Water quality data and pandas

WaterQualityData.csv

WQ_Data.ipynb

{
 "metadata": {
  "name": "",
  "signature": "sha256:999c6cb6ff3e0fb54f8e41f65b83a9feeffdc12b82589e853e77c579b447e6d9"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "heading",
     "level": 1,
     "metadata": {},
     "source": [
      "Compute water quality stats"
     ]
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Load metals data"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%matplotlib inline\n",
      "import os\n",
      "\n",
      "import numpy as np\n",
      "import matplotlib\n",
      "from scipy import stats \n",
      "import pandas\n",
      "import seaborn\n",
      "seaborn.set(style='ticks', context='talk')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 1
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Database stuff that won't work"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# import sqlalchemy\n",
      "\n",
      "# # where the output will be saved (Portland Server)\n",
      "# outputdir = r'path/on/my/network'\n",
      "\n",
      "# # where the database is\n",
      "# engine = sqlalchemy.create_engine('mssql+pyodbc://remoteserver/projectDB')\n",
      "\n",
      "# # read the all the data from the database\n",
      "# query = \"\"\"\n",
      "# select \n",
      "#     sampleloc, watershed, bmp, position, \n",
      "#     drainage_index, analyte, res, qual \n",
      "# from \n",
      "#     tidy_data\n",
      "# \"\"\"\n",
      "# data = pandas.read_sql_query(query, engine)\n",
      "\n",
      "# # list of analytes we want\n",
      "# POCs = ['Lead', 'Copper', 'Zinc']\n",
      "\n",
      "# # keep on data at a BMP and where the analytes are in our list\n",
      "# data = data.query(\"analyte in {}\".format(POCs))\n",
      "# data.to_csv('WaterQualityData.csv', index=False)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 2
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Read in a CSV instead (`res` = \"concentration\", units are \u00b5g/L)"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data = pandas.read_csv('WaterQualityData.csv')\n",
      "data.head()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>sampleloc</th>\n",
        "      <th>watershed</th>\n",
        "      <th>bmp</th>\n",
        "      <th>position</th>\n",
        "      <th>drainage_index</th>\n",
        "      <th>analyte</th>\n",
        "      <th>res</th>\n",
        "      <th>qual</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
        "      <td> A2BMP0002</td>\n",
        "      <td> Outfall 009</td>\n",
        "      <td> NaN</td>\n",
        "      <td> NaN</td>\n",
        "      <td> NaN</td>\n",
        "      <td> Copper</td>\n",
        "      <td>  2.40</td>\n",
        "      <td> =</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
        "      <td> A2BMP0002</td>\n",
        "      <td> Outfall 009</td>\n",
        "      <td> NaN</td>\n",
        "      <td> NaN</td>\n",
        "      <td> NaN</td>\n",
        "      <td>   Lead</td>\n",
        "      <td>  0.29</td>\n",
        "      <td> =</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
        "      <td> A2BMP0002</td>\n",
        "      <td> Outfall 009</td>\n",
        "      <td> NaN</td>\n",
        "      <td> NaN</td>\n",
        "      <td> NaN</td>\n",
        "      <td>   Zinc</td>\n",
        "      <td>  4.00</td>\n",
        "      <td> &lt;</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
        "      <td> A2BMP0005</td>\n",
        "      <td> Outfall 009</td>\n",
        "      <td> NaN</td>\n",
        "      <td> NaN</td>\n",
        "      <td> NaN</td>\n",
        "      <td> Copper</td>\n",
        "      <td>  8.70</td>\n",
        "      <td> =</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
        "      <td> A2BMP0005</td>\n",
        "      <td> Outfall 009</td>\n",
        "      <td> NaN</td>\n",
        "      <td> NaN</td>\n",
        "      <td> NaN</td>\n",
        "      <td>   Lead</td>\n",
        "      <td> 11.00</td>\n",
        "      <td> =</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 3,
       "text": [
        "   sampleloc    watershed  bmp position drainage_index analyte    res qual\n",
        "0  A2BMP0002  Outfall 009  NaN      NaN            NaN  Copper   2.40    =\n",
        "1  A2BMP0002  Outfall 009  NaN      NaN            NaN    Lead   0.29    =\n",
        "2  A2BMP0002  Outfall 009  NaN      NaN            NaN    Zinc   4.00    <\n",
