grammian_make.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "donee\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import csv\n",
    "import math\n",
    "import matplotlib.pyplot as plt\n",
    "from pyts.image import MarkovTransitionField\n",
    "import warnings\n",
    "import os\n",
    "\n",
    "print('donee')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2992,)\n",
      "(2962, 30, 30, 5)\n"
     ]
    }
   ],
   "source": [
    "warnings.filterwarnings('ignore') \n",
    "#SVC(kernel = 'rbf', gamma = 1e-3, C = 100), day_num = 7, WINDOW = 50\n",
    "\n",
    "def play_sound():\n",
    "    os.system(\"printf '\\a'\")\n",
    "\n",
    "def tabulate(x, y, f):\n",
    "    \"\"\"Return a table of f(x, y). Useful for the Gram-like operations.\"\"\"\n",
    "    return np.vectorize(f)(*np.meshgrid(x, y, sparse=True))\n",
    "\n",
    "def cos_sum(a, b):\n",
    "    \"\"\"To work with tabulate.\"\"\"\n",
    "    return(math.cos(a+b))\n",
    "\n",
    "def sin_diff(a, b):\n",
    "    \"\"\"To work with tabulate.\"\"\"\n",
    "    return(math.sin(a-b))\n",
    "\n",
    "def create_time_serie(size, time):\n",
    "    \"\"\"Generate a time serie of length size and dynamic with respect to time.\"\"\"\n",
    "    # Generating time-series\n",
    "    support = np.arange(0, size)\n",
    "    serie = np.cos(support + float(time))\n",
    "    return(t, serie)\n",
    "\n",
    "class GASF:\n",
    "\n",
    "    def __init__(self):\n",
    "        pass\n",
    "    def transform(self, serie):\n",
    "        \"\"\"Compute the Gramian Angular Field of an image\"\"\"\n",
    "        # Min-Max scaling\n",
    "        min_ = np.amin(serie)\n",
    "        max_ = np.amax(serie)\n",
    "        scaled_serie = (2*serie - max_ - min_)/(max_ - min_)\n",
    "\n",
    "        # Floating point inaccuracy!\n",
    "        scaled_serie = np.where(scaled_serie >= 1., 1., scaled_serie)\n",
    "        scaled_serie = np.where(scaled_serie <= -1., -1., scaled_serie)\n",
    "\n",
    "        # Polar encoding\n",
    "        phi = np.arccos(scaled_serie)\n",
    "        # Note! The computation of r is not necessary\n",
    "        r = np.linspace(0, 1, len(scaled_serie))\n",
    "\n",
    "        # GAF Computation (every term of the matrix)\n",
    "        gaf = tabulate(phi, phi, cos_sum)\n",
    "\n",
    "        return(gaf, phi, r, scaled_serie)\n",
    "    \n",
    "class GADF:\n",
    "\n",
    "    def __init__(self):\n",
    "        pass\n",
    "    def transform(self, serie):\n",
    "        \"\"\"Compute the Gramian Angular Field of an image\"\"\"\n",
    "        # Min-Max scaling\n",
    "        min_ = np.amin(serie)\n",
    "        max_ = np.amax(serie)\n",
    "        scaled_serie = (2*serie - max_ - min_)/(max_ - min_)\n",
    "\n",
    "        # Floating point inaccuracy!\n",
    "        scaled_serie = np.where(scaled_serie >= 1., 1., scaled_serie)\n",
    "        scaled_serie = np.where(scaled_serie <= -1., -1., scaled_serie)\n",
    "\n",
    "        # Polar encoding\n",
    "        phi = np.arccos(scaled_serie)\n",
    "        # Note! The computation of r is not necessary\n",
    "        r = np.linspace(0, 1, len(scaled_serie))\n",
    "\n",
    "        # GAF Computation (every term of the matrix)\n",
    "        gaf = tabulate(phi, phi, sin_diff)\n",
    "\n",
    "        return(gaf, phi, r, scaled_serie)\n",
    "\n",
    "def csv_extract(day_num, filename = \"/MERITZ_DATA.csv\"):\n",
    "    '''\n",
    "    Input: Number of samples, Number of Stocks, filename\n",
    "    Output: numpy array of (number of samples, number of stocks)\n",
    "    '''\n",
    "    df = pd.read_csv(filename)\n",
    "    sample_size = df['Open'].shape[0] - day_num\n",
    "    X = np.empty((sample_size - day_num, 0))\n",
    "    y = np.empty((sample_size - day_num))\n",
