GPU ML for Q3b--ch1.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.python.keras.layers import Input, Dense, Activation, Dropout, MaxPooling1D, Flatten, LSTM, Convolution1D\n",
    "from tensorflow.python.keras.models import Sequential\n",
    "from tensorflow.python.keras.optimizers import RMSprop\n",
    "from tensorflow.python.keras.callbacks import EarlyStopping, ModelCheckpoint\n",
    "from tensorflow.python.keras.utils.data_utils import get_file\n",
    "from timeit import default_timer as timer \n",
    "import numpy as np\n",
    "import matplotlib.mlab as mlab\n",
    "import matplotlib.pyplot as plt\n",
    "import scipy.stats as scs\n",
    "import random\n",
    "import sys\n",
    "import os\n",
    "import glob"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-20.889075 30.06359864760662\n",
      "0.013679\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY4AAAEYCAYAAABLOxEiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deXwU9f3H8dcnCXeAgGDkkiByiAioEY+2mngVbJVqrSXVgq2KirT28MBqq622ov5qq6IiVgQ8QC1WULGg1GihooByqkC4wyk34Qokn98fM9Fl2TPZ2dlNPs/HYx/Jznxn9r2zs/vZOfY7oqoYY4wxscrwO4Axxpj0YoXDGGNMXKxwGGOMiYsVDmOMMXGxwmGMMSYuVjiMMcbExQqHMcaYuFjhMMYYExcrHB4QkSUiUuB3Dr+ISDcR+UxE9ojIL/3OU1MiMlZEHvA7hzGpwgpHnERktYhcGDTsWhGZWXVfVU9W1eJ451OL3AEUq2pTVX08VAN3mS0SkX0isklEnhKR5kFtVovIfhEpc9uMFZHsoPGbRaRJwLDrRaQ4aD4DReRjEdkrIlvc/4eKiCT2adf61zUkEWkgIs+JyBr3y8JnItI/YHxLEfmXu/zXiMhPoswvT0SmisgO93UfKSJZ7riTROQ/IrJLREpE5HKvn188RORFEdkoIrtFZJmIXB80PuZlEUtbd93+wm2zQkS+48XzCmaFoxaqepP5qCOwJNxIEfkt8BBwO9AcOAvIA6aLSL2g5peqajbQBzgVuCtofBZwa5THegx4BDgOyAVuAr4F1I/5GZlIsoB1wHk4r+fvgVdFJM8d/yRQjrPsrwaeFpGTI8zvKWAL0AbndT8PGOqu15OBt4CWwBDgRRHpmuDnUxMPAnmq2gy4DHhARE4PGB/PsojYVkQuwnkf/QxoCpwLrEzs0wlDVe0Wxw1YDVwYNOxaYGaoNsCdwHpgD7AUuAB4AagE9gNlwB1u25OAYmAnzgfvZQHzPA34zJ3Pa8ArwANBj3knsBA4iPNmHg6scKf5HLg8qP3tbvu9wHM4K+g7bvv3gBYRlkPIrMB/gArggPvcugZN18wdflXQ8GycD4vB4ZY18DDwdtD44cB2IMcddj3O1g44H2J7gR/G+RqfCnzqLodXgIlByzrkco3wuoZ9Haq5Dt4NPB1wvwVwCGjo9/sjINNC4IdAE5wPv64B414ARkSY9gvgkoD7jwDPAD3d5SoB46YD91cjXz3gz+46dAhQ97YggcugG7Cxal2PZ1nE0hb4H3CdL6+v3ytYut2CP8zcYdcSonC4K846oK07PA/oHGo+7opcAvwO55vw+e4HTTf3/hqcb9b1gCvclSq4cMwHOgCN3GE/AtribFn+GOdDtE1A+9k4xaIdzof2pzgfmg1wCsC9YZZB2Kzu+GLg+jDT9gMOA1khxo0DXgq1rIH2wCLgsRDL+fWqZcGRhSPsY0V4fauW9a/d53klzgdL4LKOtlyD14+w7QPaPAU8FWPGicCNAfcLgcUh2r2FU9hD3d7y8D2Si/PFobu7Pu0PGn8b8GaE6W8CxgON3XVzMXA5cApHF453gX9VI+ND7vrfAedD+j13PTqhpsvQfS334RSiT4Fsd3jMyyJaWyAT5zNgOM57sRQYifve9/rm+QPUtpv7wVAWtALtI3ThOBHnA/lCoF6I+QQWju8Am4CMgGETgPtwNkHXB71hZnJ04fh5lOzzgQEB7a8OGDeJI7/F/gJ4I8x8wmZ1/y8mfOG4BtgUZtwIYHqIZb3HfRPOwN2yCFrOPYFdQGuOLBxHPRbOt7SdOFsF54bIcC6wIWhZ/y9wWcewXC8M1za4fTXXwSXAmQH3f01AwfX5/VEP50P4mcB1JajNDVWvUZh5nATMwyn6CowFxJ33SpxjaPWAi3E+PKfFmbGp+/p3CRh2c6RM1VgOmcC3gXtw3/vxLItobXG+iCgwF2eXXitgFvDnZLzOdoyjen6gqjlVN2BoqEaqWgL8CufDf4uITBSRtmHm2RZYp6qVAcPW4HzjagusV3eNca0LMY8jhonIIBGZLyI7RWQnzgdsq4AmmwP+3x/ifjahRcoazVagVZjjMG2Ar4KG/UBVmwIFON9gWwVPpKqLcb4ZDg8atS34sVT1HPc120boY3yhlvWawAYxLFdq0j4SEakPdMbZ+qrSG6cYJYSIFIuIhrnNjDBdBs7ulHJgmDu4DGf3ZKBmOF8Gws1jGs63/yY4y6kF8JCqHgJ+AHwP54vLb4FXcb5tx+NcYKWqLg8Y1sKdZ0KoaoWqzsTZUr7ZHRzPsojWdr/79wlV3aiqW4FHgUtqmj0WVjg8pqovq+q3cQ4YK84mMu7/gTYAHdw3TpXjcbY0NgLtgs4C6hDq4ar+EZGOwLM4b+Bj3A/LxTjf3GoqUtZoPsI5BnNF4ED3zKj+wAehJlLVD3C+ef5fmPnei/ONLLB4VT3WgBhyVQm1rI8PyBltuR7xunrwOvTAKWz73PkLTlFdENxQRN5xz0gLdXsn3AOoaoGqSpjbt0NN4+aoOk72Q/dDHmAZkCUiXQKa9yb8yRMtcdbtkap6UFW3Ac/jfiCq6kJVPU9Vj1HV7wInAJ+Eey5htAZ2BGW/HOfLR/DzqtYyDJCFU+ghvmURsa2q7sApmMGfI0lhhcND4vye4XwRaYCzz3c/zoFjcL7dnxDQ/GOcfd93iEg9cX4HcinO/uyP3OmGiUiWiAwA+kZ5+CY4K9VXbpaf4XzTTYRIWSNS1V3AH4EnRKSfO30ezgH/rcBLESb/O3CRiPQJMd8SnAPZvwwYttN9rKdE5EoRyRaRDHf6JsHzcH2Es4vkl+6yvoIjl3W05Rr8uib6dTgFOFZEOotII+B+nC8lq4Mbqmp/Vc0Oc+sf3L6GnsbZxXSpqlZ9G0ZV9+JsPfxJRJqIyLdwCvkLoWbifnNeBdzsLv8cYDBuYRSRXiLSUEQai8htOFupY6umF+eU7bFHz/kIi4HTRKSPuwwfxHmNXgmRJ+ZlKCLHuqfHZotIpoh8FyjCOV4Y17KIse3zwC/cx22Bs3fjqOLniWTsD6tNN+I7ON4L59vQHpwzf97imwPlA4C1OPvbb3OHnYzzjXsXR58FlY+zO6IM50P2deD3UXL92X3cqs3YD3CPPQS3B17EPUbh3r8eeC/CcoiUtZgwxzgC2lyH8wY+gPOmLa5aNlGe09PApDDPoYM7v+Kgaa52X4d9OB/gH+Ocylk/TLZ8vjmD7RWOPoMt0nIN9bqGbR8wz1HAqBjWv4eBf+J8I12PcyxqBTDOx/dE1dZ01Zl0Vber3fEtgTdwvmysBX4SNP07wO8C7vdx14cd7jJ7DTjWHfeIO7zMne7EoHnNAG6IIfPdOFvOG3EKT6sELIfW7mu7E9iNszvxhqA2YZdFiOUQbbnVwzkQvxNnN9vjJOnMOnEDmDQjIh/jfNA873eWmhKRn+NsGXxLVdf6nSeVubtH/qGqk/zOkmrc4z8LgF76za4y4wG/fyhmYiQi5+H8DmQrzjfoXsC/fQ2VIKo6RkQOAefgfLMy4Z2C8zsHE0RVy3F2lxmPWeFIH91wziDJxtk1caWqbvQ3UuKoash93uYb7n7sY4Hl0doa4yXbVWWMMSYudlaVMcaYuNSJXVWtWrXSvLw8v2McZe/evTRpEu6s0NSVrrkhfbOna25I3+zpmhsSl33evHlbVbV18PA6UTjy8vKYO3eu3zGOUlxcTEFBgd8x4pauuSF9