Project_Prediction.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Project.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyOMaY7bvWPMPIIW90iSjyjT",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/ishanjain1502/Youtube_adview-in-ML/blob/main/Project(revised).ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "W5byTzevWBW4"
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.cm as cm\n",
        "import matplotlib.pyplot as plt\n"
      ],
      "execution_count": 59,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-87fOGrmNiUR"
      },
      "source": [
        "data_t = pd.read_csv(\"/content/sample_data/train.csv\")"
      ],
      "execution_count": 60,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "id": "o1n1pvZqN1-9",
        "outputId": "e3b725bd-f2f8-42a1-a46b-1c6e1ae17f9b"
      },
      "source": [
        "data_t.head()\n"
      ],
      "execution_count": 61,
      "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>vidid</th>\n",
              "      <th>adview</th>\n",
              "      <th>views</th>\n",
              "      <th>likes</th>\n",
              "      <th>dislikes</th>\n",
              "      <th>comment</th>\n",
              "      <th>published</th>\n",
              "      <th>duration</th>\n",
              "      <th>category</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>VID_18655</td>\n",
              "      <td>40</td>\n",
              "      <td>1031602</td>\n",
              "      <td>8523</td>\n",
              "      <td>363</td>\n",
              "      <td>1095</td>\n",
              "      <td>2016-09-14</td>\n",
              "      <td>PT7M37S</td>\n",
              "      <td>F</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>VID_14135</td>\n",
              "      <td>2</td>\n",
              "      <td>1707</td>\n",
              "      <td>56</td>\n",
              "      <td>2</td>\n",
              "      <td>6</td>\n",
              "      <td>2016-10-01</td>\n",
              "      <td>PT9M30S</td>\n",
              "      <td>D</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>VID_2187</td>\n",
              "      <td>1</td>\n",
              "      <td>2023</td>\n",
              "      <td>25</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>2016-07-02</td>\n",
              "      <td>PT2M16S</td>\n",
              "      <td>C</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>VID_23096</td>\n",
              "      <td>6</td>\n",
              "      <td>620860</td>\n",
              "      <td>777</td>\n",
              "      <td>161</td>\n",
              "      <td>153</td>\n",
              "      <td>2016-07-27</td>\n",
              "      <td>PT4M22S</td>\n",
              "      <td>H</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>VID_10175</td>\n",
              "      <td>1</td>\n",
              "      <td>666</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2016-06-29</td>\n",
              "      <td>PT31S</td>\n",
              "      <td>D</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       vidid  adview    views likes  ... comment   published duration category\n",
              "0  VID_18655      40  1031602  8523  ...    1095  2016-09-14  PT7M37S        F\n",
              "1  VID_14135       2     1707    56  ...       6  2016-10-01  PT9M30S        D\n",
              "2   VID_2187       1     2023    25  ...       2  2016-07-02  PT2M16S        C\n",
              "3  VID_23096       6   620860   777  ...     153  2016-07-27  PT4M22S        H\n",
              "4  VID_10175       1      666     1  ...       0  2016-06-29    PT31S        D\n",
              "\n",
              "[5 rows x 9 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 61
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "U_MsqI-MOVXA",
        "outputId": "8725e0e8-3492-4b1c-bec7-d87f6d890a6e"
      },
      "source": [
        "data_t.shape"
      ],
      "execution_count": 62,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(14999, 9)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 62
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "id": "OQ1P4IsFO3zx",
        "outputId": "0ac18b0e-32fe-45ea-dd89-a4bbda2fe8fd"
      },
      "source": [
        "category = {'A': 1, 'B':2 , 'C':3, 'D':4 , 'E':5 , 'F': 6, 'G':7 , 'H':8 }\n",
        "data_t[\"category\"]= data_t[\"category\"].map(category)\n",
        "data_t.head()"
      ],
      "execution_count": 63,
      "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>vidid</th>\n",
              "      <th>adview</th>\n",
              "      <th>views</th>\n",
              "      <th>likes</th>\n",
              "      <th>dislikes</th>\n",
              "      <th>comment</th>\n",
              "      <th>published</th>\n",
              "      <th>duration</th>\n",
              "      <th>category</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>VID_18655</td>\n",
              "      <td>40</td>\n",
              "      <td>1031602</td>\n",
              "      <td>8523</td>\n",
              "      <td>363</td>\n",
              "      <td>1095</td>\n",
              "      <td>2016-09-14</td>\n",
              "      <td>PT7M37S</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>VID_14135</td>\n",
              "      <td>2</td>\n",
              "      <td>1707</td>\n",
              "      <td>56</td>\n",
              "      <td>2</td>\n",
              "      <td>6</td>\n",
              "      <td>2016-10-01</td>\n",
              "      <td>PT9M30S</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>VID_2187</td>\n",
              "      <td>1</td>\n",
              "      <td>2023</td>\n",
              "      <td>25</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>2016-07-02</td>\n",
              "      <td>PT2M16S</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>VID_23096</td>\n",
              "      <td>6</td>\n",
              "      <td>620860</td>\n",
              "      <td>777</td>\n",
              "      <td>161</td>\n",
              "      <td>153</td>\n",
              "      <td>2016-07-27</td>\n",
              "      <td>PT4M22S</td>\n",
              "      <td>8</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>VID_10175</td>\n",
              "      <td>1</td>\n",
              "      <td>666</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2016-06-29</td>\n",
              "      <td>PT31S</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       vidid  adview    views likes  ... comment   published duration category\n",
              "0  VID_18655      40  1031602  8523  ...    1095  2016-09-14  PT7M37S        6\n",
              "1  VID_14135       2     1707    56  ...       6  2016-10-01  PT9M30S        4\n",
              "2   VID_2187       1     2023    25  ...       2  2016-07-02  PT2M16S        3\n",
              "3  VID_23096       6   620860   777  ...     153  2016-07-27  PT4M22S        8\n",
              "4  VID_10175       1      666     1  ...       0  2016-06-29    PT31S        4\n",
              "\n",
