02_iDigBio_and_GBIF_Specimens_By_Continent_Over_Time.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Continuing Graphs of Specimens Over Time - Continents and GBIF\n",
    "\n",
    "This notebook continues the work done in 01_iDigBio_Specimens_Collected_Over_Time.ipynb, you may want to review that notebook before contiuing. If you are interested in seeing what kinds of outputs are possible, you are welcome to skip to the graphs at the end of this notebook.\n",
    "\n",
    "Our goal for this notebook is to see how many specimens were collected on each continent during each year from 1817 to 2017. A stacked bar graph with a different color for each continent is a good way to view this data.\n",
    "\n",
    "We will introduce some additional concepts such as user defined functions, crosstabs, and how to write SQL statements instead of using the native pyspark methods. We'll move quickly though concepts already covered in earlier notebooks."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setup\n",
    "\n",
    "Import needed functions and libraries."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# col() selects columns from a data frame, year() works on dates, and udf() creates user\n",
    "# defined functions\n",
    "from pyspark.sql.functions import col, year, udf\n",
    "\n",
    "# Plotting library and configuration to show graphs in the notebook\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Loading the data set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df = sqlContext.read.load(\"/guoda/data/idigbio-20190612T171757.parquet\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "105429263"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.count()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Year collected by continent\n",
    "\n",
    "To get a feel for what a summary of year and continent look like, we will again group by the year column but also add the continent column. The result will be a row for each unique combination of year and continent with the count of how many times that occurs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "yc_sum = (df\n",
    "          .groupBy(year(\"datecollected\").cast(\"integer\").alias(\"yearcollected\"),\n",
    "                   \"continent\")\n",
    "          .count()\n",
    "          .orderBy(\"yearcollected\")\n",
    "          )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Row(yearcollected=None, continent='australia', count=1390303),\n",
       " Row(yearcollected=None, continent='east indies', count=268),\n",
       " Row(yearcollected=None, continent='eurasie', count=33),\n",
       " Row(yearcollected=None, continent='afrotropique', count=252),\n",
       " Row(yearcollected=None, continent='middle of america, westcoast', count=1),\n",
       " Row(yearcollected=None, continent='europe, europe', count=1636),\n",
       " Row(yearcollected=None, continent='océanien', count=16203),\n",
       " Row(yearcollected=None, continent='sw pacific', count=3),\n",
       " Row(yearcollected=None, continent='arctic ocean', count=19825),\n",
       " Row(yearcollected=None, continent='northern america', count=2408),\n",
       " Row(yearcollected=None, continent='néotropique', count=384),\n",
       " Row(yearcollected=None, continent='sem informação', count=2),\n",
       " Row(yearcollected=None, continent='asia, europe', count=3),\n",
       " Row(yearcollected=None, continent='asia temperate', count=3),\n",
       " Row(yearcollected=None, continent='australia?', count=1),\n",
       " Row(yearcollected=None, continent='south american', count=4),\n",
       " Row(yearcollected=None, continent='samoa islands', count=8),\n",
       " Row(yearcollected=None, continent='tweedia', count=1),\n",
       " Row(yearcollected=None, continent='funastrum', count=1),\n",
       " Row(yearcollected=None, continent='indian o.', count=29)]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yc_sum.head(20)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "There are a lot of things that are not continents there! While iDigBio cleans stuff up a bit before indexing it, not everything is obvious to do. For the purposes of our graph, we will make an \"other\" continent and group things that are not one of the 7 continents (plus \"oceania\" which GBIF uses and we'll look at GBIF data lower down) into that category.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Fixing continent with a user defined function\n",
    "\n",
    "To create a new column that has a fixed continent field, we will write a user defined function in Python and apply that function to every row in the data frame. This is exactly how the year() function from the pyspark library works. But, since we are writing the function in Python, we need to convert it into a data frame aware function with the pyspark function udf(). Once we do that, we can use our custom function just the same way we used year() above.\n",
    "\n",
