Datasets
Models
Downloads: 1
[731764]:
/
HAR_LSTM.ipynb
760 lines (759 with data), 25.5 kB
{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"# Importing Libraries"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"import pandas as pd\n",
"import numpy as np\n",
"import sys"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
"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",
"}\n",
"\n",
"# Utility function to print the confusion matrix\n",
"def confusion_matrix(Y_true, Y_pred):\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'])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Data"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [],
"source": [
"# Data directory\n",
"DATADIR = 'UCI_HAR_Dataset'"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [],
"source": [
"# 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",
"]"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [],
"source": [
"# Utility 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",
"# Utility function to load the load\n",
"def load_signals(subset):\n",
" signals_data = []\n",
"\n",
" for signal in SIGNALS:\n",
" filename = f'UCI_HAR_Dataset/{subset}/Inertial Signals/{signal}_{subset}.txt'\n",
" signals_data.append(\n",
" _read_csv(filename).as_matrix()\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))"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
"\n",
"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'UCI_HAR_Dataset/{subset}/y_{subset}.txt'\n",
" y = _read_csv(filename)[0]\n",
"\n",
" return pd.get_dummies(y).as_matrix()"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [],
"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"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [],
"source": [
"# Importing tensorflow\n",
"np.random.seed(42)\n",
"import tensorflow as tf\n",
"tf.set_random_seed(42)"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [],
"source": [
"# Configuring a session\n",
"session_conf = tf.ConfigProto(\n",
" intra_op_parallelism_threads=1,\n",
" inter_op_parallelism_threads=1\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/home/prajin/Downloads/ENTER/envs/py36/lib/python3.6/site-packages/h5py/__init__.py:34: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.\n",
" from ._conv import register_converters as _register_converters\n",
"Using TensorFlow backend.\n"
]
}
],
"source": [
"# Import Keras\n",
"from keras import backend as K\n",
"sess = tf.Session(graph=tf.get_default_graph(), config=session_conf)\n",
"K.set_session(sess)"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [],
"source": [
"# Importing libraries\n",
"from keras.models import Sequential\n",
"from keras.layers import LSTM\n",
"from keras.layers.core import Dense, Dropout"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [],
"source": [
"# Initializing parameters\n",
"epochs = 30\n",
"batch_size = 16\n",
"n_hidden = 32"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [],
"source": [
"# Utility function to count the number of classes\n",
"def _count_classes(y):\n",
" return len(set([tuple(category) for category in y]))"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/home/prajin/Downloads/ENTER/envs/py36/lib/python3.6/site-packages/ipykernel_launcher.py:12: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
" if sys.path[0] == '':\n"
]
}
],
"source": [
