diff --git a/.ipynb_checkpoints/3a. Taxon Autocorrect with LSTM Autoencoders-checkpoint.ipynb b/.ipynb_checkpoints/3a. Taxon Autocorrect with LSTM Autoencoders-checkpoint.ipynb
new file mode 100644
index 0000000..35d545b
--- /dev/null
+++ b/.ipynb_checkpoints/3a. Taxon Autocorrect with LSTM Autoencoders-checkpoint.ipynb	
@@ -0,0 +1,1425 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# 3a. Autocorrecting Mispelt Taxon Names with Autoencoders\n",
+    "Given a list of taxon names, can we build an autocorrect model to autonomously fix erroneous records?"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 184,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "import matplotlib\n",
+    "import random\n",
+    "import string\n",
+    "from keras.models import Model\n",
+    "from keras.preprocessing import sequence\n",
+    "from keras.layers import Input, LSTM, Dense"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Exploring the dataset\n",
+    "We'll use the same dataset as last time; a publically available list of UK exports from 1975 - 2016. We'll only need the taxon names so we'll restrict our import to the taxon column."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 185,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>Taxon</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>Panthera onca</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>Varanus flavescens</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>Varanus griseus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>Branta ruficollis</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>Leopardus pardalis</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>6</th>\n",
+       "      <td>Leopardus wiedii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>7</th>\n",
+       "      <td>Diceros bicornis</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8</th>\n",
+       "      <td>Asarcornis scutulata</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9</th>\n",
+       "      <td>Branta sandvicensis</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>10</th>\n",
+       "      <td>Branta sandvicensis</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>11</th>\n",
+       "      <td>Cercopithecus diana</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>12</th>\n",
+       "      <td>Rucervus duvaucelii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>13</th>\n",
+       "      <td>Crocodylus siamensis</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>14</th>\n",
+       "      <td>Elephas maximus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>15</th>\n",
+       "      <td>Elephas maximus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>16</th>\n",
+       "      <td>Elephas maximus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>17</th>\n",
+       "      <td>Elephas maximus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>18</th>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>19</th>\n",
+       "      <td>Falco peregrinus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>20</th>\n",
+       "      <td>Acinonyx jubatus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>21</th>\n",
+       "      <td>Catopuma temminckii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>22</th>\n",
+       "      <td>Leopardus jacobitus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>23</th>\n",
+       "      <td>Leopardus pardalis mearnsi</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>24</th>\n",
+       "      <td>Panthera onca</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>25</th>\n",
+       "      <td>Panthera onca</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>26</th>\n",
+       "      <td>Panthera onca</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>27</th>\n",
+       "      <td>Panthera onca</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>28</th>\n",
+       "      <td>Panthera onca</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>29</th>\n",
+       "      <td>Panthera onca</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49339</th>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49340</th>\n",
+       "      <td>Mustela sibirica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49341</th>\n",
+       "      <td>Mustela sibirica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49342</th>\n",
+       "      <td>Mustela sibirica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49343</th>\n",
+       "      <td>Mustela sibirica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49344</th>\n",
+       "      <td>Mustela sibirica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49345</th>\n",
+       "      <td>Mustela sibirica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49346</th>\n",
+       "      <td>Mustela sibirica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49347</th>\n",
+       "      <td>Mustela sibirica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49348</th>\n",
+       "      <td>Mustela sibirica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49349</th>\n",
+       "      <td>Mustela sibirica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49350</th>\n",
+       "      <td>Mustela sibirica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49351</th>\n",
+       "      <td>Odobenus rosmarus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49352</th>\n",
+       "      <td>Odobenus rosmarus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49353</th>\n",
+       "      <td>Odobenus rosmarus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49354</th>\n",
+       "      <td>Odobenus rosmarus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49355</th>\n",
+       "      <td>Odobenus rosmarus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49356</th>\n",
+       "      <td>Odobenus rosmarus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49357</th>\n",
+       "      <td>Odobenus rosmarus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49358</th>\n",
+       "      <td>Odobenus rosmarus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49359</th>\n",
+       "      <td>Lodoicea maldivica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49360</th>\n",
+       "      <td>Pavo cristatus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49361</th>\n",
+       "      <td>Pavo cristatus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49362</th>\n",
+       "      <td>Pavo cristatus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49363</th>\n",
+       "      <td>Pavo cristatus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49364</th>\n",
+       "      <td>Pavo cristatus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49365</th>\n",
+       "      <td>Pavo cristatus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49366</th>\n",
+       "      <td>Pavo cristatus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49367</th>\n",
+       "      <td>Alligator mississippiensis</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49368</th>\n",
+       "      <td>Varanus salvator</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>49369 rows × 1 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                            Taxon\n",
+       "0               Equus przewalskii\n",
+       "1                   Panthera onca\n",
+       "2              Varanus flavescens\n",
+       "3                 Varanus griseus\n",
+       "4               Branta ruficollis\n",
+       "5              Leopardus pardalis\n",
+       "6                Leopardus wiedii\n",
+       "7                Diceros bicornis\n",
+       "8            Asarcornis scutulata\n",
+       "9             Branta sandvicensis\n",
+       "10            Branta sandvicensis\n",
+       "11            Cercopithecus diana\n",
+       "12            Rucervus duvaucelii\n",
+       "13           Crocodylus siamensis\n",
+       "14                Elephas maximus\n",
+       "15                Elephas maximus\n",
+       "16                Elephas maximus\n",
+       "17                Elephas maximus\n",
+       "18              Equus przewalskii\n",
+       "19               Falco peregrinus\n",
+       "20               Acinonyx jubatus\n",
+       "21            Catopuma temminckii\n",
+       "22            Leopardus jacobitus\n",
+       "23     Leopardus pardalis mearnsi\n",
+       "24                  Panthera onca\n",
+       "25                  Panthera onca\n",
+       "26                  Panthera onca\n",
+       "27                  Panthera onca\n",
+       "28                  Panthera onca\n",
+       "29                  Panthera onca\n",
+       "...                           ...\n",
+       "49339            Martes flavigula\n",
+       "49340            Mustela sibirica\n",
+       "49341            Mustela sibirica\n",
+       "49342            Mustela sibirica\n",
+       "49343            Mustela sibirica\n",
+       "49344            Mustela sibirica\n",
+       "49345            Mustela sibirica\n",
+       "49346            Mustela sibirica\n",
+       "49347            Mustela sibirica\n",
+       "49348            Mustela sibirica\n",
+       "49349            Mustela sibirica\n",
+       "49350            Mustela sibirica\n",
+       "49351           Odobenus rosmarus\n",
+       "49352           Odobenus rosmarus\n",
+       "49353           Odobenus rosmarus\n",
+       "49354           Odobenus rosmarus\n",
+       "49355           Odobenus rosmarus\n",
+       "49356           Odobenus rosmarus\n",
+       "49357           Odobenus rosmarus\n",
+       "49358           Odobenus rosmarus\n",
+       "49359          Lodoicea maldivica\n",
+       "49360              Pavo cristatus\n",
+       "49361              Pavo cristatus\n",
+       "49362              Pavo cristatus\n",
+       "49363              Pavo cristatus\n",
+       "49364              Pavo cristatus\n",
+       "49365              Pavo cristatus\n",
+       "49366              Pavo cristatus\n",
+       "49367  Alligator mississippiensis\n",
+       "49368            Varanus salvator\n",
+       "\n",
+       "[49369 rows x 1 columns]"
+      ]
+     },
+     "execution_count": 185,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "dataframe = pd.read_csv(\"data/goal_2_data.csv\", skipinitialspace=True, usecols=[\"Taxon\"])\n",
+    "\n",
+    "dataframe"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 186,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Loxodonta africana              3606\n",
+       "Python reticulatus              1523\n",
+       "Alligator mississippiensis      1297\n",
+       "Macaca fascicularis             1279\n",
+       "Varanus salvator                 972\n",
+       "Elephas maximus                  952\n",
+       "Cheloniidae spp.                 866\n",
+       "Varanus niloticus                744\n",
+       "Elephantidae spp.                716\n",
+       "Crocodylus niloticus             685\n",
+       "Psittacus erithacus              632\n",
+       "Crocodylus porosus               560\n",
+       "Caiman crocodilus crocodilus     524\n",
+       "Python bivittatus                501\n",
+       "Ptyas mucosus                    473\n",
+       "Chlorocebus aethiops             457\n",
+       "Falco peregrinus                 403\n",
+       "Eretmochelys imbricata           393\n",
+       "Dalbergia nigra                  354\n",
+       "Vicugna vicugna                  337\n",
+       "Panthera pardus                  325\n",
+       "Callithrix jacchus               323\n",
+       "Odobenus rosmarus                299\n",
+       "Falco rusticolus                 296\n",
