{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "a0dd99b5-ce06-403c-a941-f5e0c9adbe53",
   "metadata": {},
   "source": [
    "# pandas\n",
    "\n",
    "Når vi skal jobbe med data som passer på tabellformat kan livet vårt bli svært mye enklere om vi brukere et bibliotek som heter *pandas*. Det skal da lite kode til for å lese inn filer og for å gjøre enkle plots. Dessuten passer dataformatet i pandas sammen med flere kjente maskinlæringsbiblioteker som *tensorflow* og *scikit-learn*. Dermed slipper vi å bruke tid på å manipulere data til å passe inn i spesifikke formater for hvert bibliotek vi skal bruke. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7aaf5662-863a-4ade-a75f-3e02dce6e87c",
   "metadata": {},
   "source": [
    "Når man bruker kraftige biblioteker for databehnadling, vil det alltid være mange funksjoner man ikke kan navnet på, eller ikke vet hva gjør. Det er helt greit, og helt vanglig. Det er lov til å si at man \"kan pandas\" uten å vite hva alle funksjonene gjør, på samme måte som det er lov å si man kan engelsk selv om man ike vet betydningen av *floccinaucinihilipilification*."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a4b34131-46eb-4b44-a824-38583eae6d53",
   "metadata": {},
   "source": [
    "```{admonition} Søk i dokumentasjonen\n",
    ":class: note \n",
    "\n",
    "Når vi jobber med *pandas* kommer vi til å bruke mange forskjellige funksjoner med mange forskjellige navn. Den eneste måten å huske alle disse på, er å bruke dem jevnlig. Det er det ikke så mange av oss som gjør, og dermed kan vi like godt gi opp å huske alle funksjonene. Prøv heller å gjøre det til en vane å søke i [pandas-dokumenasjonen](https://pandas.pydata.org/docs/user_guide/10min.html) for å finne en funksjon som gjør det du ønsker å få til.\n",
    "\n",
    "``` "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "70e96009-b23e-4a08-b646-c7d5494bbbcd",
   "metadata": {},
   "source": [
    "Når vi jobber med datavitenskap er det ofte hensiktsmessig å bruke *Jupyter notebook* i *JupyterLab*. Vi kan kjøre JupyterLab både lokalt på maskinen vår og på forskjellige skytjenester. Vi anbefaler å begynne på egen maskin. Etterhvert kan det hende vi vil bruke Kaggle eller Google Colab som skytjenester, fordi de gir tilgang på en fornuftig mengde regnekraft. I del 1 av kurset skal vi jobbe med prosjekter som ikke krever noe særlig regnekraft, og da fungerer det fint å kjøre lokalt. \n",
    "\n",
    "I denne økta skal vi bruke Jupyter notebook til å trene på å lese inn og visualisere data med *Pandas* og *Matplotlib*."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "78a5fa97-e46c-4729-a83d-b27b6072c9e1",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "541f091b-9267-47c5-b7c3-5eb669d383bc",
   "metadata": {},
   "source": [
    "## Lese inn data\n",
    "Når vi skal analysere datasett, er det første vi må gjøre å lese inn dataene. Data kan finnes i mange forskjellige formater: Tabeller, bilder, lister osv. I første omgang skal vi forholde oss til tabeller. Dette svarer til excel-ark og csv-filer. \n",
    "\n",
    "La oss lese inn en tabell med data fra en csv-fil (comma separated values-fil). Vi begynner med en fil som inneholder lite nok data til at vi kan sjekke manuelt at det gikk bra. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "d1653082-8dc1-4248-bc5a-d1aa94252a25",
   "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>Rangering</th>\n",
       "      <th>Navn</th>\n",
       "      <th>År</th>\n",
       "      <th>IMDb Rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>The Shawshank Redemption</td>\n",
       "      <td>1994</td>\n",
       "      <td>9.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>The Godfather</td>\n",
       "      <td>1972</td>\n",
       "      <td>9.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>The Godfather: Part II</td>\n",
       "      <td>1974</td>\n",
       "      <td>9.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>The Dark Knight</td>\n",
       "      <td>2008</td>\n",
       "      <td>9.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>12 Angry Men</td>\n",
       "      <td>1957</td>\n",
       "      <td>9.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6</td>\n",
       "      <td>Schindler's List</td>\n",
       "      <td>1993</td>\n",
       "      <td>8.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7</td>\n",
       "      <td>The Lord of the Rings: The Return of the King</td>\n",
       "      <td>2003</td>\n",
       "      <td>8.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8</td>\n",
       "      <td>Pulp Fiction</td>\n",
       "      <td>1994</td>\n",
       "      <td>8.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>9</td>\n",
       "      <td>The Good, the Bad and the Ugly</td>\n",
       "      <td>1966</td>\n",
       "      <td>8.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>10</td>\n",
       "      <td>The Lord of the Rings: The Fellowship of the Ring</td>\n",
