{
"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"colab": {
"provenance": []
},
"kernelspec": {
"name": "python3",
"display_name": "Python 3"
},
"language_info": {
"name": "python"
},
"accelerator": "GPU",
"gpuClass": "standard"
},
"cells": [
{
"cell_type": "markdown",
"source": [
"# Applying LSTM Models on Raw Data"
],
"metadata": {
"id": "jUkGXVGfU1xN"
}
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "7USnX2QTSuKt"
},
"outputs": [],
"source": [
"# Importing Libraries\n",
"\n",
"import pandas as pd\n",
"import numpy as np\n",
"\n",
"# Import Keras\n",
"from keras import backend as K\n",
"from keras.models import Sequential\n",
"from keras.layers import LSTM\n",
"from keras.layers.core import Dense, Dropout\n",
"from keras.layers import BatchNormalization\n",
"from keras.regularizers import L1L2"
]
},
{
"cell_type": "code",
"source": [
"# Activities are the class labels\n",
"# It is a 6 class classification\n",
"ACTIVITIES = {\n",
" 0: 'WALKING',\n",
" 1: 'WALKING_UPSTAIRS',\n",
" 2: 'WALKING_DOWNSTAIRS',\n",
" 3: 'SITTING',\n",
" 4: 'STANDING',\n",
" 5: 'LAYING',\n",
"}"
],
"metadata": {
"id": "UklmP7-eU9Wm"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"import matplotlib.pyplot as plt\n",
"import seaborn as sns\n",
"\n",
"# function to print the confusion matrix\n",
"\n",
"def confusion_matrix(Y_true, Y_pred):\n",
" \n",
" Y_true = pd.Series([ACTIVITIES[y] for y in np.argmax(Y_true, axis=1)])\n",
" Y_pred = pd.Series([ACTIVITIES[y] for y in np.argmax(Y_pred, axis=1)])\n",
"\n",
" return pd.crosstab(Y_true, Y_pred, rownames=['True'], colnames=['Pred'])\n",
"\n",
" \n",
" result = confusion_matrix(Y_true, Y_pred)\n",
"\n",
" plt.figure(figsize=(10, 8))\n",
" sns.heatmap(result, \n",
" xticklabels= list(ACTIVITIES.values()), \n",
" yticklabels=list(ACTIVITIES.values()), \n",
" annot=True, fmt=\"d\");\n",
" plt.title(\"Confusion matrix\")\n",
" plt.ylabel('True label')\n",
" plt.xlabel('Predicted label')\n",
" plt.show() "
],
"metadata": {
"id": "cG79tQGXVASE"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"### Loading Data"
],
"metadata": {
"id": "dfbEhMvGVG4K"
}
},
{
"cell_type": "code",
"source": [
"# Data directory\n",
"DATADIR = 'UCI_HAR_Dataset'\n",
"\n",
"# Raw data signals\n",
"# Signals are from Accelerometer and Gyroscope\n",
"# The signals are in x,y,z directions\n",
"# Sensor signals are filtered to have only body acceleration\n",
"# excluding the acceleration due to gravity\n",
"# Triaxial acceleration from the accelerometer is total acceleration\n",
"SIGNALS = [\n",
" \"body_acc_x\",\n",
" \"body_acc_y\",\n",
" \"body_acc_z\",\n",
" \"body_gyro_x\",\n",
" \"body_gyro_y\",\n",
" \"body_gyro_z\",\n",
" \"total_acc_x\",\n",
" \"total_acc_y\",\n",
" \"total_acc_z\"\n",
" ]"
],
"metadata": {
"id": "W6E6gq2tVJjl"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"# define a function to read the data from csv file\n",
"def _read_csv(filename):\n",
" return pd.read_csv(filename, delim_whitespace=True, header=None)\n",
"\n",
"# function to load the load\n",
"def load_signals(subset):\n",
" signals_data = []\n",
"\n",
" for signal in SIGNALS:\n",
" filename = f'/content/drive/MyDrive/UCI_HAR_Dataset/{subset}/Inertial Signals/{signal}_{subset}.txt'\n",
" signals_data.append(\n",
" _read_csv(filename).to_numpy()\n",
" ) \n",
"\n",
" # Transpose is used to change the dimensionality of the output,\n",
" # aggregating the signals by combination of sample/timestep.\n",
" # Resultant shape is (7352 train/2947 test samples, 128 timesteps, 9 signals)\n",
" return np.transpose(signals_data, (1, 2, 0))"
],
"metadata": {
"id": "Gbp0kyOLVO13"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"def load_y(subset):\n",
" \"\"\"\n",
" The objective that we are trying to predict is a integer, from 1 to 6,\n",
" that represents a human activity. We return a binary representation of \n",
" every sample objective as a 6 bits vector using One Hot Encoding\n",
" (https://pandas.pydata.org/pandas-docs/stable/generated/pandas.get_dummies.html)\n",
" \"\"\"\n",
" filename = f'/content/drive/MyDrive/UCI_HAR_Dataset/{subset}/y_{subset}.txt'\n",
" y = _read_csv(filename)[0]\n",
"\n",
" return pd.get_dummies(y).to_numpy()"
],
"metadata": {
"id": "MOXrPORRVRcJ"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"def load_data():\n",
" \"\"\"\n",
" Obtain the dataset from multiple files.\n",
" Returns: X_train, X_test, y_train, y_test\n",
" \"\"\"\n",
" X_train, X_test = load_signals('train'), load_signals('test')\n",
" y_train, y_test = load_y('train'), load_y('test')\n",
"\n",
" return X_train, X_test, y_train, y_test"
],
"metadata": {
"id": "MEf5hg9lVTun"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"# Importing tensorflow\n",
"np.random.seed(42)\n",
"import tensorflow as tf\n",
"tf.random.set_seed(42)"
],
"metadata": {
"id": "Edexn_grVWug"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"# Initializing parameters\n",
"epochs = 30\n",
"batch_size = 16\n",
"n_hidden = 32"
],
"metadata": {
"id": "B4dT-A4bVapl"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"#function to count the number of classes\n",
"def _count_classes(y):\n",
" return len(set([tuple(category) for category in y]))"
],
"metadata": {
