimensionality reduction and feature decomposition\n",
    "14. Sample and balance"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dcf49658",
   "metadata": {},
   "source": [
    "## Remove strongly correlated features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "eb7d640b",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import pickle\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "dataset = pickle.load(open(\"imputed_dataset.pkl\", \"rb\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9baf06dc",
   "metadata": {},
   "source": [
    "To check for correlation we will use data from the oldest weather station, with ID"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "9af73a21",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'BRASILIA'"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "oldest_station = dataset.loc[dataset.full_time.dt.year == 2000, \"station_name\"].unique()[0]\n",
    "oldest_station"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "435fb3d7",
   "metadata": {},
   "outputs": [],
   "source": [
    "oldest_station_data = dataset.loc[dataset.station_name == oldest_station]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "573add06",
   "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>precipitation</th>\n",
       "      <th>pressure</th>\n",
       "      <th>pressure_max</th>\n",
       "      <th>pressure_min</th>\n",
       "      <th>solar_radiation</th>\n",
       "      <th>air_temperature</th>\n",
       "      <th>dp_temperature</th>\n",
       "      <th>air_temp_max</th>\n",
       "      <th>air_temp_min</th>\n",
       "      <th>dp_temp_max</th>\n",
       "      <th>dp_temp_min</th>\n",
       "      <th>rel_hum_max</th>\n",
       "      <th>rel_hum_min</th>\n",
       "      <th>Rel_humidity</th>\n",
       "      <th>wind_direction</th>\n",
       "      <th>wind_gust</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>919803</th>\n",
       "      <td>0.0</td>\n",
       "      <td>887.8</td>\n",
       "      <td>887.8</td>\n",
       "      <td>887.1</td>\n",
       "      <td>1710.329741</td>\n",
       "      <td>21.8</td>\n",
       "      <td>18.0</td>\n",
       "      <td>21.8</td>\n",
       "      <td>21.6</td>\n",
       "      <td>18.2</td>\n",
       "      <td>17.7</td>\n",
       "      <td>81.0</td>\n",
       "      <td>78.0</td>\n",
       "      <td>79.0</td>\n",
       "      <td>117.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>919804</th>\n",
       "      <td>0.0</td>\n",
       "      <td>888.2</td>\n",
       "      <td>888.3</td>\n",
       "      <td>887.8</td>\n",
       "      <td>1710.329741</td>\n",
       "      <td>21.8</td>\n",
       "      <td>17.9</td>\n",
       "      <td>21.9</td>\n",
       "      <td>21.5</td>\n",
       "      <td>18.5</td>\n",
       "      <td>17.9</td>\n",
       "      <td>83.0</td>\n",
       "      <td>78.0</td>\n",
       "      <td>79.0</td>\n",
       "      <td>135.0</td>\n",
       "      <td>2.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>919805</th>\n",
       "      <td>0.0</td>\n",
       "      <td>888.5</td>\n",
       "      <td>888.5</td>\n",
       "      <td>888.2</td>\n",
       "      <td>1710.329741</td>\n",
       "      <td>20.8</td>\n",
       "      <td>17.8</td>\n",
       "      <td>21.8</td>\n",
       "      <td>20.8</td>\n",
       "      <td>18.1</td>\n",
       "      <td>17.7</td>\n",
       "      <td>83.0</td>\n",
       "      <td>78.0</td>\n",
       "      <td>83.0</td>\n",
       "      <td>131.0</td>\n",
       "      <td>2.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>919806</th>\n",
       "      <td>0.0</td>\n",
       "      <td>888.1</td>\n",
       "      <td>888.6</td>\n",
       "      <td>888.1</td>\n",
       "      <td>1710.329741</td>\n",
       "      <td>21.2</td>\n",
       "      <td>17.9</td>\n",
       "      <td>21.3</td>\n",
       "      <td>20.5</td>\n",
       "      <td>18.2</td>\n",
       "      <td>17.8</td>\n",
       "      <td>85.0</td>\n",
       "      <td>81.0</td>\n",
       "      <td>81.0</td>\n",
       "      <td>120.0</td>\n",
       "      <td>2.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>919807</th>\n",
       "      <td>0.0</td>\n",
       "      <td>887.4</td>\n",
       "      <td>888.1</td>\n",
       "      <td>887.4</td>\n",
       "      <td>1710.329741</td>\n",
       "      <td>20.9</td>\n",
       "      <td>17.4</td>\n",
       "      <td>21.2</td>\n",
       "      <td>20.7</td>\n",
       "      <td>18.2</td>\n",
       "      <td>17.4</td>\n",
       "      <td>84.0</td>\n",
       "      <td>81.0</td>\n",
       "      <td>81.0</td>\n",
       "      <td>133.0</td>\n",
       "      <td>2.8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        precipitation  pressure  pressure_max  pressure_min  solar_radiation  \\\n",
       "919803            0.0     887.8         887.8         887.1      1710.329741   \n",
       "919804            0.0     888.2         888.3         887.8      1710.329741   \n",
       "919805            0.0     888.5         888.5         888.2      1710.329741   \n",
       "919806            0.0     888.1         888.6         888.1      1710.329741   \n",
       "919807            0.0     887.4         888.1         887.4      1710.329741   \n",
       "\n",
       "        air_temperature  dp_temperature  air_temp_max  air_temp_min  \\\n",
       "919803             21.8            18.0          21.8          21.6   \n",
       "919804             21.8            17.9          21.9          21.5   \n",
       "919805             20.8            17.8          21.8          20.8   \n",
       "919806             21.2            17.9          21.3          20.5   \n",
       "919807             20.9            17.4          21.2          20.7   \n",
       "\n",
       "        dp_temp_max  dp_temp_min  rel_hum_max  rel_hum_min  Rel_humidity  \\\n",
       "919803         18.2         17.7         81.0         78.0          79.0   \n",
       "919804         18.5         17.9         83.0         78.0          79.0   \n",
       "919805         18.1         17.7         83.0         78.0          83.0   \n",
       "919806         18.2         17.8         85.0         81.0          81.0   \n",
       "919807         18.2         17.4         84.0         81.0          81.0   \n",
       "\n",
       "        wind_direction  wind_gust  \n",
       "919803           117.0        4.0  \n",
       "919804           135.0        2.8  \n",
       "919805           131.0        2.1  \n",
       "919806           120.0        2.8  \n",
       "919807           133.0        2.8  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "station_numeric_features = oldest_station_data.iloc[:,1:17].copy()\n",
    "station_numeric_features.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "2466f533",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set_theme(style=\"white\")\n",
    "\n",
    "corr = station_numeric_features.corr()\n",
    "\n",
    "mask = np.triu(np.ones_like(corr, dtype=bool))\n",
    "\n",
    "f, ax = plt.subplots(figsize=(8, 8))\n",
    "\n",
    "cmap = sns.diverging_palette(230, 20, as_cmap=True)\n",
    "\n",
    "# Draw the heatmap with the mask and correct aspect ratio\n",
    "sns.heatmap(corr, mask=mask, cmap=cmap, vmax=.3, center=0,\n",
    "            square=True, linewidths=.5, cbar_kws={\"shrink\": .5})"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "65450087",
   "metadata": {},
   "source": [
    "There's a moderate relation between these features:\n",
    "- air_temperature\n",
    "- rel_hum_max\n",
    "- rel_hum_min\n",
    "- Rel_humidity\n",
    "\n",
    "The correlation is not very high, but for the purpose of this exercise we'll treat them as if it was high, and, in order to avoid multicollinearity when using the data for multiple linear regression, we'll drop the relative humidity features from the dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "c0c0b956",
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset.drop(columns = ['rel_hum_min', 'rel_hum_max', 'Rel_humidity'], inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "d08c869d",
   "metadata": {},
   "outputs": [],
   "source": [
    "pickle.dump(dataset, open(\"weak_correlation_dataset.pkl\", \"wb\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e1a4a5d9",
   "metadata": {},
   "source": [
    "## Remove outliers"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6bb99328",
   "metadata": {},
   "source": [
    "Let's look at a single feature, to identify if there are any outliers (observations that are distant from the other observations)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "cb520a90",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='air_temperature'>"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAV0AAAEMCAYAAABnWmXlAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAATcUlEQVR4nO3df2xV9f3H8de9Lb2FuNEi0BYaagppB3aTzjqMopHSAkMKuMioq2ykCgFEGUxCmW5u6n60c3ULgxkH08SZlRhZN37oqBRj5uKCmWwrBeyogNAWii3jR39Abz/fP/hyYoEixcv70tvn46/eH+fc9+m5eXJ6Ws71OeecAAAm/OEeAAD6EqILAIaILgAYIroAYIjoAoCh6O4eaGtrU1VVlYYMGaKoqCjLmQCg1woGg2psbFRGRoZiY2Mverzb6FZVVamgoOCaDgcAkerVV19VVlbWRfd3G90hQ4Z4CyYmJl67yQAggjQ0NKigoMBr6IW6je75UwqJiYlKTk6+NtMBQITq7rQsv0gDAENEFwAMEV0AMER0AcAQ0QUAQ0QXAAwRXQAwRHQBwBDRBQBDRBcADBFdADBEdAHAENEFAENEFwAMEV0AMER0AcAQ0QUAQ0QXAAx1+3E9QF/xu9/9TrW1tT1errm5WZIUHx8f6pGUmpqqefPmhXy9CD+iiz6vtrZWVdV7FRUb16Plgm3HJUkNzWdCOs/59SIyEV1AUlRsnAakTOzRMi0HtklSj5e70vUiMnFOFwAMEV0AMER0AcAQ0QUAQ0QXAAwRXQAwRHQBwBDRBQBDRBcADBFdADBEdAHAENEFAENEFwAMEV0AMER0AcAQ0QUAQ0QXAAwRXQAwRHQBwBDRBQBDRBcADBFdADBEdAHAENEFAENEFwAMEV0AMER0AcAQ0QUAQ0QXAAwRXQAwRHQBwBDRBQBDRBcADBFdADBEdAHAENEFAENEFwAMEV0AMER0AcAQ0QUAQ0QXV6yyslKVlZXhHgO9FO+fc6LDPQB6j4qKCklSdnZ2mCdBb8T75xyOdAHAENEFAENEFwAMEV0AMER0AcAQ0QUAQ0QXAAwRXQAwRHQBwBDRBQBDRBcADBFdADBEdAHAENEFAENEFwAMEV0AMER0AcAQ0QUAQ0QXAAwRXQAwRHQBwBDRBQBDRBcADBFdADBEdAHAENEFAENEFwAMEV0AMER0AcAQ0QUAQ0QXAAwRXQAwRHQBwBDRBQBDRBcADBFdADBEdAHAENEFAENEFwAMRV+LlTY1NenHP/6xDh8+rGAwqI6ODknS0KFD5ff71dDQoH79+ikqKkolJSUaOHCgSkpKNH/+fK1Zs0bOOeXk5GjNmjWSpEAgoCeffFKvvPKKgsGg2traVFdXp+XLl2vz5s26++679dvf/lYJCQk6ceKEiouLvXXee++9eu6557Rw4UKtXbtWZ86ckXNOgUBAS5Ys0apVqzRw4EA1NDR0uz0+n0/OuWvxreqV8vLytHHjxnCPgV6oqqpKeXl54R7jM0VHRyslJUVPPfWU4uPjQ7vukK7t/5WVlam2tvai+48ePep9ffbsWZ09e1bPPfecbr75ZlVXV+uXv/ylDh48KEmqqanxntve3q7i4mKdOnWqy/pKS0sVDAZVXV0tSTpy5IgkdVnn3r171dnZ6cX80+ssLS1VR0eHWltbL7s9BBfoWzo6OrRv3z6VlZVp4cKFIV13yE8vNDU1qaKi4oqff/DgQb311ltyznnBlS4O3YXBlc59Y5xzFz334MGDqqiokHPOO8q+VDjPP4ae6w1HK7i+VFVVhXuEHtu6dauam5tDus6QH+mWlZX1OGZnz54N9RgE1cDKlSvDPUJI1NbWqrMjKtxjeDo72lRbWxsx39/erKOjI+RHuyE/0n377bdDvUoACJvt27eHdH0hP9K955579MYbb4R6tbgO/exnPwv3CCGxcuVK7a49Eu4xPP7oWKWmJkTM9/e83npKasKECSFdX8iPdPPz8xUd3bOW9+vXL9Rj9HgGALhQdHS08vPzQ7rOkEd30KBBys3NveLnjxgxQjk5OfL5fBoxYoR3v8/n6/K8G2644aJlo6Oj5fP5LnruiBEjlJubK5/P58X3wuecXx5Xhz8ZQ09lZGSEe4QemzRpUsj/ZOya/OeI/Px8paamKhAIdAnb0KFDlZiYKOnc0W1sbKwef/xx5efna8yYMfre976n9PR0paWldTlxHQgEtGLFCqWlpWnkyJEaPny4fD6fli1bpjFjxmjBggWSpISEBPXv37/LOpctWya/369FixYpEAh48Q0EAlq2bJn69+/vzdSdSwUbQOSKjo7WyJEjQ36UK0k+180foR46dEgTJ07Utm3blJycHPIXRu9z/rfpkXau8fw53QEpE3u0XMuBbZLU4+WuZL2jI/CcbqS+fy70We3kvwEDgCGiCwCGiC4AGCK6AGCI6AKAIaILAIaILgAYIroAYIjoAoAhogsAhoguABgiugBgiOgCgCGiCwCGiC4AGCK6AGCI6AKAIaILAIaILgAYIroAYIjoAoAhogsAhoguABgiugBgiOgCgCGiCwCGiC4AGCK6AGCI6AKAIaILAIaILgAYIroAYIjoAoAhogsAhoguABgiugBgiOgCgKHocA+A3iM3NzfcI6AX4/1zDtHFFcvOzg73COjFeP+cw+kFADBEdAHAENEFAENEFwAMEV0AMER0AcAQ0QUAQ0QXAAwRXQAwRHQBwBDRBQBDRBcADBFdADBEdAHAENEFAENEFwAMEV0AMER0AcAQ0QUAQ0QXAAwRXQAwRHQBwBDRBQBDRBcADBFdADBEdAHAENEFAENEFwAMEV0AMER0AcAQ0QUAQ0QXAAwRXQAwRHQBwBDRBQBDRBcADBFdADBEdAHAENEFAENEFwAMEV0AMBQd7gGA60Gw7bhaDmzr8TKSerzcla03IaTrxPWD6KLPS01NvarlmptjJEnx8fGhHEdSwlXPhOsf0UWfN2/evHCPgD6Ec7oAYIjoAoAhogsAhoguABgiugBgiOgCgCGiCwCGiC4AGCK6AGCI6AKAIaILAIaILgAYIroAYIjoAoAhogsAhoguABgiugBgiOgCgKFuP64nGAxKkhoaGsyGAYDe7nwzzzf0Qt1Gt7GxUZJUUFBwDcYCgMjW2NiolJSUi+73OefcpRZoa2tTVVWVhgwZoqioqGs+IABEgmAwqMbGRmVkZCg2Nvaix7uNLgAg9PhFGgAYIroAYIjoAoAhogsAhoguABgiugBgiOgCgKGQRvfPf/6z8vLyNGbMGP3hD3/o8lhra6u++93vKjc3V1OmTNH27dtD+dImPvroI82ePVuTJ0/W7NmztX///nCP9LkUFxcrOztb6enp+vDDD737I2E7m5ubNW/ePE2ePFl5eXlavHixmpqaJEk7d+7U9OnTNXnyZBUWFuqTTz4J87Q9t2jRIk2fPl0zZ87Ut771Le3evVtSZOy7T/vNb37T5f0ZCftOLoT27t3rampq3PLly90rr7zS5bFVq1a5J554wjnn3EcffeTuuOMOd+rUqVC+/DU3Z84cV15e7pxzrry83M2ZMyfME30+O3bscHV1dW7ChAlu79693v2RsJ3Nzc3uvffe827//Oc/dytXrnTBYNDl5OS4HTt2OOecW716tSsqKgrXmFftxIkT3tcVFRVu5syZzrnI2HfnVVVVuYceesh7f0bKvgvpkW5aWppGjRolv//i1b7xxhuaPXu2JOmmm25SRkaG3nnnnVC+/DX1ySefqLq6WtOmTZMkTZs2TdXV1d7RU2+UlZWlpKSkLvdFynbGxcVp3Lhx3u2xY8eqrq5OVVVVCgQCysrKkiTl5+frzTffDNeYV+0LX/iC9/WpU6fk8/kiZt9J0pkzZ/T000/rRz/6kXdfpOy7bi94E2p1dXUaPny4dzspKalXXcGsvr5eCQkJ3nUooqKiNHToUNXX12vQoEFhni50InE7Ozs79cc//lHZ2dmqr6/XsGHDvMcGDRqkzs5OHT9+XHFxceEb8io88cQTevfdd+Wc09q1ayNq3/3617/W9OnTlZyc7N0XKfuuR9G97777VFdXd8nH/v73v3NhHFyXnnnmGQ0YMEAPPvigKioqwj1OyPzkJz+RJJWXl6ukpERLliwJ80Sh8cEHH6iqqkqPP/54uEe5JnoU3T/96U9X/ULDhg3T4cOHvX9x6+vru/z4d71LSkrSkSNHFAwGFRUVpWAwqKNHj17043lvF2nbWVxcrAMHDuiFF16Q3+9XUlJSlwOHpqYm+f3+XnWkdKGZM2fqhz/8oRITEyNi3+3YsUP79u3TxIkTJZ27Pu1DDz2kOXPmRMS+M/uTsSlTpmj9+vWSpP379+s///mP7rrrLquX/9xuvPFGjR49Wps2bZIkbdq0SaNHj+51P7Z9lkjaztLSUlVVVWn16tWKiYmRJGVkZKitrU3vv/++JKmsrExTpkwJ55g9dvr0adXX13u3KysrNXDgwIjZd/Pnz9ff/vY3VVZWqrKyUomJiVq3bp0efvjhXr/vpBBf2nHTpk0qKSnRiRMn1K9fP/Xv31+///3vNWrUKLW0tKioqEi7d++W3+/X8uXLlZOTE6qXNrFv3z4VFRXpxIkT+uIXv6ji4mKlpqaGe6yr9uyzz2rr1q06duyY4uPjFRcXp82bN0fEdtbU1GjatGm66aabvGuaJicna/Xq1frnP/+pp556Su3t7Ro+fLh+8YtfaPDgwWGe+ModO3ZMixYtUmtrq/x+vwYOHKgVK1bo5ptvjoh9d6Hs7Gy98MILSktL6/X7TuJ6ugBgiv+RBgCGiC4AGCK6AGCI6AKAIaILAIaILi7yl7/8RYWFheEeA4hI/MkYQio9PV1bt25VSkpKuEf53IqKipSQkKClS5eGexREEI500SMdHR3hHiEkLLYjUr5XCC2i24e9+OKLysnJUWZmpqZOnepdDGbDhg164IEHvOelp6fr1Vdf1aRJkzRp0qRu11dQUCBJmjFjhjIzM7VlyxZJ0vbt2zVjxgxlZWUpPz9fe/bs8ZbJzs7W2rVrlZeXp7Fjx+r73/++jh07pocffliZmZmaO3eu/ve//0mSDh06pPT0dK1fv17jx4/X+PHjtW7dOm9dnZ2d3jaNGzdOS5Ys0fHjx7ss+9prr+mee+7Rd77zHUnSY489pjvvvFO33nqrCgoKVFNTI0lav369Nm7cqHXr1ikzM1MLFizwvhcHDhzwXrOoqEjPP/+8JOkf//iH7r77br344ou68847tXLlysvOhD4qjNfyRZht2bLFNTQ0uGAw6DZv3uxuueUWd+TIEff666+7/Px873lpaWlu7ty5rrm52bW2tl52nWlpaW7//v3e7V27drnbb7/d7dy503V0dLgNGza4CRMmuPb2dueccxMmTHCzZs1yjY2NrqGhwd1+++1u5syZbteuXa6trc3NmTPHrVq1yjnn3Mcff+zS0tLc0qVL3enTp92ePXvcuHHj3Lvvvuucc+7ll192s2bNcvX19a69vd394Ac/cEuXLu2y7PLly93p06e97XjttdfcyZMnXXt7u3v22Wfd9OnTvdlXrFjhSktLL7t9n37Oe++950aPHu1KSkpce3u7a21tvexM6Js40u3Dvv71ryshIUF+v19Tp05VSkqK/v3vf1/yufPnz1dcXJx3HYMrtX79es2ePVu33HKLoqKidN9996lfv37auXOn95wHH3xQgwcPVkJCgrKysvSVr3xFY8aMUSAQUG5urqqrq7us85FHHtGAAQOUnp6ub3zjG94FXsrKyrR06VIlJiYqJiZGixcv1l//+tcuP+Y/+uijGjBggLcd999/v2644QbFxMTo0Ucf1Z49e3Ty5MkebeOn+f1+PfbYY4qJiVFsbOwVzYS+xewi5rj+lJeX66WXXtLhw4clSS0tLWpubr7kdZGv9vKAdXV1Ki8v7/KZeWfPntXRo0e925++YEkgEOhyOzY2Vi0tLd3OMnz4cO/zs+rq6vTII490+eQSv9/f5XO0EhMTva+DwaCef/55vfnmm95lAqVzn6/26U9m6In4+HgFAgHv9uVmSkhIuKrXQO9GdPuow4cP68knn9TLL7+szMxMRUVFacaMGd0+3+fzXdXrJCUlacGCBVq4cOHVjnqR+vp6jRw5UtK5qA0dOlTSuaD+9Kc/1a233nrRMocOHZLUdTs2btyobdu26aWXXlJycrJOnjyp2267Te7//6DnUtvcv39/tba2ercbGxu7xPPCZS43E/omTi/0Ua2trfL5fN61Vl9//XXvl0ifx+DBg/Xxxx97t2fNmqWysjL961//knNOLS0tevvtt3Xq1Kmrfo01a9aotbVVNTU12rBhg6ZOnSpJeuCBB/SrX/3KO3JvamrSW2+91e16Tp8+rZiYGMXHx6u1tVWlpaVdHr/xxhu9WJ/3pS99SZs2bVIwGNQ777yjHTt2XHbWns6EyEd0+6hRo0apsLBQ+fn5uuOOO/Thhx/qq1/96ude7+LFi1VUVKSsrCxt2bJFX/7yl/XMM8/o6aef1m233aZJkyZpw4YNn+s1vva1ryk3N1dz585VYWGhxo8fL0n69re/rezsbBUWFiozM1Pf/OY3uz1HLZ37xIVhw4bprrvu0r333quxY8d2efz+++/Xf//7X2VlZWnRokWSzn0u2fbt25WVlaWNGzd+5jWhezoTIh//OQK9xqFDhzRx4kTt2rVL0dGcGUPvxJEuABjicAE98v7772vevHmXfOyDDz4wngbofTi9AACGOL0AAIaILgAYIroAYIjoAoAhogsAhoguABj6P9vruV5DfL9FAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.boxplot(x=dataset['air_temperature'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "98541477",