+       "Panthera tigris                  283\n",
+       "Physeter macrocephalus           255\n",
+       "Hirudo medicinalis               249\n",
+       "Macaca mulatta                   232\n",
+       "Crocodylus novaeguineae          202\n",
+       "Leopardus pardalis               201\n",
+       "                                ... \n",
+       "Micrastur ruficollis               1\n",
+       "Hydnophora spp.                    1\n",
+       "Lycaste fulvescens                 1\n",
+       "Errinopora pourtalesii             1\n",
+       "Maihueniopsis darwinii             1\n",
+       "Porites divaricata                 1\n",
+       "Aloe trachyticola                  1\n",
+       "Polemaetus bellicosus              1\n",
+       "Sternbergia candida                1\n",
+       "Errinopora spp.                    1\n",
+       "Dracula tubeana                    1\n",
+       "Chinchilla lanigera                1\n",
+       "Peniocereus spp.                   1\n",
+       "Mesoplodon europaeus               1\n",
+       "Cypripedium yunnanense             1\n",
+       "Nectophrynoides minutus            1\n",
+       "Vidua paradisaea                   1\n",
+       "Bulbophyllum resupinatum           1\n",
+       "Turbinicarpus mandragora           1\n",
+       "Dalbergia retusa                   1\n",
+       "Pristis spp.                       1\n",
+       "Masdevallia andreettaeana          1\n",
+       "Dendrobium violaceum               1\n",
+       "Favites abdita                     1\n",
+       "Astrophytum myriostigma            1\n",
+       "Epiphyllum pumilum                 1\n",
+       "Pterostylis fischii                1\n",
+       "Colpophyllia amaranthus            1\n",
+       "Acineta chrysantha                 1\n",
+       "Anas spp.                          1\n",
+       "Name: Taxon, Length: 3422, dtype: int64"
+      ]
+     },
+     "execution_count": 186,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "name_distribution = dataframe[\"Taxon\"].value_counts()\n",
+    "name_distribution"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 187,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.axes._subplots.AxesSubplot at 0x12bfe1da0>"
+      ]
+     },
+     "execution_count": 187,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 720x720 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "name_distribution.head(50).plot.bar(figsize=(10, 10), title=\"Top 10 Taxon Names\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Obviously we can't use this for our training set as our model would disproportionatly learn to correct everything to \"Loxodonta africana\". We'll have to create a dataset of unique names..."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 188,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Total taxon names:  3422\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "array(['Equus przewalskii', 'Panthera onca', 'Varanus flavescens', ...,\n",
+       "       'Phaethornis longirostris', 'Mesoplodon stejnegeri',\n",
+       "       'Martes flavigula'], dtype=object)"
+      ]
+     },
+     "execution_count": 188,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "names = dataframe[\"Taxon\"].unique()\n",
+    "\n",
+    "print(\"Total taxon names: \", len(names))\n",
+    "names"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Now we have a list of 3422 names that we can train our model on, but we'll need to generate some fake spelling mistakes first, so let's write a function to do that..."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 189,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Lxodonta afkicana\n"
+     ]
+    }
+   ],
+   "source": [
+    "def spelling_mistake_generator(name):\n",
+    "    n = random.randint(0,4)\n",
+    "    if n == 0:\n",
+    "        return remove_letter(name)\n",
+    "    elif n == 1:\n",
+    "        return add_letter(name)\n",
+    "    elif n == 2:\n",
+    "        return swap_letters(name)\n",
+    "    elif n == 3:\n",
+    "        name = remove_letter(name)\n",
+    "        return add_letter(name)\n",
+    "    elif n == 4:\n",
+    "        return lowercase(name)\n",
+    "\n",
+    "def remove_letter(name):\n",
+    "    random_slice = random.randint(0, len(name))\n",
+    "    generated_name = name[:random_slice] + name[(random_slice + 1):]\n",
+    "    return generated_name\n",
+    "\n",
+    "def add_letter(name):\n",
+    "    random_slice = random.randint(0, len(name))\n",
+    "    random_letter = random.choice(string.ascii_letters)\n",
+    "    generated_name = name[:random_slice] + random_letter + name[(random_slice + 1):]\n",
+    "    return generated_name\n",
+    "\n",
+    "def swap_letters(name):\n",
+    "    random_slice = random.randint(0, len(name) - 2)\n",
+    "    generated_name = name[:random_slice] + reversed_string(name[random_slice:random_slice + 2]) + name[random_slice + 2:]\n",
+    "    return generated_name\n",
+    "    \n",
+    "def reversed_string(a_string):\n",
+    "    return a_string[::-1]\n",
+    "\n",
+    "def lowercase(name):\n",
+    "    return name.lower()\n",
+    "\n",
+    "print(spelling_mistake_generator(\"Loxodonta africana\"))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 190,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Loxodonta aLricana\n",
+      "Looxdonta africana\n",
+      "loxodonta africana\n",
+      "Loxodnta afrTcana\n",
+      "loxodonta africana\n",
+      "Loxodonta africMna\n",
+      "Loxodontaafricana\n",
+      "Loxodonta afrianaZ\n",
+      "Loxodonta africaq\n",
+      "loxodonta africana\n"
+     ]
+    }
+   ],
+   "source": [
+    "for i in range(10):\n",
+    "    print(spelling_mistake_generator(\"Loxodonta africana\"))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Generating our Rosetta Stone\n",
+    "\n",
+    "We'll use our new spelling mistake generator to generate a new dataset where we have the erroneous data in one column, and the correct data next to it. Since we get a different mistake each time we run the generator, we'll create 100 examples of each term..."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 191,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>Input</th>\n",
+       "      <th>Target</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>Equus przeawlskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>Equs przewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>Equusprzewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>Equus przealskiis</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>Equus prUewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>Equus prlewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>6</th>\n",
+       "      <td>equus przewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>7</th>\n",
+       "      <td>Equus przewlskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8</th>\n",
+       "      <td>EquusPprzewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9</th>\n",
+       "      <td>Eqlus przewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>10</th>\n",
+       "      <td>Equus pzrewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>11</th>\n",
+       "      <td>Equus przeawlskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>12</th>\n",
+       "      <td>Equusprzewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>13</th>\n",
+       "      <td>Equus przewlskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>14</th>\n",
+       "      <td>Equus przewalskji</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>15</th>\n",
+       "      <td>Equs przewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>16</th>\n",
+       "      <td>Equus przewlskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>17</th>\n",
+       "      <td>Equus pzrewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>18</th>\n",
+       "      <td>equus przewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>19</th>\n",
+       "      <td>Equus przewalsNii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>20</th>\n",
+       "      <td>Equus przewalsiki</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>21</th>\n",
+       "      <td>Equus przealskiu</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>22</th>\n",
+       "      <td>equus przewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>23</th>\n",
+       "      <td>Equus przeawlskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>24</th>\n",
+       "      <td>Equus przealWkii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>25</th>\n",
+       "      <td>equus przewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>26</th>\n",
+       "      <td>Equusp rzewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>27</th>\n",
+       "      <td>Equus prEewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>28</th>\n",
+       "      <td>equus przewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>29</th>\n",
+       "      <td>squus przewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342170</th>\n",
+       "      <td>martes flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342171</th>\n",
+       "      <td>Martes fwavgula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342172</th>\n",
+       "      <td>Martek flvigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342173</th>\n",
+       "      <td>Martes fwvigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342174</th>\n",
+       "      <td>Martes flaigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342175</th>\n",
+       "      <td>Martes flavigul</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342176</th>\n",
+       "      <td>martes flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342177</th>\n",
+       "      <td>martes flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342178</th>\n",
+       "      <td>Martes flaJigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342179</th>\n",
+       "      <td>partes flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342180</th>\n",
+       "      <td>Martes lfavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342181</th>\n",
+       "      <td>MarteP flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342182</th>\n",
+       "      <td>Martes flvibula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342183</th>\n",
+       "      <td>Martse flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342184</th>\n",
+       "      <td>Martes flaviula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342185</th>\n",
+       "      <td>Martes flavigual</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342186</th>\n",
+       "      <td>martes flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342187</th>\n",
+       "      <td>Marts flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342188</th>\n",
+       "      <td>Martesflavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342189</th>\n",
+       "      <td>Martes flNvigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342190</th>\n",
+       "      <td>martes flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342191</th>\n",
+       "      <td>martes flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342192</th>\n",
+       "      <td>Maxtes flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342193</th>\n",
+       "      <td>Martes flavigulaa</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342194</th>\n",
+       "      <td>Martes flavigXla</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342195</th>\n",
+       "      <td>aMrtes flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342196</th>\n",