       "      <td>2001</td>\n",
       "      <td>8.8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Rangering                                               Navn    År  \\\n",
       "0          1                           The Shawshank Redemption  1994   \n",
       "1          2                                      The Godfather  1972   \n",
       "2          3                             The Godfather: Part II  1974   \n",
       "3          4                                    The Dark Knight  2008   \n",
       "4          5                                      12 Angry Men   1957   \n",
       "5          6                                   Schindler's List  1993   \n",
       "6          7      The Lord of the Rings: The Return of the King  2003   \n",
       "7          8                                       Pulp Fiction  1994   \n",
       "8          9                     The Good, the Bad and the Ugly  1966   \n",
       "9         10  The Lord of the Rings: The Fellowship of the Ring  2001   \n",
       "\n",
       "   IMDb Rating  \n",
       "0          9.2  \n",
       "1          9.1  \n",
       "2          9.0  \n",
       "3          9.0  \n",
       "4          9.0  \n",
       "5          8.9  \n",
       "6          8.9  \n",
       "7          8.8  \n",
       "8          8.8  \n",
       "9          8.8  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df = pd.read_csv(\"./data/filmer.csv\")\n",
    "display(df)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c1ebdd4a-37cd-44f3-b6d7-bebb002d3366",
   "metadata": {},
   "source": [
    "```{admonition} Csv-filer\n",
    ":class: warning \n",
    "\n",
    "Siden komma er standard desimalskilletegn i Norge, mens punktum brukes i USA, oppstår det i blant trøbbel når vi skal lese inn norske CSV-filer. De bruker nemlig ofte semikolon i stedet for komma til å separere verdiene. Dette løser vi ved å bruke `delimiter=\";\"` som argument til `read_csv`. \n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6798d3e1-6585-467b-9a69-10c105e73a75",
   "metadata": {},
   "source": [
    "## Kaste ut data vi ikke skal bruke"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cc1b4689-3348-48c0-86dd-492f08931191",
   "metadata": {},
   "source": [
    "Ofte har en csv fil kollonner eller data vi ikke skal bruke, og da blir det ryddigere å ikke ha de med. Hvis vi f.eks vil undersøke hvilket år som hadde de beste filmene trenger vi ikke navnene eller rangeringene."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "3defc5d6-e0c3-4f7e-8301-4b5c64f07d86",
   "metadata": {},
   "outputs": [],
   "source": [
    "data = df.drop([\"Rangering\", \"Navn\"], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b799e847-24fa-444f-9303-7228d7b301ce",
   "metadata": {},
   "source": [
    "Man kan også fjerne rader med data, enten ved å fjerne spesifikke rader eller ved å filtrere på ulike måter. Hvis man f.eks tenker at superheltfilmer ikke er ekte filmer, kan man fjerne The Dark Knight fra datasettet med indeksen dens. Og hvis man mener det ikke ble laget ekte filmer før 70-tallet kan man filtrere de ut slik:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "e76e5037-0e4e-46d6-bce8-32fa1cf11d46",
   "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>År</th>\n",
       "      <th>IMDb Rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1994</td>\n",
       "      <td>9.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1972</td>\n",
       "      <td>9.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1974</td>\n",
       "      <td>9.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1993</td>\n",
       "      <td>8.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2003</td>\n",
       "      <td>8.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1994</td>\n",
       "      <td>8.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2001</td>\n",
       "      <td>8.8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     År  IMDb Rating\n",
       "0  1994          9.2\n",
       "1  1972          9.1\n",
       "2  1974          9.0\n",
       "5  1993          8.9\n",
       "6  2003          8.9\n",
       "7  1994          8.8\n",
       "9  2001          8.8"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data_filtered = data.drop([3])\n",
    "data_filtered = data_filtered[data_filtered[\"År\"] >= 1970]\n",
    "display(data_filtered)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9b311193-2394-42c6-8a7f-a4378a523767",
   "metadata": {},
   "source": [
    "## Lage en dataframe manuelt\n",
    "\n",
    "I blant trenger vi å skrive inn data manuelt. Da er det fint å vite hvordan vi kan lage en *Pandas* *DataFrame* fra grunnen. Dette kan også være nyttig om vi skal sette inn data fra eksempelvis *numpy*-arrayer i en *DataFrame*. En enkel og ryddig måte å lage en *DataFrame* er å først legge dataene i en *dictionary* og deretter be *Pandas* lage en *DataFrame* av denne dictionary'en."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "459cfae8-2db6-46ea-976d-9fc1398a7a92",
