predicting_n_days.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Right, this time, instead of predicting one day at a time and then only being able to predict multiple days into the future. We are going to predict n days at a time."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Preprecessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "win_len = 100  # how many days the model uses to predict the next n days\n",
    "pred_len = 100 # How many days to predict"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Fetch Latest Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import yfinance as yf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "msft = yf.Ticker(\"BTC-USD\")\n",
    "btc_usd_hist = msft.history(period=\"max\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load and split data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "dataset = btc_usd_hist['Close']\n",
    "data = dataset.values.reshape(-1,1) # get np array version"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def train_test_split(data, pred_len):\n",
    "    data_train = data[:-pred_len]\n",
    "    data_test = data[-pred_len:]\n",
    "    return data_train, data_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train: 1945 days, Test: 100 days\n"
     ]
    }
   ],
   "source": [
    "train, test = train_test_split(data, pred_len)\n",
    "print('Train: {} days, Test: {} days'.format(len(train), len(test)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Feature Scaling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "sc = MinMaxScaler(feature_range=(0,1))\n",
    "train_scaled = sc.fit_transform(train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Windowing Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1945, 1)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = np.array([train_scaled[i-win_len:i, 0] for i in range(win_len+pred_len, len(train))])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train = np.array([train_scaled[i:i+pred_len, 0] for i in range(win_len, len(train)-pred_len)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = np.reshape(X_train, (X_train.shape[0], X_train.shape[1], 1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Building the Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Activation, Dense, Dropout, LSTM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "def build_lstm_model(data, num_neurons = 50, activ_func='linear', \n",
    "                     dropout=0.2, loss='mse', optimizer='adam'):\n",
    "    model = Sequential()\n",
    "    # layer 1\n",
    "    model.add(LSTM(units=num_neurons, return_sequences=True, input_shape=(data.shape[1], 1)))\n",
    "    model.add(Dropout(dropout))\n",
    "    # layer 2\n",
    "    model.add(LSTM(units=num_neurons, return_sequences=True))\n",
    "    model.add(Dropout(dropout))\n",
    "    # layer 3\n",
    "    model.add(LSTM(units=num_neurons, return_sequences=True))\n",
    "    model.add(Dropout(dropout))\n",
    "    # layer 4\n",
    "    model.add(LSTM(units=num_neurons))\n",
    "    model.add(Dropout(dropout))\n",
    "    # ouput layer\n",
    "    model.add(Dense(units=pred_len))\n",
    "    # model.add(Activation(activ_func))\n",
    "    model.compile(optimizer=optimizer, loss=loss)\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = build_lstm_model(X_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training the Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 1745 samples\n",
      "Epoch 1/100\n",
      "1745/1745 [==============================] - 9s 5ms/sample - loss: 0.0294\n",
      "Epoch 2/100\n",
      "1745/1745 [==============================] - 1s 515us/sample - loss: 0.0089\n",
      "Epoch 3/100\n",
      "1745/1745 [==============================] - 1s 517us/sample - loss: 0.0074\n",
      "Epoch 4/100\n",
      "1745/1745 [==============================] - 1s 510us/sample - loss: 0.0065\n",
      "Epoch 5/100\n",
      "1745/1745 [==============================] - 1s 509us/sample - loss: 0.0055\n",
      "Epoch 6/100\n",
      "1745/1745 [==============================] - 1s 512us/sample - loss: 0.0043\n",
