demos/Identifikation von noch vorhandenem Wissen im Linux-Kernel.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Identifikation von noch vorhandenem Wissen im Linux-Kernel\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Der Patient\n",
    "\n",
    "#### Linux\n",
    "\n",
    "* Betriebsystem-Kernel\n",
    "* Hat verschiedene Treiberkomponenten\n",
    "* Fast ausschließlich in C geschrieben\n",
    "* Entwickelt von über 800.000 Entwicklern"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### I. Idee (1/2)\n",
    "\n",
    "<b>Fragestellung</b>\n",
    "* Gibt es besonders alte Komponenten (No-Go-Areas)?\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### I. Idee (2/2)\n",
    "#### Umsetzung\n",
    "* Werkzeuge: Jupyter, Python, pandas, matplotlib\n",
    "* Datenquelle: Git Blame Log\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "**Meta-Ziel:** Grundfunktionen anhand eines einfachen Show-Cases sehen."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<b>Git Blame Log</b>\n",
    "<div align=\"center\">\n",
    "    <img src =\"resources/linux_1.gif\" align=\"center\"/>\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<b>Git Blame Log</b>\n",
    "<div align=\"center\">\n",
    "    <img src =\"resources/linux_2.gif\" align=\"center\"/>\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<b>Git Blame Log</b>\n",
    "<div align=\"center\">\n",
    "    <img src =\"resources/linux_3.gif\" align=\"center\"/>\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### II. Datenbeschaffung\n",
    "*Wir laden Git Blame Daten aus einer CSV-Datei*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    },
    "tags": [
     "delete"
    ]
   },
   "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>path</th>\n",
       "      <th>author</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>line</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>drivers/scsi/bfa/bfad_drv.h</td>\n",
       "      <td>Anil Gurumurthy</td>\n",
       "      <td>1448528085000000000</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>drivers/scsi/bfa/bfad_drv.h</td>\n",
       "      <td>Anil Gurumurthy</td>\n",
       "      <td>1448528085000000000</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>drivers/scsi/bfa/bfad_drv.h</td>\n",
       "      <td>Anil Gurumurthy</td>\n",
       "      <td>1448528085000000000</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>drivers/scsi/bfa/bfad_drv.h</td>\n",
       "      <td>Jing Huang</td>\n",
       "      <td>1253753175000000000</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>drivers/scsi/bfa/bfad_drv.h</td>\n",
       "      <td>Anil Gurumurthy</td>\n",
       "      <td>1448528085000000000</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          path           author            timestamp  line\n",
       "0  drivers/scsi/bfa/bfad_drv.h  Anil Gurumurthy  1448528085000000000     1\n",
       "1  drivers/scsi/bfa/bfad_drv.h  Anil Gurumurthy  1448528085000000000     2\n",
       "2  drivers/scsi/bfa/bfad_drv.h  Anil Gurumurthy  1448528085000000000     3\n",
       "3  drivers/scsi/bfa/bfad_drv.h       Jing Huang  1253753175000000000     4\n",
       "4  drivers/scsi/bfa/bfad_drv.h  Anil Gurumurthy  1448528085000000000     5"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "log = pd.read_csv(\"dataset/linux_blame_log.csv.gz\")\n",
    "log.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "*Was haben wir hier eigentlich?*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    },
    "tags": [
     "delete"
    ]
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 5665947 entries, 0 to 5665946\n",
      "Data columns (total 4 columns):\n",
      " #   Column     Dtype \n",
      "---  ------     ----- \n",
      " 0   path       object\n",
      " 1   author     object\n",
      " 2   timestamp  int64 \n",
      " 3   line       int64 \n",
      "dtypes: int64(2), object(2)\n",
      "memory usage: 172.9+ MB\n"
     ]
    }
   ],
   "source": [
    "log.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "<b>1</b> **DataFrame** (~ programmierbares Excel-Arbeitsblatt), <b>4</b> **Series** (= Spalten), <b>5665947</b> **Rows** (= Einträge)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### III. Bereinigen\n",
    "* Daten sind oft nicht so, wie man sie braucht\n",
    "* Datentypen passen teilweise noch nicht"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "*Wir wandeln die Zeitstempel um*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    },
    "tags": [
     "delete"
    ]
   },
   "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>path</th>\n",
       "      <th>author</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>line</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>drivers/scsi/bfa/bfad_drv.h</td>\n",