   "metadata": {},
   "outputs": [],
   "source": [
    "data = {\"innsats\" : [2, 4, 3, 8, 4, 1, 6, 6, 10, 3, 7], \n",
    "        \"talent\" : [4, 5, 3, 4, 6, 2, 3, 5, 2, 4, 5], \n",
    "        \"resultat\" : [24, 73, 25, 185, 93, 4, 55, 110, 198, 42, 155]}\n",
    "grit_data = pd.DataFrame(data)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "24b78c98-d1fc-45c3-8dce-78e82fa94e3d",
   "metadata": {},
   "source": [
    "Vi er interresert i å se på sammenhengen mellom de ulike kolonnene i datasettet. En scatter matrise plotter fordelingen av de ulike verdiene for hver av kolonnene med søylediagrammer, og hvordan disse verdieme sammenfaller med scatter plots."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "357c440c-cf67-4f12-a3ae-45eeacc980e2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "hfig = pd.plotting.scatter_matrix(grit_data, figsize=(8,6), alpha=1)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ebf3e033-83b6-4d1b-8bbc-3fbf88ba267b",
   "metadata": {},
   "source": [
    "```{admonition} Ta imot returvariabelen fra plottingen\n",
    ":class: note\n",
    "\n",
    "Noen ganger vil plottekommandoer returnere et objekt som holder på figuren vi har laget. Dette objektet kan vi senere bruke om vi vil endre på farger, fontstørrelser og så videre. Det er ofte også slik at om vi ikke eksplisitt tar imot returvariabelen fra en plottekommando, får vi en del unyttig tekst i output-vinduet (med mindre man bruker plt.show()).\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f5c10680-48cc-46f6-b808-6e57b23bfe2e",
   "metadata": {},
   "source": [
    "## Et større datasett: Presidenvalget i USA, 2020"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "ff4594d2-0ca2-415c-9601-59f7e97beac9",
   "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>state</th>\n",
       "      <th>county</th>\n",
       "      <th>candidate</th>\n",
       "      <th>party</th>\n",
       "      <th>votes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Delaware</td>\n",
       "      <td>Kent County</td>\n",
       "      <td>Joe Biden</td>\n",
       "      <td>DEM</td>\n",
       "      <td>44552</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Delaware</td>\n",
       "      <td>Kent County</td>\n",
       "      <td>Donald Trump</td>\n",
       "      <td>REP</td>\n",
       "      <td>41009</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Delaware</td>\n",
       "      <td>Kent County</td>\n",
       "      <td>Jo Jorgensen</td>\n",
       "      <td>LIB</td>\n",
       "      <td>1044</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Delaware</td>\n",
       "      <td>Kent County</td>\n",
       "      <td>Howie Hawkins</td>\n",
       "      <td>GRN</td>\n",
       "      <td>420</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Delaware</td>\n",
       "      <td>New Castle County</td>\n",
       "      <td>Joe Biden</td>\n",
       "      <td>DEM</td>\n",
       "      <td>195034</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31107</th>\n",
       "      <td>Arizona</td>\n",
       "      <td>Maricopa County</td>\n",
       "      <td>Donald Trump</td>\n",
       "      <td>REP</td>\n",
       "      <td>993691</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31108</th>\n",
       "      <td>Arizona</td>\n",
       "      <td>Maricopa County</td>\n",
       "      <td>Jo Jorgensen</td>\n",
       "      <td>LIB</td>\n",
       "      <td>31608</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31109</th>\n",
       "      <td>Arizona</td>\n",
       "      <td>Mohave County</td>\n",
       "      <td>Donald Trump</td>\n",
       "      <td>REP</td>\n",
       "      <td>78534</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31110</th>\n",
       "      <td>Arizona</td>\n",
       "      <td>Mohave County</td>\n",
       "      <td>Joe Biden</td>\n",
       "      <td>DEM</td>\n",
       "      <td>24831</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31111</th>\n",
       "      <td>Arizona</td>\n",
       "      <td>Mohave County</td>\n",
       "      <td>Jo Jorgensen</td>\n",
       "      <td>LIB</td>\n",
       "      <td>1302</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>31112 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          state             county      candidate party   votes\n",
       "0      Delaware        Kent County      Joe Biden   DEM   44552\n",
       "1      Delaware        Kent County   Donald Trump   REP   41009\n",
       "2      Delaware        Kent County   Jo Jorgensen   LIB    1044\n",
       "3      Delaware        Kent County  Howie Hawkins   GRN     420\n",
       "4      Delaware  New Castle County      Joe Biden   DEM  195034\n",
       "...         ...                ...            ...   ...     ...\n",
       "31107   Arizona    Maricopa County   Donald Trump   REP  993691\n",
