Test_case_Bouc_Wen_Real_Data_low threshold.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pysindy as ps\n",
    "from scipy.integrate import odeint\n",
    "from math import pi\n",
    "from pysindy.differentiation import FiniteDifference\n",
    "fd = FiniteDifference(order=2, d=1)\n",
    "\n",
    "from sklearn import linear_model\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import mean_squared_error\n",
    "import glob\n",
    "from sklearn.metrics import mean_absolute_error\n",
    "import timeit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "start_train = timeit.timeit()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "       Unnamed: 0        0  -1.6174316        0.1\n",
      "0             NaN   0.0002   -1.724243    0.00000\n",
      "1             NaN   0.0004   -1.655579    0.00000\n",
      "2             NaN   0.0006   -1.754761    0.00000\n",
      "3             NaN   0.0008   -1.701355    0.00000\n",
      "4             NaN   0.0010   -1.701355    0.00000\n",
      "...           ...      ...         ...        ...\n",
      "49995         NaN   9.9992  -13.984680 -117.59304\n",
      "49996         NaN   9.9994  -13.916016 -117.70996\n",
      "49997         NaN   9.9996  -13.931274 -117.82671\n",
      "49998         NaN   9.9998  -13.839722 -117.94326\n",
      "49999         NaN  10.0000  -13.938904 -118.05964\n",
      "\n",
      "[50000 rows x 4 columns]\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "df = pd.read_csv (r'hysteresis_v_150_1hz.csv')\n",
    "print (df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(50000, 4)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(50000, 4)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = df.to_numpy()\n",
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_col = data[:,1:4]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_final = data_col[35000:45000,:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10000, 3)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_final.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10000,)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t = data_final[:,0]\n",
    "t.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10000,)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x = data_final[:,2]\n",
    "x.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10000, 1)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y = data_final[:,1]\n",
    "y = y.reshape(-1,1)\n",
    "y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "#plt.grid(linestyle='dotted')\n",
    "#t = np.linspace(0,20*pi,1000)\n",
    "#x = np.sin(t)\n",
    "#plt.plot(t, x)\n",
    "#plt.xlabel('Time', fontsize = 20)\n",
    "#plt.ylabel('Voltage', fontsize = 20)\n",
    "#plt.legend(['Actual function' , 'Learned function', 'Training Data'], prop={'size': 16})\n",
    "#plt.xticks(fontsize=20)\n",
    "#plt.yticks(fontsize=20)\n",
    "#plt.savefig(\"Results/test2_input.png\", dpi = 300, bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10000,)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "#plt.plot(t,x)\n",
    "#x_exact = 150*np.sin(2*pi*t-46.18)\n",
    "x_exact = 150*np.sin(6.28318530724*t-46.2178545458)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Relative Error Test in x:  6.977585424072029e-13 %\n"
     ]
    }
   ],
   "source": [
    "y = y.reshape(-1,1)\n",
    "# Compute the relative L2 error norm (generalization error)\n",