              "[5 rows x 9 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 63
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kWUmV-9PQMMf"
      },
      "source": [
        "data_t = data_t[data_t.views !='F']\n",
        "data_t = data_t[data_t.likes !='F']\n",
        "data_t = data_t[data_t.dislikes !='F']\n",
        "data_t = data_t[data_t.comment !='F']"
      ],
      "execution_count": 64,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8GaAQ1z1QyZd"
      },
      "source": [
        "data_t[\"views\"] = pd.to_numeric(data_t[\"views\"])\n",
        "data_t[\"likes\"] = pd.to_numeric(data_t[\"likes\"])\n",
        "data_t[\"dislikes\"] = pd.to_numeric(data_t[\"dislikes\"])\n",
        "data_t[\"comment\"] = pd.to_numeric(data_t[\"comment\"])\n",
        "data_t[\"adview\"] = pd.to_numeric(data_t[\"adview\"])"
      ],
      "execution_count": 65,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gXV8RDUbRPww"
      },
      "source": [
        "col_vid = data_t[\"vidid\"]"
      ],
      "execution_count": 66,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8kyBT1gzRplS"
      },
      "source": [
        "from sklearn.preprocessing import LabelEncoder\n",
        "data_t[\"duration\"] = LabelEncoder().fit_transform(data_t[\"duration\"])\n",
        "data_t[\"vidid\"] = LabelEncoder().fit_transform(data_t[\"vidid\"])\n",
        "data_t[\"published\"] = LabelEncoder().fit_transform(data_t[\"published\"])"
      ],
      "execution_count": 67,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "id": "u2OEPYE6pgCJ",
        "outputId": "79b55752-df3f-4b32-f3c4-387f11ba876b"
      },
      "source": [
        "data_t.head()"
      ],
      "execution_count": 68,
      "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>vidid</th>\n",
              "      <th>adview</th>\n",
              "      <th>views</th>\n",
              "      <th>likes</th>\n",
              "      <th>dislikes</th>\n",
              "      <th>comment</th>\n",
              "      <th>published</th>\n",
              "      <th>duration</th>\n",
              "      <th>category</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>5912</td>\n",
              "      <td>40</td>\n",
              "      <td>1031602</td>\n",
              "      <td>8523</td>\n",
              "      <td>363</td>\n",
              "      <td>1095</td>\n",
              "      <td>2168</td>\n",
              "      <td>2925</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2741</td>\n",
              "      <td>2</td>\n",
              "      <td>1707</td>\n",
              "      <td>56</td>\n",
              "      <td>2</td>\n",
              "      <td>6</td>\n",
              "      <td>2185</td>\n",
              "      <td>3040</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>8138</td>\n",
              "      <td>1</td>\n",
              "      <td>2023</td>\n",
              "      <td>25</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>2094</td>\n",
              "      <td>1863</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>9005</td>\n",
              "      <td>6</td>\n",
              "      <td>620860</td>\n",
              "      <td>777</td>\n",
              "      <td>161</td>\n",
              "      <td>153</td>\n",
              "      <td>2119</td>\n",
              "      <td>2546</td>\n",
              "      <td>8</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>122</td>\n",
              "      <td>1</td>\n",
              "      <td>666</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2091</td>\n",
              "      <td>1963</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   vidid  adview    views  likes  ...  comment  published  duration  category\n",
              "0   5912      40  1031602   8523  ...     1095       2168      2925         6\n",
              "1   2741       2     1707     56  ...        6       2185      3040         4\n",
              "2   8138       1     2023     25  ...        2       2094      1863         3\n",
              "3   9005       6   620860    777  ...      153       2119      2546         8\n",
              "4    122       1      666      1  ...        0       2091      1963         4\n",
              "\n",
              "[5 rows x 9 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 68
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "X5mWwpH7qtWk"
      },
      "source": [
        "import datetime\n",
        "import time"
      ],
      "execution_count": 69,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7VTb_Bu6qyjD"
      },
      "source": [
        "def checki(x):\n",
        "  y = x[2:]\n",
        "  h = ''\n",
        "  m = ''\n",
        "  s = ''\n",
        "  mm = ''\n",
        "  P = ['H','M','S']\n",
        "  for i in y:\n",
        "    if i not in P:\n",
        "      mm+=i\n",
        "    else:\n",
        "      if(i==\"H\"):\n",
        "        h = mm\n",
        "        mm = ''\n",
        "      elif(i == \"M\"):\n",
        "        m = mm\n",
        "        mm = ''\n",
        "      else:\n",
        "        s = mm\n",
        "        mm = ''\n",
        "    if(h==''):\n",
        "      h = '00'\n",
        "    if(m == ''):\n",
        "      m = '00'\n",
        "    if(s==''):\n",
        "      s='00'\n",
        "  bp = h+':'+m+':'+s\n",
        "  return bp\n",
        "train=pd.read_csv(\"/content/sample_data/train.csv\")\n",
        "mp = pd.read_csv(\"/content/sample_data/train.csv\")[\"duration\"]\n",
        "time = mp.apply(checki)"
      ],
      "execution_count": 70,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SPY6MLMstNUs"
      },
      "source": [
        "def func_sec(time_string):\n",
        "  h, m, s = time_string.split(':')\n",
        "  return int(h) * 3600 + int(m) * 60 + int(s)\n",
        "  "
      ],
      "execution_count": 71,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3Zop5yvavpKW"
      },
      "source": [
        "time1 = time.apply(func_sec)"
      ],
      "execution_count": 72,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "id": "4tri7HK4wbAo",
        "outputId": "9c7117ef-3595-49d5-8404-ac483901246f"
      },
      "source": [
        "data_t[\"duration\"] = time1\n",
        "data_t.head()"
      ],
      "execution_count": 73,
      "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>vidid</th>\n",
              "      <th>adview</th>\n",
              "      <th>views</th>\n",
              "      <th>likes</th>\n",
              "      <th>dislikes</th>\n",
              "      <th>comment</th>\n",
              "      <th>published</th>\n",
              "      <th>duration</th>\n",
              "      <th>category</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>5912</td>\n",
              "      <td>40</td>\n",
              "      <td>1031602</td>\n",
              "      <td>8523</td>\n",
              "      <td>363</td>\n",
              "      <td>1095</td>\n",
              "      <td>2168</td>\n",
              "      <td>457</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2741</td>\n",
