{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "RcxSIxsutE5p",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "ename": "ModuleNotFoundError",
     "evalue": "No module named 'tensorflow'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mModuleNotFoundError\u001b[0m                       Traceback (most recent call last)",
      "\u001b[0;32m/tmp/ipykernel_3043/1373220103.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0;32mfrom\u001b[0m \u001b[0mtensorflow\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mkeras\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      2\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mtensorflow\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mtf\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mnumpy\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mmatplotlib\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpyplot\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mpickle\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mModuleNotFoundError\u001b[0m: No module named 'tensorflow'"
     ]
    }
   ],
   "source": [
    "from tensorflow import keras\n",
    "import tensorflow as tf\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pickle\n",
    "plt.style.use('ggplot')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "ILKwfwbNtE5q"
   },
   "outputs": [],
   "source": [
    "global_random_seed = 3884 # np.random.randint(0, 99999)\n",
    "np.random.seed(global_random_seed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "id": "248sIW-3tE5q"
   },
   "outputs": [],
   "source": [
    "def get_network(input_dim, output_dim, width, depth, regularization_param,\n",
    "                act_func='tanh', output_act_func=False, random_seed=None):\n",
    "    \"\"\"Construct a sequential model based on the input configuration\n",
    "\n",
    "    Parameters\n",
    "    ----------\n",
    "    input_dim : int\n",
    "    output_dim : int\n",
    "    width : int\n",
    "    depth : int\n",
    "    regularization_param : float\n",
    "        Number of hidden layers + output layer\n",
    "    act_func : str, optional\n",
    "        Which activation function this model will be using, by default 'tanh'\n",
    "    output_act_func : bool, optional\n",
    "        Should the output layer use an activation function, by default True\n",
    "    random_seed : int\n",
    "        Seed to initialize bias and weights of the network\n",
    "\n",
    "    Returns\n",
    "    -------\n",
    "    keras.Sequential\n",
    "        The constructed model based on the input arguments\n",
    "    \"\"\"\n",
    "    # initializer = keras.initializers.RandomUniform(\n",
    "    #     minval=-1., maxval=1., seed=global_random_seed)\n",
    "    initializer = keras.initializers.RandomUniform(\n",
    "        minval=-1., maxval=1., seed=random_seed)\n",
    "    regularizer = keras.regularizers.l2(regularization_param)\n",
    "\n",
    "    if act_func == 'sin':\n",
    "        act_func = tf.math.sin\n",
    "\n",
    "    hidden_layers = [keras.layers.InputLayer(input_shape=[input_dim])]\n",
    "    hidden_layers += [keras.layers.Dense(width, activation=act_func,\n",
    "                                         kernel_initializer=initializer,\n",
    "                                         bias_initializer=initializer,\n",
    "                                         kernel_regularizer=regularizer,\n",
    "                                         bias_regularizer=regularizer) for _ in range(depth-1)]\n",
    "    if output_act_func:\n",
    "        hidden_layers.append(keras.layers.Dense(\n",
    "            output_dim, activation=act_func,\n",
    "            kernel_initializer=initializer,\n",
    "            bias_initializer=initializer))\n",
    "    else:\n",
    "        hidden_layers.append(keras.layers.Dense(\n",
    "            output_dim, kernel_initializer=initializer,\n",
    "            bias_initializer=initializer,\n",
    "            kernel_regularizer=regularizer,\n",
    "            bias_regularizer=regularizer))\n",
    "    m = keras.Sequential(hidden_layers)\n",
    "    m.compile(optimizer=keras.optimizers.Adam(learning_rate=1e-3), loss=\"mse\", metrics=[\"mse\"])\n",
    "    return m\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "id": "cR8vLGsVtE5s"
   },
   "outputs": [],
   "source": [
    "def gene_test_data(num_points: int, s_dev=0.3, low_val=0, high_val=1):\n",
    "    _x = np.linspace(low_val, high_val, num_points)\n",
    "    clean_data = [x + np.math.sin(10*x*np.pi) for x in _x]\n",
    "    clean_data_pairs = list(zip(_x, clean_data))\n",
    "    noise = np.random.normal(loc=0, scale=s_dev, size=num_points)\n",
    "    dirty_data = clean_data+noise\n",
    "    dirty_data_pairs = list(zip(_x, dirty_data))\n",
    "    np.random.shuffle(dirty_data_pairs)\n",
    "    return [clean_data_pairs, dirty_data_pairs]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 265
    },
    "id": "_iHJX4sRtE5t",
    "outputId": "151a799f-26f3-4f27-ed46-066539d9a19a"
   },
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'gene_test_data' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m/var/folders/6h/mmvp5df90fb3fsqtc5d9mxt40000gn/T/ipykernel_3949/3684707111.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      2\u001b[0m \u001b[0mnum_train\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnum_data\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0;36m0.8\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0mnum_test\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnum_data\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0mnum_train\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 4\u001b[0;31m \u001b[0mclean_all_data\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mall_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mgene_test_data\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnum_data\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      5\u001b[0m \u001b[0mtrain_data\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtest_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0msorted\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mall_data\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mnum_train\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkey\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mlambda\u001b[0m \u001b[0mpair\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mpair\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msorted\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mall_data\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mnum_train\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkey\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mlambda\u001b[0m \u001b[0mpair\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mpair\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      6\u001b[0m \u001b[0;31m# train_data, test_data = all_data[:num_train], all_data[num_train:]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mNameError\u001b[0m: name 'gene_test_data' is not defined"
     ]
    }
   ],
   "source": [
    "num_data = 125\n",
    "num_train = int(num_data*0.8)\n",
    "num_test = num_data - num_train\n",
    "clean_all_data,all_data = gene_test_data(num_data)\n",
    "train_data, test_data = sorted(all_data[:num_train], key=lambda pair: pair[1]), sorted(all_data[num_train:], key=lambda pair: pair[1])\n",
    "# train_data, test_data = all_data[:num_train], all_data[num_train:]\n",
    "x_train,y_train = zip(*train_data)\n",
    "plt.figure(figsize=(20,10))\n",
    "plt.scatter(x_train, y_train,label = \"train\", s=3, c='purple')\n",
    "x_val,y_val = list(zip(*test_data))\n",
    "plt.scatter(x_val ,y_val, label = \"test\",s=9, c='r', marker='d')\n",
    "# x_val,y_val = list(zip(*clean_all_data))\n",
    "# plt.plot(x_val ,y_val, label = \"clean\", c='g',alpha = 0.3)\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 962
    },
    "id": "wbOGtKantE5u",
    "outputId": "1edc61e8-0834-4a89-a8b8-12e4ee8dd979"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training reg:1e-06, run_num:0\n",
      "\n",
      "Training reg:1e-06, run_num:1\n",
      "\n",
      "Training reg:1e-06, run_num:2\n",
      "\n",
      "Training reg:1e-06, run_num:3\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training reg:0.001, run_num:0\n",
      "\n",
      "Training reg:0.001, run_num:1\n",
      "\n",
      "Training reg:0.001, run_num:2\n",
      "\n",
      "Training reg:0.001, run_num:3\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeEAAAFJCAYAAACsBZWNAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAA8vElEQVR4nO3deXxV9Z3/8de5W3Jv7s2+sISwhEUUUcSl1IVFBStgUay4jLaj06qjD62ddtytVqZIl2nrVmsXfzPWGaXYWjvaulQstiqKgij7viWQPblL7n5+fyS5iBASSCB88f18PHyEnHNz7jcfE958v+d7vl/Ltm0bEREROeIcfd0AERGRzyuFsIiISB9RCIuIiPQRhbCIiEgfUQiLiIj0EYWwiIhIH3Ed6TesrQ326vUKCnw0NkZ69ZqfR6pjz6mGPaca9g7Vsed6u4YlJYH9Hje+J+xyOfu6CccE1bHnVMOeUw17h+rYc0eqhsaHsIiIiKkUwiIiIn2ky3vCqVSKe+65h82bN+N0Opk3bx4VFRWZ80899RQLFy6ksLAQgAceeIBhw4YdvhaLiIgcI7oM4UWLFgHw7LPPsmTJEubNm8fPf/7zzPmVK1cyf/58xowZc/haKSIicgzqMoTPO+88Jk2aBEBVVRXFxcV7nV+5ciVPPvkktbW1TJo0ieuvv/6wNFRERORYY3V3F6Xbb7+d1157jYcffpizzjorc/zRRx/lyiuvxO/3c/PNN3PFFVcwefLkTq+TTKY0c09ERISDCGGA2tpaLrvsMl566SV8Ph+2bRMKhQgE2p5/euaZZ2hqauKmm246wDV69znhkpJAr1/z80h17DnVsOdUw96hOvZcb9ews+eEuxyOfuGFF9i9ezfXX389Xq8Xy7JwOtt6sqFQiBkzZvDyyy/j8/lYsmQJs2fP7rVGi4iIAHz44VJuueUGHnjg+5x77tTM8a9+9XJGjjyOu+++n1WrPuGXv/w5tm1j2zZf+MKZXHHFP/Hhh0u57747GTJkaObr8vMLmDt3fl98K3vpMoSnTp3KnXfeyVVXXUUymeSuu+7i1VdfJRKJMGfOHG677TauueYaPB4PEyZMYOLEiUei3SIi8jkzePAQXn/9lUwIb9y4gdbW1sz5n/zkB9xzz/cYPHgIyWSSG264lvHjTwVg/PhTeeCBeX3S7gPpMoR9Ph8/+9nPOj0/a9YsZs2a1ZttEhGRXrbgjQ28v6amV6952nGlXDZleKfnw+EQDz00l1AoSHNzEzNnXszkyedx003/wm9/+zssy+LHP57PqaeeTklJCf/5nz/A5/NRUFCAx5PF3Xffv9f1hg8fwfbt2wgGgwQCAV555WWmTv0Su3fvAqCsrD/PP/8cF154ESNGjOTnP/81brebDz9c2qvfd28yerGOWDzFG0u3E0+k+ropIiLyGTt27OC886byk588xg9+8FOee+4Z8vPzqawcwUcfLSMej7Ns2QeceebZ/OhH87jrru/y8MNPMGBAeafXPOecySxevAjbtlm9eiVjxozNnLvzznspLCzixz+ex8yZU3n00Z8Qj8cB+OCDpdx88zcy//3P//z3Yf/+u+OIb+DQm5atr+XJP63ixlljOO240r5ujojIUeuyKcMP2Gs9HIqKiliw4H/4298W4fPlkEwmAZg5cxZ//vP/UV9fz1lnnYPL5aKuro5hwyoBOOmkcfz1r6/u95rnn38BP/7xQwwYMJCTThqXOR6LxVi7dg1f+9q/8LWv/QvNzU3Mm/c9Xnzx9wwbNvyoHY42uiecSKYB1BMWETkK/e//Ps2YMWO5774HmTLlPDoexjn11NNZv34tL730IjNmfBmA0tIyNm/eBMDKlR93es2BA8tpbW1l4cJnmTr1S5njDoeDBx+8j02bNgCQl5dPv379cbs9h+vb6xVG94Sx2j50/yErERE5Us488xx+9KN5vPrqn8nLy8PpdBKPx/F4PEyadC5Ll75HefkgAP7t325n3rzv4fX6cLtdlJR0Prp57rnn88orL1NRMZiqqp0AuN1uvve9efzwh98nmUxhWRajRx/P9OkXsWLF8sxw9Kf9+McPk5WVffgK0A0H9Zxwb+jN567+vqKa37y8mn++8DjOHjug1677eaTnCntONew51bB3mFjH559fwJQp51NQUMCTTz6O2+3mn//5633WnqPmOeGjmdXeE0Y9YRERoxUWFvKtb92E1+vD7/fvMzP6WGV0CHdQBouImG3y5POYPPm8vm7GEWf0xKxMT1hERMRAZodw+8ysI3xbW0REpFcYHcKZ2dF92woREZFDYnQIZ0ajlcIiImIgsydmqScsIvK5cNNNX+faa7/B+PGnZY799Kc/orJyODNnztrn9ZdeOpNnnllIVlZW5tjNN3+DxsYGnnlmYebY3/72Bnff/e/87ncvUlbWj8ce+ykbN24gO9tDOm1x663/xsCB5dx88zeIxaJ7PVd85ZXX8MUvntWj78voELa0WoeIyOfCRRddzF/+8lImhBOJBP/4x1tcf33n+9d3Zv36tYwYMQqA119/lX79+gOwZMnb1NXV8tOfPk5JSYDnn/8Tjzzynzz00H8CZHZo6k1mh7B6wiIi3fL7Df/HsprOl4M8FONKT+SS4TM6Pd+buyhNmnQuTz75ONFolOzsbN5662+cfvoZBIMtfPe7dxGPx2hpaeZrX/s655wzqdM2nXfeNF577RVGjBhFMBgkHo9RWFgEQGlpP9asWc1f//oqU6dO5uyzJzJhwpm9Va79MvqecAd1hEVEjj69uYtSVlYWZ589kcWLFwHw8ssvctFFl7B16xYuv/wqfvrTx7nttn/n979fcMA2nXnm2bz77j+wbZs33/wrkyadmzlXWTmc22+/h8WL32TGjBlcd93VfPLJisz5uXPv22snpsbGxh7XyPCesB4UFhHpjkuGzzhgr/Vw6O1dlGbOvJjHHvsZp5xyKsFgkFGjjmPTpo3813/9mpde+iNgZd6jM1lZ2YwYMYpPPlnB4sWLeOCBefzhD233iDdsWE9FxWAeeOD7FBf7efnl17nvvjt58cVXgMMzHG10T7gjgtPqCouIHHV6exelysrhtLaGWbDgf5k+/SIAfvWrJ7jggunce++DnHLKqd1q1/nnX8Bzzz1Dbm4uPp8vc3zp0iU88cSjpFJtG0AMHTqM7GzvYe3wGd4Tbv+DMlhE5KhzOHZRmj79Ih577GGef/7/AJg8+Vx+9rMf8fTTT1FaWkZTU1OX7TrttDP4j/+4n7vuum+v45deejmPPfYzrr32KvLyckmlbO6994HM+blz79trdvS5507l4osvPdiy7MXoXZQ+WFvLY3/4mMvPHcHU0wb12nU/j0zcdeVooxr2nGrYO0yso3ZRMtCenrC6wiIiJtMuSgbSaLSIyLFBuyiZSGt1iIiIwYwOYQs9oiQiIuYyOoT3rB2trrCIiJjH6BDWLkoiImIys0NYa0eLiIjBjA7hjr7wEX7UWUREpFcYHcJaOlpERExmdgi3f1RHWERETGR0CKN7wiIiYjCjQ9jSah0iImIwo0NYPWERETGZ0SGs54RFRMRkx0QIK4NFRMRERodwxzNKek5YRERM1GUIp1Ip7rzzTi6//HKuuuoqtm3bttf5N954g9mzZzNnzhwWLFhw2Bq6P3pMWERETNZlCC9atAiAZ599lltuuYV58+ZlziUSCebNm8dvfvMbnn76aZ577jlqa2sPX2s/w9LkaBERMViXIXzeeefx4IMPAlBVVUVxcXHm3MaNG6moqCAvLw+Px8P48eNZunTp4WttJ5TBIiJiIle3XuRycfvtt/Paa6/x8MMPZ46HQiECgUDm85ycHEKh0AGvVVDgw+VyHmJz91YTjAPg83koKQl08WrpimrYc6phz6mGvUN17LkjUcNuhTDA/Pnz+fa3v81ll13GSy+9hM/nw+/3Ew6HM68Jh8N7hfL+NDZGDr21n9HUFGl/3xi1tcFeu+7nUUlJQDXsIdWw51TD3qE69lxv17CzQO9yOPqFF17gF7/4BQBerxfLsnA623qylZWVbN26laamJuLxOEuXLmXcuHG91uiuaAMHERExWZc94alTp3LnnXdy1VVXkUwmueuuu3j11VeJRCLMmTOHO+64g+uuuw7btpk9ezZlZWVHot3AnmUrNTFLRERM1GUI+3w+fvazn3V6fsqUKUyZMqVXG9VtmWUrlcIiImIeoxfr0LKVIiJiMqNDWBs4iIiIyYwOYUspLCIiBjM7hHVPWEREDGZ0CHfQ7GgRETGR0SGs54RFRMRkZoewnhMWERGDGR3CHXRPWERETGR0CGeGo5XBIiJiIMNDWMPRIiJiLrNDuP2jhqNFRMRERoew1uoQERGTGR3CWjtaRERMZnQId8zMUgaLiIiJjA7hPT1hxbCIiJjH7BDWPWERETGY0SHcQR1hERExkdEhbGm1DhERMZjZIdz+UT1hERExkdEhrOeERUTEZEaHsJ4TFhERkxkdwnueE1YKi4iIeYwOYfWERUTEZMdECCuDRUTEREaHcGZillJYREQMZHQIW+oLi4iIwcwOYT2iJCIiBjM6hDOUwiIiYiCjQ9jSVoYiImIwo0O4g62ZWSIiYiCjQzizf4OIiIiBzA7h9o/qCIuIiImMDmF0T1hERAxmdAjvWbZSMSwiIuYxOoS1laGIiJjMdaCTiUSCu+66i507dxKPx7nxxhs599xzM+efeuopFi5cSGFhIQAPPPAAw4YNO7wt/hRt4CAiIiY7YAi/+OKL5Ofn88Mf/pDGxkYuvvjivUJ45cqVzJ8/nzFjxhz2hu6PnhMWERGTHTCEL7jgAqZNm5b53Ol07nV+5cqVPPnkk9TW1jJp0iSuv/76w9PKLug5YRERMdEBQzgnJweAUCjELbfcwje/+c29zk+fPp0rr7wSv9/PzTffzKJFi5g8efIB37CgwIfL5Tzga7rLG4kD4PG4KCkJ9Mo1P89Uw55TDXtONewdqmPPHYkaHjCEAaqrq7npppu48sormTlzZua4bdt89atfJRBoa+TEiRNZtWpVlyHc2BjpYZP3iEQTAMRiSWprg7123c+jkpKAathDqmHPqYa9Q3Xsud6uYWeBfsDZ0XV1dVx77bV85zvf4dJLL93rXCgUYsaMGYTDYWzbZsmSJX1wb1hLZomIiLkO2BN+4oknaGlp4fHHH+fxxx8H4Ctf+Qqtra3MmTOH2267jWuuuQaPx8OECROYOHHiEWl0h8xWhronLCIiBjpgCN9zzz3cc889nZ6fNWsWs2bN6u02HTRFsIiImMjoxTq0gYOIiJjM8BBuS+G0hqNFRMRAZodwxx+UwSIiYiCzQ1hrR4uIiMGMDuE9OzgohkVExDxGh7B6wiIiYjKjQ7iDOsIiImIio0NYjyiJiIjJzA7h9nvCWjFLRERMZHQIa+loERExmdEh3JHB6giLiIiJzA7h9pvCymARETGR0SGcoa6wiIgYyPgQtiz1hEVExEzmhzAKYRERMZPxIayusIiImMr4EG7rCSuFRUTEPOaHsMajRUTEUMaHMFjKYBERMZLxIWxZekJJRETMZH4IAxqPFhERExkfwliWesIiImIk40NYTyiJiIipzA9hUAqLiIiRzA9hS88Ji4iImYwPYdB4tIiImMn4ENY9YRERMZX5IYyeExYRETMZH8Jat1JERExlfAgrgkVExFTmh7BSWEREDGV8CGsDBxERMZXxIdy2gYNiWEREzHNMhLCIiIiJzA9htIGDiIiYyXWgk4lEgrvuuoudO3cSj8e58cYbOffcczPn33jjDR577DFcLhezZ8/msssuO+wN3ocW6xAREUMdMIRffPFF8vPz+eEPf0hjYyMXX3xxJoQTiQTz5s1j4cKFeL1errjiCiZPnkxJSckRaXiHtg0cFMMiImKeAw5HX3DBBdx6662Zz51OZ+bPGzdupKKigry8PDweD+PHj2fp0qWHr6WdsLSfsIiIGOqAPeGcnBwAQqEQt9xyC9/85jcz50KhEIFAYK/XhkKhLt+woMCHy+Xs8nXdZVngcFqUlAS6frEckGrYc6phz6mGvUN17LkjUcMDhjBAdXU1N910E1deeSUzZ87MHPf7/YTD4czn4XB4r1DuTGNj5BCbun8WkEymqa0N9up1P29KSgKqYQ+phj2nGvYO1bHneruGnQX6AYej6+rquPbaa/nOd77DpZdeute5yspKtm7dSlNTE/F4nKVLlzJu3Lhea3C36RklEREx1AF7wk888QQtLS08/vjjPP744wB85StfobW1lTlz5nDHHXdw3XXXYds2s2fPpqys7Ig0+tMsC9LpI/62IiIiPXbAEL7nnnu45557Oj0/ZcoUpkyZ0uuNOhht/WDNzBIREfMYv1gHltaOFhERMxkfwhZ6TFhERMxkfghrXpaIiBjK/BDG0i5KIiJiJONDWGtHi4iIqYwP4ba1o/u6FSIiIgfP/BBWT1hERAxlfAi3jUcrhkVExDzGh7B6wiIiYqpjI4SVwiIiYiDjQ7hj4UoRERHTGB/CbT1hdYVFRMQ85odwXzdARETkEJkfwpale8IiImIk40MYNDtaRETMZHwIW1oyS0REDGV+CKPhaBERMZPxIawNHERExFTGh7BloRQWEREjmR/CgK0UFhERAxkfwlo8WkRETGV8CGs0WkRETGV+CGsDBxERMZT5Iay+sIiIGMr4EEY9YRERMZTxIawNHERExFTmh7A2cBAREUMZH8Kg54RFRMRMxoewVswSERFTHQMhbCmDRUTESMdACLd9tHVjWEREDGN+CLfPj1YEi4iIacwP4Y5nlJTCIiJiGONDuINmSIuIiGmMD+E994T7th0iIiIHq1sh/NFHH3H11Vfvc/ypp55i+vTpXH311Vx99dVs2rSp1xvYFUtrZomIiKFcXb3gl7/8JS+++CJer3efcytXrmT+/PmMGTPmsDSuW9QTFhERQ3XZE66oqOCRRx7Z77mVK1fy5JNPcsUVV/CLX/yi1xvXHXv6wUphERExS5c94WnTprFjx479nps+fTpXXnklfr+fm2++mUWLFjF58uQDXq+gwIfL5Ty01u6H1X5TuKg4QJa79677eVRSEujrJhhPNew51bB3qI49dyRq2GUId8a2bb761a8SCLQ1cuLEiaxatarLEG5sjBzqW+5fe1e4tjaoEO6BkpIAtbXBvm6G0VTDnlMNe4fq2HO9XcPOAv2QZ0eHQiFmzJhBOBzGtm2WLFnSJ/eGM8PRGo0WERHDHHRP+E9/+hORSIQ5c+Zw2223cc011+DxeJgwYQITJ048HG08oI7haD0nLCIipulWCJeXl7NgwQIAZs6cmTk+a9YsZs2adVgadrA0O1pERExzzCzWISIiYhrzQ7hjAwf1hEVExDDmh7BmZomIiKGMD+EOimARETGN8SGsDRxERMRUx0AIa2aWiIiYyfgQ7mCrKywiIoYxPoQzw9F92wwREZGDZn4IoxQWEREzGR/CymARETGV8SG85zFhxbCIiJjF/BDObOAgIiJiFvNDuP2jOsIiImIa40MYPSYsIiKGMj6E92zgoK6wiIiYxfwQVk9YREQMZXwIo7WjRUTEUMaHcGZiluZHi4iIYcwPYa1bKSIihjI+hDsog0VExDTGh7A6wiIiYirzQ1iPKImIiKHMD+E9M7NERESMcgyEsNaOFhERMx0DIdz2UcPRIiJiGuNDWERExFTGh3BmOFodYRERMYz5Idz+URksIiKmMT6EtaGwiIiYyvwQbg9fRbCIiJjG6BDe2LSFt1K/wcppUgqLiIhxjA7hmtY6UiRweEOkNRwtIiKGMTqEXZaz7Q9WmkQq3beNEREROUhGh7DT0RHCNrF4qm8bIyIicpDMDmGrvfmWTSyhEBYREbN0K4Q/+ugjrr766n2Ov/HGG8yePZs5c+awYMGCXm9cV5ztw9GWlVZPWEREjOPq6gW//OUvefHFF/F6vXsdTyQSzJs3j4ULF+L1erniiiuYPHkyJSUlh62xn+W09gxHR9UTFhERw3TZE66oqOCRRx7Z5/jGjRupqKggLy8Pj8fD+PHjWbp06WFpZGecjvbmO9QTFhER83TZE542bRo7duzY53goFCIQCGQ+z8nJIRQKdfmGBQU+XC7nQTZz/+ppf3/LxuVxUVISOPAXyAGpfj2nGvacatg7VMeeOxI17DKEO+P3+wmHw5nPw+HwXqHcmcbGyKG+5T6CLbG2P1hpGhoj1NYGe+3anzclJQHVr4dUw55TDXuH6thzvV3DzgL9kGdHV1ZWsnXrVpqamojH4yxdupRx48YdcgMPxZ6JWbonLCIi5jnonvCf/vQnIpEIc+bM4Y477uC6667Dtm1mz55NWVnZ4Whjpz49MSuue8IiImKYboVweXl55hGkmTNnZo5PmTKFKVOmHJ6WdcOe54TT6gmLiIhxzF6sQytmiYiIwcwO4Y57wg71hEVExDxGh7CjPYQdDnRPWEREjGN0CHcs1uFwana0iIiYx+gQ7tjK0OnUPWERETGP0SG8554w2kVJRESMY3YIOzruCbf1hG3b7uMWiYiIdJ/RIeywHFhYWA4bG4gn033dJBERkW4zOoShrTdsWW09YA1Ji4iISY6JEMbRHsKanCUiIgYxPoRdlgOstmFohbCIiJjE/BB2uDIhrGeFRUTEJMaHcNsMad0TFhER8xwbIazhaBERMZDxIeyynNgohEVExDzGh7DT8akQ1nC0iIgY5JgI4XR7CEfVExYREYMYH8Iua08IqycsIiImMT+EHU7Sdlv4KoRFRMQkxodw2z1hG9B2hiIiYpZjIoQBsGzdExYREaMYH8KuTAinNRwtIiJGMT6EndaenrBCWERETGJ+CLf3hB0Om5WbG1jwxgaaQrE+bpWIiEjXjpkQLsrzAPCX97bx4t8392WTREREusX4EHa1D0f/6yUncNX5IwFYsnq3ZkqLiMhRz/wQbu8Je7MdnDu+nIvOHEJrLMXDz69g9ZYGbNvu4xaKiIjsn6uvG9BTHcPRKbtt1azpEwazemtj5r/TR5dy4RcGE/B5yPd7sCyrL5srIiKSYXwIu51uABKpRNvnLiffuWIc67Y38Ye3NvHe6hreW10DQEEgi8qBeQzrn0tZoReAZMom4HXjcFi0xpKUFnixLIuAz403y0W4NUEyZZPn92ABadvG6TB+AEFERI4CxodwwJMDQDgRyRxzOR0cP6SQyoF5/PWDHTQFYzQEY2zc2czSNTUsXVPTrWu7XQ4SybYedm6OB6fDIplK88Ux/fC4nOxujLC7oZXjhxQwpH8ug/sFsNM2W3cH2V4TIifbzemjSykIZLGzLoxlWQwo8rFhZzO2DbsbI4wZWkRBIKv3CyMiIkc980M4qyOEw/ucy3I7ufALgzOf27ZNXXOUzdUt1Da14nBYOC2L5kicllAch8MikUrjdjqoro9Q09TKCUNycTos1m5voiUcB+CV97bv9T5bdwc7bd+CRRvI8jgzE8UsCz57m/qEIQWccXw/Ksr8RKJJivOyKcrL1tC5iMgxzvgQ9nv8AISSkS5eCZZlUZLvpSTfe9Dvk063LQaSSKWpbWwlmUrjcjlIpWyq68PUt0RpDMZIp6GizM+gUj81Ta18sLaW5lCMknwvsUSK1VsaOamyiK27gzSF2kJ95ZZGVm5p3Ov9sjxO+hf6KC3wMqjUTyKZJp5Mk5/joTA3G6fD4qQRxTj2E9S23baE587aMK3xJCcOKzro71dERA4/40M4t6MnHN+3J9ybHA4Lb5YLL5Dr8+x1buSg/P1+zfHApJMH7nUsbds4LIt02sayYEdtGI/bwTuf7CLUmsCb5aK2qZWdtWF21IbZsiuYuaf9WR63g36FPtJpcDggkUyTStnEkima2wMe4ILTK6iqD1NW4GNo/wBDB+SSl+Ohqi5CeUkOHrezJ6UREZFDZHwId/SEw93oCR8NOnquDkfbx0Glbe2fdfawfV6btm2qasOs29FEUW42LqeDhpYo4WiSVVsb2LCjmW27Q1hW233wLLcTh8Pa57Gsv7y3rf1P9fu8h9vloKLUT79iP163g7rmKNtqggztn8upo0rJ8jg5fnABTqfV5YS0ddub2NUQ4Yzjy8hSsIuIdMn4EM7Nag/hhBkhfDAclkV5qZ/y9qD+tAvOqACgpqmVgLdtJventcaSNAZjVNeHSaTSjK4ooLYpyqbqFjZXt7CrIUJZgZequjCbq4NsrGrJfG22x8kHa2v5YG3tXm0pzM2iOC+bplCcWCJFtseJ2+kglbYpystmxca2kH/t/e18cUw/ivKyKS/x43BYVNeFSaZtxg4rwuN26H63iAjdCOF0Os3999/P2rVr8Xg8zJ07l8GD90x2euqpp1i4cCGFhYUAPPDAAwwbtm+v7nDxtw9Hhw7zcPTRqrST+9veLBfeLBcDinMyx/L8WQwvz9vntem0jdvrYcOWehyWRUWZn1VbG6muC1NVF6aqPkIqnaauKcqabU04LAuP20FTKJaZZLazLkxpvpf6lig768L87s2NnbbZ5bQYVBrA43IQyPGQ63PTHI7jcjqwbZsVG+sZPbiAC86ooCWcAGxK8r0U53lZ/FEVfq+bM0/sRzyZJhpPkZfj6fS9RESOZl2G8Ouvv048Hue5555j+fLlPPTQQ/z85z/PnF+5ciXz589nzJgxh7WhnfE43XgcbsLJCKl0as/+wsewz36fiVQi87x0x1C0ZVmsblhHVWgXp/c7hV3hGpbXfszMYRcQS8UJeHJwWG3Dyw6HRWFuNkP755JKp6gK72JYeQEnDCnc573jiRRp2ybb4yKRTFNdH8bvddMYilFRGqCmqZV/rKhm6IBctu0OEm5NkEilwYZsj4uq+jC1Ta1srm7Z59qftmx9HcvW13V6fuGbG0jbEEukmHZ6BRWlfjZWNbO5OkhjMIrL6eArk4dz8vDig66viMiRYtldrOs4b948xo4dy/Tp0wE4++yzeeuttzLnv/SlLzFixAhqa2uZNGkS119//QHfsLa288d5DkVJSYDrX7iTSDKCy3JxcumJXDHqkqNiuLMh2khtpJ6RBZWZ9qTtNLZts2D9H2lNtDJtyBRak1He3PEPBub0Z3zZWFY3rOe4whGsrF9D2k4TS8bIzcrFgUVDrInXt/2NouwCxpWOpTZSx7LajzlzwBl4Xdks3vE2Zb5SSn0lvL/7QwCyndlEU1EA8jy5tMSD2NgMzx/K+NKTiKZiDC7pTyJi84eNL7MrvJuKwEDO6Hcqy2s/5viiUQzyD6TIW8iOUBX5Wbmk7bav/7TtwSoW7/gH5w2eRJmvZK9z8VScP216hcq8IYwtGUMsnsKyoDWWor4lSnFeNrFEimQyTf+iHJatr2Xllkbyczx43E5qm1qpaWrFstqGxqvqwtQ1R/db9yy3M7OtZf8iHxVlAfJyPJw8vBhvlovmcJyS/GwSyTRvLq9ixMA88gNZjKrI3+9s8+4qKQn0+s/3541q2DtUx57r7RqWlAT2e7zLEL777ruZOnUqEydOBGDSpEm8/vrruFxtnehHH32UK6+8Er/fz80338wVV1zB5MmTO71eMpnC5erd3urtr3yfzU17nt297Yv/woRB43t0Tdu22d5cRdq2GVJQTjqdZntLFTtbdjG232gi8Vbe3PIOLbEQ0WSM4YVDCMZCrNi1GhvIyw6wtGoFtm0zfeS5HFdSya8+eJZQPEy2K4twvO0etsfpJt6+2pdpSnOKyM/Ow+Vw0hILsTO4C9u2KfIVcEr/Mexo2cVxxZWMH3Ai//vxH1lZsw6AX836IV5XFs9+/CL9A2WcV3kWz3z0B1bWrOPmM76Kw3LwyobFTBxyBgNz+/HaxrcYWzaa8rz+2LaNbds42ieJbd8dZP32JhpaogwsyWH4kBwcjjTbdyb45R8/YVd9hHg395keMSifb115Ci6ng/XbmxhTWURBIPuw1U9EpFs94ZNOOokLL7wQgHPOOYfFixcDbUEVCoUIBNoS/plnnqGpqYmbbrqp0+sdjp7wj//2K5bs+iBz7JTSsVw35p+6/Nq0neaTutXURxupidRSE6kjlorjdWVTF62nJtI2HOp35xBORLBpK1W2s+0v5o7eZWe8Li/JdJJEui1kLSzKfCXsjtQyNG8wowtH8NLm1wA4Z+AXcVgWy2pWkOPOodhbRGX+EAqzC3A7XJmJZ+FEhFNKx/JJ/RqqQtUcVziCwbmDeHHjX6gO7+IbJ36VdY0b8bm9eF1eEukEjy7/FYNzB3HzSdexpnEDRdkFvLfrQ0bkD6MlHsLlcBJzRgiFY4wqqKTMV8ovVvwXpb5iJvQ/jaZYM5uat9ASDxFORAgmglg42B3Z+9GpHLeP4wtH8f7uZQesi9vhJtuZRTARAmBc6ViW1azY72vzPLk0x1twWA6+XPklFu94m8ZYM6eUjsVhOaiN1DFr+HTKfCX4XF7+472fUNtax0nFJ3DmwDMYFqikORynpiHC0rW1eFwO/D43H22oI97e63Y5LaKxFMs31GEBHb8QWW4nxw8p4Lzx5QwqC/DKe9tIJNNMOWUgpQW+fdqq3kfPqYa9Q3XsuaOmJ/zKK6+waNEiHnroIZYvX86jjz7Kr371KwCCwSAzZszg5ZdfxufzceuttzJ79uxMr3l/DkcIL924ih8sfQSA/Kw84qk4D511X+a+aU2kjnAiQkVgIKsb1rG5eSubmrdSF22gIdq43+u6HW7GFI8mkUrwSf1q8jy5lPqKKfWVsGTXB6TtNONLT2JsyQmUeIupClXjc3upzBtCPJ1gU/NWKvOG4nI4+fOW12mOtTCx/EyG5w8lbaexsLAsi3eq3md1wzquOG42Xtfh6XWtql/LkNwKfO7OFyk52B+4tJ1mbcMGygMDcDtcNMeD+FxeAh4/7+36kN3hGr444HTe372cmkgtsVSMK46bzaJtb7GibhWNsWYKsvKoDu/GxibbmU15oD/NsRZqW+spyMonnAgTT+89SuCwHHgcbqKp2D5tcjlcJNPJvY7dcvI3GFU4fJ/X2rZNNBXF69pTk6VravjdmxtoDsU5eUQxy9bXZZYt/ayCQBbXTh/N4LIAfq/7kGoo+1INe4fq2HNHTQh3zI5et24dtm3z/e9/n1WrVhGJRJgzZw4vvPACTz/9NB6PhwkTJnDLLbccsCGHI4Rra4P836ZXyc/KZVtwJ/+oWsLk8rPwurLZEtzOqvq1+/1ar8vL2OLjGZ4/lNrWes4ccAYBj5+GaCOF2QVkOdtm3UaTMTxOd2YiUzQZw7KszPljQV/90oYTEVqTUfzuHLJdWdi2ze5IDaW+EqrDu2mKtTCqoJIVdat4b9cHTB86lYKsfN6ueo+C7Hx8bh/La1awK1LDpuatOC0nOW4fFhbBRIi0nSbg8TMkdxBDcivI9eRycskJPP7RU2xu2crJJSdy8fDppOwUZb4SbNsmlbZxOR3E4ik2VTXz6vvbaYkkOGVkMW6Xkz/+fTOtsbawz83xcPMlJzJ8YJ7+4usFqmHvUB177qgJ4d52uEK4Q02kjv/88HGC8VDm2LC8wZR4i9kVqaHcP4ATi0czsmD4MRWiPXUs/NKGEmE8DjduhxvLslhe+wm//Pi/u/31Z/Qbz+WjLqYh2pSZWFYd3k2/nNLMP8A6LHhjA++v2U19S1uPPM/v4Zxx5Rw3MJePNtYzadxA+hXuO2QtB3Ys/BweDVTHnlMId9P+CtUca2FD0yb8bj95WbmU+UqOitnSR7Nj9Zd2Zf1aalvrGJlfSXO8hY9qV/LWzncA+O4X/p03tr/FJ3WrSdspmuN7vn+fy0sk2QrAmQPOYM7IWdjYpOw0K2pXMrbkBLKcHlZubuCFv29id0MrodY9Q+dFuVlcMrGSRDJNwOvOPLLVvyiHJ/64koDPTV6Oh0g0SY7Xzayzh5Lv125ax+rP4ZGmOvacQrib9MPWOz4vdbRtm5c2v0aW08P5gydljjdGm/jVJ7+lOdZCQXY+DdFGmmLNe31tUXYh2a4sdoaqGZY3mBvHXpu5z55Ipln8yS62VTVT29TKmm1NB9Uuh2Ux9bRBnDm2PwPbF1iJRBMEWxOU7WcSWIdINMH7a2rYtjtEtsfJeacOMnprzM/Lz+Hhpjr2nEK4m/TD1jtUx73Zts36po1kObP471XPsetTM8E/PQHs5JIxXD7qEgIe/1413FkX5u2Pq4kn09Q2tWaW9MzxujiuIh+3y0lTaxDLlWDMwApeX7qDxmDb0PbQ/rnUNEYIR9vew+mwGD+qhFlnD8PhsPjr0h1UN4QJtybYXL33/7Ncn5tpZ1SwdVeQSDTJ6CEFDCrx43Y5KMn3UpibTTia4MN1tbhdDk6qLN5nydN4IsWby6vYXN3CCUMKGT+qBMsCj9vZo+eou0M/h71Ddew5hXA36Yetd6iOnUukk0QSEarCu3h/1zImln+RFbUreW/3MhqijQzLG8xNJ13Hkob3sWMOJpZ/ca/bH7ZtE2xNYDtj/GzZEwzI6cd1Y/6J774zn/poA/PP+i5r6rZSvTWLpavr2V4Toig3izx/FpuqDryyGMBZY/sz5ZSB/GXJtk533OowoDiHxmAsM7EMwJvlJN+fRb4/i6ZQjIZgLLP/9adleZycOLSQPH8WRbnZxBMpHA4Lt8tBtsfJScOLycl2k7ZtPC4HadsmFk+1bcHZzaH2zn4OW2NJLKtt1bX9SabS7KgNsbmqhTHDig5pu9JjiX6fe04h3E36YesdquPBS9tpfrPyf/Z5xjnH7aMwu4CpgyezO1zL1uA28rPy2dy8lR2hKgDOGnAGf69aAkBBVj6NsSZG5A+jMLuAWCrBVcddgtvpoaEpTnM4Tks4zt8+qsLvdeIq2k2OK4fjSypJptNsZxmWZXFBxfksW19HJJrA5bE5YUgxa7c0U9cSJZFMsaU6yJptDWTlhhl/XDG+VAk768I0Btv2wg5Hk2R5nPiz3Zx+fCknVOby1w92EIulcTnc7Kxt2ze7Mw7LwrLadv9yOiySqfYlVIGAz015qZ9RFQWEIgniyRQDinKIJ1MkU20LsDgdFnXBOHa6LbRzfW4+WFtLMp1mc1UznspPGOgZyuXjJ5Lvz2LllgYCXjertjTy5vKdpNJt71ea7+WfLzyO4jwvRXnZNLTv9V1e6ucPizdRGMhi8inlOJ3WYe/Z9xX9PvecQrib9MPWO1THQ9OajPKnTX9hU/NWBuaX0RBqoTnWss9CJh2G5lawLbiTlN29VbwCbj/9ckppjDaRm5XLjuDOfZ6d7jA4dxCxZIxgom1RlfysPAb4+1HXWk8wHmZkQSUN0Ua2B3cCex7R2xGqojJvKMlUigJvHiMLKnl/9zLernqPtN32nPTY4hM4a8AE/rDqr2SniimIjqS83KLFrqMh1kB2spANO4KkmkrJ9jhJO2LEWl00BeNkZ7U9r98citEWyZ1wJnCV7CBZUwHpvVfVs3KayD7h3baavzftU9exsXwt5BYkObH4eHZYy9i63ocdzseyYGBxDjtq97+5i9/r5pKJwzh5eHGmpx5qTfD//ryG0YMLGD24ABsYUOTbZ2JnOm3z7qpd/H1FNdO/OITRgwuOqkA/0O9zKt22TkHHdqqyfwrhblJ49A7Vsec+XcO/bHmDFbUrmTLoLEYUVNIcayFppxiaW8HuSA0ftveeP6lbQzQV5cuVX+Kj2pUUZhcQT8dZtP3vFGUXkLbT1EcbcVrOToN78qCz+Lh2FXXRBtwON0XeQnaFd+/1Gr87h1CiLYyG5FawpWXb/i619/fjLSI/K4+GaCP1nSxq81lDcwfTEG2kOd6Cx+khnU4xMDCAHJePtY0bONF/GnnZAWJ2hPpQmBPzTyJpRamP1/J+wz+I2233xYdkj6DIGkx5qZ8doR18WP9BZsW6ktrzsVrzGTsqh7cjfyRitbVtVMFw1jZuINvyUd48lXWbEiSTaUYPLsCb5WJzaBO5eWlKXAPZEltN7YYy0sm2hVYcloWNzf7+NszzeygMtA2/xxJta543h+PEE3sWcikt8HL66DJi8T1roQd8bk49rhSPy0m2x8mO2hDReIqCQBYetzOzFag3y4Vt2736BEdnv89p2+a+X7+Hx+Xg25efjC/b3WvveaxRCHeTwqN3qI49dyg1/PTqaZ8WS8XxtD/vHE1GcTlcVId3s75xI2cPnMBHdSsZkNOPEm8RbqebeCrOspqPGZ4/jCJvAfWtjfx02ROMLhzJpPIz6Z9Txu5IDV6Xj7ysAOsaN/D+rmW8U72UM/qN5/iikYTaF05JpOLkZuUyof9peJxukukk97/zAxpjTZxbcQ7rGzeyI1Sd6SV/ufJLtMSCLNrxdwBclpPBuRVUhatxWs5M+PeW4flDObH4eP68+a9EU1FclpPkfv6Bclz+KOLpOEk7zvD8Yby18x0Sn1lRrcAaQHbTKAgX0pL7MSlnmAGtE9iW/pisvBb6J8axubqZaHMOHpcLjzdJKuHCOXAdlr+B03MnU78rmw837cRVsZJk9VDSoX13H+tKwOdm3IhihvTPZXBZgPrmKCX5Xgb32/MXd9ruGLZ3HOBKbfb3s1jXWs+bG5bx8ksWYDFmaCHfvOwkHJZFLJ7C7XKod/wpCuFuUnj0DtWx5462Gqbt9D6LjHxWXfsSoV1tAdoUa2ZLy3ZOKj4h8w+GhmgjzbEgQ/MqAFjfuJGV9WuZ0P9UynJKgbZJbb9dvYCCrHzOHvgF3ql+H2h7Dntd0yY2NG3m9H6nMDhQzoq6VdRGa9kZ3MXZAydQ5ivB7XCRl5XLEyv+H9C28M6m5q2Zdo3IH8Yt477B3CU/piZSx3Vj/okPaz5iVf06oqkoFlamBw3gdWWTSCf3Wt7UYTnwubwH/MdCwOPH43DTEG2iX04p1e0jDf1zynBZTra33+sH8Fm5FDOERsd2UjE3Wek8guEU/uQAssuqSSRs7LSDVMJNuD6HcGsaZ249ie0jsbLDuIesJlk1jHRzCWOGFlIQyKIhGGNHTYgsj5MLTq8gHE3gcTk5eUQxG6uaycl243BYDO0XIJZIM2hgPsHmCLFEilTaJifbzbfffIDWdJjY2vGkgwW4y9dzQs5ppONZfLypnhOGFPLPF44mL8ejMEYh3G1H2198plIde041PHifHYYNFHhYu30bA/z99nrd+sZNuJ0uclw5vL7tTQqzCzih6Li2oHa62R2uIZgIZ7bXjKfiVIV3UewtoiUW5J3q9xkcKGd82ckAbGzewjtV73N80UieXv07EukEJ5eMoTnWwuaWbRRlF9IYayJtpynMLiAYD2U2YoG2TWIsLD6o+eiw1San9hSa4k04C3eTqu9POpwHaUfbR8uG1EEMJTuSeE99HQBn3XBOPb6IJTVLSDWVEF83Hqw0rv6bSdUNwEr48PvcbfuEB2OcVFnEoDI/RbnZDB+YR2Huvmvc27ZNMpXG3ckOebZt84e3NhOJJphySjmfbG7ghCEFvPTuVjbubCYaT3HKqBImHN+PkYPyD6VcvU4h3E36i693qI49pxr2XF/UMBQP0xxvYUBOP2xsVtWvpX9OP8LJMGsa1nNexcTMPuAbmjazM1zNlEFnE0m08oOlj9AUa+Zb428kmozx3No/MKXiHByWxXEFI4mlYmxp2Y7TcvDK1jcYmjeY4wtH8vKWv7IrvBuvy0tRdkFm1jzsvf/3gXjIIZlO4nDY5KYH4I0NJBxvxZOdIu4I0xSMkkxCOpSPI7cBV3HVPtfwO3MZm3saO5Pr2RrchivtwxEPkLKTxMJZWFkR0k2lpFoKscP5eFwOKstzyfVDKJoky+GlKRSjrqmVlkiCygG5DB2QS74/i2AkTiKZZlNVC9F4il0NEXDFcRbsJtVYBsm2ZYNdTgu7cCvuQeuIrfoCU8eO5iuThhNPpnh96Q6SqTQDS/wsW1/LyEH5nHPSgH0mwaXbY6w7k+PatkMl09sPRuLUNLYytH/uXiMACuFu0l98vUN17DnVsOdMq2E0GaUlHqS0fa3x7uq4n95xu6Ax2kR9tJGAx09xdiG//OS/2dC0mQuGnMvowpFsbNpCNBmlPtrA29XvZ77+YPhcPgqy89gZqj7orwWo4GR2hqtJ5bQNxdsJN4ltx2G546SjOfiLg7SmwmA7sOPZpOoGYuU04/C1YFkWA3wDaMh/lyQJ3NES/HYxSU8LnpxW6qMNADhbBhJaM4bsbEjEHaS9TViuOK6ybSR3DSHdUgRAUW42FWV+ABKpNBuqa4m2OvB7PZxUWcSQ/rnk+z30K/TRGkuR43WBlWbV5mb+/nE19c1Rrpk2ioDPzSPPf0wkliQn20XahhHleYytLOKyqcdRVxfafzEOgUJYDkh17DnVsOdUwzZpO00indzvJjOxVBy3w0VrMsrq+rUEPAHcTjc7Q1U4cJAmzcDiEpqaI2Q7s6iPNlId3sUX+p1Kqa+ExTvf5o8b/7zPdf3t+5iPKz2RP2x4CQuLW8ddz7vVS3lv94eZ4D/QTP3e5Ez6SLkiex1zJ/KINxRDqBC7/GPSjaUkdw8he8zbuFMB2HQGLcE02A5wxnHm15JqLsGZX4O7Yg2p5mISW04ARwqHN0S61Y/LF8ZXuZZ4HBK7BuMasIH0rkp+/a/XkYzu/3HAQ6EQlgNSHXtONew51bB3dFXHDU2b+WD3chbvfIdrT7iKQYGBBDw5mf211zVuINeTS7/2CXaN0SbWNW6kxFfE0NzBALyx/S0iyVbKfCW8sf0tmmMtnFY2jhEFw2iJBXlh48vkuH1cMWp2+1ajWzildCx1rQ38fsP/Man8TC4cej4N0SZe2PASX+h/Ki9sfJlYKk7KThFPxQ/pe7ewyHcVE0oGSXDgYX075cJyJvd77ldf/gGxXvxRVAjLAamOPaca9pxq2Du6U8dEOklNpJaB/v49fj/btknb6b1m2cdScSwsPM69J5Cl7TRbW7YzJLdin0fzEqkEWBbJdIJV9WupzB/KTz98gopAOQP8/Ti5ZAy/W/ci4WSELw05l/d2LWN57cdA2/Pyy2s+wef2sitcg8fppjUZxWE5OHvgBL7Qfzzz3384815ZTg+x9qCfVXkhzfEWFm3/Ox6HmzmjLmbGiZM0HN0d+qXtHapjz6mGPaca9o7PSx3Tdpr3dy3D78nhhKLjMsdT6RSWZWFhEUvFyHa1zeh+c8c/eGHDy3xr/I0M8g9kQ9MmQokIJ5WcQCqd4tVtb3JK6Vj655QdsYlZ+18NXURE5CjnsByc0X/8Psc/3SPvCGCASeVncvaAL2TOjyio3HMtp4PpQ88/jK3dv66XXhERETlGdLUwzZGmEBYREekjCmEREZE+ohAWERHpIwphERGRPqIQFhER6SMKYRERkT6iEBYREekjCmEREZE+ohAWERHpIwphERGRPqIQFhER6SNHfBclERERaaOesIiISB9RCIuIiPQRhbCIiEgfUQiLiIj0EYWwiIhIH1EIi4iI9BFXXzfgUKXTae6//37Wrl2Lx+Nh7ty5DB48uK+bddT76KOP+NGPfsTTTz/N1q1bueOOO7AsixEjRvDd734Xh8PBggULePbZZ3G5XNx4441Mnjy5r5t9VEgkEtx1113s3LmTeDzOjTfeyPDhw1XDg5BKpbjnnnvYvHkzTqeTefPmYdu2angI6uvrueSSS/jNb36Dy+VSDQ/BrFmzCAQCAJSXl3PDDTcc+TrahnrllVfs22+/3bZt2162bJl9ww039HGLjn5PPvmkPWPGDPsrX/mKbdu2ff3119vvvvuubdu2fe+999qvvvqqXVNTY8+YMcOOxWJ2S0tL5s9i2wsXLrTnzp1r27ZtNzQ02BMnTlQND9Jrr71m33HHHbZt2/a7775r33DDDarhIYjH4/a//uu/2lOnTrU3bNigGh6CaDRqf/nLX97rWF/U0djh6A8++ICzzz4bgJNPPplPPvmkj1t09KuoqOCRRx7JfL5y5UpOP/10AM455xzefvttVqxYwbhx4/B4PAQCASoqKlizZk1fNfmocsEFF3DrrbdmPnc6narhQTrvvPN48MEHAaiqqqK4uFg1PATz58/n8ssvp7S0FNDv8qFYs2YNra2tXHvttVxzzTUsX768T+pobAiHQiH8fn/mc6fTSTKZ7MMWHf2mTZuGy7XnDoRt21iWBUBOTg7BYJBQKJQZnuk4HgqFjnhbj0Y5OTn4/X5CoRC33HIL3/zmN1XDQ+Byubj99tt58MEHmTZtmmp4kH7/+99TWFiY6YSAfpcPRXZ2Ntdddx2//vWveeCBB/j2t7/dJ3U0NoT9fj/hcDjzeTqd3itgpGsOx57//eFwmNzc3H3qGg6H9/oB/Lyrrq7mmmuu4ctf/jIzZ85UDQ/R/PnzeeWVV7j33nuJxWKZ46ph155//nnefvttrr76alavXs3tt99OQ0ND5rxq2D1Dhw7loosuwrIshg4dSn5+PvX19ZnzR6qOxobwKaecwuLFiwFYvnw5I0eO7OMWmef4449nyZIlACxevJhTTz2VsWPH8sEHHxCLxQgGg2zcuFG1bVdXV8e1117Ld77zHS699FJANTxYL7zwAr/4xS8A8Hq9WJbFmDFjVMOD8Mwzz/Db3/6Wp59+mtGjRzN//nzOOecc1fAgLVy4kIceegiA3bt3EwqFOPPMM494HY3dwKFjdvS6deuwbZvvf//7VFZW9nWzjno7duzgW9/6FgsWLGDz5s3ce++9JBIJhg0bxty5c3E6nSxYsIDnnnsO27a5/vrrmTZtWl83+6gwd+5c/vznPzNs2LDMsbvvvpu5c+eqht0UiUS48847qaurI5lM8vWvf53Kykr9HB6iq6++mvvvvx+Hw6EaHqR4PM6dd95JVVUVlmXx7W9/m4KCgiNeR2NDWERExHTGDkeLiIiYTiEsIiLSRxTCIiIifUQhLCIi0kcUwiIiIn1EISwiItJHFMIiIiJ9RCEsIiLSR/4/w/Q1qfzCjfsAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training reg:0.01, run_num:0\n",
      "\n",
      "Training reg:0.01, run_num:1\n",
      "\n",
      "Training reg:0.01, run_num:2\n",
      "\n",
      "Training reg:0.01, run_num:3\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeEAAAFJCAYAAACsBZWNAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAA790lEQVR4nO3deZxU1Z3//9etfe3qvWmWbqDZFEQRlxijCK4JYDRqUBM10cmoow8TZ5KvSzTRkW/QSTJZHE1iJvE3X5OZyGAWjSYu0WjiFlFBWUT2rel9rX27vz+aLugATUM3NAffz8fDR3fVLW+d/tDwrnPuuedYtm3biIiIyGHnGO4GiIiIfFQphEVERIaJQlhERGSYKIRFRESGiUJYRERkmCiERUREhonrcL9hc3P3kJ6vpCRAe3t8SM/5UaQ6Dp5qOHiq4dBQHQdvqGtYURHe6/PG94RdLudwN+GooDoOnmo4eKrh0FAdB+9w1dD4EBYRETGVQlhERGSYKIRFRESGiUJYRERkmCiERUREholCWEREZJgohEVERIbJYV+sQ0RE5EC9885SbrnlBu6991ucffZ5heevueZyJk2awte/fg+rVq3gpz/9EbZtY9s2H/vY6Vxxxed5552lfOMbdzB27LjC/1dcXMLChQ8Mx4/Sh0JYRESMUFs7lhdeeLYQwuvXryORSBSOf+97/8Zdd/0rtbVjyWaz3HDDtcyceRIAM2eexL33LhqWdvdHISwi8hGw+MV1vPVB05Ce8+QplXx2zoR9Ho/Fotx//0Ki0W46OzuYP/9iZs8+h5tu+gd+8Yv/xbIsvvvdBzjppFOoqKjg3//93wgEApSUlODxePn61+/pc74JEyaydesWuru7CYfDPPvsM5x33idpbGwAoKqqmieeeJxPfepCJk6cxI9+9DPcbjfvvLN0SH/uoWT0NeFUOseLS7eSzuSGuykiIvJ3tm3bxjnnnMf3vvcQ//Zv3+fxx39JcXExdXUTWb78XdLpNO+++zann34G3/nOIu6885v88Ic/ZuTI0fs855lnzuaVV17Ctm1Wr17JtGnTC8fuuONuSkvL+O53FzF//nn8x398j3Q6DcDbby/l5pv/sfDff//3/zvkP/9AGN0TfndtM488tYobL5rGyVMqh7s5IiJHrM/OmdBvr/VQKCsrY/Hi/+bll18iEAiSzWYBmD//Iv7wh9/T2trKJz5xJi6Xi5aWFsaPrwPg+ONn8Kc/PbfXc5577gV897v3M3LkKI4/fkbh+VQqxZo1H/CFL/wDX/jCP9DZ2cGiRf/Kk0/+mvHjJxyxw9FG94QzuTyAesIiIkeg//mfx5g2bTrf+MZ9zJlzDrZtA3DSSaewdu0ann76SebN+zQAlZVVbNy4AYCVK9/f5zlHjRpNIpFgyZJfcd55nyw873A4uO++b7BhwzoAIpFiRoyoxu32HKofb0gY3RO2sADI7/yDFRGRI8fpp5/Jd76ziOee+wORSASn00k6ncbj8XDWWWezdOnfGD16DAD/8i+3sWjRv+L3B3C7XVRU7Ht08+yzz+XZZ5+hpqaW+vrtALjdbv71Xxfx7W9/i2w2h2VZHHPMscydeyHvvbesMBy9u+9+94d4vb5DV4ABsGz78CbYUO4n/Or7O/jZ06v54iencMbxI4fsvB9FFRXhId/r+aNGNRw81XBomFjHJ55YzJw551JSUsIjjzyM2+3mi1/80rC1Z6hruK/9hM3uCfd0hFE/WETEbKWlpfzzP9+E3x8gFArtMTP6aGV2CO8cjj7MnXkRERlis2efw+zZ5wx3Mw47oydmoZ6wiIgYzOgQ7h2OVgqLiIiJzA5hDUeLiIjBzA5hDUeLiIjBzJ6YZfX2hIe5ISIickjddNOXuPbaf2TmzJMLz33/+9+hrm4C8+dftMfrL710Pr/85RK8Xm/huZtv/kfa29v45S+XFJ57+eUX+frX/w//+79PUlU1goce+j7r16/D5/OQz1t8+cv/wqhRo7n55n8klUr2ua/4yiuv5uMf/8Sgfi6zQ3jnVw1Hi4gc3S688GL++MenCyGcyWR49dW/cP31Nx3wudauXcPEiZMBeOGF5xgxohqAN998jZaWZr7//YepqAjzxBNP8eCD/8799/87QGGHpqFkdAj3UgSLiPTv1+t+z7tN+14O8mDMqDyOz0yYt8/jQ7mL0llnnc0jjzxMMpnE5/Pxl7+8zCmnnEp3dxff/OadpNMpuro6+cIXvsSZZ561zzadc875PP/8s0ycOJnu7m7S6RSlpWUAVFaO4IMPVvOnPz3HeefN5owzZnHaaacPVbn2yvBrwrooLCJypBrKXZS8Xi9nnDGLV155CYBnnnmSCy/8DJs3b+Lyyz/H97//MLfe+n/49a8X99um008/gzfeeBXbtvnzn//EWWedXThWVzeB2267i1de+TPz5s3juuuuYsWK9wrHFy78Rp+dmNrb2wddI6N7woUM1nC0iEi/PjNhXr+91kNhqHdRmj//Yh566AeceOJJdHd3M3nyFDZsWM9//dfPePrp3wFW4T32xev1MXHiZFaseI9XXnmJe+9dxG9+03ONeN26tdTU1HLvvd+ivDzEM8+8wDe+cQdPPvkscGiGo83uCe/8qggWETnyDPUuSnV1E0gkYixe/D/MnXshAP/5nz/mggvmcvfd93HiiScNqF3nnnsBjz/+S4qKiggEAoXnly59kx//+D/I5Xo2gBg3bjw+n3/XqOshYHhPWLOjRUSOVIdiF6W5cy/koYd+yBNP/B6A2bPP5gc/+A6PPfYolZVVdHR07LddJ598Kv/3/97DnXd+o8/zl156OQ899AOuvfZzRCJF5HI2d999b+H4woXf6DM7+uyzz+Piiy890LL0YfQuSsvWtfDDJe9x2ew6Pnlq7ZCd96PIxF1XjjSq4eCphkPDxDpqFyUDFQYI1BMWETGadlEykCZHi4gcHbSLkoF2XRNWDIuIiHnMDuGdX5XBIiJiIqNDWPsJi4iIyYwO4V0rZimGRUTEPGaH8M6vymARETHR0RHCw9oKERGRg2N2CGt2tIiIGMzwEO75qgwWERET7TeEc7kcd9xxB5dffjmf+9zn2LJlS5/jL774IpdccgkLFixg8eL+t5A6VJTBIiJiov2G8Esv9ezd+Ktf/YpbbrmFRYsWFY5lMhkWLVrEz3/+cx577DEef/xxmpubD11r/86unS0UwyIiYp79hvA555zDfffdB0B9fT3l5eWFY+vXr6empoZIJILH42HmzJksXbr00LX272g4WkRETDagtaNdLhe33XYbzz//PD/84Q8Lz0ejUcLhXTtDBINBotFov+cqKQngcjkPsrl9tUQzAPj9nn3uUCEDpxoOnmo4eKrh0FAdB+9w1HDAGzg88MADfPWrX+Wzn/0sTz/9NIFAz04XsVis8JpYLNYnlPemvT1+8K39O52dPeeKxVPGbdt1pDFx67MjjWo4eKrh0FAdB+9wbWW43+Ho3/72t/zkJz8BwO/3Y1kWTmdPT7auro7NmzfT0dFBOp1m6dKlzJgxY8gavV8ajhYREYPttyd83nnncccdd/C5z32ObDbLnXfeyXPPPUc8HmfBggXcfvvtXHfdddi2zSWXXEJVVdXhaDcAlhaPFhERg+03hAOBAD/4wQ/2eXzOnDnMmTNnSBs1ULv2E1YKi4iIebRYh4iIyDAxO4TpXbZymBsiIiJyEMwOYQ1Hi4iIwYwO4V7qCYuIiImMDmHL0uxoERExl+Eh3PNVw9EiImIis0N451cNR4uIiImMDuHerrAyWERETGR0CDvUFRYREYMZHcK98spgERExkNEhXJgdrQFpERExkOEh3PNVo9EiImIis0N451eFsIiImMjoEN41O1opLCIi5jE6hAuNVwaLiIiBjA7h3vFozY4WERETGR3CFpodLSIi5jI7hLV/g4iIGMzwEN45MUspLCIiBjI6hHvZSmERETGQ0SFcWDtaRETEQEaHcO9FYc2OFhERExkdwoWOsIajRUTEQGaHsGZHi4iIwQwPYc2OFhERcxkdwr00O1pERExkdAhrdrSIiJjM6BDunZqljrCIiJjI6BAuTMxSCouIiIGOjhAe3maIiIgcFLNDWMPRIiJiMKNDmEJPWCksIiLmMTqEHdpOWEREDGZ0CO+aHa0UFhER8xgdwpqYJSIiJjs6QlgpLCIiBjI7hDUcLSIiBnP1dzCTyXDnnXeyfft20uk0N954I2effXbh+KOPPsqSJUsoLS0F4N5772X8+PGHtsW707KVIiJisH5D+Mknn6S4uJhvf/vbtLe3c/HFF/cJ4ZUrV/LAAw8wbdq0Q97QvSlMjlZHWEREDNRvCF9wwQWcf/75hcdOp7PP8ZUrV/LII4/Q3NzMWWedxfXXX39oWrkPu7YyVAqLiIh5+g3hYDAIQDQa5ZZbbuErX/lKn+Nz587lyiuvJBQKcfPNN/PSSy8xe/bsft+wpCSAy+Xs9zUHwrLA5XZSUREesnN+VKmGg6caDp5qODRUx8E7HDXsN4QBduzYwU033cSVV17J/PnzC8/bts0111xDONzTyFmzZrFq1ar9hnB7e3yQTe7LAtKZHM3N3UN63o+aioqwajhIquHgqYZDQ3UcvKGu4b4Cvd/Z0S0tLVx77bV87Wtf49JLL+1zLBqNMm/ePGKxGLZt8+abbw7LtWHLsnSjsIiIGKnfnvCPf/xjurq6ePjhh3n44YcBuOyyy0gkEixYsIBbb72Vq6++Go/Hw2mnncasWbMOS6N3Z1m6JiwiImbqN4Tvuusu7rrrrn0ev+iii7jooouGuk0HyFJHWEREjGT0Yh3Qs4mDOsIiImIi40MYy9JwtIiIGMn4ENa8LBERMZXxIeywUAqLiIiRjA9h0HC0iIiYyfgQ1nC0iIiY6igIYUuzo0VExEjmhzBgqy8sIiIGMj+ENTFLREQMdRSEsFbMEhERMx0FIay1o0VExEzmhzCamCUiImYyP4R1i5KIiBjq6AhhdYVFRMRAxocwqCssIiJmMj6EHZbuExYRETMZH8JoxSwRETGU8SHcc014uFshIiJy4I6CENaSWSIiYibzQxjIK4NFRMRA5oewNdwtEBEROThHQQhbuk9YRESMZH4Io4lZIiJiJvNDWGt1iIiIoY6CENY9SiIiYqajIIQ1O1pERMxkfAj3XBUWERExj/Eh7NAuSiIiYijjQ9jS2tEiImIo40MYNDtaRETMZHwIO7R2tIiIGMr4EEazo0VExFDGh7A6wiIiYqqjIIQtbKWwiIgYyPwQRgtmiYiImcwPYa1aKSIihnL1dzCTyXDnnXeyfft20uk0N954I2effXbh+IsvvshDDz2Ey+Xikksu4bOf/ewhb/AeLNBFYRERMVG/Ifzkk09SXFzMt7/9bdrb27n44osLIZzJZFi0aBFLlizB7/dzxRVXMHv2bCoqKg5LwwE2dm5me8VvoPXEw/aeIiIiQ6Xf4egLLriAL3/5y4XHTqez8P369eupqakhEong8XiYOXMmS5cuPXQt3YuGWBN5Rwp80cP6viIiIkOh355wMBgEIBqNcsstt/CVr3ylcCwajRIOh/u8NhrdfxiWlARwuZz7fd1AFEX9Pd9YNhUV4f5fLPulGg6eajh4quHQUB0H73DUsN8QBtixYwc33XQTV155JfPnzy88HwqFiMVihcexWKxPKO9Le3v8IJu6p2h3CuiZmNXc3D1k5/0oqqgIq4aDpBoOnmo4NFTHwRvqGu4r0Psdjm5paeHaa6/la1/7GpdeemmfY3V1dWzevJmOjg7S6TRLly5lxowZQ9bggbCsndsYWpqYJSIi5um3J/zjH/+Yrq4uHn74YR5++GEALrvsMhKJBAsWLOD222/nuuuuw7ZtLrnkEqqqqg5Lo3s5CnsJ29i2vSuURUREDNBvCN91113cdddd+zw+Z84c5syZM+SNGqhdPeGeIWllsIiImMToxTosq7f5tpauFBER45gdwvTtCYuIiJjE7BC2dr8mPKxNEREROWBmhzC7XwRWCouIiFmMDmHHzp6wZaknLCIi5jE6hC00HC0iIuYyO4R7Z0dbaHa0iIgYx+gQdqgnLCIiBjM6hHdfrENERMQ0Zofw3y1bKSIiYhKzQ3i3DRwUwSIiYhqzQ3i3cWh1hEVExDRmh7Cl4WgRETGX0SHs2H0XpeFtioiIyAEzOoR3n5ilFBYREdOYHcJ99hNWCouIiFnMDuHdb1Ea1paIiIgcOLNDWBs4iIiIwcwO4d2ar+FoERExjdEh7NhtsQ4RERHTGB3CWqxDRERMZnQIO7RYh4iIGMzoEC70hLV2tIiIGMjsELZ2G44exnaIiIgcDLNDuLf5loajRUTEPGaH8G49YXWFRUTENGaHsK4Ji4iIwYwOYcfu14Q1HC0iIoYxOoT77ic8rE0RERE5YGaHsIajRUTEYGaHsIajRUTEYGaHMLt2UVJXWERETGN2CO/WE86rJywiIoYxOoQd7JqYpRAWERHTGB3C1m5bGebzw9sWERGRA2V2CBdmR0M+r56wiIiYxegQdli9zddwtIiImGdAIbx8+XKuuuqqPZ5/9NFHmTt3LldddRVXXXUVGzZsGPIG9qfQEwZy6gmLiIhhXPt7wU9/+lOefPJJ/H7/HsdWrlzJAw88wLRp0w5J4/anzzVh9YRFRMQw++0J19TU8OCDD+712MqVK3nkkUe44oor+MlPfjLkjduf3XvCuiYsIiKm2W9P+Pzzz2fbtm17PTZ37lyuvPJKQqEQN998My+99BKzZ8/u93wlJQFcLufBtfbv5O2dU6Itm3DYR0VFeEjO+1Gl+g2eajh4quHQUB0H73DUcL8hvC+2bXPNNdcQDvc0ctasWaxatWq/IdzeHj/Yt9xrG3Z+R3t7nObm7iE790dNRUVY9Rsk1XDwVMOhoToO3lDXcF+BftCzo6PRKPPmzSMWi2HbNm+++eZhvzasFbNERMRkB9wTfuqpp4jH4yxYsIBbb72Vq6++Go/Hw2mnncasWbMORRv7ZWGBZWt2tIiIGGdAITx69GgWL14MwPz58wvPX3TRRVx00UWHpGED1Ts5Sz1hERExjdGLdUBPCFuWrdnRIiJiHONDmJ3XhbV2tIiImMb4EO69JqzhaBERMY35IVzoCSuERUTELOaHsGZHi4iIoY6OENYuSiIiYiDjQ7h3O0MNR4uIiGmMD2HL0sQsERExk/kh3DscrZ6wiIgYxvwQtiywNBwtIiLmMT+Ed/aENTtaRERMY3wIO3p7wspgERExjPEhbOFA14RFRMRE5oewpV2URETETMaHsMPSLkoiImIm40O4tyesiVkiImIa40PYoV2URETEUMaHsGVpYpaIiJjJ+BB2aLEOERExlPEh3HNNWD1hERExj/Eh7NAGDiIiYijjQ9jauZWhZkeLiIhpjA9hZ+/ELGWwiIgYxvgQ1sQsERExlfEhrIlZIiJiKuND2GE5du6ipBAWERGzHAUhrJ6wiIiY6SgIYQeWBdl8fribIiIickCOghDu2cDB1nC0iIgYxvwQdvTuoqSesIiImMX4ELZ2/gh5WyEsIiJmMT+Edw5Ha2KWiIiYxvgQ7r0mrFuURETENMaHsEXvNWGFsIiImMX8EC7MjtY1YRERMctREMI7d1FSCIuIiGGMD2EHvdeEFcIiImKWAYXw8uXLueqqq/Z4/sUXX+SSSy5hwYIFLF68eMgbNxAOq/cWJV0TFhERs7j294Kf/vSnPPnkk/j9/j7PZzIZFi1axJIlS/D7/VxxxRXMnj2bioqKQ9bYvSncoqSesIiIGGa/PeGamhoefPDBPZ5fv349NTU1RCIRPB4PM2fOZOnSpYekkf3pDeGcesIiImKY/faEzz//fLZt27bH89FolHA4XHgcDAaJRqP7fcOSkgAul/MAm7lvjnU9IWwBFRXh/l8s/VL9Bk81HDzVcGiojoN3OGq43xDel1AoRCwWKzyOxWJ9Qnlf2tvjB/uWe9XbE87kcjQ3dw/puT9KKirCqt8gqYaDpxoODdVx8Ia6hvsK9IOeHV1XV8fmzZvp6OggnU6zdOlSZsyYcdANPFiWdlESERFDHXBP+KmnniIej7NgwQJuv/12rrvuOmzb5pJLLqGqqupQtLFfDjQ7WkREzDSgEB49enThFqT58+cXnp8zZw5z5sw5NC0bIM2OFhERUxm/WIelDRxERMRQxofw7itmKYhFRMQkxodwb08YIJ3JDWNLREREDszRE8KWTSqtEBYREXMYH8K9a0eDTVI9YRERMYjxIexxenq+ceTVExYREaMYH8JepxsAy5EjqRAWERGDmB/Crp09YadCWEREzGJ+CDu9QE9POKVrwiIiYhDzQ7i3J+zIkUxnh7cxIiIiB8D4EN41MSuniVkiImIU40O4tydsOfIajhYREaOYH8LO3YejFcIiImIO80N4t9nRGo4WERGTmB/Czt7h6JxWzBIREaMYH8IelyZmiYiImYwPYd/us6PVExYREYMYH8Ie127D0eoJi4iIQcwP4Z09YYcrTyKlxTpERMQcruFuwGC5HE6clhPblacrlh7u5oiIiAyY8T1h6OkNO3eGcC6fH+7miIiIDMjREcION5Yzjw10xTLD3RwREZEBOTpC2OkGq2dSVnt3aphbIyIiMjBHSQh7yFs9k7I6ogphERExw9ERwg4PORTCIiJilqMihL1ODzZ5sPIKYRERMcZREcI+l6/nG2eWhtb48DZGRERkgI6OEHZ6ASgrcfL+hjYtXykiIkY4OkLY1RPCU+uKSGVy/OfvV/HGqgZs2x7mlomIiOyb8Stmwa6e8HETIrz7Xpq31zTz9ppmovEM55w0ZphbJyIisndHSU+455qw12fzjWtOZt7Hx+J2OVj80joa23SNWEREjkxHRQh7d/aEk7kUZREfnzlzPF+adyzZnM0vn/9Qw9IiInJEOipCuPeacDKbLDw3c3IFU8eVsmJjGz9/ZjXdcW3uICIiR5ajIoT9O4ejk7ld9whblsXnz5tEaZGXV99v4Ms//Cv3PvoWqze3q2csIiJHhKNiYlZhODrbd6GOqpIA919/Gi8s3cbrKxvY3NjNt//nXSaPKeYzs8bT1pWiKOjhmNqS4Wi2iIh8xB0VIVwYjs4l9zjmcjq44NQaLji1hvXbO/n9a5tYvr6VRb94p/CaS2aNZ+5pYw9Xc0VERIABhHA+n+eee+5hzZo1eDweFi5cSG1tbeH4o48+ypIlSygtLQXg3nvvZfz48YeuxXvRe4tSKtv/kpV1oyJ8+bLjWbmpjTdWNBBPZVm3vZMnXt5AZyzNeSePoTziPxxNFhER2X8Iv/DCC6TTaR5//HGWLVvG/fffz49+9KPC8ZUrV/LAAw8wbdq0Q9rQ/vTeopTIDWzd6KljS5k6tudDQ2NbnO/973JeWLqNF5ZuY+LoCOURHy6ng/mnj1Uoi4jIIbPfEH777bc544wzADjhhBNYsWJFn+MrV67kkUceobm5mbPOOovrr7/+0LS0Hz5nTwjvrye8N1WlAe699hTeWNnAm6saWbOlg7XbOgF458NmPj6tmvNOHkM44Mbjdg5pu0VE5KNtvyEcjUYJhUKFx06nk2w2i8vV87/OnTuXK6+8klAoxM0338xLL73E7NmzD12L98LtcOGwHHu9JjwQXreTWSeMYtYJo+iKp0lncry5qpGnXtvE80u38vzSrQDUVIY4rq4Mt9NBOpunvTvFJbPGU1rkI5rIEPS5sCxrKH80ERE5iu03hEOhELFYrPA4n88XAti2ba655hrC4TAAs2bNYtWqVf2GcElJAJdraHuUlZVFFHlDRLNRKirCgzpXxc6vx0yo5PNzp/Knt7by1qoGuuNpPtzSzpamaJ/Xv/VBEyPKAmxrijJxTDHHjCtl+YfNhAIePj69GrfLyXF1ZdS3xKipClNR7Gf1pjY8bieTao6sWdmDrZ2ohkNBNRwaquPgHY4a7jeETzzxRF566SU+9alPsWzZMiZNmlQ4Fo1GmTdvHs888wyBQIA333yTSy65pN/ztbcP7TKSFRVhmpu7qfRVsLZjA9saWvE6PUN2/hPrSjmxruf6cSKV3XmfMbhdDlZvbmPpB01s3xnMa7d2sHZrR+H/XbmhdY/zuZwOsrk8AJUlfkaUBnA6LE4+ppKtTVHGVITwup2MqQpRGvbhcPT0rPN5m3gqy7N/20IsmWVKTTHbmqNMrikpXN8ejN46ysFTDQdPNRwaquPgDXUN9xXo+w3hc889l1dffZXLL78c27b51re+xVNPPUU8HmfBggXceuutXH311Xg8Hk477TRmzZo1ZI0+ECOCVXzYsZ7GeBM14dGH5D38XhcnTqooPJ5eV8aCORNJZ3K4XQ46Y2nau1N43U5eW9FASdjL9pYYDa0xJteUsGZLO5saujl5ShWJVI41Wztoau8J6nfXtuzxfk6HRSTkoTueIZvNs/sSI39+dzsAv39tM0UBNyVhH+3dSTxuJ7ZtU10WJBxwUzcqQiTopbosQFnEhwW6ti0icoSw7MO8fNRQfzrr/bTyyrbXePzD33L1MQs4tXrmkL7HoZK3bbrjGTZs72RdfSdup4OtTVHKI346Yyma2hNsbujG6XRgWVAU8HDM2BKKQ17eW9fCSVMqefX9HbR2pcjm8hSHPHRE+1+e0+NyMHNyJUG/i431XYwsDzKiLEBNdYTaiiCtnUlcTovKkgBbm6LUjgjR1pUiEvQUwrv3V6a9O0Vx2ItjL9fBbdtmR2ucnz29ms/OrmNsdREAWxujjKoI4vf2/fyXz9ukMrk9nj8QsWSGbU09owPDQb2PwVMNh4bqOHhHTE/YFNXBKgB2xBrJ23ni2QRBV2BIJkqlc2nebXqf6RXH4ncN7JalTD6L29F/eR2WRSToYcakCmbs1sPeXT5vY1ns8XN85szx5O08zur1TCmZSLGzipKwl/buFH6vk3gySyKd4w9vbCaby5PJ5kllcjS1J3h9ZUPhPOvru/b6vpYFu388c7scBLwuEuksmUwe585h9XHVYY6pLcWyIJrIsHxdC5PGFLOpoZum9gQAD/z3u3hcPZPZAMojPipL/HTG0mSyeSaOjrC5IUpzZ4LPnzuJqtIANZUhNtR34XBYTBpTzMYdXbzzYTOzjh9JefHe/wz+44n3WbO1g9knjmLeaWMJB9y4nA6iiQxOh9Un4G3bZmtTlJHlQTLZ/AGFf1c8TcDrYvXmdiqKey4pZLJ5rU8uIgfsqOkJxzMJbvvrvficXkp8xWyP7mBkcASfrvskyVyKUaFqHFiU+ErY3LWVEl8xXqeHjlQnv1v/B06rPhmfy0tt0Rj8Th9v7FjKGw1LmVwyga3d21nR+gEhdxCv00sql2JsUQ02Nus7NuJ2uqkOVNGV7sbv8jGheDx/2f464yK1XDbp0/xx058o8RZzYd0FZPJZHFi0pzr4/1b+io+PPJmPjzyFvJ3n5W2vUeQJsbZjI3PHnUvYs2tW+vqOTWTzWdxOFyXeYrZ2b8flcPHQ8p8B8M8n/hOjQiN4aPnPmV5+LOfWnkXezvPzFb9kTHgU54+dA/T0vjfUd5HO5KgbFWFzQzedsTTdySx/W9lAZYmf7niarliaLY1RcnmbutFhtjXF8LldlIS9uF0O2rpSJFJZ4qnswP+wrDyuqs3k2kZgp/04HRa5/IH/+o2uCFJdFsTpsEiksjS2J8hk87R27Tk7ftr4UtbtvOWsblSEju4U2bxNW1eSzM4PBS6nxVkzRnFMbQluV89ohNOyGFkepDuRIZPNFz44/PW9HTz56qbC+Z0Oi5mTK1i3vZNoPMOsE0ZRWuRlS2M3YyrDlBZ5SWfynDipnLwNIb+78P/mbZt83sbl3HMJ985YmqUfNOFyWkyvKycS8vQZcbDtnpED6Pkwl8vbgxpFOFKoBzc0VMfBO1w94aMmhAEefPenfNC+FujpGTfEmrB3u5Lqcrjwu3x0p6N7PddAeBxubCCTz+z1uIXV5z33pmeFL6twS9XI4AjqYw17vG5sUQ1VgQqWN68c0O1XDstB3u4JlsklExgTHsULW14uHC/xFvOJUR8j4PKRzmd4p+k9ppdPZUblcXiC8Pya1zhj1Mf42YpfUOor4YKac1nXUs+rzS8R8Ua4/riraU918tSGZ7Esi2unfo73d2zgvabVTA+dit/toaTc5k/r32JK0bGcNL6WTS2tvLa8mU+eMo7ffPAcb3f/hdpwDf8w+R+IhNy8Xv82sZYwZb4yQkU53v6glVTCyZamnj/TLY1RnA4Ll9PB5JpibBtWbWrrE94Br4tkOofDYXH9hVNZsbGVl5fV96lNb+BbVs/3Dssq9MwPVmmRl7augd+bblkwsiyI02lhWRYNrfGe+QRuB0UBD4lUluKQF5fLwfbmKNlc39+j3gl70XiGxvZ4n5EKv9fF7BmjqKkKUR7xE02keXlZPSNKA1SXBXl3bTP5vM3oyhCnHlOFz+sklc6RyuQLlzrKIr59tj2dydEZS1Oxj1GIA7G3kYleCo+hoToOnkJ4gHYv1N8a3uG/Vv2K6eVTuX76Naxu+5BfrfkN2XyWvJ2nK92Nz+llSulE0rkMO2KN2Ng4LSetyTbC7hCZfJZkLonf5eeLU6/AaTnpSHVybNlkbBsCLh/tqU5+v+FZSn0l+F0+Th4xgz9u+hOnjJhJsTfCr9f9npHBEXSlu3ljx1I+Vn0Sr9a/iYXFhOJxNMSa6M5EGVdUQyKXoiHWuMfP5XK4yOYH1sucVjaFNe3ryAzw9YdKhb+M9lQn2XwWC4txkVo2dG7C43AzOjyKbdF60rl04bXNiZ5JaX6Xn+PLp/JW47vk7ByjQtUksymSuSTjwuMYHa6mNdnGSVXHs7RxGSWeEqp8IxkTHEPQ6+O57c9R5a9gbFEtAbePEm+EP324jGnVdbzbsAKnO8e5486gI5qkOOQDG+LZOMlMmpYWi8piPxt2dLF82wZKPGXUVhXTGU0RS2aJBD04nRYtHUmaOhJks3lGT+rkg+h73DTzGuIxB+l0Dp/XSV1tGa++s5Xmrm5i3m00boyQTlt4XA5eXdHzIcsCPB4nuZxNVakf/84wbOtKEQp4enrquTyjK0KcemwVsWSGlRvb2NSw6++Nx+VgTFWIgNdNfUuU1gP4INCfmqoQzR1JRpQGqCzx09yRoCjgYUptCS++s43m9gSfmF7N9Lpykuksje1xRleEWL6uhQmjIvg8LqLJDAGvi/c3tOJyOvC4nXx86ggiIQ/t3Sn+vGw7f1vVRMDn4oufnILT6aCxPc5bq5tIprPMPqmG046pwGFZmkA4CArhwVMID9DuhcrbeVa3rWVSSd1er8d2pbtxO1x7va4by8QJugPk8jmWNi5jYsl4Sn2Dn+CTzmXwON2s69hIsTdCub+UvJ0nncvgc3mxbZtsPktzopV0Ps2LW/7C3PHnURWoYHt0By2JNiaX1OFxenh9x1u8Xv8WG7u2cMHYs0nn0lQGyjlj1Glkchm2x3ZQ6S/ntR1vkc/n2RrdzpjQKJwOJ1NKJ9KcaGVT5xZCniDLmlYwp+YMWhKt1EcbyDrSbGzbRnc6yrFlk6kNj6Y+1oDP6cPhcNCaaCOdS+N1ednevYPO9N6vJQOcWDmdpngL26L1lHiL8bm87Ig14rAcRDxFtKc6Bl1X6BlRKPZGaIg39Xl+9xGBXkFXgHg2gWVZOLDI2j1DuVWBCk6umkFjvJm3Gt9lXFEN54+dw/qOTbQkWplcOhGXw8WJldPZHt3BS1v/wjtN7wEwrqiW8ZHawgemkqIQpY4KVrSs5rUdf8Pj9DB79Cc4rvxYnlr/LJNK6jij+hMEfT230HWkOvnR8kcp8oY5r+Ys1rSv4+PVp1LkKcLh6Pl9TWVTFHmL+Omy/6E6XM4Ftefgd3sLQ9ipdI7OWIqQ383mhm42NrbRHO0m4okwrrqIVCbH+u1dbGvppKUjRVtXmqDfzaTREcJBD16Xk7xts2ZLB5sbuwn53UQTex/lGUp/fynC8sawc27I9tQm5Hcz64SRFIe8bKjvwuPuGQlxORwk0zkmjonQHc8QTWQYOyKMBeRtcDot/B4nTqeDbU1RbBvS2Rwb6rsYP7IIy7KoiPgI+t17vQwAkMnmyGRtAr7BD+9nsjncQ7wuwkAohAdPITxAH7Vftmw+y/LmlZxQMQ2nY+j+cvfW0bbtAU1m6/1QEXD5SWQTxLNJSnwR2pOdjA5VA7AtWk+5vwy/y0dTvBmn5SLiDbOiZTWTSyfSnuwg6A6yvnMjK1pWc2HdBazr2Ijf5SeTzxBw+VjTto4VrR8APdc+Pz7yFF7Z9jpuh5uWRCuxbJwyXynTyqdg29CV7qI10UZrsp14NsG0smNI5zN82L6OMl8pEW8ROTtHiTdC3rZZ1bZmQCMObodryEYaji+fyjm1Z/GzFb+gI9XZ55jTclITHkV9rIFUbs+JXqeMOBELi+ZEC36Xn0p/OXNqzuBvDe+ysXMTq9vWYmNz4/QvEs3EeHL9H5lTcwZ/2f46edvm8kkXM7l0Aslsko1dW6nwl9KcaGVKyURsG5wOB7FkmkwWLGeOtVs7SCRhwpggbm+epiabt9Y0EgxnqSmpoKUzSdDvpqktQXHIg8/rIpXOMWF0hKDPRXNnkif+vJ5szsbjdjBxdISJo4spCXt55o3NVJb4wcrw59yjVPjKGR+9iD+/s21I6tyfgNfFlNoSvG4HDofFsWNLyebyfLC5nWXrWkimcpx78hgmjIqQzeUJ+NwEfS66Exm642lsu+fSgsftYNWmdkZXBPF5XJRFfAS8Ll5f2cCydS28v76VEyaWc+KkCk6bNqLnVkMbvJ6eWwmbOhIkUznGVIX6XPNPpLLYtk3A5+7np9g3E/9dHK4PLPuiEB4gE3/ZjkQm1jGeSdCe6qA6WIXD2rNXE03HCHmCO18bx+v07vHBJZqJsbRhGZZlcVr1yaztWM+Wru34XF5Ghar589a/ks5nqI82EPGGyeZz1McauKjuU4TcQdpTHXSlo9SER5Nyxvjd6ufI5LNcNunTzKiYznff/g+S2RTn1p7FW43vsj26o/DeFhYnVc3g7aZllHgjTCyp440dSwEo85VSWzSaeCZRmOcQ8RT1OwIB4Ha49zlfYXd/P3dhfGQs0UyUpnjP/erTy6fyYft6nA4Hp46YyV/r3ySbz3LpxAvZ2r2d13e8RW3RGK6b+nma4s08t+XPzB13LtFMjKfW/5GIt4hJJROw7TylvhKmV0zlncblnFg1nXQuC9i0JNqoKRrNmra1/Oi9RwH4f5/5Hss2rmV7ZztFmTE0tceZOKaYZWtb6IylcTotsHuuK5cWefF5XKzf3kksmaG1MwmWhdftIJbomTRYVeJnXHURXo+T5eta8HtdBP1uWjuTtHfvfRjf5eyZg5BM5wZQR/YzA2SXsiIv7d1p8rZNWZGXTM6mK9bzQask7KUo6CEaz9AV77lrwLJgfHURmWwer8dJVWkAv8dFNp+nK5qmpipELJnF6bRwOx3Ut8bJZvMUh720dCUp8ruprQrj97po7UoyqjzIhh1dtHUlOb6unHDAjeWwsPM2Qb+bWCLD6i3tNHckwbZp605h2zB1XCmRoIfOWIrueIam9gTlER/FoZ42F4c8+L0u6kZFcLt6RiC6ExmqSwNkcnmKAh427Ohi5cY2tjZFCXhd2EBXLM0XPjmFEaUBnvzrRn736kbGVIT4+HHVTB5TzKiKnr+7jp1zOACyuTy5vM3mhm5GVQQJHuSHlIFQCA+QieFxJFIdByaZTfJ203I+NuKkPQK9oiLM5vpGopkYFf5yLMsq3FNtWRaZfJZoOsr/fvg7mhItzBt3HidUHkdjvJkiTxi/y8fr9W8RzyaYPeYThWH1pzc+T7m/jPGRWn6z7vfk8nmuPnYBLoeLpY3LeL3+Lcr9pcwddy6VgQqe3vgcH7StpdRXQsRbxKauLUwvn0qpr5jnNv+ZbdF6RgZHMLGkjle2vbbPiYQhd5B0Lk16AKF+IDwO9z7POaN6Ku/uWAnA5ZM/w2v1f2NsUQ3TK46lMd7MlJKJjAhW8mr9m8QzCT4x6lRWtX7ItPJjcFlOVrWtobZoDMlMimg6ybjiUWTtHK6dczt8Li8+p49NnVvY2N7A9NLjiCdzvLm6EYfVczvc6IoQDgv+8t4OPG4nHpeDWDJDLNkzEhLcOUy9qaGbDfVdfPLUGqKJDKlMjtauFM3tCdxuB58/dxKjKoJsaujmmdc38/6GVkaUBgn4XDR3JHA6LMZUhsjlbT7c2lGY4V4S9hLyu4kns2zc0YXTYZHP72+659DrDb78ACPC5XTg8zgP6HKG1+NkfHURqze373GsPOKjK57G53Fx7NgSfB4X73zYXPjg4nI6mDQmwtgRRUSCHvK2zYTREYqDXt7f0Eoqk8PpsGjpTJLP21x85vjCZMDeEb/3N7Ty4tvbmHf6WNq6UlSXBRhd0XNXikJ4gBQeQ0N1HDwTapjOZVjbsYEpJRNwOpzk8jka4828sWMpdcVjcVpOSnzFNMabmVo2hWg6ysvbX+PUETNJ5dK817ySMn8pp4w4kdd3vMVv1v4en8vHlNKJLG1cBsB10z5PY6yJMn8p7ckOntzwxz5tcFgOKv3le1zLH6giT5iudN86l/lKCboDbOnufyjb5XAR8RTRmmwDYP74C4h4iwi7g3idHtqSHbQm29jYuQWAC+suwO1wU+QJ8V7LKo4tm0yRJ0x3Osr7LauZXn4s6Xwav8tHc6KV6uCIwnyUvJ0nmU0ScAdoirfgspxEvEW0JNso95WSx+53LYG3G5eTydjMHDEdy4LNDd2FBW0Cvp4RgGzOxut24vVa1FQU4XY5aO5IMqWunI1b2qhvjZNIZfG6nexojeFyOhhRGqAzliaezGDTE7bt0RQ+j5Nja0sZWR4knckRCrjJ523Wbe+kM5om4HNRHvFTVuSlrTvFtuYo2VyOZCpPdzzDyk1tJFJZJo6OUBbxsaMlTjyVZVNDFydOrOCEieUcO7aUzliabU1RookMv/vrRjpjaYI+F5/+xDhSmRzPvLGZRKpnFKKsyEdHNFWYP+B0WPg8zsIHogPh9Tjxe5zYQCKZpbosyObGvr9HTodFOODG73XxvVvPIh49uE2B9kYhLP1SHQfvo1jD9mQHLoeLsCfE9ugO4pk4E0vq+rxmXcdGsvkso0MjWdO+jimlEwm6A+TtPG3JDla0rgZgTGgU3VYH1a6RvNeyir81vMNpI0/mw/Z1bO7axpmjPs66jg00J1pxWBZBd5DOVBfZfJZ4NkHO3jV8XOYrIZ5NAD0T8WLZOKNC1WTyGbrTUSaVTGBT55b9Du/vjd/lJ51L93m/Xg7LQcDlx+VwFWb4e50eUrk0FhZ+l29nu3o+TEyvmMr6nfUZHxlLQ7yJzlQXU8sm82r934CeiY6xTJy6yFiwLN5vXkmxL0LYHWZm1fFs7d7O7zc+xxkjP0Z1qIoVLR8wbeRE3FkfrYl2RoerSWSTlHiLe+ZnJFqo9JcTzcQIuPy0JNvwOr3URcayvnMjDhzUFo3hr9vfYHrFsVQFKlnZ+gGdqS5OqDyORDZJS6KVjlQnS9Y+yXk1szlv7L437dl9nknezvNe80rqiscR9oT2eq98JpvD4YB83sLtchBP9szE31DfxYyJ5ZQW9dxKF01kyObyNLTG2dEao607xfbmGE0dCeLJDMeNL6O82I/P7aS1K1m4TS+VyRPwuWhqT1BZ7KemKkRrV5JjaktZvr6FTCbPiLIA3/zSaXR2DN1eBwph6ZfqOHiq4eDtrYa2bfesDtfPRMTudJRULk2pr5j1HRupLaop/MOfy2dZ27GBY0sn9znH9ugOHl7+czK5DFPLpxDxFOF0OAl7QowtGsPbjct5Y8dSJhSPZ1XbGoKuAFXBSuqjO/C7fNRFxrGidTV1xeNoS7bTGG8mnUtT5ishm8/idXmxcNAYb2JkcAR5bGKZ2B7rFHgcbizLKgQ1sN+1Bg4nn9O333UKir0RvE4PI4MjyNpZMrksmXwGl8NFa6KNgDuA2+Ei6A7yXkvP5YZxRTW4nR4mFdcRdPt5tf5vZPJZ6iK1LG1ajtvhIuwO8bHqkyjyhCnxFVMXGctL2/6Kz+nl1OqT2Ni5Cb8rwJjwyH7bt7cJp/FkFp/HWdgk5+9pOHqA9A/f0FAdB081HLzDXcNcPtdz29peJvbBrn+8c/ncgO5G6L3VcffztyTbqArsWpY2ncvQne7GYTloTrRSEx6Ny+FkXcdGSrwRfC4fL297DYdlcUzp5J0r/o1gW3c9OTvH2KJaOlOdxLMJ3m5chsvhorZoDO+3rGJ8ZCzF3ggtuWZcWQ/l/jLeaVqOx+HB7XSTzCYZEahkQ9dmwu4gLoebIk+YznQX27q3My5SS3c6RmO8kfZkJ92ZKOX+MsYV1RDNxFjd9iHl/jJOrppBOp+mwl/Gi1v+QlNizw1o9mcgCxv9vb3dfgg9ayWkcxkml04gl88RyyZI59J4nG5KfSW8Wv8mmVyWiSXjqfRXUB2spLZoDOlcmk1dW+nORIll4rQkWjl1xEymlR+jEB4o/cM3NFTHwVMNB081HBpDVcfedQ6g9/bIFRxTOonAbh808naenJ0nb+dJ5VK4HW7cDhdOy0ksE8fn8hLLJKiP7mBbtJ6ZVccTcodwWg66M1GeWv8slYFyPlZ9Mpl8mu3RBmqLRtOdjuFyONnWXU8ql2Jj5xbeb11F0B2g0t+zjkJd8VjWd2wasrUHeo0MjmDR+bcR7Ri6SYlH/QYOIiIytHoDGHomtc2sOmGP1zgsR2Ek4e/3cu+9RTDiDRPxhjmmbFKf48XeCFcd+9k+z5X7ywrHYNfmPJ8Y9bG9tjGXz9Ge6sRhWaxu/RCfy0fEW4Tf5SNv53ll22vEMnGumXoFXaludsQa2Nq9nZe2vUq5v5SPVZ+E3+nD7XTTmepiaeMyfIXbGQ/9wjXqCQugOg4F1XDwVMOhoTruXzafxWk597k4kbYyFBEROURc+9lq9nDZ+2wEEREROeQUwiIiIsNEISwiIjJMFMIiIiLDRCEsIiIyTBTCIiIiw0QhLCIiMkwUwiIiIsNEISwiIjJMFMIiIiLDRCEsIiIyTA77Bg4iIiLSQz1hERGRYaIQFhERGSYKYRERkWGiEBYRERkmCmEREZFhohAWEREZJq7hbsDByufz3HPPPaxZswaPx8PChQupra0d7mYd8ZYvX853vvMdHnvsMTZv3sztt9+OZVlMnDiRb37zmzgcDhYvXsyvfvUrXC4XN954I7Nnzx7uZh8RMpkMd955J9u3byedTnPjjTcyYcIE1fAA5HI57rrrLjZu3IjT6WTRokXYtq0aHoTW1lY+85nP8POf/xyXy6UaHoSLLrqIcDgMwOjRo7nhhhsOfx1tQz377LP2bbfdZtu2bb/77rv2DTfcMMwtOvI98sgj9rx58+zLLrvMtm3bvv766+033njDtm3bvvvuu+3nnnvObmpqsufNm2enUim7q6ur8L3Y9pIlS+yFCxfatm3bbW1t9qxZs1TDA/T888/bt99+u23btv3GG2/YN9xwg2p4ENLptP1P//RP9nnnnWevW7dONTwIyWTS/vSnP93nueGoo7HD0W+//TZnnHEGACeccAIrVqwY5hYd+WpqanjwwQcLj1euXMkpp5wCwJlnnslrr73Ge++9x4wZM/B4PITDYWpqavjggw+Gq8lHlAsuuIAvf/nLhcdOp1M1PEDnnHMO9913HwD19fWUl5erhgfhgQce4PLLL6eyshLQ3+WD8cEHH5BIJLj22mu5+uqrWbZs2bDU0dgQjkajhEKhwmOn00k2mx3GFh35zj//fFyuXVcgbNvGsiwAgsEg3d3dRKPRwvBM7/PRaPSwt/VIFAwGCYVCRKNRbrnlFr7yla+ohgfB5XJx2223cd9993H++eerhgfo17/+NaWlpYVOCOjv8sHw+Xxcd911/OxnP+Pee+/lq1/96rDU0dgQDoVCxGKxwuN8Pt8nYGT/HI5df/yxWIyioqI96hqLxfr8An7U7dixg6uvvppPf/rTzJ8/XzU8SA888ADPPvssd999N6lUqvC8arh/TzzxBK+99hpXXXUVq1ev5rbbbqOtra1wXDUcmHHjxnHhhRdiWRbjxo2juLiY1tbWwvHDVUdjQ/jEE0/klVdeAWDZsmVMmjRpmFtknmOPPZY333wTgFdeeYWTTjqJ6dOn8/bbb5NKpeju7mb9+vWq7U4tLS1ce+21fO1rX+PSSy8FVMMD9dvf/paf/OQnAPj9fizLYtq0aarhAfjlL3/JL37xCx577DGOOeYYHnjgAc4880zV8AAtWbKE+++/H4DGxkai0Sinn376Ya+jsRs49M6O/vDDD7Ftm29961vU1dUNd7OOeNu2beOf//mfWbx4MRs3buTuu+8mk8kwfvx4Fi5ciNPpZPHixTz++OPYts3111/P+eefP9zNPiIsXLiQP/zhD4wfP77w3Ne//nUWLlyoGg5QPB7njjvuoKWlhWw2y5e+9CXq6ur0e3iQrrrqKu655x4cDodqeIDS6TR33HEH9fX1WJbFV7/6VUpKSg57HY0NYREREdMZOxwtIiJiOoWwiIjIMFEIi4iIDBOFsIiIyDBRCIuIiAwThbCIiMgwUQiLiIgME4WwiIjIMPn/AZooyle0vEaAAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "reg_params = [1e-6, 1e-3, 1e-2]\n",
    "# reg_params = [1e-3]\n",
    "epochs_num = 500\n",
    "all_history = dict()\n",
    "runs_num = 4\n",
