revision_questions.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Intro"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are a couple comments from the reviewers that I want to clear up in this notebook:\n",
    "\n",
    "1. How sensitive are the RF results to different parameter values such as number of trees? (from reviewer 1)\n",
    "2. Why not just use rainfall as a proxy for flood severity? Why go to all the work of building these more complex model if most of the information just comes from the rainfall?\n",
    "3. Not what the reviewers asked for but for my own peace of mind, are the two models being tested on the same data?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "from hr_db_scripts.main_db_script import get_db_table_as_df\n",
    "from db_scripts.main_db_script import db_filename\n",
    "from results import make_results_df\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def get_tables(table_suffix):\n",
    "\n",
    "\n",
    "    rf_trn_tbl = 'rf{}train'.format(table_suffix)\n",
    "    rf_tst_tbl = 'rf{}test'.format(table_suffix)\n",
    "    ps_trn_tbl = 'poisson{}train'.format(table_suffix)\n",
    "    ps_tst_tbl = 'poisson{}test'.format(table_suffix)\n",
    "\n",
    "    rf_trn = get_db_table_as_df(rf_trn_tbl, dbfilename=db_filename)\n",
    "    rf_tst = get_db_table_as_df(rf_tst_tbl, dbfilename=db_filename)\n",
    "    ps_trn = get_db_table_as_df(ps_trn_tbl, dbfilename=db_filename)\n",
    "    ps_tst = get_db_table_as_df(ps_tst_tbl, dbfilename=db_filename)\n",
    "    return {'rf_trn': rf_trn, 'rf_tst': rf_tst, 'ps_trn': ps_trn, 'ps_tst': ps_tst}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Question 3: are the models being tested on the same data?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "suffix = '_revisions_'\n",
    "tables = get_tables(suffix)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2541"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(tables['rf_trn']['all_trn'] != tables['ps_trn']['all_trn']).sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1255"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(tables['rf_tst']['all_tst'] != tables['ps_tst']['all_tst']).sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It looks like they weren't the same... Now we need to see how that affects the results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Now the code has been refactored so they are the same"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "suffix = '_revisions1_'\n",
    "tables1 = get_tables(suffix)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(tables1['rf_trn']['all_trn'] != tables1['ps_trn']['all_trn']).sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(tables1['rf_tst']['all_tst'] != tables1['ps_tst']['all_tst']).sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Question: do the results differ with the restructuring of the code?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(tables1['rf_trn']['all_trn'] != tables['rf_trn']['all_trn']).sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(tables1['rf_trn']['all_pred_trn'] != tables['rf_trn']['all_pred_trn']).sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(tables1['rf_tst']['all_tst'] != tables['rf_tst']['all_tst']).sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2541"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(tables1['ps_trn']['all_trn'] != tables['ps_trn']['all_trn']).sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1255"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(tables1['ps_tst']['all_tst'] != tables['ps_tst']['all_tst']).sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So now they are all the same"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Question #1 sensitivity to number of trees"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "trees = [2, 5, 10, 17, 25, 35, 50, 100, 250, 350, 500, 650, 750, 1000, 2000]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "tables = {}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "for t in trees:\n",
    "    tables[t] = 'rf_{}'.format(t)\n",
    "tables[500] = 'revisions1'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "rf train rf_rf_2_train\n",
      "rf test rf_rf_2_test\n",
      "rf train rf_rf_35_train\n",
      "rf test rf_rf_35_test\n",
      "rf train rf_rf_100_train\n",
      "rf test rf_rf_100_test\n",
      "rf train rf_rf_5_train\n",
      "rf test rf_rf_5_test\n",
      "rf train rf_rf_1000_train\n",
      "rf test rf_rf_1000_test\n",
      "rf train rf_rf_10_train\n",
      "rf test rf_rf_10_test\n",
      "rf train rf_rf_750_train\n",
      "rf test rf_rf_750_test\n",
      "rf train rf_rf_2000_train\n",
      "rf test rf_rf_2000_test\n",
      "rf train rf_rf_17_train\n",
      "rf test rf_rf_17_test\n",
      "rf train rf_rf_50_train\n",
      "rf test rf_rf_50_test\n",
      "rf train rf_revisions1_train\n",
      "rf test rf_revisions1_test\n",
      "rf train rf_rf_25_train\n",
      "rf test rf_rf_25_test\n",
      "rf train rf_rf_250_train\n",
      "rf test rf_rf_250_test\n",
      "rf train rf_rf_650_train\n",
      "rf test rf_rf_650_test\n",
      "rf train rf_rf_350_train\n",
      "rf test rf_rf_350_test\n"
     ]
    }
   ],
   "source": [
    "dfs = []\n",
    "for t in tables:\n",
    "    df = make_results_df(models=['rf'], suffix=tables[t])\n",
