SurveyResultAnalysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import pandas as pd\n",
    "import sklearn\n",
    "from sklearn import preprocessing\n",
    "from sklearn import cluster\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import numpy as np\n",
    "from scipy.cluster.hierarchy import dendrogram, linkage\n",
    "from pathlib import Path\n",
    "import librosa, librosa.display\n",
    "import IPython.lib.display as ipd\n",
    "\n",
    "\n",
    "### Setup Seaborn\n",
    "sns.set_style(\"whitegrid\")\n",
    "sns.set_context(\"poster\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "results = pd.read_csv('./surveyresult/results.csv', dtype={'happiness': str}, encoding='latin-1')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "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>file</th>\n",
       "      <th>happiness</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>61</td>\n",
       "      <td>61</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>61</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>beethoven_les_adieux_1.mp3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>1</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                              file happiness\n",
       "count                           61        61\n",
       "unique                          61         5\n",
       "top     beethoven_les_adieux_1.mp3         3\n",
       "freq                             1        21"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['2', '4', '3', '1', '5'], dtype=object)"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results.happiness.unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Using three class labels: sad, neutral and happy\n",
    "di = {'1': 'sad', '2': 'sad', '3': 'neutral', '4':'happy', '5':'happy'}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Using two class labels: sad and happy\n",
    "#di = {'1': 'sad', '2': 'sad', '3': 'happy', '4':'happy', '5':'happy'}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "results = results.replace({\"happiness\": di})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "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>file</th>\n",
       "      <th>happiness</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>beethoven_hammerklavier_3.mp3</td>\n",
       "      <td>sad</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>islamei.mp3</td>\n",
       "      <td>happy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>waldstein_3.mp3</td>\n",
       "      <td>happy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>alb_se1.mp3</td>\n",
       "      <td>happy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>beethoven_les_adieux_1.mp3</td>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                            file happiness\n",
       "0  beethoven_hammerklavier_3.mp3       sad\n",
       "1                    islamei.mp3     happy\n",
       "2                waldstein_3.mp3     happy\n",
       "3                    alb_se1.mp3     happy\n",
       "4     beethoven_les_adieux_1.mp3   neutral"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "musicdf = pd.read_csv('./musicdf30.csv', index_col=0, encoding='latin-1')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "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>composer</th>\n",
       "      <th>description</th>\n",
       "      <th>symphony</th>\n",
       "      <th>file</th>\n",
       "      <th>part</th>\n",
       "      <th>tempo</th>\n",
       "      <th>bpm</th>\n",
       "      <th>duration</th>\n",
       "      <th>zero_crossings</th>\n",
       "      <th>spectral_centroid_avg</th>\n",
       "      <th>...</th>\n",
       "      <th>Chroma_SD_B</th>\n",
       "      <th>Chroma_SD_C</th>\n",
       "      <th>Chroma_SD_C#</th>\n",
       "      <th>Chroma_SD_D</th>\n",
       "      <th>Chroma_SD_D#</th>\n",
       "      <th>Chroma_SD_E</th>\n",
       "      <th>Chroma_SD_F</th>\n",
       "      <th>Chroma_SD_F#</th>\n",
       "      <th>Chroma_SD_G</th>\n",