              "      <td>2</td>\n",
              "      <td>1707</td>\n",
              "      <td>56</td>\n",
              "      <td>2</td>\n",
              "      <td>6</td>\n",
              "      <td>2185</td>\n",
              "      <td>570</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>8138</td>\n",
              "      <td>1</td>\n",
              "      <td>2023</td>\n",
              "      <td>25</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>2094</td>\n",
              "      <td>136</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>9005</td>\n",
              "      <td>6</td>\n",
              "      <td>620860</td>\n",
              "      <td>777</td>\n",
              "      <td>161</td>\n",
              "      <td>153</td>\n",
              "      <td>2119</td>\n",
              "      <td>262</td>\n",
              "      <td>8</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>122</td>\n",
              "      <td>1</td>\n",
              "      <td>666</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2091</td>\n",
              "      <td>31</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   vidid  adview    views  likes  ...  comment  published  duration  category\n",
              "0   5912      40  1031602   8523  ...     1095       2168       457         6\n",
              "1   2741       2     1707     56  ...        6       2185       570         4\n",
              "2   8138       1     2023     25  ...        2       2094       136         3\n",
              "3   9005       6   620860    777  ...      153       2119       262         8\n",
              "4    122       1      666      1  ...        0       2091        31         4\n",
              "\n",
              "[5 rows x 9 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 73
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 524
        },
        "id": "lFHBFrquw_Ox",
        "outputId": "5b321d90-c0e4-49b9-c35e-3f8ff7bdffd5"
      },
      "source": [
        "plt.hist(data_t[\"category\"])\n",
        "plt.show()\n",
        "plt.plot(data_t[\"adview\"])\n",
        "plt.show()"
      ],
      "execution_count": 74,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "58CncRtYx02M",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 276
        },
        "outputId": "f38ef2c8-959a-4be3-9a9d-ec52436f1086"
      },
      "source": [
        "data_t = data_t[data_t[\"adview\"]<2000000]\n",
        "plt.plot(data_t[\"adview\"])\n",
        "plt.show()"
      ],
      "execution_count": 75,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iorWu2OdxMfT"
      },
      "source": [
        "import seaborn as sns"
      ],
      "execution_count": 76,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 504
        },
        "id": "pz3p_n1g5aDR",
        "outputId": "52151b97-770a-475c-ebbc-f962c94aab82"
      },
      "source": [
        "f, ax = plt.subplots(figsize=(10,8))\n",
        "corr = data_t.corr()\n",
        "sns.heatmap(corr, mask = np.zeros_like(corr, dtype=np.bool), cmap=sns.diverging_palette(220 , 10, as_cmap =True),square=True,ax=ax, annot=True)"
      ],
      "execution_count": 77,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7fce924dc0f0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 77
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x576 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_Wv6sgLxjafL"
      },
      "source": [
        "from sklearn.model_selection import train_test_split "
      ],
      "execution_count": 79,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XuzlwOmJq-J3"
      },
      "source": [
        "Y_train = pd.DataFrame(data=data_t.iloc[:, 1].values , columns=['target'])\n",
        "data_t = data_t.drop([\"adview\"], axis =1)\n",
        "data_t = data_t.drop([\"vidid\"], axis =1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4bPur_XZuR7b"
      },
      "source": [
        "X_train, X_test, y_train, y_test = train_test_split(data_t, Y_train, test_size=0.2, random_state=42)"
      ],
      "execution_count": 84,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OhPTCM7ySy7e",
        "outputId": "d260780b-92f7-41bb-ba0d-6ac5ef9af266"
      },
      "source": [
        "X_train.shape"
      ],
      "execution_count": 85,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(11708, 7)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 85
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LgYMDvBvw5qp"
      },
      "source": [
        "from sklearn.preprocessing import MinMaxScaler\n",
        "scaler = MinMaxScaler()\n",
        "X_train= scaler.fit_transform(X_train)\n",
        "X_test= scaler.fit_transform(X_test)"
      ],
      "execution_count": 89,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "d-y9tL5d5XX8"
      },
      "source": [
        "from sklearn import metrics\n",
        "def print_error(X_test , y_test , model_name):\n",
        "  prediction = model_name.predict(X_test)\n",
        "  print('Mean Absolute Error:',metrics.mean_absolute_error(y_test, prediction))\n",
        "  print('Mean Squared Error:', metrics.mean_squared_error(y_test, prediction))\n",
        "  print('RMS error: ', np.sqrt(metrics.mean_squared_error(y_test, prediction)))\n"
      ],
      "execution_count": 90,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mdUXIwPdCqlb",
        "outputId": "f4c2ac28-913f-415e-9b10-6e80a55f93eb"
      },
      "source": [
        "from sklearn import linear_model\n",
        "linear_regression = linear_model.LinearRegression()\n",
        "linear_regression.fit(X_train,y_train)\n",
        "print_error(X_test,y_test,linear_regression)"
      ],
      "execution_count": 91,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mean Absolute Error: 1439.1521541471402\n",
            "Mean Squared Error: 26304869.430857968\n",
            "RMS error:  5128.827295869687\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Yq8hMaWXDWmo",
        "outputId": "b8787810-3696-49eb-a5b9-cde590149504"
      },
      "source": [
        "from sklearn.tree import DecisionTreeRegressor\n",
        "decision_tree_reg = DecisionTreeRegressor()\n",
        "decision_tree_reg.fit(X_train,y_train)\n",
        "print_error(X_test,y_test,decision_tree_reg)"
      ],
      "execution_count": 92,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mean Absolute Error: 1468.28381147541\n",
            "Mean Squared Error: 30838755.493510928\n",
            "RMS error:  5553.265300119465\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HCDZmx-6M0dQ",
        "outputId": "e3d3aa51-3e37-47ea-e470-65e1f8d99d6b"
      },
      "source": [
        "from sklearn.ensemble import RandomForestRegressor\n",
        "n_estimators = 200\n",
        "max_d=25\n",
        "min_sample_split = 15\n",
        "min_sample_leaf = 2\n",
        "random_forest_reg = RandomForestRegressor(n_estimators = n_estimators, max_depth= max_d, min_samples_split=min_sample_split )\n",
        "random_forest_reg.fit(X_train,y_train)\n",
        "print_error(X_test,y_test, random_forest_reg)"