+       "      <td>MartesLflavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342197</th>\n",
+       "      <td>Marets flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342198</th>\n",
+       "      <td>Martes flaviula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342199</th>\n",
+       "      <td>martes flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>342200 rows × 2 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                    Input             Target\n",
+       "0       Equus przeawlskii  Equus przewalskii\n",
+       "1        Equs przewalskii  Equus przewalskii\n",
+       "2        Equusprzewalskii  Equus przewalskii\n",
+       "3       Equus przealskiis  Equus przewalskii\n",
+       "4       Equus prUewalskii  Equus przewalskii\n",
+       "5       Equus prlewalskii  Equus przewalskii\n",
+       "6       equus przewalskii  Equus przewalskii\n",
+       "7        Equus przewlskii  Equus przewalskii\n",
+       "8       EquusPprzewalskii  Equus przewalskii\n",
+       "9       Eqlus przewalskii  Equus przewalskii\n",
+       "10      Equus pzrewalskii  Equus przewalskii\n",
+       "11      Equus przeawlskii  Equus przewalskii\n",
+       "12       Equusprzewalskii  Equus przewalskii\n",
+       "13       Equus przewlskii  Equus przewalskii\n",
+       "14      Equus przewalskji  Equus przewalskii\n",
+       "15       Equs przewalskii  Equus przewalskii\n",
+       "16       Equus przewlskii  Equus przewalskii\n",
+       "17      Equus pzrewalskii  Equus przewalskii\n",
+       "18      equus przewalskii  Equus przewalskii\n",
+       "19      Equus przewalsNii  Equus przewalskii\n",
+       "20      Equus przewalsiki  Equus przewalskii\n",
+       "21       Equus przealskiu  Equus przewalskii\n",
+       "22      equus przewalskii  Equus przewalskii\n",
+       "23      Equus przeawlskii  Equus przewalskii\n",
+       "24       Equus przealWkii  Equus przewalskii\n",
+       "25      equus przewalskii  Equus przewalskii\n",
+       "26      Equusp rzewalskii  Equus przewalskii\n",
+       "27      Equus prEewalskii  Equus przewalskii\n",
+       "28      equus przewalskii  Equus przewalskii\n",
+       "29      squus przewalskii  Equus przewalskii\n",
+       "...                   ...                ...\n",
+       "342170   martes flavigula   Martes flavigula\n",
+       "342171    Martes fwavgula   Martes flavigula\n",
+       "342172    Martek flvigula   Martes flavigula\n",
+       "342173    Martes fwvigula   Martes flavigula\n",
+       "342174    Martes flaigula   Martes flavigula\n",
+       "342175    Martes flavigul   Martes flavigula\n",
+       "342176   martes flavigula   Martes flavigula\n",
+       "342177   martes flavigula   Martes flavigula\n",
+       "342178   Martes flaJigula   Martes flavigula\n",
+       "342179   partes flavigula   Martes flavigula\n",
+       "342180   Martes lfavigula   Martes flavigula\n",
+       "342181   MarteP flavigula   Martes flavigula\n",
+       "342182    Martes flvibula   Martes flavigula\n",
+       "342183   Martse flavigula   Martes flavigula\n",
+       "342184    Martes flaviula   Martes flavigula\n",
+       "342185   Martes flavigual   Martes flavigula\n",
+       "342186   martes flavigula   Martes flavigula\n",
+       "342187    Marts flavigula   Martes flavigula\n",
+       "342188    Martesflavigula   Martes flavigula\n",
+       "342189   Martes flNvigula   Martes flavigula\n",
+       "342190   martes flavigula   Martes flavigula\n",
+       "342191   martes flavigula   Martes flavigula\n",
+       "342192   Maxtes flavigula   Martes flavigula\n",
+       "342193  Martes flavigulaa   Martes flavigula\n",
+       "342194   Martes flavigXla   Martes flavigula\n",
+       "342195   aMrtes flavigula   Martes flavigula\n",
+       "342196   MartesLflavigula   Martes flavigula\n",
+       "342197   Marets flavigula   Martes flavigula\n",
+       "342198    Martes flaviula   Martes flavigula\n",
+       "342199   martes flavigula   Martes flavigula\n",
+       "\n",
+       "[342200 rows x 2 columns]"
+      ]
+     },
+     "execution_count": 191,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "repeated_names = np.repeat(names, 100)\n",
+    "spelling_errors = [spelling_mistake_generator(s) for s in repeated_names]\n",
+    "\n",
+    "corpus = np.column_stack((spelling_errors, repeated_names))\n",
+    "\n",
+    "corpus = pd.DataFrame(corpus, columns=[\"Input\", \"Target\"])\n",
+    "corpus"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We've ended up with 342200 records of names with minor typos and formatting mistakes. (You can change the number of repetitions from 100 to 1000 if you need more data, but I've left this as 100 to save time. 1000 and upwards takes a little while to generate)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Vectorise our data for learning\n",
+    "\n",
+    "We'll need to encode our data from letters to number for our model to be able to deal with it. We'll take our table, and create two lists, one of all the characters in our input dataset, and one of all the characters in our target dataset. We'll also add a start and end character to our target data as this will be useful for our model to understand when to start and stop generating..."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 192,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "input_texts = []\n",
+    "target_texts = []\n",
+    "\n",
+    "start_character = '\\t'\n",
+    "end_character = '\\n'\n",
+    "\n",
+    "input_characters = set()\n",
+    "target_characters = set()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 193,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Takes in the input and target texts and adds their characters to the list of input and target characters\n",
+    "\n",
+    "def build_character_lists(input_text, target_text):\n",
+    "    for char in input_text:\n",
+    "        if char not in input_characters:\n",
+    "            input_characters.add(char)\n",
+    "    for char in target_text:\n",
+    "        if char not in target_characters:\n",
+    "            target_characters.add(char)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 194,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "for index, row in corpus.iterrows():\n",
+    "    input_text = row[\"Input\"]\n",
+    "    target_text = row[\"Target\"]\n",
+    "    target_text = start_character + target_text + end_character\n",
+    "    build_character_lists(input_text, target_text)\n",
+    "    input_texts.append(input_text)\n",
+    "    target_texts.append(target_text)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 195,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Number of samples: 342200\n",
+      "Number of unique input tokens: 55\n",
+      "Number of unique output tokens: 56\n",
+      "Max sequence length for inputs: 36\n",
+      "Max sequence length for outputs: 38\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "['\\t',\n",
+       " '\\n',\n",
+       " ' ',\n",
+       " '-',\n",
+       " '.',\n",
+       " 'A',\n",
+       " 'B',\n",
+       " 'C',\n",
+       " 'D',\n",
+       " 'E',\n",
+       " 'F',\n",
+       " 'G',\n",
+       " 'H',\n",
+       " 'I',\n",
+       " 'J',\n",
+       " 'K',\n",
+       " 'L',\n",
+       " 'M',\n",
+       " 'N',\n",
+       " 'O',\n",
+       " 'P',\n",
+       " 'Q',\n",
+       " 'R',\n",
+       " 'S',\n",
+       " 'T',\n",
+       " 'U',\n",
+       " 'V',\n",
+       " 'W',\n",
+       " 'X',\n",
+       " 'Z',\n",
+       " 'a',\n",
+       " 'b',\n",
+       " 'c',\n",
+       " 'd',\n",
+       " 'e',\n",
+       " 'f',\n",
+       " 'g',\n",
+       " 'h',\n",
+       " 'i',\n",
+       " 'j',\n",
+       " 'k',\n",
+       " 'l',\n",
+       " 'm',\n",
+       " 'n',\n",
+       " 'o',\n",
+       " 'p',\n",
+       " 'q',\n",
+       " 'r',\n",
+       " 's',\n",
+       " 't',\n",
+       " 'u',\n",
+       " 'v',\n",
+       " 'w',\n",
+       " 'x',\n",
+       " 'y',\n",
+       " 'z']"
+      ]
+     },
+     "execution_count": 195,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "input_characters   = sorted(list(input_characters))\n",
+    "target_characters  = sorted(list(target_characters))\n",
+    "num_encoder_tokens = len(input_characters)\n",
+    "num_decoder_tokens = len(target_characters)\n",
+    "max_encoder_seq_length = max([len(txt) for txt in input_texts])\n",
+    "max_decoder_seq_length = max([len(txt) for txt in target_texts])\n",
+    "\n",
+    "print('Number of samples:', len(input_texts))\n",
+    "print('Number of unique input tokens:', num_encoder_tokens)\n",
+    "print('Number of unique output tokens:', num_decoder_tokens)\n",
+    "print('Max sequence length for inputs:', max_encoder_seq_length)\n",
+    "print('Max sequence length for outputs:', max_decoder_seq_length)\n",
+    "\n",
+    "target_characters"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We'll create two dictionaries to help us get from our characters to numbers and back for both our input and target dataset"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 198,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "{' ': 0, '-': 1, '.': 2, 'A': 3, 'B': 4, 'C': 5, 'D': 6, 'E': 7, 'F': 8, 'G': 9, 'H': 10, 'I': 11, 'J': 12, 'K': 13, 'L': 14, 'M': 15, 'N': 16, 'O': 17, 'P': 18, 'Q': 19, 'R': 20, 'S': 21, 'T': 22, 'U': 23, 'V': 24, 'W': 25, 'X': 26, 'Y': 27, 'Z': 28, 'a': 29, 'b': 30, 'c': 31, 'd': 32, 'e': 33, 'f': 34, 'g': 35, 'h': 36, 'i': 37, 'j': 38, 'k': 39, 'l': 40, 'm': 41, 'n': 42, 'o': 43, 'p': 44, 'q': 45, 'r': 46, 's': 47, 't': 48, 'u': 49, 'v': 50, 'w': 51, 'x': 52, 'y': 53, 'z': 54}\n",
+      "{'\\t': 0, '\\n': 1, ' ': 2, '-': 3, '.': 4, 'A': 5, 'B': 6, 'C': 7, 'D': 8, 'E': 9, 'F': 10, 'G': 11, 'H': 12, 'I': 13, 'J': 14, 'K': 15, 'L': 16, 'M': 17, 'N': 18, 'O': 19, 'P': 20, 'Q': 21, 'R': 22, 'S': 23, 'T': 24, 'U': 25, 'V': 26, 'W': 27, 'X': 28, 'Z': 29, 'a': 30, 'b': 31, 'c': 32, 'd': 33, 'e': 34, 'f': 35, 'g': 36, 'h': 37, 'i': 38, 'j': 39, 'k': 40, 'l': 41, 'm': 42, 'n': 43, 'o': 44, 'p': 45, 'q': 46, 'r': 47, 's': 48, 't': 49, 'u': 50, 'v': 51, 'w': 52, 'x': 53, 'y': 54, 'z': 55}\n"
+     ]
+    }
+   ],
+   "source": [
+    "input_token_index = dict([(char, i) for i, char in enumerate(input_characters)])\n",
+    "target_token_index = dict([(char, i) for i, char in enumerate(target_characters)])\n",
+    "\n",
+    "print(input_token_index)\n",
+    "print(target_token_index)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 199,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# encoder_input_data is a 3D array of shape (num_pairs, max input seq length, num input characters)\n",
+    "encoder_input_data = np.zeros((len(input_texts), max_encoder_seq_length, num_encoder_tokens), dtype='float32')\n",
+    "\n",
+    "# decoder_input_data is a 3D array of shape (num_pairs, max target seq length, num target characters)\n",
+    "decoder_input_data = np.zeros((len(input_texts), max_decoder_seq_length, num_decoder_tokens), dtype='float32')\n",
+    "\n",
+    "# decoder_target_data is the same as decoder_input_data but offset by one timestep. decoder_target_data[:, t, :] will be the same as decoder_input_data[:, t + 1, :].\n",
+    "decoder_target_data = np.zeros((len(input_texts), max_decoder_seq_length, num_decoder_tokens), dtype='float32')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Building our model"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 200,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# i = training examples\n",
+    "# t = time step\n",
