gaussian2S_change_epsilon.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Using device: <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "Using device: \u001b[1;36m1\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import torch\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "\n",
    "from torch import nn, optim\n",
    "from torch.utils.data import Dataset, DataLoader\n",
    "from tqdm.notebook import tqdm\n",
    "from scipy.stats import norm\n",
    "import plotly.graph_objects as go\n",
    "from torch.optim.lr_scheduler import CosineAnnealingLR, CosineAnnealingWarmRestarts, StepLR, OneCycleLR\n",
    "from pathlib import Path\n",
    "import matplotlib.animation as animation\n",
    "from sklearn.datasets import *\n",
    "\n",
    "\n",
    "\n",
    "from rich import print\n",
    "import imageio\n",
    "\n",
    "\n",
    "from utils.Models import MLP\n",
    "from utils.Datasets import BBdataset\n",
    "from utils.utils import *\n",
    "seed = 233\n",
    "torch.manual_seed(seed)\n",
    "if torch.cuda.is_available():\n",
    "    torch.cuda.manual_seed_all(seed)\n",
    "np.random.seed(seed)\n",
    "# os.environ[\"CUDA_VISIBLE_DEVICES\"]=\"1\"\n",
    "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
    "device = 1\n",
    "print(f\"Using device: {device}\")\n",
    "\n",
    "\n",
    "experiment_name = \"gaussian2S_2model\"  # 你可以根据需要动态设置这个变量\n",
    "# log_dir = Path('experiments') / experiment_name / 'test' / time.strftime(\"%Y-%m-%d/%H_%M_%S/\")\n",
    "log_dir = Path('experiments') / experiment_name\n",
    "log_dir.mkdir(parents=True, exist_ok=True)\n",
    "os.environ[\"CUDA_VISIBLE_DEVICES\"]=\"0\"\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">torch.Size</span><span style=\"font-weight: bold\">([</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">910</span><span style=\"font-weight: bold\">])</span>\n",
       "<span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">torch.Size</span><span style=\"font-weight: bold\">([</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">900</span><span style=\"font-weight: bold\">])</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;35mtorch.Size\u001b[0m\u001b[1m(\u001b[0m\u001b[1m[\u001b[0m\u001b[1;36m910\u001b[0m\u001b[1m]\u001b[0m\u001b[1m)\u001b[0m\n",
       "\u001b[1;35mtorch.Size\u001b[0m\u001b[1m(\u001b[0m\u001b[1m[\u001b[0m\u001b[1;36m900\u001b[0m\u001b[1m]\u001b[0m\u001b[1m)\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 生成初始和目标密度样本\n",
    "def gen_data(m=4, num_samples=1000):\n",
    "    mus = [(m, m), (m, -m), (-m, m), (-m, -m)]\n",
    "    source_dist = np.random.multivariate_normal([0, 0], [[1, 0], [0, 1]], num_samples)\n",
    "    target_dist = make_s_curve(n_samples=num_samples, noise=0.01)[0][:, ::2]\n",
    "    return source_dist, target_dist\n",
    "\n",
    "# 生成二维Brownian bridge\n",
    "def gen_bridge_2d(x, y, ts, T, num_samples):\n",
    "    sigma = 1\n",
    "    bridge = torch.zeros((ts.shape[0], num_samples, 2))\n",
    "    drift = torch.zeros((ts.shape[0], num_samples, 2))\n",
    "    bridge[0] = x\n",
    "    for i in range(len(ts) - 1):\n",
    "        dt = ts[i+1] - ts[i]\n",
    "        dydt = (y - bridge[i]) / (T - ts[i])\n",
    "        drift[i, :] = dydt\n",
    "        diffusion = sigma * torch.sqrt(dt) * torch.randn(num_samples, 2)\n",
    "        bridge[i+1] = bridge[i] + dydt * dt\n",
    "        bridge[i+1, :] += diffusion\n",
    "    return bridge, drift\n",
    "\n",
    "# 主函数\n",
    "def gen_2d_data(m=4, num_samples=1000, epsilon=0.001, T=1):\n",
    "    if not isinstance(num_samples, int):\n",
    "        num_samples = int(num_samples)\n",
    "    source_dist, target_dist = gen_data(m, num_samples)\n",
    "        \n",
    "    epsilon1 = 0.001\n",
    "    epsilon2 = 0.0001\n",
    "\n",
    "    t1 = torch.arange(0, 0.91, epsilon1)\n",
    "    t2 = torch.arange(0.91, T, epsilon2)\n",
    "    print(t1.shape, t2.shape)\n",
    "    ts = torch.concatenate((t1, t2))\n",
    "    \n",
    "    source_dist = torch.Tensor(source_dist)\n",
    "    target_dist = torch.Tensor(target_dist)\n",
    "    bridge, drift = gen_bridge_2d(source_dist, target_dist, ts, T=T, num_samples=num_samples)\n",
    "    return ts, bridge, drift, source_dist, target_dist\n",
    "\n",
    "ts, bridge, drift, source_sample, target_sample = gen_2d_data(num_samples=1000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_source_and_target(sour, targ, left_title=\"Source distribution\", right_title=\"Target distribution\", save_path=None):\n",
    "    fig, axs = plt.subplots(1, 2, figsize=(10, 5))\n",
    "    axs[0].scatter(*sour.T, c='#D27685', s=10, alpha=0.6)\n",
    "    axs[1].scatter(*targ.T, c='#37306B', s=10, alpha=0.6)\n",
    "    axs[0].set_title(left_title)\n",
    "    axs[1].set_title(right_title)\n",
    "    axs[0].grid(True, alpha=0.3)\n",
    "    axs[1].grid(True, alpha=0.3)\n",
    "    axs[0].set_xlim(-8, 8)\n",
    "    axs[0].set_ylim(-8, 8)\n",
    "    axs[1].set_xlim(-8, 8)\n",
    "    axs[1].set_ylim(-8, 8)\n",
    "    fig.show()\n",
    "    if save_path is not None:\n",
    "        fig.savefig(save_path)\n",
    "\n",
    "plot_source_and_target(source_sample, target_sample)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def draw_gaussian2d(bridge, colors=True):\n",
    "    def get_color(point):\n",
    "        x, y = point\n",
    "        if x > 0 and y > 0:\n",
    "            return 0\n",
    "        elif x > 0 and y < 0:\n",
    "            return 1\n",
    "        elif x < 0 and y > 0:\n",
    "            return 2\n",
    "        else:\n",
    "            return 3\n",
    "\n",
    "    # 生成数据\n",
    "    color_list = ['#37306B', '#66347F', '#9E4784', '#D27685']  # 对应于目标均值的颜色\n",
    "\n",
    "    # 创建3D图形\n",
    "    fig = plt.figure(figsize=(10, 8))\n",
    "    ax = fig.add_subplot(111, projection='3d')\n",
    "\n",
    "    # 为每个样本绘制bridge\n",
    "    for i in range(bridge.shape[1]):\n",
    "        x = bridge[:, i, 0].numpy()  # X坐标\n",
    "        y = bridge[:, i, 1].numpy()  # Y坐标\n",
    "        z = ts.numpy()               # 时间作为Z坐标\n",
    "        c = color_list[get_color((x[-1], y[-1]))] if colors else '#D27685'\n",
    "        ax.plot(x, y, z, color=c, label=f'Sample {i+1}', alpha=0.3)\n",
    "\n",
    "    # 添加标签和图例\n",
    "    ax.set_xlabel('X')\n",
    "    ax.set_ylabel('Y')\n",
    "    ax.set_zlabel('Time')\n",
    "\n",
    "    fig.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def draw_gaussian2d_gif(bridge):\n",
    "    def get_color(point):\n",
    "        x, y = point\n",
    "        if x > 0 and y > 0:\n",
    "            return 0\n",
    "        elif x > 0 and y < 0:\n",
    "            return 1\n",
    "        elif x < 0 and y > 0:\n",
    "            return 2\n",
    "        else:\n",
    "            return 3\n",
    "    colors = ['b', 'g', 'r', 'c']  # 对应于目标均值的颜色\n",
    "    fig, ax = plt.subplots(figsize=(10, 8))\n",
    "    ax.set_xlabel('X')\n",
    "    ax.set_ylabel('Y')\n",
    "    ax.set_xlim(-10, 10)\n",
    "    ax.set_ylim(-10, 10)\n",
    "    # 定义更新函数\n",
    "    def update(frame):\n",
    "        ax.clear()\n",
    "        ax.set_xlabel('X')\n",
    "        ax.set_ylabel('Y')\n",
    "        ax.set_xlim(-10, 10)\n",
    "        ax.set_ylim(-10, 10)\n",
    "        for i in range(bridge.shape[1]):\n",
    "            x = bridge[:frame + 1, i, 0].numpy()  # 注意：frame + 1\n",
    "            y = bridge[:frame + 1, i, 1].numpy()  # 注意：frame + 1\n",
    "            if x.size > 0 and y.size > 0:  # 确保x和y不为空\n",
    "                color_index = get_color((x[-1], y[-1]))\n",
    "                ax.plot(x, y, color=colors[color_index], alpha=0.5)\n",
    "                ax.scatter(x[-1], y[-1], color=colors[color_index], alpha=1, s=10)\n",
    "    # 创建动画\n",
    "    ani = animation.FuncAnimation(fig, update, frames=len(ts), interval=100)\n",
    "    # 保存为GIF需要花费6分钟左右\n",
    "    ani.save(log_dir / 'brownian_bridge.gif', writer='imagemagick')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">torch.Size</span><span style=\"font-weight: bold\">([</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">910</span><span style=\"font-weight: bold\">])</span>\n",
       "<span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">torch.Size</span><span style=\"font-weight: bold\">([</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">900</span><span style=\"font-weight: bold\">])</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;35mtorch.Size\u001b[0m\u001b[1m(\u001b[0m\u001b[1m[\u001b[0m\u001b[1;36m910\u001b[0m\u001b[1m]\u001b[0m\u001b[1m)\u001b[0m\n",
       "\u001b[1;35mtorch.Size\u001b[0m\u001b[1m(\u001b[0m\u001b[1m[\u001b[0m\u001b[1;36m900\u001b[0m\u001b[1m]\u001b[0m\u001b[1m)\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">====================\n",
       "</pre>\n"
      ],
      "text/plain": [
       "====================\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">torch.Size</span><span style=\"font-weight: bold\">([</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1810</span><span style=\"font-weight: bold\">])</span>\n",
       "torch.float32 cpu\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;35mtorch.Size\u001b[0m\u001b[1m(\u001b[0m\u001b[1m[\u001b[0m\u001b[1;36m1810\u001b[0m\u001b[1m]\u001b[0m\u001b[1m)\u001b[0m\n",
       "torch.float32 cpu\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">torch.Size</span><span style=\"font-weight: bold\">([</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1810</span>, <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">12000</span>, <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">2</span><span style=\"font-weight: bold\">])</span>\n",