      ],
      "execution_count": 93,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:7: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
            "  import sys\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Mean Absolute Error: 1385.5471241298492\n",
            "Mean Squared Error: 19928582.702533893\n",
            "RMS error:  4464.144117581095\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2yrTgkRoVPz3",
        "outputId": "c8ca42ec-5a6c-4f88-ff39-a4c45859bbe5"
      },
      "source": [
        "from sklearn.svm import SVR\n",
        "support_vec_reg = SVR()\n",
        "support_vec_reg.fit(X_train,y_train)\n",
        "print_error(X_test,y_test, linear_regression)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/validation.py:760: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
            "  y = column_or_1d(y, warn=True)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Mean Absolute Error: 3707.378005824532\n",
            "Mean Squared Error: 835663131.1210337\n",
            "RMS error:  28907.83857573986\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "e9dnjGWwWnxw"
      },
      "source": [
        "import keras\n",
        "from keras.layers import Dense\n",
        "\n",
        "ann = keras.models.Sequential([\n",
        "                               Dense(12, activation=\"relu\",input_shape=X_train.shape[1:]),\n",
        "                               Dense(12, activation=\"relu\"),Dense(1)\n",
        "                              ])\n",
        "optimizer=keras.optimizers.Adam()\n",
        "loss=keras.losses.mean_squared_error\n",
        "ann.compile(optimizer=optimizer,loss=loss,metrics=[\"mean_squared_error\"])"
      ],
      "execution_count": 94,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "v_ZopyCZZkPf",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "53789347-eeab-4202-9c6f-4278c469187b"
      },
      "source": [
        "history=ann.fit(X_train,y_train,epochs=100)\n",
        "ann.summary()\n",
        "print_error(X_test,y_test,ann)"
      ],
      "execution_count": 95,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/100\n",
            "366/366 [==============================] - 0s 996us/step - loss: 83908440.0000 - mean_squared_error: 83908440.0000\n",
            "Epoch 2/100\n",
            "366/366 [==============================] - 0s 984us/step - loss: 83166088.0000 - mean_squared_error: 83166088.0000\n",
            "Epoch 3/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 80964240.0000 - mean_squared_error: 80964240.0000\n",
            "Epoch 4/100\n",
            "366/366 [==============================] - 0s 995us/step - loss: 78197632.0000 - mean_squared_error: 78197632.0000\n",
            "Epoch 5/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 76334776.0000 - mean_squared_error: 76334776.0000\n",
            "Epoch 6/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 75543968.0000 - mean_squared_error: 75543968.0000\n",
            "Epoch 7/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 75242528.0000 - mean_squared_error: 75242528.0000\n",
            "Epoch 8/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 75080896.0000 - mean_squared_error: 75080896.0000\n",
            "Epoch 9/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 74958464.0000 - mean_squared_error: 74958464.0000\n",
            "Epoch 10/100\n",
            "366/366 [==============================] - 0s 951us/step - loss: 74846144.0000 - mean_squared_error: 74846144.0000\n",
            "Epoch 11/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 74739896.0000 - mean_squared_error: 74739896.0000\n",
            "Epoch 12/100\n",
            "366/366 [==============================] - 0s 955us/step - loss: 74624488.0000 - mean_squared_error: 74624488.0000\n",
            "Epoch 13/100\n",
            "366/366 [==============================] - 0s 978us/step - loss: 74509288.0000 - mean_squared_error: 74509288.0000\n",
            "Epoch 14/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 74389496.0000 - mean_squared_error: 74389496.0000\n",
            "Epoch 15/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 74274664.0000 - mean_squared_error: 74274664.0000\n",
            "Epoch 16/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 74145288.0000 - mean_squared_error: 74145288.0000\n",
            "Epoch 17/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 74011176.0000 - mean_squared_error: 74011176.0000\n",
            "Epoch 18/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 73877648.0000 - mean_squared_error: 73877648.0000\n",
            "Epoch 19/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 73725744.0000 - mean_squared_error: 73725744.0000\n",
            "Epoch 20/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 73571752.0000 - mean_squared_error: 73571752.0000\n",
            "Epoch 21/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 73411640.0000 - mean_squared_error: 73411640.0000\n",
            "Epoch 22/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 73242696.0000 - mean_squared_error: 73242696.0000\n",
            "Epoch 23/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 73063048.0000 - mean_squared_error: 73063048.0000\n",
            "Epoch 24/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 72874304.0000 - mean_squared_error: 72874304.0000\n",
            "Epoch 25/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 72676688.0000 - mean_squared_error: 72676688.0000\n",
            "Epoch 26/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 72464128.0000 - mean_squared_error: 72464128.0000\n",
            "Epoch 27/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 72237032.0000 - mean_squared_error: 72237032.0000\n",
            "Epoch 28/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 71997512.0000 - mean_squared_error: 71997512.0000\n",
            "Epoch 29/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 71740304.0000 - mean_squared_error: 71740304.0000\n",
            "Epoch 30/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 71474872.0000 - mean_squared_error: 71474872.0000\n",
            "Epoch 31/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 71189904.0000 - mean_squared_error: 71189904.0000\n",
            "Epoch 32/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 70885200.0000 - mean_squared_error: 70885200.0000\n",
            "Epoch 33/100\n",
            "366/366 [==============================] - 0s 969us/step - loss: 70570336.0000 - mean_squared_error: 70570336.0000\n",
            "Epoch 34/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 70230728.0000 - mean_squared_error: 70230728.0000\n",
            "Epoch 35/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 69879368.0000 - mean_squared_error: 69879368.0000\n",
            "Epoch 36/100\n",
            "366/366 [==============================] - 1s 2ms/step - loss: 69499496.0000 - mean_squared_error: 69499496.0000\n",
            "Epoch 37/100\n",
