diff --git a/lessons/intro/pandas/index.ipynb b/lessons/intro/pandas/index.ipynb
index 10451e3b7b..f489cbd3d7 100644
--- a/lessons/intro/pandas/index.ipynb
+++ b/lessons/intro/pandas/index.ipynb
@@ -5,7 +5,6 @@
    "metadata": {},
    "source": [
     "Na dnešní lekci si do virtuálního prostředí nainstalujte následující balíčky.\n",
-    "Můžete použít prostředí z lekce o NumPy.\n",
     "\n",
     "```console\n",
     "$ python -m pip install --upgrade pip\n",
@@ -15,8 +14,9 @@
     "Pro případ, že by vaše verze `pip`-u neuměla *wheels* nebo na PyPI nebyly příslušné *wheel* balíčky, je dobré mít na systému nainstalovaný překladač C a Fortranu (např. `gcc`, `gcc-gfortran`) a hlavičkové soubory Pythonu (např. `python3-devel`). Jestli je ale nemáte, zkuste instalaci přímo – *wheels* pro většinu operačních systémů existují – a až kdyby to nefungovalo, instalujte překladače a hlavičky.\n",
     "\n",
     "Mezitím co se instaluje, stáhněte si do adresáře `static` potřebné soubory:\n",
-    "[actors.csv](static/actors.csv) a\n",
-    "[spouses.csv](static/spouses.csv).\n",
+    "[actors.csv](static/actors.csv),\n",
+    "[spouses.csv](static/spouses.csv) a \n",
+    "[sales.csv](static/sales.csv).\n",
     "\n",
     "A až bude nainstalováno, spusťte si nový Notebook. (Viz [lekce o Notebooku](../notebook/).)\n",
     "\n",
@@ -69,15 +69,15 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Jak bylo řečeno u [NumPy](../numpy/), analytici – cílová skupina této knihovny – mají rádi zkratky. Ve spoustě materiálů na Webu proto najdete `import pandas as pd`, případně rovnou (a bez vysvětlení) použité `pd` jako zkratku pro `pandas`. Tento návod ale používá plné jméno."
+    "Čísla můžeme buď prozkoumávat, hrát si s nimi, zjišťovat zajímavé souvislosti; anebo můžeme připravovat programy, které nějaké výpočty provedou automaticky. Na obojí se používají podobné nástroje. Automaticky pouštěné skripty musí být samozřejmě robustní. Nástroje ke zkoumání dat ale bývají přívětivé k vědcům a datovým analytikům, často na úkor robustnosti nebo „dobrých programátorských mravů”. Například některé funkce tak trochu „hádají”, co uživatel chtěl, a v tutoriálech se setkáte se zkratkami jako `import pandas as pd` či dokonce `from pandas import *`.\n",
+    "\n",
+    "Toto je kurz programovací, kde nám záleží více na znovupoužitelném kódu než na jednom konkrétním výsledku. Budeme proto preferovat explicitní a jednoznačné operace. Ty jsou v použitých knihovnách vždy vedle zkratek k dispozici a popsány v dokumentaci."
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 1,
-   "metadata": {
-    "collapsed": true
-   },
+   "metadata": {},
    "outputs": [],
    "source": [
     "import pandas"
@@ -115,18 +115,18 @@
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -214,18 +214,18 @@
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -285,18 +285,18 @@
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -390,11 +390,13 @@
       "<class 'pandas.core.frame.DataFrame'>\n",
       "RangeIndex: 6 entries, 0 to 5\n",
       "Data columns (total 3 columns):\n",
-      "name     6 non-null object\n",
-      "birth    6 non-null int64\n",
-      "alive    6 non-null bool\n",
+      " #   Column  Non-Null Count  Dtype \n",
+      "---  ------  --------------  ----- \n",
+      " 0   name    6 non-null      object\n",
+      " 1   birth   6 non-null      int64 \n",
+      " 2   alive   6 non-null      bool  \n",
       "dtypes: bool(1), int64(1), object(1)\n",
-      "memory usage: 182.0+ bytes\n"
+      "memory usage: 230.0+ bytes\n"
      ]
     }
    ],
@@ -415,8 +417,8 @@
     "typu `bool` zabírá v paměti desítky bytů, ale v `bool` sloupci\n",
     "si každá hodnota vystačí s jedním bytem.\n",
     "\n",
-    "Na rozdíl od NumPy jsou typy dynamické: když do sloupce zapíšeme „nekompatibilní”\n",
-    "hodnotu, kterou Pandas neumí převést na daný typ, typ sloupce\n",
+    "Typy jsou v Pandas dynamické: když do sloupce zapíšeme „nekompatibilní”\n",
+    "hodnotu, kterou Pandas neumí převést na aktuální typ sloupce, typ sloupce\n",
     "se automaticky zobecní.\n",
     "Některé automatické převody ovšem nemusí být úplně intuitivní, např. `None` na `NaN`."