s6drbkjf7OmaGxKXXUTWhBpuu6qMMcbExdPC4f4yeKk4F1wJ7kcIcTzujl8oIqcFjBsjzkV3FoeZ921u3znV6vPHGGNM9XhWOEQkE+dCJP1x+tcpEpEeQc36A13c2xCcXwVXGYvTLXaoeXcALsLO+TfGmKTzcoujL1CiqivdH+ZM5OjO5gYA49UxG8gRkTYAqvohTjcNofwNp2tlO5fYGGOSzMuD4+04spvvUuDMGNq0w+k/JiQRuQynd9AFEuGS0SIyBGcrhtzcXIqLi+PJnhRlZWUpmSuadM0N6Zs9XXND+mZP19zgfXYvC0eoT/XgLYRY2nzTWKQxTudkF0d7cFUdDYwGyM/P11Q8OyJdz9pI19yQvtnTNTekb/Z0zQ3eZ/dyV1UpR14zoj1Ob5TxtgnUGegELBCR1W77T0XkuBqnNcYYExMvC8ccoIuIdHJ7rRwITAlqMwUY5J5ddRawK1L/S6q6SFWPVdU8Vc3DKTynqWrCrtxljDEmMs8Kh6oexrnq2TSc3jxfVdUlInKTiNzkNpuKcw3hEpyrpH19CVYRmYBzUZ1uIlIqItd5ldUYY0zsPP3luKpOxSkOgcNGBfyvwC1hpi2KYf55NYxoTEIUjivk/cHvh71vTG1ivxw3JkEKxxX6HcGYpLDCYUySWYEx6c4KhzHGmLhY4TAmgWxrwtQFVjiMSbRJkxj20nL405/gs8/8TmNMwlnhMCYRVDl3zldIpULDhmw+piGUlcG11zp/jalFrHAYU1Oq3DxxBYMnryZ732H43vd4rV8HePhhmD+f81/7Pvzyl/DVV34nNSYhrHAYU1MjRnDa5zv45V192JNd74jjHIXjz0czhPGrp0D//rB/v49BjUkMKxzG1NShQwz/TS/2NqkXtsnzV+QxI2uds+VhTJqrE9ccN8YTZWWwZAmFHd8HGhw1+ogzrET467VduYDTQZcnL6MxHrAtDmOq6/77+fdvg69NFt7+Rllw442csWg7lJd7GMwYb1nhMKY6SkvZ9eSjPHflCfFNJ8KPppfCM894k8uYJLDCYUw1vHLD2bx9Xhu2tjh6F1UkheMKGfXjzvDAA7Brl0fpjPGWFQ5jquG5KzrxwmUdqzXtyg7ZzhlWzz2X4FTGJIcdHDcmXo8+SrsDB1jdrkn15/HEE9CkBtMb4yPb4jAmHlu3wv33s7Np+FNvY1H4+mXcfuep8PrrCQpmTPJY4TAmHs88A1dcwc5m9Ws8q8OZAnffDZWVCQhmTPJY4TAmDnMn/BVuvTUh85rfPYfl+0ph+vSEzM+YZLHCYUwcbr+tF4WfJaZwIMKbhW3hww8TMz9jksQKhzGx+tWvyNuwL6GzfLOgDfzlLwmdpzFes8JhTCw2bYKxY9l0THy/24hKhHuH9YQXXkjsfI3xkKeFQ0T6ichSESkRkeEhxouIPO6OXygipwWMGyMiW0RkcdA0j4jIl277f4lIjpfPwRgAXnoJLr+cAw0Tfwb7niZZ8H//B6oJn7cxXvCscIhIJvAk0B/oARSJSI+gZv2BLu5tCPB0wLixQL8Qs34X6KmqvYBlwF2JTW5MCMuWORdl8sD87jlsWr8UFizwZP7GJJqXWxx9gRJVXamq5cBEILhHuAHAeHXMBnJEpA2Aqn4IbA+eqapOV9XD7t3ZQHvPnoExVZ55hsLV93kya80Qpn/rOB4aOdCT+RuTaF7+crwdsC7gfilwZgxt2gEbY3yMnwOvhBohIkNwtmLIzc2luLg4xlkmT1lZWUrmiiZdc0P1srebNImVjSso6lvkTSjgwCBoAWGz1bVlngrSNTd4n93LwiEhhgXvxI2lTeiZi9wNHAZeCjVeVUcDowHy8/O1oKAgltkmVXFxMamYK5p0zQ3VyK4KN9/M337UmC/KvvAsF8Dgf62i4I+ToXfvo8bVqWWeItI1N3if3ctdVaVAh4D77YEN1WhzFBEZDHwfuFrVjigaDy1ZAnv38sUJTT1/qHqHFSZM8PxxjKkpLwvHHKCLiHQSkfrAQGBKUJspwCD37KqzgF2qGnE3lYj0A+4ELlPVxJ5Ub0ywxYth0CCQUBvHiVXctzW8+qqdXWVSnmeFwz2APQyYBnwBvKqqS0TkJhG5yW02FVgJlADPAkOrpheRCcBHQDcRKRWR69xRI4GmwLsiMl9ERnn1HIwpPPiMc+2MJCg5PhtatIDS0qQ8njHV5Wm36qo6Fac4BA4bFfC/AreEmTbkkUhVPTGRGY0Ja+VKhr20nEIKo7dNBBEKf9GU9zt0iN7WGB/ZL8eNCWfKFBoerEjqQ7bYfQjuuCOpj2lMvKxwGBPOm2/yUZ9WSX3I3U2y4NlnYWOsZ6Qbk3xWOIwJpbyc0iUfMe/kFkl92IqsDOjXD956K6mPa0w8rHAYE0r9+vx0RF8ONMhM/mNfdhnMmpX8xzUmRlY4jAnl0UfJW7/Xn8e+6ip4/nl/HtuYGFjhMCZYZSWMGMF+P7Y2gMIXL+RPQ0+GefN8eXxjorHCYUywBQugRQs2t27kW4QOm/bDxIm+Pb4xkVjhMCbYRx/BxRf7GmFuzxZ2LXKTsqxwGBNs6FDnwko++rJTU1i3DrZs8TWHMaFY4TAm0L598PDD0CDBl4iNU2VmBpc+0AOOPdbXHMaEYoXDmEAffghvvUXhuCR1MxJB/UOVMGmS3zGMOYoVDmMCTZ/Oc63XRW+XBPUOVTq7zSor/Y5izBGscBgT6N13mduzpd8pAJyzunJyYOFCv6MYcwQrHMYEeu89luV5f9GmWL1+/F6YNs3vGMYcwQqHMVXmzoVNm6jM8P6iTbF6+XvHww03+B3DmCNY4TCmymOPwezZfqc4wrYWDWDOHKS83O8oxnzNCocx4FyutbiYn+4c43eSo919N82+/NLvFMZ8zQqHMQCrVkFFBaW5/nUzEs7E1lvI+ewzv2MY8zUrHMYAHHccvP02SOoc36gy/6Qc6s/9yO8YxnzN02uOG5M2vvwSunWDFDzzdUG35sxvmk87v4MY47ItDmMALr+cQU+c73eKkA40zGJrXgf46iu/oxgDeFw4RKSfiCwVkRIRGR5ivIjI4+74hSJyWsC4MSKyRUQWB03TUkTeFZHl7t/kXtvT1D6rV0N5OeuOS73jG1V6T54GI0f6HcMYwMPCISKZwJNAf6AHUCQiPYKa9Qe6uLchwNMB48YC/ULMejgwQ1W7ADPc+8ZUX3ExFBSk5PGNKht6dnNyGpMCvNzi6AuUqOpKVS0HJgIDgtoMAMarYzaQIyJtAFT1Q2B7iPkOAMa5/48DfuBJelN35OfDb37jd4qINnU/EebN47vPnut3FGM8PTjeDgjsLa4UODOGNu2AjRHmm6uqGwFUdaOIhOx3WkSG4GzFkJubS3EKflsrKytLyVzRpGtuCJ293vbtLNGtFGUX+RMqBs0yW/LBT69gYIPT027Zp+v6kq65wfvsXhaOUNv9Wo021aKqo4HRAPn5+VpQUJCI2SZUcXExqZgrmnTNDSGyr14NAwdyz4juKb2rqii7iCfPWkf9g6uZVnCr33Hikq7rS7rmBu+ze7mrqhToEHC/PbChGm2Cba7aneX+tUukmer74APe71iZ0kWjymmf72DEo4v8jmGMp4VjDtBFRDqJSH1gIDAlqM0UYJB7dtVZwK6q3VARTAEGu/8PBiYnMrSpYz74gAXdcvxOEZOleU3pvmo3WL9VxmeeFQ5VPQwMA6YBXwCvquoSEblJRG5ym00FVgIlwLPA0KrpRWQC8BHQTURKReQ6d9QI4CIRWQ5c5N43Jm6F4wrh7LP55JTUuP5GNHsbZ1Ga2xjmzfM7iqnjPP3luKpOxSkOgcNGBfyvwC1hpg15pFJVtwEXJDCmqaOkUuH669k4/mW/o8Rs8vltuS3Dfrdr/GVroKmzLpi9BW680e8YcXm7oC2cGXxyojHJZYXD1Fk9l+9iZNl//I4Rl0b7D8Mpp9h1yI2vrHCYOuuU5btY1LW53zHisr9RFuu3lMDnn/sdxdRhVjhM3VRRweq2jVnRIdvvJHFb2DUHZs70O4apw6xwmLopM5P7h55MRVb6vQU+6dUSDh50zgozxgfp964xJhEeeYTzZ2/2O0W1FPc9lsKcN/yOYeowKxymbnrrLXZn1/M7RbX9dsxSjt12wO8Ypo6ywmHqnvJymDePzzs38ztJtTXdd4heS3f6HcPUUVY4TN2zdi2cey77GqXvlZMXdc2h17JdfscwdZQVDlP3nHgiTJ0avV0KW9i1Oe027/c7hqmjrHCYuucvf2Hwg339TlEjyztm89s7+9iZVcYXVjhM3aLKzhH3sa9hpt9JakaEwo+3cFKJ7a4yyZe+O3mNqYZG69ZxoH4GW1s29DtKjbXftI+uq/f4HcPUQTFtcYjI90XEtk5M2mu6fDmLu6RXNyPhLO7SnJ7Ld9nuKpN0sRaDgcByEXlYRE7yMpAxXtpywQU8dH13v2MkxBedm9F5bRn1yiv8jmLqmJgKh6peA5wKrACeF5GPRGSIiDT1NJ0xCdZu0iQyKxJyWXvfHWiQyY8