Lyric Data Analysis using CNN.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "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": "3.8.3"
    },
    "colab": {
      "name": "Final Project Song Sentiment CNN.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2Y_g4ECWY8JF",
        "outputId": "d15593eb-960d-471d-94fa-53f03de7e49b"
      },
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "sns.set()\n",
        "import re\n",
        "import spacy\n",
        "%tensorflow_version 2.x\n",
        "import tensorflow as tf\n",
        "device_name = tf.test.gpu_device_name()\n",
        "if device_name != '/device:GPU:0':\n",
        "  raise SystemError('GPU device not found')\n",
        "print('Found GPU at: {}'.format(device_name))\n",
        "from collections import Counter\n",
        "import string\n",
        "from sklearn.metrics import accuracy_score, confusion_matrix, plot_confusion_matrix\n",
        "import sys\n",
        "import csv\n",
        "maxInt = sys.maxsize\n",
        "\n",
        "while True:\n",
        "    # decrease the maxInt value by factor 10 \n",
        "    # as long as the OverflowError occurs.\n",
        "\n",
        "    try:\n",
        "        csv.field_size_limit(maxInt)\n",
        "        break\n",
        "    except OverflowError:\n",
        "        maxInt = int(maxInt/10)"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found GPU at: /device:GPU:0\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "kGumG_7JY8JS",
        "outputId": "514176eb-4ddb-4681-d1e0-b7e1ac6e292d"
      },
      "source": [
        "# load data\n",
        "data = pd.read_csv('/content/lyric-full_df.csv', engine='python', error_bad_lines=False).sample(frac=.8, replace=True, random_state=0)\n",
        "\n",
        "data"
      ],
      "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>WeekID</th>\n",
              "      <th>Unnamed: 0</th>\n",
              "      <th>url</th>\n",
              "      <th>Week Position</th>\n",
              "      <th>Song</th>\n",
              "      <th>Performer</th>\n",
              "      <th>SongID</th>\n",
              "      <th>Instance</th>\n",
              "      <th>Previous Week Position</th>\n",
              "      <th>Peak Position</th>\n",
              "      <th>Weeks on Chart</th>\n",
              "      <th>Lyrics</th>\n",
              "      <th>Sentiment</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>9634</th>\n",
              "      <td>1975-07-26</td>\n",
              "      <td>24626</td>\n",
              "      <td>http://www.billboard.com/charts/hot-100/1975-0...</td>\n",
              "      <td>75</td>\n",
              "      <td>The Proud One</td>\n",
              "      <td>The Osmonds</td>\n",
              "      <td>The Proud OneThe Osmonds</td>\n",
              "      <td>1</td>\n",
              "      <td>NaN</td>\n",
              "      <td>75</td>\n",
              "      <td>1</td>\n",
              "      <td>Livin is dyin\\nCause love youve left me cryin\\...</td>\n",
              "      <td>0.160684</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12850</th>\n",
              "      <td>1982-11-27</td>\n",
              "      <td>18692</td>\n",
              "      <td>http://www.billboard.com/charts/hot-100/1982-1...</td>\n",
              "      <td>84</td>\n",
              "      <td>The Elvis Medley</td>\n",
              "      <td>Elvis Presley</td>\n",
              "      <td>The Elvis MedleyElvis Presley</td>\n",
              "      <td>1</td>\n",
              "      <td>NaN</td>\n",
              "      <td>84</td>\n",
              "      <td>1</td>\n",
              "      <td>825 ALL NIGHT â PAROV STELAR\\n824 BIG JET PL...</td>\n",
              "      <td>0.141266</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23257</th>\n",
              "      <td>2012-09-29</td>\n",
              "      <td>17396</td>\n",
              "      <td>http://www.billboard.com/charts/hot-100/2012-0...</td>\n",
              "      <td>95</td>\n",
              "      <td>It's Time</td>\n",
              "      <td>Glee Cast</td>\n",
              "      <td>It's TimeGlee Cast</td>\n",
              "      <td>1</td>\n",
              "      <td>NaN</td>\n",
              "      <td>95</td>\n",
              "      <td>1</td>\n",
              "      <td>[BLAINE]\\nSo this is what you meant\\nWhen you ...</td>\n",
              "      <td>-0.014922</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23126</th>\n",
              "      <td>2012-05-05</td>\n",
              "      <td>27419</td>\n",
              "      <td>http://www.billboard.com/charts/hot-100/2012-0...</td>\n",
              "      <td>97</td>\n",
              "      <td>Tonight (Best You Ever Had)</td>\n",
              "      <td>John Legend Featuring Ludacris</td>\n",
              "      <td>Tonight (Best You Ever Had)John Legend Featuri...</td>\n",
              "      <td>1</td>\n",
              "      <td>NaN</td>\n",
              "      <td>97</td>\n",
              "      <td>1</td>\n",
              "      <td>[Verse 1: John Legend]\\nAlright\\nAin't this wh...</td>\n",
              "      <td>0.483707</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14267</th>\n",
              "      <td>1986-05-31</td>\n",
              "      <td>9901</td>\n",
              "      <td>http://www.billboard.com/charts/hot-100/1986-0...</td>\n",
              "      <td>73</td>\n",
              "      <td>Stick Around</td>\n",
              "      <td>Julian Lennon</td>\n",
              "      <td>Stick AroundJulian Lennon</td>\n",
              "      <td>1</td>\n",
              "      <td>44.0</td>\n",
              "      <td>32</td>\n",
              "      <td>11</td>\n",
              "      <td>If you want to tell me something new\\nI might ...</td>\n",
              "      <td>0.197868</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16028</th>\n",
              "      <td>1991-03-23</td>\n",
              "      <td>25522</td>\n",
              "      <td>http://www.billboard.com/charts/hot-100/1991-0...</td>\n",
              "      <td>88</td>\n",
              "      <td>Wrap My Body Tight</td>\n",
              "      <td>Johnny Gill</td>\n",
              "      <td>Wrap My Body TightJohnny Gill</td>\n",
              "      <td>1</td>\n",
              "      <td>NaN</td>\n",
              "      <td>88</td>\n",
              "      <td>1</td>\n",
              "      <td>[Verse 1:]\\nI fell out of bed\\nThis morning\\nO...</td>\n",
              "      <td>0.290747</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>22339</th>\n",
              "      <td>2010-08-21</td>\n",
              "      <td>24450</td>\n",
              "      <td>http://www.billboard.com/charts/hot-100/2010-0...</td>\n",
              "      <td>96</td>\n",
              "      <td>The Man Who Can't Be Moved</td>\n",
              "      <td>The Script</td>\n",
              "      <td>The Man Who Can't Be MovedThe Script</td>\n",
              "      <td>1</td>\n",
              "      <td>NaN</td>\n",
              "      <td>96</td>\n",
              "      <td>1</td>\n",
              "      <td>[Verse 1]\\nGoing back to the corner where I fi...</td>\n",
              "      <td>0.024306</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18663</th>\n",
              "      <td>1999-07-24</td>\n",
              "      <td>12510</td>\n",
              "      <td>http://www.billboard.com/charts/hot-100/1999-0...</td>\n",
              "      <td>29</td>\n",
              "      <td>Amazed</td>\n",
              "      <td>Lonestar</td>\n",
              "      <td>AmazedLonestar</td>\n",
              "      <td>1</td>\n",
              "      <td>29.0</td>\n",
              "      <td>29</td>\n",
              "      <td>8</td>\n",
              "      <td>[Verse 1]\\nEvery time our eyes meet\\nThis feel...</td>\n",
              "      <td>0.187500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18851</th>\n",
              "      <td>2000-03-11</td>\n",
              "      <td>26687</td>\n",
              "      <td>http://www.billboard.com/charts/hot-100/2000-0...</td>\n",
              "      <td>80</td>\n",
              "      <td>The Way You Love Me</td>\n",
              "      <td>Faith Hill</td>\n",
              "      <td>The Way You Love MeFaith Hill</td>\n",
              "      <td>1</td>\n",
              "      <td>NaN</td>\n",
              "      <td>80</td>\n",
              "      <td>1</td>\n",
              "      <td>[Verse 1]\\nIf I could grant\\nYou one wish\\nI'd...</td>\n",