+    "# c = set the position representing the character to 1 (one hot encoded character)\n",
+    "\n",
+    "for i, (input_text, target_text) in enumerate(zip(input_texts, target_texts)):\n",
+    "    for t, char in enumerate(input_text):\n",
+    "        encoder_input_data[i, t, input_token_index[char]] = 1.\n",
+    "    for t, char in enumerate(target_text):\n",
+    "        # decoder_target_data is ahead of decoder_input_data by one timestep\n",
+    "        decoder_input_data[i, t, target_token_index[char]] = 1.\n",
+    "        if t > 0:\n",
+    "            # decoder_target_data will be ahead by one timestep\n",
+    "            # and will not include the start character.\n",
+    "            decoder_target_data[i, t - 1, target_token_index[char]] = 1."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 202,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "latent_dim = 256\n",
+    "\n",
+    "# Define an input sequence and process it.\n",
+    "encoder_inputs = Input(shape=(None, num_encoder_tokens))\n",
+    "encoder = LSTM(latent_dim, return_state=True)\n",
+    "encoder_outputs, state_h, state_c = encoder(encoder_inputs)\n",
+    "\n",
+    "# We discard `encoder_outputs` and only keep the states.\n",
+    "encoder_states = [state_h, state_c]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 203,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Set up the decoder, using `encoder_states` as initial state.\n",
+    "decoder_inputs = Input(shape=(None, num_decoder_tokens))\n",
+    "\n",
+    "# We set up our decoder to return full output sequences,\n",
+    "# and to return internal states as well. We don't use the\n",
+    "# return states in the training model, but we will use them in inference.\n",
+    "decoder_lstm = LSTM(latent_dim, return_sequences=True, return_state=True)\n",
+    "decoder_outputs, _, _ = decoder_lstm(decoder_inputs,\n",
+    "                                     initial_state=encoder_states)\n",
+    "\n",
+    "decoder_dense = Dense(num_decoder_tokens, activation='softmax')\n",
+    "decoder_outputs = decoder_dense(decoder_outputs)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 204,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Define the model that will turn\n",
+    "# `encoder_input_data` & `decoder_input_data` into `decoder_target_data`\n",
+    "model = Model([encoder_inputs, decoder_inputs], decoder_outputs)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 205,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Train on 273760 samples, validate on 68440 samples\n",
+      "Epoch 1/100\n",
+      "102336/273760 [==========>...................] - ETA: 6:04 - loss: 0.9042"
+     ]
+    },
+    {
+     "ename": "KeyboardInterrupt",
+     "evalue": "",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
+      "\u001b[0;32m<ipython-input-205-614d9c150bbf>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      7\u001b[0m           \u001b[0mbatch_size\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mbatch_size\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      8\u001b[0m           \u001b[0mepochs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mepochs\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 9\u001b[0;31m           validation_split=0.2)\n\u001b[0m\u001b[1;32m     10\u001b[0m \u001b[0;31m# Save model\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     11\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msave\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m's2s.h5'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.6/site-packages/keras/engine/training.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, **kwargs)\u001b[0m\n\u001b[1;32m   1703\u001b[0m                               \u001b[0minitial_epoch\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0minitial_epoch\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1704\u001b[0m                               \u001b[0msteps_per_epoch\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0msteps_per_epoch\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1705\u001b[0;31m                               validation_steps=validation_steps)\n\u001b[0m\u001b[1;32m   1706\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1707\u001b[0m     def evaluate(self, x=None, y=None,\n",
+      "\u001b[0;32m/usr/local/lib/python3.6/site-packages/keras/engine/training.py\u001b[0m in \u001b[0;36m_fit_loop\u001b[0;34m(self, f, ins, out_labels, batch_size, epochs, verbose, callbacks, val_f, val_ins, shuffle, callback_metrics, initial_epoch, steps_per_epoch, validation_steps)\u001b[0m\n\u001b[1;32m   1233\u001b[0m                         \u001b[0mins_batch\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mins_batch\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtoarray\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1234\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1235\u001b[0;31m                     \u001b[0mouts\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mins_batch\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1236\u001b[0m                     \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mouts\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlist\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1237\u001b[0m                         \u001b[0mouts\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mouts\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.6/site-packages/keras/backend/tensorflow_backend.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, inputs)\u001b[0m\n\u001b[1;32m   2476\u001b[0m         \u001b[0msession\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mget_session\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2477\u001b[0m         updated = session.run(fetches=fetches, feed_dict=feed_dict,\n\u001b[0;32m-> 2478\u001b[0;31m                               **self.session_kwargs)\n\u001b[0m\u001b[1;32m   2479\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mupdated\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2480\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36mrun\u001b[0;34m(self, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m    903\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    904\u001b[0m       result = self._run(None, fetches, feed_dict, options_ptr,\n\u001b[0;32m--> 905\u001b[0;31m                          run_metadata_ptr)\n\u001b[0m\u001b[1;32m    906\u001b[0m       \u001b[0;32mif\u001b[0m \u001b[0mrun_metadata\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    907\u001b[0m         \u001b[0mproto_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf_session\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTF_GetBuffer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrun_metadata_ptr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_run\u001b[0;34m(self, handle, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m   1135\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mfinal_fetches\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0mfinal_targets\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mhandle\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mfeed_dict_tensor\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1136\u001b[0m       results = self._do_run(handle, final_targets, final_fetches,\n\u001b[0;32m-> 1137\u001b[0;31m                              feed_dict_tensor, options, run_metadata)\n\u001b[0m\u001b[1;32m   1138\u001b[0m     \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1139\u001b[0m       \u001b[0mresults\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_do_run\u001b[0;34m(self, handle, target_list, fetch_list, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m   1353\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mhandle\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1354\u001b[0m       return self._do_call(_run_fn, self._session, feeds, fetches, targets,\n\u001b[0;32m-> 1355\u001b[0;31m                            options, run_metadata)\n\u001b[0m\u001b[1;32m   1356\u001b[0m     \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1357\u001b[0m       \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_do_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_prun_fn\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_session\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhandle\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeeds\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfetches\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_do_call\u001b[0;34m(self, fn, *args)\u001b[0m\n\u001b[1;32m   1359\u001b[0m   \u001b[0;32mdef\u001b[0m \u001b[0m_do_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1360\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1361\u001b[0;31m       \u001b[0;32mreturn\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1362\u001b[0m     \u001b[0;32mexcept\u001b[0m \u001b[0merrors\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mOpError\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1363\u001b[0m       \u001b[0mmessage\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcompat\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mas_text\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmessage\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_run_fn\u001b[0;34m(session, feed_dict, fetch_list, target_list, options, run_metadata)\u001b[0m\n\u001b[1;32m   1338\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1339\u001b[0m           return tf_session.TF_Run(session, options, feed_dict, fetch_list,\n\u001b[0;32m-> 1340\u001b[0;31m                                    target_list, status, run_metadata)\n\u001b[0m\u001b[1;32m   1341\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1342\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_prun_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msession\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhandle\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfetch_list\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
+     ]
+    }
+   ],
+   "source": [
+    "batch_size = 64  # Batch size for training.\n",
+    "epochs = 100 # Number of epochs to train for.\n",
+    "\n",
+    "# Run training\n",
+    "model.compile(optimizer='rmsprop', loss='categorical_crossentropy')\n",
+    "model.fit([encoder_input_data, decoder_input_data], decoder_target_data,\n",
+    "          batch_size=batch_size,\n",
+    "          epochs=epochs,\n",
+    "          validation_split=0.2)\n",
+    "# Save model\n",
+    "model.save('s2s.h5')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.6.5"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/2. Learning about Trade Data.ipynb b/2. Learning about Trade Data.ipynb
index fff7aaa..b92c302 100644
--- a/2. Learning about Trade Data.ipynb	
+++ b/2. Learning about Trade Data.ipynb	
@@ -4024,7 +4024,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.6.4"
+   "version": "3.6.5"
   }
  },
  "nbformat": 4,
diff --git a/3a. Taxon Autocorrect with LSTM Autoencoders.ipynb b/3a. Taxon Autocorrect with LSTM Autoencoders.ipynb
new file mode 100644
index 0000000..a6dc5f4
--- /dev/null
+++ b/3a. Taxon Autocorrect with LSTM Autoencoders.ipynb	
@@ -0,0 +1,1447 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# 3a. Autocorrecting Mispelt Taxon Names with Autoencoders\n",
+    "Given a list of taxon names, can we build an autocorrect model to autonomously fix erroneous records?"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 184,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "import matplotlib\n",
+    "import random\n",
