RNN-SineWave.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = np.linspace(0,50,501)\n",
    "y = np.sin(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0. ,  0.1,  0.2,  0.3,  0.4,  0.5,  0.6,  0.7,  0.8,  0.9,  1. ,\n",
       "        1.1,  1.2,  1.3,  1.4,  1.5,  1.6,  1.7,  1.8,  1.9,  2. ,  2.1,\n",
       "        2.2,  2.3,  2.4,  2.5,  2.6,  2.7,  2.8,  2.9,  3. ,  3.1,  3.2,\n",
       "        3.3,  3.4,  3.5,  3.6,  3.7,  3.8,  3.9,  4. ,  4.1,  4.2,  4.3,\n",
       "        4.4,  4.5,  4.6,  4.7,  4.8,  4.9,  5. ,  5.1,  5.2,  5.3,  5.4,\n",
       "        5.5,  5.6,  5.7,  5.8,  5.9,  6. ,  6.1,  6.2,  6.3,  6.4,  6.5,\n",
       "        6.6,  6.7,  6.8,  6.9,  7. ,  7.1,  7.2,  7.3,  7.4,  7.5,  7.6,\n",
       "        7.7,  7.8,  7.9,  8. ,  8.1,  8.2,  8.3,  8.4,  8.5,  8.6,  8.7,\n",
       "        8.8,  8.9,  9. ,  9.1,  9.2,  9.3,  9.4,  9.5,  9.6,  9.7,  9.8,\n",
       "        9.9, 10. , 10.1, 10.2, 10.3, 10.4, 10.5, 10.6, 10.7, 10.8, 10.9,\n",
       "       11. , 11.1, 11.2, 11.3, 11.4, 11.5, 11.6, 11.7, 11.8, 11.9, 12. ,\n",
       "       12.1, 12.2, 12.3, 12.4, 12.5, 12.6, 12.7, 12.8, 12.9, 13. , 13.1,\n",
       "       13.2, 13.3, 13.4, 13.5, 13.6, 13.7, 13.8, 13.9, 14. , 14.1, 14.2,\n",
       "       14.3, 14.4, 14.5, 14.6, 14.7, 14.8, 14.9, 15. , 15.1, 15.2, 15.3,\n",
       "       15.4, 15.5, 15.6, 15.7, 15.8, 15.9, 16. , 16.1, 16.2, 16.3, 16.4,\n",
       "       16.5, 16.6, 16.7, 16.8, 16.9, 17. , 17.1, 17.2, 17.3, 17.4, 17.5,\n",
       "       17.6, 17.7, 17.8, 17.9, 18. , 18.1, 18.2, 18.3, 18.4, 18.5, 18.6,\n",
       "       18.7, 18.8, 18.9, 19. , 19.1, 19.2, 19.3, 19.4, 19.5, 19.6, 19.7,\n",
       "       19.8, 19.9, 20. , 20.1, 20.2, 20.3, 20.4, 20.5, 20.6, 20.7, 20.8,\n",
       "       20.9, 21. , 21.1, 21.2, 21.3, 21.4, 21.5, 21.6, 21.7, 21.8, 21.9,\n",
       "       22. , 22.1, 22.2, 22.3, 22.4, 22.5, 22.6, 22.7, 22.8, 22.9, 23. ,\n",
       "       23.1, 23.2, 23.3, 23.4, 23.5, 23.6, 23.7, 23.8, 23.9, 24. , 24.1,\n",
       "       24.2, 24.3, 24.4, 24.5, 24.6, 24.7, 24.8, 24.9, 25. , 25.1, 25.2,\n",
       "       25.3, 25.4, 25.5, 25.6, 25.7, 25.8, 25.9, 26. , 26.1, 26.2, 26.3,\n",
       "       26.4, 26.5, 26.6, 26.7, 26.8, 26.9, 27. , 27.1, 27.2, 27.3, 27.4,\n",
       "       27.5, 27.6, 27.7, 27.8, 27.9, 28. , 28.1, 28.2, 28.3, 28.4, 28.5,\n",
       "       28.6, 28.7, 28.8, 28.9, 29. , 29.1, 29.2, 29.3, 29.4, 29.5, 29.6,\n",
       "       29.7, 29.8, 29.9, 30. , 30.1, 30.2, 30.3, 30.4, 30.5, 30.6, 30.7,\n",
       "       30.8, 30.9, 31. , 31.1, 31.2, 31.3, 31.4, 31.5, 31.6, 31.7, 31.8,\n",
       "       31.9, 32. , 32.1, 32.2, 32.3, 32.4, 32.5, 32.6, 32.7, 32.8, 32.9,\n",
       "       33. , 33.1, 33.2, 33.3, 33.4, 33.5, 33.6, 33.7, 33.8, 33.9, 34. ,\n",
       "       34.1, 34.2, 34.3, 34.4, 34.5, 34.6, 34.7, 34.8, 34.9, 35. , 35.1,\n",
       "       35.2, 35.3, 35.4, 35.5, 35.6, 35.7, 35.8, 35.9, 36. , 36.1, 36.2,\n",
       "       36.3, 36.4, 36.5, 36.6, 36.7, 36.8, 36.9, 37. , 37.1, 37.2, 37.3,\n",
       "       37.4, 37.5, 37.6, 37.7, 37.8, 37.9, 38. , 38.1, 38.2, 38.3, 38.4,\n",
       "       38.5, 38.6, 38.7, 38.8, 38.9, 39. , 39.1, 39.2, 39.3, 39.4, 39.5,\n",
       "       39.6, 39.7, 39.8, 39.9, 40. , 40.1, 40.2, 40.3, 40.4, 40.5, 40.6,\n",
       "       40.7, 40.8, 40.9, 41. , 41.1, 41.2, 41.3, 41.4, 41.5, 41.6, 41.7,\n",
       "       41.8, 41.9, 42. , 42.1, 42.2, 42.3, 42.4, 42.5, 42.6, 42.7, 42.8,\n",
       "       42.9, 43. , 43.1, 43.2, 43.3, 43.4, 43.5, 43.6, 43.7, 43.8, 43.9,\n",
       "       44. , 44.1, 44.2, 44.3, 44.4, 44.5, 44.6, 44.7, 44.8, 44.9, 45. ,\n",
       "       45.1, 45.2, 45.3, 45.4, 45.5, 45.6, 45.7, 45.8, 45.9, 46. , 46.1,\n",
       "       46.2, 46.3, 46.4, 46.5, 46.6, 46.7, 46.8, 46.9, 47. , 47.1, 47.2,\n",
       "       47.3, 47.4, 47.5, 47.6, 47.7, 47.8, 47.9, 48. , 48.1, 48.2, 48.3,\n",
       "       48.4, 48.5, 48.6, 48.7, 48.8, 48.9, 49. , 49.1, 49.2, 49.3, 49.4,\n",
       "       49.5, 49.6, 49.7, 49.8, 49.9, 50. ])"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0.00000000e+00,  9.98334166e-02,  1.98669331e-01,  2.95520207e-01,\n",
       "        3.89418342e-01,  4.79425539e-01,  5.64642473e-01,  6.44217687e-01,\n",
       "        7.17356091e-01,  7.83326910e-01,  8.41470985e-01,  8.91207360e-01,\n",
       "        9.32039086e-01,  9.63558185e-01,  9.85449730e-01,  9.97494987e-01,\n",
       "        9.99573603e-01,  9.91664810e-01,  9.73847631e-01,  9.46300088e-01,\n",
       "        9.09297427e-01,  8.63209367e-01,  8.08496404e-01,  7.45705212e-01,\n",
       "        6.75463181e-01,  5.98472144e-01,  5.15501372e-01,  4.27379880e-01,\n",
       "        3.34988150e-01,  2.39249329e-01,  1.41120008e-01,  4.15806624e-02,\n",
       "       -5.83741434e-02, -1.57745694e-01, -2.55541102e-01, -3.50783228e-01,\n",
       "       -4.42520443e-01, -5.29836141e-01, -6.11857891e-01, -6.87766159e-01,\n",
       "       -7.56802495e-01, -8.18277111e-01, -8.71575772e-01, -9.16165937e-01,\n",
       "       -9.51602074e-01, -9.77530118e-01, -9.93691004e-01, -9.99923258e-01,\n",
       "       -9.96164609e-01, -9.82452613e-01, -9.58924275e-01, -9.25814682e-01,\n",
       "       -8.83454656e-01, -8.32267442e-01, -7.72764488e-01, -7.05540326e-01,\n",
       "       -6.31266638e-01, -5.50685543e-01, -4.64602179e-01, -3.73876665e-01,\n",
       "       -2.79415498e-01, -1.82162504e-01, -8.30894028e-02,  1.68139005e-02,\n",
       "        1.16549205e-01,  2.15119988e-01,  3.11541364e-01,  4.04849921e-01,\n",
       "        4.94113351e-01,  5.78439764e-01,  6.56986599e-01,  7.28969040e-01,\n",
       "        7.93667864e-01,  8.50436621e-01,  8.98708096e-01,  9.37999977e-01,\n",
       "        9.67919672e-01,  9.88168234e-01,  9.98543345e-01,  9.98941342e-01,\n",
       "        9.89358247e-01,  9.69889811e-01,  9.40730557e-01,  9.02171834e-01,\n",
       "        8.54598908e-01,  7.98487113e-01,  7.34397098e-01,  6.62969230e-01,\n",
       "        5.84917193e-01,  5.01020856e-01,  4.12118485e-01,  3.19098362e-01,\n",
       "        2.22889914e-01,  1.24454424e-01,  2.47754255e-02, -7.51511205e-02,\n",
       "       -1.74326781e-01, -2.71760626e-01, -3.66479129e-01, -4.57535894e-01,\n",
       "       -5.44021111e-01, -6.25070649e-01, -6.99874688e-01, -7.67685810e-01,\n",
       "       -8.27826469e-01, -8.79695760e-01, -9.22775422e-01, -9.56635016e-01,\n",
       "       -9.80936230e-01, -9.95436253e-01, -9.99990207e-01, -9.94552588e-01,\n",
       "       -9.79177729e-01, -9.54019250e-01, -9.19328526e-01, -8.75452175e-01,\n",
       "       -8.22828595e-01, -7.61983584e-01, -6.93525085e-01, -6.18137112e-01,\n",
       "       -5.36572918e-01, -4.49647465e-01, -3.58229282e-01, -2.63231791e-01,\n",
       "       -1.65604175e-01, -6.63218974e-02,  3.36230472e-02,  1.33232041e-01,\n",
       "        2.31509825e-01,  3.27474439e-01,  4.20167037e-01,  5.08661464e-01,\n",
       "        5.92073515e-01,  6.69569762e-01,  7.40375890e-01,  8.03784427e-01,\n",
       "        8.59161815e-01,  9.05954742e-01,  9.43695669e-01,  9.72007501e-01,\n",
       "        9.90607356e-01,  9.99309389e-01,  9.98026653e-01,  9.86771964e-01,\n",
       "        9.65657777e-01,  9.34895056e-01,  8.94791172e-01,  8.45746831e-01,\n",
       "        7.88252067e-01,  7.22881350e-01,  6.50287840e-01,  5.71196870e-01,\n",
       "        4.86398689e-01,  3.96740573e-01,  3.03118357e-01,  2.06467482e-01,\n",
       "        1.07753652e-01,  7.96318379e-03, -9.19068502e-02, -1.90858581e-01,\n",
       "       -2.87903317e-01, -3.82071417e-01, -4.72421986e-01, -5.58052271e-01,\n",
       "       -6.38106682e-01, -7.11785342e-01, -7.78352079e-01, -8.37141778e-01,\n",
       "       -8.87567034e-01, -9.29124013e-01, -9.61397492e-01, -9.84065005e-01,\n",
       "       -9.96900066e-01, -9.99774431e-01, -9.92659380e-01, -9.75626005e-01,\n",
       "       -9.48844498e-01, -9.12582450e-01, -8.67202179e-01, -8.13157112e-01,\n",
       "       -7.50987247e-01, -6.81313766e-01, -6.04832822e-01, -5.22308590e-01,\n",
       "       -4.34565622e-01, -3.42480618e-01, -2.46973662e-01, -1.48999026e-01,\n",
       "       -4.95356409e-02,  5.04226878e-02,  1.49877210e-01,  2.47834208e-01,\n",
       "        3.43314929e-01,  4.35365360e-01,  5.23065765e-01,  6.05539870e-01,\n",
       "        6.81963620e-01,  7.51573415e-01,  8.13673738e-01,  8.67644101e-01,\n",
       "        9.12945251e-01,  9.49124554e-01,  9.75820518e-01,  9.92766406e-01,\n",
       "        9.99792900e-01,  9.96829794e-01,  9.83906695e-01,  9.61152725e-01,\n",
       "        9.28795234e-01,  8.87157529e-01,  8.36655639e-01,  7.77794162e-01,\n",
       "        7.11161223e-01,  6.37422596e-01,  5.57315054e-01,  4.71639003e-01,\n",
       "        3.81250492e-01,  2.87052651e-01,  1.89986676e-01,  9.10224162e-02,\n",
       "       -8.85130929e-03, -1.08636595e-01, -2.07336421e-01, -3.03964609e-01,\n",
       "       -3.97555683e-01, -4.87174512e-01, -5.71925655e-01, -6.50962306e-01,\n",
       "       -7.23494756e-01, -7.88798286e-01, -8.46220404e-01, -8.95187368e-01,\n",
       "       -9.35209915e-01, -9.65888154e-01, -9.86915558e-01, -9.98082028e-01,\n",
       "       -9.99275992e-01, -9.90485521e-01, -9.71798446e-01, -9.43401482e-01,\n",
       "       -9.05578362e-01, -8.58707003e-01, -8.03255727e-01, -7.39778585e-01,\n",
       "       -6.68909820e-01, -5.91357530e-01, -5.07896590e-01, -4.19360916e-01,\n",
       "       -3.26635126e-01, -2.30645706e-01, -1.32351750e-01, -3.27353793e-02,\n",
       "        6.72080725e-02,  1.66480004e-01,  2.64088521e-01,  3.59058354e-01,\n",
       "        4.50440594e-01,  5.37322181e-01,  6.18835022e-01,  6.94164668e-01,\n",
       "        7.62558450e-01,  8.23333001e-01,  8.75881080e-01,  9.19677645e-01,\n",
       "        9.54285094e-01,  9.79357643e-01,  9.94644774e-01,  9.99993743e-01,\n",
       "        9.95351105e-01,  9.80763248e-01,  9.56375928e-01,  9.22432817e-01,\n",
       "        8.79273062e-01,  8.27327901e-01,  7.67116353e-01,  6.99240032e-01,\n",
       "        6.24377135e-01,  5.43275669e-01,  4.56745972e-01,  3.65652620e-01,\n",
       "        2.70905788e-01,  1.73452155e-01,  7.42654456e-02, -2.56632999e-02,\n",
       "       -1.25335626e-01, -2.23755640e-01, -3.19939962e-01, -4.12927549e-01,\n",
       "       -5.01789301e-01, -5.85637340e-01, -6.63633884e-01, -7.34999618e-01,\n",
       "       -7.99021479e-01, -8.55059781e-01, -9.02554608e-01, -9.41031408e-01,\n",
       "       -9.70105734e-01, -9.89487083e-01, -9.98981805e-01, -9.98495031e-01,\n",
       "       -9.88031624e-01, -9.67696132e-01, -9.37691740e-01, -8.98318243e-01,\n",
       "       -8.49969046e-01, -7.93127239e-01, -7.28360768e-01, -6.56316756e-01,\n",
       "       -5.77715044e-01, -4.93340995e-01, -4.04037645e-01, -3.10697285e-01,\n",
       "       -2.14252540e-01, -1.15667055e-01, -1.59258626e-02,  8.39744557e-02,\n",
       "        1.83035729e-01,  2.80268170e-01,  3.74700264e-01,  4.65388476e-01,\n",
       "        5.51426681e-01,  6.31955213e-01,  7.06169457e-01,  7.73327890e-01,\n",
       "        8.32759485e-01,  8.83870424e-01,  9.26150021e-01,  9.59175833e-01,\n",
       "        9.82617877e-01,  9.96241929e-01,  9.99911860e-01,  9.93591003e-01,\n",
       "        9.77342512e-01,  9.51328739e-01,  9.15809603e-01,  8.71140000e-01,\n",
       "        8.17766255e-01,  7.56221659e-01,  6.87121146e-01,  6.11155146e-01,\n",
       "        5.29082686e-01,  4.41723807e-01,  3.49951369e-01,  2.54682333e-01,\n",
       "        1.56868595e-01,  5.74874781e-02, -4.24680347e-02, -1.41999221e-01,\n",
       "       -2.40111598e-01, -3.35824859e-01, -4.28182669e-01, -5.16262220e-01,\n",
       "       -5.99183449e-01, -6.76117835e-01, -7.46296676e-01, -8.09018766e-01,\n",
       "       -8.63657409e-01, -9.09666672e-01, -9.46586846e-01, -9.74049038e-01,\n",
       "       -9.91778853e-01, -9.99599143e-01, -9.97431767e-01, -9.85298384e-01,\n",
       "       -9.63320224e-01, -9.31716888e-01, -8.90804144e-01, -8.40990780e-01,\n",
       "       -7.82774514e-01, -7.16737023e-01, -6.43538133e-01, -5.63909223e-01,\n",