    "relative_error_test = np.mean((x - x_exact)**2)/np.mean(x_exact**2)\n",
    "print(\"Relative Error Test in x: \", relative_error_test*100, \"%\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f476838be10>]"
      ]
     },
     "execution_count": 16,
     "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(t,x)\n",
    "plt.plot(t,x_exact)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "dx = fd._differentiate(x_exact, t)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "#plt.plot(t,dx)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "mdx = np.abs(dx)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "#plt.plot(t, mdx)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "#def model(y, t):\n",
    " #   dydt = np.cos(t) - 0.5*np.abs(np.cos(t))*y - 2*np.cos(t)*np.abs(y)\n",
    "  #  return dydt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "#y0 = 0\n",
    "#y = odeint(model, y0, t)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10000, 1)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "#plt.plot(t,y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "#plt.plot(x, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "dy = fd._differentiate(y, t)\n",
    "#plt.plot(t, dy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "my = np.abs(y)\n",
    "my = my.reshape(-1,)\n",
    "mdy = np.abs(dy)\n",
    "mdy = mdy.reshape(-1,)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = y.reshape(-1,)\n",
    "dy = dy.reshape(-1,)\n",
    "t1 = dx\n",
    "t2 = mdx*y\n",
    "t3 = dx*my\n",
    "#terms = t1-0.5*t2-2*t3\n",
    "#plt.plot(t,terms)\n",
    "#plt.plot(t, dy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = np.stack((y, x, dx, mdx, my), axis=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(x0)' = -0.176 1 + -2.383 x0 + 0.580 x1 + 0.119 x2 + -0.076 x0 x4\n",
      "(x1)' = 1.000 x2\n",
      "(x2)' = -39.478 x1\n",
      "(x3)' = -111524157.422 1 + 2053.343 x0 + -255.695 x1 + -5.835 x2 + 68.333 x3 + 3043.670 x4 + -5893.724 x0^2 + 2956.032 x0 x1 + -65.388 x0 x2 + -4.187 x0 x3 + 90.798 x0 x4 + 4768.293 x1^2 + 8.205 x1 x2 + 0.539 x1 x3 + -12.532 x1 x4 + 29041.155 x2^2 + 0.469 x2 x4 + -28915.766 x3^2 + -2.766 x3 x4 + -5859.730 x4^2\n",
      "(x4)' = 143.244 x0 + -18.777 x1 + 0.453 x2 + -1.661 x3 + -89.567 x4 + -123.062 x0^2 + 62.998 x0 x1 + -1.331 x0 x2 + -8.817 x0 x4 + -3.952 x1^2 + 0.171 x1 x2 + 1.149 x1 x4 + -0.029 x2 x4 + 0.075 x3 x4 + -123.062 x4^2\n"
     ]
    }
   ],
   "source": [
    "opt = ps.STLSQ(threshold=0.01)\n",
    "model = ps.SINDy(optimizer=opt)\n",
    "model.fit(X,t)\n",
    "model.print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.006781279000279028\n"
     ]
    }
   ],
   "source": [
    "end_train = timeit.timeit()\n",
    "print(end_train - start_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "def test_model(y, t):\n",
    "    x0 = y\n",
    "    x1 = 150*np.sin(6.28318530724*t-46.2178545458)\n",
    "    x2 = 942.477796086*np.cos(6.28318530724*t-46.2178545458)\n",
    "    x3 = np.abs(x2)\n",
    "    x4 = np.abs(y)\n",
    "    dydt = -0.176 - 2.388*x0 + 0.581*x1 + 0.119*x2 - 0.076*x0*x4\n",
    "    return dydt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "start_test = timeit.timeit()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "ytest_0 = -14.030457\n",
    "y_test = odeint(test_model, ytest_0, t)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-0.0002790319995256141\n"
