Day_trader.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Train a DL model to predict whether the BTC price will go up or down from open to close each day. \n",
    "\n",
    "Then create a trading simulator to see how much money we can make."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# LSTM Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Pre-processing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Get Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import yfinance as yf\n",
    "msft = yf.Ticker(\"BTC-USD\")\n",
    "btc_usd_hist = msft.history(period=\"max\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = btc_usd_hist[['Open','High','Low','Close','Volume']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = dataset.values # get np array version"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2050, 5)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Scaling Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "sc = MinMaxScaler(feature_range=(0,1))\n",
    "# data_scaled = sc.fit_transform(data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create inputs and outputs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "wl = 256"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = np.array([sc.fit_transform(data[i-wl:i, :]) for i in range(wl, len(data))])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = []\n",
    "for i in range(wl, len(data)):\n",
    "    diff = data[i, 3] - data[i, 0]  # close - open\n",
    "    y.append(1 if diff > 0 else 0)\n",
    "y = np.array(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Days increased: 986\n",
      "Days Decreased: 808\n"
     ]
    }
   ],
   "source": [
    "print('Days increased:', sum(y))\n",
    "print('Days Decreased:', len(y) - sum(y))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((1794, 256, 5), (1794,))"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.shape, y.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For the trading simulator we want the full information of the next day, not just whether it went up or down."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_full = np.array([data[i, :] for i in range(wl, len(data))])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Split Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Get training and testing dataset for model development and hold-out dataset for trading simlation at end"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "len_holdout=100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train_and_test, X_holdout = X[:-len_holdout], X[-len_holdout:]\n",
    "y_train_and_test, y_holdout = y[:-len_holdout], y_full[-len_holdout:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X_train_and_test, y_train_and_test, test_size=0.25, random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((1270, 256, 5), (424, 256, 5), (1270,), (424,))"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape, X_test.shape, y_train.shape, y_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1270,)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model Building"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Activation, Dense, Dropout, LSTM, BatchNormalization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras import metrics\n",
    "from tensorflow.keras import initializers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "def build_lstm_model(data, dropout=0.0):\n",
    "    model = Sequential()\n",
    "    # layer \n",
    "    model.add(LSTM(units=256, return_sequences=True, input_shape=(data.shape[1], data.shape[2])))\n",
    "    model.add(Dropout(dropout))\n",
    "    model.add(BatchNormalization())\n",
    "    # layer \n",
    "    model.add(LSTM(units=256, return_sequences=True))\n",
    "    model.add(Dropout(dropout))\n",
    "    model.add(BatchNormalization())\n",
    "    # layer \n",
    "    model.add(LSTM(units=32))\n",
    "    model.add(Dropout(dropout))\n",
    "    model.add(BatchNormalization())\n",
    "    # ouput layer\n",
    "    model.add(Dense(units=1))\n",
    "    model.add(Activation('sigmoid'))\n",
    "    model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "pos = sum(y_train)\n",
    "neg = len(y_train) - sum(y_train)\n",
    "total = len(y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.25971783907728546"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "initial_bias = np.log(pos/neg)\n",
    "initial_bias  # bias_initializer = initializers.Constant(value=initial_bias)) for Dense"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = build_lstm_model(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Weight for class 0: 1.15\n",
      "Weight for class 1: 0.89\n"
     ]
    }
   ],
   "source": [
    "weight_for_0 = (1 / neg)*(total)/2.0 \n",
    "weight_for_1 = (1 / pos)*(total)/2.0\n",
    "\n",
    "class_weight = {0: weight_for_0, 1: weight_for_1}\n",
    "\n",
    "print('Weight for class 0: {:.2f}'.format(weight_for_0))\n",
    "print('Weight for class 1: {:.2f}'.format(weight_for_1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 1270 samples\n",
      "Epoch 1/50\n",