      "Epoch 7/100\n",
      "1745/1745 [==============================] - 1s 509us/sample - loss: 0.0038\n",
      "Epoch 8/100\n",
      "1745/1745 [==============================] - 1s 512us/sample - loss: 0.0037\n",
      "Epoch 9/100\n",
      "1745/1745 [==============================] - 1s 510us/sample - loss: 0.0034\n",
      "Epoch 10/100\n",
      "1745/1745 [==============================] - 1s 515us/sample - loss: 0.0032\n",
      "Epoch 11/100\n",
      "1745/1745 [==============================] - 1s 510us/sample - loss: 0.0031s - los\n",
      "Epoch 12/100\n",
      "1745/1745 [==============================] - 1s 507us/sample - loss: 0.0026\n",
      "Epoch 13/100\n",
      "1745/1745 [==============================] - 1s 512us/sample - loss: 0.0025\n",
      "Epoch 14/100\n",
      "1745/1745 [==============================] - 1s 513us/sample - loss: 0.0026\n",
      "Epoch 15/100\n",
      "1745/1745 [==============================] - 1s 514us/sample - loss: 0.0025\n",
      "Epoch 16/100\n",
      "1745/1745 [==============================] - 1s 510us/sample - loss: 0.0024\n",
      "Epoch 17/100\n",
      "1745/1745 [==============================] - 1s 513us/sample - loss: 0.0024\n",
      "Epoch 18/100\n",
      "1745/1745 [==============================] - 1s 508us/sample - loss: 0.0023\n",
      "Epoch 19/100\n",
      "1745/1745 [==============================] - 1s 510us/sample - loss: 0.0021\n",
      "Epoch 20/100\n",
      "1745/1745 [==============================] - 1s 512us/sample - loss: 0.0022\n",
      "Epoch 21/100\n",
      "1745/1745 [==============================] - 1s 510us/sample - loss: 0.0022\n",
      "Epoch 22/100\n",
      "1745/1745 [==============================] - 1s 513us/sample - loss: 0.0020\n",
      "Epoch 23/100\n",
      "1745/1745 [==============================] - 1s 510us/sample - loss: 0.0021\n",
      "Epoch 24/100\n",
      "1745/1745 [==============================] - 1s 510us/sample - loss: 0.0023\n",
      "Epoch 25/100\n",
      "1745/1745 [==============================] - 1s 514us/sample - loss: 0.0022s - loss:\n",
      "Epoch 26/100\n",
      "1745/1745 [==============================] - 1s 517us/sample - loss: 0.0020\n",
      "Epoch 27/100\n",
      "1745/1745 [==============================] - 1s 517us/sample - loss: 0.0019\n",
      "Epoch 28/100\n",
      "1745/1745 [==============================] - 1s 512us/sample - loss: 0.0018\n",
      "Epoch 29/100\n",
      "1745/1745 [==============================] - 1s 514us/sample - loss: 0.0018\n",
      "Epoch 30/100\n",
      "1745/1745 [==============================] - 1s 510us/sample - loss: 0.0018\n",
      "Epoch 31/100\n",
      "1745/1745 [==============================] - 1s 516us/sample - loss: 0.0016\n",
      "Epoch 32/100\n",
      "1745/1745 [==============================] - 1s 510us/sample - loss: 0.0017s - loss: 0\n",
      "Epoch 33/100\n",
      "1745/1745 [==============================] - 1s 510us/sample - loss: 0.0019\n",
      "Epoch 34/100\n",
      "1745/1745 [==============================] - 1s 508us/sample - loss: 0.0018\n",
      "Epoch 35/100\n",
      "1745/1745 [==============================] - 1s 510us/sample - loss: 0.0017\n",
      "Epoch 36/100\n",
      "1745/1745 [==============================] - 1s 510us/sample - loss: 0.0017\n",
      "Epoch 37/100\n",
      "1745/1745 [==============================] - 1s 512us/sample - loss: 0.0016\n",
      "Epoch 38/100\n",
      "1745/1745 [==============================] - 1s 512us/sample - loss: 0.0016s - loss: 0.0\n",
      "Epoch 39/100\n",
      "1745/1745 [==============================] - 1s 513us/sample - loss: 0.0016s - loss: \n",
      "Epoch 40/100\n",
      "1745/1745 [==============================] - 1s 514us/sample - loss: 0.0016\n",
      "Epoch 41/100\n",
      "1745/1745 [==============================] - 1s 513us/sample - loss: 0.0016\n",
      "Epoch 42/100\n",
      "1745/1745 [==============================] - 1s 510us/sample - loss: 0.0015\n",
      "Epoch 43/100\n",