+    "import string\n",
+    "from keras.models import Model\n",
+    "from keras.preprocessing import sequence\n",
+    "from keras.layers import Input, LSTM, Dense"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Exploring the dataset\n",
+    "We'll use the same dataset as last time; a publically available list of UK exports from 1975 - 2016. We'll only need the taxon names so we'll restrict our import to the taxon column."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 185,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>Taxon</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>Panthera onca</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>Varanus flavescens</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>Varanus griseus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>Branta ruficollis</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>Leopardus pardalis</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>6</th>\n",
+       "      <td>Leopardus wiedii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>7</th>\n",
+       "      <td>Diceros bicornis</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8</th>\n",
+       "      <td>Asarcornis scutulata</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9</th>\n",
+       "      <td>Branta sandvicensis</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>10</th>\n",
+       "      <td>Branta sandvicensis</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>11</th>\n",
+       "      <td>Cercopithecus diana</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>12</th>\n",
+       "      <td>Rucervus duvaucelii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>13</th>\n",
+       "      <td>Crocodylus siamensis</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>14</th>\n",
+       "      <td>Elephas maximus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>15</th>\n",
+       "      <td>Elephas maximus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>16</th>\n",
+       "      <td>Elephas maximus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>17</th>\n",
+       "      <td>Elephas maximus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>18</th>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>19</th>\n",
+       "      <td>Falco peregrinus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>20</th>\n",
+       "      <td>Acinonyx jubatus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>21</th>\n",
+       "      <td>Catopuma temminckii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>22</th>\n",
+       "      <td>Leopardus jacobitus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>23</th>\n",
+       "      <td>Leopardus pardalis mearnsi</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>24</th>\n",
+       "      <td>Panthera onca</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>25</th>\n",
+       "      <td>Panthera onca</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>26</th>\n",
+       "      <td>Panthera onca</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>27</th>\n",
+       "      <td>Panthera onca</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>28</th>\n",
+       "      <td>Panthera onca</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>29</th>\n",
+       "      <td>Panthera onca</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49339</th>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49340</th>\n",
+       "      <td>Mustela sibirica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49341</th>\n",
+       "      <td>Mustela sibirica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49342</th>\n",
+       "      <td>Mustela sibirica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49343</th>\n",
+       "      <td>Mustela sibirica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49344</th>\n",
+       "      <td>Mustela sibirica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49345</th>\n",
+       "      <td>Mustela sibirica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49346</th>\n",
+       "      <td>Mustela sibirica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49347</th>\n",
+       "      <td>Mustela sibirica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49348</th>\n",
+       "      <td>Mustela sibirica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49349</th>\n",
+       "      <td>Mustela sibirica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49350</th>\n",
+       "      <td>Mustela sibirica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49351</th>\n",
+       "      <td>Odobenus rosmarus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49352</th>\n",
+       "      <td>Odobenus rosmarus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49353</th>\n",
+       "      <td>Odobenus rosmarus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49354</th>\n",
+       "      <td>Odobenus rosmarus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49355</th>\n",
+       "      <td>Odobenus rosmarus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49356</th>\n",
+       "      <td>Odobenus rosmarus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49357</th>\n",
+       "      <td>Odobenus rosmarus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49358</th>\n",
+       "      <td>Odobenus rosmarus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49359</th>\n",
+       "      <td>Lodoicea maldivica</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49360</th>\n",
+       "      <td>Pavo cristatus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49361</th>\n",
+       "      <td>Pavo cristatus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49362</th>\n",
+       "      <td>Pavo cristatus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49363</th>\n",
+       "      <td>Pavo cristatus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49364</th>\n",
+       "      <td>Pavo cristatus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49365</th>\n",
+       "      <td>Pavo cristatus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49366</th>\n",
+       "      <td>Pavo cristatus</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49367</th>\n",
+       "      <td>Alligator mississippiensis</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49368</th>\n",
+       "      <td>Varanus salvator</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>49369 rows × 1 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                            Taxon\n",
+       "0               Equus przewalskii\n",
+       "1                   Panthera onca\n",
+       "2              Varanus flavescens\n",
+       "3                 Varanus griseus\n",
+       "4               Branta ruficollis\n",
+       "5              Leopardus pardalis\n",
+       "6                Leopardus wiedii\n",
+       "7                Diceros bicornis\n",
+       "8            Asarcornis scutulata\n",
+       "9             Branta sandvicensis\n",
+       "10            Branta sandvicensis\n",
+       "11            Cercopithecus diana\n",
+       "12            Rucervus duvaucelii\n",
+       "13           Crocodylus siamensis\n",
+       "14                Elephas maximus\n",
+       "15                Elephas maximus\n",
+       "16                Elephas maximus\n",
+       "17                Elephas maximus\n",
+       "18              Equus przewalskii\n",
+       "19               Falco peregrinus\n",
+       "20               Acinonyx jubatus\n",
+       "21            Catopuma temminckii\n",
+       "22            Leopardus jacobitus\n",
+       "23     Leopardus pardalis mearnsi\n",
+       "24                  Panthera onca\n",
+       "25                  Panthera onca\n",
+       "26                  Panthera onca\n",
+       "27                  Panthera onca\n",
+       "28                  Panthera onca\n",
+       "29                  Panthera onca\n",
+       "...                           ...\n",
+       "49339            Martes flavigula\n",
+       "49340            Mustela sibirica\n",
+       "49341            Mustela sibirica\n",
+       "49342            Mustela sibirica\n",
+       "49343            Mustela sibirica\n",
+       "49344            Mustela sibirica\n",
+       "49345            Mustela sibirica\n",
+       "49346            Mustela sibirica\n",
+       "49347            Mustela sibirica\n",
+       "49348            Mustela sibirica\n",
+       "49349            Mustela sibirica\n",
+       "49350            Mustela sibirica\n",
+       "49351           Odobenus rosmarus\n",
+       "49352           Odobenus rosmarus\n",
+       "49353           Odobenus rosmarus\n",
+       "49354           Odobenus rosmarus\n",
+       "49355           Odobenus rosmarus\n",
+       "49356           Odobenus rosmarus\n",
+       "49357           Odobenus rosmarus\n",
+       "49358           Odobenus rosmarus\n",
+       "49359          Lodoicea maldivica\n",
+       "49360              Pavo cristatus\n",
+       "49361              Pavo cristatus\n",
+       "49362              Pavo cristatus\n",
+       "49363              Pavo cristatus\n",
+       "49364              Pavo cristatus\n",
+       "49365              Pavo cristatus\n",
+       "49366              Pavo cristatus\n",
+       "49367  Alligator mississippiensis\n",
+       "49368            Varanus salvator\n",
+       "\n",
+       "[49369 rows x 1 columns]"
+      ]
+     },
+     "execution_count": 185,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "dataframe = pd.read_csv(\"data/goal_2_data.csv\", skipinitialspace=True, usecols=[\"Taxon\"])\n",
+    "\n",
+    "dataframe"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 186,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Loxodonta africana              3606\n",
+       "Python reticulatus              1523\n",
+       "Alligator mississippiensis      1297\n",
+       "Macaca fascicularis             1279\n",
+       "Varanus salvator                 972\n",
+       "Elephas maximus                  952\n",
+       "Cheloniidae spp.                 866\n",
+       "Varanus niloticus                744\n",
+       "Elephantidae spp.                716\n",
+       "Crocodylus niloticus             685\n",
+       "Psittacus erithacus              632\n",
+       "Crocodylus porosus               560\n",
+       "Caiman crocodilus crocodilus     524\n",
+       "Python bivittatus                501\n",
+       "Ptyas mucosus                    473\n",
+       "Chlorocebus aethiops             457\n",
+       "Falco peregrinus                 403\n",
+       "Eretmochelys imbricata           393\n",
+       "Dalbergia nigra                  354\n",
+       "Vicugna vicugna                  337\n",
+       "Panthera pardus                  325\n",
+       "Callithrix jacchus               323\n",
+       "Odobenus rosmarus                299\n",
+       "Falco rusticolus                 296\n",
+       "Panthera tigris                  283\n",
+       "Physeter macrocephalus           255\n",
+       "Hirudo medicinalis               249\n",
+       "Macaca mulatta                   232\n",
+       "Crocodylus novaeguineae          202\n",
+       "Leopardus pardalis               201\n",
+       "                                ... \n",
+       "Micrastur ruficollis               1\n",
+       "Hydnophora spp.                    1\n",
+       "Lycaste fulvescens                 1\n",
+       "Errinopora pourtalesii             1\n",
+       "Maihueniopsis darwinii             1\n",
+       "Porites divaricata                 1\n",
+       "Aloe trachyticola                  1\n",
