TP_Intent_Recognition.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "TP_Intent-Recognition.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "[View in Colaboratory](https://colab.research.google.com/github/youqad/Neurorobotics_Intent-Recognition/blob/master/TP_Intent_Recognition.ipynb)"
      ]
    },
    {
      "metadata": {
        "id": "WlbfbUYq3Rul",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "# TP 4: Intent recognition"
      ]
    },
    {
      "metadata": {
        "id": "XSTQ2MPC3bCc",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "# *Exercise 1*: Automatic detection of speaker’s intention from supra-segmental features\n",
        "\n",
        "The aim of this exercice is to develop a human feedback classifier : positive (approval) / negative (prohibition). This classifier might be used to teach robots and/or to guide\n",
        "robot’s learning."
      ]
    },
    {
      "metadata": {
        "id": "NxSsfl7ggNDn",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "import urllib.request\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "from google.colab import files as google_files\n",
        "\n",
        "import itertools\n",
        "import matplotlib.pyplot as plt"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "LL73eFB-I4y7",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 462
        },
        "outputId": "fa0e67ba-10f0-4b07-db6b-5e9c13ecd680"
      },
      "cell_type": "code",
      "source": [
        "!pip install ggplot\n",
        "\n",
        "import ggplot"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting ggplot\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/48/04/5c88cc51c6713583f2dc78a5296adb9741505348c323d5875bc976143db2/ggplot-0.11.5-py2.py3-none-any.whl (2.2MB)\n",
            "\u001b[K    100% |████████████████████████████████| 2.2MB 8.3MB/s \n",
            "\u001b[?25hRequirement already satisfied: scipy in /usr/local/lib/python3.6/dist-packages (from ggplot) (0.19.1)\n",
            "Requirement already satisfied: patsy>=0.4 in /usr/local/lib/python3.6/dist-packages (from ggplot) (0.5.0)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.6/dist-packages (from ggplot) (1.14.3)\n",
            "Requirement already satisfied: matplotlib in /usr/local/lib/python3.6/dist-packages (from ggplot) (2.1.2)\n",
            "Requirement already satisfied: cycler in /usr/local/lib/python3.6/dist-packages (from ggplot) (0.10.0)\n",
            "Requirement already satisfied: six in /usr/local/lib/python3.6/dist-packages (from ggplot) (1.11.0)\n",
            "Requirement already satisfied: statsmodels in /usr/local/lib/python3.6/dist-packages (from ggplot) (0.8.0)\n",
            "Requirement already satisfied: pandas in /usr/local/lib/python3.6/dist-packages (from ggplot) (0.22.0)\n",
            "Collecting brewer2mpl (from ggplot)\n",
            "  Downloading https://files.pythonhosted.org/packages/84/57/00c45a199719e617db0875181134fcb3aeef701deae346547ac722eaaf5e/brewer2mpl-1.4.1-py2.py3-none-any.whl\n",
            "Requirement already satisfied: pytz in /usr/local/lib/python3.6/dist-packages (from matplotlib->ggplot) (2018.4)\n",
            "Requirement already satisfied: python-dateutil>=2.1 in /usr/local/lib/python3.6/dist-packages (from matplotlib->ggplot) (2.5.3)\n",
            "Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /usr/local/lib/python3.6/dist-packages (from matplotlib->ggplot) (2.2.0)\n",
            "Installing collected packages: brewer2mpl, ggplot\n",
            "Successfully installed brewer2mpl-1.4.1 ggplot-0.11.5\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/ggplot/utils.py:81: FutureWarning: pandas.tslib is deprecated and will be removed in a future version.\n",
            "You can access Timestamp as pandas.Timestamp\n",
            "  pd.tslib.Timestamp,\n",
            "/usr/local/lib/python3.6/dist-packages/ggplot/stats/smoothers.py:4: FutureWarning: The pandas.lib module is deprecated and will be removed in a future version. These are private functions and can be accessed from pandas._libs.lib instead\n",
            "  from pandas.lib import Timestamp\n",
            "/usr/local/lib/python3.6/dist-packages/statsmodels/compat/pandas.py:56: FutureWarning: The pandas.core.datetools module is deprecated and will be removed in a future version. Please use the pandas.tseries module instead.\n",
            "  from pandas.core import datetools\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "metadata": {
        "id": "RJeps0G1C7zN",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "def list_from_URL(file_URL, function_applied=None):\n",
        "  lines_bytes = urllib.request.urlopen(file_URL).readlines()\n",
        "  lines = []\n",
        "\n",
        "  for line in lines_bytes:\n",
        "    line = line.decode(\"utf-8\").rstrip()\n",
        "    \n",
        "    if function_applied is not None:\n",
        "      line = function_applied(line)\n",
        "    \n",
        "    lines.append(line)\n",
        "   \n",
        "  return lines\n",
        "\n",
        "def plot_confusion_matrix(cm, classes,\n",
        "                          normalize=True,\n",
        "                          title='Confusion matrix',\n",
        "                          cmap=plt.cm.Blues):\n",
        "    \"\"\"\n",
        "    From: http://scikit-learn.org/stable/auto_examples/model_selection/plot_confusion_matrix.html\n",
        "    This function prints and plots the confusion matrix.\n",
        "    Normalization can be applied by setting `normalize=True`.\n",
        "    \"\"\"\n",
        "    if normalize:\n",
        "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
        "        print(\"Normalized confusion matrix\")\n",
        "    else:\n",
        "        print('Confusion matrix, without normalization')\n",
        "\n",
        "    print(cm)\n",
        "\n",
        "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
        "    plt.title(title)\n",
        "    plt.colorbar()\n",
        "    tick_marks = np.arange(len(classes))\n",
        "    plt.xticks(tick_marks, classes, rotation=45)\n",
        "    plt.yticks(tick_marks, classes)\n",
        "\n",
        "    fmt = '.2f' if normalize else 'd'\n",
        "    thresh = cm.max() / 2.\n",
        "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
        "        plt.text(j, i, format(cm[i, j], fmt),\n",
        "                 horizontalalignment=\"center\",\n",
        "                 color=\"white\" if cm[i, j] > thresh else \"black\")\n",
        "\n",
        "    plt.tight_layout()\n",
        "    plt.ylabel('True label')\n",
        "    plt.xlabel('Predicted label')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "IEeGWQQLeYkp",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## 1. Extraction of prosodic features ($f_0$ and energy)"
      ]
    },
    {
      "metadata": {
        "id": "cFU3YCMjHeqj",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# # /!\\ NO NEED TO EXECUTE THIS CELL AGAIN !!!\n",
        "# \n",
        "# \n",
        "# filenames = list_from_URL('https://raw.githubusercontent.com/youqad/Neurorobotics_Intent-Recognition/master/filenames.txt')\n",
        "# filenames = list(set(filenames))\n",