        "3  A2BMP0005  Outfall 009  NaN      NaN            NaN  Copper   8.70    =\n",
        "4  A2BMP0005  Outfall 009  NaN      NaN            NaN    Lead  11.00    ="
       ]
      }
     ],
     "prompt_number": 3
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Compute the median for each analyte within watersheds"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "groups = data.groupby(by=['watershed', 'analyte'])\n",
      "groups.median()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th></th>\n",
        "      <th>res</th>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>watershed</th>\n",
        "      <th>analyte</th>\n",
        "      <th></th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th rowspan=\"3\" valign=\"top\">Outfall 008</th>\n",
        "      <th>Copper</th>\n",
        "      <td>  4.10</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>Lead</th>\n",
        "      <td>  1.40</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>Zinc</th>\n",
        "      <td> 11.80</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th rowspan=\"3\" valign=\"top\">Outfall 009</th>\n",
        "      <th>Copper</th>\n",
        "      <td>  4.75</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>Lead</th>\n",
        "      <td>  2.50</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>Zinc</th>\n",
        "      <td> 27.00</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 4,
       "text": [
        "                       res\n",
        "watershed   analyte       \n",
        "Outfall 008 Copper    4.10\n",
        "            Lead      1.40\n",
        "            Zinc     11.80\n",
        "Outfall 009 Copper    4.75\n",
        "            Lead      2.50\n",
        "            Zinc     27.00"
       ]
      }
     ],
     "prompt_number": 4
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Pivot that data so that analytes are in columns "
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "groups.median().unstack(level='analyte')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr>\n",
        "      <th></th>\n",
        "      <th colspan=\"3\" halign=\"left\">res</th>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>analyte</th>\n",
        "      <th>Copper</th>\n",
        "      <th>Lead</th>\n",
        "      <th>Zinc</th>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>watershed</th>\n",
        "      <th></th>\n",
        "      <th></th>\n",
        "      <th></th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>Outfall 008</th>\n",
        "      <td> 4.10</td>\n",
        "      <td> 1.4</td>\n",
        "      <td> 11.8</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>Outfall 009</th>\n",
        "      <td> 4.75</td>\n",
        "      <td> 2.5</td>\n",
        "      <td> 27.0</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 5,
       "text": [
        "               res           \n",
        "analyte     Copper Lead  Zinc\n",
        "watershed                    \n",
        "Outfall 008   4.10  1.4  11.8\n",
        "Outfall 009   4.75  2.5  27.0"
       ]
      }
     ],
     "prompt_number": 5
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Location at the (first 10) medians by sampling location"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data.groupby(by=['sampleloc', 'analyte']).median().unstack(level='analyte').head(10)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr>\n",
        "      <th></th>\n",
        "      <th colspan=\"3\" halign=\"left\">res</th>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>analyte</th>\n",
        "      <th>Copper</th>\n",
        "      <th>Lead</th>\n",
        "      <th>Zinc</th>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>sampleloc</th>\n",
        "      <th></th>\n",
        "      <th></th>\n",
        "      <th></th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>A1BMP0001</th>\n",
        "      <td> 4.20</td>\n",
        "      <td> 0.280</td>\n",
        "      <td> 11.0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>A1BMP0002</th>\n",
        "      <td> 6.10</td>\n",
        "      <td> 1.000</td>\n",
        "      <td> 57.0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>A1SW0002</th>\n",
        "      <td>  NaN</td>\n",
        "      <td> 0.580</td>\n",
        "      <td>  NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>A1SW0003</th>\n",