      "1270/1270 [==============================] - 8s 6ms/sample - loss: 0.8035 - accuracy: 0.4961\n",
      "Epoch 2/50\n",
      "1270/1270 [==============================] - 1s 617us/sample - loss: 0.7187 - accuracy: 0.4976\n",
      "Epoch 3/50\n",
      "1270/1270 [==============================] - 1s 617us/sample - loss: 0.6944 - accuracy: 0.5457\n",
      "Epoch 4/50\n",
      "1270/1270 [==============================] - 1s 617us/sample - loss: 0.6904 - accuracy: 0.5362\n",
      "Epoch 5/50\n",
      "1270/1270 [==============================] - 1s 612us/sample - loss: 0.6926 - accuracy: 0.5165\n",
      "Epoch 6/50\n",
      "1270/1270 [==============================] - 1s 616us/sample - loss: 0.6932 - accuracy: 0.5386\n",
      "Epoch 7/50\n",
      "1270/1270 [==============================] - 1s 611us/sample - loss: 0.6859 - accuracy: 0.5528\n",
      "Epoch 8/50\n",
      "1270/1270 [==============================] - 1s 614us/sample - loss: 0.6853 - accuracy: 0.5551\n",
      "Epoch 9/50\n",
      "1270/1270 [==============================] - 1s 614us/sample - loss: 0.6857 - accuracy: 0.5512\n",
      "Epoch 10/50\n",
      "1270/1270 [==============================] - 1s 620us/sample - loss: 0.6864 - accuracy: 0.5496\n",
      "Epoch 11/50\n",
      "1270/1270 [==============================] - 1s 612us/sample - loss: 0.6882 - accuracy: 0.5165\n",
      "Epoch 12/50\n",
      "1270/1270 [==============================] - 1s 613us/sample - loss: 0.6890 - accuracy: 0.5331\n",
      "Epoch 13/50\n",
      "1270/1270 [==============================] - 1s 612us/sample - loss: 0.6908 - accuracy: 0.5402\n",
      "Epoch 14/50\n",
      "1270/1270 [==============================] - 1s 606us/sample - loss: 0.6845 - accuracy: 0.5551\n",
      "Epoch 15/50\n",
      "1270/1270 [==============================] - 1s 610us/sample - loss: 0.6829 - accuracy: 0.5457\n",
      "Epoch 16/50\n",
      "1270/1270 [==============================] - 1s 607us/sample - loss: 0.6791 - accuracy: 0.5622\n",
      "Epoch 17/50\n",
      "1270/1270 [==============================] - 1s 611us/sample - loss: 0.6786 - accuracy: 0.5685\n",
      "Epoch 18/50\n",
      "1270/1270 [==============================] - 1s 612us/sample - loss: 0.6790 - accuracy: 0.5559\n",
      "Epoch 19/50\n",
      "1270/1270 [==============================] - 1s 615us/sample - loss: 0.6785 - accuracy: 0.5780\n",
      "Epoch 20/50\n",
      "1270/1270 [==============================] - 1s 610us/sample - loss: 0.6818 - accuracy: 0.5661\n",
      "Epoch 21/50\n",
      "1270/1270 [==============================] - 1s 612us/sample - loss: 0.6744 - accuracy: 0.5606\n",
      "Epoch 22/50\n",
      "1270/1270 [==============================] - 1s 609us/sample - loss: 0.6827 - accuracy: 0.5654\n",
      "Epoch 23/50\n",
      "1270/1270 [==============================] - 1s 615us/sample - loss: 0.6799 - accuracy: 0.5724\n",
      "Epoch 24/50\n",
      "1270/1270 [==============================] - 1s 612us/sample - loss: 0.6924 - accuracy: 0.5362\n",
      "Epoch 25/50\n",
      "1270/1270 [==============================] - 1s 608us/sample - loss: 0.6806 - accuracy: 0.5583\n",
      "Epoch 26/50\n",
      "1270/1270 [==============================] - 1s 608us/sample - loss: 0.6765 - accuracy: 0.5583\n",
      "Epoch 27/50\n",
      "1270/1270 [==============================] - 1s 609us/sample - loss: 0.6774 - accuracy: 0.5575\n",
      "Epoch 28/50\n",
      "1270/1270 [==============================] - 1s 612us/sample - loss: 0.6718 - accuracy: 0.5575\n",
      "Epoch 29/50\n",
      "1270/1270 [==============================] - 1s 610us/sample - loss: 0.6734 - accuracy: 0.5661\n",
      "Epoch 30/50\n",
      "1270/1270 [==============================] - 1s 610us/sample - loss: 0.6758 - accuracy: 0.5449\n",
      "Epoch 31/50\n",
      "1270/1270 [==============================] - 1s 617us/sample - loss: 0.6738 - accuracy: 0.5646\n",
      "Epoch 32/50\n",
      "1270/1270 [==============================] - 1s 614us/sample - loss: 0.6686 - accuracy: 0.5898\n",
      "Epoch 33/50\n",
      "1270/1270 [==============================] - 1s 608us/sample - loss: 0.6730 - accuracy: 0.5654\n",
      "Epoch 34/50\n",
      "1270/1270 [==============================] - 1s 609us/sample - loss: 0.6740 - accuracy: 0.5551\n",
      "Epoch 35/50\n",
      "1270/1270 [==============================] - 1s 612us/sample - loss: 0.6716 - accuracy: 0.5748\n",
      "Epoch 36/50\n",
      "1270/1270 [==============================] - 1s 612us/sample - loss: 0.6718 - accuracy: 0.5661\n",
      "Epoch 37/50\n",
      "1270/1270 [==============================] - 1s 621us/sample - loss: 0.6735 - accuracy: 0.5528\n",
      "Epoch 38/50\n",
      "1270/1270 [==============================] - 1s 615us/sample - loss: 0.6769 - accuracy: 0.5638\n",
      "Epoch 39/50\n",
      "1270/1270 [==============================] - 1s 607us/sample - loss: 0.6712 - accuracy: 0.5677\n",
      "Epoch 40/50\n",
      "1270/1270 [==============================] - 1s 613us/sample - loss: 0.6682 - accuracy: 0.5583\n",
      "Epoch 41/50\n",
      "1270/1270 [==============================] - 1s 610us/sample - loss: 0.6660 - accuracy: 0.5701\n",
      "Epoch 42/50\n",
      "1270/1270 [==============================] - 1s 614us/sample - loss: 0.6642 - accuracy: 0.5654\n",
      "Epoch 43/50\n",
      "1270/1270 [==============================] - 1s 614us/sample - loss: 0.6632 - accuracy: 0.5827\n",
      "Epoch 44/50\n",
      "1270/1270 [==============================] - 1s 614us/sample - loss: 0.6695 - accuracy: 0.5543\n",
      "Epoch 45/50\n",
      "1270/1270 [==============================] - 1s 612us/sample - loss: 0.6722 - accuracy: 0.5740\n",
      "Epoch 46/50\n",
      "1270/1270 [==============================] - 1s 633us/sample - loss: 0.6644 - accuracy: 0.5827\n",
      "Epoch 47/50\n",
      "1270/1270 [==============================] - 1s 645us/sample - loss: 0.6614 - accuracy: 0.5858\n",
      "Epoch 48/50\n",
      "1270/1270 [==============================] - 1s 660us/sample - loss: 0.6650 - accuracy: 0.5772\n",
      "Epoch 49/50\n",
      "1270/1270 [==============================] - 1s 663us/sample - loss: 0.6641 - accuracy: 0.5803\n",
      "Epoch 50/50\n",
      "1270/1270 [==============================] - 1s 626us/sample - loss: 0.6624 - accuracy: 0.5772\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(X_train, y_train, epochs=50, batch_size=128, verbose=True, class_weight=class_weight)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training Accuracy: 0.5385827\n"
     ]
    }
   ],
   "source": [
    "train_acc = model.evaluate(X_train, y_train, verbose=False)[1]\n",
    "print('Training Accuracy:', train_acc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=[8, 4])\n",
    "plt.plot(history.history['accuracy'])\n",
    "plt.title('Training Accuracy')\n",
    "plt.xlabel('Epoch')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model Testing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Testing Accuracy: 0.509434\n"
     ]
    }
   ],
   "source": [
    "test_acc = model.evaluate(X_test, y_test, verbose=False)[1]\n",
    "print('Testing Accuracy:', test_acc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = model.predict(X_test)\n",
    "y_pred = (y_pred > 0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import confusion_matrix, classification_report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TP: 134 \t FP: 122 \n",
      "FN: 82 \t\t TN: 82 \n"
     ]
    }
   ],
   "source": [
    "cm = confusion_matrix(y_test, y_pred)\n",
    "tn, fp, fn, tp = cm.ravel()\n",
    "print(f'TP: {tp} \\t FP: {fp} ')\n",
    "print(f'FN: {tn} \\t\\t TN: {tn} ')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.49      0.40      0.44       204\n",
      "           1       0.52      0.61      0.56       220\n",
      "\n",
      "    accuracy                           0.51       424\n",
      "   macro avg       0.51      0.51      0.50       424\n",
      "weighted avg       0.51      0.51      0.50       424\n",
      "\n"
     ]
    }
   ],
   "source": [
    "cr = classification_report(y_test, y_pred)\n",
    "print(cr)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "BOOM! TIME TO MAKE SOME MONEY"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Trading Time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((100, 256, 5), (100, 5))"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_holdout.shape, y_holdout.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's see how much money we would have made if we did this over the last 100 days:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "money = 1000\n",
    "output = []\n",
    "long, short = 0, 0\n",
    "for day, data in enumerate(zip(X_holdout, y_holdout)):\n",
    "    history, _open, close = data[0], data[1][0], data[1][3]\n",
    "    pred = model.predict(np.reshape(history, (1, history.shape[0], history.shape[1])))\n",
    "    if pred > 0.5:\n",
    "        long += 1\n",
    "        money = close/_open * money\n",
    "    else:\n",
    "        short += 1\n",
    "        money = _open/close * money\n",
    "    output.append(money)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(9, 91)"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "long, short"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Profit: 106.0%\n"
     ]
    }
   ],
   "source": [
    "profit = (money - 1000)/10\n",
    "print(f'Profit: {round(profit)}%')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x22c19967b08>]"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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(output)\n",
    "plt.title(Money over time)\n",
    "plt.xlabel('Days')\n",
    "plt.ylabel('USD')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's show the BTC price over this time period for comparison"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "price = y_holdout[:,3]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x22c14212a48>]"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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(price)"
   ]
  }
 ],
 "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": 2
}