deep_02.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1 : 양성종양\n",
      "2 : 양성종양\n",
      "3 : 양성종양\n",
      "4 : 양성종양\n",
      "5 : 양성종양\n",
      "6 : 양성종양\n",
      "7 : 악성종양\n",
      "8 : 악성종양\n",
      "12 : 악성종양\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as  plt\n",
    "import numpy as np\n",
    "\n",
    "class Neuron:\n",
    "    \n",
    "    def __init__(self):\n",
    "        self.w = 1.0     # 가중치를 초기화합니다\n",
    "        self.b = 1.0     # 절편을 초기화합니다\n",
    "    \n",
    "    def forpass(self, x):\n",
    "        y_hat = x * self.w + self.b       # 직선 방정식을 계산합니다\n",
    "        return y_hat\n",
    "    \n",
    "    def backprop(self, x, err):\n",
    "        w_grad = x * err    # 가중치에 대한 그래디언트를 계산합니다\n",
    "        b_grad = 1 * err    # 절편에 대한 그래디언트를 계산합니다\n",
    "        return w_grad, b_grad\n",
    "\n",
    "    def activation(self, z):\n",
    "        return z\n",
    "    \n",
    "    def fit(self, x, y, epochs=200):\n",
    "        for i in range(epochs):           # 에포크만큼 반복합니다\n",
    "            for x_i, y_i in zip(x, y):    # 모든 샘플에 대해 반복합니다\n",
    "                y_hat = self.forpass(x_i) # 정방향 계산\n",
    "                y_hat = self.activation(y_hat)\n",
    "                err = y_hat - y_i      # 오차 계산\n",
    "                w_grad, b_grad = self.backprop(x_i, err)  # 역방향 계산\n",
    "                self.w -= 0.01*w_grad          # 가중치 업데이트\n",
    "                self.b -= 0.01*b_grad          # 절편 업데이트\n",
    "\n",
    "# x = np.array([1,2,3,4,5,6,7,8])\n",
    "# y = np.array([0,0,0,0,1,1,1,1])\n",
    "x = np.array([1,2,3,4,5,6,7,8,12])\n",
    "y = np.array([0,0,0,0,1,1,1,1,1])                \n",
    "\n",
    "neuron = Neuron()\n",
    "neuron.fit(x, y)\n",
    "\n",
    "for xi, yi in zip(x,y):\n",
    "    plt.plot(xi,yi,\"rx\")\n",
    "\n",
    "y_temp = []\n",
    "for x_i in x:\n",
    "    y_hat = neuron.forpass(x_i)\n",
    "    if( y_hat >= 0.5 ):\n",
    "        print(\"%d : 악성종양\"%x_i)\n",
    "    else:\n",
    "        print(\"%d : 양성종양\"%x_i)\n",
    "    y_temp.append(y_hat)\n",
    "\n",
    "plt.plot(x,y_temp)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 오즈비 그래프"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'numpy.ndarray'>\n",
      "1\n",
      "(100,)\n",
      "[0.   0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1  0.11 0.12 0.13\n",
      " 0.14 0.15 0.16 0.17 0.18 0.19 0.2  0.21 0.22 0.23 0.24 0.25 0.26 0.27\n",
      " 0.28 0.29 0.3  0.31 0.32 0.33 0.34 0.35 0.36 0.37 0.38 0.39 0.4  0.41\n",
      " 0.42 0.43 0.44 0.45 0.46 0.47 0.48 0.49 0.5  0.51 0.52 0.53 0.54 0.55\n",
      " 0.56 0.57 0.58 0.59 0.6  0.61 0.62 0.63 0.64 0.65 0.66 0.67 0.68 0.69\n",
      " 0.7  0.71 0.72 0.73 0.74 0.75 0.76 0.77 0.78 0.79 0.8  0.81 0.82 0.83\n",
      " 0.84 0.85 0.86 0.87 0.88 0.89 0.9  0.91 0.92 0.93 0.94 0.95 0.96 0.97\n",
      " 0.98 0.99]\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "probs = np.arange(0, 1, 0.01)\n",
    "print(type(probs))\n",
    "print(probs.ndim)\n",
    "print(probs.shape)\n",
    "print(probs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "probs = np.arange(0, 1, 0.01)\n",
    "odds = [p/(1-p) for p in probs]\n",
    "# print(odds)\n",
    "plt.plot(probs, odds)\n",
    "plt.xlabel('p')\n",
    "plt.ylabel('p/(1-p)')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 로짓 그래프"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "probs  = np.arange(0.001, 0.999, 0.001)\n",
    "logit = [np.log(p/(1-p)) for p in probs]\n",
    "plt.plot(probs, logit)\n",
    "plt.xlabel('p')\n",
    "plt.ylabel('log(p/(1-p))')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 시그모이드 그래프"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "zs = np.arange(-10., 10., 0.1)\n",
    "gs = [1/(1+np.exp(-z)) for z in zs]\n",
    "plt.plot(zs, gs)\n",
    "plt.plot(zs, (zs/20+0.5))\n",
    "plt.xlabel('z')\n",
    "plt.ylabel('1/(1+e^-z)')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 로지스틱 회귀 구현"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1 : 양성종양\n",
      "2 : 양성종양\n",
      "3 : 양성종양\n",
      "4 : 양성종양\n",
      "5 : 악성종양\n",
      "6 : 악성종양\n",
      "7 : 악성종양\n",
      "8 : 악성종양\n",
      "12 : 악성종양\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as  plt\n",
    "import numpy as np\n",
    "\n",
    "class Neuron:\n",
    "    \n",
    "    def __init__(self):\n",
    "        self.w = 1.0     # 가중치를 초기화합니다\n",
    "        self.b = 1.0     # 절편을 초기화합니다\n",
    "    \n",
    "    def forpass(self, x):\n",
    "        z = x * self.w + self.b       # 직선 방정식을 계산합니다\n",
    "        return z\n",
    "    \n",
    "    def backprop(self, x, err):\n",
    "        w_grad = x * err    # 가중치에 대한 그래디언트를 계산합니다\n",
    "        b_grad = 1 * err    # 절편에 대한 그래디언트를 계산합니다\n",
    "        return w_grad, b_grad\n",
    "\n",
    "    def activation(self, z):\n",
    "#         a = z;\n",
    "        a = 1/(1+np.exp(-z))\n",
    "        return a\n",
    "    \n",
    "    def fit(self, x, y, epochs=200):\n",
    "        for i in range(epochs):           # 에포크만큼 반복합니다\n",
    "            for x_i, y_i in zip(x, y):    # 모든 샘플에 대해 반복합니다\n",
    "                z = self.forpass(x_i) # 정방향 계산\n",
    "                a = self.activation(z)\n",
    "                err = a - y_i       # 오차 계산\n",
    "                w_grad, b_grad = self.backprop(x_i, err)  # 역방향 계산\n",
    "                self.w -= 0.01*w_grad          # 가중치 업데이트\n",
    "                self.b -= 0.01*b_grad          # 절편 업데이트\n",
    "\n",
    "# x = np.array([1,2,3,4,5,6,7,8])\n",
    "# y = np.array([0,0,0,0,1,1,1,1])\n",
    "x = np.array([1,2,3,4,5,6,7,8,12])\n",
    "y = np.array([0,0,0,0,1,1,1,1,1])                \n",
    "\n",
    "neuron = Neuron()\n",
    "neuron.fit(x, y, epochs=1000)\n",
    "\n",
    "for xi, yi in zip(x,y):\n",
    "    plt.plot(xi,yi,\"rx\")\n",
    "\n",
    "for x_i in x:\n",
    "    y_hat = neuron.forpass(x_i)\n",
    "    a = neuron.activation(y_hat)\n",
    "    if( a >= 0.5 ):\n",
    "        print(\"%d : 악성종양\"%x_i)\n",
    "    else:\n",
    "        print(\"%d : 양성종양\"%x_i)\n",
    "    y_temp.append(a)\n",
    "\n",
    "x = np.arange(0,x[-1],0.1)\n",
    "y_temp = []\n",
    "  \n",
    "for i, x_i in enumerate(x):\n",
    "    y_hat = neuron.forpass(x_i)\n",
    "    a = neuron.activation(y_hat)\n",
    "    y_temp.append(a)    \n",
    "    \n",
    "plt.plot(x,y_temp)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 로그 그래프 "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.          1.11111111  2.22222222  3.33333333  4.44444444  5.55555556\n",
      "  6.66666667  7.77777778  8.88888889 10.        ]\n",
      "[0 1 2 3 4 5 6 7 8 9]\n"
     ]
    }
   ],
   "source": [
    "a = np.linspace(0, 10, 10)\n",
    "print(a)\n",
    "\n",
    "b = np.arange(0,10,1)\n",
    "print(b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-22-8640040ed6e9>:3: RuntimeWarning: divide by zero encountered in log\n",
      "  gs = [np.log(z) for z in zs]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "zs = np.arange(0.00, 10., 0.01)\n",
    "print(len(zs))\n",
    "gs = [np.log(z) for z in zs]\n",
    "plt.plot(zs, gs)\n",
    "plt.xlim(0,1)\n",
    "plt.ylim(-5,0)\n",
    "plt.xlabel('z')\n",
    "plt.ylabel('(np.log(z)')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "zs = np.arange(0.01, 10., 0.01)\n",
    "gs = [-np.log(z) for z in zs]\n",
    "plt.plot(zs, gs)\n",
    "plt.xlim(0,1)\n",
    "plt.ylim(0,5)\n",
    "plt.xlabel('z')\n",
    "plt.ylabel('(np.log(z)')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-24-f9b7ab8d5399>:2: RuntimeWarning: divide by zero encountered in log\n",
      "  gs = [-np.log(1-z) for z in zs]\n",
      "<ipython-input-24-f9b7ab8d5399>:2: RuntimeWarning: invalid value encountered in log\n",
      "  gs = [-np.log(1-z) for z in zs]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "zs = np.arange(0.01, 10., 0.01)\n",
    "gs = [-np.log(1-z) for z in zs]\n",
    "plt.plot(zs, gs)\n",
    "plt.xlim(0,1)\n",
    "plt.ylim(0,5)\n",
    "plt.xlabel('z')\n",
