Hourly_load_forecasting_LSTM.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Deep Learning for hourly load forecasting of the building in Thane, Mumbai"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/varat/anaconda3/lib/python3.6/site-packages/h5py/__init__.py:36: 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": [
    "from __future__ import print_function\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from keras.layers.core import Dense, Activation, Dropout\n",
    "from keras.layers.recurrent import LSTM\n",
    "from keras.models import Sequential"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Data shape: (4877, 23)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([['1/1/2016', 1311693.43],\n",
       "       ['1/1/2016 1:00', 1240583.19],\n",
       "       ['1/1/2016 2:00', 1185502.49],\n",
       "       ...,\n",
       "       ['7/31/2016 21:00', 1427712.44],\n",
       "       ['7/31/2016 22:00', 1315301.39],\n",
       "       ['7/31/2016 23:00', 1204251.82]], dtype=object)"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load raw data\n",
    "df = pd.read_csv('hourly_load_mumbai.csv', header=None)\n",
    "# numpy array\n",
    "df_array = df.values\n",
    "# Daily load\n",
    "daily_load = [df_array[i,:] for i in range(0, len(df)) if i%24 == 0] # list_daily_load changed to daily_load\n",
    "# hourly load (23 hours for each day)\n",
    "hourly_load = [df_array[i,1]/100000 for i in range(0, len(df)) if i%24 != 0] # list_hourly_load changead as above\n",
    "# the length of the sequence for predicting the future value\n",
    "seq_len = 23\n",
    "load = []   # matrix changed simply to load\n",
    "\n",
    "for i in range(len(hourly_load) - seq_len + 1):\n",
    "        load.append(hourly_load[i:i + seq_len])\n",
    "final_load = load     # matrix_load changed to final_load\n",
    "\n",
    "# Shift all data by mean\n",
    "final_load = np.array(final_load)\n",
    "load_mean = final_load.mean()  # Shifted_value changed to load_mean\n",
    "final_load -= load_mean       # This is after removing all the means from the original data\n",
    "print('Data shape:', final_load.shape)\n",
    "df_array"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Split dataset: 90% for training and remaining 10% for testing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_row = int(round(0.75 * final_load.shape[0]))\n",
    "train_set = final_load[:train_row,:]\n",
    "\n",
    "# Shuffle the training set (but do not shuffle the rest set)\n",
    "np.random.shuffle(train_set)\n",
    "\n",
    "# The training Set\n",
    "X_train = train_set[:,:-1]\n",
    "# The last column is the true value to compute the mean-squared-error loss\n",
    "Y_train = train_set[:,-1]\n",
    "\n",
    "# The test set\n",
    "X_test = final_load[train_row:, :-1]\n",
    "Y_test = final_load[train_row:, -1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Input to LSTM layer needs to have the shape of (number of samples, dimension of each element)\n",
    "# Arranging the dimensions of inputs in order to feed into LSTM networks\n",
    "\n",
    "X_train = np.reshape(X_train, (X_train.shape[0], X_train.shape[1], 1))\n",
    "X_test = np.reshape(X_test, (X_test.shape[0], X_test.shape[1], 1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Build the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/varat/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:5: UserWarning: The `input_dim` and `input_length` arguments in recurrent layers are deprecated. Use `input_shape` instead.\n",
      "  \"\"\"\n",
      "/home/varat/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:5: UserWarning: Update your `LSTM` call to the Keras 2 API: `LSTM(return_sequences=True, input_shape=(None, 1), units=50)`\n",
      "  \"\"\"\n",
      "/home/varat/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:9: UserWarning: Update your `LSTM` call to the Keras 2 API: `LSTM(return_sequences=False, units=100)`\n",
      "  if __name__ == '__main__':\n",
      "/home/varat/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:13: UserWarning: Update your `Dense` call to the Keras 2 API: `Dense(activation=\"linear\", units=1)`\n",
      "  del sys.path[0]\n"
     ]
    }
   ],
   "source": [
    "# Model Definition\n",
    "model = Sequential()\n",
    "\n",
    "# Layer 1: LSTM\n",
    "model.add(LSTM(input_dim =1, output_dim = 50, return_sequences=True)) #activation by default=tanh\n",