       "torch.float32 cpu\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;35mtorch.Size\u001b[0m\u001b[1m(\u001b[0m\u001b[1m[\u001b[0m\u001b[1;36m1810\u001b[0m, \u001b[1;36m12000\u001b[0m, \u001b[1;36m2\u001b[0m\u001b[1m]\u001b[0m\u001b[1m)\u001b[0m\n",
       "torch.float32 cpu\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">torch.Size</span><span style=\"font-weight: bold\">([</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1810</span>, <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">12000</span>, <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">2</span><span style=\"font-weight: bold\">])</span>\n",
       "torch.float32 cpu\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;35mtorch.Size\u001b[0m\u001b[1m(\u001b[0m\u001b[1m[\u001b[0m\u001b[1;36m1810\u001b[0m, \u001b[1;36m12000\u001b[0m, \u001b[1;36m2\u001b[0m\u001b[1m]\u001b[0m\u001b[1m)\u001b[0m\n",
       "torch.float32 cpu\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">torch.Size</span><span style=\"font-weight: bold\">([</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">12000</span>, <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">2</span><span style=\"font-weight: bold\">])</span>\n",
       "torch.float32 cpu\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;35mtorch.Size\u001b[0m\u001b[1m(\u001b[0m\u001b[1m[\u001b[0m\u001b[1;36m12000\u001b[0m, \u001b[1;36m2\u001b[0m\u001b[1m]\u001b[0m\u001b[1m)\u001b[0m\n",
       "torch.float32 cpu\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">torch.Size</span><span style=\"font-weight: bold\">([</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">12000</span>, <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">2</span><span style=\"font-weight: bold\">])</span>\n",
       "torch.float32 cpu\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;35mtorch.Size\u001b[0m\u001b[1m(\u001b[0m\u001b[1m[\u001b[0m\u001b[1;36m12000\u001b[0m, \u001b[1;36m2\u001b[0m\u001b[1m]\u001b[0m\u001b[1m)\u001b[0m\n",
       "torch.float32 cpu\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">====================\n",
       "</pre>\n"
      ],
      "text/plain": [
       "====================\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">torch.Size</span><span style=\"font-weight: bold\">([</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">21708000</span>, <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span>, <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">2</span><span style=\"font-weight: bold\">])</span>\n",
       "torch.float32 cpu\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;35mtorch.Size\u001b[0m\u001b[1m(\u001b[0m\u001b[1m[\u001b[0m\u001b[1;36m21708000\u001b[0m, \u001b[1;36m1\u001b[0m, \u001b[1;36m2\u001b[0m\u001b[1m]\u001b[0m\u001b[1m)\u001b[0m\n",
       "torch.float32 cpu\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">torch.Size</span><span style=\"font-weight: bold\">([</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">21708000</span>, <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span>, <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span><span style=\"font-weight: bold\">])</span>\n",
       "torch.float32 cpu\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;35mtorch.Size\u001b[0m\u001b[1m(\u001b[0m\u001b[1m[\u001b[0m\u001b[1;36m21708000\u001b[0m, \u001b[1;36m1\u001b[0m, \u001b[1;36m1\u001b[0m\u001b[1m]\u001b[0m\u001b[1m)\u001b[0m\n",
       "torch.float32 cpu\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">torch.Size</span><span style=\"font-weight: bold\">([</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">21708000</span>, <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span>, <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">2</span><span style=\"font-weight: bold\">])</span>\n",
       "torch.float32 cpu\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;35mtorch.Size\u001b[0m\u001b[1m(\u001b[0m\u001b[1m[\u001b[0m\u001b[1;36m21708000\u001b[0m, \u001b[1;36m1\u001b[0m, \u001b[1;36m2\u001b[0m\u001b[1m]\u001b[0m\u001b[1m)\u001b[0m\n",
       "torch.float32 cpu\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">torch.Size</span><span style=\"font-weight: bold\">([</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">21708000</span>, <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span>, <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">2</span><span style=\"font-weight: bold\">])</span>\n",
       "torch.float32 cpu\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;35mtorch.Size\u001b[0m\u001b[1m(\u001b[0m\u001b[1m[\u001b[0m\u001b[1;36m21708000\u001b[0m, \u001b[1;36m1\u001b[0m, \u001b[1;36m2\u001b[0m\u001b[1m]\u001b[0m\u001b[1m)\u001b[0m\n",
       "torch.float32 cpu\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "torch.Size([21708000, 1, 7])"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "num_samples = 12000\n",
    "ts, bridge, drift, source_sample, target_sample = gen_2d_data(num_samples=num_samples)\n",
    "print_debug(ts, bridge, drift, source_sample, target_sample)\n",
    "start = torch.cat(torch.split(source_sample.repeat(len(ts)-1, 1, 1), 1, dim=1), dim=0)\n",
    "times = ts[:len(ts)-1].repeat(num_samples, 1).reshape(-1, 1, 1)\n",
    "positions = torch.cat(torch.split(bridge[:-1, :], 1, dim=1), dim=0)\n",
    "\n",
    "scores = torch.cat(torch.split(drift[:-1, :], 1, dim=1), dim=0)\n",
    "print_debug(start, times, positions, scores)\n",
    "raw_data = torch.concat([start, times, positions, scores], dim=-1)\n",
    "raw_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">==========<span style=\"color: #808000; text-decoration-color: #808000\">model</span>==========\n",
       "</pre>\n"
      ],
      "text/plain": [
       "==========\u001b[33mmodel\u001b[0m==========\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">batch_szi<span style=\"color: #00ff00; text-decoration-color: #00ff00; font-weight: bold\">e:1000</span>00, channel:<span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span>, length:<span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">7</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "batch_szi\u001b[1;92me:1000\u001b[0m00, channel:\u001b[1;36m1\u001b[0m, length:\u001b[1;36m7\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "epochs = 50\n",
    "batch_size = 100000\n",
    "lr = 1e-4\n",
    "\n",
    "# 组合成data\n",
    "train_ds = BBdataset(raw_data)\n",
    "train_dl = DataLoader(train_ds, batch_size=batch_size, shuffle=True, num_workers=20)\n",
    "\n",
    "model = MLP(input_dim=5, output_dim=2, hidden_layers=6, hidden_dim=512).to(device)\n",
    "loss_fn = nn.MSELoss()\n",
    "loss_list = []\n",
    "print('='*10+'model'+'='*10)\n",
    "print(\"batch_szie:{}, channel:{}, length:{}\".format(*next(train_dl.__iter__()).shape)) # batch_szie, channel, length\n",
    "# print(model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ab725adce8634fc2a747083f4b45704a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training (lr: X)  (loss= X):   0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m/tmp/ipykernel_3503475/3225742757.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m     30\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdevice\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     31\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0me\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mepoch_iterator\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 32\u001b[0;31m     \u001b[0mnow_loss\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtrain\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m \u001b[0;34m,\u001b[0m\u001b[0mtrain_dl\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moptimizer\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mscheduler\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mloss_fn\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     33\u001b[0m     \u001b[0mloss_list\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnow_loss\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     34\u001b[0m     \u001b[0mcur_lr\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0moptimizer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mparam_groups\u001b[0m\u001b[0;34m[\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'lr'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/tmp/ipykernel_3503475/3225742757.py\u001b[0m in \u001b[0;36mtrain\u001b[0;34m(model, train_dl, optimizer, scheduler, loss_fn)\u001b[0m\n\u001b[1;32m     15\u001b[0m             \u001b[0mscheduler\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstep\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     16\u001b[0m         \u001b[0moptimizer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mzero_grad\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---> 17\u001b[0;31m         \u001b[0mlosses\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0mloss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mitem\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m/\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrain_dl\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     18\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     19\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0mlosses\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "def train(model, train_dl, optimizer, scheduler, loss_fn):\n",
    "    losses = 0\n",
    "    for training_data in train_dl:\n",
    "        training_data = training_data.to(device)\n",
    "        # x: (batch_size, channel, length-2), y: (batch_size, channel, 2)\n",
    "        x, y = training_data[:, :, :-2], training_data[:, :, -2:] # 不同模型要改\n",