       "       -4.78645919e-01, -3.88600142e-01, -2.94671602e-01, -1.97798800e-01,\n",
       "       -9.89496576e-02,  8.88156806e-04,  1.00717097e-01,  1.99539705e-01,\n",
       "        2.96368579e-01,  3.90236235e-01,  4.80204780e-01,  5.65375278e-01,\n",
       "        6.44896733e-01,  7.17974593e-01,  7.83878688e-01,  8.41950526e-01,\n",
       "        8.91609873e-01,  9.32360549e-01,  9.63795386e-01,  9.85600299e-01,\n",
       "        9.97557419e-01,  9.99547275e-01,  9.91549985e-01,  9.73645456e-01,\n",
       "        9.46012583e-01,  9.08927465e-01,  8.62760644e-01,  8.07973404e-01,\n",
       "        7.45113160e-01,  6.74807993e-01,  5.97760367e-01,  5.14740117e-01,\n",
       "        4.26576754e-01,  3.34151177e-01,  2.38386872e-01,  1.40240684e-01,\n",
       "        4.06932573e-02, -5.92607627e-02, -1.58622669e-01, -2.56399670e-01,\n",
       "       -3.51614810e-01, -4.43316731e-01, -5.30589178e-01, -6.12560153e-01,\n",
       "       -6.88410630e-01, -7.57382735e-01, -8.18787322e-01, -8.72010857e-01,\n",
       "       -9.16521548e-01, -9.51874658e-01, -9.77716952e-01, -9.93790221e-01,\n",
       "       -9.99933866e-01, -9.96086503e-01, -9.82286573e-01, -9.58671960e-01,\n",
       "       -9.25478614e-01, -8.83038191e-01, -8.31774743e-01, -7.72200476e-01,\n",
       "       -7.04910637e-01, -6.30577565e-01, -5.49943970e-01, -4.63815516e-01,\n",
       "       -3.73052771e-01, -2.78562606e-01, -1.81289136e-01, -8.22042844e-02,\n",
       "        1.77019251e-02,  1.17431263e-01,  2.15987266e-01,  3.12385196e-01,\n",
       "        4.05661877e-01,  4.94885318e-01,  5.79164028e-01,  6.57655923e-01,\n",
       "        7.29576737e-01,  7.94207862e-01,  8.50903525e-01,  8.99097240e-01,\n",
       "        9.38307473e-01,  9.68142448e-01,  9.88304064e-01,  9.98590872e-01,\n",
       "        9.98900091e-01,  9.89228630e-01,  9.69673123e-01,  9.40428963e-01,\n",
       "        9.01788348e-01,  8.54137361e-01,  7.97952117e-01,  7.33793998e-01,\n",
       "        6.62304053e-01,  5.84196584e-01,  5.00252017e-01,  4.11309096e-01,\n",
       "        3.18256511e-01,  2.22024012e-01,  1.23573123e-01,  2.38875315e-02,\n",
       "       -7.60367361e-02, -1.75201270e-01, -2.72615250e-01, -3.67305349e-01,\n",
       "       -4.58325454e-01, -5.44766123e-01, -6.25763669e-01, -7.00508791e-01,\n",
       "       -7.68254661e-01, -8.28324385e-01, -8.80117764e-01, -9.23117298e-01,\n",
       "       -9.56893350e-01, -9.81108439e-01, -9.95520616e-01, -9.99985881e-01,\n",
       "       -9.94459618e-01, -9.78997043e-01, -9.53752653e-01, -9.18978681e-01,\n",
       "       -8.75022579e-01, -8.22323540e-01, -7.61408116e-01, -6.92884954e-01,\n",
       "       -6.17438715e-01, -5.35823232e-01, -4.48853980e-01, -3.57399928e-01,\n",
       "       -2.62374854e-01])"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x205ca1ea248>]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAD4CAYAAADhNOGaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOy9e5Ak2VXm+Tvxysj3u6qyqqu6qrurX1KLllRq0Gh56AWt3Rm1YATTYlgaDEbGLOzAMDOLsNmFXTHYiJ2xFbaYYLZBAjFgCI2AUQONiZZoEJjUokuo1e9HdVVXV1VmVuU78hHvuPuH+42IyorMDHe/93pUpX9maZnh4R55wv3ce875zrnnilKKBAkSJEiwf5GKW4AECRIkSBAvEkOQIEGCBPsciSFIkCBBgn2OxBAkSJAgwT5HYggSJEiQYJ8jE7cAYTA1NaWOHz8etxgJEiRIcF3h61//+qJSanr78evSEBw/fpzTp0/HLUaCBAkSXFcQkfOdjifUUIIECRLscySGIEGCBAn2ORJDkCBBggT7HIkhSJAgQYJ9jsQQJEiQIME+hxFDICKfEpErIvLsDu+LiPy/InJGRJ4Wkbe0vfeQiLzi/zxkQp4ECRIkSNA9TEUEvwPcv8v77wNO+j8fBn4DQEQmgF8EvhW4D/hFERk3JFOCBAkSJOgCRgyBUurLwPIupzwA/K7y8AQwJiIzwPcAjymllpVSK8Bj7G5QrGOzXOOzpy9wYXkrTjGuQaOheOz5yzx7aS1uUa7BV15d5K9evEyvtTQ/v7TJZ5+8QKlaj1uUq1Cs1Pndr77GlfVS3KJchUZD8egzc/zD6ytxi3INvvH6Cn/29GzcYlyDiytbfP6pSz2nY0HhakHZEeBC2+uL/rGdjl8DEfkwXjTBsWPH7EgJ/LvPfZNHn5nnlqlBHv3pbyefTVv7X0Hwe187zy98/jnSKeEv//V3cOv0UNwiAfDspTV+8De/BsB//v5v4YNvvSlmiTxslGs8+PATzK2V+PvXlvnP3/8tcYvUxP/6B9/giy9c5rf+9hxf+jffSTbdG6m63/7Ka/zSnz2PCPzlz3wHJw8Oxy0S4OnY9/76VwCo1Bp831t6Q8c2yzU++BtfZb5Q4sG3HeVj//RNcYsUGq40UDocU7scv/agUg8rpU4ppU5NT1+zQtoIXl3Y4NFn5jl5YIizi5t88YXLVv5PUCil+PXHX+XIWD+ZlPDJvzsXt0hNfPLvzjGQS3N4NM8nHj/TM1HBo0/PMbdW4vaDQ/zJNy4xt1aMWyQAzvl6deehYV5f3uIvn+sNHavVG/zml89yy/QgmZTw+197PW6RmviNv3mVob4Mh0byPPzls3GL08Qff+MS84USt04P8oenL7CwXo5bpNBwZQguAkfbXt8EzO5yPBY8/uIVAD71I2/jwHAff/70XFyiXIVnLxWYL5T42ffezrvvOsBjz1+m0Yh/wi3X6nzx+cv84zfN8BPfdSvnFjc5t7gZt1gA/OnTsxyfHODX//lbqDcUj7+4ELdIADz6jKdTv/2jno79xbO9oWPPXFpjvlDiZ95zO9999yH+7OnZnjDqlVqDv3lpgX/yLYf5ie+8hRfn1zlzZSNusQB47PnLnJga5Nc+9BaUgi/1iOMYBq4MwSPAD/vVQ98GrCml5oAvAN8tIuN+kvi7/WOx4G9eXuDkgSGOTgzwHbdP87Vzyz0xGL78ijeJvfPOA7z7zoMsrJd5+cp6zFJ5Bmq9XONddx7gnXccALx7GDdq9QZfP7/Cd94+za3TQxwayfN3Z+KXC+Br55a5/eAQM6P9vP3WSZ442xs69pVXlwB4x62TfPvJKRY3KpztAaP+jddX2CjX+M7bp3nnnZ6OPXF2KWapPAP1xNkl3nnHAe6aGebwaJ6/fWUxbrFCw1T56B8AXwXuEJGLIvJjIvITIvIT/imPAmeBM8BvAv8LgFJqGfgl4En/56P+MedQSvHUhVXedmICgLccG2d5s8L5pfiTxk9dWOWWqUEmBnO8+dgYAE9fiD9p/A0/qfjmY+McnRjg4EgfT1+MX64X5tbZqtR56/EJRIRvu2WCfzi/GrdY1BuKr7+2zLeemATg7bdMsrhR7oko6snXlrnj4DCTQ32cOu6NgSfPxTIUr8I3L3rP7b4TExybGGByMNcTyewX5wtUag3eevM4IsKbj403Zb0eYapq6ENKqRmlVFYpdZNS6pNKqf+ilPov/vtKKfWTSqlblVL3KKVOt137KaXUbf7Pb5uQJwwuLBdZL9W458goAG+52Ztwn7oQ/8N95uIa99zkyXViapCRfIZv9IBc37y4xuHRPAdH8gDcc2SUZ3qgqkkPyDcf9Z7hGw6PMl8osbQRL4d7fmmTzUqdN/nP8o2+rr0wF3909/xsgTccGQHg1mlPx56djf9ZPnupwJGxfiYGc60Jt0d0H2g+y285OsrFlSKLMetYWPRGuUIPQCv9Gw97D/bW6SEyKeGly/EO0qWNMvOFUtNAiQhvPDLKC3OFWOUCeGm+wN3+/QJvYnt1YYOtSi1GqeDly+sM9WW4abwfgDcc9ia452bjvWcvznu6dOchT57bDgyRTknsz3Jxo8yV9TJ3z3hyiQh3HBrm5fn4ufhnZ9eazw/grplhXlvaolyLt1zz+dkCo/3Zpo5po/5iDxj1MEgMgY+XL68jAicPemWZ2XSKE1ODvBKzIXh1waMNbjvQKhe97cAQr17ZiJVbrtYbnFvcbN4vgNsPDqMUnF2Il+p4+fI6Jw8OIeIVpd3lT3Avzcf7LF+cXyfVpmP5bJpbpwdjNwT6vuj7BN6zfHG+EKuOlWt1Xlvc5I5DrTLWkweHqTdU7HTaq1c2OHmgpWN6fJ7pgdxdGCSGwMe5xU2OjPVftW7g9oPDvBJzhcKrC97/bzcEt04PsV6uxVqudn5pi2pdcdv01XJBS+a48MrlDW4/0Jo8xgdzjA9kObcU/+Rx8+TgVTp2m1+qHCf0/29fm3LHoWEKpRpXYtSx15e2aCi4ZXqweeykPw5evhz/uGwfk9NDfQznM5yJWffDIjEEPs4ubHJiavCqY7ceGOL15XjD0FevbJDPpjg82t881vI+4lO6Tgbq5skBROKNCNaKVZY2K1dNHuDlVs7FHKm8trTJ8cmBq44dnxzkwvIWtXojJqng3MIm/dk0B0f6rpIL4LUYjZQ2ULdMtXTsxNQgIt64iAsrmxWWNitXGU4R4bYDQz1T2hoUiSHAqxg6t7jJLdsMwc0TAygFs6vxtQLwJo9BUqnW2rub/cnk9RjbYFxc8RZoHZ1oTWz5bJqbxvtjDdtf96u8bt424Z6YGopVLqUU55e2uHnyah07PjlIraFi1zHPiF+rY+dj1DHtUJxoM+r5bJqZkTwXVuKTS9+T49vmixOTg1xY7o2Fi0GRGAJgdavKRrnGsW2D9FgPTLiXVkvNhJTGwZE8KYHZ1fiU7tJKkf5smvGB7FXHj44PcClGufSzajdQAMcnB5gvlChW4onuljYrbJRr10QEesKNk7Z6benaaFivYj8fo1wXVrYYH8gykt+mYxMDTYMfB/S4OzJ29bg8PNbPfKEUa3QXFokhgObEdXg0f9XxYxPxG4LZ1SKHtylcNp3i0EieizFOuLOrRY6M91/lRQLMjPbHaqDOL3sT13bPW9/D+UI8nvf5ZqTS2dmIq8mhUopLK8VrnI1MOsWR8f5Y19HMrRaZGe2/5vixiYHYxyR0NgT1hoo1rxIWiSEA5ta8yWFm24OdHuqjL5Pi9Zi8oo1yjbVi9RpDAJ7SxRoRrBavGQgAR8byXI7RK7qwXGR8IMtQ39X9FGfGPCM/F9M9072OtBwaB4bzpFPC/Fo8Bmp1q0q51uBQhwn3pvH+WKO7ubVSR90/NjHAlfVybNHdpdUig7k0I/1X69gR35jGOS7DIjEEtAbp4W2DNJUSZkbzTUPhGnrS6jQYjsQ8SC/5EcF2zIz101BwOSav6HKh1NGL1Mn22ZiepZ7oZ0auli2dEg4M98WmY7NrnaNh8CjIyzHJBToavlaumZijOx2lb4+Gj/iyxjkuwyIxBHjJ4GxamBrsu+a9gyOehxsHLjVD0GsHw+GxfubXStRjaD63VamxvFnpGBFooxWXVzS/VuJQh0lNH4srIphfK9GfvdaLBJgZzTNfiClS8ZPUne7ZzGieK+vlWHRso1yjUKp1NOqH/JXscUVRs6udIxV9LDEE1ynm1oocGs1fVZmjcWg0H6Pn4f3fjhHBWD/VuoplLcFOHCm0PMvYDEGh1Gx50Y58Ns3EYC62iGCuUGJmNH+NFwleXiW2qLOws44dGslTa6hYWnO0ouFORt1z2OLa2KdT3g5gIJdhfCCbUEPXK2Z3SEqBNxguF8qxrLC8tLrlUwfXDoYjMXofl3wDtRM1BPGU3JZrdZY3K02PcTs8mq+3IhXwnY21Uiw6NrdaJJMSpoaujYZ13iAOR0gb7E4T7sEYI4JStc7SZqVjlA6evJdWEkNwXWJ2tdTRuwVP6Sq1BqtbVcdSeXIdGvGSidsRZxiqFb3TPRvqyzCSz8TiFV0peJ6r9hi3Y2a0PzY6YX6ttKuB2qrUKZTc92iaX/MiqE46puWNI1rREcFMB+M5nM8ymEvHY6B0NNzBCQJdxNFbW5B2g31vCOoN5ScYd/bWIB6vaKfKHGiFzHFMuLOrxWaSsxMOj/XH4nnrZ9SJGgLvnsVxvxq+ju0WEQCx3LNZnxbtBH08jhzZ7GoRkZ2f5cHReHJ3Tbp2BwbhSMzVfGGx7w3B4kaZWkNdUzqqoZfdx2EI5taK15Qbagznswz3ZWLxcK+sl5gaypHZYa/dw2P9TfrIJfS92Glimxntp1CqsVl263kvbvo6totcEI/nPb+2sxM0OZgjm5ZY5JpdK3FwOL/jfs6HRvKx6P7sLpV84EUw6+Ua6yX3DEIU7HtD0HywOwwG7ZG4LqNTSnGlUN7R6waYGu5jIYZE3uJGpSOnrHEoJm9N/8/tJZoaOopy7Xm3DNTOk0f7ea6glGJuF0OQ8vNTcZSQ7pZTAV3N51739Xib3mFc6nGxuFFxJpMJmNqh7H4ReUlEzojIRzq8/3ERecr/eVlEVtveq7e994gJeYJAV910Ssi2H3cdEWyUa5RrjV0n3OmhPhZjqBpaWC/vOBDAGwwrWxXni8rm10rks6mOJZrQnmR0e8+aCxZ3mNimh/tIifuIYGWXxWQacVXNLazv7gQdHMlzZb3kfO/uhfUyw32ZqzrItkOPi+ttg5rIhkBE0sAngPcBdwMfEpG7289RSv1rpdS9Sql7gV8D/rjt7aJ+Tyn1/qjyBMXSpme5J4dyHd/PZVJMDeWcex/ao9jNEEwN52JRuMWN8h4GKodSsLzp1iuaK3gJ2U4lmtC6l0ubbu+Z9vR34ruz6RRTQ33OPW+tO7tNuHFRMIsbZSZ3izpH+qjWFctbbnVscaPM1B5OEBCLgxYFJiKC+4AzSqmzSqkK8BnggV3O/xDwBwb+rxHoB7aTIYB4FpXpQbqX0rleR6CUYnFj74gAcE5bXdlhDYHGlP+MXYftSxtlUgITgzvr2ORQn3MD1au6X294E/z0LnIdiolO85ygneWaGs41z7ueYMIQHAEutL2+6B+7BiJyM3AC+Ku2w3kROS0iT4jIB3b6JyLyYf+80wsLCwbE9rC4UWYkn6Ev0znUg5gMgT9Id1W6oT4KpZrT/RLWilWqdbVrRDAZE0+6uFHZ1UCN9mfJpMT5AqnFzQoTg7mOJZoaU0M59/drc++oc3Iox2alTqnqTseWNysotbsTdMA3+K4Xle2VH5sc9Gi+ODeNCgMThqCTdu9E3D0IfE4p1a5Vx5RSp4AfBH5VRG7tdKFS6mGl1Cml1Knp6eloErdhcXP3Bwte9YRrmkN7FNO7UTD+QFlyOIFoBd89IvCMl/MJdw/KSkSYHHJPpy1tlJns0L6kHZODOecRgX4+u9KP+lk61H/9fHa7Z7odjEvdh711LJ0SJgZzLOzDZPFF4Gjb65uA2R3OfZBttJBSatb/fRb4a+DNBmTqGovruz9YgImhHEsbFacrPxc2KsgedEKrQsHdBLLQhYGaiiFhVq7VWS/Vdr1f4E0uriePpY3KrvQL+NRQDJNaSmCsP7vjORPNCdfds1xq5sd2o9LcGyi9sHSv+WJqqG9fUkNPAidF5ISI5PAm+2uqf0TkDmAc+GrbsXER6fP/ngLeATxvQKau4SV/dh+kU4N9VOoNNhzWny9ulBkf2LlWH1oDxWUY2ooIdr5nw30ZcpmUU6pjZdOr295rwp0adj9IlzYruyY+wZN7q1Jnq+JOx5Y2KkwM9nXssdUulz7XFbrJjw3k0vRlUk4jdR2x7RYNwz41BEqpGvBTwBeAF4DPKqWeE5GPikh7FdCHgM+oq93qu4DTIvJN4HHgY0opp4ZgqQtqSHuZTgfDenlXrxviiQj05D49tHNSVkScl7Z2QycATA3GwMVvlJncI1JpVjQ5nXAru3rd0EbBxEANdeoGrCHi9UdyOyb3jlTAMxTXW46gc8F1QCilHgUe3XbsF7a9/j87XPcV4B4TMoRBte6FentNHhNtYej2fUptoZtIpVWz7DZHkEvvXKuvMTWUc1o1tLxHGbCGjgiUUjuWmZqEpqz2nHDbdGz7Npu2sLS5Ny06GUO+Z3Gj0pWOTTjOq3QTqYBO/LvTMRPY1yuLm1xkF9QQuK2L36s6AbzWysN9Gafehy6f20vBvfDYfdi+l+c9OZijXGuw6Wh3q5aB2itZ7J6L36sUEloUjOuIYLILHZtwXMTRTX4MPN0vVd3pmAnsa0Ow2EXVBLRFBM4H6e5ygfs2Ewvruy+o0XDNk2qjvic15HjBj5ZrzyR2DFy8l8Te/X5pCsYt/VjeM7ID757FkrvY4541I/XriB5KDAF7c37ay3TlFW2Wa2xV6t0ZgqGcU4Vb9mvi98LUsOetuWoBsLRZIZOSPemEVrWJm3vWvY75k4cjubYqno51O+E6Tcp2EQ2D+5LbxfUKg7k0/bmd1xxBfAsqo2CfG4K9F9SAR8EM5tLOvI8mndDFhDs+kHO6V8JqscL4QBeGYKiPekOxWnQj21KXdEJzkK67eZbdRir9Obc6ttSl7oPPxTv2vLuTy6NgXFVaLW/u3vZCQztKK47XHkXBvjYEmurp6uEO5Vh25H2s+P1TxrswBBODOaf9VlY3q4wN7Fx3rqEHgytP0otUuqOswF1E0MxddOV5u6Nguo1UQK+9cCOXUsqjrLrQfdd02vJWlfEudF+P2xXHfZCiYF8bgtVilWxaGNwj1AN/MDic1AAmBvdWurGBHKtbbha7VesN1su1riKCsQG3g6GbUkhoM1AOPe9cJsVQ394Fei45724jFfDpx003OrZVqVOpN7pyglxTtqtbla7k0sZiJYZdDcNifxuCrQpjA3vTCeDzkY4GqaZ6xrqYcCcGs1TrysliNy1XN17RxIDb8Hhpc+9affC6yQ7m0s4G6aLv3XarY64iAh1FdpPvmRzKUXFUadWMhrvQsckht5VWK1vd0aL92TS5TCqhhq4XrG5Vd11e3w6XpWqtwdC95+0iT7Dqy9WNgdL0kav8xfJGd9QQtKIoF1jZ6i65Dp5ca45yKvr7j3ZF87mbcPUK8W5033VEsNIlLSoiTAzkEmroekG3Fh5abYJdhMcrm16fodEujJT2vF0YqZWt7gepS560XKuzWal3RaUBjA9mnQ3SIDo2PuBOrtWtKpmUMNwFZeWS6giaHwM3OYJKzWsxM9HlsxwbyCbU0PWC1a3uLDy0KBg34XGV0f7srm2LNVxOuCvNiGDvezaYS5NNi5PBoL3obgwneIbM1SBd26p25XWDFxGUqg0nLZ9XfN3vhrJqRXfudKwbamgg51EwLuRaLfq632V0NzGYS6ih6wVBDMFYv6Zg3AyGbj2PlrfmYDAE8NZExJtwHQyGQtHLj4x0aQhcU0PdTGrQirRcPctuKD5oUYEuaKsg+TERYaw/60QuTVl1Py4Taui6QZCwfdQh572yVQkQqeikrAu5uk8We+e5GQzBIwI3YXujoVgrVrvWMZd5lSAGSufRXMnV/j/3wthA1qlcXd+zhBq6PlCq1inXGt2H7b5iuvI+up08RvJZUuKOGsplUvTvsHH3drgapAX/mQSJCAqlKnXLq54LpSoN1Z1368nlMrqrdi3XqENDsLpVZTif2bX9ejvG+nNN2sYmdGTb7T2bGPSiTlcr66Ni3xqCIJU54LY6Z6XLemWAVEoYc+R5r256C2q67ajoKiIolIJHBErZN+pNmiNA7qL9OpsIEhFk0imG+zJOJtxuW5hojDqLCHxqKEAFWEO1dLPXsW8NQdBB2gzbXXgfAQYp+GGoE2qoeyoNdHVObyaLwb7n3aqA6S25lFKsbHUfdYI34a45o0W7l8tZjiBAoQS0FoNeL/TQvjUErQfbW+FxsVKnVO1uZaWGq5rlIMl10H2Q7K9I1RPUSD6gUbd8z4IkPr3zHOlYtU6l1gg24Q5knfSNWu2yjYOGsxzBZoX+bJp817So2xYrUWHEEIjI/SLykoicEZGPdHj/R0RkQUSe8n9+vO29h0TkFf/nIRPydIPWIO1O6fLZNPlsyrr3EZSyAk/p3KwjCBgRDOSoNRTrllc9F0rV5mrObuUC+wn2oM9S65htAxU06Q8eF+8qDxVU94vVuvWS25WtaiDKaqJJ810fhiDyDmUikgY+AbwXbyP7J0XkkQ5bTv6hUuqntl07AfwicApQwNf9a1eiyrUXWu0SgoSh9ssOg1YngBeGPnvJDU8aJCJoerib1a699TBYK1b3bD/djglHay+CtOTQcLHGYTUgzaHPnV0t2hKpiaBRp47UC8Vq1956GASp5IPWvLKfIoL7gDNKqbNKqQrwGeCBLq/9HuAxpdSyP/k/BtxvQKY9EZTzA0/pbIehYQzU+ID9DqRKqUC15+CO8y4Ua13nB8Bddc7qlrdCfDiAEXSxxiEoZeWda58aCrp6F9pzd/aju0BO46DbFitRYcIQHAEutL2+6B/bjn8qIk+LyOdE5GjAaxGRD4vIaRE5vbCwEFnotWKVfDYVyIsYHbCfmAqauwCvbLJSs7sidaNco9ZQwbzbQTcT7lqxGsgQDPVlyKTsr3oOskJcY9wB5x2Kfuy3Xw4ZdPUutC/0tHvPgqwQBxjuy5BOiZPiEhMwYQg6afl2bflT4LhS6k3AF4FPB7jWO6jUw0qpU0qpU9PT06GF1VjZDGbhwU2Fgl4lG2RiG3WwxiGcF+lmRepaMRj1JCJOPO+gXiTohUg9mCMYyNJQsGFxE5iglXzgLvFfKAVzNkSEkXzGWRPBqDBhCC4CR9te3wTMtp+glFpSSunWhb8JvLXba21hNaAXCW4qFHTdcRDO24UhCONF6snZuvEMOEjBjee9VgzGd4OmhizTjwEXR0GbjlmUTetJmByBTWpIKRU46gRPtrWim93TosKEIXgSOCkiJ0QkBzwIPNJ+gojMtL18P/CC//cXgO8WkXERGQe+2z9mHauhvDX7qxjX/M1yul29C64MQXAv0sXkATpZHGaQ2qdggni34BuoYtVqye1asdps2NYtXCyobK4QDxDdabqmYPFZlqoNqnUVuODBhY6ZQmRDoJSqAT+FN4G/AHxWKfWciHxURN7vn/avROQ5Efkm8K+AH/GvXQZ+Cc+YPAl81D9mHUErYMB7sLa7QxZ8mqPb1btaLrA74QbZi0BDt6OwORjqDcV6qRbYEIw4GKSrARdtgcd51xt2NxoKE0G5WFDZioZDcPE2DVSIKN07//oxBJHLRwGUUo8Cj2479gttf/888PM7XPsp4FMm5AiCIL1WNPRgWLNYqlYIMak5iQg2g5e1gn2vaKMUPKeiz3/58roNkZoIE6noyWatWA1UbRQEQXMq0OLtbSbYtSMTlIsf7c9aNVBBV65rjPZnubhiv+TWBPblyuJWKWTQwWA/PA5Lc+hrbWElxCDV59uUa61JJwTzaWzL1fC9+l58loViLbB320z8W0zKFnyjPhzwWY5ZLusOQ1nBPqOGrkdsVuqBSyHBTYWCRw0FD0HB7uSx5svVbVdIDduDIWjDOY2R/iwb5Zq1csj1cg2lghso/SwLFpOMhVLwiMBFi5WCn7vIBtUxy2XdYSgraOm+i10No2JfGoKgLWU1XFQoFErBI4K0v+WgbUMQpI5awzZPGiVsVwrWS3Ym3KCtsTWcRAQhcgS5TIrBXNqq7oepzAH7EUEUHas33OxqGBX70hA0y9RCJsxsJmULIfhb8CYcm5UTYeUatSzXWsgJV3vqtibcsJOHvsdW79lWcGcD7Je2holUwH41X3MHvBD0I7jZwyQq9qUhCLN6t/18W0qnlArcLkHDBQUT2hBY8rqhNWGG8dbA3iBt0glBKRhdDmmpj33DbwIYdFIDrWN2J9yguQuw3/olcnR3HbSZ2JeGIMzqXfA2ZM9YLFUr1xpU6o3Qg6HXEozgybVRrlGrNyxIFS1sb7/eNFr7KAe7Z0O5DCmxJ9dGxc9dhIoILE+4oSOCLOsluzoWKneRRAS9jbAJRq81gb3mW60KmN6MCMJFKhn/ejtRwVqxSjolDOSClfPa9rzDRgSplDCct/csw3q3YL/xXJQcAdjTsbAGykURhynsS0PQ4pVDet6WvKKwNIe+xm5EEJIaGrBPwYz2B1uAB/bbX0SZcG0+yyjOhu3+TIUQpdPQvurZjmxRomHv+sQQ9CQKxSqZVLA2Dho2E1Nhy9TAbgldtd5gs1IPJZftCXetGJ7v9q63ZwhEvJWvQWEzwR6WsoIWF2+jHDJS7sJyK+owC/DAvhNkEvvTEPglmkG9SPDCUFs7W4VdHAXeIC1bakWtSyx7ccINSycM+Pkee5FKjeG+DKkALag1bEYEYSkr8OSqNRSlqnkuPlLuwnJSNiwtajvfYxL70xCE9CJBV8HY9dbCKN2IxTA0Ks0Bdj3vMHLp1gS9Jhd43rptyiqUjunSVgv639x3uhd1LMTaHvDyPddLv6H9aQhCPliwu0AqEjVkcTBE9SKh9wwB2Pe8w27PabPkNuy6C+8ae2svouiY7aRsVMcxMQQ9irB0AnhKt1VP/aUAACAASURBVF6qUbfQmiBsTxOwbAiavHJvRSoQPmwHGLbMxUfRMZuUVdjchc3FblFzF95nmJer0VCRdCwxBD2MsBUw0FK6DQse21qxSn82WJ94DRcRQZjBkM+m6cukrMgVdsMQDZtJWa95YDgvciRvb+vRQrHKUMjcRdOo26CGIlBW2XSKgZyddudRcheQGIKehtfqOXyoB/Y8716UK0q5LdgruS1W66E2DNHoZWoI7Bn1sHJpesRGQ7wo1JC+zs6YjCiX5RYrpmDEEIjI/SLykoicEZGPdHj/Z0XkeX/z+i+JyM1t79VF5Cn/55Ht19pAlIjAZo+anp08Ig4GWwn2KMl17zq7SdkoXqT+DNMoRKRFwU5EEKUgAezpWJScClw/EUHkjWlEJA18Angv3h7ET4rII0qp59tO+wZwSim1JSL/Evi/gX/mv1dUSt0bVY5uUarWKdca0QepJaULncS2bKDCrN7VsDUYokYqI3kvKauUClVKvBNqet1FTxr18FHncDMisKFj4XMXYE/HouQu4OpW1CZ1zDRMRAT3AWeUUmeVUhXgM8AD7ScopR5XSm35L5/A26Q+FkSpzAG7i0SiJKUy6RRDllpR66qJsIps2xBEyRHYaBO83tw1LaSB6lFqqC+TJp9NWaloKhSrodddgC657T3KSq+92OrxVtQmDMER4ELb64v+sZ3wY8BftL3Oi8hpEXlCRD6w00Ui8mH/vNMLCwuhhQ3bUlbD5krZKGVqYNErilBuCza9teiGAMw/SxN0AtijYKI8y5G8Hc47slyWuHgTzkb75/QqTBiCTia8Y22liPwQcAr4T22HjymlTgE/CPyqiNza6Vql1MNKqVNKqVPT09OhhY0cEVjkb6NQQ2BvMETJqYC9csgofXPAXptgE14k2FkpG6XKCnwds2GgIkQqYK8CzETuAvaHIbgIHG17fRMwu/0kEXkP8O+B9yulyvq4UmrW/30W+GvgzQZk2hFRE5+2WhM0Gor1CNQQ2Et+Fkrha+LB3tqLKGWtYI+CibLuAlpcvGmqI2ruArxI2kbVUGQDlc+yXrahY9FzF7A/DMGTwEkROSEiOeBB4KrqHxF5M/D/4RmBK23Hx0Wkz/97CngH0J5kNo5CRP5WxM6y8c1KjYYKb6BAe0V2BmnYZBm0BsO6YU9SP4Ogm51r2BqkragznFzZtLctpGm5mj2jIjxLaxFBhCQ22NOxqLmLfWMIlFI14KeALwAvAJ9VSj0nIh8Vkff7p/0nYAj4b9vKRO8CTovIN4HHgY9tqzYyjqh0AthpARC1AgZ0FUzvUUM2ufihvgyZgBuGaNji4qPmLvS1xuWKSFnpa+1UDfWmjkXNXVwvragjl48CKKUeBR7dduwX2v5+zw7XfQW4x4QM3SIq56evtUUnROZvezRZDOYXIkVp4wD22l+YcDZ6V8cy1qqGoo5JsBPdRTWc3ufY267VBPbdyuJCqUoukyIfYi8CjZG8eS7ehLc22p9ls1I3umVfuVanVG1EqmZqrki1QA1FmTyG+zKIhTbBUdddgB1DELWaCVoRgck9CfR+F1EjKDDvbETNXQxZXHthEvvPEBRrkSZb8JRuvScHqfltIVu8coRBOmDH8/Yoq/AGKpUShvsyFuSKtu4C7FTBRM1deNd6dfFFg32Qoux3oWGPGoqWu0hrHbPUut4U9qEhiJb4BDt18SZ4ZRtUhxG5LK29iLIAT2PEQr4nKpUGvrNhWi4TlFXevOdthq61s7I+KjUEmrJNqKGegrEHWzIbHheaXpEJPtKgITAhl6WkbFRqCOwZ9cgGykITtajlttCacE0+y6iLttqvtaFjJpyNG75q6HqDiUE62p+lWjcbHmtFGYrCxVvgSU1UMw3m0qRTYiFZbGbCtZEsju5sZNgo14zme9aKBnIXFvYkiLrIE6A/myabNru+p1pvsBVyr+52jOQTaqjn4LWgjj55gNkwVNcrp0PWK4MdntQEnSAixgeDiQQj6CoY0553NF4ZWvd7o2ySgomeu7AR3bXavkTVMbOet4ncBVwfraj3nyGImGAEOxUKJnhlG2G7CW9NX2/HQEV/ljYiFRP0o/dZPaZjFvYkMJHEBvMJ9iZlNdB7+R7T2FeGQO9oZYJXBvOet6lIxWyyOLq3pq83Syf4NfERB6ktLt7YhGvU8zbDd0Pv5QjAprPRezpmGvvKEBSrdWqN8DtaadioUIjaeRS8PkjplBiPCLJpIZ+NpiqmFyKZnDyK1TqVmhkuXq+7MJGHArNG3UTuwsaeBIVilUxK6I+wtgfMUzDmomHz+R7T2FeGIOomExo2BqmJUkgRMU51aC8y6qYatsL26N6apwumetREbXOuYYWLN5C7sLEngY6gjOhYLzobFvI9prG/DIGB8rn2601GBCYoKzC/6rlQir4AD8yHxybWN4D5jYZM5lTAAv1o6Fmaph+jPkcw333XGC1qadWzSewvQ2DIixy2VDVkZJAabla2VqwybMJAWZBLf24UmO4FY45XtpOUNTHh2niWUSMoaDkbptb3mExit39eL2JfGQJTk4fpZeM1Q6WQYMNbMzVIM5SqDco1M2svTEV3pldjFwy0egYYzGVIibnJo9kzylDU2WvVTNDaetTUtpBrRS8/Fjl3YXE/cVPYV4ag1djNwMRmsELB1OShP8MGfxsVrX7xZmRbK0ZvHgjmvTVTvHIqJQwbNOqmchdgPiIwFQ2bpmy1XFFzF7a63JrE/jIEEXeOaofJCgVTdAKYb5lgokkfmOe8TfLdYFau9s+NApNG3VTuAsxHnWvF6Is8wYKOGVh8CvZarJjEPjMEJidcc+GxKZoDbNTrR2/SB+bXOHgJRjMRlP48E+jVCddEYzcNO1Fn9GdpuprPVAGHrRbZJmHEEIjI/SLykoicEZGPdHi/T0T+0H//ayJyvO29n/ePvyQi32NCnp1QKFXpz6bJZaJ/bZOet+lIpVxrUDLQB6nk19ebSTCabZFtohkYtHrUmPLWCsUauXSKPkM6ZpqyMlk1ZCIpq3Wsd6mh6AZqMJcmZWHfC5OIrK0ikgY+AbwPuBv4kIjcve20HwNWlFK3AR8HfsW/9m68PY7fANwP/Lr/eVYQde/ddpgshzTR2E2jVRcffcI1TVmBucFgylsz3aPGVE086AnXVKQSba/udowa3JPAdDQMZkuBjemY4byKaZiICO4DziilziqlKsBngAe2nfMA8Gn/788B7xZvpDwAfEYpVVZKnQPO+J9nBab4bjDrrZnYnUzDJB9pmuYAk9U5ZiICMJvvMepsGKyLN5u7MEd1mKSsWol/c6XAxnTM0l7PpmDCEBwBLrS9vugf63iOv9n9GjDZ5bUAiMiHReS0iJxeWFgIJejkUI7bDw2HunY7RvuzbFXqVA0sGze1OArMJszWDFeagNnqHFNG3euMam7yMCeXBWfDpFE3IJtJHRvOm9t6VCll1HG0tdezKZhwXTrFwNvJw53O6eZa76BSDwMPA5w6dSoUOfnL33tPmMs6on3CnRrqi/RZJvrEN+Uy6HmbnDz6Mily6ZQRL9IbpL0ZERRKZlbJgieXdjay6Wg+m6lyW08uc/2GTOpYKiUMGdp6tFxrUKk3jEV3NjZAMgkTEcFF4Gjb65uA2Z3OEZEMMAosd3ltT8JkhYK3a1q0PvEaowaTsibpBM2TmhgMG+UaDWUmggKzhmDdUIIRTOd7DHq3BiMCk9Gw/hwjY9KwXPuBGnoSOCkiJ0Qkh5f8fWTbOY8AD/l/fxD4K+WVHDwCPOhXFZ0ATgJ/b0Am6zDZgdRUrxUwHRHoBKPJ8NgEnWAuuQ7mKRiTBgoMOhvGchcWcgQGc3dmaFGzcpnUMRuIrBlKqZqI/BTwBSANfEop9ZyIfBQ4rZR6BPgk8F9F5AxeJPCgf+1zIvJZ4HmgBvykUsrc/o8WYTIxZaoCBszmCPQgHTbm4Zry1swaKN2xVSkVKSoztd9Fu1xgzvM2mVMBUwUJ5lbVg7lqPpOUlfc5ZttymIaRu6+UehR4dNuxX2j7uwR8/w7X/jLwyybkcAmT5ZAeNWRG4fL+OglTk0efIV4ZzLXlMNUzSmOkP0Ol3qBUbdAfIU9Tqjao1qPvd9GSy6znPTaQi/w50Gq6aIqCyWdT9GXM6Nhof5azixuRP8eGs6H3vTCxjsk0ek+i6wQma5ZNJj7BXP25qTpqDVP8rY2wHaJ7uKa6VWoY1TFD7RIAcpkU/dm0uWjY0HMEG9SQYTqtR+mhxBCEhEn+1uu1YkbhwBwXb2LXtHZ4XSvNhe0mvTWI/ixN890m96A2tUpWw6M6TOUuDDpBhtZeGKeGLGwjaxKJIQiJfDZNXyZlsGqo9zxv84PUS5hFbU1gchFS++eYigh6LfGvlDL/LA0lP007G6P9WSPtzu0Z9d7MEySGIAJMcN7NXivGqaHeSjCCN0irdUWpGm0RXqFYRQSG+8yWaUZ9lqZzF6b2oNa5C6P0Y78Z+tFUzygNU3mVtaK5vmSQRAQ3NEy0mTDt3erPMrKOwODiKDDHeWteOZWKvu4CzE0eJnv+g+6DFL3axHROxfssQ/SjhTwURNexgmG61sb2tiaRGIIIMJGYMrlZjoYxLt5g3xwwx3mb7OcD5so0TfPK+rN6LYntfVZvRp0mab5elMsWEkMQASY2il8z2IJawwQX36yJN+pFGkrKGo5U9DqJta3eShaDmXyPDbm8HEG0SMXLXRgulDAYddqIhnt1LUFiCCJg1ABPanopu/6saj1am+BitU6toayE7SYiApP3qy+TJp+NvvaiUKoZ5ZXBzIRrOokNraqhKM7GZqVOvWE2d2GsAswwZZXPpozue2EaiSGIALPUUG95Hy2+23x4bCpHYBImjPrallnKCsyUQ5pOYoOnF7WIG8XbiqDaPzssTFcziYixaj4bSAxBBGgKptEI7xW1ksW9xcVb4ZV1awIDUZRJLxLMlEOa5pXBTAWY6SQ2mOG87eRUzFWA2dCxJFl8A2K0P4tSsFEJP7HZqeiI7hVZkcuQt2ZlkBqK7kxOamAoWWwhIjBRBWO6jQO0aL4ocjUainULz3LYUDWfDSSGIAKaVEeEJGOhVDPazwfMcPE2chfZdIqBXDqSXKVqnbLhdRdgqhTYLJ0AnhcfdYFUoVRlIJeOvKfB1XJF130bzgZEp/k2K16bc/PRnZlqPhtIDEEEmKhQ8Hhl814kRJTLgiGA6OGxDToB9CCNnpS1EalAtD0JbOVUINpKWRu0KETXMdNtzjV6ed/ixBBEgBHP28bkYYCLt5Fg9D4v2oTbSjCaH6RGktgWDCdEo9NML44CM1y8zYigF52gJFl8g8LEln02+G4TbYJtJBi9z4vmFa1Z4JXBk2s9QuJfb5/Zk563hSS2ieocW9FdVJrPmhOUb+170WtIDEEEmEiY2TAErTbB0eQazKXJGOSVIbq3ZiN3oT+voTx+OAw2K3WPV+5Rz9t44tMELVqsMtyXIW2oVYhG1OjORhIbWvtelGvRem3ZQKRRLiITIvKYiLzi/x7vcM69IvJVEXlORJ4WkX/W9t7viMg5EXnK/7k3ijyu0fKKovK3ZicPMBMemx4IEJ0ntUlZQXjP20ZNfPvn9VrUmU4Jw33R+g3ZMFDQu85GLzeei+rufQT4klLqJPAl//V2bAE/rJR6A3A/8KsiMtb2/r9TSt3r/zwVUR6nGMxlSEnUEjpbE25ELt5C+RxET8raWCUL0atgrCXXDeShbBr1qLpvRcf6s2yUa6FpPps5AujNfkNRDcEDwKf9vz8NfGD7CUqpl5VSr/h/zwJXgOmI/7cnkEpJJA+30VCsl832zdGIzsXbG6RRuHg9Udvj4nvMEERcJV5vKNYN7k7WjqitqD0DZT4aHslnUCp8pdVasUpKPEfPqFw93IE0qiE4qJSaA/B/H9jtZBG5D8gBr7Yd/mWfMvq4iPTtcu2HReS0iJxeWFiIKLY5RAlD10s1lDJPc0B0CsZWpKK5+LCL8Aols33iNaIudlvdskNZRe1Rs24pgoLodfGFoh0nKGruTkfDptqca0St5vvqq0v84G8+wYXlLZNiAV0YAhH5oog82+HngSD/SERmgP8K/KhSSmdLfh64E3gbMAH83E7XK6UeVkqdUkqdmp7unYAiSs2yraoJiL6oxhplFZEntUZzREx+2uKVvT0Jwpcd2opU9Gf2Yh7KRHRni0qD8HJdWN7iK68umRSpiT1jH6XUe3Z6T0Qui8iMUmrOn+iv7HDeCPDnwP+ulHqi7bPn/D/LIvLbwL8NJH0PIEptsM1BGrVFto1FSNBecluDa0oL9obpvQg0opZpNp/lgCWjHlWuHow6bU+4YfXftrMReb6woGNR4+tHgIf8vx8CPr/9BBHJAX8C/K5S6r9te2/G/y14+YVnI8rjHFG8ItuDNCwXX6032KzU7Q6GkBOILTphKB9tTYjmlYcM88rg96jpQR2LovuVWoNitW7F2YhKDdl2gqIwCLZ0LKoh+BjwXhF5BXiv/xoROSUiv+Wf8wPAdwA/0qFM9PdF5BngGWAK+A8R5XEOr01wD3pr+fB18TrJZsPz7lVvLWo5pE6um+aVIVp0Z1vHtip1qvXgdfG2IygIb9Rt0aKtfS/Czxe2dCzSSFdKLQHv7nD8NPDj/t+/B/zeDte/K8r/7wVECY9tdIXUaK+LHw7o3dj2IiGa533noWGTIjURpRzSloECT65Lq8VQ19p9lq0oanJoxzoP53JFdzbsVFlBtLbiNnUsWVkcEaP9WSq1BqUQu4G5mHDD1MXb9iIhwsItS+sbIFo5pFVDkI8mF9idcMM8S5uFEoO5NOmUhDIEulWITaPea0lsSAxBZESpNlkrVkmnhMGcuRbUGlG4eJuRShQuXtfE25two1FD9iaPaHLl0inyWfNDPQoXb9NANXcDC3HPStUGlXrDCi0K0ar5bOUuIDEEkRGF6tCTh5crN4sodfE2B2k6JQyH5LzXLXqR+nOj8Mo26YSwUaeWq9d0zFa5rYaXVwk+4drUfYie70kigh5FFK+oUDK/kYlGFArG/mAI561pucYsyRW1FNgmnQDhojtbq3fBTERgy8MN+yxttTDRiJpTtOVsJIYgIqIkpmxOHlEilSZ/a2mQhuXi9erdMQuVJqANVHC5lFKOnmU4o27ToENIA7Vlf8LtNcoKwieLbetYYggiIsoqRlv9fKDFxYcdDLZ4ZfCqTcLcr9WiZUPQn2GjXKMWsBxyq1Kn1lBW6QQIGXVaWncB0SMCG61CNMLSfLZ6WWnoxYFB9yQoVutU6/Z0LDEEEdEcpCGqc2xWJ0Spi7fJK0N4r2h1qwLAaH/OtEj+53rPImizMuteZGRqyI5cug9SOFrUnlwQfnMa+88yQ72h2KwEy/fYlisxBBHRooZClNBZHKQQnoKxsbVhO0J7a7YjgpBUhws6AaIVJNhAszqnxygraK16Dup5W88RhHyWtjbL0UgMQURk0ykGc8F3A9Ocny1qCGA4ZDmk7UEalotfdcArQ3Au3oUXCcET/42Gsu55h43uXOhYta4oBqy00s9y2FYRR8joLokIrgOESUzZ5pUhfC8Ym/XK4MkVhotf3aoy1Jcha3j7zHa5IDjn3asRwXrZXptzjbBVMGuWo86wCfa1oqdjprdo1Qi7v0RiCK4DhClVsx2CQngKxroX6U8AQbn41WLFiVy95q3ls16PmqAGynatvv7scElsu9FwFKNu+37p/xMErS1a7RjPxBAYQJiIwPbkAZ73EWaXJhdhO4SYcLeq1vIDEIW/tddATWOsP8fKZiXQNU50LELU6cKoBzee9voMQXsb9t6aLxJDYABhNqexXaYGet/iYHI1Gsr31uwmiyF4eLxatGwIIvC3ttoDa4wNZFkJWJnmwhCMD2SbuZtuUas32LC0RatG2HU0BYsL8CB6QULQBpLdIjEEBjDaH9zzdjFIR/uzrJdr1APsSbBRqdFQ9uWC4INhdavCmKXSUQjfrMxme2CN8YFcs3y2W7ToBJuRipcjCKJjeqz0KgVj0zkbDrldZaFYZTifIW1JxxJDYABheNKCg8EQhurQkYrNCTds2L5WrFqlX7xtITOhEnk2nyPA+GCWlZCGwKZsYwM5lAqoY5bbS7R/dq/lCDLpFEN9GVaLwZ7lylaF8QF7YzKSIRCRCRF5TERe8X933HxQROptm9I80nb8hIh8zb/+D/3dzK47hFmR2locZTFsH/Q+ezXAYNCTjQsuPsggVUqxulW11mdII0wVjAtDMDaQC0zBuDEEwXXMVe4CQkSdxYpV3QffcQz4LFe3qoxblCtqRPAR4EtKqZPAl/zXnVBUSt3r/7y/7fivAB/3r18BfiyiPLEgzIrU1S2PV7ZVrwze5AEE8iT1ZDM+aM8mNyePAINh0y+3dTJIA3qRq1sOIoKBLKsBF0itFatkUsKAhTbnGvp5BNExm7uTaeiV9UGeZalap1RtNMeNLYSJ7la3Koz2akQAPAB82v/703j7DncFf5/idwGfC3N9LyEMH7myVWFsIGeVV9becxBuWSuoTe9D95gJIpc+1yZlBeES/7ZLIcHLEdQbivVy986GzTbnGnrSDOLhutAxCF7R1JLLsiEYyAVO/K/0eERwUCk1B+D/PrDDeXkROS0iT4iInuwngVWllNbsi8CRnf6RiHzY/4zTCwsLEcU2izBUx6rlUkhoKfTKZjC5wF4/H/C4+PGBYF6R7c6jGuOD4SgYF9QQwGqgZ2mf5mg6GwE479aztGzUA7a/aEbDtu/ZQC6ws2E7R7AnLyEiXwQOdXjr3wf4P8eUUrMicgvwV/6G9YUO5+0Y9yqlHgYeBjh16lSwBiKWoUPcIHyk7QcLbYYgRETgwkgF8YpcTR7jA1mWA9Tr224PrKEn3JWtCscmB7q6ZmWz6k7HAhgofX9t53tGA5ZPa923SVmB972DjMlavcF6qWZ1TO5pCJRS79npPRG5LCIzSqk5EZkBruzwGbP+77Mi8tfAm4E/AsZEJONHBTcBsyG+Q+wIRw1VOTKWtyUS4OUfUhKMi1/dqjJssY2DxthANhg1VHRjoMYGcs1yyG5K9dbLNWoNxYQDXhmCG/WjE90ZjbDwutQGSxavblUYydtr46Axks9yfmmr6/NbEYF9Z2Ot2L2O2d6QCaJTQ48AD/l/PwR8fvsJIjIuIn3+31PAO4DnlZf1ehz44G7XXw8IRw1VrHu3qZQwNpALSMFUGBu0O9lChIjAQVJWqe6fpaZqbCbXoY0aCsjF26Y50ilhJB/MqK9sVa3fLwjeitpVjiBoye2KgwKOqIbgY8B7ReQV4L3+a0TklIj8ln/OXcBpEfkm3sT/MaXU8/57Pwf8rIicwcsZfDKiPLEgTIMrF4MUfM87YKRiOyELuhwyeKWJ7aTsxGAwOm3ZUeIzKM2nlPITjC6eZbDVxSsOnCAIXgHmLg8VLLpba0bDMeYIdoNSagl4d4fjp4Ef9//+CnDPDtefBe6LIkMvIOgGHa7K1MDzoINRMPaT2NBqTaCU6qqqZXWrQn82TT5rrxQS2kpuNyswvff5uv+PbQ93VFMwXU64xWqdSq3hxPMeG8gFpIaqTA7Zl2ukP8tWpU613uiK6lzZdKRjvqPV7T3T+ZdepoYS0Nqgo1tD4CoE1f8jWNWQ/SQ2eHLVApRDuqiygpZn3y1tpZ+l7RxBUApmxVEFDAR3NlwUSkDwfkO2SzQ1Wutoun2W9ueLxBAYQpAVqXpidqN0wSiYlU37JYfQNhi6NFKrDipzIDgFoytg3BjP7hvP6UjFRdQZtPGcK6MetIjDRd4OgldaNSkri7m7xBAYQpCdmpqLo3ps8qg3FIVSze1g6JYndRURDLZRQ11gdavqrWK1uEJcI0jivxmpOKKGupWrUvM6j9qOoCB4Tyt3tGgw3V8tVporpW0hMQSGEIwa0lUAbia2YrVOqYst+7T8LiKVoAmz1aLdzqMag7k02bR0bTyXtyqMWe48qhEkKeuSGtLdd7vptdV0ghxVDUH3W3y6oqyClnWv+D22bK4QTwyBIQTZpcxljkAPhm6UzqVcQcshXfTzAS/fE4ROW9msOEnIgi657V4ucBd1QnfJT9cGCoJQQ24iglTKyyl2uxrbxQrxxBAYgvdgg1JD7sLQbpSu2V6ix8JjrxSywoSDShPQdFr3FIyLSQ2CRgRuVu8CTA71AXS1Ituls9Fc39PFs2w0lLNCCQi2jsbFCvHEEBjCxKDXP6Sb8Hhlq8pALk1fxm6ZGrRVwXSRmNJepAv+VpdDdjMYCqUa1bpi0qXn3WUiz8Ug1RgfyLFRrlHtioKpOlm9CzSfy9JGN86GOydIR0PLXTzL9ZK3IZMLuSDYynoX6y4SQ2AIU0PeasFuJrYVn1d2Ae1Fd+OtLW2WAZzUeAcph1zacCcXBKRgtipOErLQRsF0oWPLm+7k0hGB1p/d0KKG7MuWy6QY