examples/BasicExample.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "76ec3f97",
   "metadata": {},
   "outputs": [],
   "source": [
    "from viewser import Queryset, Column\n",
    "from functools import partial\n",
    "from viewser.operations import fetch\n",
    "from sklearn.svm import SVR\n",
    "from sklearn.metrics import r2_score\n",
    "import pandas as pd\n",
    "from matplotlib import pyplot as plt\n",
    "import views_runs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "4e8f692c",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:viewser.operations:Queryset \"fatalities_paper\" is being compiled... (0 retries)\n",
      "INFO:viewser.operations:Queryset \"fatalities_paper\" is being compiled... (1 retries)\n",
      "INFO:viewser.operations:Queryset \"fatalities_paper\" is being compiled... (2 retries)\n"
     ]
    },
    {
     "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></th>\n",
       "      <th>ged_fatalities</th>\n",
       "      <th>ged_fatalities_lagged_1</th>\n",
       "      <th>ged_fatalities_lagged_2</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>month_id</th>\n",
       "      <th>country_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">1</th>\n",
       "      <th>40</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     ged_fatalities  ged_fatalities_lagged_1  \\\n",
       "month_id country_id                                            \n",
       "1        40                     0.0                      0.0   \n",
       "         41                     0.0                      0.0   \n",
       "         42                     0.0                      0.0   \n",
       "         43                     0.0                      0.0   \n",
       "         47                     0.0                      0.0   \n",
       "\n",
       "                     ged_fatalities_lagged_2  \n",
       "month_id country_id                           \n",
       "1        40                              0.0  \n",
       "         41                              0.0  \n",
       "         42                              0.0  \n",
       "         43                              0.0  \n",
       "         47                              0.0  "
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = fetch(\"fatalities_paper\")\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "ee2798a0",
   "metadata": {},
   "outputs": [],
   "source": [
    "data_partitioner =  views_runs.DataPartitioner({\"A\":{\"train\":(1,399),\"test\":(400,450)}})\n",
    "models = views_runs.StepshiftedModels(SVR(),[1,2,3,4],\"ged_sum\")\n",
    "run = views_runs.ViewsRun(data_partitioner,models)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "295f8532",
   "metadata": {},
   "outputs": [],
   "source": [
    "run.fit(\"A\",\"train\",data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "098b337b",
   "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></th>\n",
       "      <th>ged_sum</th>\n",
       "      <th>ged_sum_tlag_1</th>\n",
       "      <th>step_pred_1</th>\n",
       "      <th>step_pred_2</th>\n",
       "      <th>step_pred_3</th>\n",
       "      <th>step_pred_4</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>month_id</th>\n",
       "      <th>country_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">400</th>\n",
       "      <th>40</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.099467</td>\n",
       "      <td>0.099956</td>\n",
       "      <td>0.099887</td>\n",
       "      <td>0.100378</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.099467</td>\n",
       "      <td>0.099956</td>\n",
       "      <td>0.099887</td>\n",
       "      <td>0.100378</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.099467</td>\n",
       "      <td>0.099956</td>\n",
       "      <td>0.099887</td>\n",
       "      <td>0.100378</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.099467</td>\n",
       "      <td>0.099956</td>\n",
       "      <td>0.099887</td>\n",
       "      <td>0.100378</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.099467</td>\n",
       "      <td>0.099956</td>\n",
       "      <td>0.099887</td>\n",
       "      <td>0.100378</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">450</th>\n",
       "      <th>242</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.099467</td>\n",
       "      <td>0.099956</td>\n",
       "      <td>0.099887</td>\n",
       "      <td>0.100378</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>243</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.099467</td>\n",
       "      <td>0.099956</td>\n",
       "      <td>0.099887</td>\n",
       "      <td>0.100378</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>244</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.099467</td>\n",
       "      <td>0.099956</td>\n",
       "      <td>0.099887</td>\n",
       "      <td>0.100378</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>245</th>\n",
       "      <td>10.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>6.133580</td>\n",
       "      <td>0.099956</td>\n",
       "      <td>0.099887</td>\n",
       "      <td>0.993295</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>246</th>\n",
       "      <td>30.0</td>\n",
       "      <td>75.0</td>\n",
       "      <td>6.809693</td>\n",
       "      <td>0.099956</td>\n",
       "      <td>0.099887</td>\n",
       "      <td>2.755230</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3366 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     ged_sum  ged_sum_tlag_1  step_pred_1  step_pred_2  \\\n",
       "month_id country_id                                                      \n",
       "400      40              0.0             0.0     0.099467     0.099956   \n",
       "         41              0.0             0.0     0.099467     0.099956   \n",
       "         42              0.0             0.0     0.099467     0.099956   \n",
       "         43              0.0             0.0     0.099467     0.099956   \n",
       "         47              0.0             0.0     0.099467     0.099956   \n",
       "...                      ...             ...          ...          ...   \n",
       "450      242             0.0             0.0     0.099467     0.099956   \n",
       "         243             0.0             0.0     0.099467     0.099956   \n",
       "         244             0.0             0.0     0.099467     0.099956   \n",
       "         245            10.0            31.0     6.133580     0.099956   \n",
       "         246            30.0            75.0     6.809693     0.099956   \n",
       "\n",
       "                     step_pred_3  step_pred_4  \n",
       "month_id country_id                            \n",
       "400      40             0.099887     0.100378  \n",
       "         41             0.099887     0.100378  \n",
       "         42             0.099887     0.100378  \n",
       "         43             0.099887     0.100378  \n",
       "         47             0.099887     0.100378  \n",
       "...                          ...          ...  \n",
       "450      242            0.099887     0.100378  \n",
       "         243            0.099887     0.100378  \n",
       "         244            0.099887     0.100378  \n",
       "         245            0.099887     0.993295  \n",
       "         246            0.099887     2.755230  \n",
       "\n",
       "[3366 rows x 6 columns]"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions = run.predict(\"A\",\"test\",data)\n",
    "predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "eb9a2513",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "          name        r2\n",
      "0  step_pred_1 -0.006194\n",
      "1  step_pred_2 -0.015726\n",
      "2  step_pred_3 -0.014874\n",
      "3  step_pred_4 -0.016166\n"
     ]
    }
   ],
   "source": [
    "\n",
    "r2_score_ged = partial(r2_score,predictions[\"ged_sum\"])\n",
    "\n",
    "step_variables = [f\"step_pred_{i}\" for i in range(1,5)]\n",
    "\n",
    "score_data = pd.DataFrame({\"name\":step_variables})\n",
    "score_data[\"r2\"] = [r2_score_ged(predictions[v]) for v in step_variables]\n",
    "\n",
    "print(score_data)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "b3e745db",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fb14eae8910>]"
      ]
     },
     "execution_count": 72,
     "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": [
    "c220 = predictions.query(\"country_id == 220\")\n",
    "plt.plot(c220[\"ged_sum\"].values,color = \"blue\")\n",
    "plt.plot(c220[\"step_pred_1\"].values, color = \"red\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "27811be3",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.8.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}