predicting_video_memorability.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "wUMpQYmVP28s"
   },
   "source": [
    "Purpose of this notebook is\n",
    "\n",
    "1) to predict the short and long term memorability of the videos using pre-extracted assorted features, as follows: Aesthetic features, C3D, Captions, ColorHistogram, HMP, HOG, InceptionV3, LRP, ORB.\n",
    "\n",
    "We have considered HMP, C3D and Captions here.\n",
    "\n",
    "2) to calculate the spearmann score of the models used in above task"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Nffhz9IsQ_ZI"
   },
   "source": [
    "As prediction of memorability is a regression problem, we have used following regression models of machine learning for this:\n",
    "\n",
    "1) Linear regresssion\n",
    "\n",
    "2) Decision Trees\n",
    "\n",
    "3) K-NN\n",
    "\n",
    "4) Ensemble models - random forests, voting regressor, gradient boost, adaboost\n",
    "\n",
    "5) Neural Networks\n",
    "\n",
    "The purpose of this was to compare all of them and use the best for test-set."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "UP6UOv_PvOlO"
   },
   "source": [
    "*Few points to be noted*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "7Qz0Xfg4SVl9"
   },
   "source": [
    " - Basically, this is an attempt to implement whatever has been learnt since the start of the course.\n",
    " - Inspite of importing them initially, necessary libraries have been imported as per the requirement...in the middle of notebook.\n",
    " - Functions to read HMP and C3D features took too much time. This is why, extracted data has been saved as csv files"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "jHG19NQXSoQf"
   },
   "source": [
    "..."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "KVGQChU-Sodx"
   },
   "source": [
    "Following function has been taken from the assignment's tutorial. This is to calculate the Spearman's score."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "PvSLln3aP1BX"
   },
   "outputs": [],
   "source": [
    "def Get_score(Y_pred,Y_true):\n",
    "    '''Calculate the Spearmann\"s correlation coefficient'''\n",
    "    Y_pred = np.squeeze(Y_pred)\n",
    "    Y_true = np.squeeze(Y_true)\n",
    "    if Y_pred.shape != Y_true.shape:\n",
    "        print('Input shapes don\\'t match!')\n",
    "    else:\n",
    "        if len(Y_pred.shape) == 1:\n",
    "            Res = pd.DataFrame({'Y_true':Y_true,'Y_pred':Y_pred})\n",
    "            score_mat = Res[['Y_true','Y_pred']].corr(method='spearman',min_periods=1)\n",
    "            print('The Spearman\\'s correlation coefficient is: %.3f' % score_mat.iloc[1][0])\n",
    "        else:\n",
    "            for ii in range(Y_pred.shape[1]):\n",
    "                Get_score(Y_pred[:,ii],Y_true[:,ii])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "_afKGaHUTPaT"
   },
   "source": [
    "*Below lines will load the data from my DCU google drive. I think, if somebody else is running, they will have to change the path*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 130
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 60214,
     "status": "ok",
     "timestamp": 1588091543499,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "8LczXyyBTH_h",
    "outputId": "94199215-64d0-4b0f-ae91-8fa0c6c59b1d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Go to this URL in a browser: https://accounts.google.com/o/oauth2/auth?client_id=947318989803-6bn6qk8qdgf4n4g3pfee6491hc0brc4i.apps.googleusercontent.com&redirect_uri=urn%3aietf%3awg%3aoauth%3a2.0%3aoob&response_type=code&scope=email%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdocs.test%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdrive%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdrive.photos.readonly%20https%3a%2f%2fwww.googleapis.com%2fauth%2fpeopleapi.readonly\n",
      "\n",
      "Enter your authorization code:\n",
      "··········\n",
      "Mounted at /content/drive/\n"
     ]
    }
   ],
   "source": [
    "from google.colab import drive\n",
    "import os\n",
    "drive.mount('/content/drive/')\n",
    "os.chdir('/content/drive/My Drive/CA684_Assignment/')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "7rY2KhUMX0JL"
   },
   "outputs": [],
   "source": [
    "#import necessary libraries\n",
    "import pandas as pd\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "gsWL1SgsXT77"
   },
   "source": [
    "**Loading Ground truth csv file**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "cw4BfUH4XQj2"
   },
   "outputs": [],
   "source": [
    "gtPath ='./Dev-set/Ground-truth/'\n",
    "dfGroundTruth = pd.read_csv(gtPath+'ground-truth.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "H79xzu1bUCW3"
   },
   "source": [
    "**Loading assorted video features**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "xMSBSv7nvAH3"
   },
   "source": [
    "below function has been referenced from the tutorial"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "hvho1TD_uTMu"
   },
   "outputs": [],
   "source": [
    "# load labels and captions\n",
    "def read_caps(fname):\n",
    "    \"\"\"Load the captions into a dataframe\"\"\"\n",
    "    vn = []\n",
    "    cap = []\n",
    "    df = pd.DataFrame();\n",
    "    with open(fname) as f:\n",
    "        for line in f:\n",
    "            pairs = line.split()\n",
    "            vn.append(pairs[0])\n",
    "            cap.append(pairs[1])\n",
    "        df['video']=vn\n",
    "        df['caption']=cap\n",
    "    return df\n",
    "\n",
    "\n",
    "# load the captions - dev set\n",
    "cap_path = './Dev-set/Captions/dev-set_video-captions.txt'\n",
    "df_captions=read_caps(cap_path)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "FREnQl4TxOvW"
   },
   "source": [
    "below function is to read the HMP files"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "CZM_UYNuUBAA"
   },
   "outputs": [],
   "source": [
    "def readHMP(fname):\n",
    "    \"\"\"Scan HMP(Histogram of Motion Patterns) features from file\"\"\"\n",
    "    with open(fname) as fileName:\n",
    "        for eachLine in fileName:\n",
    "            pairs=eachLine.split()\n",
    "            #creating a temporary dictionary\n",
    "            tempHMP = { int(p.split(':')[0]) : float(p.split(':')[1]) for p in pairs}\n",
    "    # there are 6075 bins, fill zeros\n",
    "    HMP = np.zeros(6075)\n",
    "    for index in tempHMP.keys():\n",
    "        HMP[index-1] = tempHMP[index]            \n",
    "    return HMP"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "wMCU4U1Hu2OA"
   },
   "outputs": [],
   "source": [
    "#creating a new dataframe\n",
    "dfHMP_Dev = pd.DataFrame(columns = ['video', 'arrayInfo'])\n",
    "\n",
    "#path\n",
    "dirHMP_Dev = './Dev-set/HMP/'\n",
    "\n",
    "for eachFile in os.listdir(dirHMP_Dev):\n",
    "    if eachFile.endswith(\".txt\"):\n",
    "        path = os.path.join(dirHMP_Dev, eachFile)\n",
    "        arrayFile = readHMP(path)\n",
    "        eachFile= eachFile.replace(\".txt\",\".webm\")\n",
    "        dfHMP_Dev = dfHMP_Dev.append({'video': eachFile, 'arrayInfo': arrayFile}, ignore_index=True)\n",
    "    else:\n",
    "        break"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "WNKStJ0Cu2jT"
   },
   "outputs": [],
   "source": [
    "#saving the dataframe\n",
    "dfHMP_Dev.to_csv('/content/drive/My Drive/dfHMP_Dev.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "U_8IGc1yUdDv"
   },
   "source": [
    "Below line to read the extracted features directly. This saves overall time.\n",
    "\n",
    "*Below line will load the data from my DCU google drive. I think, if somebody else is running, they will have to change the path*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "cT42XP0u4xN8"
   },
   "source": [
    "https://drive.google.com/open?id=1VKCh_yEMGwtLWkvw6Mo_dl6ywOvENkv-\n",
    "\n",
    "above is the link for this CSV which can be open by anyone with DCU email id"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "hJjUi4-_9Qhg"
   },
   "outputs": [],
   "source": [
    "dfHMP_Dev = pd.read_csv('/content/drive/My Drive/dfHMP_Dev.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "gtyHFdBc9qbx"
   },
   "source": [
    "Below function will read C3D"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "ANaOMcDwCISW"
   },
   "outputs": [],
   "source": [
    "import tqdm\n",
    "from tqdm import tqdm_notebook as tqdm\n",
    "from pathlib import Path\n",
    "#from tqdm import tqdm.notebook.tqdm as tqdm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 123,
     "referenced_widgets": [
      "4dab5ddcbe6f4292b31146b889f0e8d2",
      "84b2ed2ee2ab461a9c59a25d2b15f172",
      "684e80f5eb634a43b34f62dea0b70fa4",
      "e264ef5b6df940d7bd37c2147d50f21e",
      "b45ee9693c074e36a0af3afdccb6713e",
      "f86015d0ce8348bca4fe3e0ed24b63d7",
      "64162473341143b9899b5f93977168f1",
      "6337f6faf68b4a398ed709ff42cd000f"
     ]
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1959965,
     "status": "ok",
     "timestamp": 1587966436604,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "dGriYrdi9u95",
    "outputId": "dadf806c-6087-4cfb-b6cf-2a25e6f794fb"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:15: TqdmDeprecationWarning: This function will be removed in tqdm==5.0.0\n",
      "Please use `tqdm.notebook.tqdm` instead of `tqdm.tqdm_notebook`\n",
      "  from ipykernel import kernelapp as app\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4dab5ddcbe6f4292b31146b889f0e8d2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=6000), HTML(value='')))"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "def readC3D(fname):\n",
    "    \"\"\"Scan vectors from file\"\"\"\n",
    "    with open(fname) as file:\n",
    "        for line in file:\n",
    "            C3D =[float(item) for item in line.split()] # convert to float type, using default separator\n",
    "    return C3D\n",
    "\n",
    "dirC3D_Dev = Path('./Dev-set/C3D/')\n",
    "#path\n",
    "#dirC3D_Dev = './Dev-set/C3D/'\n",
    "\n",
    "max_iterations = 6000\n",
    "c3dDict ={}\n",
    "\n",
    "for file in tqdm (list(dirC3D_Dev.glob('*.txt'))):\n",
    "  key = file.with_suffix('.webm').name\n",
    "  c3dDict[key] = readC3D(file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "toWB9w92-AXD"
   },
   "outputs": [],
   "source": [
    "dfC3D_Dev= pd.DataFrame(c3dDict).T\n",
    "dfC3D_Dev=pd.DataFrame(dfC3D_Dev).rename_axis('video')\n",
    "dfC3D_Dev.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "iWAb_wuMNQ54"
   },
   "outputs": [],
   "source": [
    "#saving the dataframe in collab\n",
    "dfC3D_Dev.to_csv('/content/drive/My Drive/dfC3D_Dev.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "i8UkYfW2Um40"
   },
   "source": [
    "Below line to read the extracted features directly. This saves overall time.\n",
    "\n",
    "*Below line will load the data from my DCU google drive. I think, if somebody else is running, they will have to change the path*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "h81nYjsF5FFH"
   },
   "source": [
    "https://drive.google.com/open?id=1ptJuFz18sWWfJQSnz8GAZ4sTVRryshTS\n",
    "\n",
    "above is the link for this CSV which can be open by anyone with DCU email id"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "V80AWp8XUmBq"
   },
   "outputs": [],
   "source": [
    "dfC3D_Dev = pd.read_csv('/content/drive/My Drive/dfC3D_Dev.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "c3CPERb8S355"
   },
   "source": [
    "**Data cleaning of Captions using count vectorizer and TF-IDF**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "6s6HRH-CCkMd"
   },
   "source": [
    "1. Below steps are for count vectorizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 144
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1576,
     "status": "ok",
     "timestamp": 1588091592967,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "zCQnlLKzu2oi",
    "outputId": "e26086c9-f755-4cb9-b492-1d1108728e95"
   },
   "outputs": [
    {
     "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>video</th>\n",
       "      <th>caption</th>\n",
       "      <th>short-term_memorability</th>\n",
       "      <th>nb_short-term_annotations</th>\n",
       "      <th>long-term_memorability</th>\n",
       "      <th>nb_long-term_annotations</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>video3.webm</td>\n",
       "      <td>blonde-woman-is-massaged-tilt-down</td>\n",
       "      <td>0.924</td>\n",
       "      <td>34</td>\n",
       "      <td>0.846</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>video4.webm</td>\n",
       "      <td>roulette-table-spinning-with-ball-in-closeup-shot</td>\n",
       "      <td>0.923</td>\n",
       "      <td>33</td>\n",
       "      <td>0.667</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>video6.webm</td>\n",
       "      <td>khr-gangsters</td>\n",
       "      <td>0.863</td>\n",
       "      <td>33</td>\n",
       "      <td>0.700</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         video  ... nb_long-term_annotations\n",
       "0  video3.webm  ...                       13\n",
       "1  video4.webm  ...                       12\n",
       "2  video6.webm  ...                       10\n",
       "\n",
       "[3 rows x 6 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfCaptions_Dev = df_captions.merge(dfGroundTruth,on=[\"video\"],how=\"inner\")\n",
    "dfCaptions_Dev.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "e2Skua9GRimJ"
   },
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import CountVectorizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "a2dVJzR9TyeZ"
   },
   "outputs": [],
   "source": [
    "#english stopwords will be removed from here\n",
    "countVect=CountVectorizer(min_df=1, stop_words='english')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 36
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1730,
     "status": "ok",
     "timestamp": 1588091604069,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "4V6csCBERjow",
    "outputId": "f7e08a11-5f2e-426d-d4f9-f8412418aa6f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(6000, 5019)\n"
     ]
    }
   ],
   "source": [
    "xCounts=countVect.fit_transform(dfCaptions_Dev['caption'])\n",
    "print(xCounts.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "7GYgxuFHRlsv"
   },
   "outputs": [],
   "source": [
    "xCounts=countVect.fit_transform(dfCaptions_Dev['caption'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 147
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2228,
     "status": "ok",
     "timestamp": 1588091614684,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "3AQ8LDPgRoU8",
    "outputId": "cda4a986-373b-4c84-a0fd-419c6ee642ea"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0, 0, 0, ..., 0, 0, 0],\n",
       "       [0, 0, 0, ..., 0, 0, 0],\n",
       "       [0, 0, 0, ..., 0, 0, 0],\n",
       "       ...,\n",
       "       [0, 0, 0, ..., 0, 0, 0],\n",
       "       [0, 0, 0, ..., 0, 0, 0],\n",
       "       [0, 0, 0, ..., 0, 0, 0]])"
      ]
     },
     "execution_count": 17,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xCountsSeq=xCounts.toarray()\n",
    "xCountsSeq"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 36
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1967,
     "status": "ok",
     "timestamp": 1588091619144,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "TsUQKDWeRqwk",
    "outputId": "1c36dddb-c0ee-4e5f-d29d-b6e609b9b1cd"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0 0 0 ... 0 0 0]\n"
     ]
    }
   ],
   "source": [
    "print(xCountsSeq[3])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "OETyfOJqC_bd"
   },
   "source": [
    "2. Below steps are for TF-IDF"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "ecno9ivCDYfy"
   },
   "source": [
    "reference of few function has been taken from the tutorial"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 36
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1657,
     "status": "ok",
     "timestamp": 1588091623646,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "w-YjTl3zhqQl",
    "outputId": "40ef7714-0514-455b-dcc6-2c20c0060cff"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'!\"#$%&\\'()*+,-./:;<=>?@[\\\\]^_`{|}~'"
      ]
     },
     "execution_count": 19,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import string\n",
    "string.punctuation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "82hiBpwihr3Q"
   },
   "outputs": [],
   "source": [
    "#function to remove punctuation\n",
    "def fixingPunctuation(text):\n",
    "  fixedText=\"\".join([word for word in text if word not in string.punctuation])  #it will discard all punctuation\n",
    "  return fixedText"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 144
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1759,
     "status": "ok",
     "timestamp": 1588091630214,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "jT9XHk5_noX2",
    "outputId": "8deaf477-2412-4316-dbec-62bb6848c582"
   },
   "outputs": [
    {
     "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>video</th>\n",
       "      <th>caption</th>\n",
       "      <th>short-term_memorability</th>\n",
       "      <th>nb_short-term_annotations</th>\n",
       "      <th>long-term_memorability</th>\n",
       "      <th>nb_long-term_annotations</th>\n",
       "      <th>updatedCaption</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>video3.webm</td>\n",
       "      <td>blonde-woman-is-massaged-tilt-down</td>\n",
       "      <td>0.924</td>\n",
       "      <td>34</td>\n",
       "      <td>0.846</td>\n",
       "      <td>13</td>\n",
       "      <td>blondewomanismassagedtiltdown</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>video4.webm</td>\n",
       "      <td>roulette-table-spinning-with-ball-in-closeup-shot</td>\n",
       "      <td>0.923</td>\n",
       "      <td>33</td>\n",
       "      <td>0.667</td>\n",
       "      <td>12</td>\n",
       "      <td>roulettetablespinningwithballincloseupshot</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>video6.webm</td>\n",
       "      <td>khr-gangsters</td>\n",
       "      <td>0.863</td>\n",
       "      <td>33</td>\n",
       "      <td>0.700</td>\n",
       "      <td>10</td>\n",
       "      <td>khrgangsters</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         video  ...                              updatedCaption\n",
       "0  video3.webm  ...               blondewomanismassagedtiltdown\n",
       "1  video4.webm  ...  roulettetablespinningwithballincloseupshot\n",
       "2  video6.webm  ...                                khrgangsters\n",
       "\n",
       "[3 rows x 7 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfCaptions_Dev['updatedCaption']=dfCaptions_Dev['caption'].apply(lambda x: fixingPunctuation(x))\n",
    "dfCaptions_Dev.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "JwK6xB1rD3E2"
   },
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import TfidfVectorizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "nKfW_WCIEFOZ"
   },
   "outputs": [],
   "source": [
    "vectorizer=TfidfVectorizer(max_features=1000)\n",
    "#response=vectorizer.fit_transform(dfCaptions_Dev.updatedCaption).toarray()\n",
    "#above line is commented because updatedCaptions didn't perform well. I will use captions now.\n",
    "response=vectorizer.fit_transform(dfCaptions_Dev.caption).toarray()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 129
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3002,
     "status": "ok",
     "timestamp": 1588092018202,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "4KpXiprDEFTy",
    "outputId": "9e0d3056-2a58-479e-a444-498578c029e1"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "         video  ...                                  vectorizedCaption\n",
      "0  video3.webm  ...  [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ...\n",
      "1  video4.webm  ...  [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ...\n",
      "2  video6.webm  ...  [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ...\n",
      "\n",
      "[3 rows x 8 columns]\n"
     ]
    }
   ],
   "source": [
    "dfCaptions_Dev['vectorizedCaption']=pd.Series(response.tolist())\n",
    "print(dfCaptions_Dev.head(3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 166
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2954,
     "status": "ok",
     "timestamp": 1588092018209,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "JcOAgDWXEh9j",
    "outputId": "d6e15317-6d50-4687-fb07-860939298b14"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['video',\n",
       " 'caption',\n",
       " 'short-term_memorability',\n",
       " 'nb_short-term_annotations',\n",
       " 'long-term_memorability',\n",
       " 'nb_long-term_annotations',\n",
       " 'updatedCaption',\n",
       " 'vectorizedCaption']"
      ]
     },
     "execution_count": 42,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(dfCaptions_Dev.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "ay92rKSntOpA"
   },
   "outputs": [],
   "source": [
    "#dfCaptions_Dev_tfIdf = dfCaptions_Dev[['vectorizedCaption']].copy()\n",
    "dfCaptions_Dev_tfIdf=pd.DataFrame(dfCaptions_Dev['vectorizedCaption'].tolist())\n",
    "#dfCaptions_Dev_tfIdf['video']=pd.Series(dfCaptions_Dev['video'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 193
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 4152,
     "status": "ok",
     "timestamp": 1588092019476,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "YIuVapTRtO1a",
    "outputId": "bbf63704-b73e-40d1-da03-b4bfe6e93b58"
   },
   "outputs": [
    {
     "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "      <th>17</th>\n",
       "      <th>18</th>\n",
       "      <th>19</th>\n",
       "      <th>20</th>\n",
       "      <th>21</th>\n",
       "      <th>22</th>\n",
       "      <th>23</th>\n",
       "      <th>24</th>\n",
       "      <th>25</th>\n",
       "      <th>26</th>\n",
       "      <th>27</th>\n",
       "      <th>28</th>\n",
       "      <th>29</th>\n",
       "      <th>30</th>\n",
       "      <th>31</th>\n",
       "      <th>32</th>\n",
       "      <th>33</th>\n",
       "      <th>34</th>\n",
       "      <th>35</th>\n",
       "      <th>36</th>\n",
       "      <th>37</th>\n",
       "      <th>38</th>\n",
       "      <th>39</th>\n",
       "      <th>...</th>\n",
       "      <th>960</th>\n",
       "      <th>961</th>\n",
       "      <th>962</th>\n",
       "      <th>963</th>\n",
       "      <th>964</th>\n",
       "      <th>965</th>\n",
       "      <th>966</th>\n",
       "      <th>967</th>\n",
       "      <th>968</th>\n",
       "      <th>969</th>\n",
       "      <th>970</th>\n",
       "      <th>971</th>\n",
       "      <th>972</th>\n",
       "      <th>973</th>\n",
       "      <th>974</th>\n",
       "      <th>975</th>\n",
       "      <th>976</th>\n",
       "      <th>977</th>\n",
       "      <th>978</th>\n",
       "      <th>979</th>\n",
       "      <th>980</th>\n",
       "      <th>981</th>\n",
       "      <th>982</th>\n",
       "      <th>983</th>\n",
       "      <th>984</th>\n",
       "      <th>985</th>\n",
       "      <th>986</th>\n",
       "      <th>987</th>\n",
       "      <th>988</th>\n",
       "      <th>989</th>\n",
       "      <th>990</th>\n",
       "      <th>991</th>\n",
       "      <th>992</th>\n",
       "      <th>993</th>\n",
       "      <th>994</th>\n",
       "      <th>995</th>\n",
       "      <th>996</th>\n",
       "      <th>997</th>\n",
       "      <th>998</th>\n",
       "      <th>999</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.201449</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.226147</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 1000 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   0    1    2    3    4    5    6    ...  993  994  995  996  997  998  999\n",
       "0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...  0.0  0.0  0.0  0.0  0.0  0.0  0.0\n",
       "1  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...  0.0  0.0  0.0  0.0  0.0  0.0  0.0\n",
       "2  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...  0.0  0.0  0.0  0.0  0.0  0.0  0.0\n",
       "\n",
       "[3 rows x 1000 columns]"
      ]
     },
     "execution_count": 44,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfCaptions_Dev_tfIdf.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "tR5eVUwrpFAE"
   },
   "source": [
    "*below lines are to merge captions(tf-idf) with HMP*\n",
    "\n",
    "these lines are updated later after comapring the performance of count vectorizer for captions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 267,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 129
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2733,
     "status": "ok",
     "timestamp": 1588078448146,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "bNFGg-gyQzVr",
    "outputId": "da6d8e12-d94d-485d-9e9f-201f585fa29d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     0    1  ...  long-term_memorability  nb_long-term_annotations\n",
      "0  0.0  0.0  ...                   0.846                        13\n",
      "1  0.0  0.0  ...                   0.667                        12\n",
      "2  0.0  0.0  ...                   0.700                        10\n",
      "\n",
      "[3 rows x 1107 columns]\n"
     ]
    }
   ],
   "source": [
    "dfDevCaptions_HMP=pd.DataFrame(dfCaptions_Dev['vectorizedCaption'].tolist())\n",
    "dfDevCaptions_HMP['video']=pd.Series(dfCaptions_Dev['video'])\n",
    "dfDevCaptions_HMP=pd.merge(dfDevCaptions_HMP, dfHMP_Dev, on = 'video')\n",
    "\n",
    "print(dfDevCaptions_HMP.head(3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "b_ahf44x2b99"
   },
   "outputs": [],
   "source": [
    "dfDevCaptions_HMP_final = dfDevCaptions_HMP.drop(columns=['long-term_memorability', 'short-term_memorability', 'video', 'nb_long-term_annotations', 'nb_short-term_annotations'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 273,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 193
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1535,
     "status": "ok",
     "timestamp": 1588078769243,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "SBQSXwAb3tJM",
    "outputId": "c391f900-e1ba-4dc4-a7c6-9a381cd5afcd"
   },
   "outputs": [
    {
     "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "      <th>17</th>\n",
       "      <th>18</th>\n",
       "      <th>19</th>\n",
       "      <th>20</th>\n",
       "      <th>21</th>\n",
       "      <th>22</th>\n",
       "      <th>23</th>\n",
       "      <th>24</th>\n",
       "      <th>25</th>\n",
       "      <th>26</th>\n",
       "      <th>27</th>\n",
       "      <th>28</th>\n",
       "      <th>29</th>\n",
       "      <th>30</th>\n",
       "      <th>31</th>\n",
       "      <th>32</th>\n",
       "      <th>33</th>\n",
       "      <th>34</th>\n",
       "      <th>35</th>\n",
       "      <th>36</th>\n",
       "      <th>37</th>\n",
       "      <th>38</th>\n",
       "      <th>39</th>\n",
       "      <th>...</th>\n",
       "      <th>61</th>\n",
       "      <th>62</th>\n",
       "      <th>63</th>\n",
       "      <th>64</th>\n",
       "      <th>65</th>\n",
       "      <th>66</th>\n",
       "      <th>67</th>\n",
       "      <th>68</th>\n",
       "      <th>69</th>\n",
       "      <th>70</th>\n",
       "      <th>71</th>\n",
       "      <th>72</th>\n",
       "      <th>73</th>\n",
       "      <th>74</th>\n",
       "      <th>75</th>\n",
       "      <th>76</th>\n",
       "      <th>77</th>\n",
       "      <th>78</th>\n",
       "      <th>79</th>\n",
       "      <th>80</th>\n",
       "      <th>81</th>\n",
       "      <th>82</th>\n",
       "      <th>83</th>\n",
       "      <th>84</th>\n",
       "      <th>85</th>\n",
       "      <th>86</th>\n",
       "      <th>87</th>\n",
       "      <th>88</th>\n",
       "      <th>89</th>\n",
       "      <th>90</th>\n",
       "      <th>91</th>\n",
       "      <th>92</th>\n",
       "      <th>93</th>\n",
       "      <th>94</th>\n",
       "      <th>95</th>\n",
       "      <th>96</th>\n",
       "      <th>97</th>\n",
       "      <th>98</th>\n",
       "      <th>99</th>\n",
       "      <th>100</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.096647</td>\n",
       "      <td>0.004458</td>\n",
       "      <td>0.011840</td>\n",
       "      <td>0.000287</td>\n",
       "      <td>0.002195</td>\n",
       "      <td>0.002792</td>\n",
       "      <td>0.000018</td>\n",
       "      <td>0.003365</td>\n",
       "      <td>0.000061</td>\n",
       "      <td>0.000573</td>\n",
       "      <td>0.000273</td>\n",
       "      <td>0.000034</td>\n",
       "      <td>0.006737</td>\n",
       "      <td>0.000021</td>\n",
       "      <td>0.000098</td>\n",
       "      <td>0.000025</td>\n",
       "      <td>0.005768</td>\n",
       "      <td>2.990000e-06</td>\n",
       "      <td>0.000085</td>\n",
       "      <td>0.000607</td>\n",
       "      <td>0.000461</td>\n",
       "      <td>0.001189</td>\n",
       "      <td>0.000270</td>\n",
       "      <td>0.000127</td>\n",
       "      <td>0.000106</td>\n",
       "      <td>0.000154</td>\n",
       "      <td>0.000567</td>\n",
       "      <td>0.000020</td>\n",
       "      <td>0.000894</td>\n",
       "      <td>0.001079</td>\n",
       "      <td>0.001042</td>\n",
       "      <td>0.000161</td>\n",
       "      <td>0.000257</td>\n",
       "      <td>0.046617</td>\n",
       "      <td>0.000156</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>0.000537</td>\n",
       "      <td>0.000339</td>\n",
       "      <td>0.008437</td>\n",
       "      <td>0.000470</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000372</td>\n",