        "# \n",
        "# files = []\n",
        "# indices = []\n",
        "# \n",
        "# for file in filenames:\n",
        "# \n",
        "#     URL_f0 = 'https://raw.githubusercontent.com/youqad/Neurorobotics_Intent-Recognition/master/data_files/{}.f0'.format(file)\n",
        "#     file_dicts = [{key:val for key, val in zip(['time', 'f0'], map(float, l.split()))} for l in list_from_URL(URL_f0)]\n",
        "# \n",
        "#     URL_en = 'https://raw.githubusercontent.com/youqad/Neurorobotics_Intent-Recognition/master/data_files/{}.en'.format(file)\n",
        "#     for l, d in zip(list_from_URL(URL_en), file_dicts):\n",
        "#       d[\"file\"] = file\n",
        "#       d[\"en\"] = float(l.split()[1])\n",
        "#       d[\"label\"] = file[-2:]\n",
        "# \n",
        "#     files.extend(file_dicts)\n",
        "# \n",
        "# # How `files` looks like:\n",
        "# # files = [ \n",
        "# #           {\"file\": \"cy0001at\", \"time\": 0.02, \"f0\": 0., \"en\": 0.},\n",
        "# #           {\"file\": \"cy0001at\", \"time\": 1.28, \"f0\": 0., \"en\": 0.},\n",
        "# #           ...\n",
        "# #           {\"file\": \"li1450at\", \"time\": 0.02, \"f0\": 0., \"en\": 0.},\n",
        "# #           {\"file\": \"li1450at\", \"time\": 1.56, \"f0\": 404., \"en\": 65.}\n",
        "# #         ]\n",
        "# \n",
        "# pd.DataFrame(files).to_csv('data.csv', encoding='utf-8', index=False) # To reuse it next time\n",
        "# google_files.download('data.csv')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "Uf_9ZVOLjtAH",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 235
        },
        "outputId": "abb07370-43ff-44f3-9bac-557e64b29f67"
      },
      "cell_type": "code",
      "source": [
        "# loading training data\n",
        "df = pd.read_csv('https://raw.githubusercontent.com/youqad/Neurorobotics_Intent-Recognition/master/data.csv').set_index('file')\n",
        "\n",
        "df1 = df.loc[df['label'] != 'at']\n",
        "df1.head()"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>en</th>\n",
              "      <th>f0</th>\n",
              "      <th>label</th>\n",
              "      <th>time</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>file</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>li1377pw</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>pw</td>\n",
              "      <td>0.02</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>li1377pw</th>\n",
              "      <td>39.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>pw</td>\n",
              "      <td>0.04</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>li1377pw</th>\n",
              "      <td>40.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>pw</td>\n",
              "      <td>0.06</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>li1377pw</th>\n",
              "      <td>39.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>pw</td>\n",
              "      <td>0.08</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>li1377pw</th>\n",
              "      <td>39.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>pw</td>\n",
              "      <td>0.10</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "            en   f0 label  time\n",
              "file                           \n",
              "li1377pw   0.0  0.0    pw  0.02\n",
              "li1377pw  39.0  0.0    pw  0.04\n",
              "li1377pw  40.0  0.0    pw  0.06\n",
              "li1377pw  39.0  0.0    pw  0.08\n",
              "li1377pw  39.0  0.0    pw  0.10"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "metadata": {
        "id": "VS-Cg8gjelRO",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## 2. Extraction of functionals (statistics) : mean, maximum, range, variance, median, first quartile, third quartile, mean absolute of local derivate"
      ]
    },
    {
      "metadata": {
        "id": "Oi-TJTPYmDBQ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 252
        },
        "outputId": "3ab973e0-bac2-4434-8f68-ae743241c41b"
      },
      "cell_type": "code",
      "source": [
        "print(df1.columns.values)\n",
        "\n",
        "#df.groupby('file').mean().head()\n",
        "#df1.groupby('file').max().head()\n",
        "#df1.groupby('file').var().head()\n",
        "#df1.groupby('file').median().head()\n",
        "df1.groupby('file').quantile([.25, .75]).head()"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "['en' 'f0' 'label' 'time']\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th>en</th>\n",
              "      <th>f0</th>\n",
              "      <th>time</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>file</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th rowspan=\"2\" valign=\"top\">cy0007pw</th>\n",
              "      <th>0.25</th>\n",
              "      <td>41.00</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.525</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>0.75</th>\n",
              "      <td>66.00</td>\n",
              "      <td>189.5</td>\n",
              "      <td>1.535</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th rowspan=\"2\" valign=\"top\">cy0008pw</th>\n",
              "      <th>0.25</th>\n",
              "      <td>41.00</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.270</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>0.75</th>\n",
              "      <td>64.50</td>\n",
              "      <td>192.0</td>\n",
              "      <td>0.770</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>cy0009pw</th>\n",
              "      <th>0.25</th>\n",
              "      <td>40.75</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.395</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                  en     f0   time\n",
              "file                              \n",
              "cy0007pw 0.25  41.00    0.0  0.525\n",
              "         0.75  66.00  189.5  1.535\n",
              "cy0008pw 0.25  41.00    0.0  0.270\n",
              "         0.75  64.50  192.0  0.770\n",
              "cy0009pw 0.25  40.75    0.0  0.395"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "metadata": {
        "id": "fuQAy00Mshr_",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 286
        },
        "outputId": "39d9e2b8-ce96-4feb-b6e8-93ce1c4f425c"
      },
      "cell_type": "code",
      "source": [
        "list_features  = ['mean', \n",
        "                  'max',\n",
        "                  ('range', lambda x: max(x)-min(x)),\n",
        "                  'var',\n",
        "                  'median',\n",
        "                  ('1st_quantile', lambda x: x.quantile(.25)),\n",
        "                  ('3rd_quantile', lambda x: x.quantile(.75)),\n",
        "                  ('mean_absolute_local_derivate', lambda x: abs(x.diff()).mean())\n",
        "                 ]\n",
        "\n",
        "df1.groupby('file')['f0','en'].agg(list_features).head()"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead tr th {\n",
              "        text-align: left;\n",
              "    }\n",
              "\n",
              "    .dataframe thead tr:last-of-type th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr>\n",