    "all_models = dict()\n",
    "with tf.device(\"/device:cpu:0\"):\n",
    "    for reg_param in reg_params:\n",
    "        all_mse = np.zeros([epochs_num])\n",
    "        all_val_mse = np.zeros([epochs_num])\n",
    "        for runs in range(runs_num):\n",
    "            print(f\"Training reg:{reg_param}, run_num:{runs}\\n\")\n",
    "            model = get_network(1, 1, 15, 10, reg_param)\n",
    "            history = model.fit(x_train, y_train,\n",
    "                                epochs=epochs_num, batch_size=25,\n",
    "                                shuffle=True, verbose=0,\n",
    "                                validation_data=(x_val, y_val))\n",
    "            all_history[f\"{reg_param}-{runs}\"] = history\n",
    "            all_models[f\"{reg_param:.0e}-{runs}\"] = model\n",
    "            out = model.predict(x_val)\n",
    "            all_mse += history.history['mse']\n",
    "            all_val_mse += history.history['val_mse']\n",
    "        avg_mse = all_mse/4\n",
    "        avg_val_mse = all_val_mse/4\n",
    "        plt.plot(avg_mse, label='avg MSE')\n",
    "        plt.plot(avg_val_mse, label='avg Val MSE')\n",
    "        # compute the history mse and plot\n",
    "\n",
    "        # plt.figure(figsize=(20,10))\n",
    "        # plt.scatter(x_train, y_train, label=\"train\", s=5)\n",
    "        # plt.plot(np.linspace(0, 1, 250), model.predict(\n",
    "        #     np.linspace(0, 1, 250)), label=\"Model\", alpha=0.3)\n",
    "        # plt.scatter(x_val, y_val, alpha=0.7, marker='x', label=\"test_data\")\n",
    "        # plt.scatter(x_val, out, marker='^', label=\"Prediction\")\n",
    "        # plt.plot(x_val, out, label=\"Prediction\")\n",
    "\n",
    "        # plt.title(f\"Reg: {reg_param}, Loss: {loss}, MSE: {mse}\")\n",
    "        plt.legend()\n",
    "        plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: models/1e-06-0/assets\n",
      "INFO:tensorflow:Assets written to: models/1e-06-1/assets\n",
      "INFO:tensorflow:Assets written to: models/1e-06-2/assets\n",
      "INFO:tensorflow:Assets written to: models/1e-06-3/assets\n",
      "INFO:tensorflow:Assets written to: models/0.001-0/assets\n",
      "INFO:tensorflow:Assets written to: models/0.001-1/assets\n",
      "INFO:tensorflow:Assets written to: models/0.001-2/assets\n",
      "INFO:tensorflow:Assets written to: models/0.001-3/assets\n",
      "INFO:tensorflow:Assets written to: models/0.01-0/assets\n",
      "INFO:tensorflow:Assets written to: models/0.01-1/assets\n",
      "INFO:tensorflow:Assets written to: models/0.01-2/assets\n",
      "INFO:tensorflow:Assets written to: models/0.01-3/assets\n"
     ]
    }
   ],
   "source": [
    "for m in all_model.keys():\n",
    "    all_model[m].save(f\"models/{m}.model\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "with open(\"history.pickle\", 'wb') as hist:\n",
    "    extracted_hist = []\n",
    "    for k,v in all_history.items():\n",
    "        extracted_hist.append(v.history)\n",
    "    pickle.dump(extracted_hist, hist)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "with open(\"models_5000/history.pickle\", 'rb') as hist:\n",
    "    all_history = pickle.load(hist)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "reg_params = [1e-6, 1e-3, 1e-2]\n",
    "for hst_num in range(0, len(all_history), 4):\n",
    "    hist_four_runs = [all_history[hst_num+i] for i in range(4)]\n",
    "    MSE_train_four_runs = [hist_four_runs[i]['mse'] for i in range(4)]\n",
    "    MSE_val_four_runs = [hist_four_runs[i]['val_mse']for i in range(4)]\n",
    "    MSE_train_avg = np.sum(MSE_train_four_runs, axis=0)/4\n",
    "    MSE_val_avg = np.sum(MSE_val_four_runs, axis=0)/4\n",
    "    # plt.figure(figsize=(10, 5), dpi=200)\n",
    "    plt.figure(dpi=100)\n",
    "    plt.plot(MSE_train_four_runs[0], label=\"First run\", linewidth=1, alpha=0.6)\n",
    "    plt.plot(MSE_train_four_runs[1],\n",
    "             label=\"Second run\", linewidth=1, alpha=0.6)\n",
    "    plt.plot(MSE_train_four_runs[2], label=\"Third run\", linewidth=1, alpha=0.6)\n",
    "    plt.plot(MSE_train_four_runs[3],\n",
    "             label=\"Fourth run\", linewidth=1, alpha=0.6)\n",
    "    plt.plot(MSE_train_avg, linewidth=1, label=\"Training avg MSE\", alpha=0.8)\n",
    "    plt.plot(MSE_val_avg, linewidth=1, label=\"Training avg MSE\", alpha=0.8)\n",
    "    final_train_mse = hist_four_runs[3]['mse'][-1]\n",
    "    final_val_mse = hist_four_runs[3]['val_mse'][-1]\n",
    "    plt.ylim(0, 0.8)\n",
    "    plt.title(\n",
    "        f\"REG_param: {reg_params[int(hst_num/4)]: .0e}\\\n",
    "            \\n\\n\\\n",
    "        Final Training MSE:  {final_train_mse: .4f},  Final Val MSE:  {final_val_mse: .4f}\")\n",
    "    plt.legend(bbox_to_anchor=(1.25, 1))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "models_5000/0.001-0.model\n",
      "Metal device set to: Apple M1\n",
      "\n",
      "systemMemory: 16.00 GB\n",
      "maxCacheSize: 5.33 GB\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-09-21 22:29:26.090995: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:305] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.\n",
      "2021-09-21 22:29:26.091131: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:271] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>)\n"
     ]
    },
    {
     "ename": "JSONDecodeError",
     "evalue": "Expecting value: line 1 column 1 (char 0)",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mJSONDecodeError\u001b[0m                           Traceback (most recent call last)",
      "\u001b[0;32m/var/folders/6h/mmvp5df90fb3fsqtc5d9mxt40000gn/T/ipykernel_6511/404642183.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      7\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mm\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mmodel_list\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      8\u001b[0m     \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodels_path\u001b[0m\u001b[0;34m+\u001b[0m\u001b[0mm\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 9\u001b[0;31m     \u001b[0mall_models\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mm\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m6\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mkeras\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodels\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload_model\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodels_path\u001b[0m\u001b[0;34m+\u001b[0m\u001b[0mm\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m~/miniforge3/envs/tf/lib/python3.8/site-packages/tensorflow/python/keras/saving/save.py\u001b[0m in \u001b[0;36mload_model\u001b[0;34m(filepath, custom_objects, compile, options)\u001b[0m\n\u001b[1;32m    204\u001b[0m         \u001b[0mfilepath\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpath_to_string\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfilepath\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    205\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfilepath\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 206\u001b[0;31m           \u001b[0;32mreturn\u001b[0m \u001b[0msaved_model_load\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfilepath\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcompile\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moptions\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    207\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    208\u001b[0m   raise IOError(\n",
      "\u001b[0;32m~/miniforge3/envs/tf/lib/python3.8/site-packages/tensorflow/python/keras/saving/saved_model/load.py\u001b[0m in \u001b[0;36mload\u001b[0;34m(path, compile, options)\u001b[0m\n\u001b[1;32m    144\u001b[0m   \u001b[0;31m# Recreate layers and metrics using the info stored in the metadata.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    145\u001b[0m   \u001b[0mkeras_loader\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mKerasObjectLoader\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmetadata\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mobject_graph_def\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 146\u001b[0;31m   \u001b[0mkeras_loader\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload_layers\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcompile\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcompile\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    147\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    148\u001b[0m   \u001b[0;31m# Generate a dictionary of all loaded nodes.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/miniforge3/envs/tf/lib/python3.8/site-packages/tensorflow/python/keras/saving/saved_model/load.py\u001b[0m in \u001b[0;36mload_layers\u001b[0;34m(self, compile)\u001b[0m\n\u001b[1;32m    376\u001b[0m         \u001b[0;32mcontinue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    377\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 378\u001b[0;31m       self.loaded_nodes[node_metadata.node_id] = self._load_layer(\n\u001b[0m\u001b[1;32m    379\u001b[0m           \u001b[0mnode_metadata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnode_id\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnode_metadata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0midentifier\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    380\u001b[0m           node_metadata.metadata)\n",
