Logistic Regression.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.datasets._samples_generator import make_blobs\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import imageio\n",
    "import time\n",
    "import tqdm\n",
    "import glob\n",
    "import cv2\n",
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "class LogisticRegression() :\n",
    "    \n",
    "    def __init__( self, learning_rate=0.01, iterations=500 ) :\n",
    "        \n",
    "        self.learning_rate = learning_rate\n",
    "          \n",
    "        self.iterations = iterations\n",
    "        \n",
    "        self.grad_hist = [] \n",
    "          \n",
    "    # Function for model training \n",
    "    \n",
    "    def fit( self, X, Y ) :      \n",
    "        \n",
    "        # no_of_training_examples, no_of_features  \n",
    "        \n",
    "        self.m, self.n = X.shape      \n",
    "        \n",
    "        self.W = np.zeros( self.n ) \n",
    "        \n",
    "        self.b = 0\n",
    "        \n",
    "        self.X = X\n",
    "        \n",
    "        self.Y = Y\n",
    "          \n",
    "        # gradient descent learning\n",
    "        \n",
    "        param = (self.W,self.b)\n",
    "        self.grad_hist.append(param)  \n",
    "        \n",
    "        for i in range( self.iterations ):\n",
    "            \n",
    "            self.update_weights()\n",
    "            param = (self.W,self.b)\n",
    "            self.grad_hist.append(param)\n",
    "            \n",
    "        return self\n",
    "      \n",
    "    # Helper function to update weights in gradient descent\n",
    "      \n",
    "    def update_weights( self ) : \n",
    "        \n",
    "        Z = self.X.dot( self.W ) + self.b\n",
    "        \n",
    "        A = 1 / (1 + np.exp(-Z))\n",
    "          \n",
    "        # calculate gradients        \n",
    "        tmp = ( A - self.Y.T )        \n",
    "        tmp = np.reshape( tmp, self.m )        \n",
    "        dW = np.dot( self.X.T, tmp ) / self.m         \n",
    "        db = np.sum( tmp ) / self.m \n",
    "          \n",
    "        # update weights    \n",
    "        self.W = self.W - self.learning_rate * dW    \n",
    "        self.b = self.b - self.learning_rate * db\n",
    "          \n",
    "        return self\n",
    "      \n",
    "    # Hypothetical function  h( x ) \n",
    "      \n",
    "    def predict( self, X ):\n",
    "        Z = X.dot(self.W) + self.b\n",
    "        A = 1/(1 + np.exp(-Z))       \n",
    "        Y = np.where( A > 0.5, 1, 0 )        \n",
    "        return Y\n",
    "    \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [],
   "source": [
    "def predict(X,param):\n",
    "    W = param[0]\n",
    "    b = param[1]\n",
    "    Z = X.dot(W) + b\n",
    "    Z = 1/(1 + np.exp(-Z))\n",
    "    Y = np.where( Z > 0.5, 1, 0 )\n",
    "    return Y\n",
    "\n",
    "def process_frames(X,Y,grad_hist,frame_save_path):\n",
    "    i=0\n",
    "    for i in tqdm.tqdm(range(len(grad_hist)), desc='Processing Frames'):\n",
    "        w1, w2 = grad_hist[i][0]\n",
    "        b = grad_hist[i][1]\n",
    "        c = -b/w2\n",
    "        m = -w1/w2\n",
    "        fig = plt.figure(figsize=(10, 6),dpi=100, facecolor=None)\n",
    "        fig.subplots_adjust(left=0, bottom=0, right=1, top=1, wspace=0, hspace=0)\n",
    "        xmin, xmax = -20, 9\n",
    "        ymin, ymax = -23, 17\n",
    "        x = np.array([xmin, xmax])\n",
    "        y = m*x + c\n",
    "        plt.plot(x, y, color='black', lw=2, ls='-')\n",
    "        plt.fill_between(x, y, ymin, color='tab:blue', alpha=0.2)\n",
    "        plt.fill_between(x, y, ymax, color='tab:orange', alpha=0.2)\n",
    "        plt.scatter(*X[Y==0].T, s=20, alpha=0.8)\n",
    "        plt.scatter(*X[Y==1].T, s=20, alpha=0.8)\n",
    "        plt.xlim(xmin, xmax)\n",
    "        plt.ylim(ymin, ymax)\n",
    "        plt.ylabel(r'$x_2$')\n",
    "        plt.xlabel(r'$x_1$')\n",
    "        plt.xticks(())\n",
    "        plt.yticks(())\n",
    "        fpath = os.path.join(frame_save_path,f\"{i}.jpg\")\n",
    "        plt.savefig(fpath)\n",
    "        i+=1;\n",
    "        #plt.show()\n",
    "        #plt.pause(0.05)\n",
    "        plt.clf()   \n",
    "    return None\n",
    "\n",
    "\n",
    "def write_video(video_path,frame_save_path,fps):\n",
    "    \n",
    "    num_frames = len(glob.glob(frame_save_path+\"/*\"))\n",
    "\n",
    "    img=[]\n",
    "    for i in tqdm.tqdm(range(num_frames), desc='Appending Frames'):\n",
    "        fpath = os.path.join(frame_save_path,f\"{i}.jpg\")\n",
    "        img.append(cv2.imread(fpath))\n",
    "\n",
    "    height,width,layers=img[1].shape\n",
    "    fourcc = cv2.VideoWriter_fourcc(*'mp4v')\n",
    "    size = (width,height)\n",
    "    fps = fps\n",
    "    video = cv2.VideoWriter(video_path, fourcc, fps , size)\n",
    "\n",
    "    for j in tqdm.tqdm(range(len(img)), desc='Writing Video'):\n",
