genetic_programming.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "8d427c8f",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from gplearn.functions import make_function\n",
    "from gplearn.fitness import make_fitness\n",
    "from gplearn.genetic import SymbolicTransformer\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "b0152f30",
   "metadata": {},
   "outputs": [],
   "source": [
    "alpha = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "0ca802fb",
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in range(1, 6):\n",
    "    path = \"alpha\" + str(i) + \".csv\"\n",
    "    alpha.append(pd.read_csv(path, index_col=0).iloc[:, 1])\n",
    "alpha.append(pd.read_csv(path, index_col=0).iloc[:, 0])\n",
    "alpha_matrix = pd.concat(alpha, axis = 1)\n",
    "alpha_matrix.columns = [\"alpha1\", \"alpha2\", \"alpha3\", \"alpha4\", \"alpha5\", \"return\"]\n",
    "alpha_matrix.replace(-np.inf, np.nan, inplace=True)\n",
    "alpha_matrix.replace(np.inf, np.nan, inplace=True)\n",
    "alpha_matrix.dropna(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "b163647b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>alpha1</th>\n",
       "      <th>alpha2</th>\n",
       "      <th>alpha3</th>\n",
       "      <th>alpha4</th>\n",
       "      <th>alpha5</th>\n",
       "      <th>return</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Time</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2022-08-22 09:02:00</th>\n",
       "      <td>0.002226</td>\n",
       "      <td>0.000258</td>\n",
       "      <td>5.003409</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>811.301151</td>\n",
       "      <td>0.001487</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-08-22 09:03:00</th>\n",
       "      <td>-0.001487</td>\n",
       "      <td>0.000442</td>\n",
       "      <td>4.995295</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>132.249017</td>\n",
       "      <td>0.002227</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-08-22 09:04:00</th>\n",
       "      <td>-0.002227</td>\n",
       "      <td>0.001971</td>\n",
       "      <td>4.996566</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>323.119896</td>\n",
       "      <td>-0.002222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-08-22 09:05:00</th>\n",
       "      <td>0.002222</td>\n",
       "      <td>0.000389</td>\n",
       "      <td>5.003737</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>177.065677</td>\n",
       "      <td>0.000742</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-08-22 09:06:00</th>\n",
       "      <td>-0.000742</td>\n",
       "      <td>-0.000857</td>\n",
       "      <td>4.997188</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>131.048609</td>\n",
       "      <td>-0.001484</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-09-01 14:56:00</th>\n",
       "      <td>-0.000716</td>\n",
       "      <td>-0.000590</td>\n",
       "      <td>4.999999</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>409.146333</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-09-01 14:57:00</th>\n",
       "      <td>-0.000000</td>\n",
       "      <td>-0.001156</td>\n",
       "      <td>4.999572</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>949.338687</td>\n",
       "      <td>0.002146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-09-01 14:58:00</th>\n",
       "      <td>-0.002146</td>\n",
       "      <td>0.000257</td>\n",
       "      <td>4.996268</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>87.531239</td>\n",
       "      <td>-0.000714</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-09-01 14:59:00</th>\n",
       "      <td>0.000714</td>\n",
       "      <td>-0.003100</td>\n",
       "      <td>5.003117</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>256.091527</td>\n",
       "      <td>-0.001071</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-09-01 15:00:00</th>\n",
       "      <td>0.001071</td>\n",
       "      <td>-0.000003</td>\n",
       "      <td>4.998769</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>10.718116</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1640 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                       alpha1    alpha2    alpha3    alpha4      alpha5  \\\n",