       "31108   Arizona    Maricopa County   Jo Jorgensen   LIB   31608\n",
       "31109   Arizona      Mohave County   Donald Trump   REP   78534\n",
       "31110   Arizona      Mohave County      Joe Biden   DEM   24831\n",
       "31111   Arizona      Mohave County   Jo Jorgensen   LIB    1302\n",
       "\n",
       "[31112 rows x 5 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data = pd.read_csv(\"./data/president_county_candidate.csv\")\n",
    "display(data)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fe220a18-ba92-4272-a584-2f8f14f41aae",
   "metadata": {},
   "source": [
    "Når vi leser inn slik i en jupyter notebook, får vi opp en visning av deler av tabellen. Om det er en veldig liten tabell ser vi hele. I dette tilfellet har vi lastet inn en tabell med valgresultatene for presidentvalget i USA i 2020. Det er en liste med mye innhold. La oss i første omgang bare lese ut litt forskjellige data fra tabellen, for å demonstrere hvordan vi *aksesserer* deler av dataene fra tabellen."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "b7c042ad-d9d1-4129-85e4-04a04b0f6935",
   "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>state</th>\n",
       "      <th>county</th>\n",
       "      <th>candidate</th>\n",
       "      <th>party</th>\n",
       "      <th>votes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Delaware</td>\n",
       "      <td>Kent County</td>\n",
       "      <td>Joe Biden</td>\n",
       "      <td>DEM</td>\n",
       "      <td>44552</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Delaware</td>\n",
       "      <td>Kent County</td>\n",
       "      <td>Donald Trump</td>\n",
       "      <td>REP</td>\n",
       "      <td>41009</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Delaware</td>\n",
       "      <td>Kent County</td>\n",
       "      <td>Jo Jorgensen</td>\n",
       "      <td>LIB</td>\n",
       "      <td>1044</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Delaware</td>\n",
       "      <td>Kent County</td>\n",
       "      <td>Howie Hawkins</td>\n",
       "      <td>GRN</td>\n",
       "      <td>420</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Delaware</td>\n",
       "      <td>New Castle County</td>\n",
       "      <td>Joe Biden</td>\n",
       "      <td>DEM</td>\n",
       "      <td>195034</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Delaware</td>\n",
       "      <td>New Castle County</td>\n",
       "      <td>Donald Trump</td>\n",
       "      <td>REP</td>\n",
       "      <td>88364</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Delaware</td>\n",
       "      <td>New Castle County</td>\n",
       "      <td>Jo Jorgensen</td>\n",
       "      <td>LIB</td>\n",
       "      <td>2953</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Delaware</td>\n",
       "      <td>New Castle County</td>\n",
       "      <td>Howie Hawkins</td>\n",
       "      <td>GRN</td>\n",
       "      <td>1282</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Delaware</td>\n",
       "      <td>Sussex County</td>\n",
       "      <td>Donald Trump</td>\n",
       "      <td>REP</td>\n",
       "      <td>71230</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Delaware</td>\n",
       "      <td>Sussex County</td>\n",
       "      <td>Joe Biden</td>\n",
       "      <td>DEM</td>\n",
       "      <td>56682</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Delaware</td>\n",
       "      <td>Sussex County</td>\n",
       "      <td>Jo Jorgensen</td>\n",
       "      <td>LIB</td>\n",
       "      <td>1003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Delaware</td>\n",
       "      <td>Sussex County</td>\n",
       "      <td>Howie Hawkins</td>\n",
       "      <td>GRN</td>\n",
       "      <td>437</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       state             county      candidate party   votes\n",
       "0   Delaware        Kent County      Joe Biden   DEM   44552\n",
       "1   Delaware        Kent County   Donald Trump   REP   41009\n",
       "2   Delaware        Kent County   Jo Jorgensen   LIB    1044\n",
       "3   Delaware        Kent County  Howie Hawkins   GRN     420\n",
       "4   Delaware  New Castle County      Joe Biden   DEM  195034\n",
       "5   Delaware  New Castle County   Donald Trump   REP   88364\n",
       "6   Delaware  New Castle County   Jo Jorgensen   LIB    2953\n",
       "7   Delaware  New Castle County  Howie Hawkins   GRN    1282\n",
       "8   Delaware      Sussex County   Donald Trump   REP   71230\n",
       "9   Delaware      Sussex County      Joe Biden   DEM   56682\n",
       "10  Delaware      Sussex County   Jo Jorgensen   LIB    1003\n",
       "11  Delaware      Sussex County  Howie Hawkins   GRN     437"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>state</th>\n",
       "      <th>county</th>\n",
       "      <th>candidate</th>\n",
       "      <th>party</th>\n",
       "      <th>votes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Delaware</td>\n",
