project2_HAR.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "project2_HAR.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/AMEYABHAVE/Cristl/blob/main/project2_HAR.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "J_39geOCNFWK",
        "outputId": "61a93cee-24d4-4334-b7c2-c15814793e65"
      },
      "source": [
        "import tensorflow as tf\n",
        "from tensorflow.keras import Sequential\n",
        "from tensorflow.keras.layers import Flatten, Dense, Dropout, BatchNormalization\n",
        "from tensorflow.keras.layers import Conv2D, MaxPool2D\n",
        "from tensorflow.keras.optimizers import Adam\n",
        "print(tf.__version__)"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "2.6.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XOv7Iw1jOdTc"
      },
      "source": [
        "from google.colab import drive\n",
        "import pandas as pd\n",
        "file_location = '/content/drive/MyDrive/accelerometer.csv'\n",
        "accelerometer =pd.read_csv(file_location)\n",
        "accelerometer.drop(columns=['time','repetition','uuid'], inplace=True)\n",
        "accsitting= accelerometer[accelerometer['activity'] == 'CHAIR'].head(31668)\n",
        "accstanding= accelerometer[accelerometer['activity'] == 'STANDING']\n",
        "acclying= accelerometer[accelerometer['activity'] == 'LYING_UP']\n",
        "acclying= acclying.append(accelerometer[accelerometer['activity'] == 'LYING_SIDE'].head(520))\n",
        "acclying.drop(columns=['activity'], inplace=True)\n",
        "acclying['activity']='LYING_UP'\n",
        "accdataset = pd.DataFrame()\n",
        "accdataset = accdataset.append([accsitting, accstanding, acclying])\n",
        "\n",
        "file_location1 = '/content/drive/MyDrive/gyroscope.csv'\n",
        "gyroscope =pd.read_csv(file_location1)\n",
        "gyroscope.drop(columns=['time','repetition','uuid'], inplace=True)\n",
        "gyrsitting= gyroscope[gyroscope['activity'] == 'CHAIR'].head(9875)\n",
        "gyrstanding= gyroscope[gyroscope['activity'] == 'STANDING'].head(9875)\n",
        "gyrlying= gyroscope[gyroscope['activity'] == 'LYING_UP'].head(9875)\n",
        "gyrdataset = pd.DataFrame()\n",
        "gyrdataset = gyrdataset.append([gyrsitting, gyrstanding, gyrlying])"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ulpTn2jJsCkN",
        "outputId": "9c02ad5e-8f48-4576-d726-764dc61d27ec"
      },
      "source": [
        "accdataset.info()\n",
        "gyrdataset.info()"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 95004 entries, 0 to 5319\n",
            "Data columns (total 5 columns):\n",
            " #   Column    Non-Null Count  Dtype  \n",
            "---  ------    --------------  -----  \n",
            " 0   name      95004 non-null  object \n",
            " 1   acc_x     95004 non-null  float64\n",
            " 2   acc_y     95004 non-null  float64\n",
            " 3   acc_z     95004 non-null  float64\n",
            " 4   activity  95004 non-null  object \n",
            "dtypes: float64(3), object(2)\n",
            "memory usage: 4.3+ MB\n",
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 29625 entries, 0 to 89378\n",
            "Data columns (total 5 columns):\n",
            " #   Column    Non-Null Count  Dtype  \n",
            "---  ------    --------------  -----  \n",
            " 0   name      29625 non-null  object \n",
            " 1   gyr_x     29625 non-null  float64\n",
            " 2   gyr_y     29625 non-null  float64\n",
            " 3   gyr_z     29625 non-null  float64\n",
            " 4   activity  29625 non-null  object \n",
            "dtypes: float64(3), object(2)\n",
            "memory usage: 1.4+ MB\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uqoudP-YtmC3",
        "outputId": "8374dfc7-48f4-4560-a7f1-3afc92a5d46e"
      },
      "source": [
        "accdataset['activity'].value_counts()"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "STANDING    31668\n",