       "      <td>0.000154</td>\n",
       "      <td>0.000148</td>\n",
       "      <td>0.000086</td>\n",
       "      <td>0.000267</td>\n",
       "      <td>0.000014</td>\n",
       "      <td>0.001430</td>\n",
       "      <td>0.004417</td>\n",
       "      <td>0.000064</td>\n",
       "      <td>0.000105</td>\n",
       "      <td>0.003129</td>\n",
       "      <td>0.000055</td>\n",
       "      <td>0.000061</td>\n",
       "      <td>0.000025</td>\n",
       "      <td>0.004283</td>\n",
       "      <td>0.000010</td>\n",
       "      <td>0.000062</td>\n",
       "      <td>2.903000e-05</td>\n",
       "      <td>0.000374</td>\n",
       "      <td>0.000139</td>\n",
       "      <td>0.741702</td>\n",
       "      <td>0.007178</td>\n",
       "      <td>0.000346</td>\n",
       "      <td>0.000104</td>\n",
       "      <td>0.000181</td>\n",
       "      <td>0.000052</td>\n",
       "      <td>0.004839</td>\n",
       "      <td>0.000078</td>\n",
       "      <td>0.042325</td>\n",
       "      <td>0.000623</td>\n",
       "      <td>0.000582</td>\n",
       "      <td>0.000393</td>\n",
       "      <td>0.000864</td>\n",
       "      <td>0.000947</td>\n",
       "      <td>0.000136</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>0.000360</td>\n",
       "      <td>0.000159</td>\n",
       "      <td>0.001025</td>\n",
       "      <td>0.000020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.876291</td>\n",
       "      <td>0.005509</td>\n",
       "      <td>0.008852</td>\n",
       "      <td>0.000399</td>\n",
       "      <td>0.000069</td>\n",
       "      <td>0.013166</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>0.000044</td>\n",
       "      <td>0.000520</td>\n",
       "      <td>0.000046</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>0.000024</td>\n",
       "      <td>0.010491</td>\n",
       "      <td>0.000475</td>\n",
       "      <td>0.000023</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>0.005554</td>\n",
       "      <td>2.500000e-07</td>\n",
       "      <td>0.000057</td>\n",
       "      <td>0.000328</td>\n",
       "      <td>0.000149</td>\n",
       "      <td>0.001844</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>0.000103</td>\n",
       "      <td>0.000030</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000179</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>0.000215</td>\n",
       "      <td>0.000224</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000031</td>\n",
       "      <td>0.002538</td>\n",
       "      <td>0.000104</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000064</td>\n",
       "      <td>0.005380</td>\n",
       "      <td>0.001027</td>\n",
       "      <td>0.001384</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 1102 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     0    1    2    3    4  ...        96        97        98        99       100\n",
       "0  0.0  0.0  0.0  0.0  0.0  ...  0.000006  0.000537  0.000339  0.008437  0.000470\n",
       "1  0.0  0.0  0.0  0.0  0.0  ...  0.000007  0.000360  0.000159  0.001025  0.000020\n",
       "2  0.0  0.0  0.0  0.0  0.0  ...  0.000005  0.000064  0.005380  0.001027  0.001384\n",
       "\n",
       "[3 rows x 1102 columns]"
      ]
     },
     "execution_count": 273,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfDevCaptions_HMP_final.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "TU9O_Gt0wvXd"
   },
   "source": [
    "*below lines are to merge captions(tf-idf) with C3D*\n",
    "\n",
    "these lines are updated later after comapring the performance of count vectorizer for captions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 297,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 129
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2631,
     "status": "ok",
     "timestamp": 1588079795933,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "NS57jl6SQza0",
    "outputId": "173ae572-ead3-4e0a-e054-fafdc0c1f579"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     0    1  ...  long-term_memorability  nb_long-term_annotations\n",
      "0  0.0  0.0  ...                   0.846                        13\n",
      "1  0.0  0.0  ...                   0.667                        12\n",
      "2  0.0  0.0  ...                   0.700                        10\n",
      "\n",
      "[3 rows x 1106 columns]\n"
     ]
    }
   ],
   "source": [
    "dfDevCaptions_C3D=pd.DataFrame(dfCaptions_Dev['vectorizedCaption'].tolist())\n",
    "dfDevCaptions_C3D['video']=pd.Series(dfCaptions_Dev['video'])\n",
    "dfDevCaptions_C3D=pd.merge(dfDevCaptions_C3D, dfC3D_Dev, on = 'video')\n",
    "\n",
    "print(dfDevCaptions_C3D.head(3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "_LniE5-N7p51"
   },
   "outputs": [],
   "source": [
    "dfDevCaptions_C3D_final = dfDevCaptions_C3D.drop(columns=['long-term_memorability', 'short-term_memorability', 'video', 'nb_long-term_annotations', 'nb_short-term_annotations'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 299,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 256
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1467,
     "status": "ok",
     "timestamp": 1588079847690,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "oh1fdigyeN83",
    "outputId": "1c0a2049-0742-4022-c754-1c3ef123caa1"
   },
   "outputs": [
    {
     "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "      <th>17</th>\n",
       "      <th>18</th>\n",
       "      <th>19</th>\n",
       "      <th>20</th>\n",
       "      <th>21</th>\n",
       "      <th>22</th>\n",
       "      <th>23</th>\n",
       "      <th>24</th>\n",
       "      <th>25</th>\n",
       "      <th>26</th>\n",
       "      <th>27</th>\n",
       "      <th>28</th>\n",
       "      <th>29</th>\n",
       "      <th>30</th>\n",
       "      <th>31</th>\n",
       "      <th>32</th>\n",
       "      <th>33</th>\n",
       "      <th>34</th>\n",
       "      <th>35</th>\n",
       "      <th>36</th>\n",
       "      <th>37</th>\n",
       "      <th>38</th>\n",
       "      <th>39</th>\n",
       "      <th>...</th>\n",
       "      <th>61</th>\n",
       "      <th>62</th>\n",
       "      <th>63</th>\n",
       "      <th>64</th>\n",
       "      <th>65</th>\n",
       "      <th>66</th>\n",
       "      <th>67</th>\n",
       "      <th>68</th>\n",
       "      <th>69</th>\n",
       "      <th>70</th>\n",
       "      <th>71</th>\n",
       "      <th>72</th>\n",
       "      <th>73</th>\n",
       "      <th>74</th>\n",
       "      <th>75</th>\n",
       "      <th>76</th>\n",
       "      <th>77</th>\n",
       "      <th>78</th>\n",
       "      <th>79</th>\n",
       "      <th>80</th>\n",
       "      <th>81</th>\n",
       "      <th>82</th>\n",
       "      <th>83</th>\n",
       "      <th>84</th>\n",
       "      <th>85</th>\n",
       "      <th>86</th>\n",
       "      <th>87</th>\n",
       "      <th>88</th>\n",
       "      <th>89</th>\n",
       "      <th>90</th>\n",
       "      <th>91</th>\n",
       "      <th>92</th>\n",
       "      <th>93</th>\n",
       "      <th>94</th>\n",
       "      <th>95</th>\n",
       "      <th>96</th>\n",
       "      <th>97</th>\n",
       "      <th>98</th>\n",
       "      <th>99</th>\n",
       "      <th>100</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.096647</td>\n",
       "      <td>0.004458</td>\n",
       "      <td>0.011840</td>\n",
       "      <td>0.000287</td>\n",
       "      <td>0.002195</td>\n",
       "      <td>0.002792</td>\n",
       "      <td>0.000018</td>\n",
       "      <td>0.003365</td>\n",
       "      <td>0.000061</td>\n",
       "      <td>0.000573</td>\n",
       "      <td>0.000273</td>\n",
       "      <td>0.000034</td>\n",
       "      <td>0.006737</td>\n",
       "      <td>0.000021</td>\n",
       "      <td>0.000098</td>\n",
       "      <td>2.502000e-05</td>\n",
       "      <td>0.005768</td>\n",
       "      <td>2.990000e-06</td>\n",
       "      <td>0.000085</td>\n",
       "      <td>0.000607</td>\n",
       "      <td>0.000461</td>\n",
       "      <td>0.001189</td>\n",
       "      <td>0.000270</td>\n",
       "      <td>0.000127</td>\n",
       "      <td>0.000106</td>\n",
       "      <td>0.000154</td>\n",
       "      <td>0.000567</td>\n",
       "      <td>0.000020</td>\n",
       "      <td>0.000894</td>\n",
       "      <td>0.001079</td>\n",
       "      <td>0.001042</td>\n",
       "      <td>0.000161</td>\n",
       "      <td>0.000257</td>\n",
       "      <td>0.046617</td>\n",
       "      <td>0.000156</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>0.000537</td>\n",
       "      <td>0.000339</td>\n",
       "      <td>0.008437</td>\n",
       "      <td>0.000470</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000372</td>\n",
       "      <td>0.000154</td>\n",
       "      <td>0.000148</td>\n",
       "      <td>0.000086</td>\n",
       "      <td>0.000267</td>\n",
       "      <td>0.000014</td>\n",
       "      <td>0.001430</td>\n",
       "      <td>0.004417</td>\n",
       "      <td>0.000064</td>\n",
       "      <td>0.000105</td>\n",
       "      <td>0.003129</td>\n",
       "      <td>0.000055</td>\n",
       "      <td>0.000061</td>\n",
       "      <td>0.000025</td>\n",
       "      <td>0.004283</td>\n",
       "      <td>1.032000e-05</td>\n",
       "      <td>0.000062</td>\n",
       "      <td>2.903000e-05</td>\n",
       "      <td>0.000374</td>\n",
       "      <td>0.000139</td>\n",
       "      <td>0.741702</td>\n",
       "      <td>0.007178</td>\n",
       "      <td>0.000346</td>\n",
       "      <td>0.000104</td>\n",
       "      <td>0.000181</td>\n",
       "      <td>0.000052</td>\n",
       "      <td>0.004839</td>\n",
       "      <td>0.000078</td>\n",
       "      <td>0.042325</td>\n",
       "      <td>0.000623</td>\n",
       "      <td>0.000582</td>\n",
       "      <td>0.000393</td>\n",
       "      <td>0.000864</td>\n",
       "      <td>0.000947</td>\n",
       "      <td>0.000136</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>0.000360</td>\n",
       "      <td>0.000159</td>\n",
       "      <td>0.001025</td>\n",
       "      <td>0.000020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.876291</td>\n",
       "      <td>0.005509</td>\n",
       "      <td>0.008852</td>\n",
       "      <td>0.000399</td>\n",
       "      <td>0.000069</td>\n",
       "      <td>0.013166</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>0.000044</td>\n",
       "      <td>0.000520</td>\n",
       "      <td>0.000046</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>0.000024</td>\n",
       "      <td>0.010491</td>\n",
       "      <td>0.000475</td>\n",
       "      <td>0.000023</td>\n",
       "      <td>5.590000e-06</td>\n",
       "      <td>0.005554</td>\n",
       "      <td>2.500000e-07</td>\n",
       "      <td>0.000057</td>\n",
       "      <td>0.000328</td>\n",
       "      <td>0.000149</td>\n",
       "      <td>0.001844</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>0.000103</td>\n",
       "      <td>0.000030</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000179</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>0.000215</td>\n",
       "      <td>0.000224</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000031</td>\n",
       "      <td>0.002538</td>\n",
       "      <td>0.000104</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000064</td>\n",
       "      <td>0.005380</td>\n",
       "      <td>0.001027</td>\n",
       "      <td>0.001384</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.590434</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000009</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000539</td>\n",
       "      <td>0.000035</td>\n",
       "      <td>0.000391</td>\n",
       "      <td>0.000021</td>\n",
       "      <td>0.000043</td>\n",
       "      <td>0.000010</td>\n",
       "      <td>0.000028</td>\n",
       "      <td>0.012471</td>\n",
       "      <td>0.006671</td>\n",
       "      <td>0.001187</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>0.119488</td>\n",
       "      <td>5.600000e-07</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>4.250000e-06</td>\n",
       "      <td>0.000040</td>\n",
       "      <td>0.000148</td>\n",
       "      <td>0.103989</td>\n",
       "      <td>0.058989</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>0.000069</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>0.000130</td>\n",
       "      <td>0.001182</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>0.000044</td>\n",
       "      <td>0.000046</td>\n",
       "      <td>0.000009</td>\n",
       "      <td>0.000023</td>\n",
       "      <td>0.000053</td>\n",
       "      <td>0.000048</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>0.000380</td>\n",
       "      <td>0.000029</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000236</td>\n",
       "      <td>0.000046</td>\n",
       "      <td>0.000027</td>\n",
       "      <td>0.000203</td>\n",
       "      <td>0.010644</td>\n",
       "      <td>0.000252</td>\n",
       "      <td>0.000349</td>\n",
       "      <td>0.001776</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>0.001443</td>\n",
       "      <td>0.000146</td>\n",
       "      <td>0.019003</td>\n",
       "      <td>0.000584</td>\n",
       "      <td>0.000027</td>\n",
       "      <td>0.000692</td>\n",
       "      <td>5.650000e-06</td>\n",
       "      <td>0.000403</td>\n",
       "      <td>2.894700e-04</td>\n",
       "      <td>0.000197</td>\n",
       "      <td>0.001254</td>\n",
       "      <td>0.005257</td>\n",
       "      <td>0.045808</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>0.000296</td>\n",
       "      <td>0.003893</td>\n",
       "      <td>0.000318</td>\n",
       "      <td>0.000048</td>\n",
       "      <td>0.000018</td>\n",
       "      <td>0.000263</td>\n",
       "      <td>0.000026</td>\n",
       "      <td>0.000037</td>\n",
       "      <td>0.000690</td>\n",
       "      <td>0.000171</td>\n",
       "      <td>0.000231</td>\n",
       "      <td>0.000637</td>\n",
       "      <td>0.000040</td>\n",
       "      <td>0.000061</td>\n",
       "      <td>0.000075</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>0.001323</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 1101 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     0    1    2    3    4  ...        96        97        98        99       100\n",
       "0  0.0  0.0  0.0  0.0  0.0  ...  0.000006  0.000537  0.000339  0.008437  0.000470\n",
       "1  0.0  0.0  0.0  0.0  0.0  ...  0.000007  0.000360  0.000159  0.001025  0.000020\n",
       "2  0.0  0.0  0.0  0.0  0.0  ...  0.000005  0.000064  0.005380  0.001027  0.001384\n",
       "3  0.0  0.0  0.0  0.0  0.0  ...  0.000019  0.000001  0.000004  0.000380  0.000029\n",
       "4  0.0  0.0  0.0  0.0  0.0  ...  0.000040  0.000061  0.000075  0.000002  0.001323\n",
       "\n",
       "[5 rows x 1101 columns]"
      ]
     },
     "execution_count": 299,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfDevCaptions_C3D_final.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "BDpf6cJuS1oU"
   },
   "source": [
    "**HMP and joining ground truth with it**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 210
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1355,
     "status": "ok",
     "timestamp": 1588067862692,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "PJ0mfgrdODLD",
    "outputId": "af7f86ed-d264-43a2-ffd8-7c70ca719c0b"
   },
   "outputs": [
    {
     "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>Unnamed: 0</th>\n",
       "      <th>video</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "      <th>17</th>\n",
       "      <th>18</th>\n",
       "      <th>19</th>\n",
       "      <th>20</th>\n",
       "      <th>21</th>\n",
       "      <th>22</th>\n",
       "      <th>23</th>\n",
       "      <th>24</th>\n",
       "      <th>25</th>\n",
       "      <th>26</th>\n",
       "      <th>27</th>\n",
       "      <th>28</th>\n",
       "      <th>29</th>\n",
       "      <th>30</th>\n",
       "      <th>31</th>\n",
       "      <th>32</th>\n",
       "      <th>33</th>\n",
       "      <th>34</th>\n",
       "      <th>35</th>\n",
       "      <th>36</th>\n",
       "      <th>37</th>\n",
       "      <th>...</th>\n",
       "      <th>65</th>\n",
       "      <th>66</th>\n",
       "      <th>67</th>\n",
       "      <th>68</th>\n",
       "      <th>69</th>\n",
       "      <th>70</th>\n",
       "      <th>71</th>\n",
       "      <th>72</th>\n",
       "      <th>73</th>\n",
       "      <th>74</th>\n",
       "      <th>75</th>\n",
       "      <th>76</th>\n",
       "      <th>77</th>\n",
       "      <th>78</th>\n",
       "      <th>79</th>\n",
       "      <th>80</th>\n",
       "      <th>81</th>\n",
       "      <th>82</th>\n",
       "      <th>83</th>\n",
       "      <th>84</th>\n",
       "      <th>85</th>\n",
       "      <th>86</th>\n",
       "      <th>87</th>\n",
       "      <th>88</th>\n",
       "      <th>89</th>\n",
       "      <th>90</th>\n",
       "      <th>91</th>\n",
       "      <th>92</th>\n",
       "      <th>93</th>\n",
       "      <th>94</th>\n",
       "      <th>95</th>\n",
       "      <th>96</th>\n",
       "      <th>97</th>\n",
       "      <th>98</th>\n",
       "      <th>99</th>\n",
       "      <th>100</th>\n",
       "      <th>short-term_memorability</th>\n",
       "      <th>nb_short-term_annotations</th>\n",
       "      <th>long-term_memorability</th>\n",
       "      <th>nb_long-term_annotations</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>video6632.webm</td>\n",
       "      <td>0.010858</td>\n",
       "      <td>0.010386</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.700000e-07</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>3.400000e-07</td>\n",
       "      <td>8.000000e-08</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>0.000105</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4.000000e-08</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.331800e-04</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>7.600000e-07</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.800000e-07</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>6.000000e-08</td>\n",
       "      <td>0.000021</td>\n",
       "      <td>5.000000e-08</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>3.000000e-08</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000032</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>2.500000e-07</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>5.000000e-08</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>4.200000e-07</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>3.000000e-08</td>\n",
       "      <td>2.000000e-08</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>6.000000e-08</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.000000e-08</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>1.100000e-07</td>\n",
       "      <td>1.400000e-07</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.700000e-07</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>1.300000e-06</td>\n",
       "      <td>2.600000e-06</td>\n",
       "      <td>8.000000e-08</td>\n",
       "      <td>0.884</td>\n",
       "      <td>31</td>\n",
       "      <td>0.900</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>video6634.webm</td>\n",
       "      <td>0.000200</td>\n",
       "      <td>0.000065</td>\n",
       "      <td>0.993807</td>\n",
       "      <td>2.000000e-07</td>\n",
       "      <td>4.700000e-07</td>\n",
       "      <td>0.000073</td>\n",
       "      <td>3.700000e-06</td>\n",
       "      <td>0.000337</td>\n",
       "      <td>6.710000e-06</td>\n",
       "      <td>2.290000e-06</td>\n",
       "      <td>6.380000e-06</td>\n",
       "      <td>7.340000e-06</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>3.810000e-06</td>\n",
       "      <td>1.411000e-05</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>2.300000e-07</td>\n",
       "      <td>1.900000e-07</td>\n",
       "      <td>0.002938</td>\n",
       "      <td>5.200000e-07</td>\n",
       "      <td>9.200000e-07</td>\n",
       "      <td>1.136000e-05</td>\n",
       "      <td>0.000011</td>\n",
       "      <td>3.310400e-04</td>\n",
       "      <td>9.200000e-07</td>\n",
       "      <td>0.000022</td>\n",
       "      <td>8.000000e-08</td>\n",
       "      <td>0.000028</td>\n",
       "      <td>0.000013</td>\n",
       "      <td>5.778000e-05</td>\n",
       "      <td>0.000037</td>\n",
       "      <td>1.683000e-05</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>3.990000e-06</td>\n",
       "      <td>8.030000e-06</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000050</td>\n",
       "      <td>0.000138</td>\n",
       "      <td>0.000012</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>4.100000e-07</td>\n",
       "      <td>5.440000e-06</td>\n",
       "      <td>3.400000e-07</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>8.399000e-05</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>3.970000e-06</td>\n",
       "      <td>5.900000e-07</td>\n",
       "      <td>1.285300e-04</td>\n",
       "      <td>8.800000e-07</td>\n",
       "      <td>2.228000e-05</td>\n",
       "      <td>1.105000e-05</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>0.000014</td>\n",
       "      <td>4.090000e-06</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>4.920000e-06</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>6.400000e-07</td>\n",
       "      <td>2.372000e-05</td>\n",
       "      <td>6.610000e-06</td>\n",
       "      <td>0.00001</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>5.800000e-06</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>1.170000e-05</td>\n",
       "      <td>1.500000e-07</td>\n",
       "      <td>8.300000e-07</td>\n",
       "      <td>1.060000e-04</td>\n",
       "      <td>0.743</td>\n",
       "      <td>33</td>\n",
       "      <td>0.727</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>video6633.webm</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.999985</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>7.120000e-06</td>\n",
       "      <td>0.804</td>\n",
       "      <td>33</td>\n",
       "      <td>0.727</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 107 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0           video  ...  long-term_memorability  nb_long-term_annotations\n",
       "0           0  video6632.webm  ...                   0.900                        10\n",
       "1           1  video6634.webm  ...                   0.727                        11\n",
       "2           2  video6633.webm  ...                   0.727                        11\n",
       "\n",
       "[3 rows x 107 columns]"
      ]
     },
     "execution_count": 89,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfHMP_Dev = dfHMP_Dev.merge(dfGroundTruth,on=[\"video\"],how=\"inner\")\n",
    "dfHMP_Dev.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "UYw4KgPyfl0D"
   },
   "outputs": [],
   "source": [
    "dfHMP_DevFeatures = dfHMP_Dev.iloc[:,2:103].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 36
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1624,
     "status": "ok",
     "timestamp": 1588067878211,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "8sWHWyKE3ydL",
    "outputId": "2a6c5372-9dd5-48ac-8c77-1b77b24011a6"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(6000, 101)"
      ]
     },
     "execution_count": 91,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfHMP_DevFeatures.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "OPpgNkBZFf0Y"
   },
   "source": [
    "**HMP + C3D with ground truth**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "x4gjgipUGsZv"
   },
   "source": [
    "Note: We are merging dfC3D_Dev because it doesn't have ground truth columns yet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 210
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1417,
     "status": "ok",
     "timestamp": 1588067883161,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "TH3Fp-Gl7yox",
    "outputId": "3a872a4c-87b5-4dc3-93a8-2a979ac80424"
   },
   "outputs": [
    {
     "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>video</th>\n",
       "      <th>0_x</th>\n",
       "      <th>1_x</th>\n",
       "      <th>2_x</th>\n",
       "      <th>3_x</th>\n",
       "      <th>4_x</th>\n",
       "      <th>5_x</th>\n",
       "      <th>6_x</th>\n",
       "      <th>7_x</th>\n",
       "      <th>8_x</th>\n",
       "      <th>9_x</th>\n",
       "      <th>10_x</th>\n",
       "      <th>11_x</th>\n",
       "      <th>12_x</th>\n",
       "      <th>13_x</th>\n",
       "      <th>14_x</th>\n",
       "      <th>15_x</th>\n",
       "      <th>16_x</th>\n",
       "      <th>17_x</th>\n",
       "      <th>18_x</th>\n",
       "      <th>19_x</th>\n",
       "      <th>20_x</th>\n",
       "      <th>21_x</th>\n",
       "      <th>22_x</th>\n",
       "      <th>23_x</th>\n",
       "      <th>24_x</th>\n",
       "      <th>25_x</th>\n",
       "      <th>26_x</th>\n",
       "      <th>27_x</th>\n",
       "      <th>28_x</th>\n",
       "      <th>29_x</th>\n",
       "      <th>30_x</th>\n",
       "      <th>31_x</th>\n",
       "      <th>32_x</th>\n",
       "      <th>33_x</th>\n",
       "      <th>34_x</th>\n",
       "      <th>35_x</th>\n",
       "      <th>36_x</th>\n",
       "      <th>37_x</th>\n",
       "      <th>38_x</th>\n",
       "      <th>...</th>\n",
       "      <th>65_y</th>\n",
       "      <th>66_y</th>\n",
       "      <th>67_y</th>\n",
       "      <th>68_y</th>\n",
       "      <th>69_y</th>\n",
       "      <th>70_y</th>\n",
       "      <th>71_y</th>\n",
       "      <th>72_y</th>\n",
       "      <th>73_y</th>\n",
       "      <th>74_y</th>\n",
       "      <th>75_y</th>\n",
       "      <th>76_y</th>\n",
       "      <th>77_y</th>\n",
       "      <th>78_y</th>\n",
       "      <th>79_y</th>\n",
       "      <th>80_y</th>\n",
       "      <th>81_y</th>\n",
       "      <th>82_y</th>\n",
       "      <th>83_y</th>\n",
       "      <th>84_y</th>\n",
       "      <th>85_y</th>\n",
       "      <th>86_y</th>\n",
       "      <th>87_y</th>\n",
       "      <th>88_y</th>\n",
       "      <th>89_y</th>\n",
       "      <th>90_y</th>\n",
       "      <th>91_y</th>\n",
       "      <th>92_y</th>\n",
       "      <th>93_y</th>\n",
       "      <th>94_y</th>\n",
       "      <th>95_y</th>\n",
       "      <th>96_y</th>\n",
       "      <th>97_y</th>\n",
       "      <th>98_y</th>\n",
       "      <th>99_y</th>\n",
       "      <th>100_y</th>\n",
       "      <th>short-term_memorability</th>\n",
       "      <th>nb_short-term_annotations</th>\n",
       "      <th>long-term_memorability</th>\n",
       "      <th>nb_long-term_annotations</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>video6632.webm</td>\n",
       "      <td>0.010858</td>\n",
       "      <td>0.010386</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.700000e-07</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>3.400000e-07</td>\n",
       "      <td>8.000000e-08</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>0.000105</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4.000000e-08</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.331800e-04</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>7.600000e-07</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.800000e-07</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>6.000000e-08</td>\n",
       "      <td>0.000021</td>\n",
       "      <td>5.000000e-08</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>3.000000e-08</td>\n",
       "      <td>2.000000e-08</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000032</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>2.500000e-07</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>5.000000e-08</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>4.200000e-07</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>3.000000e-08</td>\n",
       "      <td>2.000000e-08</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>6.000000e-08</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.000000e-08</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>1.100000e-07</td>\n",
       "      <td>1.400000e-07</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.700000e-07</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>1.300000e-06</td>\n",
       "      <td>2.600000e-06</td>\n",
       "      <td>8.000000e-08</td>\n",
       "      <td>0.884</td>\n",
       "      <td>31</td>\n",
       "      <td>0.900</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>video6634.webm</td>\n",
       "      <td>0.000200</td>\n",
       "      <td>0.000065</td>\n",
       "      <td>0.993807</td>\n",
       "      <td>2.000000e-07</td>\n",
       "      <td>4.700000e-07</td>\n",
       "      <td>0.000073</td>\n",
       "      <td>3.700000e-06</td>\n",
       "      <td>0.000337</td>\n",
       "      <td>6.710000e-06</td>\n",
       "      <td>2.290000e-06</td>\n",
       "      <td>6.380000e-06</td>\n",
       "      <td>7.340000e-06</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>3.810000e-06</td>\n",
       "      <td>1.411000e-05</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>2.300000e-07</td>\n",
       "      <td>1.900000e-07</td>\n",
       "      <td>0.002938</td>\n",
       "      <td>5.200000e-07</td>\n",
       "      <td>9.200000e-07</td>\n",
       "      <td>1.136000e-05</td>\n",
       "      <td>0.000011</td>\n",
       "      <td>3.310400e-04</td>\n",
       "      <td>9.200000e-07</td>\n",
       "      <td>0.000022</td>\n",
       "      <td>8.000000e-08</td>\n",