       "      <td>Kent County</td>\n",
       "      <td>Joe Biden</td>\n",
       "      <td>DEM</td>\n",
       "      <td>44552</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Delaware</td>\n",
       "      <td>New Castle County</td>\n",
       "      <td>Joe Biden</td>\n",
       "      <td>DEM</td>\n",
       "      <td>195034</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Delaware</td>\n",
       "      <td>Sussex County</td>\n",
       "      <td>Joe Biden</td>\n",
       "      <td>DEM</td>\n",
       "      <td>56682</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>District of Columbia</td>\n",
       "      <td>District of Columbia</td>\n",
       "      <td>Joe Biden</td>\n",
       "      <td>DEM</td>\n",
       "      <td>35485</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>District of Columbia</td>\n",
       "      <td>Ward 2</td>\n",
       "      <td>Joe Biden</td>\n",
       "      <td>DEM</td>\n",
       "      <td>26086</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31098</th>\n",
       "      <td>Arizona</td>\n",
       "      <td>Graham County</td>\n",
       "      <td>Joe Biden</td>\n",
       "      <td>DEM</td>\n",
       "      <td>4034</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31101</th>\n",
       "      <td>Arizona</td>\n",
       "      <td>Greenlee County</td>\n",
       "      <td>Joe Biden</td>\n",
       "      <td>DEM</td>\n",
       "      <td>1182</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31104</th>\n",
       "      <td>Arizona</td>\n",
       "      <td>La Paz County</td>\n",
       "      <td>Joe Biden</td>\n",
       "      <td>DEM</td>\n",
       "      <td>2236</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31106</th>\n",
       "      <td>Arizona</td>\n",
       "      <td>Maricopa County</td>\n",
       "      <td>Joe Biden</td>\n",
       "      <td>DEM</td>\n",
       "      <td>1038364</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31110</th>\n",
       "      <td>Arizona</td>\n",
       "      <td>Mohave County</td>\n",
       "      <td>Joe Biden</td>\n",
       "      <td>DEM</td>\n",
       "      <td>24831</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>4633 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                      state                county  candidate party    votes\n",
       "0                  Delaware           Kent County  Joe Biden   DEM    44552\n",
       "4                  Delaware     New Castle County  Joe Biden   DEM   195034\n",
       "9                  Delaware         Sussex County  Joe Biden   DEM    56682\n",
       "12     District of Columbia  District of Columbia  Joe Biden   DEM    35485\n",
       "19     District of Columbia                Ward 2  Joe Biden   DEM    26086\n",
       "...                     ...                   ...        ...   ...      ...\n",
       "31098               Arizona         Graham County  Joe Biden   DEM     4034\n",
       "31101               Arizona       Greenlee County  Joe Biden   DEM     1182\n",
       "31104               Arizona         La Paz County  Joe Biden   DEM     2236\n",
       "31106               Arizona       Maricopa County  Joe Biden   DEM  1038364\n",
       "31110               Arizona         Mohave County  Joe Biden   DEM    24831\n",
       "\n",
       "[4633 rows x 5 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "display(data[data[\"state\"] == \"Delaware\"])\n",
    "display(data[data[\"candidate\"] == \"Joe Biden\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0008f7ec-63a9-4978-a810-48c4be531431",
   "metadata": {},
   "source": [
    "Det finnes mange flere måter å hente ut data på. Vi kommer til å introdusere dem i det tempoet vi trenger dem. Det er for kjedelig å liste opp alle mulige måter, og for dem som trenger å lese om alle måtene for å få sove godt om natta, finnes denne informasjonen [andre steder](https://pandas.pydata.org/pandas-docs/stable/user_guide/10min.html). \n",
    "\n",
    "La oss så begynne å regne litt. Kanskje er vi interessert i å finne ut hvor mange stemmer Donald trump fikk totalt i staten Delaware. Da kan vi summere opp stemmene for de radene i tabellen som svarer til Donald Trump og Delaware. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "9eb554d5-a073-4d51-b884-70aff158ebc5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Donald Trump fikk 200603 stemmer i Delaware\n"
     ]
    }
   ],
   "source": [
    "subset = data[data[\"state\"] == \"Delaware\"]\n",
    "subset = subset[subset[\"candidate\"] == \"Donald Trump\"]\n",
    "\n",
    "trump_votes_delaware = 0\n",
    "for number in subset[\"votes\"]: \n",
    "    trump_votes_delaware += number\n",
    "    \n",
    "print(f\"Donald Trump fikk {trump_votes_delaware} stemmer i Delaware\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "fe7e29f4-ff67-4ce2-a4b3-fe244b0009b5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "200603"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[(data.state == \"Delaware\") & (data.candidate == \"Donald Trump\")][\"votes\"].sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1ed1c51f-2572-43e0-9f0d-65c901ebd28a",