   "metadata": {},
   "source": [
    "Clearly there are many outliers, but what is an outlier in one station may be a frequent value in another.\n",
    "\n",
    "Let's look at the PDF of this feature."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "b013dbe9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:ylabel='Density'>"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dataset.air_temperature.plot.kde();"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3e066765",
   "metadata": {},
   "source": [
    "Let's boxplot all features for three randomly picked stations, to observe if outliers fall in similar ranges between stations."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "964ef048",
   "metadata": {},
   "outputs": [],
   "source": [
    "# array with station names\n",
    "stations = dataset.station_name.unique()\n",
    "\n",
    "# fix random state for repeatability\n",
    "np.random.seed(321)\n",
    "\n",
    "#  select randomly three stations to visualize\n",
    "stations_picked = np.random.randint(len(stations), size=3)\n",
    "\n",
    "# number of numeric features present in the dataset\n",
    "num_features = 14\n",
    "\n",
    "# position offset for the first numeric feature within datates\n",
    "feature_offset = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "a4d4ad54",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x2880 with 42 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# grid of stations x features\n",
    "fig, axs = plt.subplots(num_features, \n",
    "                        len(stations_picked), \n",
    "                        constrained_layout=True, \n",
    "                        figsize=(18, 40))\n",
    "\n",
    "for i in range(0,num_features): # iterate features\n",
    "    for j in range(0,len(stations_picked)): # iterate stations\n",
    "#        print('station: ' + str(j) + ' ' + stations[j] + ', feature: ' + str(i) + ' ' + dataset.columns[feature_offset + i])\n",
    "        axs[i,j].boxplot(x=dataset.loc[dataset.station_name == stations[j], dataset.columns[feature_offset + i]])\n",
    "        axs[i,j].set_ylabel(dataset.columns[feature_offset + i])\n",
    "        axs[i,j].set_title(stations[j])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "421b3c55",
   "metadata": {},
   "source": [
    "We can see that there are many outliers (circles outside the moustaches) but the ranges are specific to each station.\n",
    "\n",
    "Next will calculate the ranges of the outliers for each feature in each station.\n",
    "\n",
    "**Warning:** In a real world application we  should not use values from the test dataset for outlier removal. However, in order to keep this exercise simple, we will ignore this and remove the outliers before spliting the dataset."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e656b05d",
   "metadata": {},
   "source": [
    "The outliers fall below **Q1 - 1.5 x IQR** or above **Q3 + 1.5 x IQR**.\n",
    "\n",
    "\n",
    "```\n",
    "     Q1-1.5IQR   Q1   median  Q3   Q3+1.5IQR\n",
    "                  |-----:-----|\n",
    "  o      |--------|     :     |--------|    o  o\n",
    "                  |-----:-----|\n",
    "\n",
    "outliers          <----------->            outliers\n",
    "            inter-quartile range (IQR)\n",
    "\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6079de5d",
   "metadata": {},
   "source": [
    "As we saw in the boxplots earlier, if we use the IQR to determine outliers, some features will have a very high number of 'outliers'. We will not do outlier removal on these features. As the objective of this exercise is not to focus on outlier removal, we will not implement alternatives for these features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "3afa04e7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dataset.loc[dataset.station_name == 'PARANOA (COOPA-DF)'].precipitation.plot.kde();"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "28227018",
   "metadata": {},
   "source": [
    "Also note that wind_direction is in degrees, which rolls around in 360, so a specific strategy should be emplyed for this feature. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "99d93a1f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "processing station 1 PARANOA (COOPA-DF) with 36453 rows of data\n",
      "pressure_max: should be between 892.75 and 906.75. Removed 187 outliers (0.5%).\n",
      "pressure_min: should be between 892.10 and 906.50. Removed 3 outliers (0.0%).\n",
      "air_temperature: should be between 10.60 and 33.00. Removed 210 outliers (0.6%).\n",
      "air_temp_max: should be between 10.10 and 34.90. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 10.85 and 31.25. Removed 204 outliers (0.6%).\n",
      "dp_temp_max: should be between 2.25 and 29.05. Removed 35 outliers (0.1%).\n",
      "dp_temp_min: should be between 1.00 and 28.20. Removed 20 outliers (0.1%).\n",
      "wind_speed: should be between -1.25 and 5.55. Removed 622 outliers (1.7%).\n",
      "Removed 1281 outliers overall (3.5%).\n",
      "\n",
      "processing station 2 BELA VISTA with 77708 rows of data\n",
      "pressure_max: should be between 978.35 and 997.95. Removed 2231 outliers (2.9%).\n",
      "pressure_min: should be between 977.60 and 997.60. Removed 23 outliers (0.0%).\n",
      "air_temperature: should be between 7.85 and 40.25. Removed 366 outliers (0.5%).\n",
      "air_temp_max: should be between 7.85 and 41.85. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 7.85 and 38.65. Removed 103 outliers (0.1%).\n",
      "dp_temp_max: should be between 8.05 and 29.25. Removed 1512 outliers (1.9%).\n",
      "dp_temp_min: should be between 6.70 and 28.30. Removed 292 outliers (0.4%).\n",
      "wind_speed: should be between -0.85 and 4.35. Removed 2214 outliers (2.8%).\n",
      "Removed 6741 outliers overall (8.7%).\n",
      "\n",
      "processing station 3 NOVA UBIRATA with 96595 rows of data\n",
      "pressure_max: should be between 953.50 and 966.30. Removed 681 outliers (0.7%).\n",
      "pressure_min: should be between 952.50 and 966.10. Removed 22 outliers (0.0%).\n",
      "air_temperature: should be between 12.00 and 38.40. Removed 63 outliers (0.1%).\n",
      "air_temp_max: should be between 11.35 and 40.55. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 12.65 and 36.25. Removed 167 outliers (0.2%).\n",
      "dp_temp_max: should be between 6.90 and 30.90. Removed 816 outliers (0.8%).\n",
      "dp_temp_min: should be between 5.25 and 30.45. Removed 155 outliers (0.2%).\n",
      "wind_speed: should be between -2.35 and 6.85. Removed 1258 outliers (1.3%).\n",
      "Removed 3162 outliers overall (3.3%).\n",
      "\n",
      "processing station 4 GOIANIA with 169182 rows of data\n",
      "pressure_max: should be between 925.65 and 939.65. Removed 3987 outliers (2.4%).\n",
      "pressure_min: should be between 924.90 and 939.30. Removed 309 outliers (0.2%).\n",
      "air_temperature: should be between 10.55 and 37.35. Removed 487 outliers (0.3%).\n",
      "air_temp_max: should be between 10.05 and 39.25. Removed 13 outliers (0.0%).\n",
      "air_temp_min: should be between 10.90 and 35.70. Removed 465 outliers (0.3%).\n",
      "dp_temp_max: should be between 2.95 and 29.75. Removed 863 outliers (0.5%).\n",
      "dp_temp_min: should be between 1.30 and 29.30. Removed 173 outliers (0.1%).\n",
      "wind_speed: should be between -2.10 and 4.30. Removed 2566 outliers (1.5%).\n",
      "Removed 8863 outliers overall (5.2%).\n",
      "\n",
      "processing station 5 CARLINDA with 107197 rows of data\n",
      "pressure_max: should be between 972.40 and 984.40. Removed 603 outliers (0.6%).\n",
      "pressure_min: should be between 971.55 and 983.95. Removed 78 outliers (0.1%).\n",
      "air_temperature: should be between 13.95 and 36.75. Removed 43 outliers (0.0%).\n",
      "air_temp_max: should be between 13.20 and 38.80. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 14.70 and 34.70. Removed 365 outliers (0.3%).\n",
      "dp_temp_max: should be between 13.60 and 27.20. Removed 8099 outliers (7.6%).\n",
      "dp_temp_min: should be between 11.70 and 26.90. Removed 340 outliers (0.3%).\n",
      "wind_speed: should be between -2.85 and 4.75. Removed 1020 outliers (1.0%).\n",
      "Removed 10548 outliers overall (9.8%).\n",
      "\n",
      "processing station 6 SIDROLANDIA with 106390 rows of data\n",
      "pressure_max: should be between 950.10 and 968.50. Removed 1086 outliers (1.0%).\n",
      "pressure_min: should be between 949.35 and 968.15. Removed 48 outliers (0.0%).\n",
      "air_temperature: should be between 10.50 and 37.70. Removed 1306 outliers (1.2%).\n",
      "air_temp_max: should be between 10.40 and 39.20. Removed 1 outliers (0.0%).\n",
      "air_temp_min: should be between 10.75 and 35.95. Removed 596 outliers (0.6%).\n",
      "dp_temp_max: should be between 4.90 and 30.50. Removed 637 outliers (0.6%).\n",
      "dp_temp_min: should be between 3.40 and 29.80. Removed 154 outliers (0.1%).\n",
      "wind_speed: should be between -0.65 and 4.55. Removed 3673 outliers (3.5%).\n",
      "Removed 7501 outliers overall (7.1%).\n",
      "\n",
      "processing station 7 ITAQUIRAI with 96544 rows of data\n",
      "pressure_max: should be between 963.50 and 985.10. Removed 826 outliers (0.9%).\n",
      "pressure_min: should be between 962.75 and 984.75. Removed 27 outliers (0.0%).\n",
      "air_temperature: should be between 10.45 and 35.65. Removed 1987 outliers (2.1%).\n",
      "air_temp_max: should be between 10.35 and 37.15. Removed 28 outliers (0.0%).\n",
      "air_temp_min: should be between 10.70 and 33.90. Removed 759 outliers (0.8%).\n",
      "dp_temp_max: should be between 5.20 and 30.80. Removed 873 outliers (0.9%).\n",
      "dp_temp_min: should be between 3.95 and 29.95. Removed 227 outliers (0.2%).\n",
      "wind_speed: should be between -3.95 and 6.85. Removed 783 outliers (0.8%).\n",
      "Removed 5510 outliers overall (5.7%).\n",
      "\n",
      "processing station 8 COTRIGUACU with 101521 rows of data\n",
      "pressure_max: should be between 975.95 and 988.35. Removed 767 outliers (0.8%).\n",
      "pressure_min: should be between 975.10 and 987.90. Removed 58 outliers (0.1%).\n",
      "air_temperature: should be between 13.85 and 36.65. Removed 54 outliers (0.1%).\n",
      "air_temp_max: should be between 12.65 and 39.45. Removed 1 outliers (0.0%).\n",
      "air_temp_min: should be between 14.90 and 34.10. Removed 927 outliers (0.9%).\n",
      "dp_temp_max: should be between 16.50 and 26.90. Removed 6744 outliers (6.6%).\n",
      "dp_temp_min: should be between 15.35 and 26.15. Removed 1293 outliers (1.3%).\n",
      "wind_speed: should be between -1.40 and 3.40. Removed 1367 outliers (1.3%).\n",
      "Removed 11211 outliers overall (11.0%).\n",
      "\n",
      "processing station 9 SAO MIGUEL DO ARAGUAIA with 22531 rows of data\n",
      "pressure_max: should be between 980.65 and 994.65. Removed 84 outliers (0.4%).\n",
      "pressure_min: should be between 980.05 and 994.05. Removed 21 outliers (0.1%).\n",
      "air_temperature: should be between 12.45 and 43.25. Removed 0 outliers (0.0%).\n",
      "air_temp_max: should be between 12.35 and 44.75. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 12.70 and 41.50. Removed 0 outliers (0.0%).\n",
      "dp_temp_max: should be between 6.80 and 33.20. Removed 21 outliers (0.1%).\n",
      "dp_temp_min: should be between 4.45 and 32.85. Removed 5 outliers (0.0%).\n",
      "wind_speed: should be between -2.80 and 6.00. Removed 375 outliers (1.7%).\n",
      "Removed 506 outliers overall (2.2%).\n",
      "\n",
      "processing station 10 NHUMIRIM with 96979 rows of data\n",
      "pressure_max: should be between 990.30 and 1009.50. Removed 2456 outliers (2.5%).\n",
      "pressure_min: should be between 989.55 and 1009.15. Removed 25 outliers (0.0%).\n",
      "air_temperature: should be between 11.10 and 40.70. Removed 575 outliers (0.6%).\n",
      "air_temp_max: should be between 10.95 and 42.55. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 11.35 and 38.95. Removed 272 outliers (0.3%).\n",
      "dp_temp_max: should be between 8.65 and 32.25. Removed 884 outliers (0.9%).\n",
      "dp_temp_min: should be between 7.15 and 31.55. Removed 277 outliers (0.3%).\n",
      "wind_speed: should be between -4.35 and 7.25. Removed 428 outliers (0.4%).\n",
      "Removed 4917 outliers overall (5.1%).\n",
      "\n",
      "processing station 11 QUERENCIA with 90149 rows of data\n",
      "pressure_max: should be between 964.30 and 980.30. Removed 443 outliers (0.5%).\n",
      "pressure_min: should be between 963.70 and 979.70. Removed 70 outliers (0.1%).\n",
      "air_temperature: should be between 12.30 and 38.70. Removed 18 outliers (0.0%).\n",
      "air_temp_max: should be between 11.50 and 41.10. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 13.10 and 36.30. Removed 275 outliers (0.3%).\n",
      "dp_temp_max: should be between 13.10 and 28.30. Removed 5542 outliers (6.1%).\n",
      "dp_temp_min: should be between 10.85 and 28.05. Removed 624 outliers (0.7%).\n",
      "wind_speed: should be between -2.60 and 4.60. Removed 870 outliers (1.0%).\n",
      "Removed 7842 outliers overall (8.7%).\n",
      "\n",
      "processing station 12 RIO VERDE with 105688 rows of data\n",
      "pressure_max: should be between 919.60 and 934.80. Removed 609 outliers (0.6%).\n",
      "pressure_min: should be between 918.70 and 934.70. Removed 9 outliers (0.0%).\n",
      "air_temperature: should be between 10.50 and 36.10. Removed 475 outliers (0.4%).\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "air_temp_max: should be between 10.25 and 37.85. Removed 2 outliers (0.0%).\n",
      "air_temp_min: should be between 10.90 and 34.10. Removed 547 outliers (0.5%).\n",
      "dp_temp_max: should be between 0.50 and 30.90. Removed 270 outliers (0.3%).\n",
      "dp_temp_min: should be between -1.00 and 30.20. Removed 50 outliers (0.0%).\n",
      "wind_speed: should be between -2.55 and 5.85. Removed 1166 outliers (1.1%).\n",
      "Removed 3128 outliers overall (3.0%).\n",