    ]
@@ -432,7 +434,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Sloupec, neboli `Series`, je druhý základní datový typ v Pandas. Obsahuje sérii hodnot, jako seznam, ale navíc má jméno, datový typ a „index”, který jednotlivé hodnoty pojmenovává. Sloupce se dají získat vybráním z tabulky:"
+    "Sloupec, neboli `Series`, je druhý základní datový typ v Pandas. Obsahuje sérii hodnot, podobně jako seznam. Navíc má jméno, datový typ a „index”, který jednotlivé hodnoty pojmenovává. Sloupce se dají získat vybráním z tabulky:"
    ]
   },
   {
@@ -560,12 +562,12 @@
     {
      "data": {
       "text/plain": [
-       "0    74\n",
-       "1    73\n",
-       "2    73\n",
-       "3    75\n",
-       "4    76\n",
-       "5    77\n",
+       "0    78\n",
+       "1    77\n",
+       "2    77\n",
+       "3    79\n",
+       "4    80\n",
+       "5    81\n",
        "Name: birth, dtype: int64"
       ]
      },
@@ -575,7 +577,8 @@
     }
    ],
    "source": [
-    "ages = 2016 - birth_years\n",
+    "# Vytvoření sloupce s věkem v roce 2020\n",
+    "ages = 2020 - birth_years\n",
     "ages"
    ]
   },
@@ -587,13 +590,13 @@
     {
      "data": {
       "text/plain": [
-       "0    20\n",
-       "1    20\n",
-       "2    20\n",
-       "3    20\n",
-       "4    20\n",
-       "5    20\n",
-       "Name: birth, dtype: int64"
+       "0      (Terry)\n",
+       "1    (Michael)\n",
+       "2       (Eric)\n",
+       "3     (Graham)\n",
+       "4      (Terry)\n",
+       "5       (John)\n",
+       "Name: name, dtype: object"
       ]
      },
      "execution_count": 13,
@@ -602,15 +605,16 @@
     }
    ],
    "source": [
-    "century = birth_years // 100 + 1\n",
-    "century"
+    "# Vytvoření sloupce s příjmeními v závorkách\n",
+    "parenthesized = '(' + actors['name'] + ')'\n",
+    "parenthesized"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "To platí jak pro aritmetické operace (`+`, `-`, `*`, `/`, `//`, `%`, `**`), tak pro porovnávání:"
+    "To platí jak pro aritmetické operátory (`+`, `-`, `*`, `/`, `//`, `%`, `**`), tak pro porovnávání:"
    ]
   },
   {
@@ -798,6 +802,33 @@
    "cell_type": "code",
    "execution_count": 20,
    "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0     True\n",
+       "1     True\n",
+       "2     True\n",
+       "3     True\n",
+       "4    False\n",
+       "5    False\n",
+       "Name: birth, dtype: bool"
+      ]
+     },
+     "execution_count": 20,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Sloupec typu bool s \"maskou\"\n",
+    "birth_years > 1940"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "metadata": {},
    "outputs": [
     {
      "data": {
@@ -809,7 +840,7 @@
        "Name: birth, dtype: int64"
       ]
      },
-     "execution_count": 20,
+     "execution_count": 21,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -828,7 +859,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": 22,
    "metadata": {},
    "outputs": [
     {
@@ -839,7 +870,7 @@
        "Name: birth, dtype: int64"
       ]
      },
-     "execution_count": 21,
+     "execution_count": 22,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -858,7 +889,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": 23,
    "metadata": {},
    "outputs": [
     {
@@ -869,7 +900,7 @@
       "Průměr:  1941.3333333333333\n",
       "Medián:  1941.5\n",
       "Počet unikátních hodnot:  5\n",
-      "Koeficient špičatosti:  -1.48125\n"
+      "Koeficient špičatosti:  -1.4812500000001654\n"
      ]
     }
    ],
@@ -890,7 +921,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": 24,
    "metadata": {
     "scrolled": false
    },
@@ -907,7 +938,7 @@
        "Name: name, dtype: object"
       ]
      },
-     "execution_count": 23,
+     "execution_count": 24,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -918,7 +949,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 24,
+   "execution_count": 25,
    "metadata": {},
    "outputs": [
     {
@@ -933,7 +964,7 @@
        "Name: alive, dtype: object"
       ]
      },
-     "execution_count": 24,
+     "execution_count": 25,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -960,7 +991,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
+   "execution_count": 26,
    "metadata": {},
    "outputs": [
     {
@@ -975,7 +1006,7 @@
        "Name: name, dtype: object"
       ]
      },
-     "execution_count": 25,
+     "execution_count": 26,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -986,25 +1017,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 26,
+   "execution_count": 27,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -1052,7 +1083,7 @@
        "4    Terry   1940   True"
       ]
      },
-     "execution_count": 26,
+     "execution_count": 27,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1063,25 +1094,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 27,
+   "execution_count": 28,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -1136,7 +1167,7 @@
        "5     John   True"
       ]
      },
-     "execution_count": 27,
+     "execution_count": 28,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1171,25 +1202,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 28,
+   "execution_count": 29,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -1251,7 +1282,7 @@
        "5     John   1939   True"
       ]
      },
-     "execution_count": 28,
+     "execution_count": 29,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1262,7 +1293,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 29,
+   "execution_count": 30,
    "metadata": {},
    "outputs": [
     {
@@ -1274,7 +1305,7 @@
        "Name: 2, dtype: object"
       ]
      },
-     "execution_count": 29,
+     "execution_count": 30,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1287,19 +1318,19 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Všimněte si, že `loc` není metoda: používají se s ním hranaté závorky.\n"
+    "Všimněte si, že `loc` není metoda: používají se s ním hranaté závorky, ne kulaté.\n"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Použijeme-li k indexování *n*-tici, prvním prvkem se indexují řádky a druhým sloupce – podobně jako u NumPy:"
+    "Použijeme-li k indexování *n*-tici, prvním prvkem se indexují řádky a druhým sloupce:"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 30,
+   "execution_count": 31,
    "metadata": {},
    "outputs": [
     {
@@ -1308,7 +1339,7 @@
        "1943"
       ]
      },
-     "execution_count": 30,
+     "execution_count": 31,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1326,25 +1357,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 31,
+   "execution_count": 32,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -1381,7 +1412,7 @@
        "4   1940   True"
       ]
      },
-     "execution_count": 31,
+     "execution_count": 32,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1399,7 +1430,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 32,
+   "execution_count": 33,
    "metadata": {},
    "outputs": [
     {
@@ -1411,7 +1442,7 @@
        "Name: name, dtype: object"
       ]
      },
-     "execution_count": 32,
+     "execution_count": 33,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1422,25 +1453,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 33,
+   "execution_count": 34,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -1473,7 +1504,7 @@
        "4   Terry"
       ]
      },
-     "execution_count": 33,
+     "execution_count": 34,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1491,7 +1522,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 34,
+   "execution_count": 35,
    "metadata": {},
    "outputs": [
     {
@@ -1506,7 +1537,7 @@
        "Name: alive, dtype: bool"
       ]
      },
-     "execution_count": 34,
+     "execution_count": 35,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1524,25 +1555,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 35,
+   "execution_count": 36,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -1597,7 +1628,7 @@
        "5     John   True"
       ]
      },
-     "execution_count": 35,
+     "execution_count": 36,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1608,25 +1639,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 36,
+   "execution_count": 37,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -1674,7 +1705,7 @@
        "4   Terry   1940   True"
       ]
      },
-     "execution_count": 36,
+     "execution_count": 37,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1694,30 +1725,30 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Druhý indexer, který si v krátkosti ukážeme, je `iloc`. Umí to samé co `loc`, jen nepracuje s klíčem, ale s pozicemi řádků či sloupců. Funguje tedy jako indexování v NumPy."
