neron.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              ts_code trade_date      open      high       low     close  \\\n",
      "date                                                                       \n",
      "2018-06-11  300750.SZ   20180611   30.0498   36.0558   30.0498   36.0558   \n",
      "2018-06-12  300750.SZ   20180612   39.6614   39.6614   39.6614   39.6614   \n",
      "2018-06-13  300750.SZ   20180613   43.6255   43.6255   43.6255   43.6255   \n",
      "2018-06-14  300750.SZ   20180614   47.9880   47.9880   47.9880   47.9880   \n",
      "2018-06-15  300750.SZ   20180615   52.7888   52.7888   52.7888   52.7888   \n",
      "...               ...        ...       ...       ...       ...       ...   \n",
      "2021-12-27  300750.SZ   20211227  574.2000  587.9900  573.5000  578.6000   \n",
      "2021-12-28  300750.SZ   20211228  578.0000  587.2800  569.9100  585.0000   \n",
      "2021-12-29  300750.SZ   20211229  585.0000  585.9900  569.7800  575.5500   \n",
      "2021-12-30  300750.SZ   20211230  577.0000  596.2900  575.5700  590.0000   \n",
      "2021-12-31  300750.SZ   20211231  598.3000  605.9900  582.0000  588.0000   \n",
      "\n",
      "            pre_close   change  pct_chg        vol       amount  \n",
      "date                                                             \n",
      "2018-06-11    25.0398  11.0160  43.9940     788.29     2845.471  \n",
      "2018-06-12    36.0558   3.6056  10.0001     265.79     1058.375  \n",
      "2018-06-13    39.6614   3.9641   9.9949     450.30     1972.314  \n",
      "2018-06-14    43.6255   4.3625   9.9999     742.67     3578.184  \n",
      "2018-06-15    47.9880   4.8008  10.0042    2565.10    13595.030  \n",
      "...               ...      ...      ...        ...          ...  \n",
      "2021-12-27   576.8000   1.8000   0.3121  101835.43  5915820.121  \n",
      "2021-12-28   578.6000   6.4000   1.1061   96595.74  5589310.230  \n",
      "2021-12-29   585.0000  -9.4500  -1.6154   99639.63  5731323.286  \n",
      "2021-12-30   575.5500  14.4500   2.5106  124209.46  7323088.709  \n",
      "2021-12-31   590.0000  -2.0000  -0.3390   93711.17  5540408.920  \n",
      "\n",
      "[868 rows x 11 columns]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'\\n                  High         Low        Open       Close    Volume   Adj Close\\nDate                                                                            \\n2021-09-17  371.410004  361.589996  371.410004  364.720001  26299000  364.720001\\n2021-09-20  361.029999  349.799988  359.299988  355.700012  19822800  355.700012\\n2021-09-21  360.040009  355.190002  358.500000  357.480011  11751900  357.480011\\n'"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from pickletools import optimize\n",
    "from statistics import mode\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import datetime as dt\n",
    "\n",
    "import tushare as ts\n",
    "import pandas_datareader as web\n",
    "\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense, Dropout, LSTM\n",
    "\n",
    "stock_code = '300750.SZ'\n",
    "\n",
    "features = ['open', 'high', 'low', 'close', 'vol', 'amount']\n",
    "\n",
    "ts.set_token('51fd5a77415e6299ad8243e387472a5552e3d24f5c889781caef6d89')\n",
    "data = ts.pro_bar(ts_code=stock_code, adj='qfq', start_date='20120101', end_date='20211231')\n",
    "data['date'] = pd.to_datetime(data['trade_date'], format = \"%Y/%m/%d %H:%M:%S\")\n",
    "data.set_index('date', inplace=True)  # 设置索引覆盖原来的数据\n",
    "data = data.sort_index(ascending=True)  # 将时间顺序升序，符合时间序列\n",
