FacePoseEstimation_normalized.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Zk7O3adE6CNC"
   },
   "source": [
    "## Import Dataset\n",
    "**uncomment the code at the begining**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "qzTH_qVIFyv5"
   },
   "outputs": [],
   "source": [
    "!pip install mediapipe\n",
    "\n",
    "print(\"Downloading Dataset\")\n",
    "!curl http://www.cbsr.ia.ac.cn/users/xiangyuzhu/projects/3DDFA/Database/AFLW2000-3D.zip  --output AFLW2000-3D.zip\n",
    "\n",
    "!unzip AFLW2000-3D.zip"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ASdrRHio7OOf"
   },
   "source": [
    "## Importing liberary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "9q-wLO3j7OCo"
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import gc\n",
    "import glob\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from matplotlib import pyplot as plt\n",
    "\n",
    "\n",
    "import cv2\n",
    "import mediapipe as mp\n",
    "\n",
    "\n",
    "import math\n",
    "\n",
    "from sklearn.datasets import make_regression\n",
    "from sklearn.multioutput import MultiOutputRegressor\n",
    "from sklearn.svm import SVR\n",
    "from sklearn.linear_model import LinearRegression, Ridge, Lasso\n",
    "from sklearn.metrics import mean_squared_error, mean_absolute_error, r2_score\n",
    "from sklearn.ensemble import ExtraTreesRegressor\n",
    "from sklearn.model_selection import RandomizedSearchCV, GridSearchCV, train_test_split\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.decomposition import PCA\n",
    "import xgboost as xgb\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "import scipy.io as sio"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "teFxGdb_7Bcy"
   },
   "source": [
    "## Reading Data"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {
    "id": "emgnJDS75MkE"
   },
   "source": [
    "# create Data for faceMish\n",
    "\n",
    "\n",
    "**Read the Dataset and Create a facemish features to form a DataFrame and handling the missing values and empty files**\n",
    "**we tried to convert the angle to to polar and train but the result was bad so we keep it**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "id": "XsTaOe8G28t_"
   },
   "outputs": [],
   "source": [
    "def read_data(path='AFLW2000/'):\n",
    "    dataset_dict = []\n",
    "\n",
    "    files_name = glob.glob(path + '*.jpg')\n",
    "\n",
    "    for img_path in files_name:\n",
    "\n",
    "        label_path = img_path[:-3] + 'mat'\n",
    "\n",
    "        if os.path.exists(label_path) and os.path.getsize(label_path) > 0:\n",
    "            label = sio.loadmat(label_path)[\"Pose_Para\"][0][:3]\n",
    "            dataset_dict.append({\"image_name\":img_path, 'pitch':label[0], 'yaw':label[1], 'roll':label[2]})\n",
    "\n",
    "\n",
    "\n",
    "    dataset = pd.DataFrame(dataset_dict)\n",
    "\n",
    "    return dataset"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {
    "id": "RshV6sUxyHbA"
   },
   "source": [
    "**Convert the image to RGB if needed while trying the gray scale one**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "id": "2lXBuDxG-gtI"
   },
   "outputs": [],
   "source": [
    "def convert_RGB(img):\n",
    "    if img.ndim < 3:\n",
    "        img = cv2.cvtColor(img, cv2.COLOR_GRAY2RGB)\n",
    "    return img"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "R3u0tHN0yOTG"
   },
   "source": [
    "**Get the facemish and other spacified points to form the DataFrame**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "id": "kBZS5QQ5-ubR"
   },
   "outputs": [],
   "source": [
    "def get_facemish(img):\n",
    "    faceMesh_list = {}\n",
    "    \n",
    "    hight, width, _ = img.shape\n",
    "\n",
    "    mp_face_mesh = mp.solutions.face_mesh\n",
    "\n",
    "    with mp_face_mesh.FaceMesh(static_image_mode=True, max_num_faces=1,\n",
    "                                            min_detection_confidence=0.5) as face_mesh_images:\n",
    "        \n",
    "        face_mesh_results = face_mesh_images.process(img)\n",
    "\n",
    "        if face_mesh_results.multi_face_landmarks:\n",
    "\n",
    "            for face_landmarks in face_mesh_results.multi_face_landmarks:\n",
    "                for i, pt in enumerate(face_landmarks.landmark):\n",
    "                    x = pt.x\n",
    "                    y = pt.y\n",
    "                    faceMesh_list.update({f\"X {i}\":x, f\"Y {i}\" :y})\n",
    "\n",
    "    return faceMesh_list"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "zHkafhggyq_C"
   },
   "source": [
    "**Merging the 2 function in one**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "id": "y68eTDfPH1ti"
   },
   "outputs": [],
   "source": [
    "def multi_preprocessing(img):\n",
    "    img = convert_RGB(img)\n",
    "    \n",
    "    return get_facemish(img)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "tBLFd0CzyyTQ"
   },
   "source": [
    "**We call the create Dataframe function to call the multi_preprocessing and get the feature point and create the dataFrame**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "id": "GDV-OxDuXSq1"
   },
   "outputs": [],
   "source": [
    "def create_dataFrame(data_img):\n",
    "\n",
    "    dataFrame = pd.DataFrame()\n",
    "    \n",
    "    for _, img in data_img.iterrows():\n",
    "    \n",
    "        image, *labels = img\n",
    "        \n",
    "        features = multi_preprocessing(plt.imread(image))\n",
    "\n",
    "        features.update({'pitch':labels[0], 'yaw':labels[1], 'roll':labels[2], 'image_name':image})\n",
    "        features = pd.Series(features)\n",
    "\n",
    "        dataFrame = dataFrame.append(features, ignore_index=True)\n",
    "        \n",
    "    \n",
    "    return dataFrame"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {
    "id": "QKrYlCNBdX24"
   },
   "source": [
    "### Normalize the face feature point with respect to the nose posetion to make it invariant to face size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "id": "1ayIdYOfTg4_"
   },
   "outputs": [],
   "source": [
    "def normalise_nose(df):\n",
    "\n",
    "    # select columns whose names start with \"X, Y\" using filter() function\n",
    "    X_features = df.filter(regex='^X')\n",
    "\n",
    "    Y_features = df.filter(regex='^Y')\n",
    "\n",
    "\n",
    "    # apply your function on the selected columns using apply() function\n",
    "    X_normalized = X_features.apply(lambda x: x - X_features['X 1'])\n",
    "\n",
    "    Y_normalized = Y_features.apply(lambda y: y - Y_features['Y 1'])\n",
    "\n",
    "\n",
    "    # update the original DataFrame with the modified columns\n",
    "    df[X_normalized.columns] = X_normalized\n",
    "\n",
    "    df[Y_normalized.columns] = Y_normalized\n",
    "\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "id": "5-wPTSqO4lV-"
   },
   "outputs": [],
   "source": [
    "df  = read_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "id": "jCcws249EtEE"
   },
   "outputs": [],
   "source": [
    "train_df = create_dataFrame(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "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>X 0</th>\n",
       "      <th>Y 0</th>\n",
       "      <th>X 1</th>\n",
       "      <th>Y 1</th>\n",
       "      <th>X 2</th>\n",
       "      <th>Y 2</th>\n",
       "      <th>X 3</th>\n",
       "      <th>Y 3</th>\n",
       "      <th>X 4</th>\n",
       "      <th>Y 4</th>\n",
       "      <th>...</th>\n",
       "      <th>X 465</th>\n",
       "      <th>Y 465</th>\n",
       "      <th>X 466</th>\n",
       "      <th>Y 466</th>\n",
       "      <th>X 467</th>\n",
       "      <th>Y 467</th>\n",
       "      <th>pitch</th>\n",
       "      <th>yaw</th>\n",
       "      <th>roll</th>\n",
       "      <th>image_name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.485699</td>\n",
       "      <td>0.687087</td>\n",
       "      <td>0.489150</td>\n",
       "      <td>0.639948</td>\n",
       "      <td>0.489853</td>\n",
       "      <td>0.647495</td>\n",
       "      <td>0.479756</td>\n",
       "      <td>0.562636</td>\n",
       "      <td>0.490495</td>\n",
       "      <td>0.621423</td>\n",
       "      <td>...</td>\n",
       "      <td>0.541506</td>\n",
       "      <td>0.500291</td>\n",
       "      <td>0.665707</td>\n",
       "      <td>0.471219</td>\n",
       "      <td>0.676151</td>\n",
       "      <td>0.463839</td>\n",
       "      <td>-0.399231</td>\n",
       "      <td>0.018227</td>\n",
       "      <td>0.085676</td>\n",
       "      <td>AFLW2000\\image00002.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.442459</td>\n",
       "      <td>0.642504</td>\n",
       "      <td>0.417327</td>\n",
       "      <td>0.592812</td>\n",
       "      <td>0.449168</td>\n",
       "      <td>0.611022</td>\n",
       "      <td>0.435618</td>\n",
       "      <td>0.544955</td>\n",
       "      <td>0.415644</td>\n",
       "      <td>0.577393</td>\n",
       "      <td>...</td>\n",
       "      <td>0.500206</td>\n",
       "      <td>0.514164</td>\n",
       "      <td>0.584079</td>\n",
       "      <td>0.499727</td>\n",
       "      <td>0.593786</td>\n",
       "      <td>0.494439</td>\n",
       "      <td>0.470065</td>\n",
       "      <td>1.189533</td>\n",
       "      <td>0.300959</td>\n",