+       "Polemaetus bellicosus              1\n",
+       "Sternbergia candida                1\n",
+       "Errinopora spp.                    1\n",
+       "Dracula tubeana                    1\n",
+       "Chinchilla lanigera                1\n",
+       "Peniocereus spp.                   1\n",
+       "Mesoplodon europaeus               1\n",
+       "Cypripedium yunnanense             1\n",
+       "Nectophrynoides minutus            1\n",
+       "Vidua paradisaea                   1\n",
+       "Bulbophyllum resupinatum           1\n",
+       "Turbinicarpus mandragora           1\n",
+       "Dalbergia retusa                   1\n",
+       "Pristis spp.                       1\n",
+       "Masdevallia andreettaeana          1\n",
+       "Dendrobium violaceum               1\n",
+       "Favites abdita                     1\n",
+       "Astrophytum myriostigma            1\n",
+       "Epiphyllum pumilum                 1\n",
+       "Pterostylis fischii                1\n",
+       "Colpophyllia amaranthus            1\n",
+       "Acineta chrysantha                 1\n",
+       "Anas spp.                          1\n",
+       "Name: Taxon, Length: 3422, dtype: int64"
+      ]
+     },
+     "execution_count": 186,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "name_distribution = dataframe[\"Taxon\"].value_counts()\n",
+    "name_distribution"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 187,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.axes._subplots.AxesSubplot at 0x12bfe1da0>"
+      ]
+     },
+     "execution_count": 187,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 720x720 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "name_distribution.head(50).plot.bar(figsize=(10, 10), title=\"Top 10 Taxon Names\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Obviously we can't use this for our training set as our model would disproportionatly learn to correct everything to \"Loxodonta africana\". We'll have to create a dataset of unique names..."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 188,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Total taxon names:  3422\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "array(['Equus przewalskii', 'Panthera onca', 'Varanus flavescens', ...,\n",
+       "       'Phaethornis longirostris', 'Mesoplodon stejnegeri',\n",
+       "       'Martes flavigula'], dtype=object)"
+      ]
+     },
+     "execution_count": 188,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "names = dataframe[\"Taxon\"].unique()\n",
+    "\n",
+    "print(\"Total taxon names: \", len(names))\n",
+    "names"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Now we have a list of 3422 names that we can train our model on, but we'll need to generate some fake spelling mistakes first, so let's write a function to do that..."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 189,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Lxodonta afkicana\n"
+     ]
+    }
+   ],
+   "source": [
+    "def spelling_mistake_generator(name):\n",
+    "    n = random.randint(0,4)\n",
+    "    if n == 0:\n",
+    "        return remove_letter(name)\n",
+    "    elif n == 1:\n",
+    "        return add_letter(name)\n",
+    "    elif n == 2:\n",
+    "        return swap_letters(name)\n",
+    "    elif n == 3:\n",
+    "        name = remove_letter(name)\n",
+    "        return add_letter(name)\n",
+    "    elif n == 4:\n",
+    "        return lowercase(name)\n",
+    "\n",
+    "def remove_letter(name):\n",
+    "    random_slice = random.randint(0, len(name))\n",
+    "    generated_name = name[:random_slice] + name[(random_slice + 1):]\n",
+    "    return generated_name\n",
+    "\n",
+    "def add_letter(name):\n",
+    "    random_slice = random.randint(0, len(name))\n",
+    "    random_letter = random.choice(string.ascii_letters)\n",
+    "    generated_name = name[:random_slice] + random_letter + name[(random_slice + 1):]\n",
+    "    return generated_name\n",
+    "\n",
+    "def swap_letters(name):\n",
+    "    random_slice = random.randint(0, len(name) - 2)\n",
+    "    generated_name = name[:random_slice] + reversed_string(name[random_slice:random_slice + 2]) + name[random_slice + 2:]\n",
+    "    return generated_name\n",
+    "    \n",
+    "def reversed_string(a_string):\n",
+    "    return a_string[::-1]\n",
+    "\n",
+    "def lowercase(name):\n",
+    "    return name.lower()\n",
+    "\n",
+    "print(spelling_mistake_generator(\"Loxodonta africana\"))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 190,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Loxodonta aLricana\n",
+      "Looxdonta africana\n",
+      "loxodonta africana\n",
+      "Loxodnta afrTcana\n",
+      "loxodonta africana\n",
+      "Loxodonta africMna\n",
+      "Loxodontaafricana\n",
+      "Loxodonta afrianaZ\n",
+      "Loxodonta africaq\n",
+      "loxodonta africana\n"
+     ]
+    }
+   ],
+   "source": [
+    "for i in range(10):\n",
+    "    print(spelling_mistake_generator(\"Loxodonta africana\"))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Generating our Rosetta Stone\n",
+    "\n",
+    "We'll use our new spelling mistake generator to generate a new dataset where we have the erroneous data in one column, and the correct data next to it. Since we get a different mistake each time we run the generator, we'll create 100 examples of each term..."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 191,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>Input</th>\n",
+       "      <th>Target</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>Equus przeawlskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>Equs przewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>Equusprzewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>Equus przealskiis</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>Equus prUewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>Equus prlewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>6</th>\n",
+       "      <td>equus przewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>7</th>\n",
+       "      <td>Equus przewlskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8</th>\n",
+       "      <td>EquusPprzewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9</th>\n",
+       "      <td>Eqlus przewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>10</th>\n",
+       "      <td>Equus pzrewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>11</th>\n",
+       "      <td>Equus przeawlskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>12</th>\n",
+       "      <td>Equusprzewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>13</th>\n",
+       "      <td>Equus przewlskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>14</th>\n",
+       "      <td>Equus przewalskji</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>15</th>\n",
+       "      <td>Equs przewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>16</th>\n",
+       "      <td>Equus przewlskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>17</th>\n",
+       "      <td>Equus pzrewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>18</th>\n",
+       "      <td>equus przewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>19</th>\n",
+       "      <td>Equus przewalsNii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>20</th>\n",
+       "      <td>Equus przewalsiki</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>21</th>\n",
+       "      <td>Equus przealskiu</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>22</th>\n",
+       "      <td>equus przewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>23</th>\n",
+       "      <td>Equus przeawlskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>24</th>\n",
+       "      <td>Equus przealWkii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>25</th>\n",
+       "      <td>equus przewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>26</th>\n",
+       "      <td>Equusp rzewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>27</th>\n",
+       "      <td>Equus prEewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>28</th>\n",
+       "      <td>equus przewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>29</th>\n",
+       "      <td>squus przewalskii</td>\n",
+       "      <td>Equus przewalskii</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342170</th>\n",
+       "      <td>martes flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342171</th>\n",
+       "      <td>Martes fwavgula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342172</th>\n",
+       "      <td>Martek flvigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342173</th>\n",
+       "      <td>Martes fwvigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342174</th>\n",
+       "      <td>Martes flaigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342175</th>\n",
+       "      <td>Martes flavigul</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342176</th>\n",
+       "      <td>martes flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342177</th>\n",
+       "      <td>martes flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342178</th>\n",
+       "      <td>Martes flaJigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342179</th>\n",
+       "      <td>partes flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342180</th>\n",
+       "      <td>Martes lfavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342181</th>\n",
+       "      <td>MarteP flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342182</th>\n",
+       "      <td>Martes flvibula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342183</th>\n",
+       "      <td>Martse flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342184</th>\n",
+       "      <td>Martes flaviula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342185</th>\n",
+       "      <td>Martes flavigual</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342186</th>\n",
+       "      <td>martes flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342187</th>\n",
+       "      <td>Marts flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342188</th>\n",
+       "      <td>Martesflavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342189</th>\n",
+       "      <td>Martes flNvigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342190</th>\n",
+       "      <td>martes flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342191</th>\n",
+       "      <td>martes flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342192</th>\n",
+       "      <td>Maxtes flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342193</th>\n",
+       "      <td>Martes flavigulaa</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342194</th>\n",
+       "      <td>Martes flavigXla</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342195</th>\n",
+       "      <td>aMrtes flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342196</th>\n",
+       "      <td>MartesLflavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342197</th>\n",
+       "      <td>Marets flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342198</th>\n",
+       "      <td>Martes flaviula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>342199</th>\n",
+       "      <td>martes flavigula</td>\n",
+       "      <td>Martes flavigula</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>342200 rows × 2 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                    Input             Target\n",
+       "0       Equus przeawlskii  Equus przewalskii\n",