    "        x = x.to(device)\n",
    "        y = y.to(device)\n",
    "\n",
    "        pred = model(x)\n",
    "        loss = loss_fn(pred, y)\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "        if scheduler is not None:\n",
    "            scheduler.step()\n",
    "        optimizer.zero_grad()\n",
    "        losses += loss.item() / len(train_dl)\n",
    "        \n",
    "    return losses\n",
    "from torch.optim.lr_scheduler import OneCycleLR\n",
    "\n",
    "train_dl = DataLoader(train_ds, batch_size=batch_size, shuffle=True, num_workers=20)\n",
    "optimizer = torch.optim.Adam(model.parameters(), lr=1e-6)\n",
    "epochs = 1000\n",
    "scheduler = OneCycleLR(optimizer, max_lr=1e-3, total_steps=epochs*len(train_dl))\n",
    "scheduler = None\n",
    "\n",
    "epoch_iterator = tqdm(range(epochs), desc=\"Training (lr: X)  (loss= X)\", dynamic_ncols=True)\n",
    "model.train()\n",
    "model.to(device)\n",
    "for e in epoch_iterator:\n",
    "    now_loss = train(model ,train_dl, optimizer, scheduler, loss_fn)\n",
    "    loss_list.append(now_loss)\n",
    "    cur_lr = optimizer.param_groups[-1]['lr']\n",
    "    epoch_iterator.set_description(\"Training (lr: %2.5f)  (loss=%2.5f)\" % (cur_lr, now_loss))\n",
    "epoch_iterator.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "torch.save(model.state_dict(), log_dir / 'model.pth')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(range(1, len(loss_list)+1), loss_list)\n",
    "plt.title('Loss')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">torch.Size</span><span style=\"font-weight: bold\">([</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">910</span><span style=\"font-weight: bold\">])</span>\n",
       "<span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\">torch.Size</span><span style=\"font-weight: bold\">([</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">900</span><span style=\"font-weight: bold\">])</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1;35mtorch.Size\u001b[0m\u001b[1m(\u001b[0m\u001b[1m[\u001b[0m\u001b[1;36m910\u001b[0m\u001b[1m]\u001b[0m\u001b[1m)\u001b[0m\n",
       "\u001b[1;35mtorch.Size\u001b[0m\u001b[1m(\u001b[0m\u001b[1m[\u001b[0m\u001b[1;36m900\u001b[0m\u001b[1m]\u001b[0m\u001b[1m)\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "test_num_samples = 4000\n",
    "test_ts, test_bridge, test_drift, test_source_sample, test_target_sample = gen_2d_data(num_samples=test_num_samples)\n",
    "# print_debug(test_ts, test_bridge, test_drift, test_source_sample, test_target_sample)\n",
    "pred_bridge = torch.zeros_like(test_bridge)\n",
    "pred_bridge[0, :] = test_source_sample\n",
    "model.eval()\n",
    "model.to(device)\n",
    "sigma=1\n",
    "with torch.no_grad():\n",
    "    for i in range(len(test_ts) - 1):\n",
    "        dt = (test_ts[i+1] - test_ts[i])\n",
    "        test_source_sample_reshaped = test_source_sample.reshape(-1, 1, test_source_sample.shape[-1])\n",
    "        test_ts_reshaped = test_ts[i].reshape(-1, 1, 1).repeat(test_num_samples, 1, 1)\n",
    "        pred_bridge_reshaped = pred_bridge[i].reshape(-1, 1, pred_bridge.shape[-1])\n",
    "        x = torch.concat([test_source_sample_reshaped, test_ts_reshaped, pred_bridge_reshaped], axis=-1)\n",
    "        dydt = model(x.to(device)).cpu()\n",
    "        # print_debug(dydt)\n",
    "        dydt = dydt[:, 0, :]\n",
    "        diffusion = sigma * torch.sqrt(dt) * torch.randn(test_num_samples, 2)\n",
    "        pred_bridge[i+1, :] = pred_bridge[i, :] + dydt * dt + diffusion\n",
    "        \n",
    "        \n",
    "        \n",
    "        # dt = ts[i+1] - ts[i]\n",
    "        # dydt = (y - bridge[i]) / (T - ts[i])\n",
    "        # drift[i, :] = dydt\n",
    "        # diffusion = sigma * torch.sqrt(dt) * torch.randn(num_samples, 2)\n",
    "        # bridge[i+1] = bridge[i] + dydt * dt + diffusion"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# draw_gaussian2d(bridge, False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# draw_gaussian2d(pred_bridge, False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAz0AAAHDCAYAAAAOb3Q1AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8+yak3AAAACXBIWXMAAA9hAAAPYQGoP6dpAACcuElEQVR4nOzdd3hc5Znw/+8p02ekUXWRuwEbTI1NM4QSIDSHbLKBzYZ3A2RDGim8KZvyXpuEzW5YdvMm5JdCyptNslnSSJZACGUhoZtmMMYGbGwwLpJVRzOaftrz++NIY8mSbMn2qPn+XJcSNHNmzj0jWc+553me+9aUUgohhBBCCCGEmKH0yQ5ACCGEEEIIIapJkh4hhBBCCCHEjCZJjxBCCCGEEGJGk6RHCCGEEEIIMaNJ0iOEEEIIIYSY0STpEUIIIYQQQsxokvQIIYQQQgghZjRJeoQQQgghhBAzmiQ9QgghhBBCiBlNkh4hxH6dd955nHfeeZMdhhBCzFiLFi3i2muvrXz/yCOPoGkajzzyyKTFNNX97Gc/Q9M03nzzzckORUwTkvSICbVx40be8573sHDhQsLhMC0tLVx00UV85zvfmezQqqqrq4tPfepTLF++nEgkQnNzM6eddhqf//znyeVykx2eEEIcsQYunge+wuEwxxxzDB//+Mfp6OiY7PDG5d577+WrX/3qmI/3PI///M//5PTTT6e+vp5EIsExxxzD+9//fp5++unqBSrEJDAnOwBx5Fi7di3nn38+CxYs4Prrr2f27Nns2rWLp59+mm9/+9t84hOfmOwQqyKVSrFq1Sr6+vr4wAc+wPLly+np6eGll17itttu46Mf/SjxeHyywxRCiCPaP/3TP7F48WJKpRJPPPEEt912G/feey+bNm0iGo1OaCznnHMOxWKRYDA4rsfde++9fO973xtz4vPJT36S733ve7zzne/k6quvxjRNtmzZwn333ceSJUs444wzDiJ6IaYmSXrEhPmXf/kXamtree6550gmk0Pu6+zsnPB48vk8sVis6uf5yU9+ws6dO3nyySdZvXr1kPv6+vrGPagJIYQ4/C699FJWrVoFwAc/+EEaGhr45je/yV133cXf/u3fjviYao0juq4TDocP+/MO1tHRwfe//32uv/56fvSjHw2579Zbb6Wrq6uq5xdiosnyNjFhXn/9dVasWDEs4QFobm4e8r3jOHzta19j6dKlhEIhFi1axJe+9CXK5fKQ4zRNG/ETrX3XRw8sX3j00Uf52Mc+RnNzM/Pmzavcf99993HuueeSSCSoqanh1FNP5Ze//OWQ53zmmWe45JJLqK2tJRqNcu655/Lkk0+O6XUbhjHiJ2Y1NTVDBrbHH3+cK6+8kgULFhAKhZg/fz7/+3//b4rF4pDHXXvttcTjcXbu3MmaNWuIx+O0tLTwve99D/CXEb7tbW8jFouxcOHCYa9l4P147LHH+PCHP0xDQwM1NTW8//3vp7e394CvqVwu85WvfIWjjjqqEuc//MM/DPv5CCHEdPW2t70NgO3btwN7/+6+/vrrXHbZZSQSCa6++mrAXyZ26623smLFCsLhMLNmzeLDH/7wsL+nSin++Z//mXnz5hGNRjn//PN5+eWXh517tD09zzzzDJdddhl1dXXEYjFOPPFEvv3tb1fiGxgDBi/XG8327dtRSnHWWWcNu0/TtCHjciqV4rOf/SwnnHAC8XicmpoaLr30UjZs2DBi3L/97W+56aabaGlpIZFI8J73vIdMJkO5XObGG2+kubmZeDzOddddN+K4/vGPf5zbb7+dZcuWEQ6HWblyJY899tior2Ww++67j7e+9a3EYjESiQSXX375iO+xOPLITI+YMAsXLuSpp55i06ZNHH/88fs99oMf/CA///nPec973sNnPvMZnnnmGW6++WZeffVV7rzzzoOO4WMf+xhNTU18+ctfJp/PA34C8IEPfIAVK1bwxS9+kWQyyfr167n//vt53/veB8Bf/vIXLr30UlauXMlXvvIVdF3npz/9KW9729t4/PHHOe200/b7ul3X5Re/+AXXXHPNfuO74447KBQKfPSjH6WhoYFnn32W73znO+zevZs77rhjyLGu63LppZdyzjnn8G//9m/cfvvtfPzjHycWi/F//s//4eqrr+bd7343P/jBD3j/+9/PmWeeyeLFi4c8x8c//nGSySRf/epX2bJlC7fddhs7duyoDFwj8TyPK664gieeeIIPfehDHHvssWzcuJFvfetbvPbaa/zhD3840I9BCCGmvNdffx2AhoaGym2O43DxxRdz9tln841vfKOy7O3DH/4wP/vZz7juuuv45Cc/yfbt2/nud7/L+vXrefLJJwkEAgB8+ctf5p//+Z+57LLLuOyyy3jhhRd4+9vfjmVZB4znwQcfZM2aNcyZM4dPfepTzJ49m1dffZV77rmHT33qU3z4wx+mra2NBx98kF/84hcHfL6FCxcC/rhz5ZVX7ncJ3xtvvMEf/vAHrrzyShYvXkxHRwc//OEPOffcc3nllVeYO3fukONvvvlmIpEIX/jCF9i2bRvf+c53CAQC6LpOb28vX/3qV3n66af52c9+xuLFi/nyl7885PGPPvoov/nNb/jkJz9JKBTi+9//PpdccgnPPvvsfq8fBsbZiy++mFtuuYVCocBtt93G2Wefzfr161m0aNEB3xcxgykhJsj//M//KMMwlGEY6swzz1T/8A//oB544AFlWdaQ41588UUFqA9+8INDbv/sZz+rAPWXv/ylchugvvKVrww718KFC9U111xT+f6nP/2pAtTZZ5+tHMep3J5Op1UikVCnn366KhaLQ57D87zK/x999NHq4osvrtymlFKFQkEtXrxYXXTRRft93e3t7aqpqUkBavny5eojH/mI+uUvf6nS6fSwYwuFwrDbbr75ZqVpmtqxY0fltmuuuUYB6utf/3rltt7eXhWJRJSmaerXv/515fbNmzcPe58G3o+VK1cOef//7d/+TQHqrrvuqtx27rnnqnPPPbfy/S9+8Qul67p6/PHHh8T5gx/8QAHqySef3O/7IYQQU8nA38OHHnpIdXV1qV27dqlf//rXqqGhQUUiEbV7926l1N6/u1/4wheGPP7xxx9XgLr99tuH3H7//fcPub2zs1MFg0F1+eWXDxlLvvSlLylgyJj18MMPK0A9/PDDSimlHMdRixcvVgsXLlS9vb1DzjP4uW644QY1nku797///QpQdXV16l3vepf6xje+oV599dVhx5VKJeW67pDbtm/frkKhkPqnf/qnYXEff/zxQ8aWv/3bv1WapqlLL710yHOceeaZauHChUNuAxSg1q1bV7ltx44dKhwOq3e9612V2wZ+btu3b1dKKZXNZlUymVTXX3/9kOdrb29XtbW1w24XRx5Z3iYmzEUXXcRTTz3FFVdcwYYNG/i3f/s3Lr74YlpaWrj77rsrx917770AfPrTnx7y+M985jMA/OlPfzroGK6//noMw6h8/+CDD5LNZvnCF74wbP30wEzHiy++yNatW3nf+95HT08P3d3ddHd3k8/nueCCC3jsscfwPG/Uc86aNYsNGzbwkY98hN7eXn7wgx/wvve9j+bmZr72ta+hlKocG4lEKv+dz+fp7u5m9erVKKVYv379sOf+4Ac/WPnvZDLJsmXLiMViXHXVVZXbly1bRjKZ5I033hj2+A996EOVTyABPvrRj2KaZuVnMJI77riDY489luXLl1fei+7u7spSkIcffnjUxwohxFR14YUX0tTUxPz583nve99LPB7nzjvvpKWlZchxH/3oR4d8f8cdd1BbW8tFF1005G/iypUricfjlb+JDz30EJZl8YlPfGLITPqNN954wNjWr1/P9u3bufHGG4ctEd/fErYD+elPf8p3v/tdFi9ezJ133slnP/tZjj32WC644AJaW1srx4VCIXTdv2R0XZeenh7i8TjLli3jhRdeGPa873//+4eMLaeffjpKKT7wgQ8MOe70009n165dOI4z5PYzzzyTlStXVr5fsGAB73znO3nggQdwXXfE1/Lggw+STqf527/92yE/B8MwOP3002VsErK8TUysU089lf/+7//Gsiw2bNjAnXfeybe+9S3e85738OKLL3LcccexY8cOdF3nqKOOGvLY2bNnk0wm2bFjx0Gff9/lXQPLF/Y3Xb5161aA/S5Ny2Qy1NXVjXr/nDlzuO222/j+97/P1q1beeCBB7jlllv48pe/zJw5cyrJy86dO/nyl7/M3XffPWwteCaTGfJ9OBymqalpyG21tbXMmzdv2CBYW1s74l6do48+esj38XicOXPm7LfvwdatW3n11VeHnXvAZBSlEEKIQ/W9732PY445BtM0mTVrFsuWLatc6A8wTXPIflDw/yZmMplhe1MHDPxNHBi79v2729TUtN/xA8Y2Vh0MXde54YYbuOGGG+jp6eHJJ5/kBz/4Affddx/vfe97efzxxwF/WfO3v/1tvv/977N9+/Yhicfg5X8DFixYMOT72tpaAObPnz/sds/zyGQyQ55n3/cI4JhjjqFQKNDV1cXs2bOH3T8wVg98ALevmpqaEW8XRw5JesSkCAaDnHrqqZx66qkcc8wxXHfdddxxxx185StfqRxzKJ9ejfZJ0OCZlLEamMX593//d04++eQRjxlryWlN0zjmmGM45phjuPzyyzn66KO5/fbb+eAHP4jrulx00UWkUik+//nPs3z5cmKxGK2trVx77bXDZpMGz1iN5fbBM0qHwvM8TjjhBL75zW+OeP++g5oQQkwHp512WqV622gGz3gM8DyP5uZmbr/99hEfM9oHRFNNQ0MDV1xxBVdccQXnnXcejz76KDt27GDhwoV8/etf5x//8R/5wAc+wNe+9jXq6+vRdZ0bb7xxxJUOkzE+DcTxi1/8YsSkyDTlkvdIJ78BYtINDDJ79uwB/M2VnuexdetWjj322MpxHR0dpNPpyuZLgLq6OtLp9JDnsyyr8lwHsnTpUgA2bdo0bGZp32Nqamq48MILx/aixmDJkiXU1dVVYt24cSOvvfYaP//5z3n/+99fOe7BBx88bOfc19atWzn//PMr3+dyOfbs2cNll1026mOWLl3Khg0buOCCCw4pMRVCiJlg6dKlPPTQQ5x11ln7/WBtYOzaunUrS5Ysqdze1dV1wKqZg8eq/Y1Dh+tv8qpVq3j00UfZs2cPCxcu5He/+x3nn38+P/nJT4Ycl06naWxsPCznHGxg1maw1157jWg0OmoSOfAeNTc3H9axWswcsqdHTJiHH354xE9zBvaPLFu2DKBywX3rrbcOOW5gZuHyyy+v3LZ06dJhZSx/9KMfjTrTs6+3v/3tJBIJbr75Zkql0pD7BmJduXIlS5cu5Rvf+Aa5XG7Ycxyol8EzzzxTqRQ32LPPPktPT0/ldQ98Ajb4PVJKVcqRVsOPfvQjbNuufH/bbbfhOA6XXnrpqI+56qqraG1t5cc//vGw+4rF4oivVQghZqqrrroK13X52te+Nuw+x3EqH8xdeOGFBAIBvvOd7wz5O7/vWDeSt7zlLSxevJhbb7112Ad9g59roGfQvseMpL29nVdeeWXY7ZZl8ec//3nIMnPDMIaN33fccceQfT+H01NPPTVkr9CuXbu46667ePvb3z7qbNHFF19MTU0NX//614eMawOk75CQmR4xYT7xiU9QKBR417vexfLly