              "      <td>0.349806</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>22833</th>\n",
              "      <td>2011-08-27</td>\n",
              "      <td>25930</td>\n",
              "      <td>http://www.billboard.com/charts/hot-100/2011-0...</td>\n",
              "      <td>78</td>\n",
              "      <td>We Don't Get Down Like Y'all</td>\n",
              "      <td>T.I. Featuring B.o.B.</td>\n",
              "      <td>We Don't Get Down Like Y'allT.I. Featuring B.o.B.</td>\n",
              "      <td>1</td>\n",
              "      <td>NaN</td>\n",
              "      <td>78</td>\n",
              "      <td>1</td>\n",
              "      <td>We donât get down like y'all\\nWe donât get...</td>\n",
              "      <td>0.063779</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>20910 rows × 13 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "           WeekID  ...  Sentiment\n",
              "9634   1975-07-26  ...   0.160684\n",
              "12850  1982-11-27  ...   0.141266\n",
              "23257  2012-09-29  ...  -0.014922\n",
              "23126  2012-05-05  ...   0.483707\n",
              "14267  1986-05-31  ...   0.197868\n",
              "...           ...  ...        ...\n",
              "16028  1991-03-23  ...   0.290747\n",
              "22339  2010-08-21  ...   0.024306\n",
              "18663  1999-07-24  ...   0.187500\n",
              "18851  2000-03-11  ...   0.349806\n",
              "22833  2011-08-27  ...   0.063779\n",
              "\n",
              "[20910 rows x 13 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "b0zX2OHMY8JV"
      },
      "source": [
        "# lyrics\n",
        "#data['Lyrics'][20]"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vx3iTWEEY8JX",
        "outputId": "dd8a61f0-8fdb-4da2-9604-2d5d05339acd"
      },
      "source": [
        "data.shape"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(20910, 13)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5jsuwy5iY8JY",
        "outputId": "03da30c5-6c5c-4283-a016-5492fec204df"
      },
      "source": [
        "data.info()"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 20910 entries, 9634 to 22833\n",
            "Data columns (total 13 columns):\n",
            " #   Column                  Non-Null Count  Dtype  \n",
            "---  ------                  --------------  -----  \n",
            " 0   WeekID                  20910 non-null  object \n",
            " 1   Unnamed: 0              20910 non-null  int64  \n",
            " 2   url                     20910 non-null  object \n",
            " 3   Week Position           20910 non-null  int64  \n",
            " 4   Song                    20910 non-null  object \n",
            " 5   Performer               20910 non-null  object \n",
            " 6   SongID                  20910 non-null  object \n",
            " 7   Instance                20910 non-null  int64  \n",
            " 8   Previous Week Position  9019 non-null   float64\n",
            " 9   Peak Position           20910 non-null  int64  \n",
            " 10  Weeks on Chart          20910 non-null  int64  \n",
            " 11  Lyrics                  20910 non-null  object \n",
            " 12  Sentiment               20910 non-null  float64\n",
            "dtypes: float64(2), int64(5), object(6)\n",
            "memory usage: 2.2+ MB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hIP3jy1LY8JZ",
        "outputId": "963b28af-3c1e-4fab-a000-0ac0bbdf7bae"
      },
      "source": [
        "data.isnull().sum()"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "WeekID                        0\n",
              "Unnamed: 0                    0\n",
              "url                           0\n",
              "Week Position                 0\n",
              "Song                          0\n",
              "Performer                     0\n",
              "SongID                        0\n",
              "Instance                      0\n",
              "Previous Week Position    11891\n",
              "Peak Position                 0\n",
              "Weeks on Chart                0\n",
              "Lyrics                        0\n",
              "Sentiment                     0\n",
              "dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XJk2tt4hY8Ja"
      },
      "source": [
        "data['lyric_length'] = data['Lyrics'].apply(lambda x: len(x.split()))"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 408
        },
        "id": "WjejDuOrY8Ja",
        "outputId": "287b11e0-fef8-4119-e7ea-b7ee0fe2dc68"
      },
      "source": [
        "songs = data.copy()[['WeekID', 'Song', 'Performer', 'Lyrics', 'Sentiment', 'lyric_length']]\n",
        "\n",
        "songs.head()"
      ],
      "execution_count": 10,
      "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>WeekID</th>\n",
              "      <th>Song</th>\n",
              "      <th>Performer</th>\n",
              "      <th>Lyrics</th>\n",
              "      <th>Sentiment</th>\n",
              "      <th>lyric_length</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>9634</th>\n",
              "      <td>1975-07-26</td>\n",
              "      <td>The Proud One</td>\n",
              "      <td>The Osmonds</td>\n",
              "      <td>Livin is dyin\\nCause love youve left me cryin\\...</td>\n",
              "      <td>0.160684</td>\n",
              "      <td>112</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12850</th>\n",
              "      <td>1982-11-27</td>\n",
              "      <td>The Elvis Medley</td>\n",
              "      <td>Elvis Presley</td>\n",
              "      <td>825 ALL NIGHT â PAROV STELAR\\n824 BIG JET PL...</td>\n",
              "      <td>0.141266</td>\n",
              "      <td>5781</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23257</th>\n",
              "      <td>2012-09-29</td>\n",
              "      <td>It's Time</td>\n",
              "      <td>Glee Cast</td>\n",
              "      <td>[BLAINE]\\nSo this is what you meant\\nWhen you ...</td>\n",
              "      <td>-0.014922</td>\n",
              "      <td>318</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23126</th>\n",
              "      <td>2012-05-05</td>\n",
              "      <td>Tonight (Best You Ever Had)</td>\n",
              "      <td>John Legend Featuring Ludacris</td>\n",
              "      <td>[Verse 1: John Legend]\\nAlright\\nAin't this wh...</td>\n",
              "      <td>0.483707</td>\n",
              "      <td>404</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14267</th>\n",
              "      <td>1986-05-31</td>\n",
              "      <td>Stick Around</td>\n",
              "      <td>Julian Lennon</td>\n",
              "      <td>If you want to tell me something new\\nI might ...</td>\n",
              "      <td>0.197868</td>\n",
              "      <td>150</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "           WeekID                         Song  ... Sentiment lyric_length\n",
              "9634   1975-07-26                The Proud One  ...  0.160684          112\n",
              "12850  1982-11-27             The Elvis Medley  ...  0.141266         5781\n",
              "23257  2012-09-29                    It's Time  ... -0.014922          318\n",
              "23126  2012-05-05  Tonight (Best You Ever Had)  ...  0.483707          404\n",
              "14267  1986-05-31                 Stick Around  ...  0.197868          150\n",
              "\n",
              "[5 rows x 6 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 111
        },
        "id": "eSZtG-V9Y8Jb",
        "outputId": "a311ecb3-51b5-440e-b936-7afd5065fa52"
      },
      "source": [
        "songs.corr()"
      ],
      "execution_count": 11,
      "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>Sentiment</th>\n",
              "      <th>lyric_length</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Sentiment</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>-0.056401</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>lyric_length</th>\n",
              "      <td>-0.056401</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "              Sentiment  lyric_length\n",
              "Sentiment      1.000000     -0.056401\n",
              "lyric_length  -0.056401      1.000000"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 297
        },
        "id": "6eOXWVLTY8Jc",
        "outputId": "f263599f-a9ed-4f3c-c968-74240d27dfd7"
      },
      "source": [
        "songs[['Sentiment', 'lyric_length']].describe()"