    "#data = data[['high', 'low', 'open', 'close', 'vol']]\n",
    "print(data)\n",
    "# ['ts_code', 'open', 'high', 'low', 'close', 'pre_close', 'change', 'pct_chg', 'vol', 'amount']\n",
    "\"\"\"\n",
    "company = 'FB'\n",
    "\n",
    "start = dt.datetime(2019, 1, 1)\n",
    "end = dt.datetime(2019, 1, 5)\n",
    "data = web.DataReader(company, 'yahoo', start, end)\n",
    "\n",
    "print(list(data))\n",
    "\"\"\"\n",
    "\"\"\"\n",
    "                  High         Low        Open       Close    Volume   Adj Close\n",
    "Date                                                                            \n",
    "2021-09-17  371.410004  361.589996  371.410004  364.720001  26299000  364.720001\n",
    "2021-09-20  361.029999  349.799988  359.299988  355.700012  19822800  355.700012\n",
    "2021-09-21  360.040009  355.190002  358.500000  357.480011  11751900  357.480011\n",
    "\"\"\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[[0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "   4.07744349e-04 1.23235534e-04]\n",
      "  [1.45871167e-02 5.49680903e-03 1.49415449e-02 5.53053467e-03\n",
      "   0.00000000e+00 0.00000000e+00]\n",
      "  [2.06032628e-02 1.15401584e-02 2.11038673e-02 1.16109630e-02\n",
      "   1.43986430e-04 6.30239008e-05]\n",
      "  ...\n",
      "  [3.45100137e-02 2.55097918e-02 3.53485153e-02 2.56663070e-02\n",
      "   1.79431705e-03 8.64509448e-04]\n",
      "  [4.25215758e-02 3.35575801e-02 4.35547371e-02 3.37634724e-02\n",
      "   8.51267972e-03 4.41939276e-03]\n",
      "  [5.13343093e-02 4.24101623e-02 5.25815966e-02 4.26703696e-02\n",
      "   1.00184386e-02 5.72193307e-03]]\n",
      "\n",
      " [[1.45871167e-02 5.49680903e-03 1.49415449e-02 5.53053467e-03\n",
      "   0.00000000e+00 0.00000000e+00]\n",
      "  [2.06032628e-02 1.15401584e-02 2.11038673e-02 1.16109630e-02\n",
      "   1.43986430e-04 6.30239008e-05]\n",
      "  [2.72240436e-02 1.81908766e-02 2.78855155e-02 1.83024866e-02\n",
      "   3.72143780e-04 1.73762354e-04]\n",
      "  ...\n",
      "  [4.25215758e-02 3.35575801e-02 4.35547371e-02 3.37634724e-02\n",
      "   8.51267972e-03 4.41939276e-03]\n",
      "  [5.13343093e-02 4.24101623e-02 5.25815966e-02 4.26703696e-02\n",
      "   1.00184386e-02 5.72193307e-03]\n",
      "  [6.10237935e-02 5.21434598e-02 6.25065096e-02 5.24633857e-02\n",
      "   9.40411421e-03 5.91822349e-03]]\n",
      "\n",
      " [[2.06032628e-02 1.15401584e-02 2.11038673e-02 1.16109630e-02\n",
      "   1.43986430e-04 6.30239008e-05]\n",
      "  [2.72240436e-02 1.81908766e-02 2.78855155e-02 1.83024866e-02\n",
      "   3.72143780e-04 1.73762354e-04]\n",
      "  [3.45100137e-02 2.55097918e-02 3.53485153e-02 2.56663070e-02\n",
      "   1.79431705e-03 8.64509448e-04]\n",
      "  ...\n",
      "  [5.13343093e-02 4.24101623e-02 5.25815966e-02 4.26703696e-02\n",
      "   1.00184386e-02 5.72193307e-03]\n",
      "  [6.10237935e-02 5.21434598e-02 6.25065096e-02 5.24633857e-02\n",
      "   9.40411421e-03 5.91822349e-03]\n",
      "  [6.32306193e-02 5.54384047e-02 5.15907376e-02 4.63676186e-02\n",
      "   1.00000000e+00 5.92951785e-01]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[9.39354407e-01 9.34445643e-01 8.97944939e-01 8.80679972e-01\n",
      "   1.20838827e-01 6.65275789e-01]\n",
      "  [8.74095135e-01 8.82139365e-01 8.76803297e-01 8.71261988e-01\n",
      "   9.78524086e-02 5.20816084e-01]\n",
      "  [8.78344576e-01 8.85661006e-01 8.90669727e-01 8.80725989e-01\n",
      "   6.50460396e-02 3.51852328e-01]\n",
