multi_main.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 655,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(26638, 9)\n",
      "(6661, 9)\n"
     ]
    }
   ],
   "source": [
    "# We want 80% of the data to be used for training, and 20% for testing\n",
    "n_train_rows = int(dataset.shape[0]*.8)-1\n",
    "\n",
    "# split into train and test sets but keep all 9 columns\n",
    "train = dataset.iloc[:n_train_rows, :]\n",
    "test = dataset.iloc[n_train_rows:, :]\n",
    "\n",
    "# The total rows of the two datasets should equal the total amount of rows in your CSV\n",
    "print(train.shape)\n",
    "print(test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 656,
   "metadata": {},
   "outputs": [],
   "source": [
    "# normalize features\n",
    "sc = MinMaxScaler(feature_range = (0, 1))\n",
    "training_set_scaled = sc.fit_transform(train.values)\n",
    "test_set_scaled = sc.fit_transform(test.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 657,
   "metadata": {},
   "outputs": [],
   "source": [
    "steps = 50"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 658,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(5611, 50, 9)\n",
      "(5611, 50, 9)\n"
     ]
    }
   ],
   "source": [
    "x_train = []\n",
    "y_train = []\n",
    "\n",
    "for i in range(batch_size, test_set_scaled.shape[0]-steps):\n",
    "    x_train.append(training_set_scaled[i-steps:i, :])\n",
    "    y_train.append(training_set_scaled[i, :])\n",
    "\n",
    "x_train, y_train = np.array(x_train), np.array(y_train)\n",
    "print(x_train.shape)\n",
    "\n",
    "# Reshape\n",
    "x_train = np.reshape(x_train, (x_train.shape[0], x_train.shape[1], 9))\n",
    "print(x_train.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 659,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(26588, 50, 9)\n",
      "(26588, 50, 9)\n"
     ]
    }
   ],
   "source": [
    "x_train = []\n",
    "y_train = []\n",
    "\n",
    "for i in range(steps, training_set_scaled.shape[0]):\n",
    "    x_train.append(training_set_scaled[i-steps:i, :])\n",
    "    y_train.append(training_set_scaled[i, :])\n",
    "\n",
    "x_train, y_train = np.array(x_train), np.array(y_train)\n",
    "print(x_train.shape)\n",
    "\n",
    "# Reshape\n",
    "x_train = np.reshape(x_train, (x_train.shape[0], x_train.shape[1], 9))\n",
    "print(x_train.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 490,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 491,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "26588/26588 [==============================] - 110s 4ms/step - loss: 0.0321\n",
      "Epoch 2/50\n",
      "26588/26588 [==============================] - 97s 4ms/step - loss: 0.0251\n",
      "Epoch 3/50\n",
      "26588/26588 [==============================] - 96s 4ms/step - loss: 0.0229\n",
      "Epoch 4/50\n",
      "26588/26588 [==============================] - 97s 4ms/step - loss: 0.0222\n",
      "Epoch 5/50\n",
      "26588/26588 [==============================] - 101s 4ms/step - loss: 0.0219\n",
      "Epoch 6/50\n",
      "26588/26588 [==============================] - 98s 4ms/step - loss: 0.0218\n",
      "Epoch 7/50\n",
      "26588/26588 [==============================] - 97s 4ms/step - loss: 0.0218\n",
      "Epoch 8/50\n",
      "26588/26588 [==============================] - 99s 4ms/step - loss: 0.0216\n",
      "Epoch 9/50\n",
      "26588/26588 [==============================] - 99s 4ms/step - loss: 0.0216\n",
      "Epoch 10/50\n",
      "26588/26588 [==============================] - 108s 4ms/step - loss: 0.0215\n",
      "Epoch 11/50\n",
      "26588/26588 [==============================] - 99s 4ms/step - loss: 0.0216\n",
      "Epoch 12/50\n",
      "26588/26588 [==============================] - 97s 4ms/step - loss: 0.0215\n",
      "Epoch 13/50\n",
      "26588/26588 [==============================] - 97s 4ms/step - loss: 0.0215\n",
      "Epoch 14/50\n",
      "26588/26588 [==============================] - 98s 4ms/step - loss: 0.0215\n",
      "Epoch 15/50\n",
      "26588/26588 [==============================] - 96s 4ms/step - loss: 0.0214\n",
      "Epoch 16/50\n",
      "26588/26588 [==============================] - 96s 4ms/step - loss: 0.0214\n",
      "Epoch 17/50\n",
      "26588/26588 [==============================] - 96s 4ms/step - loss: 0.0214\n",
      "Epoch 18/50\n",
      "26588/26588 [==============================] - 97s 4ms/step - loss: 0.0214\n",
      "Epoch 19/50\n",
      "26588/26588 [==============================] - 97s 4ms/step - loss: 0.0213\n",
      "Epoch 20/50\n",
      "26588/26588 [==============================] - 98s 4ms/step - loss: 0.0214\n",
      "Epoch 21/50\n",
      "26588/26588 [==============================] - 97s 4ms/step - loss: 0.0212\n",
      "Epoch 22/50\n",
      "26588/26588 [==============================] - 97s 4ms/step - loss: 0.0213\n",
      "Epoch 23/50\n",
      "26588/26588 [==============================] - 98s 4ms/step - loss: 0.0212\n",
      "Epoch 24/50\n",
      "26588/26588 [==============================] - 95s 4ms/step - loss: 0.0211\n",
      "Epoch 25/50\n",
      "26588/26588 [==============================] - 96s 4ms/step - loss: 0.0212\n",
      "Epoch 26/50\n",
      "26588/26588 [==============================] - 103s 4ms/step - loss: 0.0211\n",