       "Time                                                                      \n",
       "2022-08-22 09:02:00  0.002226  0.000258  5.003409  8.000000  811.301151   \n",
       "2022-08-22 09:03:00 -0.001487  0.000442  4.995295  0.200000  132.249017   \n",
       "2022-08-22 09:04:00 -0.002227  0.001971  4.996566  0.500000  323.119896   \n",
       "2022-08-22 09:05:00  0.002222  0.000389  5.003737  6.000000  177.065677   \n",
       "2022-08-22 09:06:00 -0.000742 -0.000857  4.997188  0.250000  131.048609   \n",
       "...                       ...       ...       ...       ...         ...   \n",
       "2022-09-01 14:56:00 -0.000716 -0.000590  4.999999  0.666667  409.146333   \n",
       "2022-09-01 14:57:00 -0.000000 -0.001156  4.999572  1.000000  949.338687   \n",
       "2022-09-01 14:58:00 -0.002146  0.000257  4.996268  0.166667   87.531239   \n",
       "2022-09-01 14:59:00  0.000714 -0.003100  5.003117  3.000000  256.091527   \n",
       "2022-09-01 15:00:00  0.001071 -0.000003  4.998769  0.666667   10.718116   \n",
       "\n",
       "                       return  \n",
       "Time                           \n",
       "2022-08-22 09:02:00  0.001487  \n",
       "2022-08-22 09:03:00  0.002227  \n",
       "2022-08-22 09:04:00 -0.002222  \n",
       "2022-08-22 09:05:00  0.000742  \n",
       "2022-08-22 09:06:00 -0.001484  \n",
       "...                       ...  \n",
       "2022-09-01 14:56:00  0.000000  \n",
       "2022-09-01 14:57:00  0.002146  \n",
       "2022-09-01 14:58:00 -0.000714  \n",
       "2022-09-01 14:59:00 -0.001071  \n",
       "2022-09-01 15:00:00  0.000000  \n",
       "\n",
       "[1640 rows x 6 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "alpha_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "c527cc10",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = alpha_matrix.iloc[:, :5]\n",
    "y = alpha_matrix.iloc[:, 5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "16c9ab1f",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train,X_test,y_train,y_test = train_test_split(X,y,test_size=0.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "116ceca7",
   "metadata": {},
   "outputs": [],
   "source": [
    "init_function = ['add', 'sub', 'mul', 'div', 'sqrt', 'log',\n",
    "                      'abs', 'neg', 'inv', 'max', 'min']\n",
    "\n",
    "function_set = init_function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "ec544a2d",
   "metadata": {},
   "outputs": [],
   "source": [
    "def _my_metric(y, y_pred, w):\n",
    "    return np.sum((y - y_pred)**2)\n",
    "metric = make_fitness(function=_my_metric, greater_is_better=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "5ff09142",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    |   Population Average    |             Best Individual              |\n",
      "---- ------------------------- ------------------------------------------ ----------\n",
      " Gen   Length          Fitness   Length          Fitness      OOB Fitness  Time Left\n",
      "   0     3.59      3.39942e+12        3       0.00114316              N/A      0.81s\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "SymbolicTransformer(feature_names=Index(['alpha1', 'alpha2', 'alpha3', 'alpha4', 'alpha5'], dtype='object'),\n",
       "                    function_set=['add', 'sub', 'mul', 'div', 'sqrt', 'log',\n",
       "                                  'abs', 'neg', 'inv', 'max', 'min'],\n",
       "                    generations=10, init_depth=(1, 2),\n",
       "                    metric=<gplearn.fitness._Fitness object at 0x000002705E9B51C0>,\n",
       "                    n_jobs=-1, p_crossover=0.6, p_hoist_mutation=0,\n",
       "                    p_point_mutation=0.1, p_point_replace=0.4,\n",
       "                    p_subtree_mutation=0.1, verbose=1)"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    " gpmodel = SymbolicTransformer(\n",
    "    feature_names = alpha_matrix.columns[:5], \n",
    "    function_set = function_set,\n",
    "    metric = metric,\n",
    "    generations = 10,\n",
    "    init_depth = (1,2),\n",