      "\n",
      "processing station 13 APIACAS with 98811 rows of data\n",
      "pressure_max: should be between 981.10 and 993.10. Removed 702 outliers (0.7%).\n",
      "pressure_min: should be between 980.10 and 992.90. Removed 17 outliers (0.0%).\n",
      "air_temperature: should be between 14.00 and 38.80. Removed 5 outliers (0.0%).\n",
      "air_temp_max: should be between 13.10 and 41.10. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 15.15 and 36.35. Removed 88 outliers (0.1%).\n",
      "dp_temp_max: should be between 17.65 and 26.85. Removed 8423 outliers (8.5%).\n",
      "dp_temp_min: should be between 16.35 and 26.35. Removed 839 outliers (0.8%).\n",
      "wind_speed: should be between -1.95 and 3.25. Removed 1251 outliers (1.3%).\n",
      "Removed 11325 outliers overall (11.5%).\n",
      "\n",
      "processing station 14 SALTO DO CEU with 102203 rows of data\n",
      "pressure_max: should be between 969.95 and 985.55. Removed 2192 outliers (2.1%).\n",
      "pressure_min: should be between 968.95 and 985.35. Removed 9 outliers (0.0%).\n",
      "air_temperature: should be between 13.15 and 36.75. Removed 1337 outliers (1.3%).\n",
      "air_temp_max: should be between 12.35 and 39.15. Removed 2 outliers (0.0%).\n",
      "air_temp_min: should be between 13.80 and 34.60. Removed 1511 outliers (1.5%).\n",
      "dp_temp_max: should be between 11.40 and 29.00. Removed 1895 outliers (1.9%).\n",
      "dp_temp_min: should be between 9.50 and 28.70. Removed 258 outliers (0.3%).\n",
      "wind_speed: should be between -1.65 and 4.35. Removed 665 outliers (0.7%).\n",
      "Removed 7869 outliers overall (7.7%).\n",
      "\n",
      "processing station 15 COXIM with 119256 rows of data\n",
      "pressure_max: should be between 976.05 and 991.65. Removed 2628 outliers (2.2%).\n",
      "pressure_min: should be between 975.30 and 991.30. Removed 85 outliers (0.1%).\n",
      "air_temperature: should be between 11.30 and 39.30. Removed 631 outliers (0.5%).\n",
      "air_temp_max: should be between 11.30 and 40.90. Removed 4 outliers (0.0%).\n",
      "air_temp_min: should be between 11.85 and 37.05. Removed 855 outliers (0.7%).\n",
      "dp_temp_max: should be between 7.30 and 31.30. Removed 1469 outliers (1.2%).\n",
      "dp_temp_min: should be between 5.55 and 30.75. Removed 279 outliers (0.2%).\n",
      "wind_speed: should be between -2.10 and 3.50. Removed 4520 outliers (3.8%).\n",
      "Removed 10471 outliers overall (8.8%).\n",
      "\n",
      "processing station 16 SAO GABRIEL DO OESTE with 112726 rows of data\n",
      "pressure_max: should be between 932.15 and 948.55. Removed 797 outliers (0.7%).\n",
      "pressure_min: should be between 931.40 and 948.20. Removed 55 outliers (0.0%).\n",
      "air_temperature: should be between 10.60 and 36.20. Removed 1008 outliers (0.9%).\n",
      "air_temp_max: should be between 10.35 and 37.95. Removed 6 outliers (0.0%).\n",
      "air_temp_min: should be between 10.60 and 34.60. Removed 430 outliers (0.4%).\n",
      "dp_temp_max: should be between 3.35 and 30.95. Removed 1145 outliers (1.0%).\n",
      "dp_temp_min: should be between 2.00 and 30.00. Removed 210 outliers (0.2%).\n",
      "wind_speed: should be between -4.05 and 6.75. Removed 1581 outliers (1.4%).\n",
      "Removed 5232 outliers overall (4.6%).\n",
      "\n",
      "processing station 17 PONTA PORA with 146244 rows of data\n",
      "pressure_max: should be between 928.55 and 947.35. Removed 1098 outliers (0.8%).\n",
      "pressure_min: should be between 928.05 and 946.85. Removed 144 outliers (0.1%).\n",
      "air_temperature: should be between 7.95 and 36.35. Removed 1591 outliers (1.1%).\n",
      "air_temp_max: should be between 8.00 and 37.60. Removed 18 outliers (0.0%).\n",
      "air_temp_min: should be between 8.15 and 34.95. Removed 475 outliers (0.3%).\n",
      "dp_temp_max: should be between 4.05 and 29.25. Removed 1431 outliers (1.0%).\n",
      "dp_temp_min: should be between 2.75 and 28.75. Removed 374 outliers (0.3%).\n",
      "wind_speed: should be between -1.45 and 6.15. Removed 1461 outliers (1.0%).\n",
      "Removed 6592 outliers overall (4.5%).\n",
      "\n",
      "processing station 18 MARACAJU with 106918 rows of data\n",
      "pressure_max: should be between 958.40 and 979.20. Removed 822 outliers (0.8%).\n",
      "pressure_min: should be between 957.80 and 978.60. Removed 116 outliers (0.1%).\n",
      "air_temperature: should be between 7.25 and 38.85. Removed 639 outliers (0.6%).\n",
      "air_temp_max: should be between 7.25 and 40.45. Removed 2 outliers (0.0%).\n",
      "air_temp_min: should be between 7.40 and 37.00. Removed 298 outliers (0.3%).\n",
      "dp_temp_max: should be between 4.20 and 31.40. Removed 502 outliers (0.5%).\n",
      "dp_temp_min: should be between 2.60 and 30.60. Removed 128 outliers (0.1%).\n",
      "wind_speed: should be between -2.45 and 4.35. Removed 1735 outliers (1.6%).\n",
      "Removed 4242 outliers overall (4.0%).\n",
      "\n",
      "processing station 19 ALTO PARAISO DE GOIAS with 119051 rows of data\n",
      "pressure_max: should be between 871.20 and 883.20. Removed 724 outliers (0.6%).\n",
      "pressure_min: should be between 870.55 and 882.95. Removed 48 outliers (0.0%).\n",
      "air_temperature: should be between 10.20 and 31.00. Removed 408 outliers (0.3%).\n",
      "air_temp_max: should be between 9.95 and 32.75. Removed 18 outliers (0.0%).\n",
      "air_temp_min: should be between 10.45 and 29.25. Removed 599 outliers (0.5%).\n",
      "dp_temp_max: should be between 0.35 and 27.15. Removed 2087 outliers (1.8%).\n",
      "dp_temp_min: should be between -0.85 and 25.95. Removed 1603 outliers (1.3%).\n",
      "wind_speed: should be between -3.05 and 6.95. Removed 256 outliers (0.2%).\n",
      "Removed 5743 outliers overall (4.8%).\n",
      "\n",
      "processing station 20 BRASNORTE (NOVO MUNDO) with 88451 rows of data\n",
      "pressure_max: should be between 957.00 and 969.80. Removed 830 outliers (0.9%).\n",
      "pressure_min: should be between 956.15 and 969.35. Removed 71 outliers (0.1%).\n",
      "air_temperature: should be between 13.50 and 36.70. Removed 281 outliers (0.3%).\n",
      "air_temp_max: should be between 13.00 and 38.60. Removed 1 outliers (0.0%).\n",
      "air_temp_min: should be between 14.40 and 34.40. Removed 603 outliers (0.7%).\n",
      "dp_temp_max: should be between 10.05 and 29.65. Removed 1815 outliers (2.1%).\n",
      "dp_temp_min: should be between 8.65 and 29.05. Removed 261 outliers (0.3%).\n",
      "wind_speed: should be between -3.65 and 6.35. Removed 515 outliers (0.6%).\n",
      "Removed 4377 outliers overall (4.9%).\n",
      "\n",
      "processing station 21 JATAI with 118608 rows of data\n",
      "pressure_max: should be between 930.80 and 946.80. Removed 588 outliers (0.5%).\n",
      "pressure_min: should be between 930.05 and 946.45. Removed 55 outliers (0.0%).\n",
      "air_temperature: should be between 9.55 and 37.15. Removed 722 outliers (0.6%).\n",
      "air_temp_max: should be between 9.15 and 39.15. Removed 1 outliers (0.0%).\n",
      "air_temp_min: should be between 9.85 and 35.05. Removed 677 outliers (0.6%).\n",
      "dp_temp_max: should be between 3.40 and 30.60. Removed 902 outliers (0.8%).\n",
      "dp_temp_min: should be between 1.15 and 30.35. Removed 96 outliers (0.1%).\n",
      "wind_speed: should be between -2.85 and 5.55. Removed 1311 outliers (1.1%).\n",
      "Removed 4352 outliers overall (3.7%).\n",
      "\n",
      "processing station 22 GAUCHA DO NORTE with 97835 rows of data\n",
      "pressure_max: should be between 962.90 and 975.70. Removed 848 outliers (0.9%).\n",
      "pressure_min: should be between 962.15 and 975.35. Removed 54 outliers (0.1%).\n",
      "air_temperature: should be between 12.20 and 40.20. Removed 6 outliers (0.0%).\n",
      "air_temp_max: should be between 11.55 and 42.35. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 13.10 and 37.90. Removed 55 outliers (0.1%).\n",
      "dp_temp_max: should be between 7.85 and 29.85. Removed 1534 outliers (1.6%).\n",
      "dp_temp_min: should be between 6.30 and 29.50. Removed 311 outliers (0.3%).\n",
      "wind_speed: should be between -2.85 and 4.75. Removed 470 outliers (0.5%).\n",
      "Removed 3278 outliers overall (3.4%).\n",
      "\n",
      "processing station 23 SAO FELIX  DO ARAGUAIA with 83786 rows of data\n",
      "pressure_max: should be between 982.70 and 994.70. Removed 467 outliers (0.6%).\n",
      "pressure_min: should be between 981.95 and 994.35. Removed 54 outliers (0.1%).\n",
      "air_temperature: should be between 14.15 and 39.35. Removed 50 outliers (0.1%).\n",
      "air_temp_max: should be between 13.65 and 41.25. Removed 1 outliers (0.0%).\n",
      "air_temp_min: should be between 15.05 and 37.05. Removed 352 outliers (0.4%).\n",
      "dp_temp_max: should be between 16.85 and 26.85. Removed 10226 outliers (12.2%).\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dp_temp_min: should be between 15.35 and 26.15. Removed 588 outliers (0.7%).\n",
      "wind_speed: should be between -2.40 and 4.80. Removed 904 outliers (1.1%).\n",
      "Removed 12642 outliers overall (15.1%).\n",
      "\n",
      "processing station 24 SANTO ANTONIO DO LESTE with 103850 rows of data\n",
      "pressure_max: should be between 931.85 and 945.05. Removed 806 outliers (0.8%).\n",
      "pressure_min: should be between 931.10 and 944.70. Removed 49 outliers (0.0%).\n",
      "air_temperature: should be between 12.05 and 36.45. Removed 447 outliers (0.4%).\n",
      "air_temp_max: should be between 11.40 and 38.60. Removed 1 outliers (0.0%).\n",
      "air_temp_min: should be between 12.45 and 34.45. Removed 502 outliers (0.5%).\n",
      "dp_temp_max: should be between 5.80 and 29.80. Removed 1604 outliers (1.5%).\n",
      "dp_temp_min: should be between 3.90 and 29.50. Removed 120 outliers (0.1%).\n",
      "wind_speed: should be between -1.95 and 6.45. Removed 1697 outliers (1.6%).\n",
      "Removed 5226 outliers overall (5.0%).\n",
      "\n",
      "processing station 25 PARANATINGA with 79297 rows of data\n",
      "pressure_max: should be between 951.75 and 964.95. Removed 629 outliers (0.8%).\n",
      "pressure_min: should be between 950.90 and 964.50. Removed 55 outliers (0.1%).\n",
      "air_temperature: should be between 12.40 and 38.00. Removed 91 outliers (0.1%).\n",
      "air_temp_max: should be between 11.75 and 40.15. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 13.05 and 35.85. Removed 463 outliers (0.6%).\n",
      "dp_temp_max: should be between 10.90 and 28.50. Removed 4669 outliers (5.9%).\n",
      "dp_temp_min: should be between 8.50 and 28.50. Removed 175 outliers (0.2%).\n",
      "wind_speed: should be between -3.00 and 5.80. Removed 912 outliers (1.2%).\n",
      "Removed 6994 outliers overall (8.8%).\n",
      "\n",
      "processing station 26 SORRISO with 139291 rows of data\n",
      "pressure_max: should be between 962.75 and 975.95. Removed 936 outliers (0.7%).\n",
      "pressure_min: should be between 962.15 and 975.35. Removed 201 outliers (0.1%).\n",
      "air_temperature: should be between 13.25 and 38.45. Removed 98 outliers (0.1%).\n",
      "air_temp_max: should be between 12.75 and 40.35. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 14.00 and 36.40. Removed 275 outliers (0.2%).\n",
      "dp_temp_max: should be between 9.90 and 29.10. Removed 2394 outliers (1.7%).\n",
      "dp_temp_min: should be between 9.00 and 28.20. Removed 544 outliers (0.4%).\n",
      "wind_speed: should be between -3.15 and 5.25. Removed 1219 outliers (0.9%).\n",
      "Removed 5667 outliers overall (4.1%).\n",
      "\n",
      "processing station 27 DOURADOS with 110894 rows of data\n",
      "pressure_max: should be between 950.25 and 970.65. Removed 830 outliers (0.7%).\n",
      "pressure_min: should be between 949.75 and 970.15. Removed 76 outliers (0.1%).\n",
      "air_temperature: should be between 9.95 and 36.75. Removed 1532 outliers (1.4%).\n",
      "air_temp_max: should be between 9.85 and 38.25. Removed 13 outliers (0.0%).\n",
      "air_temp_min: should be between 9.90 and 35.50. Removed 359 outliers (0.3%).\n",
      "dp_temp_max: should be between 3.70 and 30.90. Removed 526 outliers (0.5%).\n",
      "dp_temp_min: should be between 2.55 and 30.15. Removed 232 outliers (0.2%).\n",
      "wind_speed: should be between -1.40 and 5.80. Removed 1403 outliers (1.3%).\n",
      "Removed 4971 outliers overall (4.5%).\n",
      "\n",
      "processing station 28 GUARANTA DO NORTE with 126762 rows of data\n",
      "pressure_max: should be between 973.70 and 985.70. Removed 609 outliers (0.5%).\n",
      "pressure_min: should be between 972.85 and 985.25. Removed 118 outliers (0.1%).\n",
      "air_temperature: should be between 13.30 and 38.10. Removed 46 outliers (0.0%).\n",
      "air_temp_max: should be between 12.50 and 40.50. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 14.35 and 35.55. Removed 662 outliers (0.5%).\n",
      "dp_temp_max: should be between 16.25 and 27.05. Removed 8660 outliers (6.8%).\n",
      "dp_temp_min: should be between 14.85 and 26.45. Removed 1777 outliers (1.4%).\n",
      "wind_speed: should be between -2.55 and 4.25. Removed 1431 outliers (1.1%).\n",
      "Removed 13303 outliers overall (10.5%).\n",
      "\n",
      "processing station 29 AGUA CLARA with 91511 rows of data\n",
      "pressure_max: should be between 966.85 and 984.85. Removed 1201 outliers (1.3%).\n",
      "pressure_min: should be between 966.10 and 984.50. Removed 26 outliers (0.0%).\n",
      "air_temperature: should be between 9.35 and 40.15. Removed 654 outliers (0.7%).\n",
      "air_temp_max: should be between 9.35 and 41.75. Removed 2 outliers (0.0%).\n",
      "air_temp_min: should be between 9.75 and 38.15. Removed 374 outliers (0.4%).\n",
      "dp_temp_max: should be between 5.35 and 31.35. Removed 810 outliers (0.9%).\n",
      "dp_temp_min: should be between 3.75 and 30.55. Removed 218 outliers (0.2%).\n",
      "wind_speed: should be between -1.60 and 4.00. Removed 1839 outliers (2.0%).\n",
      "Removed 5124 outliers overall (5.6%).\n",
      "\n",
      "processing station 30 JUTI with 104912 rows of data\n",
      "pressure_max: should be between 959.70 and 981.30. Removed 725 outliers (0.7%).\n",
      "pressure_min: should be between 958.95 and 980.95. Removed 31 outliers (0.0%).\n",
      "air_temperature: should be between 8.65 and 37.85. Removed 992 outliers (0.9%).\n",
      "air_temp_max: should be between 8.55 and 39.35. Removed 2 outliers (0.0%).\n",
      "air_temp_min: should be between 8.60 and 36.60. Removed 248 outliers (0.2%).\n",
      "dp_temp_max: should be between 3.55 and 31.15. Removed 383 outliers (0.4%).\n",
      "dp_temp_min: should be between 2.65 and 30.25. Removed 202 outliers (0.2%).\n",