    "So let's write our Python function:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "other\n",
      "europe\n",
      "oceania\n",
      "north america\n",
      "north america\n",
      "other\n",
      "asia\n"
     ]
    }
   ],
   "source": [
    "# allowed continents\n",
    "continents = set([\"africa\", \"australia\", \"asia\", \"oceania\", \"europe\", \"north america\", \"south america\"])\n",
    "\n",
    "def fix_continent(c):\n",
    "    \"\"\"\n",
    "    If the given string contains one of the continents, return which one. If not\n",
    "    then return \"other\". Lower case the string and also replace \"_\" with \" \" to\n",
    "    be able to clean up GBIF's representation of continents that contain spaces.\n",
    "    \"\"\"\n",
    "    for continent in continents:\n",
    "        if (c) and (continent in c.lower().replace(\"_\", \" \")):\n",
    "            return continent\n",
    "    return \"other\"\n",
    "\n",
    "# Some example calls to this function\n",
    "print(fix_continent(None)) \n",
    "print(fix_continent(\"europe\"))    \n",
    "print(fix_continent(\"oceania\"))\n",
    "print(fix_continent(\"NORTH_AMERICA\"))\n",
    "print(fix_continent(\"North America, Canada, Manitoba, Churchill\"))\n",
    "print(fix_continent(\"East Indies, Indonesia: Pulo Pandjang, off Sumatra\"))\n",
    "print(fix_continent(\"Asia; Thailand; Pathum Thani\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "That output looks right. So now let's make a user defined function for Spark from our Python function. Notice that we pass the Python function as an argument to udf() and udf() returns to us a new function. In Python, functions are what's called \"first class objects\" and we can pass them around and manipulate them just like a variable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "fix_continent_udf = udf(fix_continent)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "Now re-run our grouping. This time use the fix_continent_udf() function to calculate a new column and add it to the dataframe with the withColumn() method. Then we can refer to that new column in the group by method. (Yes, we could have put the fix_continent_udf() call into the group by just like we put the year() call but this is an opportunity to introduce withColumn().)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "yc_sum = (df\n",
    "          .withColumn(\"fixed\", fix_continent_udf(df.continent))\n",
    "          .groupBy(year(\"datecollected\").cast(\"integer\").alias(\"yearcollected\"),\n",
    "                   \"fixed\")\n",
    "          .count()\n",
    "          .orderBy(\"yearcollected\")\n",
    "          )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Row(yearcollected=None, fixed='africa', count=807947),\n",
       " Row(yearcollected=None, fixed='australia', count=1391203),\n",
       " Row(yearcollected=None, fixed='other', count=12824527),\n",
       " Row(yearcollected=None, fixed='south america', count=1414561),\n",
       " Row(yearcollected=None, fixed='europe', count=2324110),\n",
       " Row(yearcollected=None, fixed='north america', count=6798184),\n",
       " Row(yearcollected=None, fixed='oceania', count=446901),\n",
       " Row(yearcollected=None, fixed='asia', count=1480784),\n",
       " Row(yearcollected=100, fixed='north america', count=1),\n",
       " Row(yearcollected=100, fixed='other', count=1),\n",
       " Row(yearcollected=100, fixed='south america', count=1),\n",
       " Row(yearcollected=102, fixed='north america', count=1),\n",
       " Row(yearcollected=103, fixed='asia', count=1),\n",
       " Row(yearcollected=103, fixed='north america', count=1),\n",
       " Row(yearcollected=104, fixed='other', count=2),\n",
       " Row(yearcollected=105, fixed='other', count=5),\n",
       " Row(yearcollected=106, fixed='africa', count=3),\n",
       " Row(yearcollected=106, fixed='other', count=6),\n",
       " Row(yearcollected=107, fixed='europe', count=1),\n",
       " Row(yearcollected=107, fixed='north america', count=2)]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yc_sum.head(20)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "That looks much better. Notice now that we can actually see all of the yearcollected=None values. Before there were many continents so they went well past the top 20 results we were looking at. Now we can see all 8 counts, one for each continent."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Cross tabs to make plottable data\n",
    "\n",
    "What we have created at this point is a long list of year, continent, and count combinations. This representation of the data is normalized and can be thought of as the \"long\" form. Plotting wants to see each variable in one column though so the better represenation would the de-normalized or \"wide\" form. We can do this in Pandas with piviot() or in R you can use one of the reformating functions. We can also do this in Spark directly.\n",
    "\n",