    "plt.ylabel('-(np.log(1-z)')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(569, 30) (569,)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.datasets import load_breast_cancer\n",
    "cancer = load_breast_cancer()\n",
    "print(cancer.data.shape, cancer.target.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[1.799e+01 1.038e+01 1.228e+02 1.001e+03 1.184e-01 2.776e-01 3.001e-01\n",
      "  1.471e-01 2.419e-01 7.871e-02 1.095e+00 9.053e-01 8.589e+00 1.534e+02\n",
      "  6.399e-03 4.904e-02 5.373e-02 1.587e-02 3.003e-02 6.193e-03 2.538e+01\n",
      "  1.733e+01 1.846e+02 2.019e+03 1.622e-01 6.656e-01 7.119e-01 2.654e-01\n",
      "  4.601e-01 1.189e-01]\n",
      " [2.057e+01 1.777e+01 1.329e+02 1.326e+03 8.474e-02 7.864e-02 8.690e-02\n",
      "  7.017e-02 1.812e-01 5.667e-02 5.435e-01 7.339e-01 3.398e+00 7.408e+01\n",
      "  5.225e-03 1.308e-02 1.860e-02 1.340e-02 1.389e-02 3.532e-03 2.499e+01\n",
      "  2.341e+01 1.588e+02 1.956e+03 1.238e-01 1.866e-01 2.416e-01 1.860e-01\n",
      "  2.750e-01 8.902e-02]\n",
      " [1.969e+01 2.125e+01 1.300e+02 1.203e+03 1.096e-01 1.599e-01 1.974e-01\n",
      "  1.279e-01 2.069e-01 5.999e-02 7.456e-01 7.869e-01 4.585e+00 9.403e+01\n",
      "  6.150e-03 4.006e-02 3.832e-02 2.058e-02 2.250e-02 4.571e-03 2.357e+01\n",
      "  2.553e+01 1.525e+02 1.709e+03 1.444e-01 4.245e-01 4.504e-01 2.430e-01\n",
      "  3.613e-01 8.758e-02]]\n"
     ]
    }
   ],
   "source": [
    "print(cancer.data[:3])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.boxplot(cancer.data)\n",
    "plt.xlabel('feature')\n",
    "plt.ylabel('value')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['mean area', 'worst area'], dtype='<U23')"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cancer.feature_names[[3,23]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1,\n",
       "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0,\n",
       "       0, 0, 1, 0, 1, 1, 1, 1, 1, 0, 0, 1, 0, 0, 1, 1, 1, 1, 0, 1, 0, 0,\n",
       "       1, 1, 1, 1, 0, 1, 0, 0, 1, 0, 1, 0, 0, 1, 1, 1, 0, 0, 1, 0, 0, 0,\n",
       "       1, 1, 1, 0, 1, 1, 0, 0, 1, 1, 1, 0, 0, 1, 1, 1, 1, 0, 1, 1, 0, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 0, 0, 1, 1, 1, 0, 0, 1, 0, 1, 0,\n",
       "       0, 1, 0, 0, 1, 1, 0, 1, 1, 0, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 0, 1, 1, 1, 1, 0, 0, 1, 0, 1, 1, 0, 0, 1, 1, 0, 0, 1, 1, 1,\n",
       "       1, 0, 1, 1, 0, 0, 0, 1, 0, 1, 0, 1, 1, 1, 0, 1, 1, 0, 0, 1, 0, 0,\n",
       "       0, 0, 1, 0, 0, 0, 1, 0, 1, 0, 1, 1, 0, 1, 0, 0, 0, 0, 1, 1, 0, 0,\n",
       "       1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 0, 1, 1, 0, 1, 1, 0, 0, 1, 0, 1, 1,\n",
       "       1, 1, 0, 1, 1, 1, 1, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
       "       0, 0, 1, 1, 1, 1, 1, 1, 0, 1, 0, 1, 1, 0, 1, 1, 0, 1, 0, 0, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 0, 1, 0, 1, 1, 1, 1, 0, 0,\n",
       "       0, 1, 1, 1, 1, 0, 1, 0, 1, 0, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 0,\n",
       "       0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 0, 0, 0, 1, 0, 0,\n",
       "       1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 0, 0, 1, 1,\n",
       "       1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 1, 1, 0,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 0, 1, 0, 1, 1, 1, 1,\n",
       "       1, 0, 1, 1, 0, 1, 0, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0,\n",
       "       1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1,\n",
       "       1, 1, 1, 0, 1, 0, 1, 1, 0, 1, 1, 1, 1, 1, 0, 0, 1, 0, 1, 0, 1, 1,\n",
       "       1, 1, 1, 0, 1, 1, 0, 1, 0, 1, 0, 0, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 0, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 1])"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cancer.target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([0, 1]), array([212, 357], dtype=int64))"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.unique(cancer.target, return_counts=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(569, 30)\n",
      "(569,)\n"
     ]
    }
   ],
   "source": [
    "x = cancer.data\n",
    "print(x.shape)\n",
    "y = cancer.target\n",
    "print(y.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "x_train, x_test, y_train, y_test = train_test_split(x, y, stratify=y, \n",
    "test_size=0.2, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(455, 30) (114, 30)\n"
     ]
    }
   ],
   "source": [
    "print(x_train.shape, x_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([0, 1]), array([170, 285], dtype=int64))"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.unique(y_train, return_counts=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(455, 30)\n",
      "30\n",
      "(30,)\n",
      "[1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.\n",
      " 1. 1. 1. 1. 1. 1.]\n",
      "(30,)\n",
      "(30,)\n",
      "1 6\n",
      "2 7\n",
      "3 8\n",
      "4 9\n",
      "5 10\n"
     ]
    }
   ],
   "source": [
    "x = np.arange(455*30).reshape(455,30)\n",
    "print(x.shape)\n",
    "print(x.shape[1])\n",
    "\n",
    "w = np.ones(x.shape[1])\n",
    "print(w.shape)\n",
    "print(w)\n",
    "\n",
    "print(x[0].shape)\n",
    "for x_i in x:\n",
    "    print(x_i.shape)\n",
    "    break\n",
    "\n",
    "a = [1,2,3,4,5]\n",
    "b = [6,7,8,9,10]\n",
    "for i,j in zip(a,b):\n",
    "    print(i,j)\n",
    "\n",
    "# a = np.zeros((5,6))\n",
    "# print(a)\n",
    "# print(a.shape)\n",
    "\n",
    "# b = np.ones((5,6))\n",
    "# print(b)\n",
    "# print(b.shape)\n",
    "\n",
    "# c = np.full((5,6), 7)\n",
    "# print(c)\n",
    "# print(c.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1 2 3 4 5]\n",
      "15\n",
      "18\n"
     ]
    }
   ],
   "source": [
    "x = np.array([1,2,3,4,5])  # (5,)\n",
    "w = np.array([1,1,1,1,1])  # (5,)\n",
    "b = 3\n",
    "print(x*w) # 아다마르의 곱\n",
    "print(np.sum(x*w))  # 내적\n",
    "c = np.sum(x*w) + b\n",
    "print(c)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 3  6  9 12]\n"
     ]
    }
   ],
   "source": [
    "x = np.array([1,2,3,4])\n",
    "err = 3\n",
    "w_grad = x*err  # [1,2,3,4]*[3,3,3,3]\n",
    "print(w_grad)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0 1 2 3 4 5 6 7 8 9]\n",
      "[5 5 5 5 5 5 6 7 8 9]\n",
      "[0 1 2 3 4 5 6 7 7 7]\n",
      "[3 3 3 3 4 5 6 7 7 7]\n"
     ]
    }
   ],
   "source": [
    "a = np.arange(10)\n",
    "print(a)\n",
    "z = np.clip(a, 5, None)   # NaN\n",
    "print(z)\n",
    "z = np.clip(a, None, 7)\n",
    "print(z)\n",
    "z = np.clip(a, 3, 7)\n",
    "print(z)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2.6881171418161356e+43\n"
     ]
    }
   ],
   "source": [
    "z = np.exp(100) # e^100\n",
    "print(z)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "class LogisticNeuron:\n",
    "    \n",
    "    def __init__(self):\n",
    "        self.w = None\n",
    "        self.b = None\n",
    "\n",
    "    def forpass(self, x):  # x.shape => (30,), self.w.shape => (30,)\n",
    "        z = np.sum(x * self.w) + self.b  # 직선 방정식을 계산합니다  x1*w1 + x2*w2  + ... + b\n",
    "        return z\n",
    "\n",
    "    def backprop(self, x, err):\n",
    "        w_grad = x * err    # 가중치에 대한 그래디언트를 계산합니다\n",
    "        b_grad = 1 * err    # 절편에 대한 그래디언트를 계산합니다\n",
    "        return w_grad, b_grad\n",
    "\n",
    "    def activation(self, z):\n",
    "        z = np.clip(z, -100, None) # 안전한 np.exp() 계산을 위해\n",
    "        a = 1 / (1 + np.exp(-z))  # 시그모이드 계산\n",
    "        return a\n",
    "        \n",
    "    def fit(self, x, y, epochs=100, lr=0.01): #(455,30)\n",
    "        self.w = np.ones(x.shape[1])      # 가중치를 초기화합니다.\n",
    "        self.b = 0                        # 절편을 초기화합니다.\n",
    "        for i in range(epochs):           # epochs만큼 반복합니다\n",
    "            for x_i, y_i in zip(x, y):    # x_i.shape=>(30,)\n",
    "                z = self.forpass(x_i)     # 정방향 계산   x1*w1+x2*w2+x3*w3+...+x30*w30+b\n",
    "                a = self.activation(z)    # 활성화 함수 적용\n",
    "                err = (a - y_i)           # 0.1 - 1 => -0.9\n",
    "                w_grad, b_grad = self.backprop(x_i, err) # 역방향 계산\n",