fPZtD9dP8eI1JOzHvflLV3cAkYCLQBrgc+FREfuFRNmMSa8sW8saOpbxe7dnrmrO7nFPsh4AmyWI9xnGZiPwL+A9QD+irqv2B3sBtHuYzJnFmzWJ3jx5ohvidJGE6rd/HT95e63cMU8fE+tXrSuBvqtpLVR9R1S0AqroP+Lln6YxJpJkz2XXKKX6nSKhFXZpxcsluu7CTSapYC8dGVf0wcICIPASgqjMSnsoYL9x1FxsvvdTvFAm1I6cBu5rWsws7maSKtXBcFGJY/0QGMcZT+/fD/Pkcal47TsUNdN/QHpCX53cMU4dELBwicrOILAK6i8jCgNsqYGFyIhqTAJ98Avfcw7Jty/xOknDrj2vM9Y8W+B3D1CHRtjheBi4FJrt/q26nu6foRiQi/URkqYiUiMjwEONFRB53xy8UkdOiTSsifURktojMF5G5IpLenQ6Z5Jg5E771Lb9TeKL1tgPc/8QSv2OYOiRa4VBVXQ3cAuwJuCEiLSNNKCKZwJM4u7R6AEUi0iOoWX+gi3sbAjwdw7QPA39U1T7AH9z7xkQ2axZ8+9t+p/DEujaNaXzgMJSW+h3F1BGxbHEAzAPmun/nBdyPpC9QoqorVbUc5/cfA4LaDADGq2M2kCMibaJMq0DVFXiaAxui5DAGbrsNCgr8TuENERZ1ae4UR2OSQFS9+RWtiFwJ9FPV6937PwXOVNVhAW3eAkao6kz3/gzgTiAv3LRulyfTAMEpfOeo6poQjz8EZyuG3Nzc0ydOnOjJ86yJsrIysrOz/Y4Rt3TLnVlWRkZFBUsOb6ZlZku2V2z3O1LcouXO/aKEg9mNObZPQfJCxSjd1pcq6ZobEpe9sLBwnqrmBw+P2Dtu4DGHUFT100iTh5okxjaRpr0Z+LWqThKRq4DngAtDZBsNjAbIz8/XghT8tllcXEwq5oom7XI//TTMmcM9562iKLuICWUT/E4Ut6i5O0D23kO8WXBf0jLFKu3WF1e65gbvs0frVv2vEcYpcH6E8aVAh4D77Tl6t1K4NvUjTDsYuNX9/zXgHxEyGOMcGD//fGCV30k8k3m4kld+Oxuu2Q3N0vda6iY9RCwcqlqT/prnAF1EpBOwHqeH3Z8EtZkCDBORicCZwC5V3SgiX0WYdgNwHlCMU7iW1yCjqQtmzWJQz+XQtonfSTxTkZXB8uOz6T17Nlx8sd9xTC0XbVfV+ar6HxG5ItR4VX093LSqelhEhuEcj8gExqjqEhG5yR0/CpgKXAKUAPuAn0Wa1p31DcBjIpIFHMA9jmFMSIcPw5AhrGsz3e8knlvctTm9Z860wmE8F21X1Xk4HRuG6qdBgbCFA0BVp+IUh8BhowL+V5xTfWOa1h0+Ezg9Sm5jvvG738G4d/1O4bmZp7Xi6q7d/I5h6oBou6rudf/+LDlxjEmwX/8aevSAxn4H8d6XJzSDq68GVZDa0wOwST2xdqt+jPsL709FZJ6IPCYix3gdzpiaWj5lLPTp43eM5OnfH+ZG+4mVMTUTayeHE4GvgB/idLH+FfCKV6GMSYg9e2i/eR+cFvGs8tolL885i8wYD8VaOFqq6v2qusq9PQDkeBnMmBrbvp03LmhH4cR+fidJnm9/235BbjwXa+F4X0QGikiGe7sKeNvLYMbUWMeOjL6qs98pkurHm0dCLew63qSWaN2q7xGR3cCNOP1Wlbu3icCvvY9nTA1ccw0nrtnjd4qk2nJMQ3juOb9jmFouYuFQ1aaq2sz9m6GqWe4tQ1Xt56kmdR06BJMns6lVQ7+TJN3Iq7vApEl+xzC1WKy7qhCRFiLSV0TOrbp5GcyYmrjxz2dBx46UNannd5Skq8gA/v1vv2OYWizaDwABEJHrcfqHag/MB84CPiJyX1XG+OaYneVw6aXAbL+jJN2iLs1hgp1ZZbwT6xbHrcAZwBq3/6pTcU7JNSYlfXRqK3jwQb9j+GJVh2znR4B79/odxdRSsRaOA6p6AEBEGqjql4D1bWBSkyp3/OPLOvvBWZkhFN7VBprU3k4djb9iLRylIpIDvAG8KyKTsSvvmVS1ahX5i7dT+Nr3/E7imx4lu2D8eL9jmFoqpsKhqper6k5VvQ/4Pc7Fk37gZTBjqm3mTJZ0aV6n+2tqUF4Jo0ZFb2hMNcRzVtVpIvJLoBdQ6l4L3JjUs3Ahi7rW7R/BfdG5GSxYAPv3+x3F1EKxdnL4B2AccAzQCnheRO7xMpgx1VXYcy5vnN/W7xi+OtAgky9yM+HTSFd3NqZ6Yt3iKALOUNV73a7WzwKu9i6WMdW0YwdXvLeeysyYN6Zrrd/e0Ru+9S2/Y5haKNZ312og8Ce4DYAVCU9jTE3NmsXZ87f5nSIl1D9UCRMn+h3D1ELR+qp6QkQeBw4CS0RkrIg8DywGypIR0Ji4/Pe/zg/gDBkK3HwzVFb6HcXUMtF+OV51RZh5wL8Chhd7ksaYmvrwQxacZ4UDYEfz+qxrcJAOn38OPXv6HcfUItEuHTuu6n8RqQ90de8uVdVDXgYzplomTWLJO0V+p0gZi7s0p8PMmVY4TELFelZVAbAceBJ4ClhmnRyalLN8OWzcyOEsOzBeZdwP8hhw+AW/Y5haJtZ32F+Bi1X1PFU9F/gu8LdoE4lIPxFZKiIlIjI8xHhxr2VeIiILReS0WKYVkV+445aIyMMxPgdT240dC2+84XeKlLKlZQM6ldbNrleMd2ItHPVUdWnVHVVdBkTsr1pEMnG2UPoDPYAiEekR1Kw/0MW9DQGejjatiBQCA4Beqnoy8H8xPgdT233wAbcdmOJ3ipSiAvc99TmUlvodxdQisRaOeSLynIgUuLdncQ6YR9IXKFHVle6vzBTygmMAABg4SURBVCfifOAHGgCMV8dsIEdE2kSZ9mZghKoeBFDVLTE+B1Ob7d8P8+c7XY2Yb4iw+MRmMNO6WTeJE9P1OICbgFuAXwICfIhzrCOSdsC6gPulwJkxtGkXZdquwHdE5M/AAeA2VZ0T/OAiMgRnK4bc3FyKi4ujxE2+srKylMwVTSrmlvJymj/wAJcf0zhiu5aZLSnKTr+D5zXJndmrFUvGjuKr445LcKrYpOL6Eot0zQ3eZ49aOEQkA5inqj2BR+OYd6ge5jTGNpGmzQJa4Px6/QzgVRE5QVWPmLeqjgZGA+Tn52tBQUHsyZOkuLiYVMwVTUrmLinhkpX3s78s8ipdlF3EhLIJSQqVODXJPaPXAY5re4DHfHrNUnJ9iUG65gbvs0fdVaWqlcACETk+znmXAh0C7rfn6K7Yw7WJNG0p8Lq7e+sToBKn/yxTl91wA72W7fI7RUrackxDVrVrUmevT2ISL9ZjHG1wfjk+Q0SmVN2iTDMH6CIindzfgAwEgqeZAgxyz646C9ilqhujTPsG7iVrRaQrUB/YGuPzMLXRwYMwZ06d7xE3ktvHLIUpduKASYxYj3H8Md4Zq+phERkGTAMygTGqukREbnLHjwKmApcAJcA+4GeRpnVnPQYYIyKLgXJgcPBuKlPHfPIJdO/Ovkaxrs51z4LuOex65rd8vyj9ju+Y1BPxnSYiDXEOjJ8ILAKeU9XDsc5cVafiFIfAYaMC/lecg+4xTesOLweuiTWDqQPatGH4uXZ5mEjmd8/hBzPW+x3D1BLRdlWNA/JxikZ/nB8CGpNacnP5uPcxfqdIaSvbN2H6OblQUeF3FFMLRCscPVT1GlV9BrgS+E4SMhkTu0OHoH17muyLeUO4TtIM4YUBeWB7dU0CRCscX3dkGM8uKmOSZs4cSppXsLexHd+I5uJZm+Cmm/yOYWqBaO+23iKy2/1fgEbufcE5RNHM03TGRDNjBp/2aOF3irSwLK8p/OM/fscwtUC0btUzkxXEmGrp3Jni8tZ+p0gLq9s2Zvu2UlquWQMdO/odx6Qx63/apC9V+MlP+OJE+/1GTESYdHF72L07eltjIrDCYdLX9OkU9z3W7xRp5eXvd6Tw01/6HcOkOSscJn3NmMHqtpE7NTRHarGrnCce+NTOrjI1YoXDpC87MB63Hc3qcdzWA7Bypd9RTBqzwmHSU0UFnHACX3S2E/viIsK8k1vAu+/6ncSkMSscJj1lZsJrr9n1xath1qmteOb9R/yOYdKY/WrKpKc//IG79rwBfayrkXj9N985fflGn3OY9GVf10x6euUVvmpR3+8Uaev611Zy872n+x3DpCkrHCb9rFoFO3eyskO230nSVoYqZy7Y5ncMk6ascJj0s2IFFBWhGaGuMGxiMbdnS/KX7PA7hklTVjhM+rnwQvj73/1OkdYWndiMZmWHnKsnGhMnKxwmvZSXw1VX2XUlauhQ/UwGjzgTGjTwO4pJQ1Y4THqZNQvWrKHwxQv9TpL2uqzeAyNH+h3DpCErHCa9vPMO9Ovnd4pa4VC9DDb+8XbrfsTEzQqHSS+LF1vhSJDVbRuTWamwdKnfUUyascJh0svbb1O49C6/U9QOInzU+xj48EO/k5g042nhEJF+IrJUREpEZHiI8SIij7vjF4rIaXFMe5uIqIi08vI5mBTyzjswZQqInYabKCOvPhGGDPE7hkkznhUOEckEngT6Az2AIhHpEdSsP9DFvQ0Bno5lWhHpAFwErPUqv0lBY8bA9u1+p6hVDmdlwEMP2XI1cfFyi6MvUKKqK1W1HJgIDAhqMwAYr47ZQI6ItIlh2r8BdwB2VK+uOHjQ6dH1kkv8TlLr/G/CIzBtmt8xTBrxspPDdsC6gPulwJkxtGkXaVoRuQxYr6oLJMIuCxEZgrMVQ25uLsXFxdV6El4qKytLyVzR+JG7+aJFnNChA6/MnExRdlG159Mys2WNpveLl7krzmzH8qf+zvo2bTyZv63nyed1di8LR6hP9eAthHBtQg4XkcbA3cDF0R5cVUcDowHy8/O1oKAg2iRJV1xcTCrmisaX3AUFcMMNTHilf41mU5RdxISyCYnJlERe5m7V/QDPTPyMLueeCxmJ3wlh63nyeZ3dy11VpUCHgPvtgQ0xtgk3vDPQCVggIqvd4Z+KyHEJTW5Siyr88Y9c+NJ3/U5SK21t2ZBBD/b1pGiY2snLNWUO0EVEOolIfWAgMCWozRRgkHt21VnALlXdGG5aVV2kqseqap6q5uEUmNNUdZOHz8P4bd48mDCBCrtok2dy9hyCN9/0O4ZJE57tqlLVwyIyDJgGZAJjVHWJiNzkjh8FTAUuAUqAfcDPIk3rVVaT4iZPZkLnfX6nqNWy9x1m7ZAfc/yGvXa6s4nK0ysAqupUnOIQOGxUwP8K3BLrtCHa5NU8pUl5U6fyv/72cx0vLe3UlAbllfD553DyyX7HMSnOtv1N6vvPf/i8czO/U9RuInyY3xpef93vJCYNWOEwqe2997j+sQuotIs2ee7FyzrSL/ddv2OYNGCFw6S2e++l9Y5yv1PUCbuz69F76U5Yax0ymMiscJjUtW4dLF3K3J4t/E5SZ5y1YBu89JLfMUyKs8JhUtekSXD55U5/SiYpPsxvDf/8p98xTIqzd6RJXUOHwogRfqeoUxZ2bc6O5Ytg5Uq/o5gUZoXDpKa1a+Gttyh860q/k9QplZkZDPv9adCxo99RTAqzwmFS04svwnvv+Z2iTtrSsgF/HnqK3zFMCrPCYVLTK69wa85HfqeokyoyhOv/uRIWLvQ7iklRVjhM6lm2DLZuZVGX5n4nqZM0Q3jvnFx44QW/o5gUZYXDpJ4uXfjRbcej9qM/30w/J5et/3gcKir8jmJSkBUOk1oOH4aRI9mW08DvJHXa2rZNuPV3p0Jmpt9RTAqywmFSy/Tp8OKLtrWRAsoaZXLvsJ5+xzApyAqHSS1jxvDXHjv9TmGADIXbx3wJZWV+RzEpxgqHSR27dsGMGbzf91i/kxhgZ7P6LOyaA6+95ncUk2KscJjU0bw5Ax44mb2NPb1MjInDm4VtWfyXX/kdw6QYe4ea1KAKjzzCwRz7LpNKPunVkmV52WwfV8j7g9/3O45JEfYuNanh449h9GjK69kqmUoqM4SMSuW7/93odxSTQuxdalLDE0/w1BnY2VQp6HBWBrdMWAE7dvgdxaQIKxzGf2VlbJ36T6Z+5zi/k5gQdjarz8e9WjLyljP8jmJShBUO47/sbH7y8JnsbVLP7yQmjMkXtOOC2Vv8jmFShKeFQ0T6ichSESkRkeEhxouIPO6OXygip0WbVkQeEZEv3fb/EpEcL5+D8djBgzB0qF1TPMUtPrEZvx7eh8JxhX5HMSnAs8IhIpnAk0B/oAdQJCI9gpr1B7q4tyHA0zFM+y7QU1V7AcuAu7x6DiYJXnwRVq6kwq7yl9pEaLz/MDe8usLvJCYFePlu7QuUqOpKVS0HJgIDgtoMAMarYzaQIyJtIk2rqtNV9bA7/WygvYfPwXipogIefhjuvNPvJCYGu7Prcf7HW+CTT/yOYnzmZeFoB6wLuF/qDoulTSzTAvwceKfGSY0/li+Hzp0pXPNHv5OYGFRkZfBavw7wyCN+RzE+8/IHgKF2WmuMbaJOKyJ3A4eBl0I+uMgQnN1f5ObmUlxcHCVu8pWVlaVkrmgSklsVROCOOyjatiwhuWLRMrMlRdlFSXu8REmV3Fn9D7Llvkf58p13qGzUKKZp6vR67hOvs3tZOEqBDgH32wMbYmxTP9K0IjIY+D5wgaoGFyMAVHU0MBogPz9fCwoKqvUkvFRcXEwq5oomIbnffBOmT6cwf3FCMsWqKLuICWUTkvqYiZBKuV+4uyPv9+vnFP4Y1On13CdeZ/dyV9UcoIuIdBKR+sBAYEpQmynAIPfsqrOAXaq6MdK0ItIPuBO4TFX3eZjfeKWyEn7/e7jgAr+TmOoQYXafVlz3gP2uo67yrHC4B7CHAdOAL4BXVXWJiNwkIje5zaYCK4ES4FlgaKRp3WlGAk2Bd0VkvoiM8uo5GI+8/jpkZVG48+9+JzHV9GmPFlz7r9V+xzA+8bSTQ1WdilMcAoeNCvhfgVtindYdfmKCY5pkE3HOplp3v99JTDVNPr8tP5xeCjNnwre/7Xcck2R28rxJrpISuOIKCq1opLXy+pn8fVAX51K/ps6xwmGSZ9s2OPtsrn7kLL+TmASY3acVF664F5Yu9TuKSTIrHCZ57rkHrrqKDbmN/U5iEqTrmjK+OruXc/VGU2dY4TDJMX8+TJ7MpScv8DuJSaAvOjfjk1NaOl8KTJ1hhcMkR+/eDPzN8ZRZD7i1zjNXdWbbC6OdngBMnWCFw3jvr3+F4mI2t47tl8YmvezJrscNf8qn8H9D/I5iksQKh/HWrFls/9Nd/HDh3X4nMR7a0bw+ZyzazsRLjvc7ikkCKxzGO7t2wTXX8NefdWN7TgO/0xiPfXFCUwo/3gLvWL+jtZ0VDuOdvXt56qxM/ndqK7+TmCQoa1KPP994EtsHDoBVq/yOYzxkhcN4Y+JEaNTI6Ybb1BmLuuXw6OCuDJj6U7+jGA9Z4TCJ9+qrcPvt/HD89/xOYnww6/TWoMrL3zveuViXqXWscJjEeu89GDaM6248zo5r1GH7GmXRbfUeJl98vHPtFVOrWOEwibV2Lb+6rh0rO2T7ncT46HBWBn/4RU9OWrGb5k887Hcck2BWOExivP8+TJhAYeYLLOie43cakwL2Ncrittt7s6n7iRSOK7Qtj1rECoepuQkT4Kqr+NWCh/xOYlLMnux6rDv9FM6d8xVcfTUcOOB3JJMAVjhMzYwfD8OHw4wZLDiphd9pTIqa3bslxcvfc676uH6933FMDVnhMNWzYQNs2gQXX8wVv2lH4We3+p3IpLDy+pn8aWgPxuRugNtv9zuOqSErHCY+FRW0nTwZevfmz/eeT+G0InbY2VMmBpohvDAgj/Mv2sAVj50DQ4bAli1+xzLVYIXDxE4V+vXj2Bkz+Nmv8njvnFy/E5k0pBnC/kZZTFz3b3ac2B5GjoSDB/2OZeJghcNEpgrFxXD33SDC4PN38ervh7K6vZ1ua6rvQINMnhnYmdtv782sMfcxcGSBc4XI/fv9jmZiYIXDhDdhAvTuzbqBl/DQ5kkUji1gbdsmkGGrjUmMFcdnc8+tp7C5VUOeuPUsdrVuxusXtocFC+z03RSW5XcAkyJUYdkyp2fTTz+F8eO5c+6DHLwkg4Xd+qIZ4ndCU8u9flF7Zp3ain7/3cia/ucw5I+nM+2Mx51els88Exo29DuicXlaOESkH/AYkAn8Q1VHBI0Xd/wlwD7gWlX9NNK0ItISeAXIA1YDV6nqDi+fR61TWen0XrpkibNv+Uc/cs6xnzmTtzsd4qM+rZg1tgB6HeN3UlPHbG7VkHGXd2Lc5Z0A+MNz1/CTt9fScf1eGp1xNrz2GmRlwbp1cNJJ0MguDuYHzwqHiGQCTwIXAaXAHBGZoqqfBzTrD3Rxb2cCTwNnRpl2ODBDVUeIyHD3/p1ePY+0UV4OO3fCjh3O327dnOETJzrnzZeWQmEhDBoEJ53Epm1rWN2uCfO75/DKvqdoUFDBwe929vc5GBPkv/mt+W9+axoeOMzJJbv57J2BnLJ8N794cTntN++nwXHt4JlnnHX7nnugfXs45hjo0QNOP935glSvHjRr5myx1K/v91OqFbzc4ugLlKjqSgARmQgMAAILxwBgvKoqMFtEckSkDc7WRLhpBwAF7vTjgGK8LBxDh8KaNc6unOOOgzFjnNsrr3yzD3b0aCgrg1tv/WbYDTdAURF8//uwe7cz7JRT4Mkn4f774Z13OHXXLmeFnjYN5syBu+5y2qnCfffBhRfCOec498vLnR9PPfYYXHklzJjhbC1kZDiP/eyz7Lzr1+xpnEVZ4yz+PqgLW45pyM9fX8XWnAZsbdGAL9YuZtW456l3Wy6H6rc94mkebJDp2SI0pqYONMxiXs+WACzonsP1D5xB5uFKWu04SNnqv1Ax/kF+sG49rRccpOneQ3zeuTlvXNiOe0cu4eSSXTQ5UEFZ4yyO3XYAHnwQHn7YKSQNG8KUKZCZCQMHOu+nzEznVOFu3eCyy2D7dmd49+7Oe/2xx2Dq1G/CvfwylJTAvfd+M+wXv4DvfQ/69ftmWN++8Kc/Oe/tjz/+ZvjUqfDvf8MTT3wz7A9/gK5d4acB3dP36