06_Sampling_and_Balancing.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "bcf5d27a",
   "metadata": {},
   "source": [
    "# Data preparation and transformation exercise\n",
    "\n",
    "## Part VI - Sampling and balancing\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. Remove strongly correlated features\n",
    "9. Remove outliers\n",
    "10. Aggregate features\n",
    "11. Encode categorical features\n",
    "12. Feature scaling\n",
    "13. Dimensionality reduction and feature decomposition\n",
    "\n",
    "14. <a href=\"#Sample-and-balance\">Sample and balance</a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "id": "c60871ba",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import pickle\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "dataset = pickle.load(open(\"scaled_dataset.pkl\", \"rb\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "22e37613",
   "metadata": {},
   "source": [
    "## Sample and balance"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0b4a3547",
   "metadata": {},
   "source": [
    "The dataset is not balanced because very few stations were operational in the first years."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "id": "79360a47",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dataset['year'] = dataset.full_time.dt.year\n",
    "dataset.groupby('year')['station_name'].nunique().plot(kind='bar', figsize=(16,5))\n",
    "plt.xlabel('Year')\n",
    "plt.ylabel('Number of stations online')\n",
    "plt.title('Number of active stations per year');"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "61e0e1b9",
   "metadata": {},
   "source": [
    "Due to outlier removal (an stations coming online/offline), not all stations have the same number of samples in the last year, but this won't matter if we do regression per station.\n",
    "\n",
    "Another inbalance in the dataset, also due to outlier removal, is that each hour of the day does not necessarily have the same number of measurements."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "id": "5960b4aa",
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset['hour'] = dataset.full_time.dt.hour"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "id": "618b5317",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dataset.groupby('hour')['station_name'].count().plot(kind='bar', figsize=(16,5))\n",
    "plt.xlabel('Hour of the day')\n",
    "plt.ylabel('Number of measurements')\n",
    "plt.title('Accumulated number of measurements per hour of the day');"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "82e73e8e",
   "metadata": {},
   "source": [
    "Let's create a sub-dataset with only the 'SERRA NOVA DOURADA' station, and readings from the year 2020, and then balance these readings per hour."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "id": "1785e48d",
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset['month'] = dataset.full_time.dt.month"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "id": "ce9d02f5",
   "metadata": {},
   "outputs": [],
   "source": [
    "station_2020 = dataset.loc[(dataset.station_name == 'SERRA NOVA DOURADA') & (dataset.year == 2020)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "id": "94d3dd5a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Number of measurements')"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x360 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df1 = station_2020.groupby('month')['station_name'].count()\n",
    "df2 = station_2020.groupby('hour')['station_name'].count()\n",
    "\n",
    "fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(16,5))\n",
    "fig.suptitle('SERRA NOVA DOURADA weather station data (unbalanced)')\n",
    "\n",
    "ax1.bar(df1.index, df1)\n",
    "ax1.set_xlabel('Month')\n",
    "ax1.set_ylabel('Number of measurements')\n",
    "\n",
    "ax2.bar(df2.index, df2)\n",
    "ax2.set_xlabel('Hour of day')\n",
    "ax2.set_ylabel('Number of measurements')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1f72bfa5",
   "metadata": {},
   "source": [
    "Let's balance the number of measurements throughout the hours of the day.\n",
    "\n",
    "First we need to determine the minimum number of samples in any given hour of the day."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "id": "357c5c7e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "315"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "minimum_measurements = station_2020.groupby('hour')['station_name'].count().min()\n",
    "minimum_measurements"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "id": "47323840",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "At 0 hour(s) there are 357 measurements to sample from\n",
      "At 1 hour(s) there are 354 measurements to sample from\n",
      "At 2 hour(s) there are 348 measurements to sample from\n",
      "At 3 hour(s) there are 351 measurements to sample from\n",
      "At 4 hour(s) there are 362 measurements to sample from\n",
      "At 5 hour(s) there are 364 measurements to sample from\n",
      "At 6 hour(s) there are 363 measurements to sample from\n",
      "At 7 hour(s) there are 364 measurements to sample from\n",
      "At 8 hour(s) there are 362 measurements to sample from\n",
      "At 9 hour(s) there are 362 measurements to sample from\n",
      "At 10 hour(s) there are 360 measurements to sample from\n",
      "At 11 hour(s) there are 351 measurements to sample from\n",
      "At 12 hour(s) there are 345 measurements to sample from\n",
      "At 13 hour(s) there are 330 measurements to sample from\n",
      "At 14 hour(s) there are 315 measurements to sample from\n",
      "At 15 hour(s) there are 319 measurements to sample from\n",
      "At 16 hour(s) there are 324 measurements to sample from\n",
      "At 17 hour(s) there are 330 measurements to sample from\n",
      "At 18 hour(s) there are 335 measurements to sample from\n",
      "At 19 hour(s) there are 332 measurements to sample from\n",
      "At 20 hour(s) there are 341 measurements to sample from\n",
      "At 21 hour(s) there are 357 measurements to sample from\n",
      "At 22 hour(s) there are 364 measurements to sample from\n",
      "At 23 hour(s) there are 363 measurements to sample from\n"
     ]
    }
   ],
   "source": [
    "# For every hour we will get the same number of samples (minimum_measurements)\n",
    "\n",
    "frames = []\n",
    "\n",
    "# for each hourhour\n",
    "for i in range(0,24):\n",
    "\n",
    "    # create a dataset with all measurements taken at that hour\n",
    "    df_hour =  station_2020[station_2020.hour == i]\n",
    "    \n",
    "    print('At {} hour(s) there are {} measurements to sample from'.format(i, len(df_hour)))\n",
    "    \n",
    "    # create a frame with 219 samples from that dataframe\n",
    "    df_hour_balanced = df_hour.sample(n=minimum_measurements, random_state=1)\n",
    "\n",
    "    # append to array with previous frames\n",
    "    frames.append(df_hour_balanced)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "id": "63dffd1b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# create a balanced dataset from the frames\n",
    "balanced_station_2020 = pd.concat(frames)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "id": "d4218296",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Started with 8353 measurements and generated a balnced dataset of 7560 measurements\n"
     ]
    }
   ],
   "source": [
    "print('Started with {} measurements and generated a balnced dataset of {} measurements'.format(\n",
    "    len(station_2020),\n",
    "    len(balanced_station_2020)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "id": "e5a9f3e2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "balanced_station_2020.groupby('hour')['station_name'].count().plot(kind='bar', figsize=(16,5))\n",
    "plt.xlabel('Hour of the day')\n",
    "plt.ylabel('Number of measurements')\n",
    "plt.title('SERRA NOVA DOURADA weather station data (balanced)');"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b5f9d798",
   "metadata": {},
   "source": [
    "Thanks for your interest. Feedback is appreciated."
   ]
  }
 ],
 "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
}