7Esi7Vr1/Kb3/yGRYsWcd111wFw0kkncc011/CjH/2IdDrNueeey7PPPsvPf/5z/uqv/mrIzMQHP/hBPvKRj/DXf/3XXHTRRWzYsIEHHnhgzJ881dTU8K1vfYsPfvCDnHrqqbzvfe+jrq6ODRs2UCgU+PnPf46u6/y///f/uPTSS1mxYgXXXXcdLS0ttLa28vDDD1NTU8Mf//jHUc/xi1/8gttvv513vetdrFy5kmAwyKuvvsp//Md/EA6H+dKXvgTA8uXLWbp0KZ/97GdpbW2lpqaG3//+92Pqm3OwLMviggsu4KqrrmLLli18//vf5+yzz+aKK64Y9TF/93d/x29/+1s+8pGP8PDDD3PWWWfhui6bN2/mt7/9LQ888MABl4gIIcRMce655/LhD3+Ym2++mRdffJG3v/3tBAIBtm7dyh133MG3v/1t3vOe99DU1MRnP/tZbr75ZtasWcNll13G+vXrue+++w44Zum6zm233cY73vEOTj75ZK677jrmzJnD5s2befnll3nggQcAKpv/P/nJT3LxxRdjGAbvfe97R3zO3bt3c9ppp/G2t72NCy64gNmzZ9PZ2cmvfvUrNmzYwI033liJa82aNfzTP/0T1113HatXr2bjxo3cfvvtQ2asDqfjjz+eiy++eEjJaoCbbrpp1MfU1NRw22238Xd/93e85S1v4b3vfS9NTU3s3LmTP/3pT5x11ll897vfrUq8YpqYnKJx4kh03333qQ984ANq+fLlKh6Pq2AwqI466ij1iU98QnV0dAw51rZtddNNN6nFixerQCCg5s+fr774xS+qUqk05DjXddXnP/951djYqKLRqLr44ovVtm3bRi1Z/dxzz40Y2913361Wr16tIpGIqqmpUaeddpr61a9+NeSY9evXq3e/+92qoaFBhUIhtXDhQnXVVVepP//5z/t93S+99JL63Oc+p97ylreo+vp6ZZqmmjNnjrryyivVCy+8MOTYV155RV144YUqHo+rxsZGdf3116sNGzYoQP30pz+tHHfNNdeoWCw27FznnnuuWrFixbDbFy5cqC6//PJh78ejjz6qPvShD6m6ujoVj8fV1VdfrXp6eoY95+CS1UopZVmWuuWWW9SKFStUKBRSdXV1auXKleqmm25SmUxmv++HEEJMJQcaHwaM9nd3wI9+9CO1cuVKFYlEVCKRUCeccIL6h3/4B9XW1lY5xnVdddNNN6k5c+aoSCSizjvvPLVp06ZhY9a+JasHPPHEE+qiiy5SiURCxWIxdeKJJ6rvfOc7lfsdx1Gf+MQnVFNTk9I0bb/lq/v6+tS3v/1tdfHFF6t58+apQCCgEomEOvPMM9WPf/zjIaWwS6WS+sxnPlOJ+6yzzlJPPfXUsPFhIO477rhjyLlGe4+/8pWvKEB1dXVVbgPUDTfcoP7rv/5LHX300SoUCqlTTjll2Huxb8nqwTFcfPHFqra2VoXDYbV06VJ17bXXDimBLY5MmlKHaXezEGLaGGii99xzz8msjBBCiClD0zRuuOEGmZURh53s6RFCCCGEEELMaJL0CCGEEEIIIWY0SXqEEEIIIYQQM1pVkx7XdfnHf/xHFi9eTCQSYenSpXzta187bE0ShRAH59prr0UpJft5xBFHxiUhpjallOznEVVR1ZLVt9xyC7fddhs///nPWbFiBevWreO6666jtraWT37yk9U8tRBCCDGMjEtCCHFkqmr1tjVr1jBr1qwhHXz/+q//mkgkwn/9139V67RCCCHEiGRcEkKII1NVZ3pWr17Nj370I1577TWOOeYYNmzYwBNPPME3v/nNEY8vl8uUy+XK957nkUqlaGhoQNO0aoYqhBBiEKUU2WyWuXPnouszZ/vneMclkLFJCCGmikMam6rZBGigcaSmaco0TaVpmvr6178+6vEDTarkS77kS77ka2p87dq1q5rDxIQb77iklIxN8iVf8iVfU+3rYMamqi5v+/Wvf83nPvc5/v3f/50VK1bw4osvcuONN/LNb36Ta665Ztjx+36alslkWLBgAdu3byeZTFYrzMPG8zy6u7tpbGycFp+MSrzVJfFWl8RbXel0msWLF5NOp6mtrZ3scA6b8Y5LIGPTRJN4q0virS6Jt7oOZWyq6vK2z33uc3zhC1/gve99LwAnnHACO3bs4Oabbx5xcAmFQoRCoWG3J5PJaTOwWJZFMpmcFr84Em91SbzVJfFOjJm2fGu84xLI2DTRJN7qknirS+KdGAczNlX11RUKhWFvoGEYeJ5XzdMKIYQQI5JxSQghjkxVnel5xzvewb/8y7+wYMECVqxYwfr16/nmN7/JBz7wgWqeVgghhBiRjEtCCHFkqmrS853vfId//Md/5GMf+xidnZ3MnTuXD3/4w3z5y1+u5mmFEEKIEcm4JIQQR6aqJj2JRIJbb72VW2+9tZqnEUIIIcZExiUhhDgyTZ8dS0IIIYQQQghxECTpEUIIIYQQQsxokvQIIYQQQgghZjRJeoQQQgghhBAzmiQ9QgghhBBCiBlNkh4hhBBCCCHEjCZJjxBCCCGEEGJGk6RHCCGEEEIIMaNJ0iOEEEIIIYSY0STpEUIIIYQQQsxokvQIIYQQQgghZjRJeoQQQgghhBAzmiQ9QgghhBBCiBlNkh4hhBBCCCHEjCZJjxBCCCGEEGJGk6RHCCGEEEIIMaNJ0iOEEEIIIYSY0STpEUIIIYQQQsxokvQIIYQQQgghZjRJeoQQQgghhBAzmiQ9QgghhBBCiBlNkh4hhBBCCCHEjCZJjxBCCCGEEGJGk6RHCCGEEEIIMaNJ0iOEEEIIIYSY0STpEUIIIYQQQsxokvQIIYQQQgghZjRJeoQQQgghhBAzmiQ9QgghhBBCiBlNkh4hhBBCCCHEjCZJjxBCCCGEEGJGk6RHCCGEEEIIMaNJ0iOEEEIIIYSY0STpEUIIIYQQQsxokvQIIYQQQgghZjRJeoQQQgghhBAzmiQ9QgghhBBCiBmt6klPa2sr/+t//S8aGhqIRCKccMIJrFu3rtqnFUIIIUYk45IQQhx5zGo+eW9vL2eddRbnn38+9913H01NTWzdupW6urpqnlYIIYQYkYxLQghxZKpq0nPLLbcwf/58fvrTn1ZuW7x4cTVPKYQQQoxKxiUhhDgyVTXpufvuu7n44ou58sorefTRR2lpaeFjH/sY119//YjHl8tlyuVy5fu+vj4APM/D87xqhnpYeJ6HUmpaxAoSb7VJvNUl8VbXdIlzvMY7LoGMTRNN4q0uibe6JN7qOpQ4q5r0vPHGG9x22218+tOf5ktf+hLPPfccn/zkJwkGg1xzzTXDjr/55pu56aabht3e1dWFZVnVDPWw8DyPTCaDUgpdn/o1IiTe6pJ4q0vira5MJjPZIVTFeMclkLFpokm81SXxVpfEW12HMjZpSil1GGMZIhgMsmrVKtauXVu57ZOf/CTPPfccTz311LDjR/o0bf78+fT09JBMJqsV5mHjeR5dXV00NTVNi18cibe6JN7qknirK51O09DQQCaToaamZrLDOWzGOy6BjE0TTeKtLom3uiTe6jqUsamqMz1z5szhuOOOG3Lbsccey+9///sRjw+FQoRCoWG367o+LX4QAJqmSbxVJPFWl8RbXdMp3ukQ48EY77gEMjZNBom3uiTe6pJ4q+dQYqzqqzvrrLPYsmXLkNtee+01Fi5cWM3TCiGEECOScUkIIY5MVU16/vf//t88/fTTfP3rX2fbtm388pe/5Ec/+hE33HBDNU8rhBBCjEjGJSGEODJVNek59dRTufPOO/nVr37F8ccfz9e+9jVuvfVWrr766mqeVgghhBiRjEtCCHFkquqeHoA1a9awZs2aap9GCCGEGBMZl4QQ4sgz9XcsCSGEEEIIIcQhkKRHCCGEEEIIMaNJ0iOEEEIIIYSY0STpEUIIIYQQQsxokvQIIYQQQgghZjRJeoQQQgghhBAzmiQ9QgghhBBCiBlNkh4hhBBCCCHEjCZJjxBCCCGEEGJGk6RHCCGEEEIIMaNJ0iOEEEIIIYSY0STpEUIIIYQQQsxokvQIIYQQQgghZjRJeoQQQgghhBAzmiQ9QgghhBBCiBlNkh4hhBBCCCHEjCZJjxBCCCGEEGJGk6RHCCGEEEIIMaNJ0iOEEEIIIYSY0STpEUIIIYQQQsxokvQIIYQQQgghZjRJeoQQQgghhBAzmiQ9QgghhBBCiBlNkh4hhBBCCCHEjCZJjxBCCCGEEGJGk6RHCCGEEEIIMaNJ0iOEEEIIIYSY0STpEUIIIYQQQsxokvQIIYQQQgghZjRJeoQQQgghhBAzmiQ9QgghhBBCiBlNkh4hhBBCCCHEjCZJjxBCCCGEEGJGk6RHCCGEEEIIMaNJ0iOEEEIIIYSY0STpEUIIIYQQQsxokvQIIYQQQgghZrQJS3r+9V//FU3TuPHGGyfqlEIIIcR+ydgkhBBHhglJep577jl++MMfcuKJJ07E6YQQQogDkrFJCCGOHFVPenK5HFdffTU//vGPqaurq/bphBBCiAOSsUkIIY4sVU96brjhBi6//HIuvPDCap9KCCGEGBMZm4QQ4shiVvPJf/3rX/PCCy/w3HPPjen4crlMuVyufN/X1weA53l4nleVGA8nz/NQSk2LWEHirTaJt7ok3uqaLnEeDBmbpjaJt7ok3uqSeKvrUOKsWtKza9cuPvWpT/Hggw8SDofH9Jibb76Zm266adjtXV1dWJZ1uEM87DzPI5PJoJRC16d+YTyJt7ok3uqSeKsrk8lMdghVIWPT1P/dk3irS+KtLom3ug5lbNKUUuowxlLxhz/8gXe9610YhlG5zXVdNE1D13XK5fKQ+2DkT9Pmz59PT08PyWSyGmEeVp7n0dXVRVNT07T4xZF4q0virS6Jt7rS6TQNDQ1kMhlqamomO5zDRsamqf+7J/FWl8RbXRJvdR3K2FS1mZ4LLriAjRs3DrntuuuuY/ny5Xz+858fNqgAhEIhQqHQsNt1XZ8WPwigMnBKvNUh8VaXxFtd0yne6RDjwZCxSeKtBom3uiTe6ppO8R5KjFVLehKJBMcff/yQ22KxGA0NDcNuF0IIISaCjE1CCHFkmvopnRBCCCGEEEIcgqpWb9vXI488MpGnE0IIIQ5IxiYhhJj5ZKZHCCGEEEIIMaNJ0iOEEEIIIYSY0SZ0eZsQAqxUGiebx0zECNYnJzscIYQQQogZT5IeISZQeuNm0us24JYsjHCQ5KqTSJ6wfLLDEkIIIYSY0WR5mxATxEqlSa/bAEoRaqwDpUiv24CVSk92aEIIIYQQM5okPUIcRlYqTWFH64iJjJPN45YszEQcTdcxE3HckoWTzU98oEIIIYQQRxBZ3ibEYbLv0rXalSdCc13lfjMRwwgHcbI5zEQcJ5vDCAcxE7FJjFoIIYQQYuaTmR4hDoMRl649/xJ2Nlc5JlifJLnqJNA0yt29oGnUrTpJihkIIYQQQlSZzPQIcRgMLF0LNdZVlq6VunvRi+UhxyVPWE60ZbZUbxNCCCGEmECS9AhxGIy2dE2PhIYdG6xPSrIjhBBCCDGBZHmbEIfBSEvXkitPJJCIH/Cx+yt+IIQQQgghDp3M9AhxmOy7dM1M1tDZ2bnfx0jfHiGEEEKI6pOZHiEOo2B9kujCljEtX5O+PUIIIYQQE0OSHiEmifTtEUIIIYSYGJL0CDFJBhc/UJ4nfXuEEEIIIapEkh4hqsBKpSnsbBvSp2df0rdHCCGEEGJiSCEDIQ6zgeIETskinwgTL9rUnXjsiMdK3x4hhBBCiOqTmR4hDqOhxQmSfnGC51/ab3GC8RQ/EEIIIYQQ4ydJjxCH0b7FCYxIZEKLE0jPHyGEEEKI4WR5mxD9rFT6kJeZDS5OYCRiuKXihBUnkJ4/QgghhBAjk6RHCA5fwjBQnCC9bgPl7jQkwiRXnlD1pWv79vxxsjnS6zYQbZkty+aEEEIIccSTpEcc8Q5HwjB4lmigOIHVl6PXLlG7dEl1XwB7l9WFGusqPX/K3b042XzlNRyOmSwhhBBCiOlIkh5xxBm4+PdcF90wsPuyB0wY9me0WSIzWUOus7P6L4ihy+rMRHxYzx9Z+iaEEEKII5kkPeKIMnDxX06lcfNFjGiEQCKGZ9mjJgyDDZ4tASi2tpNauw49YA6bJTKTNf5jejN4uUJVZ1iGLqvrxQgHKz1/ZOmbEEIIIY50kvSII8bAxb9r2XhlC5TCsyxQUUDhFEpYmRxmNEzD6lXDEoLBsyWeZQEaKA8r3Ud4TjOBfWaJzGQNuTd3U3z1dbyDmGEZ73K00Xr+jGXpmxBCCCHETCZJjzhiDFz8G5EwuB56JIwqW+ihIHaugI6GpgFoqH0eO3i2xEzEyL/Rg/I8Qs0NAJQ7ujFjUZRlVWaJrN4M+W3bqTmIGZaDXY42ONEZ+P5AS9+EEEIIIWY6SXrEEWPg4t8tl8HQ8YolNMOgnMrg5vOYs5sJ1SeHJCfgJxCD9/24hSLKdfEsG6urBwDPtil3pQgmE5VlZbk3d+NZDmYijmfboGnY2cKIMyz7LptLrX0e5TiYtTUoq3zIydJoS9+EEEIIIY4EkvSII0pk4XzyW99ADwVxSmVc20az/KVuaNqQ5V+9L75Ccccu3JKFput4loWTzaHwkxwALRSCsoVmGjSccQqJZUsryYSZiKEHTYqtHbiFIp5lo+k6hbZ2ogtbKjHtm6hooRDFPR1ouo7dlyXQUIcqW+S2vkn86EVDkpXByZKV7qPn8WfB0IfNLI229O1gSSU4IYQQQkwnkvSIKW3wxfVAYYCDMTix0HSdxLKlFN7cjaaBHglT3N1Oub0LMxpFWWU0XfeTo0EFCtyShmc7OIUS6DpGwATbQQuY6MEgkblDZ2KCdbVE5s3BfqMVHBfN1DHjUXKbt1HTnxzl3thJz+PPoFwPMx7FyRUo72gFDTTDQHkepd170AydnmfW0/fKlsrszdA9RjaeVcbpy6MFTPA8Qo31Q/buDHwdKqkEJ4QQQojpRpIeMWXte3Fdu/JEaK4b9/PsW72snEqT27odgOi8Of4sju1Q2tNJqb0TMxomumAehV1tmIl4ZfbHLds0rF4FQM/a50F56KEgXv++oJH2yITq6/DCQbySBYBbKlPuT+QKre30PP4sxdYOFArdMNBME2U7hGY14uYKeLaDZzsE47VE5jRVZm/MWHTQHqM4+Td24Nk2ynXBsSns2oNbLBGsqz2se3fGWglOZoKEEEIIMZVI0iOmpBEvrp9/CfPsldDcPK7nGly9zEr3YfX04hbLaLpGQdeJzp2FETD95WhmAOV6FFvbK8vZBm/+j/Rf3Cv8fTd2JotmmoTnN1JsbcdK96EbBmYihqcU5VQvTqGErmnokTBesYTruJR7M2TWb0R5HkopUAqleSjbQSkPTdeJLJhLubMHz3UJz52FputowSDlrhSFN3dXXpOd7sNzHJTjVmZ5lOtiZ7LUn7nysCYdY6kEJzNBQgghhJhqJOkRU9JIF9el7l70YnnczzVQwKCcSmP19KJsByMSQg8GcTJZSrqOHgygmwECiWglyRlYzjb65n+/zLVbLFLq6CL11PMoBYHaOEYohEIjFzYJuC7KMFBlq7IUTpUt3JKFGY+hB0w810W5HkYkiBYOoum6P1MSi6IZOsqyKOULlDu6UZ5H3yuvoVyPQlsHTl8OZdmgFJppoOsBlKcw4jGic2cftp/J4PdytEpw0hNICCGEEFORJD1iShrt4lqPhMb9XAONO3sefxavVEYPhwg1NRBM1lDc00X9aadgRMN0P/EcWjCEWyihBUPoQX85W6AmMWSZ1uALezwX5bigPAbqXNupDE7ARAsEMBpa0AMBtIHy1kqhh4IEm+orleT0UBAsGy2kE6hLEohFaDz3zMqMUaG/AWppT6c/AzR3FrppUOpO4eYKoGn9y+zKeMUyeixKoCZGoDZx2MtS768JKkhPICGEEEJMTZL0iClppIvr5MoTKSfiB/V8yROWY8aidNz/MJquE0zW4GRzBBJR4kcvAsCzbPJv7Kg8xoxH/f/fZ19Kpd9POOjvodEAb9DJlELZDhgmmqYTmt1Eua0Dp1AimEyQWH4UumEQX34Uuc3b0ENBlONiRCIEYhHqVp1EfMmCIe+FBnT++QmCjQ2YkRCF9k5/hsfzMMJhgo11KE9R7uxGD4cI1CaqVpZ6f5XgpCeQEEIIIaYiSXrElLXvxbWZrKGzs/Ogny++ZAHOW08fcZbCSqWpTNUoz1/WlrLpfPgpAonokH0pAxf2XtlCMwzw9m1lSmUWCF3HMA0ic2bRsHoldi5P5qXNpJ5ZjxGJUHPicprnzsZz3crMzkgFAYxYlEBtAmWVKaT7KLW195fZ1vFsGzvdR7ChjmjLbBpWr6rsPaqW0SrBHWgmSAghhBBiMkjSI6a0wRfXnuft/+AxGEikiq3tAEQGNSDVg0EC9Unsnl6U6wEKz3FAqSHNSout7QSbGynt3oMeDqMVyyjHGX4yTcPpy+Jh0Lh6JUYsSvv9j+DkC/09eHI4T+UJXHTOkDgGiiH0bX2D/Gtv4DkegUSUYFMDpdZ2rM4uUGBEoyjPb5LqFkvgejS89TRqJ7lowOHuCSSEEEIIcagk6RFHnEJr+7DqYtGW2Wi6ht2VxnNd6E+wrO4UeiiArhv0rF1H36tbcbIF0MCIhKk5fhmReXPZ86eHcLN5/wTKn/nxyhZOuQyhCPkdu8m/vgMr1QuajhmL4NoO5a4U7Q88ih4wAQ3PsXGyeZRSKMtCM02MSNifNQJqTjwWJ5f3K8IZOoSC4HoE6mqZdcl5Q5bFjWYiykkfrp5AQgghDq89bT3s3t2B6xhomkaqJ0t9Q4K5LY2THZoQVVXVpOfmm2/mv//7v9m8eTORSITVq1dzyy23sGzZsmqeVohR7a+6WOzoJRTbOlC27e/T0XRAYXX2YMQipJ7rwLP6Z3SUwrNsep/bAGgkT1pB6pn1/mN1/3EowHFxdYf0C5swYlHQNPyqb8XKsjgjHMJKpf3y1Z5ftnog6VKOn4A5hSLoBoF4f6Ji5nDzBbx8Ac00aTjjLSMmPPsmOFJOWhzpZFwSM0lba/eQpGXf7weOeerJV1j7xCa2v9GBZdksXzGbndt7yRfKhENB6urjnHjyUk44cRHLls+XBEjMSFVNeh599FFuuOEGTj31VBzH4Utf+hJvf/vbeeWVV4jFZGOzmDgDF/92X3ZYdbHini5yW98ktmgefRvrKHf3+E1CLQulFJ7nobseylOga+AMFC9QeKUyPU+tI9zc6FdjUx5K08DyEyfPdvC8EspxCPRv7PdsG1wXFGihIHooiKaB63rgOP2JUT/PQwHKstFNnUjLbBSQXrcBR9fRkga1p6yg4cyVw17zvgnOQOGEsZSTluaiYqaScUnMFA8/tJ577n6afK5ELB5m4aJZbNm8i449vXieYs7cevL5Eq+9thvP3bv3VNM12tuC7NrV449rQOvubja99CaaBvFEhNVnreDsc45nztwGmQUSM0ZVk577779/yPc/+9nPaG5u5vnnn+ecc86p5qmFqBh88a/pGp5lV6qLFVvbsftypJ5djx4MYtbEsNNplFIYkTB60E9KPNvBM8r+zAtUah4A4LiU2rv8RMgdvO9IA13DyebRdB2lPMJzminsavPv0/ofUrbwPA9N+QkOytub+CiFWyihBwMkTzm+smzsQHtmrFSa1NrnUY6DWVuDsspk1r+M57hE5jTtt5x0ZtMWMs+/JLNBYkaScUnMBG2t3dxz99MopViwqJndu7r4w++fpFy2/Q/QgLa2nnE/r1KQ7SvywH3r+J8HnieRiNDUVMvbLjqFv//QZYf7ZQgxoSZ0T08mkwGgvr5+Ik8rjgCjzUyMtJzNLZXxbJfini7svhyB2gRaKEipvcsvAR2NoGmgh8IEElGMRJzsK1tRjldZdjbMwO26PvQYBcpTJI5dipvNYWfz6IaOWVuDEYtgdfdS7ujytwEp5T9W08DQKzNKmqahB0xKXSkyGzcDfgGG6MKWUd+P3hdfobino79gQpZAQx3KddFNfb/lpO1sjuLzL6FJc1FxhJBxSUxHqZ4sqZ4sDU01FIsWuq5TKIy/eff+KE/RlymQ7Suw480OOtp7+dKXrz6s5xBiIk1Y0uN5HjfeeCNnnXUWxx9//IjHlMtlyuW9/2j7+voqjz0clbuqzfO8ynKo6WA6xWv1ZrD6clh2Ca9x6DR7ZtMW0oNnJlaeSO3x/vp8qy+HU7IINSb9pWnBEErliC6Zj7Id+ra8TqChnuKuVjB0NA3MWASloObYownUJ8m8uAmzJo5TKOIUFNgjVGobEDD9+z0PpYFSHp7jUOrqQTcNf88OGkZNnHBjHXokTGH7LoxoGOW6uPmi38A0GATT/7l4to3Xl6PnyefoWbsOPRwiWJek6fwzK69z3/eq75XXUEqBpuF5HqX2LsKzmqg5YRm5La9TbO9CMwxqTz4OM1lT+TfmFkrYuQKBRBzXdjASMcrdaay+HGay5rD9PA+H6fT7C9Mz3pluLOMSyNg00STe4fa09VT26syZ28CWzTtpa+vm9W1toIFtu2i6duAnwv8gTev/QA19bOd3PcX9f3qOSy47lZPfctQhvJLxk9+H6pqO8R6sCUt6brjhBjZt2sQTTzwx6jE333wzN91007Dbu7q6sCyrmuEdFp7nkclkUEqh62P8SzKJpku8uTd3k9+2HadYphQ2Kba2U3vs0QAU27tIv7ARdI1AUxK3WKTvpU3URwIEEnFsu0Q+ESZfKuJaNk6mD8/1SG96BT0YxNM89EwvXiSAphtoQCkUwOnLkt32OlrAwC2WCTQlQSVxikXsnl5/NkaN0J9H0/3laQCahhWPoKEoFvNogQBGTQRXVxTyWULRIE65hFMTQwsY4HqokOk3PNUNdNPw9/8QHHouw0B3LQovbKCh/3UOln7pFfqsEtRGUa6HpuugaRiL5+IumYdtlci/sQPPdshufo1eq0TN8qPwPI+ezk5c14JUN5oZQA+HMBNheu0SuUPokVQN0+X3d8B0i3dgBmQmG8u4BDI2TTSJd6iXXnyd557dQrloE4oEOPa4BTz7zBaamqNAdNzPp+kas+ckQNMqe3rG6v/71q+54l2rWX7sAuobJuaDMPl9qK7pFu+hjE0TkvR8/OMf55577uGxxx5j3rx5ox73xS9+kU9/+tOV7/v6+pg/fz5NTU0kk8kJiPTQeJ6Hpmk0NTVNi1+c6RCv1Zuh+OrrRDJZ7EIRJxzA3t6OEYpgxmPk1r5AYE8XWsAk1GQTaqyj3J0mXrQJBBzM+gbiJx5P+/0PY/dmMBWgaWimgREJY4RDON1pPNdDN02/T09bF0HPIxSOYHencHJ5tPYUesAkEDAJ5kv+LMrA/p5RqIAJuk4knUfzlL/0rat3736grgyhYICg66KU8j91U8ofhJQ3dN/QvjQNo2BRFwgTbW4e8n7lt+8hliuhGbpf+tp2CDU1MH/lyQAUt7cSzZb86m9lC3t3J0YwTM0xS+js6CGOgVuy8Kwimq7TfOHZNCxdckg/x2qYDr+/g023eIPB4GSHUFVjHZdAxqaJJvHu9eIL2/j9b57GDJi0zGukuzvD73/zDG+80X7Qzzkw1ry+LeWPZeP07zffRTIZ43NfuIrzLjj5oOMYK/l9qK7pFu+hjE1VTXqUUnziE5/gzjvv5JFHHmHx4sX7PT4UChEKhYbdruv6tPhBQP/+C4n3sPFyBZxsAbdQRAf0YACcLL1Pv4AZi+IWyyjPRRVsijvbcLJ5jFCI3qeeR3kKTdfRggG8bB5N4S9xUwo8hVssEYjHCNQkCM1qorynEyfdh1cqoek6pV17wPPQABwbr/8TXS0UROtfvrbfxMR2wHXRPOU/R2Xfj+aXq1aKyNzZuPk85a6U34tH19GUwi9xsJ8nVwovX6B33QaCNfG9DVxzBfA8wrObsHt68Wy/GlztcccQbqijsKMVN+eXutYBPRbFLRTpe/FlzHgMZTlEW2ahbMdPiLIFYi1zpuzvx1T//d3XdIp3OsR4MMY7LoGMTZNB4vWrs/3qv/7C69v2EI4EKBbK6IbB69v2HNoT6/4wqAY+ZDsIvakc//jFn3Hle8/hpFOOYtnyeVWt8Ca/D9U1neI9lBirmvTccMMN/PKXv+Suu+4ikUjQ3u5/MlFbW0skEqnmqcUMYSZi6KaOZ9lgGnjFEp7tYvVmcIql/jXJul8C2lPYvWlUIkYg0YzruBR270GV+5efDPxD8RR4/r6c0p4O0HXcYgm3UEQ5jl98wB1UQW3gMYZeSVb8Km4HynpGMTDIKIWTyxGe1YyTL+IWigMj0dBzj/o8Hun1L1PY0UrD6pU0nrkSMxHDCPvL4cz5LTiZPjTTJHnycZX3UzMMvLLl9w1yHPRgAM/xEzI9aOJk8wQScbxSmUAiOqTQgRDTnYxLYjoYqM7meh5okE7nSfVkR1xVPVlc1+PXtz/C7+94giVLZnPt31/M+ReeMtlhCTGqqqZ0t912G5lMhvPOO485c+ZUvn7zm99U87RiBgnWJ6k95Xi/j02+iFL4fW1MA69k4RbLfpKg62AYYPh7cJxSmVJru9/oc4A3QvU1T4Hr4WRzgxKZfvuOLq4H+MvFKsnJIbJ6M+S2bUfZDprR/xr0UfYLjcRxsLp6aL/3Ybqfep5gfZLkqpNA0/zqbNEwDatXVmaC/PdzBZph+EkeYEYjBBJRIi2ziR21GDSNcre/b6lu1UlStU3MKDIuiekg1ZNl965u2vekKBYsXMebUgnPYLbl0Nraze9++xhtrd2THY4Qo6r68jYhDlXjmSuxenpJPfMiesDECCqCDUmcTB92tgCOi2YafkNRz0PZDsXW9v6Sz/1JTGVJ2T5Prmv+bWOd4leV/zk8bAeFRqipHqs341ffUQG8QnHsz2HoKMeh99kN1CxbSvKE5fvt41OzbClWKk1+25ug6QQSUT+5qaslvmge0aYmnO5egk31xJcsOHyvVYgpQMYlMR3YtkNfJo/tOBzqmGMGDObOqefYFQs58eQlJBIRurq76OwoUF9fw64dHbz6yi66uzJk+vIHdTrbcunrK5DqyUojUzFlTWifHiFGMlqPncEaV6/yG4maGpGmCG6+gBmPE1m8gOymLf6FjK6B4/nVzQwdz3FBKbRQsH/GZ4S/5Ae5nvmwCATAccDQ0QMBlPJQBXvvLM9oFeIG0zU0TUOhsLN9ZLe8TsOZKytNTK1UmsKO1sp7O7hRqxEOEz96EcmTVxCsT+J5Hrk3d1N89XW8/vLfTr4gjUmFEGKCBQImtckY6XQO7yDHqWgsxKxZ9SQSYT76iSs4ZaVf9dTzPDo7O2lubq7sj2hr7SbVk8W2Hf7w+yd54L514zpXuWyjAfUNiYOKVYiJIEmPmFRDL8KDJFedNOQiO/fGTqyuFMGmeurPXEnupU2UOrrw8kW0YBDPtgk2N2B19+KVLXA9tIDp71fxFJ5to/rXRKNrfiLhjlDjfaTtOQONRg0DDA2s/fTnORieC7qGsmxKHV2o0j6lb5WCYBD2VxLXU6j+/UlutkDH/zxKuaeXuWsuHPbexpcfRW7ztiGNWgs7dpM8eQXQv9Ru81bihTKB2hqUZUljUiGEmAT1DQla5jUSCgVI9+bH/fi6+gRz5tRRV59gzTvPrCQ8o5nb0jhkhubxxzZSyI+v2emb2zvY8uoumekRU5YkPWLSWKk06XUbhlyED77IbrvnIdIvbEI5DpppUnvKCmqOX47b9wKuAieXp5TODJsNUZbd30sHP5EZ0kx0lE/MRrp5YP+P68L+q1OPn6ah6QbKtvt7Jezt7ePH0x/QeHqA6Dqu49L73AYCyVryW9/AtWyMSBi3XCaz/mU8xyUypwlN1zETccrdvTjZPMH6JJkXX8ZKZbDyJZxMjkBDEjxVuV8IIcTEmNvSyJorzuBHt/1pXI8zDJ1QyOSDH76UZcvnU9+QGHcScsrKo7nworfwpz8+gzvSh4SjsCyHe+5+mmXHzpfER0xJkvSISeNk87gli1Bj3bCLcCvdR/qFTaAURjyGmy/Qu+4lzICO7ni4pZJf0W1/y7+m8tJ9pfyEp/+/jXAIp7+89LBiC+Nh23hopJ58Dtey+gsuaGgBv9mpGY/hZHOYiThONoem69h9WXJv7CS3bYc/G6b7/X3KHd2EZzVK9TYhhJgEy46dT0d775iPN02dYDDA/IXNnHnWcYeUeHzpK1cza04df7r7GfbsSY25tPXLm3bIvh4xZUnSIybNQHnlwRfhRjiImYiR2/omynEw4jGUZaNcF89xyb2xk1ghj1sq+w08D6eR9tCMZV/NYeD05Q494RnUB8hzXNxcwX9OQ0ezwTMDxI89mtKuNuxcf3lsFN1PPAfKwy6WCDTUohW7UbaD8jziRy+RWR4hhJgE3/v/7qJQGNsSs2gsRHNzkoaGGt7z3nMPS9Lx9x+6jEsvP40ffO+PPPTAC2PaW5TuzWHbh3kpuBCHydTvQiRmrMHllfctkRxsqkczTdx8AbdU8vvm6H5HNTdf9CuzVaMIQf9MR8VEVno6TOfSI2GUO2g9nuOibBevbJHb/Lr/XiqFWy75sz+JGJqu4+byKMsmNLuJUEMdkTmzKv19gEpRBCuVPixxCiGEGFlbazd//p/1YzpW02De/EYuvfw0/s9Xr+b8C04+bHHMbWnkIze8g0WLZ2MYY7tk/NTHvnfYzi/E4SRJj5hUyROWM/edFzPnsrcx950XU9tfxCC+ZAHJtxwP4PfP0TSCdclK8hNsbkQLBIa01RnRWJp8DlCqv3Gpd+DnncKUZeHZjl+AYfAaP8/D6csSqKvFK5exe/sod/dS3NmGnS/g2k5/cYNWPMcd0t8nvXEzbXc9wJ57/0LbXQ+Q3rh5Ul6bEEIcCb7xr78d87HzFjTR2FjLSxveqEosc1sauf4jl3HyKUsJBA+8QMiyHH78/XuqEosQh0KSHjHpgvVJogtbhi2jmrvmQua+61JCsxqJzJ2FWZtA2TaaaRKZ3URi+VEE65OYNQm0YAAtHPRnaTQNQkH0cAgtYI4tOdqXpk/bxEfZDsp10EPBYfuaPMel3NXtLw8E0MDzXKyeXnTTJNTcSLipHiMcxIhFKexoJffGziEFJ1CK9LoNMuMjhBBV8PvfPsZTT74ypmPnLWhk5apjaGyqJZ8vkerJViWm8y88hbdddArNzckxHX//OEteCzERJOkRU1rdycfRfOFbUa5HcXcbnuOibBs7l0dZZcx43G9MahjogQCaroOuo2saRtzfgK9cd/xFDTxvahdCOCANPRgY9G1/Bud5lDu6cYtFNMNA0w1/VkgpP4GMRQg1N2Bn83Tc/wh77v0LHfc/QrknjZmIVwpOuCXLLzghS96EEOKwaWvt5u4/PDWm1c6maVCXjON5Ht1dGWKxcNX65LS1drP2iZdJJsdW2KattZvf//axqsQixMGSQgZiyou2zMYIB4nMasQOBdCKHRR3tRGsqyU4q5FSZ7ffo8eiv1qZwrM8VMb1l8YdoR3YNV33l7i5+7wHCr96nK4It8zC6cthpzMYwQCWp7B60riFImYiRqg+STnlf2/19BJsqKsUnCi0tZPbvG3UHktCCCHG58H7n2fHmx1jOrZ5VpJgKMDOHZ3EYmHWvPPMqlVNS/VkyedKzJ5bz5bNu3Gc/fdxUAr+eOdazjzruMrjD6Z8thCHkyQ9YkqzUmnSL76Cnc0Tmt2M7pTxlAe2jZVKU06lMUJBtICJGlzCWikU/Y1FnSOwkozroodCBOp17O7ekRM/18Hq6sGMx0msWIaTzWH35Qh7nj/jU5/Esx2MUAgjHER5HuXu3lEbnUojUyGEOHhtrd08+8zmMZWH1g2NNVecwSWXnzohCUV9Q4JYPExHexozYBww6QHY9voefvDdP5LPl8jnSsTiYdZccQbnX3hK1eIUYn8k6RFTVnrjZrr+shYr1Ytn2ZR705TrEkTKNtrAci3bwXVdjHAId6C8tEal/LMRj+Hm8odWCnqaslK96MYo/8T73yvXdiFfROvpJXHCcvR4hLpYnNRjT1No68DNF/DKFpphUPuWE4jMaqo8xWg9liTpEUKI8Uv1ZHEdb0z1dxoaanh505ssXjJ7QpKIgWapv/vNY+hjLBDk2C5rn3yFY5a1sGBRM91dGWleKiaVJD1izKxUGiebx0zEqn5ha6XSpNY+j53pQw8GUK6Hsm2U4/QvYQPl9c/gKPAcb29i0798C8AtFsHQj8ikR5UtVNQYeZan/zZVLuN5LqV8HqsvS+iis4kvno/V2U3ng4+jPA89FMSMRsi+/BqF13eiPA9N1/Esa8QeS0IIIcZvYDalVLIPeOzRx7SglJrQJOL8C09h2bHzefD+5/n9HY/T2ZE+4GMK+RLRWBhd12lsqmXnjk5pXiomjRQyEGNyqCWLx7vh3cnm/YQF0EwTUKOXn1YKNbhR6eA+O7azN+EZT/nqGcIrlvZ/v+ugNA23f7lg77Mvktm0hejc2QSStcQWthBbNI9gYz1WbwblOIQa69ADBqDh2c6wHktCCCHGb25LI8cdv/CAx2ka2LZX9YptI5nb0sg1f38xP/jJjVz9/gsOeLznKTraUxNSbEGIA5GZHnFAVio9pGTxePdvpDduJr1uw7g2vJuJGEYkgt2XwyuXKw01R6RpQxOaQbM6Zl0NoOGk+/zGo+4RVtTgQEUcHA+3WOpfFqihXJeeJ56jYfUqAokoKIUeCFDu7AbArK3ZW8GtbPvH1SQmZPZPCCFmsocfWs/9f3rugMdpmoZSk5tEzG1p5Oxzjuf2//zzAY/1XDUhxRaEOBBJesQBOdn8Qe/fONiEKVifpObE5dhPZHH6cuy3frRSYI28HMDJFTGCAf+YIy3hGWxgv1P/f2uG7le2A3Bc0HU008C1LEo9faTWPkdkQQtWVw/l7l400/Sbw1plVChQWc4WkcIFQghxyNpau7nn7qfp7s4c8FgzYJLPlwiFEpOaRAQCY7uEPPW0ZVx48Uqp3iYmnSQ94oDMRAwjHDyo/RsHmzClN24mt3kbeihEsDEIpk5xd/v4g7dtXKXQQsGh1d2ONLqOFg6i8v6SQaVp/jLAgVkxz0NZHjiO39RV17G6emg890x0w8BMxCi0tpNet6FSwU2WswkhxOGR6smyY3sHxYJ1wGNPOnkxH/7YOyY9iahvSAz5PG0kgYDJrl1dkx6rECBJjxiDYH2S5KqTDuqC92ASpsGzQ8H6WpxMH57jYdbE0MNB0AvjK0ygPEINDVipNF6pzAH/Ss9EGuhoEI/iWfaoywU9xyWQrCHYUEe5uxc3X0Cv8ZdOJE9YTrRl9oQVsxBCiCOFbTv09o5tb47nqSmTRJgBA9savXx1fUOisu9oKsQrjmyS9IgxOdgL3oNJmAZmh9A17PZulOvgOS56JIQRjRJIOjipQUsADpDEaIaBHomgvF5/X8+RlvAAOC6uZhE/ahF2f7NRJ18c/l4ohVLgZHN4lkXP2udRnjdkL1awPlkpTCHJjxBCHLpAwCQYDJDPlw94rOt6UyKJSPVkaWpK0tbaM+oxmq5J8QIxZUjSI8YsWJ88qAvc8SZMZiKGpmuUOrrRDAN0HeU5uPkirmlg92YYUodttCRGHzhKw+5Joekaek0CN9037tcwIzguhe27/P/WRy/jXdrTgapL+rmk42DW1qCscmUv1sAyt/EUphBCCDG6+oYEs2Yl6e3NHfDYTLowJZKI+oYEi5fMpqM9hTvKntna2pgULxBThpSsFhMiWJ8kurBlTElTsD5J7OglKM9DuS4o0E0ddB09HgU0vwGpvk8J6n1LUivQdB09HPJb93geXv+eliOSUijHQQuYMErHb800CTfV49kWdiZLuaeX4q5WXMfFLVkU+xOegcIUKEV63YZhpcjHW6JcCCGOZHNbGjlm+fwxdVYolw48GzQRBhqWnrH6uBHvX7ComRs/827Ov+DkiQ1MiFFI0iMmxYEuiutOPo7InFkEahMYNXGU6zfJRHmV5qR4/b17dA0Cpj97oeuVEtZ6JOSXYbZtlO1UGpweyZTj4mbzeKVB/XsMw3/PDINgfS3Bhnrcgn+/Zhgoz6Pc3oXW3//ILVmYifje0tUlCyebrzzdofZ0EkKII01bazc9PX3U1cUPeKym6zz3zBbaWrsnILL9O//CU/jM56/k//38Mxy9bC6GoWMYGjW1Uc4+5wROWXn0ZIcoRIUsbxMT7kB9e6xUGiebJ9CQpPTKVjzbRrkuuhfAK5b9WZ6BiQql0EMhzLparPau/oTHv98rlkANKnbtjr7Z8ohlmn4yYxqY8SgqFPJLVOs6wcYG3L4synVRnkf86MVEWmaPWpjCSqUptraTWvs8esA4qJ5OQghxJEr1ZMnnSoQjIWD/S9wymRx/+O8neeyRl1hzxRmcf+EpExPkCNpau0n1ZLFth+bmOhKJGNFoiEKhzI7t7bS1dsvSNjFlSNIjJtSB+vZ0P/U8mfUv45bLOH1Z9FCI8Owmyl092LkCXkAHTQdDA9ffkxKoT6LKfjKkB0w8y/ZnfAaSHNuZxFc8tUVaZqEsGzNZQ2FnG+X2LoxsETMSJhANEUwmcDJ9aKZJ8uTjRixMEV9+FOkXXya39U2cQhE700d47iwC4+zpJIQQR6r6hgSxeJjs64UDHhsMBjj6mBa6uzLcc/fTLDt2/qQkFg8/tJ577n6afK6E53n09RU48eQl6LqO53ns3NE5JQouCDFAkh4xYaxUmtzWN7GzBSJzmob17enb8jqdDz7uzywohbJs3JLll1j2PDTTQA+Y6MEAumn6CU65jHI93GIJPRBAKdXfiFRmdcbCTvcRaqhHlS3CzQ048TCRWAmvbOHZLqqcx4hGhlTcG1yYotDWTt9Lr1Lc0wmAWRMHDcrtXZjRKMoqj7mnkxBCHKnmtjSy+uwVbFj/+gGPzfYVSPVkaWyqnbTEYqCZqlKKBYua2b2ri0w6z+5d3cyb30h3V0aqtokpR5IeMSEGlrQ5uQJ2OgOeS6RldmV5lOe69D67wd9oHwrh5Qt+8qJpeK6LsmyMWIRAbQ16yUGVygQa6qg95Qy8YpnedS+iNB1bNs6Pi1csEWxIkt++G6M2jh4wCTYksbrTNKxeSaAmUUlYBpeoHkiAcpu3oRwHlMKzHazuXjTTQHkOpbZ2ArU1NKxeKbM8QghxAMuWz2f+gma2bN6531Z0tu3yzFOv0thYy5Kj5kxKYjGwHG/BomZ0XWfe/CZ6erI4tsPOHZ3EYmGp2iamHEl6RNUNXtIWnt2E53nYmSzoXQQSUepWnUT+zd2Ue1L+BbTtUNmJo2lomoYC9GAQIxrBjEXQYlFmXXIeTr5Ax9rnsXr7+gscHIE9eA6BZ9mkX9qMsixUqpdSUy0BRxGsrSHSv+RwtD1YA/2U9EgY5QyaWfMUynXxPIVn25S6UkRSaUl8hBBiP+obEtQ3JAgETMrl/S/Ltm2Xzs40577tpElJLAaW43V3ZWhsqqW7K8O8eY38r2suJBAwp0zzVCEGk+ptouoGLo7NRBwnV8AIBjAiEWqPX0bdqpMwYlHyW9/oX7IW9