    "                self.w -= lr*w_grad          # 가중치 업데이트\n",
    "                self.b -= lr*b_grad          # 절편 업데이트\n",
    "    \n",
    "    def predict(self, x):\n",
    "        z = [self.forpass(x_i) for x_i in x]    # 정방향 계산\n",
    "        a = self.activation(np.array(z))        # 활성화 함수 적용\n",
    "        return a >= 0.5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "neuron = LogisticNeuron()\n",
    "neuron.fit(x_train, y_train,epochs=1000) # (455,30)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[False  True False False False False  True False False False  True False\n",
      "  True False False False False  True  True  True False False  True  True\n",
      " False False False  True  True False  True  True  True  True False  True\n",
      "  True  True False False  True  True  True False False False  True  True\n",
      "  True False  True False  True False  True  True  True False False  True\n",
      "  True  True  True False False False False  True False  True  True  True\n",
      "  True False False False  True  True  True  True False  True False  True\n",
      " False  True  True  True False False  True  True False False False  True\n",
      " False  True False False False False  True  True False False  True False\n",
      "  True False  True False False  True]\n",
      "[ True  True  True False  True False  True  True  True  True  True  True\n",
      "  True  True  True False False  True  True  True  True  True  True  True\n",
      " False False  True  True  True False  True  True  True  True  True False\n",
      "  True  True False  True  True  True  True  True  True False  True  True\n",
      "  True  True  True False  True  True  True  True  True  True  True  True\n",
      "  True  True  True False  True  True False  True False  True  True  True\n",
      "  True  True  True  True False  True  True  True False  True  True  True\n",
      "  True  True  True  True False False  True  True  True  True  True  True\n",
      "  True  True  True False  True  True False  True  True  True  True  True\n",
      "  True  True  True  True False  True]\n"
     ]
    }
   ],
   "source": [
    "print(neuron.predict(x_test))\n",
    "print(neuron.predict(x_test)== y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8157894736842105"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.mean(neuron.predict(x_test) == y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 단일층 신경망 구현"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1 5 3 2 6 7 4]\n",
      "7\n",
      "[0 1 2 3 4 5 6]\n",
      "[4 0 3 1 6 2 5]\n",
      "[6 1 2 5 4 3 7]\n"
     ]
    }
   ],
   "source": [
    "a = np.array([1,5,3,2,6,7,4])\n",
    "print(a)\n",
    "print(len(a))\n",
    "print(np.arange(len(a)))\n",
    "indexes = np.random.permutation(np.arange(len(a)))\n",
    "print(indexes)\n",
    "\n",
    "# print(a[np.arange(len(a))])\n",
    "print(a[indexes])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0 1 2 3 4 5 6 7 8]\n",
      "[8 3 4 6 2 0 5 7 1]\n",
      "[9 4 5 7 3 1 6 8 2]\n",
      "[90 40 50 70 30 10 60 80 20]\n"
     ]
    }
   ],
   "source": [
    "x = np.array([1,2,3,4,5,6,7,8,9])\n",
    "y = np.array([10,20,30,40,50,60,70,80,90])\n",
    "print(np.arange(len(x)))\n",
    "indexes = np.random.permutation(np.arange(len(x)))\n",
    "print(indexes)\n",
    "print(x[indexes])  # x[ [2 4 7 5 8 3 0 1 6] ]\n",
    "print(y[indexes])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8157894736842105\n",
      "0.8771929824561403\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.datasets import load_breast_cancer\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "cancer = load_breast_cancer()\n",
    "x = cancer.data\n",
    "y = cancer.target\n",
    "x_train, x_test, y_train, y_test = train_test_split(x, y, stratify=y, \n",
    "                                                    test_size=0.2, random_state=42)\n",
    "\n",
    "class SingleLayer:\n",
    "    \n",
    "    def __init__(self, rand_flag=0):\n",
    "        self.w = None\n",
    "        self.b = None\n",
    "        self.losses = []\n",
    "        self.rand_flag = rand_flag\n",
    "\n",
    "    def forpass(self, x):\n",
    "        z = np.sum(x * self.w) + self.b  # 직선 방정식을 계산합니다\n",
    "        return z\n",
    "\n",
    "    def backprop(self, x, err):\n",
    "        w_grad = x * err    # 가중치에 대한 그래디언트를 계산합니다\n",
    "        b_grad = 1 * err    # 절편에 대한 그래디언트를 계산합니다\n",
    "        return w_grad, b_grad\n",
    "\n",
    "    def activation(self, z):\n",
    "        z = np.clip(z, -100, None) # 안전한 np.exp() 계산을 위해\n",
    "        a = 1 / (1 + np.exp(-z))  # 시그모이드 계산\n",
    "        return a\n",
    "        \n",
    "    def fit(self, x, y, epochs=100):\n",
    "        self.w = np.ones(x.shape[1])               # 가중치를 초기화합니다.\n",
    "        self.b = 0                                 # 절편을 초기화합니다.\n",
    "        for i in range(epochs):                    # epochs만큼 반복합니다\n",
    "            loss = 0\n",
    "            # 인덱스를 섞습니다\n",
    "            \n",
    "            if( self.rand_flag == 1 ):\n",
    "                indexes = np.random.permutation(np.arange(len(x)))\n",
    "            else:\n",
    "                indexes = np.arange(len(x))\n",
    "                \n",
    "            for i in indexes:                      # 모든 샘플에 대해 반복합니다\n",
    "                z = self.forpass(x[i])             # 정방향 계산\n",
    "                a = self.activation(z)             # 활성화 함수 적용\n",
    "                err = a - y[i]                     # 오차 계산\n",
    "                w_grad, b_grad = self.backprop(x[i], err) # 역방향 계산\n",
    "                self.w -= w_grad                   # 가중치 업데이트\n",
    "                self.b -= b_grad                   # 절편 업데이트\n",
    "                # 안전한 로그 계산을 위해 클리핑한 후 손실을 누적합니다\n",
    "                a = np.clip(a, 1e-10, 1-1e-10)\n",
    "                loss += -(y[i]*np.log(a)+(1-y[i])*np.log(1-a))\n",
    "            # 에포크마다 평균 손실을 저장합니다\n",
    "            self.losses.append(loss/len(y))\n",
    "    \n",
    "    def predict(self, x):\n",
    "        z = [self.forpass(x_i) for x_i in x]     # 정방향 계산\n",
    "        return np.array(z) > 0                   # 스텝 함수 적용\n",
    "    \n",
    "    def score(self, x, y):\n",
    "        return np.mean(self.predict(x) == y)\n",
    "\n",
    "layer1 = SingleLayer(rand_flag=0)\n",
    "layer1.fit(x_train, y_train,500)\n",
    "print(layer1.score(x_test, y_test))\n",
    "\n",
    "layer2 = SingleLayer(rand_flag=1)\n",
    "layer2.fit(x_train, y_train,500)\n",
    "print(layer2.score(x_test, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(layer1.losses)\n",
    "plt.plot(layer2.losses)\n",
    "plt.xlabel('epoch')\n",
    "plt.ylabel('loss')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 스케일 테스트"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[   0]\n",
      " [1000]\n",
      " [2000]\n",
      " [3000]\n",
      " [4000]\n",
      " [5000]\n",
      " [6000]\n",
      " [7000]\n",
      " [8000]\n",
      " [9000]]\n",
      "[4500.]\n",
      "[2872.28132327]\n",
      "[[-1.5666989 ]\n",
      " [-1.21854359]\n",
      " [-0.87038828]\n",
      " [-0.52223297]\n",
      " [-0.17407766]\n",
      " [ 0.17407766]\n",
      " [ 0.52223297]\n",
      " [ 0.87038828]\n",
      " [ 1.21854359]\n",
      " [ 1.5666989 ]]\n",
      "[-4.4408921e-17]\n",
      "[1.]\n"
     ]
    }
   ],
   "source": [
    "x_train = np.arange(10).reshape(10,1)\n",
    "x_train = x_train*1000\n",
    "print(x_train)\n",
    "train_mean = np.mean(x_train, axis=0)\n",
    "print(train_mean)\n",
    "train_std = np.std(x_train, axis=0)\n",
    "print(train_std)\n",
    "x_train_scaled = (x_train - train_mean) / train_std\n",
    "print(x_train_scaled)\n",
    "print(np.mean(x_train_scaled, axis=0))\n",
    "print(np.std(x_train_scaled, axis=0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['mean perimeter' 'mean area']\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn.datasets import load_breast_cancer\n",
    "from sklearn.model_selection import train_test_split\n",