       "      <th>Chroma_SD_G#</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>albeniz</td>\n",
       "      <td>Albeniz  Isaac(1860-1909)</td>\n",
       "      <td>Espana Opus 165</td>\n",
       "      <td>alb_esp1.mp3</td>\n",
       "      <td>Prelude</td>\n",
       "      <td>Andantino</td>\n",
       "      <td>120.185320</td>\n",
       "      <td>93.570612</td>\n",
       "      <td>98094</td>\n",
       "      <td>621.574748</td>\n",
       "      <td>...</td>\n",
       "      <td>0.175779</td>\n",
       "      <td>0.193223</td>\n",
       "      <td>0.195864</td>\n",
       "      <td>0.210012</td>\n",
       "      <td>0.152202</td>\n",
       "      <td>0.207500</td>\n",
       "      <td>0.167383</td>\n",
       "      <td>0.110378</td>\n",
       "      <td>0.088626</td>\n",
       "      <td>0.167561</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>albeniz</td>\n",
       "      <td>Albeniz  Isaac(1860-1909)</td>\n",
       "      <td>Espana Opus 165</td>\n",
       "      <td>alb_esp2.mp3</td>\n",
       "      <td>Tango</td>\n",
       "      <td>Andantino</td>\n",
       "      <td>120.185320</td>\n",
       "      <td>128.705306</td>\n",
       "      <td>110224</td>\n",
       "      <td>542.039355</td>\n",
       "      <td>...</td>\n",
       "      <td>0.197480</td>\n",
       "      <td>0.148901</td>\n",
       "      <td>0.147744</td>\n",
       "      <td>0.189628</td>\n",
       "      <td>0.138035</td>\n",
       "      <td>0.170995</td>\n",
       "      <td>0.157763</td>\n",
       "      <td>0.167537</td>\n",
       "      <td>0.135463</td>\n",
       "      <td>0.170640</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>albeniz</td>\n",
       "      <td>Albeniz  Isaac(1860-1909)</td>\n",
       "      <td>Espana Opus 165</td>\n",
       "      <td>alb_esp3.mp3</td>\n",
       "      <td>Malaguena</td>\n",
       "      <td>Allegretto</td>\n",
       "      <td>117.453835</td>\n",
       "      <td>176.039184</td>\n",
       "      <td>214944</td>\n",
       "      <td>701.140112</td>\n",
       "      <td>...</td>\n",
       "      <td>0.169860</td>\n",
       "      <td>0.132680</td>\n",
       "      <td>0.120317</td>\n",
       "      <td>0.154430</td>\n",
       "      <td>0.158160</td>\n",
       "      <td>0.156281</td>\n",
       "      <td>0.122567</td>\n",
       "      <td>0.140689</td>\n",
       "      <td>0.124910</td>\n",
       "      <td>0.123183</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>albeniz</td>\n",
       "      <td>Albeniz  Isaac(1860-1909)</td>\n",
       "      <td>Espana Opus 165</td>\n",
       "      <td>alb_esp4.mp3</td>\n",
       "      <td>Serenata</td>\n",
       "      <td>Allegretto</td>\n",
       "      <td>112.347147</td>\n",
       "      <td>195.422041</td>\n",
       "      <td>203680</td>\n",
       "      <td>631.094119</td>\n",
       "      <td>...</td>\n",
       "      <td>0.166632</td>\n",
       "      <td>0.175509</td>\n",
       "      <td>0.142637</td>\n",
       "      <td>0.183892</td>\n",
       "      <td>0.158628</td>\n",
       "      <td>0.137726</td>\n",
       "      <td>0.137628</td>\n",
       "      <td>0.148855</td>\n",
       "      <td>0.166665</td>\n",
       "      <td>0.141932</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>albeniz</td>\n",
       "      <td>Albeniz  Isaac(1860-1909)</td>\n",
       "      <td>Espana Opus 165</td>\n",
       "      <td>alb_esp5.mp3</td>\n",
       "      <td>Capricho Catalan</td>\n",
       "      <td>Allegretto</td>\n",
       "      <td>120.185320</td>\n",
       "      <td>124.499592</td>\n",
       "      <td>118904</td>\n",
       "      <td>570.265597</td>\n",
       "      <td>...</td>\n",
       "      <td>0.119192</td>\n",
       "      <td>0.095741</td>\n",
       "      <td>0.103875</td>\n",
       "      <td>0.100934</td>\n",
       "      <td>0.081791</td>\n",
       "      <td>0.107883</td>\n",
       "      <td>0.133893</td>\n",
       "      <td>0.125197</td>\n",
       "      <td>0.166297</td>\n",