      "  ...\n",
      "  [9.01412969e-01 8.98238295e-01 8.28488425e-01 8.29433255e-01\n",
      "   1.91827172e-01 9.87502934e-01]\n",
      "  [8.25833626e-01 8.41434683e-01 8.44811017e-01 8.32194228e-01\n",
      "   7.92620990e-02 4.07873345e-01]\n",
      "  [8.31600725e-01 8.40352274e-01 8.39230245e-01 8.42011019e-01\n",
      "   7.51731919e-02 3.85357699e-01]]\n",
      "\n",
      " [[8.74095135e-01 8.82139365e-01 8.76803297e-01 8.71261988e-01\n",
      "   9.78524086e-02 5.20816084e-01]\n",
      "  [8.78344576e-01 8.85661006e-01 8.90669727e-01 8.80725989e-01\n",
      "   6.50460396e-02 3.51852328e-01]\n",
      "  [8.93794329e-01 9.06699381e-01 9.04784882e-01 8.98871713e-01\n",
      "   6.20024251e-02 3.41599694e-01]\n",
      "  ...\n",
      "  [8.25833626e-01 8.41434683e-01 8.44811017e-01 8.32194228e-01\n",
      "   7.92620990e-02 4.07873345e-01]\n",
      "  [8.31600725e-01 8.40352274e-01 8.39230245e-01 8.42011019e-01\n",
      "   7.51731919e-02 3.85357699e-01]\n",
      "  [8.42224327e-01 8.38385643e-01 8.39028156e-01 8.27515913e-01\n",
      "   7.75485583e-02 3.95150713e-01]]\n",
      "\n",
      " [[8.78344576e-01 8.85661006e-01 8.90669727e-01 8.80725989e-01\n",
      "   6.50460396e-02 3.51852328e-01]\n",
      "  [8.93794329e-01 9.06699381e-01 9.04784882e-01 8.98871713e-01\n",
      "   6.20024251e-02 3.41599694e-01]\n",
      "  [9.01412969e-01 8.98238295e-01 8.28488425e-01 8.29433255e-01\n",
      "   1.91827172e-01 9.87502934e-01]\n",
      "  ...\n",
      "  [8.31600725e-01 8.40352274e-01 8.39230245e-01 8.42011019e-01\n",
      "   7.51731919e-02 3.85357699e-01]\n",
      "  [8.42224327e-01 8.38385643e-01 8.39028156e-01 8.27515913e-01\n",
      "   7.75485583e-02 3.95150713e-01]\n",
      "  [8.30083067e-01 8.54088198e-01 8.48028899e-01 8.49680387e-01\n",
      "   9.67221646e-02 5.04916535e-01]]]\n",
      "[[[0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "   4.07744349e-04 1.23235534e-04]\n",
      "  [1.45871167e-02 5.49680903e-03 1.49415449e-02 5.53053467e-03\n",
      "   0.00000000e+00 0.00000000e+00]\n",
      "  [2.06032628e-02 1.15401584e-02 2.11038673e-02 1.16109630e-02\n",
      "   1.43986430e-04 6.30239008e-05]\n",
      "  ...\n",
      "  [3.45100137e-02 2.55097918e-02 3.53485153e-02 2.56663070e-02\n",
      "   1.79431705e-03 8.64509448e-04]\n",
      "  [4.25215758e-02 3.35575801e-02 4.35547371e-02 3.37634724e-02\n",
      "   8.51267972e-03 4.41939276e-03]\n",
      "  [5.13343093e-02 4.24101623e-02 5.25815966e-02 4.26703696e-02\n",
      "   1.00184386e-02 5.72193307e-03]]\n",
      "\n",
      " [[1.45871167e-02 5.49680903e-03 1.49415449e-02 5.53053467e-03\n",
      "   0.00000000e+00 0.00000000e+00]\n",
      "  [2.06032628e-02 1.15401584e-02 2.11038673e-02 1.16109630e-02\n",
      "   1.43986430e-04 6.30239008e-05]\n",
      "  [2.72240436e-02 1.81908766e-02 2.78855155e-02 1.83024866e-02\n",
      "   3.72143780e-04 1.73762354e-04]\n",
      "  ...\n",
      "  [4.25215758e-02 3.35575801e-02 4.35547371e-02 3.37634724e-02\n",
      "   8.51267972e-03 4.41939276e-03]\n",
      "  [5.13343093e-02 4.24101623e-02 5.25815966e-02 4.26703696e-02\n",
      "   1.00184386e-02 5.72193307e-03]\n",
      "  [6.10237935e-02 5.21434598e-02 6.25065096e-02 5.24633857e-02\n",
      "   9.40411421e-03 5.91822349e-03]]\n",
      "\n",
      " [[2.06032628e-02 1.15401584e-02 2.11038673e-02 1.16109630e-02\n",
      "   1.43986430e-04 6.30239008e-05]\n",
      "  [2.72240436e-02 1.81908766e-02 2.78855155e-02 1.83024866e-02\n",
      "   3.72143780e-04 1.73762354e-04]\n",