    "    population_size = 1000,\n",
    "    tournament_size = 20, \n",
    "    p_crossover = 0.6, \n",
    "    p_subtree_mutation = 0.1,\n",
    "    p_hoist_mutation = 0, \n",
    "    p_point_mutation = 0.1,\n",
    "    p_point_replace = 0.4,\n",
    "    verbose = 1,\n",
    "    n_jobs = -1\n",
    ")\n",
    "\n",
    "gpmodel.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "b3a25b2a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[mul(0.177, alpha2),\n",
      " mul(alpha2, alpha2),\n",
      " mul(mul(alpha2, alpha1), abs(alpha4)),\n",
      " abs(mul(alpha1, alpha1)),\n",
      " mul(alpha1, alpha2),\n",
      " min(min(alpha3, alpha4), mul(0.174, alpha1)),\n",
      " mul(sqrt(alpha4), div(alpha1, alpha5)),\n",
      " div(max(alpha1, alpha2), div(alpha3, alpha2)),\n",
      " sqrt(mul(alpha2, alpha1)),\n",
      " min(abs(alpha1), alpha4)]\n"
     ]
    }
   ],
   "source": [
    "print(gpmodel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2b034e49",
   "metadata": {},
   "outputs": [],
   "source": [
    "# The IC of mul(alpha2, alpha2)\n",
    "print(\"training dataset:\\n {}\".format(np.corrcoef(X_train.iloc[:, 1]**2, y_train)))\n",
    "print(\"testing dataset:\\n {}\".format(np.corrcoef(X_test.iloc[:, 1]**2, y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "05049672",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "training dataset:\n",
      " [[1.         0.10084595]\n",
      " [0.10084595 1.        ]]\n",
      "testing dataset:\n",
      " [[1.         0.10619565]\n",
      " [0.10619565 1.        ]]\n"
     ]
    }
   ],
   "source": [
    "# The IC of mul(mul(alpha2, alpha1), abs(alpha4))\n",
    "print(\"training dataset:\\n {}\".format(np.corrcoef(X_train.iloc[:, 1]*X_train.iloc[:, 0]*X_train.iloc[:, 3].abs(), y_train)))\n",
    "print(\"testing dataset:\\n {}\".format(np.corrcoef((X_test.iloc[:, 1]*X_test.iloc[:, 0]*X_test.iloc[:, 3].abs(), y_test))))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "3d226c3e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "training dataset:\n",
      " [[ 1.         -0.01632279]\n",
      " [-0.01632279  1.        ]]\n",
      "testing dataset:\n",
      " [[ 1.         -0.03278857]\n",
      " [-0.03278857  1.        ]]\n"
     ]
    }
   ],
   "source": [
    "# The IC of mul(alpha1, alpha1)\n",
    "print(\"training dataset:\\n {}\".format(np.corrcoef(X_train.iloc[:, 0]**2, y_train)))\n",
    "print(\"testing dataset:\\n {}\".format(np.corrcoef((X_test.iloc[:, 0]**2, y_test))))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "711a052e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "training dataset:\n",
      " [[ 1.         -0.04067354]\n",
      " [-0.04067354  1.        ]]\n",
      "testing dataset:\n",
      " [[1.         0.12176642]\n",
      " [0.12176642 1.        ]]\n"
     ]
    }
   ],
   "source": [
    "# The IC of mul(alpha1, alpha2)\n",
    "print(\"training dataset:\\n {}\".format(np.corrcoef(X_train.iloc[:, 1]*X_train.iloc[:, 0], y_train)))\n",
    "print(\"testing dataset:\\n {}\".format(np.corrcoef((X_test.iloc[:, 1]*X_test.iloc[:, 0], y_test))))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "1574ca17",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "training dataset:\n",
      " [[1.         0.07337739]\n",
      " [0.07337739 1.        ]]\n",
      "testing dataset:\n",
      " [[1.         0.16958539]\n",
      " [0.16958539 1.        ]]\n"
     ]
    }
   ],
   "source": [
    "# The IC of min(min(alpha3, alpha4), mul(0.174, alpha1))\n",
    "print(\"training dataset:\\n {}\".format(np.corrcoef(np.minimum(np.minimum(X_train.iloc[:, 2], X_train.iloc[:, 3]), 0.173*X_train.iloc[:, 0]), y_train)))\n",
    "print(\"testing dataset:\\n {}\".format(np.corrcoef((np.minimum(np.minimum(X_test.iloc[:, 2], X_test.iloc[:, 3]), 0.173*X_test.iloc[:, 0]), y_test))))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "401e243a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "training dataset:\n",
      " [[1.         0.02877139]\n",
      " [0.02877139 1.        ]]\n",
      "testing dataset:\n",
      " [[1.         0.06361156]\n",
      " [0.06361156 1.        ]]\n"
     ]
    }
   ],
   "source": [
