test_ppo_implementation.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First, let's run some CartPole PPO"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/Users/theeimer/anaconda3/envs/mighty/lib/python3.10/site-packages/pygame/pkgdata.py:25: DeprecationWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html\n",
      "  from pkg_resources import resource_stream, resource_exists\n",
      "/Users/theeimer/anaconda3/envs/mighty/lib/python3.10/site-packages/pkg_resources/__init__.py:2825: DeprecationWarning: Deprecated call to `pkg_resources.declare_namespace('sphinxcontrib')`.\n",
      "Implementing implicit namespace packages (as specified in PEP 420) is preferred to `pkg_resources.declare_namespace`. See https://setuptools.pypa.io/en/latest/references/keywords.html#keyword-namespace-packages\n",
      "  declare_namespace(pkg)\n",
      "/Users/theeimer/anaconda3/envs/mighty/lib/python3.10/site-packages/dacbench/envs/__init__.py:34: UserWarning: CMA-ES Benchmark not installed. If you want to use this benchmark, please follow the installation guide.\n",
      "  warnings.warn(  # noqa: B028\n",
      "/Users/theeimer/anaconda3/envs/mighty/lib/python3.10/site-packages/dacbench/benchmarks/__init__.py:26: UserWarning: CMA-ES Benchmark not installed. If you want to use this benchmark, please follow the installation guide.\n",
      "  warnings.warn(  # noqa: B028\n",
      "/Users/theeimer/anaconda3/envs/mighty/lib/python3.10/site-packages/carl/envs/__init__.py:36: UserWarning: Module py4j not found. If you want to use these environments, please follow the installation guide.\n",
      "  warnings.warn(\n",
      "/Users/theeimer/anaconda3/envs/mighty/lib/python3.10/site-packages/carl/envs/__init__.py:36: UserWarning: Module distance not found. If you want to use these environments, please follow the installation guide.\n",
      "  warnings.warn(\n",
      "[\u001b[36m2024-10-02 14:52:47,018\u001b[0m][\u001b[35mHYDRA\u001b[0m] Launching 5 jobs locally\u001b[0m\n",
      "[\u001b[36m2024-10-02 14:52:47,018\u001b[0m][\u001b[35mHYDRA\u001b[0m] \t#0 : experiment_name=cartpole_ppo algorithm=ppo environment=gymnasium/cartpole seed=0\u001b[0m\n",
      "\u001b[2;36m[10/02/24 14:52:47]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m Output will be written to       \u001b]8;id=458825;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py\u001b\\\u001b[2mmighty_runner.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=565808;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py#33\u001b\\\u001b[2m33\u001b[0m\u001b]8;;\u001b\\\n",
      "\u001b[2;36m                    \u001b[0m         runs/cartpole_ppo_0             \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                   \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m ############################### \u001b]8;id=461860;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py\u001b\\\u001b[2mmighty_runner.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=74189;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py#76\u001b\\\u001b[2m76\u001b[0m\u001b]8;;\u001b\\\n",
      "\u001b[2;36m                    \u001b[0m         ############################### \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                    \u001b[0m         ##################              \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                   \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m Using agent type                \u001b]8;id=385220;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py\u001b\\\u001b[2mmighty_runner.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=806438;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py#77\u001b\\\u001b[2m77\u001b[0m\u001b]8;;\u001b\\\n",
      "\u001b[2;36m                    \u001b[0m         \u001b[32m\"\u001b[0m\u001b[32m<\u001b[0m\u001b[32mmighty.mighty_agents.ppo.Migh\u001b[0m \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                    \u001b[0m         \u001b[32mtyPPOAgent\u001b[0m\u001b[32m object at \u001b[0m           \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                    \u001b[0m         \u001b[32m0x334f42a10\u001b[0m\u001b[32m>\u001b[0m\u001b[32m\"\u001b[0m to learn          \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                   \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m ############################### \u001b]8;id=184996;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py\u001b\\\u001b[2mmighty_runner.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=403000;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py#78\u001b\\\u001b[2m78\u001b[0m\u001b]8;;\u001b\\\n",
      "\u001b[2;36m                    \u001b[0m         ############################### \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                    \u001b[0m         ##################              \u001b[2m                   \u001b[0m\n",
      "\u001b[2KTrain Steps \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m  0%\u001b[0m Remaining: \u001b[36m-:--:--\u001b[0m Elapsed: \u001b[33m0:00:00\u001b[0m\n",
      "\u001b[?25hSteps: \u001b[1;36m5000\u001b[0m, Latest Episode Reward: \u001b[1;36m32.0\u001b[0m, Latest Step Reward: \u001b[1;36m8.0\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_0/checkpoints/5000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m10000\u001b[0m steps:\n",
      "                \u001b[1;36m16.1\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m10000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m10000\u001b[0m, Latest Episode Reward: \u001b[1;36m12.0\u001b[0m, Latest Step Reward: \u001b[1;36m3.0\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_0/checkpoints/10000\n",
      "Steps: \u001b[1;36m15000\u001b[0m, Latest Episode Reward: \u001b[1;36m15.0\u001b[0m, Latest Step Reward: \u001b[1;36m3.75\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_0/checkpoints/15000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m20000\u001b[0m steps:\n",
      "                \u001b[1;36m20.8\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m20000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m20000\u001b[0m, Latest Episode Reward: \u001b[1;36m75.0\u001b[0m, Latest Step Reward: \u001b[1;36m18.75\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_0/checkpoints/20000\n",
      "Steps: \u001b[1;36m25000\u001b[0m, Latest Episode Reward: \u001b[1;36m14.0\u001b[0m, Latest Step Reward: \u001b[1;36m3.5\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_0/checkpoints/25000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m30000\u001b[0m steps:\n",
      "                \u001b[1;36m18.3\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m30000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m30000\u001b[0m, Latest Episode Reward: \u001b[1;36m31.0\u001b[0m, Latest Step Reward: \u001b[1;36m7.75\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_0/checkpoints/30000\n",
      "Steps: \u001b[1;36m35000\u001b[0m, Latest Episode Reward: \u001b[1;36m19.0\u001b[0m, Latest Step Reward: \u001b[1;36m4.75\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_0/checkpoints/35000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m40000\u001b[0m steps:\n",
      "                \u001b[1;36m26.7\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m40000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m40000\u001b[0m, Latest Episode Reward: \u001b[1;36m35.0\u001b[0m, Latest Step Reward: \u001b[1;36m8.75\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_0/checkpoints/40000\n",
      "Steps: \u001b[1;36m45000\u001b[0m, Latest Episode Reward: \u001b[1;36m16.0\u001b[0m, Latest Step Reward: \u001b[1;36m4.0\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_0/checkpoints/45000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m50000\u001b[0m steps:\n",
      "                \u001b[1;36m23.1\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m50000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m50000\u001b[0m, Latest Episode Reward: \u001b[1;36m21.0\u001b[0m, Latest Step Reward: \u001b[1;36m5.25\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_0/checkpoints/50000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m50000\u001b[0m steps:\n",
      "                \u001b[1;36m26.6\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m50000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Training finished!\n",
      "Reached a reward of 26.6 in 50000 steps and 18.66s.\n",
      "[\u001b[36m2024-10-02 14:53:05,777\u001b[0m][\u001b[35mHYDRA\u001b[0m] \t#1 : experiment_name=cartpole_ppo algorithm=ppo environment=gymnasium/cartpole seed=1\u001b[0m\n",
      "\u001b[2;36m[10/02/24 14:53:05]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m Output will be written to       \u001b]8;id=610129;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py\u001b\\\u001b[2mmighty_runner.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=562112;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py#33\u001b\\\u001b[2m33\u001b[0m\u001b]8;;\u001b\\\n",
      "\u001b[2;36m                    \u001b[0m         runs/cartpole_ppo_1             \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                   \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m ############################### \u001b]8;id=689290;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py\u001b\\\u001b[2mmighty_runner.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=859117;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py#76\u001b\\\u001b[2m76\u001b[0m\u001b]8;;\u001b\\\n",
      "\u001b[2;36m                    \u001b[0m         ############################### \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                    \u001b[0m         ##################              \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                   \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m Using agent type                \u001b]8;id=575027;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py\u001b\\\u001b[2mmighty_runner.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=100817;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py#77\u001b\\\u001b[2m77\u001b[0m\u001b]8;;\u001b\\\n",
      "\u001b[2;36m                    \u001b[0m         \u001b[32m\"\u001b[0m\u001b[32m<\u001b[0m\u001b[32mmighty.mighty_agents.ppo.Migh\u001b[0m \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                    \u001b[0m         \u001b[32mtyPPOAgent\u001b[0m\u001b[32m object at \u001b[0m           \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                    \u001b[0m         \u001b[32m0x33fbb9ed0\u001b[0m\u001b[32m>\u001b[0m\u001b[32m\"\u001b[0m to learn          \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                   \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m ############################### \u001b]8;id=47300;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py\u001b\\\u001b[2mmighty_runner.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=336360;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py#78\u001b\\\u001b[2m78\u001b[0m\u001b]8;;\u001b\\\n",
      "\u001b[2;36m                    \u001b[0m         ############################### \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                    \u001b[0m         ##################              \u001b[2m                   \u001b[0m\n",
      "\u001b[2KTrain Steps \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m  0%\u001b[0m Remaining: \u001b[36m-:--:--\u001b[0m Elapsed: \u001b[33m0:00:00\u001b[0m\n",
      "\u001b[?25hSteps: \u001b[1;36m5000\u001b[0m, Latest Episode Reward: \u001b[1;36m26.0\u001b[0m, Latest Step Reward: \u001b[1;36m6.5\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_1/checkpoints/5000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m10000\u001b[0m steps:\n",
      "                \u001b[1;36m31.0\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m10000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m10000\u001b[0m, Latest Episode Reward: \u001b[1;36m42.0\u001b[0m, Latest Step Reward: \u001b[1;36m10.5\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_1/checkpoints/10000\n",
      "Steps: \u001b[1;36m15000\u001b[0m, Latest Episode Reward: \u001b[1;36m12.0\u001b[0m, Latest Step Reward: \u001b[1;36m3.0\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_1/checkpoints/15000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m20000\u001b[0m steps:\n",
      "                \u001b[1;36m23.1\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m20000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m20000\u001b[0m, Latest Episode Reward: \u001b[1;36m11.0\u001b[0m, Latest Step Reward: \u001b[1;36m2.75\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_1/checkpoints/20000\n",
      "Steps: \u001b[1;36m25000\u001b[0m, Latest Episode Reward: \u001b[1;36m20.0\u001b[0m, Latest Step Reward: \u001b[1;36m5.0\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_1/checkpoints/25000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m30000\u001b[0m steps:\n",
      "                \u001b[1;36m19.3\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m30000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m30000\u001b[0m, Latest Episode Reward: \u001b[1;36m19.0\u001b[0m, Latest Step Reward: \u001b[1;36m4.75\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_1/checkpoints/30000\n",
      "Steps: \u001b[1;36m35000\u001b[0m, Latest Episode Reward: \u001b[1;36m18.0\u001b[0m, Latest Step Reward: \u001b[1;36m4.5\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_1/checkpoints/35000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m40000\u001b[0m steps:\n",
      "                \u001b[1;36m19.5\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m40000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m40000\u001b[0m, Latest Episode Reward: \u001b[1;36m16.0\u001b[0m, Latest Step Reward: \u001b[1;36m4.0\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_1/checkpoints/40000\n",
      "Steps: \u001b[1;36m45000\u001b[0m, Latest Episode Reward: \u001b[1;36m11.0\u001b[0m, Latest Step Reward: \u001b[1;36m2.75\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_1/checkpoints/45000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m50000\u001b[0m steps:\n",