"# Loading the train and test data\n",
"X_train, X_test, Y_train, Y_test = load_data()"
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"128\n",
"9\n",
"7352\n"
]
}
],
"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))"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(7352, 128, 9)"
]
},
"execution_count": 19,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"X_train.shape"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(7352, 6)"
]
},
"execution_count": 24,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"Y_train.shape"
]
},
{
"cell_type": "code",
"execution_count": 34,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(2947, 128, 9)"
]
},
"execution_count": 34,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"X_test.shape"
]
},
{
"cell_type": "code",
"execution_count": 35,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"array([[[ 1.165315e-02, -2.939904e-02, 1.068262e-01, ...,\n",
" 1.041216e+00, -2.697959e-01, 2.377977e-02],\n",
" [ 1.310909e-02, -3.972867e-02, 1.524549e-01, ...,\n",
" 1.041803e+00, -2.800250e-01, 7.629271e-02],\n",
" [ 1.126885e-02, -5.240586e-02, 2.168462e-01, ...,\n",
" 1.039086e+00, -2.926631e-01, 1.474754e-01],\n",
" ...,\n",
" [ 1.291511e-03, 1.173502e-02, 3.665587e-03, ...,\n",
" 9.930164e-01, -2.599865e-01, 1.443951e-01],\n",
" [ 1.469997e-03, 9.517414e-03, 4.041945e-03, ...,\n",
" 9.932414e-01, -2.620643e-01, 1.447033e-01],\n",
" [ 2.573841e-03, 7.305069e-03, 4.888436e-03, ...,\n",
" 9.943906e-01, -2.641348e-01, 1.454939e-01]],\n",
"\n",
" [[ 9.279629e-03, 6.650520e-03, -2.631933e-02, ...,\n",
" 9.991921e-01, -2.649349e-01, 1.256164e-01],\n",
" [ 4.929711e-03, 1.864973e-02, -2.688753e-02, ...,\n",
" 9.946787e-01, -2.532142e-01, 1.256249e-01],\n",
" [ 3.953596e-03, 1.553950e-02, -3.663861e-02, ...,\n",
" 9.935518e-01, -2.565887e-01, 1.163814e-01],\n",
" ...,\n",
" [ 7.787600e-03, 4.730625e-03, 1.412899e-02, ...,\n",
" 1.001861e+00, -2.619359e-01, 1.527878e-01],\n",
" [ 3.433489e-03, -4.619849e-03, 1.338054e-03, ...,\n",
" 9.975208e-01, -2.713225e-01, 1.398428e-01],\n",
" [-1.238678e-03, -1.322889e-02, -1.703861e-02, ...,\n",
" 9.928615e-01, -2.799715e-01, 1.213135e-01]],\n",
"\n",
" [[ 5.731945e-03, 7.304842e-03, 1.021286e-02, ...,\n",
" 9.975931e-01, -2.639912e-01, 1.507741e-01],\n",
" [ 7.065650e-03, 7.330912e-03, 1.341419e-02, ...,\n",
" 9.989703e-01, -2.638194e-01, 1.539427e-01],\n",
" [ 5.109758e-03, 7.153458e-03, 3.646559e-03, ...,\n",
" 9.970574e-01, -2.638495e-01, 1.441536e-01],\n",
" ...,\n",
" [-7.428461e-04, -9.629137e-03, -2.500924e-03, ...,\n",
" 9.918802e-01, -2.836712e-01, 1.326780e-01],\n",
" [-1.923356e-03, -6.425974e-03, -2.524952e-03, ...,\n",
" 9.906626e-01, -2.805970e-01, 1.326941e-01],\n",
" [-4.304617e-03, -7.932046e-03, -3.140111e-03, ...,\n",
" 9.882446e-01, -2.822329e-01, 1.321175e-01]],\n",
"\n",
" ...,\n",
"\n",
" [[-1.476465e-01, 5.519791e-03, 1.025031e-02, ...,\n",
" 8.213505e-01, -2.484623e-01, -2.216934e-01],\n",
" [-1.699026e-01, 3.235187e-02, 2.632373e-02, ...,\n",
" 7.991996e-01, -2.232599e-01, -2.045561e-01],\n",
" [-1.686980e-01, 7.826144e-02, -2.703439e-02, ...,\n",
" 8.004623e-01, -1.790170e-01, -2.568719e-01],\n",
" ...,\n",
" [ 4.978930e-01, -3.158365e-01, -2.321939e-02, ...,\n",
" 1.463170e+00, -5.515283e-01, -2.723974e-01],\n",
" [ 2.141275e-01, -3.121422e-01, 1.814949e-01, ...,\n",