AsReAp0neCsRsJN9QQb61Eoim0dlLvTeJZN7o0dpNY+j5svoAcDk/0ypy1V7i9/6nl4lo2VylDq7qV77fPk3tg5aolqMxHzZ+mKJTRD7y8goVCe55cXNzSs7hTdjz3N7jv+JFXchBBiPwaWuO1vlmcw1/V4+M/refih9dUNbAQD5ao1TWPnjk40TWPNO8/klJVHc/yJiyXhEVOSzPSIqhu4OM5uexMnl/cLEGjQ/dQ6QvX+hbRTKBKa1Yjdk0bTwS3bBGrimMEgRjhIcFYTfS9vgWgI5ThY6T56Hn8WzTTxHAdN02SW51AMfu+Uws1m6XnyWdIbXsYIh4gtaBm2Byu6sIXkqpNIrV3nJzm6jhmPVmbdrFQGdM0ve+04UsVNCCEOYNny+SxY1MzrW9vGdHwqleN3v31sUooZnH/hKSw7dj6pnqzM7IhpQZIeUXXB+iTh+XPJbt3ur1ozdHA9f4ZmTjOe7fjJEGDWxlGOQzgcpvmic9ANA8916bj/ETzbQbkmeJ4/K+G60N+kTdKdw0zRn/zk8UplyvEYofrkkBLVsLeoweDqbSgPNN1flqg8cD30aLjSz0eSHiGEGFl9Q4JkMoZh6Ljugad8bMuhp7tv0qqkzW1plGRHTBuyvE1MiGAigWaYmIkYZjTql5RW4JUtADzLwupJUdy9B6snTbhlNvElC/z+L10pnHxx6L4RkJmdCaIcFzdfoNzdC5o2pJIb+Elt89vOYt6Vl9O4ehWBZA1GJORX0tM0MHS8QkmquAkhxAHMbWnktNOXEwqN/TPptraeIcUM2lq72fTS9inRvFSIqURmesSECDbVowdMlO2ghYL9ldkATaPUX+5YD/VfKAOFN3fT9qc/U9q9B6dQws70SZJTbZo2/Lb+4hBKQXzZUur6e/WMJFifJH70Ivpe2YJr2XhlC6s3DQo00xyWLAkhhBjuoktW8vhjG9n00ptjOt4qO7y88U3mtjQO6Z0Ti4cnvXmpEFOJzPSICRFfsoDkW473S1AXiugBE7MmgVss+z14dN2v3Oa4eKUyxT0d9D6znmJrO3amDyV9d6pvpKQHQNcxggHyW7dTbG3H2k9Z8IHmpUYwgB4IEJndTNM5ZzDvysupPWF5deIWQogZZG5LI0uWzhnXY7576x+G9c5RSnHP3U/LjI8Q/WSmR0yYuWsupOa4Y7C6UgSb6gkmayi2ttPxlycod3QP3ZjjeijD8JMd2/H3iYiq0sNhNHP4nwTN0PGUoryng46HniCYTFTKVo8k2jKbgfQpIoULhBBi3FrmNaHrGp43thUOHR1prr7q68yb11TpnTOZzUuFmIok6RETKr5kASxZUPk+WJ+k3NVDR8fjDKnTqev+7I9ly7K2CeI5NughALSAifIUKA/lepS7etANg1BTPcqyRq3ENrinj6brxI9eRPLkFZL4CCHEOJx40mIikRCFQmnMQ2CxYLH1tVY6OnqZO7eBaCxMLBauWvPSttZuqdwmphVJesSkiy6ajx4MolwPzdTxShYohVkTx+qvziaqTzd0tP7BVXkKTddQLn4yaiswDdxCkWBdbaVs9UAyY6XSFFvbSa19Hj1ggKFTau+iuKeD3Nbt1K9eNerMkBBCiKFOWXk073z3au78/RMUC9a4HtuXKZDtK5BIRHnHX50xJCFpa+1m86s7cdwSK1YYtMxrqtw+ngRmYO9QKpUlYBpcfNmp/PVV54zvRQoxwSTpERPOSqVxsnnMRIxgfRLdMDDjMTzLQjkOmmmiPM/v+1ITx8lkJzvkI4JXLOPFI+iRMJpSfknwgQpsGuC6lDu7cYtFjHC4UoltYHbHSmexM32EmhtwMlk0Q0fTkB49QghxEE44cTEvvfgGW19rpVy2x/VYpaCvr8D9f3qOWbPqaJqV5KknX2Xds1vI9uVZvLSBX5Yf45oPXAww5uIHba3dbNm8i9/99jFyfUV60zmKBYv/94N7ASTxEVPahBQy+N73vseiRYsIh8OcfvrpPPvssxNxWjGJrFSawo7WYZve0xs303bXA+y59y+03fUA6Y2bMRMxQg1JNNNA9S9x0wMmsaMXM/+978Ssrc7UvBhO1zXMmgRawATHATSMeBQ9HMazHJxcgXJPGrdkUegvapBetwGUItRUj6brlDu6ccs2oPllymtrKj16hJgqZFwSU9lAUYKa2igrTz36oJ+npyfLN//993zxsz/h7jvX0tbaQy5fplAos/W1PfzfW+7gZz95gHQ6R7wmQrFYHrX4wcMPref/3nIH//Gj+3n15Z3s2tWF57rU1cdwXY8H7n1OiiaIKa3qSc9vfvMbPv3pT/OVr3yFF154gZNOOomLL76Yzs7Oap9aTJLBic3uO/5E51+exEqlh14gN9aBUv73gJGIY6f78EoWynUxwiGszm6CyRqSJx47emUxcVh5roeTyfhryHUdPRJC13X0UBA00AMBYovmE0hESa/bQLG1HbdkYSbimJEwoVmNA0+Ecl0CDUmUZUmPHjGlyLgkprpUT5Z8rkRjUy3Ns+qYNTt52M/hKY/u7j62bN7Nyxt38ORjL/PKpp1s29pKqmfoCovBleFC4SDlsk2hUKavr0S6N084EsR23GGPE2IqqXrS881vfpPrr7+e6667juOOO44f/OAHRKNR/uM//qPapxaTYHBig6FT6uii67Fn2H3HPfS++ErlAlnTdcxEHLdkUWxtp9TajqbrGPEoZjSM8jycXAEnm6dh9SoC9bX9y6w0v7GpqA7XxStZeLk8uB6eZePaNl6hCGgEGur8vT/BIG7JX2duhIM42RzK8zACJpG5s6g7/RTCsxrBUyM2NBViMsm4JKa6+oYEsXiY7q4MnufRMq+JcDiArlfvA0ClFKWSRUd7mi987sdDZm0GkrBINERXZ5pIJIimgeM4lEoWsViI+vpE1YomCHE4VHVPj2VZPP/883zxi1+s3KbrOhdeeCFPPfVUNU8tJomTzVcSG7ujdci+jvzWN9B0HSebw0zEcbI5jHCQ/I7dWD1plOfhFUto4RCULbSkgee6FFvbCTU1YPfl/GTK80DXMGtrcHozk/2SZ46RxlLXw7NdQnW1OMUyTl/WnwnyFGZNgkjLbBSQXreBcncvRjhIw+pV1J6wfNjeLSGmAhmXxHQwt6WRNVecwT13P83OHZ3EYmH++qq3suHFN3ht824sy6nq+bs7+3j3mq8yf2Ej55x7EsuPnU8sHmb3zm4KhRLBcABN19EA23FJJKKseeeZUsVNTGlVTXq6u7txXZdZs2YNuX3WrFls3rx52PHlcplyeW+1rr6+PgA8z8Pzpn6fFs/zUEpNi1ihOvHq8Sh6OIid6cO1HTTDQDNMjNoanGyexDGLKOxspdR/gZxYtpTMxi1gGmg6eLYDhRJ6KEiguZ62Pz5IuacXZVko5bfyUQCewi4W/VmfKfp+q0Ff08GQ93fIHR5mQxKvJ42d7kO5DihQmT4ym7fRcMZbCM+dtTfBqavF8zzMZA1msgagKv8m5N9bdU2XOMdrvOMSyNg00SRe37lvO4ljls+rVFWbM7eBPW09PPTA89x151r27Ok9qOfVNK1/4YR2wPU+u3f18Mv/+gsAc+Yk6esrUixY6EWbBYuaSCbjOLbDh294Bye/5Sj5W4/EW22HEueUqt528803c9NNNw27vaurC8saX8nGyeB5HplMBqUU+jRYglWteN1jl5LbvA2rHAZdI5CswVIOJMKEFszGXDAbvVhGj4TIF8v0BXRYMBs3m8OzXVAekUXzSGWzWMrBi4UgGgTAikeGTkhEgoct7moYFu8UN2K8mk6xuxvQ0Opifg8l00RZFrs3v0apKUkgEYewCXYZJmhfhPx7q65MRmZRB8jYNLEk3r0ME5pmRQGXzs5ODBMuvvxkTj1jCWufeJltr7XS3p7CccZ+IajpGrPnJEDT/H5s4xCNxTEMHU3TME2NpuYIq996PHPn1VRtT5z8PlTXdIv3UMamqiY9jY2NGIZBR0fHkNs7OjqYPXv2sOO/+MUv8ulPf7ryfV9fH/Pnz6epqYlkMlnNUA8Lz/PQNI2mpqZp8YtTrXibm5uxjl5K+sVXyG/bjio6GGGd5MoTqF26ZMixVm8G7/lNuGUbLRTB9UoY4TD1Rx9F55+fxEj1+Q1K2TsLEenNDb0w17Qp2cB01HinqNHi1YLB/sHRBdfDrE2geRYKCOTL1AXCRJubJzxe+fdWXcHg1P5A4WCNd1wCGZsmmsR7YM3NzSw/9ihefGEbP/zeH2nf00EmUxzTY7X+MfP1bSnUIYydwYDJSScfy4UXnXHQzzEW8vtQXdMt3kMZm6qa9ASDQVauXMmf//xn/uqv/grw39w///nPfPzjHx92fCgUIhQKDbtd1/Vp8YMA/4+JxAvhhjpmX3AW1ikr9ruvI9xQR6ipgfQLm1C2A7pO5ITZxObNIRANY3W6Qy/AB31VTMGEZ8CI8U5hGvj7sLz+FEgBto2m6+imgWc7ePkCRiRMIBYlEI8SrIlP2u+7/HurnukQ48EY77gEMjZNBol3bILBAJqmc855J9G2u4cX17+O47j7f5Dev5xZqXHP9AxWLttsePF12vekqr6XR34fqms6xXsoMVZ9edunP/1prrnmGlatWsVpp53GrbfeSj6f57rrrqv2qcUUEKxP7ncTu5VKY3X1YMSiuPkCnu2QfWUrkZbZ1Jx4LPmdrTA9lpnOHLrmL18r9y/bUQotHERToAUDBJK1GMEAZjwqVdnEtCTjkpgpBld5mzuvgWxfgT3tKYIBk97eHK5b3QF0y6s7SfVkpYCBmBaqnvT8zd/8DV1dXXz5y1+mvb2dk08+mfvvv3/YJlJxZCr2N7h0CyV000APh3ALRTLrN9H41tOJzG6m1J1ClcoHfjJxeNguSleDSoMrNKXQAgGC8Siz3n4OgZqEVGUT05aMS2Km2LfKW2NzLe/869Ucs2w+Wzbv4ve/fZw9bT2Uy5Zf+FTXDmtxHdv2+Mcv/pQ7//RPh/FZhaiOCSlk8PGPf3zUZQPiyJXeuJnU2nXYmaxftS0aQcdBDwbw+jdlmvEo8ViEUm8Gqys1yRFPA1r//xzikj89FABdxyuV0UyTQF3S760UixBpmS3Jjpj2ZFwSM8X5F57CsmPnV6q8Dcy6HH/iYs486zi2bN7FppfeZMOLr5PLFrEtm9lz6nFsHdv2SNbH2flmB7ls8aCGjj1tKX78/Xs48+wVQ84vxFQzpaq3iSPHQBNTPWASbG6i1LYHt1BEi0Uw4zECiWilB0xq7fM4mez02RgzmRRgGuAcWg8Hz/XQFOAplOthdafQTJPo4vmS8AghxBQzt6VxxGRj4PbzLziFttZuUj1Z6urjZDK9uI5JQ2MNc1saaWvtZsvm3bzxehuPPbKRcsnize0dI5xpZD/9yf08v24rsXiYNVecwfkXnnI4X54Qh4UkPWJSDDQxDTXWEaitAU1R7uhGD4UJ1CYqe0WC9Uk0oHzX/+BZtr/kytChyuuUp7VDTHgAsOy9SyAcFz0RI5SsxersxkqlJfERQohpZiAB8jwPw3Rpbm6ubArfmxydzJIlc7jn7qeJRkNs2bwL1z3w9I/nwbatrZx48hLuuftplh07X2Z8xJQjSY+YFGYihhEO4mRzmIk4gUgYo2UODatXEmnxy8YWdrRiJmLYuTyeZflLtjxPEp5J4GRyhJsbcUsWTjZPsD6JlUrvtzLfSA7mMUIIISbO4OVyf/zDU9x159oxPS6XK7Hhxdc5+ph5UtxATEmS9IhJEaxPklx1Eul1Gyh392KEgzSsXkntCcsre32cQgk9YKJcDyMewy1P/SaAM5bnYfX0Emqsx0zESG/cTHrdBtyShREOklx1EskTlu/3KQ7mMUIIISbewMxPfUOCjS9tZ8eODtwxNEDN58p07OmlviExAVEKMT6S9IhJkzxhOdGW2UM++bdSabr+8iR2Joumge24eK5LbEELdiY72SEf2XSDulUnAZBetwGUItRYh5PNkV63geh+ChwM7OEaz2OEEEJMrrktjVz/kcv43W8eY/OrO8nnD1xJtbs7MwGRCTF+U78LkZjRgvVJogtbKhe+xdZ2rN4MmmGgh8NoARMcl3J3CuXJsraq2l/DL10nedKxRPqTVLdkYSbiaLqOFgxipbMUW9tHffi+jzET8cpSOSGEEFPX+Reewv/56tV8+/sfRxtDQSHbdvn9bx+vfmBCjJMkPWJSWak0hR2tWKn0PvcolOfhlf29PE6u4O/nEdUz8P7uO6qZJpppkNm4hba7HqDQ1l7Zj1XqTpF/Yyd2po+etc+T3rh5xKcevIdLeR5ONocRDmImYlV+UUIIMTO1tXaz6aXttLV2V/1cc1saOf7ExVzxV2eO6fj77312QuISYjwk6RGTJr1xM213PcCee/9C210PkN642e8BU5fEK1vYvRmUZfsHa4CugW74/y8OnaETnjtr+AzPvo0aPBelFEYsgle2yG3eRnz5UXi2Q2lPJwCRubPQAwbpdRtGSGD37uFC0yh394KmVSr0CSGEGJ+HH1rP/73lDr576x/4v7fcwcMPrZ+Q815y+WkYxoEvHXu6s/zge3+UxEdMKZL0iEmx7x4PlPK/B+pOOwnN6C9NDaBplWVUKOUnPuLQuR5WJguo0Ze2aRp4ClyPUuseyt0pyj1ponNn07B6FcFkDbElCwk11B1wyVryhOXMfefFzLnsbcx958XUShEDIYQYt7bWbu65+2mUUixY1IxSinvufnpCEoxTVh7N37zvvDEtc/vLg+v5l5tun7CETIgDkaRHTIr97fGIzp1NoK6OcHMj6DqaaaBcD90w0AyDYLKGYFMDgbrayX4Z055n2Wi64Tc03XcUM3T/PdY08Dy8so2TL+L0ZfFcl0jLbAK1CZx8Hjubp5xKo+k6dl92xNkeGL6HSwghxPikerLkcyUam2rRdZ3Gplry+RKpnokp9vPJT7+bq9533gGPcxyPra+18rvfPCYzPmJKkKRHTIr97fEwEzECiShGKIgeCqJcD5RCuS5GNATKw7MslO34F+vi4Nk2ynHAsv1ZNE1HCwYwomF0M4ARCe9d7tbfJ0m5Lm6+4DePbWqg3NFN/o0dlFrbKfem6X7iucpyRSGEEIdXfUOCWDxMd1cGz/Po7soQi4UntEz0yaccNabj+jIFXn1lBw/e/3yVIxLiwCTpEZNif3s8Bu7TQ0E00wSUP9ugaQTqk7glP+HRQkF0wxw+QyEOmh4wiMybQ6C2Bs3QcAaVCdcCJug6ylM4uTxWKk2ptZ1AbZzQrCb/Psv2ixP0L1ccbcZHCCHE2OxbsGBuSyNrrjgDTdPYuaMTTdNY884zJ7QZ6LLl85g3f6zn03j2mc0y2yMmnfTpERPOSqX9ZWwts4f16RmQPGE5ZixKx/2PoJI16MEgdrGI1ZcnEAyieS6qbKEFTbDt4ZvvxUHS/OIRmo4Rj+NZ/Q1hdR0UfhEJTcOMx0i/+DLFPZ3ohu4/zvX8/3ZdzESccncvTjYvS9mEEOIgPfzQeu65+2nyuRKxeJg1V5zB+ReewvkXnsKyY+eT6slS35CY0IQH/MTr3PNP4vb//PMBj52/oBHX9Uj1ZCc8TiEGk6RHTKj0xs2k123ALVkY4SDJVSeRHGVDu5sv+MvY0LB7O/E8heOG0Q0NMxbHUwqvVMYrlaWHz2Hi2RZ20cIIhwnEIqhEHCdfANdFMww0XSNYV4sRi5Jb+zyarvszPEqhXAfP1cEwpCS1EEIcgrbWbu7707M8cN9zJBJRlh41l+6uDPfc/TTLjp3P3JbGytdUZhgaDY21aJo2ocvvhBiJJD1iwgyu2GYm4jiZPlJrnyfaMnvYbEB642ZSa9dR6uoBxwVdQ2k6xMN4tkMx0+XfjsKvZy0ODx2vVMItFHHzefAUmqH3l6yOEkwmaFi9Ct0wUJ5HaHYTdk+vv9zQMDBjEZxsHiMclJLUQghxEB5+aD3f/Pff0dPdh+cpdF2jN5Vj1WnHsHNH55SYMWlr7WbDi68f8DjXVZTLNu/5m3MnPWYhJOkRE2agYhuGjt3Rimc7KM+j98VXmPW21ZXjBidHmmGgHNcvm2yA0sDNF8F1By1pk6Vth4umaygNcFxcr4ym6yjPQzN1ms45jcSypQTrk1j9ldqU7RCa3YRXKKGZJs0XvRXdMIYtVxRCCHFgba3d/H/fupOuzkzlNs9TtLX28OjDL3H8CYumxIxJqidLuWQTDJpYlrPfY5cfu4DzLzh5YgITYj+kkIGYMGYihqbrlNu7/Atpw0DTdfJb3xiy4X0gOdJDQXRjbzllPRwE119G5Vca2+cEUtDgkClAeXvf24Flgxr+Pp6BRKbQ2o5bKlHq6qGwoxXPcWlYvZL4kgVSkloIIQ7S73/7OO17UiPel0nniUSCU2LGpL4hQU1tFGMMFVTXPrFJihiIKUGSHjFhgvVJ4kcv8i+kPQ9N0wjNavSrgQ1qaOm5LigPJ1dAofxlbErhFUuA2juxs+8Ez2gNNsXYuS54LmgDjWH9/xv8VlupNKm169BNg+iCuYSb6jHCQSItsyc8XCGEmCnaWrt5+KH1+63L89gjG6dEAjG3pZH3XHXOmPbT5vPlCeshJMT+yFWimFDJk1cQmdNMsD5JZMFcjIA5ZMN7euNmOh98HDuTpdydwutfDqcFg+iBgF82ebRPllx3Al/JDKXhJzzKG3QDaJqGEYsCVKq2Wak05fZuMM1hiasQQojx2bJ5N11dmf0eY1kOP//J/0xQRPuXrItj2wced2PRie0hJMRoJOkREypYn6R+9SqMqL/hfXB/HiuVpusvayl1dPn7fVwPNAjPmUWgrgYMHc92pGxBNSn690uBZhoYoQBmLIJZk8DNF8hs3EzfK1uHVG0rd3Sj6ZpUahNCiEOicMfw4d369VunxGzPbd/9I6574JmeUtlmy6u7JiAiIfZPChmICZc8YTnRltkUW9sBKsuiiq3tWL1pNMMAz/P37niK0p5OUMpfYhUJoXTNP0Z5e3vHKOUXOxiNrvuTFmP4Az2jaf3vlab578dI75lSoOv+8sM5s9A8D7ds0bP2edxiESvdh1kTR5WtSjGK+NFLZB+PEEIcAn8JmMaBivOUivakV3Bb//xWXtuye0zH9qay/O43j1VKbQsxWSTpEZOi0No+rF+PBijX9RtiDk5OBj75Mg30gIFumihcdPzZBrM2gdWdYrSBQguYaIaBZ1tVf11T1kCRB0PvL0NtgNZfqMD1hjZ31TUC9bW4+SJuvogZDQMaesDAiDVg9+VwcwUi8+bgFf2qbcmTj5uUlyWEEDNBW2s3a594mZZ5jeza2bXfY9O9ObZs3sXxJy6eoOiG27mjc8yrLhzHoaenb9ITNSFkeZuYcINLUoca60Ap0us24Lke/pX4oIM1DS0Y9Ku3BYPowSAoCDfWU3f6KQRqEzh9uf3O4CjbwSuVwZ2hpa31MQw9SvlfrucvW4uGCTXUYYTD7JssaqZJIBolMmcWsy482+/LEwxgJuKYkRCh2U0oz8POZDGiYWpPXI6TzQ+pwCeEEGLsUj1Z8rkSTc21BzzW8zzWPvHypC5xW7CwmUBwbJ+bu64iFArIvh4x6STpERNuoCS1mYij6TpmIo5bslBli0BtAj0S8pejGToETDRdQw8G0HQdr1xG0yC+dBE1xx2DGYsSTNagR8KT/bImj65DILD/YzTAMNACJpH5LcQWzSPSMttPegwDTBMtHAZNRzkuCkXD6pXUnrC8/7ggTjaHUyyD7RBqqmfWhWcTX34U2c3b2HPvX2i76