7c92KZe7SAVgYrCvK8MJsLRZYXo4MQTXBSYGux8MSxsVpob7bIsEwGQAuRZ9xdTX2MbEYI6lLgyUNmKu5PL6xe892WrKygUPD+15le4oGFdyaYPTjbPhstwWPOehGx1z6QSB5zh2I1ejoVjerFjX/cQQGIJWoG6s/NJm2SHN4VEwi93ItVFhMJem3+JGJu2YHMw1vf3doGV353l7EUFjj314PZpGMeGg+kvLBd1FnbZ3tGqH/j/d6tiAQx2bGuzrSscW1j3Zp4YcOWhDOZY399Yx3ZzOtoFKDIEhTPkParELpVvaqDhTuEw6xcRAdxPu0ma5Gea7wNRQd+Gx9tbcDdI+6g21Z7WJq9bYGpq/7orm2yg3o1TbyKRTfvvu7nTM1XMEP+rsRR0b7E7HtFy2naDEEBiCDt32GqRKKZY2Kk4n3Mmh7gbD8mbFWWgMnlzdGM7lDd3Px42H261R1++7iu6muuTilVIsblSYsswrt6PbCXdxo9y8iedtqAAAFlJJREFUvy7QNTXkOOpsMgh7PEsdZdk2UIkhMITR/izplOw5GAqlGpV6w+1gGOzrOkfgiocHT7lXtvautFrarDCczzipsgKY9gfdwp6GwD2dALC43p2OTTt1Nvq6nnDdOkF9rGxVqO9BwSxtlBnJZ8hl3EyJWmf2otP0fJJQQ9cJUilhfCC354TrugJG/6+uwmPH3tpUlxVNS5vuqDSgmcjfa5DqiGDaUeI/61Mw3UYqLu9Z9/ket9RQq5pvr2fproAD2g3BHvOFTmInyeLrB1NDub0t/KZbL1L/r72822Z1glND0K1XVHYWskMbBbPHPVtYd2/Up4b6mv93J+j3nRoCP/m5G+q+jrl0Nia6XOOwuFFmymE03G1xyeJGBRH7K7ETQ2AQnue9h7e27sbCt2NyMMd6qUa5tnO/oUKpSq2h3MrVpVfklc+5mzzGfJqvG897tD/rjLICb3LvNiJwFamAVz69useCypWtCg3lOlLpzqgvOXaCxgdyiOwt1/JmmfGBnPWFgZE+XUQmROQxEXnF/z3e4Zx3ishTbT8lEfmA/97viMi5tvfujSJP3JgY7NvTK1psRgQuqaG9E9mLjrjIdkwFSJi5lCuVEiYHc3ty8Qvrbqk08GirPQ1BMyJwmYfae0FlHJRVS8f2jjpdypVOCRMDueZ8sBOWNtw4QVHNzEeALymlTgJf8l9fBaXU40qpe5VS9wLvAraAv2w75d/p95VST0WUJ1ZMdlE5oT0Al1RHN2FoM3cRR0Swy4TbaHiLtlzKBZ5s3XjeLr1u8BLZe1FDixte22JXi7agpWO7Ohvr7p2NyS5ovlq9wYqjXdPa0Q2DsOTICYpqCB4APu3//WngA3uc/0HgL5RSWxH/b09iaijHerm2a8vnpQ2vSZmLHjDtcsHuFIz2mFwOhpF8hlw6xeIuEYFeUOPScILO9+zNxbv0IgGmhnNsVuoUKzvr2KKfU3HRGltDP5/ddcx9RDDWnyUlu0cEzZXrjp/l5GDf3gUJjtb2RJ2NDiql5gD83wf2OP9B4A+2HftlEXlaRD4uIjt+YxH5sIicFpHTCwsL0aS2hFbPld0oGLeLtqA9+dlFRODQEIiIt5Zgl4hgMQa5wPO8964aqjiPCLqpNnFdmQNwYDgPsGu0ot9zWdaaSgkTg7sXcTTLgB07G91GBC7k2tMQiMgXReTZDj8PBPlHIjKDt4n9F9oO/zxwJ/A2YAL4uZ2uV0o9rJQ6pZQ6NT3dxa7iMUAr+JVCacdzXHF+7egmKbvgVye46Dzajqmh3dc4zPv3cma035VIgMfFL2yUUapz/XmxUmejXHM+4WrDc2W3CXfDbWUOwKFRzxDM76b7mxWyaWnuHOYK08N5FtZ3k8uPVGIw6rsZqHKtzlqx6sRx3POJKKXes9N7InJZRGaUUnP+RH9ll4/6AeBPlFLNbJKOJoCyiPw28G+7lLsnoQfD5V0Gw8JGmbsPj7gSCfB23RrIpXedPC6vlZge6nNKWcHeq4vn17x7eWgk70okwKOG9LaKw/lrS/fiqMyBlrOxa0SwXubWqUFXIgEw1JdhqC/TfF6dsLheZnKwz+pOW51waKSPuV3kapYBO3bQpof72CjX2KrUGMhdOxVfKXhy6XnFJqKO+keAh/y/HwI+v8u5H2IbLeQbD8TTjA8Az0aUJ1boB7aT0imlmFsrMuN4UhMRDo3mdx2kc4USMw4UbjsODPdxubCLgfKN6oGRuCiYzh6bXpfhkuaAvakhr71E2bl3C94z2ssJctn2QuPQaP+ucunx6jrqnNljvphz6ARFNQQfA94rIq8A7/VfIyKnROS39Ekichw4CvzNtut/X0SeAZ4BpoD/EFGeWDExkCOXTu0YHq8Vq5SqDScWfjtmRvPMrRV3fH9+rRiTXP0sbpSp1DrXn8+tlRgfyJLPuqvVBzjoD76djGeT73Y84epcyU5c/Hq5RrnmtoWJxqGR/K7U0NxqyflkC55cixuVHdfRzK0VGRvIOuuIqqHvxU461qJFe9wQKKWWlFLvVkqd9H8v+8dPK6V+vO2815RSR5RSjW3Xv0spdY9S6o1KqR9SSm1EkSdupFLCgZG+HR+stvCHx+IYDP27RwRrJef0C8DhsTxK7UynXS6UmpOyS7S8tc7GM47Vu+C1mZgayu14v2ZXPXnj0bE8l3fRsdm1IodjcoKgRbVsx3xMur9XRDDv6971QA0l2IaZXSiYOYcPdjtmRvNcXi93bL61Ua6xXqpxKAZvTXtFegLbjvlCKZb7pSfSneS6tFokkxLnEQF4sl1a3UHHVuNzNg6O5rmyXu7YY1/r2ExMcsHOiezZ1VI8hlPLtYOzMbdWYqgv0zFHZRqJITCMg7uEx82III7weDRPvaE6Ugrza+5C0O04PLaXV1SOxVvLZ9NMDOaY3UGuSytFZsbypB3W6mscHu3n0krnpTiXfMN1JKaIoNZQHcun53y54tCxmdHdab64nI18Ns34QHYX3S9x0FFuLDEEhnFoxIsIOpUdzq2WSMfkRe5GdTQrc2LKEYBHG2xHtd5gabMcCzUEnpHaLSKIY7IFODLez+xqZx2bXS2STYvzJDa08iqdaKtLMVJWu+V7StU6y5uVWCgr8PR/R0NQcJdTSQyBYRwazVOuNTruPDS3VuLgcF8sXuRuiSltHOLw1gb7MozkO5cdXlkvo1Q8Bgo8z3tHQ7BS5MjYgGOJPBwe66dYrXfs6zO76iX9Xa4q1ji0i+c9F2PUOZLPMJBLd4zU52OqGNLwijh2jghc6X5iCAxjtxLSuZgqc2D3xJT24OLzvD0PdzvijFTAk2uug1zVeoPL6yWOjMcUEeySv5hdLcVCPQJNGqPThDu3WiQl8eiYiDQj9e2YjdEJAvyy7mufY63e4Mp62ZlciSEwjJldElPza6VYkmXg7Xfbl+lc2hpXiabGTqWt2kDFkSMAjxpaL9colK72vD3qD26Kixry/++lDoYgTspqesiLdjs9y9m1EgeG82QdL1jUODTaOXfXjAhiumczo3lWtqrX9Cdb3PB2VXNlOBNDYBg6xLy0cvVgUEoxG8NiMg0RYWa0M+c9t1aKpWJI49AOPOnry15CNC7Pe6eKpov+s41LLp1g365j9YZivhBPBQx4m9gfHstzYbmTjnnJ9bgwM9p/zf0Ct4u2OkHr2Hb9d03XJobAMA6N5MllUs1JTMOz+o3YPA/wJq4LHQbD+aVNjk3EJ9fh0TzLm5VrOmq+vrzF+ECWEQflcx3l8p/VdnoozsQneJ0+89nUNQbqynqJekPFOuEemxi4RvchXsoK4ObJAeYLpWs874srRSYGc84Xk2loZ+LitiowfQ9vnnSTh0oMgWGkUsKxiQFeW9y86vg5//WJqXgSjADHJwc5v3S1XPWG4sJykeOTbnvTtOOYr+zbJ5DXl7Y4FqNcmmLZPki1ZxkXrywi/lqCqw1BczFZjBPusYkBLmx7jvWG8iirmCIoaE2o22V7bXHT2WTbCXrcbZ8vXlvcQgRuGk8MwXWLmzt4RdoQxDnhHp8cZHWrymrbRt5za0Uq9QY3xyjXLVNDQOseaZxf3uTYRHyD9MBwH/3ZNGe3yXVxZYvp4b7YcirgTRDX6pj3+rjjhnPtODYxyNJmhfW2vMqllSKVWoNbYpRL6/drS9sMwdImJ2LU/YMjfeSzqeaz0zi/tMnh0X5nOpYYAgu4eXKQ80tbV62wfG1xk3RKOBrjxKYniPbBcN7/+3icXpEfJbUbglK1zqWVYqxypVLCLdODnF242hC8urDBrdPxTR4At04P8urCxlU69urCBtm0cDRGz/tEh2d5dtHrHHPL9FAsMkFLv9sj4mKlztxaKVbDKSIdI/XXltxGKokhsICTB4coVutXhe5nFze4abw/tqoJgBO+wp+50mrp9MrldQBuPRDfIB3OZ5ke7uPsQkuuswubNBTcfnA4NrnAm7xebZNLKcUrVzY4eSBeuW47MESp2rhKx169ssHxyUHnrcTbcdJ/Xi9fvvpZAtwSo/EcG8gxNZTjZV/fwZtsId4ICrxxeaaDjrm8X4khsAA9eb0431K6F+fWufNQvJPH8ckB+jIpXpwrNI+9OL/O+ECWAzGsdm7H7QeHeKltkL5yZd0/Hu89u3V6kEurRbYqNcBrNrdeqnFbjIYT4Dbfu26fQM4suJ08OuHmiQFy6VTz+QG8fHmdsYGs837/23HHoWFeahuTL/jj4K6Yx+Wdh0Z4fXmLzbKnY7NrJdZLNe485G7fksQQWMAdvmK9NO8p2lalxrmlTe6eGY1TLDLpFHccGuaF+asNwR2Hhp1vFrIdd8+M8OL8OtW616D2pfl1MilpRjFx4Q2HR1EKnp/17tnz/uRx8mC8hkBPEnoy2yjXOLfYGzp2y/TgVRPuc7MF3nB4JHYdu/3gMC9fbtFpL8wVyGVSsevYXTPDKEXTEdKOmkvHMTEEFjDUl+H45ABPX1wD4IW5dZTC+c5knXDXoRGeny2glKJSa/DifIG7ZuKX6w2HR6nUGk0a5qkLq9w5M0wuE6+Kvukmb2J95pL3LJ+6sIoIvOmmsTjFYnQgy4mpQZ56fRWAZy+toRS86Wi8hgC8e/bUhdWmjr00v84bD8cv190zIxSr9WYU9dxsgTsODsdKpQHN8fecr2PPXiogArcnhuD6x6njE5w+v4JSiq+dWwLg3qPxTh4Abz0+zspWlZcvb/DMpTVK1Qb3HZ+IW6zmvXnytRVq9QZPXVjl1M3xy3VwJM+B4T7+wZ9wv/H6KrcfGGaoz+2+u53wLW0T7lMXPPnuORL/hPvWm8dZ3apydnGTZ2fXqNQbsRtOgLf5ev7355ap1Br8w+srvPXm8ZilgpvG+zk40sfXzi0D8PevLXHXoRGn62ciGQIR+X4ReU5EGiJyapfz7heRl0TkjIh8pO34CRH5moi8IiJ/KCLxkogGcd/xCZY3K7w4v85Xzixxx8HhWLqObsfbb5kE4CuvLvLEWc9A3Xci/gn35skBjoz187cvL/DUhVW2KnVOHY9/kAL8Dyen+NtXFtgo1/j7c8s9cb8Avu2WSa6sl3lhbp3HX7zCnYeGnW+U0wmn/An3715Z5G9eWiAl8I9unYxZKk/HDgz38dVXl/jG6yuUqo2ekEtEePstkzxxdolipc7Xz6/wrbe41bGoEcGzwPcBX97pBBFJA58A3gfcDXxIRO723/4V4ONKqZPACvBjEeXpGbzrrgNkUsKv/dUrfPXsEt9153TcIgFwdGKA2w8O8dnTF/mjr1/kLcfGmOyByUNEeOed03z5lQUe/vJZ8tkU33XHgbjFAuDddx5kdavK//Hfn6VYrfO+Nx6KWyQA3nP3QUTgN/7mVU6fX+E9dx2MWyQAbp0e4o6Dw/zRP1zkT785y71HxxiPOVEMno7d/8ZDPPbCZX7r786Rz6b4th4wBADvuusgixsVfu6PnqZUbfDeu90+y6hbVb6glHppj9PuA84opc4qpSrAZ4AH/A3r3wV8zj/v03gb2N8QmBrq4333zPDoM/PUG4offvvxuEVq4kf+0QlemCtwdnGzp+T60XecoFRt8JfPX+Z733ykJ+gXgPfcfYCjE/38yTcucfvBoZ6JCKaG+vgnbzrMn35zlkxKePC+o3GL1MQPvf1mnr64xtnFTf7Ft98StzhNfOi+Y1RqDR57/jLf/9ajsbUv2Y7733CIQyN5HvnmLLcdGGpG7q7gYqQdAS60vb4IfCswCawqpWptx4/s9CEi8mHgwwDHjh2zI6lh/Mfvu4cDw328684DsXWE7IQH33aUUrVOOiU8cO/huMVp4tbpIX77R9/GV19d4l+/5/a4xWmiL5Pmkw+9jc99/SI/cOpo7MnFdvzSA2/k1ukhTh0fd9aOoBv80LceY71UZXwgx/09EkGBl5j9zR8+xTcvrPIvv+vWuMVpIpdJ8Xs/fh+/+9Xz/Itvv8V5hZV02uXoqhNEvgh0epL/Xin1ef+cvwb+rVLqdIfrvx/4Hr2ZvYj8z3hRwkeBryqlbvOPHwUeVUrds5fQp06dUqdPX/OvEiRIkCDBLhCRryulrsnn7hkRKKXeE/F/XwTaY9abgFlgERgTkYwfFejjCRIkSJDAIVzEuE8CJ/0KoRzwIPCI8kKRx4EP+uc9BHzegTwJEiRIkKANUctHv1dELgJvB/5cRL7gHz8sIo8C+N7+TwFfAF4APquUes7/iJ8DflZEzuDlDD4ZRZ4ECRIkSBAce+YIehFJjiBBggQJgmOnHEHvlD8kSJAgQYJYkBiCBAkSJNjnSAxBggQJEuxzJIYgQYIECfY5rstksYgsAOdDXj6Ft4ZhPyH5zvsDyXe+8RH1+96slLqm8dl1aQiiQEROd8qa38hIvvP+QPKdb3zY+r4JNZQgQYIE+xyJIUiQIEGCfY79aAgejluAGJB85/2B5Dvf+LDyffddjiBBggQJElyN/RgRJEiQIEGCNiSGIEGCBAn2OfaVIRCR+0XkJRE5IyIfiVseGxCRT4nIFRF5tu3YhIg8JiKv+L97Y1d4AxCRoyLyuIi8ICLPichP+8dv5O+cF5G/F5Fv+t/5//KPnxCRr/nf+Q/9tu83FEQkLSLfEJE/81/f0N9ZRF4TkWdE5CkROe0fM67b+8YQiEga+ATwPuBu4EMicne8UlnB7wD3bzv2EeBLSqmTwJf81zcKasC/UUrdBXwb8JP+c72Rv3MZeJdS6luAe4H7ReTbgF8BPu5/5xXgx2KU0RZ+Gq+dvcZ++M7vVErd27Z+wLhu7xtDgLc95hml1FmlVAX4DPBAzDIZh1Lqy8DytsMPAJ/2//408AGnQlmEUmpOKfUP/t/reJPEEW7s76yUUhv+y6z/o4B3AZ/zj99Q3xlARG4C/ifgt/zXwg3+nXeAcd3eT4bgCHCh7fVF/9h+wEGl1Bx4EydwIGZ5rEBEjgNvBr7GDf6dfYrkKeAK8BjwKrDqbwQFN6Z+/yrwvwEN//UkN/53VsBfisjXReTD/jHjur3nnsU3EKTDsaR29gaBiAwBfwT8jFKq4DmLNy6UUnXgXhEZA/4EuKvTaW6lsgcR+cfAFaXU10Xku/ThDqfeMN/ZxzuUUrMicgB4TERetPFP9lNEcBE42vb6JmA2Jllc47KIzAD4v6/ELI9RiEgWzwj8vlLqj/3DN/R31lBKrQJ/jZcfGRMR7dzdaPr9DuD9IvIaHq37LrwI4Ub+ziilZv3fV/AM/n1Y0O39ZAieBE76VQY54EHgkZhlcoVHgIf8vx8CPh+jLEbh88SfBF5QSv0/bW/dyN952o8EEJF+4D14uZHHgQ/6p91Q31kp9fNKqZuUUsfxxu5fKaX+OTfwdxaRQREZ1n8D3w08iwXd3lcri0Xkf8TzItLAp5RSvxyzSMYhIn8AfBdeu9rLwC8C/x3+//bs2IbAMIjD+HNhAnqFAUxgBK3OGCqlxCqKrzGEARQ2sIXqFO+XSFAS4Z5f93bvJZf8L3d0wAS4AMvMfDwo/6SImANH4Mx9d7yh3Qn+teYZ7Ug4oA1zXWZuI2JKm5ZHwAlYZeb1ez/9jH41tM7MxT/X3Nd26J9DYJ+Zu4gY8+beLhUEkqRnlVZDkqQXDAJJKs4gkKTiDAJJKs4gkKTiDAJJKs4gkKTibjWga+uasBt+AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(x,y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's turn this into a DataFrame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.DataFrame(data=y,index=x,columns=['Sine'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>Sine</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0.0</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.1</th>\n",
       "      <td>0.099833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.2</th>\n",
       "      <td>0.198669</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.3</th>\n",
       "      <td>0.295520</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.4</th>\n",
       "      <td>0.389418</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.6</th>\n",
       "      <td>-0.617439</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.7</th>\n",
       "      <td>-0.535823</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.8</th>\n",
       "      <td>-0.448854</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.9</th>\n",
       "      <td>-0.357400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50.0</th>\n",
       "      <td>-0.262375</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>501 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          Sine\n",
       "0.0   0.000000\n",
       "0.1   0.099833\n",
       "0.2   0.198669\n",
       "0.3   0.295520\n",
       "0.4   0.389418\n",
       "...        ...\n",
       "49.6 -0.617439\n",
       "49.7 -0.535823\n",
       "49.8 -0.448854\n",
       "49.9 -0.357400\n",
       "50.0 -0.262375\n",
       "\n",
       "[501 rows x 1 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train Test Split\n",
    "\n",
    "Note! This is very different from our usual test/train split methodology!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "501"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_percent = 0.1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "50.1"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df)*test_percent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_point = np.round(len(df)*test_percent)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_ind = int(len(df) - test_point)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "451"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_ind"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "train = df.iloc[:test_ind]\n",
    "test = df.iloc[test_ind:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>Sine</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0.0</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.1</th>\n",
       "      <td>0.099833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.2</th>\n",
       "      <td>0.198669</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.3</th>\n",
       "      <td>0.295520</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.4</th>\n",
       "      <td>0.389418</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44.6</th>\n",
       "      <td>0.579164</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44.7</th>\n",
       "      <td>0.657656</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44.8</th>\n",
       "      <td>0.729577</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44.9</th>\n",
       "      <td>0.794208</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.0</th>\n",
       "      <td>0.850904</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>451 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          Sine\n",
       "0.0   0.000000\n",
       "0.1   0.099833\n",
       "0.2   0.198669\n",
       "0.3   0.295520\n",
       "0.4   0.389418\n",
       "...        ...\n",
       "44.6  0.579164\n",
       "44.7  0.657656\n",
       "44.8  0.729577\n",
       "44.9  0.794208\n",
       "45.0  0.850904\n",
       "\n",
       "[451 rows x 1 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "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>Sine</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>45.1</th>\n",