    "    df.index = [t]\n",
    "    dfs.append(df)\n",
    "df_comb = pd.concat(dfs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0xe526048>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAApUAAAGxCAYAAAAzhE0aAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nnwb7AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xe526828>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_comb.sort_index(inplace=True)\n",
    "ax = df_comb.plot(style=\"o-\", figsize = (9,7), logx=True, grid=True)\n",
    "ax.legend(bbox_to_anchor=(1, 0.5))\n",
    "ax.set_xlabel('Number of trees')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0xdeed4a8>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfcAAAEOCAYAAACU6F72AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xee5dcc0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = df_comb.rolling(5).mean().plot(style='o-', logx=True)\n",
    "ax.legend(bbox_to_anchor=(1, 0.5))\n",
    "ax.set_xlabel('Number of trees')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Question #2 what if only rainfall"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "suffixes = ['revisions1', 'only_rd', 'only_rain', 'no_rd', 'no_tides', 'top_5', 'top_2', 'revisions2', 'revisions2_no_tide']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "poisson train poisson_revisions1_train\n",
      "poisson test poisson_revisions1_test\n",
      "rf train rf_revisions1_train\n",
      "rf test rf_revisions1_test\n",
      "poisson train poisson_only_rd_train\n",
      "poisson test poisson_only_rd_test\n",
      "rf train rf_only_rd_train\n",
      "rf test rf_only_rd_test\n",
      "poisson train poisson_only_rain_train\n",
      "poisson test poisson_only_rain_test\n",
      "rf train rf_only_rain_train\n",
      "rf test rf_only_rain_test\n",
      "poisson train poisson_no_rd_train\n",
      "poisson test poisson_no_rd_test\n",
      "rf train rf_no_rd_train\n",
      "rf test rf_no_rd_test\n",
      "poisson train poisson_no_tides_train\n",
      "poisson test poisson_no_tides_test\n",
      "rf train rf_no_tides_train\n",
      "rf test rf_no_tides_test\n",
      "poisson train poisson_top_5_train\n",
      "poisson test poisson_top_5_test\n",
      "rf train rf_top_5_train\n",
      "rf test rf_top_5_test\n",
      "poisson train poisson_top_2_train\n",
      "poisson test poisson_top_2_test\n",
      "rf train rf_top_2_train\n",
      "rf test rf_top_2_test\n",
      "poisson train poisson_revisions2_train\n",
      "poisson test poisson_revisions2_test\n",
      "rf train rf_revisions2_train\n",
      "rf test rf_revisions2_test\n",
      "poisson train poisson_revisions2_no_tide_train\n",
      "poisson test poisson_revisions2_no_tide_test\n",
      "rf train rf_revisions2_no_tide_train\n",
      "rf test rf_revisions2_no_tide_test\n"
     ]
    }
   ],
   "source": [
    "dfs = []\n",
    "for s in suffixes:\n",
    "    df = make_results_df(suffix=s)\n",
    "    dfs.append(df)\n",
    "    df.index = [\"{}_{}\".format(i, s) for i in df.index]\n",
    "df_vars = pd.concat(dfs)\n",
    "\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>RMSE-Training</th>\n",
       "      <th>RMSE-Evaluation</th>\n",
       "      <th>MAE-Training</th>\n",
       "      <th>MAE-Evaluation</th>\n",
       "      <th>std-Training</th>\n",
       "      <th>std-Evaluation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>poisson_revisions1</th>\n",
       "      <td>9.95</td>\n",
       "      <td>29.36</td>\n",
       "      <td>6.36</td>\n",
       "      <td>15.90</td>\n",
       "      <td>5.26</td>\n",
       "      <td>17.66</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rf_revisions1</th>\n",
       "      <td>9.79</td>\n",
       "      <td>21.41</td>\n",
       "      <td>5.38</td>\n",
       "      <td>12.29</td>\n",
       "      <td>2.55</td>\n",
       "      <td>4.70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>poisson_only_rd</th>\n",
       "      <td>32.06</td>\n",
       "      <td>30.48</td>\n",
       "      <td>17.14</td>\n",
       "      <td>16.10</td>\n",
       "      <td>4.35</td>\n",
       "      <td>3.48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rf_only_rd</th>\n",
       "      <td>9.36</td>\n",
       "      <td>17.82</td>\n",
       "      <td>5.60</td>\n",
       "      <td>10.73</td>\n",
       "      <td>2.99</td>\n",
       "      <td>6.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>poisson_only_rain</th>\n",
       "      <td>29.00</td>\n",
       "      <td>32.01</td>\n",
       "      <td>14.19</td>\n",
       "      <td>15.61</td>\n",
       "      <td>7.65</td>\n",
       "      <td>7.86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rf_only_rain</th>\n",
       "      <td>12.21</td>\n",
       "      <td>26.83</td>\n",
       "      <td>6.72</td>\n",
       "      <td>14.09</td>\n",
       "      <td>2.37</td>\n",
       "      <td>4.62</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>poisson_no_rd</th>\n",
       "      <td>9.97</td>\n",
       "      <td>30.08</td>\n",
       "      <td>6.52</td>\n",
       "      <td>15.98</td>\n",
       "      <td>5.23</td>\n",
       "      <td>18.38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rf_no_rd</th>\n",
       "      <td>12.94</td>\n",
       "      <td>27.77</td>\n",