       "      <td>AFLW2000\\image00004.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.318026</td>\n",
       "      <td>0.799192</td>\n",
       "      <td>0.321820</td>\n",
       "      <td>0.776531</td>\n",
       "      <td>0.318601</td>\n",
       "      <td>0.782021</td>\n",
       "      <td>0.311447</td>\n",
       "      <td>0.746027</td>\n",
       "      <td>0.321967</td>\n",
       "      <td>0.767880</td>\n",
       "      <td>...</td>\n",
       "      <td>0.328722</td>\n",
       "      <td>0.723223</td>\n",
       "      <td>0.372980</td>\n",
       "      <td>0.712863</td>\n",
       "      <td>0.377189</td>\n",
       "      <td>0.709551</td>\n",
       "      <td>-0.184650</td>\n",
       "      <td>0.881137</td>\n",
       "      <td>-0.236852</td>\n",
       "      <td>AFLW2000\\image00006.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.503904</td>\n",
       "      <td>0.694327</td>\n",
       "      <td>0.469415</td>\n",
       "      <td>0.654004</td>\n",
       "      <td>0.489449</td>\n",
       "      <td>0.658819</td>\n",
       "      <td>0.444337</td>\n",
       "      <td>0.590304</td>\n",
       "      <td>0.461645</td>\n",
       "      <td>0.637688</td>\n",
       "      <td>...</td>\n",
       "      <td>0.490581</td>\n",
       "      <td>0.518596</td>\n",
       "      <td>0.595712</td>\n",
       "      <td>0.455546</td>\n",
       "      <td>0.602776</td>\n",
       "      <td>0.444499</td>\n",
       "      <td>-0.175379</td>\n",
       "      <td>0.299208</td>\n",
       "      <td>-0.373374</td>\n",
       "      <td>AFLW2000\\image00008.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.882169</td>\n",
       "      <td>1.198004</td>\n",
       "      <td>-1.033374</td>\n",
       "      <td>AFLW2000\\image00010.jpg</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 940 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        X 0       Y 0       X 1       Y 1       X 2       Y 2       X 3  \\\n",
       "0  0.485699  0.687087  0.489150  0.639948  0.489853  0.647495  0.479756   \n",
       "1  0.442459  0.642504  0.417327  0.592812  0.449168  0.611022  0.435618   \n",
       "2  0.318026  0.799192  0.321820  0.776531  0.318601  0.782021  0.311447   \n",
       "3  0.503904  0.694327  0.469415  0.654004  0.489449  0.658819  0.444337   \n",
       "4       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
       "\n",
       "        Y 3       X 4       Y 4  ...     X 465     Y 465     X 466     Y 466  \\\n",
       "0  0.562636  0.490495  0.621423  ...  0.541506  0.500291  0.665707  0.471219   \n",
       "1  0.544955  0.415644  0.577393  ...  0.500206  0.514164  0.584079  0.499727   \n",
       "2  0.746027  0.321967  0.767880  ...  0.328722  0.723223  0.372980  0.712863   \n",
       "3  0.590304  0.461645  0.637688  ...  0.490581  0.518596  0.595712  0.455546   \n",
       "4       NaN       NaN       NaN  ...       NaN       NaN       NaN       NaN   \n",
       "\n",
       "      X 467     Y 467     pitch       yaw      roll               image_name  \n",
       "0  0.676151  0.463839 -0.399231  0.018227  0.085676  AFLW2000\\image00002.jpg  \n",
       "1  0.593786  0.494439  0.470065  1.189533  0.300959  AFLW2000\\image00004.jpg  \n",
       "2  0.377189  0.709551 -0.184650  0.881137 -0.236852  AFLW2000\\image00006.jpg  \n",
       "3  0.602776  0.444499 -0.175379  0.299208 -0.373374  AFLW2000\\image00008.jpg  \n",
       "4       NaN       NaN -0.882169  1.198004 -1.033374  AFLW2000\\image00010.jpg  \n",
       "\n",
       "[5 rows x 940 columns]"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "vnSir2kTzB9X"
   },
   "source": [
    "**Some image mediaPipe can't get its features so we drop the null**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "id": "kBvXJ3tv4GwA"
   },
   "outputs": [],
   "source": [
    "train_df.dropna(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "id": "aIwc4Ak0uN-G"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1854, 940)"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "id": "siiuRDJ4Udex"
   },
   "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>X 0</th>\n",
       "      <th>Y 0</th>\n",
       "      <th>X 1</th>\n",
       "      <th>Y 1</th>\n",
       "      <th>X 2</th>\n",
       "      <th>Y 2</th>\n",
       "      <th>X 3</th>\n",
       "      <th>Y 3</th>\n",
       "      <th>X 4</th>\n",
       "      <th>Y 4</th>\n",
       "      <th>...</th>\n",
       "      <th>X 465</th>\n",
       "      <th>Y 465</th>\n",
       "      <th>X 466</th>\n",
       "      <th>Y 466</th>\n",
       "      <th>X 467</th>\n",
       "      <th>Y 467</th>\n",
       "      <th>pitch</th>\n",
       "      <th>yaw</th>\n",
       "      <th>roll</th>\n",
       "      <th>image_name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.003452</td>\n",
       "      <td>0.047139</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000702</td>\n",
       "      <td>0.007547</td>\n",
       "      <td>-0.009395</td>\n",
       "      <td>-0.077312</td>\n",
       "      <td>0.001345</td>\n",
       "      <td>-0.018525</td>\n",
       "      <td>...</td>\n",
       "      <td>0.052356</td>\n",
       "      <td>-0.139657</td>\n",
       "      <td>0.176557</td>\n",
       "      <td>-0.168729</td>\n",
       "      <td>0.187001</td>\n",
       "      <td>-0.176109</td>\n",
       "      <td>-0.399231</td>\n",
       "      <td>0.018227</td>\n",
       "      <td>0.085676</td>\n",
       "      <td>AFLW2000\\image00002.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.025133</td>\n",
       "      <td>0.049692</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.031842</td>\n",
       "      <td>0.018210</td>\n",
       "      <td>0.018291</td>\n",
       "      <td>-0.047857</td>\n",
       "      <td>-0.001682</td>\n",
       "      <td>-0.015419</td>\n",
       "      <td>...</td>\n",
       "      <td>0.082880</td>\n",
       "      <td>-0.078648</td>\n",
       "      <td>0.166752</td>\n",
       "      <td>-0.093085</td>\n",
       "      <td>0.176460</td>\n",
       "      <td>-0.098373</td>\n",
       "      <td>0.470065</td>\n",
       "      <td>1.189533</td>\n",
       "      <td>0.300959</td>\n",
       "      <td>AFLW2000\\image00004.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.003794</td>\n",
       "      <td>0.022660</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-0.003219</td>\n",
       "      <td>0.005490</td>\n",
       "      <td>-0.010373</td>\n",
       "      <td>-0.030505</td>\n",
       "      <td>0.000148</td>\n",
       "      <td>-0.008651</td>\n",
       "      <td>...</td>\n",
       "      <td>0.006902</td>\n",
       "      <td>-0.053308</td>\n",
       "      <td>0.051160</td>\n",
       "      <td>-0.063668</td>\n",
       "      <td>0.055369</td>\n",
       "      <td>-0.066981</td>\n",
       "      <td>-0.184650</td>\n",
       "      <td>0.881137</td>\n",
       "      <td>-0.236852</td>\n",
       "      <td>AFLW2000\\image00006.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.034489</td>\n",
       "      <td>0.040323</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.020034</td>\n",
       "      <td>0.004815</td>\n",
       "      <td>-0.025078</td>\n",
       "      <td>-0.063699</td>\n",
       "      <td>-0.007770</td>\n",
       "      <td>-0.016316</td>\n",
       "      <td>...</td>\n",
       "      <td>0.021166</td>\n",
       "      <td>-0.135407</td>\n",
       "      <td>0.126297</td>\n",
       "      <td>-0.198458</td>\n",
       "      <td>0.133361</td>\n",
       "      <td>-0.209504</td>\n",
       "      <td>-0.175379</td>\n",
       "      <td>0.299208</td>\n",
       "      <td>-0.373374</td>\n",
       "      <td>AFLW2000\\image00008.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.017556</td>\n",
       "      <td>0.064269</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.007057</td>\n",
       "      <td>0.019690</td>\n",
       "      <td>-0.025400</td>\n",
       "      <td>-0.062405</td>\n",
       "      <td>-0.004541</td>\n",
       "      <td>-0.020095</td>\n",
       "      <td>...</td>\n",
       "      <td>0.012831</td>\n",
       "      <td>-0.112081</td>\n",
       "      <td>0.109594</td>\n",
       "      <td>-0.152155</td>\n",
       "      <td>0.118412</td>\n",
       "      <td>-0.156283</td>\n",
       "      <td>-0.026812</td>\n",
       "      <td>0.011965</td>\n",
       "      <td>-0.220662</td>\n",
       "      <td>AFLW2000\\image00013.jpg</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 940 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        X 0       Y 0  X 1  Y 1       X 2       Y 2       X 3       Y 3  \\\n",
       "0 -0.003452  0.047139  0.0  0.0  0.000702  0.007547 -0.009395 -0.077312   \n",
       "1  0.025133  0.049692  0.0  0.0  0.031842  0.018210  0.018291 -0.047857   \n",
       "2 -0.003794  0.022660  0.0  0.0 -0.003219  0.005490 -0.010373 -0.030505   \n",
       "3  0.034489  0.040323  0.0  0.0  0.020034  0.004815 -0.025078 -0.063699   \n",
       "5  0.017556  0.064269  0.0  0.0  0.007057  0.019690 -0.025400 -0.062405   \n",
       "\n",
       "        X 4       Y 4  ...     X 465     Y 465     X 466     Y 466     X 467  \\\n",
       "0  0.001345 -0.018525  ...  0.052356 -0.139657  0.176557 -0.168729  0.187001   \n",