              "LYING_UP    31668\n",
              "CHAIR       31668\n",
              "Name: activity, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3dxiYlzHtp2f",
        "outputId": "71282af8-02ad-4e04-f6ea-280884c7feee"
      },
      "source": [
        "gyrdataset['activity'].value_counts()"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "STANDING    9875\n",
              "CHAIR       9875\n",
              "LYING_UP    9875\n",
              "Name: activity, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ROo7wt7tt_a8",
        "outputId": "d229b6cd-da56-478a-89fb-97f7cef710c8"
      },
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.preprocessing import StandardScaler, LabelEncoder\n",
        "alabel=LabelEncoder()\n",
        "accdataset['label']=alabel.fit_transform(accdataset['activity'])\n",
        "print(accdataset.head())\n",
        "print(alabel.classes_)"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "            name     acc_x     acc_y     acc_z activity  label\n",
            "0  accelerometer -0.414825  7.590469  6.235336    CHAIR      0\n",
            "1  accelerometer -0.337662  7.645508  6.125275    CHAIR      0\n",
            "2  accelerometer -0.393295  7.605423  6.231155    CHAIR      0\n",
            "3  accelerometer -0.469269  7.592865  6.281403    CHAIR      0\n",
            "4  accelerometer -0.418411  7.633530  6.317307    CHAIR      0\n",
            "['CHAIR' 'LYING_UP' 'STANDING']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "68Fq8PK3vHkE",
        "outputId": "5822b012-acea-4cfa-d89c-93822b0db5eb"
      },
      "source": [
        "glabel=LabelEncoder()\n",
        "gyrdataset['label']=glabel.fit_transform(gyrdataset['activity'])\n",
        "print(gyrdataset.head())\n",
        "print(glabel.classes_)"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "        name     gyr_x     gyr_y     gyr_z activity  label\n",
            "0  gyroscope -0.309540  7.679596  6.142029    CHAIR      0\n",
            "1  gyroscope -0.530884  7.897354  5.765747    CHAIR      0\n",
            "2  gyroscope -0.608643  8.231140  5.266846    CHAIR      0\n",
            "3  gyroscope -0.570953  8.260452  5.229172    CHAIR      0\n",
            "4  gyroscope -0.816223  8.244308  5.284195    CHAIR      0\n",
            "['CHAIR' 'LYING_UP' 'STANDING']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 417
        },
        "id": "upyuAIepv96H",
        "outputId": "e7fde735-18e6-4ece-8a87-661ec2c1110b"
      },
      "source": [
        "AX = accdataset[['acc_x','acc_y','acc_z']]\n",
        "AY = accdataset['label']\n",
        "scaler = StandardScaler()\n",
        "AX = scaler.fit_transform(AX)\n",
        "Ascaled_X = pd.DataFrame(data=AX, columns=['acc_x','acc_y','acc_z'])\n",
        "Ascaled_X['label'] = AY.values\n",
        "Ascaled_X"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>acc_x</th>\n",
              "      <th>acc_y</th>\n",
              "      <th>acc_z</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>-0.407332</td>\n",
              "      <td>0.535859</td>\n",
              "      <td>0.414297</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>-0.342817</td>\n",
              "      <td>0.557656</td>\n",
              "      <td>0.395899</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>-0.389331</td>\n",
              "      <td>0.541781</td>\n",
              "      <td>0.413598</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>-0.452851</td>\n",
              "      <td>0.536808</td>\n",
              "      <td>0.421998</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>-0.410330</td>\n",
              "      <td>0.552913</td>\n",
              "      <td>0.427999</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>94999</th>\n",
              "      <td>-0.532368</td>\n",
              "      <td>1.405810</td>\n",
              "      <td>-0.635258</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>95000</th>\n",