    "cancer = load_breast_cancer()\n",
    "\n",
    "x = cancer.data\n",
    "y = cancer.target\n",
    "x_train, x_test, y_train, y_test = train_test_split(x, y, stratify=y, \n",
    "                                                            test_size=0.2, random_state=42)\n",
    "print(cancer.feature_names[[2,3]])\n",
    "plt.boxplot(x_train[:, 2:4])\n",
    "plt.xlabel('feature')\n",
    "plt.ylabel('value')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "15\n"
     ]
    }
   ],
   "source": [
    "a = np.array([1,2,3,4,5])\n",
    "print( np.sum(a) )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[  0 100 200 300 400 500 600 700 800 900]\n",
      "287.22813232690146\n",
      "450.0\n",
      "[-1.5666989  -1.21854359 -0.87038828 -0.52223297 -0.17407766  0.17407766\n",
      "  0.52223297  0.87038828  1.21854359  1.5666989 ]\n",
      "0.9999999999999999\n",
      "-4.4408920985006264e-17\n"
     ]
    }
   ],
   "source": [
    "a = np.arange(10) # [0,1,2,3,4,5,6,7,8,9]\n",
    "a = a*100\n",
    "print(a)\n",
    "std = np.std(a)\n",
    "print(std)\n",
    "mean = np.mean(a)\n",
    "print(mean)\n",
    "z = (a - mean)/std\n",
    "print(z)\n",
    "print(np.std(z))\n",
    "print(np.mean(z)) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2, 3)\n",
      "21\n",
      "[5 7 9]\n",
      "[ 6 15]\n"
     ]
    }
   ],
   "source": [
    "a = np.array([[1,2,3],   # 6\n",
    "              [4,5,6]])  # 15\n",
    "print(a.shape)\n",
    "print(np.sum(a))\n",
    "print(np.sum(a, axis=0)) # 행끼리 더하라.  , 세로로 더하라\n",
    "print(np.sum(a, axis=1)) # 열끼리 더하라.  , 가로로 더하라"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3, 2)\n",
      "[[  1 400]\n",
      " [  2 300]\n",
      " [  3 500]]\n",
      "[  2. 400.]\n",
      "[ 0.81649658 81.64965809]\n",
      "[[  -1.    0.]\n",
      " [   0. -100.]\n",
      " [   1.  100.]]\n",
      "[[-1.22474487  0.        ]\n",
      " [ 0.         -1.22474487]\n",
      " [ 1.22474487  1.22474487]]\n"
     ]
    }
   ],
   "source": [
    "# x.shape => (455,30) \n",
    "\n",
    "a = np.array([[1,400],\n",
    "              [2,300],\n",
    "              [3,500]])\n",
    "print(a.shape)\n",
    "print(a)\n",
    "train_mean = np.mean(a, axis=0)\n",
    "print(train_mean)\n",
    "train_std = np.std(a, axis=0)\n",
    "print(train_std)\n",
    "m1 = (a - train_mean)\n",
    "print(m1)\n",
    "a_scaled = m1 / train_std\n",
    "print(a_scaled)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "364 91\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn.datasets import load_breast_cancer\n",
    "from sklearn.model_selection import train_test_split\n",
    "cancer = load_breast_cancer()\n",
    "x = cancer.data\n",
    "y = cancer.target\n",
    "x_train_all, x_test, y_train_all, y_test = train_test_split(x, y, stratify=y, \n",
    "                                                            test_size=0.2, random_state=42)\n",
    "\n",
    "x_train, x_val, y_train, y_val = train_test_split(x_train_all, y_train_all, stratify=y_train_all, \n",
    "                                                  test_size=0.2, random_state=42)\n",
    "print(len(x_train), len(x_val))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "class SingleLayer:\n",
    "    \n",
    "    def __init__(self, learning_rate=0.1, l1=0, l2=0):\n",
    "        self.w = None\n",
    "        self.b = None\n",
    "        self.losses = []\n",
    "        self.val_losses = []\n",
    "        self.w_history = []\n",
    "        self.lr = learning_rate\n",
    "        self.l1 = l1\n",
    "        self.l2 = l2\n",
    "\n",
    "    def forpass(self, x):\n",
    "        z = np.sum(x * self.w) + self.b    # 직선 방정식을 계산합니다\n",
    "        return z\n",
    "\n",
    "    def backprop(self, x, err):\n",
    "        w_grad = x * err          # 가중치에 대한 그래디언트를 계산합니다\n",
    "        b_grad = 1 * err    # 절편에 대한 그래디언트를 계산합니다\n",
    "        return w_grad, b_grad\n",
    "\n",
    "    def activation(self, z):\n",
    "        z = np.clip(z, -100, None) # 안전한 np.exp() 계산을 위해\n",
    "        a = 1 / (1 + np.exp(-z))  # 시그모이드 계산\n",
    "        return a\n",
    "        \n",
    "    def fit(self, x, y, epochs=100, x_val=None, y_val=None):\n",
    "        self.w = np.ones(x.shape[1])               # 가중치를 초기화합니다.\n",
    "        self.b = 0                                 # 절편을 초기화합니다.\n",
    "        self.w_history.append(self.w.copy())       # 가중치를 기록합니다.\n",
    "        np.random.seed(42)                         # 랜덤 시드를 지정합니다.\n",
    "        for i in range(epochs):                    # epochs만큼 반복합니다.\n",
    "            loss = 0\n",
    "            # 인덱스를 섞습니다\n",
    "            indexes = np.random.permutation(np.arange(len(x)))\n",
    "            for i in indexes:                      # 모든 샘플에 대해 반복합니다\n",
    "                z = self.forpass(x[i])             # 정방향 계산\n",
    "                a = self.activation(z)             # 활성화 함수 적용\n",
    "                err = (a - y[i])                  # 오차 계산\n",
    "                w_grad, b_grad = self.backprop(x[i], err) # 역방향 계산\n",
    "                # 그래디언트에서 페널티 항의 미분 값을 더합니다\n",
    "                w_grad += self.l1 * np.sign(self.w) + self.l2 * self.w\n",
    "                self.w -= self.lr * w_grad         # 가중치 업데이트\n",
    "                self.b -= b_grad                   # 절편 업데이트\n",
    "                # 가중치를 기록합니다.\n",
    "                self.w_history.append(self.w.copy())\n",
    "                # 안전한 로그 계산을 위해 클리핑한 후 손실을 누적합니다\n",
    "                a = np.clip(a, 1e-10, 1-1e-10)\n",
    "                loss += -(y[i]*np.log(a)+(1-y[i])*np.log(1-a))\n",
    "            # 에포크마다 평균 손실을 저장합니다\n",
    "            self.losses.append(loss/len(y) + self.reg_loss())\n",
    "            # 검증 세트에 대한 손실을 계산합니다\n",
    "            self.update_val_loss(x_val, y_val)\n",
    "    \n",
    "    def predict(self, x):\n",
    "        z = [self.forpass(x_i) for x_i in x]     # 정방향 계산\n",
    "        return np.array(z) >= 0                   # 스텝 함수 적용\n",
    "    \n",
    "    def score(self, x, y):\n",
    "        return np.mean(self.predict(x) == y)\n",
    "    \n",
    "    def reg_loss(self):\n",
    "        return self.l1 * np.sum(np.abs(self.w)) + self.l2 / 2 * np.sum(self.w**2)\n",
    "    \n",
    "    def update_val_loss(self, x_val, y_val):\n",
    "        if x_val is None:\n",
    "            return\n",
    "        val_loss = 0\n",
    "        for i in range(len(x_val)):\n",
    "            z = self.forpass(x_val[i])     # 정방향 계산\n",
    "            a = self.activation(z)         # 활성화 함수 적용\n",
    "            a = np.clip(a, 1e-10, 1-1e-10)\n",
    "            val_loss += -(y_val[i]*np.log(a)+(1-y_val[i])*np.log(1-a))\n",
    "        self.val_losses.append(val_loss/len(y_val) + self.reg_loss())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9120879120879121"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "layer1 = SingleLayer()\n",
    "layer1.fit(x_train, y_train)\n",
    "layer1.score(x_val, y_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "w2 = []\n",
    "w3 = []\n",
    "for w in layer1.w_history:\n",
    "    w2.append(w[2])\n",
    "    w3.append(w[3])\n",
    "plt.plot(w2, w3)\n",
    "plt.plot(w2[-1], w3[-1], 'ro')\n",
    "plt.xlabel('w[2]')\n",
    "plt.ylabel('w[3]')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_mean = np.mean(x_train, axis=0)\n",
    "train_std = np.std(x_train, axis=0)\n",
    "x_train_scaled = (x_train - train_mean) / train_std"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "layer2 = SingleLayer()\n",
    "layer2.fit(x_train_scaled, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "w2 = []\n",
    "w3 = []\n",
    "for w in layer2.w_history:\n",
    "    w2.append(w[2])\n",
    "    w3.append(w[3])\n",
    "plt.plot(w2, w3)\n",
    "plt.plot(w2[-1], w3[-1], 'ro')\n",
    "plt.xlabel('w[2]')\n",
    "plt.ylabel('w[3]')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.37362637362637363"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "layer2.score(x_val, y_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "val_mean = np.mean(x_val, axis=0)\n",