       "      <td>0.000028</td>\n",
       "      <td>0.000013</td>\n",
       "      <td>5.778000e-05</td>\n",
       "      <td>0.000037</td>\n",
       "      <td>1.683000e-05</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>3.990000e-06</td>\n",
       "      <td>8.030000e-06</td>\n",
       "      <td>1.450000e-06</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000050</td>\n",
       "      <td>0.000138</td>\n",
       "      <td>0.000012</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>4.100000e-07</td>\n",
       "      <td>5.440000e-06</td>\n",
       "      <td>3.400000e-07</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>8.399000e-05</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>3.970000e-06</td>\n",
       "      <td>5.900000e-07</td>\n",
       "      <td>1.285300e-04</td>\n",
       "      <td>8.800000e-07</td>\n",
       "      <td>2.228000e-05</td>\n",
       "      <td>1.105000e-05</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>0.000014</td>\n",
       "      <td>4.090000e-06</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>4.920000e-06</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>6.400000e-07</td>\n",
       "      <td>2.372000e-05</td>\n",
       "      <td>6.610000e-06</td>\n",
       "      <td>0.00001</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>5.800000e-06</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>1.170000e-05</td>\n",
       "      <td>1.500000e-07</td>\n",
       "      <td>8.300000e-07</td>\n",
       "      <td>1.060000e-04</td>\n",
       "      <td>0.743</td>\n",
       "      <td>33</td>\n",
       "      <td>0.727</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>video6633.webm</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.999985</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>7.120000e-06</td>\n",
       "      <td>0.804</td>\n",
       "      <td>33</td>\n",
       "      <td>0.727</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 208 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            video       0_x  ...  long-term_memorability  nb_long-term_annotations\n",
       "0  video6632.webm  0.010858  ...                   0.900                        10\n",
       "1  video6634.webm  0.000200  ...                   0.727                        11\n",
       "2  video6633.webm  0.000000  ...                   0.727                        11\n",
       "\n",
       "[3 rows x 208 columns]"
      ]
     },
     "execution_count": 92,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfDevHMP_C3D = dfC3D_Dev.merge(dfHMP_Dev,on=[\"video\"],how=\"inner\")\n",
    "dfDevHMP_C3D.columns\n",
    "dfDevHMP_C3D.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "rfFECVIO7zlD"
   },
   "outputs": [],
   "source": [
    "dfDevHMP_C3D_DevFeatures = dfDevHMP_C3D.iloc[:,2:204].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 36
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2778,
     "status": "ok",
     "timestamp": 1588067901513,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "0harW5LH72LJ",
    "outputId": "00542d8e-cab1-45f0-f416-0166fd8485ae"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(6000, 202)"
      ]
     },
     "execution_count": 94,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfDevHMP_C3D_DevFeatures.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "geAJ_sZRTRfO"
   },
   "source": [
    "**C3D and joining groundTruth with it**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 210
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2849,
     "status": "ok",
     "timestamp": 1588067906788,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "yGkYuMF4TMHz",
    "outputId": "ca3f4d88-f604-48df-c7fc-bb3df6c39071"
   },
   "outputs": [
    {
     "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>video</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "      <th>17</th>\n",
       "      <th>18</th>\n",
       "      <th>19</th>\n",
       "      <th>20</th>\n",
       "      <th>21</th>\n",
       "      <th>22</th>\n",
       "      <th>23</th>\n",
       "      <th>24</th>\n",
       "      <th>25</th>\n",
       "      <th>26</th>\n",
       "      <th>27</th>\n",
       "      <th>28</th>\n",
       "      <th>29</th>\n",
       "      <th>30</th>\n",
       "      <th>31</th>\n",
       "      <th>32</th>\n",
       "      <th>33</th>\n",
       "      <th>34</th>\n",
       "      <th>35</th>\n",
       "      <th>36</th>\n",
       "      <th>37</th>\n",
       "      <th>38</th>\n",
       "      <th>...</th>\n",
       "      <th>65</th>\n",
       "      <th>66</th>\n",
       "      <th>67</th>\n",
       "      <th>68</th>\n",
       "      <th>69</th>\n",
       "      <th>70</th>\n",
       "      <th>71</th>\n",
       "      <th>72</th>\n",
       "      <th>73</th>\n",
       "      <th>74</th>\n",
       "      <th>75</th>\n",
       "      <th>76</th>\n",
       "      <th>77</th>\n",
       "      <th>78</th>\n",
       "      <th>79</th>\n",
       "      <th>80</th>\n",
       "      <th>81</th>\n",
       "      <th>82</th>\n",
       "      <th>83</th>\n",
       "      <th>84</th>\n",
       "      <th>85</th>\n",
       "      <th>86</th>\n",
       "      <th>87</th>\n",
       "      <th>88</th>\n",
       "      <th>89</th>\n",
       "      <th>90</th>\n",
       "      <th>91</th>\n",
       "      <th>92</th>\n",
       "      <th>93</th>\n",
       "      <th>94</th>\n",
       "      <th>95</th>\n",
       "      <th>96</th>\n",
       "      <th>97</th>\n",
       "      <th>98</th>\n",
       "      <th>99</th>\n",
       "      <th>100</th>\n",
       "      <th>short-term_memorability</th>\n",
       "      <th>nb_short-term_annotations</th>\n",
       "      <th>long-term_memorability</th>\n",
       "      <th>nb_long-term_annotations</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>video6632.webm</td>\n",
       "      <td>0.010858</td>\n",
       "      <td>0.010386</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.700000e-07</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>3.400000e-07</td>\n",
       "      <td>8.000000e-08</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>0.000105</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4.000000e-08</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.331800e-04</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>7.600000e-07</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.800000e-07</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>6.000000e-08</td>\n",
       "      <td>0.000021</td>\n",
       "      <td>5.000000e-08</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>3.000000e-08</td>\n",
       "      <td>2.000000e-08</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000032</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>2.500000e-07</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>5.000000e-08</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>4.200000e-07</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>3.000000e-08</td>\n",
       "      <td>2.000000e-08</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>6.000000e-08</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.000000e-08</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>1.100000e-07</td>\n",
       "      <td>1.400000e-07</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.700000e-07</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000e-08</td>\n",
       "      <td>1.300000e-06</td>\n",
       "      <td>2.600000e-06</td>\n",
       "      <td>8.000000e-08</td>\n",
       "      <td>0.884</td>\n",
       "      <td>31</td>\n",
       "      <td>0.900</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>video6634.webm</td>\n",
       "      <td>0.000200</td>\n",
       "      <td>0.000065</td>\n",
       "      <td>0.993807</td>\n",
       "      <td>2.000000e-07</td>\n",
       "      <td>4.700000e-07</td>\n",
       "      <td>0.000073</td>\n",
       "      <td>3.700000e-06</td>\n",
       "      <td>0.000337</td>\n",
       "      <td>6.710000e-06</td>\n",
       "      <td>2.290000e-06</td>\n",
       "      <td>6.380000e-06</td>\n",
       "      <td>7.340000e-06</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>3.810000e-06</td>\n",
       "      <td>1.411000e-05</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>2.300000e-07</td>\n",
       "      <td>1.900000e-07</td>\n",
       "      <td>0.002938</td>\n",
       "      <td>5.200000e-07</td>\n",
       "      <td>9.200000e-07</td>\n",
       "      <td>1.136000e-05</td>\n",
       "      <td>0.000011</td>\n",
       "      <td>3.310400e-04</td>\n",
       "      <td>9.200000e-07</td>\n",
       "      <td>0.000022</td>\n",
       "      <td>8.000000e-08</td>\n",
       "      <td>0.000028</td>\n",
       "      <td>0.000013</td>\n",
       "      <td>5.778000e-05</td>\n",
       "      <td>0.000037</td>\n",
       "      <td>1.683000e-05</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>3.990000e-06</td>\n",
       "      <td>8.030000e-06</td>\n",
       "      <td>1.450000e-06</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000050</td>\n",
       "      <td>0.000138</td>\n",
       "      <td>0.000012</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>4.100000e-07</td>\n",
       "      <td>5.440000e-06</td>\n",
       "      <td>3.400000e-07</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>8.399000e-05</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>3.970000e-06</td>\n",
       "      <td>5.900000e-07</td>\n",
       "      <td>1.285300e-04</td>\n",
       "      <td>8.800000e-07</td>\n",
       "      <td>2.228000e-05</td>\n",
       "      <td>1.105000e-05</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>0.000014</td>\n",
       "      <td>4.090000e-06</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>4.920000e-06</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>6.400000e-07</td>\n",
       "      <td>2.372000e-05</td>\n",
       "      <td>6.610000e-06</td>\n",
       "      <td>0.00001</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>5.800000e-06</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>1.170000e-05</td>\n",
       "      <td>1.500000e-07</td>\n",
       "      <td>8.300000e-07</td>\n",
       "      <td>1.060000e-04</td>\n",
       "      <td>0.743</td>\n",
       "      <td>33</td>\n",
       "      <td>0.727</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>video6633.webm</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.999985</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>7.120000e-06</td>\n",
       "      <td>0.804</td>\n",
       "      <td>33</td>\n",
       "      <td>0.727</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 106 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            video         0  ...  long-term_memorability  nb_long-term_annotations\n",
       "0  video6632.webm  0.010858  ...                   0.900                        10\n",
       "1  video6634.webm  0.000200  ...                   0.727                        11\n",
       "2  video6633.webm  0.000000  ...                   0.727                        11\n",
       "\n",
       "[3 rows x 106 columns]"
      ]
     },
     "execution_count": 95,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfC3D_Dev = dfC3D_Dev.merge(dfGroundTruth,on=[\"video\"],how=\"inner\")\n",
    "dfC3D_Dev.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "h5mAEaUcTL-N"
   },
   "outputs": [],
   "source": [
    "dfC3D_DevFeatures = dfC3D_Dev.iloc[:,1:102].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 36
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 4110,
     "status": "ok",
     "timestamp": 1588067917949,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "2QjBeEtPaCQF",
    "outputId": "b388fc90-30b3-4dc4-bf4b-8536404ba3fc"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(6000, 101)"
      ]
     },
     "execution_count": 97,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfC3D_DevFeatures.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "boJvOwZT_Qr8"
   },
   "source": [
    "**Working on C3D only**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "Nk1uZl4Fu2w9"
   },
   "outputs": [],
   "source": [
    "X = dfC3D_DevFeatures\n",
    "y = dfGroundTruth[['short-term_memorability','long-term_memorability']].values #these are our target columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "PzDtocCWAOxP"
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "uW6TyQnsu2z2"
   },
   "outputs": [],
   "source": [
    "X_train,X_test,y_train,y_test = train_test_split(X,y,test_size=0.2,random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 91
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 881,
     "status": "ok",
     "timestamp": 1588065980118,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "uGp3x726ANPo",
    "outputId": "0a53240b-8d6a-40ce-e6b4-f13446503759"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train  (4800, 101)\n",
      "X_test   (1200, 101)\n",
      "Y_train  (4800, 2)\n",
      "Y_test   (1200, 2)\n"
     ]
    }
   ],
   "source": [
    "print('X_train ', X_train.shape)\n",
    "print('X_test  ', X_test.shape)\n",
    "print('Y_train ', y_train.shape)\n",
    "print('Y_test  ', y_test.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Z8Y9zW9KawQu"
   },
   "source": [
    "*Implementing regression models*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "69ty7jQVANS4"
   },
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.neighbors import KNeighborsRegressor\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.tree import DecisionTreeRegressor\n",
    "from sklearn.ensemble import VotingRegressor\n",
    "from sklearn.ensemble import AdaBoostRegressor\n",
    "from sklearn.ensemble import GradientBoostingRegressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 190,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1478,
     "status": "ok",
     "timestamp": 1588013174471,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "rQip9wthbech",
    "outputId": "01bd620b-6e2b-4649-90b7-778ffd599917"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.004\n",
      "The Spearman's correlation coefficient is: -0.008\n"
     ]
    }
   ],
   "source": [
    "#linear regression\n",
    "linearRegressor = LinearRegression()\n",
    "linearRegressor.fit(X_train, y_train)\n",
    "y_pred = linearRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "gurnlAr3_AxL"
   },
   "source": [
    "value of n has been choosen by sqrt(6000) = 77.45 \n",
    "\n",
    "https://www.quora.com/How-can-I-choose-the-best-K-in-KNN-K-nearest-neighbour-classification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 197,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1410,
     "status": "ok",
     "timestamp": 1588013552513,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "zr9w56Anb9x7",
    "outputId": "5bbc7f32-bf29-403a-c64a-33aa85165345"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.119\n",
      "The Spearman's correlation coefficient is: 0.116\n"
     ]
    }
   ],
   "source": [
    "knnRegressor = KNeighborsRegressor(n_neighbors = 77)\n",
    "knnRegressor.fit(X_test, y_test)\n",
    "y_pred = knnRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 198,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2021,
     "status": "ok",
     "timestamp": 1588013594142,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "eb3i8xbIcHzx",
    "outputId": "8e1da146-0ce6-4c37-9f67-56f4ff849e6b"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.011\n",
      "The Spearman's correlation coefficient is: 0.025\n"
     ]
    }
   ],
   "source": [
    "dtRegressor = DecisionTreeRegressor()\n",
    "dtRegressor.fit(X_train, y_train)\n",
    "y_pred = dtRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 199,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 48816,
     "status": "ok",
     "timestamp": 1588013645599,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "TcTLwTZbANWU",
    "outputId": "cce590e2-fc99-4faa-f018-eeb089ffaeeb"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.001\n",
      "The Spearman's correlation coefficient is: 0.012\n"
     ]
    }
   ],
   "source": [
    "rfRegressor = RandomForestRegressor(n_estimators=100,random_state=45)\n",
    "rfRegressor.fit(X_train,y_train)\n",
    "y_pred = rfRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "ja-w0B_WHkND"
   },
   "source": [
    "*For other ensemble and neural networking models*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "_iMLTbxfDLvN"
   },
   "source": [
    "link followed for ensembling:\n",
    "https://www.analyticsvidhya.com/blog/2018/06/comprehensive-guide-for-ensemble-models/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "5DBsmK736InC"
   },
   "outputs": [],
   "source": [
    "y_short = dfGroundTruth[['short-term_memorability']].values\n",
    "y_long = dfGroundTruth[['long-term_memorability']].values\n",
    "\n",
    "# Splitting the dataset into the Training set and Test set\n",
    "short_X_train,short_X_test,short_y_train,short_y_test = train_test_split(X,y_short,test_size=0.2,random_state=40)\n",
    "long_X_train,long_X_test,long_y_train,long_y_test = train_test_split(X,y_long,test_size=0.2,random_state=40)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 209,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 149
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 194479,
     "status": "ok",
     "timestamp": 1588014790660,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "woOT8jRkBvlV",
    "outputId": "fda9e8d8-2514-4627-d70f-f5b1574c026f"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_voting.py:406: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.006\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_voting.py:406: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.054\n"
     ]
    }
   ],
   "source": [
    "model1 = LinearRegression()\n",
    "model2 = DecisionTreeRegressor(random_state=1)\n",
    "model3 = KNeighborsRegressor(n_neighbors = 77)\n",
    "model4 = RandomForestRegressor(n_estimators=100,random_state=45)\n",
    "vRegressor = VotingRegressor(estimators=[('lr', model1), ('dt', model2), ('knn', model3), ('rf', model4)])\n",
    "\n",
    "vRegressor.fit(short_X_train,short_y_train)\n",
    "y_pred = vRegressor.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "vRegressor.fit(long_X_train,long_y_train)\n",
    "y_pred = vRegressor.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 206,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 149
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 6835,
     "status": "ok",
     "timestamp": 1588014278021,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "8Y-aNEqFA4hx",
    "outputId": "95de5a0f-2779-45ef-f2c4-a97d07e4c397"
   },
   "outputs": [
    {
     "name": "stderr",
     "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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: -0.047\n"
     ]
    },
    {
     "name": "stderr",
     "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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.019\n"
     ]
    }
   ],
   "source": [
    "abRegressor = AdaBoostRegressor()\n",
    "abRegressor.fit(short_X_train, short_y_train)\n",
    "y_pred = abRegressor.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "abRegressor = AdaBoostRegressor()\n",
    "abRegressor.fit(long_X_train, long_y_train)\n",
    "y_pred = abRegressor.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 211,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 149
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 26360,
     "status": "ok",
     "timestamp": 1588014932728,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "-bbapPifANbB",
    "outputId": "9c784f6d-d8b7-4a15-d763-adae14b0a07e"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_gb.py:1454: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.013\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_gb.py:1454: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.027\n"
     ]
    }
   ],
   "source": [
    "gbRegressor= GradientBoostingRegressor()\n",
    "\n",
    "gbRegressor.fit(short_X_train,short_y_train)\n",
    "y_pred = gbRegressor.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "gbRegressor.fit(long_X_train,long_y_train)\n",
    "y_pred = gbRegressor.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "vDZsiDzNKamX"
   },
   "source": [
    "links followed for neural network:\n",
    "\n",
    "https://stackoverflow.com/questions/49008074/how-to-create-a-neural-network-for-regression\n",
    "\n",
    "https://keras.io/activations/\n",
    "\n",
    "https://keras.io/optimizers/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 36
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3409,
     "status": "ok",
     "timestamp": 1588091748220,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "FiLMcDJ8D1R9",
    "outputId": "75621190-3b5d-42c8-9a8f-b7d2d737713d"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Dense\n",
    "from keras.wrappers.scikit_learn import KerasRegressor\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.model_selection import KFold\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.pipeline import Pipeline\n",
    "from tensorflow.keras import layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "_02N9KVIKs7a"
   },
   "outputs": [],
   "source": [
    "sc = StandardScaler()\n",
    "short_X_train = sc.fit_transform(short_X_train)\n",
    "short_X_test = sc.transform(short_X_test)\n",
    "\n",
    "long_X_train = sc.fit_transform(long_X_train)\n",
    "long_X_test = sc.transform(long_X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "x9esU4r1D1XD"
   },
   "outputs": [],
   "source": [
    "# Initialising\n",
    "neuralNetModel = Sequential()\n",
    "\n",
    "# Adding the input layer and first hidden layer\n",
    "neuralNetModel.add(Dense(32, activation = 'relu', input_dim = 101))\n",
    "\n",
    "# Adding second hidden layer\n",
    "neuralNetModel.add(Dense(units = 32, activation = 'relu'))\n",
    "\n",
    "# Adding third hidden layer\n",
    "neuralNetModel.add(Dense(units = 32, activation = 'relu'))\n",
    "\n",
    "# Adding output layer\n",
    "neuralNetModel.add(Dense(units = 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "nO4cCXz9MKC_"
   },
   "outputs": [],
   "source": [
    "neuralNetModel.compile(optimizer = 'adam',loss = 'mean_squared_error')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 256,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 122647,
     "status": "ok",
     "timestamp": 1588019108428,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "iU6ZkBiUMQG5",
    "outputId": "1320c2b4-b1f7-484a-f91f-29419adbaa24"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0076\n",
      "Epoch 2/51\n",
      "4800/4800 [==============================] - 1s 247us/step - loss: 0.0062\n",
      "Epoch 3/51\n",
      "4800/4800 [==============================] - 1s 242us/step - loss: 0.0059\n",
      "Epoch 4/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0058\n",
      "Epoch 5/51\n",
      "4800/4800 [==============================] - 1s 245us/step - loss: 0.0057\n",
      "Epoch 6/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0056\n",
      "Epoch 7/51\n",
      "4800/4800 [==============================] - 1s 247us/step - loss: 0.0055\n",
      "Epoch 8/51\n",
      "4800/4800 [==============================] - 1s 294us/step - loss: 0.0054\n",
      "Epoch 9/51\n",
      "4800/4800 [==============================] - 1s 248us/step - loss: 0.0054\n",
      "Epoch 10/51\n",
      "4800/4800 [==============================] - 1s 253us/step - loss: 0.0053\n",
      "Epoch 11/51\n",
      "4800/4800 [==============================] - 1s 254us/step - loss: 0.0053\n",
      "Epoch 12/51\n",
      "4800/4800 [==============================] - 1s 251us/step - loss: 0.0053\n",
      "Epoch 13/51\n",
      "4800/4800 [==============================] - 1s 239us/step - loss: 0.0052\n",
      "Epoch 14/51\n",
      "4800/4800 [==============================] - 1s 245us/step - loss: 0.0052\n",
      "Epoch 15/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0051\n",
      "Epoch 16/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0051\n",
      "Epoch 17/51\n",
      "4800/4800 [==============================] - 1s 246us/step - loss: 0.0050\n",
      "Epoch 18/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0050\n",
      "Epoch 19/51\n",
      "4800/4800 [==============================] - 1s 242us/step - loss: 0.0050\n",
      "Epoch 20/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0049\n",
      "Epoch 21/51\n",
      "4800/4800 [==============================] - 1s 248us/step - loss: 0.0049\n",
      "Epoch 22/51\n",
      "4800/4800 [==============================] - 1s 248us/step - loss: 0.0048\n",
      "Epoch 23/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0048\n",
      "Epoch 24/51\n",
      "4800/4800 [==============================] - 1s 240us/step - loss: 0.0048\n",
      "Epoch 25/51\n",
      "4800/4800 [==============================] - 1s 296us/step - loss: 0.0047\n",
      "Epoch 26/51\n",
      "4800/4800 [==============================] - 1s 248us/step - loss: 0.0048\n",
      "Epoch 27/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0046\n",
      "Epoch 28/51\n",
      "4800/4800 [==============================] - 1s 245us/step - loss: 0.0046\n",
      "Epoch 29/51\n",
      "4800/4800 [==============================] - 1s 238us/step - loss: 0.0046\n",
      "Epoch 30/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0046\n",
      "Epoch 31/51\n",
      "4800/4800 [==============================] - 1s 246us/step - loss: 0.0045\n",
      "Epoch 32/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0045\n",
      "Epoch 33/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0046\n",
      "Epoch 34/51\n",
      "4800/4800 [==============================] - 1s 242us/step - loss: 0.0045\n",
      "Epoch 35/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0044\n",
      "Epoch 36/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0044\n",
      "Epoch 37/51\n",
      "4800/4800 [==============================] - 1s 246us/step - loss: 0.0044\n",
      "Epoch 38/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0045\n",
      "Epoch 39/51\n",
      "4800/4800 [==============================] - 1s 253us/step - loss: 0.0045\n",
      "Epoch 40/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0044\n",
      "Epoch 41/51\n",
      "4800/4800 [==============================] - 1s 243us/step - loss: 0.0044\n",
      "Epoch 42/51\n",
      "4800/4800 [==============================] - 1s 266us/step - loss: 0.0043\n",
      "Epoch 43/51\n",
      "4800/4800 [==============================] - 1s 284us/step - loss: 0.0044\n",
      "Epoch 44/51\n",
      "4800/4800 [==============================] - 1s 251us/step - loss: 0.0043\n",
      "Epoch 45/51\n",
      "4800/4800 [==============================] - 1s 245us/step - loss: 0.0043\n",
      "Epoch 46/51\n",
      "4800/4800 [==============================] - 1s 250us/step - loss: 0.0042\n",
      "Epoch 47/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0042\n",
      "Epoch 48/51\n",
      "4800/4800 [==============================] - 1s 248us/step - loss: 0.0042\n",
      "Epoch 49/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0042\n",
      "Epoch 50/51\n",
      "4800/4800 [==============================] - 1s 245us/step - loss: 0.0042\n",
      "Epoch 51/51\n",
      "4800/4800 [==============================] - 1s 237us/step - loss: 0.0042\n",
      "Epoch 1/51\n",
      "4800/4800 [==============================] - 1s 237us/step - loss: 0.0207\n",
      "Epoch 2/51\n",
      "4800/4800 [==============================] - 1s 243us/step - loss: 0.0188\n",
      "Epoch 3/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0178\n",
      "Epoch 4/51\n",
      "4800/4800 [==============================] - 1s 240us/step - loss: 0.0169\n",
      "Epoch 5/51\n",
      "4800/4800 [==============================] - 1s 260us/step - loss: 0.0166\n",
      "Epoch 6/51\n",
      "4800/4800 [==============================] - 1s 245us/step - loss: 0.0158\n",
      "Epoch 7/51\n",
      "4800/4800 [==============================] - 1s 245us/step - loss: 0.0156\n",
      "Epoch 8/51\n",
      "4800/4800 [==============================] - 1s 242us/step - loss: 0.0151\n",
      "Epoch 9/51\n",
      "4800/4800 [==============================] - 1s 302us/step - loss: 0.0152\n",
      "Epoch 10/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0149\n",
      "Epoch 11/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0146\n",
      "Epoch 12/51\n",
      "4800/4800 [==============================] - 1s 246us/step - loss: 0.0144\n",
      "Epoch 13/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0142\n",
      "Epoch 14/51\n",
      "4800/4800 [==============================] - 1s 248us/step - loss: 0.0141\n",
      "Epoch 15/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0140\n",
      "Epoch 16/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0138\n",
      "Epoch 17/51\n",
      "4800/4800 [==============================] - 1s 245us/step - loss: 0.0138\n",
      "Epoch 18/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0138\n",
      "Epoch 19/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0137\n",
      "Epoch 20/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0135\n",
      "Epoch 21/51\n",
      "4800/4800 [==============================] - 1s 242us/step - loss: 0.0133\n",
      "Epoch 22/51\n",
      "4800/4800 [==============================] - 1s 240us/step - loss: 0.0132\n",
      "Epoch 23/51\n",
      "4800/4800 [==============================] - 1s 245us/step - loss: 0.0131\n",
      "Epoch 24/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0131\n",