      "1745/1745 [==============================] - 1s 508us/sample - loss: 0.0014\n",
      "Epoch 44/100\n",
      "1745/1745 [==============================] - 1s 509us/sample - loss: 0.0015\n",
      "Epoch 45/100\n",
      "1745/1745 [==============================] - 1s 510us/sample - loss: 0.0016\n",
      "Epoch 46/100\n",
      "1745/1745 [==============================] - 1s 514us/sample - loss: 0.0016\n",
      "Epoch 47/100\n",
      "1745/1745 [==============================] - 1s 510us/sample - loss: 0.0014\n",
      "Epoch 48/100\n",
      "1745/1745 [==============================] - 1s 512us/sample - loss: 0.0015\n",
      "Epoch 49/100\n",
      "1745/1745 [==============================] - 1s 508us/sample - loss: 0.0014\n",
      "Epoch 50/100\n",
      "1745/1745 [==============================] - 1s 504us/sample - loss: 0.0014\n",
      "Epoch 51/100\n",
      "1745/1745 [==============================] - 1s 508us/sample - loss: 0.0014\n",
      "Epoch 52/100\n",
      "1745/1745 [==============================] - 1s 507us/sample - loss: 0.0014\n",
      "Epoch 53/100\n",
      "1745/1745 [==============================] - 1s 512us/sample - loss: 0.0013\n",
      "Epoch 54/100\n",
      "1745/1745 [==============================] - 1s 514us/sample - loss: 0.0013\n",
      "Epoch 55/100\n",
      "1745/1745 [==============================] - ETA: 0s - loss: 0.001 - 1s 514us/sample - loss: 0.0013\n",
      "Epoch 56/100\n",
      "1745/1745 [==============================] - 1s 566us/sample - loss: 0.0013\n",
      "Epoch 57/100\n",
      "1745/1745 [==============================] - 1s 522us/sample - loss: 0.0015\n",
      "Epoch 58/100\n",
      "1745/1745 [==============================] - 1s 524us/sample - loss: 0.0014\n",
      "Epoch 59/100\n",
      "1745/1745 [==============================] - 1s 545us/sample - loss: 0.0013\n",
      "Epoch 60/100\n",
      "1745/1745 [==============================] - 1s 525us/sample - loss: 0.0013\n",
      "Epoch 61/100\n",
      "1745/1745 [==============================] - 1s 527us/sample - loss: 0.0013\n",
      "Epoch 62/100\n",
      "1745/1745 [==============================] - 1s 516us/sample - loss: 0.0013\n",
      "Epoch 63/100\n",
      "1745/1745 [==============================] - 1s 519us/sample - loss: 0.0012\n",
      "Epoch 64/100\n",
      "1745/1745 [==============================] - 1s 514us/sample - loss: 0.0013\n",
      "Epoch 65/100\n",
      "1745/1745 [==============================] - 1s 522us/sample - loss: 0.0012\n",
      "Epoch 66/100\n",
      "1745/1745 [==============================] - 1s 520us/sample - loss: 0.0013\n",
      "Epoch 67/100\n",
      "1745/1745 [==============================] - 1s 520us/sample - loss: 0.0013\n",
      "Epoch 68/100\n",
      "1745/1745 [==============================] - 1s 513us/sample - loss: 0.0012\n",
      "Epoch 69/100\n",
      "1745/1745 [==============================] - 1s 512us/sample - loss: 0.0012\n",
      "Epoch 70/100\n",
      "1745/1745 [==============================] - 1s 518us/sample - loss: 0.0012\n",
      "Epoch 71/100\n",
      "1745/1745 [==============================] - 1s 517us/sample - loss: 0.0014\n",
      "Epoch 72/100\n",
      "1745/1745 [==============================] - 1s 513us/sample - loss: 0.0012\n",
      "Epoch 73/100\n",
      "1745/1745 [==============================] - 1s 508us/sample - loss: 0.0012\n",
      "Epoch 74/100\n",
      "1745/1745 [==============================] - 1s 515us/sample - loss: 0.0012\n",
      "Epoch 75/100\n",
      "1745/1745 [==============================] - 1s 515us/sample - loss: 0.0012\n",
      "Epoch 76/100\n",
      "1745/1745 [==============================] - 1s 518us/sample - loss: 0.0013\n",
      "Epoch 77/100\n",
      "1745/1745 [==============================] - 1s 516us/sample - loss: 0.0013\n",
      "Epoch 78/100\n",
      "1745/1745 [==============================] - 1s 517us/sample - loss: 0.0012\n",
      "Epoch 79/100\n",
      "1745/1745 [==============================] - 1s 517us/sample - loss: 0.0012\n",
      "Epoch 80/100\n",