+fM91e/ch4PoEULeOEFJ8PLL3/dtNHAgbBiBfz3v3DttTV/AYKIenQASkSuBPqp6vXu/Z8CZ6rqsIA2bwEjVHWme38GThHICzetiOxU1ZyAeexQ1aN+siwiQ4CqiyB3A5Z68DRrqhWw1e8Q1ZCuuSF9s6drbkjf7OmaGxKXvaOqtg4e6OUWR6ijqcFVKlybWKaNSFVHA6PjmSbZRGSuqub7nSNe6Zob0jd7uuaG9M2errnB++xenldZCgRe/q09sCHGNpGm3ezuzsL9az89NcaYJPKycMwBuohIJxGpDwwEpgS1mQIMEsdZwC5V3Rhl2inAYPf/wcBkD5+DMcaYIJ7tqlLVwyIyDJiGc0rtGFVdIiI3ueNHAVNxTsUtwTkd92eRpnVnPQJ4VUSuA9YCP/LqOSRBSu9KiyBdc0P6Zk/X3JC+2dM1N3ic3bOD48YYY2on6zvCGGNMXKxwGGOMiYsVjiQQkR+JyBIRqRSR/IDheSKyX0Tmu7dRAeNOF5FFIlIiIo+73bOkRG533F1utqUi8t1Uyh1MRO4TkfUBy/mSgHEhn0cqEZF+br4St7eElCUiq93Xf76IzHWHtRSRd0Vkufs3JS4VKSJjRGSLiCwOGBY2a6qsK2FyJ3cdV1W7eXwDTsL5EWIxkB8wPA9YHGaaT4CzcX7T8g7QP4Vy9wAWAA2ATsAKIDNVcod4HvcBt4UYHvZ5pMoN5+SQFcAJQH03bw+/c0XIuxpoFTTsYWC4+/9w4CG/c7pZzgVOC3wPhsuaSutKmNxJXcdtiyMJVPULVY35l+vu71OaqepH6rz644EfeBYwjAi5BwATVfWgqq7COSuub6rkjkPI5+FzpmBfd92jquVAVfc76WQATvdAuH9TYp1Q1Q+B7UGDw2VNmXUlTO5wPMlthcN/nUTkMxH5QES+4w5rh/MjyCql7rBU0Q5YF3C/Kl8q5x4mIgvdzfyq3Q/hnkcqSYeMgRSYLiLz3G5/AHLV+X0W7t9jfUsXXbis6fA6JG0dt+txJIiIvAccF2LU3aoa7keKG4HjVXWbiJwOvCEiJ5OALldiVc3cnnUVU12RngdOr8v3u1nuB/4K/Bwf88YhHTIG+paqbhCRY4F3ReRLvwMlSKq/Dkldx61wJIiqXliNaQ4CB93/54nICqArzreC9gFNQ3XXkhDVyU3krmKSkjtYrM9DRJ4F3nLvxtItjt/SIePXVHWD+3eLiPwLZ7fIZhFpo6ob06CboHBZU/p1UNXNVf8nYx23XVU+EpHW4lx7BBE5Aee6JCvdTeQ9InKWe1bSIFKra5UpwEARaSAinXByf5Kqud0PgCqXA1Vno4R8HsnOF0UsXfekBBFpIiJNq/4HLsZZ1unUTVC4rCm9riR9HffzrIa6cnNfyFKcrYvNwDR3+A+BJThnPXwKXBowTb774q8ARuL+yj8Vcrvj7nazLSXgzKlUyB3iebwALAIWum+kNtGeRyrdcLrlWebmvNvvPBFynuCuywvc9fpud/gxwAxgufu3pd9Z3VwTcHYXH3LX8+siZU2VdSVM7qSu49bliDHGmLjYripjjDFxscJhjDEmLlY4jDHGxMUKhzHGmLhY4TDGGBMXKxzGeEREKtyeSheIyKcico47vK2I/NPvfMZUl52Oa4xHRKRMVbPd/78L/E5Vz/M5ljE1ZlscxiRHM2AHfH0dlsXu/9eKyOsi8m/3GhAPu8MzRWSsiCx2r2/xax+zG3ME66vKGO80EpH5QEOgDXB+mHZ9gFNxfqG/VESewOmVtZ2q9gQQkZwk5DUmJrbFYYx39qtqH1XtDvQDxoe5IuIMVd2lqgeAz4GOwErgBBF5QkT6AbuTF9uYyKxwGJMEqvoR0ApoHWL0wYD/K4AsVd0B9Ma5+uItwD+8zmhMrGxXlTFJICLdcS4Duw1oHEP7VkC5qk5yu9sf621CY2JnhcMY71Qd4wDngjqDVbUi9N6qo7QDnheRqr0Cd3kR0JjqsNNxjTHGxMWOcRhjjImLFQ5jjDFxscJhjDEmLlY4jDHGxMUKhzHGmLhY4TDGGBMXKxzGGGPi8v9qu8N6apC1bAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(10000000,)\n"
     ]
    }
   ],
   "source": [
    "alldata = []\n",
    "filenames = ['./170606_R_Amp_Q--ch1.dat']\n",
    "\n",
    "for filename in filenames:\n",
    "    data = np.fromfile(filename, dtype='>i2') #datatype is big-endian 2byte signed integer, as compared to '<u4'\n",
    "    data = data[2:]   # exclude first two header values \n",
    "    data = data >> 3  # shift right by 3 bits  \n",
    "    alldata.append(data)\n",
    "\n",
    "mu, sigma = np.mean(data), np.std(data)\n",
    "print(mu, sigma)\n",
    "# the histogram of the data\n",
    "n, bins, patches = plt.hist(data, list(range(np.min(data),np.max(data)+2)), density=1, facecolor='green', alpha=0.75)\n",
    "print(max(n))\n",
    "\n",
    "# add a 'best fit' line\n",
    "y = scs.norm.pdf( bins, mu, sigma)\n",
    "l = plt.plot(bins, y, 'r--', linewidth=1)\n",
    "\n",
    "plt.xlabel('Bins')\n",
    "plt.ylabel('Probability')\n",
    "plt.title(r'$\\mathrm{Histogram\\ of\\ QRNG\\ data:}\\ \\mu=%.2f,\\ \\sigma=%.2f$' % (mu,sigma))\n",
    "plt.grid(True)\n",
    "\n",
    "plt.show()\n",
    "\n",
    "del data  #delete data\n",
    "alldata = np.concatenate(alldata)    \n",
    "print(alldata.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5000000 1000000 1000000\n"
     ]
    }
   ],
   "source": [
    "text = alldata[:5000000]\n",
    "test1 = alldata[5000000:6000000]\n",
    "test2 = alldata[6000000:7000000]\n",
    "test3 = alldata[7000000:8000000]\n",
    "test4 = alldata[8000000:9000000]\n",
    "test5 = alldata[9000000:]\n",
    "text = list(text)\n",
    "test1 = list(test1)\n",
    "test2 = list(test2)\n",
    "test3 = list(test3)\n",
    "test4 = list(test4)\n",
    "test5 = list(test5)\n",
    "text = map(str,text)\n",
    "test1 = map(str,test1)\n",
    "test2 = map(str,test2)\n",
    "test3 = map(str,test3)\n",
    "test4 = map(str,test4)\n",
    "test5 = map(str,test5)\n",
    "text_list = list(text)\n",
    "test1_list = list(test1)\n",
    "test2_list = list(test2)\n",
    "test3_list = list(test3)\n",
    "test4_list = list(test4)\n",
    "test5_list = list(test5)\n",
    "\n",
    "del text\n",
    "del test1\n",
    "del test2\n",
    "del test3\n",
    "del test4\n",
    "del test5\n",
    "\n",