        "      <td>  NaN</td>\n",
        "      <td> 0.285</td>\n",
        "      <td>  NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>A1SW0005</th>\n",
        "      <td> 4.30</td>\n",
        "      <td> 0.605</td>\n",
        "      <td>  NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>A1SW0009</th>\n",
        "      <td> 7.90</td>\n",
        "      <td> 9.100</td>\n",
        "      <td>  NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>A1SW0009-C</th>\n",
        "      <td> 6.50</td>\n",
        "      <td> 2.300</td>\n",
        "      <td>  NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>A2BMP0002</th>\n",
        "      <td> 2.40</td>\n",
        "      <td> 0.290</td>\n",
        "      <td>  4.0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>A2BMP0003</th>\n",
        "      <td> 3.55</td>\n",
        "      <td> 1.375</td>\n",
        "      <td>  7.8</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>A2BMP0004</th>\n",
        "      <td> 6.70</td>\n",
        "      <td> 4.200</td>\n",
        "      <td> 19.0</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 6,
       "text": [
        "              res             \n",
        "analyte    Copper   Lead  Zinc\n",
        "sampleloc                     \n",
        "A1BMP0001    4.20  0.280  11.0\n",
        "A1BMP0002    6.10  1.000  57.0\n",
        "A1SW0002      NaN  0.580   NaN\n",
        "A1SW0003      NaN  0.285   NaN\n",
        "A1SW0005     4.30  0.605   NaN\n",
        "A1SW0009     7.90  9.100   NaN\n",
        "A1SW0009-C   6.50  2.300   NaN\n",
        "A2BMP0002    2.40  0.290   4.0\n",
        "A2BMP0003    3.55  1.375   7.8\n",
        "A2BMP0004    6.70  4.200  19.0"
       ]
      }
     ],
     "prompt_number": 6
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Aggregate data on something more complex"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def interquartile_range(group):\n",
      "    '''\n",
      "    Difference between the 75th and 25th percentiles\n",
      "    '''\n",
      "    pctl75 = stats.scoreatpercentile(group, 75)\n",
      "    pctl25 = stats.scoreatpercentile(group, 25)\n",
      "    IQR = pctl75 - pctl25\n",
      "    return IQR\n",
      "\n",
      "\n",
      "(groups.agg({'res': interquartile_range}) \n",
      "       .rename(columns={'res': 'IQR'})\n",
      "       .unstack(level='watershed')\n",
      ")"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr>\n",
        "      <th></th>\n",
        "      <th colspan=\"2\" halign=\"left\">IQR</th>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>watershed</th>\n",
        "      <th>Outfall 008</th>\n",
        "      <th>Outfall 009</th>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>analyte</th>\n",
        "      <th></th>\n",
        "      <th></th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>Copper</th>\n",
        "      <td>  5.60</td>\n",
        "      <td>  5.2075</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>Lead</th>\n",
        "      <td>  4.64</td>\n",
        "      <td>  5.3575</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>Zinc</th>\n",
        "      <td> 19.25</td>\n",
        "      <td> 53.5000</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 7,
       "text": [
        "                  IQR            \n",
        "watershed Outfall 008 Outfall 009\n",
        "analyte                          \n",
        "Copper           5.60      5.2075\n",
        "Lead             4.64      5.3575\n",
        "Zinc            19.25     53.5000"
       ]
      }
     ],
     "prompt_number": 7
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Neat Trick: pandas' `describe` method can be applied and tranposed to a `groupby` object"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "(groups.apply(lambda g: g.describe().T))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th></th>\n",
        "      <th></th>\n",
        "      <th>count</th>\n",
        "      <th>mean</th>\n",
        "      <th>std</th>\n",
        "      <th>min</th>\n",
        "      <th>25%</th>\n",
        "      <th>50%</th>\n",
        "      <th>75%</th>\n",
        "      <th>max</th>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>watershed</th>\n",
        "      <th>analyte</th>\n",
        "      <th></th>\n",
        "      <th></th>\n",
        "      <th></th>\n",
        "      <th></th>\n",
        "      <th></th>\n",