+    "Druhý indexer, který si v krátkosti ukážeme, je `iloc`. Umí to samé co `loc`, jen nepracuje s klíčem, ale s pozicemi řádků či sloupců."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 37,
+   "execution_count": 38,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -1779,7 +1810,7 @@
        "5     John   1939   True"
       ]
      },
-     "execution_count": 37,
+     "execution_count": 38,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1790,7 +1821,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 38,
+   "execution_count": 39,
    "metadata": {},
    "outputs": [
     {
@@ -1799,7 +1830,7 @@
        "'Terry'"
       ]
      },
-     "execution_count": 38,
+     "execution_count": 39,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1817,7 +1848,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 39,
+   "execution_count": 40,
    "metadata": {},
    "outputs": [
     {
@@ -1826,7 +1857,7 @@
        "1939"
       ]
      },
-     "execution_count": 39,
+     "execution_count": 40,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1837,7 +1868,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 40,
+   "execution_count": 41,
    "metadata": {
     "scrolled": false
    },
@@ -1846,18 +1877,18 @@
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -1905,7 +1936,7 @@
        "5     John"
       ]
      },
-     "execution_count": 40,
+     "execution_count": 41,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1923,25 +1954,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 41,
+   "execution_count": 42,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -1982,7 +2013,7 @@
        "3  False   1941  Graham"
       ]
      },
-     "execution_count": 41,
+     "execution_count": 42,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2000,7 +2031,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 42,
+   "execution_count": 43,
    "metadata": {},
    "outputs": [
     {
@@ -2009,7 +2040,7 @@
        "'John'"
       ]
      },
-     "execution_count": 42,
+     "execution_count": 43,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2035,25 +2066,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 43,
+   "execution_count": 44,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -2115,7 +2146,7 @@
        "5     John   1939   True"
       ]
      },
-     "execution_count": 43,
+     "execution_count": 44,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2134,7 +2165,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 44,
+   "execution_count": 45,
    "metadata": {},
    "outputs": [
     {
@@ -2143,7 +2174,7 @@
        "RangeIndex(start=0, stop=6, step=1)"
       ]
      },
-     "execution_count": 44,
+     "execution_count": 45,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2154,7 +2185,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 45,
+   "execution_count": 46,
    "metadata": {},
    "outputs": [
     {
@@ -2163,7 +2194,7 @@
        "Index(['name', 'birth', 'alive'], dtype='object')"
       ]
      },
-     "execution_count": 45,
+     "execution_count": 46,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2181,25 +2212,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 46,
+   "execution_count": 47,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -2268,7 +2299,7 @@
        "John        John   1939   True"
       ]
      },
-     "execution_count": 46,
+     "execution_count": 47,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2280,7 +2311,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 47,
+   "execution_count": 48,
    "metadata": {},
    "outputs": [
     {
@@ -2289,7 +2320,7 @@
        "Index(['Terry', 'Michael', 'Eric', 'Graham', 'Terry', 'John'], dtype='object', name='name')"
       ]
      },
-     "execution_count": 47,
+     "execution_count": 48,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2307,25 +2338,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 48,
+   "execution_count": 49,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -2394,7 +2425,7 @@
        "Terry      Terry   1940   True"
       ]
      },
-     "execution_count": 48,
+     "execution_count": 49,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2406,25 +2437,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 49,
+   "execution_count": 50,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -2465,7 +2496,7 @@
        "Graham  Graham   1941  False"
       ]
      },
-     "execution_count": 49,
+     "execution_count": 50,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2483,25 +2514,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 50,
+   "execution_count": 51,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -2542,7 +2573,7 @@
        "Terry  Terry   1940   True"
       ]
      },
-     "execution_count": 50,
+     "execution_count": 51,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2561,25 +2592,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 51,
+   "execution_count": 52,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -2639,7 +2670,7 @@
        "        1940    True"
       ]
      },
-     "execution_count": 51,
+     "execution_count": 52,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2658,18 +2689,22 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 52,
+   "execution_count": 53,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "MultiIndex(levels=[['Eric', 'Graham', 'John', 'Michael', 'Terry'], [1939, 1940, 1941, 1942, 1943]],\n",
-       "           labels=[[0, 1, 2, 3, 4, 4], [4, 2, 0, 4, 3, 1]],\n",
+       "MultiIndex([(   'Eric', 1943),\n",
+       "            ( 'Graham', 1941),\n",
+       "            (   'John', 1939),\n",
+       "            ('Michael', 1943),\n",
+       "            (  'Terry', 1942),\n",
+       "            (  'Terry', 1940)],\n",
        "           names=['name', 'birth'])"
       ]
      },
-     "execution_count": 52,
+     "execution_count": 53,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2687,25 +2722,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 53,
+   "execution_count": 54,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -2738,7 +2773,7 @@
        "1940    True"
       ]
      },
-     "execution_count": 53,
+     "execution_count": 54,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2749,7 +2784,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 54,
+   "execution_count": 55,
    "metadata": {},
    "outputs": [
     {
@@ -2759,7 +2794,7 @@
        "Name: 1940, dtype: bool"
       ]
      },
-     "execution_count": 54,
+     "execution_count": 55,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2770,7 +2805,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 55,
+   "execution_count": 56,
    "metadata": {},
    "outputs": [
     {
@@ -2780,7 +2815,7 @@