      ],
      "execution_count": 12,
      "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>Sentiment</th>\n",
              "      <th>lyric_length</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>20910.000000</td>\n",
              "      <td>20910.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>0.127870</td>\n",
              "      <td>2275.112099</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>0.170610</td>\n",
              "      <td>11318.851848</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>-0.937500</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>0.020692</td>\n",
              "      <td>200.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>0.113605</td>\n",
              "      <td>287.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>0.236962</td>\n",
              "      <td>480.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>152084.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "          Sentiment   lyric_length\n",
              "count  20910.000000   20910.000000\n",
              "mean       0.127870    2275.112099\n",
              "std        0.170610   11318.851848\n",
              "min       -0.937500       1.000000\n",
              "25%        0.020692     200.000000\n",
              "50%        0.113605     287.000000\n",
              "75%        0.236962     480.000000\n",
              "max        1.000000  152084.000000"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cLxiKPIrLjw_",
        "outputId": "a48f533a-9187-46f5-b0f5-06d92c437f07"
      },
      "source": [
        "songs['WeekID'].unique()"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array(['1975-07-26', '1982-11-27', '2012-09-29', ..., '1993-12-18',\n",
              "       '2013-04-20', '2013-03-23'], dtype=object)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "mP8h2HmQ6JZC",
        "outputId": "3f9f5f73-4bf4-477c-89ba-51e38c73f86f"
      },
      "source": [
        "from helpers_ori import lyrics_to_words, clean_lyrics\r\n",
        "import nltk\r\n",
        "\r\n",
        "nltk.download('popular')\r\n",
        "# function from helpers\r\n",
        "def unique(list1): \r\n",
        "     # intilize a null list \r\n",
        "    unique_list = [] \r\n",
        "      \r\n",
        "    # traverse for all elements \r\n",
        "    for x in list1: \r\n",
        "        # check if exists in unique_list or not \r\n",
        "        if x not in unique_list: \r\n",
        "            unique_list.append(x) \r\n",
        "    \r\n",
        "    return unique_list\r\n",
        "\r\n",
        "clean_lyrics(songs, 'Lyrics')\r\n",
        "\r\n",
        "# List used to store the words\r\n",
        "\r\n",
        "lyrica = []\r\n",
        "\r\n",
        "# Iterate each lyric and split unique words appending the result into the words list\r\n",
        "\r\n",
        "songs = songs.reset_index(drop=True)\r\n",
        "for lyric in songs['Lyrics'].tolist():\r\n",
        "    lyrica.append(unique(lyrics_to_words(lyric).split()))\r\n",
        "\r\n",
        "# Create the new column with the information of words lists \r\n",
        "\r\n",
        "new_songs = songs\r\n",
        "new_songs['words'] = lyrica\r\n",
        "\r\n",
        "new_songs.head()"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[nltk_data] Downloading collection 'popular'\n",
            "[nltk_data]    | \n",
            "[nltk_data]    | Downloading package cmudict to /root/nltk_data...\n",
            "[nltk_data]    |   Package cmudict is already up-to-date!\n",
            "[nltk_data]    | Downloading package gazetteers to /root/nltk_data...\n",
            "[nltk_data]    |   Package gazetteers is already up-to-date!\n",
            "[nltk_data]    | Downloading package genesis to /root/nltk_data...\n",
            "[nltk_data]    |   Package genesis is already up-to-date!\n",
            "[nltk_data]    | Downloading package gutenberg to /root/nltk_data...\n",
            "[nltk_data]    |   Package gutenberg is already up-to-date!\n",
            "[nltk_data]    | Downloading package inaugural to /root/nltk_data...\n",
            "[nltk_data]    |   Package inaugural is already up-to-date!\n",
            "[nltk_data]    | Downloading package movie_reviews to\n",
            "[nltk_data]    |     /root/nltk_data...\n",
            "[nltk_data]    |   Package movie_reviews is already up-to-date!\n",
            "[nltk_data]    | Downloading package names to /root/nltk_data...\n",
            "[nltk_data]    |   Package names is already up-to-date!\n",
            "[nltk_data]    | Downloading package shakespeare to /root/nltk_data...\n",
            "[nltk_data]    |   Package shakespeare is already up-to-date!\n",
            "[nltk_data]    | Downloading package stopwords to /root/nltk_data...\n",
            "[nltk_data]    |   Package stopwords is already up-to-date!\n",
            "[nltk_data]    | Downloading package treebank to /root/nltk_data...\n",
            "[nltk_data]    |   Package treebank is already up-to-date!\n",
            "[nltk_data]    | Downloading package twitter_samples to\n",
            "[nltk_data]    |     /root/nltk_data...\n",
            "[nltk_data]    |   Package twitter_samples is already up-to-date!\n",
            "[nltk_data]    | Downloading package omw to /root/nltk_data...\n",
            "[nltk_data]    |   Package omw is already up-to-date!\n",
            "[nltk_data]    | Downloading package wordnet to /root/nltk_data...\n",
            "[nltk_data]    |   Package wordnet is already up-to-date!\n",
            "[nltk_data]    | Downloading package wordnet_ic to /root/nltk_data...\n",
            "[nltk_data]    |   Package wordnet_ic is already up-to-date!\n",
            "[nltk_data]    | Downloading package words to /root/nltk_data...\n",
            "[nltk_data]    |   Package words is already up-to-date!\n",
            "[nltk_data]    | Downloading package maxent_ne_chunker to\n",
            "[nltk_data]    |     /root/nltk_data...\n",
            "[nltk_data]    |   Package maxent_ne_chunker is already up-to-date!\n",
            "[nltk_data]    | Downloading package punkt to /root/nltk_data...\n",
            "[nltk_data]    |   Package punkt is already up-to-date!\n",
            "[nltk_data]    | Downloading package snowball_data to\n",
            "[nltk_data]    |     /root/nltk_data...\n",
            "[nltk_data]    |   Package snowball_data is already up-to-date!\n",
            "[nltk_data]    | Downloading package averaged_perceptron_tagger to\n",
            "[nltk_data]    |     /root/nltk_data...\n",
            "[nltk_data]    |   Package averaged_perceptron_tagger is already up-\n",
            "[nltk_data]    |       to-date!\n",
            "[nltk_data]    | \n",
            "[nltk_data]  Done downloading collection popular\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>WeekID</th>\n",
              "      <th>Song</th>\n",
              "      <th>Performer</th>\n",
              "      <th>Lyrics</th>\n",
              "      <th>Sentiment</th>\n",
              "      <th>lyric_length</th>\n",
              "      <th>words</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1975-07-26</td>\n",
              "      <td>The Proud One</td>\n",
              "      <td>The Osmonds</td>\n",
              "      <td>livin is dyin cause love youve left me cryin n...</td>\n",
              "      <td>0.160684</td>\n",
              "      <td>112</td>\n",
              "      <td>[livin, dyin, cause, love, youve, left, cryin,...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1982-11-27</td>\n",
              "      <td>The Elvis Medley</td>\n",
              "      <td>Elvis Presley</td>\n",
              "      <td>85 all night â parov stelar 84 big jet plane â...</td>\n",
              "      <td>0.141266</td>\n",
              "      <td>5781</td>\n",
              "      <td>[85, night, â, parov, stelar, 84, big, jet, pl...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2012-09-29</td>\n",
              "      <td>It's Time</td>\n",
              "      <td>Glee Cast</td>\n",
              "      <td>blaine so this is what you meant when you said...</td>\n",
              "      <td>-0.014922</td>\n",
              "      <td>318</td>\n",
              "      <td>[blaine, meant, said, spent, time, build, bott...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2012-05-05</td>\n",
              "      <td>Tonight (Best You Ever Had)</td>\n",
              "      <td>John Legend Featuring Ludacris</td>\n",
              "      <td>john legend alright ain't this what you came f...</td>\n",