"id": "5D6pcPuVVbyl"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"# Loading the train and test data\n",
"X_train, X_test, y_train, y_test = load_data()"
],
"metadata": {
"id": "VUoSMvSfVga3"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"timesteps = len(X_train[0])\n",
"input_dim = len(X_train[0][0])\n",
"n_classes = _count_classes(y_train)\n",
"\n",
"print(timesteps)\n",
"print(input_dim)\n",
"print(len(X_train))"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "MXby1ubyVjBV",
"outputId": "8f44a692-57c1-4df3-b11f-251c4979e19e"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"128\n",
"9\n",
"7352\n"
]
}
]
},
{
"cell_type": "markdown",
"source": [
" 1. Defining the Architecture of 1-Layer of LSTM"
],
"metadata": {
"id": "m182sLnyVl0K"
}
},
{
"cell_type": "code",
"source": [
"# Initiliazing the sequential model\n",
"model = Sequential()\n",
"# Configuring the parameters\n",
"model.add(LSTM(n_hidden, input_shape=(timesteps, input_dim)))\n",
"# Adding a dropout layer\n",
"model.add(Dropout(0.5))\n",
"# Adding a dense output layer with sigmoid activation\n",
"model.add(Dense(n_classes, activation='sigmoid'))\n",
"model.summary()"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "BGD2Lt3MVnK5",
"outputId": "f7ece6dd-85c8-4d41-a713-540bea33ddf0"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"Model: \"sequential_3\"\n",
"_________________________________________________________________\n",
" Layer (type) Output Shape Param # \n",
"=================================================================\n",
" lstm_5 (LSTM) (None, 32) 5376 \n",
" \n",
" dropout_5 (Dropout) (None, 32) 0 \n",
" \n",
" dense_3 (Dense) (None, 6) 198 \n",
" \n",
"=================================================================\n",
"Total params: 5,574\n",
"Trainable params: 5,574\n",
"Non-trainable params: 0\n",
"_________________________________________________________________\n"
]
}
]
},
{
"cell_type": "code",
"source": [
"# Compiling the model\n",
"model.compile(loss='categorical_crossentropy', optimizer='rmsprop', metrics=['accuracy'])"
],
"metadata": {
"id": "MIX8hyRoVrSs"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"# Training the model\n",
"model.fit(X_train, y_train, batch_size=batch_size, validation_data=(X_test, y_test),epochs=epochs)"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "RxoV_8fdVt7u",
"outputId": "fc88555b-3bcc-47b6-f10e-1aa6d752887e"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"Epoch 1/30\n",
"460/460 [==============================] - 8s 11ms/step - loss: 1.0781 - accuracy: 0.5423 - val_loss: 0.9016 - val_accuracy: 0.6213\n",
"Epoch 2/30\n",
"460/460 [==============================] - 4s 10ms/step - loss: 0.7431 - accuracy: 0.6632 - val_loss: 0.6547 - val_accuracy: 0.7255\n",
"Epoch 3/30\n",
"460/460 [==============================] - 5s 12ms/step - loss: 0.5874 - accuracy: 0.7629 - val_loss: 0.5316 - val_accuracy: 0.7906\n",
"Epoch 4/30\n",
"460/460 [==============================] - 4s 8ms/step - loss: 0.4666 - accuracy: 0.8320 - val_loss: 1.4815 - val_accuracy: 0.6155\n",
"Epoch 5/30\n",
"460/460 [==============================] - 4s 8ms/step - loss: 0.3551 - accuracy: 0.8863 - val_loss: 0.4364 - val_accuracy: 0.8483\n",
"Epoch 6/30\n",
"460/460 [==============================] - 6s 12ms/step - loss: 0.2936 - accuracy: 0.9067 - val_loss: 0.3727 - val_accuracy: 0.8765\n",
"Epoch 7/30\n",
"460/460 [==============================] - 4s 8ms/step - loss: 0.2671 - accuracy: 0.9174 - val_loss: 0.4289 - val_accuracy: 0.8639\n",
"Epoch 8/30\n",
"460/460 [==============================] - 4s 8ms/step - loss: 0.2423 - accuracy: 0.9266 - val_loss: 0.3437 - val_accuracy: 0.8711\n",
"Epoch 9/30\n",
"460/460 [==============================] - 5s 11ms/step - loss: 0.2158 - accuracy: 0.9293 - val_loss: 0.3924 - val_accuracy: 0.8826\n",
"Epoch 10/30\n",
"460/460 [==============================] - 4s 10ms/step - loss: 0.2087 - accuracy: 0.9343 - val_loss: 0.3207 - val_accuracy: 0.8856\n",
"Epoch 11/30\n",
"460/460 [==============================] - 4s 10ms/step - loss: 0.1990 - accuracy: 0.9344 - val_loss: 0.3968 - val_accuracy: 0.8704\n",
"Epoch 12/30\n",
"460/460 [==============================] - 5s 11ms/step - loss: 0.1726 - accuracy: 0.9376 - val_loss: 0.2942 - val_accuracy: 0.8931\n",
"Epoch 13/30\n",
"460/460 [==============================] - 4s 9ms/step - loss: 0.1755 - accuracy: 0.9414 - val_loss: 0.2492 - val_accuracy: 0.9002\n",
"Epoch 14/30\n",
"460/460 [==============================] - 4s 10ms/step - loss: 0.1585 - accuracy: 0.9465 - val_loss: 0.3617 - val_accuracy: 0.8904\n",
"Epoch 15/30\n",
"460/460 [==============================] - 5s 11ms/step - loss: 0.1629 - accuracy: 0.9468 - val_loss: 0.5414 - val_accuracy: 0.8724\n",
"Epoch 16/30\n",
"460/460 [==============================] - 4s 8ms/step - loss: 0.1647 - accuracy: 0.9479 - val_loss: 0.3329 - val_accuracy: 0.8979\n",
"Epoch 17/30\n",
"460/460 [==============================] - 4s 8ms/step - loss: 0.1531 - accuracy: 0.9486 - val_loss: 0.3276 - val_accuracy: 0.9043\n",
"Epoch 18/30\n",
"460/460 [==============================] - 4s 10ms/step - loss: 0.1623 - accuracy: 0.9459 - val_loss: 0.2336 - val_accuracy: 0.9125\n",
"Epoch 19/30\n",