wHSGzdPyEsWQoiZpL4hQSwepnV3zwGPdRyP17e1TWpFtFNWHs2ixbPGdKyua5xz/kkyyyMmnSQ9YsKZiVjlIlp5Hk42hxEOEmyqJ9SQJFSXJFhfix4I+I0uaxIEG+r95CYcxojHKexqo+P+R7CzeQK1CTzbnuyXNfEGqtg5LjhO/74lzU9iBr5Mw79d7/8ecHN5lKcotrbj5PL9SxQUmvLQw0HMmjhN55xB7QnLAb/4RHLVSdjZArlt2yl19aAcFzuXJ7d527AZO5nxEUKI8Znb0siaK85AH8vMPdDZkca2998UtJraWrvZ+WbnAY8LBk1mz6nj9DOWT0BUQuyfJD1iwg1cRKNplLt7KxXc4ksWEGxqwOrN4GTzKKUIt8ym5V2XsPDv3k3juWdgRsOE6mr8i2xDxy0UKXX2HJkFCvqr2wH9xQc0EicsY9aFbyXU1IARDYPn+QUfXBcsC+W62Jk++ra8Trk7hRmPYMSi6AHTX9YWixKZ3TSs5060f7Yn3FRP/KhFBBIxMus34eQKw2bspHS1EEKM3/kXnsL/uuaiMR3reYonHpu8pp9PPfkK6fSB/9Y3NtayaPFsWdompgQpZCAmxUAFNyebx0zEKiWrra4eQrMa8coW5Z4UxZ2ttN55Pw2rVxKeMwvl+TNFnu1ghELoAQO3UPRnMZzJ+9RrUuzbwc71KLd1Ep7VhBEK4JUNfxvTwGEa6KaJWZfE6enFcxy8so0ejfgPL5VB10iuPHFYJTYn688OhZob0XQdFQpi54ropoGTzWEm4pUZOyldLYQQB+fMs46loSFBzxiWrv35wRd4c3s7a644g/MvPGUCotvrqbWvjum4+oYEa955pixtE1OCzPSISROsTxJd2FK5wK7s9YlFsVJplOWgLBuru4f2ex8mu/UN9KBJoa2T/Ju7yG/fiVuy0MMhwrMa+wse+M+txyKVi/kjgq6BrmNl+kg9+RzlnjR2Jrc3Mepf+qYAty+LFgqApuHZNl6phFmTIDy7meRbTqT2+GXDnn6kJYmBRJTaU1YMm7GT0tVCCHFw5rY0suq04X+DR1JTG0UpxT13P12VGZ+21m42vbR92HP//reP8eRjGw/4+FNPX8Y/3Xwt519w8mGPTYiDITM9YsrwXBeUotTRjVe2/KVZALqBchyyr2wluGwRzhut4HnogQCaoeOWylhlv/S1CgZQto0RDhNI1lBqbfefa99ZkenO0Icu6fMUWsh/nzzbRQ/6vYxU5T3UwPMr2SlNQzcCGGG/YIRXKqMna6g/+1TKzXUjnm5gSWJ63QbK3b0Y4SB1q06i9oTl1CxbOmTGTgghxMH7q78+iwcfeB5vf73ngNc2t9J4di35fOmwl4N++KH13HP30+RzJWLxcGU2qa21m+986w9jGlK/+I9/KzM8YkqRpEdMCemNm0mv24BTKGL3ZSsJjxbwf0WVBp7tYEYiqGQtoHD7cni2g+c4gIZu6OB5eMrDS/Vip9Jg6BjxKO5M22cyOOHR/QlbZTuAQjMNvwKb5w05XguYqP7EUrke4TnNqP4y1rMuOY/oonl0do6+MXWkJYngJ0SS7AghxOFxysqjOf7Exbz04hsHPPbpta9w9DHzDuuembbWbu65+2mUUixY1Ex3V4Z77n6aZcfO5ze/eoRS6cA972LxkPTlEVOOLG8Tk25w357YwhbCTQ1gmqBrKNdDKYVumgQSMczaBEYoiJPu8y/Y8fepaIaOEYmApoPj+ftYlALHHX/Co+to0cjY+t9MMj0SRg8G+pt4e2iGiRGN4DmuX8BgcKEDBUYoSHTxfIINSdxCEQ2NxreeRnzJgjGdb98liUIIIQ6/j378HWM6zvOgr6/Als27DtsSt4GeQY1Ntei6TmOTP5v0s588wO9+/diYnqO2Ni7FC8SUIzM9YtIN7OUJNdah6Trh2U24ZQvlOH6RAjTM2jjxoxZjxSIEj1pIcXebP2uB38PHCAb8am7WoGVx46FrlcQAQJUtv/wzyv//ai2P07W9CdpBMBNxwrMasXpSgEZixTHkX9uO05dFKb3SkBQN9GCgv5lrHs3Q+0+pmGEL/4QQYto7ZeXRnHjSYl7asP2Ax+5pS/HD791DQ0MNp52xnIsuWXlIMywDPYO6uzI0NtWye1c3O97s4MUXXh/zc6y54gyZ5RFTjsz0iEk30ib5UEOSeVe9g3lXrqH+tJMIxONkX3uD1DPr8TwPzTDQzQBmIoam+TNC4ZY56AO9akY8keE3MR3pfq8/4TENjGgEIxxCN00IBDASsb09cQ6zUHMjoaaGyhK1cdE18FxKnT3ogQD1p53M7AvfyqxLziPU1ECwvo7w7Ga0YBAjGiE8uxkjEsZK9aJpEJ03Gz1gSm8dIYSYgr76L9eMeWjYsaOTV1/dye3/+Wf+5abbefih9ZX7RitIMJqBnkGapvHSi2+w5dVd9KZyY477mOVz+cCHLh3z8UJMFJnpEZNutE3y8SULsFJpetdtQA8YBBI15EtF8lvfxIiE/Zkex0ULmGiGgd2dItBQh9WVQo3QrFQPBPpnh0AbWD5nDTrO0NEMw98X5Dgox0UPh/xZH/DLYvc/fkS6Pu5ZpnJHd/8StP5EbDyzSp7CLVkoO48bMEk9u8HP2+IxQKskkZrmv95Sewee5aBcF0/hN3TVNOxswd+nk6wZV+xCCCGqZ25LI3//4cv48W33HvBY5SmKhTK1tTGssl3Zg7Pl1V3DChKc+7aTDvh85194Csm6ODd/7VcUCqUxxxxLhPjPX31pzMcLMZEk6RFTwmib5Iut7VjpLKGmejTd37fjdmf8fSwEMEJBf9+Pp3CLRUKzmgnWJshu2wG27SciSoGGvweovxqOUh6wz+yN66FcCzdgguuimSbR+XMp7GxFx19KZvWkGHU9mOf5SYtp+uceTf+quf5A+v9b7f1+HNxi0T9v2cLNF2m/9y8E62oJJmsINtThZPoop3r7G4Yqf8+TplFq7/Rfi+uhmSaFtnbC8+eM69xCCCGq6+8/dBlPP/kKG19684DHKgWFQhnbdulo72XL5t0jFiQ4Zvk8jFGu/tpau0n1ZKlvSPDG63vYuWP04jYj+fNj/3dcxwsxkSTpEVPG4H49AIXWdlJrn8fO9OFkcwRnNWLhoWXzaBq4xRKapqMZOlowiHIcnPwOQrObCNbVYnWn/GVwIb9/j5sv7F1G5npDK6AN5rp+CWjDwCuX/UpojoOV6h094RmglP/4fUtKDzlm0H+bhp8oDTRWHbhP1yoJ2n4NPodSfl+j3gyh5ibMSAioodTRBZqGHgqjacrf92Q7eLYDuoYZDpHbvI34MUsOfD4hhBAT6qavX8vfvfdfyecOPONiWQ6bX92FpsEvf/FnDF1nwaJmdF0nEg2x881OtmzexXHHzwXgwfvXsfW1Vo4+poVNm97kiUc3USqVKeTLFIsHrtI22OrVxx7U6xNiokjSI6aMgbLVbslC0zXckkUgESU8dxbl9i5KrR04yRjRWBRVLqMHg/4DPQWeR7CxHqs7RXH3HoK1NQTragg1NwGK3Bs7+4/bz/K0fnoggOe4eJZNYfcecAYeM8Zqbgda4jZ4mZw3UGlucAA6eiSEsl2UNb5BB0A5LlZPCmPuLJxMX3/paoUql/A0be/r6U8AnVyBclePn2yG5U+CEEJMJXNbGnn/tRdx23f/OObHKAUbN2zHNA0ymTyxeIQdb3bgOi4//48HePulJ/H0E9tY/8LreJ53wJ5AY/GOd591yM8hRDVVrZDBm2++yd///d+zePFiIpEIS5cu5Stf+QrWQVzEiZlvcNnqUGOdP7PSm0YLhgg31BFbshAzHsWIhAnV14LrYUQj/hYYrb82geehmwa4Hm6pjOe4/ixRoTjiHp/RaIGAn4z07xmq6F8mNyLT8PcWmWNIGryhTUWHLWnzPLxSGU3XwNTHnGvtjdPDSqXJ72j1a7NpgKb5edXA69E0NF1DMw1QCidf9JvDCjGDybgkpquLLlnJkqXjX4LsOC7b32hn00vbyWULWJbNtq3tPHDvOtat24rjuHiHoTrpsuXzWLZ83iE/jxDVVLWPdTdv3oznefzwhz/kqKOOYtOmTVx//fXk83m+8Y1vVOu0Yprat2y1WVsDXSmcTB9GqAFllQnU1mCEdLySBYaOVyyBbqApF8/1cPqyeJaFcl2cvqz/vH1ZvziBYtiSMy0cQpXKw2Jxc/vp6zPa2OC4EAigmToKhabrQ4skjOlJBgfhoceDBGriaKEQdk8vbrF04D0/moYWDBCsT6IHAgSbGyi2dgwtwKDrlUQI1wVdQ49E0I3qVKgTYqqQcUlMV3NbGrng7afw5g/bD3pWRilwHA9N1/AGf/h2iDnPylOP5j1/c66UqBZTXtWSnksuuYRLLrmk8v2SJUvYsmULt912mwwuYpjBZavNRBxlWQTratFMs1LRrf7Mt6AKefRXX0cPBnEdFzMaRjMDeFYZuy+Pctx9KqBp/Zv3vaEJg2lgxqLYpfLeBOAQp/eVY2PW1BEMhcDQsdKZUQ4cCG0/ldoMg8iCFqyuHjQgPGcWhV2tgObPWo30OF1DD/mltoMN9di9aYo7Wv0TGvrefkC6hm6aftPX/nLWoYYkZiKGneqhUHII1sSlAamYcWRcEtPZ6Wccy5/ufpb29hTeaHtGJ9jnvngVZ551nCQ8YlqY0AX8mUyG+vr6iTylmCZGKlvd/LaziAyq6GYmayh3dpI8eileroDnuuiGgZmIYaX7aLvrAayeNMrtLwqg9yc8uoamjP49LArlOP6MiGGgmYa/ak3TUJ5zcMEPNDE1DBrfehpWTy/pFzb55wlF9h4zUNJ6IGHZ36yNoZN9eQt+0gZGJFIpp61FQnjFcmW5nVGT8GenXH9ZHCFFqa0dPRRCeQrdNPEcF9Tec0cXzccrFvEcj0AiSt2qkyi2tpN6aRPlbAkzHCS56iSSJyw/uPdEiGlCxiUxXdQ3JFi8ZBZ19TFefXnnYdmHcyje+a7V/PVV50xqDEKMx4QlPdu2beM73/nOfj9NK5fLlMt7lxv19fUB9G+ymxqfauyP53kopaZFrDD14q1ZcQzhubP2lq2uqwWo9I8ZiNesTaD33zfATNZQd/opdD74GN5A0tOfFGgAhkF0UQu6YVSWwtWddjK9z72E1ZtGeR7KZfRpftNAQ/P7/OybrPhZE5pu4OTyFHa2EahPYheLKOX5T6n1J1aG7i+FO1B1toGlcbr/YM9xiB+zhNzW7Wiajh4J41oWummiBUzUwMYfpXDLFq5lE6irRTN0PG2gRLc/42MEAniWRdPF51WSRoC2u/8HZWoEG2txswV6120gPHdW5ecw1Uy1398DmY7xznRjGZdAxqaJJvGObPacei5/x+n86Z5nWLComc6OXorFse9XHaBpmt+/TdMOame3Blz1t+fxyU+/a0J+RvL7UF3TMd6DNe6k5wtf+AK33HLLfo959dVXWb587yfEra2tXHLJJVx55ZVcf/31oz7u5ptv5qabbhp2e1dX17TYaOp5HplMxl82NNY2ypNoysYbNsEuQ+fQ/gAHjHfJPALnnkbxlddw8sX+stM6ZjyKphu4AR0jEvZ72zTWEoiG8I5djL2zDbdcxsmO0HG6P5kxYhG8chnPcvwZk5HoBuUtW/HKlv+P0nWx4hE0Lz7y8Zo++nONEIfZlMSx5/rNRV0PuzcDKPAcqI2wt+KBQjMDuMkanEIJLx5Buf2V7gyDQG0NZQ3MfI5QcwPYZcqdPWSUhxWN0adpUBPD7suht7YRsofve5oKpuzv7yimW7yZzCjLM6egao5LIGPTRJN4R3fcCXOZPfcCstkijuOSzxf59X89gjXqHtLhNF1j9pyEX+BmnLNF8USEs966gosuXkVn5/h6+Bws+X2orukW76GMTZpS4yvb0dXVRU9Pz36PWbJkCcH+csJtbW2cd955nHHGGfzsZz/b7xs60qdp8+fPp6enh2QyOZ4wJ4XneXR1ddHU1DQtfnGmY7xtb2wnGQj7e076ZyCs3syQ2SGrN0OxtZ1SRxe6bhBZNI9yRxeZF1+uLOcKNjVgdfX0l8fWcR2b0s624Sc1DepPP4XYovlkNm0h+8rWkSvBhYL+B2b9RQK8QgkFFOrjRFO5vQXYBperNvSRq7eNYvaaCwnUJuh+/BlK7V3+YKUB9gjL8gydYHMjTrqPQE0cuy+HN7AXyDDQDYNZbz+HhjPeUnkP2+7+H/pMjfpwGDdbAE1jzhVvn9IzPdPt93c6xZtOp2loaCCTyVBTUzPZ4exXNcclkLFpokm84/PIn1/kB9/7I7t2dY/peE3TWHpUPa9vSzGWS8CAaXDscfMpWza6plPXkODyNadz3gUnH2LkYzPZ7+94SbzVdShj07hnepqammhqahrTsa2trZx//vmsXLmSn/70pwd8M0OhEKFQaNjtuq5Pix8E+H9MJN7qyGzaQu9Lm7AG7TkBKr19jEH7UEptHZR378EtWfRt2gz4JZoN0yDYUE9pZyuarhNurMPq6cVq76pszRnCU+Re2Upx+y40XfeXAwx8DZpiNaNRsC1/pZtpovCbyGn9+4W0/qV2gXgU13Hw8kX//lAQVbYqBQZQauQldrqOGQyiaRpuNge27T+vYYx4uKYbOKk0eB5G2C9uUGrr6H9+haZppJ/bQO3yowjWJwk31FG38kSyL23C6s5ghoPUrTqJcEPdof7Yqmo6/f7C9Ip3OsQ4oJrjEsjYNBkk3rF720VvYflxC9iyeTe/+/WjvLRhO/ZIH4YN0AeGArXfmZ7G5hqu+/tLWLJ0Dv/184eIKUVjUy3dXRn+9MdnWH7cggkrYCC/D9U1neI9lBirtqentbWV8847j4ULF/KNb3yDrq6uyn2zZ8+u1mnFDGWl0qSffwlMjVBjEjebJ7X2eUChB0xCjXU42RzpdRswY9FKzx8zESP/hv8JcGzJQsqdXaSf24CnPMxoxG8EGgr6ky16f4WzQcmMFgpgZ7K45TLxpYvwbJtyV8o/ZqAwga6jrLI/a6PrBBuSONls5fuBLEYzTTBNNNtBCwYwwmHcUsm/v79fDro+tDeQ5vfS0YNBtFCQ1Np1uIVSf6zK32M0AuW5YPuvo7hrD5rRH4ehY8b8PTwDM2IDVdpqj19GfSRA3cBMmlRvEzOMjEtipprb0sjclkbOv+Bk2lq72bJ5F3f0J0COPb4ebKZp8LV/vZbzLzgFgE0vbSefK7FgUTO6rtPYVMvOHZ2kerJStU1MK1VLeh588EG2bdvGtm3bmDdvaMOqca6oE6LSx8doSqJpOmYiTrG1HaUgOm+239snEafc3YvVlar0/HELxUpl6HJXN+Xu3v5ERcMtlih3pwgk4hjBAJoRwskX9p40YGIEAjgly89bXJfo/Lm4xTJOsegXEdA0/9Myx8VMxNCDIfAUZk0NbrGIphvooRCB2gRuroCby6OZJsmTjsPq6aWcSuNkPb9/0MC/i4DpL4Hz/CRFDwZJrjyBUF0t5Z40nuP0l59m72MMfe/sk8LvE6Twixy4rl/Ku38WSg/6hQxGEkjEiTY3T4tPe4QYLxmXxJFgIBGJREIsXTKH3nSOvkwe1/VYsnQOV7zrTLq7u8n0riOV8m8HCAZN6hsSLD1qLsuWz688X31Dglg8THdXpjLTE4uFqW9ITMrrE+JgVS3pufbaa7n22mur9fRiBrNS6b17dPpnGwb6+LjFIiocwc3mMSIRQFV6+zjZHEY4SLCpvtLzRwv6szjK87AzWVAemmmAoaMcF7dYIpisIfmW47G6eih3p7FS/YmR6+Hk+pMgDTAMnGyOUGM9gf7lZWZtDU7eLxc965LzCCZrKrGXejN07G5l1rwWwnW1ZLe8jlssE100D90w2PX7e/EsG83QUUMaxe2dIQrUJEiuPIHZF76V3Bs7UXZ/AjYwy6Rp6BE/0fJsG9DQQwGU56EFTAKJOMp1cQtFPMf1l8cV/aV3wbokkRb5dFscOWRcEkeKVE+WfK7EsuPmUyxaFAtlUj19fPofruS44xfS2dnJmivOoTeV45mnX+WZta9iOy719QnWvPPMITM4c1saWXPFGdxz99Ps3NFJLBYedowQ08GE9ukR4kDSGzePuEcnWJ8kufJE+l7aRLk7jRkO0rB6JQqG9PapW3US8SULcPIF0us24GTzBGoTeJaN05cDXUczTYxoBDdfIJCsYdYl5xNfsgArlabY2k77/zyGne7z+/0owDAw43GcbB5jn/M62RyarhFffhTBZA3B+uTeRC1ZQyEWwu3spe2xp7GzBXRT95MUwE71ovZdd93fKFU3DD/OcIjC6zux3pJGNwyMeAylwOuv1KOZBk1nn0500TzSL75MYWcrynZxcjlQqn9GR0MzTYK1CYxwCGU7GJEIDatX/v/t3XtwVPX9//HXOXtLQhJyYVMxQADlC1+LgoYCAr+pRYp1YCq1Y2da66hlbHXQoaNVoE6HryOKiLZTbUdpnUGtDnWqrWjVqiPeL18FhIpcq6AQBEMum5DLbnbP5/dHyH4T2Nwgh92cfT5m0k42u9nXxMibl5/z+RwuYQMADzp+dabpaIuKSwq6rM4MP7NU5SPCmnjeGF06b6pqaxpVUlqQssx8Z875Gv/fI3t8DpDpKD3IGLHa+uRenM57dPLKz1CwpKjbPSd5nW5g2vFY0bkTujweq2/Q4X+9rnhLVE40qkRTsyyfT6XTL1D+2FGS2m+QGm9skj8vVznfGCanpVWOMXJaYyr7zoUKFBZ0eY+88jNUv+VTHd2zT427PlPzF/tPuKFnW+NRtWz6t+KRRsWbW+TE2tT68pvyD8k9dlPTTqxj/xOPy/h8so6tTsXqIqrbsl3Fk89RqKRI/vwhyRUby+9TKNx+Y8V4pEHBwvzkpX+x2vr2cmRZChYPPeFmrxQeAPCmnlZnUt3npGNPUG/fk7KDwYzSg4zRsW8nNKy4yx6deGNT8i/oqfacdF5d6azz48GSIsX/3zTVb9yqWH2jfLk5Kjz3v1V6YWWX1/gLhsiybcUjjfIPLZQVi8pXkKfcY8UrVluv5i+qkjf0bP7igOyAT/6CoSeUNElyWqJqa2xu31skyc4JKd54VG31cVmBgKwcW07rsaNwLVuB4vbvYxkjy+9r32eQMGrcvlvFk89R0ZRJyZUwyZKJJ3TknY8kYxRvbtGQinJZtt1+2Zrt09CJ4xUKlyTzd/wsAADexuoM0BWlBxmjo3BEv66Rf2ihTCwqX04wWTB6kmof0PGKzp2g+NEmRT7eJifuqHV/leo/2dllZaa56pASra2K1UWk6hoFi4tUNnuGgiVFJ1x6l1sxsteSZueGZPtttcXaZPl87UdVJxwZS1JcskMhydcmOUZ2bkjBwnxZlhQ/2iwn1tZ+AIFtK3qkVnVbtusbs2cor/wMtVQdUs17G2UH/O3vW1uvRHOLYjV1Cpa2r5IFCvJUNPkcSg4AZClWZ4D/wxFNyBgdhaO1ukZH/7NXbY3NKp4yqde/tNd/slMH17+sr17coIPrX1b9JztTPi9WW6+jO/8jX05IucPDkjHtKz+19cmv12/cqkDBEOWfPVo5xw5EyC0/44RL72SMmvZ83r4q1HhUxnGSByl0LmmBgnwNnfxNWZaUaG5pv+NxbkiWP9B+N+x4XJbPJysQkD8vT3YoqGGzpioULpUk2TlB+XJCsmxbTXs+V6y2XsGSIgUKC2QcI39BvizbVqikSL4huTKO035CnWX16WcHAACQDVjpQUboXDiCpcWKRxpk+f29ni7W2z6gznq7fO74r/tCweTXJaV8bcH4MWr+4kCXgxQkqfmLKtn5eZKk0ukXqK22XnUfbpHl88kOBBQ4o0hOtC15+ZlvSJ5sny+5UhWtqVP08BHJMbJ8loLfGCbjmGTWjtPsOp9cFyop0rBvX9jl+6T6ebGnBwAAZBtKDzJCT4Wjp7+c92UfUIdURaHzykxvX/flBBWrqZMdCsqJHjtdbvI3VTT5m8ki0Vx1SAfXv6xEa0x2TlBtY8qV3xpX4Tn/pdaqQzLx+LFL92LyBQMnXH4Wq63X1xveVfMXB9ovgXMc+Qrz5Qv4JctKZgmWFCX39xx/cl3n79W54HR3Mh4AAIDXUXqQEXorHAPxuu6KQucN/j1+PVyq+s3b2i9J8/tVdMHELq89ftWp6eBhRbZ8Kv17j4IFecopP0Ox6prk0dfHX35W/8lO1b63US1ffS3LthUoKlTiaLNiR2rlG16m0hlTujz/+BPqjv9enQtO/oSzdXTnf/q0IgYAAOA1lB5khN4Kx0C9rqei0NPXY7X1ilXXKPSNYfKFQkpEo4pV1yT32EhdV52ctrgSTc1STkCBoXmSMYpV16S8/Kzj/kBH3vxATiwmy1L7DUujMeWOHK54pFGlM6ZoaIpVmVQn16W65C/y8ady4gnlDg/3uiIGAADgNZQeZIzeCslAva67I657+voJl9ENyVXroWod3bNP+eNGn7DPRrZPTjQmKz9XdigoXyCg6JE62T6f8irKk9+3Y0Wm9VC12hqOyvL72vfxOO3Fx2luVWBoQa97m3rMWpCveFO1ZBxFvz7yf5fX9fFkPAAAgMGO0oOM0lshOdnXncwG/s6vOf4yuuaDhxWPNKrmfz9Ww/Zdyf0xHatO8cajsnw++UJB2QF/ysvuOlZkErE2OfGEJMk4TvvR1m0xWcYvy+/v9ylsqS75M46RE08cO4q7NnmzUlZ5AABANqD0wPM672+xbEtDxo1VcS/3r0m56f9YoWk9VK14pFGBwnzlDg932R/TedWpqeorHfx8r6JH6uVPcdldx4qMLzdHliTfkFw5za2yLEtWIKChk85R2UUX9ruYHH/Jn2XbkoxCxYXK+Ua4zyfjAQAAeAWlB57WeX+LbEuth4+o5auv1bRnr0pmVKY8vazzCowvN0eJaFT1G7fqzMsuUV75GTq6Z59q/vfjbvfHdHzkjByu1nCR8lvaZFvWCSWjY0UmEY1KPltqi8uXl3vs8ZyTKjwdOpevtoZGHXnno+Q9fXyhUvbzAACArELpgad1rKb4C4ao7dARWT6fJMnE4z3ezydaWy8nGpMSjuSzZYeCijc2Ka+iXPnjRqth+64+nRgXralXfMdnclIcE915RcYOBZWIJ2TnhBQozB+QG4t2lK9Ybf1JnYwHAADgFZQeeFpyf0ukQSYRl2xbdsAv/9BCxRuPplztcBIJJZpaJGNk5+bIaWlVIp6Qk2jfd9PXE+NidRE1/WevCns4JrrzioyTSHR7Y9Hu9iT1Za/SyZ6MBwAA4BWUHnhax1/4a9/bKCfhyDJSKFwqE4t2u9ph+3zy5eXKicVkojFZAb/sYFD2sVUiqW8nxsUbm+TE4snLyro7Jrq3Qxi6u6lox3194s2t8uflqGTGlG5vNnqyJ+MBAAB4AaUHntfxF/76LZ/q6J59MglHCnR/Kpq/YIhCpUVyojHZoWDy/48vSL2VFX/BENlBv+KNTQqc5GVlqe65U79xq/xD8lS94V21RRplWVK88aiqN7zb481GT/ZkPAAAgMGO0oOsECwpUtnsmSqa/M1+XQ7WsbpyMpeDBYuHasjZY6Qdn530ZWWp7rkTPVKn5n0HFKuLyA4EZAcDcmJtitVF1FJ1iGIDAABwHEoPskpfVzsG6nKw/NEjVDTuLDlHm0/q+6S6544vJyg7N3TsGea4/0/tZO5TBAAA4BWUHqAbA3U5WLB4qOzS4pPOkOoQgtzyM1T34Va1RRqUaGk99j5FKe+9092eIAAAgGxB6QEyXHerTuHZM1T73iYlWlrky81V6YzKlKe+pdoT1NPeHwAAAK+h9ACDQKpVp76eIJdqTxA3JgUAANmE0gNo8O556csJctyYFAAAZDtKD7Kel/e8cGNSAAAASg+yXDbseeHGpAAAINvZ6Q4ApFPHnhd/QX5yz0uiNaZ4Y1O6o52SWG29mr+oUqy2XlL7ik9eRTmFBwAAZCVWepDVvLjnxcuX6wEAAJwMVnqQ1Tr2vMiyFD1SJ1nWoN7zcvzlejJG9Ru3Jld8AAAAshErPch6XtrzwhHVAAAAJ6L0AOr96OfBwouX6wEAAJwqLm8DPMRrl+sBAAAMBFZ6AI/x0uV6AAAAA4HSA3iQVy7XAwAAGAhc3gYAAADA0yg9AAAAADyN0gMAAADA0yg9AAAAADyN0gMAAADA005L6YlGo5o8ebIsy9KWLVtOx1sCANAt5hIAZJfTUnpuu+02nXnmmafjrQAA6BVzCQCyi+ul56WXXtIrr7yi++67z+23AgCgV8wlAMg+rt6c9PDhw7ruuuv07LPPKi8vr9fnR6NRRaPR5OcNDQ2SJMdx5DiOazkHiuM4MsYMiqwSed1GXneR112DJWd/9XcuScym04287iKvu8jrrlPJ6VrpMcbommuu0fXXX68pU6Zo3759vb5m5cqVuuOOO054vLq6WrFYzIWUA8txHEUiERljZNuZf0YEed1FXneR112RSCTdEQbcycwlidl0upHXXeR1F3nddSqzqd+lZ+nSpVq1alWPz9mxY4deeeUVNTY2atmyZX3+3suWLdPNN9+c/LyhoUEjR45UOBxWUVFRf6Oedo7jyLIshcPhQfGLQ153kddd5HVXMBhMd4Q+c3MuScym04287iKvu8jrrlOZTf0uPbfccouuueaaHp8zduxYbdiwQe+//75CoVCXr02ZMkVXXnmlHnvssRNeFwqFTni+JNm2PSj+QUiSZVnkdRF53UVedw2mvIMhYwc355LEbEoH8rqLvO4ir3tOJWO/S084HFY4HO71eQ888IBWrFiR/PzgwYO65JJL9NRTT2natGn9fVsAAFJiLgEAeuPanp5Ro0Z1+Tw/P1+SdNZZZ2nEiBFuvS0AACkxlwAge2X+OhYAAAAAnAJXj6zubPTo0TLGnK63AwCgR8wlAMgerPQAAAAA8DRKDwAAAABPo/QAAAAA8DRKDwAAAABPo/QAAAAA8DRKDwAAAABPo/QAAAAA8DRKDwAAAABPo/QAAAAA8DRKDwAAAABPo/QAAAAA8DRKDwAAAABPo/QAAAAA8DRKDwAAAABPo/QAAAAA8DRKDwAAAABPo/QAAAAA8DRKDwAAAABPo/QAAAAA8DRKDwAAAABPo/QAAAAA8DRKDwAAAABPo/QAAAAA8DRKDwAAAABPo/QAAAAA8DRKDwAAAABPo/QAAAAA8DRKDwAAAABPo/QAAAAA8DRKDwAAAABPo/QAAAAA8DRKDwAAAABPo/QAAAAA8DRKDwAAAABPo/QAAAAA8DRKDwAAAABPo/QAAAAA8DRXS88LL7ygadOmKTc3V8XFxVqwYIGbbwcAQK+YTQCQffxufeNnnnlG1113ne6++27Nnj1b8Xhc27Ztc+vtAADoFbMJALKTK6UnHo9r8eLFWr16tRYuXJh8/JxzznHj7QAA6BWzCQCylyulZ/PmzaqqqpJt2zr//PN16NAhTZ48WatXr9bEiRO7fV00GlU0Gk1+3tDQIElyHEeO47gRdUA5jiNjzKDIKpHXbeR1F3ndNVhy9gezKfOzSuR1G3ndRV53nUpOV0rP559/Lkn6n//5H/32t7/V6NGjdf/99+uiiy7S7t27VVJSkvJ1K1eu1B133HHC49XV1YrFYm5EHVCO4ygSicgYI9vO/DMiyOsu8rqLvO6KRCLpjjDgmE2D43ePvO4ir7vI665TmU39Kj1Lly7VqlWrenzOjh07ki3s9ttv1w9/+ENJ0tq1azVixAj97W9/0y9+8YuUr122bJluvvnm5OcNDQ0aOXKkwuGwioqK+hM1LRzHkWVZCofDg+IXh7zuIq+7yOuuYDCY7gh9xmzq2WD73SOvu8jrLvK661RmU79Kzy233KJrrrmmx+eMHTtWX331laSu10mHQiGNHTtWX375ZbevDYVCCoVCJzxu2/ag+AchSZZlkddF5HUXed01mPIOhowdmE29G0y/exJ53UZed5HXPaeSsV+lJxwOKxwO9/q8yspKhUIh7dq1S7NmzZIktbW1ad++faqoqDi5pAAApMBsAgD0xpU9PYWFhbr++uu1fPlyjRw5UhUVFVq9erUk6YorrnDjLQEA6BGzCQCyl2v36Vm9erX8fr+uuuoqtbS0aNq0adqwYYOKi4vdeksAAHrEbAKA7ORa6QkEArrvvvt03333ufUWAAD0C7MJALJT5u9YAgAAAIBTQOkBAAAA4GmUHgAAAACeRukBAAAA4GmUHgAAAACeRukBAAAA4GmUHgAAAACeRukBAAAA4GmUHgAAAACeRukBAAAA4GmUHgAAAACeRukBAAAA4GmUHgAAAACeRukBAAAA4GmUHgAAAACeRukBAAAA4GmUHgAAAACeRukBAAAA4GmUHgAAAACeRukBAAAA4GmUHgAAAACeRukBAAAA4GmUHgAAAACeRukBAAAA4GmUHgAAAACeRukBAAAA4GmUHgAAAACeRukBAAAA4GmUHgAAAACeRukBAAAA4GmUHgAAAACeRukBAAAA4GmUHgAAAACeRukBAAAA4GmUHgAAAACeRukBAAAA4GmUHgAAAACe5lrp2b17ty677DINGzZMhYWFmjVrll5//XW33g4AgF4xmwAgO7lWeubPn694PK4NGzZo06ZNmjRpkubPn69Dhw659ZYAAPSI2QQA2cmV0nPkyBHt2bNHS5cu1Xnnnadx48bpnnvuUXNzs7Zt2+bGWwIA0CNmEwBkL78b37S0tFTjx4/X448/rgsuuEChUEhr1qxRWVmZKisru31dNBpVNBpNfh6JRCRJ9fX1bsQccI7jqKGhQcFgULad+dulyOsu8rqLvO7q+HPXGJPeIAOI2TQ4fvfI6y7yuou87jql2WRcsn//flNZWWksyzI+n88MHz7cbN68ucfXLF++3Ejigw8++OAjQz4+++wzt8ZEWjCb+OCDDz4G/8fJzCbLmL5XpaVLl2rVqlU9PmfHjh0aP368FixYoLa2Nt1+++3Kzc3VI488oueee04fffSRhg8fnvK1x//XtPr6elVUVOjLL7/U0KFD+xozbRoaGjRy5Ejt379fhYWF6Y7TK/K6i7zuIq+7IpGIRo0apbq6OhUVFaU7To+YTT0bbL975HUXed1FXnedymzqV+mprq5WTU1Nj88ZO3as3n77bc2dO1d1dXVdfoDjxo3TwoULtXTp0j69X0NDg4YOHapIJDIo/kGQ113kdRd53UVe9zCbekZed5HXXeR1Vzbl7deennA4rHA43OvzmpubJemEawNt25bjOP15SwAAesRsAgD0xpUdSxdeeKGKi4t19dVXa+vWrdq9e7duvfVW7d27V/PmzXPjLQEA6BGzCQCylyulZ9iwYfrXv/6lo0ePavbs2ZoyZYreeecdrV+/XpMmTerz9wmFQlq+fLlCoZAbMQcced1FXneR113kTT9mE3ndQF53kddd2ZS3X3t6AAAAAGCwyfwDuQEAAADgFFB6AAAAAHgapQcAAACAp1F6AAAAAHjaoCk9u3fv1mWXXaZhw4apsLBQs2bN0uuvv57uWD164YUXNG3aNOXm5qq4uFgLFixId6Q+iUajmjx5sizL0pYtW9IdJ6V9+/Zp4cKFGjNmjHJzc3XWWWdp+fLlisVi6Y6W9Mc//lGjR49WTk6Opk2bpg8//DDdkVJauXKlvvWtb6mgoEBlZWVasGCBdu3ale5YfXbPPffIsiz98pe/THeUblVVVemnP/2pSktLlZubq3PPPVcbN25Md6yUEomEfvOb33T5d+vOO+8UZ96kxmw6PZhLA4fZ5L7BMJek7JtNg6b0zJ8/X/F4XBs2bNCmTZs0adIkzZ8/X4cOHUp3tJSeeeYZXXXVVbr22mu1detWvfvuu/rJT36S7lh9ctttt+nMM89Md4we7dy5U47jaM2aNfr000/1u9/9Tg8//LB+/etfpzuaJOmpp57SzTffrOXLl2vz5s2aNGmSLrnkEn399dfpjnaCN998U4sWLdIHH3ygV199VW1tbZo7d66amprSHa1XH330kdasWaPzzjsv3VG6VVdXp5kzZyoQCOill17S9u3bdf/996u4uDjd0VJatWqVHnroIf3hD3/Qjh07tGrVKt1777168MEH0x0tIzGbTg/m0sBgNrlvMMwlKUtnkxkEqqurjSTz1ltvJR9raGgwksyrr76axmSptbW1mfLycvPII4+kO0q/vfjii2bChAnm008/NZLMxx9/nO5IfXbvvfeaMWPGpDuGMcaYqVOnmkWLFiU/TyQS5swzzzQrV65MY6q++frrr40k8+abb6Y7So8aGxvNuHHjzKuvvmq+/e1vm8WLF6c7UkpLliwxs2bNSneMPps3b5752c9+1uWxyy+/3Fx55ZVpSpS5mE2nB3Np4DCb3DVY5pIx2TmbBsVKT2lpqcaPH6/HH39cTU1NisfjWrNmjcrKylRZWZnueCfYvHmzqqqqZNu2zj//fA0fPlyXXnqptm3blu5oPTp8+LCuu+46/eUvf1FeXl664/RbJBJRSUlJumMoFotp06ZNmjNnTvIx27Y1Z84cvf/++2lM1jeRSESSMuJn2ZNFixZp3rx5XX7Omei5557TlClTdMUVV6isrEznn3++/vznP6c7VrdmzJih1157Tbt375Ykbd26Ve+8844uvfTSNCfLPMwm9zGXBg6zyX2DZS5JWTqbBrqJuWX//v2msrLSWJZlfD6fGT58uNm8eXO6Y6W0bt06I8mMGjXKPP3002bjxo3mxz/+sSktLTU1NTXpjpeS4zjme9/7nrnzzjuNMcbs3bt3UP0XtT179pjCwkLzpz/9Kd1RTFVVlZFk3nvvvS6P33rrrWbq1KlpStU3iUTCzJs3z8ycOTPdUXq0bt06M3HiRNPS0mKMMRn9X9RCoZAJhUJm2bJlZvPmzWbNmjUmJyfHPProo+mOllIikTBLliwxlmUZv99vLMsyd999d7pjZSxmk3uYSwOL2eSuwTSXjMnO2ZTW0rNkyRIjqcePHTt2GMdxzPe//31z6aWXmnfeecds2rTJ3HDDDaa8vNwcPHgw4/I++eSTRpJZs2ZN8rWtra1m2LBh5uGHHz5tefuT+fe//72ZOXOmicfjxpj0DZe+5u3swIED5qyzzjILFy48rVm7M5gHy/XXX28qKirM/v370x2lW19++aUpKyszW7duTT6WycMlEAiYCy+8sMtjN910k5k+fXqaEvVs3bp1ZsSIEWbdunXm3//+t3n88cdNSUlJxg5CNzCbMiMvc2lgMZvcM9jmkjHZOZssY9J3JE91dbVqamp6fM7YsWP19ttva+7cuaqrq1NhYWHya+PGjdPChQu1dOlSt6NK6nved999V7Nnz9bbb7+tWbNmJb82bdo0zZkzR3fddZfbUZP6mvlHP/qRnn/+eVmWlXw8kUjI5/Ppyiuv1GOPPeZ2VEl9zxsMBiVJBw8e1EUXXaTp06fr0UcflW2n/4rNWCymvLw8Pf30011ORbr66qtVX1+v9evXpy9cD2688UatX79eb731lsaMGZPuON169tln9YMf/EA+ny/5WCKRkGVZsm1b0Wi0y9fSraKiQt/97nf1yCOPJB976KGHtGLFClVVVaUxWWojR47U0qVLtWjRouRjK1as0BNPPKGdO3emMdnpw2xyF3MpPZhN7hlsc0nKztnkdytcX4TDYYXD4V6f19zcLEkn/MFh27Ycx3ElWyp9zVtZWalQKKRdu3YlB0tbW5v27duniooKt2N20dfMDzzwgFasWJH8/ODBg7rkkkv01FNPadq0aW5G7KKveaX2oxa/853vqLKyUmvXrs2YwRIMBlVZWanXXnstOVgcx9Frr72mG2+8Mb3hUjDG6KabbtI//vEPvfHGGxk7VDpcfPHF+uSTT7o8du2112rChAlasmRJxg2WmTNnnnDM6u7du0/7nwV91dzcfMK/Sz6f77T+WZtuzCZ3MZfSg9nknsE2l6QsnU0urUINqOrqalNaWmouv/xys2XLFrNr1y7zq1/9ygQCAbNly5Z0x0tp8eLFpry83Lz88stm586dZuHChaasrMzU1tamO1qfZPq10wcOHDBnn322ufjii82BAwfMV199lfzIBH/9619NKBQyjz76qNm+fbv5+c9/boqKisyhQ4fSHe0EN9xwgxk6dKh54403uvwcm5ub0x2tzzL5MoIPP/zQ+P1+c9ddd5k9e/aYJ5980uTl5Zknnngi3dFSuvrqq015ebn55z//afbu3Wv+/ve/m2HDhpnbbrst3dEyDrPp9GIunTpm0+mTyXPJmOycTYOi9BhjzEcffWTmzp1rSkpKTEFBgZk+fbp58cUX0x2rW7FYzNxyyy2mrKzMFBQUmDlz5pht27alO1afZfpwWbt2bbfXVmeKBx980IwaNcoEg0EzdepU88EHH6Q7Ukrd/RzXrl2b7mh9lunD5fnnnzcTJ040oVDITJgwIWM2NqfS0NBgFi9ebEaNGmVycnLM2LFjze23326i0Wi6o2UkZtPpw1waGMym0yPT55Ix2Teb0rqnBwAAAADcljkXmwIAAACACyg9AAAAADyN0gMAAADA0yg9AAAAADyN0gMAAADA0yg9AAAAADyN0gMAAADA0yg9AAAAADyN0gMAAADA0yg9AAAAADyN0gMAAADA0yg9AAAAADzt/wOk313t5fg42AAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 1000x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_source_and_target(pred_bridge[0, :, :], pred_bridge[-1, :, :], left_title=\"Source Sample\", right_title=\"Predict Sample\", save_path=log_dir / 'pred_bridge.jpg')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_source_and_target(test_bridge[0, :, :], test_bridge[-1, :, :], save_path=log_dir / 'gt_test_bridge.jpg')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d6862536b59e46e69e04dfee2a532eac",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Output()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"></pre>\n"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "save_gif_frame(pred_bridge, log_dir)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "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": "3.7.16"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}