       "      <td>Anil Gurumurthy</td>\n",
       "      <td>2015-11-26 08:54:45</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>drivers/scsi/bfa/bfad_drv.h</td>\n",
       "      <td>Anil Gurumurthy</td>\n",
       "      <td>2015-11-26 08:54:45</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>drivers/scsi/bfa/bfad_drv.h</td>\n",
       "      <td>Anil Gurumurthy</td>\n",
       "      <td>2015-11-26 08:54:45</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>drivers/scsi/bfa/bfad_drv.h</td>\n",
       "      <td>Jing Huang</td>\n",
       "      <td>2009-09-24 00:46:15</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>drivers/scsi/bfa/bfad_drv.h</td>\n",
       "      <td>Anil Gurumurthy</td>\n",
       "      <td>2015-11-26 08:54:45</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          path           author           timestamp  line\n",
       "0  drivers/scsi/bfa/bfad_drv.h  Anil Gurumurthy 2015-11-26 08:54:45     1\n",
       "1  drivers/scsi/bfa/bfad_drv.h  Anil Gurumurthy 2015-11-26 08:54:45     2\n",
       "2  drivers/scsi/bfa/bfad_drv.h  Anil Gurumurthy 2015-11-26 08:54:45     3\n",
       "3  drivers/scsi/bfa/bfad_drv.h       Jing Huang 2009-09-24 00:46:15     4\n",
       "4  drivers/scsi/bfa/bfad_drv.h  Anil Gurumurthy 2015-11-26 08:54:45     5"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "log['timestamp'] = pd.to_datetime(log['timestamp'])\n",
    "log.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### IV. Anreichern\n",
    "\n",
    "* Vorhandenen Daten noch zusätzlich mit anderen Datenquellen verschneiden\n",
    "* Aber auch: Teile aus vorhanden Daten extrahieren"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "=> Dadurch werden mehrere <b>Perspektiven</b> auf ein Problem möglich"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "*Wir berechnen uns das Alter jeder Codezeilenänderung*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    },
    "tags": [
     "delete"
    ]
   },
   "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>path</th>\n",
       "      <th>author</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>line</th>\n",
       "      <th>age</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>drivers/scsi/bfa/bfad_drv.h</td>\n",
       "      <td>Anil Gurumurthy</td>\n",
       "      <td>2015-11-26 08:54:45</td>\n",
       "      <td>1</td>\n",
       "      <td>1760 days 12:28:55.621148</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>drivers/scsi/bfa/bfad_drv.h</td>\n",
       "      <td>Anil Gurumurthy</td>\n",
       "      <td>2015-11-26 08:54:45</td>\n",
       "      <td>2</td>\n",
       "      <td>1760 days 12:28:55.621148</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>drivers/scsi/bfa/bfad_drv.h</td>\n",
       "      <td>Anil Gurumurthy</td>\n",
       "      <td>2015-11-26 08:54:45</td>\n",
       "      <td>3</td>\n",
       "      <td>1760 days 12:28:55.621148</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>drivers/scsi/bfa/bfad_drv.h</td>\n",
       "      <td>Jing Huang</td>\n",
       "      <td>2009-09-24 00:46:15</td>\n",
       "      <td>4</td>\n",
       "      <td>4014 days 20:37:25.621148</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>drivers/scsi/bfa/bfad_drv.h</td>\n",
       "      <td>Anil Gurumurthy</td>\n",
       "      <td>2015-11-26 08:54:45</td>\n",
       "      <td>5</td>\n",
       "      <td>1760 days 12:28:55.621148</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          path           author           timestamp  line  \\\n",
       "0  drivers/scsi/bfa/bfad_drv.h  Anil Gurumurthy 2015-11-26 08:54:45     1   \n",
       "1  drivers/scsi/bfa/bfad_drv.h  Anil Gurumurthy 2015-11-26 08:54:45     2   \n",
       "2  drivers/scsi/bfa/bfad_drv.h  Anil Gurumurthy 2015-11-26 08:54:45     3   \n",
       "3  drivers/scsi/bfa/bfad_drv.h       Jing Huang 2009-09-24 00:46:15     4   \n",
       "4  drivers/scsi/bfa/bfad_drv.h  Anil Gurumurthy 2015-11-26 08:54:45     5   \n",
       "\n",
       "                        age  \n",
       "0 1760 days 12:28:55.621148  \n",
       "1 1760 days 12:28:55.621148  \n",
       "2 1760 days 12:28:55.621148  \n",
       "3 4014 days 20:37:25.621148  \n",
       "4 1760 days 12:28:55.621148  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "log['age'] = pd.Timestamp('today') - log['timestamp']\n",
    "log.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "*Wir ordnen jeder Zeilenänderung einer Komponente zu*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    },
    "tags": [
     "delete"
    ]
   },
   "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>path</th>\n",