       "      <td>6.95</td>\n",
       "      <td>14.86</td>\n",
       "      <td>3.03</td>\n",
       "      <td>5.67</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>poisson_no_tides</th>\n",
       "      <td>12.55</td>\n",
       "      <td>37.14</td>\n",
       "      <td>8.17</td>\n",
       "      <td>18.93</td>\n",
       "      <td>6.63</td>\n",
       "      <td>12.81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rf_no_tides</th>\n",
       "      <td>10.85</td>\n",
       "      <td>24.95</td>\n",
       "      <td>5.91</td>\n",
       "      <td>13.84</td>\n",
       "      <td>2.60</td>\n",
       "      <td>4.81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>poisson_top_5</th>\n",
       "      <td>18.42</td>\n",
       "      <td>26.97</td>\n",
       "      <td>10.61</td>\n",
       "      <td>14.87</td>\n",
       "      <td>7.19</td>\n",
       "      <td>9.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rf_top_5</th>\n",
       "      <td>10.04</td>\n",
       "      <td>20.49</td>\n",
       "      <td>5.58</td>\n",
       "      <td>11.79</td>\n",
       "      <td>2.53</td>\n",
       "      <td>5.36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>poisson_top_2</th>\n",
       "      <td>22.33</td>\n",
       "      <td>29.03</td>\n",
       "      <td>13.30</td>\n",
       "      <td>16.13</td>\n",
       "      <td>5.59</td>\n",
       "      <td>6.54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rf_top_2</th>\n",
       "      <td>7.68</td>\n",
       "      <td>17.81</td>\n",
       "      <td>4.35</td>\n",
       "      <td>10.58</td>\n",
       "      <td>2.79</td>\n",
       "      <td>5.52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>poisson_revisions2</th>\n",
       "      <td>10.06</td>\n",
       "      <td>29.81</td>\n",
       "      <td>6.56</td>\n",
       "      <td>16.34</td>\n",
       "      <td>5.18</td>\n",
       "      <td>19.11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rf_revisions2</th>\n",
       "      <td>8.04</td>\n",
       "      <td>16.55</td>\n",
       "      <td>4.41</td>\n",
       "      <td>9.83</td>\n",
       "      <td>3.17</td>\n",
       "      <td>6.21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>poisson_revisions2_no_tide</th>\n",
       "      <td>15.14</td>\n",
       "      <td>33.98</td>\n",
       "      <td>9.52</td>\n",
       "      <td>16.64</td>\n",
       "      <td>7.79</td>\n",
       "      <td>11.98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rf_revisions2_no_tide</th>\n",
       "      <td>8.28</td>\n",
       "      <td>17.21</td>\n",
       "      <td>4.87</td>\n",
       "      <td>10.07</td>\n",
       "      <td>3.17</td>\n",
       "      <td>6.41</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                            RMSE-Training  RMSE-Evaluation  MAE-Training  \\\n",
       "poisson_revisions1                   9.95            29.36          6.36   \n",
       "rf_revisions1                        9.79            21.41          5.38   \n",
       "poisson_only_rd                     32.06            30.48         17.14   \n",
       "rf_only_rd                           9.36            17.82          5.60   \n",
       "poisson_only_rain                   29.00            32.01         14.19   \n",
       "rf_only_rain                        12.21            26.83          6.72   \n",
       "poisson_no_rd                        9.97            30.08          6.52   \n",
       "rf_no_rd                            12.94            27.77          6.95   \n",
       "poisson_no_tides                    12.55            37.14          8.17   \n",
       "rf_no_tides                         10.85            24.95          5.91   \n",
       "poisson_top_5                       18.42            26.97         10.61   \n",
       "rf_top_5                            10.04            20.49          5.58   \n",
       "poisson_top_2                       22.33            29.03         13.30   \n",
       "rf_top_2                             7.68            17.81          4.35   \n",
       "poisson_revisions2                  10.06            29.81          6.56   \n",
       "rf_revisions2                        8.04            16.55          4.41   \n",
       "poisson_revisions2_no_tide          15.14            33.98          9.52   \n",
       "rf_revisions2_no_tide                8.28            17.21          4.87   \n",
       "\n",
       "                            MAE-Evaluation  std-Training  std-Evaluation  \n",
       "poisson_revisions1                   15.90          5.26           17.66  \n",
       "rf_revisions1                        12.29          2.55            4.70  \n",
       "poisson_only_rd                      16.10          4.35            3.48  \n",
       "rf_only_rd                           10.73          2.99            6.02  \n",
       "poisson_only_rain                    15.61          7.65            7.86  \n",
       "rf_only_rain                         14.09          2.37            4.62  \n",
       "poisson_no_rd                        15.98          5.23           18.38  \n",
       "rf_no_rd                             14.86          3.03            5.67  \n",
       "poisson_no_tides                     18.93          6.63           12.81  \n",