            "366/366 [==============================] - 1s 2ms/step - loss: 69090152.0000 - mean_squared_error: 69090152.0000\n",
            "Epoch 38/100\n",
            "366/366 [==============================] - 1s 2ms/step - loss: 68665496.0000 - mean_squared_error: 68665496.0000\n",
            "Epoch 39/100\n",
            "366/366 [==============================] - 1s 2ms/step - loss: 68213256.0000 - mean_squared_error: 68213256.0000\n",
            "Epoch 40/100\n",
            "366/366 [==============================] - 1s 2ms/step - loss: 67735224.0000 - mean_squared_error: 67735224.0000\n",
            "Epoch 41/100\n",
            "366/366 [==============================] - 1s 2ms/step - loss: 67218152.0000 - mean_squared_error: 67218152.0000\n",
            "Epoch 42/100\n",
            "366/366 [==============================] - 1s 2ms/step - loss: 66671252.0000 - mean_squared_error: 66671252.0000\n",
            "Epoch 43/100\n",
            "366/366 [==============================] - 1s 2ms/step - loss: 66113760.0000 - mean_squared_error: 66113760.0000\n",
            "Epoch 44/100\n",
            "366/366 [==============================] - 1s 2ms/step - loss: 65543596.0000 - mean_squared_error: 65543596.0000\n",
            "Epoch 45/100\n",
            "366/366 [==============================] - 1s 2ms/step - loss: 64958564.0000 - mean_squared_error: 64958564.0000\n",
            "Epoch 46/100\n",
            "366/366 [==============================] - 1s 2ms/step - loss: 64351204.0000 - mean_squared_error: 64351204.0000\n",
            "Epoch 47/100\n",
            "366/366 [==============================] - 1s 2ms/step - loss: 63700128.0000 - mean_squared_error: 63700128.0000\n",
            "Epoch 48/100\n",
            "366/366 [==============================] - 1s 2ms/step - loss: 63015748.0000 - mean_squared_error: 63015748.0000\n",
            "Epoch 49/100\n",
            "366/366 [==============================] - 1s 2ms/step - loss: 62318540.0000 - mean_squared_error: 62318540.0000\n",
            "Epoch 50/100\n",
            "366/366 [==============================] - 1s 2ms/step - loss: 61575504.0000 - mean_squared_error: 61575504.0000\n",
            "Epoch 51/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 60786136.0000 - mean_squared_error: 60786136.0000\n",
            "Epoch 52/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 59979480.0000 - mean_squared_error: 59979480.0000\n",
            "Epoch 53/100\n",
            "366/366 [==============================] - 0s 991us/step - loss: 59152248.0000 - mean_squared_error: 59152248.0000\n",
            "Epoch 54/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 58294584.0000 - mean_squared_error: 58294584.0000\n",
            "Epoch 55/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 57422292.0000 - mean_squared_error: 57422292.0000\n",
            "Epoch 56/100\n",
            "366/366 [==============================] - 0s 997us/step - loss: 56543072.0000 - mean_squared_error: 56543072.0000\n",
            "Epoch 57/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 55638296.0000 - mean_squared_error: 55638296.0000\n",
            "Epoch 58/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 54689020.0000 - mean_squared_error: 54689020.0000\n",
            "Epoch 59/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 53707596.0000 - mean_squared_error: 53707596.0000\n",
            "Epoch 60/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 52704368.0000 - mean_squared_error: 52704368.0000\n",
            "Epoch 61/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 51673900.0000 - mean_squared_error: 51673900.0000\n",
            "Epoch 62/100\n",
            "366/366 [==============================] - 0s 990us/step - loss: 50653776.0000 - mean_squared_error: 50653776.0000\n",
            "Epoch 63/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 49607040.0000 - mean_squared_error: 49607040.0000\n",
            "Epoch 64/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 48529256.0000 - mean_squared_error: 48529256.0000\n",
            "Epoch 65/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 47415464.0000 - mean_squared_error: 47415464.0000\n",
            "Epoch 66/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 46303332.0000 - mean_squared_error: 46303332.0000\n",
            "Epoch 67/100\n",
            "366/366 [==============================] - 0s 997us/step - loss: 45215508.0000 - mean_squared_error: 45215508.0000\n",
            "Epoch 68/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 44128116.0000 - mean_squared_error: 44128116.0000\n",
            "Epoch 69/100\n",
            "366/366 [==============================] - 0s 961us/step - loss: 42961960.0000 - mean_squared_error: 42961960.0000\n",
            "Epoch 70/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 41801272.0000 - mean_squared_error: 41801272.0000\n",
            "Epoch 71/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 40620292.0000 - mean_squared_error: 40620292.0000\n",
            "Epoch 72/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 39395612.0000 - mean_squared_error: 39395612.0000\n",
            "Epoch 73/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 38189676.0000 - mean_squared_error: 38189676.0000\n",
            "Epoch 74/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 36972580.0000 - mean_squared_error: 36972580.0000\n",
            "Epoch 75/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 35746772.0000 - mean_squared_error: 35746772.0000\n",
            "Epoch 76/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 34498888.0000 - mean_squared_error: 34498888.0000\n",
            "Epoch 77/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 33270844.0000 - mean_squared_error: 33270844.0000\n",
            "Epoch 78/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 32095746.0000 - mean_squared_error: 32095746.0000\n",
            "Epoch 79/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 30906404.0000 - mean_squared_error: 30906404.0000\n",
            "Epoch 80/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 29713534.0000 - mean_squared_error: 29713534.0000\n",
            "Epoch 81/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 28494748.0000 - mean_squared_error: 28494748.0000\n",
            "Epoch 82/100\n",
            "366/366 [==============================] - 0s 968us/step - loss: 27313656.0000 - mean_squared_error: 27313656.0000\n",
            "Epoch 83/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 26156782.0000 - mean_squared_error: 26156782.0000\n",
            "Epoch 84/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 25031472.0000 - mean_squared_error: 25031472.0000\n",
            "Epoch 85/100\n",
            "366/366 [==============================] - 0s 967us/step - loss: 23936956.0000 - mean_squared_error: 23936956.0000\n",
            "Epoch 86/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 22870784.0000 - mean_squared_error: 22870784.0000\n",
            "Epoch 87/100\n",