       "1 -0.001682 -0.015419  ...  0.082880 -0.078648  0.166752 -0.093085  0.176460   \n",
       "2  0.000148 -0.008651  ...  0.006902 -0.053308  0.051160 -0.063668  0.055369   \n",
       "3 -0.007770 -0.016316  ...  0.021166 -0.135407  0.126297 -0.198458  0.133361   \n",
       "5 -0.004541 -0.020095  ...  0.012831 -0.112081  0.109594 -0.152155  0.118412   \n",
       "\n",
       "      Y 467     pitch       yaw      roll               image_name  \n",
       "0 -0.176109 -0.399231  0.018227  0.085676  AFLW2000\\image00002.jpg  \n",
       "1 -0.098373  0.470065  1.189533  0.300959  AFLW2000\\image00004.jpg  \n",
       "2 -0.066981 -0.184650  0.881137 -0.236852  AFLW2000\\image00006.jpg  \n",
       "3 -0.209504 -0.175379  0.299208 -0.373374  AFLW2000\\image00008.jpg  \n",
       "5 -0.156283 -0.026812  0.011965 -0.220662  AFLW2000\\image00013.jpg  \n",
       "\n",
       "[5 rows x 940 columns]"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df = normalise_nose(train_df)\n",
    "train_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "id": "DizqFmQ9Q7Qx"
   },
   "outputs": [],
   "source": [
    "train_df.to_csv(\"final_train.csv\", index=False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "id": "p7b355zG8Gut"
   },
   "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>X 0</th>\n",
       "      <th>Y 0</th>\n",
       "      <th>X 1</th>\n",
       "      <th>Y 1</th>\n",
       "      <th>X 2</th>\n",
       "      <th>Y 2</th>\n",
       "      <th>X 3</th>\n",
       "      <th>Y 3</th>\n",
       "      <th>X 4</th>\n",
       "      <th>Y 4</th>\n",
       "      <th>...</th>\n",
       "      <th>X 465</th>\n",
       "      <th>Y 465</th>\n",
       "      <th>X 466</th>\n",
       "      <th>Y 466</th>\n",
       "      <th>X 467</th>\n",
       "      <th>Y 467</th>\n",
       "      <th>pitch</th>\n",
       "      <th>yaw</th>\n",
       "      <th>roll</th>\n",
       "      <th>image_name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.003452</td>\n",
       "      <td>0.047139</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000702</td>\n",
       "      <td>0.007547</td>\n",
       "      <td>-0.009395</td>\n",
       "      <td>-0.077312</td>\n",
       "      <td>0.001345</td>\n",
       "      <td>-0.018525</td>\n",
       "      <td>...</td>\n",
       "      <td>0.052356</td>\n",
       "      <td>-0.139657</td>\n",
       "      <td>0.176557</td>\n",
       "      <td>-0.168729</td>\n",
       "      <td>0.187001</td>\n",
       "      <td>-0.176109</td>\n",
       "      <td>-0.399231</td>\n",
       "      <td>0.018227</td>\n",
       "      <td>0.085676</td>\n",
       "      <td>AFLW2000\\image00002.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.025133</td>\n",
       "      <td>0.049692</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.031842</td>\n",
       "      <td>0.018210</td>\n",
       "      <td>0.018291</td>\n",
       "      <td>-0.047857</td>\n",
       "      <td>-0.001682</td>\n",
       "      <td>-0.015419</td>\n",
       "      <td>...</td>\n",
       "      <td>0.082880</td>\n",
       "      <td>-0.078648</td>\n",
       "      <td>0.166752</td>\n",
       "      <td>-0.093085</td>\n",
       "      <td>0.176460</td>\n",
       "      <td>-0.098373</td>\n",
       "      <td>0.470065</td>\n",
       "      <td>1.189533</td>\n",
       "      <td>0.300959</td>\n",
       "      <td>AFLW2000\\image00004.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.003794</td>\n",
       "      <td>0.022660</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-0.003219</td>\n",
       "      <td>0.005490</td>\n",
       "      <td>-0.010373</td>\n",
       "      <td>-0.030505</td>\n",
       "      <td>0.000148</td>\n",
       "      <td>-0.008651</td>\n",
       "      <td>...</td>\n",
       "      <td>0.006902</td>\n",
       "      <td>-0.053308</td>\n",
       "      <td>0.051160</td>\n",
       "      <td>-0.063668</td>\n",
       "      <td>0.055369</td>\n",
       "      <td>-0.066981</td>\n",
       "      <td>-0.184650</td>\n",
       "      <td>0.881137</td>\n",
       "      <td>-0.236852</td>\n",
       "      <td>AFLW2000\\image00006.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.034489</td>\n",
       "      <td>0.040323</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.020034</td>\n",
       "      <td>0.004815</td>\n",
       "      <td>-0.025078</td>\n",
       "      <td>-0.063699</td>\n",
       "      <td>-0.007770</td>\n",
       "      <td>-0.016316</td>\n",
       "      <td>...</td>\n",
       "      <td>0.021166</td>\n",
       "      <td>-0.135407</td>\n",
       "      <td>0.126297</td>\n",
       "      <td>-0.198458</td>\n",
       "      <td>0.133361</td>\n",
       "      <td>-0.209504</td>\n",
       "      <td>-0.175379</td>\n",
       "      <td>0.299208</td>\n",
       "      <td>-0.373374</td>\n",
       "      <td>AFLW2000\\image00008.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.017556</td>\n",
       "      <td>0.064269</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.007057</td>\n",
       "      <td>0.019690</td>\n",
       "      <td>-0.025400</td>\n",
       "      <td>-0.062405</td>\n",
       "      <td>-0.004541</td>\n",
       "      <td>-0.020095</td>\n",
       "      <td>...</td>\n",
       "      <td>0.012831</td>\n",
       "      <td>-0.112081</td>\n",
       "      <td>0.109594</td>\n",
       "      <td>-0.152155</td>\n",
       "      <td>0.118412</td>\n",
       "      <td>-0.156283</td>\n",
       "      <td>-0.026812</td>\n",
       "      <td>0.011965</td>\n",
       "      <td>-0.220662</td>\n",
       "      <td>AFLW2000\\image00013.jpg</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 940 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        X 0       Y 0  X 1  Y 1       X 2       Y 2       X 3       Y 3  \\\n",
       "0 -0.003452  0.047139  0.0  0.0  0.000702  0.007547 -0.009395 -0.077312   \n",
       "1  0.025133  0.049692  0.0  0.0  0.031842  0.018210  0.018291 -0.047857   \n",
       "2 -0.003794  0.022660  0.0  0.0 -0.003219  0.005490 -0.010373 -0.030505   \n",
       "3  0.034489  0.040323  0.0  0.0  0.020034  0.004815 -0.025078 -0.063699   \n",
       "4  0.017556  0.064269  0.0  0.0  0.007057  0.019690 -0.025400 -0.062405   \n",
       "\n",
       "        X 4       Y 4  ...     X 465     Y 465     X 466     Y 466     X 467  \\\n",
       "0  0.001345 -0.018525  ...  0.052356 -0.139657  0.176557 -0.168729  0.187001   \n",
       "1 -0.001682 -0.015419  ...  0.082880 -0.078648  0.166752 -0.093085  0.176460   \n",
       "2  0.000148 -0.008651  ...  0.006902 -0.053308  0.051160 -0.063668  0.055369   \n",
       "3 -0.007770 -0.016316  ...  0.021166 -0.135407  0.126297 -0.198458  0.133361   \n",
       "4 -0.004541 -0.020095  ...  0.012831 -0.112081  0.109594 -0.152155  0.118412   \n",
       "\n",
       "      Y 467     pitch       yaw      roll               image_name  \n",
       "0 -0.176109 -0.399231  0.018227  0.085676  AFLW2000\\image00002.jpg  \n",
       "1 -0.098373  0.470065  1.189533  0.300959  AFLW2000\\image00004.jpg  \n",
       "2 -0.066981 -0.184650  0.881137 -0.236852  AFLW2000\\image00006.jpg  \n",
       "3 -0.209504 -0.175379  0.299208 -0.373374  AFLW2000\\image00008.jpg  \n",
       "4 -0.156283 -0.026812  0.011965 -0.220662  AFLW2000\\image00013.jpg  \n",
       "\n",
       "[5 rows x 940 columns]"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df = pd.read_csv('final_train.csv')\n",
    "train_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Z3wM1t4cY-IC"
   },
   "source": [
    "# Model Stage"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "SoF29l8-cg2k"
   },
   "source": [
    "This is a function that trains a machine learning model on the given data using scikit-learn's MultiOutputRegressor class, which is a wrapper for a single regressor that can handle multiple target variables. The function takes in three parameters:\n",
    "\n",
    "base_model: The underlying regressor model to be used, such as Linear Regression, Lasso, or Ridge.\n",
    "X_train: The input feature variables to be used for training the model.\n",
    "y_train: The target variables corresponding to X_train that the model aims to predict.\n",
    "In the first line of the function, base_model is assigned to the variable svr. Then, MultiOutputRegressor is used to create a new regressor object mor that uses base_model as its internal regressor. The fit() method of the mor regressor is called on the training data X_train and y_train to train the model. Finally, the trained model mor is returned by the function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "id": "JCQvRWBcThfr"
   },
   "outputs": [],
   "source": [
    "def model_train(base_model,X_train, y_train):\n",
    "    # Create the Model regressor\n",
    "    svr = base_model\n",
    "    # Create the Multioutput Regressor\n",
    "    mor = MultiOutputRegressor(base_model)\n",
    "    # Train the regressor\n",
    "    mor = mor.fit(X_train, y_train)\n",
    "    return mor"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "IEpQ1Dd2cjLW"
   },
   "source": [
    "This is a function that generates predictions for the testing data using the trained model. The function takes in two parameters:\n",
    "\n",
    "model: The trained machine learning model that will be used to generate predictions.\n",