       "rf_no_tides                          13.84          2.60            4.81  \n",
       "poisson_top_5                        14.87          7.19            9.31  \n",
       "rf_top_5                             11.79          2.53            5.36  \n",
       "poisson_top_2                        16.13          5.59            6.54  \n",
       "rf_top_2                             10.58          2.79            5.52  \n",
       "poisson_revisions2                   16.34          5.18           19.11  \n",
       "rf_revisions2                         9.83          3.17            6.21  \n",
       "poisson_revisions2_no_tide           16.64          7.79           11.98  \n",
       "rf_revisions2_no_tide                10.07          3.17            6.41  "
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_vars"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.16767865483418962"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(21.41-17.82)/21.41"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.042970574497898255"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(21.41-20.49)/21.41"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.1653432975245212"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(24.95-21.41)/21.41"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0xd92b198>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAl0AAAHmCAYAAABJZzsAAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nVJk+0wAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xe4f9780>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = df_vars.plot.bar(figsize=(8,6))\n",
    "ax.legend(bbox_to_anchor=(1, 0.5))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Sensitivity to number of variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "suffixes = [4, 6, 8, 10, 12, 14, 18, 20]\n",
    "suffixes = ['rf_{}v'.format(i) for i in suffixes]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "rf train rf_rf_4v_train\n",
      "rf test rf_rf_4v_test\n",
      "rf train rf_rf_6v_train\n",
      "rf test rf_rf_6v_test\n",
      "rf train rf_rf_8v_train\n",
      "rf test rf_rf_8v_test\n",
      "rf train rf_rf_10v_train\n",
      "rf test rf_rf_10v_test\n",
      "rf train rf_rf_12v_train\n",
      "rf test rf_rf_12v_test\n",
      "rf train rf_rf_14v_train\n",
      "rf test rf_rf_14v_test\n",
      "rf train rf_rf_18v_train\n",
      "rf test rf_rf_18v_test\n",
      "rf train rf_rf_20v_train\n",
      "rf test rf_rf_20v_test\n"
     ]
    }
   ],
   "source": [
    "dfs = []\n",
    "for s in suffixes:\n",
    "    df = make_results_df(suffix=s, models=['rf'])\n",
    "    dfs.append(df)\n",
    "    df.index = [\"{}_{}\".format(i, s) for i in df.index]\n",
    "df_nvars = pd.concat(dfs)\n",
    "\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>RMSE-Training</th>\n",
       "      <th>RMSE-Evaluation</th>\n",
       "      <th>MAE-Training</th>\n",
       "      <th>MAE-Evaluation</th>\n",
       "      <th>std-Training</th>\n",
       "      <th>std-Evaluation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>rf_rf_4v</th>\n",
       "      <td>10.23</td>\n",
       "      <td>22.36</td>\n",
       "      <td>5.60</td>\n",
       "      <td>12.37</td>\n",
       "      <td>2.50</td>\n",
       "      <td>5.13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rf_rf_6v</th>\n",
       "      <td>9.41</td>\n",
       "      <td>20.22</td>\n",
       "      <td>5.18</td>\n",
       "      <td>11.70</td>\n",
       "      <td>2.62</td>\n",
       "      <td>5.10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rf_rf_8v</th>\n",
       "      <td>8.75</td>\n",
       "      <td>19.06</td>\n",
       "      <td>4.85</td>\n",
       "      <td>11.25</td>\n",
       "      <td>2.61</td>\n",
       "      <td>5.34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rf_rf_10v</th>\n",
       "      <td>8.29</td>\n",
       "      <td>17.66</td>\n",
       "      <td>4.63</td>\n",
       "      <td>10.69</td>\n",
       "      <td>2.62</td>\n",
       "      <td>4.66</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rf_rf_12v</th>\n",
       "      <td>7.89</td>\n",
       "      <td>18.78</td>\n",
       "      <td>4.39</td>\n",
       "      <td>11.20</td>\n",
       "      <td>2.82</td>\n",
       "      <td>5.63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rf_rf_14v</th>\n",
       "      <td>7.78</td>\n",
       "      <td>17.97</td>\n",
       "      <td>4.22</td>\n",
       "      <td>10.66</td>\n",
       "      <td>2.84</td>\n",
       "      <td>6.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rf_rf_18v</th>\n",
       "      <td>7.74</td>\n",
       "      <td>16.81</td>\n",
       "      <td>4.30</td>\n",
       "      <td>9.95</td>\n",
       "      <td>2.97</td>\n",
       "      <td>7.05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rf_rf_20v</th>\n",
       "      <td>7.74</td>\n",
       "      <td>16.81</td>\n",
       "      <td>4.30</td>\n",
       "      <td>9.95</td>\n",
       "      <td>2.97</td>\n",
       "      <td>7.05</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           RMSE-Training  RMSE-Evaluation  MAE-Training  MAE-Evaluation  \\\n",
       "rf_rf_4v           10.23            22.36          5.60           12.37   \n",