    "val_std = np.std(x_val, axis=0)\n",
    "x_val_scaled = (x_val - val_mean) / val_std"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.967032967032967"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "layer2.score(x_val_scaled, y_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_val_scaled = (x_val - train_mean) / train_std"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.967032967032967"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "layer2.score(x_val_scaled, y_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "layer3 = SingleLayer()\n",
    "layer3.fit(x_train_scaled, y_train, x_val=x_val_scaled, y_val=y_val)\n",
    "plt.ylim(0, 0.3)\n",
    "plt.plot(layer3.losses)\n",
    "plt.plot(layer3.val_losses)\n",
    "plt.ylabel('loss')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train_loss', 'val_loss'])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "l1_list = [0.0001, 0.001, 0.01]\n",
    "\n",
    "for l1 in l1_list:\n",
    "    lyr = SingleLayer(l1=l1)\n",
    "    lyr.fit(x_train_scaled, y_train, x_val=x_val_scaled, y_val=y_val)\n",
    "    \n",
    "    plt.plot(lyr.losses)\n",
    "    plt.plot(lyr.val_losses)\n",
    "    plt.title('Learning Curve (l1={})'.format(l1))\n",
    "    plt.ylabel('loss')\n",
    "    plt.xlabel('epoch')\n",
    "    plt.legend(['train_loss', 'val_loss'])\n",
    "    plt.ylim(0, 0.3)\n",
    "    plt.show()\n",
    "    \n",
    "    plt.plot(lyr.w, 'bo')\n",
    "    plt.title('Weight (l1={})'.format(l1))\n",
    "    plt.ylabel('value')\n",
    "    plt.xlabel('weight')\n",
    "    plt.ylim(-4, 4)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEWCAYAAAB8LwAVAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAABOlklEQVR4nO3dd3gVVfrA8e970wNJCKEnQAKhdwi9qSgCKqggYkOx97KrK/52rau7uuvuqrsK9gYqiqKgCKKC9N47AUIILSQkARLSz++PM0lukhtIgAsI7+d57nNzZ87MnElu5p1TR4wxKKWUUmW5znYGlFJKnZs0QCillPJIA4RSSimPNEAopZTySAOEUkopjzRAKKWU8kgDhPpdE5G+IrLlbOfjbBCR2iKyRUQCnc9zROTOs52vkyEidUVkk4gEnO28qBIaINRJE5EEEbn0bObBGDPPGNPCW/sXkctFZK6IHBGRgyLym4gM9dbxqmgs8KExJrvsChG5QkTmi0i6iOwXkXdFJKSqBxCRABH5QEQOO/v5wwnS3ygiu0QkU0S+FZGabutGishCEckSkTnu2xljDgCzgburmkflPRog1DlNRHzO4rFHAF8BnwBRQF3gGeCqk9iXiMhp+39z7rRvBSZUkCQMeBFoALTC5v+fJ3Go54BmQGPgYuBPIjKogjy1Ad4GbsH+rrKAt9ySHAJeA16u4FgTgXtOIo/KW4wx+tLXSb2ABOBSD8td2Lvb7UAq8CVQ0239V8B+IAOYC7RxW/cRMA6YDmQClzrHeRxY62wzCQh00l8EJJXJk8e0zvo/AfuAvcCdgAFiPZyDAInAE8c5/+eACW6fo539+Tqf5wAvAQuAY8BfgOVl9vEYMNX5OQB41TnuAWA8EFTBsfsB8WWWzQHurCD9tcC6k/gb7wEGun3+K/BFBWn/Bnzm9rkpkAuElEl3JzDHw/a+2KDS+Gx/t/VlX1qCUN7wMHA10B97B5sGvOm2/kfsXWkdYCX2ztHdjdgLawgw31k2EhgExADtgduOc3yPaZ073z9gg06sk7+KtAAaApOPk6YybsFWm4QA/wVaiEgzt/U3Ap85P78CNAc6OvmLxJZYPGkHVKXtpR+woeiDiLzlVD95eq110oRj/35r3PazBmhTwTHauKc1xmzHBojmlcmgMSYfiAc6VOG8lBdpgFDecA/wZ2NMkjEmB3unPUJEfAGMMR8YY464resgImFu239njFlgjCk0JfXrbxhj9hpjDgHTsBfRilSUdiS2zn6DMSYLeP44+4hw3vdV8pwr8pFzvHxjTAbwHXADgBMoWgJTRUSAu4DHjDGHjDFHsHfkoyrYbw3gSGUyICKXYaujioONMeZ+Y0yNCl7tnWTVnfcMt91lYIOdJ9XLpD1Rek+OYM9NnQM0QChvaAxMKbojBTYBBUBdEfERkZdFZLuIHMZWCQHUctt+t4d97nf7OYuSi5cnFaVtUGbfno5TJNV5r3+cNJVR9hif4QQIbOnhWydY1QaCgRVuv7cZznJP0qjEhVdEejjHHGGM2VrFvB913kPdloVScWA6WibtidJ7EgKkVyG98iINEMobdgODy9yVBhpj9mAvisOw1Txh2Hp7sHX+Rbw1xfA+bGNtkYbHSbsFex7Dj5MmE3tRL1LPQ5qy5/ITUEtEOmIDRVH1Ugq2naKN2+8szBhTUSBcywmqbkSkEzAVuN0Y80uZdeNF5GgFrw0Axpg07O/MvcqnA25VVWVscE8rIk2w7SqVCkxOCTOW0lVa6izSAKFOlZ+IBLq9fLGNqy+JSGMo7q8/zEkfAuRg79CDsdUoZ8qXwBgRaSUiwVRcv48xxmDbK54WkTEiEioiLhHpIyLvOMlWA/1EpJFTRfbUiTLg1LNPxvYoqgnMcpYXAu8C/xGROgAiEikil1ewq6VADRGJ9LRSRNpiSyAPGWOmecjHvcaY6hW83NsYPgH+IiLhItISWw32UQV5mghc5YxNqQa8AHzjVJfhlB4DsY3RLuf74ue2fTcgwRizq4L9qzNMA4Q6VdOxd75Fr+eA17F3rj+JyBFgMdDdSf8JsAvbO2ajs+6MMMb8CLyB7W8fDyxyVuVUkH4ycD1wO7bX0wFs19HvnPWzsL2k1gIrgO8rmZXPsCWor5yAUeRJJ1+Lneq3n7GN5Z7ylou9UN9cwTH+iK2eer9syaCKnsX2RtsF/Ab80xgzo2ils9++Tp42APdiA0Uy9mbgfrd93YL9jowD+jo/v+u2/ibszYU6R4i9UVLqwiMirYD1QECZC/XvgojUBuYBnYwxx852fk6FU2r6DXsu5Qb+qbNDA4S6oIjINcAPQDXgY6DQGHP1Wc2UUucor1YxicggsXPFxIvIWA/rh4nIWhFZLSLLRaRPZbdV6iTdAxzEVpsUAPed3ewode7yWgnCmSJhK3AZkAQsA24wxmx0S1MdyDTGGBFpD3xpjGlZmW2VUkp5lzdLEN2wUwHscBrUvsB2byxmjDlqSiJUNUq6BJ5wW6WUUt7l68V9R1J6kFASJT1Zijl1wn/HTrtwRVW2dba/G2cGyGrVqnVp2bLlSWU2v9Cwad9hGtQIIqKa/0ntQymlfm9WrFiRYozxOCDTmwFCPCwrV59ljJmCHXXbDzsR2KWV3dbZ/h3gHYC4uDizfPnyk8psWmYunf46i2evas1tvWNOah9KKfV7IyIVjjvxZhVTEqVHqkZh+5J7ZIyZCzQVkVpV3fZ0cImNSQXaqUsppQDvBohlQDMRiRERf+ykY1PdE4hIrDNJGSLSGfDHjrA94banm8v5TWi3X6WUsrxWxWSMyReRB4GZgA/wgTFmg4jc66wfj53nZrSI5GFHVV7vNFp73NZbeQW3EkShBgillALvtkFgjJmOnYrBfdl4t59fwc6BX6ltvcnHZQOExgelzi15eXkkJSWRna0DrE9FYGAgUVFR+Pn5nTixw6sB4vfEKUBQqFVMSp1TkpKSCAkJITo6GhFP/VfUiRhjSE1NJSkpiZiYynfC0cn6HD7OF69QixBKnVOys7OJiIjQ4HAKRISIiIgql8I0QDhKejFpgFDqXKPB4dSdzO9QA4TDpW0QSilVigYINy7Rbq5KKVVEA4Qbl4h2c1VKlZKens5bb71V5e2GDBlCenp6lbe77bbbmDx5cpW38wYNEG5cLtEqJqVUKRUFiIKCguNuN336dGrUqOGlXJ0Z2s3VjUu0m6tS57Lnp21g497Dp3WfrRuE8uxVbSpcP3bsWLZv307Hjh3x8/OjevXq1K9fn9WrV7Nx40auvvpqdu/eTXZ2No888gh33303ANHR0SxfvpyjR48yePBg+vTpw8KFC4mMjOS7774jKCjohHn75ZdfePzxx8nPz6dr166MGzeOgIAAxo4dy9SpU/H19WXgwIG8+uqrfPXVVzz//PP4+PgQFhbG3LlzT/l3owHCjY+IdnNVSpXy8ssvs379elavXs2cOXO44oorWL9+ffF4gg8++ICaNWty7NgxunbtyvDhw4mIiCi1j23btvH555/z7rvvMnLkSL7++mtuvrmix4lb2dnZ3Hbbbfzyyy80b96c0aNHM27cOEaPHs2UKVPYvHkzIlJcjfXCCy8wc+ZMIiMjT6pqyxMNEG5cItrNValz2PHu9M+Ubt26lRps9sYbbzBlyhQAdu/ezbZt28oFiJiYGDp27AhAly5dSEhIOOFxtmzZQkxMDM2bNwfg1ltv5c033+TBBx8kMDCQO++8kyuuuIIrr7wSgN69e3PbbbcxcuRIrr322tNwptoGUYrLJWh8UEodT7Vq1Yp/njNnDj///DOLFi1izZo1dOrUyeNgtICAgOKffXx8yM/PP+FxKupR6evry9KlSxk+fDjffvstgwYNAmD8+PG8+OKL7N69m44dO5KamlrVUyt/rFPew3nEJTpZn1KqtJCQEI4cOeJxXUZGBuHh4QQHB7N582YWL1582o7bsmVLEhISiI+PJzY2lk8//ZT+/ftz9OhRsrKyGDJkCD169CA2NhaA7du30717d7p37860adPYvXt3uZJMVWmAcOPjEm2kVkqVEhERQe/evWnbti1BQUHUrVu3eN2gQYMYP3487du3p0WLFvTo0eO0HTcwMJAPP/yQ6667rriR+t577+XQoUMMGzaM7OxsjDH85z//AeCJJ55g27ZtGGMYMGAAHTp0OOU8yPk0MOxUnigH0PWln7m0VV3+fm2705grpdSp2LRpE61atTrb2TgvePpdisgKY0ycp/TaBuHGJTpZn1JKFdEqJjc+olVMSqkz44EHHmDBggWllj3yyCOMGTPmLOWoPA0QbkS7uSqlzpA333zzbGfhhLSKyY2PdnNVSqliGiDcaDdXpZQqoQHCjUu7uSqlVDENEG5c2kitlFLFNEC4sZP1ne1cKKV+z6pXr17huoSEBNq2bXsGc3NqNEC4EZ3uWymlimk3Vzc61YZS57gfx8L+dad3n/XaweCXK1z95JNP0rhxY+6//34AnnvuOUSEuXPnkpaWRl5eHi+++CLDhg2r0mGzs7O57777WL58Ob6+vvz73//m4osvZsOGDYwZM4bc3FwKCwv5+uuvadCgASNHjiQpKYmCggKefvpprr/++lM67crQAOHGtkGc7Vwopc4lo0aN4tFHHy0OEF9++SUzZszgscceIzQ0lJSUFHr06MHQoUMRkUrvt2gcxLp169i8eTMDBw5k69atjB8/nkceeYSbbrqJ3NxcCgoKmD59Og0aNOCHH34A7CSBZ4IGCDfazVWpc9xx7vS9pVOnTiQnJ7N3714OHjxIeHg49evX57HHHmPu3Lm4XC727NnDgQMHqFevXqX3O3/+fB566CHAztzauHFjtm7dSs+ePXnppZdISkri2muvpVmzZrRr147HH3+cJ598kiuvvJK+fft663RL0TYIN9rNVSnlyYgRI5g8eTKTJk1i1KhRTJw4kYMHD7JixQpWr15N3bp1PT4H4ngqmij1xhtvZOrUqQQFBXH55Zfz66+/0rx5c1asWEG7du146qmneOGFF07HaZ2QVwOEiAwSkS0iEi8iYz2sv0lE1jqvhSLSwW1dgoisE5HVInLyU7RWgXZzVUp5MmrUKL744gsmT57MiBEjyMjIoE6dOvj5+TF79mx27dpV5X3269ePiRMnArB161YSExNp0aIFO3bsoEmTJjz88MMMHTqUtWvXsnfvXoKDg7n55pt5/PHHWbly5ek+RY+8VsUkIj7Am8BlQBKwTESmGmM2uiXbCfQ3xqSJyGDgHaC72/qLjTEp3spjWdrNVSnlSZs2bThy5AiRkZHUr1+fm266iauuuoq4uDg6duxIy5Ytq7zP+++/n3vvvZd27drh6+vLRx99REBAAJMmTWLChAn4+flRr149nnnmGZYtW8YTTzyBy+XCz8+PcePGeeEsy/Pa8yBEpCfwnDHmcufzUwDGmL9XkD4cWG+MiXQ+JwBxVQkQp/o8iOvfXoQBvryn50nvQyl1eunzIE6fc+l5EJHAbrfPSc6yitwB/Oj22QA/icgKEbnbC/krx07Wp1VMSikF3u3F5Km/l8err4hcjA0QfdwW9zbG7BWROsAsEdlsjJnrYdu7gbsBGjVqdEoZ1m6uSqnTYd26ddxyyy2llgUEBLBkyZKzlKOT480AkQQ0dPscBewtm0hE2gPvAYONMalFy40xe533ZBGZAnQDygUIY8w72LYL4uLiTuny7nKJdnNV6hxkjKnSGIOzrV27dqxevfpsZ6OUk6kd8WYV0zKgmYjEiIg/MAqY6p5ARBoB3wC3GGO2ui2vJiIhRT8DA4H1XswrYMdBaBWTUueWwMBAUlNT9X/zFBhjSE1NJTAwsErbea0EYYzJF5EHgZmAD/CBMWaDiNzrrB8PPANEAG85dwf5TmNJXWCKs8wX+MwYM8NbeS3i0ifKKXXOiYqKIikpiYMHD57trPyuBQYGEhUVVaVtvDqS2hgzHZheZtl4t5/vBO70sN0OoEPZ5d7m0m6uSp1z/Pz8iImJOdvZuCDpSGo3Lp3NVSmlimmAcKOzuSqlVAkNEG5cor2YlFKqiAYINy6XoAUIpZSyNEC40TYIpZQqoQHCjY92c1VKqWIaINyIdnNVSqliGiDcaBWTUkqV0ADhRru5KqVUCQ0QbkSEAq1iUkopQANEKT4unaxPKaWKaIBwo5P1KaVUCQ0QbuxkfRoglFIKNECU4hIdSa2UUkU0QLjxcaFVTEop5dAA4cY+k1oDhFJKgQaIUlwuHUmtlFJFNEC40ZHUSilVQgOEG+3mqpRSJTRAuCnqxaSD5ZRSSgNEKS4RAHQohFJKaYAoxcf5bWg7hFJKaYAoRYpLEBoglFJKA4QbH5cTILSrq1JKaYBw58QHLUEopRQaIEopaqTWrq5KKaUBopSiAGG0ikkppTRAuCuqYtIShFJKeTlAiMggEdkiIvEiMtbD+ptEZK3zWigiHSq7rTcUN1JrgFBKKe8FCBHxAd4EBgOtgRtEpHWZZDuB/saY9sBfgXeqsK038gygDw1SSim8W4LoBsQbY3YYY3KBL4Bh7gmMMQuNMWnOx8VAVGW39YaSEoS3j6SUUuc+bwaISGC32+ckZ1lF7gB+rOq2InK3iCwXkeUHDx48hexqG4RSSrnzZoAQD8s8XnlF5GJsgHiyqtsaY94xxsQZY+Jq1659Uhkt4tIqJqWUKubrxX0nAQ3dPkcBe8smEpH2wHvAYGNMalW2Pd2Ku7lqfFBKKa+WIJYBzUQkRkT8gVHAVPcEItII+Aa4xRiztSrbekNRG4RWMSmllBdLEMaYfBF5EJgJ+AAfGGM2iMi9zvrxwDNABPCW04Mo36ku8ritt/JaRHSqDaWUKubNKiaMMdOB6WWWjXf7+U7gzspu623aBqGUUiV0JLUb7eaqlFIlNEC4Ke7mqhFCKaU0QLhz6QODlFKqmAYINxoglFKqhAYIN9oGoZRSJTRAuNFurkopVUIDhJuSZ1JrgFBKKQ0QbkraIM5yRpRS6hygAcJN8TOpNUIopZQGCHdF4yCMtkEopZQGCHcunaxPKaWKaYBwo20QSilVQgOEm6IqJu3FpJRSGiBKKRkopwFCKaU0QLjRKiallCqhAcKNdnNVSqkSGiDcuJzfhnZzVUopDRCl+Ih2c1VKqSIaINyItkEopVQxDRButJurUkqV0ADhRru5KqVUCQ0QbrQXk1JKldAA4aZoLiYtQCillAaIUlz6RDmllCpWqQAhIo+ISKhY74vIShEZ6O3MnWnazVUppUpUtgRxuzHmMDAQqA2MAV72Wq7OEu3mqpSqtPxcOHLgbOfCq3wrmc6pfGEI8KExZo0UXU3PI/pMaqXUceVmwrqvYNss2DEHCnLh3gVQu/nZzplXVLYEsUJEfsIGiJkiEgIUnmgjERkkIltEJF5ExnpY31JEFolIjog8XmZdgoisE5HVIrK8kvk8JdoGoZQ6riXjYdojsHc1tB0OLl+Y+8+znSuvqWwJ4g6gI7DDGJMlIjWx1UwVEhEf4E3gMiAJWCYiU40xG92SHQIeBq6uYDcXG2NSKpnHU1b8RDktQSilPDm8DwLD4LH1IAKBobDoTej/J6jV7Gzn7rSrbAmiJ7DFGJMuIjcDfwEyTrBNNyDeGLPDGJMLfAEMc09gjEk2xiwD8qqYb68oGgehBQillEdZKVCttg0OAL0eAZ+A87YUUdkAMQ7IEpEOwJ+AXcAnJ9gmEtjt9jnJWVZZBvhJRFaIyN0VJRKRu0VkuYgsP3jwYBV2X15RFZP2YlJKeZSZAsG1Sj5Xrw3d7rTtEinxZy9fXlLZAJFv7BzYw4DXjTGvAyEn2MZTI3ZVrry9jTGdgcHAAyLSz1MiY8w7xpg4Y0xc7dq1q7D78koeGKQBQinlQVYqVKtVelmvh8/bUkRlA8QREXkKuAX4wWlf8DvBNklAQ7fPUcDeymbMGLPXeU8GpmCrrLxKq5iUUseVmQLBEaWXVa8DXe+AdV/aNorzSGUDxPVADnY8xH5sVdGJwuUyoJmIxIiIPzAKmFqZg4lINaenFCJSDTv+Yn0l83rSfLSRWilVkcJCzyUIgGYDwRRC6rYzny8vqlQvJmPMfhGZCHQVkSuBpcaY47ZBGGPyReRBYCbgA3xgjNkgIvc668eLSD1gORAKFIrIo0BroBYwxRlq4Qt8ZoyZcVJnWAXazVUpVaHsdDAFpdsgioQ6zauHK11J8rtQqQAhIiOxJYY52LaF/4rIE8aYycfbzhgzHZheZtl4t5/3Y6ueyjoMdKhM3k4nEUFEB8qp35FtsyDvGLQeerZzcv7LSrXvnkoQoQ3se0bSmcvPGVDZcRB/Bro67QGISG3gZ+C4AeL3yCWiU22oqtk0DX59Ee7+DfwCz+yxf/sHZGdogDgTMp1ekmXbIAD8gyEo/LwrQVS2DcJVFBwcqVXY9nfFR0S7uaqqSZgPBzfDAa83k5WXsdu+9DvrfZnOmF1PJQiA0Cg4vOfM5ecMqOxFfoaIzBSR20TkNuAHylQdnS9EtA1CVVF6on3fs8J7x1j9OSQuKb0sPweO7IO8LMg65L1jVyR5M2QfPvPHPVuynADhqQ0CICwSMi7AAGGMeQJ4B2iPbRt4xxjzpDczdra4RLQNogryCgoZ/Po8vl97fhWtqyRtl31P8tKUYRlJ8N0DsOC18suLpO/yzrErUpAP7w2AhW+c2eOeTZnHaYMA2w7hzRJEdgYcS/fe/j2odDWRMeZrY8wfjDGPGWOmeDNTZ5OPS9sgqiI++Sib9h1m+rrfX//vA4ezyczJP7WdGFNycfZWCWLZe7b3zKGdpZcXlVzAVjOdSYf3QO5RSN1eft2it2D9N2c2P2dCVgr4h4BvgOf1oZFw7BDkZlVufxu+haOVnP3h6EEY3xcmH3cKvNPuuAFCRI6IyGEPryMicl6WLbWKqWo27LVfg2UJaZjf0e+tsNAw9H/z+efMLae2o6xD9kJZvS4c2n76q3pys2DFR/bntJ22L34R96CQfoYDRJoTrDz12ln4Biz/4Mzm50zITIFqHhqoixR1dT1SiZulzFT46lZY9L8Tp807Bp+Psjci+9ZWLq+nyXEDhDEmxBgT6uEVYowJPVOZPJN8XFrFVBUb9to5Gw8eySHxUCXvnM4BO1IyOXA4h7VJ6ae2o6LSQ5tr7PveVae2v7LWToJjadDuOsjPLn3xSU8EcYFftTNfgkhLsO9lA0RRu8iZzs+ZkJVScfsD2DYIqFxX16IAe6JqycJC+OZuWzptcpHNw7G0SmX3dDgveyKdCu3mWjUb9hwmopo/YEsRvxcrd9m8bj1w9NRKPkUBovUwQE5vNZMxsORtqNcOOt5ol6W5VTOl77Z3reGNz0IJIsG+H9kHBW6TMRddHDOSoLDgzObJ2zIrGEVdpCqD5Yp+f3tXHf/3NOdvsGkqXP436HaPXeapWs9LNECU4dJurpVWWGjYuO8wg9vVIyzIj+UJZ6EnzUla4QSIozn57Ek/VmG6/IJC8gqO82ysogbqum2hVvPTGyB2zIGDm6D7fVCzqV12aEfJ+vRECGtoX+7tEZVUWGjYvP8ka4qL20NM6QtiUcAszK9cVcu5qrCwdHUenLgEUTRY7nAVShB5mZC8qeJ0Kz6GlldCj/tKnjeRcuam89AAUYZL+F3VpZ9NiYeyOJqTT9sGYcQ1DmfZ7yhArExMo6ZT8tl64EiF6Z76Zh2j319a8Y7Sd9kBUoGhENnFBojT9f1ZMt5ekNoOh7AocPmVDhAZu6FGQ6jRCDKqHiCmrtnLoNfmsWnfSQSJtATwdQYFuvfccS/JnETQOicYA9/cBR8OKr3sRG0QfkF2EF1lurqmJdi/J0DSMs9pjiZDZjI07m0bR8Oj7RPszuB8TxogyvDzcbEvI1uDRCUUNVC3aRBGXHRNth/MJPVozlnO1YllZOWxLfkowzvbKoEt+496TFdYaPhp4wGWJhziWG4F1QDpiVCjsf05srMdbXs66t/zsiH+Z+h4gx2d7fKxVUlFAaIgz16YazSyQSI7o8pjEhbE2379c7acxHNU0hKgoTPBcqnutomef/49WTcZ1k+27QP5uXZZzmEozDt+CQKcrq6VqWLaZb8vQTVhTwXtEPvX2fd6be27j58NElqCOHuGd4lizpaDTFhyZr/cR7LzuOm9xazfc6IH9Z07NuzNwNclNK9Xna7R4UBJ1U1ZaZm5HM4+Jx4cyMrdNo8Xt6hDg7DACksQG/cdJuNYHgWFprgxvpy0XfbCDRAVZ989NTxmpsAXN8H6rytXwkjeYKtpotxmua/ZpCRAHN5rZw8tqmKCKgempU6Jb358FQPEsTQ7cV103/LHTU+EanVKfv69ObwPpj8OfsFO12Knvr94FPUJnjlT2dHUh3ZCeIwtdSZVUC1ZNDK/btuSZRHNIPXMPZhIA0QZjwxoxiUt6/D81A0s3Vm1KpNZGw/wwMSVJ1X6mLs1hQXxqbwzd8eJE58j1u89TGyd6gT4+tAuKgx/XxfLKwgQt3ywhCe+WnOGc+jZql1puAQ6NKxB83ohbNnvOUAs2p5a/PPq3emwaiLMeKokQWGhU4JoZD/XaWMfHOOpHWLzD7D5e5h8O3xw+YnbKvautu/13easrNkEDiXYAFN0Ua7RqOT47tU78b/AL3+tcPcHDmezKzWL0EBfliWkkbN/M/yva3GbSsax4wTzogbWOq1t9Zp7CSJjN8fCmlAQXOfMD947VcbAtEdsb7Gh/7XLDjrdoI83UZ+7ygyWy8+xacKjIaqrnabFU+lv/3rb8B1cs2RZrVjbSH2GOgBogCjDxyX85/qONKoZzP0TV7Avo+IGTHfGGP45czM/rNvH1gOeqyyOp+gubsaG/aRn5VZ5+1K2zoQ1k05tHydgjGHj3gzaNAgDIMDXhw5RYR7bIZLSsli/5zCLdxw6J6ruViSm0bJeKNUCfGlRN4T4g0fJ99AQvXB7Ck1rVyOyRhAJO7baO8vF40rGOmQmQ0FOSRWTrz/Ubw97VpY/6K6Ftnriqjfs3eO7l8DOeRVnct8aCKxRcvEHGyByj9i72aK78xqNSkoQ7nfsi96Eea9W2CWy6Obnrr5NyM0vZM/iyZCyFbb9xIpdaXT56yw+XVzBBb4oQIRH27YRtwBh0nbx875ANhyrQU7KqQeIM/p9Wf0ZbJsJlz4HLQbbZSlb7XtRCcLTRH3uwiLt7/x4g+XSdwPGCRBd7M97PXxnDqwvXXoAiIi137kzNGusBgiwd4J5JYEgLMiPd0Z3ITuvkD9NrtzAlCU7DxUHhvlO3W5lGWOYt81ejHLzC/l2VSWH62+fDQc2ll8++yWYMbZ8Lww3m/Yd5m/TN5Gbf5weOu4KC2Hz9OIvfvKRHFKO5tI2smQ4TFx0TdbvyShXXz/bqePOOJbHzpTMyh3PSwoKDasT0+nS2FaJNa8bQm5+IbvKjOHIKyhk6c5D9Gpaiw4Nw7g48Q075xEGdi2wiYp6MIVHl2wY2QX2rbZTUbhLXAiNe0KXW+GhFXZE7rqvMMYwa+MBMrLK3LHvWwMNOtrGySLhMfb90I6S0kJYlH2imW9gSUN1QR4kLrY/V1B9sSzhEMH+PtzWOxp/HxeFO+fbFbsW8MGCneQXGv72wyYSPP29inowhTe2wamoUTY/F47sY3teTXblR3BoTzzZeZW40zXGY7Xbj+v20f65n9h9psbXLHsP6rWHbvdw1ASQFxJFYXEJ4gQT9RVxuroeS3X+FrlZ8NPTxdN0xCcfIbeom2pNp4oJyldL5mXb4FSvbIBwejKdoYZqDRB5x+DfLWFB6TllYuuE8NAlsczblsKqxBP37/9kUQI1gv2ICg9iYRUDxK7ULJLSjnFrr2jaR4XxxbLdJ75z2jQNJlwLM58qvTzvGObABjvkv+jux4O/Td/EO3N38MYvlfyirfkcvrgBZj0NlAyQKypBAMQ1DievwLCmzOCzOZuTCfb3AZyqGneZKaV75njZlv1HyMwtKA4QLerZR6tvLVPNtDYpg8zcAno2jWBItS0MKFhAVo8/2LrpHb/ZREVVKO53+Q2720CStAxjDAeP5NgLaHoiNOpl0wSGQuwlsHUmszbs565PlvOHL1eX/M3zcyF5Y+nqJbAlCHACRCJUr2enfRCxgaIoaOxdbbtPAiR57oG1dOchujQOJyTQj7hGoTTIWA1AQcJCZq7fx9AODfD1ER7/ak35JyymJdi6+ICQ0iWIw3sQDEcC6tO6dVsiCpL589erT/xdXvQ/eK19qfEUxhi++mUR9xd8ykdzKzfa3RjDt6v2cOBwdqnlhYWGDxfsZE3Z7567vGzbKNz0EnC5ePSLVSxIj2DLuuWMHL+IX1faG7EUE3L8TDgB4u7/TbXH2zTVjixfP5ltB45w2X/m8t53v9q04dG2ii6iWfkqx4ObbRtU2RJEcVfXM9MOoQHCL8j+Ubf/Um7VTT0aUyPYjzdnH/+PsS/jGDM3HOD6uIZc1KI2i3ekHr/vfBnznIDSJ7YW13dtyOb9R1ibZC/AhYWGb1YmsSvV7U5u1yKYfIf9OWl56bvVfWuRQvs5e7vnKoxdqZnM25ZCRDV/3poTz4pdJdVC8clHePnHzczcsL/47i/pYBoZP75APi7Msvdhzwo27LF1pq3ql/zDFF10F7rV3WfnFbBgewrDO0dRzd+HVYnpJRk5mgzvXATvD4T8XIwxHi8mhzJzmb05uXJ3oyewwgn2RXmNrVMdEUhMSoSckqrBRdvt36RH41AG7PwXuwrrsCRqDDTqCTvn2kRpHgJE7KXg44/ZNI3np22k60s/M+6TT+26xr1K0jUfBEf388306QT4uvhlczLfrnbuxA9ugoLc8gGiRiM7cvrQDltaqNGIrQeOsO3AEedO3gkQCc7fPawh7C4fINKzctm8/wjdom3d9rD6hwjmGLlRvfDJPECU2c8fBzbn+aFtWL4rjffnlwngaQklpaawKMjJgOwMDu21/yfNmremabPW+EsBC1ZvYPxvJ7gBWDPJnk/iouJFKxPT6Z4ymft8p+G/8j3SMk9c7fr50t08Omk1I8YvJCnNljqMMTw7dQPPT9vIHR8vr7iX3f51tpdSVBwb9x7m503JmFrNiXXto6Agn4TERLJMAHGvLGTk+EXk5Hv+LhaE2ABRx6QyZdUe2OBMW5e4mB+c+cqqZyWRbfyYvtO5RkTF2a6u7t/9ogbqeu1KH6BabQgI0xLEGRU7wF5oy8yUWD3Al9t7x/DzpuSKe7EAny1JpNAYbu7RmD6xtcjMLTj+3UoZ87cdJLJGEDG1qjG0QwOC/HyYtHw3mTn53DdxBX/4cg0j315ki9rJm+Dz6+3FYtDLdh4gt+cQpMfbf7LDJoj45T9VmF8flzDpnp40qBHEY5PWkJmTz5RVSVz13wWM/20793y6gi5/ncXI8Yt477VnCcvdz/P+j3PQhHHoywfZuCeN6IhgQgL9ivdbI9ifPrG1+GzJruJqpsU7UsnOK2RAqzp0aFijpASRl2179WQkQeZBZk/7lItfnUPHF2YxYfGu4ulOftt6kIH/mcuYj5bR8++/8I8Zm9l7nIFtYC8Kc7Ykl7uTBDuCunZIAFHhQQAEZh/k39UncMeSIfD1ncXpFm5PpVX9UGpu+IigjHheLBjNqr3HoEl/SNnCt3NXMHvJckz1uvYmo0hgKIVNLiJtxdd8tHAnl7aqS0TqCo6YIO6ZlWNLFACxl2EQmmcs5PVRHencqAbPTd1I8pFsW70EUL9j6cz7+tuLftpOCtMSWXc0lEGvzWX4uIUcDapfUoJImA+1W0Kzy+ydaZmqxuXOiPeuMTZA9PWzd+iLIm8DYHSDPTSOqMY1nSIZ2Lour/60lfhkt3a1tJ0lAaJo9HDGHtass9Wxvbt2KQ6aI5sZXpmxmb9N3+R5Cpv0RDjgdOfcUvJU4Y8W7GSwr612uVe+4at5x6/qTUrL4qUfNtI+KoyMrDyuf3sxuw9l8cL3G/l08S6u7RTJ4ew8nvx6necSTVFX08g4xv22nWr+PnTv1hM/k8PXNzbk1g7V8Qmpxf0XNWVpwqEKq4G/22H33Swwg/nrtmPinRvPxEXMWLePro1rcl3TfJJ963P/Z6t4YdpGCho43aPd25D2rwffoJJSYxER21B9hrq6aoAAW6w0BSV3hm5u7RVNSIBvcSkiO6+A//6yjbFfr2XxjlRy8gv4fGkiA1rWoWHNYHo2qYVI5dsh8gsKWRifSp/YWogIIYF+DGlXn6mr9zJi/CJmbTzAfRc1JSe/kJvfXUT+ZzfYL87NX0OLIXYnu0ueE5C8aSF7TU02VetGRMpytpUZKZuTX8CXy3dzWau6xNapzn+u78jutCyueGMej01aQ7uoMBaMvYRP7+jGVR0akJ11mCeCp5IT1Ys/P/4kX9W6n5oZG6kf/1mp6qUiDw9oRsrRXCYusXfXc7YcJNDPRY8mEXRsWINN+w6TnZsPUx+CpKWs6PoqyYSTv3ICNYL9aVkvhL98u55rxi3kz1PWcesHSwkP9uP1UR3pFlOT8b9tp9fLv3LZv3/jqW/WMW3N3nIXnu9W7+W2D5fR6+VfuW/CCuZuPciC+BQ+XpjAvG0pdG5UAxGBBa/D6x0ZmjeTBFdD2DoDDu0gO6+A5bvS6BtTHea/Bk0uIql2f1btToeYfgAsm/0t/kd2cySwQaljZ+cV8PGhdtTM288rvQzvju7C8FqJHKrZkbnxaQwft5CElEyy/MNZJ824Kmgtl7epxz9GdOBYXgHPfLsBs3c1BISWtDm4yQ1rTOquDRSkJzE/JZiRcQ0