       "      <td>0.131250</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 59 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  composer                description         symphony          file  \\\n",
       "0  albeniz  Albeniz  Isaac(1860-1909)  Espana Opus 165  alb_esp1.mp3   \n",
       "1  albeniz  Albeniz  Isaac(1860-1909)  Espana Opus 165  alb_esp2.mp3   \n",
       "2  albeniz  Albeniz  Isaac(1860-1909)  Espana Opus 165  alb_esp3.mp3   \n",
       "3  albeniz  Albeniz  Isaac(1860-1909)  Espana Opus 165  alb_esp4.mp3   \n",
       "4  albeniz  Albeniz  Isaac(1860-1909)  Espana Opus 165  alb_esp5.mp3   \n",
       "\n",
       "               part       tempo         bpm    duration  zero_crossings  \\\n",
       "0           Prelude   Andantino  120.185320   93.570612           98094   \n",
       "1             Tango   Andantino  120.185320  128.705306          110224   \n",
       "2         Malaguena  Allegretto  117.453835  176.039184          214944   \n",
       "3          Serenata  Allegretto  112.347147  195.422041          203680   \n",
       "4  Capricho Catalan  Allegretto  120.185320  124.499592          118904   \n",
       "\n",
       "   spectral_centroid_avg      ...       Chroma_SD_B  Chroma_SD_C  \\\n",
       "0             621.574748      ...          0.175779     0.193223   \n",
       "1             542.039355      ...          0.197480     0.148901   \n",
       "2             701.140112      ...          0.169860     0.132680   \n",
       "3             631.094119      ...          0.166632     0.175509   \n",
       "4             570.265597      ...          0.119192     0.095741   \n",
       "\n",
       "   Chroma_SD_C#  Chroma_SD_D  Chroma_SD_D#  Chroma_SD_E  Chroma_SD_F  \\\n",
       "0      0.195864     0.210012      0.152202     0.207500     0.167383   \n",
       "1      0.147744     0.189628      0.138035     0.170995     0.157763   \n",
       "2      0.120317     0.154430      0.158160     0.156281     0.122567   \n",
       "3      0.142637     0.183892      0.158628     0.137726     0.137628   \n",
       "4      0.103875     0.100934      0.081791     0.107883     0.133893   \n",
       "\n",
       "   Chroma_SD_F#  Chroma_SD_G  Chroma_SD_G#  \n",
       "0      0.110378     0.088626      0.167561  \n",
       "1      0.167537     0.135463      0.170640  \n",
       "2      0.140689     0.124910      0.123183  \n",
       "3      0.148855     0.166665      0.141932  \n",
       "4      0.125197     0.166297      0.131250  \n",
       "\n",
       "[5 rows x 59 columns]"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "musicdf.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(61, 2)"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(61, 59)"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "musicdf.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "results = pd.merge(results, musicdf, on='file')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['file', 'happiness', 'composer', 'description', 'symphony', 'part',\n",
       "       'tempo', 'bpm', 'duration', 'zero_crossings', 'spectral_centroid_avg',\n",
       "       'spectral_bw_avg', 'Chroma_A', 'Chroma_A#', 'Chroma_B', 'Chroma_C',\n",
       "       'Chroma_C#', 'Chroma_D', 'Chroma_D#', 'Chroma_E', 'Chroma_F',\n",
       "       'Chroma_F#', 'Chroma_G', 'Chroma_G#', 'MFCC_0', 'MFCC_1', 'MFCC_10',\n",
       "       'MFCC_11', 'MFCC_2', 'MFCC_3', 'MFCC_4', 'MFCC_5', 'MFCC_6', 'MFCC_7',\n",
       "       'MFCC_8', 'MFCC_9', 'MFCC_SD_0', 'MFCC_SD_1', 'MFCC_SD_10',\n",
       "       'MFCC_SD_11', 'MFCC_SD_2', 'MFCC_SD_3', 'MFCC_SD_4', 'MFCC_SD_5',\n",
       "       'MFCC_SD_6', 'MFCC_SD_7', 'MFCC_SD_8', 'MFCC_SD_9', 'Chroma_SD_A',\n",
       "       'Chroma_SD_A#', 'Chroma_SD_B', 'Chroma_SD_C', 'Chroma_SD_C#',\n",
       "       'Chroma_SD_D', 'Chroma_SD_D#', 'Chroma_SD_E', 'Chroma_SD_F',\n",
       "       'Chroma_SD_F#', 'Chroma_SD_G', 'Chroma_SD_G#'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#mfcc_columns = (['MFCC_0', 'MFCC_1', 'MFCC_10',\n",
    "#       'MFCC_11', 'MFCC_2', 'MFCC_3', 'MFCC_4', 'MFCC_5', 'MFCC_6', 'MFCC_7',\n",
    "#       'MFCC_8', 'MFCC_9', 'MFCC_SD_0', 'MFCC_SD_1', 'MFCC_SD_10',\n",