       "rf_rf_6v            9.41            20.22          5.18           11.70   \n",
       "rf_rf_8v            8.75            19.06          4.85           11.25   \n",
       "rf_rf_10v           8.29            17.66          4.63           10.69   \n",
       "rf_rf_12v           7.89            18.78          4.39           11.20   \n",
       "rf_rf_14v           7.78            17.97          4.22           10.66   \n",
       "rf_rf_18v           7.74            16.81          4.30            9.95   \n",
       "rf_rf_20v           7.74            16.81          4.30            9.95   \n",
       "\n",
       "           std-Training  std-Evaluation  \n",
       "rf_rf_4v           2.50            5.13  \n",
       "rf_rf_6v           2.62            5.10  \n",
       "rf_rf_8v           2.61            5.34  \n",
       "rf_rf_10v          2.62            4.66  \n",
       "rf_rf_12v          2.82            5.63  \n",
       "rf_rf_14v          2.84            6.50  \n",
       "rf_rf_18v          2.97            7.05  \n",
       "rf_rf_20v          2.97            7.05  "
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_nvars"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0xe65d390>"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmcAAAFpCAYAAADZUI/8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/wNluZXTI3H18AAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xe65d828>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = df_nvars.plot(figsize=(8,6))\n",
    "ax.legend(bbox_to_anchor=(1, 0.5))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df_tune = get_db_table_as_df(\"tuning_mtry\", dbfilename=db_filename)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "del df_tune['row_names']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0xd773cf8>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEWCAYAAAB2X2wCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/g9xa2DW+YrnogAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xf9897b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_tune_piv = df_tune.pivot(columns=\"mtry\", values=\"OOBError\")\n",
    "df_tune_piv.mean().plot.bar()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "j = 0\n",
    "l = []\n",
    "m = []\n",
    "first_time = True\n",
    "for i in df_tune.iterrows():\n",
    "    if i[1]['mtry'] == 1 and first_time == False:\n",
    "        sub_df = pd.DataFrame(m)\n",
    "        sub_df['num_run'] = j\n",
    "        sub_df['rank'] = sub_df['OOBError'].rank()\n",
    "        l.append(sub_df)\n",
    "        j += 1\n",
    "        m = []\n",
    "    first_time = False\n",
    "    m.append(i[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_tune_counts = pd.concat(l)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>mtry</th>\n",
       "      <th>OOBError</th>\n",
       "      <th>num_run</th>\n",
       "      <th>rank</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>46.183688</td>\n",
       "      <td>0</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2.0</td>\n",
       "      <td>34.617008</td>\n",
       "      <td>0</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.0</td>\n",
       "      <td>28.417848</td>\n",
       "      <td>0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>8.0</td>\n",
       "      <td>24.093781</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>16.0</td>\n",
       "      <td>22.988226</td>\n",
       "      <td>0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   mtry   OOBError  num_run  rank\n",
       "0   1.0  46.183688        0   6.0\n",
       "1   2.0  34.617008        0   5.0\n",
       "2   4.0  28.417848        0   4.0\n",
       "3   8.0  24.093781        0   3.0\n",
       "4  16.0  22.988226        0   2.0"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_tune_counts.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df_tune_rank_pivot = df_tune_counts.pivot_table(values='rank', columns='num_run', index='mtry')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>num_run</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>89</th>\n",
       "      <th>90</th>\n",
       "      <th>91</th>\n",
       "      <th>92</th>\n",
       "      <th>93</th>\n",
       "      <th>94</th>\n",
       "      <th>95</th>\n",
       "      <th>96</th>\n",
       "      <th>97</th>\n",
       "      <th>98</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mtry</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",
       "      <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",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1.0</th>\n",
       "      <td>6.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2.0</th>\n",
       "      <td>5.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>...</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4.0</th>\n",
       "      <td>4.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8.0</th>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>...</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16.0</th>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17.0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>6 rows × 99 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "num_run   0    1    2    3    4    5    6    7    8    9  ...    89   90   91  \\\n",
       "mtry                                                      ...                   \n",