    "print(len(text_list), len(test1_list), len(test5_list))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['-1', '-10', '-100', '-101', '-102', '-103', '-104', '-105', '-106', '-107', '-108', '-109', '-11', '-110', '-111', '-112', '-113', '-114', '-115', '-116', '-117', '-118', '-119', '-12', '-120', '-121', '-122', '-123', '-124', '-125', '-126', '-127', '-128', '-129', '-13', '-130', '-131', '-132', '-133', '-134', '-135', '-136', '-137', '-138', '-139', '-14', '-140', '-141', '-142', '-143', '-144', '-145', '-146', '-147', '-148', '-149', '-15', '-150', '-151', '-152', '-153', '-154', '-155', '-156', '-157', '-158', '-159', '-16', '-160', '-161', '-162', '-163', '-164', '-166', '-169', '-17', '-174', '-18', '-19', '-2', '-20', '-21', '-22', '-23', '-24', '-25', '-26', '-27', '-28', '-29', '-3', '-30', '-31', '-32', '-33', '-34', '-35', '-36', '-37', '-38', '-39', '-4', '-40', '-41', '-42', '-43', '-44', '-45', '-46', '-47', '-48', '-49', '-5', '-50', '-51', '-52', '-53', '-54', '-55', '-56', '-57', '-58', '-59', '-6', '-60', '-61', '-62', '-63', '-64', '-65', '-66', '-67', '-68', '-69', '-7', '-70', '-71', '-72', '-73', '-74', '-75', '-76', '-77', '-78', '-79', '-8', '-80', '-81', '-82', '-83', '-84', '-85', '-86', '-87', '-88', '-89', '-9', '-90', '-91', '-92', '-93', '-94', '-95', '-96', '-97', '-98', '-99', '0', '1', '10', '100', '101', '102', '103', '104', '105', '106', '107', '108', '109', '11', '110', '111', '112', '113', '114', '115', '116', '117', '118', '119', '12', '120', '121', '122', '124', '126', '127', '129', '13', '131', '134', '14', '15', '152', '16', '17', '18', '19', '2', '20', '21', '22', '23', '24', '25', '26', '27', '28', '29', '3', '30', '31', '32', '33', '34', '35', '36', '37', '38', '39', '4', '40', '41', '42', '43', '44', '45', '46', '47', '48', '49', '5', '50', '51', '52', '53', '54', '55', '56', '57', '58', '59', '6', '60', '61', '62', '63', '64', '65', '66', '67', '68', '69', '7', '70', '71', '72', '73', '74', '75', '76', '77', '78', '79', '8', '80', '81', '82', '83', '84', '85', '86', '87', '88', '89', '9', '90', '91', '92', '93', '94', '95', '96', '97', '98', '99']\n",
      "Total words: 297\n"
     ]
    }
   ],
   "source": [
    "# Treating each number as a \"word\". Creating a dictionary.\n",
    "alldata = alldata.astype(np.str)\n",
    "chars = sorted(list(set(alldata)))\n",
    "\n",
    "print(chars)\n",
    "del alldata\n",
    "print('Total words:', len(chars))\n",
    "\n",
    "char_indices = dict((c, i) for i, c in enumerate(chars))\n",
    "indices_char = dict((i, c) for i, c in enumerate(chars))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "nb sequences: 1666634\n",
      "Start vectorization...\n",
      "Done vectorization!\n"
     ]
    }
   ],
   "source": [
    "# Length of input. Treating each input that consists of 100 \"words\" as a \"sentence\".\n",
    "maxlen = 100\n",
    "# Distance between 2 consecutive \"sentences\"\n",
    "step = 3\n",
    "\n",
    "sentences = []\n",
    "next_chars = []\n",
    "for i in range(0, len(text_list) - maxlen, step):\n",
    "    sentences.append(text_list[i: (i + maxlen)])\n",
    "    next_chars.append(text_list[(i + maxlen)])\n",
    "print('nb sequences:', len(sentences))\n",
    "\n",
    "\n",
    "print('Start vectorization...')\n",
    "X = np.zeros((len(sentences), maxlen, len(chars)), dtype=np.bool)\n",
    "y = np.zeros((len(sentences), len(chars)), dtype=np.bool)\n",
    "for i, sentence in enumerate(sentences):\n",
    "    for t, char in enumerate(sentence):        \n",
    "        X[i, t, char_indices[char]] = 1\n",
    "    y[i, char_indices[next_chars[i]]] = 1\n",
    "print('Done vectorization!')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Build model...\n",
      "WARNING:tensorflow:From C:\\Users\\e0031292\\Anaconda3\\lib\\site-packages\\tensorflow\\python\\ops\\init_ops.py:1251: calling VarianceScaling.__init__ (from tensorflow.python.ops.init_ops) with dtype is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Call initializer instance with the dtype argument instead of passing it to the constructor\n",
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv1d (Conv1D)              (None, 100, 64)           171136    \n",
      "_________________________________________________________________\n",
      "max_pooling1d (MaxPooling1D) (None, 50, 64)            0         \n",
      "_________________________________________________________________\n",
      "conv1d_1 (Conv1D)            (None, 50, 128)           24704     \n",
      "_________________________________________________________________\n",
      "max_pooling1d_1 (MaxPooling1 (None, 25, 128)           0         \n",