              "      <td>-0.584891</td>\n",
              "      <td>1.400359</td>\n",
              "      <td>-0.640058</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>95001</th>\n",
              "      <td>-0.561876</td>\n",
              "      <td>1.406281</td>\n",
              "      <td>-0.642660</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>95002</th>\n",
              "      <td>-0.572389</td>\n",
              "      <td>1.409841</td>\n",
              "      <td>-0.661458</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>95003</th>\n",
              "      <td>-0.586881</td>\n",
              "      <td>1.394672</td>\n",
              "      <td>-0.664057</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>95004 rows × 4 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "          acc_x     acc_y     acc_z  label\n",
              "0     -0.407332  0.535859  0.414297      0\n",
              "1     -0.342817  0.557656  0.395899      0\n",
              "2     -0.389331  0.541781  0.413598      0\n",
              "3     -0.452851  0.536808  0.421998      0\n",
              "4     -0.410330  0.552913  0.427999      0\n",
              "...         ...       ...       ...    ...\n",
              "94999 -0.532368  1.405810 -0.635258      1\n",
              "95000 -0.584891  1.400359 -0.640058      1\n",
              "95001 -0.561876  1.406281 -0.642660      1\n",
              "95002 -0.572389  1.409841 -0.661458      1\n",
              "95003 -0.586881  1.394672 -0.664057      1\n",
              "\n",
              "[95004 rows x 4 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 417
        },
        "id": "t9qh_ah7xvwO",
        "outputId": "239781a4-df6d-4836-b35f-a670ba1c3eb7"
      },
      "source": [
        "GX = gyrdataset[['gyr_x','gyr_y','gyr_z']]\n",
        "GY = gyrdataset['label']\n",
        "scaler = StandardScaler()\n",
        "GX = scaler.fit_transform(GX)\n",
        "Gscaled_X = pd.DataFrame(data=GX, columns=['gyr_x','gyr_y','gyr_z'])\n",
        "Gscaled_X['label'] = GY.values\n",
        "Gscaled_X"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>gyr_x</th>\n",
              "      <th>gyr_y</th>\n",
              "      <th>gyr_z</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>-0.511514</td>\n",
              "      <td>0.743974</td>\n",
              "      <td>0.380241</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>-0.751855</td>\n",
              "      <td>0.819761</td>\n",
              "      <td>0.323345</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>-0.836288</td>\n",
              "      <td>0.935931</td>\n",
              "      <td>0.247909</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>-0.795364</td>\n",
              "      <td>0.946133</td>\n",
              "      <td>0.242213</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>-1.061684</td>\n",
              "      <td>0.940514</td>\n",
              "      <td>0.250533</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>29620</th>\n",
              "      <td>0.393923</td>\n",
              "      <td>1.434373</td>\n",
              "      <td>-0.432256</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>29621</th>\n",
              "      <td>0.417320</td>\n",
              "      <td>1.446039</td>\n",
              "      <td>-0.434066</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>29622</th>\n",
              "      <td>0.326331</td>\n",
              "      <td>1.445205</td>\n",
              "      <td>-0.435876</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>29623</th>\n",
              "      <td>0.383524</td>\n",
              "      <td>1.451038</td>\n",
              "      <td>-0.441668</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>29624</th>\n",
              "      <td>0.388723</td>\n",
              "      <td>1.450205</td>\n",