    "model.add(Dropout(0.2))\n",
    "\n",
    "# Layer 2: LSTM\n",
    "model.add(LSTM(output_dim=100, return_sequences=False))\n",
    "model.add(Dropout(0.2))\n",
    "\n",
    "# Layer 3: Dense\n",
    "model.add(Dense(output_dim=1, activation='linear'))\n",
    "\n",
    "#Compile the model\n",
    "model.compile(loss='mse', optimizer='adam' )  # optimizer='rmsprop'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Train the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/varat/anaconda3/lib/python3.6/site-packages/keras/models.py:942: UserWarning: The `nb_epoch` argument in `fit` has been renamed `epochs`.\n",
      "  warnings.warn('The `nb_epoch` argument in `fit` '\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 3475 samples, validate on 183 samples\n",
      "Epoch 1/50\n",
      "3475/3475 [==============================] - 4s 1ms/step - loss: 7.9799 - val_loss: 5.8805\n",
      "Epoch 2/50\n",
      "3475/3475 [==============================] - 2s 499us/step - loss: 6.1906 - val_loss: 5.4110\n",
      "Epoch 3/50\n",
      "3475/3475 [==============================] - 2s 513us/step - loss: 5.5766 - val_loss: 4.5324\n",
      "Epoch 4/50\n",
      "3475/3475 [==============================] - 2s 541us/step - loss: 4.7423 - val_loss: 3.8810\n",
      "Epoch 5/50\n",
      "3475/3475 [==============================] - 2s 537us/step - loss: 3.8201 - val_loss: 2.8953\n",
      "Epoch 6/50\n",
      "3475/3475 [==============================] - 2s 547us/step - loss: 2.7819 - val_loss: 2.1540\n",
      "Epoch 7/50\n",
      "3475/3475 [==============================] - 2s 617us/step - loss: 2.2321 - val_loss: 1.7900\n",
      "Epoch 8/50\n",
      "3475/3475 [==============================] - 2s 593us/step - loss: 1.8370 - val_loss: 1.4028\n",
      "Epoch 9/50\n",
      "3475/3475 [==============================] - 2s 547us/step - loss: 1.5184 - val_loss: 1.1201\n",
      "Epoch 10/50\n",
      "3475/3475 [==============================] - 2s 543us/step - loss: 1.2401 - val_loss: 0.9044\n",
      "Epoch 11/50\n",
      "3475/3475 [==============================] - 2s 574us/step - loss: 1.0968 - val_loss: 0.7677\n",
      "Epoch 12/50\n",
      "3475/3475 [==============================] - 2s 534us/step - loss: 0.9798 - val_loss: 0.7565\n",
      "Epoch 13/50\n",
      "3475/3475 [==============================] - 2s 524us/step - loss: 0.8793 - val_loss: 0.6179\n",
      "Epoch 14/50\n",
      "3475/3475 [==============================] - 2s 526us/step - loss: 0.7600 - val_loss: 0.5266\n",
      "Epoch 15/50\n",
      "3475/3475 [==============================] - 2s 545us/step - loss: 0.6721 - val_loss: 0.4849\n",
      "Epoch 16/50\n",
      "3475/3475 [==============================] - 2s 582us/step - loss: 0.6553 - val_loss: 0.4329\n",
      "Epoch 17/50\n",
      "3475/3475 [==============================] - 2s 524us/step - loss: 0.6259 - val_loss: 0.3898\n",
      "Epoch 18/50\n",
      "3475/3475 [==============================] - 2s 518us/step - loss: 0.5994 - val_loss: 0.3517\n",
      "Epoch 19/50\n",
      "3475/3475 [==============================] - 2s 528us/step - loss: 0.5446 - val_loss: 0.3351\n",
      "Epoch 20/50\n",
      "3475/3475 [==============================] - 2s 537us/step - loss: 0.5111 - val_loss: 0.3105\n",
      "Epoch 21/50\n",
      "3475/3475 [==============================] - 2s 630us/step - loss: 0.4906 - val_loss: 0.2622\n",
      "Epoch 22/50\n",
      "3475/3475 [==============================] - 2s 511us/step - loss: 0.4744 - val_loss: 0.2515\n",
      "Epoch 23/50\n",
      "3475/3475 [==============================] - 2s 509us/step - loss: 0.4599 - val_loss: 0.2191\n",
      "Epoch 24/50\n",
      "3475/3475 [==============================] - 2s 605us/step - loss: 0.4344 - val_loss: 0.2253\n",
      "Epoch 25/50\n",
      "3475/3475 [==============================] - 2s 509us/step - loss: 0.4361 - val_loss: 0.2331\n",
      "Epoch 26/50\n",
      "3475/3475 [==============================] - 2s 519us/step - loss: 0.4279 - val_loss: 0.1931\n",
      "Epoch 27/50\n",
      "3475/3475 [==============================] - 2s 560us/step - loss: 0.4154 - val_loss: 0.1883\n",
      "Epoch 28/50\n",
      "3475/3475 [==============================] - 2s 517us/step - loss: 0.3788 - val_loss: 0.1708\n",
      "Epoch 29/50\n",
      "3475/3475 [==============================] - 2s 511us/step - loss: 0.3659 - val_loss: 0.1766\n",