      "                \u001b[1;36m22.5\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m50000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m50000\u001b[0m, Latest Episode Reward: \u001b[1;36m12.0\u001b[0m, Latest Step Reward: \u001b[1;36m3.0\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_1/checkpoints/50000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m50000\u001b[0m steps:\n",
      "                \u001b[1;36m20.8\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m50000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Training finished!\n",
      "Reached a reward of 20.8 in 50000 steps and 18.94s.\n",
      "[\u001b[36m2024-10-02 14:53:24,799\u001b[0m][\u001b[35mHYDRA\u001b[0m] \t#2 : experiment_name=cartpole_ppo algorithm=ppo environment=gymnasium/cartpole seed=2\u001b[0m\n",
      "\u001b[2;36m[10/02/24 14:53:24]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m Output will be written to       \u001b]8;id=310513;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py\u001b\\\u001b[2mmighty_runner.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=301609;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py#33\u001b\\\u001b[2m33\u001b[0m\u001b]8;;\u001b\\\n",
      "\u001b[2;36m                    \u001b[0m         runs/cartpole_ppo_2             \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                   \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m ############################### \u001b]8;id=9726;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py\u001b\\\u001b[2mmighty_runner.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=259740;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py#76\u001b\\\u001b[2m76\u001b[0m\u001b]8;;\u001b\\\n",
      "\u001b[2;36m                    \u001b[0m         ############################### \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                    \u001b[0m         ##################              \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                   \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m Using agent type                \u001b]8;id=598773;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py\u001b\\\u001b[2mmighty_runner.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=45250;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py#77\u001b\\\u001b[2m77\u001b[0m\u001b]8;;\u001b\\\n",
      "\u001b[2;36m                    \u001b[0m         \u001b[32m\"\u001b[0m\u001b[32m<\u001b[0m\u001b[32mmighty.mighty_agents.ppo.Migh\u001b[0m \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                    \u001b[0m         \u001b[32mtyPPOAgent\u001b[0m\u001b[32m object at \u001b[0m           \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                    \u001b[0m         \u001b[32m0x33fb84760\u001b[0m\u001b[32m>\u001b[0m\u001b[32m\"\u001b[0m to learn          \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                   \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m ############################### \u001b]8;id=448494;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py\u001b\\\u001b[2mmighty_runner.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=882221;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py#78\u001b\\\u001b[2m78\u001b[0m\u001b]8;;\u001b\\\n",
      "\u001b[2;36m                    \u001b[0m         ############################### \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                    \u001b[0m         ##################              \u001b[2m                   \u001b[0m\n",
      "\u001b[2KTrain Steps \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m  0%\u001b[0m Remaining: \u001b[36m-:--:--\u001b[0m Elapsed: \u001b[33m0:00:00\u001b[0m\n",
      "\u001b[?25hSteps: \u001b[1;36m5000\u001b[0m, Latest Episode Reward: \u001b[1;36m26.0\u001b[0m, Latest Step Reward: \u001b[1;36m6.5\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_2/checkpoints/5000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m10000\u001b[0m steps:\n",
      "                \u001b[1;36m22.3\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m10000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m10000\u001b[0m, Latest Episode Reward: \u001b[1;36m26.0\u001b[0m, Latest Step Reward: \u001b[1;36m6.5\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_2/checkpoints/10000\n",
      "Steps: \u001b[1;36m15000\u001b[0m, Latest Episode Reward: \u001b[1;36m18.0\u001b[0m, Latest Step Reward: \u001b[1;36m4.5\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_2/checkpoints/15000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m20000\u001b[0m steps:\n",
      "                \u001b[1;36m21.3\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m20000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m20000\u001b[0m, Latest Episode Reward: \u001b[1;36m16.0\u001b[0m, Latest Step Reward: \u001b[1;36m4.0\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_2/checkpoints/20000\n",
      "Steps: \u001b[1;36m25000\u001b[0m, Latest Episode Reward: \u001b[1;36m12.0\u001b[0m, Latest Step Reward: \u001b[1;36m3.0\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_2/checkpoints/25000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m30000\u001b[0m steps:\n",
      "                \u001b[1;36m17.6\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m30000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m30000\u001b[0m, Latest Episode Reward: \u001b[1;36m48.0\u001b[0m, Latest Step Reward: \u001b[1;36m12.0\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_2/checkpoints/30000\n",
      "Steps: \u001b[1;36m35000\u001b[0m, Latest Episode Reward: \u001b[1;36m25.0\u001b[0m, Latest Step Reward: \u001b[1;36m6.25\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_2/checkpoints/35000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m40000\u001b[0m steps:\n",
      "                \u001b[1;36m25.0\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m40000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m40000\u001b[0m, Latest Episode Reward: \u001b[1;36m16.0\u001b[0m, Latest Step Reward: \u001b[1;36m4.0\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_2/checkpoints/40000\n",
      "Steps: \u001b[1;36m45000\u001b[0m, Latest Episode Reward: \u001b[1;36m18.0\u001b[0m, Latest Step Reward: \u001b[1;36m4.5\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_2/checkpoints/45000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m50000\u001b[0m steps:\n",
      "                \u001b[1;36m22.0\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m50000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m50000\u001b[0m, Latest Episode Reward: \u001b[1;36m15.0\u001b[0m, Latest Step Reward: \u001b[1;36m3.75\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_2/checkpoints/50000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m50000\u001b[0m steps:\n",
      "                \u001b[1;36m22.6\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m50000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Training finished!\n",
      "Reached a reward of 22.6 in 50000 steps and 18.61s.\n",
      "[\u001b[36m2024-10-02 14:53:43,494\u001b[0m][\u001b[35mHYDRA\u001b[0m] \t#3 : experiment_name=cartpole_ppo algorithm=ppo environment=gymnasium/cartpole seed=3\u001b[0m\n",
      "\u001b[2;36m[10/02/24 14:53:43]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m Output will be written to       \u001b]8;id=152505;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py\u001b\\\u001b[2mmighty_runner.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=880599;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py#33\u001b\\\u001b[2m33\u001b[0m\u001b]8;;\u001b\\\n",
      "\u001b[2;36m                    \u001b[0m         runs/cartpole_ppo_3             \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                   \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m ############################### \u001b]8;id=926609;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py\u001b\\\u001b[2mmighty_runner.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=876475;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py#76\u001b\\\u001b[2m76\u001b[0m\u001b]8;;\u001b\\\n",
      "\u001b[2;36m                    \u001b[0m         ############################### \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                    \u001b[0m         ##################              \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                   \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m Using agent type                \u001b]8;id=170175;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py\u001b\\\u001b[2mmighty_runner.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=299291;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py#77\u001b\\\u001b[2m77\u001b[0m\u001b]8;;\u001b\\\n",
      "\u001b[2;36m                    \u001b[0m         \u001b[32m\"\u001b[0m\u001b[32m<\u001b[0m\u001b[32mmighty.mighty_agents.ppo.Migh\u001b[0m \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                    \u001b[0m         \u001b[32mtyPPOAgent\u001b[0m\u001b[32m object at \u001b[0m           \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                    \u001b[0m         \u001b[32m0x33fced3f0\u001b[0m\u001b[32m>\u001b[0m\u001b[32m\"\u001b[0m to learn          \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                   \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m ############################### \u001b]8;id=939485;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py\u001b\\\u001b[2mmighty_runner.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=918336;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py#78\u001b\\\u001b[2m78\u001b[0m\u001b]8;;\u001b\\\n",
      "\u001b[2;36m                    \u001b[0m         ############################### \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                    \u001b[0m         ##################              \u001b[2m                   \u001b[0m\n",
      "\u001b[2KTrain Steps \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m  0%\u001b[0m Remaining: \u001b[36m-:--:--\u001b[0m Elapsed: \u001b[33m0:00:00\u001b[0m\n",
      "\u001b[?25hSteps: \u001b[1;36m5000\u001b[0m, Latest Episode Reward: \u001b[1;36m19.0\u001b[0m, Latest Step Reward: \u001b[1;36m4.75\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_3/checkpoints/5000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m10000\u001b[0m steps:\n",
      "                \u001b[1;36m27.2\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m10000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m10000\u001b[0m, Latest Episode Reward: \u001b[1;36m17.0\u001b[0m, Latest Step Reward: \u001b[1;36m4.25\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_3/checkpoints/10000\n",
      "Steps: \u001b[1;36m15000\u001b[0m, Latest Episode Reward: \u001b[1;36m34.0\u001b[0m, Latest Step Reward: \u001b[1;36m8.5\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_3/checkpoints/15000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m20000\u001b[0m steps:\n",
      "                \u001b[1;36m26.3\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m20000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m20000\u001b[0m, Latest Episode Reward: \u001b[1;36m24.0\u001b[0m, Latest Step Reward: \u001b[1;36m6.0\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_3/checkpoints/20000\n",
      "Steps: \u001b[1;36m25000\u001b[0m, Latest Episode Reward: \u001b[1;36m16.0\u001b[0m, Latest Step Reward: \u001b[1;36m4.0\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_3/checkpoints/25000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m30000\u001b[0m steps:\n",
      "                \u001b[1;36m21.2\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m30000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m30000\u001b[0m, Latest Episode Reward: \u001b[1;36m10.0\u001b[0m, Latest Step Reward: \u001b[1;36m2.5\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_3/checkpoints/30000\n",
      "Steps: \u001b[1;36m35000\u001b[0m, Latest Episode Reward: \u001b[1;36m20.0\u001b[0m, Latest Step Reward: \u001b[1;36m5.0\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_3/checkpoints/35000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m40000\u001b[0m steps:\n",
      "                \u001b[1;36m32.3\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m40000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m40000\u001b[0m, Latest Episode Reward: \u001b[1;36m17.0\u001b[0m, Latest Step Reward: \u001b[1;36m4.25\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_3/checkpoints/40000\n",
      "Steps: \u001b[1;36m45000\u001b[0m, Latest Episode Reward: \u001b[1;36m13.0\u001b[0m, Latest Step Reward: \u001b[1;36m3.25\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_3/checkpoints/45000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m50000\u001b[0m steps:\n",
      "                \u001b[1;36m21.1\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m50000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m50000\u001b[0m, Latest Episode Reward: \u001b[1;36m18.0\u001b[0m, Latest Step Reward: \u001b[1;36m4.5\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_3/checkpoints/50000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m50000\u001b[0m steps:\n",
      "                \u001b[1;36m17.6\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m50000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Training finished!\n",
      "Reached a reward of 17.6 in 50000 steps and 18.94s.\n",
      "[\u001b[36m2024-10-02 14:54:02,516\u001b[0m][\u001b[35mHYDRA\u001b[0m] \t#4 : experiment_name=cartpole_ppo algorithm=ppo environment=gymnasium/cartpole seed=4\u001b[0m\n",
      "\u001b[2;36m[10/02/24 14:54:02]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m Output will be written to       \u001b]8;id=379081;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py\u001b\\\u001b[2mmighty_runner.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=804087;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py#33\u001b\\\u001b[2m33\u001b[0m\u001b]8;;\u001b\\\n",
      "\u001b[2;36m                    \u001b[0m         runs/cartpole_ppo_4             \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                   \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m ############################### \u001b]8;id=53612;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py\u001b\\\u001b[2mmighty_runner.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=698031;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py#76\u001b\\\u001b[2m76\u001b[0m\u001b]8;;\u001b\\\n",