      "Epoch 25/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0129\n",
      "Epoch 26/51\n",
      "4800/4800 [==============================] - 1s 295us/step - loss: 0.0129\n",
      "Epoch 27/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0127\n",
      "Epoch 28/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0129\n",
      "Epoch 29/51\n",
      "4800/4800 [==============================] - 1s 238us/step - loss: 0.0129\n",
      "Epoch 30/51\n",
      "4800/4800 [==============================] - 1s 252us/step - loss: 0.0127\n",
      "Epoch 31/51\n",
      "4800/4800 [==============================] - 1s 247us/step - loss: 0.0126\n",
      "Epoch 32/51\n",
      "4800/4800 [==============================] - 1s 249us/step - loss: 0.0126\n",
      "Epoch 33/51\n",
      "4800/4800 [==============================] - 1s 246us/step - loss: 0.0124\n",
      "Epoch 34/51\n",
      "4800/4800 [==============================] - 1s 247us/step - loss: 0.0123\n",
      "Epoch 35/51\n",
      "4800/4800 [==============================] - 1s 250us/step - loss: 0.0123\n",
      "Epoch 36/51\n",
      "4800/4800 [==============================] - 1s 249us/step - loss: 0.0123\n",
      "Epoch 37/51\n",
      "4800/4800 [==============================] - 1s 247us/step - loss: 0.0123\n",
      "Epoch 38/51\n",
      "4800/4800 [==============================] - 1s 247us/step - loss: 0.0123\n",
      "Epoch 39/51\n",
      "4800/4800 [==============================] - 1s 251us/step - loss: 0.0122\n",
      "Epoch 40/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0122\n",
      "Epoch 41/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0119\n",
      "Epoch 42/51\n",
      "4800/4800 [==============================] - 1s 250us/step - loss: 0.0120\n",
      "Epoch 43/51\n",
      "4800/4800 [==============================] - 1s 277us/step - loss: 0.0123\n",
      "Epoch 44/51\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 0.0119\n",
      "Epoch 45/51\n",
      "4800/4800 [==============================] - 1s 238us/step - loss: 0.0119\n",
      "Epoch 46/51\n",
      "4800/4800 [==============================] - 1s 242us/step - loss: 0.0118\n",
      "Epoch 47/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0117\n",
      "Epoch 48/51\n",
      "4800/4800 [==============================] - 1s 245us/step - loss: 0.0118\n",
      "Epoch 49/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0117\n",
      "Epoch 50/51\n",
      "4800/4800 [==============================] - 1s 250us/step - loss: 0.0117\n",
      "Epoch 51/51\n",
      "4800/4800 [==============================] - 1s 248us/step - loss: 0.0116\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.callbacks.History at 0x7ff076582860>"
      ]
     },
     "execution_count": 256,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "neuralNetModel.fit(short_X_train, short_y_train, batch_size = 10, epochs = 51)\n",
    "neuralNetModel.fit(long_X_train, long_y_train, batch_size = 10, epochs = 51)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 257,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1242,
     "status": "ok",
     "timestamp": 1588019124528,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "SMMvpKdLNLF0",
    "outputId": "acdc79c4-4e42-43d7-b19e-824e5d198eed"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.040\n",
      "The Spearman's correlation coefficient is: 0.017\n"
     ]
    }
   ],
   "source": [
    "y_pred = model.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "y_pred = model.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "Ia259mnqNmNG"
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 259,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 281
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1598,
     "status": "ok",
     "timestamp": 1588019137919,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "c1nVR2GJNiwb",
    "outputId": "4a4411d6-18ce-4af6-c45d-a695d1f56c86"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(y_test, color = 'green', label = 'Real data')\n",
    "plt.plot(y_pred, color = 'blue', label = 'Predicted data')\n",
    "plt.title('Prediction')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "HMgYce30fWgM"
   },
   "source": [
    "**Working on HMP only**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "bqY2GU2xANZq"
   },
   "outputs": [],
   "source": [
    "X = dfHMP_DevFeatures\n",
    "y = dfGroundTruth[['short-term_memorability','long-term_memorability']].values #these are our target columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "ZGSTOrL1ANNb"
   },
   "outputs": [],
   "source": [
    "#from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "2FWGmSa-u22v"
   },
   "outputs": [],
   "source": [
    "X_train,X_test,y_train,y_test = train_test_split(X,y,test_size=0.2,random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 91
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1542,
     "status": "ok",
     "timestamp": 1588066615106,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "k2b_YUEkb4Wq",
    "outputId": "c2b58b8e-969f-4539-80d7-56d87c6d9b9b"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train  (4800, 101)\n",
      "X_test   (1200, 101)\n",
      "Y_train  (4800, 2)\n",
      "Y_test   (1200, 2)\n"
     ]
    }
   ],
   "source": [
    "print('X_train ', X_train.shape)\n",
    "print('X_test  ', X_test.shape)\n",
    "print('Y_train ', y_train.shape)\n",
    "print('Y_test  ', y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1705,
     "status": "ok",
     "timestamp": 1588066619056,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "ol05n5yrb4f3",
    "outputId": "7d9fdc58-2540-4ca6-a7f7-1dd0ed856fe2"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.004\n",
      "The Spearman's correlation coefficient is: -0.008\n"
     ]
    }
   ],
   "source": [
    "#linear regression\n",
    "linearRegressor = LinearRegression()\n",
    "linearRegressor.fit(X_train, y_train)\n",
    "y_pred = linearRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1285,
     "status": "ok",
     "timestamp": 1588066631566,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "KLqkuNgPb4qD",
    "outputId": "8e1f8e09-6539-4f05-ee43-591e57fab020"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.119\n",
      "The Spearman's correlation coefficient is: 0.116\n"
     ]
    }
   ],
   "source": [
    "knnRegressor = KNeighborsRegressor(n_neighbors = 77)\n",
    "knnRegressor.fit(X_test, y_test)\n",
    "y_pred = knnRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1812,
     "status": "ok",
     "timestamp": 1588066635468,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "XaMER8SQb4m4",
    "outputId": "028692e0-8c24-4fa7-d269-48075907a79f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.018\n",
      "The Spearman's correlation coefficient is: 0.012\n"
     ]
    }
   ],
   "source": [
    "dtRegressor = DecisionTreeRegressor()\n",
    "dtRegressor.fit(X_train, y_train)\n",
    "y_pred = dtRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 50870,
     "status": "ok",
     "timestamp": 1588066689188,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "KUJPZ_pob4k6",
    "outputId": "1adc3423-7ca9-45d8-eec2-7ddca0975c13"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.001\n",
      "The Spearman's correlation coefficient is: 0.012\n"
     ]
    }
   ],
   "source": [
    "rfRegressor = RandomForestRegressor(n_estimators=100,random_state=45)\n",
    "rfRegressor.fit(X_train,y_train)\n",
    "y_pred = rfRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "w3j9EKuUHU1o"
   },
   "source": [
    "*For other ensemble and neural networking models*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "ZRfhqDvsb4d0"
   },
   "outputs": [],
   "source": [
    "y_short = dfGroundTruth[['short-term_memorability']].values\n",
    "y_long = dfGroundTruth[['long-term_memorability']].values\n",
    "\n",
    "# Splitting the dataset into the Training set and Test set\n",
    "short_X_train,short_X_test,short_y_train,short_y_test = train_test_split(X,y_short,test_size=0.2,random_state=40)\n",
    "long_X_train,long_X_test,long_y_train,long_y_test = train_test_split(X,y_long,test_size=0.2,random_state=40)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 149
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 106013,
     "status": "ok",
     "timestamp": 1588067170884,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "6aiiPgF-HrYM",
    "outputId": "baaf4705-b191-43f4-e882-0aa7db88e62b"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_voting.py:406: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.007\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_voting.py:406: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.051\n"
     ]
    }
   ],
   "source": [
    "model1 = LinearRegression()\n",
    "model2 = DecisionTreeRegressor(random_state=1)\n",
    "model3 = KNeighborsRegressor(n_neighbors = 77)\n",
    "model4 = RandomForestRegressor(n_estimators=100,random_state=45)\n",
    "vRegressor = VotingRegressor(estimators=[('lr', model1), ('dt', model2), ('knn', model3), ('rf', model4)])\n",
    "\n",
    "vRegressor.fit(short_X_train,short_y_train)\n",
    "y_pred = vRegressor.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "vRegressor.fit(long_X_train,long_y_train)\n",
    "y_pred = vRegressor.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 149
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 6452,
     "status": "ok",
     "timestamp": 1588067177371,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "1x9wkUnlHrcP",
    "outputId": "0e9c0146-17a6-489f-8264-e6c86a8d3b2b"
   },
   "outputs": [
    {
     "name": "stderr",
     "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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: -0.054\n"
     ]
    },
    {
     "name": "stderr",
     "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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: -0.015\n"
     ]
    }
   ],
   "source": [
    "abRegressor = AdaBoostRegressor()\n",
    "abRegressor.fit(short_X_train, short_y_train)\n",
    "y_pred = abRegressor.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "abRegressor = AdaBoostRegressor()\n",
    "abRegressor.fit(long_X_train, long_y_train)\n",
    "y_pred = abRegressor.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 149
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 33705,
     "status": "ok",
     "timestamp": 1588067204647,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "MSoDE4H_HrgV",
    "outputId": "cef58df1-8621-4c3a-9fc7-aa90f8bdad95"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_gb.py:1454: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.008\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_gb.py:1454: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.028\n"
     ]
    }
   ],
   "source": [
    "gbRegressor= GradientBoostingRegressor()\n",
    "\n",
    "gbRegressor.fit(short_X_train,short_y_train)\n",
    "y_pred = gbRegressor.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "gbRegressor.fit(long_X_train,long_y_train)\n",
    "y_pred = gbRegressor.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "36cReFVNH1A0"
   },
   "outputs": [],
   "source": [
    "sc = StandardScaler()\n",
    "short_X_train = sc.fit_transform(short_X_train)\n",
    "short_X_test = sc.transform(short_X_test)\n",
    "\n",
    "long_X_train = sc.fit_transform(long_X_train)\n",
    "long_X_test = sc.transform(long_X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "Wq5d9VibH1GA"
   },
   "outputs": [],
   "source": [
    "# Initialising\n",
    "neuralNetModel = Sequential()\n",
    "\n",
    "# Adding the input layer and first hidden layer\n",
    "neuralNetModel.add(Dense(32, activation = 'relu', input_dim = 101))\n",
    "\n",
    "# Adding second hidden layer\n",
    "neuralNetModel.add(Dense(units = 32, activation = 'relu'))\n",
    "\n",
    "# Adding third hidden layer\n",
    "neuralNetModel.add(Dense(units = 32, activation = 'relu'))\n",
    "\n",
    "# Adding output layer\n",
    "neuralNetModel.add(Dense(units = 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "Y5x-XDOHIaw8"
   },
   "outputs": [],
   "source": [
    "neuralNetModel.compile(optimizer = 'adam',loss = 'mean_squared_error')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 171878,
     "status": "ok",
     "timestamp": 1588067342871,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "5JoQJsy2Ia1p",
    "outputId": "5c125032-15ea-487a-a0e7-ded2e4787995"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/51\n",
      "4800/4800 [==============================] - 3s 628us/step - loss: 0.0968\n",
      "Epoch 2/51\n",
      "4800/4800 [==============================] - 1s 245us/step - loss: 0.0195\n",
      "Epoch 3/51\n",
      "4800/4800 [==============================] - 1s 271us/step - loss: 0.0158\n",
      "Epoch 4/51\n",
      "4800/4800 [==============================] - 1s 301us/step - loss: 0.0121\n",
      "Epoch 5/51\n",
      "4800/4800 [==============================] - 1s 289us/step - loss: 0.0095\n",
      "Epoch 6/51\n",
      "4800/4800 [==============================] - 1s 258us/step - loss: 0.0087\n",
      "Epoch 7/51\n",
      "4800/4800 [==============================] - 1s 299us/step - loss: 0.0103\n",
      "Epoch 8/51\n",
      "4800/4800 [==============================] - 1s 242us/step - loss: 0.0092\n",
      "Epoch 9/51\n",
      "4800/4800 [==============================] - 1s 246us/step - loss: 0.0082\n",
      "Epoch 10/51\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 0.0078\n",
      "Epoch 11/51\n",
      "4800/4800 [==============================] - 1s 264us/step - loss: 0.0069\n",
      "Epoch 12/51\n",
      "4800/4800 [==============================] - 1s 240us/step - loss: 0.0066\n",
      "Epoch 13/51\n",
      "4800/4800 [==============================] - 1s 248us/step - loss: 0.0066\n",
      "Epoch 14/51\n",
      "4800/4800 [==============================] - 1s 276us/step - loss: 0.0069\n",
      "Epoch 15/51\n",
      "4800/4800 [==============================] - 1s 264us/step - loss: 0.0068\n",
      "Epoch 16/51\n",
      "4800/4800 [==============================] - 1s 243us/step - loss: 0.0067\n",
      "Epoch 17/51\n",
      "4800/4800 [==============================] - 1s 246us/step - loss: 0.0064\n",
      "Epoch 18/51\n",
      "4800/4800 [==============================] - 1s 254us/step - loss: 0.0062\n",
      "Epoch 19/51\n",
      "4800/4800 [==============================] - 1s 242us/step - loss: 0.0063\n",
      "Epoch 20/51\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 0.0064\n",
      "Epoch 21/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0060\n",
      "Epoch 22/51\n",
      "4800/4800 [==============================] - 1s 239us/step - loss: 0.0061\n",
      "Epoch 23/51\n",
      "4800/4800 [==============================] - 1s 240us/step - loss: 0.0060\n",
      "Epoch 24/51\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 0.0060\n",
      "Epoch 25/51\n",
      "4800/4800 [==============================] - 1s 242us/step - loss: 0.0059\n",
      "Epoch 26/51\n",
      "4800/4800 [==============================] - 1s 245us/step - loss: 0.0058\n",
      "Epoch 27/51\n",
      "4800/4800 [==============================] - 1s 258us/step - loss: 0.0057\n",
      "Epoch 28/51\n",
      "4800/4800 [==============================] - 1s 238us/step - loss: 0.0058\n",
      "Epoch 29/51\n",
      "4800/4800 [==============================] - 1s 240us/step - loss: 0.0058\n",
      "Epoch 30/51\n",
      "4800/4800 [==============================] - 1s 266us/step - loss: 0.0057\n",
      "Epoch 31/51\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 0.0055\n",
      "Epoch 32/51\n",
      "4800/4800 [==============================] - 1s 245us/step - loss: 0.0055\n",
      "Epoch 33/51\n",
      "4800/4800 [==============================] - 1s 243us/step - loss: 0.0053\n",
      "Epoch 34/51\n",
      "4800/4800 [==============================] - 1s 249us/step - loss: 0.0053\n",
      "Epoch 35/51\n",
      "4800/4800 [==============================] - 1s 248us/step - loss: 0.0053\n",
      "Epoch 36/51\n",
      "4800/4800 [==============================] - 1s 250us/step - loss: 0.0053\n",
      "Epoch 37/51\n",
      "4800/4800 [==============================] - 1s 242us/step - loss: 0.0051\n",
      "Epoch 38/51\n",
      "4800/4800 [==============================] - 1s 248us/step - loss: 0.0052\n",
      "Epoch 39/51\n",
      "4800/4800 [==============================] - 1s 243us/step - loss: 0.0050\n",
      "Epoch 40/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0048\n",
      "Epoch 41/51\n",
      "4800/4800 [==============================] - 1s 249us/step - loss: 0.0050\n",
      "Epoch 42/51\n",
      "4800/4800 [==============================] - 1s 255us/step - loss: 0.0049\n",
      "Epoch 43/51\n",
      "4800/4800 [==============================] - 1s 240us/step - loss: 0.0047\n",
      "Epoch 44/51\n",
      "4800/4800 [==============================] - 1s 253us/step - loss: 0.0048\n",
      "Epoch 45/51\n",
      "4800/4800 [==============================] - 1s 243us/step - loss: 0.0047\n",
      "Epoch 46/51\n",
      "4800/4800 [==============================] - 1s 239us/step - loss: 0.0047\n",
      "Epoch 47/51\n",
      "4800/4800 [==============================] - 1s 266us/step - loss: 0.0046\n",
      "Epoch 48/51\n",
      "4800/4800 [==============================] - 1s 242us/step - loss: 0.0047\n",
      "Epoch 49/51\n",
      "4800/4800 [==============================] - 1s 245us/step - loss: 0.0046\n",
      "Epoch 50/51\n",
      "4800/4800 [==============================] - 1s 245us/step - loss: 0.0044\n",
      "Epoch 51/51\n",
      "4800/4800 [==============================] - 1s 248us/step - loss: 0.0045\n",
      "Epoch 1/51\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 0.0215\n",
      "Epoch 2/51\n",
      "4800/4800 [==============================] - 1s 250us/step - loss: 0.0202\n",
      "Epoch 3/51\n",
      "4800/4800 [==============================] - 1s 240us/step - loss: 0.0195\n",
      "Epoch 4/51\n",
      "4800/4800 [==============================] - 1s 250us/step - loss: 0.0191\n",
      "Epoch 5/51\n",
      "4800/4800 [==============================] - 1s 272us/step - loss: 0.0183\n",
      "Epoch 6/51\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 0.0180\n",
      "Epoch 7/51\n",
      "4800/4800 [==============================] - 1s 253us/step - loss: 0.0173\n",
      "Epoch 8/51\n",
      "4800/4800 [==============================] - 1s 260us/step - loss: 0.0173\n",
      "Epoch 9/51\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 0.0166\n",
      "Epoch 10/51\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 0.0163\n",
      "Epoch 11/51\n",
      "4800/4800 [==============================] - 1s 268us/step - loss: 0.0162\n",
      "Epoch 12/51\n",
      "4800/4800 [==============================] - 1s 267us/step - loss: 0.0157\n",
      "Epoch 13/51\n",
      "4800/4800 [==============================] - 1s 287us/step - loss: 0.0155\n",
      "Epoch 14/51\n",
      "4800/4800 [==============================] - 1s 260us/step - loss: 0.0154\n",
      "Epoch 15/51\n",
      "4800/4800 [==============================] - 1s 258us/step - loss: 0.0153\n",
      "Epoch 16/51\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 0.0149\n",
      "Epoch 17/51\n",
      "4800/4800 [==============================] - 1s 255us/step - loss: 0.0147\n",
      "Epoch 18/51\n",
      "4800/4800 [==============================] - 1s 250us/step - loss: 0.0144\n",
      "Epoch 19/51\n",
      "4800/4800 [==============================] - 1s 252us/step - loss: 0.0143\n",
      "Epoch 20/51\n",
      "4800/4800 [==============================] - 1s 240us/step - loss: 0.0143\n",
      "Epoch 21/51\n",
      "4800/4800 [==============================] - 1s 250us/step - loss: 0.0139\n",
      "Epoch 22/51\n",
      "4800/4800 [==============================] - 1s 282us/step - loss: 0.0139\n",
      "Epoch 23/51\n",
      "4800/4800 [==============================] - 1s 258us/step - loss: 0.0138\n",
      "Epoch 24/51\n",
      "4800/4800 [==============================] - 1s 240us/step - loss: 0.0137\n",
      "Epoch 25/51\n",
      "4800/4800 [==============================] - 1s 298us/step - loss: 0.0135\n",
      "Epoch 26/51\n",
      "4800/4800 [==============================] - 1s 255us/step - loss: 0.0132\n",
      "Epoch 27/51\n",
      "4800/4800 [==============================] - 1s 255us/step - loss: 0.0132\n",
      "Epoch 28/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0130\n",
      "Epoch 29/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0131\n",
      "Epoch 30/51\n",
      "4800/4800 [==============================] - 1s 238us/step - loss: 0.0131\n",
      "Epoch 31/51\n",
      "4800/4800 [==============================] - 1s 243us/step - loss: 0.0128\n",
      "Epoch 32/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0128\n",
      "Epoch 33/51\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 0.0127\n",
      "Epoch 34/51\n",
      "4800/4800 [==============================] - 1s 276us/step - loss: 0.0129\n",
      "Epoch 35/51\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 0.0128\n",
      "Epoch 36/51\n",
      "4800/4800 [==============================] - 1s 242us/step - loss: 0.0125\n",
      "Epoch 37/51\n",
      "4800/4800 [==============================] - 1s 239us/step - loss: 0.0125\n",
      "Epoch 38/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0123\n",
      "Epoch 39/51\n",
      "4800/4800 [==============================] - 1s 239us/step - loss: 0.0121\n",
      "Epoch 40/51\n",
      "4800/4800 [==============================] - 1s 238us/step - loss: 0.0123\n",
      "Epoch 41/51\n",
      "4800/4800 [==============================] - 1s 254us/step - loss: 0.0121\n",
      "Epoch 42/51\n",
      "4800/4800 [==============================] - 1s 238us/step - loss: 0.0119\n",
      "Epoch 43/51\n",
      "4800/4800 [==============================] - 1s 239us/step - loss: 0.0120\n",
      "Epoch 44/51\n",
      "4800/4800 [==============================] - 1s 284us/step - loss: 0.0120\n",
      "Epoch 45/51\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 0.0118\n",
      "Epoch 46/51\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 0.0119\n",
      "Epoch 47/51\n",
      "4800/4800 [==============================] - 2s 319us/step - loss: 0.0116\n",
      "Epoch 48/51\n",
      "4800/4800 [==============================] - 1s 265us/step - loss: 0.0115\n",
      "Epoch 49/51\n",
      "4800/4800 [==============================] - 1s 287us/step - loss: 0.0117\n",
      "Epoch 50/51\n",
      "4800/4800 [==============================] - 1s 245us/step - loss: 0.0117\n",
      "Epoch 51/51\n",
      "4800/4800 [==============================] - 1s 246us/step - loss: 0.0117\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.callbacks.History at 0x7fab244dfb00>"
      ]
     },
     "execution_count": 81,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "neuralNetModel.fit(short_X_train, short_y_train, batch_size = 10, epochs = 51)\n",
    "neuralNetModel.fit(long_X_train, long_y_train, batch_size = 10, epochs = 51)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1319,
     "status": "ok",
     "timestamp": 1588067436130,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "roJjv5LOIiu5",
    "outputId": "394a0996-1244-454b-e7f1-1ecfd5c1fb68"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.004\n",
      "The Spearman's correlation coefficient is: -0.016\n"
     ]
    }
   ],
   "source": [
    "y_pred = neuralNetModel.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "y_pred = neuralNetModel.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 281
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1931,
     "status": "ok",
     "timestamp": 1588067443800,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "UNnRVgt1Ii0Z",
    "outputId": "4ce2522d-fdb7-45e8-f9f7-07930e691000"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(y_test, color = 'green', label = 'Actual dataset')\n",
    "plt.plot(y_pred, color = 'peachpuff', label = 'Predicted dataset')\n",
    "plt.title('Prediction')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Qnv18OUl59Eu"
   },
   "source": [
    "**Working on HMP + C3D**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "n1EKyRz6b4bY"
   },
   "outputs": [],
   "source": [
    "X = dfDevHMP_C3D_DevFeatures\n",
    "y = dfGroundTruth[['short-term_memorability','long-term_memorability']].values #these are our target columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "musIA7Rhb4Un"
   },
   "outputs": [],
   "source": [
    "X_train,X_test,y_train,y_test = train_test_split(X,y,test_size=0.2,random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 91
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 4293,
     "status": "ok",
     "timestamp": 1588067943123,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "s5HFGtJGb4SN",
    "outputId": "3c0fab65-ae95-49bd-aaa4-db2aac198953"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train  (4800, 202)\n",
      "X_test   (1200, 202)\n",
      "Y_train  (4800, 2)\n",
      "Y_test   (1200, 2)\n"
     ]
    }
   ],
   "source": [
    "print('X_train ', X_train.shape)\n",
    "print('X_test  ', X_test.shape)\n",
    "print('Y_train ', y_train.shape)\n",
    "print('Y_test  ', y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 4056,
     "status": "ok",
     "timestamp": 1588067945413,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "dWDFL3MGb4QZ",
    "outputId": "cc749274-ecaa-4390-f10b-6dcdbe737ce6"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.005\n",
      "The Spearman's correlation coefficient is: -0.006\n"
     ]
    }
   ],
   "source": [
    "#linear regression\n",
    "linearRegressor = LinearRegression()\n",
    "linearRegressor.fit(X_train, y_train)\n",
    "y_pred = linearRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2040,
     "status": "ok",
     "timestamp": 1588068051016,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "wqLmraUf854y",
    "outputId": "8a041029-c583-46ea-8679-d6259ad0b0ca"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.082\n",
      "The Spearman's correlation coefficient is: 0.112\n"
     ]
    }
   ],
   "source": [
    "knnRegressor = KNeighborsRegressor(n_neighbors = 77)\n",
    "knnRegressor.fit(X_test, y_test)\n",
    "y_pred = knnRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 4695,
     "status": "ok",
     "timestamp": 1588067949996,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "2Q5eyam286DW",
    "outputId": "f00c0052-14f9-4bfc-87e9-7021e07a58c1"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: -0.019\n",
      "The Spearman's correlation coefficient is: 0.005\n"
     ]
    }
   ],
   "source": [
    "dtRegressor = DecisionTreeRegressor()\n",
    "dtRegressor.fit(X_train, y_train)\n",
    "y_pred = dtRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 102112,
     "status": "ok",
     "timestamp": 1588068048953,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "PHPhPBii86Bs",
    "outputId": "909e928b-9789-449d-89ac-30340f6ed910"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.000\n",
      "The Spearman's correlation coefficient is: 0.005\n"
     ]
    }
   ],
   "source": [
    "rfRegressor = RandomForestRegressor(n_estimators=100,random_state=45)\n",
    "rfRegressor.fit(X_train,y_train)\n",
    "y_pred = rfRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "wEdyE41CO-No"
   },
   "source": [
    "*For other ensemble and neural networking models*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "AJOfvYlr85_3"
   },
   "outputs": [],
   "source": [
    "y_short = dfGroundTruth[['short-term_memorability']].values\n",
    "y_long = dfGroundTruth[['long-term_memorability']].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "wnAUnh9yOejb"
   },
   "outputs": [],
   "source": [
    "# Splitting the dataset into the Training set and Test set\n",
    "short_X_train,short_X_test,short_y_train,short_y_test = train_test_split(X,y_short,test_size=0.2,random_state=40)\n",
    "long_X_train,long_X_test,long_y_train,long_y_test = train_test_split(X,y_long,test_size=0.2,random_state=40)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 149
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 200039,
     "status": "ok",
     "timestamp": 1588068355350,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "Xg50FqQGOe5j",
    "outputId": "4ffe0400-4eaf-46d0-af69-0b2c013a25f3"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_voting.py:406: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: -0.009\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_voting.py:406: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.040\n"
     ]
    }
   ],
   "source": [
    "model1 = LinearRegression()\n",
    "model2 = DecisionTreeRegressor(random_state=1)\n",
    "model3 = KNeighborsRegressor(n_neighbors = 77)\n",
    "model4 = RandomForestRegressor(n_estimators=100,random_state=45)\n",
    "vRegressor = VotingRegressor(estimators=[('lr', model1), ('dt', model2), ('knn', model3), ('rf', model4)])\n",
    "\n",
    "vRegressor.fit(short_X_train,short_y_train)\n",
    "y_pred = vRegressor.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "vRegressor.fit(long_X_train,long_y_train)\n",
    "y_pred = vRegressor.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 149
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 15843,
     "status": "ok",
     "timestamp": 1588068372097,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "FaClu8dqOe-y",
    "outputId": "c71ab94c-c5c4-4e20-9ce4-687d454a00a1"
   },
   "outputs": [
    {
     "name": "stderr",
     "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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: -0.058\n"
     ]
    },
    {
     "name": "stderr",