    "X_val: The input feature variables for the validation dataset that the model will be used to predict the target variable for.\n",
    "The predict() method of the model object is called on the validation feature data X_val to generate the predictions for the target variable. These predictions are returned as the output of the function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "id": "vuGCdIgUThcZ"
   },
   "outputs": [],
   "source": [
    "def model_predict(model,X_val):\n",
    "  # Generate predictions for testing data\n",
    "  return model.predict(X_val)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "SZ3DKFWtcrOc"
   },
   "source": [
    "This is a function that evaluates the performance of a trained machine learning model by calculating the mean squared error (MSE) and mean absolute error (MAE) for each target variable in the validation dataset. The function takes in two parameters:\n",
    "\n",
    "y_val: The actual target variables for the validation dataset.\n",
    "y_pred: The predicted target variables for the validation dataset.\n",
    "The function first calculates the MSE and MAE for each target variable by calling the mean_squared_error() and mean_absolute_error() functions from the scikit-learn library. The first argument for each of these functions is the actual target variable for the specific target column in the validation dataset, and the second argument is the predicted target variable for the same target column. The calculated MSE and MAE for each target variable are stored in the variables mse_one, mae_one, mse_two, mae_two, mse_third, and mae_third.\n",
    "\n",
    "The function then prints out the calculated MSE and MAE for each target variable in a human-readable format. The output includes the MSE and MAE for each of the three target variables, with the target variables labeled as the first, second, and third regressor."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "id": "NY_lJNF1ThYg"
   },
   "outputs": [],
   "source": [
    "def evaluate_model(y_val, y_pred):\n",
    "    # Evaluate the regressor\n",
    "    mse_one = mean_squared_error(y_val.iloc[:,0], y_pred[:,0])\n",
    "    mse_two = mean_squared_error(y_val.iloc[:,1], y_pred[:,1])\n",
    "    mse_third = mean_squared_error(y_val.iloc[:,2], y_pred[:,2])\n",
    "    print(f'MSE for first regressor: {mse_one} - second regressor: {mse_two} - Thrid regressor: {mse_third}')\n",
    "    mae_one = mean_absolute_error(y_val.iloc[:,0], y_pred[:,0])\n",
    "    mae_two = mean_absolute_error(y_val.iloc[:,1], y_pred[:,1])\n",
    "    mae_third = mean_absolute_error(y_val.iloc[:,2], y_pred[:,2])\n",
    "    print(f'MAE for first regressor: {mae_one} - second regressor: {mae_two} - Thrid regressor: {mae_third}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "id": "VpY90Q8oZWgb"
   },
   "outputs": [],
   "source": [
    "#Train\n",
    "train_df.drop(['image_name'], axis = 1, inplace = True)\n",
    "X = train_df.iloc[:,:-3]\n",
    "y = train_df.iloc[:,-3:] "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "id": "8ST083ZbZO56"
   },
   "outputs": [],
   "source": [
    "#Split\n",
    "X_train, X_val, y_train, y_val = train_test_split(X, y, test_size=0.20, random_state=33)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "_DYUuL79xCgG"
   },
   "source": [
    "# LinearRegression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "nISU_5R3dI3d"
   },
   "source": [
    "To establish a baseline model, we utilized linear regression. However, the results were unsatisfactory and the model appeared to be overfitting the data. Consequently, we decided to explore other models to improve the performance. To that end, we examined the effectiveness of Lasso and Ridge models."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "id": "EYJnPGGHxNPA"
   },
   "outputs": [],
   "source": [
    "scaler = StandardScaler().fit(X_train)\n",
    "X_train_std = scaler.transform(X_train)\n",
    "X_val_std = scaler.transform(X_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "id": "7wKUQ_W7xCDe"
   },
   "outputs": [],
   "source": [
    "reg = LinearRegression()\n",
    "multi_reg = model_train(reg,X_train_std, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "id": "O6GfFCY3xB6I"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE for first regressor: 0.48350345938637374 - second regressor: 0.08415366971078136 - Thrid regressor: 0.2674762631851598\n",
      "MAE for first regressor: 0.3298580496193222 - second regressor: 0.1965866114272592 - Thrid regressor: 0.3022119531042025\n"
     ]
    }
   ],
   "source": [
    "# Create the Multioutput Regressor\n",
    "y_pred = model_predict(multi_reg, X_val_std)\n",
    "evaluate_model(y_val, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "id": "XtFJKX0xd-y0"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE for first regressor: 0.015619950071530343 - second regressor: 0.007948736509907931 - Thrid regressor: 0.01526612471257049\n",
      "MAE for first regressor: 0.09651022980466784 - second regressor: 0.06352893441303328 - Thrid regressor: 0.09177044126373564\n"
     ]
    }
   ],
   "source": [
    "y_pred_train = model_predict(multi_reg, X_train_std)\n",
    "evaluate_model(y_train, y_pred_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "XSyHcTlieFhL"
   },
   "source": [
    "Initially, we used Linear Regression as the baseline model, but the results were not very good, and the model was overfitting. So, we decided to use Lasso and Ridge, which helped to reduce overfitting. However, the model's performance was still not satisfactory. Therefore, we decided to use standard scaling to preprocess the data, hoping to improve accuracy."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "pQOitwE9I0Im"
   },
   "source": [
    "## Ridge"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "id": "B0NIL97zI1KY"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE for first regressor: 0.1361706409292237 - second regressor: 0.016445832255705145 - Thrid regressor: 0.1621473304689493\n",
      "MAE for first regressor: 0.12337417647891517 - second regressor: 0.08018525276881318 - Thrid regressor: 0.1634811288121667\n"
     ]
    }
   ],
   "source": [
    "clf = Ridge(alpha=0.1, tol = 1e-4)\n",
    "multi_clf = model_train(clf,X_train, y_train)\n",
    "# Generate predictions for testing data\n",
    "y_pred = model_predict(multi_clf,X_val)\n",
    "evaluate_model(y_val, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "id": "q3bZp0KKenat"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE for first regressor: 0.290093899159002 - second regressor: 0.05621287000402126 - Thrid regressor: 0.32294136698254555\n",
      "MAE for first regressor: 0.13496774715890708 - second regressor: 0.10242326517705093 - Thrid regressor: 0.1632385100371764\n"
     ]
    }
   ],
   "source": [
    "y_pred_train = model_predict(multi_clf,X_train)\n",
    "evaluate_model(y_train, y_pred_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "sx3Q-RZGJEu3"
   },
   "source": [
    "#  Lasso"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "id": "dPqmyWUZI1F6"
   },
   "outputs": [],
   "source": [
    "lasso = Lasso(alpha=0.2, tol = 1e-5)\n",
    "multi_clf_lasso = model_train(lasso,X_train_std, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "id": "gjm8x4UqI1Cl"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE for first regressor: 0.14810335218368514 - second regressor: 0.06865956028335772 - Thrid regressor: 0.14767108716956434\n",
      "MAE for first regressor: 0.18941937207447387 - second regressor: 0.16049937604030348 - Thrid regressor: 0.13915031450669327\n"
     ]
    }
   ],
   "source": [
    "# Create the Multioutput Regressor\n",
    "y_pred = model_predict(multi_clf_lasso,X_val_std)\n",
    "evaluate_model(y_val, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "id": "xtS6_kVyetcL"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE for first regressor: 0.3505012275466951 - second regressor: 0.11276553502777385 - Thrid regressor: 0.41736497380961884\n",
      "MAE for first regressor: 0.21366443126510387 - second regressor: 0.17964723380838307 - Thrid regressor: 0.15168968155963586\n"
     ]
    }
   ],
   "source": [
    "# Create the Multioutput Regressor\n",
    "y_pred_train = model_predict(multi_clf_lasso,X_train_std)\n",
    "evaluate_model(y_train, y_pred_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "G8k_LiGwJLp9"
   },
   "source": [
    "\n",
    "# Normalization are important for model such as Ridge & Lasso"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "QwowV0FsfGKM"
   },
   "source": [
    "After applying standard scaling to preprocess the data, we observed a slight decrease in the mean absolute error, but the model's performance was still not satisfactory. Therefore, we decided to use Support Vector Machines (SVM) for the next step of our solution."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "id": "LBwFVvTVJJv_"
   },
   "outputs": [],
   "source": [
    "scaler = StandardScaler().fit(X_train)\n",
    "X_train_std = scaler.transform(X_train)\n",
    "X_val_std = scaler.transform(X_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "id": "9d2taIORJJse"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE for first regressor: 0.21279597174630593 - second regressor: 0.021092109322207366 - Thrid regressor: 0.20340993091916543\n",