" 1.179223e+00, -5.472997e-01, -6.773376e-02],\n",
" [-1.145089e-01, -2.553472e-01, 3.870347e-01, ...,\n",
" 8.504963e-01, -4.900368e-01, 1.378256e-01]],\n",
"\n",
" [[ 7.122683e-02, -1.498122e-01, -1.659306e-01, ...,\n",
" 1.037668e+00, -3.971532e-01, -3.940817e-01],\n",
" [-8.866530e-02, -3.755543e-02, -8.708159e-02, ...,\n",
" 8.780725e-01, -2.848634e-01, -3.151097e-01],\n",
" [-7.067473e-02, -1.615178e-02, 1.401189e-02, ...,\n",
" 8.963897e-01, -2.635297e-01, -2.139040e-01],\n",
" ...,\n",
" [ 1.859878e-01, 7.344366e-03, 2.383924e-01, ...,\n",
" 1.156389e+00, -2.283478e-01, -3.512052e-03],\n",
" [ 2.737114e-01, -2.279012e-02, 1.302276e-01, ...,\n",
" 1.243857e+00, -2.583220e-01, -1.117857e-01],\n",
" [ 3.536738e-01, -1.118625e-01, -3.402252e-02, ...,\n",
" 1.323546e+00, -3.472416e-01, -2.760682e-01]],\n",
"\n",
" [[-1.936425e-01, -1.907511e-01, 1.958357e-01, ...,\n",
" 7.713622e-01, -4.250499e-01, -5.327655e-02],\n",
" [-6.498738e-02, -2.035990e-01, -1.531400e-01, ...,\n",
" 9.000949e-01, -4.375916e-01, -4.020727e-01],\n",
" [-9.712210e-02, -2.083832e-01, -2.710627e-01, ...,\n",
" 8.681034e-01, -4.421595e-01, -5.197379e-01],\n",
" ...,\n",
" [-5.075521e-02, -1.047171e-01, 1.732707e-01, ...,\n",
" 9.188616e-01, -3.516799e-01, -7.253919e-02],\n",
" [-1.980675e-02, -2.076396e-02, 1.956384e-01, ...,\n",
" 9.494752e-01, -2.675260e-01, -5.097549e-02],\n",
" [-1.104015e-02, 5.243883e-02, 2.184321e-01, ...,\n",
" 9.578348e-01, -1.941603e-01, -2.892477e-02]]])"
]
},
"execution_count": 35,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"X_test"
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(2947, 6)"
]
},
"execution_count": 26,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"Y_test.shape"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"- Defining the Architecture of LSTM"
]
},
{
"cell_type": "code",
"execution_count": 27,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"_________________________________________________________________\n",
"Layer (type) Output Shape Param # \n",
"=================================================================\n",
"lstm_2 (LSTM) (None, 32) 5376 \n",
"_________________________________________________________________\n",
"dropout_2 (Dropout) (None, 32) 0 \n",
"_________________________________________________________________\n",
"dense_2 (Dense) (None, 6) 198 \n",
"=================================================================\n",
"Total params: 5,574\n",
"Trainable params: 5,574\n",
"Non-trainable params: 0\n",
"_________________________________________________________________\n"
]
}
],
"source": [
"# Initiliazing the sequential model\n",
"\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()"
]
},
{
"cell_type": "code",
"execution_count": 28,
"metadata": {},
"outputs": [],
"source": [
"# Compiling the model\n",
"model.compile(loss='categorical_crossentropy',\n",
" optimizer='rmsprop',\n",
" metrics=['accuracy'])"
]
},
{
"cell_type": "code",
"execution_count": 29,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Train on 7352 samples, validate on 2947 samples\n",
"Epoch 1/30\n",
"7352/7352 [==============================] - 30s 4ms/step - loss: 1.3992 - acc: 0.3528 - val_loss: 1.3149 - val_acc: 0.3485\n",
"Epoch 2/30\n",
"7352/7352 [==============================] - 29s 4ms/step - loss: 1.1923 - acc: 0.4475 - val_loss: 1.1875 - val_acc: 0.4523\n",