      "\u001b[2;36m                    \u001b[0m         ############################### \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                    \u001b[0m         ##################              \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                   \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m Using agent type                \u001b]8;id=328663;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py\u001b\\\u001b[2mmighty_runner.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=265789;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py#77\u001b\\\u001b[2m77\u001b[0m\u001b]8;;\u001b\\\n",
      "\u001b[2;36m                    \u001b[0m         \u001b[32m\"\u001b[0m\u001b[32m<\u001b[0m\u001b[32mmighty.mighty_agents.ppo.Migh\u001b[0m \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                    \u001b[0m         \u001b[32mtyPPOAgent\u001b[0m\u001b[32m object at \u001b[0m           \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                    \u001b[0m         \u001b[32m0x33fd5ba30\u001b[0m\u001b[32m>\u001b[0m\u001b[32m\"\u001b[0m to learn          \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                   \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m ############################### \u001b]8;id=580858;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py\u001b\\\u001b[2mmighty_runner.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=306795;file:///Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_runners/mighty_runner.py#78\u001b\\\u001b[2m78\u001b[0m\u001b]8;;\u001b\\\n",
      "\u001b[2;36m                    \u001b[0m         ############################### \u001b[2m                   \u001b[0m\n",
      "\u001b[2;36m                    \u001b[0m         ##################              \u001b[2m                   \u001b[0m\n",
      "\u001b[2KTrain Steps \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[35m  0%\u001b[0m Remaining: \u001b[36m-:--:--\u001b[0m Elapsed: \u001b[33m0:00:00\u001b[0m\n",
      "\u001b[?25hSteps: \u001b[1;36m5000\u001b[0m, Latest Episode Reward: \u001b[1;36m46.0\u001b[0m, Latest Step Reward: \u001b[1;36m11.5\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_4/checkpoints/5000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m10000\u001b[0m steps:\n",
      "                \u001b[1;36m20.8\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m10000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m10000\u001b[0m, Latest Episode Reward: \u001b[1;36m34.0\u001b[0m, Latest Step Reward: \u001b[1;36m8.5\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_4/checkpoints/10000\n",
      "Steps: \u001b[1;36m15000\u001b[0m, Latest Episode Reward: \u001b[1;36m13.0\u001b[0m, Latest Step Reward: \u001b[1;36m3.25\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_4/checkpoints/15000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m20000\u001b[0m steps:\n",
      "                \u001b[1;36m19.8\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m20000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m20000\u001b[0m, Latest Episode Reward: \u001b[1;36m17.0\u001b[0m, Latest Step Reward: \u001b[1;36m4.25\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_4/checkpoints/20000\n",
      "Steps: \u001b[1;36m25000\u001b[0m, Latest Episode Reward: \u001b[1;36m31.0\u001b[0m, Latest Step Reward: \u001b[1;36m7.75\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_4/checkpoints/25000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m30000\u001b[0m steps:\n",
      "                \u001b[1;36m23.8\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m30000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m30000\u001b[0m, Latest Episode Reward: \u001b[1;36m17.0\u001b[0m, Latest Step Reward: \u001b[1;36m4.25\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_4/checkpoints/30000\n",
      "Steps: \u001b[1;36m35000\u001b[0m, Latest Episode Reward: \u001b[1;36m16.0\u001b[0m, Latest Step Reward: \u001b[1;36m4.0\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_4/checkpoints/35000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m40000\u001b[0m steps:\n",
      "                \u001b[1;36m18.6\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m40000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m40000\u001b[0m, Latest Episode Reward: \u001b[1;36m16.0\u001b[0m, Latest Step Reward: \u001b[1;36m4.0\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_4/checkpoints/40000\n",
      "Steps: \u001b[1;36m45000\u001b[0m, Latest Episode Reward: \u001b[1;36m14.0\u001b[0m, Latest Step Reward: \u001b[1;36m3.5\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_4/checkpoints/45000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m50000\u001b[0m steps:\n",
      "                \u001b[1;36m25.0\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m50000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Steps: \u001b[1;36m50000\u001b[0m, Latest Episode Reward: \u001b[1;36m16.0\u001b[0m, Latest Step Reward: \u001b[1;36m4.0\u001b[0m\n",
      "Saved checkpoint at runs/cartpole_ppo_4/checkpoints/50000\n",
      "\n",
      "------------------------------------------------------------------------------\n",
      "Evaluation performance after \u001b[1;36m50000\u001b[0m steps:\n",
      "                \u001b[1;36m20.3\u001b[0m\n",
      "Evaluation performance per step after \u001b[1;36m50000\u001b[0m steps:\n",
      "                \u001b[1;36m1.0\u001b[0m\n",
      "------------------------------------------------------------------------------\n",
      "\n",
      "Training finished!\n",
      "Reached a reward of 20.3 in 50000 steps and 18.91s.\n"
     ]
    }
   ],
   "source": [
    "!python mighty/run_mighty.py experiment_name=cartpole_ppo_latest algorithm=ppo environment=gymnasium/cartpole seed=0,1,2,3,4 -m"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then let's load the data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "import pandas as pd\n",
    "from mighty.mighty_utils.logger import to_dataframe\n",
    "\n",
    "def load_results(experiment_name, seeds=[0,1,2,3,4]):\n",
    "    dfs = []\n",
    "    for s in seeds:\n",
    "        logs = []\n",
    "        with open(f\"runs/{experiment_name}_{s}/eval.jsonl\", \"r\") as f:\n",
    "            for line in f:\n",
    "                logs.append(json.loads(line))\n",
    "        df = to_dataframe(logs)\n",
    "        df[\"seed\"] = s\n",
    "        dfs.append(df)\n",
    "    return pd.concat(dfs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:193: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][(dataframe[\"total_steps\"] < i) & (dataframe[\"total_steps\"] > last_step)] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: FutureWarning: ChainedAssignmentError: behaviour will change in pandas 3.0!\n",
      "You are setting values through chained assignment. Currently this works in certain cases, but when using Copy-on-Write (which will become the default behaviour in pandas 3.0) this will never work to update the original DataFrame or Series, because the intermediate object on which we are setting values will behave as a copy.\n",
      "A typical example is when you are setting values in a column of a DataFrame, like:\n",
      "\n",
      "df[\"col\"][row_indexer] = value\n",
      "\n",
      "Use `df.loc[row_indexer, \"col\"] = values` instead, to perform the assignment in a single step and ensure this keeps updating the original `df`.\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n",
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/logger.py:196: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  dataframe[column][dataframe[\"total_steps\"] > last_step] = last_value\n"
     ]
    }
   ],
   "source": [
    "ppo_cartpole = load_results(\"cartpole_ppo\")\n",
    "ppo_cartpole[\"algorithm\"] = \"baseline\"\n",
    "\n",
    "#ppo_cartpole_latest = load_results(\"cartpole_ppo_latest\")\n",
    "#ppo_cartpole_latest[\"algorithm\"] = \"update\"\n",
    "#ppo_cartpole = pd.concat([ppo_cartpole, ppo_cartpole_latest])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And check how performance looks:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/theeimer/Documents/git/Mighty-DACS/mighty/mighty_utils/plotting.py:200: UserWarning: The palette list has more values (10) than needed (1), which may not be intended.\n",
      "  ax = sns.lineplot(data=data, x=x, y=y, ax=ax, marker=marker, hue=hue, errorbar=errorbar, estimator=aggregation,palette=sns.color_palette('colorblind', as_cmap = True))\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[[255, 255, 255],\n",
       "        [255, 255, 255],\n",
       "        [255, 255, 255],\n",
       "        ...,\n",
       "        [255, 255, 255],\n",
       "        [255, 255, 255],\n",
       "        [255, 255, 255]],\n",
       "\n",
       "       [[255, 255, 255],\n",
       "        [255, 255, 255],\n",
       "        [255, 255, 255],\n",
       "        ...,\n",
       "        [255, 255, 255],\n",
       "        [255, 255, 255],\n",
       "        [255, 255, 255]],\n",
       "\n",
       "       [[255, 255, 255],\n",
       "        [255, 255, 255],\n",
       "        [255, 255, 255],\n",
       "        ...,\n",
       "        [255, 255, 255],\n",
       "        [255, 255, 255],\n",
       "        [255, 255, 255]],\n",
       "\n",
       "       ...,\n",
       "\n",
       "       [[255, 255, 255],\n",
       "        [255, 255, 255],\n",
       "        [255, 255, 255],\n",
       "        ...,\n",
       "        [255, 255, 255],\n",
       "        [255, 255, 255],\n",
       "        [255, 255, 255]],\n",
       "\n",
       "       [[255, 255, 255],\n",
       "        [255, 255, 255],\n",
       "        [255, 255, 255],\n",
       "        ...,\n",
       "        [255, 255, 255],\n",
       "        [255, 255, 255],\n",
       "        [255, 255, 255]],\n",
       "\n",
       "       [[255, 255, 255],\n",
       "        [255, 255, 255],\n",
       "        [255, 255, 255],\n",
       "        ...,\n",
       "        [255, 255, 255],\n",
       "        [255, 255, 255],\n",
       "        [255, 255, 255]]], dtype=uint8)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABJIAAASTCAYAAAAhsRe6AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8hTgPZAAAACXBIWXMAAC4jAAAuIwF4pT92AAEAAElEQVR4nOzdd3xT1eP/8XdaWqBllFF2kQ3KHqIgS0A2Cqg4EFSGCCofHKC4UPyILPmoqAiKIAgyBBGQPQVlCshGZM/SUgp0r/z+6K/325uk7U0pNC2v5+Phw5ybc889KUnavHOGzW632wUAAAAAAABkwCu7OwAAAAAAAICcgSAJAAAAAAAAlhAkAQAAAAAAwBKCJAAAAAAAAFhCkAQAAAAAAABLCJIAAAAAAABgCUESAAAAAAAALCFIAgAAAAAAgCUESQAAAAAAALCEIAkAAAAAAACWECQBAAAAAADAEoIkAAAAAAAAWEKQBAAAAAAAAEsIkgAAAAAAAGAJQRIAAAAAAAAsIUgCAAAAAACAJQRJAAAAAAAAsIQgCQAAAAAAAJYQJAEAAAAAAMASgiQAAAAAAABYQpAEAAAAAAAASwiSAAAAAAAAYAlBEgAAAAAAACwhSAIAAAAAAIAlBEkAAAAAAACwhCAJAAAAAAAAlhAkAQAAAAAAwBKCJAAAAAAAAFhCkAQAAAAAAABLCJIAAAAAAABgCUESAAAAAAAALCFIAgAAAAAAgCUESQAAAAAAALCEIAkAAAAAAACWECQBAAAAAADAEoIkAAAAAAAAWEKQBAAAAAAAAEsIkgAAAAAAAGAJQRIAAAAAAAAsIUgCAAAAAACAJQRJAAAAAAAAsIQgCQAAAAAAAJYQJAEAAAAAAMASgiQAAAAAAABYQpAEAAAAAAAASwiSAAAAAAAAYAlBEgAAwB1i+/btql69uvHfokWLsrtLuc65c+dMP+NJkya5rNe6dWujTu/evW9zLwEAyDyCJAAAAAAAAFhCkAQAAAAAAABLCJIAAAAAAABgSZ7s7gAAAABwp1m/fn12dwEAgExhRBIAAAAAAAAsIUgCAAAAAACAJQRJAAAAAAAAsIQ1kgAAAO5wERERWrhwoVatWqVTp07pxo0bKl68uCpVqqS2bduqc+fOKlSokOX24uPjtWnTJm3fvl179uxRaGiorl27ptjYWOXPn18lSpRQ9erV1apVK3Xq1Em+vr6W+7lixQr9/vvvOnTokMLCwpSQkKCAgAAFBgaqQYMGatasmZo3by5vb2+3fgZHjx7VypUrtXXrVp0/f15Xr16Vv7+/ihcvroYNG6pNmzZq2bKlW22mp3Xr1jp//rwkqXHjxpo1a5ZTnbfeeku//PKLJKlr166aMGGCJOn8+fNasmSJNm7cqAsXLujq1asKCAhQUFCQWrVqpW7duqlkyZJu9ScxMVHr1q3Txo0btXfvXl25ckWRkZEqWrSoypYtq2bNmqlLly666667bvKRAwByOoIkAACAO9j27dv15ptv6uLFi6bjFy5c0IULF7RlyxZ9/vnnev311/X4449n2N68efM0efJkp/ZSREREKCIiQidOnNCKFSs0ceJEffDBB2rdunW67f72228aNWqUwsPDne67fPmyLl++rIMHD2rWrFm66667NGrUKN1///0Z9jckJERjxozRb7/9JrvdbrovPDxc4eHh+vfffzVv3jzVqVNHI0eOVK1atTJs91ZISkrS9OnT9fnnnys2NtZ0X0hIiEJCQrR7925NnjxZ//nPf/T8889banfz5s0aPXq0Tpw44XRfcHCwgoODtXv3bn399dfq2bOnhg0bJj8/vyx5TACAnIepbQAAAHeo7du364UXXkgz9Elx9epVvfvuu5o4cWKadZKSkvT+++/r/fffz7C91IKDg/XSSy9pw4YNadaZO3euXnvtNZchkiunT59Wv379tHnz5nTrHTlyRI8//riWLVvmFCK5sm/fPj3zzDNau3atpX5ktdGjR2vcuHFOIZKj6OhojRkzRt9//32GbU6fPl0vvPCCyxDJUUJCgubMmaNnnnlGly9fttxvAEDuwogkAAAgSYqJT9SlG+l/QIV7ShXMq3w+7k2xup0WL15s3G7fvr169+6tu+++W5J06NAhzZw5U2vWrDHqTJkyRZUqVVK3bt2c2po3b57mzZtnlAsWLKjnnntODz74oO666y7lz59f0dHROnnypDZs2KAffvhBERERkpJDqI8++kitWrWSzWYztXv+/HmNHj3a1O4LL7ygVq1aqVy5csqTJ4+Cg4O1f/9+TZ06VYcPH5aUHHqMGjVKy5cvl4+Pj1N/L1++rH79+ik0NNQ4VrlyZfXt21dNmjRRYGCgIiIidOjQIc2fP1+rVq2SlBzSDB06VLNnz1bdunWt/qhv2u+//65r165JksqUKaO+ffuqZcuWKlWqlK5du6Zdu3Zp8uTJOnr0qHHO559/rs6dO6c5ze3nn3/WmDFjjLLNZlOnTp306KOPqmbNmvLz89Ply5f1559/6rvvvtPp06clSQcPHtSgQYP0008/WZ6WCADIPWx2K1+/AACAXCsqLkH95v+tXw9cUkxCUnZ3J1fJl8dLj9QqpWk968rPN/u/v9u+fbv69OljOubl5aX//ve/evTRR12eM3v2bH300UfGiJ3ChQtr9erVCggIMOpER0erZcuWRtDh7++v+fPnq0qVKmn25ezZs3r88cd19epV49jcuXNVv359U71Jkybpyy+/lCR5e3trwYIFqlmzpss24+Li1LdvX+3cudM49tVXX6lt27ZOdZ977jlt3brVKPfs2VPvv/++y9BJklasWKFhw4YpPj5eUnKYs2LFCuXLl89U79y5c2rTpo1Rfvnll/XKK684tefuGkkpmjdvrs8++0wFChRwqh8dHa2+fftq9+7dpjZcTXE7deqUHnnkEcXExEiS8ufPr88//zzNdaDi4uL05ptvavny5caxfv36afjw4S7rAwByL6a2AQBwh+s3/2/N23uBEOkWiElI0ry9F9Rv/t/Z3ZU0vfTSS2mGSJLUq1cvPf3000b52rVrWrBgganOli1bjBBJkl588cV0QyRJCgoK0pNPPmk6lno0TYqUEUaSdM8996QZIkmSr6+v3n33XdOxbdu2OdXbtWuXKURq2bKlRo0alWaIJEkdO3bUiBEjjPKFCxecQp5bLTAwMM0QSUoOg9555x3TMVePX5KmTp1qhEhS8rS59BYT9/X11YQJE0yjsObMmWN5uiEAIPcgSAIA4A4WE5+oXw9cyu5u5Hq/HrikmPjE7O6GkxIlSujFF1/MsN6QIUNMU5h+/vln0/158+ZVjx491LhxY5UtWzbdYCo1x1Do+vXr6dY/e/asoqKi0q1To0YNffrpp/rxxx+1ceNGvf322051HEf/DBs2zGlKnStPPPGEaZrYnDlzMjwnK/Xs2TPNEClFrVq1VKRIEaMcHBzsVOfatWtasmSJUa5du7Y6deqU4fW9vb318ssvG+Xo6OjbHqYBALIfQRIAAMAdqlu3bsqTJ+MpdwEBAWrevLlRPnXqlCmgaNGihT755BPNmjVL69evV7FixSxd39/f31SOi4tzqlO1alXjdnh4uAYOHKhjx46l226XLl107733qnTp0vLyMv+5a7fbTaN07rrrLtM10pMnTx7TqJ1jx44pLCzM0rlZoXHjxpbqlSlTxrgdHR3tdP9ff/1lTNGTZJqKl5EmTZqYdmzbvn275XMBALkDQRIAAHewfD7eeqRWqezuRq73SK1SHrnodoMGDSzXrV27tqm8b9++TF3z4sWLWr9+vSZOnKhPPvkkw/qPP/64KezasWOHunTpoi5dumjs2LHasmWLaYpWRk6ePGmajlWuXDm3+l+jRg3jtt1uz/TPITMqV65sqV7+/PmN2wkJCU7379mzx1R252fg4+OjSpUqGeW9e/daPhcAkDtk/6qPAAAgW03rmbzmCYttZ73Ui217ovLly1uuGxQUZCqHhISkWTchIUEHDhzQ4cOHderUKZ09e1bnz5/XmTNn0p2a5moPmHLlymnkyJF67733TMePHTumY8eO6fvvv5evr68aNmyoFi1aqFWrVqagw5HjVK8//vhD1atXT7N+RtL7OWS1ggULWqqXehSWq5/ppUvm6axvvPGG3njjjUz16erVq0pISLA0sg0AkDvwjg8AwB3OzzePfnqmoWLiE3XpRmx2dydXKVUwr0eOREphNZhwVdfVekZXr17V1KlT9csvv5h2Y0tLnjx5XI6YcdSzZ0+VKFFCo0aNMnY6Sy0uLk5bt27V1q1bNXbsWFWvXl19+vRRjx49nKa2pV4UPCvczsWmHXeIy6yM1qJy17Vr1yxPZwQA5HwESQAAQFLyNLcKRf0yrohcw9VolbQkJZlHqzmGGn/99ZdeeumldAMkf39/1ahRQ3Xr1tV9990nSRo4cKCl67dq1UrNmzfXli1btHLlSm3atElXrlxxWffo0aN65513tHDhQn377bemBaqtBFfuiI3NeeFrVv8M3JlaCADI+QiSAAAA7lCuFmJOy40bN0zlwoULG7fPnDmjAQMGKDIy0nR/ixYtVLduXVWpUkWVKlUy7XgmSZs3b3arv97e3mrZsqVatmwpu92uw4cPa/v27dq6dat27dplur4k7d69W2+88Ya++eYb41ihQoVMdfr3769hw4a51Y+czvFnsHz5csvrLwEAQJAEAABwh7p48aIqVKhgqe7JkydN5VKl/m+R9nHjxplCnI4dO+q///1vhlvVR0REWO+sA5vNpnvuuUf33HOPnn/+ecXFxWnbtm368ccftWnTJqPehg0bdOTIEWORbMcpWLdz1zVP4epnQJAEALCKXdsAAADuUIcOHbJcN/VOX97e3qpTp46k5JFKGzZsMO4rWbKkxo0bl2GIJEmnT582ldOaahcWFqa//vorzalskuTr66sWLVpo6tSp6t69u+m+v//+27hdtWpV07Q8d3ddCw4O1pkzZxQfH+/WeZ7EcQe+1D8fK44dO6bQ0NCs7BIAIAchSAIAALhDrVmzxlK98+fPa/v27Ua5Tp068vf3l5Q8rS31mjv169eXr6+vpXbXrVtnKicmJprKO3fuVOPGjdWkSRM9/fTTWr58uaV2O3bsaCqnDj18fX3VoEEDo/zvv//q8OHDltqVpGHDhumhhx5SnTp11KJFC7dDGE/QuHFj2Ww2o7xs2TLL5549e1aPPPKIHnjgAdWpU0fPPPPMregiAMCDESQBAADcofbs2aMVK1ZkWG/06NGmxbZThweOo4is7NYmSTNmzHAaDeQ4yqdq1aqm6W8LFixwCptccdzZrXTp0qbyE088YSqPHj3aUrvr1683ArWkpCTFx8frnnvuyfA8T1O6dGm1bNnSKB8+fFg///yzpXPHjRtn/KxiY2Nz5OMHANwcgiQAAIA72DvvvKO//vrL5X12u11jxozR2rVrjWNVqlRRhw4djPJdd90lHx8fo7xz507t2rUrzevZ7XZNmzZNEyZMcLrPcQe0gIAAtWnTxigfPXpU48ePT3e3ueDgYE2ZMsUo+/j4qHnz5qY67dq1U9WqVY3yjh07NHz48HR3YDtw4IDeeust07GBAweaHntO8uKLL8rL6/8+Cnz44YcZjvj64osvtHr1aqOcL18+9evX75b1EQDgmQiSAAAA7lDe3t6KjIzUs88+q08++URHjhxRdHS0wsPDtX79ej399NOaPn26UT9fvnz63//+pzx5/m+/loIFC+rBBx80yklJSRowYICmTJmikydPKi4uThERETp27Jhmz56tHj16aNy4cS7XGHLcGU6SXn75ZVNYM336dD3zzDNauXKlgoODlZCQoJiYGB0/flwzZsxQjx49dOnSJaN+7969FRgYaGrTy8tLn332mfz8/Ixjy5YtU6dOnfTjjz/q9OnTio2NVVhYmHbv3q133nlHTz75pK5du2bUb9CggXr16mX1R+1x6tevr5deeskox8XF6dVXX9WLL76o9evXKywsTHFxcTp79qyWL1+uxx57TF999ZWpjeHDhzvtxAcAyP3YtQ0AAOAO9eabb+qTTz5RfHy8ZsyYoRkzZqRZ18/PT1988YWqVavmdN/w4cO1c+dOY1pbVFSUJk6cqIkTJ6Z7/YYNG+r48eMKDw+XlLz+jqPq1avr/fff1/vvv2+MRNq1a1e6o55StG3bVq+//rrL+6pUqaKvv/5aQ4cONa5/7tw5ffTRRxm2e/fdd2vSpEk5djRSipdeekk3btww/btv2LDBtHh6WgYOHJijgzQAQOYxIgkAAOAO1b59e3366acqWLBguvVq1aqln376yWmKWIqgoCDNmDFDFSpUsHRdPz8/vfrqq5o5c6Zq1aplHN+/f78iIyOd6vfs2VNffPGF08iijNp3HD3lqEmTJlqwYIFatWplqV0vLy89/fTT+vHHH1W8eHFL53gym82mESNGaOLEiSpbtqylcwIDAzVx4kS99tprt7h3AABPxYgkAACAO1jnzp3VqFEjzZ8/X+vWrdP58+cVFxenwMBA1alTRx07dlTbtm1Nu3y5UqNGDS1ZskS//fab1qxZo8OHD+vq1atKTEyUv7+/AgMDVa1aNTVs2FCdO3dWQECApOQwa8uWLZKSF9tetGiRevfu7dR+u3bt1Lx5c61du1abNm3SkSNHdPnyZUVGRipfvnwqXry4KlasqFatWumhhx5SsWLFLD3+8uXLa8qUKTp48KBWr16t7du368KFC8YopcKFC6tKlSq699579cgjj1gOXHKSzp07q127dlqzZo22bNmiv//+W6GhoYqIiDB+trVq1VLLli3Vrl075cuXL7u7DADIRjZ7eqsVAgAAAAAAAP8fU9sAAAAAAABgCUESAAAAAAAALCFIAgAAAAAAgCUESQAAAAAAALCEIAkAAAAAAACWECQBAAAAAADAEoIkAAAAAAAAWEKQBAAAAAAAAEsIkgAAAAAAAGAJQRIAAAAAAAAsIUgCAAAAAACAJQRJAAAAAAAAsIQgCQAAAAAAAJYQJAEAAAAAAMASgiQAAAAAAABYQpAEAAAAAAAASwiSAAAAAAAAYAlBEgAAAAAAACwhSAIAAAAAAIAlBEkAAAAAAACwhCAJAAAAAAAAlhAkAQAAAAAAwBKCJAAAAAAAAFhCkAQAAAAAAABLCJIAAAAAAABgCUESAAAAAAAALCFIAgAAAAAAgCUESQAAAAAAALCEIAkAAAAAAACWECQBAAAAAADAEoIkAAAAAAAAWEKQBAAAAAAAAEsIkgAAAAAAAGAJQRIAAAAAAAAsIUgCAAAAAACAJQRJAAAAAAAAsIQgCQAAAAAAAJYQJAEAAAAAAMASgiQAAAAAAABYQpAEAAAAAAAASwiSAAAAAAAAYAlBEgAAAAAAACwhSAIAAAAAAIAlBEkAAAAAAACwhCAJAAAAAAAAlhAkAQAAAAAAwBKCJAAAAAAAAFhCkAQAAAAAAABLCJIAAAAAAABgCUESAAAAAAAALCFIAgAAAAAAgCUESQAAAAAAALCEIAkAAAAAAACWECQBAAAAAADAEoIkAAAAAAAAWEKQBAAAAAAAAEsIkgAAAAAAAGAJQRIAAAAAAAAsIUgCAAAAAACAJQRJAAAAAAAAsIQgCQAAAAAAAJYQJAEAAAAAAMASgiQAAAAAAABYQpAEAACAbPXjjz+qevXqql69up544ons7g4AZJvjx4+rZs2aql69uho2bKgLFy5kd5cAJza73W7P7k4AgCvbt29Xnz59XN6XL18+bd26VX5+fll2vbfeeku//PKL0/GyZctq/fr1WXYd4E7Xu3dv7dix47Zdr3v37hozZowkqXXr1jp//rxx38svv6xXXnnltvUFzk6dOqVu3bopOjpakjR37lzVr18/m3sFQJIWLVqkESNGmI4dPXo0m3qT/ZKSkvTaa68pKSnpptopXbq00881tZEjR2ru3LmSpPvvv18zZsyQzWa7qWsCWSlPdncAADIjJiZGGzduVKdOnbKkvbi4OK1duzZL2gIAWJOYmKg333zTCJE6dOhAiATAY508eVIrVqy46XaqVq2a7v1DhgzR0qVLFRkZqW3btunHH39U7969b/q6QFZhahuAHGvlypVZ1tbvv/+uGzduZFl7AICMzZo1S3v37pUk+fj46LXXXsveDgFAOg4dOnRbrlOsWDH17dvXKH/66ae6ePHibbk2YAVBEoAca9OmTYqKisqStn777bcsaQcAYM2VK1f05ZdfGuXHHntMd911Vzb2CADSd/jw4dt2reeff17FihWTJEVHRxtTtAFPwNQ2ADlWVk1vi4qK0oYNG7KoVwAyMmvWLLfqV69e3VRu3Lix223A80yYMMEYCZo3b14NGjQom3sEAOk7cuSIqTx06NBb9t7l7++v/v37a+zYsZKSR+Jv375d99133y25HuAORiQByFEKFy5sKmfF9LYNGzYY63MAyN3Wr1+vo0ePGv+x0Hb2OHLkiGlzg8cee0wlS5bMxh4BQMYcRyTdc889t/R6Tz31lIoUKWKUR48eLfbKgicgSAKQo7Rt29ZUzorpbY7T2urWrXtT7QEA0vfVV18ZH4ZsNpuef/75bO4RAKQvODhYYWFhpmN33333Lb1m/vz59fTTTxvlI0eOaN26dbf0moAVBEkAcpTWrVsrX758Rjlleltm3bhxQ5s3bzbKtWrVUvny5W+miwCAdPzzzz9as2aNUW7cuLGCgoKysUcAkDHHhbaLFy+uEiVK3PLr9ujRQzabzSh//fXXt/yaQEYIkgDkKH5+fmrZsqXp2M1Mb1uzZo3i4uKMcufOnTPdFgAgY1OnTjVNzejRo0c29gYArHGc1narRyOlKFeunGldpIMHD5q+BAWyA4ttA8hxOnbsqFWrVhnllOltfn5+bre1fPly47bNZlOnTp2ydGvXCxcu6NChQ7py5YrCw8NVoEABFStWTNWqVVOlSpWy7DqSFBsbq/379+vy5csKDw/X9evXZbPZ5OfnpxIlSqhq1aqqWLGi6VutrBAeHq69e/fqzJkzioyMVKFChVS8eHHVq1fPI9Y8uXTpkvbt26fQ0FBdv35d/v7+KlasmCpWrKgaNWpk+c/jdridzyspeTj/1q1bdfnyZQUGBqpRo0aMIJGUlJSk/fv36+jRowoLC5OXl5eKFy+ue+65RzVq1Mjw/EuXLmn37t26ePGiEhISVLhwYVWqVEn16tWTr69vlvTx5MmT+ueff3TlyhVdv35dhQsXVvHixVWzZk2VKVMmS67hjqtXr5rev/39/dW+ffssa//48eM6dOiQgoODlZiYqGLFiqlMmTJq2LCh8ubNm2XXyQ43btzQX3/9pVOnThm/88qVK6d69eqpePHi6Z5rt9u1f/9+HTlyRGFhYfL19VXRokVVp06dW/KecSuFhYVp3759unDhgm7cuKG8efMqICBAxYsXV926dVWwYMEsu1Z0dLT27t2r4OBghYaGysvLS0WLFlXZsmVVt27dLHudprhx44b27NmjkJAQXblyRb6+vipWrJjKly+v2rVry8sr68YAXLlyRXv37tX58+cVHR2t4sWLq2zZsmrYsKF8fHyy7Dp2u10nTpzQoUOHFBYWpujoaBUsWFBFihRRUFCQatasmaWP61a63esjpdajRw9t27bNKM+dO1fNmze/bdcHHBEkAchxWrVqJT8/P2NtpMzu3nb16lVt3brVKDds2FClSpW66f7FxMRo1qxZWrJkif75558065UvX15dunRRv379VKBAgUxd69q1a1q4cKHWrVunffv2mUZXuRIYGKjOnTvr+eeft/RYJ02aZNqee8SIEXruueckSXv27NHkyZO1efNmJSUlOZ1rs9lUs2ZN9e3b97aP9IqIiNC8efM0f/58nTp1Ks16gYGBatOmjQYOHJjhh+qxY8fq+++/N8re3t7atGmTAgMD3e7fa6+9Zlqbq0KFCqYP167cyufV9u3b1adPH6Pcpk0bY+j8l19+qalTpyo2NtZ0TuPGjTV06FA1bNjQ0jU8RevWrXX+/Hmj/PLLL6e54LZj3X379ilv3ryKi4vT9OnTNXPmTIWGhro8t3z58nrllVf08MMPO923a9cuff7559q5c6fLRVP9/Pz05JNP6qWXXsrUe8P169c1bdo0rVixQqdPn06zXrVq1dS9e3c988wzWf6BOC2//vqr6X2qWbNmyp8/v6VzHf89du7cqUKFCikhIUHz5s3TDz/8kObj9fPz04MPPqhXXnlFFStWzPBa586dU5s2bUzHZs6c6dZuSY7vn2XLltX69etd1nV8DXbq1En/+9//JEmnT5/WZ599pjVr1ig+Pt7pXC8vL7Vs2VLDhw93CoXi4+P1448/avr06QoODnZ57UqVKumVV1656R1Qb6X4+HgtWLBAixYt0v79+9Os5+3trTp16qhDhw56/PHH5e/vn6nr/fnnn5o+fbq2b9/u9N6Xwt/fX82bN9eLL7540yNTVqxYoTlz5mj37t1KSEhwWScgIECtW7fW4MGDbyrI/+OPP/Ttt99q+/btLn93Fy1aVI8++qgGDx6cqS/oUpw/f14zZszQb7/9pitXrqRZr3Dhwrr//vvVs2dPNWvWLNPXux2ya0SSlPx7OU+ePMbzY+PGjQoJCcnU3yBAVsgZ8S8ApJI/f361atXKdCwz09tWrVpl+oMtK8KOtWvXqn379powYUK6H/Yl6cyZM/r666/Vtm1bLVmyxO1rzZ8/Xw899JDGjh2rXbt2ZRgiSVJISIhmzJih9u3b6+eff3b7mlLyKIyxY8fqqaee0qZNm1z+ISolfwt54MABvfbaa+rdu7ciIiIydT13LV++XK1bt9a4cePSDZGk5J/H3Llz1b59e3366adKTExMs+6jjz5qKicmJjot1G5FRESE00KZ3bt3T/ec2/m8Su3LL7/UpEmTXH6Q2rFjR6afQznZyZMn9eSTT2rixIlphkhS8r/DsGHD9N577xlhUcpr55lnntGOHTvS3HknKipK33//vR5//HFTcGJFyvvCN998k26IJCWvVTR27Fh16NBBf/zxh1vXySzH58zNfqN+/vx59ezZU6NGjUr38UZFRem3335T165dTYGwp/vll1/UrVs3LV++3GWIJCU/rzZs2KBHH33UNN3l/PnzevzxxzVmzJg0QyRJOnHihF599VV98MEHWd39LHHixAl17dpVH374YbohkpT8vrxnzx598sknat++vTZt2uTWtYKDg9W/f389//zz+v3339MMkSQpMjJSK1euVI8ePTRixIhM7f7677//qmfPnho6dKh27NiRZogkJY/+XbRokTp27Kjx48en+/vKlYiICP3nP/9R3759tXXr1jR/d4eFhenbb7/VI488ooMHD7p1jRQ//fSTOnbsqJkzZ6YbIknJX4itWrVK/fr1U79+/TKs/9Zbb6l69eq37L+0RERE6Ny5c6Zjt3NEUoECBUybwSQkJJh2vgRuN4IkADmS4zenv//+u9t/xKUOAfLkyaMOHTrcVJ++//57vfzyy7p06ZJb5129elXDhg3TpEmTLJ8zevRovffee7p27Zq73ZSUPLrlnXfeMS14a9WIESP0/fffu7X97I4dO9S/f/90/0jOCv/973/16quvuv1ziYuL09SpU/XCCy/oxo0bLutUqVJFderUMR3LTFCzatUqxcTEGGUvLy9169Ytzfq383mV2r59+/TVV1+lW+dmXzM5zblz59SrVy+3PlzNnz9f3333nex2u9566y23XjsnTpzQ0KFD0wwQUktKSjLeF8LDwy33T0oOHF544QUtWLDArfPcdfLkSR07dsx07IEHHsh0excuXHD73yM+Pl5jx47VnDlzMn3d22XOnDl66623LO9MGhUVpSFDhujs2bM6d+6cnn76aacRFOn56aef9MMPP2S2u7fEqVOn9NRTT+nkyZNunxsSEqJBgwalOQrM0dGjR9WzZ0+3155JSkrSokWL1KdPnwxDkNS2bdump556Sn///bdb14uPj9d3332nwYMHW35uhIeH66mnnnLrS7czZ86ob9++OnHihFv9mz59uj744IN0Q7i0bNmyRb169cr03za30uHDh03v3f7+/ipfvrySkpK0ceNGvf/+++ratasaN26sWrVqqXnz5nrsscc0fvx47d69O0v60LRpU1M5M3/DAVmFqW0AcqQWLVrI399fkZGRkpLXMdiwYYPlofmXL1/Wrl27jPL999+vokWLZro/s2bN0tixY52O33PPPerYsaNq1aqlgIAAXbt2TcePH9eqVau0Y8cOU90vv/xS/v7+6tu3b7rXWrhwocs/9hs3bqy2bduqSpUqKly4sKTkb/pSdkj666+/nM4ZNWqUWrZsaXlay7x580x/VJYoUULdu3dXw4YNVaxYMV2/fl1Hjx7V3LlznUYD7dmzR/PnzzdtY5uVvvjiC82aNcvpePny5dW5c2fVr19fxYoV040bN3Ts2DGtWLHC6Y+7LVu2aPDgwZoxY4a8vb2d2urRo4f27dtnlA8ePKjjx4+rcuXKlvv566+/mspNmzZNc5rh7XxeORo1alSa31hLydMsHP+oze0GDx5sfFC02Wxq2bKlOnTooPLly8vb21v//POPZs+erSNHjpjOmzx5ssLCwkz/9o0aNVLnzp1VrVo1+fr66syZM/rll1+0ZcsW07n79u3TihUrXE6RS23cuHEu3xfuvfdePfTQQ6pevboKFiyo8PBwHT58WMuXLzcFMAkJCXr33XdVqFChLF2zKDXHHTYrVKhwU+s0vfrqq7p48aJRbty4sdq1a6caNWoof/78unTpkrZs2aJff/3V6QP3hAkT1L59exUrVizT17+V9u/fb5ruWqhQIT3++OOqX7++SpcurRs3bmjbtm2aOXOm6bFFRUVp/PjxOnXqlBE++/r66pFHHlHz5s1VunRpRUVF6eDBg5o9e7bTiLcvv/xSjz/++E1Nacoqdrtdw4YNcwpG69Wrp44dO6pq1aoqXLiwbDabQkNDtWfPHi1cuFCXL1826iYmJhpfmqQ3TfT8+fPq3bu3U4BRuHBhPfzww2rcuLFKlSqlxMREXbx4UZs3b9bKlStNP/t9+/ZpwIABmjt3boa/U/fv36/+/fs7hcQlSpTQww8/rIYNGyowMFCxsbE6d+6cNm3a5DS1cePGjXrttdf0zTffpHutpKQkvfzyy06jWb28vNSuXTu1a9dOQUFBio2N1fHjx7Vs2TLt3LlTUnIA9d1336XbfmqHDx/Wp59+ajrm4+OjLl266P7771dQUJD8/f0VExOjs2fPauPGjVq5cqXpS6aTJ0/q008/1ahRoyxf93ZwDGVr1KihX3/9VV988YXLkaOXL1/W5cuXtX//fn333Xe677779Oabb6pmzZqZ7sMDDzxg+nJo//79Cg0NzXCNNOBWIEgCkCPlzZtXrVu31tKlS41jK1eutBwkrVixwvQhuUuXLpnuy4EDB5w+7BcoUECjRo1yOV2uSZMmeuaZZ/T777/rzTffVFhYmHHfxIkT1bBhQ9Pw5dQiIyM1YcIE07F8+fLp008/Vdu2bV2e88ADDxjD9F9//XVdv37duO/y5cvavHmz01ogaUkJkWw2m/r3768hQ4Y4/cHctGlT9enTRx9//LFmz55tum/GjBm3JEhau3at0+gZLy8v/ec//1G/fv2cFg5t0qSJ+vTpo7Vr1+rtt982fXjYsWOHJk6cqGHDhjldp3Pnzvrkk09M37QuXbpUQ4cOtdTPixcvOgU9ae1YdTufV462b99umopYpUoVY7HiEydOaNu2bWrbtm2WLsiaE6SEo4GBgfr666+dRqjVq1dP3bt318CBA01TxSIjI43pVPnz59fHH3/s9G9Yp04ddenSRTNmzNAnn3xium/BggXpBknr16/X9OnTTccCAwM1btw4l2HfAw88oP79++vXX3/VyJEjTaM533nnHd1zzz23ZDF1x5DM6vMxLSnvR4GBgRozZozT+iq1atVS27Zt1a9fP7344ov6999/jfsiIyM1f/58DRo06Kb6cKucPXvWuN2mTRuNHz/eab2fJk2aqFOnTnryySdNgUbqAOqee+7RpEmTVK5cOdO5999/v5544gm9+OKLRmggJa+vtWrVqgyn294O27ZtMwX3kvTmm2+mGYq3bNlS/fv31xtvvKENGzYYx8PCwowRQ67Ex8frtddecwqRHn/8cb311ltOAVT9+vXVqVMnDR06VO+8845pBNPBgwc1btw4vfvuu2k+rhs3bujVV181hUI2m00DBw7U4MGDnRaFb9Sokbp166bjx4/rzTffNE3v27Bhg6ZPn67nn38+zevNmjXL9G8sSSVLltSkSZOcXoP33nuvnnzySS1fvlxvv/22oqOj3Rp9/P3335seV2BgoKZPn66qVas61a1Xr566du2qZ599Vi+88IKuXr1q3Ldo0SINHTrU5Rd8tWvXtjwSKys5Bkl79+51+QVdWrZv364nnnhCH3zwgR577LFM9aFmzZqmdZLsdru2bNmS7qhm4FYhSAKQY3Xq1MkUJKVMb7OycGvq3dp8fX310EMPZbofI0eONP3h5O/vr9mzZ2e4a1OLFi00e/ZsPfPMM8Yoh/j4eH3yySeaO3euy3OWLl1qCgik5D+s0wqRHK/3/vvv64033jAd//PPPy0HSSleffVVDRw4MM37vb299f777+vYsWOm4OT06dNuj+DJSFJSkiZOnGg6ZrPZNG7cOHXt2jXdc9u2bavy5curd+/epm+9p0+frp49e+quu+4y1S9UqJAeeughLVu2zDjmTpC0dOlS0x/lhQoVSvPf7nY+rxylhEheXl5677339NRTT5l2twsODjZNz7uT+Pn5ac6cOSpfvrzL+318fPTBBx+oXbt2Lj+ATZgwId3X63PPPafff//dFETt379fCQkJypPH+c+2uLg4jRw50nSsZMmSmj9/foYL6j/yyCMKCgrS888/b/x73rhxQ5999pnTqIKbZbfbnUKBrFhfpECBAvrhhx/SfU8JCgrSl19+qa5du5peUxs2bPDYIClFy5YtNWnSJJcjJCWpevXqev75511OQy1btqx++OEHFSpUyOW5BQoU0JgxY9ShQwfTz2X37t0eESQ5riXXtGnTDEdWFihQQBMnTlSXLl1Mo0TWrFmTZpA0b9487d2713TspZde0pAhQ9K9VsmSJfXNN9/otddeM4V3P/74o55++uk0d8P7+uuvTUGhlDwtO6NwoXLlyvrhhx/Uv39/02jaL774Qj169DBGIqcWHR2tKVOmmI4FBARo1qxZTr/fUuvUqZOKFSumfv36WZpam8Lx32z48OEuQ6TU6tSpozFjxpj+poiPj9fGjRtdftHSq1cv9erVy3KfsopjkOS4RlWpUqVUtmxZFSpUSGFhYTp69KjT78n4+Hi98847unbtmvr16+d2H3x9fVWlShXTqNf9+/cTJCFbsEYSgByrWbNmpj+Qo6OjnaZOuHL+/HnTH40tW7bM9K5pO3bs0IEDB0zH3n33XUtbf0vJu+U4fgjcs2dPmvPpHdc3KFu2rJ544gnL/e3YsaPTlAV3196pWrVquiFSai+88ILTMcdpPzdr6dKlOn78uOnY008/nWGIlKJatWpOQ+gTExP17bffuqzv+IftuXPnLK9/4DitrXPnzi63JL/dz6u0DB06VE8//bQpRJKSP0Cl9yEkN+vXr1+aIVKK8uXLq1atWk7HW7VqZSn0dRwhGR0drQsXLris+9tvv5mm8kjS+PHjLe9A2aBBA/3nP/8xHVu5cmWa18usU6dOmUZDSskj3W7WoEGDLAXTFStWVLt27UzH3Fk/KDv4+Pjo7bffTjNESpHWSNzhw4enGSKlKFeunOrXr2865vh+ml0yu7Cxn5+f6feit7e3rly54jLYTUpKcpoS2rhx4wxDpBR58uTR6NGjTTtn2e12pxGCKSIiIjR//nzTse7du1seoeLv76///e9/ypcvn3EsKipKP/30k8v6K1eudFq36c0337T0/n3fffe5FXZcvXrVWG4ghdVpXK1atVKFChWMsp+fX5a/B92MuLg404jG1B5++GEtWLBAmzZt0pw5c/TNN99o/vz52rFjhz777DOXgeKECRPcXosrheP7pmNAD9wuBEkAcixfX1+nkTQrVqzI8DzHnbZuZstjxz/eypYt6/Y3Q+3atTP9ASUlr4PkStOmTfXwww+rfv36CgwMVNeuXTP8kJFanjx5nK7l7hDxnj17Wq7r+AFFkluLkVrh+G+eN29epw/GGWnfvr1TX5csWeJysdAmTZqodOnSTnUzcvDgQac/RNOa1na7n1euBAQEpPkN/p3Ky8vL8jfhrkI/qx8WXW1Pn9bis47PlUaNGrm1Rb0kPfXUUypYsKBRTkhI0OLFi91qIyOOi2xLuukwMk+ePG5NEalXr56pHBcXd9t2k8yMFi1aOL2GXalUqZLTFOOAgADLI20dn2+OgV92cZw66zhqKD3dunXTtGnTtHr1av39999avny5UyAuJU+3PHPmjOnY4MGD3epngQIFnN4XlixZ4nIn1aVLlzo951566SW3rleqVCmnL0rSem9PPXpWSv4SwJ3fJQMHDrS8Xparqc7ufHkxZswYzZ07V3/88Yf27Nmjl19+2fK5t9rx48edRmb5+/vryy+/1Pjx452mOUvJf4t07NhRixcvVseOHU33JSUlacSIEZlakNzxfdPVeytwOxAkAcjRHH85W9m9LXXw4Ofnp9atW2f6+o7r3bRo0UJeXu69tdpsNj344IOmY6kXAk/thRde0Pjx4zV37lxt2bJFr776qnsdlpzW2XBn2LqU/EHVqgIFCjiN9srMFslpSUxMdPpZtWvXzuUQ/4w4juyKjY11uZuOq13WVqxYkeHP0XE0UtWqVV3+8Snd/ueVK/fdd5+laaJ3kipVqqhIkSKW6rqq5ypYdSWtKSqOIiMjnUautWrVytI1UsufP7+aNGliOubOc8UKx8VobTabUyDrrsqVKysgIMByfVfXy461Vqy69957LdXz8vJyes7Url3b8pcMjud6ys/E8QPzrl279N///tfStNqSJUuqWbNmuuuuu9Jdy2379u2msr+/v+Wfe2qOf0fExMQ4vTZdXa9y5cqZWo/M8XpnzpxRcHCw6Vh8fLzTGj4dO3Z063eJn5+f00i+tBQoUMBp0eexY8dq06ZNls6vX7++6tev75ELR0dGRqpevXrG4vxeXl769NNPLYW1efPm1aeffur0vAoJCUlzJFl6HNc7i46OdlryALgdCJIA5GhNmzY1fZDIaHrbyZMndejQIaPcunVr0xBxd5w6dUqhoaGmY1a+PXbl7rvvdmo7q/8wiIiI0Nq1a52Gi6e3M5cjLy+vDNc7cOT4bWbq3Vlu1uHDh3Xjxg3TMXdHY6Ro3Lix07G0Pkw7jiQKDw9Pd5h6YmKi00i4tNYg8ZTnldXQ406S1ponrjiOEPHz87P8AcnVB19X03L27t3rtE5HVj1X9u7d69Z7Q0Ycg6TChQu7XPPJHdWrV3ervquRFY4/P09yM8+3jKZfpuapi+a72gRj1qxZatOmjT766CP9/vvvN71Wm2PQUq5cuUw9L6tWrer0c3Q1GsfxWGZH5bma5ufY9rFjx5wC6MwscO/Ol0eO/2Y3btzQCy+8oMcff1zffvutjh496vb1PUGjRo00b948/fnnn9qzZ49Wr17t9EVNery9vfXxxx87hbs///yz231xtdOk4zRQ4HZgsW0AOZqPj4/atm1r+mW8cuVKp5FKKRw/zN/Mbm2u5ssvWbLE7XVoJLn8cH/u3DmXO5ZkJDo6WmfPntXZs2d18uRJ/fvvvzp06JCOHTvm8oOhOzuy+Pn5uf2hw7G+O9fLiKv1nTK7tW7ZsmVVpEgR084xaa3RUL58ed17772mnXCWLl2a5ui2LVu2mMKhPHny6JFHHnFZ11OeV1mxfk1uk3odFHe5sw6bqyk4rria0jBjxgzTJgRWOX4QiYyM1NWrV11+aMmM1K8rSZl6b3Pk7shDVwFBVoZlWe1mRmaknqqYEavPt9utRo0a6tixo9P05dDQUP3444/68ccflTdvXjVq1EjNmjVT8+bN3f6iw/H9Njg42PL6SI68vb1NI1MdX1PXr193GjV05MiRTF3P1e9Rx+ul7DKZmrs/H8m9wPb555/X4sWLTZtXSMnr+Ozbt08TJkxQiRIl1KxZMzVr1kxNmza1PMrTU/j5+Vme7pfaXXfdpQcffFBr1641jh07dkzBwcEqWbKk5XZc/bwc31+B24EgCUCO16lTJ1OQtGnTpjR3b0v9B2lAQIDTdtHucPxDSUpeB+fgwYOZbjOj9h1FR0dr/fr1+uuvv/TPP//o1KlTCgkJyZLru5LZRclTy8ogydUfTzfz4atYsWKmNtNal0ZKHpWUOkhav369IiIiXP6MHNdQat68eZr99ITnleT+h/