       "      <th>author</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>line</th>\n",
       "      <th>age</th>\n",
       "      <th>component</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>drivers/scsi/bfa/bfad_drv.h</td>\n",
       "      <td>Anil Gurumurthy</td>\n",
       "      <td>2015-11-26 08:54:45</td>\n",
       "      <td>1</td>\n",
       "      <td>1760 days 12:28:55.621148</td>\n",
       "      <td>drivers:scsi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>drivers/scsi/bfa/bfad_drv.h</td>\n",
       "      <td>Anil Gurumurthy</td>\n",
       "      <td>2015-11-26 08:54:45</td>\n",
       "      <td>2</td>\n",
       "      <td>1760 days 12:28:55.621148</td>\n",
       "      <td>drivers:scsi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>drivers/scsi/bfa/bfad_drv.h</td>\n",
       "      <td>Anil Gurumurthy</td>\n",
       "      <td>2015-11-26 08:54:45</td>\n",
       "      <td>3</td>\n",
       "      <td>1760 days 12:28:55.621148</td>\n",
       "      <td>drivers:scsi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>drivers/scsi/bfa/bfad_drv.h</td>\n",
       "      <td>Jing Huang</td>\n",
       "      <td>2009-09-24 00:46:15</td>\n",
       "      <td>4</td>\n",
       "      <td>4014 days 20:37:25.621148</td>\n",
       "      <td>drivers:scsi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>drivers/scsi/bfa/bfad_drv.h</td>\n",
       "      <td>Anil Gurumurthy</td>\n",
       "      <td>2015-11-26 08:54:45</td>\n",
       "      <td>5</td>\n",
       "      <td>1760 days 12:28:55.621148</td>\n",
       "      <td>drivers:scsi</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          path           author           timestamp  line  \\\n",
       "0  drivers/scsi/bfa/bfad_drv.h  Anil Gurumurthy 2015-11-26 08:54:45     1   \n",
       "1  drivers/scsi/bfa/bfad_drv.h  Anil Gurumurthy 2015-11-26 08:54:45     2   \n",
       "2  drivers/scsi/bfa/bfad_drv.h  Anil Gurumurthy 2015-11-26 08:54:45     3   \n",
       "3  drivers/scsi/bfa/bfad_drv.h       Jing Huang 2009-09-24 00:46:15     4   \n",
       "4  drivers/scsi/bfa/bfad_drv.h  Anil Gurumurthy 2015-11-26 08:54:45     5   \n",
       "\n",
       "                        age     component  \n",
       "0 1760 days 12:28:55.621148  drivers:scsi  \n",
       "1 1760 days 12:28:55.621148  drivers:scsi  \n",
       "2 1760 days 12:28:55.621148  drivers:scsi  \n",
       "3 4014 days 20:37:25.621148  drivers:scsi  \n",
       "4 1760 days 12:28:55.621148  drivers:scsi  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "log['component'] = log['path'].str.split(\"/\").str[:2].str.join(\":\")\n",
    "log.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "<br/> <small><i>String-Operationen...die dauern. Gibt aber diverse Optimierungsmöglichkeiten!</i></small>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### V. Aggregieren\n",
    "* Vorhandene Daten sind oft zu viel für manuelle Sichtung\n",
    "* Neue Einsichten über Problem aber oft auf hoher Flugbahn möglich"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "*Wir fassen nach Komponenten zusammen und arbeiten mit der jeweils jüngsten Zeilenänderung weiter*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    },
    "tags": [
     "delete"
    ]
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "component\n",
       "drivers:scsi    893\n",
       "drivers:i2c     893\n",
       "drivers:net     893\n",
       "drivers:of      893\n",
       "drivers:pci     894\n",
       "Name: age, dtype: int64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "age_per_component = log.groupby(\"component\")['age'].min().sort_values().dt.days\n",
    "age_per_component.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### IV. Visualisieren\n",
    "* Grafische Darstellung geben Analysen den letzten Schliff\n",
    "* Probleme können Außenstehenden visuell dargestellt besser kommuniziert werden"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "*Wir bauen ein Diagramm mit min. Alter pro Komponente*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    },
    "tags": [
     "delete"
    ]
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "age_per_component.plot.bar(figsize=[15,5], rot=90);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Vielen Dank! Fragen?\n",
    "\n",
    "<b>Markus Harrer</b><br/>\n",
    "innoQ Deutschland GmbH\n",
    "  \n",
    "markus.harrer@innoq.com \n",
    "\n",
    "`@feststelltaste`"
   ]
  }
 ],
 "metadata": {
  "celltoolbar": "Slideshow",
  "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": "3.8.3"
  },
  "rise": {
   "controls": false,
   "showSlideNumber": "speaker",
   "slideNumber ": "false",
   "transition": "none"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}