+       "1        Equs przewalskii  Equus przewalskii\n",
+       "2        Equusprzewalskii  Equus przewalskii\n",
+       "3       Equus przealskiis  Equus przewalskii\n",
+       "4       Equus prUewalskii  Equus przewalskii\n",
+       "5       Equus prlewalskii  Equus przewalskii\n",
+       "6       equus przewalskii  Equus przewalskii\n",
+       "7        Equus przewlskii  Equus przewalskii\n",
+       "8       EquusPprzewalskii  Equus przewalskii\n",
+       "9       Eqlus przewalskii  Equus przewalskii\n",
+       "10      Equus pzrewalskii  Equus przewalskii\n",
+       "11      Equus przeawlskii  Equus przewalskii\n",
+       "12       Equusprzewalskii  Equus przewalskii\n",
+       "13       Equus przewlskii  Equus przewalskii\n",
+       "14      Equus przewalskji  Equus przewalskii\n",
+       "15       Equs przewalskii  Equus przewalskii\n",
+       "16       Equus przewlskii  Equus przewalskii\n",
+       "17      Equus pzrewalskii  Equus przewalskii\n",
+       "18      equus przewalskii  Equus przewalskii\n",
+       "19      Equus przewalsNii  Equus przewalskii\n",
+       "20      Equus przewalsiki  Equus przewalskii\n",
+       "21       Equus przealskiu  Equus przewalskii\n",
+       "22      equus przewalskii  Equus przewalskii\n",
+       "23      Equus przeawlskii  Equus przewalskii\n",
+       "24       Equus przealWkii  Equus przewalskii\n",
+       "25      equus przewalskii  Equus przewalskii\n",
+       "26      Equusp rzewalskii  Equus przewalskii\n",
+       "27      Equus prEewalskii  Equus przewalskii\n",
+       "28      equus przewalskii  Equus przewalskii\n",
+       "29      squus przewalskii  Equus przewalskii\n",
+       "...                   ...                ...\n",
+       "342170   martes flavigula   Martes flavigula\n",
+       "342171    Martes fwavgula   Martes flavigula\n",
+       "342172    Martek flvigula   Martes flavigula\n",
+       "342173    Martes fwvigula   Martes flavigula\n",
+       "342174    Martes flaigula   Martes flavigula\n",
+       "342175    Martes flavigul   Martes flavigula\n",
+       "342176   martes flavigula   Martes flavigula\n",
+       "342177   martes flavigula   Martes flavigula\n",
+       "342178   Martes flaJigula   Martes flavigula\n",
+       "342179   partes flavigula   Martes flavigula\n",
+       "342180   Martes lfavigula   Martes flavigula\n",
+       "342181   MarteP flavigula   Martes flavigula\n",
+       "342182    Martes flvibula   Martes flavigula\n",
+       "342183   Martse flavigula   Martes flavigula\n",
+       "342184    Martes flaviula   Martes flavigula\n",
+       "342185   Martes flavigual   Martes flavigula\n",
+       "342186   martes flavigula   Martes flavigula\n",
+       "342187    Marts flavigula   Martes flavigula\n",
+       "342188    Martesflavigula   Martes flavigula\n",
+       "342189   Martes flNvigula   Martes flavigula\n",
+       "342190   martes flavigula   Martes flavigula\n",
+       "342191   martes flavigula   Martes flavigula\n",
+       "342192   Maxtes flavigula   Martes flavigula\n",
+       "342193  Martes flavigulaa   Martes flavigula\n",
+       "342194   Martes flavigXla   Martes flavigula\n",
+       "342195   aMrtes flavigula   Martes flavigula\n",
+       "342196   MartesLflavigula   Martes flavigula\n",
+       "342197   Marets flavigula   Martes flavigula\n",
+       "342198    Martes flaviula   Martes flavigula\n",
+       "342199   martes flavigula   Martes flavigula\n",
+       "\n",
+       "[342200 rows x 2 columns]"
+      ]
+     },
+     "execution_count": 191,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "repeated_names = np.repeat(names, 100)\n",
+    "spelling_errors = [spelling_mistake_generator(s) for s in repeated_names]\n",
+    "\n",
+    "corpus = np.column_stack((spelling_errors, repeated_names))\n",
+    "\n",
+    "corpus = pd.DataFrame(corpus, columns=[\"Input\", \"Target\"])\n",
+    "corpus"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We've ended up with 342200 records of names with minor typos and formatting mistakes. (You can change the number of repetitions from 100 to 1000 if you need more data, but I've left this as 100 to save time. 1000 and upwards takes a little while to generate)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Vectorise our data for learning\n",
+    "\n",
+    "We'll need to encode our data from letters to number for our model to be able to deal with it. We'll take our table, and create two lists, one of all the characters in our input dataset, and one of all the characters in our target dataset. We can use these later on to one hot encode our characters to vectors before we feed them to our model. We'll also add a start and end character to our target data as this will be useful for our model to understand when to start and stop generating..."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 192,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "input_texts = []\n",
+    "target_texts = []\n",
+    "\n",
+    "start_character = '\\t'\n",
+    "end_character = '\\n'\n",
+    "\n",
+    "input_characters = set()\n",
+    "target_characters = set()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 193,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Takes in the input and target texts and adds their characters to the list of input and target characters\n",
+    "\n",
+    "def build_character_lists(input_text, target_text):\n",
+    "    for char in input_text:\n",
+    "        if char not in input_characters:\n",
+    "            input_characters.add(char)\n",
+    "    for char in target_text:\n",
+    "        if char not in target_characters:\n",
+    "            target_characters.add(char)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 194,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "for index, row in corpus.iterrows():\n",
+    "    input_text = row[\"Input\"]\n",
+    "    target_text = row[\"Target\"]\n",
+    "    target_text = start_character + target_text + end_character\n",
+    "    build_character_lists(input_text, target_text)\n",
+    "    input_texts.append(input_text)\n",
+    "    target_texts.append(target_text)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 195,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Number of samples: 342200\n",
+      "Number of unique input tokens: 55\n",
+      "Number of unique output tokens: 56\n",
+      "Max sequence length for inputs: 36\n",
+      "Max sequence length for outputs: 38\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "['\\t',\n",
+       " '\\n',\n",
+       " ' ',\n",
+       " '-',\n",
+       " '.',\n",
+       " 'A',\n",
+       " 'B',\n",
+       " 'C',\n",
+       " 'D',\n",
+       " 'E',\n",
+       " 'F',\n",
+       " 'G',\n",
+       " 'H',\n",
+       " 'I',\n",
+       " 'J',\n",
+       " 'K',\n",
+       " 'L',\n",
+       " 'M',\n",
+       " 'N',\n",
+       " 'O',\n",
+       " 'P',\n",
+       " 'Q',\n",
+       " 'R',\n",
+       " 'S',\n",
+       " 'T',\n",
+       " 'U',\n",
+       " 'V',\n",
+       " 'W',\n",
+       " 'X',\n",
+       " 'Z',\n",
+       " 'a',\n",
+       " 'b',\n",
+       " 'c',\n",
+       " 'd',\n",
+       " 'e',\n",
+       " 'f',\n",
+       " 'g',\n",
+       " 'h',\n",
+       " 'i',\n",
+       " 'j',\n",
+       " 'k',\n",
+       " 'l',\n",
+       " 'm',\n",
+       " 'n',\n",
+       " 'o',\n",
+       " 'p',\n",
+       " 'q',\n",
+       " 'r',\n",
+       " 's',\n",
+       " 't',\n",
+       " 'u',\n",
+       " 'v',\n",
+       " 'w',\n",
+       " 'x',\n",
+       " 'y',\n",
+       " 'z']"
+      ]
+     },
+     "execution_count": 195,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "input_characters   = sorted(list(input_characters))\n",
+    "target_characters  = sorted(list(target_characters))\n",
+    "num_encoder_tokens = len(input_characters)\n",
+    "num_decoder_tokens = len(target_characters)\n",
+    "max_encoder_seq_length = max([len(txt) for txt in input_texts])\n",
+    "max_decoder_seq_length = max([len(txt) for txt in target_texts])\n",
+    "\n",
+    "print('Number of samples:', len(input_texts))\n",
+    "print('Number of unique input tokens:', num_encoder_tokens)\n",
+    "print('Number of unique output tokens:', num_decoder_tokens)\n",
+    "print('Max sequence length for inputs:', max_encoder_seq_length)\n",
+    "print('Max sequence length for outputs:', max_decoder_seq_length)\n",
+    "\n",
+    "target_characters"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We'll create two dictionaries to help us get from our characters to numbers and back for both our input and target dataset"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 198,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "{' ': 0, '-': 1, '.': 2, 'A': 3, 'B': 4, 'C': 5, 'D': 6, 'E': 7, 'F': 8, 'G': 9, 'H': 10, 'I': 11, 'J': 12, 'K': 13, 'L': 14, 'M': 15, 'N': 16, 'O': 17, 'P': 18, 'Q': 19, 'R': 20, 'S': 21, 'T': 22, 'U': 23, 'V': 24, 'W': 25, 'X': 26, 'Y': 27, 'Z': 28, 'a': 29, 'b': 30, 'c': 31, 'd': 32, 'e': 33, 'f': 34, 'g': 35, 'h': 36, 'i': 37, 'j': 38, 'k': 39, 'l': 40, 'm': 41, 'n': 42, 'o': 43, 'p': 44, 'q': 45, 'r': 46, 's': 47, 't': 48, 'u': 49, 'v': 50, 'w': 51, 'x': 52, 'y': 53, 'z': 54}\n",
+      "{'\\t': 0, '\\n': 1, ' ': 2, '-': 3, '.': 4, 'A': 5, 'B': 6, 'C': 7, 'D': 8, 'E': 9, 'F': 10, 'G': 11, 'H': 12, 'I': 13, 'J': 14, 'K': 15, 'L': 16, 'M': 17, 'N': 18, 'O': 19, 'P': 20, 'Q': 21, 'R': 22, 'S': 23, 'T': 24, 'U': 25, 'V': 26, 'W': 27, 'X': 28, 'Z': 29, 'a': 30, 'b': 31, 'c': 32, 'd': 33, 'e': 34, 'f': 35, 'g': 36, 'h': 37, 'i': 38, 'j': 39, 'k': 40, 'l': 41, 'm': 42, 'n': 43, 'o': 44, 'p': 45, 'q': 46, 'r': 47, 's': 48, 't': 49, 'u': 50, 'v': 51, 'w': 52, 'x': 53, 'y': 54, 'z': 55}\n"
+     ]
+    }
+   ],
+   "source": [
+    "input_token_index = dict([(char, i) for i, char in enumerate(input_characters)])\n",
+    "target_token_index = dict([(char, i) for i, char in enumerate(target_characters)])\n",
+    "\n",
+    "print(input_token_index)\n",
+    "print(target_token_index)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 199,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# encoder_input_data is a 3D array of shape (num_pairs, max input seq length, num input characters)\n",
+    "encoder_input_data = np.zeros((len(input_texts), max_encoder_seq_length, num_encoder_tokens), dtype='float32')\n",
+    "\n",
+    "# decoder_input_data is a 3D array of shape (num_pairs, max target seq length, num target characters)\n",
+    "decoder_input_data = np.zeros((len(input_texts), max_decoder_seq_length, num_decoder_tokens), dtype='float32')\n",
+    "\n",
+    "# decoder_target_data is the same as decoder_input_data but offset by one timestep. decoder_target_data[:, t, :] will be the same as decoder_input_data[:, t + 1, :].\n",
+    "decoder_target_data = np.zeros((len(input_texts), max_decoder_seq_length, num_decoder_tokens), dtype='float32')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Building our model"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 200,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# i = training examples\n",
+    "# t = time step\n",
+    "# c = set the position representing the character to 1 (one hot encoded character)\n",
+    "\n",
+    "for i, (input_text, target_text) in enumerate(zip(input_texts, target_texts)):\n",
+    "    for t, char in enumerate(input_text):\n",
+    "        encoder_input_data[i, t, input_token_index[char]] = 1.\n",
+    "    for t, char in enumerate(target_text):\n",
+    "        # decoder_target_data is ahead of decoder_input_data by one timestep\n",
+    "        decoder_input_data[i, t, target_token_index[char]] = 1.\n",