pNDA1wdc2mmcftu0P0X1sF9mcw8xeMJfbPlzKIecufGnCIfx9XHRsWAOA+ukr2U1dnlhRk1QTwuBQ28YgIrx0TTsCfFy8WjSpYUGeDepFeXMayE3GbhJ3bKEQFw0bx0KYDRCPdglgdM/GvDN3Bw98trJ8KbAoKNRqDlumgzHsz8gmfsMyGnEAut5FiGQTvOS1CkuQxhjGfr0OA7x5Y2cm3tmDozn5DHptLh8uSGBM72j+NbIDTw5qyc+bDjB5wUZ7g+IuaTmERpKQG8oPa/dyc8/GBDdobdcd3IrPsVQCQuvwxOUtaBsZyvjfdpSresvOK+DVXxJIlzAub5hPh8wFSGEedLgBjuzjaPJOBrWtR9DRJKKatGJM72g+WLCT/1vmDDh0Hw9xYD3UbW3Hv5QV0Yxj+7fw0g8bmb8tpfLtiCdBAwTYrmb+IR6rmcKC/BjdqzE/rt/PZ0sSGfz6PP41ayvfrd7LqHcW0/vlX0k5msvontE2fbAf7SPDWBhfUs2yZf8R/m/KOo9TS69JyuBITj59m5c0fo3q1pCjOfkkpWXx4ZhuPDmoJR+P6UZ05hp803eyvu0TfJvgy8cbC8iv3qC4aG6MoXrKGhIDW9Gm5yDqyyH+9+2vpe6YZqzfT1pWHjf1sP/AXaNrcl//piSkZvFUj0C+qDuByPc60nfbP3j5klCmdl1Ptbw0AgY+R6C/L3fe+0c2BnbmUfmCuIjyxfVuMTXp2SSCt+faC+2vm5Pp3bQWgX4+dGoUTn5hIYd+eA7WfUnBRX/mgTUxzAm4hEt91/Dt6Fi+uLsHr13fkT1pWUxcksitPRsz7aE+DOsYydu3xPHbExfzxOUtiAwP4vu1e3no81X8e1ZJW8uhzFxe+H4jHaLCuKNPDIt2pDL6g6Xc9N4Snp26gdz8AoZ1jLTjCH5+DqJ781Hnrxid/ThGXLD8Q1YmppGbX8jVPovsRbf3o3RsVIM1u9MxdduR4xtC+7y1NHIdJD63Zqnzf/u3HbyxpxmFuLi+2mrkWBo+KZtp3GkAn93VnSPZeQwft5C/TFnPjNyONMvfihxNJrZOdR67tDkzNuwnY8dy21jqKvn3nLF+PyPfXsRXO/wIy9iMH/kM7d+dl4e358Wr27I0rbpNuPl72/4Q3ad4INsvP33PnC0Huf2jZWTl5rN05yHaR4UR6OcDhYVI4iKSQjuRfDSXDb5tqJde0qOmdkgAt/eJYcaG/XaMTsZuO/7CvYoJ2LVzG9Wz95IdWMcO6KphA4crYzfPD23DX65oxYwN+7lvQpm69i3TKYxoRlrbMbZkkrKViUt2cZksxSDQ73EOxo5gZOF0Zs5fXO77BrZqaX58Ck8NaUXDmsG0iwpj4p3dqR7oy229onnmytaICGN6RTOkiS/9Zl1B/Pu389Q367jqv/P5w6TVFOxeDpFdeHvuDnx9XNzRJwZqt7AHSNli28qC7U3cff1j2ZmSycwN+0vl49NFu9ibkY1veBRRvmlc5buEdP960ON+ALrIFga1qQtpCbhqNuHZq9rw+qiO/HAgnEwCObz8K7sjY2yVV9nqJUeyfxRBx/YzYd4mbn5/CZ3/Oov7Jqzw2BPvVGmAAPuFbtIf4n/12Jvi9t4xBPr68H9T1lFoDBPu6M7Kpy/j3yM70KR2dXo1jaBPbMkFvldsLVYmppGZk092XgEPfraSz5YkMsFDt8H521IQgd5Na9k53zNTiGtUg38Mb8+3D/Smf3M7MKdDwxr8o8VWskwA180J59FJq3l26gZmZzWhcNciMIbN23fSwOwnILor1ZvZO11X4mKmryv5Ik9YvIvGEcH2eI4n+tVlU5dp3LPmOlzrvrI9J5Z/CG90gjmv2PryRt0BCPDzJXbM2wT7GEZvfdDjYKhHLm3GwSM5vPTDJhIPZXFRSztwqmNkCM/7fkSDNf+Fjjfza+3R7D+cTf3+dyCmANZOQkS4ulMk8/uuZ1PkSzx/UQ17IXM0rBnMAxfH8tGYbqx+ZiDXxzXkf7PjWTjnB5j2KK9/O4/Dx/J4ZUR7/m9IKxY/NYB3bunCxDu7s/T/BrDm2YEMaVcf1n5pL3SD/0Hd6FbsLQznSOOBsGoCS7ftxccFLRI+sY2ETS6iY8MaHM7OZ+OBTBYUtGJAwAYaSCrLMkKLA/+x3AI+XpRAp5axuKJ7244ERb2JGvWiU6NwJt/Xi0A/H75ZtYfsmEvtum0zAbirbwzNagUSdGgTxq39YX9GNg9/vorkw9lENW2Lr9gLQWS0vYBd3SmSps1aApCx+BO7UeM+HAluSAYhdPfbzivD27E2KZ27P1nB+j0ZdHOqlzi4GY4dQqJ7A+DftA+SllBquojb+8QQGujLaz9vc+vBFG3fQ+phxIfNmzfSyJVCQG2nZOEXZAfMZSQiItzZtwlPXN6C2VsOlnT6yM6AhPnMc8UxZIYNcJt/+5LPliQyPHgNEtUVQupRZ+jzFIovIQv+VqoUYYzh+7V7eemHjfRqGsFN3UragtpGhrH4qQE8N7SNLS1ip9H5T8A71JU0IvfN4te1Owj292HB2s34ZCSwjmZ8vSKJ67pEUSck0I4/CK5lx0JklUzUN6htPWJqVWPcnO3FN1+HMnP53+x4+jevTfXajfFL2UIf1zp+KOhOYe3WZEowg0ITaOCXaauFnd/fsI6RfHV/fya4hhGa8CNm5zw7RiI7vXwDNZCZk8+49fayPXtMQ94bHcdVHRqQV2Dw9Tn9l3Pf077H36uml9i7r9Ttto7PTUT1AP52bVv2pmdzR5+Y4gvWtZ2juLZzVLld9Ymtxbg521m68xCLdqSyLfkojSOCeXfeTm7tFV3qgjc//iBtG4QRvnOarQM3hYhPACNrNILar0HtPjZhfi51k2aQ1XIIE3tfQo0gP/ZlZPPThz9xWeF8SE9k9eJfaQXEdroI6rTGBIYx0H8HD32xis+W7qJL45qsSEhl7JDWuFwl3Sfll+cI2vQVdLsLej8KofXtiNIl4+2YikufK3V+/nWbw63fwmfXw/uXwy3fQJ1Wxet7NImgW0zN4n70F7eoDXnZ1J5xD7f6zuLnGtdz6dD/8vknK6gdEkCP7j1hY1dYPRF6PQSbphE42znmhBFw+4+2t4fze2DPcvCvhk9QTf7a0zB86xt0m2MvxK3yd3Jv/3/Qsp7tfhvo58PANvVK/4GMsb2yGnaHiKa0yLc9mFbVG07/hB85sGgSt9aph8/BTXD1OBChg1Md839T1tMhtxWX+NnG3x35tZixfj/Du0QxeWUShzJzubtfE0gZZsdNrPgIfPyLuzM2rV2db+7vxZuz47m9bwx8FGV/x51H4+vj4o+dDAHzctnq05TmTnbfm7eDAmP49I7uNEzOggRnhVvj+O1X9Ie3IGz/Ig4ERHPoSAD//XUdIwpjuTQskaCujTAGxn5j6/uL2h+Kuuy2730FfwgrpEOzerD1VVsqbTcCsKXou/o24V+ztpLUYCdRQFpAJK98vZZ521KYVBhOVsouegSl4RPu1rBeo1GpG4jRPaMZN2c778zdwbibu9jBfIV5vLm3OY1imrF1fwyH104jMLchjQO3QasX7PcztAF7Wt3BJZveIumllhyt3YWwJp1ZuWk7pCXwRcBR6l7831LfaaA4MBRbMp6AnT9zrOW1BG3+hsXX5iLte7Jr0V6YCX9dHUw+hdzTr2nJNrVb2N6AmSnFYyB8XMI9/Zow9pt1LIhPJb+wkLFf2xqCPw1qAasawJbp+AKTsuII25hMSH4s3YK3lh5D4mhVP5S1l/6RPTN/JvS7JwgZ9IxdUa8dS3akkn4sj77NahHs78vT361n8+Ga4A/18nZTr203Lm1dF2/REkSRppfYdw/VTADXdIrigYtjS13cy9nxG/z0NF0ahRHg62LcnO28O28HN3ZvxCvD25NyNIdJy0pGux7JzmNVYjo31kmAKffaC9agV6D7PXa05bRHSrr+bf8VjqUR3OUGOjcKp0nt6vSOrUVUh4sB2Lr8Z47uWEwhLkJiuoLLhTTqyWXVtnNXvyYcOJxD+G9/YW7Ao1zXyq1RNXU7rJoAcbfD4FdscAD7ftnz8MDiUhf/Yo17wpjptu3mg0HlBoc9OsB2x2tetzpR/sfg02tg01S+rn0/z2aPYs/hHOZsSeb6uIb4+big4032bnblJzDlHlvtd9PXdqqDz2+03ZE3TYM3u8GHg+HtfvBaW/zf7UccG3jL50Ym5g/gOt+5PNRRyufX3b7V9lgdRgEQXasafj7CnfOC2V5YnzsCfuGJ0Fm2G2lbe5FsVieEYH8f1uxOJ7NBn+Jd5VSP4uuVSRQUGt6bt4MOUWF0j6kJLa+wCbbNtMHBL7B4m7qhgbwwrC1RNatB88vteBanTvySGra098EOG+DSMnP5bGkiQzs0oGHN4NLtEmElNyfVIqIwYr+bPx9rzuDX5zF93X6qN+1JUEY8HEtjVLdGjB3cknqhgcQ5vbjYtRBCIwmu04SHBzQjKKqjrW7dtbDUr+y23tHUCPZj3brVFLj8uezdLXyzcg+dGtXAJ7whA+sfIyzvYHHVElAuQFQP8OXmHo2ZsWG/HV+x5UeO+oSx3tWC/97Yidg+I+jq2sYbLZ1OFy2vLN626fDn2Rn3DPuDmxOevIj6S15iYMZX9K2+h7aykzpLXj7B33wtzHoGmg8maOR7ENIA2WCfetf42CaM+NCr7wD+NKgljSKCS7ar1dw2GOcfKzUG4prOkdQJCeCRL1Zx24fLqB7oy9f39bLtck7DfWFYIza5mvLsdxtYVtiCmpnb7XcPSo+dAa7p1oy3/W8jJH0T5hcbGNfnR3HL+0u559MVdP7rLEa+vYhvVu5hUL/egJyRrq4aIIrUjLE9BuI9B4hSjh4s/8c5tAMm3QIL3yAw/ke6RtdkacIhosKD+POQVnSPqUnX6HDG/7ad3PxCcvILePjzVTQzCYyMf9JO53zD59DjXhj4VxjyTzvnyvIP7f7XfWUn9ioKZI6bhw4hkyBWzf+RZnlbyQyLhQCnPrpRT3zT4nmyTwQ/X57GGN+ZREkK4b/+qaQqbc7LdlbJvo9X/XdWry3cPtM2qH52fanRnT2bRnBDt4Y82hH77OI9K2DEB6R3uJs96cf436/bMMD1XZ0LSttrbeP7tIftCOLrJ0CzS+Gat+3d7GvtYdLNtt//iA/g+ol2VPKQV3E9upaet/6NqeGjwTeAgPn/KMljYaF9/OXR5JJla76wU2I4D/nx83HRNbomUeHVKOgyhtjcTQQl/mZLVL52xlcfl9Au0jbKj7j8Ehs8gA5t27NoRyofLUxgV2oWd/drau9cQxvYIOf8HSrUfJBtM1j1KRiDf/I68lxBfLkzkM37D/PRwgSycgu47yLnrrbowhJcC/yrlezHxxdxHlhz9dUjGTu4JQ9eHEvXvpfb9c6AuXv7N2XRU5fY3mfGGfTXuFfJgDyXj61OLAoQhYWQup0QP7F31mkJ7MyvRb3wYL5/uA//u7Ez9RvGUj11va0mdO/yW6OR7Vnl1otqTK9o/FwuPpi3lYKtM5mR24FbejahTkggrpaDEQrpsvsTqN0KItzu5H39ibnyj8T96Qcy7lvHx31/JfmRRMKeXI/0fcwGYk8j2MEOUps8xpYAhr1pz7HttbBtlh31nLQcqdOaR4d04t7+TUtvW7tFyZgSt4n6Anx9uO+ipqRl5XJP/yZ8/1Cf4lJmUeB2tbmavs3qkJqZS3J4JwRjqzahZP4uh7+vizaX3crSwhbIwU0UhEVzz5dbqR0SwPu3xjGqayP2pB2jf/PaPDCwne0csGe5nQ78y9Hw4RWez/0UaYBw13SA7UOef5wZSfOOwUdXwFs97IWmaNmXo+0/WY3GMOdlLmoegQi8OqID1QJ8EREeuDiWfRnZTFqWyH0TVpK8dSlfh/4bn6Aw2yupqBoF7J1lTD+Y83fboLplOrS52raXuAkK9CevfhwdzGY6+mwnOKZ7ycqifvfrvrQX3gad4eK/2DvxdZPtKOx1X9kSS8hJFlNrxsBNX9r5az4bVTyWQET4e7tkhiy5BXKOwG3fQ9vhxT1nPl+6m37Natu7YoDAMCdIBMKoiRDiVAu1vdYGS78guOJfcO8CO/11qyvtqORud0FwTTo1CmfSH6/Gp/s9JeeWn2OnbZ48Bj4cYoNEQZ495xaDS/2+J9zRnV//2J/mA++2gco3yJaq3NzcozG39YqmZ2wt+7cRFwO6d8EYO/CwUc1gBrV1q85qdVXpv4MnMf2gQSdbHfXJUNgxB6nfjgA/P17/eRsfLUzg0lZ1aV7XGW/iHwwhDUrfqRdxeg5Va34R9/ZvyuOXt0Ai4+zYCbcBc1KQa6s6Nn8PRw/Y6aTdNe5lx2L88Dj8pw38tzO83Y/bIvfQKuAQrogYptzfuyRPYVF2+gcoHyAK8+BoSRtYndBArukUyc6Vv+CTk8FcibNVcgD1O9lHtxbklJTAPGheL5RbB3QhsqZzI9TtHntT8ds/yifOOQqfXWcD1YgPSqbrbnutzdumaTawRHXxfLBazUt+LjOK+rZe0ax8+jKeGtyqdM1C/Y42GHW80bZ3AdHt+9mpupOWQkj90l2jHdd2acg7wXdRiLAqN5KDR3J466bODGhVl+eGtmHB2Ev4+PZu+LjEDpjb9pP9v969zP4flh29fxpoG4S72AGw7F3bbdTpzljOL3+1vRrqtbNVIanx9sKzfx3cMMlOC/zNXdwavp5L/3gZ0bWq2eH2iYvoH9medpFhPP3dem7zmcm7gZ/j41/bBoeixxUWEYGBL8Lb/WHCcDs6t911HrNUo2U/auxzBkc1jCtZUb+jvdDN/DP4V4cR79sAtu0nmP5H21MmIAR6P3Jqv7c6rWDEh/Yf8es7bd314nH2Dqd2S7hxUvGdb5sGofj5CHkFhhvcGhUBuOLfcMnTJdVcRbrdZV+V0fsR+zzkWc/YoJUwD+LusG0On1wNPR+wDY4dbii1WXH9dVC4LcG5fEpPkgZc1aEBV3VwurVeNBaaX06jOmF0c0qLd/aNsf+8RbqMAcQ+KrIifoFwx8+w4kP49UXITse32z1cX7chHy1MAOD+i8vc1XYYBUE1yu+rYVfb8F7dbcbRgOp2EsGdc+2Fad1kp92hqDOGlP+ux1wEvGCr+2IvhUb3wtJ3CZp4FdGILdm5N4i6VXUVzwsFbpMIJpY8TAe4q280G9b8Qo7xpUmPq4ioHmBXuFy2RLXyY3sDUFmBofbvOvslO4q8QUe7PD/Hljr3rrIlUvdA3aCz/U7O/48d6FdU2iurqCcTlJvqW0SoEezvYZvm8Cc7OHBIjXzW78ngup6xsKuDLUmXqV4q4ufj4vJLB/HoNzvZntOAZ4a1LimVlDXgaWg20Hauqd2y9JQsp5EGCHfRfWyUX/kJNO5TqpshYAcgLX4Lut4Jl/8dfngM5v7TruvzB2gxyM6yOPef+M17heh7h9rGrYnXwd6VCMKEmu1Y719Ib9d6aDYIhr1V8UNI6newF4M1n9tphBv28JyuoVupIdItQPj621GaCfPs3XfRoJurx8H4Pnb5Rf9X7kJ4Uppdagfu/fgn2PqjrTIb8qp9VKZbVUignw+tG4SxL/0YA1rVKb0Pv0DwKxMcqiq4pr1YzPm7/Vte87b9HbYeChNHwnf323/02AEV76MywSiiaXEVyN39mlBoDNd1KXNXHxgKvR8+8b58fO0x21xrA0Wba7hD6vPJogS6x0TQuVF46fSXPut5P5c+53kq8ag4u9/ERfaOuO8fbFtGaH07+VvZC1ZUF7hrtl0X6My11fUuezFd8Hr5i2lxUJDSwcI9QDQq+e7Gxn9IrM9CxjOc2y4q01On9yM2P2UHCZ5I93vslB2//cMGg6SlNr87ZttqpbIlEhFbEp33L/vZ/f/GXWikvbnKPXr8hwVVINjfl+eGtrEfGvU8boAAuKZTJBOXXEHHBqHc1L1Rhelo0Mm+vK1oegNvvIBBwBYgHhjrYX1LYBGQAzxelW09vbp06WJO2U/PGPNsqDGf32hMztGS5dmHjflPW2Ne71iyvLDQmEVvGfPdQ8bk55WkXfuV3ce8/9j0f61jzLIPjJnzijHvDjCFf4syZuH/7PYnkr7bmBfrG/PzCxWnyTlqzHPhxrxYr3Q+jDFm+2xj5r9efpuVE4x5d4AxxzJOnIfKKiy0+938ozEFBRUm27Qvw6zdnX