   "metadata": {},
   "source": [
    "Over skjer det litt av hvert. For det første bruker vi et sammensatt logisk uttrykk for å hente ut de linjene i tabellen som inneholder *både* staten \"Delaware\" *og* kandidaten \"Donald Trump\". Når vi har hentet ut linjene som oppfyller disse kriteriene, velger vi kolonnen \"votes\", siden vi er interessert i antallet stemmer. Til slutt bruker vi `sum`-funksjonen på DataFrame-objektet. DataFrame-objektet er pandas sin vanlige måte å holde på data. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3585ac30-a5e2-4364-831f-4c560e50a310",
   "metadata": {},
   "source": [
    "## Gruppere data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "f9c4ee89-cf72-4ef6-b318-932a61a0462c",
   "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>candidate</th>\n",
       "      <th>votes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>Zachary Scalf</td>\n",
       "      <td>29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>Keith McCormic</td>\n",
       "      <td>127</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Gary Swing</td>\n",
       "      <td>143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>Jordan Scott</td>\n",
       "      <td>173</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>Joseph Kishore</td>\n",
       "      <td>192</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         candidate  votes\n",
       "37   Zachary Scalf     29\n",
       "26  Keith McCormic    127\n",
       "13      Gary Swing    143\n",
       "23    Jordan Scott    173\n",
       "24  Joseph Kishore    192"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>candidate</th>\n",
       "      <th>votes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Write-ins</td>\n",
       "      <td>155165</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Howie Hawkins</td>\n",
       "      <td>376714</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>Jo Jorgensen</td>\n",
       "      <td>1808321</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Donald Trump</td>\n",
       "      <td>72691128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>Joe Biden</td>\n",
       "      <td>78220235</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        candidate     votes\n",
       "1       Write-ins    155165\n",
       "15  Howie Hawkins    376714\n",
       "19   Jo Jorgensen   1808321\n",
       "12   Donald Trump  72691128\n",
       "20      Joe Biden  78220235"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "by_candidate_sum = data.groupby(\"candidate\", as_index=False).sum()\n",
    "by_candidate_sum = by_candidate_sum.sort_values(by=\"votes\")\n",
    "\n",
    "display(by_candidate_sum.head()) #viser kun de fem første radene\n",
    "\n",
    "display(by_candidate_sum.tail()) #viser kun de fem siste radene"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d59635ce-de40-4312-99f2-e1841ce033de",
   "metadata": {},
   "source": [
    "Her innser vi for det første at det ar veldig mange kandidater. La oss så plotte resultatene for alle disse kandidatene:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "2d99fe0f-5266-48d9-861e-6d079017f0e8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,6))\n",
    "plt.bar(by_candidate_sum[\"candidate\"], by_candidate_sum[\"votes\"])\n",
    "plt.xticks(rotation=90)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3eec9e84-2481-4a3c-aa9b-18eefb5a9961",
   "metadata": {},
   "source": [
    "## Henting av subsets og kollaps av subsets"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "df50bef0-ed8e-4437-9cf9-008e84c155c6",
   "metadata": {},
   "source": [
    "Hvis vi vil hente ut en del av dataene til en ny dataframe, som f.eks alle radene med Biden stemmer gjør vi det slik:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "abadd328-2cad-498a-83c1-0caf5d7add0a",
   "metadata": {},
   "outputs": [],
   "source": [
    "biden = data[data[\"candidate\"] == \"Joe Biden\"]\n",
    "trump = data[data[\"candidate\"] == \"Donald Trump\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "09e27489-3adc-4d79-97a3-0ac263659a47",
   "metadata": {},
   "source": [
    "Hvis vi så vil slå sammen alle delene av dataene som har samme verdi for \"state\" gjør vi det slik. Her slår vi de sammen ved å summere verdiene, men det går også ann å ta gjennomsnitt eller bare telle antall rader."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "317e4b50-9897-489e-8830-e1fa3d8443ae",
   "metadata": {},
   "outputs": [],
   "source": [
    "biden_states = biden.groupby(\"state\").sum() #.mean() og .size() er også nyttige ofte!\n",
    "trump_states = trump.groupby(\"state\").sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "199c8d20-6c02-471e-84ed-290b0b6df9fa",
   "metadata": {},