       "1.0      6.0  5.0  6.0  6.0  6.0  6.0  6.0  6.0  6.0  6.0 ...   6.0  6.0  6.0   \n",
       "2.0      5.0  4.0  5.0  5.0  5.0  5.0  5.0  5.0  5.0  5.0 ...   5.0  5.0  5.0   \n",
       "4.0      4.0  3.0  4.0  4.0  4.0  4.0  4.0  4.0  4.0  4.0 ...   4.0  4.0  4.0   \n",
       "8.0      3.0  1.0  3.0  3.0  3.0  3.0  3.0  3.0  2.0  3.0 ...   3.0  3.0  3.0   \n",
       "16.0     2.0  2.0  2.0  2.0  1.0  1.0  1.0  2.0  1.0  1.0 ...   2.0  1.0  1.0   \n",
       "17.0     1.0  NaN  1.0  1.0  2.0  2.0  2.0  1.0  3.0  2.0 ...   1.0  2.0  2.0   \n",
       "\n",
       "num_run   92   93   94   95   96   97   98  \n",
       "mtry                                        \n",
       "1.0      6.0  6.0  5.0  6.0  6.0  6.0  6.0  \n",
       "2.0      5.0  5.0  4.0  5.0  5.0  5.0  5.0  \n",
       "4.0      4.0  4.0  3.0  4.0  4.0  4.0  4.0  \n",
       "8.0      3.0  3.0  1.0  3.0  3.0  3.0  3.0  \n",
       "16.0     1.0  2.0  2.0  2.0  1.0  1.0  1.0  \n",
       "17.0     2.0  1.0  NaN  1.0  2.0  2.0  2.0  \n",
       "\n",
       "[6 rows x 99 columns]"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_tune_rank_pivot.head(6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x106ceeb8>"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEWCAYAAAB2X2wCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xe01efd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "(df_tune_rank_pivot == 1).sum(1).plot.bar()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## number of predictions outside 159 poisson"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "suffix = '_revisions2_'\n",
    "tables_rev2 = get_tables(suffix)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'ps_trn':       row_names  all_trn  all_pred_trn\n",
       " 0             1    101.0     87.781775\n",
       " 1             2      2.0      0.018436\n",
       " 2             3    110.0    110.538831\n",
       " 3             4    105.0    106.236694\n",
       " 4            10      2.0      1.499750\n",
       " 5            11     39.0     38.354316\n",
       " 6            12     18.0      5.004708\n",
       " 7            13      1.0      1.052861\n",
       " 8            14      5.0      0.344056\n",
       " 9            15     31.0     31.826234\n",
       " 10           16      3.0      0.084996\n",
       " 11           17      1.0      0.229902\n",
       " 12           19      1.0      0.040780\n",
       " 13           20      4.0      2.546393\n",
       " 14           21     10.0      5.886819\n",
       " 15           22      7.0     10.155219\n",
       " 16           23     10.0      3.277841\n",
       " 17           24      1.0      0.922080\n",
       " 18           26     16.0      0.854746\n",
       " 19           27    159.0    160.480709\n",
       " 20           28      1.0      0.053965\n",
       " 21           29      3.0      0.089965\n",
       " 22           30      6.0      0.015120\n",
       " 23           32      5.0      6.849457\n",
       " 24           34      8.0      2.549178\n",
       " 25           35      6.0     11.752951\n",
       " 26           36      2.0      0.150452\n",
       " 27           37      9.0      1.022975\n",
       " 28           40      8.0     11.464867\n",
       " 29           42     11.0     10.729371\n",
       " ...         ...      ...           ...\n",
       " 57439      2049      0.0      0.054119\n",
       " 57440      2050      0.0      0.050904\n",
       " 57441      2057      0.0      0.007860\n",
       " 57442      2058      0.0      0.159903\n",
       " 57443      2059      0.0      0.007988\n",
       " 57444      2062      0.0      0.035797\n",
       " 57445      2063      0.0      0.132407\n",
       " 57446      2067      0.0      0.017867\n",
       " 57447      2078      0.0      0.029878\n",
       " 57448      2080      0.0      0.017520\n",
       " 57449      2085      0.0      0.019815\n",
       " 57450      2086      0.0      0.038028\n",
       " 57451      2087      0.0      0.206184\n",
       " 57452      2088      0.0      0.018823\n",
       " 57453      2090      0.0      0.027487\n",
       " 57454      2091      0.0      0.080834\n",
       " 57455      2092      0.0      0.082915\n",
       " 57456      2095      0.0      0.015520\n",
       " 57457      2100      0.0      0.007682\n",
       " 57458      2101      0.0      0.012631\n",
       " 57459      2111      0.0      0.840060\n",
       " 57460      2112      0.0      0.016895\n",
       " 57461      2116      0.0      0.019428\n",
       " 57462      2119      0.0      0.011437\n",
       " 57463      2132      0.0      0.190289\n",
       " 57464      2143      0.0      0.127912\n",
       " 57465      2147      0.0      0.060336\n",
       " 57466      2164      0.0      0.358587\n",
       " 57467      2166      0.0      0.044481\n",
       " 57468      2167      0.0      0.027649\n",
       " \n",
       " [57469 rows x 3 columns], 'ps_tst':       row_names  all_tst  all_pred_tst\n",
       " 0             5     36.0     16.005220\n",
       " 1             6     35.0      1.357693\n",
       " 2             7      1.0      0.019771\n",
       " 3             8      5.0      1.168173\n",
       " 4             9      8.0      0.360320\n",