        "Name: (Terry, 1942), dtype: bool"
       ]
      },
-     "execution_count": 55,
+     "execution_count": 56,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2798,7 +2833,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 56,
+   "execution_count": 57,
    "metadata": {
     "scrolled": false
    },
@@ -2807,18 +2842,18 @@
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -2878,7 +2913,7 @@
        "        1940    True"
       ]
      },
-     "execution_count": 56,
+     "execution_count": 57,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2889,7 +2924,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 57,
+   "execution_count": 58,
    "metadata": {
     "scrolled": true
    },
@@ -2903,7 +2938,7 @@
        "dtype: object"
       ]
      },
-     "execution_count": 57,
+     "execution_count": 58,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2916,25 +2951,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 58,
+   "execution_count": 59,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -3002,7 +3037,7 @@
        "        1940    True   Gilliam"
       ]
      },
-     "execution_count": 58,
+     "execution_count": 59,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3028,7 +3063,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 59,
+   "execution_count": 60,
    "metadata": {
     "scrolled": true
    },
@@ -3046,7 +3081,7 @@
        "Name: last_name, dtype: object"
       ]
      },
-     "execution_count": 59,
+     "execution_count": 60,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3064,7 +3099,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 60,
+   "execution_count": 61,
    "metadata": {},
    "outputs": [
     {
@@ -3080,7 +3115,7 @@
        "Name: last_name, dtype: bool"
       ]
      },
-     "execution_count": 60,
+     "execution_count": 61,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3098,25 +3133,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 61,
+   "execution_count": 62,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -3184,7 +3219,7 @@
        "        1940    True   Gilliam"
       ]
      },
-     "execution_count": 61,
+     "execution_count": 62,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3202,25 +3237,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 62,
+   "execution_count": 63,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -3260,7 +3295,7 @@
        "      1940    True   Gilliam"
       ]
      },
-     "execution_count": 62,
+     "execution_count": 63,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3287,25 +3322,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 63,
+   "execution_count": 64,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -3367,7 +3402,7 @@
        "5     John   1939   True"
       ]
      },
-     "execution_count": 63,
+     "execution_count": 64,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3379,7 +3414,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 64,
+   "execution_count": 65,
    "metadata": {
     "scrolled": false
    },
@@ -3388,18 +3423,18 @@
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -3496,7 +3531,7 @@
        "10  Michael   1943       Helen Gibbins"
       ]
      },
-     "execution_count": 64,
+     "execution_count": 65,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3508,25 +3543,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 65,
+   "execution_count": 66,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -3635,7 +3670,7 @@
        "10     John   1939   True       Jennifer Wade"
       ]
      },
-     "execution_count": 65,
+     "execution_count": 66,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3666,26 +3701,35 @@
    "source": [
     "Dostáváme se do bodu, kdy nám jednoduchá tabulka přestává stačit. Pojďme si vytvořit tabulku větší: fiktivních prodejů v e-shopu, ve formátu jaký bychom mohli dostat z SQL databáze nebo datového souboru.\n",
     "\n",
-    "Použijeme k tomu mimo jiné `date_range`, která vytváří kalendářní intervaly. Zde, i v jiných případech, kdy je jasné, že se má nějaká hodnota interpretovat jako datum, nám Pandas dovolí místo objektů `datetime` zadávat data řetězcem:"
+    "Tabulka obsahuje kalendářní data. Výchozí chování fuunkce `read_csv` tato data automaticky detekuje (což se dá vypnout). A v dalším kódu nám v případech, kdy je jasné že se má nějaká hodnota interpretovat jako datum, Pandas dovolí místo objektů `datetime` zadávat data řetězcem."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "> [note]\n",
+    "> Mimochodem, tabulka byla získána následujícím kódem:\n",
+    "> ```\n",
+    "> import itertools\n",
+    "> import random\n",
+    "> random.seed(0)\n",
+    "> \n",
+    "> months = pandas.date_range('2015-01', '2016-12', freq='M')\n",
+    "> categories = ['Electronics', 'Power Tools', 'Clothing']\n",
+    "> data = pandas.DataFrame([{'month': a, 'category': b, 'sales': random.randint(-1000, 10000)}\n",
+    ">                           for a, b in itertools.product(months, categories)\n",
+    ">                           if random.randrange(20) > 0])\n",
+    "> ```"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 66,
-   "metadata": {
-    "collapsed": true
-   },
+   "execution_count": 67,
+   "metadata": {},
    "outputs": [],
    "source": [
-    "import itertools\n",
-    "import random\n",
-    "random.seed(0)\n",
-    "\n",
-    "months = pandas.date_range('2015-01', '2016-12', freq='M')\n",
-    "categories = ['Electronics', 'Power Tools', 'Clothing']\n",
-    "data = pandas.DataFrame([{'month': a, 'category': b, 'sales': random.randint(-1000, 10000)}\n",
-    "                          for a, b in itertools.product(months, categories)\n",
-    "                          if random.randrange(20) > 0])"
+    "data = pandas.read_csv('static/sales.csv', index_col=0)"
    ]
   },
   {
@@ -3697,64 +3741,64 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 67,
+   "execution_count": 68,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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>category</th>\n",
        "      <th>month</th>\n",
+       "      <th>category</th>\n",
        "      <th>sales</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
-       "      <td>Electronics</td>\n",
        "      <td>2015-01-31</td>\n",
+       "      <td>Electronics</td>\n",
        "      <td>5890</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
-       "      <td>Power Tools</td>\n",
        "      <td>2015-01-31</td>\n",
+       "      <td>Power Tools</td>\n",
        "      <td>3242</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
-       "      <td>Clothing</td>\n",
        "      <td>2015-01-31</td>\n",
+       "      <td>Clothing</td>\n",