      "MAE for first regressor: 0.1846777982031816 - second regressor: 0.09756691728701503 - Thrid regressor: 0.22810042659333535\n"
     ]
    }
   ],
   "source": [
    "# Create the Ridge regressor\n",
    "clf_n = Ridge(alpha=0.2, tol = 1e-4)\n",
    "multi_clf_n =  model_train(clf,X_train_std, y_train)\n",
    "# Create the Multioutput Regressor\n",
    "y_pred =model_predict(multi_clf_n,X_val_std)\n",
    "evaluate_model(y_val, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "id": "Gnki4ynKJRdv"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE for first regressor: 0.14810335218368514 - second regressor: 0.06866254878567718 - Thrid regressor: 0.14766955374098947\n",
      "MAE for first regressor: 0.18941937207447387 - second regressor: 0.1605015448242217 - Thrid regressor: 0.13914781171864934\n"
     ]
    }
   ],
   "source": [
    "# Create the Ridge regressor\n",
    "Lasso_n = Lasso(alpha=0.2, tol = 1e-4)\n",
    "multi_Lasso_n =  model_train(Lasso_n ,X_train_std, y_train)\n",
    "# Create the Multioutput Regressor\n",
    "y_pred =model_predict(multi_Lasso_n,X_val_std)\n",
    "evaluate_model(y_val, y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "VzO8p5sAJxD3"
   },
   "source": [
    "# SVM"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "AEiIgW-df34E"
   },
   "source": [
    "For the next step in our analysis, we will use Support Vector Machines (SVM) with the radial basis function (RBF) kernel. This choice is based on the fact that the RBF kernel is capable of capturing complex, nonlinear relationships between input features and output targets. Additionally, the RBF kernel has a local nature, which means that it only considers nearby points in the feature space when making predictions. This property allows it to handle high-dimensional data effectively"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "id": "7aO9ucNlJjT1"
   },
   "outputs": [],
   "source": [
    "# Create the SVR regressor\n",
    "svr = SVR()\n",
    "svr = model_train(svr,X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "id": "aC2CjVVfJjQz"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE for first regressor: 0.11887540566526163 - second regressor: 0.01206745417938321 - Thrid regressor: 0.1304682770242424\n",
      "MAE for first regressor: 0.0980016876764528 - second regressor: 0.06302759742017414 - Thrid regressor: 0.07643389345441091\n"
     ]
    }
   ],
   "source": [
    "# Create the Multioutput Regressor\n",
    "y_pred =model_predict(svr,X_val)\n",
    "evaluate_model(y_val, y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "mEq5kgqhKJk-"
   },
   "source": [
    "Normalization are important for model such as SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "id": "9sRBpe28JjM1"
   },
   "outputs": [],
   "source": [
    "# Create the SVR regressor\n",
    "svrn = SVR()\n",
    "svrn = model_train(svrn,X_train_std, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "id": "zG-4KuLaJjJ-"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE for first regressor: 0.12030577148920273 - second regressor: 0.012006977895257223 - Thrid regressor: 0.13246570657683968\n",
      "MAE for first regressor: 0.10083724257425755 - second regressor: 0.06972732619570149 - Thrid regressor: 0.08133431559865023\n"
     ]
    }
   ],
   "source": [
    "# Create the Multioutput Regressor\n",
    "y_pred =model_predict(svrn,X_val_std)\n",
    "evaluate_model(y_val, y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "MVrPuvG0gQNT"
   },
   "source": [
    "We found that applying Standard Scale did not have a significant impact on the performance of our SVM model. The model was able to achieve a satisfactory level of accuracy. However, for the next iteration of our analysis, we plan to explore tree-based models as they may provide a different perspective and potentially lead to further improvements in performance."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "qh3xJrSkQs7n"
   },
   "source": [
    "# Try Non-Linear Models"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "MPzmUoXnQy2G"
   },
   "source": [
    "Why Extra Tree ?\n",
    "Extra Trees not only randomly select the subset of features to split on at each node, but also use random thresholds for the splitting. This makes the trees more random and therefore more diverse, which can help to reduce overfitting and improve generalization.\n",
    "\n",
    "in addition to Extra Trees can be particularly useful when dealing with high-dimensional datasets this is perfect in our situation.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "L87Tri3QQwao"
   },
   "outputs": [],
   "source": [
    "# Create the Etra regressor\n",
    "etra = ExtraTreesRegressor()\n",
    "# Create the Multioutput Regressor\n",
    "etra = model_train(etra,X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "FtxefbUiQwXm"
   },
   "outputs": [],
   "source": [
    "# Generate predictions for testing data\n",
    "y_pred = model_predict(etra,X_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "KDEs-j0bQwU3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE for first regressor: 0.3457157629352173 - second regressor: 0.033822608693491976 - Thrid regressor: 0.4451662723571256\n",
      "MAE for first regressor: 0.14734835246377315 - second regressor: 0.07786276856380575 - Thrid regressor: 0.11791809484860996\n"
     ]
    }
   ],
   "source": [
    "evaluate_model(y_val, y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "1KnzCKmhR4Ut"
   },
   "source": [
    "Result Navie ExtraTreesRegressor it's seem accuracy as svm in validation try optimze the result on the bad esimator."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "KqmuojxxR30Q"
   },
   "outputs": [],
   "source": [
    "# Create the Etra regressor\n",
    "etra = ExtraTreesRegressor()\n",
    "param_grid = {\n",
    "    'n_estimators': [50,100,300],\n",
    "    'criterion': ['squared_error', 'absolute_error','friedman_mse', 'poisson'],\n",
    "    'max_depth': [2,8,16,32,50],\n",
    "    'max_features': ['auto','sqrt','log2'],    \n",
    "    'bootstrap': [True, False],\n",
    "}\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "6PyXuUYlR8J3"
   },
   "outputs": [],
   "source": [
    "gcv = RandomizedSearchCV(etra,param_grid,n_jobs=-1).fit(X_train, y_train.iloc[:,0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "IA9P9r2VR8G1"
   },
   "outputs": [],
   "source": [
    "# Create the Etra regressor\n",
    "best_etra = gcv.best_estimator_  \n",
    "# Train the regressor\n",
    "morextr = model_train(best_etra,X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "g2_Ka6dISGU3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE for first regressor: 0.1352833234791707 - second regressor: 0.030764972823884754 - Thrid regressor: 0.14146367842963686\n",
      "MAE for first regressor: 0.1444936499209394 - second regressor: 0.11571015104216542 - Thrid regressor: 0.15003954851381124\n"
     ]
    }
   ],
   "source": [
    "# Generate predictions for testing data\n",
    "y_pred = model_predict(morextr, X_val)\n",
    "evaluate_model(y_val, y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Sfhd9zX3hiSU"
   },
   "source": [
    "Finally we will try use xgb\n",
    "\n",
    "why xgb?\n",
    "\n",
    "popular tree-based ensemble model that uses gradient boosting algorithm. It is known for its high accuracy, speed and performance in large datasets, and ability to handle missing values, making it a popular choice for machine learning tasks. Additionally, XGBRegressor has a number of tunable hyperparameters that can be adjusted to optimize performance, which can make it a powerful tool for improving the accuracy of regression models.\n",
    "\n",
    "our result almost see as SVR"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "WXM0iRyaRB-G"
   },
   "source": [
    "# xgb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "3co8MuSP_bjJ"
   },
   "outputs": [],
   "source": [
    "xg = xgb.XGBRegressor(random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "wC0WggsZ_zHe"
   },
   "outputs": [],
   "source": [
    "multi_xg = model_train( xg,X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "Gp_wnAlS_bgH"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE for first regressor: 0.5272059378067118 - second regressor: 0.12627696590500903 - Thrid regressor: 0.4886777621549199\n",
      "MAE for first regressor: 0.14585931524348814 - second regressor: 0.0966081212926154 - Thrid regressor: 0.11353331774751992\n"
     ]
    }
   ],
   "source": [
    "# Create the Multioutput Regressor\n",
    "y_pred = model_predict(multi_xg, X_val)\n",
    "evaluate_model(y_val, y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "rkL1sgxmh8gV"
   },
   "source": [
    "We may want to consider using PCA on our models as our data has high dimensionality. PCA can help reduce the dimensionality of our feature space, allowing us to simplify our models and potentially improve their performance. By using PCA, we can identify the most important components that capture the most variation in our data and only keep those components for training our models, which can lead to better accuracy, faster training time, and less overfitting. However, it's worth noting that PCA can also result in loss of information, so it's important to carefully choose the number of components to keep and evaluate the impact on model performance"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "FrU0qagNvtxU"