     "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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.040\n"
     ]
    }
   ],
   "source": [
    "abRegressor = AdaBoostRegressor()\n",
    "abRegressor.fit(short_X_train, short_y_train)\n",
    "y_pred = abRegressor.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "abRegressor = AdaBoostRegressor()\n",
    "abRegressor.fit(long_X_train, long_y_train)\n",
    "y_pred = abRegressor.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 149
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 68025,
     "status": "ok",
     "timestamp": 1588068424291,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "cOj9nru2Oe25",
    "outputId": "bbc6414f-622d-46a3-fa94-512232f88eae"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_gb.py:1454: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: -0.004\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_gb.py:1454: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.021\n"
     ]
    }
   ],
   "source": [
    "gbRegressor= GradientBoostingRegressor()\n",
    "\n",
    "gbRegressor.fit(short_X_train,short_y_train)\n",
    "y_pred = gbRegressor.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "gbRegressor.fit(long_X_train,long_y_train)\n",
    "y_pred = gbRegressor.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "QMXz2KwVOe0m"
   },
   "outputs": [],
   "source": [
    "sc = StandardScaler()\n",
    "short_X_train = sc.fit_transform(short_X_train)\n",
    "short_X_test = sc.transform(short_X_test)\n",
    "\n",
    "long_X_train = sc.fit_transform(long_X_train)\n",
    "long_X_test = sc.transform(long_X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "CqrFmyy9OeyQ"
   },
   "outputs": [],
   "source": [
    "# Initialising\n",
    "neuralNetModel = Sequential()\n",
    "\n",
    "# Adding the input layer and first hidden layer\n",
    "neuralNetModel.add(Dense(32, activation = 'relu', input_dim = 202))\n",
    "\n",
    "# Adding second hidden layer\n",
    "neuralNetModel.add(Dense(units = 32, activation = 'relu'))\n",
    "\n",
    "# Adding third hidden layer\n",
    "neuralNetModel.add(Dense(units = 32, activation = 'relu'))\n",
    "\n",
    "# Adding output layer\n",
    "neuralNetModel.add(Dense(units = 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 126317,
     "status": "ok",
     "timestamp": 1588068576354,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "ayLr3L65Oewd",
    "outputId": "58a05369-1336-47aa-af20-c8a89dd5c472"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/51\n",
      "4800/4800 [==============================] - 1s 266us/step - loss: 0.0752\n",
      "Epoch 2/51\n",
      "4800/4800 [==============================] - 1s 238us/step - loss: 0.0146\n",
      "Epoch 3/51\n",
      "4800/4800 [==============================] - 1s 283us/step - loss: 0.0228\n",
      "Epoch 4/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0111\n",
      "Epoch 5/51\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 0.0106\n",
      "Epoch 6/51\n",
      "4800/4800 [==============================] - 1s 283us/step - loss: 0.0076\n",
      "Epoch 7/51\n",
      "4800/4800 [==============================] - 1s 290us/step - loss: 0.0069\n",
      "Epoch 8/51\n",
      "4800/4800 [==============================] - 1s 290us/step - loss: 0.0069\n",
      "Epoch 9/51\n",
      "4800/4800 [==============================] - 1s 238us/step - loss: 0.0075\n",
      "Epoch 10/51\n",
      "4800/4800 [==============================] - 1s 280us/step - loss: 0.0066\n",
      "Epoch 11/51\n",
      "4800/4800 [==============================] - 1s 240us/step - loss: 0.0066\n",
      "Epoch 12/51\n",
      "4800/4800 [==============================] - 1s 234us/step - loss: 0.0064\n",
      "Epoch 13/51\n",
      "4800/4800 [==============================] - 1s 255us/step - loss: 0.0064\n",
      "Epoch 14/51\n",
      "4800/4800 [==============================] - 1s 251us/step - loss: 0.0064\n",
      "Epoch 15/51\n",
      "4800/4800 [==============================] - 1s 245us/step - loss: 0.0063\n",
      "Epoch 16/51\n",
      "4800/4800 [==============================] - 1s 281us/step - loss: 0.0063\n",
      "Epoch 17/51\n",
      "4800/4800 [==============================] - 1s 248us/step - loss: 0.0063\n",
      "Epoch 18/51\n",
      "4800/4800 [==============================] - 1s 243us/step - loss: 0.0063\n",
      "Epoch 19/51\n",
      "4800/4800 [==============================] - 1s 242us/step - loss: 0.0061\n",
      "Epoch 20/51\n",
      "4800/4800 [==============================] - 1s 269us/step - loss: 0.0063\n",
      "Epoch 21/51\n",
      "4800/4800 [==============================] - 1s 245us/step - loss: 0.0060\n",
      "Epoch 22/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0060\n",
      "Epoch 23/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0060\n",
      "Epoch 24/51\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 0.0057\n",
      "Epoch 25/51\n",
      "4800/4800 [==============================] - 1s 251us/step - loss: 0.0060\n",
      "Epoch 26/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0058\n",
      "Epoch 27/51\n",
      "4800/4800 [==============================] - 1s 245us/step - loss: 0.0056\n",
      "Epoch 28/51\n",
      "4800/4800 [==============================] - 1s 239us/step - loss: 0.0057\n",
      "Epoch 29/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0055\n",
      "Epoch 30/51\n",
      "4800/4800 [==============================] - 1s 246us/step - loss: 0.0054\n",
      "Epoch 31/51\n",
      "4800/4800 [==============================] - 1s 243us/step - loss: 0.0054\n",
      "Epoch 32/51\n",
      "4800/4800 [==============================] - 1s 254us/step - loss: 0.0054\n",
      "Epoch 33/51\n",
      "4800/4800 [==============================] - 1s 279us/step - loss: 0.0052\n",
      "Epoch 34/51\n",
      "4800/4800 [==============================] - 1s 255us/step - loss: 0.0053\n",
      "Epoch 35/51\n",
      "4800/4800 [==============================] - 1s 258us/step - loss: 0.0053\n",
      "Epoch 36/51\n",
      "4800/4800 [==============================] - 1s 242us/step - loss: 0.0052\n",
      "Epoch 37/51\n",
      "4800/4800 [==============================] - 1s 274us/step - loss: 0.0052\n",
      "Epoch 38/51\n",
      "4800/4800 [==============================] - 1s 240us/step - loss: 0.0051\n",
      "Epoch 39/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0051\n",
      "Epoch 40/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0049\n",
      "Epoch 41/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0050\n",
      "Epoch 42/51\n",
      "4800/4800 [==============================] - 1s 239us/step - loss: 0.0050\n",
      "Epoch 43/51\n",
      "4800/4800 [==============================] - 1s 239us/step - loss: 0.0049\n",
      "Epoch 44/51\n",
      "4800/4800 [==============================] - 1s 242us/step - loss: 0.0048\n",
      "Epoch 45/51\n",
      "4800/4800 [==============================] - 1s 242us/step - loss: 0.0048\n",
      "Epoch 46/51\n",
      "4800/4800 [==============================] - 1s 246us/step - loss: 0.0048\n",
      "Epoch 47/51\n",
      "4800/4800 [==============================] - 1s 276us/step - loss: 0.0047\n",
      "Epoch 48/51\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 0.0048\n",
      "Epoch 49/51\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 0.0047\n",
      "Epoch 50/51\n",
      "4800/4800 [==============================] - 1s 258us/step - loss: 0.0047\n",
      "Epoch 51/51\n",
      "4800/4800 [==============================] - 1s 294us/step - loss: 0.0047\n",
      "Epoch 1/51\n",
      "4800/4800 [==============================] - 1s 279us/step - loss: 0.0213\n",
      "Epoch 2/51\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 0.0204\n",
      "Epoch 3/51\n",
      "4800/4800 [==============================] - 1s 242us/step - loss: 0.0199\n",
      "Epoch 4/51\n",
      "4800/4800 [==============================] - 1s 254us/step - loss: 0.0196\n",
      "Epoch 5/51\n",
      "4800/4800 [==============================] - 1s 245us/step - loss: 0.0189\n",
      "Epoch 6/51\n",
      "4800/4800 [==============================] - 1s 242us/step - loss: 0.0184\n",
      "Epoch 7/51\n",
      "4800/4800 [==============================] - 1s 238us/step - loss: 0.0182\n",
      "Epoch 8/51\n",
      "4800/4800 [==============================] - 1s 246us/step - loss: 0.0181\n",
      "Epoch 9/51\n",
      "4800/4800 [==============================] - 1s 239us/step - loss: 0.0178\n",
      "Epoch 10/51\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 0.0174\n",
      "Epoch 11/51\n",
      "4800/4800 [==============================] - 1s 279us/step - loss: 0.0170\n",
      "Epoch 12/51\n",
      "4800/4800 [==============================] - 1s 258us/step - loss: 0.0167\n",
      "Epoch 13/51\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 0.0169\n",
      "Epoch 14/51\n",
      "4800/4800 [==============================] - 1s 265us/step - loss: 0.0166\n",
      "Epoch 15/51\n",
      "4800/4800 [==============================] - 1s 284us/step - loss: 0.0164\n",
      "Epoch 16/51\n",
      "4800/4800 [==============================] - 1s 264us/step - loss: 0.0163\n",
      "Epoch 17/51\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 0.0162\n",
      "Epoch 18/51\n",
      "4800/4800 [==============================] - 1s 253us/step - loss: 0.0159\n",
      "Epoch 19/51\n",
      "4800/4800 [==============================] - 1s 253us/step - loss: 0.0156\n",
      "Epoch 20/51\n",
      "4800/4800 [==============================] - 1s 243us/step - loss: 0.0154\n",
      "Epoch 21/51\n",
      "4800/4800 [==============================] - 1s 237us/step - loss: 0.0153\n",
      "Epoch 22/51\n",
      "4800/4800 [==============================] - 1s 285us/step - loss: 0.0152\n",
      "Epoch 23/51\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 0.0153\n",
      "Epoch 24/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0149\n",
      "Epoch 25/51\n",
      "4800/4800 [==============================] - 1s 251us/step - loss: 0.0160\n",
      "Epoch 26/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0150\n",
      "Epoch 27/51\n",
      "4800/4800 [==============================] - 1s 278us/step - loss: 0.0148\n",
      "Epoch 28/51\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 0.0146\n",
      "Epoch 29/51\n",
      "4800/4800 [==============================] - 1s 286us/step - loss: 0.0144\n",
      "Epoch 30/51\n",
      "4800/4800 [==============================] - 1s 250us/step - loss: 0.0148\n",
      "Epoch 31/51\n",
      "4800/4800 [==============================] - 1s 265us/step - loss: 0.0142\n",
      "Epoch 32/51\n",
      "4800/4800 [==============================] - 1s 248us/step - loss: 0.0146\n",
      "Epoch 33/51\n",
      "4800/4800 [==============================] - 1s 245us/step - loss: 0.0143\n",
      "Epoch 34/51\n",
      "4800/4800 [==============================] - 1s 250us/step - loss: 0.0146\n",
      "Epoch 35/51\n",
      "4800/4800 [==============================] - 1s 239us/step - loss: 0.0144\n",
      "Epoch 36/51\n",
      "4800/4800 [==============================] - 1s 237us/step - loss: 0.0140\n",
      "Epoch 37/51\n",
      "4800/4800 [==============================] - 1s 252us/step - loss: 0.0140\n",
      "Epoch 38/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0139\n",
      "Epoch 39/51\n",
      "4800/4800 [==============================] - 1s 241us/step - loss: 0.0140\n",
      "Epoch 40/51\n",
      "4800/4800 [==============================] - 1s 237us/step - loss: 0.0138\n",
      "Epoch 41/51\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 0.0138\n",
      "Epoch 42/51\n",
      "4800/4800 [==============================] - 1s 251us/step - loss: 0.0144\n",
      "Epoch 43/51\n",
      "4800/4800 [==============================] - 1s 252us/step - loss: 0.0139\n",
      "Epoch 44/51\n",
      "4800/4800 [==============================] - 1s 249us/step - loss: 0.0140\n",
      "Epoch 45/51\n",
      "4800/4800 [==============================] - 1s 258us/step - loss: 0.0135\n",
      "Epoch 46/51\n",
      "4800/4800 [==============================] - 1s 255us/step - loss: 0.0133\n",
      "Epoch 47/51\n",
      "4800/4800 [==============================] - 1s 282us/step - loss: 0.0132\n",
      "Epoch 48/51\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 0.0133\n",
      "Epoch 49/51\n",
      "4800/4800 [==============================] - 1s 272us/step - loss: 0.0134\n",
      "Epoch 50/51\n",
      "4800/4800 [==============================] - 1s 247us/step - loss: 0.0132\n",
      "Epoch 51/51\n",
      "4800/4800 [==============================] - 1s 244us/step - loss: 0.0132\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.callbacks.History at 0x7faace2e0cc0>"
      ]
     },
     "execution_count": 118,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "neuralNetModel.compile(optimizer = 'adam',loss = 'mean_squared_error')\n",
    "neuralNetModel.fit(short_X_train, short_y_train, batch_size = 10, epochs = 51)\n",
    "neuralNetModel.fit(long_X_train, long_y_train, batch_size = 10, epochs = 51)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 36
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2311,
     "status": "ok",
     "timestamp": 1588068578687,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "VhVn-jFpOet0",
    "outputId": "bfa73256-a9ea-4cc2-bfb0-16800f0eefd0"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: -0.037\n"
     ]
    }
   ],
   "source": [
    "y_pred = neuralNetModel.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 36
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2299,
     "status": "ok",
     "timestamp": 1588068578689,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "4hW4ehbHOeru",
    "outputId": "b1ac76cd-f075-40a5-94bb-20442e14e751"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: -0.029\n"
     ]
    }
   ],
   "source": [
    "y_pred = neuralNetModel.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 281
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2293,
     "status": "ok",
     "timestamp": 1588068578693,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "G4R56QHwOepJ",
    "outputId": "e8792783-3642-4d0c-bcf1-56f30e79aa3b"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(y_test, color = 'green', label = 'Actual dataset')\n",
    "plt.plot(y_pred, color = 'mistyrose', label = 'Predicted dataset')\n",
    "plt.title('Prediction')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "4fR5C74LEkSX"
   },
   "source": [
    "**Working on captions + count vectorizer**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "CgHI7PL-859s"
   },
   "outputs": [],
   "source": [
    "X = xCountsSeq\n",
    "y = dfGroundTruth[['short-term_memorability','long-term_memorability']].values #these are our target columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "sOfaVD0iU5TS"
   },
   "outputs": [],
   "source": [
    "X_train,X_test,y_train,y_test = train_test_split(X,y,test_size=0.2,random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 91
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1000,
     "status": "ok",
     "timestamp": 1588070591078,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "jvxXk3psU9_t",
    "outputId": "04194f78-d18a-4b58-884f-671cedb411ba"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train  (4800, 5019)\n",
      "X_test   (1200, 5019)\n",
      "Y_train  (4800, 2)\n",
      "Y_test   (1200, 2)\n"
     ]
    }
   ],
   "source": [
    "print('X_train ', X_train.shape)\n",
    "print('X_test  ', X_test.shape)\n",
    "print('Y_train ', y_train.shape)\n",
    "print('Y_test  ', y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 52211,
     "status": "ok",
     "timestamp": 1588070645587,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "HRsWIMg7U-JJ",
    "outputId": "361537bf-c63a-4034-c57a-aa5981132027"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.110\n",
      "The Spearman's correlation coefficient is: 0.030\n"
     ]
    }
   ],
   "source": [
    "#linear regression\n",
    "linearRegressor = LinearRegression()\n",
    "linearRegressor.fit(X_train, y_train)\n",
    "y_pred = linearRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 62199,
     "status": "ok",
     "timestamp": 1588070660562,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "zCttUuMCU-NP",
    "outputId": "8855df89-e78e-4a5e-f64a-e89f16c892e8"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.402\n",
      "The Spearman's correlation coefficient is: 0.231\n"
     ]
    }
   ],
   "source": [
    "knnRegressor = KNeighborsRegressor(n_neighbors = 77)\n",
    "knnRegressor.fit(X_test, y_test)\n",
    "y_pred = knnRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 82314,
     "status": "ok",
     "timestamp": 1588070683658,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "LlVQBRgUU-He",
    "outputId": "701b8a26-6a8a-40c8-a29e-6f257a661b0f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.267\n",
      "The Spearman's correlation coefficient is: 0.110\n"
     ]
    }
   ],
   "source": [
    "dtRegressor = DecisionTreeRegressor()\n",
    "dtRegressor.fit(X_train, y_train)\n",
    "y_pred = dtRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 169,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1000726,
     "status": "ok",
     "timestamp": 1588071608055,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "ZAKbosR1VGny",
    "outputId": "a3e86345-e760-4698-9d0d-80851a51b33e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.396\n",
      "The Spearman's correlation coefficient is: 0.172\n"
     ]
    }
   ],
   "source": [
    "rfRegressor = RandomForestRegressor(n_estimators=100,random_state=45)\n",
    "rfRegressor.fit(X_train,y_train)\n",
    "y_pred = rfRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "7Va1wIdSdA4H"
   },
   "source": [
    "*For other ensemble and neural networking models*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "0K814jsCTqqp"
   },
   "outputs": [],
   "source": [
    "y_short = dfGroundTruth[['short-term_memorability']].values\n",
    "y_long = dfGroundTruth[['long-term_memorability']].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "piPVLd7ZTqvI"
   },
   "outputs": [],
   "source": [
    "# Splitting the dataset into the Training set and Test set\n",
    "short_X_train,short_X_test,short_y_train,short_y_test = train_test_split(X,y_short,test_size=0.2,random_state=40)\n",
    "long_X_train,long_X_test,long_y_train,long_y_test = train_test_split(X,y_long,test_size=0.2,random_state=40)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 149
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2120688,
     "status": "ok",
     "timestamp": 1588073922759,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "s2CtsPgAdHAG",
    "outputId": "a745951b-3af6-4f8f-8224-108a4e7098cf"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_voting.py:406: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.129\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_voting.py:406: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.075\n"
     ]
    }
   ],
   "source": [
    "model1 = LinearRegression()\n",
    "model2 = DecisionTreeRegressor(random_state=1)\n",
    "model3 = KNeighborsRegressor(n_neighbors = 77)\n",
    "model4 = RandomForestRegressor(n_estimators=100,random_state=45)\n",
    "vRegressor = VotingRegressor(estimators=[('lr', model1), ('dt', model2), ('knn', model3), ('rf', model4)])\n",
    "\n",
    "vRegressor.fit(short_X_train,short_y_train)\n",
    "y_pred = vRegressor.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "vRegressor.fit(long_X_train,long_y_train)\n",
    "y_pred = vRegressor.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 173,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 149
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 31415,
     "status": "ok",
     "timestamp": 1588073954244,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "2GRve5ovdKtL",
    "outputId": "59484700-687a-4405-9bb6-b1476c35d6d8"
   },
   "outputs": [
    {
     "name": "stderr",
     "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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.185\n"
     ]
    },
    {
     "name": "stderr",
     "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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.120\n"
     ]
    }
   ],
   "source": [
    "abRegressor = AdaBoostRegressor()\n",
    "abRegressor.fit(short_X_train, short_y_train)\n",
    "y_pred = abRegressor.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "abRegressor = AdaBoostRegressor()\n",
    "abRegressor.fit(long_X_train, long_y_train)\n",
    "y_pred = abRegressor.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 149
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 168979,
     "status": "ok",
     "timestamp": 1588074091857,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "30uUeyzNdHHe",
    "outputId": "fca99a00-2b82-4119-d30d-1c43405c5fe5"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_gb.py:1454: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.322\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_gb.py:1454: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.170\n"
     ]
    }
   ],
   "source": [
    "gbRegressor= GradientBoostingRegressor()\n",
    "\n",
    "gbRegressor.fit(short_X_train,short_y_train)\n",
    "y_pred = gbRegressor.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "gbRegressor.fit(long_X_train,long_y_train)\n",
    "y_pred = gbRegressor.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "gPQ5b8L1dHFN"
   },
   "outputs": [],
   "source": [
    "sc = StandardScaler()\n",
    "short_X_train = sc.fit_transform(short_X_train)\n",
    "short_X_test = sc.transform(short_X_test)\n",
    "\n",
    "long_X_train = sc.fit_transform(long_X_train)\n",
    "long_X_test = sc.transform(long_X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "7Oa7sFVJdUl8"
   },
   "outputs": [],
   "source": [
    "# Initialising\n",
    "neuralNetModel = Sequential()\n",
    "\n",
    "# Adding the input layer and first hidden layer\n",
    "neuralNetModel.add(Dense(32, activation = 'relu', input_dim = 5019))\n",
    "\n",
    "# Adding second hidden layer\n",
    "neuralNetModel.add(Dense(units = 32, activation = 'relu'))\n",
    "\n",
    "# Adding third hidden layer\n",
    "neuralNetModel.add(Dense(units = 32, activation = 'relu'))\n",
    "\n",
    "# Adding output layer\n",
    "neuralNetModel.add(Dense(units = 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 185,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 262808,
     "status": "ok",
     "timestamp": 1588074666429,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "QRZOm5BddUsi",
    "outputId": "02dbd702-175a-4bcc-e0e6-07a1ee6c9d86"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/101\n",
      "4800/4800 [==============================] - 1s 296us/step - loss: 0.0063\n",
      "Epoch 2/101\n",
      "4800/4800 [==============================] - 1s 265us/step - loss: 0.0038\n",
      "Epoch 3/101\n",
      "4800/4800 [==============================] - 1s 260us/step - loss: 0.0028\n",
      "Epoch 4/101\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 0.0022\n",
      "Epoch 5/101\n",
      "4800/4800 [==============================] - 1s 289us/step - loss: 0.0018\n",
      "Epoch 6/101\n",
      "4800/4800 [==============================] - 1s 275us/step - loss: 0.0016\n",
      "Epoch 7/101\n",
      "4800/4800 [==============================] - 1s 269us/step - loss: 0.0014\n",
      "Epoch 8/101\n",
      "4800/4800 [==============================] - 1s 288us/step - loss: 0.0013\n",
      "Epoch 9/101\n",
      "4800/4800 [==============================] - 1s 282us/step - loss: 0.0012\n",
      "Epoch 10/101\n",
      "4800/4800 [==============================] - 1s 252us/step - loss: 0.0012\n",
      "Epoch 11/101\n",
      "4800/4800 [==============================] - 1s 305us/step - loss: 0.0012\n",
      "Epoch 12/101\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 0.0011\n",
      "Epoch 13/101\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 0.0011\n",
      "Epoch 14/101\n",
      "4800/4800 [==============================] - 1s 264us/step - loss: 0.0010\n",
      "Epoch 15/101\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 9.9111e-04\n",
      "Epoch 16/101\n",
      "4800/4800 [==============================] - 1s 260us/step - loss: 9.9349e-04\n",
      "Epoch 17/101\n",
      "4800/4800 [==============================] - 1s 295us/step - loss: 9.6424e-04\n",
      "Epoch 18/101\n",
      "4800/4800 [==============================] - 1s 285us/step - loss: 9.3842e-04\n",
      "Epoch 19/101\n",
      "4800/4800 [==============================] - 1s 276us/step - loss: 9.3659e-04\n",
      "Epoch 20/101\n",
      "4800/4800 [==============================] - 2s 316us/step - loss: 8.9886e-04\n",
      "Epoch 21/101\n",
      "4800/4800 [==============================] - 1s 271us/step - loss: 9.4349e-04\n",
      "Epoch 22/101\n",
      "4800/4800 [==============================] - 1s 272us/step - loss: 9.0350e-04\n",
      "Epoch 23/101\n",
      "4800/4800 [==============================] - 1s 273us/step - loss: 8.8165e-04\n",
      "Epoch 24/101\n",
      "4800/4800 [==============================] - 1s 284us/step - loss: 8.8031e-04\n",
      "Epoch 25/101\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 8.4057e-04\n",
      "Epoch 26/101\n",
      "4800/4800 [==============================] - 1s 252us/step - loss: 8.5517e-04\n",
      "Epoch 27/101\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 8.7266e-04\n",
      "Epoch 28/101\n",
      "4800/4800 [==============================] - 1s 254us/step - loss: 8.5679e-04\n",
      "Epoch 29/101\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 8.3094e-04\n",
      "Epoch 30/101\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 7.8824e-04\n",
      "Epoch 31/101\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 7.9423e-04\n",
      "Epoch 32/101\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 7.9389e-04\n",
      "Epoch 33/101\n",
      "4800/4800 [==============================] - 1s 273us/step - loss: 7.8228e-04\n",
      "Epoch 34/101\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 7.8025e-04\n",
      "Epoch 35/101\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 7.4708e-04\n",
      "Epoch 36/101\n",
      "4800/4800 [==============================] - 1s 308us/step - loss: 7.7917e-04\n",
      "Epoch 37/101\n",
      "4800/4800 [==============================] - 2s 331us/step - loss: 7.9650e-04\n",
      "Epoch 38/101\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 7.5601e-04\n",
      "Epoch 39/101\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 7.5729e-04\n",
      "Epoch 40/101\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 7.4816e-04\n",
      "Epoch 41/101\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 7.7376e-04\n",
      "Epoch 42/101\n",
      "4800/4800 [==============================] - 1s 255us/step - loss: 7.4304e-04\n",
      "Epoch 43/101\n",
      "4800/4800 [==============================] - 1s 258us/step - loss: 7.2315e-04\n",
      "Epoch 44/101\n",
      "4800/4800 [==============================] - 1s 254us/step - loss: 7.7280e-04\n",
      "Epoch 45/101\n",
      "4800/4800 [==============================] - 1s 304us/step - loss: 7.3456e-04\n",
      "Epoch 46/101\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 7.4096e-04\n",
      "Epoch 47/101\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 7.2876e-04\n",
      "Epoch 48/101\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 7.0474e-04\n",
      "Epoch 49/101\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 7.0368e-04\n",
      "Epoch 50/101\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 7.0526e-04\n",
      "Epoch 51/101\n",
      "4800/4800 [==============================] - 1s 264us/step - loss: 7.2946e-04\n",
      "Epoch 52/101\n",
      "4800/4800 [==============================] - 1s 285us/step - loss: 7.2171e-04\n",
      "Epoch 53/101\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 6.9472e-04\n",
      "Epoch 54/101\n",
      "4800/4800 [==============================] - 1s 275us/step - loss: 6.9917e-04\n",
      "Epoch 55/101\n",
      "4800/4800 [==============================] - 1s 282us/step - loss: 7.0716e-04\n",
      "Epoch 56/101\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 7.0717e-04\n",
      "Epoch 57/101\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 6.8609e-04\n",
      "Epoch 58/101\n",
      "4800/4800 [==============================] - 1s 260us/step - loss: 7.3139e-04\n",
      "Epoch 59/101\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 7.2229e-04\n",
      "Epoch 60/101\n",
      "4800/4800 [==============================] - 1s 296us/step - loss: 7.0390e-04\n",
      "Epoch 61/101\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 6.8536e-04\n",
      "Epoch 62/101\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 6.8742e-04\n",
      "Epoch 63/101\n",
      "4800/4800 [==============================] - 1s 260us/step - loss: 6.5986e-04\n",
      "Epoch 64/101\n",
      "4800/4800 [==============================] - 1s 258us/step - loss: 6.6603e-04\n",
      "Epoch 65/101\n",
      "4800/4800 [==============================] - 1s 254us/step - loss: 6.7783e-04\n",
      "Epoch 66/101\n",
      "4800/4800 [==============================] - 1s 252us/step - loss: 6.6181e-04\n",
      "Epoch 67/101\n",
      "4800/4800 [==============================] - 1s 285us/step - loss: 7.0073e-04\n",
      "Epoch 68/101\n",
      "4800/4800 [==============================] - 1s 255us/step - loss: 6.7752e-04\n",
      "Epoch 69/101\n",
      "4800/4800 [==============================] - 1s 276us/step - loss: 6.9543e-04\n",