        "      <th></th>\n",
        "      <th></th>\n",
        "      <th></th>\n",
        "      <th></th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th rowspan=\"3\" valign=\"top\">Outfall 008</th>\n",
        "      <th>Copper</th>\n",
        "      <th>res</th>\n",
        "      <td>  77</td>\n",
        "      <td>  5.687013</td>\n",
        "      <td>  4.503969</td>\n",
        "      <td> 0.49</td>\n",
        "      <td>  2.3000</td>\n",
        "      <td>  4.10</td>\n",
        "      <td>  7.90</td>\n",
        "      <td>  19</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>Lead</th>\n",
        "      <th>res</th>\n",
        "      <td>  79</td>\n",
        "      <td>  5.411139</td>\n",
        "      <td> 14.049663</td>\n",
        "      <td> 0.17</td>\n",
        "      <td>  0.5100</td>\n",
        "      <td>  1.40</td>\n",
        "      <td>  5.15</td>\n",
        "      <td> 120</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>Zinc</th>\n",
        "      <th>res</th>\n",
        "      <td>  35</td>\n",
        "      <td> 19.794286</td>\n",
        "      <td> 22.036026</td>\n",
        "      <td> 2.50</td>\n",
        "      <td>  4.8500</td>\n",
        "      <td> 11.80</td>\n",
        "      <td> 24.10</td>\n",
        "      <td> 110</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th rowspan=\"3\" valign=\"top\">Outfall 009</th>\n",
        "      <th>Copper</th>\n",
        "      <th>res</th>\n",
        "      <td> 364</td>\n",
        "      <td>  7.027830</td>\n",
        "      <td>  7.442934</td>\n",
        "      <td> 0.60</td>\n",
        "      <td>  3.1925</td>\n",
        "      <td>  4.75</td>\n",
        "      <td>  8.40</td>\n",
        "      <td>  86</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>Lead</th>\n",
        "      <th>res</th>\n",
        "      <td> 430</td>\n",
        "      <td>  6.592888</td>\n",
        "      <td> 16.149913</td>\n",
        "      <td> 0.04</td>\n",
        "      <td>  0.9425</td>\n",
        "      <td>  2.50</td>\n",
        "      <td>  6.30</td>\n",
        "      <td> 260</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>Zinc</th>\n",
        "      <th>res</th>\n",
        "      <td> 191</td>\n",
        "      <td> 50.619895</td>\n",
        "      <td> 63.762495</td>\n",
        "      <td> 4.00</td>\n",
        "      <td> 13.0000</td>\n",
        "      <td> 27.00</td>\n",
        "      <td> 66.50</td>\n",
        "      <td> 310</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 8,
       "text": [
        "                         count       mean        std   min      25%    50%  \\\n",
        "watershed   analyte                                                          \n",
        "Outfall 008 Copper  res     77   5.687013   4.503969  0.49   2.3000   4.10   \n",
        "            Lead    res     79   5.411139  14.049663  0.17   0.5100   1.40   \n",
        "            Zinc    res     35  19.794286  22.036026  2.50   4.8500  11.80   \n",
        "Outfall 009 Copper  res    364   7.027830   7.442934  0.60   3.1925   4.75   \n",
        "            Lead    res    430   6.592888  16.149913  0.04   0.9425   2.50   \n",
        "            Zinc    res    191  50.619895  63.762495  4.00  13.0000  27.00   \n",
        "\n",
        "                           75%  max  \n",
        "watershed   analyte                  \n",
        "Outfall 008 Copper  res   7.90   19  \n",
        "            Lead    res   5.15  120  \n",
        "            Zinc    res  24.10  110  \n",
        "Outfall 009 Copper  res   8.40   86  \n",
        "            Lead    res   6.30  260  \n",
        "            Zinc    res  66.50  310  "
       ]
      }
     ],
     "prompt_number": 8
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Some seaborn stuff"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# log data makes smoother densities\n",
      "data['logres'] = np.log10(data['res'])\n",
      "fg = seaborn.FacetGrid(data=data, row='watershed', col='analyte', margin_titles=True)\n",
      "fg.map(seaborn.kdeplot, 'logres', shade=True)\n",
      "fg.set_xlabels(r'$\\log_{10}$ conc.')\n",
      "fg.set_ylabels('density')\n",
      "fg.fig.tight_layout()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAo0AAAGqCAYAAABu7H5eAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0xb7ea2b0>"
       ]
      }
     ],
     "prompt_number": 13
    }
   ],
   "metadata": {}
  }
 ]
}