"460/460 [==============================] - 4s 9ms/step - loss: 0.1494 - accuracy: 0.9489 - val_loss: 0.3656 - val_accuracy: 0.9013\n",
"Epoch 20/30\n",
"460/460 [==============================] - 4s 8ms/step - loss: 0.1543 - accuracy: 0.9480 - val_loss: 0.3468 - val_accuracy: 0.9016\n",
"Epoch 21/30\n",
"460/460 [==============================] - 4s 9ms/step - loss: 0.1398 - accuracy: 0.9490 - val_loss: 0.3211 - val_accuracy: 0.9057\n",
"Epoch 22/30\n",
"460/460 [==============================] - 5s 10ms/step - loss: 0.1520 - accuracy: 0.9479 - val_loss: 0.2789 - val_accuracy: 0.9067\n",
"Epoch 23/30\n",
"460/460 [==============================] - 4s 8ms/step - loss: 0.1479 - accuracy: 0.9475 - val_loss: 0.2733 - val_accuracy: 0.9074\n",
"Epoch 24/30\n",
"460/460 [==============================] - 4s 10ms/step - loss: 0.1425 - accuracy: 0.9493 - val_loss: 0.3381 - val_accuracy: 0.9036\n",
"Epoch 25/30\n",
"460/460 [==============================] - 5s 12ms/step - loss: 0.1437 - accuracy: 0.9504 - val_loss: 0.3459 - val_accuracy: 0.8921\n",
"Epoch 26/30\n",
"460/460 [==============================] - 4s 8ms/step - loss: 0.1390 - accuracy: 0.9497 - val_loss: 0.3192 - val_accuracy: 0.9074\n",
"Epoch 27/30\n",
"460/460 [==============================] - 4s 10ms/step - loss: 0.1361 - accuracy: 0.9520 - val_loss: 0.3324 - val_accuracy: 0.9189\n",
"Epoch 28/30\n",
"460/460 [==============================] - 6s 12ms/step - loss: 0.1469 - accuracy: 0.9509 - val_loss: 0.3501 - val_accuracy: 0.9097\n",
"Epoch 29/30\n",
"460/460 [==============================] - 5s 10ms/step - loss: 0.1439 - accuracy: 0.9438 - val_loss: 0.2556 - val_accuracy: 0.9308\n",
"Epoch 30/30\n",
"460/460 [==============================] - 4s 10ms/step - loss: 0.1434 - accuracy: 0.9486 - val_loss: 0.3141 - val_accuracy: 0.9165\n"
]
},
{
"output_type": "execute_result",
"data": {
"text/plain": [
"<keras.callbacks.History at 0x7f78fa564e80>"
]
},
"metadata": {},
"execution_count": 52
}
]
},
{
"cell_type": "code",
"source": [
"# Confusion Matrix\n",
"confusion_matrix(y_test, model.predict(X_test))"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 286
},
"id": "Pl6Y6R3zVwzE",
"outputId": "d50b38f5-1af7-4cd8-c5f5-771190191675"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"93/93 [==============================] - 1s 4ms/step\n"
]
},
{
"output_type": "execute_result",
"data": {
"text/plain": [
"Pred LAYING SITTING STANDING WALKING WALKING_DOWNSTAIRS \\\n",
"True \n",
"LAYING 511 0 26 0 0 \n",
"SITTING 1 409 76 4 0 \n",
"STANDING 0 95 433 4 0 \n",
"WALKING 0 0 0 480 15 \n",
"WALKING_DOWNSTAIRS 0 0 0 0 419 \n",
"WALKING_UPSTAIRS 0 1 0 12 9 \n",
"\n",
"Pred WALKING_UPSTAIRS \n",
"True \n",
"LAYING 0 \n",
"SITTING 1 \n",
"STANDING 0 \n",
"WALKING 1 \n",
"WALKING_DOWNSTAIRS 1 \n",
"WALKING_UPSTAIRS 449 "
],
"text/html": [
"\n",
" <div id=\"df-ebce1837-967c-4ffc-872b-b24d45f84c89\">\n",
" <div class=\"colab-df-container\">\n",
" <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>Pred</th>\n",
" <th>LAYING</th>\n",
" <th>SITTING</th>\n",
" <th>STANDING</th>\n",
" <th>WALKING</th>\n",
" <th>WALKING_DOWNSTAIRS</th>\n",
" <th>WALKING_UPSTAIRS</th>\n",
" </tr>\n",
" <tr>\n",
" <th>True</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>LAYING</th>\n",
" <td>511</td>\n",
" <td>0</td>\n",
" <td>26</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>SITTING</th>\n",
" <td>1</td>\n",
" <td>409</td>\n",
" <td>76</td>\n",
" <td>4</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <th>STANDING</th>\n",
" <td>0</td>\n",
" <td>95</td>\n",
" <td>433</td>\n",
" <td>4</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>WALKING</th>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>480</td>\n",
" <td>15</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <th>WALKING_DOWNSTAIRS</th>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>419</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <th>WALKING_UPSTAIRS</th>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>12</td>\n",
" <td>9</td>\n",
" <td>449</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>\n",
" <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-ebce1837-967c-4ffc-872b-b24d45f84c89')\"\n",
" title=\"Convert this dataframe to an interactive table.\"\n",
" style=\"display:none;\">\n",
" \n",
" <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
" width=\"24px\">\n",
" <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
" <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
" </svg>\n",
" </button>\n",
" \n",
" <style>\n",
" .colab-df-container {\n",
" display:flex;\n",
" flex-wrap:wrap;\n",
" gap: 12px;\n",
" }\n",
"\n",
" .colab-df-convert {\n",
" background-color: #E8F0FE;\n",
" border: none;\n",
" border-radius: 50%;\n",
" cursor: pointer;\n",
" display: none;\n",
" fill: #1967D2;\n",
" height: 32px;\n",
" padding: 0 0 0 0;\n",
" width: 32px;\n",
" }\n",
"\n",
" .colab-df-convert:hover {\n",
" background-color: #E2EBFA;\n",
" box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
" fill: #174EA6;\n",
" }\n",
"\n",
" [theme=dark] .colab-df-convert {\n",
" background-color: #3B4455;\n",
" fill: #D2E3FC;\n",
" }\n",
"\n",
" [theme=dark] .colab-df-convert:hover {\n",