      "Epoch 70/101\n",
      "4800/4800 [==============================] - 1s 252us/step - loss: 6.7564e-04\n",
      "Epoch 71/101\n",
      "4800/4800 [==============================] - 1s 258us/step - loss: 7.0161e-04\n",
      "Epoch 72/101\n",
      "4800/4800 [==============================] - 1s 258us/step - loss: 6.7401e-04\n",
      "Epoch 73/101\n",
      "4800/4800 [==============================] - 1s 254us/step - loss: 6.6512e-04\n",
      "Epoch 74/101\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 6.7075e-04\n",
      "Epoch 75/101\n",
      "4800/4800 [==============================] - 1s 258us/step - loss: 6.4930e-04\n",
      "Epoch 76/101\n",
      "4800/4800 [==============================] - 1s 267us/step - loss: 6.6448e-04\n",
      "Epoch 77/101\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 6.4903e-04\n",
      "Epoch 78/101\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 6.7112e-04\n",
      "Epoch 79/101\n",
      "4800/4800 [==============================] - 1s 271us/step - loss: 6.7638e-04\n",
      "Epoch 80/101\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 6.8038e-04\n",
      "Epoch 81/101\n",
      "4800/4800 [==============================] - 1s 270us/step - loss: 6.5699e-04\n",
      "Epoch 82/101\n",
      "4800/4800 [==============================] - 1s 301us/step - loss: 7.0847e-04\n",
      "Epoch 83/101\n",
      "4800/4800 [==============================] - 1s 281us/step - loss: 6.4471e-04\n",
      "Epoch 84/101\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 6.4728e-04\n",
      "Epoch 85/101\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 6.2789e-04\n",
      "Epoch 86/101\n",
      "4800/4800 [==============================] - 1s 260us/step - loss: 6.5315e-04\n",
      "Epoch 87/101\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 6.4809e-04\n",
      "Epoch 88/101\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 6.5416e-04\n",
      "Epoch 89/101\n",
      "4800/4800 [==============================] - 1s 269us/step - loss: 6.3197e-04\n",
      "Epoch 90/101\n",
      "4800/4800 [==============================] - 1s 267us/step - loss: 6.2745e-04\n",
      "Epoch 91/101\n",
      "4800/4800 [==============================] - 1s 287us/step - loss: 6.3583e-04\n",
      "Epoch 92/101\n",
      "4800/4800 [==============================] - 1s 277us/step - loss: 6.3281e-04\n",
      "Epoch 93/101\n",
      "4800/4800 [==============================] - 1s 308us/step - loss: 6.2470e-04\n",
      "Epoch 94/101\n",
      "4800/4800 [==============================] - 1s 304us/step - loss: 6.4087e-04\n",
      "Epoch 95/101\n",
      "4800/4800 [==============================] - 1s 305us/step - loss: 6.3761e-04\n",
      "Epoch 96/101\n",
      "4800/4800 [==============================] - 1s 260us/step - loss: 6.3328e-04\n",
      "Epoch 97/101\n",
      "4800/4800 [==============================] - 1s 265us/step - loss: 6.3786e-04\n",
      "Epoch 98/101\n",
      "4800/4800 [==============================] - 2s 322us/step - loss: 6.4835e-04\n",
      "Epoch 99/101\n",
      "4800/4800 [==============================] - 1s 291us/step - loss: 6.3880e-04\n",
      "Epoch 100/101\n",
      "4800/4800 [==============================] - 1s 312us/step - loss: 6.4735e-04\n",
      "Epoch 101/101\n",
      "4800/4800 [==============================] - 1s 289us/step - loss: 6.4370e-04\n",
      "Epoch 1/101\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 0.0144\n",
      "Epoch 2/101\n",
      "4800/4800 [==============================] - 1s 294us/step - loss: 0.0094\n",
      "Epoch 3/101\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 0.0069\n",
      "Epoch 4/101\n",
      "4800/4800 [==============================] - 1s 282us/step - loss: 0.0057\n",
      "Epoch 5/101\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 0.0051\n",
      "Epoch 6/101\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 0.0047\n",
      "Epoch 7/101\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 0.0043\n",
      "Epoch 8/101\n",
      "4800/4800 [==============================] - 1s 255us/step - loss: 0.0042\n",
      "Epoch 9/101\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 0.0040\n",
      "Epoch 10/101\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 0.0040\n",
      "Epoch 11/101\n",
      "4800/4800 [==============================] - 1s 269us/step - loss: 0.0039\n",
      "Epoch 12/101\n",
      "4800/4800 [==============================] - 1s 301us/step - loss: 0.0037\n",
      "Epoch 13/101\n",
      "4800/4800 [==============================] - 1s 271us/step - loss: 0.0036\n",
      "Epoch 14/101\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 0.0036\n",
      "Epoch 15/101\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 0.0036\n",
      "Epoch 16/101\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 0.0035\n",
      "Epoch 17/101\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 0.0034\n",
      "Epoch 18/101\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 0.0034\n",
      "Epoch 19/101\n",
      "4800/4800 [==============================] - 1s 260us/step - loss: 0.0034\n",
      "Epoch 20/101\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 0.0034\n",
      "Epoch 21/101\n",
      "4800/4800 [==============================] - 1s 300us/step - loss: 0.0034\n",
      "Epoch 22/101\n",
      "4800/4800 [==============================] - 1s 273us/step - loss: 0.0033\n",
      "Epoch 23/101\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 0.0033\n",
      "Epoch 24/101\n",
      "4800/4800 [==============================] - 1s 265us/step - loss: 0.0032\n",
      "Epoch 25/101\n",
      "4800/4800 [==============================] - 1s 253us/step - loss: 0.0032\n",
      "Epoch 26/101\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 0.0032\n",
      "Epoch 27/101\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 0.0032\n",
      "Epoch 28/101\n",
      "4800/4800 [==============================] - 1s 277us/step - loss: 0.0032\n",
      "Epoch 29/101\n",
      "4800/4800 [==============================] - 1s 258us/step - loss: 0.0032\n",
      "Epoch 30/101\n",
      "4800/4800 [==============================] - 1s 273us/step - loss: 0.0031\n",
      "Epoch 31/101\n",
      "4800/4800 [==============================] - 1s 265us/step - loss: 0.0032\n",
      "Epoch 32/101\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 0.0031\n",
      "Epoch 33/101\n",
      "4800/4800 [==============================] - 1s 258us/step - loss: 0.0031\n",
      "Epoch 34/101\n",
      "4800/4800 [==============================] - 1s 260us/step - loss: 0.0030\n",
      "Epoch 35/101\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 0.0030\n",
      "Epoch 36/101\n",
      "4800/4800 [==============================] - 1s 258us/step - loss: 0.0030\n",
      "Epoch 37/101\n",
      "4800/4800 [==============================] - 1s 291us/step - loss: 0.0031\n",
      "Epoch 38/101\n",
      "4800/4800 [==============================] - 1s 255us/step - loss: 0.0030\n",
      "Epoch 39/101\n",
      "4800/4800 [==============================] - 1s 266us/step - loss: 0.0030\n",
      "Epoch 40/101\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 0.0030\n",
      "Epoch 41/101\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 0.0030\n",
      "Epoch 42/101\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 0.0030\n",
      "Epoch 43/101\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 0.0030\n",
      "Epoch 44/101\n",
      "4800/4800 [==============================] - 1s 283us/step - loss: 0.0029\n",
      "Epoch 45/101\n",
      "4800/4800 [==============================] - 1s 269us/step - loss: 0.0029\n",
      "Epoch 46/101\n",
      "4800/4800 [==============================] - 1s 272us/step - loss: 0.0029\n",
      "Epoch 47/101\n",
      "4800/4800 [==============================] - 1s 285us/step - loss: 0.0030\n",
      "Epoch 48/101\n",
      "4800/4800 [==============================] - 1s 272us/step - loss: 0.0030\n",
      "Epoch 49/101\n",
      "4800/4800 [==============================] - 1s 274us/step - loss: 0.0029\n",
      "Epoch 50/101\n",
      "4800/4800 [==============================] - 1s 286us/step - loss: 0.0029\n",
      "Epoch 51/101\n",
      "4800/4800 [==============================] - 1s 258us/step - loss: 0.0029\n",
      "Epoch 52/101\n",
      "4800/4800 [==============================] - 1s 267us/step - loss: 0.0029\n",
      "Epoch 53/101\n",
      "4800/4800 [==============================] - 1s 254us/step - loss: 0.0029\n",
      "Epoch 54/101\n",
      "4800/4800 [==============================] - 1s 282us/step - loss: 0.0029\n",
      "Epoch 55/101\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 0.0029\n",
      "Epoch 56/101\n",
      "4800/4800 [==============================] - 1s 294us/step - loss: 0.0029\n",
      "Epoch 57/101\n",
      "4800/4800 [==============================] - 1s 265us/step - loss: 0.0028\n",
      "Epoch 58/101\n",
      "4800/4800 [==============================] - 1s 271us/step - loss: 0.0029\n",
      "Epoch 59/101\n",
      "4800/4800 [==============================] - 2s 324us/step - loss: 0.0029\n",
      "Epoch 60/101\n",
      "4800/4800 [==============================] - 1s 281us/step - loss: 0.0028\n",
      "Epoch 61/101\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 0.0028\n",
      "Epoch 62/101\n",
      "4800/4800 [==============================] - 1s 264us/step - loss: 0.0029\n",
      "Epoch 63/101\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 0.0028\n",
      "Epoch 64/101\n",
      "4800/4800 [==============================] - 1s 299us/step - loss: 0.0028\n",
      "Epoch 65/101\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 0.0028\n",
      "Epoch 66/101\n",
      "4800/4800 [==============================] - 1s 270us/step - loss: 0.0028\n",
      "Epoch 67/101\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 0.0028\n",
      "Epoch 68/101\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 0.0028\n",
      "Epoch 69/101\n",
      "4800/4800 [==============================] - 1s 254us/step - loss: 0.0028\n",
      "Epoch 70/101\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 0.0028\n",
      "Epoch 71/101\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 0.0028\n",
      "Epoch 72/101\n",
      "4800/4800 [==============================] - 1s 258us/step - loss: 0.0028\n",
      "Epoch 73/101\n",
      "4800/4800 [==============================] - 1s 271us/step - loss: 0.0028\n",
      "Epoch 74/101\n",
      "4800/4800 [==============================] - 1s 288us/step - loss: 0.0028\n",
      "Epoch 75/101\n",
      "4800/4800 [==============================] - 1s 267us/step - loss: 0.0028\n",
      "Epoch 76/101\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 0.0028\n",
      "Epoch 77/101\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 0.0027\n",
      "Epoch 78/101\n",
      "4800/4800 [==============================] - 1s 255us/step - loss: 0.0027\n",
      "Epoch 79/101\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 0.0028\n",
      "Epoch 80/101\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 0.0028\n",
      "Epoch 81/101\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 0.0027\n",
      "Epoch 82/101\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 0.0028\n",
      "Epoch 83/101\n",
      "4800/4800 [==============================] - 1s 255us/step - loss: 0.0028\n",
      "Epoch 84/101\n",
      "4800/4800 [==============================] - 1s 258us/step - loss: 0.0027\n",
      "Epoch 85/101\n",
      "4800/4800 [==============================] - 1s 255us/step - loss: 0.0027\n",
      "Epoch 86/101\n",
      "4800/4800 [==============================] - 1s 297us/step - loss: 0.0027\n",
      "Epoch 87/101\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 0.0027\n",
      "Epoch 88/101\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 0.0027\n",
      "Epoch 89/101\n",
      "4800/4800 [==============================] - 1s 258us/step - loss: 0.0027\n",
      "Epoch 90/101\n",
      "4800/4800 [==============================] - 1s 287us/step - loss: 0.0027\n",
      "Epoch 91/101\n",
      "4800/4800 [==============================] - 1s 296us/step - loss: 0.0027\n",
      "Epoch 92/101\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 0.0027\n",
      "Epoch 93/101\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 0.0027\n",
      "Epoch 94/101\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 0.0027\n",
      "Epoch 95/101\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 0.0027\n",
      "Epoch 96/101\n",
      "4800/4800 [==============================] - 1s 260us/step - loss: 0.0027\n",
      "Epoch 97/101\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 0.0027\n",
      "Epoch 98/101\n",
      "4800/4800 [==============================] - 1s 278us/step - loss: 0.0027\n",
      "Epoch 99/101\n",
      "4800/4800 [==============================] - 1s 296us/step - loss: 0.0027\n",
      "Epoch 100/101\n",
      "4800/4800 [==============================] - 1s 266us/step - loss: 0.0026\n",
      "Epoch 101/101\n",
      "4800/4800 [==============================] - 1s 267us/step - loss: 0.0027\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.callbacks.History at 0x7fab4522fef0>"
      ]
     },
     "execution_count": 185,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "neuralNetModel.compile(optimizer = 'adam',loss = 'mean_squared_error')\n",
    "neuralNetModel.fit(short_X_train, short_y_train, batch_size = 10, epochs = 101)\n",
    "neuralNetModel.fit(long_X_train, long_y_train, batch_size = 10, epochs = 101)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 186,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 259004,
     "status": "ok",
     "timestamp": 1588074666439,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "j0ZHB2V8Tqnl",
    "outputId": "1243b04e-d274-4a2e-bc3e-6666d2fdade9"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.172\n",
      "The Spearman's correlation coefficient is: 0.164\n"
     ]
    }
   ],
   "source": [
    "y_pred = neuralNetModel.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "y_pred = neuralNetModel.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 187,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 281
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 258257,
     "status": "ok",
     "timestamp": 1588074667308,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "IqPXznwBddoK",
    "outputId": "37c8b388-efac-464b-d7ea-ddf3433a1c7b"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(y_test, color = 'green', label = 'Actual dataset')\n",
    "plt.plot(y_pred, color = 'lightgrey', label = 'Predicted dataset')\n",
    "plt.title('Prediction')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "vKr3M0tcEpLF"
   },
   "source": [
    "**Working on captions + tf-idf**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "ZrWtnhXc9K7W"
   },
   "outputs": [],
   "source": [
    "X = dfCaptions_Dev_tfIdf\n",
    "y = dfGroundTruth[['short-term_memorability','long-term_memorability']].values #these are our target columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "cYyMlPOT8YgZ"
   },
   "outputs": [],
   "source": [
    "X_train,X_test,y_train,y_test = train_test_split(X,y,test_size=0.2,random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 91
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 956,
     "status": "ok",
     "timestamp": 1588092064311,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "kWHigsxU8fJG",
    "outputId": "febc78cf-862f-43a1-b933-f1a4476531a1"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train  (4800, 1000)\n",
      "X_test   (1200, 1000)\n",
      "Y_train  (4800, 2)\n",
      "Y_test   (1200, 2)\n"
     ]
    }
   ],
   "source": [
    "print('X_train ', X_train.shape)\n",
    "print('X_test  ', X_test.shape)\n",
    "print('Y_train ', y_train.shape)\n",
    "print('Y_test  ', y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2255,
     "status": "ok",
     "timestamp": 1588092069391,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "I_2xFa1a8f9Y",
    "outputId": "ad5c27da-d2c1-408b-8034-78b03bd00c9f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.381\n",
      "The Spearman's correlation coefficient is: 0.165\n"
     ]
    }
   ],
   "source": [
    "#linear regression\n",
    "linearRegressor = LinearRegression()\n",
    "linearRegressor.fit(X_train, y_train)\n",
    "y_pred = linearRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 4565,
     "status": "ok",
     "timestamp": 1588092077131,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "NO-FBx798gDV",
    "outputId": "7b4328c8-7c81-47e2-bbc8-b704d5fc3b77"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.371\n",
      "The Spearman's correlation coefficient is: 0.209\n"
     ]
    }
   ],
   "source": [
    "knnRegressor = KNeighborsRegressor(n_neighbors = 77)\n",
    "knnRegressor.fit(X_test, y_test)\n",
    "y_pred = knnRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3381,
     "status": "ok",
     "timestamp": 1588092081921,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "wPiCrTG58gVN",
    "outputId": "296ef0ca-d72a-44cf-e782-b267f7a7cc02"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.283\n",
      "The Spearman's correlation coefficient is: 0.093\n"
     ]
    }
   ],
   "source": [
    "dtRegressor = DecisionTreeRegressor()\n",
    "dtRegressor.fit(X_train, y_train)\n",
    "y_pred = dtRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 111875,
     "status": "ok",
     "timestamp": 1588092196121,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "2Atxlzkr8gjf",
    "outputId": "6697670f-c0e4-42ec-e3a4-7abc3c27f2e6"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.406\n",
      "The Spearman's correlation coefficient is: 0.142\n"
     ]
    }
   ],
   "source": [
    "rfRegressor = RandomForestRegressor(n_estimators=100,random_state=45)\n",
    "rfRegressor.fit(X_train,y_train)\n",
    "y_pred = rfRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "4stA6rXmveP-"
   },
   "source": [
    "*For other ensemble and neural networking models*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "HSf4isAWvfW4"
   },
   "outputs": [],
   "source": [
    "y_short = dfGroundTruth[['short-term_memorability']].values\n",
    "y_long = dfGroundTruth[['long-term_memorability']].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "35efBSKBvf0U"
   },
   "outputs": [],
   "source": [
    "# Splitting the dataset into the Training set and Test set\n",
    "short_X_train,short_X_test,short_y_train,short_y_test = train_test_split(X,y_short,test_size=0.2,random_state=40)\n",
    "long_X_train,long_X_test,long_y_train,long_y_test = train_test_split(X,y_long,test_size=0.2,random_state=40)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 252,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 149
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 310999,
     "status": "ok",
     "timestamp": 1588077039697,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "ti9p0i9evfyd",
    "outputId": "b0bff0ed-311c-4675-e9ff-27e7d3b0b12d"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_voting.py:406: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.363\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_voting.py:406: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.155\n"
     ]
    }
   ],
   "source": [
    "model1 = LinearRegression()\n",
    "model2 = DecisionTreeRegressor(random_state=1)\n",
    "model3 = KNeighborsRegressor(n_neighbors = 77)\n",
    "model4 = RandomForestRegressor(n_estimators=100,random_state=45)\n",
    "vRegressor = VotingRegressor(estimators=[('lr', model1), ('dt', model2), ('knn', model3), ('rf', model4)])\n",
    "\n",
    "vRegressor.fit(short_X_train,short_y_train)\n",
    "y_pred = vRegressor.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "vRegressor.fit(long_X_train,long_y_train)\n",
    "y_pred = vRegressor.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 253,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 149
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 318381,
     "status": "ok",
     "timestamp": 1588077047090,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "zSebZXx-vfwG",
    "outputId": "01b27793-86e9-48cb-deb9-7e1b6812b341"
   },
   "outputs": [
    {
     "name": "stderr",
     "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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.192\n"
     ]
    },
    {
     "name": "stderr",
     "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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.144\n"
     ]
    }
   ],
   "source": [
    "abRegressor = AdaBoostRegressor()\n",
    "abRegressor.fit(short_X_train, short_y_train)\n",
    "y_pred = abRegressor.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "abRegressor = AdaBoostRegressor()\n",
    "abRegressor.fit(long_X_train, long_y_train)\n",
    "y_pred = abRegressor.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 254,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 149
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 328857,
     "status": "ok",
     "timestamp": 1588077057576,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "7sC4p2DpvfuS",
    "outputId": "5c9a6b61-2559-4d8e-c74d-32b59a68bd68"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_gb.py:1454: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.319\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_gb.py:1454: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.164\n"
     ]
    }
   ],
   "source": [
    "gbRegressor= GradientBoostingRegressor()\n",
    "\n",
    "gbRegressor.fit(short_X_train,short_y_train)\n",
    "y_pred = gbRegressor.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "gbRegressor.fit(long_X_train,long_y_train)\n",
    "y_pred = gbRegressor.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "iPU0drO-vfsY"
   },
   "outputs": [],
   "source": [
    "sc = StandardScaler()\n",
    "short_X_train = sc.fit_transform(short_X_train)\n",
    "short_X_test = sc.transform(short_X_test)\n",
    "\n",
    "long_X_train = sc.fit_transform(long_X_train)\n",
    "long_X_test = sc.transform(long_X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "84z00cogvfp4"
   },
   "outputs": [],
   "source": [
    "# Initialising\n",
    "neuralNetModel = Sequential()\n",
    "\n",
    "# Adding the input layer and first hidden layer\n",
    "neuralNetModel.add(Dense(32, activation = 'relu', input_dim = 1000))\n",
    "\n",
    "# Adding second hidden layer\n",
    "neuralNetModel.add(Dense(units = 32, activation = 'relu'))\n",
    "\n",
    "# Adding third hidden layer\n",
    "neuralNetModel.add(Dense(units = 32, activation = 'relu'))\n",
    "\n",
    "# Adding output layer\n",
    "neuralNetModel.add(Dense(units = 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 260,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 135900,
     "status": "ok",
     "timestamp": 1588077245216,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "m3QE5T0rwK_6",
    "outputId": "2f713274-e262-4ee0-a132-47c3081c71e3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/51\n",
      "4800/4800 [==============================] - 1s 297us/step - loss: 0.0816\n",
      "Epoch 2/51\n",
      "4800/4800 [==============================] - 1s 272us/step - loss: 0.0116\n",
      "Epoch 3/51\n",
      "4800/4800 [==============================] - 1s 272us/step - loss: 0.0056\n",
      "Epoch 4/51\n",
      "4800/4800 [==============================] - 1s 293us/step - loss: 0.0042\n",
      "Epoch 5/51\n",
      "4800/4800 [==============================] - 1s 307us/step - loss: 0.0038\n",
      "Epoch 6/51\n",
      "4800/4800 [==============================] - 1s 308us/step - loss: 0.0038\n",
      "Epoch 7/51\n",
      "4800/4800 [==============================] - 1s 294us/step - loss: 0.0035\n",
      "Epoch 8/51\n",
      "4800/4800 [==============================] - 1s 267us/step - loss: 0.0035\n",
      "Epoch 9/51\n",
      "4800/4800 [==============================] - 1s 268us/step - loss: 0.0032\n",
      "Epoch 10/51\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 0.0031\n",
      "Epoch 11/51\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 0.0032\n",
      "Epoch 12/51\n",
      "4800/4800 [==============================] - 1s 264us/step - loss: 0.0031\n",
      "Epoch 13/51\n",
      "4800/4800 [==============================] - 1s 264us/step - loss: 0.0029\n",
      "Epoch 14/51\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 0.0028\n",
      "Epoch 15/51\n",
      "4800/4800 [==============================] - 1s 273us/step - loss: 0.0027\n",
      "Epoch 16/51\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 0.0025\n",
      "Epoch 17/51\n",
      "4800/4800 [==============================] - 1s 264us/step - loss: 0.0025\n",
      "Epoch 18/51\n",
      "4800/4800 [==============================] - 1s 270us/step - loss: 0.0025\n",
      "Epoch 19/51\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 0.0024\n",
      "Epoch 20/51\n",
      "4800/4800 [==============================] - 1s 253us/step - loss: 0.0024\n",
      "Epoch 21/51\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 0.0022\n",
      "Epoch 22/51\n",
      "4800/4800 [==============================] - 1s 269us/step - loss: 0.0021\n",
      "Epoch 23/51\n",
      "4800/4800 [==============================] - 1s 268us/step - loss: 0.0021\n",
      "Epoch 24/51\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 0.0022\n",
      "Epoch 25/51\n",
      "4800/4800 [==============================] - 1s 279us/step - loss: 0.0020\n",
      "Epoch 26/51\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 0.0021\n",
      "Epoch 27/51\n",
      "4800/4800 [==============================] - 1s 260us/step - loss: 0.0020\n",
      "Epoch 28/51\n",
      "4800/4800 [==============================] - 1s 255us/step - loss: 0.0020\n",
      "Epoch 29/51\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 0.0019\n",
      "Epoch 30/51\n",
      "4800/4800 [==============================] - 1s 274us/step - loss: 0.0018\n",
      "Epoch 31/51\n",
      "4800/4800 [==============================] - 1s 287us/step - loss: 0.0017\n",
      "Epoch 32/51\n",
      "4800/4800 [==============================] - 1s 277us/step - loss: 0.0017\n",
      "Epoch 33/51\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 0.0018\n",
      "Epoch 34/51\n",
      "4800/4800 [==============================] - 1s 260us/step - loss: 0.0018\n",
      "Epoch 35/51\n",
      "4800/4800 [==============================] - 1s 255us/step - loss: 0.0017\n",
      "Epoch 36/51\n",
      "4800/4800 [==============================] - 1s 269us/step - loss: 0.0016\n",
      "Epoch 37/51\n",
      "4800/4800 [==============================] - 1s 277us/step - loss: 0.0015\n",
      "Epoch 38/51\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 0.0016\n",
      "Epoch 39/51\n",
      "4800/4800 [==============================] - 1s 299us/step - loss: 0.0015\n",
      "Epoch 40/51\n",
      "4800/4800 [==============================] - 1s 304us/step - loss: 0.0015\n",
      "Epoch 41/51\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 0.0015\n",
      "Epoch 42/51\n",
      "4800/4800 [==============================] - 1s 267us/step - loss: 0.0015\n",
      "Epoch 43/51\n",
      "4800/4800 [==============================] - 2s 320us/step - loss: 0.0015\n",
      "Epoch 44/51\n",
      "4800/4800 [==============================] - 1s 292us/step - loss: 0.0014\n",
      "Epoch 45/51\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 0.0014\n",
      "Epoch 46/51\n",
      "4800/4800 [==============================] - 1s 270us/step - loss: 0.0014\n",
      "Epoch 47/51\n",
      "4800/4800 [==============================] - 1s 288us/step - loss: 0.0014\n",
      "Epoch 48/51\n",
      "4800/4800 [==============================] - 1s 279us/step - loss: 0.0014\n",
      "Epoch 49/51\n",
      "4800/4800 [==============================] - 2s 351us/step - loss: 0.0014\n",
      "Epoch 50/51\n",
      "4800/4800 [==============================] - 2s 337us/step - loss: 0.0014\n",
      "Epoch 51/51\n",
      "4800/4800 [==============================] - 2s 339us/step - loss: 0.0014\n",
      "Epoch 1/51\n",
      "4800/4800 [==============================] - 1s 267us/step - loss: 0.0203\n",
      "Epoch 2/51\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 0.0167\n",
      "Epoch 3/51\n",
      "4800/4800 [==============================] - 1s 272us/step - loss: 0.0134\n",
      "Epoch 4/51\n",
      "4800/4800 [==============================] - 1s 276us/step - loss: 0.0108\n",
      "Epoch 5/51\n",
      "4800/4800 [==============================] - 1s 274us/step - loss: 0.0095\n",
      "Epoch 6/51\n",
      "4800/4800 [==============================] - 1s 260us/step - loss: 0.0084\n",
      "Epoch 7/51\n",
      "4800/4800 [==============================] - 1s 266us/step - loss: 0.0076\n",
      "Epoch 8/51\n",
      "4800/4800 [==============================] - 1s 264us/step - loss: 0.0071\n",
      "Epoch 9/51\n",
      "4800/4800 [==============================] - 1s 267us/step - loss: 0.0066\n",
      "Epoch 10/51\n",
      "4800/4800 [==============================] - 1s 286us/step - loss: 0.0064\n",
      "Epoch 11/51\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 0.0060\n",
      "Epoch 12/51\n",
      "4800/4800 [==============================] - 1s 272us/step - loss: 0.0059\n",
      "Epoch 13/51\n",
      "4800/4800 [==============================] - 1s 253us/step - loss: 0.0058\n",
      "Epoch 14/51\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 0.0055\n",
      "Epoch 15/51\n",
      "4800/4800 [==============================] - 1s 256us/step - loss: 0.0055\n",