              "      <td>-0.443116</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>29625 rows × 4 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "          gyr_x     gyr_y     gyr_z  label\n",
              "0     -0.511514  0.743974  0.380241      0\n",
              "1     -0.751855  0.819761  0.323345      0\n",
              "2     -0.836288  0.935931  0.247909      0\n",
              "3     -0.795364  0.946133  0.242213      0\n",
              "4     -1.061684  0.940514  0.250533      0\n",
              "...         ...       ...       ...    ...\n",
              "29620  0.393923  1.434373 -0.432256      1\n",
              "29621  0.417320  1.446039 -0.434066      1\n",
              "29622  0.326331  1.445205 -0.435876      1\n",
              "29623  0.383524  1.451038 -0.441668      1\n",
              "29624  0.388723  1.450205 -0.443116      1\n",
              "\n",
              "[29625 rows x 4 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rxKOFIVopfnx"
      },
      "source": [
        "import scipy.stats as stats\n",
        "Fs = 20\n",
        "frame_s = Fs*2\n",
        "hop_s = Fs"
      ],
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "n-UOtxBJpw43"
      },
      "source": [
        "def aget_frames(df, frame_s, hop_s):\n",
        "  N = 3\n",
        "  frames=[]\n",
        "  labels=[]\n",
        "  for i in range(0, len(df) - frame_s, hop_s):\n",
        "    x = df['acc_x'].values[i: i+frame_s]\n",
        "    y = df['acc_y'].values[i: i+frame_s]\n",
        "    z = df['acc_z'].values[i: i+frame_s]\n",
        "\n",
        "    label = stats.mode(df['label'][i:i+frame_s])[0][0]\n",
        "    frames.append([x,y,z])\n",
        "    labels.append(label)\n",
        "  \n",
        "  frames = np.asarray(frames).reshape(-1, frame_s, N)\n",
        "  labels = np.asarray(labels)\n",
        "  \n",
        "  return frames, labels"
      ],
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0-OoJK-Pp4Vt"
      },
      "source": [
        "def gget_frames(df, frame_s, hop_s):\n",
        "  N = 3\n",
        "  frames=[]\n",
        "  labels=[]\n",
        "  for i in range(0, len(df) - frame_s, hop_s):\n",
        "    x = df['gyr_x'].values[i: i+frame_s]\n",
        "    y = df['gyr_y'].values[i: i+frame_s]\n",
        "    z = df['gyr_z'].values[i: i+frame_s]\n",
        "\n",
        "    label = stats.mode(df['label'][i:i+frame_s])[0][0]\n",
        "    frames.append([x,y,z])\n",
        "    labels.append(label)\n",
        "  \n",
        "  frames = np.asarray(frames).reshape(-1, frame_s, N)\n",
        "  labels = np.asarray(labels)\n",
        "  \n",
        "  return frames, labels"
      ],
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yONi7nuqp9XK"
      },
      "source": [
        "AX, AY = aget_frames(Ascaled_X, frame_s, hop_s)\n",
        "GX, GY = gget_frames(Gscaled_X, frame_s, hop_s)"
      ],
      "execution_count": 13,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lQYycQoSqA4D",
        "outputId": "6ca4cc4c-68db-4751-d72d-1ebc74b41579"
      },
      "source": [
        "AX.shape, AY.shape"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((4749, 40, 3), (4749,))"
            ]
          },
          "metadata": {},
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Dfg47wwDqFG7",
        "outputId": "a5c0a4e3-acce-4bcb-afda-80120169816a"
      },
      "source": [
        "GX.shape, GY.shape"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((1480, 40, 3), (1480,))"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hlXj1J7BqQxT"
      },
      "source": [
        "AX_train, AX_test, AY_train, AY_test = train_test_split(AX, AY, test_size=0.2, stratify=AY)\n",
        "GX_train, GX_test, GY_train, GY_test = train_test_split(GX, GY, test_size=0.2, stratify=GY)"
      ],
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6hnsjypsqbVq",
        "outputId": "fc6c6513-4d83-4c19-a2cd-b2890622f700"
      },
      "source": [