      "Epoch 27/50\n",
      "26588/26588 [==============================] - 122s 5ms/step - loss: 0.0212\n",
      "Epoch 28/50\n",
      "26588/26588 [==============================] - 172s 6ms/step - loss: 0.0211\n",
      "Epoch 29/50\n",
      "26588/26588 [==============================] - 185s 7ms/step - loss: 0.0211\n",
      "Epoch 30/50\n",
      "26588/26588 [==============================] - 185s 7ms/step - loss: 0.0210\n",
      "Epoch 31/50\n",
      "26588/26588 [==============================] - 28696s 1s/step - loss: 0.0211\n",
      "Epoch 32/50\n",
      "26588/26588 [==============================] - 361s 14ms/step - loss: 0.0209\n",
      "Epoch 33/50\n",
      "26588/26588 [==============================] - 194s 7ms/step - loss: 0.0210\n",
      "Epoch 34/50\n",
      "26588/26588 [==============================] - 193s 7ms/step - loss: 0.0209\n",
      "Epoch 35/50\n",
      "26588/26588 [==============================] - 191s 7ms/step - loss: 0.0209\n",
      "Epoch 36/50\n",
      "26588/26588 [==============================] - 111s 4ms/step - loss: 0.0209\n",
      "Epoch 37/50\n",
      "26588/26588 [==============================] - 108s 4ms/step - loss: 0.0209\n",
      "Epoch 38/50\n",
      "26588/26588 [==============================] - 109s 4ms/step - loss: 0.0207\n",
      "Epoch 39/50\n",
      "26588/26588 [==============================] - 107s 4ms/step - loss: 0.0208\n",
      "Epoch 40/50\n",
      "26588/26588 [==============================] - 103s 4ms/step - loss: 0.0207\n",
      "Epoch 41/50\n",
      "26588/26588 [==============================] - 98s 4ms/step - loss: 0.0207\n",
      "Epoch 42/50\n",
      "26588/26588 [==============================] - 101s 4ms/step - loss: 0.0207\n",
      "Epoch 43/50\n",
      "26588/26588 [==============================] - 111s 4ms/step - loss: 0.0207\n",
      "Epoch 44/50\n",
      "26588/26588 [==============================] - 108s 4ms/step - loss: 0.0207\n",
      "Epoch 45/50\n",
      "26588/26588 [==============================] - 100s 4ms/step - loss: 0.0205\n",
      "Epoch 46/50\n",
      "26588/26588 [==============================] - 100s 4ms/step - loss: 0.0205\n",
      "Epoch 47/50\n",
      "26588/26588 [==============================] - 100s 4ms/step - loss: 0.0205\n",
      "Epoch 48/50\n",
      "26588/26588 [==============================] - 100s 4ms/step - loss: 0.0204\n",
      "Epoch 49/50\n",
      "26588/26588 [==============================] - 100s 4ms/step - loss: 0.0204\n",
      "Epoch 50/50\n",
      "26588/26588 [==============================] - 100s 4ms/step - loss: 0.0204\n",
      "Saved model to disk\n"
     ]
    }
   ],
   "source": [
    "model.add(LSTM(units=50, return_sequences = True, input_shape = (x_train.shape[1],9)))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(LSTM(units=50, return_sequences = True))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(LSTM(units=50, return_sequences = True))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(LSTM(units=50))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(Dense(units=9))\n",
    "model.compile(loss=\"mse\", optimizer=\"adam\")\n",
    "model.fit(x_train, y_train, batch_size = 32, epochs = 50)\n",
    "model.summary\n",
    "\n",
    "model.save(\"multiple_features_\"+str(steps)+\".h5\")\n",
    "print(\"Saved model to disk\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 660,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(6561, 50, 9)\n"
     ]
    }
   ],
   "source": [
    "# Generate the inputs for the model from the training data and transform them back to their original values\n",
    "inputs = test.reset_index(drop=True).values\n",
    "inputs = sc.transform(inputs)\n",
    "\n",
    "# Initialise the parameters needed to build the batches\n",
    "x_test = []\n",
    "prediction_window = inputs.shape[0]\n",
    "\n",
    "# Build the test set and reshape as necessary as per the model requirements\n",
    "for i in range(steps, prediction_window-steps):\n",
    "    x_test.append(inputs[i-steps:i, :])\n",
    "x_test = np.array(x_test)\n",
    "x_test = np.reshape(x_test, (x_test.shape[0], x_test.shape[1], 9))\n",
    "print(x_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 662,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Predict values from test data trained using training data\n",
    "y_hat = model.predict(x_test)\n",
    "y_hat = sc.inverse_transform(y_hat)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 663,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Initialise the range over which the test data was taken\n",
    "x_data = range(dataset_total.shape[0]-x_test.shape[0],dataset_total.shape[0])\n",
    "\n",
    "# Transform the numpy array into a one-dimensional list\n",
    "pred_ask = []\n",
    "for i in y_hat:\n",
    "    pred_ask.append(i[0])\n",
    "\n",
    "# Visualise the results\n",
    "plt.figure(figsize = (18,9))\n",
    "plt.plot(test['ask_price'], color = 'red', label = 'y_test')\n",
    "plt.plot(x_data,pred_ask, color = 'blue', label = 'y_hat')\n",
    "plt.title('Test data prediction')\n",
    "plt.xlabel('Time')\n",
    "plt.ylabel('ask_price')\n",
    "plt.gca().set_xlim([train.shape[0],train.shape[0]+prediction_window])\n",
    "\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}