Q7QWY+PKS4FaM+XD0/s3JKWnh4eJYFSY7TpbJi2qSnvR9ltZt5vt3saC9P8cknn+jixYtpro8UGxurP/74Q3/88YfGjh2rsmXLqlWrVmrfvr3uvffedKdxJSYmOq1XFB4ebtqB7WY4vj5dvV4vXLiQZYtKO7bvuAi/5DqEyIjVhftT6n711VcaMGBAmlMkL1++rEWLFmnRokXy9vZW3bp11aZNG3Xo0MFp2lZu88ADD5iCJCl51zV3giRX751ZuWQAYFXu+C0D4I52//33q2jRosboi5TpbY6jko4cOWL69vGhhx66qQ936YUMWSG9BU+vXbumyZMna968eW6tZ2Gz2ZQnTx6310VK4c7C3reDq1DEnW/iMzo3vdClQ4cO+uijj4yff0xMjNasWeM0ZS1lSmFqaS2yLWXv8yo1giRnnjYFyFOeK1Y4TkHKip+lp70fZTVPe75lh/z58+vHH3/Ut99+q6lTp2b4gfn8+fOaPXu2Zs+erTJlyqhfv3566qmnXD5Xrl27dkuDRMfXj9UQP6uu5+pn5bhGohXuntOoUSMtXbpUH374oX7//fd06yYmJmr37t3avXu3JkyYoPvuu0+vv/56musHStLs2bOd1prKSl988cUta9vV6C7HqewZcZzGKhEkIXsQJAHI8by9vfXQQw9p3rx5xjFX09uWL19uKnfu3PmmrpvZMMYqVzu+SNKBAwf04osvWh55VLx4cdWpU0f333+/2rRpoxEjRjgt5pxTufoQcDMfDBzXb0rvW30/Pz917NjRtFvOkiVLnIKk1atXmz5EFylSJN21FbLreeUot39Izw085bliheP0KU8eCQTP4uPjo8GDB6t379767bfftGLFCu3atSvD9fYuXLigjz76SAsXLtS0adOcplPe7tfP7b5eVk1ZzJs3r9vnlCtXTt9++63++ecfLVmyRKtXr9bp06fTPcdut2vbtm3q2bOnBg4cmOZmIvv378+yUWO3m6vNAdwNGF29d3rq9FTkbgRJAHKFTp06mYIkV9PbUgdJgYGBmV6UOYWrkS/Lly9X5cqVb6rd9Jw8eVK9e/d2OQopf/78uueee1StWjVVrlxZFStWVNWqVZ2GTOemD3CuRs1ERERkekqI48LdGS3E3qNHD1OQtG3bNl2+fFklSpQwjjluo961a9d0Rxpkx/MKOZOr58q+ffsy9cHvVnN8Td4J36Bn5cYCSH6+P/nkk3ryySd148YNbdmyRVu3btXWrVt15syZNM87dOiQ+vfvr59++sn02ihUqJBT3QEDBuiNN964Jf13db0PPvhATz311C25nqspUDdu3HB7WmlkZGSm+1CtWjW98cYbeuONN3TixAlt3rxZW7du1c6dO52mFaaw2+365ptvVKBAAQ0YMCDT1/ZErtZkc/ffw9UC85kZaQbcLIIkALlC48aNFRgYaIzSiY6O1qZNm9ShQwdJyd9gnT171qjfoUMHt7dTd+Tqm6WsnAriyvDhw51CpCpVqujVV19VixYtXA55dnSzu9x4Eld/mIeGhpqCHKvsdrvTEPOM1ltq1KiRKlSoYCxqmpSUpOXLl+u5556TlLwYuOPor0cffTTdNrPjeYWcydVz5dq1a5l6/t9qjkFSbGxsNvUk89xdmDs3vdd6moIFC6pjx47GyOPz58/rjz/+0Pr167Vlyxan0T8HDx7UTz/9ZLw3S8kf4PPmzWt6Lt7K91pXX3zcyuu52hwgLCzM7feHrOpjpUqVVKlSJT377LNKSEjQ/v37tXnzZq1Zs0b//POPU/0vvvhC3bp1u6lNDm6VhISETK1D5mr0kbvrVmXVlEXgZt3cpygA8BBeXl5q166d6Vjqoc+Ou77c7LQ2KXmXL0eudknJKrt27dK+fftMx2rWrKkFCxaobdu2lkIkyfVOXjmVqy2uM7so9enTp52+IQ0KCsrwPFdT2VIsW7bMNALsnnvuUY0aNdJt73Y/r5Bz5aTniuOH6Fu9XszNcjVVxN2pSVeuXMmq7iADZcuWVc+ePfXNN9/o999/d7kjq+OurZJUpkwZU/nkyZO3rI9FixZ1ClRv5eu1UqVKTscOHTrkdjuudoe8WXny5FH9+vU1ZMgQLV26VHPmzHF6P4uLi9OaNWuczh0zZoyOHj16y/5zdOXKFfXt21ePPPKImjVrplq1amnixImZetyu2q9YsaJbbbh672RNQ2QHgiQAuUanTp1M5Y0bNxrfNK5cudI4XrZsWdWvX/+mr3fPPfc4TX3K7I5J+/bt0/z587VlyxadPHnS5bf1rv6gev31192axnXjxg2nHWJy8lS3GjVqOD3+bdu2ZaotV4t3ZhT6SFK3bt1Mo9v27NljfIB0/ODiGDq5crufV8i5GjZs6HQss8+VP//8UwsXLtTWrVt19uzZLF/PxTH0vXbtmlsbBdxurkYbuNvfEydOZFV37kiJiYk6ffq0NmzYoDlz5lg+r2jRoho/frzuuece0/HUm22kcHwN7d+/P1Nrg8XGxmrGjBlas2aNDh486HJH0ZQdylLL7Os1NDRUs2bN0rp163TkyBGnadmSVLVqVafpr6l3GrXK8Qus9ERFRenAgQNasmSJtm7davm8hg0b6ssvv3Q67urf7HYLCAjQzp07deTIEYWEhCg+Pl579uzJVFuOf5/4+fk5PU8zcvHiRadjrr5UA241prYByDUaNmyokiVLKjg4WFLyHzSbN29WiRIldP78eaNeVoxGkpJ3zqhbt64pgFi1apVGjBjh9rbU48ePN02B8vb21qZNm0xDus+dO+d0Xr169dy6zrp165yCo5y8joe3t7caNGigLVu2GMfWrFmja9euuf0N3YIFC0zlPHnyuPyg7qhUqVJq2rSp0YekpCStX79ejRo1Mn376+Pjo65du2bY3u1+XiHnKlWqlMqVK2d6b1i8eLFefPFFt6buJiYm6v333zdN/y1cuLC2bNlieaRjRu666y6nYxcvXvTYtb9cTRVxZ3el69evu5yuA2tWrFih4cOHm0Kddu3aZTjdOIWXl5eaNm1qeg92NdXw3nvv1c8//2yUo6OjtXz5cnXr1s2t/v7222/65JNPTMeGDRum/v37m441atTIFLCcOXNGO3fu1L333uvW9X766Sen4OXzzz83pvNLye/3TZs2NY3OXrVqld577z3LX0DZ7XYtW7bMUt2XXnrJ9DdGixYt1KRJE0vnSslfohQrVsw0ks8T1lLz9vZWjRo1TIHanj17dOHCBacRbek5f/68NmzYYDrWokULtze2uHTpkqlctGjRm9qtFsgsRiQByDVsNpvat29vOrZmzRrTVCMp64IkyXm9mxs3bujbb791q43Nmzc7raPTsGFDpw/7rr4ldWeESVRUlMtv/G71TjK3muO/QWxsrD7//HO32li+fLn2799vOta6dWvL6w449mHNmjVOI8hat25teS2E2/m8Qs7m+Fw5ffq06YOxFQsWLDCFSFLy8zWrQiTJ9fSNWzmN6GYVKFDAaRFcd0ZY/Pzzzzn+vTU7VatWzel33q+//upWG47TuIsVK+ZU56GHHnJaa2/y5MlujT6LjY11+bvV8e8RKXkEq2Nw8NlnnykxMdHy9a5cuaIZM2aYjuXPn18tWrRwqtujRw9TOTIyUlOmTLF8rWXLljm9N6SlYsWKpi+qtm7dqsuXL1u+VlxcnNP0cqvB4a3mOOI9ZUFwd4wZM8bpi7vevXu73RfH9013p8YBWYUgCUCu4mp6W+pv4ypXrmxpupJVnTt3VunSpU3Hpk6dqrVr11o6PyQkRO+8847T8dQLgqZw3H1NMk/ZS09sbKxee+01l38Q5vQFYdu3b++0tsKcOXMsf4v6zz//6MMPP3Q6/uyzz1ruQ9u2bU0joLZu3aqlS5ea6jj+QZ+e2/m8Qs729NNPO40u+Pjjj/X3339bOv/ff//VuHHjTMdsNptbz38rgoKCnELMw4cPZ+k1slrNmjVN5Q0bNlha0+bo0aOaNGnSLerVnaFy5cqqUqWK6diUKVOMEccZCQ0NdXq/dDXC1N/fX08++aTp2KlTp/T2229bHq07atQo06hnKTmgcrXGXrly5UyjhqTk6W3jx4+3dK3ExEQNHz7cKXB54oknXI4yatGihapVq2Y69t133+nPP//M8FpnzpzRf//7X0v9kuS0TmV8fLxGjx5t+fxFixY5fTnWqFEjy+ffSl27dnUKln/++Wdt3LjR0vlTpkxx+lLz/vvvz9Tjc1znKiuWagAygyAJQK5Sr14901Dj8PBw07bAjkHTzcqTJ49GjBhhOpaUlKQhQ4Zo6tSp6a61cOTIEfXq1cvpD+OGDRuqTZs2TvXvv/9+p2Offvqp9u7dm24f9+/fr6eeesppSHWKm9na1xN4e3s7/RvY7XYNGzZMU6ZMSXdUwLp169S7d2+nxSu7d+/u1h94vr6+ppFucXFxpmktgYGBat68ueX2bufzCjlbQECAhgwZYjoWExOjPn366Oeff053p7Ft27apT58+Tu8BXbt21d13353lfXX8IO/pQVLr1q1N5cTERA0YMMBpaklqa9euVZ8+fTx6/aecok+fPqbytWvX9Oyzz2a4+HNwcLAGDhzotNuYY2CUon///k5fRqxYsUIDBgxwCohSi4yM1IgRI5xGAPr4+Gjo0KFpnvfqq686TUWaPn26Xn/99XQ3wwgLC9OgQYNMU7ml5GmoAwcOdHmOl5eX3n77bdPi8QkJCXrxxRe1ePHiNK+1e/du9erVy61F8evUqeMUaqxYsULvvfdehlPU1q5d6zQ1sGzZsmrWrJnl699KxYsX16BBg0zHEhMTNXTo0HR/jlFRUfrvf//rtDi3n5+fWyFdiuvXrzstc9C4cWO32wGyAmskAchVbDabOnTooO+//97l/a52crlZ7du3V+/evTVr1izjWGJioj799FPNnTtXXbp0UcOGDVW8eHElJCTo5MmTWr9+vdauXes0nD0gIEATJkxweZ22bduqdOnSpoUWIyIi1KtXL3Xu3FmtWrVSuXLl5OPjo/DwcB05ckTr1693mt7k5+dn+pBz9epVJSYmuj1P35M89NBDevbZZ/XDDz8Yx5KSkjRx4kT9/PPP6tq1q+rXr6+iRYsqIiJCx44d0/Lly/XXX385tXX33Xfr/fffd7sPPXr0SHNB2Icfftjtn+/tel4h53v++ee1c+dOrVu3zjgWExOjd955RzNmzFCnTp1Up04dFS1aVDExMfr333+1atUqpw+kUvKirZl5/lvRuHFj0yjKvXv3ym63u9whzRP06NFD33zzjSmQOHPmjDp06KDHH39c9957r8qUKaOYmBgdPHhQv/32m2kkWKNGjTK9mDKkxx57TAsWLDBNOz558qS6d++u1q1bq3nz5qpYsaIKFCigmJgYnT9/Xtu3b9eyZcucwtG2bdumuV5P4cKF9b///U+9evUyffHw559/qkOHDurQoYNatGihsmXLKm/evAoJCdHOnTu1ePFil+tmjRgxwmk0VWpBQUH65JNP9PLLL5uOL1u2TBs3blTnzp3VtGlTlS5dWt7e3rp48aK2bt2qJUuWOC2qbbPZNGbMGBUtWjTN6zVp0kT9+vXTd999ZxyLjY3Vm2++qblz56pbt26qUqWKfHx8dObMGa1evVpr1qwxpqkVLlxY165dS7P91N5991098cQTptFc8+fP14YNG9S1a1c1bNhQJUqUkI+Pj65fv65///1Xa9ascTlt9O233/aov0uef/55bdmyxfQ3VXR0tN58803Nnj1bDz/8sO6++27lz59foaGh2r59u5YsWaKQkBBTOzabTf/9738t7QrryHGRb29vb0trOQK3AkESgFynY8eOLoOkmjVrqkKFCrfkmiNGjFBiYqJTkHD+/HnL6xEEBARo6tSpaS7emC9fPn344YcaNGiQKShISEjQr7/+amn9iE6dOumxxx5T3759TecfOnRItWvXttRPTzV8+HDFx8c7/RucOXNGX331laU26tatq2+++catnfBS1K5dW9WqVXO5wK7jOjZW3Y7nFXKHTz/9VMOGDXNam+vYsWOW1wwrV66cvvvuu1u2cGvr1q310UcfGR9Qr1y5oqNHj2bpdOOsVKRIEb3++usaOXKk6Xh0dLRmzpypmTNnpnlutWrVNHHiRJfr1sAab29vTZw4Uc8884xphGV8fLxWrVplmraennvuucdptIujunXravLkyRo6dKhp2lhcXJyWLFmiJUuWWLrWoEGD1KtXrwzrPfTQQxozZozee+89U3gVERGhefPmad68eRm24eXlpQ8++MBp5Jwrb7zxhsLCwrRo0SLT8T179qS7A5mvr6/GjRuX5ognR7Vq1dI777yjUaNGmdZLCgkJ0ffff5/ml3yOXnnlFbVt29ZS3dvF19dXX3/9tZ577jkdOHDAdN++ffss7W6X8m+W2bU6HQO3++67z+1NOICswtQ2ALlOnTp1VK5cOafjWbnItiNvb2+NHDlSo0aNUkBAgNvnN2jQQHPnznXaGthRy5YtNWbMGKe5+hkpXry4xo0bp//973+qV6+efHx8TPdv2rTJ7T57mjx58mjkyJH68MMP3d6xLU+ePHrhhRc0a9asdL/ZzYirdZDq1q2b6Z2pbtfzCjlf/vz5NWnSJA0ZMsTt9wdJevDBBzV37lyXu6tlldKlS6tBgwamY3/88cctu15WePLJJzVixAin98z0NGvWTLNmzcrUaxZm5cuX1+zZs93eoTRF165dNWvWLKcFtV1p3ry55s2bpzp16rh9nYCAAI0dOzbdKW2OunfvrhkzZqhSpUpuX6906dKaMmWKnnjiCUv1bTabPvnkEw0fPtzyIvoFCxbU//73P7d/Hk8//bTGjRvn9u9hKXmR+9GjRzuN1vIUBQsW1Jw5c/T000+7fW7p0qX1/fffW/43c8Vxbatb+XctkBGCJAC5kuNaSDabLcvXR3LliSee0Lp16zRkyJB0h7ZLyd9M3Xffffrf//6nOXPmWN554+GHH9Yvv/yibt26pfsHoZeXl2rWrKm33npLq1ev1iOPPCIpeXFRx28w586dm+66OznJk08+qbVr12rw4MEZjkALDAxUnz59tHr1ar3++uvKmzfvTV374YcfdvrA2b1795tqU7o9zyvkfDabTS+99JLWrl2rvn37Zjh1Ik+ePGrdurW+++47ffPNN7dlRz/HDz5prd3mSZ577jktWbJEPXr0SDeQqF69uiZMmKBp06YRImWhoKAgzZ07V+PHj9d9990nL6/0P77kz59fnTp10rx58zRhwgS3RmxUqVJFCxYs0Ndff60HHnggw98JZcuW1SuvvKIVK1aoW7dulq+TolGjRlq2bJnGjh2rhg0bKk+e9CeLVKlSRW+99ZaWL1+eqdFu/fr10/Lly/Xoo4+mOfLW19dXHTt21OLFizM9Kujhhx/WihUrNGjQIKf1p1wpW7asBg8erDVr1mR6BO/tkjdvXo0cOVJLlizRI488kuHurpUrV9aIESO0atWqNKdXWnH+/HnTiGcfHx+nBc6B28lmTz3uEACQpYKDg7V//36FhYUpPDxcNptNhQoVUvny5VW7du2bHpIcHR2tAwcO6OTJk7p27ZrsdruKFCmiEiVKqG7dunyYkXT27FkdPnxYV65c0bVr1+Tr66tixYqpRo0aqlq1aoYfSjzRrX5eIfc4ffq0jh49ajxXfHx8VKhQIVWsWFG1atVSvnz5bmt/rl27ppYtWxqL73p5eWnjxo0ud6X0RPHx8dq9e7fOnTunK1euyGazqVSpUqpTp84tHc2F/xMREaEjR47o1KlTioiIUFRUlHx8fFSsWDFVrFhRNWvWtDzqJiMxMTHau3evLl26pPDwcMXExKhAgQIqXry4atasmal1btITERGhPXv2KCQkRFevXlVCQoIKFCigkiVLqnbt2ln6OomOjtbff/+t48eP68aNGypUqJBKlSqlhg0bZmo0UXouXLigI0eO6NKlS4qMjFR8fLz8/PxUunRp1ahRI0e/dlKWBzhx4oTCw8MVFRUlPz8/lSlTRjVr1rQUpFnx7bffmtY67NKliz799NMsaRvIDIIkAAAA3DYjR47U3LlzjfKIESP03HPPZV+HAMDDde/eXYcOHTLKCxYsyNQ0TCCr5LyvYQEAAJBj9enTx7RTm+MCwACA/3PkyBFTiFS/fn1CJGQ7giQAAADcNpUrVzatvXL06FHt2rUrG3sEAJ5r9uzZpvKAAQOyqSfA/yFIAgAAwG31yiuvmEYlzZkzJxt7AwCe6fr161q6dKlRrl27ttq0aZONPQKSESQBAADgtqpevbppB7dVq1bp3Llz2dgjAPA8P/30k7E5gSQNHTo0+zoDpMJi2wAAALjtzp8/r86dOxsfknr27KmPPvoom3uF1GbPnq3t27ff1mt26NBBnTp1uq3XBDxRZGSkWrdurfDwcElSq1atNGXKlOztFPD/5cnuDgAAAODOU7ZsWQ0aNEgTJ06UJP3yyy8aNGiQypQpk809Q4r9+/dr1apVt/WaVatWva3XAzzV7NmzjRApb968evfdd7O3Q0AqTG0DAABAtujbt68RHMTHx+uLL77I5h4BQPa7du2apk2bZpRffPFFBQUFZWOPADOCJAAAAGQLHx8fTZgwQb6+vpKkxYsXa//+/dncKwDIXpMmTTJGI9WvX18DBw7M3g4BDlgjCQAAANlqxowZ+uSTTyQlf2iaO3duNvcIALLHiRMn1LVrVyUkJKhAgQJavHgxo5HgcQiSAAAAAAAAYAlT2wAAAAAAAGAJQRIAAAAAAAAsIUgCAAAAAACAJQRJAAAAAAAAsIQgCQAAAAAAAJYQJAEAAAAAAMASgiQAAAAAAABYQpAEAAAAAAAASwiSAAAAAAAAYAlBEgAAAAAAACwhSAIAAAAAAIAlebK7A4ArCQkJunHjhlH29fWVlxe5JwAAAADgzpaUlKS4uDijXLBgQeXJc/viHYIkeKQbN27oxIkT2d0NAAAAAAA8WqVKlVSkSJHbdj2GeAAAAAAAAMASgiQAAAAAAABYwtQ2eCRfX98076tUqZLy5ct3G3sD4GbFxMSYpqvyOgZyHl7HQO7AaxnI+Rxfx+l9fr4VCJLgkdJbWDtfvnzKnz//bewNgKzG6xjI+XgdA7kDr2Ug57vdG1MxtQ0AAAAAAACWECQBAAAAAADAEoIkAAAAAAAAWEKQBAAAAAAAAEsIkgAAAAAAAGAJQRIAAAAAAAAsIUgCAAAAAACAJQRJAAAAAAAAsIQgCQAAAAAAAJYQJAEAAAAAAMASgiQAAAAAAABYQpAEAAAAAAAASwiSAAAAAAAAYAlBEgAAAAAAACwhSAIAAAAAAIAlBEkAAAAAAACwhCAJAAAAAAAAlhAkAQAAAAAAwBKCJAAAAAAAAFhCkAQAAAAAAABLCJIAAAAAAABgCUESAAAAAAAALCFIAgAAAAAAgCUESQAAAAAAALCEIAkAAAAAAACWECQBAAAAAADAEoIkAAAAAAAAWEKQBAAAAAAAAEsIkgAAAAAAAGAJQRIAAAAAAAAsIUgCAAAAAACAJQRJAAAAAAAAsIQgCQAAAAAAAJYQJAEAAAAAAMASgiQAAAAAAABYQpAEAAAAAAAASwiSAAAAAAAAYAlBEgAAAAAAACwhSAIAAAAAAIAlBEkAAAAAAACwhCAJAAAAAAAAlhAkAQAAAAAAwBKCJAAAAAAAAFhCkAQAAAAAAABLCJIAAAAAAABgCUESAAAAAAAALCFIAgAAAAAAgCUESQAAAAAAALCEIAkAAAAAAACWECQBAAAAAADAEoIkAAAAAAAAWEKQBAAAAAAAAEsIkgAAAAAAAGAJQRIAAAAAAAAsIUgCAAAAAACAJXmyuwMw27Fjh3777Tft3r1bwcHBioqKUsGCBVWuXDk1atRIPXr0UNWqVd1uNzg4WPPnz9cff/yhkydPKjIyUkWKFFGpUqXUokULdevWTUFBQbfgEQEAAAAAgNyCIMlDXL58WcOGDdO2bduc7gsLC1NYWJj27dun6dOn6+GHH9bIkSPl7+9vqe1p06bps88+U1xcnNM1L1++rH379mny5MkaPHiwBg0aJG9v7yx5TAAAAAAAIHdhapsHOHv2rHr06OEyRHJkt9v166+/6sknn9S1a9cyrD9q1CiNGzfOKURylJiYqEmTJmno0KGy2+2W+w4AAAAAAO4cBEnZLC4uTi+++KJCQkKMY3fddZfGjh2rLVu26MCBA1q3bp1GjBihgIAAo84///yjoUOHptv2nDlzNHv2bKPs7++vN954Q2vXrtX+/fu1du1avfHGGypQoIBRZ/Xq1Zo0aVKWPT4AAAAAAJB7ECRls7lz5+rff/81yvXr19eiRYvUrVs3BQYGysfHR+XKldNzzz2nhQsXqlSpUkbdP//8U2vWrHHZbmhoqMaPH2+UAwICtGDBAg0YMEBBQUHy9fVVUFCQBgwYoIULFyowMNCoO3XqVJ05c+YWPFoAAAAAAJCTESRls8WLFxu3fXx8NHHiRNMIodTKlSun0aNHm4798ssvLutOmzZNUVFRRvnjjz9W5cqVXdatUKGCvvzyS3l5JT8d4uPjGZUEAAAAAACcECRlo9jYWB08eNAoN2rUSGXKlEn3nAceeEAlSpQwyn///bdTnbi4OC1atMgoV6tWTW3btk233Xr16ql9+/ZGeeXKlYqIiMjwMQAAAAAAgDsHQVI2CgsLM5WLFStm6byyZcum2YYk7d69W+Hh4Ua5c+fOltrt0qWLcTsuLk7r1q2zdB4AAAAAALgzECRlI19fX1P54sWLls5LvTC3q2lwjru/3XfffZbabdy4sWw2m1H+/fffLZ0HAAAAAADuDARJ2ahYsWIqXry4Uf7777919uzZdM/5+++/de7cOaNcr149pzpHjx41bttsNtWoUcNSfwoVKmSaWnfgwAFL5wEAAAAAgDsDQVI269q1q3E7ISFBw4cPNy2SnVpYWJhGjBhhOta7d2+neidOnDBulyhRQvnz57fcn6CgIOP2uXPnlJiYaPlcAAAAAACQuxEkZbMXX3xRpUuXNsq7d+9Wjx499Ouvvyo0NFQJCQkKDg7WggUL1L17dx0/ftyo27t3b7Vo0cKpzdRT31IvzG1F6hFSCQkJCg0Ndet8AAAAAACQe+XJ7g7c6QICAjRjxgwNHjzYCIlOnjyp4cOHp3mOn5+f/vOf/+i5555zui82NlaRkZFGuXDhwm71x7F+eHi4SpYs6VYbt1pMTEx2dwGAmxxft7yOgZyH1zGQO/BaBnK+7H7dEiR5gAoVKmjx4sWaM2eOPv/88zSntklSwYIFNW7cOLVu3drl/alDJEny9/d3qy9+fn7ptucJUk/dA5Az8ToGcj5ex0DuwGsZgLuY2uYBYmJiNG3aNH377bfphkiSdOPGDQ0aNEgDBgxQcHCw0/3x8fGmso+Pj1t9yZPHnC0mJCS4dT4AAAAAAMi9GJGUzS5duqQBAwbon3/+MY41b95cvXr1Ut26dVWgQAGFhoZq69at+v777/Xvv/9Kkn7//Xc99thjmjVrlipUqGCc67g4ts1mc6s/Xl7mbDEpKcnNRwQAAAAAAHIrgqRsFBMTo/79++vYsWOSkkOfkSNH6qmnnjLVK1OmjB599FE98sgj+uCDD7RgwQJJ0uXLlzVo0CD9+uuv8vX1lSR5e3ubznU3CHIMolLa9SSVKlVSvnz5srsbANwQExNjGjrP6xjIeXgdA7kDr2Ug53N8Hd9uBEnZ6NtvvzVCJCl5BzfHECm1PHny6KOPPlJISIg2btwoKXlO8+zZs/X8889LkvLnz286JzY21q0+Odb3xCApX758To8TQM7C6xjI+XgdA7kDr2UA7mKNpGySmJiouXPnGuVixYrpxRdfzPA8m82md99917SW0bx584zbBQsWNI1KioiIcKtfjvUDAgLcOh8AAAAAAOReBEnZ5NixYwoNDTXKrVu3tjykNCgoSPXr1zfKJ0+eNNqy2WwqWrSocV9YWJhb/UrdJyk54AIAAAAAAJAIkrLNmTNnTOXKlSu7dX6NGjVM5QsXLhi3y5UrZ9x2tbNbelLXL1asGMNcAQAAAACAgSApmziuRZR6qpoVfn5+pnJcXJxxu1q1asbta9euOY0ySktSUpJOnTpllKtWrepWnwAAAAAAQO5GkJRNihQpYiqnHlFkRXpT0OrUqWO6b//+/ZbaPHbsmKKiooxyvXr13OoTAAAAAADI3QiSsknFihVN5S1btrh1/q5du4zbfn5+Klu2rFF+4IEHTHVTdnjLiGM9x3YAAAAAAMCdjSApm5QtW9Y0deyff/7R1q1bLZ27YcMGnT592ig3adJEvr6+Rrl06dKmxbiXLVumq1evpttmXFycafe3MmXKqFGjRpb6AwAAAAAA7gwESdmod+/epvI777yjK1eupHtOcHCw3n//fdOxvn37OtXr06ePcTsiIkJvv/22kpKS0mz3448/1vnz543yM888Iy8vnh4AAAAAAOD/kBRko8cee0w1a9Y0yufPn1ePHj3SnOb2xx9/qGfPnrp8+bJxrEuXLi5HDnXo0EG1a9c2yuvXr9fQoUN1/fp1U73Y2Fh98MEHmjt3rnGsfPnyTiEXAAAAAACAe1uFIUt5e3vr66+/Vs+ePRUcHCxJunTpkvr166cKFSqoQYMGCgwM1LVr1/TXX3/p2LFjpvNr166tjz/+2GXbXl5eGj9+vJ566iljWtuqVau0efNmtW7dWmXKlFFISIg2bNig8PBw47y8efNq4sSJpqlyAAAAAAAAEkFStitVqpTmz5+vV155Rfv27TOOnzp1SqdOnUrzvHbt2mnMmDHKly9fmnUqVqyoGTNmqH///goJCZEkRUVFadmyZS7rFyhQQJMmTTKNZAIAAAAAAEjB1DYPUKpUKc2ZM0cff/yxKleunGY9m82m+vXra/LkyZo0aZL8/f0zbLtGjRpavny5evXqlWZ9Hx8fde7cWUuXLlXTpk0z/TgAAAAAAEDuxogkD+Hj46PHHntMjz32mM6fP6+///5boaGhioyMVKFChVSiRAk1bNhQRYsWdbvtQoUK6f3339dbb72lHTt26Ny5cwoPD1e+fPlUvnx5NWjQQAEBAVn/oAAAAAAAQK5CkOSBypYtq7Jly2Z5u76+vmrWrFmWtwsAAAAAAO4MTG0DAAAAAACAJQRJAAAAAAAAsIQgCQAAAAAAAJYQJAEAAAAAAMASgiQAAAAAAABYQpAEAAAAAAAASwiSAAAAAAAAYAlBEgAAAAAAACwhSAIAAAAAAIAlBEkAAAAAAACwhCAJAAAAAAAAlhAkAQAAAAAAwBKCJAAAAAAAAFhCkAQAAAAAAABLCJIAAAAAAABgCUESAAAAAAAALCFIAgAAAAAAgCUESQAAAAAAALCEIAkAAAAAAACWECQBAAAAAADAEoIkAAAAAAAAWEKQBAAAAAAAAEsIkgAAAAAAAGAJQRIAAAAAAAAsIUgCAAAAAACAJQRJAAAAAAAAsIQgCQAAAAAAAJYQJAEAAAAAAMASgiQAAAAAAABYQpAEAAAAAAAASwiSAAAAAAAAYAlBEgAAAAAAACwhSAIAAAAAAIAlBEkAAAAAAACwhCAJAAAAAAAAlhAkAQAAAAAAwBKCJAAAAAAAAFhCkAQAAAAAAABLCJIAAAAAAABgCUESAAAAAAAALCFIAgAAAAAAgCUESQAAAAAAALCEIAkAAAAAAACWECQBAAAAAADAEoIkAAAAAAAAWEKQBAAAAAAAAEsIkgAAAAAAAGAJQRIAAAAAAAAsIUgCAAAAAACAJQRJAAAAAAAAsIQgCQAAAAAAAJYQJAEAAAAAAMASgiQAAAAAAABYQpAEAAAAAAAASwiSAAAAAAAAYAlBEgAAAAAAACwhSAIAAAAAAIAlBEkAAAAAAACwhCAJAAAAAAAAlhAkAQAAAAAAwBKCJAAAAAAAAFhCkAQAAAAAAABLCJIAAAAAAABgCUESAAAAAAAALCFIAgAAAAAAgCUESQAAAAAAALCEIAkAAAAAAACWECQBAAAAAADAEoIkAAAAAAAAWEKQBAAAAAAAAEsIkgAAAAAAAGAJQRIAAAAAAAAsIUgCAAAAAACAJXmyuwN3surVq2dZWzNnztR9992Xbp3g4GDNnz9ff/zxh06ePKnIyEgVKVJEpUqVUosWLdStWzcFBQVlWZ8AAAAAAEDuQpB0h5g2bZo+++wzxcXFmY5fvnxZly9f1r59+zR58mQNHjxYgwYNkre3dzb1FAAAAAAAeCqmtuUCefPmValSpdK8f9SoURo3bpxTiOQoMTFRkyZN0tChQ2W327O6mwAAAAAAIIdjRFI2Onr0qNvnxMXFqU+fPtqzZ49x7JNPPtFdd93lsv6cOXM0e/Zso+zv769BgwapQ4cOKlmypIKDg7Vy5Up98803ioiIkCStXr1akyZN0pAhQ9zuHwAAAAAAyL0YkZTDfPLJJ6YQqW/fvurcubPLuqGhoRo/frxRDggI0IIFCzRgwAAFBQXJ19dXQUFBGjBggBYuXKjAwECj7tSpU3XmzJlb90AAAAAAAECOQ5CUg2zYsEFz5swxyjVr1tSrr76aZv1p06YpKirKKH/88ceqXLmyy7oVKlTQl19+KS+v5KdEfHy8Jk2alEU9BwAAAAAAuQFBUg5x9epVvfvuu0bZx8dH48ePl6+vr8v6cXFxWrRokVGuVq2a2rZtm+416tWrp/bt2xvllStXGtPdAAAAAAAACJJyiAkTJig0NNQov/jii2mOLpKk3bt3Kzw83CinNf3NUZcuXYzbcXFxWrdunfudBQAAAAAAuRJBUg6wd+9eLVy40ChXqFBBL7zwQrrnbNu2zVS+7777LF2rcePGstlsRvn33393o6cAAAAAACA3I0jKAcaMGSO73W6U33zzzTSntKVIvSOczWZTjRo1LF2rUKFCKlOmjFE+cOCAm70FAAAAAAC5FUGSh1u9erVpl7YmTZqodevWGZ534sQJ43aJEiWUP39+y9cMCgoybp87d06JiYmWzwUAAAAAALkXQZIHs9vt+uKLL0zHXnvtNUvnhoSEGLdLlCjh1nWLFy9u3E5ISDCtzQQAAAAAAO5cebK7A0jbunXrdOzYMaPcqlUr1alTJ8PzYmNjFRkZaZQLFy7s1nUd64eHh6tkyZJutXErxcTEZHcXALjJ8XXL6xjIeXgdA7kDr2Ug58vu1y1Bkgf79ttvTeWXXnrJ0nmpQyRJ8vf3d+u6fn5+6baX3VJP2wOQM/E6BnI+XsdA7sBrGYC7mNrmoQ4cOKC9e/ca5QYNGlgajSRJ8fHxprKPj49b186Tx5wvJiQkuHU+AAAAAADInQiSPNRPP/1kKj/77LOWz3VcHNtms7l1bS8v89MiKSnJrfMBAAAAAEDuxNQ2DxQTE6Ply5cb5cDAQLVt29by+d7e3qayu0GQYxDl6+vr1vm3WqVKlZQvX77s7gYAN8TExJiGzvM6BnIeXsdA7sBrGcj5HF/HtxtBkgdav369oqKijHLXrl2dppulJ3/+/KZybGysW9d3rO9pQVK+fPmcHiOAnIXXMZDz8ToGcgdeywDcxdQ2D7Ry5UpTuXv37m6dX7BgQdOopIiICLfOd6wfEBDg1vkAAAAAACB3IkjyMPHx8dqyZYtRrlixoqpVq+ZWGzabTUWLFjXKYWFhbp0fGhpqKhcrVsyt8wEAAAAAQO5EkORh/vrrL0VGRhrlNm3aZKqdcuXKGbeDg4PdOjd1/WLFijHUFQAAAAAASCJI8jhbt241lR988MFMtZN6FNO1a9ecRhmlJSkpSadOnTLKVatWzdT1AQAAAABA7kOQ5GF2795t3M6fP7/q1q2bqXbq1KljKu/fv9/SeceOHTMt9F2vXr1MXR8AAAAAAOQ+BEkeJDEx0RT41K5dWz4+Pplq64EHHjCVN27caOk8x3qO7QAAAAAAgDsXQZIHOXnypKKjo42yu4tsp1a6dGnVr1/fKC9btkxXr15N95y4uDjNmzfPKJcpU0aNGjXKdB8AAAAAAEDuQpDkQf755x9TuVKlSjfVXp8+fYzbERERevvtt5WUlJRm/Y8//ljnz583ys8884y8vHiKAAAAAACAZKQEHiT1IteSVLJkyZtqr0OHDqpdu7ZRXr9+vYYOHarr16+b6sXGxuqDDz7Q3LlzjWPly5dX7969b+r6AAAAAAAgd8mT3R3A/7l48aKpHBgYeFPteXl5afz48XrqqaeMaW2rVq3S5s2b1bp1a5UpU0YhISHasGGDwsPDjfPy5s2riRMnytfX96auDwAAAAAAcheCJA8SFhZmKvv5+d10mxUrVtSMGTPUv39/hYSESJKioqK0bNkyl/ULFCigSZMmmUYyAQAAAAAASExt8yhRUVGmct68ebOk3Ro1amj58uXq1auX/P39Xdbx8fFR586dtXTpUjVt2jRLrgsAAAAAAHIXRiR5kOnTp9+ytgsVKqT3339fb731lnbs2KFz584pPDxc+fLlU/ny5dWgQQMFBATcsusDAAAAAICcjyDpDuPr66tmzZpldzcAAAAAAEAOxNQ2AAAAAAAAWEKQBAAAAAAAAEsIkgAAAAAAAGAJQRIAAAAAAAAsIUgCAAAAAACAJQRJAAAAAAAAsIQgCQAAAAAAAJYQJAEAAAAAAMASgiQAAAAAAABYQpAEAAAAAAAASwiSAAAAAAAAYAlBEgAAAAAAACwhSAIAAAAAAIAlBEkAAAAAAACwhCAJAAAAAAAAlhAkAQAAAAAAwBKCJAAAAAAAAFhCkAQAAAAAAABLCJIAAAAAAABgCUESAAAAAAAALCFIAgAAAAAAgCUESQAAAAAAALCEIAkAAAAAAACWECQBAAAAAADAEoIkAAAAAAAAWEKQBAAAAAAAAEsIkgAAAAAAAGAJQRIAAAAAAAAsIUgCAAAAAACAJQRJAAAAAAAAsIQgCQAAAAAAAJYQJAEAAAAAAMASgiQAAAAAAABYQpAEAAAAAAAASwiSAAAAAAAAYAlBEgAAAAAAACwhSAIAAAAAAIAlBEkAAAAAAACwhCAJAAAAAAAAlhAkAQAAAAAAwBKCJAAAAAAAAFhCkAQAAAAAAABLCJIAAAAAAABgCUESAAAAAAAALCFIAgAAAAAAgCUESQAAAAAAALCEIAkAAAAAAACWECQBAAAAAADAEoIkAAAAAAAAWEKQBAAAAAAAAEsIkgAAAAAAAGAJQRIAAAAAAAAsIUgCAAAAAACAJQRJAAAAAAAAsIQgCQAAAAAAAJYQJAEAAAAAAMASgiQAAAAAAABYkic7LjpixIjsuKxLNptNo0ePzu5uAAAAAAAAeLxsCZJ++eUX2Wy27Li0id1uJ0gCAAAAAACwKFuCpBR2uz1L2kkrlEqvfU8IsgAAAAAAAHKSbA2S3JFeWORuYJTROQAAAAAAAHCWLUHSvffe61b9pKQk7d6923TMbrcrf/78uvvuu1WjRg2VKlVK/v7+8vX1VWRkpG7cuKFjx45p//79unjxoqTkUMlut6tw4cLq27evfHx8suwxAQAAAAAA5HbZEiTNmjXLct24uDi9+uqrxnpGdrtdd911lwYNGqR27drJz88vwzZ27NihKVOm6I8//pDNZtP169e1fv16ffXVVypevPjNPBQAAAAAAIA7hld2dyAj77zzjtavX29MUevZs6eWLl2qbt26WQqRJKlx48aaNm2a3nrrLWNK2759+zRkyBAlJSXdsr4DAAAAAADkJh4dJK1evVpLly41wp/u3btr1KhR8vX1zVR7zz33nIYMGWKskbRnzx599913WdllAAAAAACAXMujg6SpU6cat0uUKKF33333ptscOHCgqlSpYkyT++mnn266TQAAAAAAgDuBxwZJx48f14EDB2Sz2WSz2dS9e3fLU9nS4+3trR49ehijnC5duuS0kDcAIGvtOX9dh8PiFJPAdGIAAAAgJ/PYIOmff/6RJCPwcXent/TUqlXLVD5z5kyWtQ0AMPvrXLhx++T1BCUk2bOvMwAAAABuiscGSZcuXTKVixQpkmVt+/v7S5KxgHdISEiWtQ0A+D8pXwYk307+/7Hw+GzqDQAAAICb5bFBkuNualevXs2yti9evGgq+/j4ZFnbAID/k3rwkZ2BSAAAAECO57FBUrFixST936ihHTt2ZFnbGzZskPR/35SXKVMmy9oGALhGjgQAAADkfB4bJNWsWdO4bbfbtXDhQkVERNx0u6dOndLSpUuNgEqS6tate9PtAgAAAAAA5HYeGyRVrVpV5cqVM8pXrlzRe++9Z1pvw103btzQa6+9pri4OEnJo50aNGigkiVL3nR/AQAAAAAAcrs82d2B9PTs2VMTJ06UzWaT3W7XypUrlZiYqFGjRikgIMCtto4fP65hw4bp0KFDRns2m039+vW7NZ3PpKSkJG3ZskWrV6/W33//rcuXLysyMlIBAQGqWrWqWrRooccff1wFChRwq93g4GDNnz9ff/zxh06ePKnIyEgVKVJEpUqVUosWLdStWzcFBQXdokcFAAAAAAByA48Okp599lnNnTtXFy9eNMKfNWvWaNeuXerZs6c6deqkatWqpXl+QkKC9u7dq19++UXLli0zRiKlhEjNmjVT69atb9fDydDBgwc1cuRI7d+/3+m+kJAQhYSE6M8//9TXX3+tUaNGqWPHjpbanTZtmj777DPj8ae4fPmyLl++rH379mny5MkaPHiwBg0aJG9v7yx5PACQGottAwAAADmfRwdJefPm1cSJE/XMM88oMTHRCJPCwsI0ZcoUTZkyRfny5VOVKlVUpEgR+fv7S0qewhYWFqZ///1X8fHJ20ynTIlLaaNMmTIaN25ctj02R5s2bdLLL7/sFPa4cv36dQ0dOlShoaHq3bt3unVHjRql2bNnZ9hmYmKiJk2apKNHj+qLL74wrSEFAFmBHAkAAADI+Tw6SJKkevXq6csvv9R//vMfxcXFGQFHSjAUHR2t/fv3OwUfjmsppT7vrrvu0vTp01WkSJHb8AgytnPnTqcQqUWLFurbt6+qV6+uvHnz6tChQ/r++++1fv16o87o0aN19913q1GjRi7bnTNnjilE8vf316BBg9ShQweVLFlSwcHBWrlypb755htjIfPVq1dr0qRJGjJkyC16tAAAAAAAIKfy2MW2U2vVqpUWLFigu+++W3a73Zialvq/lOPp3S9JjzzyiBYuXKgyZcpk86NKFhUVpWHDhpkWAP/www/17bffqkmTJipatKj8/f117733avLkyXrppZeMc5OSkjR69GiX7YaGhmr8+PFGOSAgQAsWLNCAAQMUFBQkX19fBQUFacCAAVq4cKECAwONulOnTtWZM2du0SMGcCe5mQ0SAAAAAHieHBEkSVK1atW0aNEiffbZZ2rcuLFTeCTJCI0kme7z9vZW+/btNX/+fI0dO9bthapvpZkzZ+rixYtGefDgwXryySfTrD9kyBA1bNjQKB88eNDlmkrTpk1TVFSUUf74449VuXJll21WqFBBX375pby8kp8O8fHxmjRpktuPBQDSQ6QEAAAA5HweP7XNUYcOHdShQwddvnxZ27dv1759+3T69GldvnxZ0dHRkiQ/Pz8FBgaqUqVKql+/vu6//34VLlw4m3vuLCEhwTT1rHLlyho0aFCG5/Xp00d//fWXUd68ebNq165tlOPi4rRo0SKjXK1aNbVt2zbdNuvVq6f27dtrxYoVkqSVK1dq5MiRHhW6AQAAAACA7JXjgqQUJUqUUNeuXdW1a9fs7kqm7dixQ5cvXzbKL7zwgnx8fDI8r1mzZgoMDFThwoVVtGhR+fn5me7fvXu3wsPDjXLnzp0t9adLly5GkBQXF6d169bpkUcesXQuAAAAAADI/Tw6SFq8eLG++uordezYUR07dtTdd9+d3V3KUps2bTJu58+fX+3bt7d0XoECBbRly5Y079+2bZupfN9991lqN/WUQUn6/fffCZIAZJkku10SO0ICAAAAOZlHr5E0efJknT17Vt9++6169OihDh066ODBg9ndrSyze/du43b9+vWVP3/+LGn36NGjxm2bzaYaNWpYOq9QoUKmRcgPHDiQJf0BAIk1kgAAAIDcwGODpAMHDuj06dOmRbWvXr2a5oLROY3dbjcFPlk52urEiRPG7RIlSrgVUAUFBRm3z507p8TExCzrFwAAAAAAyNk8dmqb405kNptN999/v/Lly5dNPcpawcHBio2NNcply5Y1boeEhOi3337T2rVrdfbsWV25ckUBAQGqVKmSWrdurcceeyzdRbBDQkKM2yVKlHCrX8WLFzduJyQkKDQ0VCVLlnSrDQAAAAAAkDt5bJAUERHhdKxcuXLZ0JNb49KlS6Zy8eLFlZSUpJkzZ+qzzz4zdqBLERISopCQEG3fvl1TpkzRiBEj9PDDDzu1Gxsbq8jISKPs7m51jvXDw8M9LkiKiYnJ7i4AsCghMUmxMbGKjYuTPdXcNl7HQM7j+LrldQzkTLyWgZwvu1+3Hhsk3XXXXU7Hrly5kg09uTVS76omSfny5dPrr7+u5cuXZ3huWFiYhg0bpjNnzujll1823Zc6RJIkf39/t/rluAOcY3ueIPXUPQCeLSHJrhPh8aZj3jYbr2MgF+B1DOQOvJYBuMtjg6TmzZsrMDBQoaGhxjpJW7ZsUUxMTK6Y3uYY0Hz99dfau3evpOTRSf3799eDDz6oMmXKKDIyUvv27dPMmTNNu7VNmjRJ5cqVU7du3Yxj8fHmD2w+Pj5u9StPHvNTIiEhwa3zASAjXmzcBgAAAORYHrvYdv78+TVy5Eh5eSV30Waz6cqVKxo9enQ29yxrOAY+KSFSw4YNtWzZMj3//POqUKGCfH19VaRIEbVs2VLTpk3T8OHDTed9+OGHCgsLM8qOi2PbbO59Ykv5eadISkpy63wAAAAAAJB7eeyIJElq27atvvvuOw0fPtxYQHrBggW6dOmSXnnlFdWuXTube5i1ypYtq8mTJ6e7rlG/fv108uRJLViwQJIUFRWlH374Qa+++qokydvb21Tf3SDIMYjy9fV16/zboVKlSrliVBpwJ0hITFLipQjFxsXp2KmzxnFex0DOExMTY5oCw+sYyJl4LQM5n+Pr+Hbz6CBJkpo0aaIVK1bo66+/1uzZsxUbG6vNmzdr8+bNKlKkiOrVq6eqVauqcOHCKlSokNPULCtSTw27XVwFNC+99JKlxbGHDBmixYsXG6Oa1qxZYwRJ+fPnN9VNvTOcFY71PTFIypcvn9PjBOCZEhKTlDdfvNNxXsdAzsfrGMgdeC0DcJdHB0mNGzc2lZOSkoz1kqTkRac3bNigDRs23NR1siNIclwE29vbW+3bt7d0bokSJVSnTh399ddfkqTjx48rLCxMRYsWVcGCBeXt7W2MLHK1+116HOsHBAS4dT4ApObu9FoAAAAAns1j10iSpOvXr5v+S0hIkN1ul81mM/6z2+039V92KVasmKlcpkwZFShQwPL51atXN5UvX74sKflDW9GiRY3jqddPsiI0NDTdfgIAAAAAgDuXRwdJkkyhUUo5rfvd/S87lS1b1lR2dzhpoUKFTOXr168bt8uVK2fcDg4Odqvd1PWLFSvGMFcAAAAAAGDw+CDpZkcceeJoJEkqUqSIihcvbpQdRwJlJCoqylROHfhUq1bNuH3t2jXLbSclJenUqVNGuWrVqm71CQAAAAAA5G4evUbSkSNHsrsLt1TdunW1bt06SdLVq1cVGhpqCpfSc/bsWVM5KCjIuF2nTh3NmzfPKO/fv18PPvhghm0eO3bMFFDVq1fPUl8AAAAAAMCdweNHJOVmLVq0MG7b7XatXLnS0nlxcXHas2ePUa5QoYJpUewHHnjAVH/jxo2W2nWs59gOAAAAAAC4sxEkZaP27dsrb968Rnn69OmKjo7O8LzFixcrPDzcKLdr1850f+nSpVW/fn2jvGzZMl29ejXdNuPi4kyjmMqUKaNGjRpl2BcAAAAAAHDnIEjKRkWKFNFjjz1mlM+dO6ePPvpISUlJaZ5z+vRpjR8/3ij7+Pjo6aefdqrXp08f43ZERITefvvtdNv9+OOPdf78eaP8zDPPyMuLpwcAAAAAAPg/JAXZ7JVXXlGJEiWM8sKFCzVkyBBdunTJqe7WrVvVq1cv0w5t/fr1U+nSpZ3qdujQQbVr1zbK69ev19ChQ03nSlJsbKw++OADzZ071zhWvnx59e7d+6YeFwAAAAAAyH08erHtWy0mJkYLFy5Ur169sq0PRYoU0TfffKO+ffsa09XWrFmjjRs3qmnTpqpUqZISExO1Z88e7d+/33Ru48aN9fLLL7ts18vLS+PHj9dTTz1lTGtbtWqVNm/erNatW6tMmTIKCQnRhg0bTNPk8ubNq4kTJ8rX1/eWPF4AAAAAAJBz5bggKTw8XMHBwYqKilJ8fLySkpJkt9vTrG+325WQkKD4+HjFxsYqMjJSly9f1v79+7Vz505FRUVla5AkSTVr1tSPP/6ol156SadPn5YkxcfHa9OmTdq0aZPLcx588EFNnDhRPj4+abZbsWJFzZgxQ/3791dISIgkKSoqSsuWLXNZv0CBApo0aZJpJBMAAAAAAECKHBEkJSUl6aefftLChQt1+PDhLGvXbrfLZrNlWXs3o2rVqlq2bJmmT5+uuXPn6sKFCy7rValSRS+88IIeeeQRS+3WqFFDy5cv12effabFixcrMjLSqY6Pj4/atWunN954Q2XKlLmpxwEAGcmfxzPedwEAAAC4z+ODpCtXrmjgwIE6ePCgJKU7+sgdnhIgpebr66uBAwdq4MCBOnDggP7991+FhoYqKSlJgYGBql27tqpUqeJ2u4UKFdL777+vt956Szt27NC5c+cUHh6ufPnyqXz58mrQoIECAgKy/gEBgAt5vDzv/RcAAACANR4fJA0dOlQHDhwwylkRANnt9iwLpG6VWrVqqVatWlnapq+vr5o1a5albQIAAAAAgDuHRwdJq1at0s6dO53CI3dDoNTn2+125cuXTw0bNlStWrXUpUuXLOkrAAAAAABAbufRQdLs2bON2ynhUZUqVfTwww+rSpUq8vf317p16zRz5kxJyTuOTZ48Wd7e3oqJiVFYWJj++ecfrV+/XqdPnzYCpYSEBPXt21cPPPDA7X9QAAAAAAAAOZTHBkkRERH666+/ZLPZjEWxO3XqpHHjxsnb29uoV7x4cSNIiouLU2Jiopo2bWpqa/jw4Zo7d67GjBmjuLg4JSQkaNiwYVq2bJmKFi16Wx8XAAAAAABATuWV3R1Iy6FDh5SYmGiUCxcurFGjRplCJEmqXLmyAgICjNFGW7ZscWrLZrPpqaee0jfffCNvb2/ZbDZdvXpVo0aNurUPAgAAAAAAIBfx2CDp/Pnzxm2bzaYuXbrI39/fZd3atWsbU9927tyZZptNmjTRoEGDjMW2V61apX379mVtxwEAAAAAAHIpjw2Srl69Kun/1kZq0KBBmnWrVatm1D127Jji4+PTrNu/f38VLVrUGMH0448/ZlWXAQAAAAAAcjWPDZJiY2NN5dKlS6dZt0qVKsbthIQEHT9+PM26efPmVYcOHYxRSZs3b775zgIAAAAAANwBPDZIyp8/v6ns5+eXZt2goCBTOb0gSZIaNWpk3A4PD9eRI0cy0UMAQGb8/4GmAAAAAHIgjw2SChcubCrHxMSkWTclSEqZrnby5Ml02y5XrpypfOLEicx0EQCQCVdjE41pywAAAAByFo8NkgICAkzl1ItvOypZsqTy5s1rlI8dO5Zu276+vpL+L3gKCwvLZC8BAJkREZeYcSUAAAAAHsdjg6SqVatK+r+wZ8+ePenWL1++vKTkBbcPHTqUbt3w8HBTOTIyMpO9BAC4Iykp+f/XYxKytyMAAAAAMsVjg6Ry5cqpTJkykpLDod9++00RERFp1q9YsaIxVeLcuXPpjmA6cOCA0a4k+fv7Z1W3AQAWMLENAAAAyJk8NkiSpPvuu092u102m01Xr17VG2+8keZaSTVr1pT0fyOYfvzxR5f14uPj9fPPPxv1JKl48eJZ3HMAQHpYIgkAAADImTw6SHr00UeN23a7XZs2bVLXrl21aNEiRUVFmeo++OCDprqzZs3S8uXLTXXi4+P17rvv6tSpU6bjtWrVyvrOAwCcpORHoZFx2doPAAAAAJmTJ7s7kJ5GjRrpgQce0B9//CGbzSa73a6zZ8/qnXfe0dixY7V9+3ajbrVq1VSzZk0dOnRINptNCQkJev311/Xjjz+qXr16io2N1YYNG3Tx4kWjLUmqXLmy0y5uAICsZxcjkQAAAICczqNHJEnS6NGjVaxYMUnJ09ZSQqDAwECnuq+++qoREKXU27Nnj6ZPn645c+bowoULpi2nbTab+vTpc3seCAAAAAAAQA7n8UFSyZIlNWvWLFWoUMEUErkaRdSsWTM9+eSTTmFSyn8pQVSKxo0bq2fPnrfngQAAAAAAAORwHh8kSVKlSpW0ePFivfbaaypRooTsdruCgoJc1n3//ffVs2dPp/AoJUBKOd6oUSN9+eWXt/NhAMAdj5ltAAAAQM7m0WskpZY3b1698MILGjBggHbv3q28efO6rOfl5aVRo0apTZs2+v7777Vjxw7TdLYKFSqod+/eevLJJ+Xt7X27ug8AAAAAAJDj5ZggKYXNZlPDhg0zrNeyZUu1bNlSEREROnfunOLi4lSyZEmVLFnyNvQSAOAKI5IAAACAnC3HBUnuKlCggGrUqJHd3QAAAAAAAMjxPHqNpJkzZ+rChQvZ3Q0AAAAAAADIw4Ok0aNHq02bNurevbu+/PJLHTlyJLu7BAC4CXYmtwEAAAA5msdPbbPb7Tp8+LCOHDmir776SmXLllWbNm3Upk0b3XvvvcZubAAAz5dEjgQAAADkaB4fJKUERSk7r507d04zZ87UzJkzVbhwYT344INq06aNmjdvnuZObgAAAAAAALh5Hh0kVa1aVceOHTPKjqFSeHi4Fi9erMWLFytfvnxq2rSp2rRpowcffFBFihTJlj4DAAAAAADkVh4dJC1dulQXL17Uhg0btHHjRm3fvl2xsbGSZJrSZrfbFR0drfXr12v9+vXy9vZW/fr11aZNG7Vt21blypXLrocAAAAAAACQa3h0kCRJpUuX1tNPP62nn35aMTEx+vPPP7Vx40Zt2rRJwcHBRr3Uo5USEhK0a9cu7dq1S2PHjlW1atXUtm1btWnTRvfcc092PRQAAAAAAIAczeODpNTy5cun1q1bq3Xr1pKkw4cPG6OV9u/fb0x5c5wCd/ToUf3zzz/6+uuvVbp0abVu3Vpt27ZV48aN5eXl0RvXAQAAAAAAeIwcFSQ5uvvuu3X33Xdr8ODBCgsL06ZNm7Rhwwb98ccfioyMlOQcKl24cEGzZ8/W7NmzVahQIT344IMaM2ZMtj0GAAAAAACAnCLXDMcpWrSounfvri+++ELbt2/X9OnT1adPH5UvX940Uslms8lut8tut+vatWv69ddfs7nnAHDnsGd3BwAAAADclFwTJKWWJ08eNWnSRG+//bZWrVqlJUuW6MEHH3QKlAAAAAAAAGBdjp7alp5Dhw5p69at2rp1q/766y/FxMQYo5EIkQAAAAAAQE4TE5+o6LjEbO1DrgmS/v33X23btk3btm3Tzp07df36deM+VyORUo4BAAAAAAB4uovXY3TheoyUEJet/cixQdKZM2e0bds2bd++Xdu3b9eVK1eM+1KHRI7rIkmSr6+v6tevr6ZNm6pJkya3ve8AAAAAAABWXYmMSw6RlJx5ZOc8qxwTJF26dMkUHF28eNG4z1VwlHLcbrfLy8tLd999t5o0aaKmTZuqUaNGyps3721/DAAAAAAAAO6IiE3Q6atRkqSQiDgF5LHLJxv749FB0vLly43w6MyZM8bxjIIjSQoKCjKCo/vvv18BAQG3te8AABeYVQwAAABYFpuQqONXImWXdD0mXqGRsQoonJ0xkocHSa+99poxLS01V8FR0aJFdf/996tJkyZq0qSJypUrd9v7CwAAAAAAkBUSk+w6HhqlhCS7YuITdfF6rKTs/27Wo4OkFKl3WUsJj8qVK6f69eurfv36atCggWrUqJGNPQQAAAAAAMgadrtdJ8OiFJ2QqIREu86GxyjJQzYNyxFBUgq73S6bzaZGjRrp4Ycf1v3336+goKDs7hYAAAAAAECWOX8tRtdi4pWUZNfZ8GglJCVld5cMOSpIShmZtGvXLu3atUuSVKZMGWM6W5MmTVS0aNHs7CIAAAAAAECmhUbGKjgieRrbxesxiklI1P9j787j5Kjr/PG/6q6+5j6SyUUIhBsCAgoCagAR0K+wrgdyqLvCV11Q3HVdBRdZXVyBXb5q4vHTxXVFkEMUEeQWlEvu+ybJJJn76Lur667fH9XTcyeZZGa6e+b1fDzyYKpTVfPpMD1d/a7P+/OSRAGezxlJO/X1r38djz/+OJ588kkYhlF+fOy6Sd3d3bj11ltx6623AgD23XffclHpqKOOQiwWq8jYiYiIiIiIiIhmIm+52JYqAggT2rKWC0EQsLw+Uk5uq3SLW1UXkj71qU/hU5/6FFzXxXPPPYdHH30Ujz32GF5++eVyIWnswtsA8Oabb+Ktt97CL3/5S0iShEMOOaRcWFq3bh0UpbKrmxMRERERERERTTRVQhsALEloiKpSeb9Kz0uq6kLSCFmWcdRRR+Goo47CxRdfjGw2i8cff7xcWOrq6irvO7aw5Lounn/+eTz//PP48Y9/DF3XccQRR+DYY4/FMcccgwMPPLBST4mIiIiIiIiICECY0PZ2KaGtaI8mtDVFVTREqmtCTE0Ukiaqq6vDKaecglNOOQUAsG3bNjz66KN49NFH8cQTTyCXy006JggCFItFPPbYY3jsscfK53nXu96F73//+/M6fiIiIiIiIiIiYDShzSwltHVlwoS2uCajLa5WeniT1GQhaaKVK1di5cqVOOuss+D7Pl588UU89thjeOKJJ/D888/DsqzyvsGYXsJMJoN77723EkMmIiIiIiIiIpoyoU2TRXTU6eWuq7EqvETSwigkjSWKItatW4d169bhC1/4AmzbxgsvvIC//vWvuOmmmzA8PFzeN6j0vz4RERERERERLVrTJbStaIhAEicXkUJcbHvODA0N4ZFHHsEjjzyCxx57DKlUCkBYQJqqqkdERERERERENB9y5vQJbYokTntcpefELKhCkm3beOaZZ/DII4/g0UcfxRtvvFH+u7HFIxaRiIjmXxAA4K9fIiIiIiJYrofNyTChLbODhLapeBWuJNV8IWnTpk14+OGH8eijj+Lpp5+GaZoAxretCYIAQRDGPbb//vtj/fr1OPHEE+d9zERERERERES0OE1MaOubYUKb78/1CHes5gpJmUwGjz76aLldrb+/v/x3E4tHI48FQQBFUXD00Udj/fr1WL9+PZYuXTrvYyciIiIiIiKixSsIAmweLsB0PTju7iW0VXq156ovJHmeh+eee6681tFrr70Gv1R+m2rW0cjjQRCgvr4exx9/PNavX48TTjgB8Xi8Is+BiIhCle7nJiIiIiKqpK6MiazlwvcDdGV2ntA2Fa/CU5KqupD0+c9/Hk8++SQMwwAwOWVtYuEIAFasWFGedXTkkUdCknbcW0hERERERERENNeGChYGZpzQNpntcY2kaT344IPjtsdW50aKR4Ig4LDDDiuvd7TPPvvM9zCJiIiIiIiIiKa1uwltUwkq3NxW1YUkYHLxCAB0XccxxxxTnnnU3NxcqeEREREREREREU1rTxLaqlHVF5JGikfNzc1473vfi/Xr1+Pd7343dF2v8MiIiIiIiIiIiKY3XUJbc2zXEtrGGkmjr/S6o1VdSNpnn33Ks44OO+ywXV54ioiIiIiIiIioknaU0NYa27WEtrFaY2p5jaVKqupC0h133FHpIRARERERERERzdhsJLSNpckzW0tprlTHKIiIiIiIiIiIFoixCW09e5DQVo2qekbSjriui5dffhnPPvssurq6kMlkkMlk4DgO/vd//3fcvj//+c9xyimnYNmyZRUaLRERAahwvgQRERER0dybmNCW24OEtmpUc4Wkzs5O/PznP8cf/vAHmKY57u+CIJg0PSydTuOqq67C1VdfjZNPPhlf+tKXsGbNmvkcMhEREREREREtAgstoW0qNVMKc10XV1xxBU477TTccsstKBaLpdXKR/9MpaenB0BYZLrvvvtw5pln4te//vV8Dp2IiMo4J4mIiIiIFqbZTGirZjUxIymZTOKCCy7AK6+8Ui4YTbUw1VTFpJFC0khMnm3b+Na3voWhoSFcdNFFcztwIiIiIiIiIlrwpktoS2gy2uJapYc3q6p+RpJpmvjc5z6Hl19+udy6NlIUCoIAsixD06b/n9Ld3V3+euyxP/rRj3D77bfPx1MgIiIiIiIiogVsuoS2pXV6pYc266q+kPStb30LL7744rgiUEtLCy6++GLcfvvtePHFF3HCCSdM29p21lln4Tvf+Q46OjrGzWYKggDf+c53kE6n5/HZEBEREREREdFCMphfuAltU6nqQtLrr7+O2267rVz4AYD3v//9uPvuu/G5z30Oa9eunbLFbSxVVfE3f/M3uOuuu/CRj3xkXMEpk8nghhtumNPnQEREREREREQLU850sT0dJrQN5K0Fl9A2lap+Vj/5yU/g+z6AcBbR8ccfj+9973uIxWIzPpeqqrjiiitw5plnllvkgiDArbfeOtvDJiKiaTj+6Neu50+/IxERERFRlbNcD5uGRxPahgs2gIWV0DaVqi0kOY6Dhx9+uFzwiUQi+Pd//3eI4p4N+V//9V/R0tJS3u7p6Rm3jhIREc2dnOXD9sKvt5Xu3BARERER1ZqRhDYvCBPaejMmgIWX0DaVqi0kvfDCCygUCgDC2Uinnnoq2tra9vi80WgUZ5xxxrgWt1deeWWPz0tERDvnAxgquggQIFV0kDLsSg+JiIiIiGhGpkpoC4AFmdA2laotJI3MEhop+Bx//PGzdu6jjjoKAMrrKw0ODs7auYmIaMcsDxg2HADhrCSHLW5EREREVEPGJrRtT48mtHXUL7yEtqlUbSEpmUyO2166dOmsnbu1tXXc9sjMJyIimh9DhgPL9eH6Abal2OJGRERERLVhYkKb5Y0mtIk7CQNbKKq2kDSyyPaIPV0baSzLssZta9rCn3pGRFRVAqA7bSIIAqRNB0m2uBERERFRlZsuoW1Fw8JNaJtK1T7T5ubmcdv9/f2zdu6JbXONjY2zdm4iIto5AYDleRgqlFrcUmxxIyIiIqLqNV1C29I6DRFl4Sa0TUWu9ACm097eDmB0HaMnnngCJ5100qyc+y9/+cu47SVLlszKeffUhz/8Ybz++uu7dewjjzwyqWVvov7+ftx888149NFHsWXLFhQKBTQ2NmLJkiU44YQTcMYZZ2DFihW79f2JiHYmGPN1W1xFyg4wXLCQ0CToioStqSL2aYlVbHxERERERFPZUUJbvb6wE9qmUrUzkg4//PByy1kQBLj99ttnZS2jnp4e3H333eUClaZpWLdu3R6fd085joNNmzbN2fmvvfZanHTSSdi4cSOee+45pNNpOI6DgYEBvPjii9i4cSNOOeUUbNy4EZ7nzdk4iGgxGy0lJTQZCV1GAKAnYyEIgnF3doiIiIiIqsFiT2ibStUWknRdx9FHH11uP8tms7jyyiv36Jye5+Eb3/gGbDv8oCIIAt7xjndAVdU9Hu+e2rRpExzHmZNzf+tb38JVV11Vft7T8TwPGzZswMUXX1z+dycimitLEjokUYDleRgsFZC2M8WNiIiIiKrIYk9om0rVtrYBwGc+8xk8/PDDEAQBQRDglltuQVtbGy688MIZn8s0TVxyySV47LHHyucTBAHnnHPOHIx85ia2tD3wwANYvnz5Hp/3hhtuwPXXX1/ejsVi+PznP48PfOADaG9vR39/P+6++2785Cc/QT6fBwDce++92LBhA774xS/u8fcnIpqOLApYmtDRlSkiWbCRUGVEVLa4EREREVF1mCqhTRbFRZXQNpWqnZEEAMceeyyOO+64ctEnCAL88Ic/xKc//Wk8+eSTk5LdpuK6Ln7/+9/jzDPPxF133QUA5fMdeuiheN/73jfXT2OXvPbaa+Wv6+rqZqWINDQ0hKuvvrq83dDQgFtuuQXnn38+VqxYAVVVsWLFCpx//vm49dZbx62x9NOf/hTbtm3b4zEQEe1IQpdRpythi1vWgu+HLW5DBWunxxIRERERzZWs6UyZ0La8QV9UCW1TqeoZSQBwxRVX4GMf+xgGBgbKxaQnnngCTzzxBGKxGJYvXz4p0e2aa65BJpNBZ2cnXnzxRZimWW7VGjlHPB7Hd77znUo8pSmNnZG03377zco5r732WhiGUd6+4oorsGbNmin33WuvvbBx40acddZZ8H0fjuNgw4YN4wpRRER7YrqO2faEBsP2YJda3NoTGrrSJuo0Baq8uN+kiYiIiGj+mY6HzcNGmNBWXNwJbVOp+iv09vZ2/PSnP0UikQAwWggKggD5fB6vv/460ul0ef8gCPCzn/0MN998M5588kkUi8XyDKSRY2VZxve+971piyqVMLaQtP/+++/x+Wzbxm9/+9vy9tq1a3eaerdu3Tqccsop5e2777673O5GRDRXZFHA0rpwocKkYcOwPXhBgK0pYydHEhERERHNLs8PsGl4TEJbdnEntE2l6gtJQDhD53e/+x0OPfTQcUWhkT8Ayv8FUC40jd135PGOjg7ccMMNePe7312R5zKVvr6+ccWw2ZiR9Oyzz4475+mnn75Lx33wgx8sf23bNh544IE9HgsR0c7ENbn8xtybNeH7AbKWi8E8W9yIiIiIaH5MTGjbniku+oS2qdREIQkAli1bhuuvvx7f+MY3sGLFinHFohETC0xjZy8lEglccMEFuO2223DooYdW8JlMNnGh7dmYkfTXv/513PY73/nOXTru6KOPHleU+8tf/rLHYyEi2hXtCQ2yKML2/HKKW1fGhOV6FR4ZERERES0G29PFcQltnh9AV6RFndA2lapfI2ksWZZxzjnn4Oyzz8YjjzyCJ554As888wzefvtt5PP5cUUlTdPQ0dGBd7zjHTjyyCPx/ve/H9FotIKjn97YhbYlScLatWv3+JxvvPFG+WtBEHa5OFVXV4eOjg50d3cDAF5++eU9HgsREQBMs0RSmVRqcdueLiJp2EhoMqKlFLe1rfF5GSMRERERLU6Deat8M3NsQtvyen1RJ7RNpaYKSSMEQcDxxx+P448/vvyY7/vIZDJwHAd1dXXQ9dqpGI6dkbR69WpomoZnnnkGf/jDH/DMM8+gp6cHtm2jubkZa9euxXvf+16cccYZOyyMbd68ufx1W1sbIpHILo9nxYoV5UJSV1cXPM+DJHFBMSKaPdI078VxTUZDREG66KA3a2J1UxQ5y8VAzkJbgtOJiYiIiGj2MaFtZmqykDQVURTR2NhY6WHslrGFJE3TcPbZZ+Ppp5+etF9vby96e3vx5z//GT/60Y/wL//yL/jQhz405TkHBwfLX7e1tc1oPC0tLeWvXdfF0NAQ2tvbZ3QOIqIdEXZwV6ctrqFge7A9HwN5C0vqdHRnTdRHZGgyi9pERERENHuY0DZzC6aQtLvS6TQaGhoq9v0Nw8C2bdvK26+88souHTc4OIivfOUrePvtt/HlL3953N9ZloVCoVDerq+vn9GYJu6fTqerqpBkmmalh0BEu8jzA1imBcu2xz0+cXuiJlXA9oyDftuBKviIKRJe73GwtjU2l8Mloh2Y+P7L92Oi2sTXMtEozw/w+mABtuvDcDxsy5hAALREFejwYZnVFfxi2x5s24Ej+xVd8XpRF5J+97vf4aqrrsLjjz9esTG88cYb8H1/3GOiKOLUU0/F3/7t32LfffdFfX09BgcH8cQTT+B//ud/8Oabb5b3/clPfoKWlhace+655cfGFpEAIBab2QeviS1zE89XaWPb9oiounl+gM1pZ9Lj3d1dOz02W/SQswP09wLLYjJEERjYLqFJ550homrA92OihYGvZVqsgiDAtpwHw/XhekBPwYUXAFFFgBqVkK30AKdQdH30FXwUdQGrl+768jWzbVE2+3V2duJTn/oULrnkEqTT6YqOZWJiW0NDA/77v/8b11xzDY499li0trZCVVUsW7YMf/M3f4Pf//7344pGAPDd734XW7ZsKW87zvgPbYqizGhMsjy+vui67oyOJyKaDU26CFkEXB9IWmFy26Dhw/J2tmw3EREREdGO9RthEckPgP5iWERSJaA1sijLJDNSdf9CPT09+POf/4w777wTDz300KRCy55wHAcbN27Ehz/8YTz55JPjUt4qpa2tDaeeeioOPvhgtLS0YOPGjXj3u9897f6iKOIb3/gGTjnllPJjruvihz/8YXnb88ZHZe9oLZLpvsdYE2dMERHNB1EQ0FJ6I8/ZAQw3gI8AvQW3Kn5/ExEREVFtSpkeUlb4OXfQ8GB7gCQC7VGJCW27oCpa24rFIq677jr85je/wfbt2yf9fX19Pf7P//k/OP/889Ha2rpb3+Opp57CN7/5TWzZsgVBEMy4uDJXTjzxRJx44okzPu7rX/86/vSnP5VnH917772wLAuapk1KWJtpIWhiIUpV1RmPby7tvffeNZXKR7SYeX4AtzcHy7aR2Ty6HtyyZcuh7eLvlpa8hVTRhSwKWNUUgSQIaKnX0B5nihvRfDJNc1wLDN+PiWoTX8u02GVNF1bSwN4BMFiwYRsOmgCsbNCrfnHtgu1Bypiok30AmYqNo+KFpEcffRRf+9rXMDQ0NO0d5nQ6jeuuuw633norvvvd7+Lkk0/e5fOn02lceeWVuO222wCgqopIe2Lp0qV417vehYcffhhAuMD2s88+i2OOOQaRyPheScua2QJhE/evtkKSruuTniMRVSfPD6DpkxfW1lQVmr5rhaDlmgp72IDj+UjbQEe9hqQFLGlQoVf5mz3RQsb3Y6KFga9lWkxMx0OvYUPTNGSKDnIuoKoKOup11OszWxKmEhzBhap6UCQf8Ha+/1ypaGvb73//e1xwwQUYHBwsF3im+xMEAQqFAr74xS+Wi0I787vf/Q6nnnoqbrvtNgRBMGURqa2tbQ6e2fw47LDDxm13dYWL1yYSiXGzkvL5/IzOO3H/SqbaERGJgoCOuvBOacZ0kDddBAA6UwZb3IiIiIhol7iej7eHCvCCAIbtoTcbJha2xNSaKCJVk4oVkl599VVceuml8DyvXCwaMVL0GfkDYFxB6bLLLsOrr7467bmTySQ+97nP4ZJLLkEqlRpXpBo5vyAIOOecc3DnnXfO7ROdQ83NzeO2U6kUgPDfqqmpqfx4Mpmc0XmHhoZ2+H2IiOZbVJXQFA1nR/bmLHh+gILtoT9XXZGsRERERFR9giDA5qQBy/PhuAG6MkUEABK6jFYulzBjFWtt+4//+A+4rjupgNTa2opjjz0W7e3tSKVSeP311/HSSy8BGC0m2baN//iP/8B111036bwPP/wwvv71r2N4eHjKGUhBEODAAw/Ev/3bv+GQQw6Z2yc5xybeiR/bgrZ8+XIMDg4CAPr7+2d03rH7Nzc3c6orEVWF1riKvO3Cdn305yx01OvoyZqojyhV389ORERERJWzPV1EznLh+wG2p4vw/AC6IpVnvdPMVKSQtGnTJjz11FPjZggpioKvfvWr+OQnPzlpseiXX34Z3/zmN/HKK6+Uj3n66afxzDPP4B3veEd5v1tuuQWXX355ebHoiUWqaDSKL33pSzj33HMnJZNVSqFQwPDwMIaGhhCJRHDAAQfs8rETZxqNbUFbu3YtnnvuOQBAJpPB0NAQWlpadnpO3/fR2dlZ3t533313eTxERHNppMWtM2kgYzpIaDISuozOpIH92+ILYv07IiIiIppdAzkLg4Vwzc7urAnL8yCLIpbX60xo200VqaaMLBANhAUeURSxceNGnHvuuZOKSABw8MEH41e/+hWOOuqocbNw7rjjjvLXP/vZz3DZZZdN2yp30kkn4a677sKnPvWpqikivf322zjiiCNw8skn46yzzsK3v/3tGR0/MlNrxEEHHVT++tBDD93hvtN56623YBhGeXvdunUzGhMR0VyKKBKaY+Hsy76cBc8LYDge+tjiRkREREQTZE0HXZkiAKA/ZyFvuRAFAcsbdChSddQFdkellwmtyL/cK6+8AmB0raIzzzwT73nPe3Z4TCQSwb//+79DUZTyWkkPPvggAOCuu+7Cf/3Xf01qZQuCAO3t7fjRj36EjRs3or29fe6e1G7Ya6+9xrWjvfjii8hkdi3CL5VK4Yknnihvt7S0YJ999ilvv/vd7x63/0MPPbRL552438TzEBFVWktMhSqLcH2/XEDqzZowbLfCIyMiIiKiamE6HjYPGwgAZIoOkkY4K2lJncZlEfZQRQpJY1unAOCcc87ZpeNWrVqFD37wg+VZSf39/di8eTMuv/xyABjXKjdy3jvvvBPr16+fnYHPMlmWcfzxx5e3HcfBLbfcskvH/vSnP4VpmuXtj370o+OKaEuXLsXhhx9e3r7jjjvKi3FPx7Zt3HTTTeXtjo4OHHnkkbs0HiKi+TLS4iYIArKWg6zphCluySJT3IiIiIiICW1zrCKFpOHh4XLRI5FIzGhdoOOOOw7AaNHon/7pn5DJZMYVkZYtW4brrrsO3/jGNxCLxWZ59LPr4x//+LjtH//4x9i0adMOj7n//vvxi1/8orxdV1c3ZTHuvPPOK3+dz+dxySWXwPf9ac97xRVXoLu7u7x9zjnnVE0bIBHRWBFFQnM0vAgYaXEruh56s2xxIyIiIlrMmNA29ypSJRhp3xIEAXvvvfeMjl27du247ddff31cEen000/H7bffXjMzad7znveMm5WUz+fxqU99Co888sikfR3Hwc9+9jN86UtfGlcQuuSSS6ZcSPsDH/jAuGS6P/3pT7j44ouRzWbH7WdZFi6//HLceOON5cdWrlyJc889d4+eGxHRWLM9V6glpkKTRXh+gL5ceJepL8cWNyIiIqLFbCShzWNC25ypSGqbbdvlr+vr62d0bFtb26THRtZG+pd/+Rd85jOf2ePxzberrroKZ511Vrnlb3BwEH//93+PAw44AO94xzsQi8XQ29uLRx55ZFJS2xe+8AWceeaZU55XFEVcffXVOOuss8ptbffccw8efvhhrF+/Hh0dHRgcHMSDDz6IdDpdPk7TNFxzzTXj1m8iIqo2wkiKW6qIrOUiYTqo0xVsSRZxYDtT3IiIiIgWm7EJbT1jEtpWMKFtVlWkkOQ4TvkCX9dnVhWc2Ko2UkS6/PLLJ7WJ1Yqmpib87//+L7785S/j2WefLT/+2muv4bXXXpvyGFVVcdFFF+GCCy7Y4blXr16NX/ziF/jsZz+LwcFBAIBhGOMS78aKx+PYsGHDuJlMRETVSi+1uA0VbPTlLEQVGSY89GRNLKuPVHp4RERERDRPsqaD7dMktMk1nNBWjWruX1OWR2tfI0WkU045pWaLSCOWLFmC6667Dt/+9rexevXqafeLRCL4wAc+gN///vc7LSKN2H///fHHP/4RZ5999rRrRimKgtNPPx1/+MMfcOyxx+7WcyAiqoSWmApdkSa0uFkoWGxxIyIiIloMRhLaACA9JqFtKRPa5kRFZiTNti9+8YuVHsKskGUZH/vYx/Cxj30M27ZtwwsvvIDh4WFYloWmpiYsWbIERx55JCKRmd9lr6urw2WXXYavfe1rePLJJ9HV1YV0Og1d17Fy5UocccQRaGhomP0nRUQ0xwRBwNI6DZ3JsB8+Yzqo1xV0poo4oC0OUeQ0ZiIiIqKFamJCW9+YhLY6JrTNiZovJK1Zs2bGC3bXgpUrV2LlypWzfl5VVcvJd0REC4UuS2iJqRjMW+jLWoiNaXFb3sAWNyIiIqKFaGxCm+36TGibJzXX2jaWIAjo6Oio9DCIiKgKNEcV6IoEPwjQW2px68+H/fFEREREtPBsS40mtHWlTSa0zZOaLiQBM1+sm4iIFqaRFjdBEJC3XGSKDgCgM2nA94MKj46IiIiIZtNAzsKQwYS2Sqj5QhLjnYmIaMRIixsQLrjtuAEsz0d3qVeeiIiIiGrfdAltK5jQNi/4L0xERAtKc1RBZEKL20DeQs5kixsRERFRrdtRQpvOhLZ5wUISEREtKGGLWziluWC7SJda3Lam2OJGREREVMumS2hrjWtMaJtHLCQREdGCo8liucWtf0yLW1dpCjQRERER1ZYdJbSNXPfR/GAhiYiIFqSmMS1uPaUWt8GCzRY3IiIiohrEhLbqIVd6AJs2bcLGjRsrdjwAXHjhhXt0PBERVR9BENBRp2NL0oBhu0gZDhqjCjpTBg5sT0ASGdZAREREVAuY0FZdKlpICoIAmzZtwg9/+MPdOhbAbh8/FgtJREQLkyqLaI2r6M9ZGMhbiKnhAoxd6SJWNUUrPDoiIiIi2plMkQlt1abi/+pBEMz4z54eP925iIho4WmKqoiqIyluFgBgyLCRNZ0Kj4yIiIiIdsR0PGxJMqGt2lS0kCQIQkX/EBHR4rA0EU57NmwXyUJYQOpMFuExxY2IiIioKo1NaCtYTGirJhUrJO3pTKLZ+ENERIuDKotoi4dpHoMFC7brw/F9bE8zxY2IiIio2kxMaOvOMqGtmlRkjaRf/vKXlfi2RES0iDVGVWQtF4btoTdrYVVTBMOGjcaIgvoI72oRERERVYuxCW3bSwltESa0VY2KFJKOPvroSnxbIiJa5DrqdGweNmA4LoYLDppjCramijhIk5niRkRERFQFRhLagiBAT9aEXUpoW86EtqpR8cW2iYiI5osiiWhPaACAoYIFq9Titi3FFjciIiKiShub0DaQt5nQVqX4f4KIiBaVhoiCmCqHKW5ZE0EQIFm0kS4yxY2IiIioUopMaKsZLCQREdGis7ROgygIKDpeOcVta8qA6/kVHhkRERHR4uN6PjYxoa1msJBERESLztgWt8FSi5tbWsyRiIiIiOZPEATYNDw5oa1OV5jQVqVYSCIiokWpIaIgrskIAPRkRlvcUqVp1EREREQ097alisjbkxPaltZplR4aTYOFJCIiWrSWJsIWN9P1MFxqcduWLrLFjYiIiGge9I9JaOvOMKGtVrCQREREi5YsiVhSNz7FzfUDbEszxY2IiIhoLmWKDrrGJLQVbCa01Qr+3yEiokWtXleQmNDilhqTFEJEREREs4sJbbWNhSQiIlr0liQ0SGLY4jZUanHbni7CYYsbERER0axiQlvtYyGJiIgWPVkSsaSU4jZcsGA6XtjilmKLGxEREdFsYULbwsBCEhEREcILmIQ+0uJmIQgCpE0HwwW2uBERERHNBia0LQwsJBEREZUsSeiQRAGW52GwVEBiixsRERHRnpsqoU2RmNBWi1hIIiIiKpFFodzilizYKNoevCDAVra4EREREe226RLaltczoa0W8f8YERHRGHW6gjpdCVvcsmGLW4YtbkRERES7ZWxCW8oYTWjrqNOZ0FajWEgiIiKaoD2hQRZF2J6Hgfxoi5vtssWNiIiIaFe5no+3xyS09edGE9oSulzh0dHuYiGJiIhognEtbsbYFjejwiMjIiIiqg0jCW02E9oWHBaSiIiIppDQZdTrCoCwxc33A2QtF0MFq8IjIyIiIqp+W5nQtmCxkERERDSNsS1uoyluJlvciIiIiHagP2dhmAltCxYLSURERNOQRKF81yxp2DBsD34QoJMtbkRERERTYkLbwldzq1tt2bIFr732GpLJJAzDgOd58H0fQRDs9jkvvPDCWRwhEREtJHFNRkNEQbrooDdrYnVTFDnLxWDeQmucU7OJiIiIRhQdD5uZ0Lbg1UQhyfd93Hzzzfif//kfbNu2bdbPz0ISERHtSFtcQ8H2YHs+Bgs22hMaujIm6nQZmsyLIiIiIqKRhDafCW0LXtX/38zn87jgggvw3HPP7dGso+kI7M8kIqKdkEopbtvTRSQNG3FVRkyT0JksYr+2eKWHR0RERFRRTGhbXKq+QfELX/gCnn32WQRBAEEQZvUPERHRroprMhojYYpbb86E7wfI2y4GckxxIyIiosWNCW2LS1XPSPrjH/+IJ598clzRZ+yspHg8jmg0Clmu6qdBREQLRGtcQ9724Hg+BvIWltTp6M6GLW7s+yciIqLFiAlti09VV2B+/vOfl78eKSAdddRR+L//9/9i3bp1iMfZTkBERPNnJMVtW6qIVNFBQpMR02RsTRWxtjXG2a5ERES0qDChbXGq2kJSf38/XnnlFQiCUG5rO/vss/GNb3yj0kMjIqJFLKbKaIyqSBk2enMWVitS2OKWDxfhJiIiIloMmNC2eFVtifCFF14Y18a211574dJLL63giIiIiEJtcRWKJIYtbqU1krozRZiOV+GREREREc09JrQtblVbSBocHASA8mykj3/842wZICKiqiAKAjrqdABA2nSQN10EADpTxpwkjBIRERFVi4kJbV0ZJrQtNlVbSMrlcuO299tvvwqNhIiIaLKoKqExGl4s9eYseH6Agu2hnyluREREtIBNTGjzAya0LTZVW0iKxWLjtuvr6ys0EiIioqmNtLi5vl8uIPVkTRTZ4kZEREQLUF/WZEIbVW8haeXKleO2h4aGKjQSIiKiqYmCgGX1YYtbZmyLW5ItbkRERLSwpIsOurPhWkj9OWs0oa2BCW2LTdX+3163bh1EUSyvi/TMM89UeERERESTRRQJTWNb3LwAhuOhjy1uREREtEAUHQ9bxiS0pYoOgFJCm8yEtsWmagtJ9fX1OO644xAEAYIgwG233QbHcSo9LCIiokla4ypUOWxxGykg9WZNGLZb4ZERERER7RlnXEKby4Q2qt5CEgB8/vOfhyAIEAQBAwMD2LhxY6WHRERENMnYFLes5SBXbnErssWNiIiIalYQBNhcSmizXB9dGRMBgHomtC1qVV1IOvzww/HpT3+6PCvpZz/7GX72s59VelhERESTRBQJzbGRFjcTnheg6HrozbLFjYiIiGrT2IS2Lia0UUlVF5IA4Ktf/So+/OEPAwB838c111yDD33oQ/jVr36F119/HZ7HZBwiIqoOrTEVmizC8wP0laZ99+XY4kZERES1Z9qEtoZIeS1jWpyquqHxl7/8JQDgoIMOwksvvYTNmzcjCAK89dZbuOKKKwAAoigiGo0iHo9DFGdeFxMEAffff/+sjpuIiBYnodTi1pkqImu5SJgO6nQFnckiDmiP86KLiIiIasIOE9pEXs8sdlVdSPrOd74z7qJ75OuRVjcA8DwPuVwOuVxut74HL+qJiGg26YqE5qiCoYKNvpyFmCKjCA89WRPL6iOVHh4RERHRDjGhrXpVy9KbVd/aBmDSQqUjC3Dv6R8iIqK50DKmxa233OJmoWCxxY2IiIiqFxPaqptXJZWkqi8kjRSRRmYhzeYfIiKiuSAIAjrqdQiCgJzlImOGd/I6U0X4Pt9/qHZZXgDH488wEdFC5PsBNg0XmNBGO1XVJcULL7yw0kMgIiLaLbosoSWmYjBvob/U4maWWtyWN7DFjWrPc91ZbM6ERdG4IuLACo+HiIhm17Z0EQXbKyW0FZnQRtNiIYmIiGiONEcV5CwXpuOhN2diRUME/XkLDREFca2q34KJxpmYPJh3fM7uJiJaQMYmtHVlTNiez4Q2mlbVt7YRERHVKkEQsLROgyAIyFsuMqXFKjuTBlvcqKZ0Z8I1MsbWjlz+DBMRLQgTE9oMJrTRTrCQRERENIdGWtyAcMFt1wtgeX75go2oFvBuNBHRwjQ2oS1ZGE1oW1bPhDaaHgtJREREc6w5qiCiSPCDAD2lAtJA3kKeKW5UI0bKSGPnILGzjYiotk1MaBvIh9cobXGNLfi0Q4u+kOS6vIgnIqK5Fba4hSluBdtFmi1uVKuCKb8kIqIas6OEtmYmtNFOLMpCUhAEeOWVV/CjH/0I69evr/RwiIhoEdBkEa2lC7P+nAXHZYsb1Y5gzH9Hap8+pyQREdWsralSQpvHhDaauZqZr+Y4Dh555BE8+uij6O/vh2EYsO1wVfkdpYb4vg/P8+A4DkzTRKFQQDKZhOd58zh6IiIioKmU4lYspbitbIxgIG+hQVeQ0GvmLZkWIYWLrRIRLRh9WRPJYimhLcuENpq5mrhqvf/++/Htb38bAwMDu3X8jgpNfKEQEdF8GWlx60waKNguUoaDxqiCzpSBA9sTkPhhnYiIiOYQE9poNlR9a9utt96Kiy66CP39/eXZR2MLQ2Mfm1gwmviYIAjj/hAREc03TRbRGg9b3AbyFmzXh+356M4UKzwyop1jNxsRUe0ybJcJbTQrqrqQ1N/fj29/+9sIgmBSEWikSDSxMDS2qDRd4Wjk7xVFwcknn1ypp0dERItUU1Qtp7j15iwAwGDBRtZ0KjwyIiIiWogcz8emYYMJbTWuWubDVPVPzP/8z//ANM1JRaBIJILVq1cjFouhp6cH3d3dAABVVXHYYYcBAEzTxPDwMHp7e8tFpZHjo9EofvjDH+Kwww5DNBqd/ydGRESLXkedji1JA4btIllw0BRTsDVVxIHtMlvcqOpkzDEptwEA/ogSEdUMJrTRbKvqQtKdd945rgBUV1eHyy67DKeeeiokKZx6d//99+PCCy8EALiuiyuvvBIdHR3lc+Tzedx///344Q9/iO3bt0MQBBSLRfzmN7/BMcccM/9PioiICIAqi2iLq+jLWRgsWIhr4ftaV7qIVU28yUHVxfF9AICHACKrSERENYUJbTTbqra1bcuWLRgcHASA8oyia665Bh/84AfLRSQAOOaYYyCKYrnd7U9/+tO488TjcZxxxhm4/fbb8d73vrfc1vbHP/4Rd9xxx7w+JyIiorEaoyqiaqnFLRu2uA0ZNjJFtrhRdfK4RhIRUU1hQhvNhaotJL311lvlrwVBwDve8Q4cd9xxk/aLxWLYd999y4tqP/7441OeLxKJ4Ac/+AEOPPDActHpu9/9LopFLm5KRESV01GnQxQEGI6L4UJYQNqaKsLz+YmdqhB/LImIasbYhLa+LBPaaPZUbSFpYGAAAMoFopNOOmnafQ888MDyvi+88MK0+6mqin/7t38rz3AaHh7GzTffPIujJiIimhlFEtGeCKeWDxUsWK4Px/exPc0bHVR9WEciIqoNExPa0iYT2mj2VG0hyTCMcdv77bfftPvus88+5a+Hh4fLLXFTOeSQQ3D00UeXC1S33377Ho6UiIhozzREFMRUudTiZiIIAgwbNtJscaMq4wejxSTOmiMiqk5MaFv4Kv0OXLWFpJFCz4j6+vpp9129evW47bFtcVNZv359+Xu8/vrryGazuzlKIiKi2bG0ToMoCCg6HpKlFrdtqSJcz6/wyIiABl0BgNJak+FjItfWICKqOkxoo/lQtYWkurq6Xd53xYoVAFBeLGzLli073H/s7Cbf9/H888/PfIBERESzaGyL2+C4FjezwiMjAiIK2yCIiGoBE9poPlTtvLaGhoZx25lMZtp9RwpJIzZt2rTDczc2NgIYLTztqBWuWrz22mv46Ec/CscJ71I/8MADWL58+S4f39/fj5tvvhmPPvootmzZgkKhgMbGRixZsgQnnHACzjjjjEn/jkRENL8aIgpylou85aInY2KvpgiSRRsNhozGKO8iEhER0fR6mdBG86RqZyS1tbUBGC32vP3229Puq+s6WlpayttvvvnmDs89UowZkUqldneY88K2bXz1q1+dNO5dde211+Kkk07Cxo0b8dxzzyGdTsNxHAwMDODFF1/Exo0bccopp2Djxo3wPG+WR09ERDOxNBG2uJmuV05x25ZmixsRERFNL2XY6GFCG82Tqi0kHXLIIVDV0buv999//w73X716dalvP8Arr7wC13Wn3Xf79u3jtqu9Ovv9739/p8Wx6XzrW9/CVVddBdu2d7if53nYsGEDLr744knrUxER0fyRJRFL6sanuLl+gG1McSMiIqIpGLaLzlR4ncCENpoPVVtIUlUVhx56aLk49OSTT+LBBx+cdv+1a9eWvzZNEw8//PC0+/75z38GMLqg90zWY5pvzzzzDH7+85/v1rE33HADrr/++vJ2LBbDV77yFdx///146aWXcP/99+MrX/kK4vF4eZ97770XGzZs2ONxExHR7qvXFSQ0GQGAnkyY4pYqOkgZO74pQERERIuL4/l4e4gJbTS/qraQBIymqwmCgCAI8I//+I+46aabppwx8453vGPcvtdccw1Mc/ICpS+88ALuvPPOcbOQJqa+VQvDMPC1r30Nvj/zdoahoSFcffXV5e2GhgbccsstOP/887FixQqoqooVK1bg/PPPx6233orW1tbyvj/96U+xbdu2WXkORES0e5YkNEhi2OI2NKbFzWGLGxEREWE0oc3xxye0NUSY0EZzq6oLSX/7t3+L+vp6AGGBqFgs4vLLL8fJJ5+M73//++P2Xb9+PaLRaHn77bffxjnnnIPHH38cxWIRqVQKN954Iz772c+Oa3tTVRWHHHLI/DyhGbrqqqt2u6Bz7bXXwjCM8vYVV1yBNWvWTLnvXnvthY0bN0IUwx8Hx3E4K4mIqMLkMSluwwULpuOFLW4ptrgRERHR9AltSxJMaKO5VdWFpLq6OvzTP/1TeQbSyGyjrq4u/OY3vxm3r67r+NjHPoYgCMr7vfzyy/i7v/s7HHHEETj22GPxb//2b8jlcuW/FwQBp512GjSt+l5ojzzyCH79618DAERRxHHHHbfLx9q2jd/+9rfl7bVr1+Kkk07a4THr1q3DKaecUt6+++67kc/nZzhqIiKaTfW6goQ+0uJmIQgCpE0HSba4ERERLWpMaKNKqupCEgB87GMfw3nnnTeumCQIwpRR9V/4whfK6W0jL56RNZZG/ox9Uem6js997nPz8CxmJpvN4tJLLy1vf+Yzn8G6det2+fhnn30W6XS6vH366afv0nEf/OAHy1/bto0HHnhgl78nERHNjSUJHZIowPLGtLil2OJGRES0WDGhjSqt6gtJAHDJJZfgiiuuQGNjY7mgNFUhqa6uDj/5yU8QiUQAjBadxv4BwuKSJEn49re/jVWrVs3fE9lF3/72t9HX1wcAWLNmDS6++OIZHf/Xv/513PY73/nOXTru6KOPHldo+8tf/jKj70tERLNPFoXyFPWRFjcvCLCVLW5ERESLDhPaqBrURCEJAD7ykY/g/vvvx+WXX453vetd2Geffabc7+CDD8att96Kgw8+eNJspJE/S5cuxY9+9KNxM3Cqxb333ovbb78dACBJEr773e9CVWe2UNobb7xR/loQBOy///67dFxdXR06OjrK2y+//PKMvi8REc2NOl1Bna4gANBdanHLmA6GC2xxIyIiWizGJrTlTSa0UeXU1E9bNBrFJz7xCXziE5/Y4X6