result_analyze.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Result Analyze\n",
    "\n",
    "Analyze forecast result using sMAPE."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import numpy as np\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load Result\n",
    "\n",
    "Load result using pandas.read_csv."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>date</th>\n",
       "      <th>store_id</th>\n",
       "      <th>y_label</th>\n",
       "      <th>y_pred</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2019-09-24</td>\n",
       "      <td>0</td>\n",
       "      <td>157.2</td>\n",
       "      <td>12.363743</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2019-09-25</td>\n",
       "      <td>0</td>\n",
       "      <td>125.2</td>\n",
       "      <td>8.149283</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2019-09-26</td>\n",
       "      <td>0</td>\n",
       "      <td>154.4</td>\n",
       "      <td>17.012093</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2019-09-27</td>\n",
       "      <td>0</td>\n",
       "      <td>174.1</td>\n",
       "      <td>10.019235</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2019-09-28</td>\n",
       "      <td>0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>12.105393</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         date  store_id  y_label     y_pred\n",
       "0  2019-09-24         0    157.2  12.363743\n",
       "1  2019-09-25         0    125.2   8.149283\n",
       "2  2019-09-26         0    154.4  17.012093\n",
       "3  2019-09-27         0    174.1  10.019235\n",
       "4  2019-09-28         0     27.0  12.105393"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result = pd.read_csv('data/deepar-predict-result.csv')\n",
    "result.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Group Result\n",
    "\n",
    "Group result using pandas.groupby."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "result_group = result.groupby(['store_id'])\n",
    "tss = []\n",
    "forecasts = []\n",
    "for name, group in result_group:\n",
    "    ts = list(group['y_label'].values)\n",
    "    forecast = list(group['y_pred'].values)\n",
    "    tss.append(ts)\n",
    "    forecasts.append(forecast)\n",
    "target = np.array(tss)\n",
    "forecast = np.array(forecasts)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Calculate sMAPE\n",
    "\n",
    "Define a function called smape, and calculate sMAPE."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def smape(target, forecast):\n",
    "    r\"\"\"\n",
    "    .. math::\n",
    "        smape = mean(2 * |Y - Y_hat| / (|Y| + |Y_hat|))\n",
    "    https://www.m4.unic.ac.cy/wp-content/uploads/2018/03/M4-Competitors-Guide.pdf\n",
    "    \"\"\"\n",
    "\n",
    "    denominator = np.abs(target) + np.abs(forecast)\n",
    "    flag = denominator == 0\n",
    "\n",
    "    smape = 2 * np.mean(\n",
    "        (np.abs(target - forecast) * (1 - flag)) / (denominator + flag)\n",
    "    )\n",
    "    return smape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "smape_values = []\n",
    "for i in range(len(target)):\n",
    "    smape_value = smape(target[i], forecast[i])\n",
    "    smape_values.append(smape_value)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Show Result\n",
    "\n",
    "Show sMAPE result using bar plot."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x102650890>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "item_metrics = pd.DataFrame({'sMAPE': smape_values})\n",
    "item_metrics['sMAPE'].plot(kind='bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x110aa2ed0>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "quartiles = pd.cut(item_metrics.sMAPE, [0, .1, .2, .3, .4, .5, .6, .7, .8, .9, 1, 2])\n",
    "def get_stats(group):\n",
    "    return {'sMAPE': group.count()}\n",
    "grouped = item_metrics.sMAPE.groupby(quartiles)\n",
    "target_bucket_amount = grouped.apply(get_stats).unstack()\n",
    "target_bucket_amount.plot(kind='bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}