" background-color: #434B5C;\n",
" box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
" filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
" fill: #FFFFFF;\n",
" }\n",
" </style>\n",
"\n",
" <script>\n",
" const buttonEl =\n",
" document.querySelector('#df-ebce1837-967c-4ffc-872b-b24d45f84c89 button.colab-df-convert');\n",
" buttonEl.style.display =\n",
" google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
"\n",
" async function convertToInteractive(key) {\n",
" const element = document.querySelector('#df-ebce1837-967c-4ffc-872b-b24d45f84c89');\n",
" const dataTable =\n",
" await google.colab.kernel.invokeFunction('convertToInteractive',\n",
" [key], {});\n",
" if (!dataTable) return;\n",
"\n",
" const docLinkHtml = 'Like what you see? Visit the ' +\n",
" '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
" + ' to learn more about interactive tables.';\n",
" element.innerHTML = '';\n",
" dataTable['output_type'] = 'display_data';\n",
" await google.colab.output.renderOutput(dataTable, element);\n",
" const docLink = document.createElement('div');\n",
" docLink.innerHTML = docLinkHtml;\n",
" element.appendChild(docLink);\n",
" }\n",
" </script>\n",
" </div>\n",
" </div>\n",
" "
]
},
"metadata": {},
"execution_count": 53
}
]
},
{
"cell_type": "code",
"source": [
"score = model.evaluate(X_test, y_test)\n",
"\n",
"print(\"\\n categorical_crossentropy || accuracy \")\n",
"print(\" ____________________________________\")\n",
"print(score)"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "hwFZ64M7V0dN",
"outputId": "7d42242a-fe62-4172-ca6b-f328c8e9412a"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"93/93 [==============================] - 0s 5ms/step - loss: 0.3141 - accuracy: 0.9165\n",
"\n",
" categorical_crossentropy || accuracy \n",
" ____________________________________\n",
"[0.3141055107116699, 0.9165253043174744]\n"
]
}
]
},
{
"cell_type": "markdown",
"source": [
" 2. Defining the Architecture of 2-Layer of LSTM with more hyperparameter tunning"
],
"metadata": {
"id": "EFsp0FcXV9HG"
}
},
{
"cell_type": "markdown",
"source": [
"#### 2.1 First Model for 2-Layer of LSTM with more hyperparameter tunning"
],
"metadata": {
"id": "8p4j41aSV_Uv"
}
},
{
"cell_type": "code",
"source": [
"# Initializing parameters\n",
"n_epochs = 30\n",
"n_batch = 16\n",
"n_classes = _count_classes(y_train)\n",
"\n",
"# Bias regularizer value - we will use elasticnet\n",
"reg = L1L2(0.01, 0.01)"
],
"metadata": {
"id": "O594yNV2WSRj"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"# Model execution\n",
"model = Sequential()\n",
"model.add(LSTM(48, input_shape=(timesteps, input_dim), return_sequences=True,bias_regularizer=reg ))\n",
"model.add(BatchNormalization())\n",
"model.add(Dropout(0.50))\n",
"model.add(LSTM(32))\n",
"model.add(Dropout(0.50))\n",
"model.add(Dense(n_classes, activation='sigmoid'))\n",
"print(\"Model Summary: \")\n",
"model.summary()"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "17yKQHdAWVWs",
"outputId": "48627f4b-09b1-465c-d1ab-cd2142af8642"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"Model Summary: \n",
"Model: \"sequential_4\"\n",
"_________________________________________________________________\n",
" Layer (type) Output Shape Param # \n",
"=================================================================\n",
" lstm_6 (LSTM) (None, 128, 48) 11136 \n",
" \n",
" batch_normalization_2 (Batc (None, 128, 48) 192 \n",
" hNormalization) \n",
" \n",
" dropout_6 (Dropout) (None, 128, 48) 0 \n",
" \n",
" lstm_7 (LSTM) (None, 32) 10368 \n",
" \n",
" dropout_7 (Dropout) (None, 32) 0 \n",
" \n",
" dense_4 (Dense) (None, 6) 198 \n",
" \n",
"=================================================================\n",
"Total params: 21,894\n",
"Trainable params: 21,798\n",
"Non-trainable params: 96\n",
"_________________________________________________________________\n"
]
}
]
},
{
"cell_type": "code",
"source": [
"model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])"
],
"metadata": {
"id": "HfXE9HTbWYMc"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"# Training the model\n",
"model.fit(X_train, y_train, batch_size=n_batch, validation_data=(X_test, y_test), epochs=n_epochs)"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "TQxn8R_VWbJA",
"outputId": "b9186f7f-87a8-4e68-f35e-63fd06ca7efb"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"Epoch 1/30\n",
"460/460 [==============================] - 14s 17ms/step - loss: 1.5191 - accuracy: 0.6938 - val_loss: 0.9805 - val_accuracy: 0.8310\n",
"Epoch 2/30\n",
"460/460 [==============================] - 7s 16ms/step - loss: 0.7182 - accuracy: 0.8690 - val_loss: 0.5508 - val_accuracy: 0.8904\n",
"Epoch 3/30\n",
"460/460 [==============================] - 6s 13ms/step - loss: 0.3774 - accuracy: 0.9128 - val_loss: 0.3188 - val_accuracy: 0.8965\n",
"Epoch 4/30\n",
"460/460 [==============================] - 8s 17ms/step - loss: 0.2595 - accuracy: 0.9225 - val_loss: 0.4659 - val_accuracy: 0.8405\n",
"Epoch 5/30\n",
"460/460 [==============================] - 6s 14ms/step - loss: 0.2368 - accuracy: 0.9187 - val_loss: 0.4207 - val_accuracy: 0.8565\n",
"Epoch 6/30\n",
"460/460 [==============================] - 8s 17ms/step - loss: 0.2269 - accuracy: 0.9234 - val_loss: 0.2667 - val_accuracy: 0.9036\n",
