Added autoregressive LSTM

lstm_energy.ipynb

@@ -2,7 +2,7 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 98,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -25,99 +25,21 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 102,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>date</th>\n",
-       "      <th>hvac_N</th>\n",
-       "      <th>hvac_S</th>\n",
-       "      <th>air_temp_set_1</th>\n",
-       "      <th>solar_radiation_set_1</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>0</th>\n",
-       "      <td>2018-01-01 00:00:00</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>11.64</td>\n",
-       "      <td>86.7</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1</th>\n",
-       "      <td>2018-01-01 00:01:00</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>11.64</td>\n",
-       "      <td>86.7</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2</th>\n",
-       "      <td>2018-01-01 00:02:00</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>11.64</td>\n",
-       "      <td>86.7</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>3</th>\n",
-       "      <td>2018-01-01 00:03:00</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>11.64</td>\n",
-       "      <td>86.7</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>4</th>\n",
-       "      <td>2018-01-01 00:04:00</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>11.64</td>\n",
-       "      <td>86.7</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "                  date  hvac_N  hvac_S  air_temp_set_1  solar_radiation_set_1\n",
-       "0  2018-01-01 00:00:00     NaN     NaN           11.64                   86.7\n",
-       "1  2018-01-01 00:01:00     NaN     NaN           11.64                   86.7\n",
-       "2  2018-01-01 00:02:00     NaN     NaN           11.64                   86.7\n",
-       "3  2018-01-01 00:03:00     NaN     NaN           11.64                   86.7\n",
-       "4  2018-01-01 00:04:00     NaN     NaN           11.64                   86.7"
-      ]
-     },
-     "execution_count": 102,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
+   "outputs": [],
    "source": [
+    "# Prepar energy data set with extended features\n",
     "feature_list = ['date', 'hvac_N', 'hvac_S', 'air_temp_set_1', 'solar_radiation_set_1']\n",
     "extended_energy_data = all_data[feature_list]\n",
+    "\n",
+    "extended_energy_data['date'] = pd.to_datetime(extended_energy_data['date'])\n",
+    "extended_energy_data.set_index('date', inplace=True)\n",
+    "\n",
+    "extended_energy_data = extended_energy_data.resample('15T').mean()\n",
+    "# extended_energy_data = extended_energy_data.interpolate(method='linear')\n",
+    "\n",
+    "extended_energy_data = extended_energy_data.reset_index(drop=False)\n",
     "extended_energy_data.head()"
    ]
   },
@@ -127,7 +49,8 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "energy_data = pd.read_csv(dataPATH + r\"\\hvac_data_1h.csv\")\n",
+    "# energy_data = pd.read_csv(dataPATH + r\"\\hvac_data_1h.csv\")\n",
+    "energy_data = extended_energy_data\n",
     "\n",
     "# Convert the date column to datetime\n",
     "energy_data['date'] = pd.to_datetime(energy_data['date'], format = \"%Y-%m-%d %H:%M:%S\")\n",
@@ -168,8 +91,8 @@
     "traindataset = traindataset.astype('float32')\n",
     "testdataset = testdataset.astype('float32')\n",
     "\n",
-    "mintest = np.min(testdataset)\n",
-    "maxtest = np.max(testdataset)\n",
+    "mintest = np.min(testdataset[:,0:2])\n",
+    "maxtest = np.max(testdataset[:,0:2])\n",
     "\n",
     "scaler = MinMaxScaler(feature_range=(0, 1))\n",
     "traindataset = scaler.fit_transform(traindataset)\n",
@@ -178,63 +101,29 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 101,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 1/5\n",
-      "57/58 [============================>.] - ETA: 0s - loss: 0.0238\n",
-      "Epoch 1: val_loss improved from inf to 0.01717, saving model to lstm_energy_01.keras\n",
-      "58/58 [==============================] - 11s 62ms/step - loss: 0.0238 - val_loss: 0.0172\n",
-      "Epoch 2/5\n",
-      "56/58 [===========================>..] - ETA: 0s - loss: 0.0139\n",
-      "Epoch 2: val_loss improved from 0.01717 to 0.01117, saving model to lstm_energy_01.keras\n",
-      "58/58 [==============================] - 2s 42ms/step - loss: 0.0139 - val_loss: 0.0112\n",
-      "Epoch 3/5\n",
-      "57/58 [============================>.] - ETA: 0s - loss: 0.0116\n",
-      "Epoch 3: val_loss improved from 0.01117 to 0.00990, saving model to lstm_energy_01.keras\n",
-      "58/58 [==============================] - 2s 36ms/step - loss: 0.0116 - val_loss: 0.0099\n",
-      "Epoch 4/5\n",
-      "57/58 [============================>.] - ETA: 0s - loss: 0.0084\n",
-      "Epoch 4: val_loss improved from 0.00990 to 0.00889, saving model to lstm_energy_01.keras\n",
-      "58/58 [==============================] - 2s 40ms/step - loss: 0.0084 - val_loss: 0.0089\n",
-      "Epoch 5/5\n",