       " 5            18      2.0      0.312223\n",
       " 6            31     26.0      3.514888\n",
       " 7            33      1.0      0.111078\n",
       " 8            41     40.0     17.914064\n",
       " 9            43    101.0      7.369082\n",
       " 10           57      0.0      0.804607\n",
       " 11           59      0.0      0.169936\n",
       " 12           85      0.0      0.169134\n",
       " 13          119      0.0      0.060930\n",
       " 14          122      0.0      0.105334\n",
       " 15          130      0.0      0.029571\n",
       " 16          131      0.0      0.037860\n",
       " 17          143      0.0      0.029638\n",
       " 18          156      0.0      0.053849\n",
       " 19          165      0.0      0.021434\n",
       " 20          166      0.0      0.214093\n",
       " 21          186      0.0      0.009195\n",
       " 22          203      0.0      0.022997\n",
       " 23          204      0.0      0.021036\n",
       " 24          207      0.0      0.107954\n",
       " 25          217      0.0      0.034595\n",
       " 26          230      0.0      0.023577\n",
       " 27          231      0.0      0.083405\n",
       " 28          234      0.0      0.008479\n",
       " 29          238      0.0      0.020092\n",
       " ...         ...      ...           ...\n",
       " 24513      1861      0.0      0.016700\n",
       " 24514      1870      0.0      0.040685\n",
       " 24515      1890      0.0      0.126902\n",
       " 24516      1942      0.0      0.005625\n",
       " 24517      1957      0.0      0.005095\n",
       " 24518      1958      0.0      0.017680\n",
       " 24519      1966      0.0      0.003645\n",
       " 24520      1973      0.0      0.010163\n",
       " 24521      1994      0.0      0.015978\n",
       " 24522      1996      0.0      0.013618\n",
       " 24523      1997      0.0      0.074629\n",
       " 24524      2002      0.0      0.008843\n",
       " 24525      2007      0.0      0.047357\n",
       " 24526      2012      0.0      0.004064\n",
       " 24527      2032      0.0      0.144229\n",
       " 24528      2042      0.0      0.043040\n",
       " 24529      2051      0.0      0.040552\n",
       " 24530      2070      0.0      0.014525\n",
       " 24531      2071      0.0      0.034900\n",
       " 24532      2072      0.0      1.497156\n",
       " 24533      2083      0.0      0.047064\n",
       " 24534      2098      0.0      0.011140\n",
       " 24535      2099      0.0      0.066064\n",
       " 24536      2113      0.0      4.736392\n",
       " 24537      2120      0.0      0.394358\n",
       " 24538      2126      0.0      0.041315\n",
       " 24539      2130      0.0      0.228343\n",
       " 24540      2162      0.0      0.023815\n",
       " 24541      2163      0.0      0.022382\n",
       " 24542      2169      0.0      0.084426\n",
       " \n",
       " [24543 rows x 3 columns], 'rf_trn':       row_names  all_trn  all_pred_trn\n",
       " 0             1    101.0  8.669317e+01\n",
       " 1             2      2.0  9.853333e-01\n",
       " 2             3    110.0  1.083062e+02\n",
       " 3             4    105.0  9.547217e+01\n",
       " 4            10      2.0  4.698167e+00\n",
       " 5            11     39.0  2.602500e+01\n",
       " 6            12     18.0  1.111183e+01\n",
       " 7            13      1.0  7.846667e-01\n",
       " 8            14      5.0  2.565500e+00\n",
       " 9            15     31.0  2.933783e+01\n",
       " 10           16      3.0  1.519000e+00\n",
       " 11           17      1.0  7.441667e-01\n",
       " 12           19      1.0  4.880000e-01\n",
       " 13           20      4.0  5.524833e+00\n",
       " 14           21     10.0  8.021833e+00\n",
       " 15           22      7.0  1.061550e+01\n",
       " 16           23     10.0  1.106900e+01\n",
       " 17           24      1.0  6.356667e-01\n",
       " 18           26     16.0  1.380883e+01\n",
       " 19           27    159.0  1.252338e+02\n",
       " 20           28      1.0  5.973333e-01\n",
       " 21           29      3.0  1.889000e+00\n",
       " 22           30      6.0  3.765167e+00\n",
       " 23           32      5.0  1.028150e+01\n",
       " 24           34      8.0  1.027550e+01\n",
       " 25           35      6.0  5.982333e+00\n",
       " 26           36      2.0  1.132500e+00\n",
       " 27           37      9.0  4.922833e+00\n",
       " 28           40      8.0  1.330717e+01\n",
       " 29           42     11.0  1.185367e+01\n",
       " ...         ...      ...           ...\n",
       " 57439      2049      0.0 -3.097522e-15\n",
       " 57440      2050      0.0 -3.070877e-15\n",
       " 57441      2057      0.0 -2.049472e-15\n",
       " 57442      2058      0.0  6.000000e-02\n",
       " 57443      2059      0.0 -2.977618e-15\n",
       " 57444      2062      0.0 -3.108624e-15\n",
       " 57445      2063      0.0 -3.097522e-15\n",
       " 57446      2067      0.0 -2.997602e-15\n",
       " 57447      2078      0.0 -2.904343e-15\n",
       " 57448      2080      0.0 -2.993161e-15\n",
       " 57449      2085      0.0 -3.093081e-15\n",
       " 57450      2086      0.0 -2.837730e-15\n",
       " 57451      2087      0.0  3.000000e-02\n",
       " 57452      2088      0.0 -1.694200e-15\n",