       "      <td>0.899097</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.2</th>\n",
       "      <td>0.938307</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.3</th>\n",
       "      <td>0.968142</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.4</th>\n",
       "      <td>0.988304</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.5</th>\n",
       "      <td>0.998591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.6</th>\n",
       "      <td>0.998900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.7</th>\n",
       "      <td>0.989229</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.8</th>\n",
       "      <td>0.969673</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.9</th>\n",
       "      <td>0.940429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.0</th>\n",
       "      <td>0.901788</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.1</th>\n",
       "      <td>0.854137</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.2</th>\n",
       "      <td>0.797952</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.3</th>\n",
       "      <td>0.733794</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.4</th>\n",
       "      <td>0.662304</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.5</th>\n",
       "      <td>0.584197</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.6</th>\n",
       "      <td>0.500252</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.7</th>\n",
       "      <td>0.411309</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.8</th>\n",
       "      <td>0.318257</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.9</th>\n",
       "      <td>0.222024</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.0</th>\n",
       "      <td>0.123573</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.1</th>\n",
       "      <td>0.023888</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.2</th>\n",
       "      <td>-0.076037</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.3</th>\n",
       "      <td>-0.175201</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.4</th>\n",
       "      <td>-0.272615</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.5</th>\n",
       "      <td>-0.367305</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.6</th>\n",
       "      <td>-0.458325</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.7</th>\n",
       "      <td>-0.544766</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.8</th>\n",
       "      <td>-0.625764</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.9</th>\n",
       "      <td>-0.700509</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.0</th>\n",
       "      <td>-0.768255</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.1</th>\n",
       "      <td>-0.828324</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.2</th>\n",
       "      <td>-0.880118</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.3</th>\n",
       "      <td>-0.923117</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.4</th>\n",
       "      <td>-0.956893</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.5</th>\n",
       "      <td>-0.981108</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.6</th>\n",
       "      <td>-0.995521</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.7</th>\n",
       "      <td>-0.999986</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.8</th>\n",
       "      <td>-0.994460</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.9</th>\n",
       "      <td>-0.978997</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.0</th>\n",
       "      <td>-0.953753</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.1</th>\n",
       "      <td>-0.918979</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.2</th>\n",
       "      <td>-0.875023</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.3</th>\n",
       "      <td>-0.822324</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.4</th>\n",
       "      <td>-0.761408</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.5</th>\n",
       "      <td>-0.692885</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.6</th>\n",
       "      <td>-0.617439</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.7</th>\n",
       "      <td>-0.535823</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.8</th>\n",
       "      <td>-0.448854</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.9</th>\n",
       "      <td>-0.357400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50.0</th>\n",
       "      <td>-0.262375</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Sine\n",
       "45.1  0.899097\n",
       "45.2  0.938307\n",
       "45.3  0.968142\n",
       "45.4  0.988304\n",
       "45.5  0.998591\n",
       "45.6  0.998900\n",
       "45.7  0.989229\n",
       "45.8  0.969673\n",
       "45.9  0.940429\n",
       "46.0  0.901788\n",
       "46.1  0.854137\n",
       "46.2  0.797952\n",
       "46.3  0.733794\n",
       "46.4  0.662304\n",
       "46.5  0.584197\n",
       "46.6  0.500252\n",
       "46.7  0.411309\n",
       "46.8  0.318257\n",
       "46.9  0.222024\n",
       "47.0  0.123573\n",
       "47.1  0.023888\n",
       "47.2 -0.076037\n",
       "47.3 -0.175201\n",
       "47.4 -0.272615\n",
       "47.5 -0.367305\n",
       "47.6 -0.458325\n",
       "47.7 -0.544766\n",
       "47.8 -0.625764\n",
       "47.9 -0.700509\n",
       "48.0 -0.768255\n",
       "48.1 -0.828324\n",
       "48.2 -0.880118\n",
       "48.3 -0.923117\n",
       "48.4 -0.956893\n",
       "48.5 -0.981108\n",
       "48.6 -0.995521\n",
       "48.7 -0.999986\n",
       "48.8 -0.994460\n",
       "48.9 -0.978997\n",
       "49.0 -0.953753\n",
       "49.1 -0.918979\n",
       "49.2 -0.875023\n",
       "49.3 -0.822324\n",
       "49.4 -0.761408\n",
       "49.5 -0.692885\n",
       "49.6 -0.617439\n",
       "49.7 -0.535823\n",
       "49.8 -0.448854\n",
       "49.9 -0.357400\n",
       "50.0 -0.262375"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Scale Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "scaler = MinMaxScaler()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MinMaxScaler(copy=True, feature_range=(0, 1))"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# IGNORE WARNING ITS JUST CONVERTING TO FLOATS\n",
    "# WE ONLY FIT TO TRAININ DATA, OTHERWISE WE ARE CHEATING ASSUMING INFO ABOUT TEST SET\n",
    "scaler.fit(train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "scaled_train = scaler.transform(train)\n",
    "scaled_test = scaler.transform(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.preprocessing.sequence import TimeseriesGenerator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# scaled_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# define generator\n",
    "length = 2 # Length of the output sequences (in number of timesteps)\n",
    "batch_size = 1 #Number of timeseries samples in each batch\n",
    "generator = TimeseriesGenerator(scaled_train, scaled_train, length=length, batch_size=batch_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "451"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(scaled_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "449"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(generator) # n_input = 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# scaled_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "X,y = generator[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Given the Array: \n",
      "[0.49999912 0.54991622]\n",
      "Predict this y: \n",
      " [[0.59933458]]\n"
     ]
    }
   ],
   "source": [
    "print(f'Given the Array: \\n{X.flatten()}')\n",
    "print(f'Predict this y: \\n {y}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Let's redefine to get 10 steps back and then predict the next step out\n",
    "length = 10 # Length of the output sequences (in number of timesteps)\n",
    "generator = TimeseriesGenerator(scaled_train, scaled_train, length=length, batch_size=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "X,y = generator[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Given the Array: \n",
      "[0.49999912 0.54991622 0.59933458 0.64776041 0.69470985 0.73971381\n",
      " 0.78232262 0.82211054 0.85868004 0.89166571]\n",
      "Predict this y: \n",
      " [[0.92073798]]\n"
     ]
    }
   ],
   "source": [
    "print(f'Given the Array: \\n{X.flatten()}')\n",
    "print(f'Predict this y: \\n {y}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "length = 50 # Length of the output sequences (in number of timesteps)\n",
    "generator = TimeseriesGenerator(scaled_train, scaled_train, length=length, batch_size=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now you will be able to edit the length so that it makes sense for your time series!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create the Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense,LSTM,SimpleRNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "# We're only using one feature in our time series\n",
    "n_features = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# define model\n",
    "model = Sequential()\n",
    "\n",
    "# Simple RNN layer\n",
    "model.add(SimpleRNN(50,input_shape=(length, n_features)))\n",
    "\n",
    "# Final Prediction\n",
    "model.add(Dense(1))\n",
    "\n",
    "model.compile(optimizer='adam', loss='mse')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "simple_rnn (SimpleRNN)       (None, 50)                2600      \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 1)                 51        \n",
      "=================================================================\n",
      "Total params: 2,651\n",
      "Trainable params: 2,651\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n",
      "401/401 [==============================] - 18s 44ms/step - loss: 0.0078\n",
      "Epoch 2/5\n",
      "401/401 [==============================] - 17s 43ms/step - loss: 1.0471e-04\n",
      "Epoch 3/5\n",
      "401/401 [==============================] - 17s 42ms/step - loss: 1.2039e-04\n",
      "Epoch 4/5\n",
      "401/401 [==============================] - 17s 42ms/step - loss: 2.6780e-04\n",
      "Epoch 5/5\n",
      "401/401 [==============================] - 17s 43ms/step - loss: 1.0440e-05\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x207f6992188>"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# fit model\n",
    "model.fit_generator(generator,epochs=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['loss'])"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.history.history.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x205d64d7c08>"
      ]
     },
     "execution_count": 39,
     "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": [
    "losses = pd.DataFrame(model.history.history)\n",
    "losses.plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Evaluate on Test Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "first_eval_batch = scaled_train[-length:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[8.37405820e-01],\n",
       "       [7.98881698e-01],\n",
       "       [7.57371240e-01],\n",
       "       [7.13289205e-01],\n",
       "       [6.67076045e-01],\n",
       "       [6.19193508e-01],\n",
       "       [5.70120021e-01],\n",
       "       [5.20345908e-01],\n",
       "       [4.70368497e-01],\n",
       "       [4.20687145e-01],\n",
       "       [3.71798252e-01],\n",
       "       [3.24190300e-01],\n",
       "       [2.78338972e-01],\n",
       "       [2.34702398e-01],\n",
       "       [1.93716581e-01],\n",
       "       [1.55791039e-01],\n",
       "       [1.21304709e-01],\n",
       "       [9.06021693e-02],\n",
       "       [6.39901882e-02],\n",
       "       [4.17346643e-02],\n",
       "       [2.40579672e-02],\n",
       "       [1.11367167e-02],\n",
       "       [3.10001778e-03],\n",
       "       [2.81703720e-05],\n",
       "       [1.95186738e-03],\n",
       "       [8.85188786e-03],\n",
       "       [2.06592891e-02],\n",