      "  [3.45100137e-02 2.55097918e-02 3.53485153e-02 2.56663070e-02\n",
      "   1.79431705e-03 8.64509448e-04]\n",
      "  ...\n",
      "  [5.13343093e-02 4.24101623e-02 5.25815966e-02 4.26703696e-02\n",
      "   1.00184386e-02 5.72193307e-03]\n",
      "  [6.10237935e-02 5.21434598e-02 6.25065096e-02 5.24633857e-02\n",
      "   9.40411421e-03 5.91822349e-03]\n",
      "  [6.32306193e-02 5.54384047e-02 5.15907376e-02 4.63676186e-02\n",
      "   1.00000000e+00 5.92951785e-01]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[9.39354407e-01 9.34445643e-01 8.97944939e-01 8.80679972e-01\n",
      "   1.20838827e-01 6.65275789e-01]\n",
      "  [8.74095135e-01 8.82139365e-01 8.76803297e-01 8.71261988e-01\n",
      "   9.78524086e-02 5.20816084e-01]\n",
      "  [8.78344576e-01 8.85661006e-01 8.90669727e-01 8.80725989e-01\n",
      "   6.50460396e-02 3.51852328e-01]\n",
      "  ...\n",
      "  [9.01412969e-01 8.98238295e-01 8.28488425e-01 8.29433255e-01\n",
      "   1.91827172e-01 9.87502934e-01]\n",
      "  [8.25833626e-01 8.41434683e-01 8.44811017e-01 8.32194228e-01\n",
      "   7.92620990e-02 4.07873345e-01]\n",
      "  [8.31600725e-01 8.40352274e-01 8.39230245e-01 8.42011019e-01\n",
      "   7.51731919e-02 3.85357699e-01]]\n",
      "\n",
      " [[8.74095135e-01 8.82139365e-01 8.76803297e-01 8.71261988e-01\n",
      "   9.78524086e-02 5.20816084e-01]\n",
      "  [8.78344576e-01 8.85661006e-01 8.90669727e-01 8.80725989e-01\n",
      "   6.50460396e-02 3.51852328e-01]\n",
      "  [8.93794329e-01 9.06699381e-01 9.04784882e-01 8.98871713e-01\n",
      "   6.20024251e-02 3.41599694e-01]\n",
      "  ...\n",
      "  [8.25833626e-01 8.41434683e-01 8.44811017e-01 8.32194228e-01\n",
      "   7.92620990e-02 4.07873345e-01]\n",
      "  [8.31600725e-01 8.40352274e-01 8.39230245e-01 8.42011019e-01\n",
      "   7.51731919e-02 3.85357699e-01]\n",
      "  [8.42224327e-01 8.38385643e-01 8.39028156e-01 8.27515913e-01\n",
      "   7.75485583e-02 3.95150713e-01]]\n",
      "\n",
      " [[8.78344576e-01 8.85661006e-01 8.90669727e-01 8.80725989e-01\n",
      "   6.50460396e-02 3.51852328e-01]\n",
      "  [8.93794329e-01 9.06699381e-01 9.04784882e-01 8.98871713e-01\n",
      "   6.20024251e-02 3.41599694e-01]\n",
      "  [9.01412969e-01 8.98238295e-01 8.28488425e-01 8.29433255e-01\n",
      "   1.91827172e-01 9.87502934e-01]\n",
      "  ...\n",
      "  [8.31600725e-01 8.40352274e-01 8.39230245e-01 8.42011019e-01\n",
      "   7.51731919e-02 3.85357699e-01]\n",
      "  [8.42224327e-01 8.38385643e-01 8.39028156e-01 8.27515913e-01\n",
      "   7.75485583e-02 3.95150713e-01]\n",
      "  [8.30083067e-01 8.54088198e-01 8.48028899e-01 8.49680387e-01\n",
      "   9.67221646e-02 5.04916535e-01]]]\n"
     ]
    }
   ],
   "source": [
    "# prepare data\n",
    "# scale down the \n",
    "scaler = MinMaxScaler(feature_range=(0,1))\n",
    "raw_data = data[features]\n",
    "\n",
    "before_scale = raw_data.values.reshape(-1, len(features))\n",
    "\n",
    "scaled_data = scaler.fit_transform(before_scale)\n",
    "\n",
    "prediction_days = 7\n",
    "#print(scaled_data)\n",
    "#print('------')\n",
    "\n",
    "x_train = []\n",
    "y_train = []\n",
    "\n",
    "for x in range(prediction_days, len(scaled_data)):\n",
    "    x_train.append(scaled_data[x-prediction_days: x, :])\n",
    "    y_train.append(scaled_data[x, 0])\n",
    "\n",
    "x_train, y_train = np.array(x_train), np.array(y_train)\n",
    "print(x_train)\n",
    "x_train = np.reshape(x_train, (x_train.shape[0], x_train.shape[1], len(features)))\n",
    "\n",