      "wind_speed: should be between -2.70 and 5.30. Removed 633 outliers (0.6%).\n",
      "Removed 3216 outliers overall (3.1%).\n",
      "\n",
      "processing station 31 TRES LAGOAS with 131936 rows of data\n",
      "pressure_max: should be between 966.80 and 986.00. Removed 987 outliers (0.7%).\n",
      "pressure_min: should be between 966.05 and 985.65. Removed 21 outliers (0.0%).\n",
      "air_temperature: should be between 11.95 and 38.75. Removed 1023 outliers (0.8%).\n",
      "air_temp_max: should be between 11.85 and 40.25. Removed 9 outliers (0.0%).\n",
      "air_temp_min: should be between 12.05 and 37.25. Removed 417 outliers (0.3%).\n",
      "dp_temp_max: should be between 6.35 and 29.95. Removed 970 outliers (0.7%).\n",
      "dp_temp_min: should be between 5.10 and 29.10. Removed 288 outliers (0.2%).\n",
      "wind_speed: should be between -2.10 and 4.30. Removed 1065 outliers (0.8%).\n",
      "Removed 4780 outliers overall (3.6%).\n",
      "\n",
      "processing station 32 MONTE ALEGRE DE GOIAS with 115065 rows of data\n",
      "pressure_max: should be between 944.20 and 957.80. Removed 853 outliers (0.7%).\n",
      "pressure_min: should be between 943.45 and 957.45. Removed 66 outliers (0.1%).\n",
      "air_temperature: should be between 12.80 and 38.40. Removed 40 outliers (0.0%).\n",
      "air_temp_max: should be between 12.80 and 40.00. Removed 5 outliers (0.0%).\n",
      "air_temp_min: should be between 12.95 and 36.55. Removed 177 outliers (0.2%).\n",
      "dp_temp_max: should be between -0.40 and 33.20. Removed 41 outliers (0.0%).\n",
      "dp_temp_min: should be between -1.85 and 32.15. Removed 18 outliers (0.0%).\n",
      "wind_speed: should be between -2.75 and 5.65. Removed 352 outliers (0.3%).\n",
      "Removed 1552 outliers overall (1.3%).\n",
      "\n",
      "processing station 33 TANGARA DA SERRA with 130275 rows of data\n",
      "pressure_max: should be between 955.15 and 969.95. Removed 1815 outliers (1.4%).\n",
      "pressure_min: should be between 954.15 and 969.75. Removed 12 outliers (0.0%).\n",
      "air_temperature: should be between 13.45 and 36.25. Removed 1281 outliers (1.0%).\n",
      "air_temp_max: should be between 12.95 and 38.15. Removed 5 outliers (0.0%).\n",
      "air_temp_min: should be between 13.95 and 34.35. Removed 1269 outliers (1.0%).\n",
      "dp_temp_max: should be between 8.30 and 29.90. Removed 1567 outliers (1.2%).\n",
      "dp_temp_min: should be between 7.00 and 29.40. Removed 333 outliers (0.3%).\n",
      "wind_speed: should be between -1.85 and 6.55. Removed 1771 outliers (1.4%).\n",
      "Removed 8053 outliers overall (6.2%).\n",
      "\n",
      "processing station 34 CAIAPONIA with 115980 rows of data\n",
      "pressure_max: should be between 924.05 and 938.05. Removed 883 outliers (0.8%).\n",
      "pressure_min: should be between 923.30 and 937.70. Removed 78 outliers (0.1%).\n",
      "air_temperature: should be between 11.10 and 36.70. Removed 249 outliers (0.2%).\n",
      "air_temp_max: should be between 10.60 and 38.60. Removed 1 outliers (0.0%).\n",
      "air_temp_min: should be between 11.60 and 34.80. Removed 277 outliers (0.2%).\n",
      "dp_temp_max: should be between 0.55 and 31.35. Removed 269 outliers (0.2%).\n",
      "dp_temp_min: should be between -1.10 and 30.90. Removed 60 outliers (0.1%).\n",
      "wind_speed: should be between -2.20 and 5.00. Removed 1142 outliers (1.0%).\n",
      "Removed 2959 outliers overall (2.6%).\n",
      "\n",
      "processing station 35 PRIMAVERA DO LESTE with 23889 rows of data\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "pressure_max: should be between 930.80 and 944.40. Removed 173 outliers (0.7%).\n",
      "pressure_min: should be between 930.05 and 944.05. Removed 14 outliers (0.1%).\n",
      "air_temperature: should be between 13.15 and 35.15. Removed 291 outliers (1.2%).\n",
      "air_temp_max: should be between 12.50 and 37.30. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 13.80 and 33.00. Removed 248 outliers (1.0%).\n",
      "dp_temp_max: should be between 13.04 and 21.62. Removed 2669 outliers (11.2%).\n",
      "dp_temp_min: should be between 11.97 and 20.34. Removed 435 outliers (1.8%).\n",
      "wind_speed: should be between -1.80 and 7.00. Removed 233 outliers (1.0%).\n",
      "Removed 4063 outliers overall (17.0%).\n",
      "\n",
      "processing station 36 POSSE with 116336 rows of data\n",
      "pressure_max: should be between 915.10 and 928.70. Removed 661 outliers (0.6%).\n",
      "pressure_min: should be between 914.60 and 928.20. Removed 112 outliers (0.1%).\n",
      "air_temperature: should be between 12.80 and 36.00. Removed 97 outliers (0.1%).\n",
      "air_temp_max: should be between 12.85 and 37.25. Removed 5 outliers (0.0%).\n",
      "air_temp_min: should be between 13.00 and 34.60. Removed 158 outliers (0.1%).\n",
      "dp_temp_max: should be between -0.70 and 30.50. Removed 77 outliers (0.1%).\n",
      "dp_temp_min: should be between -1.66 and 29.64. Removed 47 outliers (0.0%).\n",
      "wind_speed: should be between -1.40 and 5.80. Removed 1303 outliers (1.1%).\n",
      "Removed 2460 outliers overall (2.1%).\n",
      "\n",
      "processing station 37 SETE QUEDAS with 104088 rows of data\n",
      "pressure_max: should be between 956.40 and 978.80. Removed 689 outliers (0.7%).\n",
      "pressure_min: should be between 955.80 and 978.20. Removed 125 outliers (0.1%).\n",
      "air_temperature: should be between 8.40 and 36.40. Removed 1153 outliers (1.1%).\n",
      "air_temp_max: should be between 8.30 and 37.90. Removed 4 outliers (0.0%).\n",
      "air_temp_min: should be between 8.20 and 35.40. Removed 185 outliers (0.2%).\n",
      "dp_temp_max: should be between 3.25 and 30.85. Removed 440 outliers (0.4%).\n",
      "dp_temp_min: should be between 2.00 and 30.00. Removed 163 outliers (0.2%).\n",
      "wind_speed: should be between -3.20 and 7.20. Removed 689 outliers (0.7%).\n",
      "Removed 3448 outliers overall (3.3%).\n",
      "\n",
      "processing station 38 BATAGUASSU with 61523 rows of data\n",
      "pressure_max: should be between 959.00 and 979.00. Removed 320 outliers (0.5%).\n",
      "pressure_min: should be between 958.25 and 978.65. Removed 22 outliers (0.0%).\n",
      "air_temperature: should be between 11.25 and 37.25. Removed 569 outliers (0.9%).\n",
      "air_temp_max: should be between 11.30 and 38.50. Removed 6 outliers (0.0%).\n",
      "air_temp_min: should be between 11.35 and 35.75. Removed 206 outliers (0.3%).\n",
      "dp_temp_max: should be between 4.70 and 30.30. Removed 368 outliers (0.6%).\n",
      "dp_temp_min: should be between 3.75 and 28.95. Removed 247 outliers (0.4%).\n",
      "wind_speed: should be between -0.50 and 6.70. Removed 1074 outliers (1.7%).\n",
      "Removed 2812 outliers overall (4.6%).\n",
      "\n",
      "processing station 39 S.J. DO RIO CLARO with 129847 rows of data\n",
      "pressure_max: should be between 966.85 and 980.85. Removed 1268 outliers (1.0%).\n",
      "pressure_min: should be between 965.85 and 980.65. Removed 30 outliers (0.0%).\n",
      "air_temperature: should be between 11.55 and 39.15. Removed 202 outliers (0.2%).\n",
      "air_temp_max: should be between 10.90 and 41.30. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 12.35 and 36.75. Removed 640 outliers (0.5%).\n",
      "dp_temp_max: should be between 12.05 and 28.45. Removed 4763 outliers (3.7%).\n",
      "dp_temp_min: should be between 9.90 and 28.30. Removed 732 outliers (0.6%).\n",
      "wind_speed: should be between -2.55 and 4.25. Removed 896 outliers (0.7%).\n",
      "Removed 8531 outliers overall (6.6%).\n",
      "\n",
      "processing station 40 PORTO ESTRELA with 90738 rows of data\n",
      "pressure_max: should be between 986.40 and 1003.20. Removed 2040 outliers (2.2%).\n",
      "pressure_min: should be between 985.80 and 1002.60. Removed 162 outliers (0.2%).\n",
      "air_temperature: should be between 13.40 and 39.00. Removed 540 outliers (0.6%).\n",
      "air_temp_max: should be between 12.75 and 41.15. Removed 2 outliers (0.0%).\n",
      "air_temp_min: should be between 14.05 and 36.85. Removed 914 outliers (1.0%).\n",
      "dp_temp_max: should be between 10.75 and 31.15. Removed 1140 outliers (1.3%).\n",
      "dp_temp_min: should be between 9.10 and 30.70. Removed 187 outliers (0.2%).\n",
      "wind_speed: should be between -2.40 and 4.00. Removed 905 outliers (1.0%).\n",
      "Removed 5890 outliers overall (6.5%).\n",
      "\n",
      "processing station 41 RONDONOPOLIS with 139534 rows of data\n",
      "pressure_max: should be between 971.40 and 987.40. Removed 1639 outliers (1.2%).\n",
      "pressure_min: should be between 970.40 and 987.20. Removed 31 outliers (0.0%).\n",
      "air_temperature: should be between 12.50 and 38.90. Removed 725 outliers (0.5%).\n",
      "air_temp_max: should be between 12.10 and 40.90. Removed 19 outliers (0.0%).\n",
      "air_temp_min: should be between 13.25 and 36.85. Removed 1270 outliers (0.9%).\n",
      "dp_temp_max: should be between 7.85 and 30.65. Removed 2639 outliers (1.9%).\n",
      "dp_temp_min: should be between 6.20 and 30.20. Removed 374 outliers (0.3%).\n",
      "wind_speed: should be between -1.45 and 4.55. Removed 3103 outliers (2.2%).\n",
      "Removed 9800 outliers overall (7.0%).\n",
      "\n",
      "processing station 42 IPORA with 63822 rows of data\n",
      "pressure_max: should be between 938.15 and 952.15. Removed 516 outliers (0.8%).\n",
      "pressure_min: should be between 937.40 and 951.80. Removed 19 outliers (0.0%).\n",
      "air_temperature: should be between 11.65 and 37.65. Removed 111 outliers (0.2%).\n",
      "air_temp_max: should be between 11.25 and 39.65. Removed 2 outliers (0.0%).\n",
      "air_temp_min: should be between 11.90 and 35.90. Removed 159 outliers (0.2%).\n",
      "dp_temp_max: should be between 2.80 and 31.60. Removed 297 outliers (0.5%).\n",
      "dp_temp_min: should be between 1.20 and 30.80. Removed 42 outliers (0.1%).\n",
      "wind_speed: should be between -0.85 and 3.55. Removed 2846 outliers (4.5%).\n",
      "Removed 3992 outliers overall (6.3%).\n",
      "\n",
      "processing station 43 CACERES with 69966 rows of data\n",
      "pressure_max: should be between 989.70 and 1006.50. Removed 1722 outliers (2.5%).\n",
      "pressure_min: should be between 988.95 and 1006.15. Removed 51 outliers (0.1%).\n",
      "air_temperature: should be between 14.60 and 38.60. Removed 678 outliers (1.0%).\n",
      "air_temp_max: should be between 14.50 and 40.10. Removed 9 outliers (0.0%).\n",
      "air_temp_min: should be between 15.10 and 36.70. Removed 609 outliers (0.9%).\n",
      "dp_temp_max: should be between 11.40 and 30.60. Removed 1180 outliers (1.7%).\n",
      "dp_temp_min: should be between 9.65 and 30.05. Removed 199 outliers (0.3%).\n",
      "wind_speed: should be between -0.70 and 3.30. Removed 776 outliers (1.1%).\n",
      "Removed 5224 outliers overall (7.5%).\n",
      "\n",
      "processing station 44 AMAMBAI with 102345 rows of data\n",
      "pressure_max: should be between 953.15 and 974.35. Removed 832 outliers (0.8%).\n",
      "pressure_min: should be between 952.40 and 974.00. Removed 26 outliers (0.0%).\n",
      "air_temperature: should be between 6.35 and 37.95. Removed 758 outliers (0.7%).\n",
      "air_temp_max: should be between 6.35 and 39.55. Removed 2 outliers (0.0%).\n",
      "air_temp_min: should be between 6.35 and 36.35. Removed 258 outliers (0.3%).\n",
      "dp_temp_max: should be between 2.90 and 30.90. Removed 531 outliers (0.5%).\n",
      "dp_temp_min: should be between 1.15 and 30.35. Removed 199 outliers (0.2%).\n",
      "wind_speed: should be between -3.20 and 6.40. Removed 1008 outliers (1.0%).\n",
      "Removed 3614 outliers overall (3.5%).\n",
      "\n",
      "processing station 45 SAPEZAL with 27313 rows of data\n",
      "pressure_max: should be between 944.15 and 956.55. Removed 390 outliers (1.4%).\n",
      "pressure_min: should be between 943.15 and 956.35. Removed 2 outliers (0.0%).\n",
      "air_temperature: should be between 13.35 and 35.35. Removed 238 outliers (0.9%).\n",
      "air_temp_max: should be between 12.85 and 37.25. Removed 2 outliers (0.0%).\n",
      "air_temp_min: should be between 14.25 and 33.05. Removed 326 outliers (1.2%).\n",
      "dp_temp_max: should be between 17.42 and 24.27. Removed 3038 outliers (11.1%).\n",
      "dp_temp_min: should be between 16.08 and 23.63. Removed 132 outliers (0.5%).\n",
      "wind_speed: should be between -1.55 and 6.85. Removed 351 outliers (1.3%).\n",
      "Removed 4479 outliers overall (16.4%).\n",
      "\n",
      "processing station 46 BRAZLANDIA with 32436 rows of data\n",
      "pressure_max: should be between 882.40 and 895.20. Removed 193 outliers (0.6%).\n",
      "pressure_min: should be between 881.90 and 894.70. Removed 34 outliers (0.1%).\n",
      "air_temperature: should be between 11.25 and 32.45. Removed 128 outliers (0.4%).\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "air_temp_max: should be between 11.00 and 34.20. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 11.60 and 30.80. Removed 118 outliers (0.4%).\n",
      "dp_temp_max: should be between -0.25 and 28.95. Removed 32 outliers (0.1%).\n",
      "dp_temp_min: should be between -1.65 and 28.35. Removed 25 outliers (0.1%).\n",
      "wind_speed: should be between -1.55 and 6.05. Removed 228 outliers (0.7%).\n",
      "Removed 758 outliers overall (2.3%).\n",
      "\n",
      "processing station 47 ITIQUIRA with 90801 rows of data\n",
      "pressure_max: should be between 898.20 and 991.00. Removed 816 outliers (0.9%).\n",
      "pressure_min: should be between 902.60 and 977.80. Removed 3097 outliers (3.4%).\n",
      "air_temperature: should be between 12.75 and 36.35. Removed 880 outliers (1.0%).\n",
      "air_temp_max: should be between 12.35 and 38.35. Removed 2 outliers (0.0%).\n",
      "air_temp_min: should be between 13.40 and 34.20. Removed 756 outliers (0.8%).\n",
      "dp_temp_max: should be between 7.00 and 30.20. Removed 2116 outliers (2.3%).\n",
      "dp_temp_min: should be between 5.75 and 29.35. Removed 220 outliers (0.2%).\n",
      "wind_speed: should be between -3.95 and 6.85. Removed 798 outliers (0.9%).\n",
      "Removed 8685 outliers overall (9.6%).\n",
      "\n",
      "processing station 48 CUIABA with 142283 rows of data\n",
      "pressure_max: should be between 972.40 and 1006.00. Removed 5184 outliers (3.6%).\n",
      "pressure_min: should be between 971.90 and 1004.70. Removed 223 outliers (0.2%).\n",
      "air_temperature: should be between 14.80 and 38.80. Removed 1441 outliers (1.0%).\n",
      "air_temp_max: should be between 14.30 and 40.70. Removed 4 outliers (0.0%).\n",
      "air_temp_min: should be between 15.15 and 37.15. Removed 916 outliers (0.6%).\n",