       "       [3.72560954e-02],\n",
       "       [5.84764771e-02],\n",
       "       [8.41084070e-02],\n",
       "       [1.13895779e-01],\n",
       "       [1.47540969e-01],\n",
       "       [1.84707803e-01],\n",
       "       [2.25024924e-01],\n",
       "       [2.68089497e-01],\n",
       "       [3.13471233e-01],\n",
       "       [3.60716695e-01],\n",
       "       [4.09353821e-01],\n",
       "       [4.58896644e-01],\n",
       "       [5.08850150e-01],\n",
       "       [5.58715219e-01],\n",
       "       [6.07993616e-01],\n",
       "       [6.56192968e-01],\n",
       "       [7.02831682e-01],\n",
       "       [7.47443761e-01],\n",
       "       [7.89583454e-01],\n",
       "       [8.28829716e-01],\n",
       "       [8.64790412e-01],\n",
       "       [8.97106234e-01],\n",
       "       [9.25454293e-01]])"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "first_eval_batch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "first_eval_batch = first_eval_batch.reshape((1, length, n_features))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.9492969]], dtype=float32)"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.predict(first_eval_batch)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.94955134])"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scaled_test[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let's put this logic in a for loop to predict into the future for the entire test range.\n",
    "\n",
    "----"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_predictions = []\n",
    "\n",
    "first_eval_batch = scaled_train[-length:]\n",
    "current_batch = first_eval_batch.reshape((1, length, n_features))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1, 50, 1)"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "current_batch.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[8.37405820e-01],\n",
       "        [7.98881698e-01],\n",
       "        [7.57371240e-01],\n",
       "        [7.13289205e-01],\n",
       "        [6.67076045e-01],\n",
       "        [6.19193508e-01],\n",
       "        [5.70120021e-01],\n",
       "        [5.20345908e-01],\n",
       "        [4.70368497e-01],\n",
       "        [4.20687145e-01],\n",
       "        [3.71798252e-01],\n",
       "        [3.24190300e-01],\n",
       "        [2.78338972e-01],\n",
       "        [2.34702398e-01],\n",
       "        [1.93716581e-01],\n",
       "        [1.55791039e-01],\n",
       "        [1.21304709e-01],\n",
       "        [9.06021693e-02],\n",
       "        [6.39901882e-02],\n",
       "        [4.17346643e-02],\n",
       "        [2.40579672e-02],\n",
       "        [1.11367167e-02],\n",
       "        [3.10001778e-03],\n",
       "        [2.81703720e-05],\n",
       "        [1.95186738e-03],\n",
       "        [8.85188786e-03],\n",
       "        [2.06592891e-02],\n",
       "        [3.72560954e-02],\n",
       "        [5.84764771e-02],\n",
       "        [8.41084070e-02],\n",
       "        [1.13895779e-01],\n",
       "        [1.47540969e-01],\n",
       "        [1.84707803e-01],\n",
       "        [2.25024924e-01],\n",
       "        [2.68089497e-01],\n",
       "        [3.13471233e-01],\n",
       "        [3.60716695e-01],\n",
       "        [4.09353821e-01],\n",
       "        [4.58896644e-01],\n",
       "        [5.08850150e-01],\n",
       "        [5.58715219e-01],\n",
       "        [6.07993616e-01],\n",
       "        [6.56192968e-01],\n",
       "        [7.02831682e-01],\n",
       "        [7.47443761e-01],\n",
       "        [7.89583454e-01],\n",
       "        [8.28829716e-01],\n",
       "        [8.64790412e-01],\n",
       "        [8.97106234e-01],\n",
       "        [9.25454293e-01]]])"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "current_batch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[7.98881698e-01],\n",
       "        [7.57371240e-01],\n",
       "        [7.13289205e-01],\n",
       "        [6.67076045e-01],\n",
       "        [6.19193508e-01],\n",
       "        [5.70120021e-01],\n",
       "        [5.20345908e-01],\n",
       "        [4.70368497e-01],\n",
       "        [4.20687145e-01],\n",
       "        [3.71798252e-01],\n",
       "        [3.24190300e-01],\n",
       "        [2.78338972e-01],\n",
       "        [2.34702398e-01],\n",
       "        [1.93716581e-01],\n",
       "        [1.55791039e-01],\n",
       "        [1.21304709e-01],\n",
       "        [9.06021693e-02],\n",
       "        [6.39901882e-02],\n",
       "        [4.17346643e-02],\n",
       "        [2.40579672e-02],\n",
       "        [1.11367167e-02],\n",
       "        [3.10001778e-03],\n",
       "        [2.81703720e-05],\n",
       "        [1.95186738e-03],\n",
       "        [8.85188786e-03],\n",
       "        [2.06592891e-02],\n",
       "        [3.72560954e-02],\n",
       "        [5.84764771e-02],\n",
       "        [8.41084070e-02],\n",
       "        [1.13895779e-01],\n",
       "        [1.47540969e-01],\n",
       "        [1.84707803e-01],\n",
       "        [2.25024924e-01],\n",
       "        [2.68089497e-01],\n",
       "        [3.13471233e-01],\n",
       "        [3.60716695e-01],\n",
       "        [4.09353821e-01],\n",
       "        [4.58896644e-01],\n",
       "        [5.08850150e-01],\n",
       "        [5.58715219e-01],\n",
       "        [6.07993616e-01],\n",
       "        [6.56192968e-01],\n",
       "        [7.02831682e-01],\n",
       "        [7.47443761e-01],\n",
       "        [7.89583454e-01],\n",
       "        [8.28829716e-01],\n",
       "        [8.64790412e-01],\n",
       "        [8.97106234e-01],\n",
       "        [9.25454293e-01],\n",
       "        [9.90000000e+01]]])"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.append(current_batch[:,1:,:],[[[99]]],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_predictions = []\n",
    "\n",
    "first_eval_batch = scaled_train[-length:]\n",
    "current_batch = first_eval_batch.reshape((1, length, n_features))\n",
    "\n",
    "for i in range(len(test)):\n",
    "    \n",
    "    # get prediction 1 time stamp ahead ([0] is for grabbing just the number instead of [array])\n",
    "    current_pred = model.predict(current_batch)[0]\n",
    "    \n",
    "    # store prediction\n",
    "    test_predictions.append(current_pred) \n",
    "    \n",
    "    # update batch to now include prediction and drop first value\n",
    "    current_batch = np.append(current_batch[:,1:,:],[[current_pred]],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[array([0.9492969], dtype=float32),\n",
       " array([0.96829283], dtype=float32),\n",
       " array([0.9825035], dtype=float32),\n",
       " array([0.9918724], dtype=float32),\n",
       " array([0.99646246], dtype=float32),\n",
       " array([0.99637645], dtype=float32),\n",
       " array([0.99164665], dtype=float32),\n",
       " array([0.98222077], dtype=float32),\n",
       " array([0.96802306], dtype=float32),\n",
       " array([0.94905436], dtype=float32),\n",
       " array([0.92531675], dtype=float32),\n",
       " array([0.8969388], dtype=float32),\n",
       " array([0.8642019], dtype=float32),\n",
       " array([0.82754314], dtype=float32),\n",
       " array([0.7874816], dtype=float32),\n",
       " array([0.74457157], dtype=float32),\n",
       " array([0.69942284], dtype=float32),\n",
       " array([0.6525521], dtype=float32),\n",
       " array([0.60444236], dtype=float32),\n",
       " array([0.5555317], dtype=float32),\n",
       " array([0.5061999], dtype=float32),\n",
       " array([0.45680717], dtype=float32),\n",
       " array([0.40768504], dtype=float32),\n",
       " array([0.35925782], dtype=float32),\n",
       " array([0.31197262], dtype=float32),\n",
       " array([0.2663339], dtype=float32),\n",
       " array([0.22290482], dtype=float32),\n",
       " array([0.18226038], dtype=float32),\n",
       " array([0.1449609], dtype=float32),\n",
       " array([0.11144441], dtype=float32),\n",
       " array([0.08208905], dtype=float32),\n",
       " array([0.05711444], dtype=float32),\n",
       " array([0.0366171], dtype=float32),\n",
       " array([0.02066744], dtype=float32),\n",
       " array([0.00932245], dtype=float32),\n",
       " array([0.00268826], dtype=float32),\n",
       " array([0.00087248], dtype=float32),\n",
       " array([0.00407234], dtype=float32),\n",
       " array([0.01242992], dtype=float32),\n",
       " array([0.02597789], dtype=float32),\n",
       " array([0.04464158], dtype=float32),\n",
       " array([0.06815765], dtype=float32),\n",
       " array([0.09613431], dtype=float32),\n",
       " array([0.12800996], dtype=float32),\n",
       " array([0.16326386], dtype=float32),\n",
       " array([0.20140329], dtype=float32),\n",
       " array([0.24198247], dtype=float32),\n",
       " array([0.28470013], dtype=float32),\n",
       " array([0.329278], dtype=float32),\n",
       " array([0.37546846], dtype=float32)]"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[9.49551344e-01],\n",
       "       [9.69156618e-01],\n",
       "       [9.84074225e-01],\n",
       "       [9.94155114e-01],\n",
       "       [9.99298559e-01],\n",
       "       [9.99453170e-01],\n",
       "       [9.94617400e-01],\n",
       "       [9.84839568e-01],\n",
       "       [9.70217371e-01],\n",
       "       [9.50896908e-01],\n",
       "       [9.27071224e-01],\n",
       "       [8.98978376e-01],\n",
       "       [8.66899060e-01],\n",
       "       [8.31153800e-01],\n",
       "       [7.92099752e-01],\n",
       "       [7.50127132e-01],\n",
       "       [7.05655314e-01],\n",
       "       [6.59128649e-01],\n",
       "       [6.11012013e-01],\n",
       "       [5.61786173e-01],\n",
       "       [5.11942978e-01],\n",
       "       [4.61980443e-01],\n",
       "       [4.12397778e-01],\n",
       "       [3.63690397e-01],\n",
       "       [3.16344967e-01],\n",
       "       [2.70834550e-01],\n",
       "       [2.27613868e-01],\n",
       "       [1.87114770e-01],\n",
       "       [1.49741909e-01],\n",
       "       [1.15868702e-01],\n",
       "       [8.58335998e-02],\n",
       "       [5.99367021e-02],\n",
       "       [3.84367625e-02],\n",
       "       [2.15486014e-02],\n",
       "       [9.44095974e-03],\n",
       "       [2.23481297e-03],\n",
       "       [2.16257798e-06],\n",
       "       [2.76531647e-03],\n",
       "       [1.04966661e-02],\n",
       "       [2.31189624e-02],\n",
       "       [4.05060876e-02],\n",
       "       [6.24843152e-02],\n",
       "       [8.88340461e-02],\n",
       "       [1.19292002e-01],\n",
       "       [1.53553858e-01],\n",
       "       [1.91277281e-01],\n",
       "       [2.32085350e-01],\n",
       "       [2.75570325e-01],\n",
       "       [3.21297718e-01],\n",
       "       [3.68810636e-01]])"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scaled_test"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Inverse Transformations and Compare"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "true_predictions = scaler.inverse_transform(test_predictions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 0.89858834],\n",
       "       [ 0.93657992],\n",
       "       [ 0.96500097],\n",
       "       [ 0.98373861],\n",
       "       [ 0.99291873],\n",
       "       [ 0.99274671],\n",
       "       [ 0.98328717],\n",
       "       [ 0.96443557],\n",
       "       [ 0.93604038],\n",
       "       [ 0.89810328],\n",
       "       [ 0.85062844],\n",
       "       [ 0.793873  ],\n",
       "       [ 0.72839973],\n",
       "       [ 0.65508279],\n",
       "       [ 0.57496037],\n",
       "       [ 0.48914098],\n",
       "       [ 0.39884424],\n",
       "       [ 0.30510358],\n",
       "       [ 0.20888481],\n",
       "       [ 0.11106424],\n",
       "       [ 0.01240146],\n",
       "       [-0.08638321],\n",
       "       [-0.18462667],\n",
       "       [-0.28148034],\n",
       "       [-0.37604998],\n",
       "       [-0.46732667],\n",
       "       [-0.55418415],\n",
       "       [-0.63547237],\n",