      "Epoch 16/51\n",
      "4800/4800 [==============================] - 1s 260us/step - loss: 0.0055\n",
      "Epoch 17/51\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 0.0053\n",
      "Epoch 18/51\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 0.0052\n",
      "Epoch 19/51\n",
      "4800/4800 [==============================] - 1s 269us/step - loss: 0.0051\n",
      "Epoch 20/51\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 0.0050\n",
      "Epoch 21/51\n",
      "4800/4800 [==============================] - 2s 313us/step - loss: 0.0050\n",
      "Epoch 22/51\n",
      "4800/4800 [==============================] - 1s 255us/step - loss: 0.0050\n",
      "Epoch 23/51\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 0.0048\n",
      "Epoch 24/51\n",
      "4800/4800 [==============================] - 1s 251us/step - loss: 0.0047\n",
      "Epoch 25/51\n",
      "4800/4800 [==============================] - 1s 258us/step - loss: 0.0049\n",
      "Epoch 26/51\n",
      "4800/4800 [==============================] - 1s 293us/step - loss: 0.0047\n",
      "Epoch 27/51\n",
      "4800/4800 [==============================] - 1s 260us/step - loss: 0.0046\n",
      "Epoch 28/51\n",
      "4800/4800 [==============================] - 1s 255us/step - loss: 0.0045\n",
      "Epoch 29/51\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 0.0045\n",
      "Epoch 30/51\n",
      "4800/4800 [==============================] - 1s 255us/step - loss: 0.0045\n",
      "Epoch 31/51\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 0.0046\n",
      "Epoch 32/51\n",
      "4800/4800 [==============================] - 1s 255us/step - loss: 0.0045\n",
      "Epoch 33/51\n",
      "4800/4800 [==============================] - 1s 265us/step - loss: 0.0045\n",
      "Epoch 34/51\n",
      "4800/4800 [==============================] - 1s 260us/step - loss: 0.0044\n",
      "Epoch 35/51\n",
      "4800/4800 [==============================] - 1s 311us/step - loss: 0.0045\n",
      "Epoch 36/51\n",
      "4800/4800 [==============================] - 2s 330us/step - loss: 0.0044\n",
      "Epoch 37/51\n",
      "4800/4800 [==============================] - 2s 327us/step - loss: 0.0043\n",
      "Epoch 38/51\n",
      "4800/4800 [==============================] - 1s 277us/step - loss: 0.0044\n",
      "Epoch 39/51\n",
      "4800/4800 [==============================] - 1s 285us/step - loss: 0.0043\n",
      "Epoch 40/51\n",
      "4800/4800 [==============================] - 1s 295us/step - loss: 0.0043\n",
      "Epoch 41/51\n",
      "4800/4800 [==============================] - 2s 329us/step - loss: 0.0042\n",
      "Epoch 42/51\n",
      "4800/4800 [==============================] - 1s 294us/step - loss: 0.0043\n",
      "Epoch 43/51\n",
      "4800/4800 [==============================] - 1s 272us/step - loss: 0.0043\n",
      "Epoch 44/51\n",
      "4800/4800 [==============================] - 1s 277us/step - loss: 0.0042\n",
      "Epoch 45/51\n",
      "4800/4800 [==============================] - 1s 270us/step - loss: 0.0042\n",
      "Epoch 46/51\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 0.0042\n",
      "Epoch 47/51\n",
      "4800/4800 [==============================] - 1s 268us/step - loss: 0.0041\n",
      "Epoch 48/51\n",
      "4800/4800 [==============================] - 1s 278us/step - loss: 0.0041\n",
      "Epoch 49/51\n",
      "4800/4800 [==============================] - 1s 267us/step - loss: 0.0042\n",
      "Epoch 50/51\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 0.0041\n",
      "Epoch 51/51\n",
      "4800/4800 [==============================] - 1s 255us/step - loss: 0.0041\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.callbacks.History at 0x7faace0e2518>"
      ]
     },
     "execution_count": 260,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "neuralNetModel.compile(optimizer = 'adam',loss = 'mean_squared_error')\n",
    "neuralNetModel.fit(short_X_train, short_y_train, batch_size = 10, epochs = 51)\n",
    "neuralNetModel.fit(long_X_train, long_y_train, batch_size = 10, epochs = 51)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 261,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2725,
     "status": "ok",
     "timestamp": 1588077861691,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "F_qSyOq1vfoC",
    "outputId": "b16b69cb-b67b-4cb6-e8f6-9366650b1ec7"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.183\n",
      "The Spearman's correlation coefficient is: 0.106\n"
     ]
    }
   ],
   "source": [
    "y_pred = neuralNetModel.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "y_pred = neuralNetModel.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 262,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 281
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3360,
     "status": "ok",
     "timestamp": 1588077867457,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "2ZYaKKJovfmA",
    "outputId": "d5941f30-9258-4e17-dbde-e3d354821b31"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(y_test, color = 'green', label = 'Actual dataset')\n",
    "plt.plot(y_pred, color = 'peachpuff', label = 'Predicted dataset')\n",
    "plt.title('Prediction')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "hvr0Mm6R9qIP"
   },
   "source": [
    "**Working on captions(tf-idf)+ HMP**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "zYK6ogf538Da"
   },
   "outputs": [],
   "source": [
    "X = dfDevCaptions_HMP_final\n",
    "y = dfGroundTruth[['short-term_memorability','long-term_memorability']].values #these are our target columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "00NoRFF69xMi"
   },
   "outputs": [],
   "source": [
    "X_train,X_test,y_train,y_test = train_test_split(X,y,test_size=0.2,random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 277,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 91
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1435,
     "status": "ok",
     "timestamp": 1588078884457,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "fwpvySOm9xKI",
    "outputId": "ea038c3d-a2f8-4ce9-b1b8-bb42e1f70cb2"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train  (4800, 1102)\n",
      "X_test   (1200, 1102)\n",
      "Y_train  (4800, 2)\n",
      "Y_test   (1200, 2)\n"
     ]
    }
   ],
   "source": [
    "print('X_train ', X_train.shape)\n",
    "print('X_test  ', X_test.shape)\n",
    "print('Y_train ', y_train.shape)\n",
    "print('Y_test  ', y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 278,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3129,
     "status": "ok",
     "timestamp": 1588078889566,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "qzfSZ3gqcsOp",
    "outputId": "03a837c9-3e0a-4cc2-e6fc-5e603f1eab75"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.369\n",
      "The Spearman's correlation coefficient is: 0.154\n"
     ]
    }
   ],
   "source": [
    "#linear regression\n",
    "linearRegressor = LinearRegression()\n",
    "linearRegressor.fit(X_train, y_train)\n",
    "y_pred = linearRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 280,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1453,
     "status": "ok",
     "timestamp": 1588078902967,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "t-fcJX2KcsYk",
    "outputId": "f9797047-4fd0-408c-a96c-91fd6e5b7ab3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.090\n",
      "The Spearman's correlation coefficient is: 0.108\n"
     ]
    }
   ],
   "source": [
    "knnRegressor = KNeighborsRegressor(n_neighbors = 77)\n",
    "knnRegressor.fit(X_test, y_test)\n",
    "y_pred = knnRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 281,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3409,
     "status": "ok",
     "timestamp": 1588078909090,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "xypgw61Ccsid",
    "outputId": "39b85c8d-3c70-440f-b72d-32ae577d7917"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.157\n",
      "The Spearman's correlation coefficient is: 0.017\n"
     ]
    }
   ],
   "source": [
    "dtRegressor = DecisionTreeRegressor()\n",
    "dtRegressor.fit(X_train, y_train)\n",
    "y_pred = dtRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 282,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 81443,
     "status": "ok",
     "timestamp": 1588078992551,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "mld37jC8csq0",
    "outputId": "869d330a-ba56-4e84-a350-4537d9f3c8a9"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.354\n",
      "The Spearman's correlation coefficient is: 0.136\n"
     ]
    }
   ],
   "source": [
    "rfRegressor = RandomForestRegressor(n_estimators=100,random_state=45)\n",
    "rfRegressor.fit(X_train,y_train)\n",
    "y_pred = rfRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "h8j_eO1i4TAR"
   },
   "source": [
    "*For other ensemble and neural networking models*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "LGW8j7_W4Tm0"
   },
   "outputs": [],
   "source": [
    "y_short = dfGroundTruth[['short-term_memorability']].values\n",
    "y_long = dfGroundTruth[['long-term_memorability']].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "stoZm_864TtQ"
   },
   "outputs": [],
   "source": [
    "# Splitting the dataset into the Training set and Test set\n",
    "short_X_train,short_X_test,short_y_train,short_y_test = train_test_split(X,y_short,test_size=0.2,random_state=40)\n",
    "long_X_train,long_X_test,long_y_train,long_y_test = train_test_split(X,y_long,test_size=0.2,random_state=40)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 286,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 149
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 165245,
     "status": "ok",
     "timestamp": 1588079169243,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "syGaWTrh4T6g",
    "outputId": "e16d3228-d884-4c58-f8ed-32682fd5497b"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_voting.py:406: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.322\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_voting.py:406: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.140\n"
     ]
    }
   ],
   "source": [
    "model1 = LinearRegression()\n",
    "model2 = DecisionTreeRegressor(random_state=1)\n",
    "model3 = KNeighborsRegressor(n_neighbors = 77)\n",
    "model4 = RandomForestRegressor(n_estimators=100,random_state=45)\n",
    "vRegressor = VotingRegressor(estimators=[('lr', model1), ('dt', model2), ('knn', model3), ('rf', model4)])\n",
    "\n",
    "vRegressor.fit(short_X_train,short_y_train)\n",
    "y_pred = vRegressor.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "vRegressor.fit(long_X_train,long_y_train)\n",
    "y_pred = vRegressor.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 287,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 149
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 161918,
     "status": "ok",
     "timestamp": 1588079181349,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "dAekSzpQ4T4Y",
    "outputId": "a8be0f6b-df2f-42b9-9a69-e33eba45ac45"
   },
   "outputs": [
    {
     "name": "stderr",
     "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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.258\n"
     ]
    },
    {
     "name": "stderr",
     "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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.089\n"
     ]
    }
   ],
   "source": [
    "abRegressor = AdaBoostRegressor()\n",
    "abRegressor.fit(short_X_train, short_y_train)\n",
    "y_pred = abRegressor.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "abRegressor = AdaBoostRegressor()\n",
    "abRegressor.fit(long_X_train, long_y_train)\n",
    "y_pred = abRegressor.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 288,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 149
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 38725,
     "status": "ok",
     "timestamp": 1588079220085,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "1cQlWwAb4T2J",
    "outputId": "a973a500-367e-4ade-9e97-541c3087d7e8"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_gb.py:1454: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.340\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_gb.py:1454: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.170\n"
     ]
    }
   ],
   "source": [
    "gbRegressor= GradientBoostingRegressor()\n",
    "\n",
    "gbRegressor.fit(short_X_train,short_y_train)\n",
    "y_pred = gbRegressor.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "gbRegressor.fit(long_X_train,long_y_train)\n",
    "y_pred = gbRegressor.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "BN7iG7NG4Tz7"
   },
   "outputs": [],
   "source": [
    "sc = StandardScaler()\n",
    "short_X_train = sc.fit_transform(short_X_train)\n",
    "short_X_test = sc.transform(short_X_test)\n",
    "\n",
    "long_X_train = sc.fit_transform(long_X_train)\n",
    "long_X_test = sc.transform(long_X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "3Pr81O-E4Txt"
   },
   "outputs": [],
   "source": [
    "# Initialising\n",
    "neuralNetModel = Sequential()\n",
    "\n",
    "# Adding the input layer and first hidden layer\n",
    "neuralNetModel.add(Dense(32, activation = 'relu', input_dim = 1102))\n",
    "\n",
    "# Adding second hidden layer\n",
    "neuralNetModel.add(Dense(units = 32, activation = 'relu'))\n",
    "\n",
    "# Adding third hidden layer\n",
    "neuralNetModel.add(Dense(units = 32, activation = 'relu'))\n",
    "\n",
    "# Adding output layer\n",
    "neuralNetModel.add(Dense(units = 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 293,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 136400,
     "status": "ok",
     "timestamp": 1588079391131,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "Bf4bLWiU4Tj8",
    "outputId": "27832922-2936-4e88-d785-c7ae0f90f2d4"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/51\n",
      "4800/4800 [==============================] - 1s 296us/step - loss: 0.0457\n",
      "Epoch 2/51\n",
      "4800/4800 [==============================] - 1s 266us/step - loss: 0.0083\n",
      "Epoch 3/51\n",
      "4800/4800 [==============================] - 1s 277us/step - loss: 0.0044\n",
      "Epoch 4/51\n",
      "4800/4800 [==============================] - 1s 266us/step - loss: 0.0033\n",
      "Epoch 5/51\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 0.0031\n",
      "Epoch 6/51\n",
      "4800/4800 [==============================] - 1s 268us/step - loss: 0.0032\n",
      "Epoch 7/51\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 0.0030\n",
      "Epoch 8/51\n",
      "4800/4800 [==============================] - 1s 264us/step - loss: 0.0030\n",
      "Epoch 9/51\n",
      "4800/4800 [==============================] - 1s 270us/step - loss: 0.0029\n",
      "Epoch 10/51\n",
      "4800/4800 [==============================] - 1s 273us/step - loss: 0.0028\n",
      "Epoch 11/51\n",
      "4800/4800 [==============================] - 1s 308us/step - loss: 0.0026\n",
      "Epoch 12/51\n",
      "4800/4800 [==============================] - 1s 295us/step - loss: 0.0026\n",
      "Epoch 13/51\n",
      "4800/4800 [==============================] - 1s 283us/step - loss: 0.0025\n",
      "Epoch 14/51\n",
      "4800/4800 [==============================] - 1s 282us/step - loss: 0.0023\n",
      "Epoch 15/51\n",
      "4800/4800 [==============================] - 1s 274us/step - loss: 0.0022\n",
      "Epoch 16/51\n",
      "4800/4800 [==============================] - 1s 265us/step - loss: 0.0021\n",
      "Epoch 17/51\n",
      "4800/4800 [==============================] - 1s 271us/step - loss: 0.0021\n",
      "Epoch 18/51\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 0.0020\n",
      "Epoch 19/51\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 0.0018\n",
      "Epoch 20/51\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 0.0016\n",
      "Epoch 21/51\n",
      "4800/4800 [==============================] - 1s 271us/step - loss: 0.0017\n",
      "Epoch 22/51\n",
      "4800/4800 [==============================] - 1s 270us/step - loss: 0.0018\n",
      "Epoch 23/51\n",
      "4800/4800 [==============================] - 1s 290us/step - loss: 0.0015\n",
      "Epoch 24/51\n",
      "4800/4800 [==============================] - 1s 276us/step - loss: 0.0014\n",
      "Epoch 25/51\n",
      "4800/4800 [==============================] - 1s 269us/step - loss: 0.0014\n",
      "Epoch 26/51\n",
      "4800/4800 [==============================] - 1s 273us/step - loss: 0.0013\n",
      "Epoch 27/51\n",
      "4800/4800 [==============================] - 1s 290us/step - loss: 0.0013\n",
      "Epoch 28/51\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 0.0013\n",
      "Epoch 29/51\n",
      "4800/4800 [==============================] - 1s 271us/step - loss: 0.0012\n",
      "Epoch 30/51\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 0.0012\n",
      "Epoch 31/51\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 0.0011\n",
      "Epoch 32/51\n",
      "4800/4800 [==============================] - 1s 266us/step - loss: 0.0011\n",
      "Epoch 33/51\n",
      "4800/4800 [==============================] - 1s 257us/step - loss: 0.0011\n",
      "Epoch 34/51\n",
      "4800/4800 [==============================] - 1s 271us/step - loss: 0.0011\n",
      "Epoch 35/51\n",
      "4800/4800 [==============================] - 1s 270us/step - loss: 9.6682e-04\n",
      "Epoch 36/51\n",
      "4800/4800 [==============================] - 1s 264us/step - loss: 9.5893e-04\n",
      "Epoch 37/51\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 9.3571e-04\n",
      "Epoch 38/51\n",
      "4800/4800 [==============================] - 1s 271us/step - loss: 9.1873e-04\n",
      "Epoch 39/51\n",
      "4800/4800 [==============================] - 1s 271us/step - loss: 9.0985e-04\n",
      "Epoch 40/51\n",
      "4800/4800 [==============================] - 1s 274us/step - loss: 9.1419e-04\n",
      "Epoch 41/51\n",
      "4800/4800 [==============================] - 1s 276us/step - loss: 8.5550e-04\n",
      "Epoch 42/51\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 7.8167e-04\n",
      "Epoch 43/51\n",
      "4800/4800 [==============================] - 1s 278us/step - loss: 8.2782e-04\n",
      "Epoch 44/51\n",
      "4800/4800 [==============================] - 1s 273us/step - loss: 7.4075e-04\n",
      "Epoch 45/51\n",
      "4800/4800 [==============================] - 1s 300us/step - loss: 8.1310e-04\n",
      "Epoch 46/51\n",
      "4800/4800 [==============================] - 1s 266us/step - loss: 7.5892e-04\n",
      "Epoch 47/51\n",
      "4800/4800 [==============================] - 1s 307us/step - loss: 7.9156e-04\n",
      "Epoch 48/51\n",
      "4800/4800 [==============================] - 1s 265us/step - loss: 7.2566e-04\n",
      "Epoch 49/51\n",
      "4800/4800 [==============================] - 1s 273us/step - loss: 6.7891e-04\n",
      "Epoch 50/51\n",
      "4800/4800 [==============================] - 2s 317us/step - loss: 6.4939e-04\n",
      "Epoch 51/51\n",
      "4800/4800 [==============================] - 1s 297us/step - loss: 6.6815e-04\n",
      "Epoch 1/51\n",
      "4800/4800 [==============================] - 1s 309us/step - loss: 0.0203\n",
      "Epoch 2/51\n",
      "4800/4800 [==============================] - 1s 265us/step - loss: 0.0156\n",
      "Epoch 3/51\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 0.0119\n",
      "Epoch 4/51\n",
      "4800/4800 [==============================] - 1s 268us/step - loss: 0.0092\n",
      "Epoch 5/51\n",
      "4800/4800 [==============================] - 1s 281us/step - loss: 0.0074\n",
      "Epoch 6/51\n",
      "4800/4800 [==============================] - 1s 273us/step - loss: 0.0061\n",
      "Epoch 7/51\n",
      "4800/4800 [==============================] - 1s 266us/step - loss: 0.0054\n",
      "Epoch 8/51\n",
      "4800/4800 [==============================] - 1s 265us/step - loss: 0.0047\n",
      "Epoch 9/51\n",
      "4800/4800 [==============================] - 1s 265us/step - loss: 0.0043\n",
      "Epoch 10/51\n",
      "4800/4800 [==============================] - 1s 267us/step - loss: 0.0038\n",
      "Epoch 11/51\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 0.0036\n",
      "Epoch 12/51\n",
      "4800/4800 [==============================] - 1s 265us/step - loss: 0.0034\n",
      "Epoch 13/51\n",
      "4800/4800 [==============================] - 1s 272us/step - loss: 0.0032\n",
      "Epoch 14/51\n",
      "4800/4800 [==============================] - 1s 310us/step - loss: 0.0031\n",
      "Epoch 15/51\n",
      "4800/4800 [==============================] - 1s 264us/step - loss: 0.0030\n",
      "Epoch 16/51\n",
      "4800/4800 [==============================] - 1s 270us/step - loss: 0.0028\n",
      "Epoch 17/51\n",
      "4800/4800 [==============================] - 1s 274us/step - loss: 0.0029\n",
      "Epoch 18/51\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 0.0028\n",
      "Epoch 19/51\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 0.0026\n",
      "Epoch 20/51\n",
      "4800/4800 [==============================] - 1s 310us/step - loss: 0.0025\n",
      "Epoch 21/51\n",
      "4800/4800 [==============================] - 1s 264us/step - loss: 0.0025\n",
      "Epoch 22/51\n",
      "4800/4800 [==============================] - 2s 330us/step - loss: 0.0024\n",
      "Epoch 23/51\n",
      "4800/4800 [==============================] - 1s 266us/step - loss: 0.0024\n",
      "Epoch 24/51\n",
      "4800/4800 [==============================] - 1s 295us/step - loss: 0.0023\n",
      "Epoch 25/51\n",
      "4800/4800 [==============================] - 1s 282us/step - loss: 0.0022\n",
      "Epoch 26/51\n",
      "4800/4800 [==============================] - 1s 287us/step - loss: 0.0022\n",
      "Epoch 27/51\n",
      "4800/4800 [==============================] - 1s 264us/step - loss: 0.0021\n",
      "Epoch 28/51\n",
      "4800/4800 [==============================] - 1s 272us/step - loss: 0.0022\n",
      "Epoch 29/51\n",
      "4800/4800 [==============================] - 1s 277us/step - loss: 0.0022\n",
      "Epoch 30/51\n",
      "4800/4800 [==============================] - 1s 274us/step - loss: 0.0021\n",
      "Epoch 31/51\n",
      "4800/4800 [==============================] - 1s 266us/step - loss: 0.0019\n",
      "Epoch 32/51\n",
      "4800/4800 [==============================] - 1s 274us/step - loss: 0.0021\n",
      "Epoch 33/51\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 0.0020\n",
      "Epoch 34/51\n",
      "4800/4800 [==============================] - 1s 266us/step - loss: 0.0018\n",
      "Epoch 35/51\n",
      "4800/4800 [==============================] - 1s 265us/step - loss: 0.0018\n",
      "Epoch 36/51\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 0.0019\n",
      "Epoch 37/51\n",
      "4800/4800 [==============================] - 1s 290us/step - loss: 0.0018\n",
      "Epoch 38/51\n",
      "4800/4800 [==============================] - 1s 264us/step - loss: 0.0018\n",
      "Epoch 39/51\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 0.0017\n",
      "Epoch 40/51\n",
      "4800/4800 [==============================] - 1s 284us/step - loss: 0.0017\n",
      "Epoch 41/51\n",
      "4800/4800 [==============================] - 1s 280us/step - loss: 0.0017\n",
      "Epoch 42/51\n",
      "4800/4800 [==============================] - 1s 269us/step - loss: 0.0017\n",
      "Epoch 43/51\n",
      "4800/4800 [==============================] - 1s 261us/step - loss: 0.0017\n",
      "Epoch 44/51\n",
      "4800/4800 [==============================] - 1s 270us/step - loss: 0.0016\n",
      "Epoch 45/51\n",
      "4800/4800 [==============================] - 1s 274us/step - loss: 0.0016\n",
      "Epoch 46/51\n",
      "4800/4800 [==============================] - 1s 284us/step - loss: 0.0016\n",
      "Epoch 47/51\n",
      "4800/4800 [==============================] - 1s 290us/step - loss: 0.0015\n",
      "Epoch 48/51\n",
      "4800/4800 [==============================] - 1s 298us/step - loss: 0.0016\n",
      "Epoch 49/51\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 0.0016\n",
      "Epoch 50/51\n",
      "4800/4800 [==============================] - 1s 259us/step - loss: 0.0016\n",
      "Epoch 51/51\n",
      "4800/4800 [==============================] - 1s 258us/step - loss: 0.0016\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.callbacks.History at 0x7fab45173ef0>"
      ]
     },
     "execution_count": 293,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "neuralNetModel.compile(optimizer = 'adam',loss = 'mean_squared_error')\n",
    "neuralNetModel.fit(short_X_train, short_y_train, batch_size = 10, epochs = 51)\n",
    "neuralNetModel.fit(long_X_train, long_y_train, batch_size = 10, epochs = 51)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 294,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 960,
     "status": "ok",
     "timestamp": 1588079392153,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "FPsPcseg44OG",
    "outputId": "328c9854-1b67-4ab6-d607-bce567702323"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.140\n",
      "The Spearman's correlation coefficient is: 0.103\n"
     ]
    }
   ],
   "source": [
    "y_pred = neuralNetModel.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "y_pred = neuralNetModel.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 295,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 281
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2010,
     "status": "ok",
     "timestamp": 1588079393230,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "_OFb5hrj44cn",
    "outputId": "9a24093f-59f7-4fef-d7f8-a4304a530af9"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(y_test, color = 'green', label = 'Actual dataset')\n",
    "plt.plot(y_pred, color = 'peachpuff', label = 'Predicted dataset')\n",
    "plt.title('Prediction')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "ZdXu4lLVEsLJ"
   },
   "source": [
    "**Working on captions(tf-idf)+ C3D**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "BANxpAm28ms1"
   },
   "outputs": [],
   "source": [
    "X = dfDevCaptions_C3D_final\n",
    "y = dfGroundTruth[['short-term_memorability','long-term_memorability']].values #these are our target columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "tN4pa8dUdqA2"
   },
   "outputs": [],
   "source": [
    "X_train,X_test,y_train,y_test = train_test_split(X,y,test_size=0.2,random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 304,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 91
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1003,
     "status": "ok",
     "timestamp": 1588080074393,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "Zakr08OodqUq",
    "outputId": "a612a94e-6812-4aeb-e066-87c02e8a13a0"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train  (4800, 1101)\n",
      "X_test   (1200, 1101)\n",
      "Y_train  (4800, 2)\n",
      "Y_test   (1200, 2)\n"
     ]
    }
   ],
   "source": [
    "print('X_train ', X_train.shape)\n",
    "print('X_test  ', X_test.shape)\n",
    "print('Y_train ', y_train.shape)\n",
    "print('Y_test  ', y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 305,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2924,
     "status": "ok",
     "timestamp": 1588080079784,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "Z5hjliDbdqSC",
    "outputId": "8dfeeee6-daa1-47a0-d60a-80ffc1f2e01e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.370\n",
      "The Spearman's correlation coefficient is: 0.154\n"
     ]
    }
   ],
   "source": [
    "#linear regression\n",
    "linearRegressor = LinearRegression()\n",
    "linearRegressor.fit(X_train, y_train)\n",
    "y_pred = linearRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 307,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 4577,
     "status": "ok",
     "timestamp": 1588080100877,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "sd1lII6RdqP1",
    "outputId": "8a65f380-3aee-4a5b-a369-8ff5e73adc46"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.354\n",
      "The Spearman's correlation coefficient is: 0.145\n"
     ]
    }
   ],
   "source": [
    "knnRegressor = KNeighborsRegressor(n_neighbors = 77)\n",
    "knnRegressor.fit(X_test, y_test)\n",
    "y_pred = knnRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 308,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3493,
     "status": "ok",
     "timestamp": 1588080112220,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "wHfAZ0EKdqNn",
    "outputId": "0e468096-4a83-4637-e167-3b36f550ced0"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.122\n",
      "The Spearman's correlation coefficient is: 0.029\n"
     ]
    }
   ],
   "source": [
    "dtRegressor = DecisionTreeRegressor()\n",
    "dtRegressor.fit(X_train, y_train)\n",
    "y_pred = dtRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 309,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 84105,
     "status": "ok",
     "timestamp": 1588080199537,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "SK12Bi0CdqLB",
    "outputId": "ca780ac8-76ec-4f2a-d6a7-b804a324f5e3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.351\n",
      "The Spearman's correlation coefficient is: 0.140\n"
     ]
    }
   ],
   "source": [
    "rfRegressor = RandomForestRegressor(n_estimators=100,random_state=45)\n",
    "rfRegressor.fit(X_train,y_train)\n",
    "y_pred = rfRegressor.predict(X_test)\n",
    "Get_score(y_pred, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "xW_M5qYl88DP"
   },
   "source": [
    "*For other ensemble and neural networking models*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "aeeZzHY6861S"
   },
   "outputs": [],
   "source": [
    "y_short = dfGroundTruth[['short-term_memorability']].values\n",