      "1745/1745 [==============================] - 1s 508us/sample - loss: 0.0012\n",
      "Epoch 81/100\n",
      "1745/1745 [==============================] - 1s 521us/sample - loss: 0.0011\n",
      "Epoch 82/100\n",
      "1745/1745 [==============================] - 1s 518us/sample - loss: 0.0012\n",
      "Epoch 83/100\n",
      "1745/1745 [==============================] - 1s 513us/sample - loss: 0.0012\n",
      "Epoch 84/100\n",
      "1745/1745 [==============================] - 1s 516us/sample - loss: 0.0012\n",
      "Epoch 85/100\n",
      "1745/1745 [==============================] - 1s 516us/sample - loss: 0.0011\n",
      "Epoch 86/100\n",
      "1745/1745 [==============================] - 1s 516us/sample - loss: 0.0011\n",
      "Epoch 87/100\n",
      "1745/1745 [==============================] - 1s 518us/sample - loss: 0.0011\n",
      "Epoch 88/100\n",
      "1745/1745 [==============================] - 1s 518us/sample - loss: 0.0012\n",
      "Epoch 89/100\n",
      "1745/1745 [==============================] - 1s 516us/sample - loss: 0.0012\n",
      "Epoch 90/100\n",
      "1745/1745 [==============================] - 1s 522us/sample - loss: 0.0011\n",
      "Epoch 91/100\n",
      "1745/1745 [==============================] - 1s 527us/sample - loss: 0.0011\n",
      "Epoch 92/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1745/1745 [==============================] - 1s 520us/sample - loss: 0.0011\n",
      "Epoch 93/100\n",
      "1745/1745 [==============================] - 1s 512us/sample - loss: 0.0011\n",
      "Epoch 94/100\n",
      "1745/1745 [==============================] - 1s 512us/sample - loss: 0.0012\n",
      "Epoch 95/100\n",
      "1745/1745 [==============================] - 1s 504us/sample - loss: 0.0012\n",
      "Epoch 96/100\n",
      "1745/1745 [==============================] - 1s 510us/sample - loss: 0.0012\n",
      "Epoch 97/100\n",
      "1745/1745 [==============================] - 1s 517us/sample - loss: 0.0011\n",
      "Epoch 98/100\n",
      "1745/1745 [==============================] - 1s 517us/sample - loss: 0.0011\n",
      "Epoch 99/100\n",
      "1745/1745 [==============================] - 1s 508us/sample - loss: 0.0011\n",
      "Epoch 100/100\n",
      "1745/1745 [==============================] - 1s 509us/sample - loss: 0.0011\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x1b35226c348>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(X_train, y_train, epochs=100, batch_size=32)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Testing the Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Get Predicted Values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs = sc.transform(train[-win_len:])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_test = np.reshape(inputs, (1, inputs.shape[0], inputs.shape[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "prediction = model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "preds = sc.inverse_transform(prediction.reshape(-1,1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plot predicted values versus expected values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(test, color = 'red', label = 'Real BTC Price')\n",
    "plt.plot(preds, color = 'blue', label = 'Predicted BTC Price')\n",
    "plt.title('BTC Price Prediction')\n",
    "plt.xlabel('Time')\n",
    "plt.ylabel('BTC Price (USD)')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Prediction the future"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs = sc.transform(data[-win_len:])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs = np.reshape(inputs, (1, inputs.shape[0], inputs.shape[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "prediction = model.predict(inputs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "preds = sc.inverse_transform(prediction.reshape(-1,1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(preds, color = 'blue', label = 'Predicted BTC Price')\n",
    "plt.title('BTC Price Prediction')\n",
    "plt.xlabel('Time')\n",
    "plt.ylabel('BTC Price (USD)')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}