"Epoch 7/30\n",
"460/460 [==============================] - 6s 14ms/step - loss: 0.1687 - accuracy: 0.9374 - val_loss: 0.2469 - val_accuracy: 0.9125\n",
"Epoch 8/30\n",
"460/460 [==============================] - 7s 16ms/step - loss: 0.1691 - accuracy: 0.9387 - val_loss: 0.3808 - val_accuracy: 0.8884\n",
"Epoch 9/30\n",
"460/460 [==============================] - 7s 14ms/step - loss: 0.1535 - accuracy: 0.9422 - val_loss: 0.2922 - val_accuracy: 0.9060\n",
"Epoch 10/30\n",
"460/460 [==============================] - 7s 16ms/step - loss: 0.1839 - accuracy: 0.9350 - val_loss: 0.3028 - val_accuracy: 0.8935\n",
"Epoch 11/30\n",
"460/460 [==============================] - 7s 15ms/step - loss: 0.1648 - accuracy: 0.9408 - val_loss: 0.3396 - val_accuracy: 0.8860\n",
"Epoch 12/30\n",
"460/460 [==============================] - 7s 16ms/step - loss: 0.1487 - accuracy: 0.9442 - val_loss: 0.2619 - val_accuracy: 0.9148\n",
"Epoch 13/30\n",
"460/460 [==============================] - 8s 18ms/step - loss: 0.1564 - accuracy: 0.9393 - val_loss: 0.2611 - val_accuracy: 0.9131\n",
"Epoch 14/30\n",
"460/460 [==============================] - 7s 14ms/step - loss: 0.1492 - accuracy: 0.9418 - val_loss: 0.3017 - val_accuracy: 0.9155\n",
"Epoch 15/30\n",
"460/460 [==============================] - 7s 16ms/step - loss: 0.1759 - accuracy: 0.9370 - val_loss: 0.3169 - val_accuracy: 0.9182\n",
"Epoch 16/30\n",
"460/460 [==============================] - 7s 14ms/step - loss: 0.1700 - accuracy: 0.9396 - val_loss: 0.3099 - val_accuracy: 0.9030\n",
"Epoch 17/30\n",
"460/460 [==============================] - 8s 17ms/step - loss: 0.1506 - accuracy: 0.9403 - val_loss: 0.3593 - val_accuracy: 0.8965\n",
"Epoch 18/30\n",
"460/460 [==============================] - 6s 13ms/step - loss: 0.1462 - accuracy: 0.9461 - val_loss: 0.3433 - val_accuracy: 0.9074\n",
"Epoch 19/30\n",
"460/460 [==============================] - 8s 17ms/step - loss: 0.1372 - accuracy: 0.9459 - val_loss: 0.2816 - val_accuracy: 0.9206\n",
"Epoch 20/30\n",
"460/460 [==============================] - 6s 13ms/step - loss: 0.1436 - accuracy: 0.9429 - val_loss: 0.2907 - val_accuracy: 0.9196\n",
"Epoch 21/30\n",
"460/460 [==============================] - 8s 17ms/step - loss: 0.1311 - accuracy: 0.9480 - val_loss: 0.2638 - val_accuracy: 0.9223\n",
"Epoch 22/30\n",
"460/460 [==============================] - 6s 13ms/step - loss: 0.1498 - accuracy: 0.9410 - val_loss: 0.3071 - val_accuracy: 0.9030\n",
"Epoch 23/30\n",
"460/460 [==============================] - 8s 16ms/step - loss: 0.1399 - accuracy: 0.9472 - val_loss: 0.3322 - val_accuracy: 0.9060\n",
"Epoch 24/30\n",
"460/460 [==============================] - 6s 14ms/step - loss: 0.1619 - accuracy: 0.9397 - val_loss: 0.2797 - val_accuracy: 0.9179\n",
"Epoch 25/30\n",
"460/460 [==============================] - 7s 16ms/step - loss: 0.1452 - accuracy: 0.9446 - val_loss: 0.2839 - val_accuracy: 0.9148\n",
"Epoch 26/30\n",
"460/460 [==============================] - 7s 14ms/step - loss: 0.1296 - accuracy: 0.9494 - val_loss: 0.3043 - val_accuracy: 0.8982\n",
"Epoch 27/30\n",
"460/460 [==============================] - 7s 16ms/step - loss: 0.1334 - accuracy: 0.9460 - val_loss: 0.2795 - val_accuracy: 0.9080\n",
"Epoch 28/30\n",
"460/460 [==============================] - 7s 14ms/step - loss: 0.1310 - accuracy: 0.9463 - val_loss: 0.2821 - val_accuracy: 0.8945\n",
"Epoch 29/30\n",
"460/460 [==============================] - 7s 16ms/step - loss: 0.1253 - accuracy: 0.9489 - val_loss: 0.2743 - val_accuracy: 0.9165\n",
"Epoch 30/30\n",
"460/460 [==============================] - 7s 14ms/step - loss: 0.1407 - accuracy: 0.9406 - val_loss: 0.2640 - val_accuracy: 0.9131\n"
]
},
{
"output_type": "execute_result",
"data": {
"text/plain": [
"<keras.callbacks.History at 0x7f78fa3f00a0>"
]
},
"metadata": {},
"execution_count": 58
}
]
},
{
"cell_type": "code",
"source": [
"# Confusion Matrix\n",
"confusion_matrix(y_test, model.predict(X_test))"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 286
},
"id": "abLtI6JsWeM1",
"outputId": "1e5b55ea-0793-463a-df25-7596e18e36c8"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"93/93 [==============================] - 2s 7ms/step\n"
]
},
{
"output_type": "execute_result",
"data": {
"text/plain": [
"Pred LAYING SITTING STANDING WALKING WALKING_DOWNSTAIRS \\\n",
"True \n",
"LAYING 537 0 0 0 0 \n",
"SITTING 2 419 67 0 0 \n",
"STANDING 0 128 404 0 0 \n",
"WALKING 0 0 0 472 19 \n",
"WALKING_DOWNSTAIRS 0 0 0 2 415 \n",
"WALKING_UPSTAIRS 0 0 0 11 16 \n",
"\n",
"Pred WALKING_UPSTAIRS \n",
"True \n",
"LAYING 0 \n",
"SITTING 3 \n",
"STANDING 0 \n",
"WALKING 5 \n",
"WALKING_DOWNSTAIRS 3 \n",
"WALKING_UPSTAIRS 444 "
],
"text/html": [
"\n",
" <div id=\"df-03249097-2495-4177-b033-477f5b7ebb57\">\n",
" <div class=\"colab-df-container\">\n",
" <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>Pred</th>\n",
" <th>LAYING</th>\n",
" <th>SITTING</th>\n",
" <th>STANDING</th>\n",
" <th>WALKING</th>\n",
" <th>WALKING_DOWNSTAIRS</th>\n",
" <th>WALKING_UPSTAIRS</th>\n",
" </tr>\n",
" <tr>\n",
" <th>True</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>LAYING</th>\n",