    "# The IC of mul(sqrt(alpha4), div(alpha1, alpha5))\n",
    "print(\"training dataset:\\n {}\".format(np.corrcoef(X_train.iloc[:, 3]**(1/2)*X_train.iloc[:, 0]/X_train.iloc[:, 4], y_train)))\n",
    "print(\"testing dataset:\\n {}\".format(np.corrcoef(X_test.iloc[:, 3]**(1/2)*X_test.iloc[:, 0]/X_test.iloc[:, 4], y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "970dd40d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "training dataset:\n",
      " [[1.         0.11352925]\n",
      " [0.11352925 1.        ]]\n",
      "testing dataset:\n",
      " [[1.        0.0674851]\n",
      " [0.0674851 1.       ]]\n"
     ]
    }
   ],
   "source": [
    "# The IC of div(max(alpha1, alpha2), div(alpha3, alpha2))\n",
    "print(\"training dataset:\\n {}\".format(np.corrcoef(np.maximum(X_train.iloc[:, 0], X_train.iloc[:, 2])/(X_train.iloc[:, 2]/X_train.iloc[:, 1]), y_train)))\n",
    "print(\"testing dataset:\\n {}\".format(np.corrcoef(np.maximum(X_test.iloc[:, 0], X_test.iloc[:, 2])/(X_test.iloc[:, 2]/X_test.iloc[:, 1]), y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "10f23364",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "training dataset:\n",
      " [[1.         0.07264207]\n",
      " [0.07264207 1.        ]]\n",
      "testing dataset:\n",
      " [[1.         0.06705202]\n",
      " [0.06705202 1.        ]]\n"
     ]
    }
   ],
   "source": [
    "# The IC of abs(sqrt(mul(alpha1, alpha2)))\n",
    "print(\"training dataset:\\n {}\".format(np.corrcoef((X_train.iloc[:, 1]*X_train.iloc[:, 0]).abs()**(1/2), y_train)))\n",
    "print(\"testing dataset:\\n {}\".format(np.corrcoef(((X_test.iloc[:, 1]*X_test.iloc[:, 0]).abs()**(1/2), y_test))))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "e5e7d85b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "training dataset:\n",
      " [[1.         0.02388513]\n",
      " [0.02388513 1.        ]]\n",
      "testing dataset:\n",
      " [[1.         0.00979593]\n",
      " [0.00979593 1.        ]]\n"
     ]
    }
   ],
   "source": [
    "# The IC of min(abs(alpha1), alpha4)\n",
    "print(\"training dataset:\\n {}\".format(np.corrcoef(np.minimum(X_train.iloc[:, 0].abs(), X_train.iloc[:, 3]), y_train)))\n",
    "print(\"testing dataset:\\n {}\".format(np.corrcoef(np.minimum(X_test.iloc[:, 0].abs(), X_test.iloc[:, 3]), y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "ae43db84",
   "metadata": {},
   "outputs": [],
   "source": [
    "alpha_matrix.loc[:, \"alpha6\"] = alpha_matrix.iloc[:, 1]**2\n",
    "alpha_matrix.loc[:, \"alpha7\"] = alpha_matrix.iloc[:, 1]*alpha_matrix.iloc[:, 0]*alpha_matrix.iloc[:, 3].abs()\n",
    "alpha_matrix.loc[:, \"alpha8\"] = np.minimum(np.minimum(alpha_matrix.iloc[:, 2], alpha_matrix.iloc[:, 3]), 0.173*alpha_matrix.iloc[:, 0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "cee0457c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>alpha1</th>\n",
       "      <th>alpha2</th>\n",
       "      <th>alpha3</th>\n",
       "      <th>alpha4</th>\n",
       "      <th>alpha5</th>\n",
       "      <th>return</th>\n",
       "      <th>alpha6</th>\n",
       "      <th>alpha7</th>\n",
       "      <th>alpha8</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Time</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2022-08-22 09:02:00</th>\n",
       "      <td>0.002226</td>\n",
       "      <td>0.000258</td>\n",
       "      <td>5.003409</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>811.301151</td>\n",
       "      <td>0.001487</td>\n",
       "      <td>6.634759e-08</td>\n",
       "      <td>4.585999e-06</td>\n",
       "      <td>0.000385</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-08-22 09:03:00</th>\n",
       "      <td>-0.001487</td>\n",
       "      <td>0.000442</td>\n",
       "      <td>4.995295</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>132.249017</td>\n",
       "      <td>0.002227</td>\n",
       "      <td>1.949258e-07</td>\n",
       "      <td>-1.313023e-07</td>\n",
       "      <td>-0.000257</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-08-22 09:04:00</th>\n",
       "      <td>-0.002227</td>\n",
       "      <td>0.001971</td>\n",
       "      <td>4.996566</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>323.119896</td>\n",