              "      <td>0.483707</td>\n",
              "      <td>404</td>\n",
              "      <td>[john, legend, alright, aint, came, wish, girl...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1986-05-31</td>\n",
              "      <td>Stick Around</td>\n",
              "      <td>Julian Lennon</td>\n",
              "      <td>if you want to tell me something new i might s...</td>\n",
              "      <td>0.197868</td>\n",
              "      <td>150</td>\n",
              "      <td>[want, tell, something, new, might, stick, aro...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       WeekID  ...                                              words\n",
              "0  1975-07-26  ...  [livin, dyin, cause, love, youve, left, cryin,...\n",
              "1  1982-11-27  ...  [85, night, â, parov, stelar, 84, big, jet, pl...\n",
              "2  2012-09-29  ...  [blaine, meant, said, spent, time, build, bott...\n",
              "3  2012-05-05  ...  [john, legend, alright, aint, came, wish, girl...\n",
              "4  1986-05-31  ...  [want, tell, something, new, might, stick, aro...\n",
              "\n",
              "[5 rows x 7 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 561
        },
        "id": "qk0i2WfFR58K",
        "outputId": "260f6047-1c12-4da5-e917-6310f4751b7b"
      },
      "source": [
        "songs.head()"
      ],
      "execution_count": 15,
      "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>WeekID</th>\n",
              "      <th>Song</th>\n",
              "      <th>Performer</th>\n",
              "      <th>Lyrics</th>\n",
              "      <th>Sentiment</th>\n",
              "      <th>lyric_length</th>\n",
              "      <th>words</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1975-07-26</td>\n",
              "      <td>The Proud One</td>\n",
              "      <td>The Osmonds</td>\n",
              "      <td>livin is dyin cause love youve left me cryin n...</td>\n",
              "      <td>0.160684</td>\n",
              "      <td>112</td>\n",
              "      <td>[livin, dyin, cause, love, youve, left, cryin,...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1982-11-27</td>\n",
              "      <td>The Elvis Medley</td>\n",
              "      <td>Elvis Presley</td>\n",
              "      <td>85 all night â parov stelar 84 big jet plane â...</td>\n",
              "      <td>0.141266</td>\n",
              "      <td>5781</td>\n",
              "      <td>[85, night, â, parov, stelar, 84, big, jet, pl...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2012-09-29</td>\n",
              "      <td>It's Time</td>\n",
              "      <td>Glee Cast</td>\n",
              "      <td>blaine so this is what you meant when you said...</td>\n",
              "      <td>-0.014922</td>\n",
              "      <td>318</td>\n",
              "      <td>[blaine, meant, said, spent, time, build, bott...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2012-05-05</td>\n",
              "      <td>Tonight (Best You Ever Had)</td>\n",
              "      <td>John Legend Featuring Ludacris</td>\n",
              "      <td>john legend alright ain't this what you came f...</td>\n",
              "      <td>0.483707</td>\n",
              "      <td>404</td>\n",
              "      <td>[john, legend, alright, aint, came, wish, girl...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1986-05-31</td>\n",
              "      <td>Stick Around</td>\n",
              "      <td>Julian Lennon</td>\n",
              "      <td>if you want to tell me something new i might s...</td>\n",
              "      <td>0.197868</td>\n",
              "      <td>150</td>\n",
              "      <td>[want, tell, something, new, might, stick, aro...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       WeekID  ...                                              words\n",
              "0  1975-07-26  ...  [livin, dyin, cause, love, youve, left, cryin,...\n",
              "1  1982-11-27  ...  [85, night, â, parov, stelar, 84, big, jet, pl...\n",
              "2  2012-09-29  ...  [blaine, meant, said, spent, time, build, bott...\n",
              "3  2012-05-05  ...  [john, legend, alright, aint, came, wish, girl...\n",
              "4  1986-05-31  ...  [want, tell, something, new, might, stick, aro...\n",
              "\n",
              "[5 rows x 7 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 835
        },
        "id": "8R_ZBhZhY8Jd",
        "outputId": "bdecf14b-4d4f-4fba-ff30-9ba104771d48"
      },
      "source": [
        "new_songs['Sentiment'].plot()\n",
        "plt.ylabel('Sentiment Score')\n",
        "plt.title('Lineplot of Sentiment Score')\n",
        "plt.show();\n",
        "new_songs['lyric_length'].plot()\n",
        "plt.ylabel('Lyric Length')\n",
        "plt.title('Lineplot of Lyric Length')\n",
        "plt.show();\n",
        "new_songs[['Sentiment', 'lyric_length']].plot(kind='area', y='lyric_length', x='Sentiment', stacked=False, color='green')\n",
        "plt.title('Areaplot of Sentiment vs Lyric Length')\n",
        "plt.show();"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "iq1-KXLMY8Jf"
      },
      "source": [
        "## Convert Sentiments to Discrete Values"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 666
        },
        "id": "9K4LOqpOY8Jg",
        "outputId": "086cb898-a543-4cf3-92ec-230de754ab3d"
      },
      "source": [
        "# transform sentiments to discrete variable\n",
        "\n",
        "labels = []\n",
        "\n",
        "for i in new_songs['Sentiment']:\n",
        "  if i < .075:\n",
        "    labels.append(0)\n",
        "  elif i > 0.15:\n",
        "    labels.append(2)\n",
        "  else: \n",
        "    labels.append(1)\n",
        "\n",
        "new_songs['label'] = labels\n",
        "\n",
        "\n",
        "#new_songs['Sentiment'].loc[new_songs['Sentiment'] < -0.25] = 'neg'\n",
        "#new_songs['Sentiment'][~new_songs['Sentiment'].between(-0.25,0.25)] = 'neu'\n",
        "#new_songs['Sentiment'].loc[new_songs['Sentiment'] > 0.25] = 'pos'\n",
        "\n",
        "new_songs['label'] = new_songs['label'].astype('category')\n",
        "\n",
        "new_songs.head()"
      ],
      "execution_count": 17,
      "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>WeekID</th>\n",
              "      <th>Song</th>\n",
              "      <th>Performer</th>\n",
              "      <th>Lyrics</th>\n",
              "      <th>Sentiment</th>\n",
              "      <th>lyric_length</th>\n",
              "      <th>words</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1975-07-26</td>\n",
              "      <td>The Proud One</td>\n",
              "      <td>The Osmonds</td>\n",
              "      <td>livin is dyin cause love youve left me cryin n...</td>\n",
              "      <td>0.160684</td>\n",
              "      <td>112</td>\n",
              "      <td>[livin, dyin, cause, love, youve, left, cryin,...</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1982-11-27</td>\n",
              "      <td>The Elvis Medley</td>\n",
              "      <td>Elvis Presley</td>\n",
              "      <td>85 all night â parov stelar 84 big jet plane â...</td>\n",
              "      <td>0.141266</td>\n",
              "      <td>5781</td>\n",
              "      <td>[85, night, â, parov, stelar, 84, big, jet, pl...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2012-09-29</td>\n",
              "      <td>It's Time</td>\n",
              "      <td>Glee Cast</td>\n",
              "      <td>blaine so this is what you meant when you said...</td>\n",
              "      <td>-0.014922</td>\n",
              "      <td>318</td>\n",
              "      <td>[blaine, meant, said, spent, time, build, bott...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2012-05-05</td>\n",
              "      <td>Tonight (Best You Ever Had)</td>\n",
              "      <td>John Legend Featuring Ludacris</td>\n",
              "      <td>john legend alright ain't this what you came f...</td>\n",
              "      <td>0.483707</td>\n",
              "      <td>404</td>\n",
              "      <td>[john, legend, alright, aint, came, wish, girl...</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1986-05-31</td>\n",
              "      <td>Stick Around</td>\n",
              "      <td>Julian Lennon</td>\n",
              "      <td>if you want to tell me something new i might s...</td>\n",
              "      <td>0.197868</td>\n",
              "      <td>150</td>\n",