    "We will also take this opportunity to show how more typically a full data pipeline is shown in Spark. We will calculate the Pandas data frame all at once starting from the original df we loaded, filtering on years, creating the new fixed continents, and summarizing with group by. Each line in the pipeline below becomes the input for the next lower line. \n",
    "\n",
    "Notice that we first select only the columns of interest to let Spark know we don't need the other 200+ fields in iDigBio. Then we filter on year to cut down on the rows we need to work with. Only then do we run our udf which only runs on the filtered rows.\n",
    "\n",
    "The new step in here is the crosstab() method. This is the method that takes all values in the column that is the second argument and makes them in to columns. You'll see the output after."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "pandas_yc_cross = (df\n",
    "            .select(year(\"datecollected\").cast(\"integer\").alias(\"yearcollected\"),\n",
    "                    df.continent)\n",
    "            .filter(col(\"yearcollected\") >= 1817)\n",
    "            .filter(col(\"yearcollected\") <= 2017)\n",
    "            .withColumn(\"fixed\", fix_continent_udf(df.continent))\n",
    "            .crosstab(\"yearcollected\", \"fixed\")\n",
    "            .orderBy(col(\"yearcollected_fixed\"))\n",
    "            .toPandas()\n",
    "            )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>yearcollected_fixed</th>\n",
       "      <th>africa</th>\n",
       "      <th>asia</th>\n",
       "      <th>australia</th>\n",
       "      <th>europe</th>\n",
       "      <th>north america</th>\n",
       "      <th>oceania</th>\n",
       "      <th>other</th>\n",
       "      <th>south america</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1817</td>\n",
       "      <td>77</td>\n",
       "      <td>139</td>\n",
       "      <td>391</td>\n",
       "      <td>534</td>\n",
       "      <td>471</td>\n",
       "      <td>100</td>\n",
       "      <td>290</td>\n",
       "      <td>1454</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1818</td>\n",
       "      <td>43</td>\n",
       "      <td>250</td>\n",
       "      <td>480</td>\n",
       "      <td>805</td>\n",
       "      <td>5365</td>\n",
       "      <td>16</td>\n",
       "      <td>4246</td>\n",
       "      <td>361</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1819</td>\n",
       "      <td>76</td>\n",
       "      <td>333</td>\n",
       "      <td>488</td>\n",
       "      <td>1065</td>\n",
       "      <td>560</td>\n",
       "      <td>42</td>\n",
       "      <td>873</td>\n",
       "      <td>313</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1820</td>\n",
       "      <td>51</td>\n",
       "      <td>156</td>\n",
       "      <td>218</td>\n",
       "      <td>1745</td>\n",
       "      <td>644</td>\n",
       "      <td>8</td>\n",
       "      <td>1001</td>\n",
       "      <td>211</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1821</td>\n",
       "      <td>79</td>\n",
       "      <td>715</td>\n",
       "      <td>115</td>\n",
       "      <td>1870</td>\n",
       "      <td>466</td>\n",
       "      <td>35</td>\n",
       "      <td>848</td>\n",
       "      <td>653</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  yearcollected_fixed  africa  asia  australia  europe  north america  \\\n",
       "0                1817      77   139        391     534            471   \n",
       "1                1818      43   250        480     805           5365   \n",
       "2                1819      76   333        488    1065            560   \n",
       "3                1820      51   156        218    1745            644   \n",
       "4                1821      79   715        115    1870            466   \n",
       "\n",
       "   oceania  other  south america  \n",
       "0      100    290           1454  \n",
       "1       16   4246            361  \n",
       "2       42    873            313  \n",
       "3        8   1001            211  \n",
       "4       35    848            653  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pandas_yc_cross.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "That looks pretty intuitive and we made the whole thing with one Spark statement. You can tell it's a Pandas dataframe now because the display is nice and tabular."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Making the graph\n",
    "\n",
    "Plotting isn't the focus of this notebook. Briefly, we'll make a function that stacks up multiple variables (continent counts) on top of each other and shades them different colors."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "def stack_bar_plot(pdf, columns, colors):\n",
    "    \"\"\"\n",
    "    Matplotlib needs to have the bars stacked on top of each other by hand.\n",
    "    First, we plot one column of data, then we make a ...\n",
    "    \"\"\"\n",
    "    plots = []\n",