    "print(x_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/25\n",
      "27/27 [==============================] - 5s 10ms/step - loss: 0.0448\n",
      "Epoch 2/25\n",
      "27/27 [==============================] - 0s 8ms/step - loss: 0.0041\n",
      "Epoch 3/25\n",
      "27/27 [==============================] - 0s 8ms/step - loss: 0.0026\n",
      "Epoch 4/25\n",
      "27/27 [==============================] - 0s 8ms/step - loss: 0.0029\n",
      "Epoch 5/25\n",
      "27/27 [==============================] - 0s 8ms/step - loss: 0.0028\n",
      "Epoch 6/25\n",
      "27/27 [==============================] - 0s 8ms/step - loss: 0.0025\n",
      "Epoch 7/25\n",
      "27/27 [==============================] - 0s 8ms/step - loss: 0.0022\n",
      "Epoch 8/25\n",
      "27/27 [==============================] - 0s 8ms/step - loss: 0.0018\n",
      "Epoch 9/25\n",
      "27/27 [==============================] - 0s 9ms/step - loss: 0.0024\n",
      "Epoch 10/25\n",
      "27/27 [==============================] - 0s 8ms/step - loss: 0.0020\n",
      "Epoch 11/25\n",
      "27/27 [==============================] - 0s 8ms/step - loss: 0.0019\n",
      "Epoch 12/25\n",
      "27/27 [==============================] - 0s 8ms/step - loss: 0.0019\n",
      "Epoch 13/25\n",
      "27/27 [==============================] - 0s 8ms/step - loss: 0.0018\n",
      "Epoch 14/25\n",
      "27/27 [==============================] - 0s 8ms/step - loss: 0.0017\n",
      "Epoch 15/25\n",
      "27/27 [==============================] - 0s 8ms/step - loss: 0.0018\n",
      "Epoch 16/25\n",
      "27/27 [==============================] - 0s 8ms/step - loss: 0.0019\n",
      "Epoch 17/25\n",
      "27/27 [==============================] - 0s 8ms/step - loss: 0.0017\n",
      "Epoch 18/25\n",
      "27/27 [==============================] - 0s 8ms/step - loss: 0.0017\n",
      "Epoch 19/25\n",
      "27/27 [==============================] - 0s 8ms/step - loss: 0.0016\n",
      "Epoch 20/25\n",
      "27/27 [==============================] - 0s 8ms/step - loss: 0.0016\n",
      "Epoch 21/25\n",
      "27/27 [==============================] - 0s 8ms/step - loss: 0.0018\n",
      "Epoch 22/25\n",
      "27/27 [==============================] - 0s 8ms/step - loss: 0.0024\n",
      "Epoch 23/25\n",
      "27/27 [==============================] - 0s 8ms/step - loss: 0.0017\n",
      "Epoch 24/25\n",
      "27/27 [==============================] - 0s 8ms/step - loss: 0.0016\n",
      "Epoch 25/25\n",
      "27/27 [==============================] - 0s 8ms/step - loss: 0.0016\n"
     ]
    }
   ],
   "source": [
    "# build the model\n",
    "model = Sequential()\n",
    "\n",
    "model.add(LSTM(units=50, return_sequences=True, input_shape=(x_train.shape[1], len(features))))\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=1))\n",
    "\n",
    "model.compile(optimizer='adam', loss='mean_squared_error')\n",
    "model.fit(x_train, y_train, epochs=25, batch_size=32)\n",
    "\n",
    "model.save('ningdeshidai.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_data = ts.pro_bar(ts_code=stock_code, adj='qfq', start_date='20220101')\n",
    "test_data['date'] = pd.to_datetime(test_data['trade_date'], format = \"%Y/%m/%d %H:%M:%S\")\n",
    "test_data.set_index('date', inplace=True)  # 设置索引覆盖原来的数据\n",
    "test_data = test_data.sort_index(ascending=True)  # 将时间顺序升序，符合时间序列\n",
    "#test_data = test_data[['high', 'low', 'open', 'close', 'vol']]\n",
    "actual_prices = test_data['close']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "def inverse_predictions(predictions,scaler,prediction_index=2):\n",