              "      <td>[want, tell, something, new, might, stick, aro...</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       WeekID  ... label\n",
              "0  1975-07-26  ...     2\n",
              "1  1982-11-27  ...     1\n",
              "2  2012-09-29  ...     0\n",
              "3  2012-05-05  ...     2\n",
              "4  1986-05-31  ...     2\n",
              "\n",
              "[5 rows x 8 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pin-3B92Y8Jh",
        "outputId": "e73934dd-cfc8-445d-9669-8c2e16221e40"
      },
      "source": [
        "# Count of Sentiment Classes\n",
        "\n",
        "Counter(new_songs['label'])"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Counter({0: 8203, 1: 4090, 2: 8617})"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "scrolled": true,
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GT9ZP8dMY8Ji",
        "outputId": "4b383e0b-c44d-441c-b887-f125cd07213f"
      },
      "source": [
        "new_songs['label'].dtype"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "CategoricalDtype(categories=[0, 1, 2], ordered=False)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 19
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vl8PFKw9Y8Jj"
      },
      "source": [
        "## Baseline Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IKJP_vn6Y8Jj"
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "lyrics = new_songs['words']\n",
        "y = new_songs['label']\n",
        "\n",
        "lyrics_train, lyrics_test, y_train, y_test = train_test_split(lyrics,y,\n",
        "                                                             test_size=.25,\n",
        "                                                             random_state=0)"
      ],
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NP2-g74UY8Jk",
        "outputId": "2dd62755-4fce-4344-8966-af4494f39d7d"
      },
      "source": [
        "# Create BoW\n",
        "\n",
        "from sklearn.feature_extraction.text import CountVectorizer\n",
        "\n",
        "vectorizer = CountVectorizer()\n",
        "\n",
        "train_bow = vectorizer.fit_transform(' '.join(text) for text in lyrics_train)\n",
        "test_bow = vectorizer.transform(' '.join(text) for text in lyrics_test)\n",
        "\n",
        "X_train = train_bow\n",
        "X_test  = test_bow\n",
        "X_train"
      ],
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<15682x226678 sparse matrix of type '<class 'numpy.int64'>'\n",
              "\twith 5786702 stored elements in Compressed Sparse Row format>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 21
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "scrolled": true,
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bWTJrZIBY8Jl",
        "outputId": "1c7b4404-7504-4e9b-d25d-0e727cfe8287"
      },
      "source": [
        "# baseline logistic regression\n",
        "\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.metrics import classification_report, roc_auc_score, accuracy_score\n",
        "\n",
        "clf = LogisticRegression()\n",
        "clf.fit(X_train, y_train)\n",
        "\n",
        "pred = clf.predict(X_test)\n",
        "\n",
        "print('Accuracy', accuracy_score(y_test, pred), '\\n')\n",
        "#print('ROC AUC:', roc_auc_score(y_test, pred))\n",
        "print(classification_report(y_test, pred))"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Accuracy 0.8031752104055088 \n",
            "\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.83      0.84      0.84      2079\n",
            "           1       0.67      0.62      0.65       997\n",
            "           2       0.83      0.85      0.84      2152\n",
            "\n",
            "    accuracy                           0.80      5228\n",
            "   macro avg       0.78      0.77      0.77      5228\n",
            "weighted avg       0.80      0.80      0.80      5228\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
            "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
            "\n",
            "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
            "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
            "Please also refer to the documentation for alternative solver options:\n",
            "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
            "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "povyEimdoJMV"
      },
      "source": [
        "# Unigram & Bigrams Logistic Reg\r\n",
        "\r\n",
        "vectorizer = CountVectorizer(ngram_range=(1,2))\r\n",
        "\r\n",
        "train_bow = vectorizer.fit_transform(' '.join(text) for text in lyrics_train)\r\n",
        "test_bow = vectorizer.transform(' '.join(text) for text in lyrics_test)\r\n",
        "\r\n",
        "X_train = train_bow\r\n",
        "X_test  = test_bow\r\n",
        "X_train\r\n",
        "\r\n",
        "clf = LogisticRegression()\r\n",
        "clf.fit(X_train, y_train)\r\n",
        "\r\n",
        "pred = clf.predict(X_test)\r\n",
        "\r\n",
        "print('Accuracy', accuracy_score(y_test, pred), '\\n')\r\n",
        "#print('ROC AUC:', roc_auc_score(y_test, pred))\r\n",
        "print(classification_report(y_test, pred))\r\n",
        "pickle.dump(clf, open('uni & bi log', 'wb'))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nLU14f6HoJif"
      },
      "source": [
        "# Bigrams Logistic Reg\r\n",
        "\r\n",
        "vectorizer = CountVectorizer(ngram_range=(2,2))\r\n",
        "\r\n",
        "train_bow = vectorizer.fit_transform(' '.join(text) for text in lyrics_train)\r\n",
        "test_bow = vectorizer.transform(' '.join(text) for text in lyrics_test)\r\n",
        "\r\n",
        "X_train = train_bow\r\n",
        "X_test  = test_bow\r\n",
        "X_train\r\n",
        "\r\n",
        "clf = LogisticRegression()\r\n",
        "clf.fit(X_train, y_train)\r\n",
        "\r\n",
        "pred = clf.predict(X_test)\r\n",
        "\r\n",
        "print('Accuracy', accuracy_score(y_test, pred), '\\n')\r\n",
        "#print('ROC AUC:', roc_auc_score(y_test, pred))\r\n",
        "print(classification_report(y_test, pred))\r\n",
        "pickle.dump(clf, open('bigram log', 'wb'))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PTV9ptTBoKbQ"
      },
      "source": [
        "#Grid Search\r\n",
        "\r\n",
        "from sklearn.model_selection import GridSearchCV\r\n",
        "\r\n",
        "grid_values = {'penalty': ['l1', 'l2'],\r\n",
        "               'C':[0.001,.009,0.01,.09,1,5,10,25],\r\n",
        "               'class_weight': ['balanced', None]}\r\n",
        "\r\n",
        "log_grid = GridSearchCV(estimator = clf, \r\n",
        "                            param_grid = grid_values,\r\n",
        "                            scoring = 'accuracy',\r\n",
        "                            cv = 3,\r\n",
        "                            verbose = 2,\r\n",
        "                            n_jobs = 2)\r\n",
        "\r\n",
        "log_grid.fit(X_train, y_train.values.ravel())\r\n",
        "\r\n",
        "# Confusion Matrix Plot\r\n",
        "\r\n",
        "plot_confusion_matrix(log_grid, X_test, y_test, cmap=plt.cm.Blues)\r\n",
        "plt.grid(False)\r\n",
        "plt.show();\r\n",
        "\r\n",
        "#Predict values based on new parameters\r\n",
        "\r\n",
        "pred = log_grid.predict(X_test)\r\n",
        "\r\n",
        "con_mat = confusion_matrix(y_test, pred)\r\n",
        "\r\n",
        "print('Confusion Matrix:\\n', con_mat)\r\n",
        "\r\n",
        "print('Misclassification Rate:', mis_class(con_mat))\r\n",
        "\r\n",
        "print('Training Accuracy:', log_grid.score(X_train,y_train))\r\n",
        "print('Testing Accuracy:', accuracy_score(y_test, pred))\r\n",
        "print('Classification Report:\\n', classification_report(y_test, pred))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mE0nk7fhY8Jm"
      },
      "source": [
        "## CNN's"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9YWsT06RY8Jv"
      },
      "source": [
        "**Note:** Pay close attention to the difference between this technique and the X_train that was produced by scikit-learn’s CountVectorizer.\n",
        "\n",
        "With CountVectorizer, we had stacked vectors of word counts, and each vector was the same length (the size of the total corpus vocabulary). With Tokenizer, the resulting vectors equal the length of each text, and the numbers don’t denote counts, but rather correspond to the word values from the dictionary tokenizer.word_index."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nfocMTaLZ6pV",
        "outputId": "2c550b0f-eff3-417e-e980-8620bcfea90d"