        "AX_train.shape, AX_test.shape"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((3799, 40, 3), (950, 40, 3))"
            ]
          },
          "metadata": {},
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pMlWrEyrqfOV",
        "outputId": "be12b06f-56ec-49be-e473-d73bdc3d6f03"
      },
      "source": [
        "GX_train.shape, GX_test.shape"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((1184, 40, 3), (296, 40, 3))"
            ]
          },
          "metadata": {},
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sbDxn1WWqh7g"
      },
      "source": [
        "AX_train = AX_train.reshape(len(AX_train), 40, 3, 1)\n",
        "AX_test = AX_test.reshape(len(AX_test), 40, 3, 1)\n",
        "GX_train = GX_train.reshape(len(GX_train), 40, 3, 1)\n",
        "GX_test = GX_test.reshape(len(GX_test), 40, 3, 1)"
      ],
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ws06E_QNqssU"
      },
      "source": [
        "amodel = Sequential()\n",
        "amodel.add(Conv2D(16, (2,2), activation='relu', input_shape = AX_train[0].shape))\n",
        "amodel.add(Dropout(0.1))\n",
        "\n",
        "amodel.add(Conv2D(32, (2,2), activation='relu'))\n",
        "amodel.add(Dropout(0.2))\n",
        "\n",
        "amodel.add(Flatten())\n",
        "\n",
        "amodel.add(Dense(64, activation='relu'))\n",
        "amodel.add(Dropout(0.5))\n",
        "amodel.add(Dense(6, activation='softmax'))"
      ],
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7uI5Mnmgqvuz"
      },
      "source": [
        "amodel.compile(optimizer=Adam(learning_rate = 0.001), loss='sparse_categorical_crossentropy', metrics=['accuracy'])"
      ],
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cUeRbuszqzF2",
        "outputId": "f9c41d70-5052-4849-b19e-5e62a32f4b07"
      },
      "source": [
        "ahistory = amodel.fit(AX_train, AY_train, epochs=20, validation_data=(AX_test, AY_test), verbose=1)"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/20\n",
            "119/119 [==============================] - 2s 9ms/step - loss: 0.7306 - accuracy: 0.6352 - val_loss: 0.5342 - val_accuracy: 0.6495\n",
            "Epoch 2/20\n",
            "119/119 [==============================] - 1s 8ms/step - loss: 0.5712 - accuracy: 0.6957 - val_loss: 0.5062 - val_accuracy: 0.7305\n",
            "Epoch 3/20\n",
            "119/119 [==============================] - 1s 8ms/step - loss: 0.5403 - accuracy: 0.7089 - val_loss: 0.4901 - val_accuracy: 0.7537\n",
            "Epoch 4/20\n",
            "119/119 [==============================] - 1s 7ms/step - loss: 0.5169 - accuracy: 0.7215 - val_loss: 0.4739 - val_accuracy: 0.7547\n",
            "Epoch 5/20\n",
            "119/119 [==============================] - 1s 8ms/step - loss: 0.5017 - accuracy: 0.7394 - val_loss: 0.4720 - val_accuracy: 0.7705\n",
            "Epoch 6/20\n",
            "119/119 [==============================] - 1s 8ms/step - loss: 0.5010 - accuracy: 0.7299 - val_loss: 0.4585 - val_accuracy: 0.7768\n",
            "Epoch 7/20\n",
            "119/119 [==============================] - 1s 7ms/step - loss: 0.4839 - accuracy: 0.7439 - val_loss: 0.4525 - val_accuracy: 0.7832\n",
            "Epoch 8/20\n",
            "119/119 [==============================] - 1s 8ms/step - loss: 0.4850 - accuracy: 0.7470 - val_loss: 0.4586 - val_accuracy: 0.7937\n",
            "Epoch 9/20\n",
            "119/119 [==============================] - 1s 8ms/step - loss: 0.4742 - accuracy: 0.7523 - val_loss: 0.4544 - val_accuracy: 0.7695\n",
            "Epoch 10/20\n",
            "119/119 [==============================] - 1s 8ms/step - loss: 0.4731 - accuracy: 0.7510 - val_loss: 0.4351 - val_accuracy: 0.8000\n",
            "Epoch 11/20\n",
            "119/119 [==============================] - 1s 7ms/step - loss: 0.4656 - accuracy: 0.7465 - val_loss: 0.4300 - val_accuracy: 0.8000\n",
            "Epoch 12/20\n",