              "      <th></th>\n",
              "      <th colspan=\"8\" halign=\"left\">f0</th>\n",
              "      <th colspan=\"8\" halign=\"left\">en</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th></th>\n",
              "      <th>mean</th>\n",
              "      <th>max</th>\n",
              "      <th>range</th>\n",
              "      <th>var</th>\n",
              "      <th>median</th>\n",
              "      <th>1st_quantile</th>\n",
              "      <th>3rd_quantile</th>\n",
              "      <th>mean_absolute_local_derivate</th>\n",
              "      <th>mean</th>\n",
              "      <th>max</th>\n",
              "      <th>range</th>\n",
              "      <th>var</th>\n",
              "      <th>median</th>\n",
              "      <th>1st_quantile</th>\n",
              "      <th>3rd_quantile</th>\n",
              "      <th>mean_absolute_local_derivate</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>file</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>cy0007pw</th>\n",
              "      <td>92.284314</td>\n",
              "      <td>257.0</td>\n",
              "      <td>257.0</td>\n",
              "      <td>10372.542128</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>189.50</td>\n",
              "      <td>13.683168</td>\n",
              "      <td>52.313725</td>\n",
              "      <td>71.0</td>\n",
              "      <td>71.0</td>\n",
              "      <td>228.455057</td>\n",
              "      <td>52.0</td>\n",
              "      <td>41.00</td>\n",
              "      <td>66.00</td>\n",
              "      <td>2.970297</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>cy0008pw</th>\n",
              "      <td>78.431373</td>\n",
              "      <td>250.0</td>\n",
              "      <td>250.0</td>\n",
              "      <td>9930.090196</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>192.00</td>\n",
              "      <td>26.440000</td>\n",
              "      <td>47.725490</td>\n",
              "      <td>70.0</td>\n",
              "      <td>70.0</td>\n",
              "      <td>321.963137</td>\n",
              "      <td>43.0</td>\n",
              "      <td>41.00</td>\n",
              "      <td>64.50</td>\n",
              "      <td>3.960000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>cy0009pw</th>\n",
              "      <td>69.065789</td>\n",
              "      <td>243.0</td>\n",
              "      <td>243.0</td>\n",
              "      <td>8927.182281</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>182.25</td>\n",
              "      <td>12.853333</td>\n",
              "      <td>49.473684</td>\n",
              "      <td>74.0</td>\n",
              "      <td>74.0</td>\n",
              "      <td>260.839298</td>\n",
              "      <td>42.0</td>\n",
              "      <td>40.75</td>\n",
              "      <td>66.00</td>\n",
              "      <td>3.520000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>cy0010pw</th>\n",
              "      <td>29.196078</td>\n",
              "      <td>221.0</td>\n",
              "      <td>221.0</td>\n",
              "      <td>4696.178994</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.00</td>\n",
              "      <td>15.267327</td>\n",
              "      <td>46.049020</td>\n",
              "      <td>77.0</td>\n",
              "      <td>77.0</td>\n",
              "      <td>165.789652</td>\n",
              "      <td>42.0</td>\n",
              "      <td>41.00</td>\n",
              "      <td>50.75</td>\n",
              "      <td>3.306931</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>cy0011pw</th>\n",
              "      <td>110.743590</td>\n",
              "      <td>230.0</td>\n",
              "      <td>230.0</td>\n",
              "      <td>9290.400932</td>\n",
              "      <td>172.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>192.50</td>\n",
              "      <td>7.506494</td>\n",
              "      <td>53.653846</td>\n",
              "      <td>71.0</td>\n",
              "      <td>71.0</td>\n",
              "      <td>258.125375</td>\n",
              "      <td>62.0</td>\n",
              "      <td>41.25</td>\n",
              "      <td>66.00</td>\n",
              "      <td>2.337662</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                  f0                                                  \\\n",
              "                mean    max  range           var median 1st_quantile   \n",
              "file                                                                   \n",
              "cy0007pw   92.284314  257.0  257.0  10372.542128    0.0          0.0   \n",
              "cy0008pw   78.431373  250.0  250.0   9930.090196    0.0          0.0   \n",
              "cy0009pw   69.065789  243.0  243.0   8927.182281    0.0          0.0   \n",
              "cy0010pw   29.196078  221.0  221.0   4696.178994    0.0          0.0   \n",
              "cy0011pw  110.743590  230.0  230.0   9290.400932  172.0          0.0   \n",
              "\n",
              "                                                           en              \\\n",
              "         3rd_quantile mean_absolute_local_derivate       mean   max range   \n",
              "file                                                                        \n",
              "cy0007pw       189.50                    13.683168  52.313725  71.0  71.0   \n",
              "cy0008pw       192.00                    26.440000  47.725490  70.0  70.0   \n",
              "cy0009pw       182.25                    12.853333  49.473684  74.0  74.0   \n",
              "cy0010pw         0.00                    15.267327  46.049020  77.0  77.0   \n",
              "cy0011pw       192.50                     7.506494  53.653846  71.0  71.0   \n",
              "\n",
              "                                                       \\\n",
              "                 var median 1st_quantile 3rd_quantile   \n",
              "file                                                    \n",
              "cy0007pw  228.455057   52.0        41.00        66.00   \n",
              "cy0008pw  321.963137   43.0        41.00        64.50   \n",
              "cy0009pw  260.839298   42.0        40.75        66.00   \n",
              "cy0010pw  165.789652   42.0        41.00        50.75   \n",
              "cy0011pw  258.125375   62.0        41.25        66.00   \n",
              "\n",
              "                                       \n",
              "         mean_absolute_local_derivate  \n",
              "file                                   \n",
              "cy0007pw                     2.970297  \n",
              "cy0008pw                     3.960000  \n",
              "cy0009pw                     3.520000  \n",
              "cy0010pw                     3.306931  \n",
              "cy0011pw                     2.337662  "
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "metadata": {
        "id": "-NwHHkQfe1GF",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## 3. Check functionals for both voiced (i.e. $f_0\\neq 0$) and unvoiced segments. Which segments are suited for the approach ?"