   "source": [
    "Nå har vi to dataframes som holder antall stemmer hver kandidat har i hver stat, så vi kan f.eks regne ut forskjellen og plotte den:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "1a1c636f-124b-4cb7-91cf-2d51143e0ca5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "diff_states = biden_states - trump_states\n",
    "\n",
    "plt.figure(figsize=(12,6))\n",
    "plt.bar(diff_states.index, diff_states[\"votes\"])\n",
    "plt.xticks(rotation=90)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "75515492-9718-4040-9797-893224d6183e",
   "metadata": {},
   "source": [
    "## Masker for å effektivt hente subsets av subsets"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2983ea9c-9538-497b-8b9d-9642743afb8a",
   "metadata": {},
   "source": [
    "I forrige eksempel brukte vi en maske(rekke med `True` og `False` verdier) for å plukke ut verdiene vi ville ha med i dataframen `biden`.\n",
    "\n",
    "```\n",
    "biden = data[data[\"candidate\"] == \"Joe Biden\"]\n",
    "```\n",
    "\n",
    "Masken er det inni hakeparantesen, og kan bli lagret i en egen variabel:\n",
    "\n",
    "```\n",
    "biden_mask = data[\"candidate\"] == \"Joe Biden\"\n",
    "```\n",
    "Dette blir utrolig nyttig om vi skal ha tak i mange ulike subsets av subsets av dataen. Dette eksempelet illustrerer ikke dette veldig godt, siden vi har få kollonner å jobbe med, men i datasett med egenskaper som kjønn, alder og rase blir dette helt nødvendig for å effektivt hente alle mulige kombinasjoner som hvite menn, asiatiske kvinner etc."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "5a45b2bc-95aa-4856-816a-0091a5843d74",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "db057b8c-b3ef-46d6-b70a-665c1e32001f",
   "metadata": {},
   "outputs": [],
   "source": [
    "biden_mask = data[\"candidate\"] == \"Joe Biden\"\n",
    "trump_mask = data[\"candidate\"] == \"Donald Trump\"\n",
    "other = np.logical_not(np.logical_or(biden_mask, trump_mask)) #evt. ~(biden_mask | trump_mask)\n",
    "\n",
    "even = data[\"votes\"] % 2 == 0\n",
    "more10k = data[\"votes\"] >= 10000\n",
    "more100k = data[\"votes\"] >= 100000"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "983e8315-2805-40d8-b79b-2a69fc487336",
   "metadata": {},
   "source": [
    "Vi kombinerer masker med numpy sine logiske funksjoner. Man kan også bruke python sine innebygde [bitwise operatorer](https://realpython.com/python-bitwise-operators/#overview-of-pythons-bitwise-operators).\n",
    "\n",
    "\n",
    "Hvis vi bare vil telle antall `True`, kan vi ta summen av masken. Hugs at masker også kan brukes til å hente ut en subset av dataen."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4c96c690-adc6-42cc-acba-5e8f7f0050a3",
   "metadata": {},
   "source": [
    "Trump fikk et partall antall stemmer i flere fylker:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "c83d5cbf-8978-42ca-a2ad-ba8b0f86f33a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2354\n",
      "2381\n"
     ]
    }
   ],
   "source": [
    "print(np.logical_and(biden_mask, even).sum()) #evt. (biden_mask & even).sum()\n",
    "print(np.logical_and(trump_mask, even).sum())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "665a3547-5c85-447e-8c6b-d4775829e174",
   "metadata": {},
   "source": [
    "Trump fikk mer enn 10000 stemmer i flere fylker enn Biden:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "595c6e76-c460-4f95-a8fb-73cc3f6d4744",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "932\n",
      "1347\n"
     ]
    }
   ],
   "source": [
    "print(np.logical_and(biden_mask, more10k).sum()) #evt. (biden_mask & more10k).sum()\n",
    "print(np.logical_and(trump_mask, more10k).sum())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0f3594dd-6fda-4162-a96a-0ba9fb994fbd",
   "metadata": {},
   "source": [
    "Men Biden fikk med enn 100000 stemmer i flere fylker enn Trump!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "610ed233-2c7e-41e6-9a34-e5ea2e737915",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "168\n",
      "136\n"
     ]
    }
   ],
   "source": [
    "print(np.logical_and(biden_mask, more100k).sum()) #evt. (biden_mask & more10k).sum()\n",
    "print(np.logical_and(trump_mask, more100k).sum())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1fdcb738-1c96-4365-92d9-934ddf36c0b3",
   "metadata": {},
   "source": [
    "Vi kan lage en funksjon som sammenligner hvor mange fylker en maske gjelder for for hver kandidat. Denne kan vi bruke til å finne når Biden tok igjen Trump"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "d0a51432-0b15-4d2b-87bb-0bf90fa1a82b",
   "metadata": {},
   "outputs": [],
   "source": [
    "def trumpVbiden(subset_mask=True):\n",
    "    return np.logical_and(biden_mask, subset_mask).sum() < np.logical_and(trump_mask, subset_mask).sum()\n",
    "    #evt. (biden_mask & subset_mask).sum() < (trump_mask & subset_mask).sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cd6f0c39-a475-44b5-85c8-64eeff37253d",