            "119/119 [==============================] - 1s 7ms/step - loss: 0.4595 - accuracy: 0.7563 - val_loss: 0.4615 - val_accuracy: 0.7684\n",
            "Epoch 13/20\n",
            "119/119 [==============================] - 1s 8ms/step - loss: 0.4508 - accuracy: 0.7599 - val_loss: 0.4207 - val_accuracy: 0.8000\n",
            "Epoch 14/20\n",
            "119/119 [==============================] - 1s 8ms/step - loss: 0.4449 - accuracy: 0.7676 - val_loss: 0.4222 - val_accuracy: 0.7884\n",
            "Epoch 15/20\n",
            "119/119 [==============================] - 1s 7ms/step - loss: 0.4379 - accuracy: 0.7713 - val_loss: 0.4140 - val_accuracy: 0.7979\n",
            "Epoch 16/20\n",
            "119/119 [==============================] - 1s 7ms/step - loss: 0.4324 - accuracy: 0.7736 - val_loss: 0.4046 - val_accuracy: 0.8074\n",
            "Epoch 17/20\n",
            "119/119 [==============================] - 1s 7ms/step - loss: 0.4351 - accuracy: 0.7773 - val_loss: 0.4088 - val_accuracy: 0.7884\n",
            "Epoch 18/20\n",
            "119/119 [==============================] - 1s 7ms/step - loss: 0.4254 - accuracy: 0.7789 - val_loss: 0.3994 - val_accuracy: 0.7947\n",
            "Epoch 19/20\n",
            "119/119 [==============================] - 1s 8ms/step - loss: 0.4180 - accuracy: 0.7844 - val_loss: 0.3936 - val_accuracy: 0.8063\n",
            "Epoch 20/20\n",
            "119/119 [==============================] - 1s 7ms/step - loss: 0.4136 - accuracy: 0.7860 - val_loss: 0.4022 - val_accuracy: 0.7958\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WeXCjRplriJ1"
      },
      "source": [
        "gmodel = Sequential()\n",
        "gmodel.add(Conv2D(16, (2,2), activation='relu', input_shape = GX_train[0].shape))\n",
        "gmodel.add(Dropout(0.1))\n",
        "\n",
        "gmodel.add(Conv2D(32, (2,2), activation='relu'))\n",
        "gmodel.add(Dropout(0.2))\n",
        "\n",
        "gmodel.add(Flatten())\n",
        "\n",
        "gmodel.add(Dense(64, activation='relu'))\n",
        "gmodel.add(Dropout(0.5))\n",
        "gmodel.add(Dense(6, activation='softmax'))"
      ],
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mMZFJxxTrtlR"
      },
      "source": [
        "gmodel.compile(optimizer=Adam(learning_rate = 0.001), loss='sparse_categorical_crossentropy', metrics=['accuracy'])"
      ],
      "execution_count": 24,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "i9U3lITTryLQ",
        "outputId": "947b52fc-89b4-4f56-c2f2-cd72aa205f6a"
      },
      "source": [
        "ghistory = gmodel.fit(GX_train, GY_train, epochs=20, validation_data=(GX_test, GY_test), verbose=1)"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/20\n",
            "37/37 [==============================] - 1s 16ms/step - loss: 0.9979 - accuracy: 0.5777 - val_loss: 0.6525 - val_accuracy: 0.7027\n",
            "Epoch 2/20\n",
            "37/37 [==============================] - 0s 8ms/step - loss: 0.7023 - accuracy: 0.6512 - val_loss: 0.5661 - val_accuracy: 0.6723\n",
            "Epoch 3/20\n",
            "37/37 [==============================] - 0s 8ms/step - loss: 0.6369 - accuracy: 0.6647 - val_loss: 0.5465 - val_accuracy: 0.6993\n",
            "Epoch 4/20\n",
            "37/37 [==============================] - 0s 6ms/step - loss: 0.5988 - accuracy: 0.6951 - val_loss: 0.5119 - val_accuracy: 0.7500\n",
            "Epoch 5/20\n",
            "37/37 [==============================] - 0s 7ms/step - loss: 0.5445 - accuracy: 0.7221 - val_loss: 0.5193 - val_accuracy: 0.7534\n",
            "Epoch 6/20\n",
            "37/37 [==============================] - 0s 6ms/step - loss: 0.5289 - accuracy: 0.7348 - val_loss: 0.5107 - val_accuracy: 0.7264\n",
            "Epoch 7/20\n",
            "37/37 [==============================] - 0s 7ms/step - loss: 0.5015 - accuracy: 0.7542 - val_loss: 0.5107 - val_accuracy: 0.7196\n",
            "Epoch 8/20\n",