      ]
    },
    {
      "metadata": {
        "id": "58s4COw1Srye",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 286
        },
        "outputId": "45290e78-1e38-4814-99a1-987d14565b27"
      },
      "cell_type": "code",
      "source": [
        "voiced = df1.loc[df1['f0']!=0].groupby('file')['f0','en'].agg(list_features)\n",
        "voiced.head()"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead tr th {\n",
              "        text-align: left;\n",
              "    }\n",
              "\n",
              "    .dataframe thead tr:last-of-type th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr>\n",
              "      <th></th>\n",
              "      <th colspan=\"8\" halign=\"left\">f0</th>\n",
              "      <th colspan=\"8\" halign=\"left\">en</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th></th>\n",
              "      <th>mean</th>\n",
              "      <th>max</th>\n",
              "      <th>range</th>\n",
              "      <th>var</th>\n",
              "      <th>median</th>\n",
              "      <th>1st_quantile</th>\n",
              "      <th>3rd_quantile</th>\n",
              "      <th>mean_absolute_local_derivate</th>\n",
              "      <th>mean</th>\n",
              "      <th>max</th>\n",
              "      <th>range</th>\n",
              "      <th>var</th>\n",
              "      <th>median</th>\n",
              "      <th>1st_quantile</th>\n",
              "      <th>3rd_quantile</th>\n",
              "      <th>mean_absolute_local_derivate</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>file</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>cy0007pw</th>\n",
              "      <td>200.276596</td>\n",
              "      <td>257.0</td>\n",
              "      <td>90.0</td>\n",
              "      <td>675.987049</td>\n",
              "      <td>191.0</td>\n",
              "      <td>182.50</td>\n",
              "      <td>213.00</td>\n",
              "      <td>5.869565</td>\n",
              "      <td>65.914894</td>\n",
              "      <td>71.0</td>\n",
              "      <td>16.0</td>\n",
              "      <td>17.775208</td>\n",
              "      <td>66.0</td>\n",
              "      <td>63.5</td>\n",
              "      <td>70.00</td>\n",
              "      <td>1.652174</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>cy0008pw</th>\n",
              "      <td>200.000000</td>\n",
              "      <td>250.0</td>\n",
              "      <td>83.0</td>\n",
              "      <td>538.421053</td>\n",
              "      <td>198.5</td>\n",
              "      <td>179.50</td>\n",
              "      <td>210.00</td>\n",
              "      <td>10.368421</td>\n",
              "      <td>61.000000</td>\n",
              "      <td>70.0</td>\n",
              "      <td>70.0</td>\n",
              "      <td>242.736842</td>\n",
              "      <td>66.0</td>\n",
              "      <td>61.5</td>\n",
              "      <td>68.00</td>\n",
              "      <td>5.842105</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>cy0009pw</th>\n",
              "      <td>194.407407</td>\n",
              "      <td>243.0</td>\n",
              "      <td>77.0</td>\n",
              "      <td>446.943020</td>\n",
              "      <td>190.0</td>\n",
              "      <td>180.00</td>\n",
              "      <td>209.00</td>\n",
              "      <td>7.192308</td>\n",
              "      <td>67.333333</td>\n",
              "      <td>74.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>17.923077</td>\n",
              "      <td>68.0</td>\n",
              "      <td>66.0</td>\n",
              "      <td>70.00</td>\n",
              "      <td>2.884615</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>cy0010pw</th>\n",
              "      <td>186.125000</td>\n",
              "      <td>221.0</td>\n",
              "      <td>67.0</td>\n",
              "      <td>465.316667</td>\n",
              "      <td>178.5</td>\n",
              "      <td>171.25</td>\n",
              "      <td>204.25</td>\n",
              "      <td>6.466667</td>\n",
              "      <td>65.750000</td>\n",
              "      <td>77.0</td>\n",
              "      <td>25.0</td>\n",
              "      <td>50.466667</td>\n",
              "      <td>64.0</td>\n",
              "      <td>62.0</td>\n",
              "      <td>70.75</td>\n",
              "      <td>4.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>cy0011pw</th>\n",
              "      <td>191.955556</td>\n",
              "      <td>230.0</td>\n",
              "      <td>66.0</td>\n",
              "      <td>314.816162</td>\n",
              "      <td>190.0</td>\n",
              "      <td>179.00</td>\n",
              "      <td>204.00</td>\n",
              "      <td>4.113636</td>\n",
              "      <td>65.288889</td>\n",
              "      <td>71.0</td>\n",
              "      <td>19.0</td>\n",
              "      <td>14.710101</td>\n",
              "      <td>65.0</td>\n",
              "      <td>63.0</td>\n",
              "      <td>68.00</td>\n",
              "      <td>0.954545</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                  f0                                               \\\n",
              "                mean    max range         var median 1st_quantile   \n",
              "file                                                                \n",
              "cy0007pw  200.276596  257.0  90.0  675.987049  191.0       182.50   \n",
              "cy0008pw  200.000000  250.0  83.0  538.421053  198.5       179.50   \n",
              "cy0009pw  194.407407  243.0  77.0  446.943020  190.0       180.00   \n",
              "cy0010pw  186.125000  221.0  67.0  465.316667  178.5       171.25   \n",
              "cy0011pw  191.955556  230.0  66.0  314.816162  190.0       179.00   \n",
              "\n",
              "                                                           en              \\\n",
              "         3rd_quantile mean_absolute_local_derivate       mean   max range   \n",
              "file                                                                        \n",