    "        video.write(img[j])\n",
    "\n",
    "    cv2.destroyAllWindows()\n",
    "    video.release()\n",
    "    return None\n",
    "\n",
    "\n",
    "def write_gif(gif_path,frame_save_path,fps):\n",
    "    d = 1/fps\n",
    "    num_frames = len(glob.glob(frame_save_path+\"/*\"))\n",
    "    with imageio.get_writer(gif_path, mode='I', duration=d) as writer:\n",
    "        for i in tqdm.tqdm(range(num_frames), desc='Writing GIF'):\n",
    "            fpath = os.path.join(frame_save_path,f\"{i}.jpg\")\n",
    "            image = imageio.imread(fpath)\n",
    "            writer.append_data(image)\n",
    "    return None\n",
    "\n",
    "\n",
    "def play_video(video_path,fps):   \n",
    "    cap = cv2.VideoCapture(video_path)\n",
    "\n",
    "    while cap.isOpened():\n",
    "\n",
    "        ret,frame = cap.read()\n",
    "\n",
    "        if ret == True:\n",
    "\n",
    "            cv2.imshow('frame',frame)\n",
    "            wait_ms = int(1000/fps)\n",
    "            if cv2.waitKey(wait_ms) & 0xFF == ord(\"q\"):    # Don't use \"and\" instead of \"&\"\n",
    "                break\n",
    "        else:\n",
    "            break\n",
    "\n",
    "    cap.release()\n",
    "    cv2.destroyAllWindows()\n",
    "    \n",
    "    return None\n",
    "\n",
    "\n",
    "def plot_data(X,Y,grad_hist=False):\n",
    "    \n",
    "    fig = plt.figure(figsize=(10, 6),dpi=100, facecolor=None, edgecolor='red')\n",
    "    fig.subplots_adjust(left=0, bottom=0, right=1, top=1, wspace=0, hspace=0)\n",
    "    \n",
    "    if not type(grad_hist)==bool:\n",
    "        b = grad_hist[-1][1]\n",
    "        w1, w2 = grad_hist[-1][0]\n",
    "        c = -b/w2\n",
    "        m = -w1/w2\n",
    "        xmin, xmax = -20, 9\n",
    "        ymin, ymax = -23, 17\n",
    "        x = np.array([xmin, xmax])\n",
    "        y = m*x + c\n",
    "        plt.plot(x, y, color='green', lw=2, ls='-')\n",
    "        plt.fill_between(x, y, ymin, color='blue', alpha=0.2)\n",
    "        plt.fill_between(x, y, ymax, color='orange', alpha=0.2)\n",
    "        plt.xlim(xmin, xmax)\n",
    "        plt.ylim(ymin, ymax)\n",
    "    \n",
    "    plt.scatter(*X[Y==0].T, s=20, alpha=0.8)\n",
    "    plt.scatter(*X[Y==1].T, s=20, alpha=0.8)\n",
    "    plt.ylabel(r'$x_2$')\n",
    "    plt.xlabel(r'$x_1$')\n",
    "    plt.xticks(())\n",
    "    plt.yticks(())\n",
    "    plt.show()\n",
    "    return None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "X, Y = make_blobs(n_samples=200, centers=2, n_features=2, cluster_std=5 , random_state=11)\n",
    "plot_data(X,Y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "X, Y = make_blobs(n_samples=200, centers=2, n_features=2, cluster_std=5 , random_state=11)\n",
    "\n",
    "model = LogisticRegression( iterations = 200, learning_rate = 0.3 ) \n",
    "model.fit( X,Y)  \n",
    "grad_hist = model.grad_hist\n",
    "Y_pred = model.predict(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABA8AAAJ/CAYAAAAXsvogAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD+naQAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nOzdb2wcd57n9091d7HZzX8SZVp/LNo3K3ml27MusbPxjjQHzxnG3J1vscYCe0EOMJAHB/lJgMDIDrBBgCSGsckhGcAbTIIkD1YBkgdONvcnOHiSDE672rlhcvKMdjLarL07kj20h6L+UW1JI3azm93F7sqDH5tkU2Sz/1R1/arq/XrSUJEiq5tV3fX71Pf3/Tm+7wsAAAAAAGA/mah3AAAAAAAA2I3wAAAAAAAAdEV4AAAAAAAAuiI8AAAAAAAAXREeAAAAAACArggPAAAAAABAV4QHAAAAAACgq1wUv9RxHEfSCUnlKH4/AAAAAADYMiXpru/7/n7fEEl4IBMc3I7odwMAAAAAgE4nJd3Z74tRhQdlSVr+6R9penomol0AAAAAACDdVp880vy/9bZ0wMyAqMIDSdL0zKymD81GuQsAAAAAAOAANEwEAAAAAABdER4AAAAAAICuCA8AAAAAAEBXhAcAAAAAAKArwgMAAAAAANAV4QEAAAAAAOiK8AAAAAAAAHRFeAAAAAAAALoiPAAAAAAAAF0RHgAAAAAAgK4IDwAAAAAAQFeEBwAAAAAAoCvCAwAAAAAA0BXhAQAAAAAA6IrwAAAAAAAAdEV4AAAAAAAAuiI8AAAAAAAAXREeAAAAAACArggPAAAAAABAV4QHAAAAAACgK8IDAAAAAADQFeEBAAAAAADoivAAAAAAAAB0RXgAAAAAAAC6ykW9AwAABK68IlVK0uScNHU06r0BAACIPcIDAEBy1Nekq5ekpWtS05OyrvTCq9KFi1J+Iuq9AwAAiC2mLQAAkuPqJWlxQXIyUmHaPC4umO0AAAAYGOEBACAZyium4sAtmiqDTM48ugWzvbwS9R4CAADEFuEBACAZKiUzVcHNd253x832Sima/QIAAEgAwgMAQDJMzpkeB169c7u3brZPzkWzXwAAAAlAeAAASIapo6Y5oleV6hWptWEevZrZzqoLAAAAA2O1BQBAcly4aB6Xrkm1VVNxcOq17e0AAAAYCOEBACA58hPS6++a5oiVkpmqQMUBAADA0AgPAADJM3WU0AAAACBA9DwAAAAAAABdER4AAAAAAICuCA8AAAAAAEBXhAcAAAAAAKArwgMAAAAAANAV4QEAAAAAAOiK8AAAAAAAAHSVi3oHAAAAgJEor0iVkjQ5J00djXpvACBWCA8AAACQbPU16eolaema1PSkrCu98Kp04aKUn4h67wAgFpi2AAAAgGS7eklaXJCcjFSYNo+LC2Y7AKAnhAcAAABIrvKKqThwi6bKIJMzj27BbC+vRL2HABALhAcAAABIrkrJTFVw853b3XGzvVKKZr8AIGYIDwAAAJBck3Omx4FX79zurZvtk3PR7BcAxAzhAQAAAJJr6qhpjuhVpXpFam2YR69mtrPqAgD0hNUWAAAAkGwXLprHpWtSbdVUHJx6bXs7AOBAhAcAAABItvyE9Pq7pjlipWSmKlBxAAB9ITwAAABAOkwdJTQAgAHR8wAAAAAAAHRFeAAAAAAAALoiPAAAAAAAAF0RHgAAAAAAgK4IDwAAAAAAQFestgAAAJBkLE8IAAgA4QEAAEAS1dekq5ekpWtS05OyrvTCq9KFi1J+Iuq9AwDEDNMWAABAp/KKdO9T84j4unpJWlyQnIxUmDaPiwtme9g4hgAgcag8AAAABneqk6O8Yv6ObnH7b5fPSfLN9vJKOFMYOIYAILGoPAAAAEaUd6oRrErJDN7dfOd2d9xsr5TC+b0cQwCQWIQHQNxQCgogDLvvVGdy5tEtbN+pRnxMzpm7/l69c7u3brbLCf6zhGMIABKNaQtAXFAKCiBM7TvVhenO7e64VFs1X6dTf3xMHTWfEYsLknzzd/TWpUZVKhyS/uU/Dv6zhGMIABKNygMgLigFBRCmg+5UT85Fs18Y3IWL0qnXJN83g3ffN8FB7ZfhfJZwDI0e1YgARojKAyAOomp8BSA99rtT7dXMAJT3mPjJT0ivv2s+IyolSY6pOBibCOezhGNodKhGBBABKg+AOIiq8RWAdNnrTvWp18x2xNfUUen4S5L8gz9Lhr2TzTE0GlQjAogAlQdAHOwsBc3vOG0pBQUQpN13qifnuFucJN0+SzJZ6ZPvSXc/He5ONsdQ+KhGBBARKg+AOGiXgnpVqV6RWhvm0auZ7VwkAAhS+0417y3J0u2zxC1It34S3J1sjqHwUI0IICKEB0BcUAoKABjWXp8l879uqg9YYjEeaEwJICJMWwDiglJQAMCw9vosqZSk29elfLHze1li0U40pgQQEcIDIG6mjnJhAAAYzu7PEvrqxEu76nDpmgl4si7ViABCR3gAAACQZtzJjh+qEQFEgPAAAAAg7YK4k81AdvSoRgQwQoQHAAAAaTfMnez6mlmVYenacMs8AgCsRngAAAAAY5A72VcvmSkPbtEs8+jVN6dAyAQSAIBEYKlGAAAADKa8YioOWOYRABKP8AAAAACDqZTMVAU337ndHTfbK6Vo9gsAEDjCAwAAAAxmcm57mcedWOYRABKH8AAAAACDaS/z6FWlekVqbZhHr2a2sxIAACQGDRMBAAAwuCCWeQQAWI/wAAAAAIMbZplHAEBsEB4AAABgeIMs8wgAiA16HgAAAAAAgK6oPAAAAECwmMIAAIlDeAAAAIBg1Nekq5dM88SmZ5onvvCqaZ6Yn4h67wAAQ2DaAgAAAIJx9ZK0uCA5GakwbR4XF8z2JCuvSPc+NY8AkFBUHgAAAGB45RVTceAWt6sM8jlJvtleXkneFAYqLUaDaTCAFQgPAAAAMLxKyQygC9Od291xqbZqvp60gV+70sItmuft1c2/JbN8JYZDOANYhWkLAAAAGN7knBncefXO7d662T45F81+hWV3pUUmZx7dwnalBYaT1mkwgKUIDwAAADC8qaPmrrBXleoVqbVhHr2a2Z60qoN2pYWb79zujpvtlVI0+5UUhDOAdQgPAAAAEIwLF6VTr0m+b6Yq+L7594WLUe9Z8NJWaTFqhDOAdeh5AAAAgGDkJ8xc/zQ0uGtXWiwuSPLNoNZbN5UWp15L7vMelZ3hTH7HkIVwBogM4QEAAACCNXU0HYPndkXF0jVTaZF1k1tp0Y8gwiPCmfhKQ3iYUoQHAAAAwCDSVGnRi2FWR9jrNSSciRdWx0g8wgMAAABgGGmptDjIIEtXHjTgJJyJD5YuTTzCAwAAAADD2b06grTZq8DfXh1hr4F/LwNOwhn7Dfr3R6yw2gIAAACA4QyyOgLLMSYHq2OkAuEBAAAAgOEMsnQlA87kYOnSVCA8AAAAADCc9uoIXlWqV6TWhnn0amb7XiXrDDiTY5C/P2KH8AAAAADA8C5cNKsh+L5ZHcH3u6+OwIAzWfr9+yN2aJgIAAAAYHiDrI7AcozJweoYiUd4AAAAACA4/ayOwIAzeVgdI7EIDwAAAABEiwEnYD16HgAAAAAAgK6oPAAAAAAAdGIqCXYhPAAAAAAAGPU16eol08Sy6Zkmli+8appY5iei3jtEiGkLAAAAAADj6iVpcUFyMlJh2jwuLpjtSDXCAwAAAACAmaqwdE1yi6bKIJMzj27BbC+vRL2HiBDhAQAAAADA9DhoepKb79zujpvtlVI0+wUrEB4AAAAAAExzxKwrefXO7d662T45F81+wQqEBwAAAAAAs6rCC69KXlWqV6TWhnn0amY7qy6kGqstAAAAAACMCxfN49I1qbZqKg5Ovba9HalFeAAAAAAAMPIT0uvvmuaIlZKZqkDFAUR4AAAAAADYbeoooQE60PMAAAAAAAB0RXgAAAAAAAC6YtoCACD+mJcJAAAQKsIDAEB81dekq5dMR+imZzpCv/Cq6Qidn4h67wAAABKDaQsAgPi6eklaXJCcjFSYNo+LC2Y7AAAAAkN4AACIp/KKqThwi6bKIJMzj27BbC+vRL2HAAAcrLwi3fuUzy1Yj2kLAIB4qpTMVIXCdOd2d1yqrZqv0/8AAGArpt4hZqg8AADE0+ScudDy6p3bvXWzfXIumv0CAKAXTL1DzBAeAADiaeqouUPjVaV6RWptmEevZrZTdQAAsBVT7xBDhAcAgPi6cFE69Zrk+2aqgu+bf1+4GPWeAQCwv/bUOzffud0dN9srpWj2C+iCngcAgPjKT0ivv2vu0FRKZqoCFQcAANvtnHqX3zEkY+odLEblAQAg/qaOSsdfIjgA+kWXdyAaTL1DDFF5AAAAgkc1iN3o8h5fnFvJ0Z5it3TNTL3Luky9g9UIDwAAQHAYlMZDu8u7WzRd3r26+bdkpgLBPpxbycPUO8QM0xYAAEBwWHrMfnR5jyfOreRi6h1igvAAAAAEg0FpPNDlPX44twBYgPAAAAAEg0FpPOzs8r4TXd7txbkFwAKEBwAAIBgMSuOBLu/xw7kFwAKEBwAAIBgMSuPjwkXT1d33TZd336fLu804twBYgNUWAABAcFh6LB7o8h4/nFsAIub4vj/6X+o405KePFm8rOlDsyP//QAAIGQMSoFwcG4BCNjq45JmTr8pSTO+76/u931UHgAAgOBNHWVgEwYGjuDcAhARwgMAAGzC4DC9uv3t62vS1UumZL3pmZL1F141Jev5iWj2NyycA8DwOI8QAsIDAABskKbBITr18re/eklaXJDcolSYNl33FxfM115/N7p9DxLnADA8ziOEiNUWAACwQXtw6GTM4NDJmH9fvRT1niFsB/3tyytmIOAWzcV/Jmce3YLZXl6Jdv+DwjkADI/zCCEiPAAAIGppGRziab387SslcwfRzXf+X3fcbK+Uotn3IHEOAMPjPELICA8ADK+8It37lA8lYFBpGBxib7387SfnTOmxV+/8Hm/dbJ+cG93+hqVSMs9HvnnebZwD6cM1xeD4LEHI6HkAYHDMqwOCsXNwmN/x0ZykwSH21svffuqoeW9dXJDkm4GAty55NenUa/FvhlZfkz75nrS+Kq0/kZyslC9KE89wDqQJ1xTD47MEIaPyAMDgmFcHBKM9OPSqUr0itTbMo1cz2+M+OMT+ev3bX7hoggLfl2qr5vHUa2Z73F29JN36yY67pS1pvSKt3uMcSBOuKYbHZwlCRuUBgMHsnlcnbabc/va8Oj6kgN61B4FL18zgMOsmZ3CI7nr52+cnzKoKSVt+bednSfGItFaS6lVJTXP39GvfiMc5kLS/y6hxTREcPksQIsIDAINpz6srTHdud8fNh1WlxAc90I+kDg5xsH7+9lNHk3Vc7PwsyWTMcyt60kbd3C39m79ld8k6pfbB4JoiOHyWIESEBwAGw7w6IBxJGxyid2n82+/1WZJ1TXiQG7f/s6Rdau8WzcDXq2/2ppAZwCVNWANSrimCl8b3E4SO8ADAYJLewAsAEL44f5akqdQ+7AqLOB8HtqHiACEiPAAwOObVIc24QAOCEdfPkjSV2o+iwiKux4Etegl4+NzCkAgPAAyOeXVII+Y4A8GK62dJWkrtR1VhEdfjwBbdAp4LF/ncQiAIDwAMj3l1SJO0zXEGRiVunyVpKbUfdYVF3I4DGxwU8HjVzeVQ+dzCcDJR7wAAALGx+wItkzOPbmH7DhyA9Lhw0QQFvm8G0r6fvFL7nRUWOyWtwiLO2gGPm+/c3g60lq/zuYVAUHkAAECv0jTHGcDB0lBqn5YKizjrNoXGyUh+c+9ggc8t9InKAwAAesUdOAB7mToqHX8puYOwNFRYxFk74PGqUr0itTbMo1eT5l8xy57yuYUAUHkAAECvuAMHII3SUGERd91Wq2j36uFzC0MiPAAAoB8sJzZ6DFiGw+uHoNDM0F7dAh4+txAQx/f90f9Sx5mW9OTJ4mVNH5od+e8HEDEuZJEEHMfhY1nM4fD6AdiJzy3sY/VxSTOn35SkGd/3V/f7PioPAIwOF7JIEu7AhY9lMYfD6wdgJz63MCQaJgIYnfaFrJMxF7JOxvz76qWo9wyAbcJaFrO8It37NPnLk7GsKDB6aXl/2S2tzzuFqDwAMBq7L2SlzeWE/O0LWdJwAG1BL4uZtsonlhUFRidt7y9taX3eKUblAYDRaF/I7rXOcNMzXweAtqCXxUxb5RPLigKjk7b3l7a4PG8qIwJD5QGA0dh5IZvf8dbDhSyAvQS5LGYaK59YVhQYjTS+v0jxeN5URgSOygMAo9G+kPWqUr0itTbMo1cz26P+gAFgnwsXzUDX902pve8PtrxYWiufgnr9AOwvre8vcXjecamMiBEqDwCMDusMA5B6Xy6s27rl/Uhr5VNQrx+A/aX1/cX25x2HyogYIjwAMDpcyALpNmgJ6bDLi6W9hJ/l2YDwpPX9xfbnTdPYUBAeABg9LmSBdNgdFLZLSN2iuaDz6psXnjLBYpiofAIQlrS+v9j8vG2vjIgpwgMAABCsvSoMTrwk3fmL6EpIqXwCEJa0vr/Y/Lxtr4yIKcIDAAAQrL0qDL78kQkSDp3o/N5Rl5BS+QQgLGl9f7H1edtcGRFThAcAACA4+zWpam1Ila9MVUJhZvv7KSEFgKfZeDc/bmyujIgpwgMAAOLG5guh/ZpU5Sek2hOpUZUyWUpIAWAvgzaWxf5srYyIIcIDAADiIg4Xld2aVBUOSSfOSfc+pYQUSBqbQ804ibKxLHAAwgMAAOIiDheVBzWpooQUSJY4hJpxsd+0r1E1lgUOkIl6BwAAQA92X1RmcubRLWxfVNriwkUTFPi+qTDw/c4Kg6mj0vGXuAgGkqAdajoZE2o6GfPvq5ei3rP4aU/7cvOd291xs71Sima/gE1UHgAAEAf79RIY9WoFvaBJFZAO3CkPVrdpXzSWhQWoPAAAIA52XlTuZPNFJRUGQLJxpzxY7WlfXlWqV8wqNfWKmfb1wqu8lyJyhAcAAMQBF5UAbBPHUNN2B037AiLEtAUAAOKiffG4dC2eqxUwjQFIloMapHKe949pX7AY4QEAAHER14tKurEj7uJ2zo1S3ENNW00d5ViDdQgPAACIm7hdVMZhiUlgLwRfB4trqJk2/H0QAMIDAAAQHrqxp0cSBydJC77C/BvFLdRMCwIwBIjwAAAAhCdOS0xiMEkdnCQp+Erq3wgHS1oAhkix2gIAAAgP3diTrz04cTJmcOJkzL9/+N9I9z41g+w4StIyhPv9ja5einrPEKbdAVgmZx7dwnYAlhbllXi/H1mCygMAABAeurEn2153592MVHss3fxTaekn5m8ex7vcO4Ov/I5L5rgFX0mqoEB/qPyi6iZgVB4AAIBwsW55cu11d36tZAbY8qWxQnzvcreDL68q1StSa8M8ejWzPS6DriRVUKA/VH5RdRMwKg8AAEC46MaeXLvvzjc9qV6VHEdSRsrlzdfjepc7CcsQJqWCAv1Le+UXVTeBIzwAAACjQTf25Nk9OJEkvynJkcaLm8GB4lsmnYTgK+0DyLRLQgA2KKZtBI7wAACiFOcLUgCQOgcnG+uSHHNxPvHM9vfE/S533IOvNA8g0y4JAdigqLoJHOEBAESBBj4AkmL34OQvvict/8Tc2eYutx3SPICEEfcAbBBU3QSO8AAAosC6ywCSpj04mf3adjjKXW67pHEAGSeEO8Gj6iZQhAcAMGo08AGQZNzlBvpDNWJ4eD8KFOEBAIwaDXwApAF3uYHeUI0YPt6PApGJegcAIHVYdxkAAEhPVyNmcubRLWxXIwKWIDwAgFFrN/DxqlK9IrU2zKNXM9tJxgEASId2NaKb79zujpvtlVI0+wXsgfAAAKJw4aJp2OP7ZqqC79PABwCAtKEaETFCzwMAiAINfAAAAMsJIkYIDwAgSjTwAQAg3VhOEDFBeAAAAIB0oeoLNqEaETFBeAAAAIB0qK+ZZfGWrplmdFnXlIxfuGgGcECUqEaE5WiYCAAAgHS4esnMLXcyUmHaPC4umO0AgK4IDwAAABB/5RXp3qfmcb+vL12T3KKpMsjkzKNbMNv3+38AAElMWwAAAECc9ToVoVIyXy9Md/5/d9w0qauUKBkHgC6oPAAAAEB89ToVYXLOBAtevXO7t262T86Nbp/3clDlBBAnHM+JROUBAAAA4mn3VARJyuck+dtTEdrVBFNHTUXC4oL5ujtuggOvZpbFi6rqgCaOSBKO50Sj8gAAAADx1J6K4OY7t7vjZnul1Ln9wkUTFPi+marg++bfFy6Obp93o4kjkoTjOdEirTz4/Je3dG5qSmNZN8rdAAAAQD9sWY9+51SE/I7L2v2mIuQnpNfftWf/+6mcQO9s+fumDcdz4kUaHvz6P72oTCGjF6aO68VD8zo9M68XD83rxZnndfrQSX1t+jmCBQAAAFvYVpI86FSEqaN2DGJo4hgs247PtOF4TrzIex60/Ja+XL2jL1fv6LJ+1PG1jEOwAACAdbirl17tkmS3aAYIXn1z4C5zRz8K7SkHS9fMACXrRj8VoVf9Vk6kVa/vOTYen2nC8Zx4kYYHf+vw39eKs6w761+q2qo89XWCBQAALMJdvXSztSTZtqkI/bC1iaMt+nnPsfX4TBOO58SLNDz4z17+XzU7Oy3f9/W4UdKttc+1tHpDS6s3tbz2mW7XviBYAADAFtzVSzfbS5JtmYrQrzhXToStn/cc24/PtOB4TrTIpy1IkuM4ms0/q9n8s/o3Z7/R8bVRBAvtR4IFAAD2wV09dCtJdjLS2mOOg0HEuXIiTP2+51AybweO50SzIjzohmABAFKAiwz7cVcPe5UkN6rS2iMpm5N++N8ylWUYca2cCEu/7zmUzNuF4zmRrA8PuiFYAICYYw59fHBXD9LTJcmNquQ40viMNFZgKguCM8h7TppL5gnhMQKxDg+66TVYuLV6Q78gWACAaDCHPj64qwepsyT5wU1p4b+XMi5TWRC8Qd5z0lgyTwiPEUpseNANwQIAWIA59PGT5rt66DR11AzOWi0pn+/8GlNZEJRB33PSVDJPCI8RSmV40A3BAgCMCHPo4yeNd/WwP6ayIGy853RHCI8RIzzoA8ECAASIgUd8pemuHvbHVBaMCu85eyOETyaLwzLCg4D0GyzcXvtcy7VFggUA6cXAA4g/prIA0SGET5YY9K9wfN8f/S91nGlJTy5ffqLZ2ekDvz/JBgkWuiFYABArMfigBNADi++UAYn2g+9u9jwoPB3C0/MgXrb+lkXJzZtQyKuO5G+5+rikmdNvStKM7/ur+30f4YHFwgoWTs+c1IuHnidYAGAPBh69S9prlbTnAwCjRAifDOUV6Z//ruRkOv9u9Yrk+9Lv/EGon5G9hgdMW7BYmFMh/nj5xx1fI1gAECnmsx7M9gvEfkMA258PMAjCMIwaTSWTISb9KwgPYopgAQBSxtbluAYNAWx9PsAgCMMQNUL4eItJ/wrCgwQiWACAhLF5Oa5BQgCbnw8wCMIwAMOISRNpwoOUIVgAgBiytZxx0BDA1ucDDIIwDEAQYrB6DeEBtkQZLOwMGAgWAGAXW8sZBw0BbH0+wCAIwwAEIQb9KwgP0BOCBQCIkK3ljIOGALY+H2AQhGEAgmRx/wrCAwytl2Bhee3nWlr9GcECAAzKxnLGYUIAG58PMAjCMAAp4fi+P/pf6jjTkp5cvvxEs7PTB34/kmmQYKEbggUAqWBbOeOwXeZtez7AIFhtAUCMrT4uaeb0m5I04/v+6n7fR3gAK40qWDg9M69fmSFYAIChEQIAnAcAYonwAIlFsAAAAAAAwSA8QCoRLAAAAABA7wgPgF12BwtL5c+0XPmMYAEAAABAahEeAH0gWAAAYEjM9weAWCI8AAISRrDw/NQxvTgzT7AAID4YGGI/rDQAALFGeACMAMECgMRjYIiD/OC70uKC5BYlNy95dcmrSqdek15/N+q9AwAcoNfwIDe6XQKSx3Eczeaf1Wz+Wf0bsxc6vjZIsNDyW/rF6l39YvWu/nj5xx1fI1gAEImrl7YHhoVpMzBcXDBfY2CI8ooJltzidpiUz0nyzfbyCpUqAJAQhAdASAYJFm7XvtDt9S8IFgDYgYEhDlIpmYqUwq5KUndcqq2ar3OMAEAiEB4AEYgqWDh9aH5HwECwAOAADAxxkMk5M5XFq28GS5u8dbN9ci66fRsV+oEA6FXM3y8IDwDLECwAsAYDQxxk6qjpgbG4IMk3wZK3Lnk10/MghhfHPaMfCIBeJeT9gvAAiBGCBQAjleaBIXp34aJ5XLpmKlKyrjk+2tuTin4gAHqVkPcLwgMgIQ4KFn7Z+Eq31j4nWADQn7QODNG7/IS5+I15OW5f6AcCoFcJer8gPABSwHEcHc7P6XB+jmABQH/SODDEYKaOpufYoB8IkA5BfPYl6P2C8ABIuaiDBfP4vL42c0L57FiozxXAENI0MAQOEkQ/EAI5wF5B9ihIUP8gwgMA+yJYAHAgBkBIo2H6gSSkcRqQaEH2KEhQ/yDCAwADIVgAUo4BENJu0H4gCWmcBiRWGD0KEtI/iPAAQOB6DRZurd7QL8o3CRaAOGIAhLQbpB9IghqnAYkVRo+ChPQPIjwAMFIEC0ACMAACtvXTDyRBjdOAxAqzR0HM+wcRHgCwBsECEBMMgIDBJKhxGpBYCepREDTCAwCxQLAAWIQBEDAYBiVAPCSkR0HQHN/3R/9LHWda0pPLl59odnb6wO8HgEHtGSysfa7b1cV9g4VuCBaATT/47mbPg8LTAyB6HkQv5vNqE41mo0B8pOS9dPVxSTOn35SkGd/3V/f7PsIDAKlFsAAMgQGQnfi7xEdKBiUA7Ed4AABDCCtYOD1zUi8eep5gAcnBAMguWxUhRcnNm6klXpWKEADAvnoND+h5AAB76LvHwgHBws4eC3+yfK3jawQLiLWYd45OFFbBAACEiPAAAPpEsADASqyCAQAIEeEBAASIYAFAZFgFAwAQIsIDABgRggUEgh4D2A/LAAIAQkTDRACwXNjNG7cDBoIFq9FFH70I+Di5s9rU/XJTx6ayem46G8IOAwCixmoLAJACBAspYmkXfS0vZkkAACAASURBVAaXlhqyQqVcb+k7C2X98MuG6hu+8jlH3/zamH7vtSlN5TMh7DAAICqstgAAKRDlVAiChRGysIu+VYNLpnI8bchVML6zUNb3b9Y1OSYdHne0vuHr+zfrkqTf/9ZMUHtpP44tANhCeAAACdVLsLBc/bmWnvyMYMF2FnbRt2JwmZCpHLZVb9xZbeqHXzY0OaatIMjNOvLV0g+/bOjOatOK/QxVQo4tAAgS4QEApNDOYOFvHj7f8bVRBQunZ+b1KzPPESwc4M5qUw8rh/RrGVc5S7roWzO4vHppeypHYdpM5VhcMF+LcCpHr6yq3tjhfrmp+oavw+NOx/ZCztHjdV/3yykID2J+bAFAGAgPAAAdCBbs0DmwHNN/5J/Tt/RjFX1f2bFou+hbMbi0cCpHv6yo3tjDsams8jmzP252+29c2ww4jk0lPDhIwLEFAGEgPAAA9IxgYXR2Dywvef9QG3Xpjdr/pyPNVVNxcOo1U0Y9YlYMLi2cytEPa6o39vDcdFbf/NqYvn+zLl8tFXKOahu+1hrSm2fGkl91EPNjCwDCQngAAAhEP8HCUvkz3Vr7jGBhH3sNLJWd1H/n/CP9E/+h/odvepo7OlxDvGFYMbicnDMBiiVTOfplRfVGF7/32pQk6YdfNvR43YRCb54Z29qeaDE/tgAgLIQHAIDQESz0p9vAcnn9iG4VDmluKtrnEfngcuqoaWC3uCDJN3eFI5zK0S8rqje6mMpn9PvfmrGumeNIxPzYAoCwEB4AACI1ymDBkaPnp47pxUPzVgcLtg8sJUsGl+0pG0vXTDl5hFM5+mVF9UYPnptOUWiwU4yPrdhhOUwgNhzf90f/Sx1nWtKTy5efaHZ2+sDvBwBgt0GChW5sCxb+0z9+ou/frGtiTLsGlvlIm+lZKaaDD1tXW8AOMT22BjXSMJDlMAFrrD4uaeb0m5I04/v+6n7fR3gAAEicJAQLDCzTI5VTA2CVSN5vfvDd7eUw3bzpMeFVTYUHy2ECI0V4AADAHvYLFu5Uv9Dt9S+01ir39fPCDhYYWAIIW7vSaXJMGt+cMlUJs9KpvCL989+VnExnlUG9Ivm+9Dt/kIpKD8AWvYYH9DwAAKRKrz0Wbj2+rl/e+p90p/qZFlvrWpSnJ2o99fN8+Voq39NS+V4oPRZSO+ccwEhEsmwoy2ECsUR4AADApp3Bwm/e/mc6XH+sZv5ramYLymxU9XhjVdef+Q392TPf6qlioddg4fTM/I6Awa7mjQCSLZJlQ1kOE4glwgMAHdzqssbW76oxfkJecT7q3QEi4VaXNf3gipq5STVdc2esNTajWcfRG09+pvlf+468v/aPtr5/kKkQBAsAbBDJ6i4shwnEEuEBAElSxlvVczfe1/SDK8q01tXKjGv12Td05+x7arn0JkG6jK3fVaa1Ls+d7djezBbkeo80tn63I1w7aCrEE++hbq19TrAAwDqRLRvKcphA7BAeAJAkPXfjfR2++5GauUl57qyyzZoO3/1IkrR87oOI9w4Yrcb4CbUy48o2a2pm3K3t2WZNrcy4GuMnev5ZjuPo0NgzOjT2DMECgMh0a776e69NSZJ++GVDj9dNxcGbZ8a2tociP2FWVUjZcphAnLHaAgC51WX96se/JcnZKtGWpKy3KsnXZ+e/xxQGpM78J9/eCtSa2YKyzZqyGxU9PvHWSAK1QYKFbvYKFszj8wQLQIL1swwjq7sA6cRqCwB61m+JNpAGd86+J0mafnBFrvdIrcy4Hp94a2t72KKuWCBYAJLhOwvlrWUYD4+b3gbfv1mXpKeWYWR1FwDdEB4ACLREG0iKljut5XMfWNlElGABEneJcbBIlmEEkFiEBwDkFee1+uwbWz0Odpdo2zJgQnyUqjU9rNZ0pFjQXLEQ9e4MxSvOx+oc6DlYWL2hpdWbBAsx1E8ZOtItkmUYASQW4QEASdGXaCMZqp6nDz+5qev3S/KaTbnZrF4+Nqe3z51R0XUP/gEIFcFCMrz3J6u6sljX5JhzYBk60i2SZRgBJBbhAQBJdpdoIz4+/OSmPr59TwU3p6n8mOrNpj6+fU+S9M4rLwX2e8KqbEhSxUS/CBbsV6639N6fPNFHP6ur5Us1z1fNMwNAytCxl8iWYQSQSIQHADrErUQb9ihVa7p+v6SCm9PEZpVBLmNKqK/fL6lUrQ09IA+rsoGKie76CRZuP/qJblf+SkuNFS3XbxMsBOg7C2Vd+XlDvi+5GcmXtFr3JTX17ESGMnTsKZJlGAEkEuEBACAQD6s1ec2mpvKdg7t8Nqtyo6GHAYQHYVU2jKpiIonawcKsM6bfvP3PNP3oXyvTWlcrM64nJy/qL0/9B1pq3KdiYUjtxndTeam2Ifm+lN1sb1Bp+Cq4lKFjb1P5jH7/WzM02AQwNMIDAEAgjhQLcrNZ1ZvNrYoDSao3m3IzWR0ZMjgIq7JhFBUTafDcjfd1+O5HauYm5bmzyjZrmr33Pb3kOJo590FPUyGW1z7X7eoiwcIethvfZVT1WqbioCU5juS1TIDw1l/PMyjEvliGEcCwCA+AFEjzPG6MzlyxoJePzW3dsc9vBgk1b0PnTx4f+tgLq7JhFBUTSedWlzX94IqauUk13WlJ2lr2dfrBFbnV5aemQ/XbYyGIYKEdKuwMFr42fULjuXyQL0codja+OzaZkdRSpeFroyVlHOmNU3nK0AEAoSI8ABKMedwYtbfPnZFk7tiXGw25mazOnzy+tX0YYVU2hF0xkQZj63eVaa3Lc2c7tjezBbneI42t3+2rl0qYwcKV2/EMFjob3/l6diKjgttSpS69cXpMf/Cbh6LeRQBAwhEeAAnGPG6MWtF19c4rL4VS7RJWZUPYFRNp0Bg/oVZmXNlmbaviQJKyzZpamXE1xk8E9rvSHCw83fguo7d+zaLGd+UVqVKSJuekqaNR7w2AKPF+kEiEB0BCMY8bUZoLaYpMWJUNYVZMpIFXnNfqs2/o8N2PJJmKg2yzpuxGRY9PvNVT1UEQy8QmPViwtvFdfU26eklauiY1PSnrSi+8Kl24KOUnot47AG2jGNDzfpBoju/7o/+ljjMt6cnly080Ozs98t8PpMGNrx7pg49/qqn8WEcp9karpXKjoW9//RWdfWa2y08A7BVWHw/6gwwu463quRvva/rBla3VFlaffUN3zr6nlrv/Z/2g/y9IgwQL3dhWsRC6H3xXWlyQ3KLk5iWvLnlV6dRr0uvvRr13AEY5oB/k/YAqhcitPi5p5vSbkjTj+/7qft9H5QGQUMzjRpKFVdkQ1s9Ng5Y7reVzH/RdQbDXKg3tCoblcx+EvduSoq1Y2A4YYhoslFfMgMQtbg9C8jlJvtleXmEwAETt6qXtAX1h2gzoFxfM14IM+Pp9P6BKIXYID4CEYh43gCh4xfmepx3sXqVho9XSujOh8Wxr31UaRo1g4QCVkrnoL+yqEnHHpdqq+TrhARCdUQZ8/b4fjCrUQGAID4AEYx43AJu1V2lo5A7rYbWmqrchX1JOvuYya2pVbkkRhwfdECzIlBlnXXPRn99xWemtm+2Tc9HsFwBjlAFfP+8HVC3FEuEBkGBhdr4HgGG1V2lYqz1RZSOvjOMo4zga9+uqNDP6335R0z94Nuq9HEwvwcLy2s/1i9WfxTtYmDpqyowXFyT5ZkDirUtezcxx5uIfiNYoA75+3g+oWoolwgMgBZjHDcBGXnFeK4e/qck7/0JTTlMNjWvcr6uodf1p5rz+9KusvpnAlWF2BgvnDn+942uxDBYuXDSPS9fMRX/WNQOF9nYA0Rl1wNfr+wFVS7FEeAAAACLzoxO/K925r7+lP9eMymo4rhYy39D/mP335HlNPUxgeNCNDcHC6UPz+pXp53oPFvITZn4yHdMBO40y4Ov1/YCqpVgaaKlGx3EKkmZ937+za/vf8H3/L3v4/yzVCAAAVKrW9N6/+pGOqaTns7/UV84zepB5VmueJ9+X3v/bX09VeDCoQYKFbgILFgDYw7aAj9UWrNHrUo19hweO4/wDSf+1pEeSHEnv+L7/482v/dT3/Vd6+BmEBwAAQJL0hz/9VB/fvqeCm3tqZZh3Xnkp6t2Lvd3Bwq3Vz3Rr7TOCBQB2sC3USKEww4M/l/Qt3/dLjuP8uqT/WdJ/4fv+/+I4znXf91/u4WcQHgA70NAQQJpVPU8ffnJT1++X5LWacjNZvXxsTm+fO6Oi60a9e4lGsAAACDM8+Evf9//Gjn8fkfS/S7oi6bepPAB613HB3GzKzW5fMK95GwQKAFKFINUuYQQL81NH9eLM8wQLAGCRMMODH0h61/f9v9ixbUymAuHf8X3/wCaMhAeAsVepbtXb0Ex+TI1m66lAgTtwAAAbECwAQHIEHh44jjPl+37ZcZyTkjZ837+/x/d8w/f9f93DzyI8QOq1m4Q5jjSxIxS4W15TfWNDR4rjKrouc38BYJNbXdbY+l01xk/IK85HvTvYB8ECAMRLr+FBP0s1/t+O4/w93/dv7/cNvQQHAIyH1Zq8ZlNT+bGtbV6rpUazKTmOxrJZ5TIZ5TIZSdL1+yWVUrZkGQBIUsZb1XM33tf0gyvKtNbVyoxr9dk3dOfse2q53ISwTT/LTW4FC7UvdGf9C1WaT1+z+vJ1q3xft8r391xukmABAEajn/DgJ5J+7DjO3/V9/0Z7o+M4L8s0TPz7ge8dkGBHigW5m1MV2gHBRqsl3/eVyTjKbm6TpHw2q3Kjkbr1zgFAkp678b4O3/1IzdykPHdW2WZNh+9+JElaPvdBxHuHfhAsAEB89Rwe+L5/0XGc9yT9P47j/LakB5L+c0m/I+mjkPYPSKy5YkEvH5vTx7fvSTIBgddsSZLcTEbujvCg3jTdx48QHABIGbe6rOkHV9TMTaq5WWXQzJipXtMPrsitLjOFISGiChZOHzq5HTAQLADAvvqpPJDv++87jtOQ9MeSspL+paR/2/f9n4axc0DSvX3ujCQzJaHcaMjNZPXc9KSerNe15nlPrXdO1QGAtBlbv6tMa12eO9uxvZktyPUeaWz9LuFBChAsAED0eg4PHMc5Luk/lnRR0l9JOivpjwgOgMEVXVfvvPJSx/JkE25ua/nGdqBw/uTxraABANKkMX5Crcy4ss3aVsWBJGWbNbUy42qMn4hw75Injk0pCRYAYDT6WW2hJumGpP/E9/3/03Gcvyvpn0j6x77v/1d9/VJWWwAOxHrnAGDMf/LtrZ4HzWxB2WZN2Y2KHp94i54HAUljU0oTLDzS8trnPQcL3RAsAAlQXpEqJWlyTpo6GvXejEwYSzX+Q9/3/2jXtlck/R+S/oXv+/9+rztHeAAAAHqVxoHtqBHQdCJYAFKmviZdvSQtXZOanpR1pRdelS5clPITUe9d6AIPD/b9AY7z1yT9X77v/1of/4fwAAAA9CWOJfVx4FaX9asf/5YkZ6sppSRlvVVJvj47/z1e7x1GESycPnRSLx56nmABGJUffFdaXJDcouTmJa8ueVXp1GvS6+9GvXeh6zU86Kth4l583/+F4zjfGPbnAAAAdOMV51MziB1lUEJTyv6YHgtHdGjsyD49FvoLFvrtsUCwAASsvGIqDtzidpVBPifJN9vLK6mawtDN0OGBJPm+/ziInwMAAJBmUUzRoCllcHoPFm7q1uoNggXABpWSmapQ2PUe645LtVXzdcIDSQGFBwAAABjeczfe3+o94LmzyjZrOnz3I0kKrfeAV5zX6rNvbP2e3T0PqDoIBsECYKnJOdPjwKtvVhxs8tbN9sm56PbNMoQHAAAAFnCry5p+cMU0LdysMmhXAkw/uCK3uhzaQP7O2fe2f4/3SK3MuB6feGtrO8JFsABEaOqoaY64uCDJNxUH3rrk1UzPA6oOthAeAAAAWCDK3gMtd1rL5z6gKaWFCBaAEbhw0TwuXTNTFbKuCQ7a2yGJ8AAAAMAKNvQeSFNTyiQgWAACkp8wqyqUV0yPg8k5Kg72QHgAAABgAXoPIEj9BgvLa59rubZIsIB0mzpKaNCF4/v+6H+p40xLenL58hPNzobTORgAACBuolhtAdhpkGChG4IFwH6rj0uaOf2mJM34vr/vSU54AAAAYBl6D8BGYQULp2fm9eKheYIFICKEBwAAALAOwUgyESwA8UV4AAAAAGukcUpGqVrTw2pNR4oFzRULUe9OZAgWALsRHgAAAMAa8598W4fvfqRmbvKpZpDL5z6IevcCVfU8ffjJTV2/X5LXbMrNZvXysTm9fe6Miq578A9ImV82HhIsABHqNTxgtQUAAACEyq0ua/rBFRMcbFYZtJejnH5wRW51OVFTGD785KY+vn1PBTenqfyY6s2mPr59T5L0zisvRbx39tlvVQip/2Bh56oQf3r7zzq+tjtY2AoYCBaAnhAeAAAA4EDD9CoYW7+rTGtdnjvbsb2ZLcj1Hmls/W5iwoNStabr90squDlNbFYZ5DIZSdL1+yWVqrVUT2HoF8ECYA/CAwAAAOwriF4FjfETamXGlW3WtioOJCnbrKmVGVdj/ERYuz9yD6s1ec2mpvJjHdvz2azKjYYeEh4EhmABGC3CA8BSNFkCANjguRvvb/Uq8NxZZZs1Hb77kST13KvAK85r9dk3tv7f7p4HSak6kKQjxYLcbFb1ZnOr4kCS6s2m3ExWR/hMH4mDgoXbaz/Xl6s3CBaAPhAeAJahyRIAwBZB9iq4c/a97f/nPVIrM67HJ97a2p4Uc8WCXj42t9XjIL8ZJNS8DZ0/eZwbAhZoBwsvHf6Np75GsBBj5RWpUpIm56Spo1HvTSIRHgCWockS0DsqdPrHa4Z+BNmroOVOa/ncB0P1ToiLt8+dkWR6HJQbDbmZrM6fPL61HfaKOlg4PTOvUzMnCRb6UV+Trl6Slq5JTU/KutILr0oXLkr5iaj3LlEIDwCL0GQJ6A0VOv3jNcMgwuhV4BXnExsatBVdV++88hJhXcIQLFjq6iVpcUFyi1JhWvLq5t+S9Pq70e5bwhAeABahyRLQGyp0+sdrhkGkqVdBGOYIDVKDYCEi5RVTceAWt6sM8jlJvtleXmEKQ4AIDwCL0GQJOBgVOv3jNcMw0tKrAAhL78HCTS2vfUaw0I9KyUxVKOxa+cUdl2qr5uuEB4EhPMDQKMkLDk2WgINRodM/XjMMI029CoBRI1gY0uSc6XHg1TcrDjZ562b75Fx0+5ZAhAcYGPNnw0GTJaA7KnT6x2uGIKShVwFgk1EGCycnn9WLh56PX7AwddQ0R1xckOSbigNvXfJq0qnXqDoIGOEBBsb82XDQZAnojgqd/vGaAUCyBB0sLFdWtFxZiWewcOGieVy6ZqYqZF0THLS3IzCO7/uj/6WOMy3pyeXLTzQ7O33g98M+pWpN7/2rH8lxtDV/VpLWPE++L73/t7/OxSiAroYJyDoqn1rm7jmVT93xmgEA9goWbte+0J31L1RuPunrZ1kXLJRXTI+DyTkqDvq0+rikmdNvStKM7/tPJ0ybCA8wkBtfPdIHH/9UU/mxjhLYjVZL5UZD3/76Kzr7zGyXnwAgrYKc8kSFTv94zQAAe0l0sICueg0PmLaAgTB/FqPGgCc5gpzyxDJo/eM1AwDspd+pEN2ChV6nQpyeOdkRMBAs2I3wAANh/ixGhcacycKSgQAAxA/BAiTCAwyBVQEwCjTmTBaWDAQAIFkIFtKD8AADY1UAhI271MnDlCcAANKjW7DwpPFIy2ufEyzECOEBhsb8WYSFu9TJw5QnABg9t7qssfW7aoyfkFecj3p3AEnSzNisZsZ+g2AhRggPAFiLu9TJxJQnABiNjLeq5268r+kHV5RprauVGdfqs2/oztn31HJZ8Qz2ijJYaK8MQbDwNJZqBGC1P/zpp1s9D3bfpabnQbwx5QkAwjX/ybd1+O5HauYm1cwWlG3WlN2o6PGJt7R87oOodw8IXL/BQjdpChZ6XaqR8ACA1TpWW2iZigNWWwAAoDu3uqxf/fi3JDlq7qgyyHqrknx9dv57TGFAqhAs7K/X8IBpCwCsRmPO0eD1BcLBuYWojK3fVaa1Ls+d7djezBbkeo80tn6X8ACp0stUiF+s3tTS6g2mQuyD8ABALNCYMxwdlR3NptwslR1AEDi3ELXG+Am1MuPKNmtqZraPuWyzplZmXI3xExHuHWAXG4KFX5l5ToXceGjPMQiEBwCQYh9+cnOrp8RUfkz1ZnNrJQR6SgCD49xC1LzivFaffUOH734kSU/1PKDqAOgNwcI2wgMASKlStabr90squDlNbN4Jba9qcf1+SSWWwgQGwrkFW9w5+54kafrBFbneI7Uy43p84q2t7QCGk7ZggfAAAFLqYbUmr9nUVH6sY3s+m1W50dBDBjjAQDi3YIuWO63lcx/IrS5rbP2uGuMnqDgARqS/YOFz3a4tWh8sEB4AQEodKRbkbi5/2b4rKkn1plnV4giDG2AgnFuwjVecJzQALBLXYIHwAABSaq5Y0MvH5rbmYec3Bzs1b0PnTx7nzigwIM4tAMCgRh0snD40rxdyz/S0b4QHAJBib587I8nMwy43GnIzWZ0/eXxrO4DBcG4BAIIWVrCg9d5+v+P7/rDPoW+O40xLenL58hPNzk6P/PcDADqxFj0QDs4tAEDUdgYLt8o3dKuyK1hYl/RfSpJmfN9f3e/nEB4AALCJgR4AAEiTJ41H+ovlP9d/+O++IR0QHjBtAQCQelXP04ef3NT1+yV5zabcbFYvH5vT2+fOqLi51B4waoRZAICwzYzN6uz0r/f0vYQHAIDU+/CTm/r49j0V3Jym8mOqN5tbze7eeeWliPcOaUOYBQCwUebgbwEAILlK1Zqu3y+p4OY04brKZTKacF0V3Jyu3y+pVK1FvYtImXaY5TjSVH5MjiN9fPuePvzkZtS7FrlStaYbXz3ivASACFB5AABItYfVmrxmU1P5sY7t+WxW5UZDD6s1SsYxMrvDLEnKZcy9nnaYlcbjkWoMAIgelQcAgJGw9Y7hkWJBbjarerPZsb3ebMrNZHUkhQM1RKcdZuWz2Y7t+WxWXquph5adP6NCNQYARI/KAwBAqGy/YzhXLOjlY3NbPQ7ym0FCzdvQ+ZPHU3mXF9HZGWa1Kw6kdIdZVGMAgB2oPAAAhCoOdwzfPndG508el+9L5UZDvi+dP3lcb587E/WuIWXaYVbN29Ca52mj1dKa56nmbejlY3OpHCRTjQEAdqDyAAAQmrjcMSy6rt555SWWxoMV2qHV9fsllRsNuZlsqsMsqjEAwA6EBwCA0MStGeEcoQH2MOpQiTCrE1OLAMAOhAcAgNBwxxBxFnW/DsKsbVRjAED0CA8AAKHhjiHirN2vo+DmNJUfU73Z3DqW33nlpYj3Ll32q8YoVWu69aSc+uoMABgFwgMAQKi4Y4g4iku/jrRpV2NUPU9/+NNPrV3FBQCSiPAAABAq5m8jjuLWryNtqAoBgNEjPAAAjERa528TmsQT/TrsRVUIAESD8AAAgBBE3WwPwxm2XwehUXioCokexzeQToQHAAAEqH1RffmLW/rzzbujlFXH0yD9OgiNwkdVSHQ4voF0IzwAACAAOy+q1zc2VGl4GstmNVvIKeM4lFXH0CD9OpiLHz5WcYkOxzeQbpmDvwUAABykfVHtONJ4LidfUn1jQw9r61vfk89m5bWaelitRbej6Es/wcHuufi5TEYTrquCm9sKjRCMt8+d0fmTx+X7UrnRkO+LVVxCxvENgMoDAACGtPui2mu1lHEctSRVvQ15rZbcTIay6hgZpDybufijwyouo8fxDYDKAwBAoErVmm589ShVd6HaF9X5bFaS5GYyKuZMPt9qtVTfaGrN81TzNvTysTkusGNgZyXJVH5MjiN9fPuePvzk5r7/Z+dc/J0IjcIzVyzo7DOznFMjwPENgMoDAEAg0txIa68GbkeK4/IqTTWaLdU3NpTP5SirjolBlwJkLj6SjOMbAOEBACAQaW6ktd9F9Vg2q1dPHNPfOfU8ZdUxMkx59iArNABxwfENpBvhAQBgaIPeqU2SbhfVNlVeMEf8YMMsBchcfCQZxzeQboQHAICh0UjL/ovqNE8r6dfuSpKs46i2sSGv2dI35k/09Heds+zvDwSJ4xtIJ8IDAMDQhrlTmzS2XlSneVrJIN4+d0Zeq6mry/c3G8T5ymez2mi1VPU8AhcAQOqw2gIAYGjtO7U1b0NrnqeNVovVBSzC+uz9K7qu3ExWuYyjQ+NjOj45oZnxvP7s7krXFRcAAEgqwgMgBtK49B3i5+1zZ3T+5HH5vlRuNOT7opGWJXYvJdmWz2bltZp6yHvLU9qBy8SYq8Pj4xrP5QhcAACpxrQFwGLMUUac2D7nP82YVtI/+ngAANCJygPAYu05yo4jTeXH5DjSx7fvUTILq80VCzr7zCwDqx2irh6yeVpJ1K/NfnYGLjsRuAAA0orKA8BSLH0HxJ9N1UO2rc9u02uzl90rLuQ3g4Sat6HzJ4/z/gsASB3CA8BSlMwC8WfTCge2TSux6bXZj22BCwAAUSI8ACzFHGWkhS2D2aDZWj1kw1KStr42u0UZuCT1vAAAxBfhAWApSmb3x0V1Mthetj4sqof2N+hrE9W5P8rAJennBQAgvggPAItRMtuJi+pkiUPZ+jDSXj3UbaDf72vTz7kf93Ax6ecFACC+CA8Ai9k2RzlqXFQnR1zK1oeR1uqhXgb6/b42vZz7SQgX03BeAADii6UagRhg6bunL6pzmYwmXFcFN7d1UY34aJet57PZju35bFZeq6mHMfx77rXk4Nvnzuj8yePyfancaMj3lfjqoV6XmO31ten13E/C0rZJPC8AAMlB5QGAWGD+eLIkqaT/oDveaaoe6ufOea+vTS/nfvvnx/2OfZLOCwBA8lB5ACAWdl5U78RFdTy1y9Zr3obWPE8brZbWPE81b0MvH5uLxUCvrZc73rZWD+1VLTGMQe6cH/TaQS0C8gAAF0ZJREFU9HLux+2O/X6vu23nRdDHBwAg3qg8ABALaZ0/nmRJaAga1znqYfUHCOPOea/nfhzu2PfyuttwXiShfwQAIHiEBwBiw4aLagQnCSX9cZ1OE1bz0bBCvoPO/biEi7287qM4Lw762TSnBQDshfAAQGwkYbCJp83F+O8YxznqYVdLhBHy9XLu2x4u9vu6h3Fe9FJRENdqGgBA+AgPAMRO+8K1PY+ZC1lEJS53vHcKu1oizJCv24Da9nDRhiqVXioKbNhPAICdCA8AxApzcWEb2+947zaqaomoKkpsrWSJukql14qCqPcTAGAvwgMAscJcXNjG9jveu8WxWiIJon7de60oiHo/AQD2IjwA8BRbB0HMxYXNbL3jvZe4VUskRZSvez8VBRwfAIC9EB4A2GL7lADm4gLBiFu1RFJE+br3U1HA8QHYiXMSUSM8ALDF9ikBzMXtT5IuMpL0XGwSp2qJJInqde+lomD3ucbxAUTP9ps7SA/CAwCS4jElgLm4vUnSRUaSngsQtW4VBZxrgL1sv7mD9Mgc/C0A0qA9JSCfzXZsz2ez8lrNrWURo/b2uTM6f/K4fF8qNxryfSV6Lm6pWtONrx6p1Mfr377IcBxpKj8mx5E+vn1PH35yM8Q9DUeSngtgi7liQWefme0IXDnXADvtvrmTy2Q04boquLmtmzvAqFB5AEBSfKYEpGUu7qB3AeNQQdKrJD0XxFvS32841wB70e8JNiE8ACApflMCkj4Xd9ASxSRdZCTpuSCe0lLKz7kG2CsuN3eQDkxbALAlbVMCbDVMieLOi4yd4niRkaTngnhKSyk/5xpgr/bNnZq3oTXP00arpTXPU83b0MvH5gj2MFJUHgDYkpYpAVHq5bUd5i5g3CpIujHzsg/r/737QM2Wr6Kbi+1zQfykqZQ/Se8bQBL1slIKMAqEBwCekvQpAVHop/x52BLFJFxktF+vvyo9ktdqqlbb0Go9o5n8WOyeC+IpbaX8SXjfAJKKmzuwBeEBAIxAPz0Mhr0LGJeLjG77t/P1OjY5oarnqept6K8/M8uyVBiJtM0zjsv7BpBm3NxB1AgPACBkg5Q/B3EX0NaLjIOqMPZ6vabzeWUzGd14+DhR5eKwV1pL+W193wAARI/wAABCNkj5c5LvAh5UhZG2cnHYi1J+AAC2ER4AQMiGKX9O2l3AXqow0lYuDnslOcQDAKBfLNUIACFjmaVt7aqCfDbbsT2fzcprNbeqCni9YBOz8scsxx4AINWoPACAEaD82ei1qoDXCwAAwC6EBwAwApQ/G702oeP1AgAAsAvhAQCMUNJ6GHSz38C/n6qCNL1eAAAANiM8AAAE6qClGKkqAAAAiB/CAwBAoA5airGNqgIAAID4YLUFAEBgdi/FmMtkNOG6Kri5raUYAQAAED+EBwCAwPSyFCMADKNUrenGV48IIwFgxJi2AAAITK9LMSIY9I1AmhzUTwUAEC7CAwBAYHpdihHDYRCFNOq1nwoAIBxMWwAABOrtc2d0/uRx+b5UbjTk+9p3KUYMpj2IchxpKj8mx5E+vn1PH35yM+pdA0JBPxUAiB6VBwCAQLEUY7h2D6IkbU0RaQ+ieL2RNO1+KlP5sY7t+WxW5UZDDznuASB0hAcAgFCwFGM4GEQhjeinAgDRY9oCAABDGHXn952DqJ0YRCHJ2v1Uat6G1jxPG62W1jxPNW9DLx+bIzADgBGg8gAAgAFE1bSQppRIq3bflOv3Syo3GnIzWfqpAMAIER4AADCAsDq/99IrgkEU0oh+KgAQLcIDAAD6FEbTwn4qGRhEIc3opwIA0aDnAYDEGvVcdKRHu2lhPpvt2J7PZuW1mno4wDE3yPKLc8WCzj4zy0AKAACEjsoDAIkT1Vx0pEfQnd9ZfhEAANiOygMAiTPIHVygH0F3fg+jkgEAACBIVB4ASBTb7uAyJz25gmxayBr2AADAdoQHABKlfQd3Kj/WsT2fzarcaOjhiMIDpk6MRlThTPv3/vbZU/rts6eG3geWXwQAALYjPACQKLbcwQ1rGT8YUYUzYf5ell8EAAA2IzwAkCg23MG1bepEEkUVzoT5e1l+EQAA2IzwAEDiRH0H15apE0kVVTgzqt/LGvYAAMBGhAcAEifqO7i2TJ1IqqjCGUIhAACQZizVCCCx5ooFnX1mduQDuqCX8UOnneHMTmGHM1H9XgAAABsQHgBACN4+9/+3dz+/caRpHcC/5XKl1Q7ZFZNtyWFz4JZFJAeHUw5IcGG1J4TEBeXAKeLIGXHggED8EdyQJSSQkEAI7UoIcfJhBUZYWo0PXJB37VWPg2ajdE9cbheHibPx7KQSO91d/ePzOVYi9Zspj7vebz3v8zzIk/v30jTJi7OzNE00v5uSrsKZts/9zrd+OaejcYaj8Uw+GwCga44tAMxA10cnVl1XfS2++rmbxUa+0buVHw2f5+CnnxnJCQCsrKJpmvl/aFF8I8nnP/jB5/nkk2/M/fMBWA1dhTOXn/uD//nf/NdPv2yi+NXJHkZyAgDLYDj8Wb73vW8myTebpvnZu/6eygMAllZXkwkuP/PT0/8zkhMAWAt6HgDADVxOX+iV5ZXrvbJMfTHJqf4HAMAKER4AwA2YvkCXhqNxPv3suSadAMyNYwswQ5rlweq6nL6wd3ScJL/Q88D/88zCqK6ze3CY/ZNh6slEk04A5kZ4ADPg4Q7WQ1dTH1hfuweH2Ts6Tr/azJ3erbyaTN4EWJp0AjBLwgOYAQ93sB7mPZJTNdNymfb9Go7G2T8ZLmSTTj+bAKtPeABTtsgPd+vMgy2zNOupD6qZlsus7tdlk847vVtXrvfKMi/OznLawfeLn02A9SE8gClbxIe7debBllUw72omYdvHmdX9ertJ52UonXTbpFOlHcD6EB7AlC3iw92sLfJGw4Mty26e1UzCto83y/u1aE06VdoBrBfhAUzZoj3czdKibzQ82LIK5lnNJGz7eLO+X4vUpFOlHcB6ER7ADCzSw90sLfpGw4Mtq2Be1UzCtumY9f2ad5PONutYaQewzoQHMAOL9HA3K8uw0fBgyyqYVzWTsG065nW/Zt2k80PXsC6VdgAkG+//K8BNDbb6+c63PlnJB6jLjUavLK9c75Vl6otJTkfjjlb2c5cPtuP6PC/rOucXF3lZ1xnX59nZHqzkfWE1PX30IE/u30vTJC/OztI0mXo109th29uEbdc3j/u1KNbp3wqw7lQeADeyLG/11+UICattHtVM3iJPzzpUn11ap38rwLoTHgA3siwbDQ+2rJJZl6oL26ZrEY4WzMs6/VsB1pXwALixZdpoeLCF9xO2AQDvIjwAbsxGA1aTsA1g/jxPseiEB8BHs9EAYFnZsNG1UV1n9+Aw+yfD1JNJqrLMzvYgTx89yNbriVawCIQHAACsHRs2FsXuwWH2jo7TrzZzp3crryaTNz2lnj1+2PHq4OeMagRg6oajcT797HmGCzCyE+DrXG7YiiK507uVokj2jo6ze3DY9dJYI8PROPsnw/SrzdyuqmxubOR2VaVfbWb/ZOh7lIWi8gCAqfEmD1gGX92wJXkzdvhyw+YIA/NwOhqnnkxyp3fryvVeWebF2VlO/SyyQFQeADA13uQBy+Byw9YryyvXe2WZ+mKSU297mZO7W/1Ur8ddv+3VZJJqo8xdwQELRHgAwFQovQSWhQ0bi2Kw1c/O9iDj+jwv6zrnFxd5WdcZ1+fZ2R6oOmChCA8AmApv8oBlYcPGInn66EGe3L+XpklenJ2laZIn9+/l6aMHXS8NrtDzAICpePtN3uXZ4cSbPGAxXW7M9k+GeXF2lmqjtGGjE1tVlWePHxobysITHgAwFZdv8i7HS/VeBwnj+jxP7t/zIAQsFBs2Fs3AzyALTngAwNR4kwcsGxs2gA8jPABgarzJA9aF33PAuhEeADB13uQBq2pU19k9OMz+yTD1ZJKqLLOzPcjTRw+yVVVdLw9gZkxbAABYEcPROJ9+9txo1BnaPTjM3tFxiiK507uVokj2jo6ze3DY9dIAZkrlAQDAkvM2fD6Go3H2T4bpV5u5/fq/6+V0mf2TYYajsaorYGWpPAAAWHLehs/H6WicejJJryyvXO+VZeqLSU5VfAArTHgAALDEvvo2fHNjI7erKv1q883bcKbj7lY/1esxtG97NZmk2ihzV9UBsMKEB8DCcnYX4P28DZ+fwVY/O9uDjOvzvKzrnF9c5GVdZ1yfZ2d74MgCsNL0PAAWjrO7wDSt+ki9t9+GX56/T1bjbfgi3runjx4k+bLHwYuzs1QbZZ7cv/fmOsCqEh4AC+fy7G6/2syd3q28mkyyd3ScJHn2+GHHqwOWxboEkZdvwy9/T/ZeBwnj+jxP7t9bmE33dSzyvduqqjx7/HAhgw2AWXJsAVgozu4C07JOTQSfPnqQJ/fvpWmSF2dnaZos9dvwZbh3g61+vvOtTwQHwNpQeQAslMuzu3d6t65c75VlXpyd5dQYLOADrNtIvVV6G75u9w5gWag8AD7KtJsa6mQNTMO6NhFchbfh63rvbkpzYWBeVB4ANzKr86ireHYXmL9VbiK46ty7D7PIfSGA1aTyALiRWZ5HXbWzu8D8Gam3vNy7D7MMfSGA1aLyALi2WZ9HXaWzu0B3jNRbXu5dO30hgC4ID4Brm1dTw4HQAPgIgsjl5d6101wY6ILwAD7Cuj7UOI8KLBNB5PJy776e72GgC8IDuIF1b1KkqSEAdMf3MNAFDRPhBjQp0tQQALrkexiYN5UHcE2aFH3JeVQA6I7vYWDehAdwTZoUXeU8KqvMQzmw6HwPA/MiPIBr0qQIVt+69zUBAPgqPQ/gmi6bFI3r87ys65xfXORlXWdcn2dneyD9hxWgrwnrajga59PPnmc4Gne9FAAWjMoDuIHLZkT7J8O8ODtLtVFqUgQrQl8T1pFqGwDeR3gAN6BJEawufU1YR5fVNv1qM3d6t/JqMnkzBvDZ44cdrw6ARSA8gI+gSRGsHn1NWDeqbQD4EHoeAMBb9DVh3VxW2/TK8sr1XlmmvpjkVP8DAKLyAAB+gb4mrBPVNgB8COEBAHyFviask8tqm8seB73XQcK4Ps+T+/f87AOQRHgAAO+krwnrQrUNAO8jPAAAWHOqbQB4H+EBAABJVNsA8G6mLQAAAACthAcAAABAK+EBAAAA0Ep4AAAAALQSHgAAAACthAcAAABAK+EBAAAA0Ep4AAAAALQSHgAAAACthAcAAABAK+EBAAAA0Ep4AAAAALQSHgAAAACthAcAAABAq82uFwAA62g4Gud0NM7drX4GW/2ulwMA0Ep4AABzNKrr7B4cZv9kmHoySVWW2dke5OmjB9mqqq6XBwDwtRxbAIA52j04zN7RcYoiudO7laJI9o6Os3tw2PXSAADeSXgAAHMyHI2zfzJMv9rM7arK5sZGbldV+tVm9k+GGY7GXS8RAOBrCQ8AYE5OR+PUk0l6ZXnleq8sU19Mcio8AAAWlPAAAObk7lY/VVnm1WRy5fqrySTVRpm7GicCAAtKeAAAczLY6mdne5BxfZ6XdZ3zi4u8rOuM6/PsbA9MXQAAFpZpCwAwR08fPUiS7J8M8+LsLNVGmSf37725DgCwiIQHADBHW1WVZ48fZjga53Q0zt2tvooDAGDhCQ/ghjz4Ax9j4HcHALBEhAdwTaO6zu7BYfZPhqknk1RlmZ3tQZ4+epCtqup6eQAAAFOnYSJc0+7BYfaOjlMUyZ3erRRFsnd0nN2Dw66XBgAAMBPCA7iG4Wic/ZNh+tVmbldVNjc2cruq0q82s38yzNCMdgAAYAUJD+AaTkfj1JNJemV55XqvLFNfTHIqPAAAAFaQ8ACu4e5WP1VZ5tVkcuX6q8kk1UaZu5qfAQAAK0h4ANcw2OpnZ3uQcX2el3Wd84uLvKzrjOvz7GwPdE4HAABWkmkLcE1PHz1IkuyfDPPi7CzVRpkn9++9uQ4AALBqhAdwTVtVlWePH2Y4Gud0NM5ds9oBAIAVJzyAGxoIDQCYISE1AItEeAAAsEBGdZ3dg8PsnwxTTyapyjI724M8ffQgW1XV9fIAWFMaJgIALJDdg8PsHR2nKJI7vVspimTv6Di7B4ddLw2ANSY8AABYEMPROPsnw/SrzdyuqmxubOR2VaVfbWb/ZJjhaNz1EgFYU8IDAIAFcToap55M0ivLK9d7ZZn6YpJT4QEAHREeAAAsiLtb/VRlmVeTyZXrryaTVBtl7mqcCEBHhAcAK2g4GufTz54rcYYlM9jqZ2d7kHF9npd1nfOLi7ys64zr8+xsD0xdAKAzpi0ArBBd2mH5PX30IEmyfzLMi7OzVBtlnty/9+Y6AHRBeACwQi67tPerzdzp3cqrySR7R8dJkmePH3a8OuBDbFVVnj1+mOFonNPROHe3+ioOAOic8ABgRXy1S3uSbG58eTrtsku7DQgsj4HQAIAFoucBwIrQpR0AgFkRHgCsCF3aAQCYFeEBwIrQpR0AgFnR8wBghejSDgDvphEp3JzwAGCF6NIOAL/IKGP4eMIDgBWkSzsA/JxRxvDx9DwAAABW1ldHGW9ubOR2VaVfbb4ZZQy8n/AAAABYWUYZw3QIDwAAgJVllDFMh/AAAABYWUYZw3RomAgAAKw0o4zh4wkPAACAlWaUMXw84QEAALAWjDKGm9PzAAAAAGglPAAAAABaCQ8AAACAVsIDAAAAoJXwAAAAAGglPAAAAABaCQ8AAACAVsIDAAAAoJXwAAAAAGglPAAAAABaCQ8AAACAVsIDAAAAoJXwAAAAAGglPAAAAABaCQ8AAACAVsIDAAAAoJXwAAAAAGglPAAAAABaCQ8AAACAVsIDAAAAoJXwAAAAAGglPAAAAABaCQ8AAACAVsIDAAAAoJXwAAAAAGglPAAAAABaCQ8AAACAVsIDAAAAoJXwAAAAAGglPAAAAABaCQ8AAACAVsIDAAAAoJXwAAAAAGglPAAAAABaCQ8AAACAVsIDAAAAoJXwAAAAAGglPAAAAABaCQ8AAACAVptdLwCA9TMcjXM6GufuVj+DrX7XywEA4D2EBwDMzaius3twmP2TYerJJFVZZmd7kKePHmSrqrpeHgAA7+DYAgBzs3twmL2j4xRFcqd3K0WR7B0dZ/fgsOulAQDQQngAwFwMR+PsnwzTrzZzu6qyubGR21WVfrWZ/ZNhhqNx10sEAOAdhAcAzMXpaJx6MkmvLK9c75Vl6otJToUHAAALq9OeB59//rMuPx6AOSq/qFOcfZGXZ1+k/1Z/g3Fdp3j958+f+14AAJinD92XF03TzHgpX/OhRfHtJEdz/2AAAADg69xvmubH7/rDrsKDIsmvJHkx9w8HAAAA3nYnyU+aloCgk/AAAAAAWB4aJgIAAACthAcAAABAK+EBAAAA0Ep4AAAAALQSHgAAAACthAcAwHsVRfEHRVF8URTFt9+69tdFUfx3URTf7HJtAMDsCQ8AgA/xt0kOk/xJkhRF8WdJvpvke03TfN7lwgCA2dvsegEAwOJrmqYpiuJPk/x9URQ/SfLHSX6zaZofJ0lRFP+Q5LeS/GvTNL/f3UoBgFkomqbpeg0AwJIoiuI/k/x6kt9pmubf37r+20l+KckfCg8AYPU4tgAAfJCiKL6b5DtJyiQ/ffvPmqb5tyQvulgXADB7wgMA4L2Konic5O+S/FGS7yf5825XBADMk54HAECroih+Nck/J/mrpmn+piiKHyX5YVEUv9E0zX90ujgAYC5UHgAA71QUxSdJ/iXJPzZN85dJ8jow+Kckf9Hl2gCA+VF5AAC8U9M0z5P82tdc/90OlgMAdMS0BQDgoxVF8f0kj5PcTvI8ye81TfPDblcFAEyL8AAAAABopecBAAAA0Ep4AAAAALQSHgAAAACthAcAAABAK+EBAAAA0Ep4AAAAALQSHgAAAACthAcAAABAK+EBAAAA0Ep4AAAAALQSHgAAAACt/h/Q//kHoRbIiwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_data(X,Y,grad_hist)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Processing Frames: 100%|█████████████████████████████████████████████████████████████| 201/201 [00:56<00:00,  3.57it/s]\n",
      "Appending Frames: 100%|██████████████████████████████████████████████████████████████| 201/201 [00:07<00:00, 26.66it/s]\n",
      "Writing Video: 100%|████████████████████████████████████████████████████████████████| 201/201 [00:01<00:00, 122.29it/s]\n",
      "Writing GIF: 100%|███████████████████████████████████████████████████████████████████| 201/201 [00:37<00:00,  5.34it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total elapsed time = 103.29527115821838 seconds\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "X, Y = make_blobs(n_samples=200, centers=2, n_features=2, cluster_std=5 , random_state=11)\n",
    "\n",
    "model = LogisticRegression( iterations = 200, learning_rate = 0.3) \n",
    "model.fit( X,Y)  \n",
    "grad_hist = model.grad_hist\n",
    "Y_pred = model.predict(X)\n",
    "\n",
    "\n",
    "frame_save_path = 'logistic frames'\n",
    "video_path = 'logistic.mp4'\n",
    "gif_path = 'logistic.gif'\n",
    "fps = 20\n",
    "\n",
    "if not os.path.exists(frame_save_path):\n",
    "    os.mkdir(frame_save_path)\n",
    "\n",
    "\n",
    "def main(play=True):\n",
    "    \n",
    "    start = time.time()\n",
    "    process_frames(X,Y,grad_hist,frame_save_path)\n",
    "    write_video(video_path,frame_save_path,fps)\n",
    "    write_gif(gif_path,frame_save_path,fps)\n",
    "    end = time.time()\n",
    "    \n",
    "    print(f\"Total elapsed time = {end-start} seconds\")\n",
    "    \n",
    "    if play:\n",
    "        play_video(video_path,fps)\n",
    "        \n",
    "    return None\n",
    "\n",
    "\n",
    "if __name__== '__main__':\n",
    "    main()"
   ]
  },
  {
   "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}