   },
   "source": [
    "# Using PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "ivexs8RnvTHQ"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PCA(n_components=0.99)"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pca = PCA(n_components= 0.99)\n",
    "pca.fit(X_train_std)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "gHg_s10BwJIL"
   },
   "outputs": [],
   "source": [
    "train = pca.transform(X_train_std)\n",
    "test = pca.transform(X_val_std)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "rWxbwP__eA5O"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1483, 7)"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "HKFvS6ubwTAb"
   },
   "outputs": [],
   "source": [
    "# Create the Ridge regressor\n",
    "clf_n = Ridge(alpha=0.1, tol = 1e-4)\n",
    "multi_clf_n = model_train(clf_n,train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "drPlqI-JwS9T"
   },
   "outputs": [],
   "source": [
    "# Generate predictions for testing data\n",
    "y_pred = model_predict(multi_clf_n,test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "xB-U5anAx190"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE for first regressor: 0.12579648744532052 - second regressor: 0.02913881154810364 - Thrid regressor: 0.15124727609747457\n",
      "MAE for first regressor: 0.11811263793105023 - second regressor: 0.11791127550893984 - Thrid regressor: 0.14702144033786482\n"
     ]
    }
   ],
   "source": [
    "evaluate_model(y_val, y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ZFbMZJ00za7T"
   },
   "source": [
    "Linear"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "6kQs-SzczVLb"
   },
   "outputs": [],
   "source": [
    "reg = LinearRegression()\n",
    "multi_reg = model_train(reg,train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "FRmMjNl2zVIC"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE for first regressor: 0.12579648660181156 - second regressor: 0.029138806107966174 - Thrid regressor: 0.15124733572146132\n",
      "MAE for first regressor: 0.1181126468714062 - second regressor: 0.11791128134744851 - Thrid regressor: 0.1470215989411869\n"
     ]
    }
   ],
   "source": [
    "# Generate predictions for testing data\n",
    "y_pred = model_predict(multi_reg,test)\n",
    "evaluate_model(y_val, y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "7i-RHi4wyOkL"
   },
   "source": [
    "EXtra Tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "4RzaVBnzyNYa"
   },
   "outputs": [],
   "source": [
    "# Create the Etra regressor\n",
    "etra = ExtraTreesRegressor()\n",
    "# Create the Multioutput Regressor\n",
    "etra = model_train(etra,train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "Hse013N3yNST"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE for first regressor: 0.1815300290438273 - second regressor: 0.028634237273992395 - Thrid regressor: 0.3837715245905265\n",
      "MAE for first regressor: 0.1278134193184645 - second regressor: 0.07429417234658765 - Thrid regressor: 0.12316859265231456\n"
     ]
    }
   ],
   "source": [
    "# Generate predictions for testing data\n",
    "y_pred = model_predict(etra,test)\n",
    "evaluate_model(y_val, y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "P7zTaSrWFQNe"
   },
   "source": [
    "Lasso with PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "FQ6YKp-QFTYd"
   },
   "outputs": [],
   "source": [
    "lasso = Lasso(alpha=0.2, tol = 1e-5)\n",
    "multi_clf_lasso1 = model_train(lasso,train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "jBkOn3LmFTWF"
   },
   "outputs": [],
   "source": [
    "# Generate predictions for testing data\n",
    "y_pred = model_predict(multi_clf_lasso1,test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "9MHj7yeJFcJG"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE for first regressor: 0.12757766602157308 - second regressor: 0.03051401851771512 - Thrid regressor: 0.14413215387744738\n",
      "MAE for first regressor: 0.12007814152909187 - second regressor: 0.11976122127370428 - Thrid regressor: 0.12983896144535056\n"
     ]
    }
   ],
   "source": [
    "evaluate_model(y_val, y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "JP5Ei1jTi4IU"
   },
   "source": [
    "XGBOOST with PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "1jnxgVaEixqm"
   },
   "outputs": [],
   "source": [
    "xg = xgb.XGBRegressor(random_state=42)\n",
    "multi_xg = model_train( xg,train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "drhQg4uVixl1"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE for first regressor: 0.3063239907442858 - second regressor: 0.016698722211344892 - Thrid regressor: 1.4182959989488049\n",
      "MAE for first regressor: 0.13840581419249504 - second regressor: 0.07943705044940334 - Thrid regressor: 0.17393321084181654\n"
     ]
    }
   ],
   "source": [
    "# Generate predictions for testing data\n",
    "y_pred = model_predict(multi_xg,test)\n",
    "evaluate_model(y_val, y_pred)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "    SVR With PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [],
   "source": [
    "svr_pca = SVR()\n",
    "multi_svr_pca = model_train( svr_pca,train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE for first regressor: 0.12123246228866742 - second regressor: 0.012336476962170677 - Thrid regressor: 0.1327945134977274\n",
      "MAE for first regressor: 0.10383843064339275 - second regressor: 0.06964454700541187 - Thrid regressor: 0.08098357221802185\n"
     ]
    }
   ],
   "source": [
    "# Generate predictions for testing data\n",
    "y_pred = model_predict(multi_svr_pca,test)\n",
    "evaluate_model(y_val, y_pred)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {
    "id": "H5kYyNaBJ2O1"
   },
   "source": [
    "As a conclusion, using PCA to reduce the dimensionality of the data to only 7 components, did not lead to a decrease in accuracy in most cases. This approach can be useful when working with high-dimensional data, as it can help to reduce the computational cost and potentially improve the performance of the model. However, it is important to note that the optimal number of components to use may vary depending on the specific problem and data set, and it may require some experimentation to find the best approach.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "IYNF3J7bKdpH"
   },
   "source": [
    "# Make Model Pipeline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "VXRsC7p3j4Js"
   },
   "source": [
    "We continued our analysis by exploring the combination of PCA and SVR, and found that using PCA with SVR resulted in the best accuracy for most models. This approach allowed us to reduce the dimensionality of the data while maintaining high accuracy, demonstrating the effectiveness of this technique in improving the performance of our models."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {
    "colab": {
     "background_save": true
    },
    "id": "yIBNPB4Aj3e1"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE for first regressor: 0.11887540566526163 - second regressor: 0.01206745417938321 - Thrid regressor: 0.1304682770242424\n",
      "MAE for first regressor: 0.0980016876764528 - second regressor: 0.06302759742017414 - Thrid regressor: 0.07643389345441091\n"
     ]
    }
   ],
   "source": [
    "# Create the SVR regressor\n",
    "svr = SVR()\n",
    "svr = model_train(svr,X_train, y_train)\n",
    "# Create the Multioutput Regressor\n",
    "y_pred = model_predict(svr,X_val)\n",
    "evaluate_model(y_val, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {
    "id": "HZiB-ZzDKhbl"
   },
   "outputs": [],
   "source": [
    "pipe = Pipeline([('scaler', StandardScaler()),('pca',PCA(n_components= 0.99)), ('svr', SVR())])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {
    "id": "6PmZNz3yKhVV"
   },
   "outputs": [],
   "source": [
    "morsvr = model_train(pipe,X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {
    "id": "wmhfX9n3KhSn"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE for first regressor: 0.12123246228866742 - second regressor: 0.012336476962170656 - Thrid regressor: 0.13279451349772745\n",
      "MAE for first regressor: 0.10383843064339299 - second regressor: 0.0696445470054118 - Thrid regressor: 0.08098357221802191\n"
     ]
    }
   ],
   "source": [
    "y_pred = model_predict(morsvr,X_val)\n",
    "evaluate_model(y_val, y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Y1v9I70ubTlM"
   },
   "source": [
    "# Draw"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "KhNEwm4dlW0F"
   },
   "source": [
    "This function takes in Euler angles (roll, yaw, pitch) and returns a 3x3 rotation matrix that represents the same rotation as the Euler angles. The rotation matrix is obtained by computing the product of three separate rotation matrices, one for each Euler angle, which are constructed using trigonometric functions.\n",
    "\n",
    "The resulting rotation matrix can be used to transform a 3D object, such as an image or point cloud, to the orientation specified by the input Euler angles."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "id": "-tNNM2MbQAfL"
   },
   "outputs": [],
   "source": [
    "def EulerToMatrix(roll, yaw, pitch):\n",
    "    # roll - z axis\n",