            "366/366 [==============================] - 0s 999us/step - loss: 21799820.0000 - mean_squared_error: 21799820.0000\n",
            "Epoch 88/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 20777528.0000 - mean_squared_error: 20777528.0000\n",
            "Epoch 89/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 19797084.0000 - mean_squared_error: 19797084.0000\n",
            "Epoch 90/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 18882846.0000 - mean_squared_error: 18882846.0000\n",
            "Epoch 91/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 17985174.0000 - mean_squared_error: 17985174.0000\n",
            "Epoch 92/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 17143026.0000 - mean_squared_error: 17143026.0000\n",
            "Epoch 93/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 16356071.0000 - mean_squared_error: 16356071.0000\n",
            "Epoch 94/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 15602117.0000 - mean_squared_error: 15602117.0000\n",
            "Epoch 95/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 14903076.0000 - mean_squared_error: 14903076.0000\n",
            "Epoch 96/100\n",
            "366/366 [==============================] - 0s 994us/step - loss: 14279301.0000 - mean_squared_error: 14279301.0000\n",
            "Epoch 97/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 13685695.0000 - mean_squared_error: 13685695.0000\n",
            "Epoch 98/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 13129589.0000 - mean_squared_error: 13129589.0000\n",
            "Epoch 99/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 12608768.0000 - mean_squared_error: 12608768.0000\n",
            "Epoch 100/100\n",
            "366/366 [==============================] - 0s 1ms/step - loss: 12135162.0000 - mean_squared_error: 12135162.0000\n",
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "dense (Dense)                (None, 12)                96        \n",
            "_________________________________________________________________\n",
            "dense_1 (Dense)              (None, 12)                156       \n",
            "_________________________________________________________________\n",
            "dense_2 (Dense)              (None, 1)                 13        \n",
            "=================================================================\n",
            "Total params: 265\n",
            "Trainable params: 265\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "Mean Absolute Error: 1455.7371369630016\n",
            "Mean Squared Error: 21020553.40580252\n",
            "RMS error:  4584.81770693258\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qLIDkKUaYcT1",
        "outputId": "b3495bab-796e-472f-b77d-9b5b07040170"
      },
      "source": [
        "import joblib\n",
        "joblib.dump(decision_tree_reg, \"decision_tree_youtube_adview.pkl\")"
      ],
      "execution_count": 96,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['decision_tree_youtube_adview.pkl']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 96
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sW6sPZnNbXLX"
      },
      "source": [
        "ann.save(\"ann_youtube_adview.h5\")"
      ],
      "execution_count": 97,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "id": "MVfa1QXRjoSQ",
        "outputId": "4ded1c26-020f-4d03-acdf-7aaf25d565a4"
      },
      "source": [
        "test_data= pd.read_csv(\"/content/sample_data/test.csv\")\n",
        "test_data.head()"
      ],
      "execution_count": 98,
      "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>vidid</th>\n",
              "      <th>views</th>\n",
              "      <th>likes</th>\n",
              "      <th>dislikes</th>\n",
              "      <th>comment</th>\n",
              "      <th>published</th>\n",
              "      <th>duration</th>\n",
              "      <th>category</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>VID_1054</td>\n",
              "      <td>440238</td>\n",
              "      <td>6153</td>\n",
              "      <td>218</td>\n",
              "      <td>1377</td>\n",
              "      <td>2017-02-18</td>\n",
              "      <td>PT7M29S</td>\n",
              "      <td>B</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>VID_18629</td>\n",
              "      <td>1040132</td>\n",
              "      <td>8171</td>\n",
              "      <td>340</td>\n",
              "      <td>1047</td>\n",
              "      <td>2016-06-28</td>\n",
              "      <td>PT6M29S</td>\n",
              "      <td>F</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>VID_13967</td>\n",
              "      <td>28534</td>\n",
              "      <td>31</td>\n",
              "      <td>11</td>\n",
              "      <td>1</td>\n",
              "      <td>2014-03-10</td>\n",
              "      <td>PT37M54S</td>\n",
              "      <td>D</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>VID_19442</td>\n",
              "      <td>1316715</td>\n",
              "      <td>2284</td>\n",
              "      <td>250</td>\n",
              "      <td>274</td>\n",
              "      <td>2010-06-05</td>\n",
              "      <td>PT9M55S</td>\n",
              "      <td>G</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>VID_770</td>\n",
              "      <td>1893173</td>\n",
              "      <td>2519</td>\n",
              "      <td>225</td>\n",
              "      <td>116</td>\n",
              "      <td>2016-09-03</td>\n",
              "      <td>PT3M8S</td>\n",
              "      <td>B</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       vidid    views likes dislikes comment   published  duration category\n",
              "0   VID_1054   440238  6153      218    1377  2017-02-18   PT7M29S        B\n",
              "1  VID_18629  1040132  8171      340    1047  2016-06-28   PT6M29S        F\n",
              "2  VID_13967    28534    31       11       1  2014-03-10  PT37M54S        D\n",
              "3  VID_19442  1316715  2284      250     274  2010-06-05   PT9M55S        G\n",
              "4    VID_770  1893173  2519      225     116  2016-09-03    PT3M8S        B"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 98
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "V52USV9PkB3m",
        "outputId": "c8ecb94d-375f-4658-ae9a-d5a70a8dce9a"
      },
      "source": [
        "test_data.shape"
      ],
      "execution_count": 99,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(8764, 8)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 99
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "id": "GiZ7dzim0wF0",
        "outputId": "0950f44a-6d19-464d-bd59-241d6deb45e2"
      },
      "source": [
        "category = {'A': 1, 'B':2 , 'C':3, 'D':4 , 'E':5 , 'F': 6, 'G':7 , 'H':8 }\n",
        "test_data[\"category\"]= test_data[\"category\"].map(category)\n",
        "test_data.head()"
      ],
      "execution_count": 100,
      "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>vidid</th>\n",
              "      <th>views</th>\n",