      "_________________________________________________________________\n",
      "lstm (LSTM)                  (None, 128)               131584    \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 128)               0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 297)               38313     \n",
      "_________________________________________________________________\n",
      "activation (Activation)      (None, 297)               0         \n",
      "=================================================================\n",
      "Total params: 365,737\n",
      "Trainable params: 365,737\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "# build the RCNN model\n",
    "print('Build model...')\n",
    "model = Sequential()\n",
    "model.add(Convolution1D(filters=64, kernel_size=9, padding='same', activation='relu', input_shape=(maxlen, len(chars))))\n",
    "model.add(MaxPooling1D(pool_size=2))\n",
    "model.add(Convolution1D(filters=128, kernel_size=3, padding='same', activation='relu'))\n",
    "model.add(MaxPooling1D(pool_size=2))\n",
    "model.add(LSTM(128, return_sequences=False))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(Dense(len(chars)))\n",
    "model.add(Activation('softmax'))\n",
    "\n",
    "model.compile(loss='categorical_crossentropy', optimizer=tf.keras.optimizers.RMSprop(lr=0.0005), metrics=['accuracy'])\n",
    "print(model.summary())\n",
    "early_stopping = EarlyStopping(monitor='val_loss', patience=4, verbose=1)\n",
    "monitoring = ModelCheckpoint('weights_Q3b_ch1_gpu.hdf5', monitor='val_loss', verbose=1, save_best_only=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 1333307 samples, validate on 333327 samples\n",
      "WARNING:tensorflow:From C:\\Users\\e0031292\\Anaconda3\\lib\\site-packages\\tensorflow\\python\\ops\\math_grad.py:1250: add_dispatch_support.<locals>.wrapper (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.where in 2.0, which has the same broadcast rule as np.where\n",
      "Epoch 1/50\n",
      "1333248/1333307 [============================>.] - ETA: 0s - loss: 4.8300 - acc: 0.0134\n",
      "Epoch 00001: val_loss improved from inf to 4.82817, saving model to weights_Q3b_ch1_gpu.hdf5\n",
      "1333307/1333307 [==============================] - 3357s 3ms/sample - loss: 4.8300 - acc: 0.0134 - val_loss: 4.8282 - val_acc: 0.0137\n",
      "Epoch 2/50\n",
      "1333248/1333307 [============================>.] - ETA: 0s - loss: 4.8250 - acc: 0.0137\n",
      "Epoch 00002: val_loss did not improve from 4.82817\n",
      "1333307/1333307 [==============================] - 3297s 2ms/sample - loss: 4.8250 - acc: 0.0137 - val_loss: 4.8284 - val_acc: 0.0133\n",
      "Epoch 3/50\n",
      "1333248/1333307 [============================>.] - ETA: 0s - loss: 4.8239 - acc: 0.0138\n",
      "Epoch 00003: val_loss improved from 4.82817 to 4.82779, saving model to weights_Q3b_ch1_gpu.hdf5\n",
      "1333307/1333307 [==============================] - 3479s 3ms/sample - loss: 4.8239 - acc: 0.0138 - val_loss: 4.8278 - val_acc: 0.0133\n",
      "Epoch 4/50\n",
      "1333248/1333307 [============================>.] - ETA: 0s - loss: 4.8234 - acc: 0.0138\n",
      "Epoch 00004: val_loss did not improve from 4.82779\n",
      "1333307/1333307 [==============================] - 3475s 3ms/sample - loss: 4.8234 - acc: 0.0138 - val_loss: 4.8296 - val_acc: 0.0136\n",
      "Epoch 5/50\n",
      "1333248/1333307 [============================>.] - ETA: 0s - loss: 4.8268 - acc: 0.0138\n",
      "Epoch 00005: val_loss did not improve from 4.82779\n",
      "1333307/1333307 [==============================] - 3492s 3ms/sample - loss: 4.8268 - acc: 0.0138 - val_loss: 4.8351 - val_acc: 0.0131\n",
      "Epoch 6/50\n",
      "1333248/1333307 [============================>.] - ETA: 0s - loss: 4.8273 - acc: 0.0138\n",
      "Epoch 00006: val_loss did not improve from 4.82779\n",
      "1333307/1333307 [==============================] - 3474s 3ms/sample - loss: 4.8273 - acc: 0.0138 - val_loss: 4.8381 - val_acc: 0.0136\n",
      "Epoch 7/50\n",
      "1333248/1333307 [============================>.] - ETA: 0s - loss: 4.8314 - acc: 0.0138\n",
      "Epoch 00007: val_loss did not improve from 4.82779\n",
      "1333307/1333307 [==============================] - 3548s 3ms/sample - loss: 4.8314 - acc: 0.0138 - val_loss: 4.8508 - val_acc: 0.0136\n",
      "Epoch 00007: early stopping\n"
     ]
    }
   ],
   "source": [
    "model.fit(X, y, epochs=50, batch_size=128, validation_split=0.2, verbose=1, callbacks=[early_stopping,monitoring])\n",