     ]
    }
   ],
   "source": [
    "end_test = timeit.timeit()\n",
    "print(end_test - start_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f4767b03b10>]"
      ]
     },
     "execution_count": 36,
     "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(t, y_test)\n",
    "plt.plot(t, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.grid(linestyle='dotted')\n",
    "plt.plot(x, y, 'r')\n",
    "plt.plot(x, y_test, linewidth=2, linestyle=':')\n",
    "plt.xlabel('Voltage [V]', fontsize = 20)\n",
    "plt.ylabel('Displacement [$\\mu$m]', fontsize = 20)\n",
    "plt.legend(['Experimental data' , 'Learned relation'], loc='upper left', prop={'size': 12})\n",
    "plt.xticks(fontsize=20, rotation='45')\n",
    "plt.yticks(fontsize=20)\n",
    "plt.savefig(\"Results/pic3.png\", dpi=1000,bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import r2_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "r2 = r2_score(y, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9999562944015838"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Relative Error Test:  0.004370415810827789 %\n"
     ]
    }
   ],
   "source": [
    "y = y.reshape(-1,1)\n",
    "# Compute the relative L2 error norm (generalization error)\n",
    "relative_error_test = np.mean((y_test - y)**2)/np.mean(y**2)\n",
    "print(\"Relative Error Test: \", relative_error_test*100, \"%\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f47676e5410>]"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD4CAYAAADiry33AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAA2PElEQVR4nO2deXgUVdbG32NkERBZVZQlKIjiyhhh3EDHDVccdWZwRgdHHUTF0fEbHVxARR1RHB0XHFfGBRUQFFE22UR2EiCQsISEBEIChJCd7Emf74+uDpVOL9Xdtff5PU+eVN+6VXX6dtVb955777nEzBAEQRDcyzFWGyAIgiAYiwi9IAiCyxGhFwRBcDki9IIgCC5HhF4QBMHlHGu1Af5069aNExMTrTZDEATBUWzcuPEwM3cPtM92Qp+YmIiUlBSrzRAEQXAURLQ32D5x3QiCILgcEXpBEASXI0IvCILgckToBUEQXI4IvSAIgssRoRcEQXA5IvSCIAguR4ReEFwIM2PWxjzU1DdabYpgA0ToBcGF/JxRiH98swWvLcyw2hTBBojQC4ILKa+pBwAUHqm12BLBDojQC4IguBwRekEQBJcjQi8ILkbWhBYAEXpBcCVEZLUJgo0QoRcEQXA5IvSC4ELEZSOoEaEXBBcjLhwBEKEXBFcjNXsBEKEXBFciNXlBjSahJ6LhRJRBRFlENC7A/jFElEZEqUS0iogGKumJRFStpKcS0ft6fwFBEAQhNGEXByeiBABTAFwDIA9AMhHNZebtqmxfMfP7Sv5bALwBYLiybzczX6Cr1YIgCIJmtNToBwPIYuZsZq4DMB3ACHUGZi5XfWwPQByDgmAh4psX1GgR+lMB7FN9zlPSmkFEDxPRbgCvAfibaldfItpMRCuI6PJAFyCi0USUQkQphYWFEZgvCIIghEO3zlhmnsLMpwP4J4BnleQDAHoz8yAAjwP4iog6Bjj2Q2ZOYuak7t2762WSIMQt0hkrqNEi9PkAeqk+91TSgjEdwK0AwMy1zFykbG8EsBvAGVFZKgiCIESFFqFPBtCfiPoSUWsAIwHMVWcgov6qjzcCyFTSuyuduSCi0wD0B5Cth+GCIAiCNsKOumHmBiIaC2ARgAQAU5l5GxFNBJDCzHMBjCWiqwHUAygBMEo5fCiAiURUD8ADYAwzFxvxRQRBaIl0yQqABqEHAGaeD2C+X9oE1fajQY6bDWB2LAYKghA54qEX1MjMWEFwIVKTF9SI0AuCi5GavQCI0AuC7tQ3erAhR7qiBPsgQi8IOvP6Txn4/QdrsWVfqdWmiAtHACBCLwi6s+tgBQCgqLLWMhvEZSOoEaEXBIOQcDOCXRChFwSdkfADgt0QoRcEg5AavWAXROgFQWekPi/YDRF6QTAIqdALdkGEXhB0Rlz05pJfWo2SyjqrzbA1IvSCYBCyypM5XDppGS6etNRqM2yNCL0g6I6NqvRx8q6pqfdYbYKtEaEXBIOwUmPFfSSoEaHXSEOjB3UNUmsQnIF4jQQ1IvQaufW91Tjj2QVWmyE4CFtUqm1hhGA1mhYeEYD0/HKrTRBsDjOjVtXqk0q1YBc01eiJaDgRZRBRFhGNC7B/DBGlEVEqEa0iooGqfU8px2UQ0XV6Gi8IduLdZVk4c/xClNfUW23KUeRtI0CD0CuLe08BcD2AgQDuVAu5wlfMfC4zXwDgNQBvKMcOhHcx8bMBDAfwnm+xcEFwG99tzgcAFB3xRq2cmbwPT3271RJbpDNWUKOlRj8YQBYzZzNzHYDpAEaoMzCz2q/RHkfrESMATGfmWmbOAZClnE8QXM/SnYfw9YZ9VpshCJp89KcCUN+teQCG+GcioocBPA6gNYDfqI5d53fsqQGOHQ1gNAD07t1bi92CYD+kFi3YFN1G3TDzFGY+HcA/ATwb4bEfMnMSMyd1795dL5MEwRKK/Kbjy7KC7qXRw3hj8S6UVdmoXyYAWoQ+H0Av1eeeSlowpgO4NcpjBcHxlPo99L//YK1FlghGs2RHAd5emokXfthmtSkh0SL0yQD6E1FfImoNb+fqXHUGIuqv+ngjgExley6AkUTUhoj6AugPYEPsZguC/RDPTfzR0OjtjqxpaLTYktCE9dEzcwMRjQWwCEACgKnMvI2IJgJIYea5AMYS0dUA6gGUABilHLuNiGYC2A6gAcDDzGzvEhEEh/LLrkKcfUpHdO3QxmpT4gZ2yPhVTROmmHk+gPl+aRNU24+GOPZlAC9Ha6AgOAUrlxBsaPTgz1M3YMBJx2PR34daZke8QjZvz0kIBEFwAR6lYrm78EizdKfUOAVjEaEXBBfgE3SZKGUuTgkeJ0IvCC7C34Vgd5eCa7B5MYvQm8TC9INIHDcPOYcrAQD9n5mP1xdlWGyV4BacUrN0G04pdhF6k/hh634AQHp+GQCgvpHx7vIsK00SdMYWlTo/I8RHbw62+O1DIEIvCC5EXDaCGhF6QdAJKzpCx3yxEdPW7T1qg/kmxDVOWQBehF4QHMzCbQfx7Jx08dFbjJVzKLQgQq+B7ftldSkhPFa6S/yHV074Pt20a+8uPIIJ36fD45G3jV0RodfA6C9SdDuXPAqCGfhH0DSS0Z+n4PO1e1tM1oon6hoaMeDZBfhRGXRhN0ToTcLeDTtBDzIKKqw2wVJs7r0wlILyWtQ2eDBpwU6rTQmICL0gCILLEaHXgHR0CYIQCru3ZkToTULeFYIZWFkpifTaOw+WI3HcPGzNKzXEHjNwSiVQhN5kbP7iF1yGGUIU7dDCpTsOAQAWpB/U0xxLCFQC6flleH7uNluMtRehNxnrf3JB0Bc7CJkdufPDdfh0zR5U1DZYbYoIvVlITV4wA3/JNdN3HO21nPyecEosIU1CT0TDiSiDiLKIaFyA/Y8T0XYi2kpES4moj2pfIxGlKn9z/Y8VBEE/6ho8VpugGbt3YEaDXb9T2KUEiSgBwBQA1wDIA5BMRHOZebsq22YAScxcRUQPAngNwB+UfdXMfIG+ZguCEAy1K8XM2rKTa+bR4vvOvn4Ku5aBlhr9YABZzJzNzHUApgMYoc7AzMuZuUr5uA5AT33NFAQhFGqB+WRVjq7nrmvwYML36SgOMts21jgvTnF/hMKmFfkmtAj9qQD2qT7nKWnBuA/AAtXntkSUQkTriOjWQAcQ0WglT0phYaEGkwR/DpbVoNEmsUbmbM7Hykx3/o5frNuLhTYfJfLSvB26nm9+2gF8vnYvXpq3PXzmCJBQyuaha2csEd0FIAnAZFVyH2ZOAvBHAP8hotP9j2PmD5k5iZmTunfvrqdJtqOmrhFpeWW6nvNAWTV+/cpSvLl4l67njZbHZqTi7k82WG2GIYyfk44x0zY2S1uecQjb9uv7m2rhC1V4YiPxKM2FYG6JeB5145SvHtZHDyAfQC/V555KWjOI6GoAzwAYxsy1vnRmzlf+ZxPRzwAGAdgdg82OxNe8fXL2Vt3PfajcW9y/ZBbiH9cN0P38QnBGvLsKW3R+cWtl/BzzIlQG46NfsrG7sDK2kzhELKPFDi8DLTX6ZAD9iagvEbUGMBJAs9EzRDQIwAcAbmHmQ6r0zkTURtnuBuBSAPq2/wRT2VdchflpB6w2wzZYJfJaWZB+EP+cFVvlIpRQffBL3NXZmjhQVo1GX2tHSWvWXWEjz1RYoWfmBgBjASwCsAPATGbeRkQTiegWJdtkAB0AfOM3jPIsAClEtAXAcgCT/EbrOAI9mqZuad7e8NZKPPTlJqvNECJgRsq+8JmiRJfb2kaCqJXCilpc/MqypmiVVXWNFlsUGi2uGzDzfADz/dImqLavDnLcGgDnxmKgYC/sMMtPaMk3Bop5qEE1sej81xtyAQDl1c67p0qqvCOQfCORdhyw9+JEMjNWA+6oiwtu5vkfjGsoG9UYzS32jsg+WFat63lLKusMffGFwq4NdxF6DRwoq7HahKjJLKhAblFV2HzPz92G05+eHzaf4KXRw/jTx+s053eD6y5QxV79vcqq680zJgSPzkjFE7O2IuuQxSte2egnF6E3CasWD77mzV8wdPLysPk+XbPHNuPwnUBhRS1WZxVpzj9tfa6B1phDoLujpOqouL+/IrqOWb3vukPl3oqZXcJB2CEsggi9IJhATqxDEB2BDRTNYgKJuh0acyL0LsCK++jaN1egtsHeIw2MZE+Rs4R7c24JEsfNQ9ah6Ne1FRk/iibxtlGBidC7CDPvq10FR7Bxb4mJV7QXIz/U7p+3Az9s8c59+DnDuNAUDZ7oXCV2qPG6HRF6QRB04eeMQqTn23sCWbwiQi8Igm4sSI981rRenZUllXWuGN1kBCL0giYOH6nFOc8tstoMweZMWW5dSIRBLy7G+yuyTbmWHUbSRIIIvYswsi6zOuswjsisWMcSS8x3o+vIelbCF22zdwhpq4grod9XXIWaevNHiqTnl+GHLfsNO79VlYuGRmkmOw2r5nOEQs+JTan7Spu2bfhVLSNuhL6+0YPLX1uOv3292fRr3/TOKtOvqTdFR1quLjRT4zTzgvIanDl+QbOY7asyDyOvJPyMXUFfovFhG62X+aX6hkAwg4iK0Qb1obgRet+szxW73LnykdFM/LFlLJWaem3D6ZbtPISaeg++WHt0oYy7PlmP3/x7hW72CaHRuppTo4dbhMywgU45Ejs1KOJG6AX7YZcp6vHExr0lIZdCfHPxLgydvDxwfCQ7KZcGZADOUUTobYDHw5i2bm/UM03lfhbC4euMXZB+sMVSiGrW7D4MACg8EiCQn9xoEWGn4oo7oY+18I2I0Pf9lnw8Oycd7y7Liuk8DqtwCTFg5ggou95Xm3JLsDWvNOh+23TG2sAOTQuPCEfxGBDhsaLG+9CWVsX2EtmSV4a3lmTi0av762GWYBAeD6ORGa0Soq9neUz0SxhxpYbG2N12t723JmD6zoPRx/PRSrBwD1r7QsxG051GRMOJKIOIsohoXID9jxPRdiLaSkRLiaiPat8oIspU/kbpaXwsNDR68PCXmyJeGcaIm17PZ/bNJbv0O5lgCI9M34z+zyyI6Rxa1hhQ801KnqZ8IW9FjRq2qyC80NY4vH8mWOs7lvkKRhJW6IkoAcAUANcDGAjgTiIa6JdtM4AkZj4PwCwArynHdgHwHIAhAAYDeI6IOutnfhQwcKiiBt9uyse8tAP4+4xUS81RY5umpkE4vXMslmGAU1fn4OOV3lmb87bGvrj63AjnZUTu6gkUb1fbkde++UuE13IehypqtWe2wX2vpUY/GEAWM2czcx2A6QBGqDMw83Jm9lUx1gHoqWxfB2AxMxczcwmAxQCG62N69Fz17xV4cvbWqI6VWBqR45b31ycrc2I6/qV5O3SyBKioscdqTtES7Dlau7tI8/wMu2On+16L0J8KQF3yeUpaMO4D4GuXRnqs8dBRn7hdsNvLw+Nh/Pfn3Y4XEzfz9QYLxNAE5brzo3V4clZ0lTAhOLqOuiGiuwAkAZgc4XGjiSiFiFIKC+09oclISbZLDWDZzkN4deFOjP3K/FnEWtldeAR7Dodf/GPj3mKJfyLojt0qZ+HQIvT5AHqpPvdU0ppBRFcDeAbALcxcG8mxzPwhMycxc1L37t212h4dNvt9EsfNw49R+GxLKuvwfWqLotSFWqWjLNwsYiv7FK769wpc8frPYfPd/t+1eOCL4OPGheY4TL90p6a+0ZXB+7QIfTKA/kTUl4haAxgJYK46AxENAvABvCJ/SLVrEYBriaiz0gl7rZJmOnqJkhEPQkoUKzWN/XoTHp2ein3F1sWL0VoWca4djsTsl7gVwQYDccNbK2MKx11aaU93Z9hx9MzcQERj4RXoBABTmXkbEU0EkMLMc+F11XQA8I0SHS+XmW9h5mIiehHelwUATGTmYkO+iQ2ZkZyLp75N05w/ksiCB0qVle51GI/c0o7mn3OCukgik3C3jyoSoqe+0YO2rRKsNgPZGtyBoaiwaWtA04QpZp4PYL5f2gTV9tUhjp0KYGq0BjqZl+ftgAHzq0znSg0uEi0Y5RZYnnEIU1fl4PN7B9syDK/TcZo/2oeRt0Ik95kdxtbHYQgE6wvd7qhv4dkbtU200Xo+QP8wEg98vhErMw8b0rpxK8H6d3xPh9GvS/+n0Kin0uNhvPTj9ognmIVjowZ3q50qHXEn9OEoqWwZd11NJC8KO/3Q0fJ/32zR/ZwHymKLP36oIkDALYeQOG6e1SYAAL5clxt4h1J7V9+7we7jV+brNy/gvOd/wgcr9F+GMKOgAh+vymkK5FZcWYfJi3Y2hS03Eju1hOJO6EPFoliYfhCDXlyM9dlFJloUHYYO89T4flqy41D4TH6szy7CO0ujD97GzHjar9/DrFaaVe/tHQfKkThuHjbnRt5prwfBBOuDX/Rdn/XrDUFePjHgM90XG2j8nHRMWb4bvwQYUVap8q8Pm7xcNxt8mvPxymy8vTRTt/NGQtwJfShR2JDj7SdOyy8LmkeLpsxIzsUlryyN1DTXUeg3TdzDjD98uA7z0qIPATBtfW7QF4zRAaWsqqD9nOEVpYV6zgeIoqj0LF+rytI3uidQjb5RZdTeGF09T3yzBeV+EzNfmrcDbyy2JhZV3Am9Gfxzdhr2l9UYEtLYR6SPnNZFPmrqG/HT9oLIDQrA6C9Smn3W+mx/n5qPR6cHnqzlexk3O699WsjGouP3jEayZ2/Kw8xkd4QnCESCjk22b3To29ITEfoIMVJTPl2zJ2wfQbR2TPg+XVO+f83fgW83aZ+IdfrT84O+RPxr9Fr4bM0ePDo9Fd+nRr6YeqjndGbyPlP8sk4hWFmFK6FoY0Q1nZ8Zry7cif0mrBM7eWEG0vcHbp0zvLXunzOOtg5d0KUWlLgX+khrg4UVtdi417ipAL9kGhMCYp3GfodIJ2A1elhzy0VLWT83d1tE1we0vfSenL0VU5bHtrCLVVghQEZdcseBCvz35914+MtNBl3hKEt3HmoRN0ddlt9szMM9/0tGPBB3Qh9rM3/ElNW4/b9rUVPfqPuQLcDrOtHSW++7X5kZB8v0G4WyvzTyc4XrDPU9XEYvlhFOnHYXHjH0+kbjhvaI7x4INBQ2klFqGSYsLqIXdhjSHTdCH0pj5mzOx380Ltjha/7/38wtGDp5ue5Tt/85Ow1faRh94Ps67y7LCrkGaKRkaFg0Qiv+Zf7d5shi82zf33JRmFiGrDm1ZW6E3bF2rJqxsHtmQUVQd1vWodhe2n/9PCV8piBEWrFas9v6UXxxI/Q+AlUaHpuRiv8siWzYk8+312CA33dhuvbRFSszDwfd1+/po5OZDR3TH6YIotXmao0vUX/xZ2ZMWZ6FyYt2Rndhm6LnuOygPnq/SxSU1+BfAcbLj5+jrc8nWnYeLMc1b/6Cd5Y1fy6nrduLi15eEoM7K/bnIDdC9+ZDJripwhE3Qp9d6I1hUd8Y/GFxUtPed7uGEnD1S6i2vhEDJyzUZXWjSPj9+2uxbGfk4+1jYX1OMSYvysCU5c0n4IQqq/zS6qg6j83AiHd0uHP69j8xaytKAqxlvDaCuSaJ4+bhka8jC3nte14355Y2S392TnqMv5P1bhQriBuhX7Ij/JBBX1heLbVfI28XLR2nvutrXV96f1kNquoa8coC/WYz+tsyb+uBZqsDVdU1YMOe4qiHa2oVOP/foj6KUAiXTlqGi15eEnT/R79kY+rq2FaYihVfbbug3LyZwXos4g0AP0S49GG4WnCk7z4zh+DaccEeTUHNhOAY4RAJ1erw5xgdqnt6DTt8+Ctrm6jhvkUsJfWyjtP9I8Xfnz7kX7FPxovVR29EB2MkAmmE60Zz2O0wGc974adIDDKFuBF6ozzUVjUEfd9HD6F/Y3FGTMcHu+8DNfkjQes3I8Rrgzx6/G8bZsa0dXsNneTX/Hot0w4f0TaHJBrMHKJqxwl88SP0Ov/QVo/g8N1Lx+jgfPP3g9oFrR3Imp8rq380G5O6rxTjvz86h8HocBLlMb5Qxkwzv/V4oKwaJ3dsa/p19SBufPTRjjoJ5i+3y0s70gfSzbP/BO34Pw+1fsMlQ90nJZV1MddatS7QofdzFu39v+NAOS5+ZRk+W7NHV3vMIm6EPhLUPriPV4aO0BfovjG1mWibV47+aC1Go5vKdhmNw9BvyT3NbrEAGWsbPMgrMT6EgZ3wLUS/Ljv2WfHVdeYvmxj3Qh9KKC+dtCxsKF6rZXZfcXQPXKOHUVrl9YnG+mIy6mVz1yfrcfrT88Nn1Ei07oidB1tO3DIT9e8zw6CgYoFKZnNuCVZntWzROrFyYSe/+fVv/WL6NTUJPRENJ6IMIsoionEB9g8lok1E1EBEd/jtaySiVOVvrv+xevN9aj6+XL83pnP4mrX5JgReCob/SBhmxhuLd2mKRZM4bh7+EWTBEN+L4ZX5O3DBxMW6DAUz6iGqqGloWQ4ajgsm6FpfaJtyS5qVs9H+aq0wRzd0NBBaJv0szzAm7lIk2KPkj953ucVVKNIYeDAYewwInRKOsEJPRAkApgC4HsBAAHcS0UC/bLkA7gHwVYBTVDPzBcrfLTHaG5ZHp6fime9aztozyp1i1I14xC+W9Z6iKry9NFPz1O1ZIcKk1tQ34sv13jALf/xofYtr2Rkt0T1jrXHe9t4aXP6afgtP2BH/Bd8jWgNVQ/E+OStwRcPpfUTbD5TbYqZrpGip0Q8GkMXM2cxcB2A6gBHqDMy8h5m3ArB00c4/T91g5eWb0KOG++WG5q0SXzConToEc7rj/TVN4QXS8suwJS/EQisaMKtVXFFTb0ncELuIE4MNC2Wh92lnpgSuaET6bOh1bx3R2PnrVrQI/akA1I7BPCVNK22JKIWI1hHRrYEyENFoJU9KYWH0zcVAy4OFozhIDfGQibMPA/HawtjGtociPd9an7OaSMIib9X4QloSZCauT8uKjtRikcbVmqauyjEldnoo1OJup3VInYQWN6ybS9aMcfR9mDmfiE4DsIyI0pi5WRASZv4QwIcAkJSUpEt5L91RgKvOOqnpczA/a7BJGje/u0oPM+KCrENHUF0XXY2ppKoOvbq005RXa6z6z9YG7qPx6eVfPk3G1rwybH3+WnRs2yroeQ5V1GDij9s1XdPpRFKhj/YBTQ+1RKdJ2KRxZjpaavT5AHqpPvdU0jTBzPnK/2wAPwMYFIF9UfO/1XuafY60aVpQHnpInRFRK53KqKkbcPUbxo8kCFSbVc/kDNY687FAiQrq64hsDBNqwmOpI7IlzAZHIVWRX1rdNKQwFpL3HB2OeNM7q2zjBos3tAh9MoD+RNSXiFoDGAlA0+gZIupMRG2U7W4ALgVgSRVJ74UKzIjHHQxmbtGZFq88/V1a03awkUY+KhzU6azGp40NHg9eNKiF4S/AD3yxEXMjDETmT15JFX73/tpmaeJ5soawQs/MDQDGAlgEYAeAmcy8jYgmEtEtAEBEFxFRHoDfAfiAiHxt7LMApBDRFgDLAUxiZkuEPtJFL7Syp8h8wc2McdEFpxJII9SjcLQKeSixybZxqOol280N9xwrgTpAjV5lLBzx2qLQ5KNn5vkA5vulTVBtJ8Pr0vE/bg2Ac2O00XQiuRduemcVcl650RA7Dh+pRbcObVqkN0QQ3TKe0PIQq1tCgfLf/t81EZ3PTMptEv62OIbgY++v2B0+UxD0CJkc6j0Ta/wdOxP3M2NjhRlYszv4Kk+xEGzBDifOTAxGyp4SzXkD6W6kFcTXfzo6mik5wLXt4t4pKK9B4rh5mKmaCesfj0Yvxs3eisURtBZWZkU/Mi7ncPSThV5daOyKYY/PTDX0/FYiQq8Df/xovTEndo+eB2XSgvAP72hlkpje8eb/+nkKDlUEH0b7twhXRdKLBWkHmoYKf7s5z/CWxfTkfTHVtCMhmuGhHg/j8Zmp+Ghl7Au/hCrLAyHWgnV634JrhX7jXu01RX8c/ps6ijoNzXGtK1RFMyLlqtdX+J3j6Pb6nNgDWEXDg19uwhOztrZIt5knyTS+WLcX324ypo8NQNOcCj3WdrArrhX66vpGy4NRxUowF41eK0LZhYMhalI+ymvqm9YRDUY0j6nWcLnCUbTG/gmUL5qZ3XsNjg3zpPJSDTWSzenvANcKPRB+XHUw3l6aGT6TH7kmBip6a0nk9tmZvJLwZfeoBjeK5jVmQ7TD7RLAzAfz0ReY017vevQlMbOufVKBft+y6nqMm92yBdXcDt1MsARXC320RLOc2tDJ+gfBCnZzbd5Xqvu17I6WeQOaRJpDC6YWV5LZVOkUg14vzFpuEND2u2vF42EsDBL6YrpB4Z+DsX1/uamhNVwt9PtLvSMXFqYfMOV6egz/UqMWpLWqYF4S7yQ27DKyRiu+uEd2cdlp7bjVo3WUV1KNQzot/GKnl/gNb6/EJZOWmXY9Vwu9L7bGrI3GdeSoMfIxfHZOy9DL8YSmOPRadMVenhlBA/O2mlNRC4XThzS7WuitwIzatrNvOePQJPQ2LLxGD0srzeZoWajFzojQ68x7Pxs/Hlk0ITBVdY2Yn2Z97S9STn96vuYFZQRrcPozFydCb86vRADm6BhTx+k3l1Z8NfFQk5eCoW5Sb80rC7/6j01dN+HWJhZipyCGNSYmLzJufQgzcLXQWzH21YxrOn1Mrz/fbc5HZW0DDlfEthanJhz28jR60tbGvdZMCrMCowIb6sWbi3fhzcW7DDm3q4Xe7BoxEek6uy5YB5DbavrT1uVi/PfpITu8gk2aWZcdP0JlBE6vqRrJ8p2HTB1K+tbSTLwVxRweLZixwpRl+EKiOlUYnWp3NBwKs9CLbgR4DyeOm2fOtTVgZqfsD1v2G/qibLDZyi2RFu1fPk0GAHwyKskAa8zF1TX6z5Ul5ZZlmOf/NGMFIDNrGW7DygVjAlGjmgxVY/LEqEcMDtr2RZAlHa0i2kfzvs+c31HuaqH3YVYl6aKXl6CsSj8/cxxV6EEELNYYvCwWKmwS092HWgzPHL/Q8cP41EiF5ChWtxrjQujNoriyDvs1BOgSApNXYt6UcDvAzC3WHt5t4xWuIsW3Rq8QnNveWx3VaLNI0ST0RDSciDKIKIuIxgXYP5SINhFRAxHd4bdvFBFlKn+j9DI8Hhg/Jx0em0x7NwMzlpmzU7/HptxStEpo7k+491PnuwkE7WzKLcVX63Px49bY1ucNR1ihJ6IEAFMAXA9gIIA7iWigX7ZcAPcA+Mrv2C4AngMwBMBgAM8RUefYzY4f7BSfw3BMEGGrYswHoqHR4+oY6HbDzAizkeDxMKauin1RlVBoqdEPBpDFzNnMXAdgOoAR6gzMvIeZtwLwV6XrACxm5mJmLgGwGMBwHeyOG84cv9BqE0zDRpVtU2C4b06EndmqxL6yG28vy8Km3FJDr6FF6E8FoI7hmaekaUHTsUQ0mohSiCilsDD69SgFZ2P3CS16w2wvV5LbyTrknv6PSLFFZywzf8jMScyc1L17d6vNEQRTYDAm/rjdajOEOECL0OcD6KX63FNJ00IsxwqCu5HavGASWoQ+GUB/IupLRK0BjAQwV+P5FwG4log6K52w1yppgtAMMyaa2Q3RecEswgo9MzcAGAuvQO8AMJOZtxHRRCK6BQCI6CIiygPwOwAfENE25dhiAC/C+7JIBjBRSROEZjTabLq8Gdht8pbgXjTFumHm+QDm+6VNUG0nw+uWCXTsVABTY7BRiANWZxWFz+QyxkwLE1JZEHTCFp2xgiAIgnGI0DuAtDx7jv8VBMEZiNA7gJvfXWW1CYIgOBgRekEQBJfjGqE3c8EGQRAEJ+EaoRcEQRAC4xqhlwq9IAhCYFwj9IIgCEJgXCP0UqEX4pUHhp1mtQmCzXGN0AtCvHLWyR2tNkGwOa4ReieNuhk9VGpggn4MPEWEXgiNa4TeSTxx3QCrTRBcxBknHW+1CYLNcY3QO6c+D7RKcE2xC3707dbe0uvfc0mipdcX7IkojiDoiNVR9Y89xmoLBDviGqF3kIteEAxD/Rh8cPeFltkh2AvXCL0gCM0Z1LuT1SYINsE1Qs+O8tILrsVGnhOykzGCpWgSeiIaTkQZRJRFROMC7G9DRDOU/euJKFFJTySiaiJKVf7e19l+QRBUEIBuHVpbbYZgM8IKPRElAJgC4HoAAwHcSUQD/bLdB6CEmfsBeBPAq6p9u5n5AuVvjE52C4IjWPfUVaZf84v7huCeSxJF8IUmtNToBwPIYuZsZq4DMB3ACL88IwB8pmzPAnAVEZnabpTOWMGOnHxCW9OveVaPjnj+lrNh8iMo2BgtQn8qgH2qz3lKWsA8zNwAoAxAV2VfXyLaTEQriOjyGO0VBEEQIuRYg89/AEBvZi4iogsBzCGis5m5XJ2JiEYDGA0AvXv3NtgkQRCE+EJLjT4fQC/V555KWsA8RHQsgBMAFDFzLTMXAQAzbwSwG8AZ/hdg5g+ZOYmZk7p37x75txCEOGXXS9dbbYLgALQIfTKA/kTUl4haAxgJYK5fnrkARinbdwBYxsxMRN2VzlwQ0WkA+gPI1sf05jjNRz/87JOtNkEwALO94q2Pdc0IacFAwt4lis99LIBFAHYAmMnM24hoIhHdomT7BEBXIsoC8DgA3xDMoQC2ElEqvJ20Y5i5WOfv4EiSEjtbbYIgCHGCJh89M88HMN8vbYJquwbA7wIcNxvA7Bht1IRMmDKeLu1bo7iyzmozhAA8dMXpeO/n3VabIdgUafcJmpn3t8usNsH2XJTYpUWaGaMcO7eTMfNCcFwj9E7z0V/Qq5PVJkRMjxOOs9oE29G5Xatmn18YcTYWPTa0WdoxcTCevX3rBKtNEELgGqF3GkkBan6C8xjUu3lfS5tjEzDg5OYLgZze3bwY9Q6r7wgm4RqhlxtcsCtf/fXXePCK0602Q4hjXCP0TkTWiHA+Wn7Cbh3a4AGT1gmWW0oIhGuE3kmLg8c7Pz4Sf526ndq1xr9+e67VZhjGQ1f2s9oEIQSuEXon4qSgU//7y0W6nevUTuZ06nZpb/xIlEh+wtt+5R8iyj08fGU/7Jl0Y8THOXFQghNxjdA7sT7vHJkHrhxwom7nctD7TTCYVglyM6gZ2KOjIec1OqiZEIJO7Vrh8JH4m4AkXjb3kDrhGlvdw2uf+g0A4OJXlllsib1wT43egeIx+8FLrDZBiJJ7L+2LDU8HX1RkwaOXY/Hfm4+nd2NLplO71uh3Yoeoj9d7ucMeJxxn+nyP7se3Qce29q4zu0bonUifruaNr7YTx5v4UPTqYsxD/9QNZ+LEjm0RzAF3Vo+O6H9S8/H0soZrS9wQuuTOwb11628zypXlHqF3yP0y4oJTrDYhJpY8PizmcxybYN5td++lfXU/5zEEtIriO1hdo3/4SneO5e/dpV2LtBvONS867N+v7q/bucYMM+Y3co/QR0FrEwUHAPZMuhFvjRxk6jX1pt+JHdDJb9p/NFxyetfwmXTgmoEn6X7OP1+c2LQ9ZthpmmthVtfno3k5OYGnrj+zRdoVZ+g3eCAcRIQ2OoWLbmtQKAnX/PKRNgEfv+YMy2tYTqGrAcMU27U23n1z/Tkno2fndlEN+wvFczcPbNpOSuyCzJdv0HSck4bTOolAxWq2S+irv/4al/Xrhv/dcxFOOC76ipBRd4hrhD5SHvmNuyZ4rHzySt1qFf4kP3N1s896dHw3ejyxnyQML9xytiHnFcHWj+N0eeG3/D1GXGDunIV+J3bAtPuH4MozT7TljHfXCH2k4kNE6NahjTHGWECvLu3wpyF9rDZDM3+7Sj+/ZiAG9uhoal/Adw9dgrljLw2Zx4rn/7uHzB3Z1eOEthHl//fvzo/5moEqOG1bmRNN88URLSsTCcdEf98Z1Q5xjdBHQ99u1o96MTOyYTCMGpkSikG9O+PtO43rrxhjchCxQb0747yenULmMbIh4BvieKZBE24CMcBvVBEAfHD3hRGdo/vxbfDq7bGFhhh6hjXrTO+ZdCPuVvXX+Hj9d+eZb0wYXCP00bwJI3nwzu/VCUseH4p37hyE9Beui+JqgbFDB1li1/YYf9NAXNhH2/KGvrhC/7i2xTrvtuH6c+y3Jq+RLp8rzzwRix4bittNDLPwj+sGtEjrGqKVfEqQ2v4fLuod1fUfvOJ0JHZtZztXiR09BZpUhoiGE1EGEWUR0bgA+9sQ0Qxl/3oiSlTte0pJzyAi/RQyQi7t1xWX9evWLO38MDUwf/qdeDxuPv8UdGijX0difWPsvuqFj10OABg2IHDNpr9S27vxvB4B9x97DOG+y/riFCUGzVl+tcJgL9FhIUY2DOrdKYTF4UmI8ekN9wL94r7BOC6K5n2s38tIBpx8fMiXyW2DeuLSfvqNdgoUSDDUr/a+qrZ/xkkdMPzs2F7GT143AD8/caVuL9BITrNp/DVB9zV47DfWO6zQE1ECgCkArgcwEMCdRDTQL9t9AEqYuR+ANwG8qhw7EMBIAGcDGA7gPeV8uuN/0/nfRF/e/2tMu38Idr44HGnPXwsA+Ps1Z4QdW3zOqR1xQa9OeP5m/6+sD4/85qiv+qozoxsSdubJXmEedkZ3fHbv4Bb7OygTlB7yc2c8MOw0PDDsNLx6e/Om5phhp2Hni8Nx5sktm+YAmiYC9e3eHi//9hy8f1fL5vp3DwX2V/8+qWfTdii3ld5DX/3HJ1/evzu6doh8NFHbY525ktL5vTqhd9d2+PL+X0d03Kd/uQh9urZr9rv5CLRyVqjVtHzLHd45uDd++vuwZsKvlc7tWiH9heuQ/MzVUQt8sPs6kn6+UJP+TjzefjV6LVXTwQCymDkbAIhoOoARALar8owA8LyyPQvAu+T9FUYAmM7MtQByiChLOd9afcwPzvt3X4gBzy5AbUPzGnPbVglNHTUJxxCuP6cHpiwPvqjywB4d8dodLTuMNo2/Br96cXHQ427/VU/M3pTX9Pm3gwI3qW8ddCoem5GK41ol4JN7LkJJZR1mb8pDUWUdNu4twYacYgw46Xhcf+7J+M+SzKbjHhh2Gj5YkY3xNzV/AQ3z81d2P74N3v3jr/D1+lwM7NERL956Dtq3TsAVA05Ep+Na4RhVzdnXqZVwDKFtqwR8ef8Q7DhQ0aJ2/cmoJGzNK0OHNsc2dQD/756L8JdPk/HHIb1xbYCx6yd1bIOC8tpm/tSzTzkBf0jqheKqOnz05yQkjpsHAFj2f8Mw7ts0bMgpDlq+kdKhjfc3H9K3S5PL4ZXbzsXdn2xoyjPvb5ehZ+d22FdchZzDlcg4WIF3l2fhx0cuQ8e2rXDfZ8l45sazYrLjvsv64pNVOSHz3Dm4N3p2Pg6TF2UAAH554koMnby8WZ6ZD1ysafz+WT064qweHTHhpqN2n9yxLQ6W1zTL9+eL+6DoSB3uvawv5mzOR99u7fGHi3qhfZtjseKJE1FSWYeMgxXYklfWdMwVAVqQJx7fBjee2wNtWh2DbzflA/C2KCeOOAe9urTDD2MvwxkntwyZMO2+Iaisa0BSn8648KUleOnWc5CWV4a/Du2Lvt06oLy6Hm1bJeA4ZZy5llb1l/cPwZ8+Xt/0+dJ+XfHWyEHo1qENEsfNQ9f2rfHgFafjpvNOQUF5De76eD0qahsw4aaBmPjjdtxzSSI+XbOn2TlHXdwH89IOIiHES+aUTsfhs3sHo7quAWOmbQLgHaJcVBk+JpBW92nEMHPIPwB3APhY9fluAO/65UkH0FP1eTeAbgDeBXCXKv0TAHcEuMZoACkAUnr37s3RUF5dx/f+bwNf9NJiXrv7cFN6QXk1HyyrDnqcx+Phd5dl8uSFOzktr5SZmfccPsJpeaW8KrOQq+sagh47Y0Mup+WVcll1HW/LL+PSyjoePyeN/7N4FzMzz03N57ySKv5uUx7XNzQGPc/KXYWcV1LVIr2mvoGX7yxo+lx0pJanb9jbZNO+4sqA5yusqOHHpm/m0qq6oNcMRGllHU9asCOkraGorK1v9nnxtoOcWVDR9Lmipt7/kGbMSM7llD1FzMxcVl3HqzMLuehILX++dg9vyCnibfll/PX6vVxZW8+frs7h6roGLiir5m9S9nFBWTXP2ZzHaXmlvONAWYtzV9c18LvLMrm2vvl3yywo5zmb83h1ZmFAm/zzx0pdQyNv2VfC+4orufhILafnl3JpZR0XlFVzyp5irqk/er81NHrY4/EwM/OOA2U8MzmXZ2/cF/R310ppZR3P3riPU3NLeEXGIT5cUaP52Nr6Rk7OKeLvU/PD5l287SBv2VcSg6Xa+WnbQV6642CL9OScIt55oJzT80ubypLZW7aNjZ5meTMLKnjqquwWaTOSc3lbfst7Sgszk3P5prdXcnVdAy/dcZD/vWgnMzNX1TZwZkEFP/3tVp6+YS+/tWQXl1dH9rz6AyCFg+g4cZj2ChHdAWA4M9+vfL4bwBBmHqvKk67kyVM+7wYwBN5a/jpmnqakfwJgATPPCna9pKQkTklJ0fiaEgRBEACAiDYyc1KgfVocofkAeqk+91TSAuYhomMBnACgSOOxgiAIgoFoEfpkAP2JqC8RtYa3c3WuX565AEYp23cAWKY0JeYCGKmMyukLoD+ADRAEQRBMI2yPBjM3ENFYAIsAJACYyszbiGgivD6hufD63r9QOluL4X0ZQMk3E96O2wYADzNzo0HfRRAEQQhAWB+92YiPXhAEIXJi9dELgiAIDkaEXhAEweWI0AuCILgcEXpBEASXY7vOWCIqBLA3hlN0A3BYJ3P0ROyKDLErMsSuyHCjXX2YOWBkQ9sJfawQUUqwnmcrEbsiQ+yKDLErMuLNLnHdCIIguBwRekEQBJfjRqH/0GoDgiB2RYbYFRliV2TElV2u89ELgiAIzXFjjV4QBEFQIUIvCILgchwh9EQ0gIhSVX/lRPSYXx4ioreVhci3EtGvVPtGEVGm8jeqxQWMtetPij1pRLSGiM5X7dujpKcSkW6R3DTadQURlanyTFDtC7kYvMF2PaHan05EjUTURdlnSHkp5/47EW1Trvk1EbX129+GiGYoZbKeiBJV+55S0jOI6DqT7XqciLYr99hSIuqj2teoKkv/0OJG23UPERWqrn+/ap8hz6NGu95U2bSLiEpV+wwrL+X8jyp2bfO/75X9xmlYsKWn7PoHb6jkg/BODlCn3wBgAbwL0f8awHolvQuAbOV/Z2W7s4l2XeK7HrwLrK9X7dsDoJtF5XUFgB+D5N8N4DQArQFsATDQLLv88twM79oGhpYXgFMB5AA4Tvk8E8A9fnkeAvC+sj0SwAxle6BSRm0A9FXKLsFEu64E0E7ZftBnl/L5iEH3lBa77oHfkqNKumHPoxa7/PI/Am/YdUPLSzn3OfAuudoO3vDwSwD088tjmIY5okbvx1UAdjOz/+zZEQA+Zy/rAHQioh4ArgOwmJmLmbkEwGIAw82yi5nXKNcFgHXwrrJlJsHKKxhNi8Ezcx0A32LwVth1J4CvDbh2II4FcBx5V0hrB2C/3/4RAD5TtmcBuIqISEmfzsy1zJwDIAveMjTFLmZezsxVykcz769w5RUMo5/HSOwy8/46C17hrmLmBgArANzml8cwDXOi0I9E4B/nVAD7VJ/zlLRg6WbZpeY+eN/YPhjAT0S0kYhGG2BTOLsuJqItRLSAiM5W0mxRXkTUDt6bebYq2ZDyYuZ8AK8DyAVwAEAZM//kl62pXJQHtQxAVxhYXhrtUuN/f7UlohQiWkdEt+phU4R23a64IGYRkW9JUVuUl+Li6gtgmSrZkPJSSAdwORF1Ve7tG9B8mVXAQA1zlNCTdynDWwB8Y7UtarTYRURXwvsg/lOVfBkz/wpel87DRDTURLs2wes2OR/AOwDm6HntGOzycTOA1cxcrEozpLyIqDO8tam+AE4B0J6I7tLj3LEQiV1KehKAyarkPuydTv9HAP8hotNNtOsHAInMfB68NdDPYDAR/o4jAczi5iveGVJeAMDMOwC8CuAnAAsBpAIwbbU9Rwk9vA/4JmYuCLAv2ELkZixQHsouENF5AD4GMIKZi3zpSg0EzHwIwHfQt8kf0i5mLmfmI8r2fACtiKgbbFBeCi1q/AaW19UAcpi5kJnrAXwLb9+KmqZyUdwCJwAogrHlpcUuENHVAJ4BcAsz1/rSVeWVDeBnAIPMsouZi1S2fAzgQmXb8vJSCHV/6V1evvN/wswXMvNQACUAdvllMU7DjOp8MOIPXn/xX4LsuxHNOzI28NGOjBx4OzE6K9tdTLSrN7x+20v80tsDOF61vQbAcBPtOhlHJ8wNhre5S/D6OLPhrRX5OmPPNssuZf8J8K493N6M8gIwBMA2eH26BG/t8xG/PA+jeWfsTGX7bDTvjM2Gfp2xWuwaBG8HcH+/9M4A2ijb3QBkQqdOdY129VBt/xbAOmXbsOdRi11KvjPh7dgnM8pLdY0Tlf+9AewE0Mlvv2EaptuXMPpPebiLAJygShsDYIyyTQCmKDd9GoAkVb574RXbrFACY5BdH8P79k5V/lKU9NMUgdii3JzPmGzXWOW6W+DtxLtEle8GeGsbu822S/l8D7wdnOrjjC6vF5SHLx3AF/AK90R4a8kA0BZeV1MWgA0ATlMd+4xSVhkArjfZriUAClT311wl/RLlOdii/L/PZLteUd1fywGcqTrWyOcxpF1KnucBTPI7ztDyUq6xEsB25RpX+d/7MFDDJASCIAiCy3Gaj14QBEGIEBF6QRAElyNCLwiC4HJE6AVBEFyOCL0gCILLEaEXBEFwOSL0giAILuf/AQNtY9L0Eq84AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "err_t = np.abs(y-y_test)\n",
    "plt.plot(t, err_t)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.23283050730987565"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import mean_squared_error\n",
    "rmse = np.sqrt(mean_squared_error(y, y_test))\n",
    "nrmse = rmse*100/(np.max(y)-np.min(y))\n",
    "nrmse"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "sindy",
   "language": "python",
   "name": "sindy"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}