      },
      "source": [
        "from pickle import load\r\n",
        "from numpy import array\r\n",
        "from keras.preprocessing.text import Tokenizer\r\n",
        "from keras.preprocessing.sequence import pad_sequences\r\n",
        "from keras.utils.vis_utils import plot_model\r\n",
        "from keras.models import Model\r\n",
        "from keras.layers import Input\r\n",
        "from keras.layers import Dense\r\n",
        "from keras.layers import Flatten\r\n",
        "from keras.layers import Dropout\r\n",
        "from keras.layers import Embedding\r\n",
        "from keras.layers.convolutional import Conv1D\r\n",
        "from keras.layers.convolutional import MaxPooling1D\r\n",
        "from keras.layers.merge import concatenate\r\n",
        "\r\n",
        "# fit a tokenizer\r\n",
        "def create_tokenizer(lines):\r\n",
        "\ttokenizer = Tokenizer()\r\n",
        "\ttokenizer.fit_on_texts(lines)\r\n",
        "\treturn tokenizer\r\n",
        " \r\n",
        "# calculate the maximum document length\r\n",
        "def max_length(lines):\r\n",
        "\treturn max([len(s) for s in lines])\r\n",
        " \r\n",
        "# encode a list of lines\r\n",
        "def encode_text(tokenizer, lines, length):\r\n",
        "\t# integer encode\r\n",
        "\tencoded = tokenizer.texts_to_sequences(lines)\r\n",
        "\t# pad encoded sequences\r\n",
        "\tpadded = pad_sequences(encoded, maxlen=length, padding='post')\r\n",
        "\treturn padded\r\n",
        " \r\n",
        "# define the model\r\n",
        "def define_model(length, vocab_size):\r\n",
        "\t# channel 1\r\n",
        "\tinputs1 = Input(shape=(length,))\r\n",
        "\tembedding1 = Embedding(vocab_size, 100)(inputs1)\r\n",
        "\tconv1 = Conv1D(filters=32, kernel_size=4, activation='relu')(embedding1)\r\n",
        "\tdrop1 = Dropout(0.5)(conv1)\r\n",
        "\tpool1 = MaxPooling1D(pool_size=2)(drop1)\r\n",
        "\tflat1 = Flatten()(pool1)\r\n",
        "\t# channel 2\r\n",
        "\tinputs2 = Input(shape=(length,))\r\n",
        "\tembedding2 = Embedding(vocab_size, 100)(inputs2)\r\n",
        "\tconv2 = Conv1D(filters=32, kernel_size=6, activation='relu')(embedding2)\r\n",
        "\tdrop2 = Dropout(0.5)(conv2)\r\n",
        "\tpool2 = MaxPooling1D(pool_size=2)(drop2)\r\n",
        "\tflat2 = Flatten()(pool2)\r\n",
        "\t# channel 3\r\n",
        "\tinputs3 = Input(shape=(length,))\r\n",
        "\tembedding3 = Embedding(vocab_size, 100)(inputs3)\r\n",
        "\tconv3 = Conv1D(filters=32, kernel_size=8, activation='relu')(embedding3)\r\n",
        "\tdrop3 = Dropout(0.5)(conv3)\r\n",
        "\tpool3 = MaxPooling1D(pool_size=2)(drop3)\r\n",
        "\tflat3 = Flatten()(pool3)\r\n",
        "\t# merge\r\n",
        "\tmerged = concatenate([flat1, flat2, flat3])\r\n",
        "\t# interpretation\r\n",
        "\tdense1 = Dense(10, activation='relu')(merged)\r\n",
        "\toutputs = Dense(1, activation='softmax')(dense1)\r\n",
        "\tmodel = Model(inputs=[inputs1, inputs2, inputs3], outputs=outputs)\r\n",
        "\t# compile\r\n",
        "\tmodel.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])\r\n",
        "\t# summarize\r\n",
        "\tprint(model.summary())\r\n",
        "\tplot_model(model, show_shapes=True, to_file='multichannel.png')\r\n",
        "\treturn model\r\n",
        " \r\n",
        "# create tokenizer\r\n",
        "tokenizer = create_tokenizer(lyrics_train)\r\n",
        "# calculate max document length\r\n",
        "length = max_length(lyrics_train)\r\n",
        "# calculate vocabulary size\r\n",
        "vocab_size = len(tokenizer.word_index) + 1\r\n",
        "print('Max document length: %d' % length)\r\n",
        "print('Vocabulary size: %d' % vocab_size)\r\n",
        "# encode data\r\n",
        "trainX = encode_text(tokenizer, lyrics_train, length)\r\n",
        "print(trainX.shape)\r\n",
        " \r\n",
        "# define model\r\n",
        "model = define_model(length, vocab_size)\r\n",
        "# fit model\r\n",
        "model.fit([trainX,trainX,trainX], array(y_train), epochs=10, batch_size=16)\r\n",
        "# save the model\r\n",
        "model.save('Kim/Cano CNN')"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Max document length: 11489\n",
            "Vocabulary size: 226727\n",
            "(15682, 11489)\n",
            "Model: \"functional_1\"\n",
            "__________________________________________________________________________________________________\n",
            "Layer (type)                    Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            "input_1 (InputLayer)            [(None, 11489)]      0                                            \n",
            "__________________________________________________________________________________________________\n",
            "input_2 (InputLayer)            [(None, 11489)]      0                                            \n",
            "__________________________________________________________________________________________________\n",
            "input_3 (InputLayer)            [(None, 11489)]      0                                            \n",
            "__________________________________________________________________________________________________\n",
            "embedding (Embedding)           (None, 11489, 100)   22672700    input_1[0][0]                    \n",
            "__________________________________________________________________________________________________\n",
            "embedding_1 (Embedding)         (None, 11489, 100)   22672700    input_2[0][0]                    \n",
            "__________________________________________________________________________________________________\n",
            "embedding_2 (Embedding)         (None, 11489, 100)   22672700    input_3[0][0]                    \n",
            "__________________________________________________________________________________________________\n",
            "conv1d (Conv1D)                 (None, 11486, 32)    12832       embedding[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "conv1d_1 (Conv1D)               (None, 11484, 32)    19232       embedding_1[0][0]                \n",
            "__________________________________________________________________________________________________\n",
            "conv1d_2 (Conv1D)               (None, 11482, 32)    25632       embedding_2[0][0]                \n",
            "__________________________________________________________________________________________________\n",
            "dropout (Dropout)               (None, 11486, 32)    0           conv1d[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "dropout_1 (Dropout)             (None, 11484, 32)    0           conv1d_1[0][0]                   \n",
            "__________________________________________________________________________________________________\n",
            "dropout_2 (Dropout)             (None, 11482, 32)    0           conv1d_2[0][0]                   \n",
            "__________________________________________________________________________________________________\n",
            "max_pooling1d (MaxPooling1D)    (None, 5743, 32)     0           dropout[0][0]                    \n",
            "__________________________________________________________________________________________________\n",
            "max_pooling1d_1 (MaxPooling1D)  (None, 5742, 32)     0           dropout_1[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "max_pooling1d_2 (MaxPooling1D)  (None, 5741, 32)     0           dropout_2[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "flatten (Flatten)               (None, 183776)       0           max_pooling1d[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "flatten_1 (Flatten)             (None, 183744)       0           max_pooling1d_1[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "flatten_2 (Flatten)             (None, 183712)       0           max_pooling1d_2[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "concatenate (Concatenate)       (None, 551232)       0           flatten[0][0]                    \n",
            "                                                                 flatten_1[0][0]                  \n",
            "                                                                 flatten_2[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "dense (Dense)                   (None, 10)           5512330     concatenate[0][0]                \n",
            "__________________________________________________________________________________________________\n",