        "      <td>6961</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
-       "      <td>Electronics</td>\n",
        "      <td>2015-02-28</td>\n",
+       "      <td>Electronics</td>\n",
        "      <td>3969</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
-       "      <td>Power Tools</td>\n",
        "      <td>2015-02-28</td>\n",
+       "      <td>Power Tools</td>\n",
        "      <td>4866</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
@@ -3762,15 +3806,15 @@
        "</div>"
       ],
       "text/plain": [
-       "      category      month  sales\n",
-       "0  Electronics 2015-01-31   5890\n",
-       "1  Power Tools 2015-01-31   3242\n",
-       "2     Clothing 2015-01-31   6961\n",
-       "3  Electronics 2015-02-28   3969\n",
-       "4  Power Tools 2015-02-28   4866"
+       "        month     category  sales\n",
+       "0  2015-01-31  Electronics   5890\n",
+       "1  2015-01-31  Power Tools   3242\n",
+       "2  2015-01-31     Clothing   6961\n",
+       "3  2015-02-28  Electronics   3969\n",
+       "4  2015-02-28  Power Tools   4866"
       ]
      },
-     "execution_count": 67,
+     "execution_count": 68,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3782,7 +3826,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 68,
+   "execution_count": 69,
    "metadata": {},
    "outputs": [
     {
@@ -3791,7 +3835,7 @@
        "67"
       ]
      },
-     "execution_count": 68,
+     "execution_count": 69,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3803,7 +3847,34 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 69,
+   "execution_count": 70,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "<class 'pandas.core.frame.DataFrame'>\n",
+      "Int64Index: 67 entries, 0 to 66\n",
+      "Data columns (total 3 columns):\n",
+      " #   Column    Non-Null Count  Dtype \n",
+      "---  ------    --------------  ----- \n",
+      " 0   month     67 non-null     object\n",
+      " 1   category  67 non-null     object\n",
+      " 2   sales     67 non-null     int64 \n",
+      "dtypes: int64(1), object(2)\n",
+      "memory usage: 2.1+ KB\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Informace např. o datových typech\n",
+    "data.info()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 71,
    "metadata": {},
    "outputs": [
     {
@@ -3820,12 +3891,13 @@
        "Name: sales, dtype: float64"
       ]
      },
-     "execution_count": 69,
+     "execution_count": 71,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
+    "# Statistické informace (u číselných sloupců)\n",
     "data['sales'].describe()"
    ]
   },
@@ -3838,25 +3910,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 70,
+   "execution_count": 72,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -3911,7 +3983,7 @@
        "Power Tools 2015-02-28   4866"
       ]
      },
-     "execution_count": 70,
+     "execution_count": 72,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3932,24 +4004,28 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 71,
+   "execution_count": 73,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
        "    }\n",
        "\n",
-       "    .dataframe thead th {\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead tr th {\n",
        "        text-align: left;\n",
        "    }\n",
        "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
+       "    .dataframe thead tr:last-of-type th {\n",
+       "        text-align: right;\n",
        "    }\n",
        "</style>\n",
        "<table border=\"1\" class=\"dataframe\">\n",
@@ -4093,12 +4169,12 @@
        "Electronics     5890.0     3969.0     1281.0     7725.0     4409.0     4180.0   \n",
        "Power Tools     3242.0     4866.0     1289.0     1407.0     8171.0     9492.0   \n",
        "\n",
-       "                                                           ...                 \\\n",
-       "month       2015-07-31 2015-08-31 2015-09-30 2015-10-31    ...     2016-02-29   \n",
-       "category                                                   ...                  \n",
-       "Clothing        7989.0        NaN       31.0     7896.0    ...         4194.0   \n",
-       "Electronics     6253.0        NaN     7086.0     8298.0    ...         6290.0   \n",
-       "Power Tools     3267.0     5534.0     2996.0     2909.0    ...         8769.0   \n",
+       "                                                         ...             \\\n",
+       "month       2015-07-31 2015-08-31 2015-09-30 2015-10-31  ... 2016-02-29   \n",
+       "category                                                 ...              \n",
+       "Clothing        7989.0        NaN       31.0     7896.0  ...     4194.0   \n",
+       "Electronics     6253.0        NaN     7086.0     8298.0  ...     6290.0   \n",
+       "Power Tools     3267.0     5534.0     2996.0     2909.0  ...     8769.0   \n",
        "\n",
        "                                                                               \\\n",
        "month       2016-03-31 2016-04-30 2016-05-31 2016-06-30 2016-07-31 2016-08-31   \n",
@@ -4117,7 +4193,7 @@
        "[3 rows x 23 columns]"
       ]
      },
-     "execution_count": 71,
+     "execution_count": 73,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -4136,51 +4212,51 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 72,
+   "execution_count": 74,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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>month</th>\n",
-       "      <th>2015-01-31 00:00:00</th>\n",
-       "      <th>2015-02-28 00:00:00</th>\n",
-       "      <th>2015-03-31 00:00:00</th>\n",
-       "      <th>2015-04-30 00:00:00</th>\n",
-       "      <th>2015-05-31 00:00:00</th>\n",
-       "      <th>2015-06-30 00:00:00</th>\n",
-       "      <th>2015-07-31 00:00:00</th>\n",
-       "      <th>2015-08-31 00:00:00</th>\n",
-       "      <th>2015-09-30 00:00:00</th>\n",
-       "      <th>2015-10-31 00:00:00</th>\n",
+       "      <th>2015-01-31</th>\n",
+       "      <th>2015-02-28</th>\n",
+       "      <th>2015-03-31</th>\n",
+       "      <th>2015-04-30</th>\n",
+       "      <th>2015-05-31</th>\n",
+       "      <th>2015-06-30</th>\n",
+       "      <th>2015-07-31</th>\n",
+       "      <th>2015-08-31</th>\n",
+       "      <th>2015-09-30</th>\n",
+       "      <th>2015-10-31</th>\n",
        "      <th>...</th>\n",
-       "      <th>2016-02-29 00:00:00</th>\n",
-       "      <th>2016-03-31 00:00:00</th>\n",
-       "      <th>2016-04-30 00:00:00</th>\n",
-       "      <th>2016-05-31 00:00:00</th>\n",
-       "      <th>2016-06-30 00:00:00</th>\n",
-       "      <th>2016-07-31 00:00:00</th>\n",
-       "      <th>2016-08-31 00:00:00</th>\n",
-       "      <th>2016-09-30 00:00:00</th>\n",