    "    print(y.shape)\n",
    "\n",
    "    #Feature Processing. Trying to keep in range [-1, 1]\n",
    "    \n",
    "    features = ['Open', 'High', 'Low', 'Close', 'Price_5', 'Price_10', 'Price_20',\n",
    "                'Price_60', 'Price_120', 'Volume', 'Volume_5', 'Volume_10',\n",
    "                'Volume_20', 'Volume_60', 'Volume_120', 'Volume Ratio', 'OBV',\n",
    "                'MACD', 'MACD_Signal', 'Momentum', 'Momentum_Signal', 'William_R',\n",
    "                'William_D', 'Psychology', 'Psychology_Signal', 'MFI',\n",
    "                'MFI_Signal', 'RSI', 'RSI_Signal']    \n",
    "    for feature in features:\n",
    "        if feature in ['Open', 'High', 'Low', 'Close', 'Volume']:\n",
    "            X_col = df[feature]\n",
    "        else:\n",
    "            continue\n",
    "        X = np.column_stack((X, X_col[0 : sample_size - day_num]))\n",
    "        \n",
    "    for n in range(sample_size - day_num):\n",
    "        close = df['Close']\n",
    "        pct = 100 * (close[n + day_num] - close[n]) / close[n]\n",
    "        if pct > 1.:\n",
    "            y[n] = 1\n",
    "        else:\n",
    "            y[n] = 0\n",
    "        #y[n] = 1 if df['Close'][n + day_num] >= df['Close'][n] else 0\n",
    "    return X, y\n",
    "\n",
    "def feature_extract(X, window = 10):\n",
    "    feature_num = X.shape[1]\n",
    "    for i in range(window, X.shape[0]):f\n",
    "        x = X[i - window: i]\n",
    "        x = x.T.reshape(x.shape[0] * x.shape[1])\n",
    "        for j in np.arange(0, window*feature_num, window):\n",
    "            #x[j : j + window] = (x[j : j + window])\n",
    "            x[j : j + window] = (x[j : j + window])/np.max(x[j : j + window])\n",
    "        if i == window:\n",
    "            X_ = x\n",
    "            continue\n",
    "        X_ = np.vstack((X_, x))\n",
    "    return X_\n",
    "\n",
    "filename = \"./Data/HYNIX_DATA.csv\"\n",
    "WINDOW = 30\n",
    "FEATURE_NUM = 5\n",
    "feature_list = ['_open', '_high', '_low', '_close', '_volume']\n",
    "window = WINDOW\n",
    "feature_num = FEATURE_NUM\n",
    "day_num = 4\n",
    "start_lim = 0\n",
    "X, y = csv_extract(day_num, filename = filename)\n",
    "X, y = feature_extract(X[start_lim:], window = WINDOW), y[start_lim + WINDOW:]\n",
    "path = './Data/Grammian3/'\n",
    "np.savetxt(path + 'labels', y)\n",
    "for c in range(X.shape[0]):\n",
    "    mtf = MarkovTransitionField(image_size = WINDOW)\n",
    "    gasf, gadf = GASF(), GADF()\n",
    "    for n, j in enumerate(np.arange(0, window*feature_num, window)):\n",
    "        x = X[c, j : j + window]\n",
    "        x_gasf = gasf.transform(x)[0]\n",
    "        feature_name = feature_list[n]\n",
    "        gasf_name = path + str(c) + feature_name + '_gasf'\n",
    "        try:\n",
    "            np.savetxt(gasf_name, x_gasf)\n",
    "        except Exception:\n",
    "            os.mkdir(path)\n",
    "            np.savetxt(gasf_name, x_gasf)\n",
    "        if n == 0:\n",
    "            X_gasf_ = x_gasf\n",
    "        elif n == 1:\n",
    "            X_gasf_ = np.stack((X_gasf_, x_gasf), axis = -1)\n",
    "        else:\n",
    "            X_gasf_ = np.append(X_gasf_, np.expand_dims(x_gasf, axis= -1), axis = -1)\n",
    "\n",
    "    if c == 0:\n",
    "        X_gasf = X_gasf_\n",
    "    elif c == 1:\n",
    "        X_gasf = np.stack((X_gasf, X_gasf_), axis = 0)\n",
    "    else:\n",
    "        X_gasf = np.append(X_gasf, np.array([X_gasf_]), axis = 0 )\n",
    "#X_tot = np.concatenate((X_gadf, X_gasf, X_mtf), axis = -1)\n",
    "X_tot = X_gasf\n",
    "print(X_tot.shape)\n",
    "np.save(path + 'X_tot', X_tot)\n"
   ]
  }
 ],
 "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": "3.9.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}