" <td>537</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>SITTING</th>\n",
" <td>2</td>\n",
" <td>419</td>\n",
" <td>67</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>3</td>\n",
" </tr>\n",
" <tr>\n",
" <th>STANDING</th>\n",
" <td>0</td>\n",
" <td>128</td>\n",
" <td>404</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>WALKING</th>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>472</td>\n",
" <td>19</td>\n",
" <td>5</td>\n",
" </tr>\n",
" <tr>\n",
" <th>WALKING_DOWNSTAIRS</th>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>2</td>\n",
" <td>415</td>\n",
" <td>3</td>\n",
" </tr>\n",
" <tr>\n",
" <th>WALKING_UPSTAIRS</th>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>11</td>\n",
" <td>16</td>\n",
" <td>444</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>\n",
" <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-03249097-2495-4177-b033-477f5b7ebb57')\"\n",
" title=\"Convert this dataframe to an interactive table.\"\n",
" style=\"display:none;\">\n",
" \n",
" <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
" width=\"24px\">\n",
" <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
" <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
" </svg>\n",
" </button>\n",
" \n",
" <style>\n",
" .colab-df-container {\n",
" display:flex;\n",
" flex-wrap:wrap;\n",
" gap: 12px;\n",
" }\n",
"\n",
" .colab-df-convert {\n",
" background-color: #E8F0FE;\n",
" border: none;\n",
" border-radius: 50%;\n",
" cursor: pointer;\n",
" display: none;\n",
" fill: #1967D2;\n",
" height: 32px;\n",
" padding: 0 0 0 0;\n",
" width: 32px;\n",
" }\n",
"\n",
" .colab-df-convert:hover {\n",
" background-color: #E2EBFA;\n",
" box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
" fill: #174EA6;\n",
" }\n",
"\n",
" [theme=dark] .colab-df-convert {\n",
" background-color: #3B4455;\n",
" fill: #D2E3FC;\n",
" }\n",
"\n",
" [theme=dark] .colab-df-convert:hover {\n",
" background-color: #434B5C;\n",
" box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
" filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
" fill: #FFFFFF;\n",
" }\n",
" </style>\n",
"\n",
" <script>\n",
" const buttonEl =\n",
" document.querySelector('#df-03249097-2495-4177-b033-477f5b7ebb57 button.colab-df-convert');\n",
" buttonEl.style.display =\n",
" google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
"\n",
" async function convertToInteractive(key) {\n",
" const element = document.querySelector('#df-03249097-2495-4177-b033-477f5b7ebb57');\n",
" const dataTable =\n",
" await google.colab.kernel.invokeFunction('convertToInteractive',\n",
" [key], {});\n",
" if (!dataTable) return;\n",
"\n",
" const docLinkHtml = 'Like what you see? Visit the ' +\n",
" '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
" + ' to learn more about interactive tables.';\n",
" element.innerHTML = '';\n",
" dataTable['output_type'] = 'display_data';\n",
" await google.colab.output.renderOutput(dataTable, element);\n",
" const docLink = document.createElement('div');\n",
" docLink.innerHTML = docLinkHtml;\n",
" element.appendChild(docLink);\n",
" }\n",
" </script>\n",
" </div>\n",
" </div>\n",
" "
]
},
"metadata": {},
"execution_count": 59
}
]
},
{
"cell_type": "code",
"source": [
"score = model.evaluate(X_test, y_test)\n",
"\n",
"print(\"\\n categorica_crossentropy || accuracy \")\n",
"print(\" ____________________________________\")\n",
"print(score)"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "sYNMpcu_WjB-",
"outputId": "e43835c8-4b1b-4994-9e9e-67cf6f208d59"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"93/93 [==============================] - 1s 6ms/step - loss: 0.2640 - accuracy: 0.9131\n",
"\n",
" categorica_crossentropy || accuracy \n",
" ____________________________________\n",
"[0.2640216052532196, 0.9131320118904114]\n"
]
}
]
},
{
"cell_type": "markdown",
"source": [
"#### 2.2 Second Model for 2-Layer of LSTM with more hyperparameter tunning"
],
"metadata": {
"id": "gPw5BDcNWocp"
}
},
{
"cell_type": "code",
"source": [
"# Model execution\n",
"model = Sequential()\n",
"model.add(LSTM(64, input_shape=(timesteps, input_dim), return_sequences=True, bias_regularizer=reg))\n",
"model.add(BatchNormalization())\n",
"model.add(Dropout(0.50))\n",
"model.add(LSTM(48))\n",
"model.add(Dropout(0.50))\n",
"model.add(Dense(n_classes, activation='sigmoid'))\n",
"print(\"Model Summary: \")\n",
"model.summary()"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "KgIyrMM0WsEY",
"outputId": "414d300c-10de-40e0-b5e5-73de5ff3e97f"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"Model Summary: \n",
"Model: \"sequential_5\"\n",
"_________________________________________________________________\n",
" Layer (type) Output Shape Param # \n",
"=================================================================\n",
" lstm_8 (LSTM) (None, 128, 64) 18944 \n",
" \n",
" batch_normalization_3 (Batc (None, 128, 64) 256 \n",
" hNormalization) \n",
" \n",
" dropout_8 (Dropout) (None, 128, 64) 0 \n",
" \n",
" lstm_9 (LSTM) (None, 48) 21696 \n",
" \n",
" dropout_9 (Dropout) (None, 48) 0 \n",
" \n",
" dense_5 (Dense) (None, 6) 294 \n",
" \n",
"=================================================================\n",
"Total params: 41,190\n",
"Trainable params: 41,062\n",
"Non-trainable params: 128\n",