              "cy0007pw       213.00                     5.869565  65.914894  71.0  16.0   \n",
              "cy0008pw       210.00                    10.368421  61.000000  70.0  70.0   \n",
              "cy0009pw       209.00                     7.192308  67.333333  74.0  20.0   \n",
              "cy0010pw       204.25                     6.466667  65.750000  77.0  25.0   \n",
              "cy0011pw       204.00                     4.113636  65.288889  71.0  19.0   \n",
              "\n",
              "                                                       \\\n",
              "                 var median 1st_quantile 3rd_quantile   \n",
              "file                                                    \n",
              "cy0007pw   17.775208   66.0         63.5        70.00   \n",
              "cy0008pw  242.736842   66.0         61.5        68.00   \n",
              "cy0009pw   17.923077   68.0         66.0        70.00   \n",
              "cy0010pw   50.466667   64.0         62.0        70.75   \n",
              "cy0011pw   14.710101   65.0         63.0        68.00   \n",
              "\n",
              "                                       \n",
              "         mean_absolute_local_derivate  \n",
              "file                                   \n",
              "cy0007pw                     1.652174  \n",
              "cy0008pw                     5.842105  \n",
              "cy0009pw                     2.884615  \n",
              "cy0010pw                     4.000000  \n",
              "cy0011pw                     0.954545  "
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "metadata": {
        "id": "R-XyYtohGW2R",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 235
        },
        "outputId": "7a682a8b-b436-4060-cc1a-f1348acd72da"
      },
      "cell_type": "code",
      "source": [
        "unvoiced = df1.loc[df1['f0']==0].groupby('file')['en'].agg(list_features)\n",
        "unvoiced.head()"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>mean</th>\n",
              "      <th>max</th>\n",
              "      <th>range</th>\n",
              "      <th>var</th>\n",
              "      <th>median</th>\n",
              "      <th>1st_quantile</th>\n",
              "      <th>3rd_quantile</th>\n",
              "      <th>mean_absolute_local_derivate</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>file</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>cy0007pw</th>\n",
              "      <td>40.690909</td>\n",
              "      <td>58.0</td>\n",
              "      <td>58.0</td>\n",
              "      <td>113.550842</td>\n",
              "      <td>41.0</td>\n",
              "      <td>40.5</td>\n",
              "      <td>43.5</td>\n",
              "      <td>3.740741</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>cy0008pw</th>\n",
              "      <td>39.161290</td>\n",
              "      <td>58.0</td>\n",
              "      <td>58.0</td>\n",
              "      <td>189.606452</td>\n",
              "      <td>42.0</td>\n",
              "      <td>41.0</td>\n",
              "      <td>43.0</td>\n",
              "      <td>5.200000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>cy0009pw</th>\n",
              "      <td>39.632653</td>\n",
              "      <td>56.0</td>\n",
              "      <td>56.0</td>\n",
              "      <td>119.570578</td>\n",
              "      <td>41.0</td>\n",
              "      <td>40.0</td>\n",
              "      <td>42.0</td>\n",
              "      <td>3.625000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>cy0010pw</th>\n",
              "      <td>42.383721</td>\n",
              "      <td>68.0</td>\n",
              "      <td>68.0</td>\n",
              "      <td>101.439261</td>\n",
              "      <td>41.0</td>\n",
              "      <td>40.0</td>\n",
              "      <td>43.0</td>\n",
              "      <td>3.105882</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>cy0011pw</th>\n",
              "      <td>37.787879</td>\n",
              "      <td>51.0</td>\n",
              "      <td>51.0</td>\n",
              "      <td>150.922348</td>\n",
              "      <td>41.0</td>\n",
              "      <td>40.0</td>\n",
              "      <td>42.0</td>\n",
              "      <td>4.062500</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "               mean   max  range         var  median  1st_quantile  \\\n",
              "file                                                                 \n",
              "cy0007pw  40.690909  58.0   58.0  113.550842    41.0          40.5   \n",
              "cy0008pw  39.161290  58.0   58.0  189.606452    42.0          41.0   \n",
              "cy0009pw  39.632653  56.0   56.0  119.570578    41.0          40.0   \n",
              "cy0010pw  42.383721  68.0   68.0  101.439261    41.0          40.0   \n",
              "cy0011pw  37.787879  51.0   51.0  150.922348    41.0          40.0   \n",
              "\n",
              "          3rd_quantile  mean_absolute_local_derivate  \n",
              "file                                                  \n",
              "cy0007pw          43.5                      3.740741  \n",
              "cy0008pw          43.0                      5.200000  \n",
              "cy0009pw          42.0                      3.625000  \n",
              "cy0010pw          43.0                      3.105882  \n",
              "cy0011pw          42.0                      4.062500  "
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "metadata": {
        "id": "xEovkR10fBUe",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## 4. Build two databases by randomly extracting examples : Learning database ($60\\%$) and Test database"
      ]
    },
    {
      "metadata": {
        "id": "mVX-LyhLTAgN",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# Question 3\n",
        "\n",
        "## Explanation: A sound can be either voiced, like a vowel, or unvoiced, like a consonant. We want to see which of them contains more intention\n",
        " \n",
        "## ps: change \"data\" by the correct name\n",
        "\n",
        "def train_test(df=df1, train_percentage=.6, seed=1):\n",
        "  \n",