"Epoch 3/30\n",
"7352/7352 [==============================] - 27s 4ms/step - loss: 1.0586 - acc: 0.4977 - val_loss: 1.1083 - val_acc: 0.5124\n",
"Epoch 4/30\n",
"7352/7352 [==============================] - 27s 4ms/step - loss: 0.9001 - acc: 0.6019 - val_loss: 0.9712 - val_acc: 0.5898\n",
"Epoch 5/30\n",
"7352/7352 [==============================] - 27s 4ms/step - loss: 0.8077 - acc: 0.6205 - val_loss: 0.8670 - val_acc: 0.5769\n",
"Epoch 6/30\n",
"7352/7352 [==============================] - 29s 4ms/step - loss: 0.7221 - acc: 0.6443 - val_loss: 0.7999 - val_acc: 0.6108\n",
"Epoch 7/30\n",
"7352/7352 [==============================] - 34s 5ms/step - loss: 0.7032 - acc: 0.6481 - val_loss: 0.8130 - val_acc: 0.6067\n",
"Epoch 8/30\n",
"7352/7352 [==============================] - 37s 5ms/step - loss: 0.6789 - acc: 0.6590 - val_loss: 0.7781 - val_acc: 0.6118\n",
"Epoch 9/30\n",
"7352/7352 [==============================] - 38s 5ms/step - loss: 0.6733 - acc: 0.6549 - val_loss: 0.8595 - val_acc: 0.6033\n",
"Epoch 10/30\n",
"7352/7352 [==============================] - 38s 5ms/step - loss: 0.6385 - acc: 0.6714 - val_loss: 0.8202 - val_acc: 0.6043\n",
"Epoch 11/30\n",
"7352/7352 [==============================] - 37s 5ms/step - loss: 0.5983 - acc: 0.6918 - val_loss: 0.7822 - val_acc: 0.6586\n",
"Epoch 12/30\n",
"7352/7352 [==============================] - 28s 4ms/step - loss: 0.5781 - acc: 0.7304 - val_loss: 0.7093 - val_acc: 0.7503\n",
"Epoch 13/30\n",
"7352/7352 [==============================] - 23s 3ms/step - loss: 0.5395 - acc: 0.7752 - val_loss: 0.6877 - val_acc: 0.7503\n",
"Epoch 14/30\n",
"7352/7352 [==============================] - 22s 3ms/step - loss: 0.5074 - acc: 0.7888 - val_loss: 0.5969 - val_acc: 0.7621\n",
"Epoch 15/30\n",
"7352/7352 [==============================] - 24s 3ms/step - loss: 0.4639 - acc: 0.7983 - val_loss: 0.6399 - val_acc: 0.7574\n",
"Epoch 16/30\n",
"7352/7352 [==============================] - 27s 4ms/step - loss: 0.4533 - acc: 0.8041 - val_loss: 0.5525 - val_acc: 0.7625\n",
"Epoch 17/30\n",
"7352/7352 [==============================] - 25s 3ms/step - loss: 0.4612 - acc: 0.8166 - val_loss: 0.5325 - val_acc: 0.7679\n",
"Epoch 18/30\n",
"7352/7352 [==============================] - 27s 4ms/step - loss: 0.3810 - acc: 0.8595 - val_loss: 0.5302 - val_acc: 0.8385\n",
"Epoch 19/30\n",
"7352/7352 [==============================] - 24s 3ms/step - loss: 0.3549 - acc: 0.8924 - val_loss: 0.7042 - val_acc: 0.8246\n",
"Epoch 20/30\n",
"7352/7352 [==============================] - 25s 3ms/step - loss: 0.3123 - acc: 0.9124 - val_loss: 0.5711 - val_acc: 0.8456\n",
"Epoch 21/30\n",
"7352/7352 [==============================] - 35s 5ms/step - loss: 0.2819 - acc: 0.9136 - val_loss: 0.5149 - val_acc: 0.8636\n",
"Epoch 22/30\n",
"7352/7352 [==============================] - 55s 7ms/step - loss: 0.2355 - acc: 0.9249 - val_loss: 0.5110 - val_acc: 0.8646\n",
"Epoch 23/30\n",
"7352/7352 [==============================] - 43s 6ms/step - loss: 0.2248 - acc: 0.9290 - val_loss: 0.6960 - val_acc: 0.8524\n",
"Epoch 24/30\n",
"7352/7352 [==============================] - 32s 4ms/step - loss: 0.2245 - acc: 0.9314 - val_loss: 0.6003 - val_acc: 0.8687\n",
"Epoch 25/30\n",
"7352/7352 [==============================] - 31s 4ms/step - loss: 0.2142 - acc: 0.9312 - val_loss: 0.4520 - val_acc: 0.8809\n",