       "      <td>-0.002222</td>\n",
       "      <td>3.885850e-06</td>\n",
       "      <td>-2.195162e-06</td>\n",
       "      <td>-0.000385</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-08-22 09:05:00</th>\n",
       "      <td>0.002222</td>\n",
       "      <td>0.000389</td>\n",
       "      <td>5.003737</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>177.065677</td>\n",
       "      <td>0.000742</td>\n",
       "      <td>1.509589e-07</td>\n",
       "      <td>5.180457e-06</td>\n",
       "      <td>0.000384</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-08-22 09:06:00</th>\n",
       "      <td>-0.000742</td>\n",
       "      <td>-0.000857</td>\n",
       "      <td>4.997188</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>131.048609</td>\n",
       "      <td>-0.001484</td>\n",
       "      <td>7.338000e-07</td>\n",
       "      <td>1.589869e-07</td>\n",
       "      <td>-0.000128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-09-01 14:56:00</th>\n",
       "      <td>-0.000716</td>\n",
       "      <td>-0.000590</td>\n",
       "      <td>4.999999</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>409.146333</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.477241e-07</td>\n",
       "      <td>2.814037e-07</td>\n",
       "      <td>-0.000124</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-09-01 14:57:00</th>\n",
       "      <td>-0.000000</td>\n",
       "      <td>-0.001156</td>\n",
       "      <td>4.999572</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>949.338687</td>\n",
       "      <td>0.002146</td>\n",
       "      <td>1.335380e-06</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>-0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-09-01 14:58:00</th>\n",
       "      <td>-0.002146</td>\n",
       "      <td>0.000257</td>\n",
       "      <td>4.996268</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>87.531239</td>\n",
       "      <td>-0.000714</td>\n",
       "      <td>6.604965e-08</td>\n",
       "      <td>-9.191747e-08</td>\n",
       "      <td>-0.000371</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-09-01 14:59:00</th>\n",
       "      <td>0.000714</td>\n",
       "      <td>-0.003100</td>\n",
       "      <td>5.003117</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>256.091527</td>\n",
       "      <td>-0.001071</td>\n",
       "      <td>9.610507e-06</td>\n",
       "      <td>-6.638291e-06</td>\n",
       "      <td>0.000123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-09-01 15:00:00</th>\n",
       "      <td>0.001071</td>\n",
       "      <td>-0.000003</td>\n",
       "      <td>4.998769</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>10.718116</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.058189e-11</td>\n",
       "      <td>-2.323559e-09</td>\n",
       "      <td>0.000185</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1640 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                       alpha1    alpha2    alpha3    alpha4      alpha5  \\\n",
       "Time                                                                      \n",
       "2022-08-22 09:02:00  0.002226  0.000258  5.003409  8.000000  811.301151   \n",
       "2022-08-22 09:03:00 -0.001487  0.000442  4.995295  0.200000  132.249017   \n",
       "2022-08-22 09:04:00 -0.002227  0.001971  4.996566  0.500000  323.119896   \n",
       "2022-08-22 09:05:00  0.002222  0.000389  5.003737  6.000000  177.065677   \n",
       "2022-08-22 09:06:00 -0.000742 -0.000857  4.997188  0.250000  131.048609   \n",
       "...                       ...       ...       ...       ...         ...   \n",
       "2022-09-01 14:56:00 -0.000716 -0.000590  4.999999  0.666667  409.146333   \n",
       "2022-09-01 14:57:00 -0.000000 -0.001156  4.999572  1.000000  949.338687   \n",
       "2022-09-01 14:58:00 -0.002146  0.000257  4.996268  0.166667   87.531239   \n",
       "2022-09-01 14:59:00  0.000714 -0.003100  5.003117  3.000000  256.091527   \n",
       "2022-09-01 15:00:00  0.001071 -0.000003  4.998769  0.666667   10.718116   \n",
       "\n",
       "                       return        alpha6        alpha7    alpha8  \n",