    "    # yaw  - y axis\n",
    "    # pitch - x axis\n",
    "    roll = roll / 180 * np.pi\n",
    "    yaw = yaw / 180 * np.pi\n",
    "    pitch = pitch / 180 * np.pi\n",
    "\n",
    "    Rz = [[math.cos(roll), -math.sin(roll), 0],\n",
    "          [math.sin(roll), math.cos(roll), 0],\n",
    "          [0, 0, 1]]\n",
    "\n",
    "    Ry = [[math.cos(yaw), 0, math.sin(yaw)],\n",
    "          [0, 1, 0],\n",
    "          [-math.sin(yaw), 0, math.cos(yaw)]]\n",
    "\n",
    "    Rx = [[1, 0, 0],\n",
    "          [0, math.cos(pitch), -math.sin(pitch)],\n",
    "          [0, math.sin(pitch), math.cos(pitch)]]\n",
    "\n",
    "    matrix = np.matmul(Rx, Ry)\n",
    "    matrix = np.matmul(matrix, Rz)\n",
    "\n",
    "    return matrix"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "10u4Dx9qmddF"
   },
   "source": [
    "The draw_axis function is used to draw a set of 3D axes on an input image, with the orientation of the axes determined by the input pitch, yaw, and roll angles. The function takes in the following parameters:\n",
    "\n",
    "img: The input image on which to draw the axes.\n",
    "pitch: The pitch angle in degrees.\n",
    "yaw: The yaw angle in degrees.\n",
    "roll: The roll angle in degrees.\n",
    "tdx: The x-coordinate of the center point of the axes. If not provided, the center of the image is used.\n",
    "tdy: The y-coordinate of the center point of the axes. If not provided, the center of the image is used.\n",
    "size: The length of the axes to be drawn in pixels.\n",
    "The function first adjusts the input yaw value by negating it, and then calculates the 3D coordinates of the endpoints of the three axes using the input pitch, yaw, and roll angles. The function then draws the three axes on the input image using OpenCV's cv2.line function, with red representing the x-axis, green representing the y-axis, and blue representing the z-axis.\n",
    "\n",
    "The function returns the input image with the 3D axes drawn on it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "id": "KaRy8XE4lEOl"
   },
   "outputs": [],
   "source": [
    "def draw_axis(img, pitch,yaw,roll, tdx=None, tdy=None, size = 100):\n",
    "\n",
    "    yaw = -yaw\n",
    "    if tdx != None and tdy != None:\n",
    "        tdx = tdx\n",
    "        tdy = tdy\n",
    "    else:\n",
    "        height, width = img.shape[:2]\n",
    "        tdx = width / 2\n",
    "        tdy = height / 2\n",
    "\n",
    "    # X-Axis pointing to right. drawn in red\n",
    "    x1 = size * (math.cos(yaw) * math.cos(roll)) + tdx\n",
    "    y1 = size * (math.cos(pitch) * math.sin(roll) + math.cos(roll) * math.sin(pitch) * math.sin(yaw)) + tdy\n",
    "\n",
    "    # Y-Axis | drawn in green\n",
    "    #        v\n",
    "    x2 = size * (-math.cos(yaw) * math.sin(roll)) + tdx\n",
    "    y2 = size * (math.cos(pitch) * math.cos(roll) - math.sin(pitch) * math.sin(yaw) * math.sin(roll)) + tdy\n",
    "\n",
    "    # Z-Axis (out of the screen) drawn in blue\n",
    "    x3 = size * (math.sin(yaw)) + tdx\n",
    "    y3 = size * (-math.cos(yaw) * math.sin(pitch)) + tdy\n",
    "\n",
    "    cv2.line(img, (int(tdx), int(tdy)), (int(x1),int(y1)),(0,0,255),3)\n",
    "    cv2.line(img, (int(tdx), int(tdy)), (int(x2),int(y2)),(0,255,0),3)\n",
    "    cv2.line(img, (int(tdx), int(tdy)), (int(x3),int(y3)),(255,0,0),2)\n",
    "\n",
    "    return img"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "j2jcVyjRv8U2"
   },
   "source": [
    "# Position Nose"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "2ur2DjjhzVJg"
   },
   "outputs": [],
   "source": [
    "!curl  'https://raw.githubusercontent.com/adobe/SimpleSensor/master/simplesensor/collection_modules/demographic_camera/classifiers/haarcascades/haarcascade_mcs_nose.xml' > haarcascade_mcs_nose.xml\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "83_Qh1bNlvSd"
   },
   "source": [
    "This function takes an image as input and detects the position of the nose in the image. It returns the x and y coordinates of the center of the nose.\n",
    "\n",
    "\n",
    "**But we replace it using the media pipe feature point of the nose**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "OyuDdYhpwAIv"
   },
   "outputs": [],
   "source": [
    "\"\"\"def position_nose(img):\n",
    "  # Load the Haar cascade classifier XML file\n",
    "  nose_cascade = cv2.CascadeClassifier(\"/content/haarcascade_mcs_nose.xml\")\n",
    "  if nose_cascade.empty():\n",
    "    return 0,0\n",
    "  img = convert_RGB(img)\n",
    "  gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)\n",
    "  nose_rects = nose_cascade.detectMultiScale(gray, 1.3, 5)\n",
    "  for (x,y,w,h) in nose_rects:\n",
    "    return int((x+x+w)/2), int((y+y+h)/2)\n",
    "  return 0,0\"\"\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Q_yrzbccdBb7"
   },
   "source": [
    "# Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {
    "id": "AnLNf0-NIHEd"
   },
   "outputs": [],
   "source": [
    "def imge_prediction(img, model):\n",
    "\n",
    "    if type(img) is str:\n",
    "        img = plt.imread(img)\n",
    "    \n",
    "    list_mesh = multi_preprocessing(img)\n",
    "    df = pd.DataFrame(pd.Series(list_mesh))\n",
    "    \n",
    "    if df.empty == True:\n",
    "        return img\n",
    "    \n",
    "    nose_x, nose_y = df.T['X 1'][0], df.T['Y 1'][0]\n",
    "    \n",
    "    list_mesh = normalise_nose(df.T)\n",
    "     \n",
    "    \n",
    "    if list_mesh.shape[1] == model.n_features_in_:\n",
    "        y_pred = model.predict(list_mesh)\n",
    "        \n",
    "        if list_mesh['X 1'][0] is not None:\n",
    "            \n",
    "                img = convert_RGB(img)   \n",
    "                return draw_axis(img, y_pred[0][0], y_pred[0][1], y_pred[0][2], tdx=nose_x*img.shape[1], tdy=nose_y*img.shape[0], size=100)\n",
    "\n",
    "    return img\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {
    "id": "aViEtlBtJTkG"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x1a3accaf850>"
      ]
     },
     "execution_count": 117,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(imge_prediction('AFLW2000/image00002.jpg', morsvr))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "dHziXkCALJel"
   },
   "source": [
    "# video"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "moBchhFtQp3V"
   },
   "outputs": [],
   "source": [
    "#from google.colab import drive\n",
    "#drive.mount('/content/drive')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "kyBqPVlo0SEN"
   },
   "outputs": [],
   "source": [
    "#!cp '/content/drive/MyDrive/dataset/WhatsApp Video 2023-02-19 at 8.19.33 PM.mp4'  \"/content/uploaded.mp4\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {
    "id": "g_2HblxvYkN2"
   },
   "outputs": [],
   "source": [
    "video_path = \"FINAL-VIDEO.mp4\"\n",
    "\n",
    "cap = cv2.VideoCapture(video_path)\n",
    "\n",
    "rotateCode = False\n",
    "\n",
    "video_height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))\n",
    "video_width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))\n",
    "\n",
    "video_fps = cap.get(cv2.CAP_PROP_FPS)\n",
    "\n",
    "out = cv2.VideoWriter(\"final.mp4\", fourcc=cv2.VideoWriter_fourcc(*'mp4v'), fps=int(video_fps), frameSize=(video_width, video_height))\n",
    "\n",
    "# Check if the video is opened successfully\n",
    "if not cap.isOpened():\n",
    "\n",
    "    print(\"Error opening video file\")\n",
    "\n",
    "# Read the frames of the video\n",
    "while True:\n",
    "    ret, frame = cap.read()\n",
    "\n",
    "    # Break the loop if the video is over\n",
    "    if not ret:\n",
    "        break\n",
    "\n",
    "    \n",
    "    if rotateCode:\n",
    "         frame = cv2.rotate(frame, cv2.ROTATE_180)\n",
    "\n",
    "    \n",
    "    img = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)\n",
    "\n",
    "    \n",
    "    result = imge_prediction(frame, morsvr)\n",
    "\n",
    "    if type(result) is type(None):\n",
    "        continue\n",
    "\n",
    "    \n",
    "    frame = np.uint8(result)\n",
    "    \n",
    "    out.write(frame)\n",
    "    # Break the loop if the user presses \"q\"\n",
    "    if cv2.waitKey(25) & 0xFF == ord(\"q\"):\n",
    "        break\n",
    "\n",
    "# Release the video capture object and destroy all windows\n",
    "out.release()\n",
    "cap.release()\n",
    "#cv2.destroyAllWindows()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "lVhYL4Bj0xC3"
   },
   "source": [
    "# Run Time video"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "WBu32rC6fRXz"
   },
   "outputs": [],
   "source": [
    "from IPython.display import display, Javascript, Image\n",
    "from google.colab.output import eval_js\n",
    "from base64 import b64decode, b64encode\n",
    "import PIL\n",
    "import io\n",
    "import html\n",
    "import time\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "H4xnerBjfIs5"
   },
   "outputs": [],
   "source": [
    "# function to convert the JavaScript object into an OpenCV image\n",
    "def js_to_image(js_reply):\n",
    "  \"\"\"\n",
    "  Params:\n",
    "          js_reply: JavaScript object containing image from webcam\n",
    "  Returns:\n",
    "          img: OpenCV RGB image\n",
    "  \"\"\"\n",
    "  # decode base64 image\n",