      "\u001b[0;32m~/miniforge3/envs/tf/lib/python3.8/site-packages/tensorflow/python/keras/saving/saved_model/load.py\u001b[0m in \u001b[0;36m_load_layer\u001b[0;34m(self, node_id, identifier, metadata)\u001b[0m\n\u001b[1;32m    418\u001b[0m     \u001b[0;31m# Detect whether this object can be revived from the config. If not, then\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    419\u001b[0m     \u001b[0;31m# revive from the SavedModel instead.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 420\u001b[0;31m     \u001b[0mobj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msetter\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_revive_from_config\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0midentifier\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmetadata\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnode_id\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    421\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mobj\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    422\u001b[0m       \u001b[0mobj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msetter\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mrevive_custom_object\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0midentifier\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmetadata\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/miniforge3/envs/tf/lib/python3.8/site-packages/tensorflow/python/keras/saving/saved_model/load.py\u001b[0m in \u001b[0;36m_revive_from_config\u001b[0;34m(self, identifier, metadata, node_id)\u001b[0m\n\u001b[1;32m    442\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    443\u001b[0m     \u001b[0msetter\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_config_node_setter\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_revive_setter\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 444\u001b[0;31m     self._add_children_recreated_from_config(\n\u001b[0m\u001b[1;32m    445\u001b[0m         obj, self._proto.nodes[node_id], node_id)\n\u001b[1;32m    446\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0mobj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msetter\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/miniforge3/envs/tf/lib/python3.8/site-packages/tensorflow/python/keras/saving/saved_model/load.py\u001b[0m in \u001b[0;36m_add_children_recreated_from_config\u001b[0;34m(self, obj, proto, node_id)\u001b[0m\n\u001b[1;32m    303\u001b[0m     \u001b[0mobj\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_maybe_initialize_trackable\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    304\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mobj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbase_layer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mLayer\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mobj\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbuilt\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 305\u001b[0;31m       \u001b[0mmetadata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mjson_utils\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdecode\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mproto\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0muser_object\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmetadata\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    306\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_try_build_layer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mobj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnode_id\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmetadata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'build_input_shape'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    307\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/miniforge3/envs/tf/lib/python3.8/site-packages/tensorflow/python/keras/saving/saved_model/json_utils.py\u001b[0m in \u001b[0;36mdecode\u001b[0;34m(json_string)\u001b[0m\n\u001b[1;32m     60\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     61\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mdecode\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mjson_string\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 62\u001b[0;31m   \u001b[0;32mreturn\u001b[0m \u001b[0mjson\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloads\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mjson_string\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mobject_hook\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0m_decode_helper\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     63\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     64\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/miniforge3/envs/tf/lib/python3.8/json/__init__.py\u001b[0m in \u001b[0;36mloads\u001b[0;34m(s, cls, object_hook, parse_float, parse_int, parse_constant, object_pairs_hook, **kw)\u001b[0m\n\u001b[1;32m    368\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mparse_constant\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    369\u001b[0m         \u001b[0mkw\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'parse_constant'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mparse_constant\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 370\u001b[0;31m     \u001b[0;32mreturn\u001b[0m \u001b[0mcls\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0mkw\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdecode\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0ms\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m~/miniforge3/envs/tf/lib/python3.8/json/decoder.py\u001b[0m in \u001b[0;36mdecode\u001b[0;34m(self, s, _w)\u001b[0m\n\u001b[1;32m    335\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    336\u001b[0m         \"\"\"\n\u001b[0;32m--> 337\u001b[0;31m         \u001b[0mobj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mend\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mraw_decode\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0ms\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0midx\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0m_w\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0ms\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    338\u001b[0m         \u001b[0mend\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_w\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0ms\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mend\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    339\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mend\u001b[0m \u001b[0;34m!=\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0ms\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/miniforge3/envs/tf/lib/python3.8/json/decoder.py\u001b[0m in \u001b[0;36mraw_decode\u001b[0;34m(self, s, idx)\u001b[0m\n\u001b[1;32m    353\u001b[0m             \u001b[0mobj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mend\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mscan_once\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0ms\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0midx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    354\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mStopIteration\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0merr\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 355\u001b[0;31m             \u001b[0;32mraise\u001b[0m \u001b[0mJSONDecodeError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Expecting value\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0ms\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0merr\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalue\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    356\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mobj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mend\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mJSONDecodeError\u001b[0m: Expecting value: line 1 column 1 (char 0)"
     ]
    }
   ],
   "source": [
    "import os\n",
    "models_path = \"models_5000/\"\n",
    "file_list = os.listdir(models_path)\n",
    "model_list = [\n",
    "    file_name for file_name in file_list if file_name[-6:] == \".model\"]\n",
    "all_models = dict()\n",
    "for m in model_list:\n",
    "    print(models_path+m)\n",
    "    all_models[m[:-6]] = keras.models.load_model(models_path+m)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "ename": "JSONDecodeError",
     "evalue": "Expecting value: line 1 column 1 (char 0)",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mJSONDecodeError\u001b[0m                           Traceback (most recent call last)",
      "\u001b[0;32m/var/folders/6h/mmvp5df90fb3fsqtc5d9mxt40000gn/T/ipykernel_6511/4113717988.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mtest_model\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mkeras\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodels\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload_model\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"/Volumes/GoogleDrive/我的云端硬盘/Colab Notebooks/Assignment2/models_5000/1e-06-0\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m~/miniforge3/envs/tf/lib/python3.8/site-packages/tensorflow/python/keras/saving/save.py\u001b[0m in \u001b[0;36mload_model\u001b[0;34m(filepath, custom_objects, compile, options)\u001b[0m\n\u001b[1;32m    204\u001b[0m         \u001b[0mfilepath\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpath_to_string\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfilepath\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    205\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfilepath\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 206\u001b[0;31m           \u001b[0;32mreturn\u001b[0m \u001b[0msaved_model_load\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfilepath\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcompile\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moptions\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    207\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    208\u001b[0m   raise IOError(\n",