            "37/37 [==============================] - 0s 8ms/step - loss: 0.5021 - accuracy: 0.7348 - val_loss: 0.4956 - val_accuracy: 0.7331\n",
            "Epoch 9/20\n",
            "37/37 [==============================] - 0s 9ms/step - loss: 0.4993 - accuracy: 0.7508 - val_loss: 0.4988 - val_accuracy: 0.7365\n",
            "Epoch 10/20\n",
            "37/37 [==============================] - 0s 10ms/step - loss: 0.4736 - accuracy: 0.7610 - val_loss: 0.4768 - val_accuracy: 0.7703\n",
            "Epoch 11/20\n",
            "37/37 [==============================] - 0s 7ms/step - loss: 0.4703 - accuracy: 0.7627 - val_loss: 0.4667 - val_accuracy: 0.7770\n",
            "Epoch 12/20\n",
            "37/37 [==============================] - 0s 8ms/step - loss: 0.4419 - accuracy: 0.7889 - val_loss: 0.4680 - val_accuracy: 0.7703\n",
            "Epoch 13/20\n",
            "37/37 [==============================] - 0s 8ms/step - loss: 0.4522 - accuracy: 0.7863 - val_loss: 0.4527 - val_accuracy: 0.7973\n",
            "Epoch 14/20\n",
            "37/37 [==============================] - 0s 9ms/step - loss: 0.4176 - accuracy: 0.8066 - val_loss: 0.4747 - val_accuracy: 0.7872\n",
            "Epoch 15/20\n",
            "37/37 [==============================] - 0s 9ms/step - loss: 0.4148 - accuracy: 0.8083 - val_loss: 0.4165 - val_accuracy: 0.8277\n",
            "Epoch 16/20\n",
            "37/37 [==============================] - 0s 9ms/step - loss: 0.4279 - accuracy: 0.7965 - val_loss: 0.4662 - val_accuracy: 0.8243\n",
            "Epoch 17/20\n",
            "37/37 [==============================] - 0s 8ms/step - loss: 0.4210 - accuracy: 0.8074 - val_loss: 0.4405 - val_accuracy: 0.7838\n",
            "Epoch 18/20\n",
            "37/37 [==============================] - 0s 8ms/step - loss: 0.4096 - accuracy: 0.7981 - val_loss: 0.4498 - val_accuracy: 0.8074\n",
            "Epoch 19/20\n",
            "37/37 [==============================] - 0s 8ms/step - loss: 0.4047 - accuracy: 0.8142 - val_loss: 0.4394 - val_accuracy: 0.8142\n",
            "Epoch 20/20\n",
            "37/37 [==============================] - 0s 8ms/step - loss: 0.3751 - accuracy: 0.8252 - val_loss: 0.4643 - val_accuracy: 0.8074\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2Pud3oIitfXT"
      },
      "source": [
        "def plot_accuracycurve(a, history, epochs):\n",
        "  ep = range(1, epochs+1)\n",
        "  plt.title(a+' Model Accuracy')\n",
        "  plt.plot(ep, history.history['accuracy'])\n",
        "  plt.plot(ep, history.history['val_accuracy'])\n",
        "  plt.ylabel('Accuracy')\n",
        "  plt.ylabel('Epochs')\n",
        "  plt.legend(['Train','Val'])\n",
        "  plt.show()\n",
        "\n",
        "def plot_losscurve(a, history, epochs):\n",
        "  ep = range(1, epochs+1)\n",
        "  plt.title(a+' Model Loss')\n",
        "  plt.plot(ep, history.history['loss'])\n",
        "  plt.plot(ep, history.history['val_loss'])\n",
        "  plt.ylabel('Loss')\n",
        "  plt.ylabel('Epochs')\n",
        "  plt.legend(['Train','Val'])\n",
        "  plt.show()"
      ],
      "execution_count": 26,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 545
        },
        "id": "CIeeu22fumO4",
        "outputId": "19ec281c-cb76-4c44-da8f-204333f5cb91"
      },
      "source": [
        "plot_accuracycurve('Accelerometer',ahistory, 20)\n",
        "plot_losscurve('Accelerometer', ahistory, 20)"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 545
        },
        "id": "1QfJARCtvmgN",
        "outputId": "5ce674ef-eb55-4b22-d185-536926a39b34"
      },
      "source": [
        "plot_accuracycurve('Gyroscope',ghistory, 20)\n",
        "plot_losscurve('Gyroscope', ghistory, 20)"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3WD0GIFgwMRh",
        "outputId": "54299735-ce02-4350-fcb8-de731f901b8c"
      },
      "source": [
        "pip install mlxtend"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: mlxtend in /usr/local/lib/python3.7/dist-packages (0.14.0)\n",