      "dp_temp_max: should be between 10.15 and 29.75. Removed 4394 outliers (3.1%).\n",
      "dp_temp_min: should be between 8.75 and 29.15. Removed 519 outliers (0.4%).\n",
      "wind_speed: should be between -0.55 and 3.85. Removed 3262 outliers (2.3%).\n",
      "Removed 15943 outliers overall (11.2%).\n",
      "\n",
      "processing station 49 ALTO TAQUARI with 110694 rows of data\n",
      "pressure_max: should be between 909.70 and 924.90. Removed 867 outliers (0.8%).\n",
      "pressure_min: should be between 909.20 and 924.40. Removed 133 outliers (0.1%).\n",
      "air_temperature: should be between 10.75 and 34.35. Removed 761 outliers (0.7%).\n",
      "air_temp_max: should be between 10.35 and 36.35. Removed 4 outliers (0.0%).\n",
      "air_temp_min: should be between 11.55 and 31.95. Removed 1187 outliers (1.1%).\n",
      "dp_temp_max: should be between 3.95 and 29.15. Removed 852 outliers (0.8%).\n",
      "dp_temp_min: should be between 2.20 and 28.60. Removed 208 outliers (0.2%).\n",
      "wind_speed: should be between -4.80 and 8.00. Removed 232 outliers (0.2%).\n",
      "Removed 4244 outliers overall (3.8%).\n",
      "\n",
      "processing station 50 PORTO MURTINHO with 111487 rows of data\n",
      "pressure_max: should be between 990.20 and 1014.20. Removed 2497 outliers (2.2%).\n",
      "pressure_min: should be between 989.45 and 1013.85. Removed 42 outliers (0.0%).\n",
      "air_temperature: should be between 9.10 and 41.90. Removed 409 outliers (0.4%).\n",
      "air_temp_max: should be between 9.15 and 43.15. Removed 12 outliers (0.0%).\n",
      "air_temp_min: should be between 9.05 and 40.65. Removed 108 outliers (0.1%).\n",
      "dp_temp_max: should be between 4.70 and 32.70. Removed 736 outliers (0.7%).\n",
      "dp_temp_min: should be between 3.30 and 32.10. Removed 162 outliers (0.1%).\n",
      "wind_speed: should be between -2.50 and 5.50. Removed 971 outliers (0.9%).\n",
      "Removed 4937 outliers overall (4.4%).\n",
      "\n",
      "processing station 51 GOIAS with 117883 rows of data\n",
      "pressure_max: should be between 948.25 and 961.45. Removed 1229 outliers (1.0%).\n",
      "pressure_min: should be between 947.50 and 961.10. Removed 159 outliers (0.1%).\n",
      "air_temperature: should be between 12.25 and 39.05. Removed 114 outliers (0.1%).\n",
      "air_temp_max: should be between 11.85 and 41.05. Removed 7 outliers (0.0%).\n",
      "air_temp_min: should be between 12.50 and 37.30. Removed 112 outliers (0.1%).\n",
      "dp_temp_max: should be between 2.70 and 32.30. Removed 463 outliers (0.4%).\n",
      "dp_temp_min: should be between 1.10 and 31.50. Removed 137 outliers (0.1%).\n",
      "wind_speed: should be between -2.70 and 4.50. Removed 2027 outliers (1.7%).\n",
      "Removed 4248 outliers overall (3.6%).\n",
      "\n",
      "processing station 52 ALTA FLORESTA with 94457 rows of data\n",
      "pressure_max: should be between 972.60 and 984.60. Removed 646 outliers (0.7%).\n",
      "pressure_min: should be between 971.75 and 984.15. Removed 97 outliers (0.1%).\n",
      "air_temperature: should be between 13.15 and 37.55. Removed 31 outliers (0.0%).\n",
      "air_temp_max: should be between 12.10 and 40.10. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 14.20 and 35.00. Removed 547 outliers (0.6%).\n",
      "dp_temp_max: should be between 16.50 and 26.90. Removed 6867 outliers (7.3%).\n",
      "dp_temp_min: should be between 15.25 and 26.05. Removed 1439 outliers (1.5%).\n",
      "wind_speed: should be between -0.95 and 3.45. Removed 2839 outliers (3.0%).\n",
      "Removed 12466 outliers overall (13.2%).\n",
      "\n",
      "processing station 53 LUZIANIA with 122379 rows of data\n",
      "pressure_max: should be between 896.95 and 910.95. Removed 820 outliers (0.7%).\n",
      "pressure_min: should be between 896.45 and 910.45. Removed 141 outliers (0.1%).\n",
      "air_temperature: should be between 11.20 and 33.60. Removed 343 outliers (0.3%).\n",
      "air_temp_max: should be between 10.70 and 35.50. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 11.30 and 32.10. Removed 262 outliers (0.2%).\n",
      "dp_temp_max: should be between 1.25 and 28.85. Removed 377 outliers (0.3%).\n",
      "dp_temp_min: should be between -0.15 and 28.25. Removed 101 outliers (0.1%).\n",
      "wind_speed: should be between -1.00 and 6.20. Removed 1669 outliers (1.4%).\n",
      "Removed 3713 outliers overall (3.0%).\n",
      "\n",
      "processing station 54 CATALAO with 112116 rows of data\n",
      "pressure_max: should be between 907.55 and 922.35. Removed 1006 outliers (0.9%).\n",
      "pressure_min: should be between 906.65 and 922.25. Removed 13 outliers (0.0%).\n",
      "air_temperature: should be between 11.80 and 34.20. Removed 460 outliers (0.4%).\n",
      "air_temp_max: should be between 11.60 and 35.60. Removed 10 outliers (0.0%).\n",
      "air_temp_min: should be between 12.00 and 32.80. Removed 257 outliers (0.2%).\n",
      "dp_temp_max: should be between 0.80 and 29.60. Removed 218 outliers (0.2%).\n",
      "dp_temp_min: should be between -0.20 and 28.60. Removed 144 outliers (0.1%).\n",
      "wind_speed: should be between -1.30 and 5.10. Removed 1290 outliers (1.2%).\n",
      "Removed 3398 outliers overall (3.0%).\n",
      "\n",
      "processing station 55 MINEIROS with 107518 rows of data\n",
      "pressure_max: should be between 915.20 and 947.20. Removed 3252 outliers (3.0%).\n",
      "pressure_min: should be between 914.45 and 946.85. Removed 177 outliers (0.2%).\n",
      "air_temperature: should be between 9.70 and 36.90. Removed 813 outliers (0.8%).\n",
      "air_temp_max: should be between 9.30 and 38.90. Removed 75 outliers (0.1%).\n",
      "air_temp_min: should be between 10.25 and 34.65. Removed 993 outliers (0.9%).\n",
      "dp_temp_max: should be between 5.55 and 29.15. Removed 1365 outliers (1.3%).\n",
      "dp_temp_min: should be between 3.65 and 28.85. Removed 150 outliers (0.1%).\n",
      "wind_speed: should be between -3.30 and 5.50. Removed 766 outliers (0.7%).\n",
      "Removed 7591 outliers overall (7.1%).\n",
      "\n",
      "processing station 56 CHAPADAO DO SUL with 106629 rows of data\n",
      "pressure_max: should be between 915.45 and 930.25. Removed 745 outliers (0.7%).\n",
      "pressure_min: should be between 914.55 and 930.15. Removed 18 outliers (0.0%).\n",
      "air_temperature: should be between 11.15 and 34.75. Removed 1009 outliers (0.9%).\n",
      "air_temp_max: should be between 10.90 and 36.50. Removed 4 outliers (0.0%).\n",
      "air_temp_min: should be between 11.65 and 32.85. Removed 791 outliers (0.7%).\n",
      "dp_temp_max: should be between 3.00 and 29.40. Removed 955 outliers (0.9%).\n",
      "dp_temp_min: should be between 1.50 and 28.70. Removed 180 outliers (0.2%).\n",
      "wind_speed: should be between -2.40 and 6.40. Removed 1148 outliers (1.1%).\n",
      "Removed 4850 outliers overall (4.5%).\n",
      "\n",
      "processing station 57 GOIANESIA with 109014 rows of data\n",
      "pressure_max: should be between 931.80 and 945.40. Removed 949 outliers (0.9%).\n",
      "pressure_min: should be between 931.05 and 945.05. Removed 50 outliers (0.0%).\n",
      "air_temperature: should be between 12.35 and 36.75. Removed 128 outliers (0.1%).\n",
      "air_temp_max: should be between 12.35 and 38.35. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 12.75 and 34.75. Removed 388 outliers (0.4%).\n",
      "dp_temp_max: should be between 1.95 and 30.35. Removed 313 outliers (0.3%).\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dp_temp_min: should be between 0.10 and 29.70. Removed 101 outliers (0.1%).\n",
      "wind_speed: should be between -2.45 and 4.35. Removed 1986 outliers (1.8%).\n",
      "Removed 3915 outliers overall (3.6%).\n",
      "\n",
      "processing station 58 RIO BRILHANTE with 111849 rows of data\n",
      "pressure_max: should be between 965.40 and 986.20. Removed 1034 outliers (0.9%).\n",
      "pressure_min: should be between 964.65 and 985.85. Removed 25 outliers (0.0%).\n",
      "air_temperature: should be between 8.30 and 38.70. Removed 965 outliers (0.9%).\n",
      "air_temp_max: should be between 8.15 and 40.55. Removed 2 outliers (0.0%).\n",
      "air_temp_min: should be between 8.30 and 37.10. Removed 441 outliers (0.4%).\n",
      "dp_temp_max: should be between 7.45 and 29.45. Removed 1407 outliers (1.3%).\n",
      "dp_temp_min: should be between 5.95 and 28.75. Removed 357 outliers (0.3%).\n",
      "wind_speed: should be between -3.30 and 5.50. Removed 2079 outliers (1.9%).\n",
      "Removed 6310 outliers overall (5.6%).\n",
      "\n",
      "processing station 59 ALTO ARAGUAIA with 76858 rows of data\n",
      "pressure_max: should be between 922.20 and 936.60. Removed 418 outliers (0.5%).\n",
      "pressure_min: should be between 921.30 and 936.50. Removed 7 outliers (0.0%).\n",
      "air_temperature: should be between 10.95 and 36.15. Removed 404 outliers (0.5%).\n",
      "air_temp_max: should be between 10.30 and 38.30. Removed 1 outliers (0.0%).\n",
      "air_temp_min: should be between 11.20 and 34.40. Removed 277 outliers (0.4%).\n",
      "dp_temp_max: should be between 3.75 and 29.75. Removed 737 outliers (1.0%).\n",
      "dp_temp_min: should be between 2.00 and 29.20. Removed 96 outliers (0.1%).\n",
      "wind_speed: should be between -0.10 and 4.70. Removed 2606 outliers (3.4%).\n",
      "Removed 4546 outliers overall (5.9%).\n",
      "\n",
      "processing station 60 GAMA (PONTE ALTA) with 56577 rows of data\n",
      "pressure_max: should be between 899.05 and 911.45. Removed 775 outliers (1.4%).\n",
      "pressure_min: should be between 898.15 and 911.35. Removed 5 outliers (0.0%).\n",
      "air_temperature: should be between 10.15 and 34.55. Removed 156 outliers (0.3%).\n",
      "air_temp_max: should be between 10.15 and 36.15. Removed 1 outliers (0.0%).\n",
      "air_temp_min: should be between 10.15 and 32.95. Removed 209 outliers (0.4%).\n",
      "dp_temp_max: should be between 2.30 and 27.90. Removed 199 outliers (0.4%).\n",
      "dp_temp_min: should be between -0.05 and 27.55. Removed 41 outliers (0.1%).\n",
      "wind_speed: should be between -2.30 and 6.50. Removed 940 outliers (1.7%).\n",
      "Removed 2326 outliers overall (4.1%).\n",
      "\n",
      "processing station 61 SONORA with 68089 rows of data\n",
      "pressure_max: should be between 949.60 and 964.80. Removed 736 outliers (1.1%).\n",
      "pressure_min: should be between 949.00 and 964.20. Removed 154 outliers (0.2%).\n",
      "air_temperature: should be between 12.25 and 37.45. Removed 499 outliers (0.7%).\n",
      "air_temp_max: should be between 12.25 and 39.05. Removed 10 outliers (0.0%).\n",
      "air_temp_min: should be between 12.90 and 35.30. Removed 588 outliers (0.9%).\n",
      "dp_temp_max: should be between 5.40 and 31.00. Removed 757 outliers (1.1%).\n",
      "dp_temp_min: should be between 3.90 and 30.30. Removed 185 outliers (0.3%).\n",
      "wind_speed: should be between -0.50 and 6.70. Removed 1797 outliers (2.6%).\n",
      "Removed 4726 outliers overall (6.9%).\n",
      "\n",
      "processing station 62 AGUAS EMENDADAS with 108893 rows of data\n",
      "pressure_max: should be between 894.15 and 907.35. Removed 819 outliers (0.8%).\n",
      "pressure_min: should be between 893.65 and 906.85. Removed 121 outliers (0.1%).\n",
      "air_temperature: should be between 8.15 and 34.95. Removed 143 outliers (0.1%).\n",
      "air_temp_max: should be between 7.75 and 36.95. Removed 3 outliers (0.0%).\n",
      "air_temp_min: should be between 8.40 and 33.20. Removed 207 outliers (0.2%).\n",
      "dp_temp_max: should be between 2.45 and 27.65. Removed 463 outliers (0.4%).\n",
      "dp_temp_min: should be between 0.95 and 26.95. Removed 232 outliers (0.2%).\n",
      "wind_speed: should be between -1.45 and 5.35. Removed 403 outliers (0.4%).\n",
      "Removed 2391 outliers overall (2.2%).\n",
      "\n",
      "processing station 63 BANDEIRANTES with 6738 rows of data\n",
      "pressure_max: should be between 934.45 and 953.25. Removed 16 outliers (0.2%).\n",
      "pressure_min: should be between 933.95 and 952.75. Removed 1 outliers (0.0%).\n",
      "air_temperature: should be between 7.80 and 37.40. Removed 42 outliers (0.6%).\n",
      "air_temp_max: should be between 7.75 and 39.35. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 8.00 and 35.20. Removed 19 outliers (0.3%).\n",
      "dp_temp_max: should be between 1.15 and 31.95. Removed 12 outliers (0.2%).\n",
      "dp_temp_min: should be between -0.15 and 31.45. Removed 14 outliers (0.2%).\n",
      "wind_speed: should be between -1.20 and 7.60. Removed 91 outliers (1.4%).\n",
      "Removed 195 outliers overall (2.9%).\n",
      "\n",
      "processing station 64 CASSILANDIA with 105525 rows of data\n",
      "pressure_max: should be between 948.90 and 966.50. Removed 563 outliers (0.5%).\n",
      "pressure_min: should be between 948.15 and 966.15. Removed 35 outliers (0.0%).\n",
      "air_temperature: should be between 10.30 and 39.10. Removed 545 outliers (0.5%).\n",
      "air_temp_max: should be between 10.05 and 40.85. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 10.45 and 37.25. Removed 268 outliers (0.3%).\n",
      "dp_temp_max: should be between 2.85 and 31.25. Removed 366 outliers (0.3%).\n",
      "dp_temp_min: should be between 1.35 and 30.55. Removed 98 outliers (0.1%).\n",
      "wind_speed: should be between -2.70 and 4.50. Removed 1275 outliers (1.2%).\n",
      "Removed 3150 outliers overall (3.0%).\n",
      "\n",
      "processing station 65 NOVA MARINGA with 106278 rows of data\n",
      "pressure_max: should be between 967.00 and 980.60. Removed 1101 outliers (1.0%).\n",
      "pressure_min: should be between 966.25 and 980.25. Removed 65 outliers (0.1%).\n",
      "air_temperature: should be between 12.30 and 39.50. Removed 332 outliers (0.3%).\n",
      "air_temp_max: should be between 11.40 and 41.80. Removed 28 outliers (0.0%).\n",
      "air_temp_min: should be between 12.95 and 37.35. Removed 463 outliers (0.4%).\n",
      "dp_temp_max: should be between 11.40 and 29.00. Removed 5627 outliers (5.3%).\n",
      "dp_temp_min: should be between 10.00 and 28.40. Removed 648 outliers (0.6%).\n",
      "wind_speed: should be between -2.55 and 4.25. Removed 1604 outliers (1.5%).\n",
      "Removed 9868 outliers overall (9.3%).\n",
      "\n",
      "processing station 66 SELVIRIA with 17584 rows of data\n",
      "pressure_max: should be between 962.70 and 980.30. Removed 222 outliers (1.3%).\n",
      "pressure_min: should be between 961.80 and 980.20. Removed 2 outliers (0.0%).\n",
      "air_temperature: should be between 12.90 and 37.70. Removed 161 outliers (0.9%).\n",
      "air_temp_max: should be between 12.65 and 39.45. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 13.00 and 36.20. Removed 60 outliers (0.3%).\n",