    "    '''This function uses the fitted scaler to inverse predictions, \n",
    "    the index should be set to the position of the target variable'''\n",
    "    \n",
    "    max_val = scaler.data_max_[prediction_index]\n",
    "    min_val = scaler.data_min_[prediction_index]\n",
    "    original_values = (predictions*(max_val - min_val )) + min_val\n",
    "    \n",
    "    return original_values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6/6 [==============================] - 0s 4ms/step\n",
      "(1, 7, 6)\n",
      "1/1 [==============================] - 0s 32ms/step\n",
      "[[454.84143]]\n"
     ]
    }
   ],
   "source": [
    "total_dataset = pd.concat((raw_data, test_data[features]), axis=0)\n",
    "\n",
    "model_inputs = total_dataset[len(total_dataset) - len(test_data) - prediction_days:].values\n",
    "model_inputs = model_inputs.reshape(-1, len(features))\n",
    "model_inputs = scaler.transform(model_inputs)\n",
    "\n",
    "x_test = []\n",
    "for x in range(prediction_days, len(model_inputs)):\n",
    "    x_test.append(model_inputs[x-prediction_days:x, :])\n",
    "\n",
    "x_test = np.array(x_test)\n",
    "x_test = np.reshape(x_test, (x_test.shape[0], x_test.shape[1], len(features)))\n",
    "\n",
    "predicted_prices = model.predict(x_test)\n",
    "#print(predicted_prices.shape)\n",
    "#predicted_prices = predicted_prices.reshape(-1, 1)\n",
    "#predicted_prices = scaler.inverse_transform(predicted_prices)\n",
    "predicted_prices = inverse_predictions(predicted_prices, scaler, 0)\n",
    "#print(predicted_prices)\n",
    "#print(type(predicted_prices))\n",
    "actual_prices = test_data['close']\n",
    "#print(actual_prices.values)\n",
    "#print(type(actual_prices.values))\n",
    "\n",
    "real_data = [model_inputs[-prediction_days : , :]]\n",
    "\n",
    "\n",
    "real_data = np.array(real_data)\n",
    "print(real_data.shape)\n",
    "real_data = np.reshape(real_data, (real_data.shape[0], real_data.shape[1], len(features)))\n",
    "\n",
    "predict_tomorrow = model.predict(real_data)\n",
    "predict_tomorrow = inverse_predictions(predict_tomorrow, scaler, 0) \n",
    "\n",
    "print(predict_tomorrow)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.4682080924855491\n"
     ]
    }
   ],
   "source": [
    "from operator import xor\n",
    "#print(actual_prices.values)\n",
    "predicted_prices = predicted_prices.reshape((-1, ))\n",
    "\n",
    "real_up_down = np.diff(actual_prices.values) > 0\n",
    "pred_up_down = np.diff(predicted_prices) > 0\n",
    "pred_up_down = pred_up_down.reshape((-1, ))\n",
    "\n",
    "t = (real_up_down == pred_up_down).astype(int)\n",
    "\n",
    "print(float(sum(t)) / len(t))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# create a dataframe and use the index from another dataframe\n",
    "predicted_prices = pd.DataFrame(predicted_prices, index = actual_prices.index)\n",
    "plt.plot(actual_prices, color='black', label=f'Actual price')\n",
    "plt.plot(predicted_prices, color='green', label='Predicted price')\n",
    "plt.title('Share Price')\n",
    "plt.xlabel('Time')\n",
    "plt.ylabel('share price')\n",
    "plt.gcf().autofmt_xdate()\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.7.13 ('py37_vnpy')",
   "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": "3.7.13"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "aba7986fadd20fda8637e8ea5870161e5c73789f03e96b3ff06d86921c0a4a03"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}