            "Requirement already satisfied: pandas>=0.17.1 in /usr/local/lib/python3.7/dist-packages (from mlxtend) (1.1.5)\n",
            "Requirement already satisfied: scikit-learn>=0.18 in /usr/local/lib/python3.7/dist-packages (from mlxtend) (0.22.2.post1)\n",
            "Requirement already satisfied: numpy>=1.10.4 in /usr/local/lib/python3.7/dist-packages (from mlxtend) (1.19.5)\n",
            "Requirement already satisfied: setuptools in /usr/local/lib/python3.7/dist-packages (from mlxtend) (57.4.0)\n",
            "Requirement already satisfied: scipy>=0.17 in /usr/local/lib/python3.7/dist-packages (from mlxtend) (1.4.1)\n",
            "Requirement already satisfied: matplotlib>=1.5.1 in /usr/local/lib/python3.7/dist-packages (from mlxtend) (3.2.2)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib>=1.5.1->mlxtend) (1.3.1)\n",
            "Requirement already satisfied: python-dateutil>=2.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib>=1.5.1->mlxtend) (2.8.2)\n",
            "Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib>=1.5.1->mlxtend) (2.4.7)\n",
            "Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.7/dist-packages (from matplotlib>=1.5.1->mlxtend) (0.10.0)\n",
            "Requirement already satisfied: six in /usr/local/lib/python3.7/dist-packages (from cycler>=0.10->matplotlib>=1.5.1->mlxtend) (1.15.0)\n",
            "Requirement already satisfied: pytz>=2017.2 in /usr/local/lib/python3.7/dist-packages (from pandas>=0.17.1->mlxtend) (2018.9)\n",
            "Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.7/dist-packages (from scikit-learn>=0.18->mlxtend) (1.0.1)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rvO-dmkRwYIq"
      },
      "source": [
        "from mlxtend.plotting import plot_confusion_matrix\n",
        "from sklearn.metrics import confusion_matrix"
      ],
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZVuSLYbiw25_"
      },
      "source": [
        "AY_P = amodel.predict(AX_test)\n",
        "AY_predict=np.argmax(AY_P,axis=1)\n",
        "GY_P = gmodel.predict(GX_test)\n",
        "GY_predict=np.argmax(GY_P,axis=1)"
      ],
      "execution_count": 31,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 316
        },
        "id": "lxalTJUGxzjN",
        "outputId": "274d1023-025d-4de1-c44a-1a2c6981de88"
      },
      "source": [
        "amat = confusion_matrix(AY_test, AY_predict)\n",
        "plot_confusion_matrix(conf_mat=amat, show_normed=True, colorbar=True)"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(<Figure size 432x288 with 2 Axes>,\n",
              " <matplotlib.axes._subplots.AxesSubplot at 0x7f9562015590>)"
            ]
          },
          "metadata": {},
          "execution_count": 32
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Z_3SclqLt8ks",
        "outputId": "4b456796-e52e-4a24-b608-2cc5c00d5485"
      },
      "source": [
        "print(alabel.classes_)"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "['CHAIR' 'LYING_UP' 'STANDING']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 313
        },
        "id": "dS2e6jk30R2M",
        "outputId": "e550c64a-1466-47eb-c333-f7d3410d1306"
      },
      "source": [
        "gmat = confusion_matrix(GY_test, GY_predict)\n",
        "plot_confusion_matrix(conf_mat=gmat, show_normed=True, colorbar=True)"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(<Figure size 432x288 with 2 Axes>,\n",
              " <matplotlib.axes._subplots.AxesSubplot at 0x7f95618bdad0>)"
            ]
          },
          "metadata": {},
          "execution_count": 34
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EAubsRItKAGR",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "cc731116-6e30-4c8a-f9a1-646314204351"
      },
      "source": [
        "print(glabel.classes_)"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "['CHAIR' 'LYING_UP' 'STANDING']\n"
          ]
        }
      ]
    }
  ]
}