      "dp_temp_max: should be between 6.65 and 31.85. Removed 78 outliers (0.4%).\n",
      "dp_temp_min: should be between 4.90 and 31.30. Removed 18 outliers (0.1%).\n",
      "wind_speed: should be between 0.05 and 4.45. Removed 1117 outliers (6.4%).\n",
      "Removed 1658 outliers overall (9.4%).\n",
      "\n",
      "processing station 67 CRISTALINA (FAZENDA SANTA MONICA) with 28225 rows of data\n",
      "pressure_max: should be between 902.35 and 916.35. Removed 201 outliers (0.7%).\n",
      "pressure_min: should be between 901.45 and 916.25. Removed 1 outliers (0.0%).\n",
      "air_temperature: should be between 9.45 and 34.65. Removed 126 outliers (0.4%).\n",
      "air_temp_max: should be between 9.20 and 36.40. Removed 1 outliers (0.0%).\n",
      "air_temp_min: should be between 9.45 and 33.05. Removed 122 outliers (0.4%).\n",
      "dp_temp_max: should be between 3.05 and 29.05. Removed 59 outliers (0.2%).\n",
      "dp_temp_min: should be between 1.15 and 28.75. Removed 11 outliers (0.0%).\n",
      "wind_speed: should be between -2.75 and 6.45. Removed 297 outliers (1.1%).\n",
      "Removed 818 outliers overall (2.9%).\n",
      "\n",
      "processing station 68 JUARA with 103858 rows of data\n",
      "pressure_max: should be between 973.60 and 988.80. Removed 5604 outliers (5.4%).\n",
      "pressure_min: should be between 972.75 and 988.35. Removed 76 outliers (0.1%).\n",
      "air_temperature: should be between 13.75 and 38.15. Removed 110 outliers (0.1%).\n",
      "air_temp_max: should be between 13.10 and 40.30. Removed 1 outliers (0.0%).\n",
      "air_temp_min: should be between 14.65 and 35.85. Removed 502 outliers (0.5%).\n",
      "dp_temp_max: should be between 14.15 and 28.15. Removed 5698 outliers (5.5%).\n",
      "dp_temp_min: should be between 12.65 and 27.45. Removed 573 outliers (0.6%).\n",
      "wind_speed: should be between -2.70 and 4.50. Removed 572 outliers (0.6%).\n",
      "Removed 13136 outliers overall (12.6%).\n",
      "\n",
      "processing station 69 MIRANDA with 106691 rows of data\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "pressure_max: should be between 986.15 and 1006.55. Removed 2087 outliers (2.0%).\n",
      "pressure_min: should be between 985.30 and 1006.10. Removed 457 outliers (0.4%).\n",
      "air_temperature: should be between 11.15 and 39.55. Removed 550 outliers (0.5%).\n",
      "air_temp_max: should be between 10.90 and 41.30. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 11.00 and 38.20. Removed 169 outliers (0.2%).\n",
      "dp_temp_max: should be between 7.20 and 32.00. Removed 1132 outliers (1.1%).\n",
      "dp_temp_min: should be between 5.20 and 31.60. Removed 164 outliers (0.2%).\n",
      "wind_speed: should be between -2.40 and 4.00. Removed 1295 outliers (1.2%).\n",
      "Removed 5854 outliers overall (5.5%).\n",
      "\n",
      "processing station 70 CAMPO NOVO DOS PARECIS with 125056 rows of data\n",
      "pressure_max: should be between 945.45 and 958.65. Removed 1477 outliers (1.2%).\n",
      "pressure_min: should be between 944.45 and 958.45. Removed 27 outliers (0.0%).\n",
      "air_temperature: should be between 12.70 and 36.70. Removed 511 outliers (0.4%).\n",
      "air_temp_max: should be between 11.80 and 39.00. Removed 2 outliers (0.0%).\n",
      "air_temp_min: should be between 13.20 and 34.80. Removed 605 outliers (0.5%).\n",
      "dp_temp_max: should be between 12.80 and 26.40. Removed 9226 outliers (7.4%).\n",
      "dp_temp_min: should be between 11.40 and 25.80. Removed 804 outliers (0.6%).\n",
      "wind_speed: should be between -2.45 and 5.15. Removed 572 outliers (0.5%).\n",
      "Removed 13224 outliers overall (10.6%).\n",
      "\n",
      "processing station 71 ITAPACI with 117207 rows of data\n",
      "pressure_max: should be between 944.55 and 957.75. Removed 1016 outliers (0.9%).\n",
      "pressure_min: should be between 943.80 and 957.40. Removed 76 outliers (0.1%).\n",
      "air_temperature: should be between 10.75 and 38.35. Removed 108 outliers (0.1%).\n",
      "air_temp_max: should be between 10.25 and 40.25. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 11.40 and 36.20. Removed 470 outliers (0.4%).\n",
      "dp_temp_max: should be between 7.10 and 29.50. Removed 612 outliers (0.5%).\n",
      "dp_temp_min: should be between 4.45 and 29.65. Removed 494 outliers (0.4%).\n",
      "wind_speed: should be between -2.40 and 4.00. Removed 2624 outliers (2.2%).\n",
      "Removed 5400 outliers overall (4.6%).\n",
      "\n",
      "processing station 72 MORRINHOS with 144884 rows of data\n",
      "pressure_max: should be between 922.55 and 938.15. Removed 743 outliers (0.5%).\n",
      "pressure_min: should be between 922.05 and 937.65. Removed 159 outliers (0.1%).\n",
      "air_temperature: should be between 9.25 and 36.85. Removed 429 outliers (0.3%).\n",
      "air_temp_max: should be between 9.40 and 38.20. Removed 11 outliers (0.0%).\n",
      "air_temp_min: should be between 9.25 and 35.25. Removed 313 outliers (0.2%).\n",
      "dp_temp_max: should be between 1.90 and 29.90. Removed 300 outliers (0.2%).\n",
      "dp_temp_min: should be between 0.30 and 29.10. Removed 122 outliers (0.1%).\n",
      "wind_speed: should be between -3.55 and 6.45. Removed 769 outliers (0.5%).\n",
      "Removed 2846 outliers overall (2.0%).\n",
      "\n",
      "processing station 73 PONTES E LACERDA with 105992 rows of data\n",
      "pressure_max: should be between 972.25 and 988.65. Removed 2079 outliers (2.0%).\n",
      "pressure_min: should be between 971.50 and 988.30. Removed 36 outliers (0.0%).\n",
      "air_temperature: should be between 13.35 and 37.75. Removed 556 outliers (0.5%).\n",
      "air_temp_max: should be between 12.80 and 40.00. Removed 1 outliers (0.0%).\n",
      "air_temp_min: should be between 14.25 and 35.45. Removed 1192 outliers (1.1%).\n",
      "dp_temp_max: should be between 10.95 and 28.95. Removed 2708 outliers (2.6%).\n",
      "dp_temp_min: should be between 9.30 and 28.50. Removed 318 outliers (0.3%).\n",
      "wind_speed: should be between -2.60 and 4.60. Removed 491 outliers (0.5%).\n",
      "Removed 7381 outliers overall (7.0%).\n",
      "\n",
      "processing station 74 ARAL MOREIRA with 24370 rows of data\n",
      "pressure_max: should be between 935.50 and 953.90. Removed 256 outliers (1.1%).\n",
      "pressure_min: should be between 935.00 and 953.40. Removed 11 outliers (0.0%).\n",
      "air_temperature: should be between 8.55 and 36.95. Removed 234 outliers (1.0%).\n",
      "air_temp_max: should be between 8.60 and 38.20. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 8.75 and 35.55. Removed 101 outliers (0.4%).\n",
      "dp_temp_max: should be between 6.55 and 26.15. Removed 405 outliers (1.7%).\n",
      "dp_temp_min: should be between 5.50 and 25.50. Removed 96 outliers (0.4%).\n",
      "wind_speed: should be between -0.70 and 6.50. Removed 336 outliers (1.4%).\n",
      "Removed 1439 outliers overall (5.9%).\n",
      "\n",
      "processing station 75 BRASILANDIA with 10987 rows of data\n",
      "pressure_max: should be between 964.90 and 984.10. Removed 47 outliers (0.4%).\n",
      "pressure_min: should be between 964.15 and 983.75. Removed 2 outliers (0.0%).\n",
      "air_temperature: should be between 9.35 and 38.55. Removed 54 outliers (0.5%).\n",
      "air_temp_max: should be between 9.30 and 40.50. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 9.25 and 36.85. Removed 21 outliers (0.2%).\n",
      "dp_temp_max: should be between 4.35 and 31.95. Removed 20 outliers (0.2%).\n",
      "dp_temp_min: should be between 3.25 and 30.85. Removed 7 outliers (0.1%).\n",
      "wind_speed: should be between 2.50 and 2.50. Removed 3718 outliers (33.8%).\n",
      "Removed 3869 outliers overall (35.2%).\n",
      "\n",
      "processing station 76 SILVANIA with 91709 rows of data\n",
      "pressure_max: should be between 902.15 and 916.15. Removed 750 outliers (0.8%).\n",
      "pressure_min: should be between 901.40 and 915.80. Removed 30 outliers (0.0%).\n",
      "air_temperature: should be between 10.05 and 34.45. Removed 297 outliers (0.3%).\n",
      "air_temp_max: should be between 9.55 and 36.35. Removed 4 outliers (0.0%).\n",
      "air_temp_min: should be between 10.55 and 32.55. Removed 358 outliers (0.4%).\n",
      "dp_temp_max: should be between 2.40 and 28.80. Removed 461 outliers (0.5%).\n",
      "dp_temp_min: should be between 0.65 and 28.25. Removed 81 outliers (0.1%).\n",
      "wind_speed: should be between -1.15 and 3.25. Removed 2556 outliers (2.8%).\n",
      "Removed 4537 outliers overall (4.9%).\n",
      "\n",
      "processing station 77 ITUMBIARA with 114109 rows of data\n",
      "pressure_max: should be between 949.60 and 966.40. Removed 758 outliers (0.7%).\n",
      "pressure_min: should be between 948.85 and 966.05. Removed 55 outliers (0.0%).\n",
      "air_temperature: should be between 11.55 and 37.55. Removed 835 outliers (0.7%).\n",
      "air_temp_max: should be between 11.45 and 39.05. Removed 12 outliers (0.0%).\n",
      "air_temp_min: should be between 11.80 and 35.80. Removed 659 outliers (0.6%).\n",
      "dp_temp_max: should be between 4.55 and 30.55. Removed 556 outliers (0.5%).\n",
      "dp_temp_min: should be between 2.95 and 29.75. Removed 224 outliers (0.2%).\n",
      "wind_speed: should be between -3.60 and 6.00. Removed 1837 outliers (1.6%).\n",
      "Removed 4936 outliers overall (4.3%).\n",
      "\n",
      "processing station 78 CAMPO GRANDE with 162286 rows of data\n",
      "pressure_max: should be between 944.35 and 961.55. Removed 1658 outliers (1.0%).\n",
      "pressure_min: should be between 943.85 and 961.05. Removed 163 outliers (0.1%).\n",
      "air_temperature: should be between 11.55 and 36.75. Removed 2173 outliers (1.3%).\n",
      "air_temp_max: should be between 11.45 and 38.25. Removed 7 outliers (0.0%).\n",
      "air_temp_min: should be between 11.80 and 35.00. Removed 837 outliers (0.5%).\n",
      "dp_temp_max: should be between 3.55 and 30.35. Removed 1023 outliers (0.6%).\n",
      "dp_temp_min: should be between 2.15 and 29.75. Removed 207 outliers (0.1%).\n",
      "wind_speed: should be between -0.90 and 7.10. Removed 3438 outliers (2.1%).\n",
      "Removed 9506 outliers overall (5.9%).\n",
      "\n",
      "processing station 79 AQUIDAUANA with 102114 rows of data\n",
      "pressure_max: should be between 984.30 and 1004.30. Removed 1866 outliers (1.8%).\n",
      "pressure_min: should be between 983.55 and 1003.95. Removed 52 outliers (0.1%).\n",
      "air_temperature: should be between 10.80 and 40.40. Removed 473 outliers (0.5%).\n",
      "air_temp_max: should be between 10.65 and 42.25. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 10.80 and 38.80. Removed 164 outliers (0.2%).\n",
      "dp_temp_max: should be between 7.45 and 31.85. Removed 999 outliers (1.0%).\n",
      "dp_temp_min: should be between 5.60 and 31.20. Removed 267 outliers (0.3%).\n",
      "wind_speed: should be between -3.00 and 5.00. Removed 663 outliers (0.6%).\n",
      "Removed 4484 outliers overall (4.4%).\n",
      "\n",
      "processing station 80 NOVA ALVORADA DO SUL with 21419 rows of data\n",
      "pressure_max: should be between 959.10 and 978.30. Removed 265 outliers (1.2%).\n",
      "pressure_min: should be between 958.35 and 977.95. Removed 7 outliers (0.0%).\n",
      "air_temperature: should be between 10.10 and 40.50. Removed 200 outliers (0.9%).\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "air_temp_max: should be between 10.30 and 41.50. Removed 1 outliers (0.0%).\n",
      "air_temp_min: should be between 9.70 and 39.30. Removed 46 outliers (0.2%).\n",
      "dp_temp_max: should be between 5.45 and 32.25. Removed 67 outliers (0.3%).\n",
      "dp_temp_min: should be between 4.20 and 31.40. Removed 12 outliers (0.1%).\n",
      "wind_speed: should be between -1.05 and 4.95. Removed 579 outliers (2.7%).\n",
      "Removed 1177 outliers overall (5.5%).\n",
      "\n",
      "processing station 81 COSTA RICA with 72961 rows of data\n",
      "pressure_max: should be between 924.85 and 939.65. Removed 601 outliers (0.8%).\n",
      "pressure_min: should be between 924.10 and 939.30. Removed 30 outliers (0.0%).\n",
      "air_temperature: should be between 10.30 and 36.70. Removed 392 outliers (0.5%).\n",
      "air_temp_max: should be between 9.75 and 38.95. Removed 1 outliers (0.0%).\n",
      "air_temp_min: should be between 11.10 and 34.30. Removed 454 outliers (0.6%).\n",
      "dp_temp_max: should be between 3.10 and 30.30. Removed 592 outliers (0.8%).\n",
      "dp_temp_min: should be between 1.60 and 29.60. Removed 106 outliers (0.1%).\n",
      "wind_speed: should be between -0.80 and 7.20. Removed 795 outliers (1.1%).\n",
      "Removed 2971 outliers overall (4.1%).\n",
      "\n",
      "processing station 82 CRISTALINA with 116805 rows of data\n",
      "pressure_max: should be between 875.95 and 889.15. Removed 977 outliers (0.8%).\n",
      "pressure_min: should be between 875.30 and 888.90. Removed 40 outliers (0.0%).\n",
      "air_temperature: should be between 10.60 and 31.40. Removed 504 outliers (0.4%).\n",
      "air_temp_max: should be between 10.25 and 33.05. Removed 10 outliers (0.0%).\n",
      "air_temp_min: should be between 10.80 and 30.00. Removed 329 outliers (0.3%).\n",
      "dp_temp_max: should be between 1.40 and 27.00. Removed 566 outliers (0.5%).\n",
      "dp_temp_min: should be between 0.15 and 26.15. Removed 185 outliers (0.2%).\n",
      "wind_speed: should be between -4.40 and 7.60. Removed 242 outliers (0.2%).\n",
      "Removed 2853 outliers overall (2.4%).\n",
      "\n",
      "processing station 83 SINOP with 101013 rows of data\n",
      "pressure_max: should be between 963.80 and 976.60. Removed 755 outliers (0.7%).\n",
      "pressure_min: should be between 963.05 and 976.25. Removed 51 outliers (0.1%).\n",
      "air_temperature: should be between 12.80 and 38.40. Removed 72 outliers (0.1%).\n",
      "air_temp_max: should be between 12.30 and 40.30. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 13.85 and 35.85. Removed 402 outliers (0.4%).\n",
      "dp_temp_max: should be between 11.90 and 28.70. Removed 5620 outliers (5.6%).\n",
      "dp_temp_min: should be between 10.15 and 28.15. Removed 358 outliers (0.4%).\n",
      "wind_speed: should be between -1.50 and 4.90. Removed 1451 outliers (1.4%).\n",
      "Removed 8709 outliers overall (8.6%).\n",
      "\n",
      "processing station 84 JARDIM with 77400 rows of data\n",
      "pressure_max: should be between 976.55 and 990.55. Removed 5089 outliers (6.6%).\n",
      "pressure_min: should be between 975.80 and 990.20. Removed 187 outliers (0.2%).\n",
      "air_temperature: should be between 9.90 and 40.30. Removed 300 outliers (0.4%).\n",