-       "      <th>2016-10-31 00:00:00</th>\n",
-       "      <th>2016-11-30 00:00:00</th>\n",
+       "      <th>2016-02-29</th>\n",
+       "      <th>2016-03-31</th>\n",
+       "      <th>2016-04-30</th>\n",
+       "      <th>2016-05-31</th>\n",
+       "      <th>2016-06-30</th>\n",
+       "      <th>2016-07-31</th>\n",
+       "      <th>2016-08-31</th>\n",
+       "      <th>2016-09-30</th>\n",
+       "      <th>2016-10-31</th>\n",
+       "      <th>2016-11-30</th>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>category</th>\n",
@@ -4292,34 +4368,28 @@
        "Electronics      5890.0      3969.0      1281.0      7725.0      4409.0   \n",
        "Power Tools      3242.0      4866.0      1289.0      1407.0      8171.0   \n",
        "\n",
-       "month        2015-06-30  2015-07-31  2015-08-31  2015-09-30  2015-10-31  \\\n",
-       "category                                                                  \n",
-       "Clothing         8052.0      7989.0         NaN        31.0      7896.0   \n",
-       "Electronics      4180.0      6253.0         NaN      7086.0      8298.0   \n",
-       "Power Tools      9492.0      3267.0      5534.0      2996.0      2909.0   \n",
-       "\n",
-       "month           ...      2016-02-29  2016-03-31  2016-04-30  2016-05-31  \\\n",
-       "category        ...                                                       \n",
-       "Clothing        ...          4194.0      2059.0       471.0      5410.0   \n",
-       "Electronics     ...          6290.0      2966.0      9039.0      1450.0   \n",
-       "Power Tools     ...          8769.0      2012.0      6807.0       314.0   \n",
+       "month        2015-06-30  2015-07-31  2015-08-31  2015-09-30  2015-10-31  ...  \\\n",
+       "category                                                                 ...   \n",
+       "Clothing         8052.0      7989.0         NaN        31.0      7896.0  ...   \n",
+       "Electronics      4180.0      6253.0         NaN      7086.0      8298.0  ...   \n",
+       "Power Tools      9492.0      3267.0      5534.0      2996.0      2909.0  ...   \n",
        "\n",
-       "month        2016-06-30  2016-07-31  2016-08-31  2016-09-30  2016-10-31  \\\n",
+       "month        2016-02-29  2016-03-31  2016-04-30  2016-05-31  2016-06-30  \\\n",
        "category                                                                  \n",
-       "Clothing         8663.0      9817.0      6969.0      -735.0      4448.0   \n",
-       "Electronics      3515.0      8497.0       349.0      9324.0       919.0   \n",
-       "Power Tools      2858.0      6382.0      9039.0      2119.0      5095.0   \n",
+       "Clothing         4194.0      2059.0       471.0      5410.0      8663.0   \n",
+       "Electronics      6290.0      2966.0      9039.0      1450.0      3515.0   \n",
+       "Power Tools      8769.0      2012.0      6807.0       314.0      2858.0   \n",
        "\n",
-       "month        2016-11-30  \n",
-       "category                 \n",
-       "Clothing         -259.0  \n",
-       "Electronics        18.0  \n",
-       "Power Tools      1397.0  \n",
+       "month        2016-07-31  2016-08-31  2016-09-30  2016-10-31  2016-11-30  \n",
+       "category                                                                 \n",
+       "Clothing         9817.0      6969.0      -735.0      4448.0      -259.0  \n",
+       "Electronics      8497.0       349.0      9324.0       919.0        18.0  \n",
+       "Power Tools      6382.0      9039.0      2119.0      5095.0      1397.0  \n",
        "\n",
        "[3 rows x 23 columns]"
       ]
      },
-     "execution_count": 72,
+     "execution_count": 74,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -4338,7 +4408,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 73,
+   "execution_count": 75,
    "metadata": {},
    "outputs": [
     {
@@ -4347,7 +4417,7 @@
        "103742.0"
       ]
      },
-     "execution_count": 73,
+     "execution_count": 75,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -4365,39 +4435,37 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 74,
+   "execution_count": 76,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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>month</th>\n",
-       "      <th>2016-03-31 00:00:00</th>\n",
-       "      <th>2016-04-30 00:00:00</th>\n",
-       "      <th>2016-05-31 00:00:00</th>\n",
+       "      <th>2016-03-31</th>\n",
+       "      <th>2016-04-30</th>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>category</th>\n",
        "      <th></th>\n",
        "      <th></th>\n",
-       "      <th></th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
@@ -4405,26 +4473,24 @@
        "      <th>Electronics</th>\n",
        "      <td>2966.0</td>\n",
        "      <td>9039.0</td>\n",
-       "      <td>1450.0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>Power Tools</th>\n",
        "      <td>2012.0</td>\n",
        "      <td>6807.0</td>\n",
-       "      <td>314.0</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "text/plain": [
-       "month        2016-03-31  2016-04-30  2016-05-31\n",
-       "category                                       \n",
-       "Electronics      2966.0      9039.0      1450.0\n",
-       "Power Tools      2012.0      6807.0       314.0"
+       "month        2016-03-31  2016-04-30\n",
+       "category                           \n",
+       "Electronics      2966.0      9039.0\n",
+       "Power Tools      2012.0      6807.0"
       ]
      },
-     "execution_count": 74,
+     "execution_count": 76,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -4442,7 +4508,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 75,
+   "execution_count": 77,
    "metadata": {
     "scrolled": false
    },
@@ -4477,7 +4543,7 @@
        "Name: Clothing, dtype: float64"
       ]
      },
-     "execution_count": 75,
+     "execution_count": 77,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -4504,55 +4570,43 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Je-li nainstalována knihovna `matplotlib`, Pandas ji umí využít k tomu, aby kreslil grafy. Nastavení je trochu jiné pro Jupyter Notebook a pro příkazovou řádku.\n",
-    "\n",
-    "Používáte-li Jupyter Notebook, zapněte integraci pro kreslení grafů pomocí:"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 76,
-   "metadata": {
-    "collapsed": true
-   },
-   "outputs": [],
-   "source": [
-    "import matplotlib\n",
-    "\n",
-    "# Zapnout zobrazování grafů (procento uvozuje „magickou” zkratku IPythonu):\n",
-    "%matplotlib inline"
+    "Je-li nainstalována knihovna `matplotlib`, Pandas ji umí využít k tomu, aby kreslil grafy. Nastavení je trochu jiné pro Jupyter Notebook a pro příkazovou řádku."