    "    bottoms = pd.DataFrame(np.zeros((len(pdf['yearcollected_fixed']), 1)))\n",
    "    for c in columns:\n",
    "        plots.append(\n",
    "          plt.bar(pdf['yearcollected_fixed'], pdf[c], \n",
    "                  color=colors[len(plots)], edgecolor='none',\n",
    "                  width=1.0, bottom=bottoms[0])\n",
    "        )\n",
    "        bottoms[0] += pdf[c]\n",
    "    return plots"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "and then we'll call our function to make the graph and add some axes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABA0AAAGHCAYAAADbbqv0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nIEmSJEmSBhkaSJIkSZKkQf8FprHnbxGkrssAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f3da5cd0668>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# set up ordering of continents and their colors\n",
    "continents_list = sorted(continents)\n",
    "continents_list.insert(0, \"other\")\n",
    "colors = [\"#888888\", \"#a75902\", \"#00e5c8\", \"#1bcda7\", \"#37b686\", \"#539f65\", \"#51827c\", \"#bf6603\"]\n",
    "\n",
    "plots = stack_bar_plot(pandas_yc_cross, continents_list, colors)\n",
    "        \n",
    "# Start of WWI\n",
    "plt.axvline(x=1914)\n",
    "# Start of WWII\n",
    "plt.axvline(x=1939)\n",
    "# \"1988 - October 31: President Reagan signs the NSF Authorization Act of 1988,\n",
    "# thereby authorizing the doubling of the NSF budget over the next five years.\"\n",
    "plt.axvline(x=1988)\n",
    "\n",
    "plt.legend(plots, continents_list, loc=2)\n",
    "plt.title(\"Specimens in iDigBio by Collection Year and Continent\")\n",
    "plt.ylabel(\"Number of Specimen Records\")\n",
    "plt.xlabel(\"Year\")\n",
    "axes = plt.gca()\n",
    "axes.set_xlim([1815, 2020])\n",
    "axes.set_ylim([0, 1200000])\n",
    "fig = plt.gcf()\n",
    "fig.set_size_inches(12, 4)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There it is, the number of specimen records in iDigBio by collection date and continent. We chose to add some vertical lines to the graph at particularly interesting dates. You can read the code comments to see what those are."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## Doing the same with GBIF data\n",
    "\n",
    "GBIF periocdically produces a similar graph to the above on their [analytics page](http://www.gbif.org/analytics/global) titled \"Records by year of occurrence\" but it's not very comparable because it include observation records as well. It would be great if we could run the same process on their data after filtering it for only specimens so we can look at how the contents of different occurrance record aggregators differs. (This is an expected and healthy thing. One of the goals of the GUODA collaboration is to begin to characterize the relationships between sets of aggregated data across domains without that work being done by one entity. Notice too that this copy of GBIF is pretty old. Again this is GUODA's fault as we have not been running our import regularly while we worked on improving other parts of our system.)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can start by reading in the GBIF data that is stored in GUODA."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "gbif_df = sqlContext.read.parquet(\"/guoda/data/gbif-idigbio.parquet/source=gbif/date=20160825\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We use a different process for formating GBIF data and as part of GUODA's process, the columns are named with a syntax that can be difficult to work with in Python. First let's use the SQL \"as\" syntax to rename the columns.\n",
    "\n",
    "Spark data frames can also be manupulated with SQL statements instead of the pyspark functions. Note that Spark does not support the full SQL standard, it only supports syntx similar to the native functions.\n",
    "\n",
    "The first step is to tell Spark what name we want to use to refer to our data frame with in the SQL. That is the createOrReplaceTempView() method. After than we can send our SQL as a string to the SQL Context associated with our Spark Context and we will get back a data frame. Because we'll use this data frame more than once, we'll persist it so it will not be recalculated."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As part of our SQL statement we will also filter for only those records that contain the string \"SPECIMEN\" in the basisOfRecord field. This may or may not be the best way to select only specimens from GBIF but for our graph it's good enough."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "gbif_df.createOrReplaceTempView(\"gbif_df\")\n",
    "\n",
    "renamed_df = sqlContext.sql(\"\"\"\n",
    "SELECT `http://rs.tdwg.org/dwc/terms/eventDate` as eventDate,\n",
    "       `http://rs.tdwg.org/dwc/terms/continent` as continent\n",
    "FROM gbif_df\n",
    "WHERE `http://rs.tdwg.org/dwc/terms/basisOfRecord` LIKE \"%SPECIMEN%\"\n",