      "air_temp_max: should be between 10.20 and 41.40. Removed 5 outliers (0.0%).\n",
      "air_temp_min: should be between 10.00 and 38.80. Removed 106 outliers (0.1%).\n",
      "dp_temp_max: should be between 6.55 and 30.95. Removed 422 outliers (0.5%).\n",
      "dp_temp_min: should be between 5.30 and 30.10. Removed 221 outliers (0.3%).\n",
      "wind_speed: should be between -1.55 and 4.45. Removed 2051 outliers (2.6%).\n",
      "Removed 8381 outliers overall (10.8%).\n",
      "\n",
      "processing station 85 VILA BELA DA SANTISSIMA TRINDADE with 107874 rows of data\n",
      "pressure_max: should be between 979.05 and 995.45. Removed 2301 outliers (2.1%).\n",
      "pressure_min: should be between 978.45 and 994.85. Removed 152 outliers (0.1%).\n",
      "air_temperature: should be between 13.55 and 37.95. Removed 758 outliers (0.7%).\n",
      "air_temp_max: should be between 12.75 and 40.35. Removed 1 outliers (0.0%).\n",
      "air_temp_min: should be between 14.20 and 35.80. Removed 967 outliers (0.9%).\n",
      "dp_temp_max: should be between 12.65 and 29.05. Removed 2476 outliers (2.3%).\n",
      "dp_temp_min: should be between 11.00 and 28.60. Removed 481 outliers (0.4%).\n",
      "wind_speed: should be between -2.55 and 4.25. Removed 1170 outliers (1.1%).\n",
      "Removed 8306 outliers overall (7.7%).\n",
      "\n",
      "processing station 86 CAMAPUA with 9117 rows of data\n",
      "pressure_max: should be between 944.65 and 960.25. Removed 118 outliers (1.3%).\n",
      "pressure_min: should be between 943.90 and 959.90. Removed 0 outliers (0.0%).\n",
      "air_temperature: should be between 10.70 and 37.10. Removed 109 outliers (1.2%).\n",
      "air_temp_max: should be between 10.30 and 39.10. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 10.85 and 35.25. Removed 38 outliers (0.4%).\n",
      "dp_temp_max: should be between 20.01 and 20.01. Removed 606 outliers (6.6%).\n",
      "dp_temp_min: should be between 19.05 and 19.05. Removed 0 outliers (0.0%).\n",
      "wind_speed: should be between -0.90 and 6.30. Removed 120 outliers (1.3%).\n",
      "Removed 991 outliers overall (10.9%).\n",
      "\n",
      "processing station 87 CAMPO VERDE with 113219 rows of data\n",
      "pressure_max: should be between 923.20 and 936.80. Removed 746 outliers (0.7%).\n",
      "pressure_min: should be between 922.45 and 936.45. Removed 48 outliers (0.0%).\n",
      "air_temperature: should be between 12.95 and 34.15. Removed 1105 outliers (1.0%).\n",
      "air_temp_max: should be between 12.45 and 36.05. Removed 2 outliers (0.0%).\n",
      "air_temp_min: should be between 13.45 and 32.25. Removed 964 outliers (0.9%).\n",
      "dp_temp_max: should be between 8.70 and 27.90. Removed 5532 outliers (4.9%).\n",
      "dp_temp_min: should be between 7.20 and 27.20. Removed 376 outliers (0.3%).\n",
      "wind_speed: should be between -1.40 and 6.60. Removed 1441 outliers (1.3%).\n",
      "Removed 10214 outliers overall (9.0%).\n",
      "\n",
      "processing station 88 BRASILIA with 179133 rows of data\n",
      "pressure_max: should be between 880.55 and 894.55. Removed 5030 outliers (2.8%).\n",
      "pressure_min: should be between 880.05 and 894.05. Removed 795 outliers (0.4%).\n",
      "air_temperature: should be between 10.85 and 32.05. Removed 458 outliers (0.3%).\n",
      "air_temp_max: should be between 10.65 and 33.45. Removed 8 outliers (0.0%).\n",
      "air_temp_min: should be between 11.20 and 30.40. Removed 601 outliers (0.3%).\n",
      "dp_temp_max: should be between 1.70 and 27.30. Removed 534 outliers (0.3%).\n",
      "dp_temp_min: should be between 0.55 and 26.55. Removed 198 outliers (0.1%).\n",
      "wind_speed: should be between -1.05 and 5.75. Removed 1095 outliers (0.6%).\n",
      "Removed 8719 outliers overall (4.9%).\n",
      "\n",
      "processing station 89 AGUA BOA with 107916 rows of data\n",
      "pressure_max: should be between 955.75 and 968.95. Removed 750 outliers (0.7%).\n",
      "pressure_min: should be between 955.00 and 968.60. Removed 74 outliers (0.1%).\n",
      "air_temperature: should be between 12.70 and 38.30. Removed 44 outliers (0.0%).\n",
      "air_temp_max: should be between 12.55 and 40.15. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 13.20 and 36.40. Removed 177 outliers (0.2%).\n",
      "dp_temp_max: should be between 6.55 and 30.95. Removed 1346 outliers (1.2%).\n",
      "dp_temp_min: should be between 4.05 and 30.85. Removed 87 outliers (0.1%).\n",
      "wind_speed: should be between -1.30 and 5.90. Removed 1748 outliers (1.6%).\n",
      "Removed 4226 outliers overall (3.9%).\n",
      "\n",
      "processing station 90 CORUMBA with 109455 rows of data\n",
      "pressure_max: should be between 987.85 and 1008.25. Removed 2595 outliers (2.4%).\n",
      "pressure_min: should be between 987.10 and 1007.90. Removed 19 outliers (0.0%).\n",
      "air_temperature: should be between 15.75 and 37.75. Removed 2275 outliers (2.1%).\n",
      "air_temp_max: should be between 15.70 and 38.90. Removed 59 outliers (0.1%).\n",
      "air_temp_min: should be between 15.70 and 36.50. Removed 477 outliers (0.4%).\n",
      "dp_temp_max: should be between 8.90 and 31.30. Removed 1430 outliers (1.3%).\n",
      "dp_temp_min: should be between 6.90 and 30.90. Removed 323 outliers (0.3%).\n",
      "wind_speed: should be between -2.75 and 5.65. Removed 990 outliers (0.9%).\n",
      "Removed 8168 outliers overall (7.5%).\n",
      "\n",
      "processing station 91 COMODORO with 113305 rows of data\n",
      "pressure_max: should be between 939.70 and 954.10. Removed 928 outliers (0.8%).\n",
      "pressure_min: should be between 939.10 and 953.50. Removed 170 outliers (0.2%).\n",
      "air_temperature: should be between 11.55 and 34.35. Removed 1138 outliers (1.0%).\n",
      "air_temp_max: should be between 11.00 and 36.60. Removed 3 outliers (0.0%).\n",
      "air_temp_min: should be between 12.20 and 32.20. Removed 1488 outliers (1.3%).\n",
      "dp_temp_max: should be between 13.75 and 26.15. Removed 5416 outliers (4.8%).\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dp_temp_min: should be between 12.65 and 25.05. Removed 1387 outliers (1.2%).\n",
      "wind_speed: should be between -3.30 and 5.50. Removed 553 outliers (0.5%).\n",
      "Removed 11083 outliers overall (9.8%).\n",
      "\n",
      "processing station 92 PARANAIBA with 98290 rows of data\n",
      "pressure_max: should be between 957.95 and 976.75. Removed 1175 outliers (1.2%).\n",
      "pressure_min: should be between 957.35 and 976.15. Removed 199 outliers (0.2%).\n",
      "air_temperature: should be between 11.50 and 37.90. Removed 961 outliers (1.0%).\n",
      "air_temp_max: should be between 11.40 and 39.40. Removed 3 outliers (0.0%).\n",
      "air_temp_min: should be between 11.60 and 36.40. Removed 523 outliers (0.5%).\n",
      "dp_temp_max: should be between 5.30 and 30.90. Removed 565 outliers (0.6%).\n",
      "dp_temp_min: should be between 3.40 and 30.60. Removed 139 outliers (0.1%).\n",
      "wind_speed: should be between -1.65 and 5.15. Removed 2158 outliers (2.2%).\n",
      "Removed 5723 outliers overall (5.8%).\n",
      "\n",
      "processing station 93 ITAPORA with 26391 rows of data\n",
      "pressure_max: should be between 961.85 and 980.65. Removed 451 outliers (1.7%).\n",
      "pressure_min: should be between 961.10 and 980.30. Removed 2 outliers (0.0%).\n",
      "air_temperature: should be between 10.50 and 38.50. Removed 209 outliers (0.8%).\n",
      "air_temp_max: should be between 10.50 and 40.10. Removed 3 outliers (0.0%).\n",
      "air_temp_min: should be between 10.60 and 37.00. Removed 112 outliers (0.4%).\n",
      "dp_temp_max: should be between 6.10 and 30.10. Removed 146 outliers (0.6%).\n",
      "dp_temp_min: should be between 4.85 and 29.25. Removed 38 outliers (0.1%).\n",
      "wind_speed: should be between -1.20 and 5.20. Removed 751 outliers (2.8%).\n",
      "Removed 1712 outliers overall (6.5%).\n",
      "\n",
      "processing station 94 PARAUNA with 107730 rows of data\n",
      "pressure_max: should be between 930.35 and 945.15. Removed 781 outliers (0.7%).\n",
      "pressure_min: should be between 929.60 and 944.80. Removed 38 outliers (0.0%).\n",
      "air_temperature: should be between 12.15 and 36.55. Removed 474 outliers (0.4%).\n",
      "air_temp_max: should be between 11.65 and 38.45. Removed 4 outliers (0.0%).\n",
      "air_temp_min: should be between 12.50 and 34.90. Removed 316 outliers (0.3%).\n",
      "dp_temp_max: should be between 1.60 and 31.20. Removed 304 outliers (0.3%).\n",
      "dp_temp_min: should be between 0.20 and 30.60. Removed 66 outliers (0.1%).\n",
      "wind_speed: should be between -2.45 and 4.35. Removed 1475 outliers (1.4%).\n",
      "Removed 3458 outliers overall (3.2%).\n",
      "\n",
      "processing station 95 ANGELICA with 25606 rows of data\n",
      "pressure_max: should be between 964.00 and 983.20. Removed 343 outliers (1.3%).\n",
      "pressure_min: should be between 963.25 and 982.85. Removed 4 outliers (0.0%).\n",
      "air_temperature: should be between 10.45 and 38.85. Removed 165 outliers (0.6%).\n",
      "air_temp_max: should be between 10.45 and 40.45. Removed 1 outliers (0.0%).\n",
      "air_temp_min: should be between 10.70 and 37.10. Removed 109 outliers (0.4%).\n",
      "dp_temp_max: should be between 4.90 and 31.30. Removed 76 outliers (0.3%).\n",
      "dp_temp_min: should be between 3.65 and 30.45. Removed 31 outliers (0.1%).\n",
      "wind_speed: should be between -1.25 and 5.55. Removed 412 outliers (1.6%).\n",
      "Removed 1141 outliers overall (4.5%).\n",
      "\n",
      "processing station 96 CAARAPO with 21975 rows of data\n",
      "pressure_max: should be between 952.00 and 972.00. Removed 210 outliers (1.0%).\n",
      "pressure_min: should be between 951.25 and 971.65. Removed 4 outliers (0.0%).\n",
      "air_temperature: should be between 8.85 and 38.85. Removed 178 outliers (0.8%).\n",
      "air_temp_max: should be between 9.00 and 40.20. Removed 2 outliers (0.0%).\n",
      "air_temp_min: should be between 8.85 and 37.25. Removed 66 outliers (0.3%).\n",
      "dp_temp_max: should be between 4.35 and 31.15. Removed 76 outliers (0.3%).\n",
      "dp_temp_min: should be between 3.10 and 30.30. Removed 23 outliers (0.1%).\n",
      "wind_speed: should be between -1.45 and 6.95. Removed 209 outliers (1.0%).\n",
      "Removed 768 outliers overall (3.5%).\n",
      "\n",
      "processing station 97 ARAGARCAS with 109659 rows of data\n",
      "pressure_max: should be between 967.90 and 983.10. Removed 802 outliers (0.7%).\n",
      "pressure_min: should be between 967.15 and 982.75. Removed 225 outliers (0.2%).\n",
      "air_temperature: should be between 12.65 and 40.25. Removed 135 outliers (0.1%).\n",
      "air_temp_max: should be between 12.15 and 42.15. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 13.15 and 38.35. Removed 261 outliers (0.2%).\n",
      "dp_temp_max: should be between 4.80 and 32.00. Removed 799 outliers (0.7%).\n",
      "dp_temp_min: should be between 3.15 and 31.55. Removed 170 outliers (0.2%).\n",
      "wind_speed: should be between -1.30 and 3.50. Removed 1907 outliers (1.7%).\n",
      "Removed 4299 outliers overall (3.9%).\n",
      "\n",
      "processing station 98 SAO SIMAO with 116461 rows of data\n",
      "pressure_max: should be between 949.15 and 966.35. Removed 741 outliers (0.6%).\n",
      "pressure_min: should be between 948.25 and 966.25. Removed 12 outliers (0.0%).\n",
      "air_temperature: should be between 11.75 and 37.75. Removed 652 outliers (0.6%).\n",
      "air_temp_max: should be between 11.50 and 39.50. Removed 5 outliers (0.0%).\n",
      "air_temp_min: should be between 12.15 and 35.75. Removed 670 outliers (0.6%).\n",
      "dp_temp_max: should be between 2.65 and 31.85. Removed 448 outliers (0.4%).\n",
      "dp_temp_min: should be between 1.15 and 31.15. Removed 126 outliers (0.1%).\n",
      "wind_speed: should be between -2.80 and 6.80. Removed 1047 outliers (0.9%).\n",
      "Removed 3701 outliers overall (3.2%).\n",
      "\n",
      "processing station 99 PIRES DO RIO with 116068 rows of data\n",
      "pressure_max: should be between 922.50 and 937.70. Removed 945 outliers (0.8%).\n",
      "pressure_min: should be between 921.75 and 937.35. Removed 34 outliers (0.0%).\n",
      "air_temperature: should be between 10.15 and 36.95. Removed 350 outliers (0.3%).\n",
      "air_temp_max: should be between 10.05 and 38.45. Removed 5 outliers (0.0%).\n",
      "air_temp_min: should be between 10.65 and 35.05. Removed 467 outliers (0.4%).\n",
      "dp_temp_max: should be between 2.60 and 29.80. Removed 361 outliers (0.3%).\n",
      "dp_temp_min: should be between 0.95 and 29.35. Removed 79 outliers (0.1%).\n",
      "wind_speed: should be between -2.40 and 4.00. Removed 1155 outliers (1.0%).\n",
      "Removed 3396 outliers overall (2.9%).\n",
      "\n",
      "processing station 100 GUIRATINGA with 111089 rows of data\n",
      "pressure_max: should be between 947.60 and 960.40. Removed 2173 outliers (2.0%).\n",
      "pressure_min: should be between 946.85 and 960.05. Removed 138 outliers (0.1%).\n",
      "air_temperature: should be between 13.25 and 38.45. Removed 431 outliers (0.4%).\n",
      "air_temp_max: should be between 12.75 and 40.35. Removed 3 outliers (0.0%).\n",
      "air_temp_min: should be between 13.75 and 36.55. Removed 560 outliers (0.5%).\n",
      "dp_temp_max: should be between 5.05 and 31.85. Removed 872 outliers (0.8%).\n",
      "dp_temp_min: should be between 3.90 and 31.10. Removed 267 outliers (0.2%).\n",
      "wind_speed: should be between -2.25 and 3.75. Removed 1338 outliers (1.2%).\n",
      "Removed 5782 outliers overall (5.2%).\n",
      "\n",
      "processing station 101 PORANGATU with 16328 rows of data\n",
      "pressure_max: should be between 964.20 and 977.00. Removed 67 outliers (0.4%).\n",
      "pressure_min: should be between 963.45 and 976.65. Removed 10 outliers (0.1%).\n",
      "air_temperature: should be between 13.75 and 39.75. Removed 6 outliers (0.0%).\n",
      "air_temp_max: should be between 13.25 and 41.65. Removed 2 outliers (0.0%).\n",
      "air_temp_min: should be between 14.25 and 37.85. Removed 18 outliers (0.1%).\n",
      "dp_temp_max: should be between 10.80 and 30.00. Removed 555 outliers (3.4%).\n",
      "dp_temp_min: should be between 9.15 and 29.55. Removed 43 outliers (0.3%).\n",
      "wind_speed: should be between -2.35 and 4.45. Removed 94 outliers (0.6%).\n",
      "Removed 795 outliers overall (4.9%).\n",
      "\n",
      "processing station 102 IGUATEMI with 25118 rows of data\n",
      "pressure_max: should be between 965.90 and 985.90. Removed 401 outliers (1.6%).\n",
      "pressure_min: should be between 965.15 and 985.55. Removed 14 outliers (0.1%).\n",
      "air_temperature: should be between 8.30 and 38.70. Removed 280 outliers (1.1%).\n",