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "a pak můžete přímo použít metodu `plot()`, která bez dalších argumentů vynese data z tabulky proti indexu:"
+    "V Notebooku můžete přímo použít metodu `plot()`, která bez dalších argumentů vynese data z tabulky proti indexu:"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 77,
-   "metadata": {},
+   "execution_count": 78,
+   "metadata": {
+    "scrolled": true
+   },
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "<matplotlib.axes._subplots.AxesSubplot at 0x7f4ab57c0470>"
+       "<AxesSubplot:xlabel='month'>"
       ]
      },
-     "execution_count": 77,
+     "execution_count": 78,
      "metadata": {},
      "output_type": "execute_result"
     },
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYcAAAEICAYAAAC0+DhzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
+      "image/png": "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\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x7f4ab5a56908>"
+       "<Figure size 432x288 with 1 Axes>"
       ]
      },
-     "metadata": {},
+     "metadata": {
+      "needs_background": "light"
+     },
      "output_type": "display_data"
     }
    ],
@@ -4560,6 +4614,19 @@
     "unstacked.loc['Clothing'].dropna().plot()"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "> [note]\n",
+    "> Ve starších verzích matplotlib/Notebooku je potřeba integraci pro kreslení grafů zapnout pomocí „magické” zkratky IPythonu:\n",
+    "> ```python\n",
+    "> import matplotlib\n",
+    "> \n",
+    "> %matplotlib inline\n",
+    "> ```"
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {},
@@ -4588,27 +4655,29 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 78,
+   "execution_count": 79,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "<matplotlib.axes._subplots.AxesSubplot at 0x7f4ab5733160>"
+       "<AxesSubplot:xlabel='month'>"
       ]
      },
-     "execution_count": 78,
+     "execution_count": 79,
      "metadata": {},
      "output_type": "execute_result"
     },
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY0AAAEVCAYAAAAckrn/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
+      "image/png": "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\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x7f4ab57c03c8>"
+       "<Figure size 432x288 with 1 Axes>"
       ]
      },
-     "metadata": {},
+     "metadata": {
+      "needs_background": "light"
+     },
      "output_type": "display_data"
     }
    ],
@@ -4621,27 +4690,29 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 79,
+   "execution_count": 80,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "<matplotlib.axes._subplots.AxesSubplot at 0x7f4ab567ae10>"
+       "<AxesSubplot:xlabel='category'>"
       ]
      },
-     "execution_count": 79,
+     "execution_count": 80,
      "metadata": {},
      "output_type": "execute_result"
     },
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAE7CAYAAADUylYJAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n1WkeGrroAAAAAElFTkSuQmCC\n",
+      "image/png": "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\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x7f4ab5a3de48>"
+       "<Figure size 432x288 with 1 Axes>"
       ]
      },
-     "metadata": {},
+     "metadata": {
+      "needs_background": "light"
+     },
      "output_type": "display_data"
     }
    ],
@@ -4668,64 +4739,64 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 80,
+   "execution_count": 81,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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>category</th>\n",
        "      <th>month</th>\n",
+       "      <th>category</th>\n",
        "      <th>sales</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
-       "      <td>Electronics</td>\n",
        "      <td>2015-01-31</td>\n",
+       "      <td>Electronics</td>\n",
        "      <td>5890</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
-       "      <td>Power Tools</td>\n",
        "      <td>2015-01-31</td>\n",
+       "      <td>Power Tools</td>\n",
        "      <td>3242</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
-       "      <td>Clothing</td>\n",
        "      <td>2015-01-31</td>\n",
+       "      <td>Clothing</td>\n",
        "      <td>6961</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
-       "      <td>Electronics</td>\n",
        "      <td>2015-02-28</td>\n",
+       "      <td>Electronics</td>\n",
        "      <td>3969</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
-       "      <td>Power Tools</td>\n",
        "      <td>2015-02-28</td>\n",
+       "      <td>Power Tools</td>\n",
        "      <td>4866</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
@@ -4733,15 +4804,15 @@
        "</div>"
       ],
       "text/plain": [
-       "      category      month  sales\n",
-       "0  Electronics 2015-01-31   5890\n",
-       "1  Power Tools 2015-01-31   3242\n",
-       "2     Clothing 2015-01-31   6961\n",
-       "3  Electronics 2015-02-28   3969\n",
-       "4  Power Tools 2015-02-28   4866"
+       "        month     category  sales\n",
+       "0  2015-01-31  Electronics   5890\n",
+       "1  2015-01-31  Power Tools   3242\n",
+       "2  2015-01-31     Clothing   6961\n",
+       "3  2015-02-28  Electronics   3969\n",
+       "4  2015-02-28  Power Tools   4866"
       ]
      },
-     "execution_count": 80,
+     "execution_count": 81,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -4759,16 +4830,16 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 81,
+   "execution_count": 82,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "<pandas.core.groupby.DataFrameGroupBy object at 0x7f4ab8363f28>"
+       "<pandas.core.groupby.generic.DataFrameGroupBy object at 0x7f3661840970>"
       ]
      },
-     "execution_count": 81,
+     "execution_count": 82,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -4786,25 +4857,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 82,
+   "execution_count": 83,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -4842,7 +4913,7 @@
        "Power Tools  104859"
       ]
      },
-     "execution_count": 82,
+     "execution_count": 83,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -4860,7 +4931,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 83,
+   "execution_count": 84,
    "metadata": {
     "scrolled": false
    },
@@ -4869,18 +4940,18 @@
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -4923,7 +4994,7 @@
        "Power Tools     23     23"
       ]
      },
-     "execution_count": 83,
+     "execution_count": 84,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -4941,25 +5012,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 84,
+   "execution_count": 85,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: left;\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",