            "dense_1 (Dense)                 (None, 1)            11          dense[0][0]                      \n",
            "==================================================================================================\n",
            "Total params: 73,588,137\n",
            "Trainable params: 73,588,137\n",
            "Non-trainable params: 0\n",
            "__________________________________________________________________________________________________\n",
            "None\n",
            "Epoch 1/10\n",
            "981/981 [==============================] - 659s 672ms/step - loss: 1.2180e-07 - accuracy: 0.3701\n",
            "Epoch 2/10\n",
            "981/981 [==============================] - 658s 671ms/step - loss: 1.2180e-07 - accuracy: 0.3633\n",
            "Epoch 3/10\n",
            "981/981 [==============================] - 659s 672ms/step - loss: 1.2180e-07 - accuracy: 0.3691\n",
            "Epoch 4/10\n",
            "981/981 [==============================] - 656s 669ms/step - loss: 1.2180e-07 - accuracy: 0.3675\n",
            "Epoch 5/10\n",
            "981/981 [==============================] - 655s 668ms/step - loss: 1.2180e-07 - accuracy: 0.3697\n",
            "Epoch 6/10\n",
            "981/981 [==============================] - 657s 670ms/step - loss: 1.2180e-07 - accuracy: 0.3708\n",
            "Epoch 7/10\n",
            "981/981 [==============================] - 657s 670ms/step - loss: 1.2180e-07 - accuracy: 0.3672\n",
            "Epoch 8/10\n",
            "981/981 [==============================] - 664s 676ms/step - loss: 1.2180e-07 - accuracy: 0.3663\n",
            "Epoch 9/10\n",
            "981/981 [==============================] - 662s 675ms/step - loss: 1.2180e-07 - accuracy: 0.3661\n",
            "Epoch 10/10\n",
            "981/981 [==============================] - 662s 675ms/step - loss: 1.2180e-07 - accuracy: 0.3679\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/training/tracking/tracking.py:111: Model.state_updates (from tensorflow.python.keras.engine.training) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "This property should not be used in TensorFlow 2.0, as updates are applied automatically.\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/training/tracking/tracking.py:111: Layer.updates (from tensorflow.python.keras.engine.base_layer) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "This property should not be used in TensorFlow 2.0, as updates are applied automatically.\n",
            "INFO:tensorflow:Assets written to: Kim/Cano CNN/assets\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "eVH67DxsslPq",
        "outputId": "b2bf7a81-218b-4572-fac9-8cde9b640fd8"
      },
      "source": [
        "# report\r\n",
        "\r\n",
        "y_pred = model.predict(X_test, batch_size=64, verbose=1)\r\n",
        "y_pred_bool = np.argmax(y_pred, axis=1)\r\n",
        "\r\n",
        "print(classification_report(y_test, y_pred))"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "error",
          "ename": "TypeError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-27-adae8a2f48b2>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;31m# report\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      2\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 3\u001b[0;31m \u001b[0my_pred\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpredict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_test\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch_size\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m64\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mverbose\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      4\u001b[0m \u001b[0my_pred_bool\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0margmax\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0my_pred\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py\u001b[0m in \u001b[0;36m_method_wrapper\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m    128\u001b[0m       raise ValueError('{} is not supported in multi-worker mode.'.format(\n\u001b[1;32m    129\u001b[0m           method.__name__))\n\u001b[0;32m--> 130\u001b[0;31m     \u001b[0;32mreturn\u001b[0m \u001b[0mmethod\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    131\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    132\u001b[0m   return tf_decorator.make_decorator(\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py\u001b[0m in \u001b[0;36mpredict\u001b[0;34m(self, x, batch_size, verbose, steps, callbacks, max_queue_size, workers, use_multiprocessing)\u001b[0m\n\u001b[1;32m   1597\u001b[0m           \u001b[0;32mfor\u001b[0m \u001b[0mstep\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mdata_handler\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msteps\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1598\u001b[0m             \u001b[0mcallbacks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mon_predict_batch_begin\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mstep\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1599\u001b[0;31m             \u001b[0mtmp_batch_outputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpredict_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0miterator\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1600\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mdata_handler\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshould_sync\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1601\u001b[0m               \u001b[0mcontext\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0masync_wait\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/def_function.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, *args, **kwds)\u001b[0m\n\u001b[1;32m    778\u001b[0m       \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    779\u001b[0m         \u001b[0mcompiler\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m\"nonXla\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 780\u001b[0;31m         \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwds\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    781\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    782\u001b[0m       \u001b[0mnew_tracing_count\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_get_tracing_count\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/def_function.py\u001b[0m in \u001b[0;36m_call\u001b[0;34m(self, *args, **kwds)\u001b[0m\n\u001b[1;32m    812\u001b[0m       \u001b[0;31m# In this case we have not created variables on the first call. So we can\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    813\u001b[0m       \u001b[0;31m# run the first trace but we should fail if variables are created.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 814\u001b[0;31m       \u001b[0mresults\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_stateful_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwds\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    815\u001b[0m       \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_created_variables\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    816\u001b[0m         raise ValueError(\"Creating variables on a non-first call to a function\"\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/function.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   2826\u001b[0m     \u001b[0;34m\"\"\"Calls a graph function specialized to the inputs.