        "  voiced = df.loc[df['f0']!=0].groupby('file')['f0','en'].agg(list_features)\n",
        "  unvoiced = df.loc[df['f0']==0].groupby('file')['en'].agg(list_features)\n",
        "\n",
        "  X, Y = {}, {}\n",
        "\n",
        "  X['voiced'], Y['voiced'] = {}, {}\n",
        "  X['unvoiced'], Y['unvoiced'] = {}, {}\n",
        "\n",
        "\n",
        "  X['voiced']['all'] = np.array(df.groupby('file')['f0','en'].agg(list_features))\n",
        "  Y['voiced']['all'] = np.array(df.loc[df['f0']!=0].groupby(['file']).min().label.values)\n",
        "\n",
        "  X['unvoiced']['all'] = np.array(unvoiced)\n",
        "  Y['unvoiced']['all'] = np.array(df.loc[df['f0']==0].groupby(['file']).min().label.values)\n",
        "  \n",
        "  np.random.seed(seed)\n",
        "  \n",
        "  for type in ['voiced', 'unvoiced']:\n",
        "    n = len(X[type]['all'])\n",
        "    ind_rand = np.random.randint(n, size=int(train_percentage*n)) # random indices\n",
        "    train_mask = np.zeros(n, dtype=bool)\n",
        "    train_mask[ind_rand] = True\n",
        "    X[type]['train'], X[type]['test'] = X[type]['all'][train_mask],  X[type]['all'][~train_mask]\n",
        "    Y[type]['train'], Y[type]['test'] = Y[type]['all'][train_mask],  Y[type]['all'][~train_mask]\n",
        "  \n",
        "  return X, Y\n",
        "\n",
        "X1, Y1 = train_test()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "lAckn6PVgfaf",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 581
        },
        "outputId": "600cabfc-e48a-45ef-f069-26dd9e413294"
      },
      "cell_type": "code",
      "source": [
        "col = ['mean', 'max', 'range', 'var', 'median', '1st_quantile', '3rd_quantile', 'mean_absolute_local_derivate']\n",
        "col = ['f0_'+c for c in col]+['en_'+c for c in col]\n",
        "\n",
        "voi = pd.DataFrame(X1['voiced']['all'], columns=col).assign(label=Y1['voiced']['all'])\n",
        "\n",
        "ggplot.ggplot(voi, ggplot.aes(x='f0_var', y='f0_mean_absolute_local_derivate', color='label')) +\\\n",
        "    ggplot.geom_point() +\\\n",
        "    ggplot.scale_color_brewer(type='qual', palette='Set1') +\\\n",
        "    ggplot.xlab(\"Variance\") + ggplot.ylab(\"Mean absolute of local derivate\") + ggplot.ggtitle(\"Voiced: $f_0$\")"
      ],
      "execution_count": 45,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtQAAAIjCAYAAAAjuoxRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nIQIgCbUQQgghhBABkIRaCCGEEEKIAPx/M9FWr130Q8AAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fa88dbfb278>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<ggplot: (8772619415223)>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 45
        }
      ]
    },
    {
      "metadata": {
        "id": "__zkaXZ8PJQS",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 579
        },
        "outputId": "ada03828-c13e-47d4-bdf0-88edb51958cd"
      },
      "cell_type": "code",
      "source": [
        "col = ['mean', 'max', 'range', 'var', 'median', '1st_quantile', '3rd_quantile', 'mean_absolute_local_derivate']\n",
        "\n",
        "unvoi = pd.DataFrame(X1['unvoiced']['all'], columns=col).assign(label=Y1['unvoiced']['all'])\n",
        "\n",
        "ggplot.ggplot(unvoi, ggplot.aes(x='var', y='mean_absolute_local_derivate', color='label')) +\\\n",
        "    ggplot.geom_point() +\\\n",
        "    ggplot.scale_color_brewer(type='qual', palette='Set1') +\\\n",
        "    ggplot.xlab(\"Variance\") + ggplot.ylab(\"Mean absolute of local derivate\") + ggplot.ggtitle(\"Voiced: $en$\")"
      ],
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtQAAAIhCAYAAABuci1aAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEgAl1CIiIiIiAfj/guLog8uqcm4AAAAASUVORK5CYII=\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fa88db481d0>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<ggplot: (8772619328376)>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 47
        }
      ]
    },
    {
      "metadata": {
        "id": "kCplkLkFrLEy",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## 5. Train a classifer (k-NN method)"
      ]
    },
    {
      "metadata": {
        "id": "CE7xzKUrPrcH",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "89be66b0-7890-434f-de3a-7d012651b5dd"
      },
      "cell_type": "code",
      "source": [
        "# Scikit Learn's kNN classifier:\n",
        "# Just to test, but we will implement it ourselves of course!\n",
        "from sklearn.neighbors import KNeighborsClassifier\n",
        "from sklearn.metrics import accuracy_score\n",
        "\n",
        "def sklearn_knn(k, X, Y):\n",
        "  for type in ['voiced', 'unvoiced']:\n",
        "    kNN = KNeighborsClassifier(n_neighbors=k)\n",
        "    kNN.fit(X[type]['train'], Y[type]['train'])\n",
        "\n",
        "    print(\"Accuracy score for {}: {:.2f}\".format(type, accuracy_score(Y[type]['test'],\n",
        "                                                                  kNN.predict(X[type]['test']))))\n",
        "sklearn_knn(3, X1, Y1)"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Accuracy score for voiced: 0.91\n",
            "Accuracy score for unvoiced: 0.61\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "cWqW53zM2yAT",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 915
        },
        "outputId": "670efcd1-4079-42d9-a4f4-d7b1102991ce"
      },
      "cell_type": "code",
      "source": [
        "# Our own implementation!\n",
        "from scipy.spatial.distance import cdist\n",
        "from sklearn.metrics import confusion_matrix\n",
        "from collections import Counter\n",
        "\n",
        "def kNN(k, X, Y, labels=[\"pw\", \"ap\"]):\n",
        "    # auxiliary function: label prediction (by majority vote)\n",
        "    # based on the nearest neighbors\n",
        "    def predicted_label(ind_neighbors):\n",
        "        label_neighbors = tuple(Y['train'][ind_neighbors])\n",
        "        return Counter(label_neighbors).most_common(1)[0][0]\n",
        "    \n",
        "    # Pairwise distances between test and train data points\n",
        "    dist_matrix = cdist(X['test'], X['train'], 'euclidean')\n",
        "    y_predicted = []\n",
        "\n",