              "      <th>likes</th>\n",
              "      <th>dislikes</th>\n",
              "      <th>comment</th>\n",
              "      <th>published</th>\n",
              "      <th>duration</th>\n",
              "      <th>category</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>VID_1054</td>\n",
              "      <td>440238</td>\n",
              "      <td>6153</td>\n",
              "      <td>218</td>\n",
              "      <td>1377</td>\n",
              "      <td>2017-02-18</td>\n",
              "      <td>PT7M29S</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>VID_18629</td>\n",
              "      <td>1040132</td>\n",
              "      <td>8171</td>\n",
              "      <td>340</td>\n",
              "      <td>1047</td>\n",
              "      <td>2016-06-28</td>\n",
              "      <td>PT6M29S</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>VID_13967</td>\n",
              "      <td>28534</td>\n",
              "      <td>31</td>\n",
              "      <td>11</td>\n",
              "      <td>1</td>\n",
              "      <td>2014-03-10</td>\n",
              "      <td>PT37M54S</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>VID_19442</td>\n",
              "      <td>1316715</td>\n",
              "      <td>2284</td>\n",
              "      <td>250</td>\n",
              "      <td>274</td>\n",
              "      <td>2010-06-05</td>\n",
              "      <td>PT9M55S</td>\n",
              "      <td>7</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>VID_770</td>\n",
              "      <td>1893173</td>\n",
              "      <td>2519</td>\n",
              "      <td>225</td>\n",
              "      <td>116</td>\n",
              "      <td>2016-09-03</td>\n",
              "      <td>PT3M8S</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       vidid    views likes dislikes comment   published  duration  category\n",
              "0   VID_1054   440238  6153      218    1377  2017-02-18   PT7M29S         2\n",
              "1  VID_18629  1040132  8171      340    1047  2016-06-28   PT6M29S         6\n",
              "2  VID_13967    28534    31       11       1  2014-03-10  PT37M54S         4\n",
              "3  VID_19442  1316715  2284      250     274  2010-06-05   PT9M55S         7\n",
              "4    VID_770  1893173  2519      225     116  2016-09-03    PT3M8S         2"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 100
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aYMi-4It1N35"
      },
      "source": [
        "test_data = test_data[test_data.views !='F']\n",
        "test_data = test_data[test_data.likes !='F']\n",
        "test_data = test_data[test_data.dislikes !='F']\n",
        "test_data = test_data[test_data.comment !='F']"
      ],
      "execution_count": 101,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HJlaRkYL1Zl_"
      },
      "source": [
        "test_data[\"views\"] = pd.to_numeric(test_data[\"views\"])\n",
        "test_data[\"likes\"] = pd.to_numeric(test_data[\"likes\"])\n",
        "test_data[\"dislikes\"] = pd.to_numeric(test_data[\"dislikes\"])\n",
        "test_data[\"comment\"] = pd.to_numeric(test_data[\"comment\"])"
      ],
      "execution_count": 102,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "id": "7TANWvUP1lK0",
        "outputId": "0019fda1-e6c4-45c3-9714-a48dbcc1bc3f"
      },
      "source": [
        "test_data.head()"
      ],
      "execution_count": 103,
      "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>vidid</th>\n",
              "      <th>views</th>\n",
              "      <th>likes</th>\n",
              "      <th>dislikes</th>\n",
              "      <th>comment</th>\n",
              "      <th>published</th>\n",
              "      <th>duration</th>\n",
              "      <th>category</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>VID_1054</td>\n",
              "      <td>440238</td>\n",
              "      <td>6153</td>\n",
              "      <td>218</td>\n",
              "      <td>1377</td>\n",
              "      <td>2017-02-18</td>\n",
              "      <td>PT7M29S</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>VID_18629</td>\n",
              "      <td>1040132</td>\n",
              "      <td>8171</td>\n",
              "      <td>340</td>\n",
              "      <td>1047</td>\n",
              "      <td>2016-06-28</td>\n",
              "      <td>PT6M29S</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>VID_13967</td>\n",
              "      <td>28534</td>\n",
              "      <td>31</td>\n",
              "      <td>11</td>\n",
              "      <td>1</td>\n",
              "      <td>2014-03-10</td>\n",
              "      <td>PT37M54S</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>VID_19442</td>\n",
              "      <td>1316715</td>\n",
              "      <td>2284</td>\n",
              "      <td>250</td>\n",
              "      <td>274</td>\n",
              "      <td>2010-06-05</td>\n",
              "      <td>PT9M55S</td>\n",
              "      <td>7</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>VID_770</td>\n",
              "      <td>1893173</td>\n",
              "      <td>2519</td>\n",
              "      <td>225</td>\n",
              "      <td>116</td>\n",
              "      <td>2016-09-03</td>\n",
              "      <td>PT3M8S</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       vidid    views  likes  dislikes  comment   published  duration  category\n",
              "0   VID_1054   440238   6153       218     1377  2017-02-18   PT7M29S         2\n",
              "1  VID_18629  1040132   8171       340     1047  2016-06-28   PT6M29S         6\n",
              "2  VID_13967    28534     31        11        1  2014-03-10  PT37M54S         4\n",
              "3  VID_19442  1316715   2284       250      274  2010-06-05   PT9M55S         7\n",
              "4    VID_770  1893173   2519       225      116  2016-09-03    PT3M8S         2"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 103
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "18vrMues2L8L"
      },
      "source": [
        "from sklearn.preprocessing import LabelEncoder\n",
        "test_data[\"duration\"] = LabelEncoder().fit_transform(test_data[\"duration\"])\n",
        "test_data[\"published\"] = LabelEncoder().fit_transform(test_data[\"published\"])"
      ],
      "execution_count": 104,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "id": "GWf6g6R_2rke",
        "outputId": "854d728a-38f0-4c0c-8968-e84264b57f2b"
      },
      "source": [
        "test_data.head()"
      ],
      "execution_count": 105,
      "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>vidid</th>\n",
              "      <th>views</th>\n",
              "      <th>likes</th>\n",
              "      <th>dislikes</th>\n",
              "      <th>comment</th>\n",
              "      <th>published</th>\n",
              "      <th>duration</th>\n",
              "      <th>category</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>VID_1054</td>\n",