       "       [-0.71007074],\n",
       "       [-0.77710318],\n",
       "       [-0.83581342],\n",
       "       [-0.88576223],\n",
       "       [-0.9267566 ],\n",
       "       [-0.95865565],\n",
       "       [-0.98134546],\n",
       "       [-0.99461373],\n",
       "       [-0.99824526],\n",
       "       [-0.99184558],\n",
       "       [-0.97513057],\n",
       "       [-0.94803485],\n",
       "       [-0.91070775],\n",
       "       [-0.863676  ],\n",
       "       [-0.80772312],\n",
       "       [-0.74397234],\n",
       "       [-0.67346511],\n",
       "       [-0.59718686],\n",
       "       [-0.51602914],\n",
       "       [-0.43059453],\n",
       "       [-0.34143951],\n",
       "       [-0.24905931]])"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "true_predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "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>Sine</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>45.1</th>\n",
       "      <td>0.899097</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.2</th>\n",
       "      <td>0.938307</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.3</th>\n",
       "      <td>0.968142</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.4</th>\n",
       "      <td>0.988304</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.5</th>\n",
       "      <td>0.998591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.6</th>\n",
       "      <td>0.998900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.7</th>\n",
       "      <td>0.989229</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.8</th>\n",
       "      <td>0.969673</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.9</th>\n",
       "      <td>0.940429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.0</th>\n",
       "      <td>0.901788</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.1</th>\n",
       "      <td>0.854137</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.2</th>\n",
       "      <td>0.797952</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.3</th>\n",
       "      <td>0.733794</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.4</th>\n",
       "      <td>0.662304</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.5</th>\n",
       "      <td>0.584197</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.6</th>\n",
       "      <td>0.500252</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.7</th>\n",
       "      <td>0.411309</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.8</th>\n",
       "      <td>0.318257</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.9</th>\n",
       "      <td>0.222024</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.0</th>\n",
       "      <td>0.123573</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.1</th>\n",
       "      <td>0.023888</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.2</th>\n",
       "      <td>-0.076037</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.3</th>\n",
       "      <td>-0.175201</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.4</th>\n",
       "      <td>-0.272615</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.5</th>\n",
       "      <td>-0.367305</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.6</th>\n",
       "      <td>-0.458325</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.7</th>\n",
       "      <td>-0.544766</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.8</th>\n",
       "      <td>-0.625764</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.9</th>\n",
       "      <td>-0.700509</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.0</th>\n",
       "      <td>-0.768255</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.1</th>\n",
       "      <td>-0.828324</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.2</th>\n",
       "      <td>-0.880118</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.3</th>\n",
       "      <td>-0.923117</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.4</th>\n",
       "      <td>-0.956893</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.5</th>\n",
       "      <td>-0.981108</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.6</th>\n",
       "      <td>-0.995521</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.7</th>\n",
       "      <td>-0.999986</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.8</th>\n",
       "      <td>-0.994460</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.9</th>\n",
       "      <td>-0.978997</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.0</th>\n",
       "      <td>-0.953753</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.1</th>\n",
       "      <td>-0.918979</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.2</th>\n",
       "      <td>-0.875023</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.3</th>\n",
       "      <td>-0.822324</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.4</th>\n",
       "      <td>-0.761408</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.5</th>\n",
       "      <td>-0.692885</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.6</th>\n",
       "      <td>-0.617439</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.7</th>\n",
       "      <td>-0.535823</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.8</th>\n",
       "      <td>-0.448854</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.9</th>\n",
       "      <td>-0.357400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50.0</th>\n",
       "      <td>-0.262375</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Sine\n",
       "45.1  0.899097\n",
       "45.2  0.938307\n",
       "45.3  0.968142\n",
       "45.4  0.988304\n",
       "45.5  0.998591\n",
       "45.6  0.998900\n",
       "45.7  0.989229\n",
       "45.8  0.969673\n",
       "45.9  0.940429\n",
       "46.0  0.901788\n",
       "46.1  0.854137\n",
       "46.2  0.797952\n",
       "46.3  0.733794\n",
       "46.4  0.662304\n",
       "46.5  0.584197\n",
       "46.6  0.500252\n",
       "46.7  0.411309\n",
       "46.8  0.318257\n",
       "46.9  0.222024\n",
       "47.0  0.123573\n",
       "47.1  0.023888\n",
       "47.2 -0.076037\n",
       "47.3 -0.175201\n",
       "47.4 -0.272615\n",
       "47.5 -0.367305\n",
       "47.6 -0.458325\n",
       "47.7 -0.544766\n",
       "47.8 -0.625764\n",
       "47.9 -0.700509\n",
       "48.0 -0.768255\n",
       "48.1 -0.828324\n",
       "48.2 -0.880118\n",
       "48.3 -0.923117\n",
       "48.4 -0.956893\n",
       "48.5 -0.981108\n",
       "48.6 -0.995521\n",
       "48.7 -0.999986\n",
       "48.8 -0.994460\n",
       "48.9 -0.978997\n",
       "49.0 -0.953753\n",
       "49.1 -0.918979\n",
       "49.2 -0.875023\n",
       "49.3 -0.822324\n",
       "49.4 -0.761408\n",
       "49.5 -0.692885\n",
       "49.6 -0.617439\n",
       "49.7 -0.535823\n",
       "49.8 -0.448854\n",
       "49.9 -0.357400\n",
       "50.0 -0.262375"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'true_predictions' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-1-e47f77af07ee>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m      1\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m----> 2\u001b[1;33m \u001b[0mtest\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;34m'Predictions'\u001b[0m\u001b[1;33m]\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mtrue_predictions\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m: name 'true_predictions' is not defined"
     ]
    }
   ],
   "source": [
    "test['Predictions'] = true_predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "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>Sine</th>\n",
       "      <th>Predictions</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>45.1</th>\n",
       "      <td>0.899097</td>\n",
       "      <td>0.898588</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.2</th>\n",
       "      <td>0.938307</td>\n",
       "      <td>0.936580</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.3</th>\n",
       "      <td>0.968142</td>\n",
       "      <td>0.965001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.4</th>\n",
       "      <td>0.988304</td>\n",
       "      <td>0.983739</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.5</th>\n",
       "      <td>0.998591</td>\n",
       "      <td>0.992919</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.6</th>\n",
       "      <td>0.998900</td>\n",
       "      <td>0.992747</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.7</th>\n",
       "      <td>0.989229</td>\n",
       "      <td>0.983287</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.8</th>\n",
       "      <td>0.969673</td>\n",
       "      <td>0.964436</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45.9</th>\n",
       "      <td>0.940429</td>\n",
       "      <td>0.936040</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.0</th>\n",
       "      <td>0.901788</td>\n",
       "      <td>0.898103</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.1</th>\n",
       "      <td>0.854137</td>\n",
       "      <td>0.850628</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.2</th>\n",
       "      <td>0.797952</td>\n",
       "      <td>0.793873</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.3</th>\n",
       "      <td>0.733794</td>\n",
       "      <td>0.728400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.4</th>\n",
       "      <td>0.662304</td>\n",
       "      <td>0.655083</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.5</th>\n",
       "      <td>0.584197</td>\n",
       "      <td>0.574960</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.6</th>\n",
       "      <td>0.500252</td>\n",
       "      <td>0.489141</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.7</th>\n",
       "      <td>0.411309</td>\n",
       "      <td>0.398844</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.8</th>\n",
       "      <td>0.318257</td>\n",
       "      <td>0.305104</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46.9</th>\n",
       "      <td>0.222024</td>\n",
       "      <td>0.208885</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.0</th>\n",
       "      <td>0.123573</td>\n",
       "      <td>0.111064</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.1</th>\n",
       "      <td>0.023888</td>\n",
       "      <td>0.012401</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.2</th>\n",
       "      <td>-0.076037</td>\n",
       "      <td>-0.086383</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.3</th>\n",
       "      <td>-0.175201</td>\n",
       "      <td>-0.184627</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.4</th>\n",
       "      <td>-0.272615</td>\n",
       "      <td>-0.281480</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.5</th>\n",
       "      <td>-0.367305</td>\n",
       "      <td>-0.376050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.6</th>\n",
       "      <td>-0.458325</td>\n",
       "      <td>-0.467327</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.7</th>\n",
       "      <td>-0.544766</td>\n",
       "      <td>-0.554184</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.8</th>\n",
       "      <td>-0.625764</td>\n",
       "      <td>-0.635472</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.9</th>\n",
       "      <td>-0.700509</td>\n",
       "      <td>-0.710071</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.0</th>\n",
       "      <td>-0.768255</td>\n",
       "      <td>-0.777103</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.1</th>\n",
       "      <td>-0.828324</td>\n",
       "      <td>-0.835813</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.2</th>\n",
       "      <td>-0.880118</td>\n",
       "      <td>-0.885762</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.3</th>\n",
       "      <td>-0.923117</td>\n",
       "      <td>-0.926757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.4</th>\n",
       "      <td>-0.956893</td>\n",