+    "        if t > 0:\n",
+    "            # decoder_target_data will be ahead by one timestep\n",
+    "            # and will not include the start character.\n",
+    "            decoder_target_data[i, t - 1, target_token_index[char]] = 1."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 202,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "latent_dim = 256\n",
+    "\n",
+    "# Define an input sequence and process it.\n",
+    "encoder_inputs = Input(shape=(None, num_encoder_tokens))\n",
+    "encoder = LSTM(latent_dim, return_state=True)\n",
+    "encoder_outputs, state_h, state_c = encoder(encoder_inputs)\n",
+    "\n",
+    "# We discard `encoder_outputs` and only keep the states.\n",
+    "encoder_states = [state_h, state_c]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 203,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Set up the decoder, using `encoder_states` as initial state.\n",
+    "decoder_inputs = Input(shape=(None, num_decoder_tokens))\n",
+    "\n",
+    "# We set up our decoder to return full output sequences,\n",
+    "# and to return internal states as well. We don't use the\n",
+    "# return states in the training model, but we will use them in inference.\n",
+    "decoder_lstm = LSTM(latent_dim, return_sequences=True, return_state=True)\n",
+    "decoder_outputs, _, _ = decoder_lstm(decoder_inputs,\n",
+    "                                     initial_state=encoder_states)\n",
+    "\n",
+    "decoder_dense = Dense(num_decoder_tokens, activation='softmax')\n",
+    "decoder_outputs = decoder_dense(decoder_outputs)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 204,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Define the model that will turn\n",
+    "# `encoder_input_data` & `decoder_input_data` into `decoder_target_data`\n",
+    "model = Model([encoder_inputs, decoder_inputs], decoder_outputs)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 206,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Train on 273760 samples, validate on 68440 samples\n",
+      "Epoch 1/100\n",
+      "273760/273760 [==============================] - 743s 3ms/step - loss: 0.2597 - val_loss: 1.3135\n",
+      "Epoch 2/100\n",
+      "273760/273760 [==============================] - 702s 3ms/step - loss: 0.0587 - val_loss: 1.5200\n",
+      "Epoch 3/100\n",
+      "273760/273760 [==============================] - 658s 2ms/step - loss: 0.0232 - val_loss: 1.6334\n",
+      "Epoch 4/100\n",
+      "273760/273760 [==============================] - 655s 2ms/step - loss: 0.0141 - val_loss: 1.7563\n",
+      "Epoch 5/100\n",
+      "273760/273760 [==============================] - 648s 2ms/step - loss: 0.0101 - val_loss: 1.7978\n",
+      "Epoch 6/100\n",
+      "273760/273760 [==============================] - 648s 2ms/step - loss: 0.0078 - val_loss: 1.8374\n",
+      "Epoch 7/100\n",
+      "273760/273760 [==============================] - 645s 2ms/step - loss: 0.0062 - val_loss: 1.8761\n",
+      "Epoch 8/100\n",
+      "273760/273760 [==============================] - 658s 2ms/step - loss: 0.0052 - val_loss: 1.9187\n",
+      "Epoch 9/100\n",
+      "273760/273760 [==============================] - 779s 3ms/step - loss: 0.0045 - val_loss: 1.9649\n",
+      "Epoch 10/100\n",
+      "273760/273760 [==============================] - 706s 3ms/step - loss: 0.0039 - val_loss: 1.9650\n",
+      "Epoch 11/100\n",
+      "273760/273760 [==============================] - 640s 2ms/step - loss: 0.0035 - val_loss: 1.9923\n",
+      "Epoch 12/100\n",
+      "242432/273760 [=========================>....] - ETA: 1:06 - loss: 0.0032"
+     ]
+    },
+    {
+     "ename": "KeyboardInterrupt",
+     "evalue": "",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
+      "\u001b[0;32m<ipython-input-206-614d9c150bbf>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      7\u001b[0m           \u001b[0mbatch_size\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mbatch_size\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      8\u001b[0m           \u001b[0mepochs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mepochs\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 9\u001b[0;31m           validation_split=0.2)\n\u001b[0m\u001b[1;32m     10\u001b[0m \u001b[0;31m# Save model\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     11\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msave\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m's2s.h5'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.6/site-packages/keras/engine/training.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, **kwargs)\u001b[0m\n\u001b[1;32m   1703\u001b[0m                               \u001b[0minitial_epoch\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0minitial_epoch\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1704\u001b[0m                               \u001b[0msteps_per_epoch\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0msteps_per_epoch\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1705\u001b[0;31m                               validation_steps=validation_steps)\n\u001b[0m\u001b[1;32m   1706\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1707\u001b[0m     def evaluate(self, x=None, y=None,\n",
+      "\u001b[0;32m/usr/local/lib/python3.6/site-packages/keras/engine/training.py\u001b[0m in \u001b[0;36m_fit_loop\u001b[0;34m(self, f, ins, out_labels, batch_size, epochs, verbose, callbacks, val_f, val_ins, shuffle, callback_metrics, initial_epoch, steps_per_epoch, validation_steps)\u001b[0m\n\u001b[1;32m   1233\u001b[0m                         \u001b[0mins_batch\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mins_batch\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtoarray\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1234\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1235\u001b[0;31m                     \u001b[0mouts\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mins_batch\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1236\u001b[0m                     \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mouts\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlist\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1237\u001b[0m                         \u001b[0mouts\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mouts\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.6/site-packages/keras/backend/tensorflow_backend.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, inputs)\u001b[0m\n\u001b[1;32m   2476\u001b[0m         \u001b[0msession\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mget_session\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2477\u001b[0m         updated = session.run(fetches=fetches, feed_dict=feed_dict,\n\u001b[0;32m-> 2478\u001b[0;31m                               **self.session_kwargs)\n\u001b[0m\u001b[1;32m   2479\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mupdated\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2480\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36mrun\u001b[0;34m(self, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m    903\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    904\u001b[0m       result = self._run(None, fetches, feed_dict, options_ptr,\n\u001b[0;32m--> 905\u001b[0;31m                          run_metadata_ptr)\n\u001b[0m\u001b[1;32m    906\u001b[0m       \u001b[0;32mif\u001b[0m \u001b[0mrun_metadata\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    907\u001b[0m         \u001b[0mproto_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf_session\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTF_GetBuffer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrun_metadata_ptr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_run\u001b[0;34m(self, handle, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m   1135\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mfinal_fetches\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0mfinal_targets\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mhandle\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mfeed_dict_tensor\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1136\u001b[0m       results = self._do_run(handle, final_targets, final_fetches,\n\u001b[0;32m-> 1137\u001b[0;31m                              feed_dict_tensor, options, run_metadata)\n\u001b[0m\u001b[1;32m   1138\u001b[0m     \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1139\u001b[0m       \u001b[0mresults\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_do_run\u001b[0;34m(self, handle, target_list, fetch_list, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m   1353\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mhandle\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1354\u001b[0m       return self._do_call(_run_fn, self._session, feeds, fetches, targets,\n\u001b[0;32m-> 1355\u001b[0;31m                            options, run_metadata)\n\u001b[0m\u001b[1;32m   1356\u001b[0m     \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1357\u001b[0m       \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_do_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_prun_fn\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_session\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhandle\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeeds\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfetches\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_do_call\u001b[0;34m(self, fn, *args)\u001b[0m\n\u001b[1;32m   1359\u001b[0m   \u001b[0;32mdef\u001b[0m \u001b[0m_do_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1360\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1361\u001b[0;31m       \u001b[0;32mreturn\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1362\u001b[0m     \u001b[0;32mexcept\u001b[0m \u001b[0merrors\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mOpError\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1363\u001b[0m       \u001b[0mmessage\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcompat\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mas_text\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmessage\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_run_fn\u001b[0;34m(session, feed_dict, fetch_list, target_list, options, run_metadata)\u001b[0m\n\u001b[1;32m   1338\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1339\u001b[0m           return tf_session.TF_Run(session, options, feed_dict, fetch_list,\n\u001b[0;32m-> 1340\u001b[0;31m                                    target_list, status, run_metadata)\n\u001b[0m\u001b[1;32m   1341\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1342\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_prun_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msession\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhandle\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfetch_list\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
+     ]
+    }
+   ],
+   "source": [
+    "batch_size = 64  # Batch size for training.\n",
+    "epochs = 100 # Number of epochs to train for.\n",
+    "\n",
+    "# Run training\n",
+    "model.compile(optimizer='rmsprop', loss='categorical_crossentropy')\n",
+    "model.fit([encoder_input_data, decoder_input_data], decoder_target_data,\n",
+    "          batch_size=batch_size,\n",
+    "          epochs=epochs,\n",
+    "          validation_split=0.2)\n",
+    "# Save model\n",
+    "model.save('s2s.h5')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.6.5"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}