    "  image_bytes = b64decode(js_reply.split(',')[1])\n",
    "  # convert bytes to numpy array\n",
    "  jpg_as_np = np.frombuffer(image_bytes, dtype=np.uint8)\n",
    "  # decode numpy array into OpenCV RGB image\n",
    "  img = cv2.imdecode(jpg_as_np, flags=1)\n",
    "\n",
    "  return img\n",
    "\n",
    "# function to convert OpenCV Rectangle bounding box image into base64 byte string to be overlayed on video stream\n",
    "def bbox_to_bytes(bbox_array):\n",
    "  \"\"\"\n",
    "  Params:\n",
    "          bbox_array: Numpy array (pixels) containing rectangle to overlay on video stream.\n",
    "  Returns:\n",
    "        bytes: Base64 image byte string\n",
    "  \"\"\"\n",
    "\n",
    "  # convert array into PIL image\n",
    "  bbox_PIL = PIL.Image.fromarray(bbox_array, 'RGB')\n",
    "  iobuf = io.BytesIO()\n",
    "  # format bbox into png for return\n",
    "  bbox_PIL.save(iobuf, format='png')\n",
    "  # format return string\n",
    "  bbox_bytes = 'data:image/png;base64,{}'.format((str(b64encode(iobuf.getvalue()), 'utf-8')))\n",
    "\n",
    "  return bbox_bytes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "qT5TWs5kfIqI"
   },
   "outputs": [],
   "source": [
    "# JavaScript to properly create our live video stream using our webcam as input\n",
    "def video_stream():\n",
    "\n",
    "    js = Javascript('''\n",
    "        var video;\n",
    "        var div = null;\n",
    "        var stream;\n",
    "        var captureCanvas;\n",
    "        var imgElement;\n",
    "        var labelElement;\n",
    "        \n",
    "        var pendingResolve = null;\n",
    "        var shutdown = false;\n",
    "        \n",
    "        \n",
    "        function onAnimationFrame() {\n",
    "            if (!shutdown) {\n",
    "                window.requestAnimationFrame(onAnimationFrame);\n",
    "            }\n",
    "            if (pendingResolve) {\n",
    "\n",
    "                var result = \"\";\n",
    "\n",
    "                if (!shutdown) {\n",
    "                    captureCanvas.getContext('2d').drawImage(video, 0, 0, 640, 480);\n",
    "                    result = captureCanvas.toDataURL('image/jpeg', 0.8)\n",
    "                }\n",
    "\n",
    "                var lp = pendingResolve;\n",
    "\n",
    "                pendingResolve = null;\n",
    "                lp(result);\n",
    "            }\n",
    "        }\n",
    "        \n",
    "        async function createDom() {\n",
    "            if (div !== null) {\n",
    "                return stream;\n",
    "            }\n",
    "\n",
    "            div = document.createElement('div');\n",
    "            div.style.border = '2px solid black';\n",
    "            div.style.padding = '3px';\n",
    "            div.style.width = '100%';\n",
    "            div.style.maxWidth = '600px';\n",
    "            document.body.appendChild(div);\n",
    "            \n",
    "            const modelOut = document.createElement('div');\n",
    "            modelOut.innerHTML = \"<span>Status:</span>\";\n",
    "            labelElement = document.createElement('span');\n",
    "            labelElement.innerText = 'No data';\n",
    "            labelElement.style.fontWeight = 'bold';\n",
    "            modelOut.appendChild(labelElement);\n",
    "            div.appendChild(modelOut);\n",
    "                \n",
    "            video = document.createElement('video');\n",
    "            video.style.display = 'block';\n",
    "            video.width = div.clientWidth - 6;\n",
    "            video.setAttribute('playsinline', '');\n",
    "            video.onclick = () => { shutdown = true; };\n",
    "            stream = await navigator.mediaDevices.getUserMedia(\n",
    "                        {video: {facingMode: \"environment\"}}\n",
    "                    );\n",
    "            div.appendChild(video);\n",
    "\n",
    "            imgElement = document.createElement('img');\n",
    "            imgElement.style.position = 'absolute';\n",
    "            imgElement.style.zIndex = 1;\n",
    "            imgElement.onclick = () => { shutdown = true; };\n",
    "            div.appendChild(imgElement);\n",
    "            \n",
    "            const instruction = document.createElement('div');\n",
    "            instruction.innerHTML = \n",
    "                '<span style=\"color: red; font-weight: bold;\">' +\n",
    "                'When finished, click here or on the video to stop this demo</span>';\n",
    "            div.appendChild(instruction);\n",
    "            instruction.onclick = () => {shutdown = true;};\n",
    "            \n",
    "            video.srcObject = stream;\n",
    "            await video.play();\n",
    "\n",
    "            captureCanvas = document.createElement('canvas');\n",
    "            captureCanvas.width = 640; //video.videoWidth;\n",
    "            captureCanvas.height = 480; //video.videoHeight;\n",
    "            window.requestAnimationFrame(onAnimationFrame);\n",
    "            \n",
    "            return stream;\n",
    "        }\n",
    "\n",
    "        function removeDom() {\n",
    "            stream.getVideoTracks()[0].stop();\n",
    "            video.remove();\n",
    "            div.remove();\n",
    "            video = null;\n",
    "            div = null;\n",
    "            stream = null;\n",
    "            imgElement = null;\n",
    "            captureCanvas = null;\n",
    "            labelElement = null;\n",
    "        }\n",
    "\n",
    "        async function stream_frame(label, imgData) {\n",
    "            if (shutdown) {\n",
    "                removeDom();\n",
    "                shutdown = false;\n",
    "                return '';\n",
    "            }\n",
    "\n",
    "            var preCreate = Date.now();\n",
    "            stream = await createDom();\n",
    "            \n",
    "            var preShow = Date.now();\n",
    "\n",
    "            if (label != \"\") {\n",
    "                labelElement.innerHTML = label;\n",
    "            }\n",
    "                    \n",
    "            if (imgData != \"\") {\n",
    "                var videoRect = video.getClientRects()[0];\n",
    "                imgElement.style.top = videoRect.top + \"px\";\n",
    "                imgElement.style.left = videoRect.left + \"px\";\n",
    "                imgElement.style.width = videoRect.width + \"px\";\n",
    "                imgElement.style.height = videoRect.height + \"px\";\n",
    "                imgElement.src = imgData;\n",
    "            }\n",
    "            \n",
    "            var preCapture = Date.now();\n",
    "            var result = await new Promise(function(resolve, reject) {\n",
    "                        pendingResolve = resolve;\n",
    "                    }\n",
    "            );\n",
    "            shutdown = false;\n",
    "            \n",
    "            return {'create': preShow - preCreate, \n",
    "                    'show': preCapture - preShow, \n",
    "                    'capture': Date.now() - preCapture,\n",
    "                    'img': result};\n",
    "        }\n",
    "\n",
    "    ''')\n",
    "\n",
    "    display(js)\n",
    "  \n",
    "def video_frame(label, bbox):\n",
    "  data = eval_js('stream_frame(\"{}\", \"{}\")'.format(label, bbox))\n",
    "  return data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "dU-rvexufWPA"
   },
   "outputs": [],
   "source": [
    "frames = np.empty(shape=(0,480,640,3))\n",
    "# start streaming video from webcam\n",
    "video_stream()\n",
    "# label for video\n",
    "label_html = 'Capturing...'\n",
    "# initialze bounding box to empty\n",
    "bbox = ''\n",
    "\n",
    "while True:\n",
    "    js_reply = video_frame(label_html, bbox)\n",
    "\n",
    "    if not js_reply:\n",
    "        break\n",
    "\n",
    "    # convert JS response to OpenCV Image\n",
    "    img = js_to_image(js_reply[\"img\"])\n",
    "\n",
    "\n",
    "    result = imge_prediction(img, morsvr)\n",
    "\n",
    "\n",
    "\n",
    "    if  result:\n",
    "        frames = np.append(frames, np.expand_dims(result, axis=0), axis=0)\n",
    "        res_bytes = bbox_to_bytes(result)\n",
    "        # update bbox so next frame gets new overlay\n",
    "        bbox = res_bytes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "Hi_POHKTTID3"
   },
   "outputs": [],
   "source": [
    "height, width = frames.shape[1:3]\n",
    "print(f\"number of frames: {frames.shape[0]} and width: {width} and height: {height}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "o7aIFo4cfjCo"
   },
   "outputs": [],
   "source": [
    "out = cv2.VideoWriter(\"videoStream.mp4\", cv2.VideoWriter_fourcc(*'mp4v'), 1, (width, height))\n",
    "for frame in frames:\n",
    "    \n",
    "    frame = np.uint8(frame)\n",
    "    out.write(frame) \n",
    "out.release()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "xng9SHusg1jV"
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "provenance": []
  },
  "gpuClass": "standard",
  "kernelspec": {
   "display_name": "base",
   "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": "3.9.13"
  },
  "vscode": {
   "interpreter": {
    "hash": "4c3bb0dd0e872a9fa34dcec78a4cd41b8cc2ac5cbe6ae443bdb0029bdd2040d5"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}