   "metadata": {},
   "source": [
    "Biden tar igjen Trump i antall fylker når man ser på fylker med 67114 stemmer eller flere."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "ef2a2f05-8a07-46b4-88d8-bef3e2b80eb0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "67114\n"
     ]
    }
   ],
   "source": [
    "votes = 10000\n",
    "while trumpVbiden(data[\"votes\"] >= votes):\n",
    "    votes += 1\n",
    "    if votes > 100000:\n",
    "        break\n",
    "print(votes)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a25d788f-d6da-45da-ae6a-2424628d0c10",
   "metadata": {},
   "source": [
    "For å finne hvilket fylke som var vippepunktet er det bare å lage en maske for 67113 stemmer. Trump mistet ledelsen i antall fylker når man ser på fylker med mer enn 67113 stemmer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "e2c85868-5118-4d61-8b61-90de0874fa73",
   "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>state</th>\n",
       "      <th>county</th>\n",
       "      <th>candidate</th>\n",
       "      <th>party</th>\n",
       "      <th>votes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>18799</th>\n",
       "      <td>Texas</td>\n",
       "      <td>Bell County</td>\n",
       "      <td>Donald Trump</td>\n",
       "      <td>REP</td>\n",
       "      <td>67113</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       state       county     candidate party  votes\n",
       "18799  Texas  Bell County  Donald Trump   REP  67113"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[data[\"votes\"] == votes-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "9906d1e4-5bb2-4c6d-bb3b-70191eb1a3af",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "227\n",
      "227\n"
     ]
    }
   ],
   "source": [
    "print(np.logical_and(biden_mask, data[\"votes\"] >= votes).sum()) #evt. (biden_mask & (data[\"votes\"] >= votes)).sum()\n",
    "print(np.logical_and(trump_mask, data[\"votes\"] >= votes).sum())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "83a30beb-b796-44e0-b265-d89609e19622",
   "metadata": {},
   "source": [
    "## Kombinere DataFrames (Bonus)\n",
    " \n",
    "Noen ganger ønsker vi å kombinere tabeller. \n",
    "Vi er nå klare til å gjøre litt mer omfattende oppgaver. For eksempel om vi skal legge data fra presidentvalget inn i et kart. Da må vi kombinere kartdata og stemmedata. Vi kan bruke modulen `geopandas` til å lage figurer som viser fordelingen av stemmer rundt omkring i USA."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "5d158a9f-8970-48d8-8198-6ca2d8778552",
   "metadata": {},
   "outputs": [],
   "source": [
    "import geopandas as gp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "1b5c66f9-4e6e-4208-9e88-807255e0fda7",
   "metadata": {},
   "outputs": [],
   "source": [
    "states = gp.read_file(\"https://www2.census.gov/geo/tiger/GENZ2018/shp/cb_2018_us_state_500k.zip\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d426683f-896a-4702-8052-0814de4f6758",
   "metadata": {},
   "source": [
    "Først fjerner vi alle små kandidater fra dataen. Vi må så summere stemmene hver kandidat har per stat. Så må vi vri om dataframen slik at statene blir kolonnene, hver kandidat har en rad og antall stemmer er verdiene. Så gjør vi antall stemmer om til andel av alle stemmen i den staten. Til slutt slår vi sammen kartet og dataframen med alle stemmene slik at de kan dele data om hver stat. Da kan vi plotte og få andel av stemmer inn i kartet."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "8b0b554d-2d48-44e5-8c15-bd350d8edb7b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "trumpOrBiden = data[(data[\"candidate\"] == \"Donald Trump\") | (data[\"candidate\"] == \"Joe Biden\")] #used bitwise or\n",
    "votes_by_state = trumpOrBiden.groupby([\"state\", \"candidate\"], as_index=False)[\"votes\"].sum()\n",
    "votes_pivot = votes_by_state.pivot(index=\"state\", columns=\"candidate\", values=\"votes\")\n",
    "votes_share = votes_pivot.div(votes_pivot.sum(axis=1), axis=0)\n",
    "\n",
    "plotting_df = pd.merge(states, votes_share, left_on=\"NAME\", right_on=\"state\")\n",
    "\n",
    "plotting_df.plot(figsize=(16, 6), column=\"Donald Trump\", cmap=\"bwr\", legend=True, vmin=0.4, vmax=0.6)\n",
    "plt.xlim(-180, -63)\n",
    "plt.title(\"Donal Trump\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2fd3cdab-7b79-4089-a55b-b426bea49705",
   "metadata": {},
   "source": [
    "Større bredde av verdier lar oss se forskjellene mellom de røde statene og mellom de blå statene bedre, men gjør vippestatene vanskeligere å avklare."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "d20b55ff-bb1b-49ef-864f-4af930c10474",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plotting_df.plot(figsize=(16, 6), column=\"Donald Trump\", cmap=\"bwr\", legend=True, vmin=0.2, vmax=0.8)\n",
    "plt.xlim(-180, -63)\n",
    "plt.title(\"Donal Trump\")\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}