       "Time                                                                 \n",
       "2022-08-22 09:02:00  0.001487  6.634759e-08  4.585999e-06  0.000385  \n",
       "2022-08-22 09:03:00  0.002227  1.949258e-07 -1.313023e-07 -0.000257  \n",
       "2022-08-22 09:04:00 -0.002222  3.885850e-06 -2.195162e-06 -0.000385  \n",
       "2022-08-22 09:05:00  0.000742  1.509589e-07  5.180457e-06  0.000384  \n",
       "2022-08-22 09:06:00 -0.001484  7.338000e-07  1.589869e-07 -0.000128  \n",
       "...                       ...           ...           ...       ...  \n",
       "2022-09-01 14:56:00  0.000000  3.477241e-07  2.814037e-07 -0.000124  \n",
       "2022-09-01 14:57:00  0.002146  1.335380e-06  0.000000e+00 -0.000000  \n",
       "2022-09-01 14:58:00 -0.000714  6.604965e-08 -9.191747e-08 -0.000371  \n",
       "2022-09-01 14:59:00 -0.001071  9.610507e-06 -6.638291e-06  0.000123  \n",
       "2022-09-01 15:00:00  0.000000  1.058189e-11 -2.323559e-09  0.000185  \n",
       "\n",
       "[1640 rows x 9 columns]"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "alpha_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "941741ae",
   "metadata": {},
   "outputs": [],
   "source": [
    "alpha_matrix_2 = alpha_matrix.drop(\"return\", axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "f47b37ff",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>alpha1</th>\n",
       "      <th>alpha2</th>\n",
       "      <th>alpha3</th>\n",
       "      <th>alpha4</th>\n",
       "      <th>alpha5</th>\n",
       "      <th>alpha6</th>\n",
       "      <th>alpha7</th>\n",
       "      <th>alpha8</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Time</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2022-08-22 09:02:00</th>\n",
       "      <td>0.002226</td>\n",
       "      <td>0.000258</td>\n",
       "      <td>5.003409</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>811.301151</td>\n",
       "      <td>6.634759e-08</td>\n",
       "      <td>4.585999e-06</td>\n",
       "      <td>0.000385</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-08-22 09:03:00</th>\n",
       "      <td>-0.001487</td>\n",
       "      <td>0.000442</td>\n",
       "      <td>4.995295</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>132.249017</td>\n",
       "      <td>1.949258e-07</td>\n",
       "      <td>-1.313023e-07</td>\n",
       "      <td>-0.000257</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-08-22 09:04:00</th>\n",
       "      <td>-0.002227</td>\n",
       "      <td>0.001971</td>\n",
       "      <td>4.996566</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>323.119896</td>\n",
       "      <td>3.885850e-06</td>\n",
       "      <td>-2.195162e-06</td>\n",
       "      <td>-0.000385</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-08-22 09:05:00</th>\n",
       "      <td>0.002222</td>\n",
       "      <td>0.000389</td>\n",
       "      <td>5.003737</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>177.065677</td>\n",
       "      <td>1.509589e-07</td>\n",
       "      <td>5.180457e-06</td>\n",
       "      <td>0.000384</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-08-22 09:06:00</th>\n",
       "      <td>-0.000742</td>\n",
       "      <td>-0.000857</td>\n",
       "      <td>4.997188</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>131.048609</td>\n",
       "      <td>7.338000e-07</td>\n",
       "      <td>1.589869e-07</td>\n",
       "      <td>-0.000128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-09-01 14:56:00</th>\n",
       "      <td>-0.000716</td>\n",
       "      <td>-0.000590</td>\n",
       "      <td>4.999999</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>409.146333</td>\n",
       "      <td>3.477241e-07</td>\n",
       "      <td>2.814037e-07</td>\n",
       "      <td>-0.000124</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-09-01 14:57:00</th>\n",
       "      <td>-0.000000</td>\n",
       "      <td>-0.001156</td>\n",
       "      <td>4.999572</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>949.338687</td>\n",
       "      <td>1.335380e-06</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>-0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-09-01 14:58:00</th>\n",