      "\u001b[0;32m~/miniforge3/envs/tf/lib/python3.8/site-packages/tensorflow/python/keras/saving/saved_model/load.py\u001b[0m in \u001b[0;36mload\u001b[0;34m(path, compile, options)\u001b[0m\n\u001b[1;32m    144\u001b[0m   \u001b[0;31m# Recreate layers and metrics using the info stored in the metadata.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    145\u001b[0m   \u001b[0mkeras_loader\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mKerasObjectLoader\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmetadata\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mobject_graph_def\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 146\u001b[0;31m   \u001b[0mkeras_loader\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload_layers\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcompile\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcompile\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    147\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    148\u001b[0m   \u001b[0;31m# Generate a dictionary of all loaded nodes.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/miniforge3/envs/tf/lib/python3.8/site-packages/tensorflow/python/keras/saving/saved_model/load.py\u001b[0m in \u001b[0;36mload_layers\u001b[0;34m(self, compile)\u001b[0m\n\u001b[1;32m    376\u001b[0m         \u001b[0;32mcontinue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    377\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 378\u001b[0;31m       self.loaded_nodes[node_metadata.node_id] = self._load_layer(\n\u001b[0m\u001b[1;32m    379\u001b[0m           \u001b[0mnode_metadata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnode_id\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnode_metadata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0midentifier\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    380\u001b[0m           node_metadata.metadata)\n",
      "\u001b[0;32m~/miniforge3/envs/tf/lib/python3.8/site-packages/tensorflow/python/keras/saving/saved_model/load.py\u001b[0m in \u001b[0;36m_load_layer\u001b[0;34m(self, node_id, identifier, metadata)\u001b[0m\n\u001b[1;32m    418\u001b[0m     \u001b[0;31m# Detect whether this object can be revived from the config. If not, then\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    419\u001b[0m     \u001b[0;31m# revive from the SavedModel instead.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 420\u001b[0;31m     \u001b[0mobj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msetter\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_revive_from_config\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0midentifier\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmetadata\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnode_id\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    421\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mobj\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    422\u001b[0m       \u001b[0mobj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msetter\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mrevive_custom_object\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0midentifier\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmetadata\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/miniforge3/envs/tf/lib/python3.8/site-packages/tensorflow/python/keras/saving/saved_model/load.py\u001b[0m in \u001b[0;36m_revive_from_config\u001b[0;34m(self, identifier, metadata, node_id)\u001b[0m\n\u001b[1;32m    442\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    443\u001b[0m     \u001b[0msetter\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_config_node_setter\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_revive_setter\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 444\u001b[0;31m     self._add_children_recreated_from_config(\n\u001b[0m\u001b[1;32m    445\u001b[0m         obj, self._proto.nodes[node_id], node_id)\n\u001b[1;32m    446\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0mobj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msetter\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/miniforge3/envs/tf/lib/python3.8/site-packages/tensorflow/python/keras/saving/saved_model/load.py\u001b[0m in \u001b[0;36m_add_children_recreated_from_config\u001b[0;34m(self, obj, proto, node_id)\u001b[0m\n\u001b[1;32m    303\u001b[0m     \u001b[0mobj\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_maybe_initialize_trackable\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    304\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mobj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbase_layer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mLayer\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mobj\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbuilt\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 305\u001b[0;31m       \u001b[0mmetadata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mjson_utils\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdecode\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mproto\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0muser_object\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmetadata\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    306\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_try_build_layer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mobj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnode_id\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmetadata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'build_input_shape'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    307\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/miniforge3/envs/tf/lib/python3.8/site-packages/tensorflow/python/keras/saving/saved_model/json_utils.py\u001b[0m in \u001b[0;36mdecode\u001b[0;34m(json_string)\u001b[0m\n\u001b[1;32m     60\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     61\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mdecode\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mjson_string\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 62\u001b[0;31m   \u001b[0;32mreturn\u001b[0m \u001b[0mjson\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloads\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mjson_string\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mobject_hook\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0m_decode_helper\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     63\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     64\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/miniforge3/envs/tf/lib/python3.8/json/__init__.py\u001b[0m in \u001b[0;36mloads\u001b[0;34m(s, cls, object_hook, parse_float, parse_int, parse_constant, object_pairs_hook, **kw)\u001b[0m\n\u001b[1;32m    368\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mparse_constant\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    369\u001b[0m         \u001b[0mkw\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'parse_constant'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mparse_constant\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 370\u001b[0;31m     \u001b[0;32mreturn\u001b[0m \u001b[0mcls\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0mkw\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdecode\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0ms\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m~/miniforge3/envs/tf/lib/python3.8/json/decoder.py\u001b[0m in \u001b[0;36mdecode\u001b[0;34m(self, s, _w)\u001b[0m\n\u001b[1;32m    335\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    336\u001b[0m         \"\"\"\n\u001b[0;32m--> 337\u001b[0;31m         \u001b[0mobj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mend\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mraw_decode\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0ms\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0midx\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0m_w\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0ms\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    338\u001b[0m         \u001b[0mend\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_w\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0ms\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mend\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    339\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mend\u001b[0m \u001b[0;34m!=\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0ms\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/miniforge3/envs/tf/lib/python3.8/json/decoder.py\u001b[0m in \u001b[0;36mraw_decode\u001b[0;34m(self, s, idx)\u001b[0m\n\u001b[1;32m    353\u001b[0m             \u001b[0mobj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mend\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mscan_once\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0ms\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0midx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    354\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mStopIteration\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0merr\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 355\u001b[0;31m             \u001b[0;32mraise\u001b[0m \u001b[0mJSONDecodeError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Expecting value\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0ms\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0merr\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalue\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    356\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mobj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mend\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mJSONDecodeError\u001b[0m: Expecting value: line 1 column 1 (char 0)"
     ]
    }
   ],
   "source": [
    "test_model = tf.keras.models.load_model(\"/Volumes/GoogleDrive/我的云端硬盘/Colab Notebooks/Assignment2/models_5000/1e-06-0\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'2.5.0'"
      ]
     },
     "execution_count": 120,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.version.VERSION"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "Zhilong_Assignment2.ipynb",
   "provenance": []
  },
  "interpreter": {
   "hash": "859cadc33cf9c93d9354f622d0bfdc75f7d75fc23a9b22a18dfc646072acf83e"
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.8.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}