    "\"\"\").persist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "63184115"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "renamed_df.count()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will again run the crosstab all at once on this data frame with renamed columns. This is exactly the same summarization we ran on the iDigBio data frame."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "pandas_yc_cross_gbif = (renamed_df\n",
    "            .select(year(\"eventDate\").cast(\"integer\").alias(\"yearcollected\"),\n",
    "                    renamed_df.continent)\n",
    "            .filter(col(\"yearcollected\") >= 1817)\n",
    "            .filter(col(\"yearcollected\") <= 2017)\n",
    "            .withColumn(\"fixed\", fix_continent_udf(renamed_df.continent))\n",
    "            .crosstab(\"yearcollected\", \"fixed\")\n",
    "            .orderBy(col(\"yearcollected_fixed\"))\n",
    "            .toPandas()\n",
    "            )\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>yearcollected_fixed</th>\n",
       "      <th>africa</th>\n",
       "      <th>asia</th>\n",
       "      <th>europe</th>\n",
       "      <th>north america</th>\n",
       "      <th>oceania</th>\n",
       "      <th>other</th>\n",
       "      <th>south america</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1817</td>\n",
       "      <td>5</td>\n",
       "      <td>7</td>\n",
       "      <td>67</td>\n",
       "      <td>14</td>\n",
       "      <td>23</td>\n",
       "      <td>344</td>\n",
       "      <td>72</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1818</td>\n",
       "      <td>5</td>\n",
       "      <td>108</td>\n",
       "      <td>69</td>\n",
       "      <td>7</td>\n",
       "      <td>43</td>\n",
       "      <td>423</td>\n",
       "      <td>112</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1819</td>\n",
       "      <td>14</td>\n",
       "      <td>140</td>\n",
       "      <td>91</td>\n",
       "      <td>19</td>\n",
       "      <td>45</td>\n",
       "      <td>400</td>\n",
       "      <td>83</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1820</td>\n",
       "      <td>4</td>\n",
       "      <td>31</td>\n",
       "      <td>265</td>\n",
       "      <td>225</td>\n",
       "      <td>38</td>\n",
       "      <td>505</td>\n",
       "      <td>79</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1821</td>\n",
       "      <td>5</td>\n",
       "      <td>121</td>\n",
       "      <td>273</td>\n",
       "      <td>203</td>\n",
       "      <td>10</td>\n",
       "      <td>297</td>\n",
       "      <td>81</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  yearcollected_fixed  africa  asia  europe  north america  oceania  other  \\\n",
       "0                1817       5     7      67             14       23    344   \n",
       "1                1818       5   108      69              7       43    423   \n",
       "2                1819      14   140      91             19       45    400   \n",
       "3                1820       4    31     265            225       38    505   \n",
       "4                1821       5   121     273            203       10    297   \n",
       "\n",
       "   south america  \n",
       "0             72  \n",
       "1            112  \n",
       "2             83  \n",
       "3             79  \n",
       "4             81  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pandas_yc_cross_gbif.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And we'll use the exact same code for graphing as well. We just need to remove the \"australia\" and it's color from our continent list because GBIF doesn't use it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABA0AAAGHCAYAAADbbqv0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f3da5a96eb8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# GBIF doesn't have \"australia\"\n",
    "try:\n",
    "    continents_list.remove(\"australia\")\n",
    "    colors.remove(\"#1bcda7\")\n",
    "except:\n",
    "    pass\n",
    "\n",
    "plots = stack_bar_plot(pandas_yc_cross_gbif, continents_list, colors)\n",
    "\n",
    "# Start of WWI\n",
    "plt.axvline(x=1914)\n",
    "\n",
    "# Start of WWII\n",
    "plt.axvline(x=1939)\n",
    "\n",
    "# \"1988 - October 31: President Reagan signs the NSF Authorization Act of 1988, thereby authorizing the doubling of the NSF budget over the next five years.\"\n",
    "plt.axvline(x=1988)\n",
    "\n",
    "plt.legend(plots, continents_list, loc=2)\n",
    "plt.title(\"Specimens in GBIF by Collection Year and Continent\")\n",
    "plt.ylabel(\"Number of Specimen Records\")\n",
    "plt.xlabel(\"Year\")\n",
    "axes = plt.gca()\n",
    "axes.set_xlim([1815, 2020])\n",
    "axes.set_ylim([0, 1200000])\n",
    "fig = plt.gcf()\n",
    "fig.set_size_inches(12, 4)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "3 - PySpark Python3 Medium",
   "language": "python",
   "name": "pyspark3-med"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}