    "#       'MFCC_SD_11', 'MFCC_SD_2', 'MFCC_SD_3', 'MFCC_SD_4', 'MFCC_SD_5',\n",
    "#       'MFCC_SD_6', 'MFCC_SD_7', 'MFCC_SD_8', 'MFCC_SD_9']);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "feature_cols = results.columns.drop(['composer', 'description', 'symphony', 'tempo', 'file', 'part', 'duration']);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df = results[feature_cols]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['happiness', 'bpm', 'zero_crossings', 'spectral_centroid_avg',\n",
       "       'spectral_bw_avg', 'Chroma_A', 'Chroma_A#', 'Chroma_B', 'Chroma_C',\n",
       "       'Chroma_C#', 'Chroma_D', 'Chroma_D#', 'Chroma_E', 'Chroma_F',\n",
       "       'Chroma_F#', 'Chroma_G', 'Chroma_G#', 'MFCC_0', 'MFCC_1', 'MFCC_10',\n",
       "       'MFCC_11', 'MFCC_2', 'MFCC_3', 'MFCC_4', 'MFCC_5', 'MFCC_6', 'MFCC_7',\n",
       "       'MFCC_8', 'MFCC_9', 'MFCC_SD_0', 'MFCC_SD_1', 'MFCC_SD_10',\n",
       "       'MFCC_SD_11', 'MFCC_SD_2', 'MFCC_SD_3', 'MFCC_SD_4', 'MFCC_SD_5',\n",
       "       'MFCC_SD_6', 'MFCC_SD_7', 'MFCC_SD_8', 'MFCC_SD_9', 'Chroma_SD_A',\n",
       "       'Chroma_SD_A#', 'Chroma_SD_B', 'Chroma_SD_C', 'Chroma_SD_C#',\n",
       "       'Chroma_SD_D', 'Chroma_SD_D#', 'Chroma_SD_E', 'Chroma_SD_F',\n",
       "       'Chroma_SD_F#', 'Chroma_SD_G', 'Chroma_SD_G#'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X, y = df.loc[:, df.columns != 'happiness'].values, df.loc[:, 'happiness'].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.cross_validation import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(45, 52)"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(16, 52)"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "forest = RandomForestClassifier(n_estimators=10000, random_state=0, n_jobs=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
       "            max_depth=None, max_features='auto', max_leaf_nodes=None,\n",
       "            min_impurity_split=1e-07, min_samples_leaf=1,\n",
       "            min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
       "            n_estimators=10000, n_jobs=-1, oob_score=False, random_state=0,\n",
       "            verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "forest.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "y_train_pred = forest.predict(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "y_test_pred = forest.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "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>preds</th>\n",
       "      <th>neutral</th>\n",
       "      <th>sad</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>actual</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>happy</th>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>neutral</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sad</th>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>6</td>\n",
       "      <td>10</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "preds    neutral  sad  All\n",
       "actual                    \n",
       "happy          4    3    7\n",
       "neutral        1    3    4\n",
       "sad            1    4    5\n",
       "All            6   10   16"
      ]
     },
     "execution_count": 93,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(y_test, y_test_pred, rownames=['actual'], colnames=['preds'],margins=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean accuracy score: 0.312\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "accuracy = accuracy_score(y_test, y_test_pred)\n",
    "print(f'Mean accuracy score: {accuracy:.3}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['happy', 'happy', 'sad', 'neutral', 'neutral', 'happy', 'happy',\n",
       "       'neutral', 'happy', 'sad', 'neutral', 'happy', 'sad', 'sad', 'sad',\n",
       "       'happy'], dtype=object)"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1d8f71ecf98>"
      ]
     },
     "execution_count": 96,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtcAAAIiCAYAAAAD0eT3AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nuySB5wZlAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1d8f71ec160>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "cm = pd.DataFrame(confusion_matrix(y_test, y_test_pred), columns=['happy','neutral','sad'], index=['happy','neutral','sad']).rename_axis('Actual').rename_axis(\"Predicted\", axis=\"columns\")\n",
    "sns.heatmap(cm, annot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "importances = forest.feature_importances_\n",
    "std = np.std([tree.feature_importances_ for tree in forest.estimators_],\n",
    "             axis=0)\n",
    "indices = np.argsort(importances)[::-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "total_features = X_test.shape[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "52"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "forest.n_features_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "column_names = df.loc[:, df.columns != 'happiness'].columns.tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Feature ranking:\n",
      "1. feature spectral_centroid_avg (0.048722)\n",
      "2. feature MFCC_2 (0.042173)\n",
      "3. feature bpm (0.037655)\n",
      "4. feature Chroma_SD_C (0.035174)\n",
      "5. feature MFCC_0 (0.032522)\n",
      "6. feature MFCC_1 (0.030969)\n",
      "7. feature MFCC_4 (0.029513)\n",
      "8. feature MFCC_SD_3 (0.027937)\n",
      "9. feature MFCC_3 (0.027018)\n",
      "10. feature Chroma_C# (0.025312)\n",
      "11. feature Chroma_SD_C# (0.025057)\n",
      "12. feature Chroma_SD_D# (0.023877)\n",
      "13. feature MFCC_SD_6 (0.023525)\n",
      "14. feature Chroma_SD_E (0.022699)\n",
      "15. feature MFCC_5 (0.021270)\n",
      "16. feature Chroma_SD_G# (0.021170)\n",
      "17. feature MFCC_6 (0.019858)\n",
      "18. feature zero_crossings (0.019687)\n",
      "19. feature Chroma_SD_F# (0.018897)\n",
      "20. feature spectral_bw_avg (0.018896)\n",
      "21. feature Chroma_SD_A# (0.018816)\n",
      "22. feature MFCC_SD_4 (0.018719)\n",
      "23. feature MFCC_SD_7 (0.017353)\n",
      "24. feature MFCC_10 (0.017268)\n",
      "25. feature Chroma_G (0.017235)\n",
      "26. feature MFCC_SD_5 (0.016779)\n",
      "27. feature MFCC_11 (0.016752)\n",
      "28. feature MFCC_9 (0.016119)\n",
      "29. feature Chroma_SD_B (0.016101)\n",
      "30. feature MFCC_7 (0.016058)\n",
      "31. feature Chroma_SD_A (0.014894)\n",
      "32. feature Chroma_SD_G (0.014631)\n",
      "33. feature Chroma_F# (0.014582)\n",
      "34. feature MFCC_8 (0.014329)\n",
      "35. feature Chroma_D# (0.014063)\n",
      "36. feature MFCC_SD_2 (0.013949)\n",
      "37. feature Chroma_C (0.013634)\n",
      "38. feature Chroma_F (0.013175)\n",
      "39. feature MFCC_SD_9 (0.013161)\n",
      "40. feature Chroma_A (0.012737)\n",
      "41. feature Chroma_SD_D (0.012734)\n",
      "42. feature MFCC_SD_0 (0.012226)\n",
      "43. feature MFCC_SD_10 (0.012174)\n",
      "44. feature Chroma_B (0.012011)\n",
      "45. feature Chroma_SD_F (0.011932)\n",
      "46. feature Chroma_A# (0.011908)\n",
      "47. feature MFCC_SD_1 (0.011899)\n",
      "48. feature MFCC_SD_11 (0.011541)\n",
      "49. feature MFCC_SD_8 (0.011404)\n",
      "50. feature Chroma_D (0.011245)\n",
      "51. feature Chroma_G# (0.009661)\n",
      "52. feature Chroma_E (0.008979)\n"
     ]
    }
   ],
   "source": [
    "print(\"Feature ranking:\")\n",
    "\n",
    "for f in range(forest.n_features_):\n",
    "    print(\"%d. feature %s (%f)\" % (f + 1, column_names[indices[f]], importances[indices[f]]))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "column_names = np.array(column_names)[indices].tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAwIAAAKpCAYAAAAPEJshAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1d8f7302d68>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure()\n",
    "plt.title(\"Feature importances\")\n",
    "plt.bar(range(forest.n_features_), importances[indices],\n",
    "       color=\"r\",  align=\"center\")\n",
    "plt.xticks(range(forest.n_features_), column_names, rotation='vertical')\n",
    "plt.xlim([-1, forest.n_features_])\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}