       " 57453      2090      0.0 -2.835510e-15\n",
       " 57454      2091      0.0 -3.079759e-15\n",
       " 57455      2092      0.0 -3.033129e-15\n",
       " 57456      2095      0.0 -3.093081e-15\n",
       " 57457      2100      0.0 -2.235989e-15\n",
       " 57458      2101      0.0  1.000000e-02\n",
       " 57459      2111      0.0  3.980000e-01\n",
       " 57460      2112      0.0 -1.558753e-15\n",
       " 57461      2116      0.0 -2.091660e-15\n",
       " 57462      2119      0.0 -1.993961e-15\n",
       " 57463      2132      0.0 -3.022027e-15\n",
       " 57464      2143      0.0 -2.533529e-15\n",
       " 57465      2147      0.0 -3.097522e-15\n",
       " 57466      2164      0.0  1.510333e+00\n",
       " 57467      2166      0.0 -2.915446e-15\n",
       " 57468      2167      0.0 -2.247091e-15\n",
       " \n",
       " [57469 rows x 3 columns], 'rf_tst':       row_names  all_tst  all_pred_tst\n",
       " 0             5     36.0  4.886733e+01\n",
       " 1             6     35.0  8.086000e+00\n",
       " 2             7      1.0 -5.926371e-15\n",
       " 3             8      5.0  3.775000e-01\n",
       " 4             9      8.0  1.580000e-01\n",
       " 5            18      2.0  1.141083e+01\n",
       " 6            31     26.0  3.024400e+01\n",
       " 7            33      1.0  3.000000e-02\n",
       " 8            41     40.0  5.570400e+01\n",
       " 9            43    101.0  8.506633e+01\n",
       " 10           57      0.0  1.008667e+00\n",
       " 11           59      0.0 -2.275957e-15\n",
       " 12           85      0.0  1.792000e+00\n",
       " 13          119      0.0 -3.814726e-15\n",
       " 14          122      0.0  9.000000e-02\n",
       " 15          130      0.0 -5.744294e-15\n",
       " 16          131      0.0 -5.988543e-15\n",
       " 17          143      0.0  2.500000e-02\n",
       " 18          156      0.0  9.000000e-02\n",
       " 19          165      0.0 -4.249934e-15\n",
       " 20          166      0.0  2.800000e-01\n",
       " 21          186      0.0  1.340000e-01\n",
       " 22          203      0.0 -4.987122e-15\n",
       " 23          204      0.0 -4.596323e-15\n",
       " 24          207      0.0 -3.288481e-15\n",
       " 25          217      0.0 -5.950795e-15\n",
       " 26          230      0.0 -6.199485e-15\n",
       " 27          231      0.0  2.950000e-02\n",
       " 28          234      0.0 -6.388223e-15\n",
       " 29          238      0.0 -6.330492e-15\n",
       " ...         ...      ...           ...\n",
       " 24513      1861      0.0 -2.857714e-15\n",
       " 24514      1870      0.0 -3.097522e-15\n",
       " 24515      1890      0.0 -2.933209e-15\n",
       " 24516      1942      0.0  2.600000e-01\n",
       " 24517      1957      0.0 -2.551293e-15\n",
       " 24518      1958      0.0  1.000000e-02\n",
       " 24519      1966      0.0 -2.338130e-15\n",
       " 24520      1973      0.0 -2.755574e-15\n",
       " 24521      1994      0.0 -3.108624e-15\n",
       " 24522      1996      0.0  1.358333e-01\n",
       " 24523      1997      0.0  1.000000e-02\n",
       " 24524      2002      0.0  2.366667e-01\n",
       " 24525      2007      0.0 -2.660094e-15\n",
       " 24526      2012      0.0  2.000000e-02\n",
       " 24527      2032      0.0  6.533333e-02\n",
       " 24528      2042      0.0  8.500000e-02\n",
       " 24529      2051      0.0 -2.324807e-15\n",
       " 24530      2070      0.0 -3.093081e-15\n",
       " 24531      2071      0.0 -3.108624e-15\n",
       " 24532      2072      0.0  3.640000e-01\n",
       " 24533      2083      0.0 -3.044232e-15\n",
       " 24534      2098      0.0 -2.891021e-15\n",
       " 24535      2099      0.0  2.000000e-02\n",
       " 24536      2113      0.0  2.043333e+00\n",
       " 24537      2120      0.0  3.106667e-01\n",
       " 24538      2126      0.0  1.286667e-01\n",
       " 24539      2130      0.0  3.186667e-01\n",
       " 24540      2162      0.0 -2.084999e-15\n",
       " 24541      2163      0.0  5.000000e-03\n",
       " 24542      2169      0.0  1.300000e+00\n",
       " \n",
       " [24543 rows x 3 columns]}"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tables_rev2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "ps_prd = tables_rev2['ps_tst']['all_pred_tst']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Percent above 159"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.41559711526708226"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(ps_prd > 159).sum()/float(len(ps_prd)) * 100"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Percent above 159 when prediction is at least 0.5 (rounds to 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5.8553386911595871"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(ps_prd > 159).sum()/float((ps_prd>0.5).sum()) * 100"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "### predictions when true flooding is 31"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "29.2925403801\n",
      "91.6333979186\n",
      "3.50829001951\n"
     ]
    }
   ],
   "source": [
    "ps_tst = tables_rev2['ps_tst']\n",
    "pred_31 = ps_tst[(ps_tst['all_tst']==31) & (ps_tst['all_pred_tst']<159)]['all_pred_tst']\n",
    "print pred_31.mean()\n",
    "print pred_31.max()\n",
    "print pred_31.min()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}