        "    for i in range(len(X['test'])):\n",
        "        ind_k_smallest = np.argpartition(dist_matrix[i, :], k)[:k]\n",
        "        y_predicted.append(predicted_label(ind_k_smallest))\n",
        "    \n",
        "    # Confusion matrix: C[i, j] is the number of observations \n",
        "    # known to be in group i but predicted to be in group j\n",
        "    return confusion_matrix(Y['test'], np.array(y_predicted), labels=labels)\n",
        "\n",
        "plt.figure()\n",
        "cm = kNN(3, X1['voiced'], Y1['voiced'])\n",
        "plot_confusion_matrix(cm, classes=[\"pw\", \"ap\"],\n",
        "                      title='Confusion matrix, without normalization')\n",
        "plt.show()\n",
        "\n",
        "cm2 = kNN(3, X1['unvoiced'], Y1['unvoiced'])\n",
        "plot_confusion_matrix(cm2, classes=[\"pw\", \"ap\"],\n",
        "                      title='Confusion matrix, without normalization')\n",
        "plt.show()"
      ],
      "execution_count": 56,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Normalized confusion matrix\n",
            "[[0.92       0.08      ]\n",
            " [0.09473684 0.90526316]]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAb0AAAGOCAYAAAADwN1HAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f58cbad9668>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Normalized confusion matrix\n",
            "[[0.48648649 0.51351351]\n",
            " [0.22727273 0.77272727]]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAb0AAAGOCAYAAAADwN1HAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAwDYBqEHALANQg8AYBuEHgDANv4/b9pxm603VtcAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f58cbe72240>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "3d1lmW3swpLV",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "# *Exercice 2*: Detection of multiple intents\n",
        "\n",
        "We consider the following intents : \"Approval\", \"Prohibition\" and \"Attention\""
      ]
    },
    {
      "metadata": {
        "id": "juGCnAZsxJcU",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## 1. Extract the prosodic features ($f_0$ and energy) and their functionals"
      ]
    },
    {
      "metadata": {
        "id": "gA2VmRgGaq-N",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 286
        },
        "outputId": "e68c17ca-a000-47be-8b0f-523af438e467"
      },
      "cell_type": "code",
      "source": [
        "# Easy-peasy! All the work has been done before: all we have to do now is to use \n",
        "# the DataFrame `df` instead of `df1`\n",
        "\n",
        "df.groupby('file')['f0','en'].agg(list_features).head()"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead tr th {\n",
              "        text-align: left;\n",
              "    }\n",
              "\n",
              "    .dataframe thead tr:last-of-type th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr>\n",
              "      <th></th>\n",
              "      <th colspan=\"8\" halign=\"left\">f0</th>\n",
              "      <th colspan=\"8\" halign=\"left\">en</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th></th>\n",
              "      <th>mean</th>\n",
              "      <th>max</th>\n",
              "      <th>range</th>\n",
              "      <th>var</th>\n",
              "      <th>median</th>\n",
              "      <th>1st_quantile</th>\n",
              "      <th>3rd_quantile</th>\n",
              "      <th>mean_absolute_local_derivate</th>\n",
              "      <th>mean</th>\n",
              "      <th>max</th>\n",
              "      <th>range</th>\n",
              "      <th>var</th>\n",
              "      <th>median</th>\n",
              "      <th>1st_quantile</th>\n",
              "      <th>3rd_quantile</th>\n",
              "      <th>mean_absolute_local_derivate</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>file</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>cy0001at</th>\n",
              "      <td>110.609375</td>\n",
              "      <td>402.0</td>\n",
              "      <td>402.0</td>\n",
              "      <td>27607.511657</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>331.0</td>\n",
              "      <td>23.777778</td>\n",
              "      <td>47.296875</td>\n",
              "      <td>73.0</td>\n",
              "      <td>73.0</td>\n",
              "      <td>327.323165</td>\n",
              "      <td>41.5</td>\n",
              "      <td>40.0</td>\n",
              "      <td>67.0</td>\n",
              "      <td>3.809524</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>cy0002at</th>\n",
              "      <td>105.640449</td>\n",
              "      <td>430.0</td>\n",
              "      <td>430.0</td>\n",
              "      <td>27108.528345</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>251.0</td>\n",
              "      <td>15.636364</td>\n",
              "      <td>47.337079</td>\n",
              "      <td>76.0</td>\n",
              "      <td>76.0</td>\n",
              "      <td>269.112360</td>\n",
              "      <td>41.0</td>\n",
              "      <td>40.0</td>\n",
              "      <td>62.0</td>\n",
              "      <td>3.272727</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>cy0005at</th>\n",
              "      <td>110.609375</td>\n",
              "      <td>402.0</td>\n",
              "      <td>402.0</td>\n",
              "      <td>27607.511657</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>331.0</td>\n",
              "      <td>23.777778</td>\n",
              "      <td>47.296875</td>\n",
              "      <td>73.0</td>\n",
              "      <td>73.0</td>\n",
              "      <td>327.323165</td>\n",
              "      <td>41.5</td>\n",
              "      <td>40.0</td>\n",
              "      <td>67.0</td>\n",
              "      <td>3.809524</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>cy0006at</th>\n",
              "      <td>105.640449</td>\n",
              "      <td>430.0</td>\n",
              "      <td>430.0</td>\n",
              "      <td>27108.528345</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>251.0</td>\n",
              "      <td>15.636364</td>\n",
              "      <td>47.337079</td>\n",