    "y_long = dfGroundTruth[['long-term_memorability']].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "cdq1nwJDdqIv"
   },
   "outputs": [],
   "source": [
    "# Splitting the dataset into the Training set and Test set\n",
    "short_X_train,short_X_test,short_y_train,short_y_test = train_test_split(X,y_short,test_size=0.2,random_state=40)\n",
    "long_X_train,long_X_test,long_y_train,long_y_test = train_test_split(X,y_long,test_size=0.2,random_state=40)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 312,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 149
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 193007,
     "status": "ok",
     "timestamp": 1588080392608,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "lGvQdTpadqGT",
    "outputId": "b99447c3-118a-44d8-edac-6fe52d3c311a"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_voting.py:406: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.348\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_voting.py:406: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.176\n"
     ]
    }
   ],
   "source": [
    "model1 = LinearRegression()\n",
    "model2 = DecisionTreeRegressor(random_state=1)\n",
    "model3 = KNeighborsRegressor(n_neighbors = 77)\n",
    "model4 = RandomForestRegressor(n_estimators=100,random_state=45)\n",
    "vRegressor = VotingRegressor(estimators=[('lr', model1), ('dt', model2), ('knn', model3), ('rf', model4)])\n",
    "\n",
    "vRegressor.fit(short_X_train,short_y_train)\n",
    "y_pred = vRegressor.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "vRegressor.fit(long_X_train,long_y_train)\n",
    "y_pred = vRegressor.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 313,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 149
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 206222,
     "status": "ok",
     "timestamp": 1588080405836,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "KdV-iDpT9LM0",
    "outputId": "dc39985a-aa21-45d6-f21a-38675b462bda"
   },
   "outputs": [
    {
     "name": "stderr",
     "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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.244\n"
     ]
    },
    {
     "name": "stderr",
     "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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.098\n"
     ]
    }
   ],
   "source": [
    "abRegressor = AdaBoostRegressor()\n",
    "abRegressor.fit(short_X_train, short_y_train)\n",
    "y_pred = abRegressor.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "abRegressor = AdaBoostRegressor()\n",
    "abRegressor.fit(long_X_train, long_y_train)\n",
    "y_pred = abRegressor.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 314,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 149
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 238335,
     "status": "ok",
     "timestamp": 1588080442662,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "h3M_RgAo9K4X",
    "outputId": "7fc923b1-e369-4fe5-e716-c013e7b636df"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_gb.py:1454: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.344\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/ensemble/_gb.py:1454: 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": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.171\n"
     ]
    }
   ],
   "source": [
    "gbRegressor= GradientBoostingRegressor()\n",
    "\n",
    "gbRegressor.fit(short_X_train,short_y_train)\n",
    "y_pred = gbRegressor.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "gbRegressor.fit(long_X_train,long_y_train)\n",
    "y_pred = gbRegressor.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "1BJ3pz2V9LKu"
   },
   "outputs": [],
   "source": [
    "sc = StandardScaler()\n",
    "short_X_train = sc.fit_transform(short_X_train)\n",
    "short_X_test = sc.transform(short_X_test)\n",
    "\n",
    "long_X_train = sc.fit_transform(long_X_train)\n",
    "long_X_test = sc.transform(long_X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "Fy8T7n7W9LH4"
   },
   "outputs": [],
   "source": [
    "# Initialising\n",
    "neuralNetModel = Sequential()\n",
    "\n",
    "# Adding the input layer and first hidden layer\n",
    "neuralNetModel.add(Dense(32, activation = 'relu', input_dim = 1101))\n",
    "\n",
    "# Adding second hidden layer\n",
    "neuralNetModel.add(Dense(units = 32, activation = 'relu'))\n",
    "\n",
    "# Adding third hidden layer\n",
    "neuralNetModel.add(Dense(units = 32, activation = 'relu'))\n",
    "\n",
    "# Adding output layer\n",
    "neuralNetModel.add(Dense(units = 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 319,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 142549,
     "status": "ok",
     "timestamp": 1588080601275,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "mnBpynk_9LFt",
    "outputId": "459ebb4c-941d-44be-a8b1-4eddb12e7b15"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/51\n",
      "4800/4800 [==============================] - 2s 340us/step - loss: 0.1072\n",
      "Epoch 2/51\n",
      "4800/4800 [==============================] - 1s 287us/step - loss: 0.0163\n",
      "Epoch 3/51\n",
      "4800/4800 [==============================] - 1s 289us/step - loss: 0.0075\n",
      "Epoch 4/51\n",
      "4800/4800 [==============================] - 1s 293us/step - loss: 0.0051\n",
      "Epoch 5/51\n",
      "4800/4800 [==============================] - 1s 278us/step - loss: 0.0048\n",
      "Epoch 6/51\n",
      "4800/4800 [==============================] - 1s 277us/step - loss: 0.0039\n",
      "Epoch 7/51\n",
      "4800/4800 [==============================] - 1s 284us/step - loss: 0.0034\n",
      "Epoch 8/51\n",
      "4800/4800 [==============================] - 1s 298us/step - loss: 0.0033\n",
      "Epoch 9/51\n",
      "4800/4800 [==============================] - 1s 284us/step - loss: 0.0037\n",
      "Epoch 10/51\n",
      "4800/4800 [==============================] - 2s 339us/step - loss: 0.0033\n",
      "Epoch 11/51\n",
      "4800/4800 [==============================] - 1s 284us/step - loss: 0.0031\n",
      "Epoch 12/51\n",
      "4800/4800 [==============================] - 1s 278us/step - loss: 0.0032\n",
      "Epoch 13/51\n",
      "4800/4800 [==============================] - 1s 284us/step - loss: 0.0033\n",
      "Epoch 14/51\n",
      "4800/4800 [==============================] - 1s 281us/step - loss: 0.0030\n",
      "Epoch 15/51\n",
      "4800/4800 [==============================] - 1s 278us/step - loss: 0.0028\n",
      "Epoch 16/51\n",
      "4800/4800 [==============================] - 1s 306us/step - loss: 0.0026\n",
      "Epoch 17/51\n",
      "4800/4800 [==============================] - 1s 279us/step - loss: 0.0025\n",
      "Epoch 18/51\n",
      "4800/4800 [==============================] - 1s 282us/step - loss: 0.0024\n",
      "Epoch 19/51\n",
      "4800/4800 [==============================] - 1s 287us/step - loss: 0.0023\n",
      "Epoch 20/51\n",
      "4800/4800 [==============================] - 2s 316us/step - loss: 0.0022\n",
      "Epoch 21/51\n",
      "4800/4800 [==============================] - 1s 277us/step - loss: 0.0022\n",
      "Epoch 22/51\n",
      "4800/4800 [==============================] - 1s 272us/step - loss: 0.0022\n",
      "Epoch 23/51\n",
      "4800/4800 [==============================] - 1s 277us/step - loss: 0.0020\n",
      "Epoch 24/51\n",
      "4800/4800 [==============================] - 1s 287us/step - loss: 0.0020\n",
      "Epoch 25/51\n",
      "4800/4800 [==============================] - 1s 297us/step - loss: 0.0018\n",
      "Epoch 26/51\n",
      "4800/4800 [==============================] - 1s 282us/step - loss: 0.0018\n",
      "Epoch 27/51\n",
      "4800/4800 [==============================] - 1s 298us/step - loss: 0.0016\n",
      "Epoch 28/51\n",
      "4800/4800 [==============================] - 1s 304us/step - loss: 0.0016\n",
      "Epoch 29/51\n",
      "4800/4800 [==============================] - 1s 281us/step - loss: 0.0014\n",
      "Epoch 30/51\n",
      "4800/4800 [==============================] - 1s 306us/step - loss: 0.0015\n",
      "Epoch 31/51\n",
      "4800/4800 [==============================] - 1s 304us/step - loss: 0.0014\n",
      "Epoch 32/51\n",
      "4800/4800 [==============================] - 2s 340us/step - loss: 0.0013\n",
      "Epoch 33/51\n",
      "4800/4800 [==============================] - 1s 297us/step - loss: 0.0014\n",
      "Epoch 34/51\n",
      "4800/4800 [==============================] - 1s 281us/step - loss: 0.0014\n",
      "Epoch 35/51\n",
      "4800/4800 [==============================] - 1s 269us/step - loss: 0.0012\n",
      "Epoch 36/51\n",
      "4800/4800 [==============================] - 1s 279us/step - loss: 0.0012\n",
      "Epoch 37/51\n",
      "4800/4800 [==============================] - 1s 306us/step - loss: 0.0012\n",
      "Epoch 38/51\n",
      "4800/4800 [==============================] - 1s 276us/step - loss: 0.0012\n",
      "Epoch 39/51\n",
      "4800/4800 [==============================] - 1s 286us/step - loss: 0.0011\n",
      "Epoch 40/51\n",
      "4800/4800 [==============================] - 1s 283us/step - loss: 0.0011\n",
      "Epoch 41/51\n",
      "4800/4800 [==============================] - 1s 288us/step - loss: 0.0011\n",
      "Epoch 42/51\n",
      "4800/4800 [==============================] - 2s 314us/step - loss: 9.9999e-04\n",
      "Epoch 43/51\n",
      "4800/4800 [==============================] - 1s 285us/step - loss: 0.0010\n",
      "Epoch 44/51\n",
      "4800/4800 [==============================] - 1s 277us/step - loss: 9.3435e-04\n",
      "Epoch 45/51\n",
      "4800/4800 [==============================] - 1s 298us/step - loss: 8.8729e-04\n",
      "Epoch 46/51\n",
      "4800/4800 [==============================] - 1s 282us/step - loss: 9.7306e-04\n",
      "Epoch 47/51\n",
      "4800/4800 [==============================] - 1s 283us/step - loss: 9.4610e-04\n",
      "Epoch 48/51\n",
      "4800/4800 [==============================] - 1s 278us/step - loss: 8.7838e-04\n",
      "Epoch 49/51\n",
      "4800/4800 [==============================] - 1s 281us/step - loss: 7.9874e-04\n",
      "Epoch 50/51\n",
      "4800/4800 [==============================] - 1s 284us/step - loss: 8.5745e-04\n",
      "Epoch 51/51\n",
      "4800/4800 [==============================] - 1s 280us/step - loss: 8.5921e-04\n",
      "Epoch 1/51\n",
      "4800/4800 [==============================] - 1s 289us/step - loss: 0.0206\n",
      "Epoch 2/51\n",
      "4800/4800 [==============================] - 1s 297us/step - loss: 0.0162\n",
      "Epoch 3/51\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 0.0124\n",
      "Epoch 4/51\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 0.0097\n",
      "Epoch 5/51\n",
      "4800/4800 [==============================] - 1s 304us/step - loss: 0.0079\n",
      "Epoch 6/51\n",
      "4800/4800 [==============================] - 1s 268us/step - loss: 0.0066\n",
      "Epoch 7/51\n",
      "4800/4800 [==============================] - 1s 268us/step - loss: 0.0058\n",
      "Epoch 8/51\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 0.0055\n",
      "Epoch 9/51\n",
      "4800/4800 [==============================] - 1s 268us/step - loss: 0.0050\n",
      "Epoch 10/51\n",
      "4800/4800 [==============================] - 1s 268us/step - loss: 0.0048\n",
      "Epoch 11/51\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 0.0046\n",
      "Epoch 12/51\n",
      "4800/4800 [==============================] - 1s 262us/step - loss: 0.0040\n",
      "Epoch 13/51\n",
      "4800/4800 [==============================] - 1s 270us/step - loss: 0.0044\n",
      "Epoch 14/51\n",
      "4800/4800 [==============================] - 1s 269us/step - loss: 0.0041\n",
      "Epoch 15/51\n",
      "4800/4800 [==============================] - 1s 290us/step - loss: 0.0040\n",
      "Epoch 16/51\n",
      "4800/4800 [==============================] - 1s 295us/step - loss: 0.0036\n",
      "Epoch 17/51\n",
      "4800/4800 [==============================] - 1s 308us/step - loss: 0.0040\n",
      "Epoch 18/51\n",
      "4800/4800 [==============================] - 1s 264us/step - loss: 0.0036\n",
      "Epoch 19/51\n",
      "4800/4800 [==============================] - 1s 264us/step - loss: 0.0034\n",
      "Epoch 20/51\n",
      "4800/4800 [==============================] - 1s 312us/step - loss: 0.0032\n",
      "Epoch 21/51\n",
      "4800/4800 [==============================] - 1s 300us/step - loss: 0.0031\n",
      "Epoch 22/51\n",
      "4800/4800 [==============================] - 1s 270us/step - loss: 0.0032\n",
      "Epoch 23/51\n",
      "4800/4800 [==============================] - 1s 264us/step - loss: 0.0030\n",
      "Epoch 24/51\n",
      "4800/4800 [==============================] - 1s 292us/step - loss: 0.0029\n",
      "Epoch 25/51\n",
      "4800/4800 [==============================] - 1s 274us/step - loss: 0.0028\n",
      "Epoch 26/51\n",
      "4800/4800 [==============================] - 1s 279us/step - loss: 0.0028\n",
      "Epoch 27/51\n",
      "4800/4800 [==============================] - 1s 307us/step - loss: 0.0028\n",
      "Epoch 28/51\n",
      "4800/4800 [==============================] - 1s 288us/step - loss: 0.0028\n",
      "Epoch 29/51\n",
      "4800/4800 [==============================] - 1s 288us/step - loss: 0.0028\n",
      "Epoch 30/51\n",
      "4800/4800 [==============================] - 1s 296us/step - loss: 0.0028\n",
      "Epoch 31/51\n",
      "4800/4800 [==============================] - 1s 269us/step - loss: 0.0026\n",
      "Epoch 32/51\n",
      "4800/4800 [==============================] - 1s 283us/step - loss: 0.0025\n",
      "Epoch 33/51\n",
      "4800/4800 [==============================] - 1s 305us/step - loss: 0.0023\n",
      "Epoch 34/51\n",
      "4800/4800 [==============================] - 1s 306us/step - loss: 0.0024\n",
      "Epoch 35/51\n",
      "4800/4800 [==============================] - 1s 263us/step - loss: 0.0024\n",
      "Epoch 36/51\n",
      "4800/4800 [==============================] - 1s 272us/step - loss: 0.0022\n",
      "Epoch 37/51\n",
      "4800/4800 [==============================] - 1s 266us/step - loss: 0.0022\n",
      "Epoch 38/51\n",
      "4800/4800 [==============================] - 1s 273us/step - loss: 0.0023\n",
      "Epoch 39/51\n",
      "4800/4800 [==============================] - 1s 275us/step - loss: 0.0023\n",
      "Epoch 40/51\n",
      "4800/4800 [==============================] - 1s 266us/step - loss: 0.0025\n",
      "Epoch 41/51\n",
      "4800/4800 [==============================] - 1s 270us/step - loss: 0.0023\n",
      "Epoch 42/51\n",
      "4800/4800 [==============================] - 1s 271us/step - loss: 0.0022\n",
      "Epoch 43/51\n",
      "4800/4800 [==============================] - 1s 280us/step - loss: 0.0022\n",
      "Epoch 44/51\n",
      "4800/4800 [==============================] - 1s 292us/step - loss: 0.0021\n",
      "Epoch 45/51\n",
      "4800/4800 [==============================] - 1s 297us/step - loss: 0.0021\n",
      "Epoch 46/51\n",
      "4800/4800 [==============================] - 1s 270us/step - loss: 0.0021\n",
      "Epoch 47/51\n",
      "4800/4800 [==============================] - 1s 275us/step - loss: 0.0020\n",
      "Epoch 48/51\n",
      "4800/4800 [==============================] - 1s 274us/step - loss: 0.0020\n",
      "Epoch 49/51\n",
      "4800/4800 [==============================] - 1s 295us/step - loss: 0.0020\n",
      "Epoch 50/51\n",
      "4800/4800 [==============================] - 1s 266us/step - loss: 0.0020\n",
      "Epoch 51/51\n",
      "4800/4800 [==============================] - 1s 285us/step - loss: 0.0019\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.callbacks.History at 0x7faacdccaf28>"
      ]
     },
     "execution_count": 319,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "neuralNetModel.compile(optimizer = 'adam',loss = 'mean_squared_error')\n",
    "neuralNetModel.fit(short_X_train, short_y_train, batch_size = 10, epochs = 51)\n",
    "neuralNetModel.fit(long_X_train, long_y_train, batch_size = 10, epochs = 51)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 320,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 140081,
     "status": "ok",
     "timestamp": 1588080601283,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "KrrOwYzB9WYO",
    "outputId": "bb677397-6a00-4b44-f3ea-98924401be49"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Spearman's correlation coefficient is: 0.193\n",
      "The Spearman's correlation coefficient is: 0.144\n"
     ]
    }
   ],
   "source": [
    "y_pred = neuralNetModel.predict(short_X_test)\n",
    "Get_score(y_pred, short_y_test)\n",
    "\n",
    "y_pred = neuralNetModel.predict(long_X_test)\n",
    "Get_score(y_pred, long_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 321,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 281
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 138618,
     "status": "ok",
     "timestamp": 1588080601290,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "wRu_3w5f9Wdt",
    "outputId": "8303fa9e-5ed9-4cda-e647-fa2e3196845a"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(y_test, color = 'green', label = 'Actual dataset')\n",
    "plt.plot(y_pred, color = 'peachpuff', label = 'Predicted dataset')\n",
    "plt.title('Prediction')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "2OA8b0jNpQdo"
   },
   "source": [
    "## **Prediction of Test Results**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "7iKE6KQqxzza"
   },
   "source": [
    "We have considered whole captions(tf-idf) dev set as X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "ze-tYvSO9LDj"
   },
   "outputs": [],
   "source": [
    "X = dfCaptions_Dev_tfIdf\n",
    "y = dfGroundTruth[['short-term_memorability','long-term_memorability']].values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "A40XVYxoyA9q"
   },
   "source": [
    "loading the test set data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "cLR1x3Cg9K_h"
   },
   "outputs": [],
   "source": [
    "gtPath ='./Test-set/Ground-truth_test/'\n",
    "dfGroundTruth_test = pd.read_csv(gtPath+'ground_truth_template.csv')\n",
    "\n",
    "cap_path = './Test-set/Captions_test/test-set-1_video-captions.txt'\n",
    "df_captions_Test=read_caps(cap_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 144
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2597,
     "status": "ok",
     "timestamp": 1588092285613,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "rfNhIZ7nH_rH",
    "outputId": "7dbd262f-ecf8-4940-f2f2-c4224ed3f41c"
   },
   "outputs": [
    {
     "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>video</th>\n",
       "      <th>caption</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>video7494.webm</td>\n",
       "      <td>green-jeep-struggling-to-drive-over-huge-rocks</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>video7495.webm</td>\n",
       "      <td>hiking-woman-tourist-is-walking-forward-in-mou...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>video7496.webm</td>\n",
       "      <td>close-up-of-african-american-doctors-hands-usi...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            video                                            caption\n",
       "0  video7494.webm     green-jeep-struggling-to-drive-over-huge-rocks\n",
       "1  video7495.webm  hiking-woman-tourist-is-walking-forward-in-mou...\n",
       "2  video7496.webm  close-up-of-african-american-doctors-hands-usi..."
      ]
     },
     "execution_count": 54,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_captions_Test.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 144
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3639,
     "status": "ok",
     "timestamp": 1588092292241,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "LvJ4iZe4ID-U",
    "outputId": "8a1873fa-9a29-492d-9d52-6ee8bfa4aa92"
   },
   "outputs": [
    {
     "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>video</th>\n",
       "      <th>short-term_memorability</th>\n",
       "      <th>nb_short-term_annotations</th>\n",
       "      <th>long-term_memorability</th>\n",
       "      <th>nb_long-term_annotations</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>7494</td>\n",
       "      <td>NaN</td>\n",
       "      <td>33</td>\n",
       "      <td>NaN</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>7495</td>\n",
       "      <td>NaN</td>\n",
       "      <td>34</td>\n",
       "      <td>NaN</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7496</td>\n",
       "      <td>NaN</td>\n",
       "      <td>32</td>\n",
       "      <td>NaN</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   video  ...  nb_long-term_annotations\n",
       "0   7494  ...                        12\n",
       "1   7495  ...                        10\n",
       "2   7496  ...                        13\n",
       "\n",
       "[3 rows x 5 columns]"
      ]
     },
     "execution_count": 55,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfGroundTruth_test.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1474,
     "status": "ok",
     "timestamp": 1588092297133,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "1Pdu5WUXrO9n",
    "outputId": "0b496db9-5152-4ce4-d198-ef3217f369fa"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X: (6000)\n",
      "y:(6000, 2)\n"
     ]
    }
   ],
   "source": [
    "print(f'X: ({len(X)})')\n",
    "print(f'y:{y.shape}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "5TYeG_XNIvmG"
   },
   "source": [
    "*TF-IDF on test captions*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "BfDiheorI0pJ"
   },
   "outputs": [],
   "source": [
    "vectorizer=TfidfVectorizer(max_features=1000)\n",
    "response=vectorizer.fit_transform(df_captions_Test.caption).toarray()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 129
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1533,
     "status": "ok",
     "timestamp": 1588092395923,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "LQAraBh0I4BT",
    "outputId": "a06bb701-3d11-42a7-d3d1-dc3fddfda417"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "            video  ...                                  vectorizedCaption\n",
      "0  video7494.webm  ...  [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ...\n",
      "1  video7495.webm  ...  [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ...\n",
      "2  video7496.webm  ...  [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ...\n",
      "\n",
      "[3 rows x 3 columns]\n"
     ]
    }
   ],
   "source": [
    "df_captions_Test['vectorizedCaption']=pd.Series(response.tolist())\n",
    "print(df_captions_Test.head(3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 36
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1691,
     "status": "ok",
     "timestamp": 1588092400266,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "jPeNC9p_I9WV",
    "outputId": "216c07a4-3dbf-4043-eda4-8153fb091db3"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['video', 'caption', 'vectorizedCaption']"
      ]
     },
     "execution_count": 60,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(df_captions_Test.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "_2EmOodwJALU"
   },
   "outputs": [],
   "source": [
    "df_captions_Test_tfIdf=pd.DataFrame(df_captions_Test['vectorizedCaption'].tolist())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "5IKf_Hdcry3T"
   },
   "source": [
    "Training Random forest regressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 147
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 150791,
     "status": "ok",
     "timestamp": 1588092559719,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "DIOZwy_cpv9-",
    "outputId": "e5014a65-d543-41c5-f220-bf6f1f522039"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(bootstrap=True, ccp_alpha=0.0, criterion='mse',\n",
       "                      max_depth=None, max_features='auto', max_leaf_nodes=None,\n",
       "                      max_samples=None, min_impurity_decrease=0.0,\n",
       "                      min_impurity_split=None, min_samples_leaf=1,\n",
       "                      min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
       "                      n_estimators=100, n_jobs=None, oob_score=False,\n",
       "                      random_state=45, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 62,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfRegressor = RandomForestRegressor(n_estimators=100,random_state=45)\n",
    "rfRegressor.fit(X,y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "6iRZ7b09pwC3"
   },
   "outputs": [],
   "source": [
    "testSet_pred = rfRegressor.predict(df_captions_Test_tfIdf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "5zqt7tNYpwHm"
   },
   "outputs": [],
   "source": [
    "finalPred = pd.DataFrame()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 36
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1873,
     "status": "ok",
     "timestamp": 1588092847398,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "qujy5eTOpwTd",
    "outputId": "05fefbb9-0af2-4a68-f5f7-ce7940ae044a"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "numpy.ndarray"
      ]
     },
     "execution_count": 66,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(testSet_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "JTSAVcAmpwRa"
   },
   "outputs": [],
   "source": [
    "finalPred['short-term'] = testSet_pred[:,0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "TJaCvBDnpwPi"
   },
   "outputs": [],
   "source": [
    "finalPred['long-term'] = testSet_pred[:,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3148,
     "status": "ok",
     "timestamp": 1588092858181,
     "user": {
      "displayName": "Raghvendra Pratap Singh",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gj1jQJ5n0fdqt2IyBbxgNbDiN_ZfLwM20c8-dFk=s64",
      "userId": "16108454137484926510"
     },
     "user_tz": -330
    },
    "id": "rasJC1OCpwNg",
    "outputId": "01c8a740-9d3f-4cec-fa8c-bf122dc4927d"
   },
   "outputs": [
    {
     "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>short-term</th>\n",
       "      <th>long-term</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1995</th>\n",
       "      <td>0.856382</td>\n",
       "      <td>0.751422</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1996</th>\n",
       "      <td>0.754677</td>\n",
       "      <td>0.734267</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1997</th>\n",
       "      <td>0.829638</td>\n",
       "      <td>0.761398</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998</th>\n",
       "      <td>0.855205</td>\n",
       "      <td>0.722449</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1999</th>\n",
       "      <td>0.851584</td>\n",
       "      <td>0.717439</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      short-term  long-term\n",
       "1995    0.856382   0.751422\n",
       "1996    0.754677   0.734267\n",
       "1997    0.829638   0.761398\n",
       "1998    0.855205   0.722449\n",
       "1999    0.851584   0.717439"
      ]
     },
     "execution_count": 69,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "finalPred.tail()  #tail shows the row number. We have predicted values of 2000 test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "UNQJ8swIJuEG"
   },
   "outputs": [],
   "source": [
    "finalPred.to_csv(\"/content/drive/My Drive/Raghvendra_Pratap_Singh_19210261_predictions.csv\",index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "Qkq2d9LClvJc"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "qxyKNnZxm252"
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "authorship_tag": "ABX9TyMvvm7/injax3Ci/aKHfr+Q",
   "collapsed_sections": [],
   "name": "Raghvendra_Pratap_Singh_19210261_predicting_memorability.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.9"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "4dab5ddcbe6f4292b31146b889f0e8d2": {
     "model_module": "@jupyter-widgets/controls",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_684e80f5eb634a43b34f62dea0b70fa4",
       "IPY_MODEL_e264ef5b6df940d7bd37c2147d50f21e"
      ],
      "layout": "IPY_MODEL_84b2ed2ee2ab461a9c59a25d2b15f172"
     }
    },
    "6337f6faf68b4a398ed709ff42cd000f": {
     "model_module": "@jupyter-widgets/base",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "64162473341143b9899b5f93977168f1": {
     "model_module": "@jupyter-widgets/controls",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "684e80f5eb634a43b34f62dea0b70fa4": {
     "model_module": "@jupyter-widgets/controls",
     "model_name": "IntProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "IntProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "100%",
      "description_tooltip": null,
      "layout": "IPY_MODEL_f86015d0ce8348bca4fe3e0ed24b63d7",
      "max": 6000,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_b45ee9693c074e36a0af3afdccb6713e",
      "value": 6000
     }
    },
    "84b2ed2ee2ab461a9c59a25d2b15f172": {
     "model_module": "@jupyter-widgets/base",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "b45ee9693c074e36a0af3afdccb6713e": {
     "model_module": "@jupyter-widgets/controls",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": "initial"
     }
    },
    "e264ef5b6df940d7bd37c2147d50f21e": {
     "model_module": "@jupyter-widgets/controls",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_6337f6faf68b4a398ed709ff42cd000f",
      "placeholder": "​",
      "style": "IPY_MODEL_64162473341143b9899b5f93977168f1",
      "value": " 6000/6000 [39:44&lt;00:00,  2.52it/s]"
     }
    },
    "f86015d0ce8348bca4fe3e0ed24b63d7": {
     "model_module": "@jupyter-widgets/base",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    }
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}