              "      <td>440238</td>\n",
              "      <td>6153</td>\n",
              "      <td>218</td>\n",
              "      <td>1377</td>\n",
              "      <td>2053</td>\n",
              "      <td>2115</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>VID_18629</td>\n",
              "      <td>1040132</td>\n",
              "      <td>8171</td>\n",
              "      <td>340</td>\n",
              "      <td>1047</td>\n",
              "      <td>1825</td>\n",
              "      <td>2055</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>VID_13967</td>\n",
              "      <td>28534</td>\n",
              "      <td>31</td>\n",
              "      <td>11</td>\n",
              "      <td>1</td>\n",
              "      <td>1009</td>\n",
              "      <td>1506</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>VID_19442</td>\n",
              "      <td>1316715</td>\n",
              "      <td>2284</td>\n",
              "      <td>250</td>\n",
              "      <td>274</td>\n",
              "      <td>116</td>\n",
              "      <td>2265</td>\n",
              "      <td>7</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>VID_770</td>\n",
              "      <td>1893173</td>\n",
              "      <td>2519</td>\n",
              "      <td>225</td>\n",
              "      <td>116</td>\n",
              "      <td>1892</td>\n",
              "      <td>1625</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       vidid    views  likes  dislikes  comment  published  duration  category\n",
              "0   VID_1054   440238   6153       218     1377       2053      2115         2\n",
              "1  VID_18629  1040132   8171       340     1047       1825      2055         6\n",
              "2  VID_13967    28534     31        11        1       1009      1506         4\n",
              "3  VID_19442  1316715   2284       250      274        116      2265         7\n",
              "4    VID_770  1893173   2519       225      116       1892      1625         2"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 105
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ibI4j6hD2zut"
      },
      "source": [
        "mp = pd.read_csv(\"/content/sample_data/test.csv\")[\"duration\"]\n",
        "time = mp.apply(checki)"
      ],
      "execution_count": 106,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ez_AQ25HSLT6"
      },
      "source": [
        "test_data = test_data.drop([\"vidid\"], axis =1)"
      ],
      "execution_count": 107,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "De3rFiwa3kpn"
      },
      "source": [
        "time2 = time.apply(func_sec)"
      ],
      "execution_count": 108,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "id": "duKr_v0A3mzs",
        "outputId": "03ff05a8-fc52-43ad-a731-5f9af605089c"
      },
      "source": [
        "test_data[\"duration\"] = time2\n",
        "test_data.head()"
      ],
      "execution_count": 109,
      "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>views</th>\n",
              "      <th>likes</th>\n",
              "      <th>dislikes</th>\n",
              "      <th>comment</th>\n",
              "      <th>published</th>\n",
              "      <th>duration</th>\n",
              "      <th>category</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>440238</td>\n",
              "      <td>6153</td>\n",
              "      <td>218</td>\n",
              "      <td>1377</td>\n",
              "      <td>2053</td>\n",
              "      <td>449</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1040132</td>\n",
              "      <td>8171</td>\n",
              "      <td>340</td>\n",
              "      <td>1047</td>\n",
              "      <td>1825</td>\n",
              "      <td>389</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>28534</td>\n",
              "      <td>31</td>\n",
              "      <td>11</td>\n",
              "      <td>1</td>\n",
              "      <td>1009</td>\n",
              "      <td>2274</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1316715</td>\n",
              "      <td>2284</td>\n",
              "      <td>250</td>\n",
              "      <td>274</td>\n",
              "      <td>116</td>\n",
              "      <td>595</td>\n",
              "      <td>7</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1893173</td>\n",
              "      <td>2519</td>\n",
              "      <td>225</td>\n",
              "      <td>116</td>\n",
              "      <td>1892</td>\n",
              "      <td>188</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "     views  likes  dislikes  comment  published  duration  category\n",
              "0   440238   6153       218     1377       2053       449         2\n",
              "1  1040132   8171       340     1047       1825       389         6\n",
              "2    28534     31        11        1       1009      2274         4\n",
              "3  1316715   2284       250      274        116       595         7\n",
              "4  1893173   2519       225      116       1892       188         2"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 109
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2RMboh9GVYBg"
      },
      "source": [
        "test_data = scaler.fit_transform(test_data)"
      ],
      "execution_count": 110,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RY1UixV6PR0y",
        "outputId": "8c7b53ea-36ae-4e05-aa83-6014017aab63"
      },
      "source": [
        "Y_pred = decision_tree_reg.predict(test_data)\n",
        "predictions = pd.DataFrame(data = Y_pred)\n",
        "predictions.shape"
      ],
      "execution_count": 118,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(8549, 1)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 118
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "id": "7K6dbDYzWeit",
        "outputId": "9e522348-7955-4697-f46d-5c03b062a24b"
      },
      "source": [
        "predictions = predictions.rename(columns = {0:'predictions'})\n",
        "predictions.head()"
      ],
      "execution_count": 119,
      "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>predictions</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>5467.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>7256.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>27.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2023.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2233.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   predictions\n",
              "0       5467.0\n",
              "1       7256.0\n",
              "2         27.0\n",
              "3       2023.0\n",
              "4       2233.0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 119
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "clcJGRiZehgv"
      },
      "source": [
        "predictions.to_csv('predictions.csv', encoding = 'utf-8' , index=True)"
      ],
      "execution_count": 127,
      "outputs": []
    }
  ]
}