\"\"\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2827\u001b[0m     \u001b[0;32mwith\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_lock\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2828\u001b[0;31m       \u001b[0mgraph_function\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_maybe_define_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2829\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0mgraph_function\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_filtered_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m  \u001b[0;31m# pylint: disable=protected-access\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2830\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/function.py\u001b[0m in \u001b[0;36m_maybe_define_function\u001b[0;34m(self, args, kwargs)\u001b[0m\n\u001b[1;32m   3208\u001b[0m           \u001b[0;32mand\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0minput_signature\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3209\u001b[0m           and call_context_key in self._function_cache.missed):\n\u001b[0;32m-> 3210\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_define_function_with_shape_relaxation\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   3211\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3212\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_function_cache\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmissed\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0madd\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcall_context_key\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/function.py\u001b[0m in \u001b[0;36m_define_function_with_shape_relaxation\u001b[0;34m(self, args, kwargs)\u001b[0m\n\u001b[1;32m   3140\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3141\u001b[0m     graph_function = self._create_graph_function(\n\u001b[0;32m-> 3142\u001b[0;31m         args, kwargs, override_flat_arg_shapes=relaxed_arg_shapes)\n\u001b[0m\u001b[1;32m   3143\u001b[0m     \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_function_cache\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0marg_relaxed\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mrank_only_cache_key\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mgraph_function\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3144\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/function.py\u001b[0m in \u001b[0;36m_create_graph_function\u001b[0;34m(self, args, kwargs, override_flat_arg_shapes)\u001b[0m\n\u001b[1;32m   3073\u001b[0m             \u001b[0marg_names\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0marg_names\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3074\u001b[0m             \u001b[0moverride_flat_arg_shapes\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0moverride_flat_arg_shapes\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 3075\u001b[0;31m             capture_by_value=self._capture_by_value),\n\u001b[0m\u001b[1;32m   3076\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_function_attributes\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3077\u001b[0m         \u001b[0mfunction_spec\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfunction_spec\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/func_graph.py\u001b[0m in \u001b[0;36mfunc_graph_from_py_func\u001b[0;34m(name, python_func, args, kwargs, signature, func_graph, autograph, autograph_options, add_control_dependencies, arg_names, op_return_value, collections, capture_by_value, override_flat_arg_shapes)\u001b[0m\n\u001b[1;32m    984\u001b[0m         \u001b[0m_\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moriginal_func\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf_decorator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0munwrap\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpython_func\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    985\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 986\u001b[0;31m       \u001b[0mfunc_outputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpython_func\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mfunc_args\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mfunc_kwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    987\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    988\u001b[0m       \u001b[0;31m# invariant: `func_outputs` contains only Tensors, CompositeTensors,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/def_function.py\u001b[0m in \u001b[0;36mwrapped_fn\u001b[0;34m(*args, **kwds)\u001b[0m\n\u001b[1;32m    598\u001b[0m         \u001b[0;31m# __wrapped__ allows AutoGraph to swap in a converted function. We give\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    599\u001b[0m         \u001b[0;31m# the function a weak reference to itself to avoid a reference cycle.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 600\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mweak_wrapped_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__wrapped__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwds\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    601\u001b[0m     \u001b[0mweak_wrapped_fn\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mweakref\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mref\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mwrapped_fn\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    602\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/func_graph.py\u001b[0m in \u001b[0;36mwrapper\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    971\u001b[0m           \u001b[0;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m  \u001b[0;31m# pylint:disable=broad-except\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    972\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mhasattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"ag_error_metadata\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 973\u001b[0;31m               \u001b[0;32mraise\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mag_error_metadata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto_exception\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    974\u001b[0m             \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    975\u001b[0m               \u001b[0;32mraise\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mTypeError\u001b[0m: in user code:\n\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:1462 predict_function  *\n        return step_function(self, iterator)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:1452 step_function  **\n        outputs = model.distribute_strategy.run(run_step, args=(data,))\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/distribute/distribute_lib.py:1211 run\n        return self._extended.call_for_each_replica(fn, args=args, kwargs=kwargs)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/distribute/distribute_lib.py:2585 call_for_each_replica\n        return self._call_for_each_replica(fn, args, kwargs)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/distribute/distribute_lib.py:2945 _call_for_each_replica\n        return fn(*args, **kwargs)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:1445 run_step  **\n        outputs = model.predict_step(data)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:1418 predict_step\n        return self(x, training=False)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/base_layer.py:985 __call__\n        outputs = call_fn(inputs, *args, **kwargs)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/functional.py:386 call\n        inputs, training=training, mask=mask)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/functional.py:508 _run_internal_graph\n        outputs = node.layer(*args, **kwargs)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/base_layer.py:985 __call__\n        outputs = call_fn(inputs, *args, **kwargs)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/layers/embeddings.py:189 call\n        out = embedding_ops.embedding_lookup_v2(self.embeddings, inputs)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/util/dispatch.py:201 wrapper\n        return target(*args, **kwargs)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/embedding_ops.py:394 embedding_lookup_v2\n        return embedding_lookup(params, ids, \"div\", name, max_norm=max_norm)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/util/dispatch.py:201 wrapper\n        return target(*args, **kwargs)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/embedding_ops.py:328 embedding_lookup\n        transform_fn=None)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/embedding_ops.py:134 _embedding_lookup_and_transform\n        ids = ops.convert_to_tensor(ids, name=\"ids\")\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/ops.py:1499 convert_to_tensor\n        ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/constant_op.py:338 _constant_tensor_conversion_function\n        return constant(v, dtype=dtype, name=name)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/constant_op.py:264 constant\n        allow_broadcast=True)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/constant_op.py:282 _constant_impl\n        allow_broadcast=allow_broadcast))\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/tensor_util.py:552 make_tensor_proto\n        \"supported type.\" % (type(values), values))\n\n    TypeError: Failed to convert object of type <class 'tensorflow.python.framework.sparse_tensor.SparseTensor'> to Tensor. Contents: SparseTensor(indices=Tensor(\"DeserializeSparse:0\", shape=(None, 2), dtype=int64), values=Tensor(\"functional_1/embedding/Cast:0\", shape=(None,), dtype=int32), dense_shape=Tensor(\"stack:0\", shape=(2,), dtype=int64)). Consider casting elements to a supported type.\n"
          ]
        }
      ]
    }
  ]
}