              "      <td>76.0</td>\n",
              "      <td>76.0</td>\n",
              "      <td>269.112360</td>\n",
              "      <td>41.0</td>\n",
              "      <td>40.0</td>\n",
              "      <td>62.0</td>\n",
              "      <td>3.272727</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>cy0007pw</th>\n",
              "      <td>92.284314</td>\n",
              "      <td>257.0</td>\n",
              "      <td>257.0</td>\n",
              "      <td>10372.542128</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>189.5</td>\n",
              "      <td>13.683168</td>\n",
              "      <td>52.313725</td>\n",
              "      <td>71.0</td>\n",
              "      <td>71.0</td>\n",
              "      <td>228.455057</td>\n",
              "      <td>52.0</td>\n",
              "      <td>41.0</td>\n",
              "      <td>66.0</td>\n",
              "      <td>2.970297</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                  f0                                                  \\\n",
              "                mean    max  range           var median 1st_quantile   \n",
              "file                                                                   \n",
              "cy0001at  110.609375  402.0  402.0  27607.511657    0.0          0.0   \n",
              "cy0002at  105.640449  430.0  430.0  27108.528345    0.0          0.0   \n",
              "cy0005at  110.609375  402.0  402.0  27607.511657    0.0          0.0   \n",
              "cy0006at  105.640449  430.0  430.0  27108.528345    0.0          0.0   \n",
              "cy0007pw   92.284314  257.0  257.0  10372.542128    0.0          0.0   \n",
              "\n",
              "                                                           en              \\\n",
              "         3rd_quantile mean_absolute_local_derivate       mean   max range   \n",
              "file                                                                        \n",
              "cy0001at        331.0                    23.777778  47.296875  73.0  73.0   \n",
              "cy0002at        251.0                    15.636364  47.337079  76.0  76.0   \n",
              "cy0005at        331.0                    23.777778  47.296875  73.0  73.0   \n",
              "cy0006at        251.0                    15.636364  47.337079  76.0  76.0   \n",
              "cy0007pw        189.5                    13.683168  52.313725  71.0  71.0   \n",
              "\n",
              "                                                       \\\n",
              "                 var median 1st_quantile 3rd_quantile   \n",
              "file                                                    \n",
              "cy0001at  327.323165   41.5         40.0         67.0   \n",
              "cy0002at  269.112360   41.0         40.0         62.0   \n",
              "cy0005at  327.323165   41.5         40.0         67.0   \n",
              "cy0006at  269.112360   41.0         40.0         62.0   \n",
              "cy0007pw  228.455057   52.0         41.0         66.0   \n",
              "\n",
              "                                       \n",
              "         mean_absolute_local_derivate  \n",
              "file                                   \n",
              "cy0001at                     3.809524  \n",
              "cy0002at                     3.272727  \n",
              "cy0005at                     3.809524  \n",
              "cy0006at                     3.272727  \n",
              "cy0007pw                     2.970297  "
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 34
        }
      ]
    },
    {
      "metadata": {
        "id": "Fx2bHYUhxvJu",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## 2. Develop a classifier for these three classes"
      ]
    },
    {
      "metadata": {
        "id": "N4zymBwlxud-",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "X, Y = train_test(df=df)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "Z1Hquq0xxQ48",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "89b02edf-7dac-4a21-a6b7-ca1be25ad2b1"
      },
      "cell_type": "code",
      "source": [
        "sklearn_knn(3, X, Y)"
      ],
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Accuracy score for voiced: 0.74\n",
            "Accuracy score for unvoiced: 0.45\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "6BD7YXRi0AqF",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 949
        },
        "outputId": "e4b67b5c-4bf7-4469-cf39-d14267a6047c"
      },
      "cell_type": "code",
      "source": [
        "plt.figure()\n",
        "cm = kNN(3, X['voiced'], Y['voiced'], labels=[\"pw\", \"ap\", \"at\"])\n",
        "plot_confusion_matrix(cm, classes=[\"pw\", \"ap\", \"at\"],\n",
        "                      title='Confusion matrix, without normalization')\n",
        "plt.show()\n",
        "\n",
        "cm2 = kNN(3, X['unvoiced'], Y['unvoiced'], labels=[\"pw\", \"ap\", \"at\"])\n",
        "plot_confusion_matrix(cm2, classes=[\"pw\", \"ap\", \"at\"],\n",
        "                      title='Confusion matrix, without normalization')\n",
        "plt.show()"
      ],
      "execution_count": 57,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Normalized confusion matrix\n",
            "[[0.86138614 0.08910891 0.04950495]\n",
            " [0.11111111 0.61111111 0.27777778]\n",
            " [0.02325581 0.30232558 0.6744186 ]]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAb0AAAGOCAYAAAADwN1HAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nxv8BqZa2qMLFbWsAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f58cbad9b70>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Normalized confusion matrix\n",
            "[[0.51401869 0.28971963 0.19626168]\n",
            " [0.23529412 0.45098039 0.31372549]\n",
            " [0.22826087 0.40217391 0.36956522]]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAcIAAAGOCAYAAAAEggNSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQiIi0mpMhEREpNWYCImISKv9H1bAHjzlxFDmAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f58cdea2630>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "8XprAjU8-uUo",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}