      "Epoch 30/50\n",
      "3475/3475 [==============================] - 2s 505us/step - loss: 0.3776 - val_loss: 0.1627\n",
      "Epoch 31/50\n",
      "3475/3475 [==============================] - 2s 497us/step - loss: 0.3507 - val_loss: 0.1521\n",
      "Epoch 32/50\n",
      "3475/3475 [==============================] - 2s 537us/step - loss: 0.3452 - val_loss: 0.1624\n",
      "Epoch 33/50\n",
      "3475/3475 [==============================] - 2s 598us/step - loss: 0.3499 - val_loss: 0.1471\n",
      "Epoch 34/50\n",
      "3475/3475 [==============================] - 2s 494us/step - loss: 0.3255 - val_loss: 0.1404\n",
      "Epoch 35/50\n",
      "3475/3475 [==============================] - 2s 503us/step - loss: 0.3354 - val_loss: 0.1437\n",
      "Epoch 36/50\n",
      "3475/3475 [==============================] - 2s 516us/step - loss: 0.3085 - val_loss: 0.1416\n",
      "Epoch 37/50\n",
      "3475/3475 [==============================] - 2s 531us/step - loss: 0.3201 - val_loss: 0.1368\n",
      "Epoch 38/50\n",
      "3475/3475 [==============================] - 2s 527us/step - loss: 0.3129 - val_loss: 0.1326\n",
      "Epoch 39/50\n",
      "3475/3475 [==============================] - 2s 528us/step - loss: 0.3058 - val_loss: 0.1346\n",
      "Epoch 40/50\n",
      "3475/3475 [==============================] - 2s 524us/step - loss: 0.2992 - val_loss: 0.1359\n",
      "Epoch 41/50\n",
      "3475/3475 [==============================] - 2s 589us/step - loss: 0.3047 - val_loss: 0.1289\n",
      "Epoch 42/50\n",
      "3475/3475 [==============================] - 2s 582us/step - loss: 0.2967 - val_loss: 0.1348\n",
      "Epoch 43/50\n",
      "3475/3475 [==============================] - 2s 512us/step - loss: 0.2804 - val_loss: 0.1284\n",
      "Epoch 44/50\n",
      "3475/3475 [==============================] - 2s 509us/step - loss: 0.2869 - val_loss: 0.1312\n",
      "Epoch 45/50\n",
      "3475/3475 [==============================] - 2s 511us/step - loss: 0.2891 - val_loss: 0.1270\n",
      "Epoch 46/50\n",
      "3475/3475 [==============================] - 2s 515us/step - loss: 0.2954 - val_loss: 0.1228\n",
      "Epoch 47/50\n",
      "3475/3475 [==============================] - 2s 533us/step - loss: 0.2790 - val_loss: 0.1379\n",
      "Epoch 48/50\n",
      "3475/3475 [==============================] - 2s 559us/step - loss: 0.2850 - val_loss: 0.1256\n",
      "Epoch 49/50\n",
      "3475/3475 [==============================] - 2s 554us/step - loss: 0.2672 - val_loss: 0.1247\n",
      "Epoch 50/50\n",
      "3475/3475 [==============================] - 2s 601us/step - loss: 0.2737 - val_loss: 0.1303\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x7f4e77b92c50>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(X_train, Y_train, batch_size=512, nb_epoch=50, validation_split=0.05, verbose=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Evaluate the result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1219/1219 [==============================] - 0s 366us/step\n",
      "\n",
      " mean squared error(MSE) on the test data set is 0.422 over 1219 test samples.\n"
     ]
    }
   ],
   "source": [
    "mse_test = model.evaluate(X_test, Y_test, verbose=1)\n",
    "print('\\n mean squared error(MSE) on the test data set is %.3f over %d test samples.' %(mse_test, len(Y_test)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###  Get the predicted values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "predicted_values = model.predict(X_test)\n",
    "num_test_samples = len(predicted_values)\n",
    "predicted_values = np.reshape(predicted_values, (num_test_samples, 1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plot the results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f4e54063588>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig= plt.figure()\n",
    "plt.plot(Y_test + load_mean)\n",
    "plt.plot(predicted_values + load_mean)\n",
    "plt.xlabel('hours (hrs)')\n",
    "plt.ylabel('Electricity Load (*1e5)')\n",
    "plt.show()\n",
    "\n",
    "# Save the plot into a jpg file\n",
    "fig.savefig('Forecasted_Load_Results.jpg', bbox_inches='tight')\n",
    "\n",
    "# Save the results into txt file\n",
    "test_result = np.vstack((predicted_values + load_mean))\n",
    "np.savetxt('Forecasted_Load__Results.txt', test_result)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}