76jlvWsQxjvr7Lnru7LTONeT7CmJl/8d6xT6PZmw+YxNTMU9/RwW3GzP67MXtWVe47dzwFBeX3sX+9/b6/2rL08pxMu/y3f5Ys2/CdMc+GmdzPbzGJKUdOLS9lzf67Pd4/Yu378xHGLHyz4vRF+f5b1HG/r+bt/jbdqf6vbPjO7mf234+brPBU/0ZVBCw3FVxTxXhpOl0R8QG2ApcBScAy4AZjzEa3NHWAxsDVQJox5tXKbutJXFycWb68gqczVcWSt+1sqHXb2mfBHtgAO3+z72NmFI8JqFBhAbzV01ZFBYXDjV+WPH3rZBxNtvsp0/5QyvuX23rN0d+e/HEuBEf22xKEjxaeT5tj6fBKY1vCvWNm6XX/jIWY/tDzfjs9xLRHbeeGW6d5rIc/5XxMHGHv+ltdZacaCSw/2r+YMTC+ry1J3fRVxem+uQc2fAN/ST61qpysQzCuF1wz/vjVjmeYiKwwxngsQnnzv6QbEG+M2eFk4gtgGFB8kTfGJAPJIlK2ReqE23pV93tsEXzyGPj2PttwWrcNXPnaiYMD2PrrQX+3xdcrX7N1kqeiep0Tpxk1EfBOPeR5JaTeidOoqgkMsz3ZPDWch0fbx3eun1zyedRnpz84gG2XufPnyqcXgVumeJ7Owl2vByGm76nX8wfXhD9uPrV9nGHeDBCRgPszEJOASlxdq7atiNwN3A3QqNFx6uyqqvlAeHiVbdCMaFb1O87YAcev5z7dTjRPvVLeImLr+ctOLAe2B9qeFfauPqS+bQPxDy6f7mxxb9CvSL12FU57cb7zZoDwFG4rW59V6W2NMe8A74CtYqrk/iunep3K3b0rdaFrPdTz8jPVmKq8wpvjIJIA9zJnFLD3DGyrlFLqNPBmgFgGNBORGBHxB0YBU8/AtkoppU4Dr1UxGWPyReRBYCbgA3xgjNkgIvc668eLSD1gORAKFIrIo0BrY8xhT9t6K69KKaXK81o317PhtHVzVUqpC8TxurnqXExKKaU80gChlFLKIw0QSimlPNIAoZRSyiMNEEoppTzSAKGUUsojDRBKKaU80gChlFLKIw0QSimlPNIAoZRSyiMNEEoppTzSAKGUUsojDRBKKaU80gChlFLKIw0QSimlPNIAoZRSyiMNEEoppTzSAKGUUsojDRBKKaU80gChlFLKIw0QSimlPNIAoZRSyiMNEEoppTzSAKGUUsojDRBKKaU80gChlFLKIw0QSimlPPJqgBCRQSKyRUTiRWSsh/UiIm8469eKSGe3dQkisk5EVovIcm/mUymlVHm+3tqxiPgAbwKXAUnAMhGZaozZ6JZsMNDMeXUHxjnvRS42xqR4K49KKaUq5s0SRDcg3hizwxiTC3wBDCuTZhjwibEWAzVEpL4X86SUUqqSvBkgIoHdbp+TnGWVTWOAn0RkhYjc7bVcKqWU8shrVUyAeFhmqpCmtzFmr4jUAWaJyGZjzNxyB7HB426ARo0anUp+lVJKufFmCSIJaOj2OQrYW9k0xpii92RgCrbKqhxjzDvGmDhjTFzt2rVPU9aVUkp5M0AsA5qJSIyI+AOjgKll0kwFRju9mXoAGcaYfSJSTURCAESkGjAQWO/FvCqllCrDa1VMxph8EXkQmAn4AB8YYzaIyL3O+vHAdGAIEA9kAWOczesCU0SkKI+fGWNmeCuvSimlyhNjyjYL/H7FxcWZ5ct1yIRSSlWWiKwwxsR5WqcjqZVSSnmkAUIppZRHGiCUUkp5pAFCKaWURxoglFJKeaQBQimllEcaIJRSSnmkAUIppZRHGiCUUkp5pAFCKaWURxoglFJKeaQBQimllEcaIJRSSnmkAUIppZRHGiCUUkp5pAFCKaWURxoglFJKeaQBQimllEcaIJRSSnmkAUIppZRHGiCUUkp5pAFCKaWURxoglFJKeaQBQimllEcaIJRSSnmkAUIppZRHGiCUUkp5pAFCKaWUR14NECIySES2iEi8iIz1sF5E5A1n/VoR6VzZbZVSSnmX1wKEiPgAbwKDgdbADSLSukyywUAz53U3MK4K2yqllPIib5YgugHxxpgdxphc4AtgWJk0w4BPjLUYqCEi9Su5rVJKKS/y9eK+I4Hdbp+TgO6VSBNZyW0BEJG7saUPgKMisuUk81sLSDnJbX+vLsRzhgvzvC/Ec4YL87yres6NK1rhzQAhHpaZSqapzLZ2oTHvAO9ULWvlichyY0zcqe7n9+RCPGe4MM/7QjxnuDDP+3SeszcDRBLQ0O1zFLC3kmn8K7GtUkopL/JmG8QyoJmIxIiIPzAKmFomzVRgtNObqQeQYYzZV8ltlVJKeZHXShDGmHwReRCYCfgAHxhjNojIvc768cB0YAgQD2QBY463rbfy6jjlaqrfoQvxnOHCPO8L8Zzhwjzv03bOYozHqn2llFIXOB1JrZRSyiMNEEoppTy64APEhTKlh4g0FJHZIrJJRDaIyCPO8poiMktEtjnv4Wc7r6ebiPiIyCoR+d75fCGccw0RmSwim52/ec/z/bxF5DHnu71eRD4XkcDz8ZxF5AMRSRaR9W7LKjxPEXnKub5tEZHLq3KsCzpAXGBTeuQDfzTGtAJ6AA845zoW+MUY0wz4xfl8vnkE2OT2+UI459eBGcaYlkAH7Pmft+ctIpHAw0CcMaYttnPLKM7Pc/4IGFRmmcfzdP7HRwFtnG3ecq57lXJBBwguoCk9jDH7jDErnZ+PYC8Ykdjz/dhJ9jFw9VnJoJeISBRwBfCe2+Lz/ZxDgX7A+wDGmFxjTDrn+Xlje2UGiYgvEIwdO3XenbMxZi5wqMziis5zGPCFMSbHGLMT22O0W2WPdaEHiIqm+jiviUg00AlYAtR1xp7gvNc5i1nzhteAPwGFbsvO93NuAhwEPnSq1t4TkWqcx+dtjNkDvAokAvuwY6p+4jw+5zIqOs9TusZd6AGi0lN6nC9EpDrwNfCoMebw2c6PN4nIlUCyMWbF2c7LGeYLdAbGGWM6AZmcH1UrFXLq3IcBMUADoJqI3Hx2c3VOOKVr3IUeICozHch5Q0T8sMFhojHmG2fxAWcGXZz35LOVPy/oDQwVkQRs9eElIjKB8/ucwX6vk4wxS5zPk7EB43w+70uBncaYg8aYPOAboBfn9zm7q+g8T+kad6EHiAtmSg8REWyd9CZjzL/dVk0FbnV+vhX47kznzVuMMU8ZY6KMMdHYv+2vxpibOY/PGcAYsx/YLSItnEUDgI2c3+edCPQQkWDnuz4A2852Pp+zu4rOcyowSkQCRCQG++ydpZXeqzHmgn5hp/rYCmwH/ny28+PF8+yDLVquBVY7ryFABLbXwzbnvebZzquXzv8i4Hvn5/P+nIGOwHLn7/0tEH6+nzfwPLAZWA98CgScj+cMfI5tZ8nDlhDuON55An92rm9bgMFVOZZOtaGUUsqjC72KSSmlVAU0QCillPJIA4RSSimPNEAopZTySAOEUkopjzRAKHUOEJGLimabVepcoQFCKaWURxoglKoCEblZRJaKyGoRedt51sRREfmXiKwUkV9EpLaTtqOILBaRtSIypWiOfhGJFZGfRWSNs01TZ/fV3Z7hMNEZEazUWaMBQqlKEpFWwPVAb2NMR6AAuAmoBqw0xnQGfgOedTb5BHjSGNMeWOe2fCLwpjGmA3a+oH3O8k7Ao9hnkzTBziWl1Fnje7YzoNTvyACgC7DMubkPwk6KVghMctJMAL4RkTCghjHmN2f5x8BXIhICRBpjpgAYY7IBnP0tNcYkOZ9XA9HAfK+flVIV0AChVOUJ8LEx5qlSC0WeLpPuePPXHK/aKMft5wL0/1OdZVrFpFTl/QKMEJE6UPwc4MbY/6MRTpobgfnGmAwgTUT6OstvAX4z9hkcSSJytbOPABEJPpMnoVRl6R2KUpVkjNkoIn8BfhIRF3Y2zQewD+RpIyIrgAxsOwXYaZfHOwFgBzDGWX4L8LaIvODs47ozeBpKVZrO5qrUKRKRo8aY6mc7H0qdblrFpJRSyiMtQSillPJISxBKKaU80gChlFLKIw0QSimlPNIAoZRSyiMNEEoppTz6f6BAzf+LKbwFAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "l2_list = [0.0001, 0.001, 0.01]\n",
    "\n",
    "for l2 in l2_list:\n",
    "    lyr = SingleLayer(l2=l2)\n",
    "    lyr.fit(x_train_scaled, y_train, x_val=x_val_scaled, y_val=y_val)\n",
    "    \n",
    "    plt.plot(lyr.losses)\n",
    "    plt.plot(lyr.val_losses)\n",
    "    plt.title('Learning Curve (l2={})'.format(l2))\n",
    "    plt.ylabel('loss')\n",
    "    plt.xlabel('epoch')\n",
    "    plt.legend(['train_loss', 'val_loss'])\n",
    "    plt.ylim(0, 0.3)\n",
    "    plt.show()\n",
    "    \n",
    "    plt.plot(lyr.w, 'bo')\n",
    "    plt.title('Weight (l2={})'.format(l2))\n",
    "    plt.ylabel('value')\n",
    "    plt.xlabel('weight')\n",
    "    plt.ylim(-4, 4)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "range(0, 10)\n",
      "0 1 2 3 4 5 6 7 8 9 [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]\n",
      "[5 7 9]\n",
      "[0, 1, 2, 6, 7, 8]\n",
      "[1 2 3 7 8 9]\n"
     ]
    }
   ],
   "source": [
    "print(range(10))\n",
    "for i in range(10):\n",
    "    print(i, end=\" \")\n",
    "    \n",
    "l1 = list(range(10))\n",
    "print(l1)\n",
    "a = np.array([1,2,3])\n",
    "b = np.array([4,5,6])\n",
    "print(a+b)\n",
    "\n",
    "array = np.array([1,2,3,4,5,6,7,8,9])\n",
    "a= [0,1,2]\n",
    "b= [6,7,8]\n",
    "index = a+b\n",
    "print(index)\n",
    "print(array[index])  # a[ [0,1,2,6,7,8] ]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[  0  10  20  30  40  50  60  70  80  90 100 110 120 130 140 150 160 170\n",
      " 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350\n",
      " 360 370 380 390 400 410 420 430 440 450 460 470 480 490 500 510 520 530\n",
      " 540 550 560 570 580 590 600 610 620 630 640 650 660 670 680 690 700 710\n",
      " 720 730 740 750 760 770 780 790 800 810 820 830 840 850 860 870 880 890\n",
      " 900 910 920 930 940 950 960 970 980 990]\n",
      "[100 110 120 130 140 150 160 170 180 190]\n",
      "[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99]\n",
      "[  0  10  20  30  40  50  60  70  80  90 200 210 220 230 240 250 260 270\n",
      " 280 290 300 310 320 330 340 350 360 370 380 390 400 410 420 430 440 450\n",
      " 460 470 480 490 500 510 520 530 540 550 560 570 580 590 600 610 620 630\n",
      " 640 650 660 670 680 690 700 710 720 730 740 750 760 770 780 790 800 810\n",
      " 820 830 840 850 860 870 880 890 900 910 920 930 940 950 960 970 980 990]\n"
     ]
    }
   ],
   "source": [
    "a = np.arange(100)\n",
    "a *= 10\n",
    "print(a)\n",
    "start=10\n",
    "end = 20\n",
    "val = a[start:end]\n",
    "print(val)\n",
    "index = list(range(0, start))+list(range(end, len(a)))\n",
    "print(index)\n",
    "# # # b = list(a)\n",
    "fold = a[np.array(index)]\n",
    "print(fold)\n",
    "# fold[0] = 999\n",
    "# print(fold)\n",
    "# print(a)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(455, 30)\n",
      "0.9711111111111113\n"
     ]
    }
   ],
   "source": [
    "print(x_train_all.shape)\n",
    "validation_scores = []\n",
    "k = 10\n",
    "bins = len(x_train_all) // k\n",
    "\n",
    "for i in range(k):\n",
    "    start = i*bins\n",
    "    end = (i+1)*bins\n",
    "    val_fold = x_train_all[start:end]\n",
    "    val_target = y_train_all[start:end]\n",
    "    \n",
    "    train_index = list(range(0, start))+list(range(end, len(x_train)))\n",
    "    train_fold = x_train_all[train_index]\n",
    "    train_target = y_train_all[train_index]\n",
    "    train_mean = np.mean(train_fold, axis=0)\n",
    "    train_std = np.std(train_fold, axis=0)\n",
    "    train_fold_scaled = (train_fold - train_mean) / train_std\n",
    "    val_fold_scaled = (val_fold - train_mean) / train_std\n",
    "    \n",
    "    lyr = SingleLayer(l2=0.01)\n",
    "    lyr.fit(train_fold_scaled, train_target, epochs=50)\n",
    "    score = lyr.score(val_fold_scaled, val_target)\n",
    "    validation_scores.append(score)\n",
    "\n",
    "print(np.mean(validation_scores))"
   ]
  },
  {
   "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}