       "      <td>-0.958656</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.5</th>\n",
       "      <td>-0.981108</td>\n",
       "      <td>-0.981345</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.6</th>\n",
       "      <td>-0.995521</td>\n",
       "      <td>-0.994614</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.7</th>\n",
       "      <td>-0.999986</td>\n",
       "      <td>-0.998245</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.8</th>\n",
       "      <td>-0.994460</td>\n",
       "      <td>-0.991846</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48.9</th>\n",
       "      <td>-0.978997</td>\n",
       "      <td>-0.975131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.0</th>\n",
       "      <td>-0.953753</td>\n",
       "      <td>-0.948035</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.1</th>\n",
       "      <td>-0.918979</td>\n",
       "      <td>-0.910708</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.2</th>\n",
       "      <td>-0.875023</td>\n",
       "      <td>-0.863676</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.3</th>\n",
       "      <td>-0.822324</td>\n",
       "      <td>-0.807723</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.4</th>\n",
       "      <td>-0.761408</td>\n",
       "      <td>-0.743972</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.5</th>\n",
       "      <td>-0.692885</td>\n",
       "      <td>-0.673465</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.6</th>\n",
       "      <td>-0.617439</td>\n",
       "      <td>-0.597187</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.7</th>\n",
       "      <td>-0.535823</td>\n",
       "      <td>-0.516029</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.8</th>\n",
       "      <td>-0.448854</td>\n",
       "      <td>-0.430595</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.9</th>\n",
       "      <td>-0.357400</td>\n",
       "      <td>-0.341440</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50.0</th>\n",
       "      <td>-0.262375</td>\n",
       "      <td>-0.249059</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Sine  Predictions\n",
       "45.1  0.899097     0.898588\n",
       "45.2  0.938307     0.936580\n",
       "45.3  0.968142     0.965001\n",
       "45.4  0.988304     0.983739\n",
       "45.5  0.998591     0.992919\n",
       "45.6  0.998900     0.992747\n",
       "45.7  0.989229     0.983287\n",
       "45.8  0.969673     0.964436\n",
       "45.9  0.940429     0.936040\n",
       "46.0  0.901788     0.898103\n",
       "46.1  0.854137     0.850628\n",
       "46.2  0.797952     0.793873\n",
       "46.3  0.733794     0.728400\n",
       "46.4  0.662304     0.655083\n",
       "46.5  0.584197     0.574960\n",
       "46.6  0.500252     0.489141\n",
       "46.7  0.411309     0.398844\n",
       "46.8  0.318257     0.305104\n",
       "46.9  0.222024     0.208885\n",
       "47.0  0.123573     0.111064\n",
       "47.1  0.023888     0.012401\n",
       "47.2 -0.076037    -0.086383\n",
       "47.3 -0.175201    -0.184627\n",
       "47.4 -0.272615    -0.281480\n",
       "47.5 -0.367305    -0.376050\n",
       "47.6 -0.458325    -0.467327\n",
       "47.7 -0.544766    -0.554184\n",
       "47.8 -0.625764    -0.635472\n",
       "47.9 -0.700509    -0.710071\n",
       "48.0 -0.768255    -0.777103\n",
       "48.1 -0.828324    -0.835813\n",
       "48.2 -0.880118    -0.885762\n",
       "48.3 -0.923117    -0.926757\n",
       "48.4 -0.956893    -0.958656\n",
       "48.5 -0.981108    -0.981345\n",
       "48.6 -0.995521    -0.994614\n",
       "48.7 -0.999986    -0.998245\n",
       "48.8 -0.994460    -0.991846\n",
       "48.9 -0.978997    -0.975131\n",
       "49.0 -0.953753    -0.948035\n",
       "49.1 -0.918979    -0.910708\n",
       "49.2 -0.875023    -0.863676\n",
       "49.3 -0.822324    -0.807723\n",
       "49.4 -0.761408    -0.743972\n",
       "49.5 -0.692885    -0.673465\n",
       "49.6 -0.617439    -0.597187\n",
       "49.7 -0.535823    -0.516029\n",
       "49.8 -0.448854    -0.430595\n",
       "49.9 -0.357400    -0.341440\n",
       "50.0 -0.262375    -0.249059"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x2080265dbc8>"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "test.plot(figsize=(12,8))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Adding in Early Stopping and Validation Generator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.callbacks import EarlyStopping"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "early_stop = EarlyStopping(monitor='val_loss',patience=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "length = 49\n",
    "generator = TimeseriesGenerator(scaled_train,scaled_train,\n",
    "                               length=length,batch_size=1)\n",
    "\n",
    "\n",
    "validation_generator = TimeseriesGenerator(scaled_test,scaled_test,\n",
    "                                          length=length,batch_size=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# LSTMS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "# define model\n",
    "model = Sequential()\n",
    "\n",
    "# Simple RNN layer\n",
    "model.add(LSTM(50,input_shape=(length, n_features)))\n",
    "\n",
    "# Final Prediction\n",
    "model.add(Dense(1))\n",
    "\n",
    "model.compile(optimizer='adam', loss='mse')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "402/402 [==============================] - 6s 16ms/step - loss: 0.0158 - val_loss: 1.3247e-04\n",
      "Epoch 2/20\n",
      "402/402 [==============================] - 5s 14ms/step - loss: 1.3544e-04 - val_loss: 5.6189e-05\n",
      "Epoch 3/20\n",
      "402/402 [==============================] - 5s 13ms/step - loss: 3.0088e-04 - val_loss: 2.1999e-05\n",
      "Epoch 4/20\n",
      "402/402 [==============================] - 6s 14ms/step - loss: 7.6084e-05 - val_loss: 8.2873e-05\n",
      "Epoch 5/20\n",
      "402/402 [==============================] - 6s 14ms/step - loss: 4.9175e-05 - val_loss: 1.1015e-04\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x208026c92c8>"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit_generator(generator,epochs=20,\n",
    "                   validation_data=validation_generator,\n",
    "                   callbacks=[early_stop])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_predictions = []\n",
    "\n",
    "first_eval_batch = scaled_train[-length:]\n",
    "current_batch = first_eval_batch.reshape((1, length, n_features))\n",
    "\n",
    "for i in range(len(test)):\n",
    "    \n",
    "    # get prediction 1 time stamp ahead ([0] is for grabbing just the number instead of [array])\n",
    "    current_pred = model.predict(current_batch)[0]\n",
    "    \n",
    "    # store prediction\n",
    "    test_predictions.append(current_pred) \n",
    "    \n",
    "    # update batch to now include prediction and drop first value\n",
    "    current_batch = np.append(current_batch[:,1:,:],[[current_pred]],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Marcial\\Anaconda3\\envs\\tf2gpu\\lib\\site-packages\\ipykernel_launcher.py:3: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  This is separate from the ipykernel package so we can avoid doing imports until\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x20834c08b08>"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "true_predictions = scaler.inverse_transform(test_predictions)\n",
    "test['LSTM Predictions'] = true_predictions\n",
    "test.plot(figsize=(12,8))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Forecasting\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "full_scaler = MinMaxScaler()\n",
    "scaled_full_data = full_scaler.fit_transform(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "length = 50 # Length of the output sequences (in number of timesteps)\n",
    "generator = TimeseriesGenerator(scaled_full_data, scaled_full_data, length=length, batch_size=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/6\n",
      "451/451 [==============================] - 6s 14ms/step - loss: 0.0148\n",
      "Epoch 2/6\n",
      "451/451 [==============================] - 6s 13ms/step - loss: 4.9990e-05\n",
      "Epoch 3/6\n",
      "451/451 [==============================] - 6s 14ms/step - loss: 1.9414e-04\n",
      "Epoch 4/6\n",
      "451/451 [==============================] - 6s 14ms/step - loss: 1.2191e-04\n",
      "Epoch 5/6\n",
      "451/451 [==============================] - 6s 13ms/step - loss: 4.0607e-05\n",
      "Epoch 6/6\n",
      "451/451 [==============================] - 6s 13ms/step - loss: 1.8571e-04\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x20835075608>"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = Sequential()\n",
    "model.add(LSTM(50, input_shape=(length, n_features)))\n",
    "model.add(Dense(1))\n",
    "model.compile(optimizer='adam', loss='mse')\n",
    "model.fit_generator(generator,epochs=6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "forecast = []\n",
    "\n",
    "first_eval_batch = scaled_full_data[-length:]\n",
    "current_batch = first_eval_batch.reshape((1, length, n_features))\n",
    "\n",
    "for i in range(len(test)):\n",
    "    \n",
    "    # get prediction 1 time stamp ahead ([0] is for grabbing just the number instead of [array])\n",
    "    current_pred = model.predict(current_batch)[0]\n",
    "    \n",
    "    # store prediction\n",
    "    forecast.append(current_pred) \n",
    "    \n",
    "    # update batch to now include prediction and drop first value\n",
    "    current_batch = np.append(current_batch[:,1:,:],[[current_pred]],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "forecast = scaler.inverse_transform(forecast)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "# forecast"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "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>Sine</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0.0</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.1</th>\n",
       "      <td>0.099833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.2</th>\n",
       "      <td>0.198669</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.3</th>\n",
       "      <td>0.295520</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.4</th>\n",
       "      <td>0.389418</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.6</th>\n",
       "      <td>-0.617439</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.7</th>\n",
       "      <td>-0.535823</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.8</th>\n",
       "      <td>-0.448854</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49.9</th>\n",
       "      <td>-0.357400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50.0</th>\n",
       "      <td>-0.262375</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>501 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          Sine\n",
       "0.0   0.000000\n",
       "0.1   0.099833\n",
       "0.2   0.198669\n",
       "0.3   0.295520\n",
       "0.4   0.389418\n",
       "...        ...\n",
       "49.6 -0.617439\n",
       "49.7 -0.535823\n",
       "49.8 -0.448854\n",
       "49.9 -0.357400\n",
       "50.0 -0.262375\n",
       "\n",
       "[501 rows x 1 columns]"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "50"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(forecast)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5.0"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "50*0.1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [],
   "source": [
    "forecast_index = np.arange(50.1,55.1,step=0.1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "50"
      ]
     },
     "execution_count": 93,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(forecast_index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x2083cb5b248>]"
      ]
     },
     "execution_count": 94,
     "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(df.index,df['Sine'])\n",
    "plt.plot(forecast_index,forecast)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}