@@ -5010,7 +5081,7 @@
        "         2015-05-31   4796"
       ]
      },
-     "execution_count": 84,
+     "execution_count": 85,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -5028,24 +5099,28 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 85,
+   "execution_count": 86,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
        "    }\n",
        "\n",
-       "    .dataframe thead th {\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead tr th {\n",
        "        text-align: left;\n",
        "    }\n",
        "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
+       "    .dataframe thead tr:last-of-type th {\n",
+       "        text-align: right;\n",
        "    }\n",
        "</style>\n",
        "<table border=\"1\" class=\"dataframe\">\n",
@@ -5104,7 +5179,7 @@
        "Power Tools  4559.086957  3769.0  104859 -1.044767"
       ]
      },
-     "execution_count": 85,
+     "execution_count": 86,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -5122,7 +5197,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 86,
+   "execution_count": 87,
    "metadata": {
     "scrolled": false
    },
@@ -5131,17 +5206,21 @@
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
        "    }\n",
        "\n",
-       "    .dataframe thead th {\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead tr th {\n",
        "        text-align: left;\n",
        "    }\n",
        "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
+       "    .dataframe thead tr:last-of-type th {\n",
+       "        text-align: right;\n",
        "    }\n",
        "</style>\n",
        "<table border=\"1\" class=\"dataframe\">\n",
@@ -5487,7 +5566,7 @@
        "2016-11-30  1397.0  "
       ]
      },
-     "execution_count": 86,
+     "execution_count": 87,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -5506,24 +5585,28 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 87,
+   "execution_count": 88,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
        "<div>\n",
-       "<style>\n",
-       "    .dataframe thead tr:only-child th {\n",
-       "        text-align: right;\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
        "    }\n",
        "\n",
-       "    .dataframe thead th {\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead tr th {\n",
        "        text-align: left;\n",
        "    }\n",
        "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
+       "    .dataframe thead tr:last-of-type th {\n",
+       "        text-align: right;\n",
        "    }\n",
        "</style>\n",
        "<table border=\"1\" class=\"dataframe\">\n",
@@ -5572,7 +5655,7 @@
        "20000     3   71781"
       ]
      },
-     "execution_count": 87,
+     "execution_count": 88,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -5586,32 +5669,34 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 88,
+   "execution_count": 89,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "<matplotlib.axes._subplots.AxesSubplot at 0x7f4ab5588240>"
+       "<AxesSubplot:xlabel='sales'>"
       ]
      },
-     "execution_count": 88,
+     "execution_count": 89,
      "metadata": {},
      "output_type": "execute_result"
     },
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZkAAAEKCAYAAADAVygjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYwLm/wFs+GhAV4F94AAAAABJRU5ErkJggg==\n",
+      "image/png": "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\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x7f4ab557b7f0>"
+       "<Figure size 432x288 with 1 Axes>"
       ]
      },
-     "metadata": {},
+     "metadata": {
+      "needs_background": "light"
+     },
      "output_type": "display_data"
     }
    ],
    "source": [
-    "by_thousands[('sales', 'sum')].plot()"
+    "by_thousands[('sales', 'sum')].plot.bar()"
    ]
   }
  ],
@@ -5631,7 +5716,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.6.2"
+   "version": "3.8.5"
   }
  },
  "nbformat": 4,
diff --git a/lessons/intro/pandas/static/sales.csv b/lessons/intro/pandas/static/sales.csv
new file mode 100644
index 0000000000..ce7be70fad
--- /dev/null
+++ b/lessons/intro/pandas/static/sales.csv
@@ -0,0 +1,68 @@
+,month,category,sales
+0,2015-01-31,Electronics,5890
+1,2015-01-31,Power Tools,3242
+2,2015-01-31,Clothing,6961
+3,2015-02-28,Electronics,3969
+4,2015-02-28,Power Tools,4866
+5,2015-02-28,Clothing,2578
+6,2015-03-31,Electronics,1281
+7,2015-03-31,Power Tools,1289
+8,2015-03-31,Clothing,9131
+9,2015-04-30,Electronics,7725
+10,2015-04-30,Power Tools,1407
+11,2015-04-30,Clothing,618
+12,2015-05-31,Electronics,4409
+13,2015-05-31,Power Tools,8171
+14,2015-05-31,Clothing,4796
+15,2015-06-30,Electronics,4180
+16,2015-06-30,Power Tools,9492
+17,2015-06-30,Clothing,8052
+18,2015-07-31,Electronics,6253
+19,2015-07-31,Power Tools,3267
+20,2015-07-31,Clothing,7989
+21,2015-08-31,Power Tools,5534
+22,2015-09-30,Electronics,7086
+23,2015-09-30,Power Tools,2996
+24,2015-09-30,Clothing,31
+25,2015-10-31,Electronics,8298
+26,2015-10-31,Power Tools,2909
+27,2015-10-31,Clothing,7896
+28,2015-11-30,Electronics,494
+29,2015-11-30,Power Tools,4243
+30,2015-11-30,Clothing,7016
+31,2015-12-31,Electronics,3938
+32,2015-12-31,Power Tools,3769
+33,2015-12-31,Clothing,7969
+34,2016-01-31,Electronics,7852
+35,2016-01-31,Power Tools,8882
+36,2016-01-31,Clothing,8627
+37,2016-02-29,Electronics,6290
+38,2016-02-29,Power Tools,8769
+39,2016-02-29,Clothing,4194
+40,2016-03-31,Electronics,2966
+41,2016-03-31,Power Tools,2012
+42,2016-03-31,Clothing,2059
+43,2016-04-30,Electronics,9039
+44,2016-04-30,Power Tools,6807
+45,2016-04-30,Clothing,471
+46,2016-05-31,Electronics,1450
+47,2016-05-31,Power Tools,314
+48,2016-05-31,Clothing,5410
+49,2016-06-30,Electronics,3515
+50,2016-06-30,Power Tools,2858
+51,2016-06-30,Clothing,8663
+52,2016-07-31,Electronics,8497
+53,2016-07-31,Power Tools,6382
+54,2016-07-31,Clothing,9817
+55,2016-08-31,Electronics,349
+56,2016-08-31,Power Tools,9039
+57,2016-08-31,Clothing,6969
+58,2016-09-30,Electronics,9324
+59,2016-09-30,Power Tools,2119
+60,2016-09-30,Clothing,-735
+61,2016-10-31,Electronics,919
+62,2016-10-31,Power Tools,5095
+63,2016-10-31,Clothing,4448
+64,2016-11-30,Electronics,18
+65,2016-11-30,Power Tools,1397
+66,2016-11-30,Clothing,-259
diff --git a/runs/2020/nipyt-zima/info.yml b/runs/2020/nipyt-zima/info.yml
index f9f8f98bbf..683e96dfa6 100644
--- a/runs/2020/nipyt-zima/info.yml
+++ b/runs/2020/nipyt-zima/info.yml
@@ -25,12 +25,20 @@ plan:
 
 - base: pandas
   date: 2020-10-07
+  materials:
+  - lesson: intro/notebook
+  - lesson: intro/pandas
 
 - base: distribution
   date: 2020-10-14
 
 - base: numpy
   date: 2020-10-21
+  materials:
+  - lesson: intro/numpy
+  - title: Tahák na NumPy
+    url: https://pyvec.github.io/cheatsheets/numpy/numpy-cs.pdf
+    type: cheatsheet
 
 - base: testing
   date: 2020-11-04