    "model.load_weights('weights_Q3b_ch1_gpu.hdf5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "nb sequences: 399900\n",
      "Vectorization...\n",
      "Processed 0 0\n",
      "Processed 100000 1325\n",
      "Processed 200000 2705\n",
      "Processed 300000 4053\n",
      "5383_399900_0.01346\n",
      "nb sequences: 399900\n",
      "Vectorization...\n",
      "Processed 0 0\n",
      "Processed 100000 1333\n",
      "Processed 200000 2703\n",
      "Processed 300000 4066\n",
      "5437_399900_0.01360\n",
      "nb sequences: 399900\n",
      "Vectorization...\n",
      "Processed 0 0\n",
      "Processed 100000 1368\n",
      "Processed 200000 2729\n",
      "Processed 300000 4145\n",
      "5465_399900_0.01367\n",
      "nb sequences: 399900\n",
      "Vectorization...\n",
      "Processed 0 0\n",
      "Processed 100000 1308\n",
      "Processed 200000 2666\n",
      "Processed 300000 3997\n",
      "5343_399900_0.01336\n",
      "nb sequences: 399900\n",
      "Vectorization...\n",
      "Processed 0 0\n",
      "Processed 100000 1366\n",
      "Processed 200000 2736\n",
      "Processed 300000 4101\n",
      "5453_399900_0.01364\n",
      "nb sequences: 399900\n",
      "Vectorization...\n",
      "Processed 0 0\n",
      "Processed 100000 1425\n",
      "Processed 200000 2804\n",
      "Processed 300000 4142\n",
      "5455_399900_0.01364\n",
      "nb sequences: 399900\n",
      "Vectorization...\n",
      "Processed 0 0\n",
      "Processed 100000 1385\n",
      "Processed 200000 2716\n",
      "Processed 300000 4139\n",
      "5517_399900_0.01380\n",
      "nb sequences: 399900\n",
      "Vectorization...\n",
      "Processed 0 0\n",
      "Processed 100000 1379\n",
      "Processed 200000 2777\n",
      "Processed 300000 4112\n",
      "5434_399900_0.01359\n",
      "nb sequences: 399900\n",
      "Vectorization...\n",
      "Processed 0 0\n",
      "Processed 100000 1375\n",
      "Processed 200000 2732\n",
      "Processed 300000 4102\n",
      "5473_399900_0.01369\n",
      "nb sequences: 399900\n",
      "Vectorization...\n",
      "Processed 0 0\n",
      "Processed 100000 1374\n",
      "Processed 200000 2713\n",
      "Processed 300000 4058\n",
      "5355_399900_0.01339\n",
      "Time taken: 6.05041 hours, P_ML = 0.01358\n"
     ]
    }
   ],
   "source": [
    "tests = [test1_list,test2_list,test3_list,test4_list,test5_list]\n",
    "del test1_list\n",
    "del test2_list\n",
    "del test3_list\n",
    "del test4_list\n",
    "del test5_list\n",
    "new_size = 400000\n",
    "start = timer()\n",
    "average = 0\n",
    "for test in tests:\n",
    "    n_batch = int(len(test)/new_size)\n",
    "    for ib in range(n_batch):\n",
    "        test_ = test[ib*new_size:(ib+1)*new_size]\n",
    "    \n",
    "        maxlen = 100\n",
    "        step = 1\n",
    "        sentences = []\n",
    "        next_chars = []\n",
    "        for i in range(0, len(test_) - maxlen, step):\n",
    "            sentences.append(test_[i: (i + maxlen)])\n",
    "            next_chars.append(test_[(i + maxlen)])\n",
    "        print('nb sequences:', len(sentences))\n",
    "\n",
    "        print('Vectorization...')\n",
    "        Xt = np.zeros((len(sentences), maxlen, len(chars)), dtype=np.bool)\n",
    "        yt = np.zeros((len(sentences), len(chars)), dtype=np.bool)\n",
    "        for i, sentence in enumerate(sentences):\n",
    "\n",
    "            for t, char in enumerate(sentence):        \n",
    "                Xt[i, t, char_indices[char]] = 1       \n",
    "            yt[i, char_indices[next_chars[i]]] = 1\n",
    "        n_true = 0\n",
    "        diversity = 1\n",
    "        for i,x in enumerate(Xt):\n",
    "            if i % 100000 == 0:\n",
    "                print (\"Processed %d %d\" % (i,n_true))\n",
    "            x = x.reshape(1,maxlen,-1)\n",
    "            preds = model.predict(x, verbose=0)[0]\n",
    "            next_index = np.argmax(preds)      \n",
    "            next_char = indices_char[next_index]\n",
    "            if next_char == indices_char[np.argmax(yt[i])]:\n",
    "                n_true += 1\n",
    "        print (\"%d_%d_%.5f\" % (n_true,yt.shape[0],(float(n_true)/yt.shape[0])))\n",
    "        average = average + (float(n_true)/yt.shape[0])\n",
    "print(\"Time taken: %.5f hours, P_ML = %.5f\" % (float(timer()-start)/3600, average/(5*n_batch)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}