"_________________________________________________________________\n"
]
}
]
},
{
"cell_type": "code",
"source": [
"model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])"
],
"metadata": {
"id": "cJ4wvkh5WycW"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"# Training the model\n",
"model.fit(X_train, y_train, batch_size=n_batch, validation_data=(X_test, y_test), epochs=n_epochs)"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "VKrGoJiuW3tK",
"outputId": "86011b01-e928-434b-a373-102ff9c27ec2"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"Epoch 1/30\n",
"460/460 [==============================] - 13s 19ms/step - loss: 1.6734 - accuracy: 0.7078 - val_loss: 1.5870 - val_accuracy: 0.5945\n",
"Epoch 2/30\n",
"460/460 [==============================] - 6s 14ms/step - loss: 0.7557 - accuracy: 0.8898 - val_loss: 0.4793 - val_accuracy: 0.9179\n",
"Epoch 3/30\n",
"460/460 [==============================] - 8s 16ms/step - loss: 0.4011 - accuracy: 0.9135 - val_loss: 0.3697 - val_accuracy: 0.8761\n",
"Epoch 4/30\n",
"460/460 [==============================] - 6s 14ms/step - loss: 0.2216 - accuracy: 0.9287 - val_loss: 0.3722 - val_accuracy: 0.8829\n",
"Epoch 5/30\n",
"460/460 [==============================] - 8s 17ms/step - loss: 0.1946 - accuracy: 0.9334 - val_loss: 0.4488 - val_accuracy: 0.8446\n",
"Epoch 6/30\n",
"460/460 [==============================] - 6s 13ms/step - loss: 0.1719 - accuracy: 0.9381 - val_loss: 0.2355 - val_accuracy: 0.9128\n",
"Epoch 7/30\n",
"460/460 [==============================] - 7s 16ms/step - loss: 0.1631 - accuracy: 0.9399 - val_loss: 0.2094 - val_accuracy: 0.9325\n",
"Epoch 8/30\n",
"460/460 [==============================] - 6s 13ms/step - loss: 0.1798 - accuracy: 0.9317 - val_loss: 0.2030 - val_accuracy: 0.9189\n",
"Epoch 9/30\n",
"460/460 [==============================] - 8s 17ms/step - loss: 0.1874 - accuracy: 0.9312 - val_loss: 0.2831 - val_accuracy: 0.9165\n",
"Epoch 10/30\n",
"460/460 [==============================] - 6s 14ms/step - loss: 0.1565 - accuracy: 0.9387 - val_loss: 0.2430 - val_accuracy: 0.9192\n",
"Epoch 11/30\n",
"460/460 [==============================] - 7s 16ms/step - loss: 0.1611 - accuracy: 0.9384 - val_loss: 0.2457 - val_accuracy: 0.9080\n",
"Epoch 12/30\n",
"460/460 [==============================] - 6s 14ms/step - loss: 0.1424 - accuracy: 0.9445 - val_loss: 0.3006 - val_accuracy: 0.8992\n",
"Epoch 13/30\n",
"460/460 [==============================] - 7s 16ms/step - loss: 0.1442 - accuracy: 0.9440 - val_loss: 0.2308 - val_accuracy: 0.9162\n",
"Epoch 14/30\n",
"460/460 [==============================] - 6s 13ms/step - loss: 0.1882 - accuracy: 0.9353 - val_loss: 0.2404 - val_accuracy: 0.9199\n",
"Epoch 15/30\n",
"460/460 [==============================] - 8s 16ms/step - loss: 0.1385 - accuracy: 0.9448 - val_loss: 0.2316 - val_accuracy: 0.9094\n",
"Epoch 16/30\n",
"460/460 [==============================] - 6s 14ms/step - loss: 0.1704 - accuracy: 0.9393 - val_loss: 0.2102 - val_accuracy: 0.9257\n",
"Epoch 17/30\n",
"460/460 [==============================] - 7s 16ms/step - loss: 0.1349 - accuracy: 0.9465 - val_loss: 0.2214 - val_accuracy: 0.9165\n",
"Epoch 18/30\n",
"460/460 [==============================] - 6s 13ms/step - loss: 0.1502 - accuracy: 0.9400 - val_loss: 0.2715 - val_accuracy: 0.9101\n",
"Epoch 19/30\n",
"460/460 [==============================] - 7s 16ms/step - loss: 0.1281 - accuracy: 0.9472 - val_loss: 0.2358 - val_accuracy: 0.9179\n",
"Epoch 20/30\n",
"460/460 [==============================] - 6s 14ms/step - loss: 0.1380 - accuracy: 0.9448 - val_loss: 0.2803 - val_accuracy: 0.9080\n",
"Epoch 21/30\n",
"460/460 [==============================] - 7s 16ms/step - loss: 0.1440 - accuracy: 0.9429 - val_loss: 0.2399 - val_accuracy: 0.9253\n",
"Epoch 22/30\n",
"460/460 [==============================] - 7s 14ms/step - loss: 0.1608 - accuracy: 0.9416 - val_loss: 0.2226 - val_accuracy: 0.9216\n",
"Epoch 23/30\n",
"460/460 [==============================] - 7s 16ms/step - loss: 0.1291 - accuracy: 0.9489 - val_loss: 0.2334 - val_accuracy: 0.9257\n",
"Epoch 24/30\n",
"460/460 [==============================] - 7s 14ms/step - loss: 0.1311 - accuracy: 0.9471 - val_loss: 0.2140 - val_accuracy: 0.9267\n",
"Epoch 25/30\n",
"460/460 [==============================] - 8s 16ms/step - loss: 0.1324 - accuracy: 0.9475 - val_loss: 0.2815 - val_accuracy: 0.9165\n",
"Epoch 26/30\n",
"460/460 [==============================] - 6s 13ms/step - loss: 0.1284 - accuracy: 0.9484 - val_loss: 0.2534 - val_accuracy: 0.9325\n",
"Epoch 27/30\n",
"460/460 [==============================] - 7s 16ms/step - loss: 0.1268 - accuracy: 0.9486 - val_loss: 0.2600 - val_accuracy: 0.9220\n",
"Epoch 28/30\n",
"460/460 [==============================] - 6s 13ms/step - loss: 0.1290 - accuracy: 0.9501 - val_loss: 0.2439 - val_accuracy: 0.9192\n",
"Epoch 29/30\n",
"460/460 [==============================] - 8s 17ms/step - loss: 0.1354 - accuracy: 0.9486 - val_loss: 0.5618 - val_accuracy: 0.8320\n",
"Epoch 30/30\n",
"460/460 [==============================] - 7s 14ms/step - loss: 0.1541 - accuracy: 0.9411 - val_loss: 0.3802 - val_accuracy: 0.9125\n"
]
},
{
"output_type": "execute_result",