"Epoch 26/30\n",
"7352/7352 [==============================] - 63s 9ms/step - loss: 0.2139 - acc: 0.9340 - val_loss: 0.4768 - val_acc: 0.8643\n",
"Epoch 27/30\n",
"7352/7352 [==============================] - 38s 5ms/step - loss: 0.2048 - acc: 0.9316 - val_loss: 0.4726 - val_acc: 0.8795\n",
"Epoch 28/30\n",
"7352/7352 [==============================] - 34s 5ms/step - loss: 0.1946 - acc: 0.9369 - val_loss: 0.4605 - val_acc: 0.8765\n",
"Epoch 29/30\n",
"7352/7352 [==============================] - 63s 9ms/step - loss: 0.2185 - acc: 0.9327 - val_loss: 0.4615 - val_acc: 0.8768\n",
"Epoch 30/30\n",
"7352/7352 [==============================] - 76s 10ms/step - loss: 0.1809 - acc: 0.9374 - val_loss: 0.4475 - val_acc: 0.8843\n"
]
},
{
"data": {
"text/plain": [
"<keras.callbacks.History at 0x7fd101658ef0>"
]
},
"execution_count": 29,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Training the model\n",
"model.fit(X_train,\n",
" Y_train,\n",
" batch_size=batch_size,\n",
" validation_data=(X_test, Y_test),\n",
" epochs=epochs)"
]
},
{
"cell_type": "code",
"execution_count": 30,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"2947/2947 [==============================] - 2s 821us/step\n"
]
}
],
"source": [
"score = model.evaluate(X_test, Y_test)"
]
},
{
"cell_type": "code",
"execution_count": 33,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"array([[6.73119284e-05, 1.16691635e-05, 6.29200213e-06, 3.27186123e-03,\n",
" 3.09266567e-01, 6.14459668e-06],\n",
" [6.55044132e-05, 2.05861852e-05, 1.43757034e-05, 6.77454285e-03,\n",
" 4.01470065e-01, 6.11893711e-06],\n",
" [6.84985353e-05, 1.96996807e-05, 1.35612627e-05, 6.61419239e-03,\n",
" 4.27904457e-01, 6.24169252e-06],\n",
" ...,\n",
" [1.55011995e-03, 7.93270528e-01, 3.01599008e-04, 2.30963960e-05,\n",
" 5.54955914e-05, 1.02767759e-08],\n",
" [4.90763341e-04, 3.85723859e-01, 1.03853172e-05, 6.35696642e-06,\n",
" 2.14066167e-05, 1.14835030e-08],\n",
" [7.23787583e-04, 6.95120990e-01, 2.18840923e-05, 8.08145796e-06,\n",
" 6.29514252e-05, 5.91321019e-08]], dtype=float32)"
]
},
"execution_count": 33,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
" model.predict(X_test)"
]
},
{
"cell_type": "code",
"execution_count": 31,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Pred LAYING SITTING STANDING WALKING WALKING_DOWNSTAIRS \\\n",
"True \n",
"LAYING 510 0 27 0 0 \n",
"SITTING 0 375 110 3 0 \n",
"STANDING 0 80 446 2 0 \n",
"WALKING 0 0 0 410 27 \n",
"WALKING_DOWNSTAIRS 0 0 0 2 407 \n",
"WALKING_UPSTAIRS 0 0 0 3 10 \n",
"\n",
"Pred WALKING_UPSTAIRS \n",
"True \n",
"LAYING 0 \n",
"SITTING 3 \n",
"STANDING 4 \n",
"WALKING 59 \n",
"WALKING_DOWNSTAIRS 11 \n",
"WALKING_UPSTAIRS 458 \n"
]
}
],
"source": [
"# Confusion Matrix\n",
"print(confusion_matrix(Y_test, model.predict(X_test)))"
]
},
{
"cell_type": "code",
"execution_count": 32,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[0.44746464555687265, 0.8842891075670173]"
]
},
"execution_count": 32,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"score"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"- With a simple 2 layer architecture we got 90.09% accuracy and a loss of 0.30\n",
"- We can further imporve the performace with Hyperparameter tuning"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.6.7"
}
},
"nbformat": 4,
"nbformat_minor": 2
}