       "      <td>-0.002146</td>\n",
       "      <td>0.000257</td>\n",
       "      <td>4.996268</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>87.531239</td>\n",
       "      <td>6.604965e-08</td>\n",
       "      <td>-9.191747e-08</td>\n",
       "      <td>-0.000371</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-09-01 14:59:00</th>\n",
       "      <td>0.000714</td>\n",
       "      <td>-0.003100</td>\n",
       "      <td>5.003117</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>256.091527</td>\n",
       "      <td>9.610507e-06</td>\n",
       "      <td>-6.638291e-06</td>\n",
       "      <td>0.000123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-09-01 15:00:00</th>\n",
       "      <td>0.001071</td>\n",
       "      <td>-0.000003</td>\n",
       "      <td>4.998769</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>10.718116</td>\n",
       "      <td>1.058189e-11</td>\n",
       "      <td>-2.323559e-09</td>\n",
       "      <td>0.000185</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1640 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                       alpha1    alpha2    alpha3    alpha4      alpha5  \\\n",
       "Time                                                                      \n",
       "2022-08-22 09:02:00  0.002226  0.000258  5.003409  8.000000  811.301151   \n",
       "2022-08-22 09:03:00 -0.001487  0.000442  4.995295  0.200000  132.249017   \n",
       "2022-08-22 09:04:00 -0.002227  0.001971  4.996566  0.500000  323.119896   \n",
       "2022-08-22 09:05:00  0.002222  0.000389  5.003737  6.000000  177.065677   \n",
       "2022-08-22 09:06:00 -0.000742 -0.000857  4.997188  0.250000  131.048609   \n",
       "...                       ...       ...       ...       ...         ...   \n",
       "2022-09-01 14:56:00 -0.000716 -0.000590  4.999999  0.666667  409.146333   \n",
       "2022-09-01 14:57:00 -0.000000 -0.001156  4.999572  1.000000  949.338687   \n",
       "2022-09-01 14:58:00 -0.002146  0.000257  4.996268  0.166667   87.531239   \n",
       "2022-09-01 14:59:00  0.000714 -0.003100  5.003117  3.000000  256.091527   \n",
       "2022-09-01 15:00:00  0.001071 -0.000003  4.998769  0.666667   10.718116   \n",
       "\n",
       "                           alpha6        alpha7    alpha8  \n",
       "Time                                                       \n",
       "2022-08-22 09:02:00  6.634759e-08  4.585999e-06  0.000385  \n",
       "2022-08-22 09:03:00  1.949258e-07 -1.313023e-07 -0.000257  \n",
       "2022-08-22 09:04:00  3.885850e-06 -2.195162e-06 -0.000385  \n",
       "2022-08-22 09:05:00  1.509589e-07  5.180457e-06  0.000384  \n",
       "2022-08-22 09:06:00  7.338000e-07  1.589869e-07 -0.000128  \n",
       "...                           ...           ...       ...  \n",
       "2022-09-01 14:56:00  3.477241e-07  2.814037e-07 -0.000124  \n",
       "2022-09-01 14:57:00  1.335380e-06  0.000000e+00 -0.000000  \n",
       "2022-09-01 14:58:00  6.604965e-08 -9.191747e-08 -0.000371  \n",
       "2022-09-01 14:59:00  9.610507e-06 -6.638291e-06  0.000123  \n",
       "2022-09-01 15:00:00  1.058189e-11 -2.323559e-09  0.000185  \n",
       "\n",
       "[1640 rows x 8 columns]"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "alpha_matrix_2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "03e54639",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: >"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.heatmap(alpha_matrix_2.corr(),  annot=True, cmap=\"Blues\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "9758d217",
   "metadata": {},
   "outputs": [],
   "source": [
    "alpha_matrix_2 = alpha_matrix_2.drop(\"alpha8\", axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "b4058a10",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = alpha_matrix_2\n",
    "y = alpha_matrix.iloc[:, 5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "e0ade975",
   "metadata": {},
   "outputs": [],
   "source": [
    "X.to_csv(\"X.csv\")\n",
    "y.to_csv(\"y.csv\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.16"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}