"data": {
"text/plain": [
"<keras.callbacks.History at 0x7f78f3474a30>"
]
},
"metadata": {},
"execution_count": 63
}
]
},
{
"cell_type": "code",
"source": [
"# Confusion Matrix\n",
"confusion_matrix(y_test, model.predict(X_test))"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 286
},
"id": "65v6IGSdW66a",
"outputId": "e9876301-35b6-4a9e-f364-4580cd009f5d"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"93/93 [==============================] - 2s 7ms/step\n"
]
},
{
"output_type": "execute_result",
"data": {
"text/plain": [
"Pred LAYING SITTING STANDING WALKING WALKING_DOWNSTAIRS \\\n",
"True \n",
"LAYING 537 0 0 0 0 \n",
"SITTING 0 407 82 0 0 \n",
"STANDING 0 94 438 0 0 \n",
"WALKING 0 0 0 447 46 \n",
"WALKING_DOWNSTAIRS 0 0 0 1 419 \n",
"WALKING_UPSTAIRS 0 2 0 1 27 \n",
"\n",
"Pred WALKING_UPSTAIRS \n",
"True \n",
"LAYING 0 \n",
"SITTING 2 \n",
"STANDING 0 \n",
"WALKING 3 \n",
"WALKING_DOWNSTAIRS 0 \n",
"WALKING_UPSTAIRS 441 "
],
"text/html": [
"\n",
" <div id=\"df-af1f7d68-479a-4380-870e-0e82389a4389\">\n",
" <div class=\"colab-df-container\">\n",
" <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>Pred</th>\n",
" <th>LAYING</th>\n",
" <th>SITTING</th>\n",
" <th>STANDING</th>\n",
" <th>WALKING</th>\n",
" <th>WALKING_DOWNSTAIRS</th>\n",
" <th>WALKING_UPSTAIRS</th>\n",
" </tr>\n",
" <tr>\n",
" <th>True</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>LAYING</th>\n",
" <td>537</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>SITTING</th>\n",
" <td>0</td>\n",
" <td>407</td>\n",
" <td>82</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>2</td>\n",
" </tr>\n",
" <tr>\n",
" <th>STANDING</th>\n",
" <td>0</td>\n",
" <td>94</td>\n",
" <td>438</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>WALKING</th>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>447</td>\n",
" <td>46</td>\n",
" <td>3</td>\n",
" </tr>\n",
" <tr>\n",
" <th>WALKING_DOWNSTAIRS</th>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>419</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>WALKING_UPSTAIRS</th>\n",
" <td>0</td>\n",
" <td>2</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>27</td>\n",
" <td>441</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>\n",
" <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-af1f7d68-479a-4380-870e-0e82389a4389')\"\n",
" title=\"Convert this dataframe to an interactive table.\"\n",
" style=\"display:none;\">\n",
" \n",
" <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
" width=\"24px\">\n",
" <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
" <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
" </svg>\n",
" </button>\n",
" \n",
" <style>\n",
" .colab-df-container {\n",
" display:flex;\n",
" flex-wrap:wrap;\n",
" gap: 12px;\n",
" }\n",
"\n",
" .colab-df-convert {\n",
" background-color: #E8F0FE;\n",
" border: none;\n",
" border-radius: 50%;\n",
" cursor: pointer;\n",
" display: none;\n",
" fill: #1967D2;\n",
" height: 32px;\n",
" padding: 0 0 0 0;\n",
" width: 32px;\n",
" }\n",
"\n",
" .colab-df-convert:hover {\n",
" background-color: #E2EBFA;\n",
" box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
" fill: #174EA6;\n",
" }\n",
"\n",
" [theme=dark] .colab-df-convert {\n",
" background-color: #3B4455;\n",
" fill: #D2E3FC;\n",
" }\n",
"\n",
" [theme=dark] .colab-df-convert:hover {\n",
" background-color: #434B5C;\n",
" box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
" filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
" fill: #FFFFFF;\n",
" }\n",
" </style>\n",
"\n",
" <script>\n",
" const buttonEl =\n",
" document.querySelector('#df-af1f7d68-479a-4380-870e-0e82389a4389 button.colab-df-convert');\n",
" buttonEl.style.display =\n",
" google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
"\n",
" async function convertToInteractive(key) {\n",
" const element = document.querySelector('#df-af1f7d68-479a-4380-870e-0e82389a4389');\n",
" const dataTable =\n",
" await google.colab.kernel.invokeFunction('convertToInteractive',\n",
" [key], {});\n",
" if (!dataTable) return;\n",
"\n",
" const docLinkHtml = 'Like what you see? Visit the ' +\n",
" '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
" + ' to learn more about interactive tables.';\n",
" element.innerHTML = '';\n",
" dataTable['output_type'] = 'display_data';\n",
" await google.colab.output.renderOutput(dataTable, element);\n",
" const docLink = document.createElement('div');\n",
" docLink.innerHTML = docLinkHtml;\n",
" element.appendChild(docLink);\n",
" }\n",
" </script>\n",
" </div>\n",
" </div>\n",
" "
]
},
"metadata": {},
"execution_count": 64
}
]
},
{
"cell_type": "code",
"source": [
"score = model.evaluate(X_test, y_test)\n",
"\n",
"print(\"\\n categorical_crossentropy || accuracy \")\n",
"print(\" ____________________________________\")\n",
"print(score)"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "N7i84QfGW9Er",
"outputId": "18a8a70f-2896-45a2-a22b-cef28557cfe3"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"93/93 [==============================] - 1s 9ms/step - loss: 0.3802 - accuracy: 0.9125\n",
"\n",
" categorical_crossentropy || accuracy \n",
" ____________________________________\n",
"[0.38021206855773926, 0.9124533534049988]\n"
]
}
]
}
]
}