-      "57/58 [============================>.] - ETA: 0s - loss: 0.0066\n",
-      "Epoch 5: val_loss did not improve from 0.00889\n",
-      "58/58 [==============================] - 2s 37ms/step - loss: 0.0066 - val_loss: 0.0103\n"
-     ]
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<keras.callbacks.History at 0x1f4931dc790>"
-      ]
-     },
-     "execution_count": 101,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
+   "outputs": [],
    "source": [
     "train,test = traindataset,testdataset\n",
-    "days_in_past = 20\n",
+    "steps_in_past = 24*4\n",
     "time_step = 1\n",
+    "no_inputs = 5\n",
+    "no_outputs = 5\n",
     "def create_dataset(dataset,time_step):\n",
-    "    x = [[] for _ in range(3)] \n",
-    "    Y = [[] for _ in range(2)]\n",
-    "    for i in range(time_step * days_in_past, len(dataset) - time_step * days_in_past): # -time_step is to ensure that the Y value has enough values\n",
-    "        for j in range(3):\n",
-    "            x[j].append(dataset[(i-time_step*days_in_past):i, j])\n",
-    "        for j in range(2):\n",
+    "    x = [[] for _ in range(no_inputs)] \n",
+    "    Y = [[] for _ in range(no_outputs)]\n",
+    "    for i in range(time_step * steps_in_past, len(dataset) - time_step * steps_in_past): # -time_step is to ensure that the Y value has enough values\n",
+    "        for j in range(no_inputs):\n",
+    "            x[j].append(dataset[(i-time_step*steps_in_past):i, j])\n",
+    "        for j in range(no_outputs):\n",
     "            Y[j].append([dataset[x + i, j] for x in range(0,time_step)])    \n",
     "    x = [np.array(feature_list) for feature_list in x]\n",
+    "    x = np.stack(x,axis=2)\n",
     "    Y = [np.array(feature_list) for feature_list in Y] \n",
-    "    Y = np.stack(Y,axis=1)\n",
-    "    Y = np.reshape(Y, (Y.shape[0], time_step*2))\n",
-    "    return np.stack(x,axis=2), Y\n",
+    "    Y = np.stack(Y,axis=2)\n",
+    "    Y = np.reshape(Y, (Y.shape[0], time_step*no_outputs))\n",
+    "    return x, Y\n",
     "\n",
     "\n",
     "X_train, y_train = create_dataset(train, time_step)\n",
@@ -245,7 +134,7 @@
     "model.add(LSTM(units=50, return_sequences=True, input_shape=(X_train.shape[1], X_train.shape[2])))\n",
     "model.add(LSTM(units=50, return_sequences=True))\n",
     "model.add(LSTM(units=30))\n",
-    "model.add(Dense(units=time_step*2))\n",
+    "model.add(Dense(units=time_step*no_outputs))\n",
     "\n",
     "model.compile(optimizer='adam', loss='mean_squared_error')\n",
     "\n",
@@ -259,23 +148,63 @@
    "execution_count": null,
    "metadata": {},
    "outputs": [],
-   "source": []
+   "source": [
+    "# Autoregressive prediction\n",
+    "X_pred = testdataset.copy()\n",
+    "for i in range(steps_in_past,steps_in_past*2):\n",
+    "    xin = X_pred[i-steps_in_past:i].reshape((1, steps_in_past, no_outputs)) \n",
+    "    X_pred[i] = model.predict(xin, verbose = 0)"
+   ]
   },
   {
    "cell_type": "code",
-   "execution_count": 95,
+   "execution_count": 54,
    "metadata": {},
    "outputs": [
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "22/22 [==============================] - 0s 8ms/step - loss: 0.0126\n",
-      "22/22 [==============================] - 1s 7ms/step\n",
-      "Loss:  0.01257658563554287\n"
-     ]
+     "data": {
+      "text/plain": [
+       "<matplotlib.legend.Legend at 0x25d1c90b4c0>"
+      ]
+     },
+     "execution_count": 54,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
+   "source": [
+    "# Plot prediction vs actual for test data\n",
+    "plt.figure()\n",
+    "plt.plot(X_pred[steps_in_past:steps_in_past*2,0],':',label='LSTM')\n",
+    "plt.plot(testdataset[steps_in_past:steps_in_past*2,0],'--',label='Actual')\n",
+    "plt.legend()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "test = model.predict(X_test[0].reshape((1,X_test.shape[1],X_test.shape[2])))\n",
+    "test"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
    "source": [
     "loss            = model.evaluate(X_test, y_test)\n",
     "test_predict1   = model.predict(X_test)\n",
@@ -288,7 +217,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 96,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -321,19 +250,9 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "%matplotlib qt\n",
-    "index = 100\n",
-    "\n",
-    "\n",
-    "plt.plot(y_test[index,0:24], label='Original Testing Data', color='blue')\n",
-    "plt.plot(test_predict1[index,0:24], label='Predicted Testing Data', color='red',alpha=0.8)\n",
-    "\n",
-    "\n",
-    "plt.title('Testing Data - Predicted vs Actual')\n",
-    "plt.xlabel('Time [hours]')\n",
-    "plt.ylabel('Energy [kW]')\n",
-    "plt.legend()\n",
-    "plt.show()"
+    "plt.plot(testdataset_df['date'], testdataset_df['hvac_N'], alpha = 0.8, label='Original Testing Data', color='blue')\n",
+    "plt.plot(test_predict2[])\n",
+    "plt.show()\n"
    ]
   },
   {
@@ -350,7 +269,9 @@
    "execution_count": null,
    "metadata": {},
    "outputs": [],
-   "source": []
+   "source": [
+    "maxtest"
+   ]
   },
   {
    "cell_type": "code",