      "air_temp_max: should be between 8.40 and 40.40. Removed 2 outliers (0.0%).\n",
      "air_temp_min: should be between 8.05 and 37.25. Removed 88 outliers (0.4%).\n",
      "dp_temp_max: should be between 5.00 and 30.60. Removed 53 outliers (0.2%).\n",
      "dp_temp_min: should be between 3.50 and 29.90. Removed 35 outliers (0.1%).\n",
      "wind_speed: should be between -2.25 and 7.75. Removed 350 outliers (1.4%).\n",
      "Removed 1223 outliers overall (4.9%).\n",
      "\n",
      "processing station 103 PEDRO GOMES with 17648 rows of data\n",
      "pressure_max: should be between 975.85 and 993.05. Removed 234 outliers (1.3%).\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "pressure_min: should be between 975.10 and 992.70. Removed 1 outliers (0.0%).\n",
      "air_temperature: should be between 14.40 and 33.60. Removed 850 outliers (4.8%).\n",
      "air_temp_max: should be between 15.00 and 34.20. Removed 91 outliers (0.5%).\n",
      "air_temp_min: should be between 13.70 and 32.90. Removed 78 outliers (0.4%).\n",
      "dp_temp_max: should be between 11.95 and 28.35. Removed 286 outliers (1.6%).\n",
      "dp_temp_min: should be between 10.70 and 28.30. Removed 21 outliers (0.1%).\n",
      "wind_speed: should be between -0.90 and 3.10. Removed 841 outliers (4.8%).\n",
      "Removed 2402 outliers overall (13.6%).\n",
      "\n",
      "processing station 104 FATIMA DO SUL with 8540 rows of data\n",
      "pressure_max: should be between 965.05 and 985.45. Removed 104 outliers (1.2%).\n",
      "pressure_min: should be between 964.05 and 985.25. Removed 0 outliers (0.0%).\n",
      "air_temperature: should be between 9.05 and 38.25. Removed 80 outliers (0.9%).\n",
      "air_temp_max: should be between 9.05 and 39.85. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 8.90 and 36.90. Removed 16 outliers (0.2%).\n",
      "dp_temp_max: should be between 15.98 and 15.98. Removed 2047 outliers (24.0%).\n",
      "dp_temp_min: should be between 14.89 and 14.89. Removed 0 outliers (0.0%).\n",
      "wind_speed: should be between -1.35 and 6.25. Removed 162 outliers (1.9%).\n",
      "Removed 2409 outliers overall (28.2%).\n",
      "\n",
      "processing station 105 RIBAS DO RIO PARDO with 24838 rows of data\n",
      "pressure_max: should be between 954.50 and 971.30. Removed 321 outliers (1.3%).\n",
      "pressure_min: should be between 953.75 and 970.95. Removed 14 outliers (0.1%).\n",
      "air_temperature: should be between 10.80 and 39.60. Removed 225 outliers (0.9%).\n",
      "air_temp_max: should be between 10.40 and 41.60. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 11.10 and 37.50. Removed 132 outliers (0.5%).\n",
      "dp_temp_max: should be between 6.15 and 31.35. Removed 173 outliers (0.7%).\n",
      "dp_temp_min: should be between 4.90 and 30.50. Removed 48 outliers (0.2%).\n",
      "wind_speed: should be between -0.85 and 5.15. Removed 699 outliers (2.8%).\n",
      "Removed 1612 outliers overall (6.5%).\n",
      "\n",
      "processing station 106 EDEIA with 27724 rows of data\n",
      "pressure_max: should be between 937.40 and 952.60. Removed 196 outliers (0.7%).\n",
      "pressure_min: should be between 936.65 and 952.25. Removed 8 outliers (0.0%).\n",
      "air_temperature: should be between 11.95 and 38.75. Removed 76 outliers (0.3%).\n",
      "air_temp_max: should be between 11.85 and 40.25. Removed 2 outliers (0.0%).\n",
      "air_temp_min: should be between 12.45 and 36.85. Removed 79 outliers (0.3%).\n",
      "dp_temp_max: should be between 1.45 and 33.05. Removed 74 outliers (0.3%).\n",
      "dp_temp_min: should be between 0.20 and 32.20. Removed 18 outliers (0.1%).\n",
      "wind_speed: should be between -1.85 and 5.75. Removed 441 outliers (1.6%).\n",
      "Removed 894 outliers overall (3.2%).\n",
      "\n",
      "processing station 107 LAGUNA CARAPA with 21443 rows of data\n",
      "pressure_max: should be between 946.80 and 966.00. Removed 296 outliers (1.4%).\n",
      "pressure_min: should be between 946.05 and 965.65. Removed 2 outliers (0.0%).\n",
      "air_temperature: should be between 7.80 and 39.00. Removed 71 outliers (0.3%).\n",
      "air_temp_max: should be between 7.80 and 40.60. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 7.80 and 37.40. Removed 41 outliers (0.2%).\n",
      "dp_temp_max: should be between 4.30 and 30.70. Removed 101 outliers (0.5%).\n",
      "dp_temp_min: should be between 2.95 and 29.75. Removed 25 outliers (0.1%).\n",
      "wind_speed: should be between -1.15 and 8.05. Removed 378 outliers (1.8%).\n",
      "Removed 914 outliers overall (4.3%).\n",
      "\n",
      "processing station 108 NOVA ANDRADINA with 8147 rows of data\n",
      "pressure_max: should be between 961.55 and 981.95. Removed 52 outliers (0.6%).\n",
      "pressure_min: should be between 960.95 and 981.35. Removed 11 outliers (0.1%).\n",
      "air_temperature: should be between 9.95 and 37.55. Removed 81 outliers (1.0%).\n",
      "air_temp_max: should be between 9.80 and 39.40. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 10.10 and 35.70. Removed 32 outliers (0.4%).\n",
      "dp_temp_max: should be between 3.12 and 32.53. Removed 19 outliers (0.2%).\n",
      "dp_temp_min: should be between 2.30 and 31.10. Removed 9 outliers (0.1%).\n",
      "wind_speed: should be between -0.53 and 5.28. Removed 213 outliers (2.6%).\n",
      "Removed 417 outliers overall (5.1%).\n",
      "\n",
      "processing station 109 SANTA RITA DO PARDO with 24781 rows of data\n",
      "pressure_max: should be between 961.00 and 979.40. Removed 304 outliers (1.2%).\n",
      "pressure_min: should be between 960.25 and 979.05. Removed 9 outliers (0.0%).\n",
      "air_temperature: should be between 9.05 and 39.85. Removed 221 outliers (0.9%).\n",
      "air_temp_max: should be between 9.40 and 41.40. Removed 2 outliers (0.0%).\n",
      "air_temp_min: should be between 8.95 and 38.15. Removed 75 outliers (0.3%).\n",
      "dp_temp_max: should be between 4.50 and 32.50. Removed 28 outliers (0.1%).\n",
      "dp_temp_min: should be between 3.05 and 31.45. Removed 12 outliers (0.0%).\n",
      "wind_speed: should be between -1.90 and 6.90. Removed 296 outliers (1.2%).\n",
      "Removed 947 outliers overall (3.8%).\n",
      "\n",
      "processing station 110 JUINA with 105222 rows of data\n",
      "pressure_max: should be between 963.85 and 977.05. Removed 787 outliers (0.7%).\n",
      "pressure_min: should be between 963.25 and 976.45. Removed 141 outliers (0.1%).\n",
      "air_temperature: should be between 14.30 and 35.90. Removed 317 outliers (0.3%).\n",
      "air_temp_max: should be between 13.55 and 37.95. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 15.05 and 33.85. Removed 860 outliers (0.8%).\n",
      "dp_temp_max: should be between 15.15 and 26.75. Removed 8320 outliers (7.9%).\n",
      "dp_temp_min: should be between 13.75 and 26.15. Removed 613 outliers (0.6%).\n",
      "wind_speed: should be between -2.70 and 4.50. Removed 414 outliers (0.4%).\n",
      "Removed 11452 outliers overall (10.9%).\n",
      "\n",
      "processing station 111 BONITO with 21175 rows of data\n",
      "pressure_max: should be between 965.55 and 982.75. Removed 595 outliers (2.8%).\n",
      "pressure_min: should be between 964.80 and 982.40. Removed 8 outliers (0.0%).\n",
      "air_temperature: should be between 10.15 and 40.15. Removed 116 outliers (0.5%).\n",
      "air_temp_max: should be between 9.90 and 41.90. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 10.15 and 38.55. Removed 57 outliers (0.3%).\n",
      "dp_temp_max: should be between 8.15 and 30.95. Removed 79 outliers (0.4%).\n",
      "dp_temp_min: should be between 6.40 and 30.40. Removed 40 outliers (0.2%).\n",
      "wind_speed: should be between -1.55 and 6.85. Removed 414 outliers (2.0%).\n",
      "Removed 1309 outliers overall (6.2%).\n",
      "\n",
      "processing station 112 IVINHEMA with 122572 rows of data\n",
      "pressure_max: should be between 959.10 and 979.90. Removed 803 outliers (0.7%).\n",
      "pressure_min: should be between 958.35 and 979.55. Removed 40 outliers (0.0%).\n",
      "air_temperature: should be between 10.75 and 37.55. Removed 1141 outliers (0.9%).\n",
      "air_temp_max: should be between 10.80 and 38.80. Removed 4 outliers (0.0%).\n",
      "air_temp_min: should be between 10.95 and 36.15. Removed 487 outliers (0.4%).\n",
      "dp_temp_max: should be between 4.05 and 30.85. Removed 638 outliers (0.5%).\n",
      "dp_temp_min: should be between 3.15 and 29.95. Removed 263 outliers (0.2%).\n",
      "wind_speed: should be between -1.10 and 6.10. Removed 2451 outliers (2.0%).\n",
      "Removed 5827 outliers overall (4.8%).\n",
      "\n",
      "processing station 113 ROSARIO OESTE with 5761 rows of data\n",
      "pressure_max: should be between 980.55 and 998.55. Removed 90 outliers (1.6%).\n",
      "pressure_min: should be between 979.70 and 998.10. Removed 5 outliers (0.1%).\n",
      "air_temperature: should be between 10.35 and 41.95. Removed 11 outliers (0.2%).\n",
      "air_temp_max: should be between 9.90 and 44.30. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 11.05 and 39.45. Removed 12 outliers (0.2%).\n",
      "dp_temp_max: should be between 20.09 and 20.09. Removed 209 outliers (3.6%).\n",
      "dp_temp_min: should be between 18.90 and 18.90. Removed 0 outliers (0.0%).\n",
      "wind_speed: should be between -1.30 and 3.50. Removed 53 outliers (0.9%).\n",
      "Removed 380 outliers overall (6.6%).\n",
      "\n",
      "processing station 114 SAO JOSE DO XINGU with 19077 rows of data\n",
      "pressure_max: should be between 972.25 and 983.85. Removed 110 outliers (0.6%).\n",
      "pressure_min: should be between 971.25 and 983.65. Removed 5 outliers (0.0%).\n",
      "air_temperature: should be between 14.55 and 37.35. Removed 155 outliers (0.8%).\n",
      "air_temp_max: should be between 13.80 and 39.40. Removed 0 outliers (0.0%).\n",
      "air_temp_min: should be between 15.55 and 35.15. Removed 281 outliers (1.5%).\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dp_temp_max: should be between 13.95 and 28.75. Removed 1240 outliers (6.5%).\n",
      "dp_temp_min: should be between 12.30 and 28.30. Removed 95 outliers (0.5%).\n",
      "wind_speed: should be between -0.85 and 3.55. Removed 514 outliers (2.7%).\n",
      "Removed 2400 outliers overall (12.6%).\n",
      "\n",
      "processing station 115 SERRA NOVA DOURADA with 14126 rows of data\n",
      "pressure_max: should be between 955.80 and 967.80. Removed 60 outliers (0.4%).\n",
      "pressure_min: should be between 955.05 and 967.45. Removed 7 outliers (0.0%).\n",
      "air_temperature: should be between 15.80 and 35.80. Removed 91 outliers (0.6%).\n",
      "air_temp_max: should be between 15.45 and 37.45. Removed 4 outliers (0.0%).\n",
      "air_temp_min: should be between 16.00 and 34.40. Removed 48 outliers (0.3%).\n",
      "dp_temp_max: should be between 21.65 and 21.65. Removed 64 outliers (0.5%).\n",
      "dp_temp_min: should be between 20.77 and 20.77. Removed 0 outliers (0.0%).\n",
      "wind_speed: should be between -0.90 and 6.30. Removed 270 outliers (1.9%).\n",
      "Removed 544 outliers overall (3.9%).\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# For each station, iterate through the numeric features and calculate \n",
    "# the lower and upper limits of the feature, and remove all measurements\n",
    "# that fall outside those limits (i.e. outliers)\n",
    "\n",
    "frames = []\n",
    "num_features = 14\n",
    "feature_offset = 1\n",
    "stations = dataset['station_name'].unique().tolist() # list of stations\n",
    "\n",
    "a = 1\n",
    "\n",
    "# for each station\n",
    "for i in stations:\n",
    "\n",
    "    # create dataset with data of that station only\n",
    "    df_station = dataset[dataset['station_name']== i].copy()\n",
    "    \n",
    "    print('processing station {} {}'.format(str(a), i), end='')\n",
    "    a +=1\n",
    "    \n",
    "    # start with all data rows\n",
    "    outlier_free_index = df_station.index\n",
    "    previous_OFI_size = len(outlier_free_index)\n",
    "\n",
    "    print(' with {} rows of data'.format(len(outlier_free_index)))\n",
    "    \n",
    "    # remove outliers of each feature form the outlier_free_index\n",
    "    for j in range(feature_offset, feature_offset + num_features): # iterate features\n",
    "\n",
    "        # skip outlier removal for features that don't have normal distribution)\n",
    "        if j in [1, 2, 5, 7, 12, 13]:\n",
    "            continue\n",
    "        Q1, Q3 = np.percentile(df_station.iloc[:,j], [25, 75])\n",
    "        IQR = Q3 - Q1\n",
    "        # determine outlier fences\n",
    "        minimum = Q1 - 1.5 * IQR\n",
    "        maximum = Q3 + 1.5 * IQR\n",
    "        \n",
    "        print('{}: should be between {:.2f} and {:.2f}.'.format(dataset.columns[j], minimum, maximum), end='')\n",
    "        \n",
    "        # Keep only rows that are not outliers of this feature\n",
    "        outlier_free_index = [x for x in outlier_free_index\n",
    "                              if ((df_station.loc[x,df_station.columns[j]] >= minimum) &\n",
    "                                  (df_station.loc[x,df_station.columns[j]] <= maximum))]\n",
    "        \n",
    "        original = len(df_station)\n",
    "        current = len(outlier_free_index)\n",
    "        \n",
    "        print(\" Removed {} outliers ({:.1f}%).\".format(previous_OFI_size - current, \n",
    "                                                       (previous_OFI_size - current)/original*100))\n",
    "        \n",
    "        previous_OFI_size = len(outlier_free_index)\n",
    "\n",
    "    # save frame with data from the station (without outliers)\n",
    "    frames.append(df_station.loc[outlier_free_index])\n",
    "\n",
    "    original = len(df_station)\n",
    "    current = len(outlier_free_index)\n",
    "    print(\"Removed {} outliers overall ({:.1f}%).\\n\".format(original - current, (original - current)/original*100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "2db4bf3e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# recreate the dataset from the cleaned frames\n",
    "dataset = pd.concat(frames)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "97d914ca",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2058300 outliers were removed from the dataset (0.2%)\n"
     ]
    }
   ],
   "source": [
    "print(\"{} outliers were removed from the dataset ({:.1f}%)\".format(11427120-len(dataset), 1-(len(dataset)/11427120)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "34415ea6",
   "metadata": {},
   "outputs": [],
   "source": [
    "pickle.dump(dataset, open(\"no_outliers_dataset.pkl\", \"wb\"))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}