rV6/GLbfcgqeffhr3338/urq6YNs22tvb8a53vQsnnXRSVa6LNDw8jG9+85vl7fPPPx+HHnrojM+zefPm8tdtbW3lGVq7YsWKFeju7gYAdHV1wfM8SBIr20REldae0GDYHmzPw2DBRltcw/Z0EQlNhirXzH0hIiIi2g0TE9q6s0xoo8qpqULSTB155JE48sgjKz2MXfav//qvSCaTAMIFsv/hH/5ht84zODhY/rqtrW1Gx46sMQUArutiaGgI7e3tuzUOIiKaPSMtbl2ZIpIFGwlVRkSVsDVlYN/WeKWHR0RERHNoqoS2qMqENqqMBV1IqiW/+93vyotbK4qCK6+8csYtbQBgWRYKhUJ5u76+fkbHT9w/nU5XXSHJNM1KD4GIdpHnB7BMC5Y9vgVr4jbtGhVARAiQsVx0DmWxV2MElilAFzy08G4kzSFzitexaZkoFjlrmajWTLyW5rV19evNWejLWggQYFvaRNHxoUoCWjQFtsVrqsXENl3YtgNV8oAKvgWzkFQFent7ccUVV5S3/+///b848MADd+tcY4tIABCLxWZ0fDQa3eH5qsHY1j0iqm6eH2Bz2pn0eHd3VwVGszD4QYC+vAfPB5L9Ipp0EZ2CgL3rZCgSk1pobgwaHoZMb9xjnVu2oI9tlUQ1j9fW1S1r++jOuwCAoaKHnB1AFIClMQnbMnzfX2zyto/Boo8GXcCapbu+hM1sq/lCUiqVQldXFzKZDLLZLGzbxhlnnDFun+HhYTQ3N1dmgDsRBAG+/vWvI5fLAQAOPPBAfO5zn9vt8znO+A9siqLM6HhZHv8j4brubo9lNiWLHkQRaNB455OIFjdRENCii+g3fGQsH1FZgC4DPQUPq+pq/m2diIiISkzXR28+LOJnLB85O0wwb42IUHnziCqoJq84n3/+edx666146qmnsHXr1kl/P7aQZNs23vOe9+Ad73gHPvOZz+C9733v/A10F1x//fV4/PHHAYRFn+9+97szLv6M5Xnj7xYKwsx+wYji+DuLvu/v9lhmU8r2YXsBXB9oibCYRESLW1QREVcD5O0Ag0UPy+IyDNdHyvTQqPN3JBERUa1z/QDbcx58BDCcAEkz/FzWrIuIKpwNSpVVU4WkF198Ef/xH/+B559/HkA4m2eiiYWTnp4euK6LJ598Ek8++SSOPPJIXH311ViyZMl8DHmHOjs78Z//+Z/l7X/4h3/Afvvtt0fnnJiwNtNC0MRC1O6s0zSXBosejj5oX+i6XumhENEu8PwAbm8Olm0jvXlb+fFly5ZDq7LfL7VmlR+gM1WE4wdIRBS0x1UIArCmLQ6N7UY0y3qzFrThHLJjXsd7rV6NpsTMWuiJqPJM0xzXzrb33nvz2rrK+H6AN4cMrGz0YLo+tqVNLGsM0KDLXFx7kcuaLtSchbjkAchWbBw1U0j6xS9+gf/6r/+C67rlAtLEotFUhaWRKPuRv3/qqafwN3/zN/j//r//D4cccsjcDnoHPM/D1772NRSLRQDAIYccggsuuGCPzxuJjO+TtCxrRsdP3L/aCkkAoOv6pOdJRNXJ8wNo+uRFIDVVhabzQmhPrZRkbE8XUfAAV5AR0yQMmAHWtvJ3JM0u3RagqeOvEXSN78dECwGvravP5uECfEmGDAlDeQOyIiOqSljZEJlxxwktLEogQrV8KKIEVLB5qCZuWW7cuBFXXnklHMdBEAQQBAGCICAIgvKf6fT09JS/HnnRJZNJXHDBBVO2xc2X//7v/8Zzzz0HICzWXHnllZNmE+2ORCIx7jz5fH5Gx0/cv6GhYY/HREREcyOuyWiIhO3QfTkTvh8gZ7kYyM3sJgIRERFVh96siVQx/NzblTFhez4UScSyehaRqHpUfSHpD3/4AzZu3AgA4wpIK1euxGc+8xlcddVVOOqoo6YtJu2///7j/n7kxZdKpXDppZfOz5OY4PXXX8eGDRvK21/84hexZs2aWTm3IAhoamoqbyeTyRkdPzQ0NG67WhcpJyKiUFtcgyKJsD0fA/mwgNSdNWG53k6OJCIiomqSMmz0ZE0AQF/WguG4EAUBKxoikEUWkah6VHVrWz6fx3e+853ydhAEaG5uxje/+U28//3vLz9+//33T3uOQw45BNdddx0efvhhfOMb38DAwEC5GPXMM8/gnnvuwSmnnDKnz2Oi++67b1y62n/+53+OWytpV5x44onjth944AEsX74cALB8+XIMDg4CAPr7+2d03rH7Nzc3c5orEVGVk0QBS+s0bEsVkSo6SGgyYpqMzmQRa1tjvHtJRERUAwzbRWcqXPZkuOAgbYafF5fV61z7kKpOVf9E/vznP0cqlSoXfpqbm/GrX/1qXBFpVx1//PH4zW9+gxUrVox7/Ne//vVsDbdqrF27tvx1JpOZNMtoOr7vo7Ozs7y97777zvbQiIhoDsRUGY2lFrfenAXPD5C3XQzkJ69PRURERNXF8Xy8PWTADwLkTRcD+XBWUntCQ1yr6rkftEhVdSHpjjvuKBeRBEHAlVdeidWrV+/2+VpbW/GDH/wAoiiWz/vUU0+hUCjM4qgr79BDDx23/dJLL+3ScW+99RYMwyhvr1u3bjaHRUREc6gtEba4OZ5fXiOpJ2vCdNjiRkREVK18P8DbQwU4vg/L9dFdam1riChoilZf8BFVluuPLOkz/TrR86Fqy5tbtmzBtm3byusiHXXUUTjuuOP2+Lz7778/TjrpJNxzzz0Awlk4L7/8Mt75znfu8bl31UUXXYSLLrpoRsds2LChvFYUML6VbaJ3v/vd47YfeughvO9979vp93jooYd2eB4iIqpeoiCgo07H1pSBtOmgTi+1uKUM7NcaZ4sbERFRFepMGTAcD54XoCtdhB8EiKoSliSYbkuTWW4Y1baDvLF5UbUzkt58800AKC+SPZvrGE0sqmzbtm3Wzl0Nli5disMPP7y8fccddyCVSu3wGNu2cdNNN5W3Ozo6cOSRR87ZGImIaPZFVQmNpbuXIy1uBdtDP1PciIiIqg4T2mimpNKPRYXrSNVbSJq4rs8+++wza+fee++9AYwmuGWz2Vk7d7U477zzyl/n83lccskl8H1/2v2vuOIKdHd3l7fPOecciGLV/ngQEdE02uJqucWtf0yLW5EtbkRERFVjbEJbLxPaqMZUbaVg4rpF9fX1s3ZuWR7f0bejAkut+sAHPoBDDjmkvP2nP/0JF1988aSimWVZuPzyy3HjjTeWH1u5ciXOPffceRsrERHNnpEWNwDImA7yposAQGfSKM/yJSIiosopWOMT2jJMaKMZqvQlXdWukdTQ0DBue2etWTMxODgIAOVFvCd+r4VAFEVcffXVOOuss8r/dvfccw8efvhhrF+/Hh0dHRgcHMSDDz6IdDpdPk7TNFxzzTVQ1Spb2I2ffYiIdllUldAUVZE0bPTmLOytSjAcD305C0tLRSYiIiKaf47nY9MwE9qotlXtT2pTUxOA0fazl19+Gcccc8ysnPvZZ58dt93c3Dwr5602q1evxi9+8Qt89rOfLRfPDMPAHXfcMeX+8XgcGzZsGDeTiYiIalNrXEXedmG7PvqyFpY16OjNmmiIKIgoUqWHR0REtOhMl9DWyIQ2qjFVO2/uoIMOKn8dBAHuuuuuWTmv4zi44447xi1edvDBB8/KuavR/vvvjz/+8Y84++yzEYvFptxHURScfvrp+MMf/oBjjz12nkdIRERzYWyLW9ZykGOLG82ydKkVg4iIds3YhLbtYxLa2pnQRruqShZhr9oZSUuXLsXatWvx1ltvAQBee+013HXXXTj11FP36Ly/+MUv0N3dXS4k7bPPPmhra9vj8c61iy66CBdddNFuHVtXV4fLLrsMX/va1/Dkk0+iq6sL6XQauq5j5cqVOOKIIxZkex8R0WIXUSQ0x1QMF2z05kxElRgMeOjNWuioZ4sb7Zn+nA1RNrC8QYciVe29SSKiqtCTGZ/Q5jChjfZAUOG1X6q2kAQAH/rQh/Bf//VfEAQBQRDgsssuw5o1a7B27drdOt/999+P733ve+XzCYKA0047bZZHXb1UVcVxxx1X6WEQEdE8ao2pyFsuLNdHX87EsoYI+nImGiIyompVXwZQlbKDAIYVXsAmizYypoOOOh2tcZUfhoiIphCuWciENtpzTREFKcOu+BLCVX376LzzzsOSJUsAhGsl5XI5fPKTn8Rtt902o/MYhoHvfe97+NKXvgTPG40/rq+vx3nnnTebQyYiIqoqQqnFTRAEZC0XWdMptbgV2eJGuyUIgGHTR2eqiKLtwQsCbM8U8Vp/HnnLrfTwiIiqSsFysXVCQpsgCExoo92iVsnPTFXfitQ0Dd/4xjfKLV2CICCfz+PrX/86Nm7ciBNOOAGrVq1Cd3f3uOM6OzuRyWSwZcsWPPPMM7jnnnuQy+XKs5BG/vtP//RP064bREREtFDoioTmqIKhgo2+nIWYIqPIFjfaQ6brozNloEFX0BbXUISHNwbzaI6qWFbPdjciItudOqGtLa4yoY1qWtX/9J500kn453/+Z1x11VUQBKFcCOrq6sKvf/3rSfsHQTBpHaWRO65jp1t/8pOfxEc/+tG5HTwREVGVaImpyJVa3HpzJpaXWtzqdRkxXszSDEycyJY2HeRsF21xDQ0RBcOGjXSR7W5EtLj5foBNw0xoo4WpJm4V/d3f/R0uvfRSyHJ4oTu2oDTyZ6yxj4/MPhq7/7nnnotLL720Ek+FiIioIsa2uOUsF5mRFrdUEb7PFjfaPasao9BkEZ4foDdrojNpwHRG291eH8ijwHY3IlqEmNBGC1lNFJIA4Nxzz8Wvf/1rHHjggZMKRDv7M7L/0qVL8YMf/ACXXnopRLFmnjoREdGs0BUJLbHwLmh/zoLrBTBdr7wAKNGuGFt2jKoSVjdF0Z7QIAoCio6HLUkDfVkTnh/AcDy8PpjH1qQB1/MrNmYiovnEhDZa6GpqLvvBBx+MW2+9FU8++SRuvPFGPPnkkxgaGtrhMbqu47DDDsNHPvIRnH766ZAkaZ5GS0REVH2aowpylgvT8dBXbnGz0KArbHGjXTR+BpsgCGiKqqjTZPTnbWRNB6mig6w12u42ZNhIFR0sq9fREmO7GxEtXExoo8WgJq8Yjz76aBx99NEAgK1bt+Ktt95CJpNBJpOB4zhIJBJoaGjAsmXLcOCBB0JRlAqPmIiIqDoIgoCldRo6k8Wwxa3ooD6ioDNVxAFtcYi8yKWdmC7sT5ZELKvX0RhR0Jczw/W4siYyRRftCRW6ImFbuoihgo2VDREWLolowSlYLjqTBgBgOG8zoY0WrJp/B1+1ahVWrVpV6WEQERHVDF0OW9wG81aY4qbKMOGhJxvOUCLaEyPtbknDwVDBhuG46Ey6aIyqaImp5Xa31piKjjodMtPdiGgBGEloC4Awoa1gAQDamdBGCxDfuYmIiBah5qiCiCLBD4LyFPz+vIU8F0amWSAIAppjKtY0R1GnKwgQtntsHjaQKToAgMGCjVf6cxgqfdgiIqpVO0poa2RCGy1ALCQREREtQmGLW5jilrdcpEsf7juTBlPcaNaMtLutbIxAlUW4vo+erImtySIs14frB9iaKuL1gRwMm0VMIqpNTGijxYaFJCIiokVKk0W0jklxc9wAljd6J5VotsRUGXs3RdEaD9PdDMfFluEC+nMWPD9Awfbw2kAe21IGPBYyiaiGTJfQtpwJbbSAsZBERERzJphuVV6qGk1TtLgN5C3kTM4OodklCAJaYir2bo4iocvj293M0Xa3l/uyGC7YlR0sEdEu2FFCm8TwClrAKrLq14knnliJbzslQRBw//33V3oYREREFTHS4taZNFCwXaQMB41RBZ0pAwe1J5jiRrNu5E59XnfRl7PgeD56MibSRRdLEhogi+hMGWG6W2MEEUWq9JCJiCZhQhstZhUpJHV3d0MQhKq4U83phkREtNhpsojWuIr+nIWBvIW4KgPw0ZUpYmVjtNLDoyozW1dvcU3GGlXCsOFguGDDsF1sGQ7T3VpjKvK2i9f6c2iNa+io03l3n4iqxtiEthwT2mgRqmipVBCEiv4hIiKiUFNULbe49ZSm6Q8WbGRLLUdEc2Fcu5s2pt0taSBrOggQtlq+0pdD0mC7GxFV3sSEth4mtNEiVFVz7oIgmNc/RERENKqjTg8XQrZdJAthAWlrqsjFj2nOKZKI5Q0RrGiIQJFEOJ6P7oyJbakw3c3xfWxJGnhzMI+i41V6uEQ1jZ+D9syWZJjQ5o5JaIupMhPaaFGp6Ly7IAjKLW6CIGDdunVYsWJFJYdERES0aKmyiLa4ir6chcGChbgWrk3TlS5iVRNb3CjkzeFn0LgmY29VQtJwMFSwUSi1uzXFVLREVeSssN2tLa5hKdvdiGbE9wM815OBZVrYnLSxIiEjrlTVvIKq15MxkTZHEtqKcDwfqixiWb3OjhdaVCpaSBp5sY0Uk1544QV4nofTTz8dp556Ktrb2ys5PCIiokWnMaoia7kwbA+9WQurmiIYMmw0RhXU6Uqlh0eLgFhqd6vTZPTnLeQtF8MFG1nTRXtcQ0IPH08aDlY06GwlIdpF29LFcdvbcy4OaOLrZ1dNTGgrOh5EQcDyeia00eJTkRL0GWecgXg8Pq7FbKSY9PLLL+PKK6/E+973Ppx77rm46aabkE6nKzFMIiKiRanc4ua4GC61uHUm2eJGofnqilFlESsaIlheP9ru1pUpYnu6CLvU7rY5aeCtwTxMtrsR7dRwaZ0xy/UrPJLaw4Q2ovEq8lP/3e9+F48++ig2btyI0047Dbquj+vVDYIAvu/j6aefxuWXX47jjjsO559/Pm677TYUCoVKDJmIiGjRUKSwxQ0AhgpWeY2a7RPuZtNiNb8FxYQuY+/mKFpiKgRBQN5ysSVpYDBvw/cDZC0Xr/bn0J0pwmexk2in+CqZGSa0EU1WsZ98VVVx0kkn4aSTTkKxWMSDDz6IO++8Ew8//DBsezSVIwgCuK6LRx55BI888gi++c1v4j3veQ9OO+00rF+/HqrK6ZhERESzrTGqImd5KNguerMm9mqKYtiw0RhRUB9hi9tiVokPoaIgoDWuoU5X0J+zULBdDBUsZE0H7XENcV1GX26k3S2CBv6MEk2LhaRdx4Q2oqlVRQk1EongtNNOw2mnnYZ8Po/77rsPf/zjH/H444/Ddd3yfkEQwLIs3HfffbjvvvsQjUaxfv16fPCDH8Rxxx0HSZIq+CxoLvENj4ho/i2t07B52EPR8TBccNAcU7A1VcSBqgRZ4lT+xaqSgU+aLGJlYwRZ08FA3obt+dieKSJuylgS1wH42DRcQJ0mY2VjBJrMa0Mi2n1MaCOaWlUUksaKx+M488wzceaZZyKdTuOee+7BnXfeiaeffnpS+1uhUMAdd9yBO+64A/X19Xj/+9+P008/He985zsr+AyIiIgWBkUS0Z7Q0Js1MTSS4iYD29MmVjczxW2xqoabO3W6grgmY6hgI2k4yFsuNtsFNEdVNMcUZC0Xr/TlsCShY0lCg8iFcInKgkpWg2tId6bIhDaiaVRdIWmshoYGfPzjH8fHP/5xDA4O4q677sIf//hHPP/88wBGU9+CIEA6ncYtt9yCW265BS0tLTj11FNx+umn47DDDqvgM6C5wDc/IqL50xBRkLNc5K2wxW1VYwTJoo3GosL2IaooURDQFtdQryvoy5kwbA+DBQsZ08GShIaYJqM3Z2LYsLGyIcKWTCLaZUnDRl8uXAupJzOa0LaCCW1EACq02PbuaG1txXnnnYcbb7wRf/rTn/CP//iPOOCAA8alvgFhkWFwcBDXXXcdPvGJT+Ckk07C//t//w9vvPFGJYdPsyhnMZmFiGg+LU1oEAUBRcdDspTitjVlwPWY/EOVp8kiVjVGsaxehyyKsD0f29JFdKWLcNwAtufj7eEC3h4qwHJ5DUHEW7I7NjGhLWuFCW3LG3SoTGgjAlBDhaSxOjo6cMEFF+C3v/0t7r77blx44YVYs2bNlEWlrq4u/PSnP8UZZ5yBD37wg/jxj3+Mbdu2VXL4tIf45kdENL/kUosbAAyWUtxcP8A2prgtTlX6RlynK9i7OYqmaJjulrNcbE4WMJy3EQQBMqaDV/vz6M2aTHejRY2T+6e3o4S2mFrVzTxE86omC0lj7bXXXrjwwgtxxx134Pe//z0uuOACLF++fMqi0ttvv40f/OAHOOWUU/CRj3wEv/jFL9Df31/J4dMMjLzpsbWNiGj+NUQUJDQZAYCejIkgCJAqOkgZ9k6PJZovkiigPaFhr6YIIooEPwgwULCwebiIguXBDwL0ZE282p9DpuhUerhEVEV8P8DbQ2FCm+l4owltUZUJbUQT1Hwhaaz99tsP//iP/4j77rsPv/nNb/D3f//36OjomLRIdxAEePXVV3HllVdi/fr1FRwx7Y4tySJeH8ihO1NEpujA411FIqJ5sSShQRIFmG6Y4gYA29LhAqRE1USXJezVFEVHud3Nw7a0ge50Ea4XwCq1u20aKsB2+fNLRGFCW9ENE9q6MuZoQlucRSSiiRbs/LyDDz4YBx98MP75n/8ZL730Eu69917cc8892LZt27hZSr7Pi4daUXAC6FL4/65geyjYHoBwumlUkZDQZMQ1GXHGUhMRzYmRFreezGiKm65I2JYqYk1LrNLDI5qkXlcQV2UMFmykDBtZy0XeLqAlpqEpKiNtOsj2u1ia0NCe0JjERIsCZ/dPxoQ2oplZsIWkEbZtY3BwEENDQ8hkMpUeDu2B4aIHH4A4bKAxHiCqSogqElRZhOF4MBwP/fmwsBSRJST0sKiU0GQWloiIZkm9riBnushZLnoyFlY3R5A2HSQNG02c+k9VSBIFLEloaIjI6MuG6UsDeROZooT2hIaYJqE7a2LYcLCiQUedznQ3osVkuMCENqKZWpCFJMuy8Oc//xn33HMPHnroIRiGUf47VpRrn+uHC2ZmzLCtQhZFRBUpLCypEjRZRNH1UMx7GCgdo8sSEpqEuCYjoclQWFgiItptS+p0GMMFWJ6HoYKD1riKbakif78uErU6l2Gk3S1ddDCQt2CV2t3qNAXtCQ0mPLw1VEBjRMHy+gjTmYgWgYLlYmuKCW1EM7VgCkmWZeGhhx7C3XffjYceegimGS6ONnbq5tiWNgBYvnw5TjvttPkfLO2RJXEVoqKgYHswXR+u7yNr+chaYWFJEgXEFAkRRUZMCwtLpuvBdD0MFsJFYTVJLLfCJTSZbxRERDMgl2Z4dGdMDBcsJEotbltTRezDFrcFr9YXBRhZOH4wbyFVdJC1HORtF60xDY1RGamig4zpoqNOR1tc5U1IWnBqtRg822w3XCuNCW1EM1fTr5CdFY8EQRhXPAqCAB0dHTjllFNw2mmn4ZBDDqnY2Gn3RVUJiZiGVgB+EKDoeCg6Pgq2C9Px4fkBspaLrOUC+bCwFB0pLKkiNFmE5fmwDBtDpbQhtVxYClvhNFmq7JMkIqpydbqCrOUiZ462uGVMB8MFG80xtrhRdZNEAUvqdDREFPTlwlaW/ryJdFHCkjoNUVVCV6aIoYKNlQ0RJPSavmQmGsfnGknlhDbXD5jQRrQbau5dcaR4dNddd+HPf/7zTotHANDe3l4uHq1bt64i46bZM/a+oCgIiKkyYirQElMRBAGKjh+umWR7KDoePD9AzgrX8xg5JqpKiMphK5yuiLA9H8OGjeFSF6Qiji8s6QoLS0REEy1J6DDssMVtsGCjLa5he7qIOp0tbgvZQvoIqiuT2922pgzU6wra42G725tDeTRFVCxv0PlzTbRAMKGNaM/URCHJsiw8+OCDuPvuu3e5eNTS0oJTTjkFp556Ko488sjKDJzmnTBSJFIlIBb+PJiuD8P2ysUlPwiQt1zkLRcohIWliCKV11mKKCIc30eyaCNZDM8riwLiatgGl9BlRFhYIiKCLApYmtDRlSkiWbCRUGVEVAmdSQP7tsYrPTyiXdYQURAvtbuli+E6jDnLRVtcQ2NUQbJoI2M6WMp2N6KaN5LQ5vtMaCPaXVVbSDJNs9y2tqvFo6amJpx88sk47bTTcPTRR/MXAUEoFYkiioRmhD8rljs6Y8kozVgq2C4K9mhhSR8pLClhYckFkDYdpE0HyACSIJTXV0po4fn580ZEi1FCl1FnKciaDnqyFlY3RZC1XAwVLLTEtEoPj2iXyaKApWPa3UzHQ1/ORLroYElCQ6TU7jZcsLGyMYK4VrWX0UQ0jbEJbb2lFEdJZEIb0UxV1TvgSPHorrvuwl/+8pddKh7V19eXi0fvete7IIqcckzTE0pFIl2R0BQNHzNdD0XbQ8HxULTDxbsN24Vhh61wAoCIIiOiioiVjoWIcclxkiAgpkrlBbxjKgtLRLR4tCc0GLYHu9Ti1p7Q0JU2UacpDDNYwBLqwpydG1Ek7NUYKbW72TBdD50pAw26gra4hiI8vDGYR3NUxbJ6trsR1Yr8mIS2oTEJbcvqmdBGNFMVLySZplluW9vV4lFdXR1OPPFEnHrqqTj22GMhyxV/GlTDdFmCLktoLG1bro+i46FQWmPJ8XwYjgvDAYYRFpa00vpKMUVCRA0LS+UFvjGydlOpsKSGhSWRdzloEevNWnisx4QfAHvX83f2QhPO5AjXR0oaNhKajKgqYWuKLW4L2UK+YSIIAhqjKhK6goGchUxpVnKunO6mYNiwkS466KjT0cp2N6KqZrs+No1JaBssJbQtSWhMaCPaDRV51exO8SgWi2H9+vU47bTTcNxxx0FRlEoMnRYBTQ6T3Roi4c+YPaEVzvF8mK4H0/WQHDlGkhDTwja4mCIDEsYt8C0AiJXWWIprEuKqzMISLRp/3jSMr935Gly/NJNUFXDEWgurdbY9LSRxTUa9riBjOujNmljdFEXWcjGYt9Aa5/9rqk2yKKCjfqTdzYTl+ujLmciYYbubrkjYnili2AjT3WJsdyOqOt4OEtpGrveJaGYq8m73rne9C5YVVoGDMfGTE4tHkUgE73vf+3DaaafhhBNOgKpyFf3FqpIppaosQh1TWApnKI0WlmzXh+V5sAxv9BhJQkwVEVElxBQZsiQgb7vI2y6QCwtL0TEzluKazL5sWpAcz8dX/vDquMcydoANj23FNWccUqFR0VxpT2go2B5szx9tccuYqNNlaPLCbIOixSGqSljdFEWq6GAwb6PoeNiSNNAYUdAa12A4Hl4fzKOl1O4ms92NqGp0MqGNaNZVbEaSIAgIgmBc8SgIAui6jhNOOAGnnXYa3vve90LX9UoMkaqMX0VZw4okol4SUa+HhSV3pLDkhGsrWa4P2/NgFz2kiuEaSqokllPhoooMRRZQsMP2OSAsqkaV0TWW4qrEi1BaEC696/UpH3+4Mz2/A6F5IU3b4lbEWra4UY0TBAFNURV1moyBfJjilio6yJbS3RoiCoYMG6mig2X1OlpibHej6lRFl9VzjgltRHOjovNvR4pJwOii2e973/sQjYarID/33HPzMo5jjjlmXr4P7b5qfsOTJRF1kog6HQA0uH6AYmm2kuF4MJ3w7rzt+WHqG8Ji1GhhSYIqi+X9+/NhYSkiS0joYVEpocksLFHNcT0f33t4S6WHQfMsrsloiChIF0db3HKWi4GchbYEW9yo9smSOKndrTc7mu6mKxK2pYsYKrDdjapfUMlp/3OMCW1Ec6ei72xjf3Fls1nceuutuPXWW+d1DIIg4NVXX935jlRxtfI2J4sCErqMhB6+vDw/QLFUJDJsD6brw/F8ZDy/nPomi2MKS6oETRZRdD0U8x4GSufVZQkJTUJcC9daYkoMVbv73hys9BCoQtrioy1uA3kLS+p0dGdN1EfY4kYLx0i7W9JwMFQI2906kwYaoypaYmq53a01pqKjju1uVD3G1o6GTb9yA5lDTGgjmltVc4tkIVfDaTbU7s+HJAphu1rpjqQfBDBKiXCFUmHJ9X1kLR9ZyykfE1MkRBQZMS0sLI0s8D1YsAEAmiSWW+ESmsw3Rao697CQtGhJooAlibDFLVV0kNAUxDQJncki9mtjixstHIIgoDmmol6X0Z+3kTUdJA0bWdNFW1xFfUTBYGFsuxtn5VHl+UEAPwBEARgqejs/oMaMTWjLmg4T2ojmQMVb2yqJxSuqBFEYLSy1InwzLzoeio6Pgu3CdHx4foCs5SJruUA+/FAWHSksqWGqnOX5sAwbQ0ZYWFLLhaWwFY53/anS+Ct2cYtrMhojClJFB705E6uVKPI2W9xoYZKlcM2VhoiMvpwF2/XRkzWRLrpYUqcBsoitqdF2tyg/zFKlBQCEWr5VO7WJCW292bCIxIQ2otlVsXcxFnGIQqIgIKbKiKlAS0xFEAQoOqPJcEXHg+cHyFkucpZbPiaqSojKYSucroiwPR/Dho3hcBYvFHF8YUlXWFii+cXf8tSW0JC3PTiej4GchaX1YYtbnc7fSbUtfHVzhZHJYqqMvZskDBsOhgs2DMfFlmG33O5WsD28NhC2uy3jOi1EsyoIAmwZk9C2Pc2ENqK5UpFC0i9/+ctKfFuimiCMFIlUCYiFb4qm68OwR9dZ8oMAectF3nKBQlhYiihSeZ2liCLC8X0kizaSxfC8siggroZtcAldRoQf4miO8YYBiUKY4rYtFabm1OkyYpqMzpSB/VrjFZ+ZTDQXBEFAS7ndzULOdEfb3RIq6vXRdrfl9RE0x/gBl+bZAn177smayIxJaHN9JrQRzZWKFJKOPvroSnxbqmGL+fOoUCoSRRQJzQg/nFvu6IwlozRjqWC7KNijhSVdHk2FiygiXABp0wmT4zKAVGqxS2hhMlxUlfgmS7NqMb9uaVRMldEYVZEybPTmLKxWJBRsDwN5G+1scaMFTJFELK+PIK+76MtZcDwfPZlSu1sibHfrTBlhu1tjhDd4iPYAE9qI5hcbtIlqjCAI0BUJuiKhKRo+ZroeiraHguOhaIeLdxuOC8MJW+EEAHpptlKsdCxEIGM65eQ4SRAQU6XyAt4xFpaIaJa0xVXkLReO56M/Z6GjXkd3poh6trjRIhDXZKxRx7S72aPtbq0xFXnbxWv9ObTGNXTU6fzQS3MuwMKalMSENqL5x0IS0QKgyxJ0WUJjadty/XIqXNEJ1ycJF/T2MIywsKSNmbEUVcPCUnmBb4ys3VQqLKlhYUnkxS3NwEK6SKU9IwoCOup0bE0ZyJgO6jQZcZ0tbrR4jGt3y1nIWWG7W84K093qdAUDeQspw8HyBh1NUba70dxZSDOGmdBGVBl8dVFNWEDvd/NCk8Nkt5F0CsfzUbBHW+Ecz4fpejBdD8mRYyQJMS1sg4spMiBh3ALfAsIWlZEFvOOqzMIS7RDXSKKxoqqEpqiKZKnFbW81bHHrz1lYUqdXenhE80KRRCxviCBvjba7dZfa3doTGiADW5Jhu9uKBra7Ee3IdAltTUxoI5pzLCRRTeDH0T2jSCIaIuMLS4Yz2g5nuz4sz4NleOVjVElCTBURUSXEFBmyJCBvu8jbLpALC0vRMTOW4prM6fg0Dl+3NFFrPGzjsV0ffVkLyxp09GRN1EcUfmCmRSWuydhblZA0HAwVbBRK7W5NMRUtURU5K2x3a4trWMp2N5plPgAvAEZ+69bijZ/pEtrimow2JrQRzTkWkqg21N77W1VTJBH1koh6PSwsuaXCkuH4MGwXluvD9jzYRQ+pYriGkiqJ5Ta4qCJDkQUU7LB9DgjvAEWV0TWW4qoEWWJf+mJWg9elNMdGWtw6kwayloM6M0yR7Ewa2L+NLW60uIildrc6LUx3y1suhgthult7XEOilPqWNBysaNDRyHY3mjUBXD+AKtXu79zuzNQJbR11TGgjmg8sJBERZElEnSQi7C7R4PoBiqU2OMPxYDoebM+H7flh6hvCYtRoYUmCKovl/fvzYWEpIktI6GFRKaHJLCwtMpbr7fDvgyDgxd4iFFEkNMfUcsJOVJFgwENfzsJStrjRIqTKIlY0RJAzXfTnw3a3rkwRcVNGezxsd9ucNFBXanfjAvW0p2r9Rs9wwS5fa/ZmTSa0EVUAC0lUc/j2MPdkUUBCD2cKAGEPerFUJDJsD6brw/F8ZDy/nPomi2MKS6oETRZRdD0U8x4GSufVZQkJTUJcC9daUlhYWtAe2jRc6SFQlWqJha07tuujL2diWUMEvVkT9bqMKBdHpUUqocuIaRKGCzaGDQd5y0XBctEc09AcVZC1XLzan0N7QsPShM51CmlRmpzQ5kIQBCyvjzChjWge8WqNiHZKEoWwXU0Lf2X4QQCjlAhXKBWWXN9H1vKRtZzyMTFFQkQJL4w1WSwv8D1YsAEAmiSWW+ESmswLgAWmKapga6pY6WFQFRIFAcvqdHSmishaLhKmgzpdQWeyiAPa2eJGi5coCGiNa6jTFfTnLBRsF0MFC1nTQXtcQ1yX0ZcbaXeLcEFh2m21OCnJcr1pE9qiKmfqEc0nFpKIaMZEYbSw1IqwsFR0PBQdHwXbhen48PwAWctF1nKBfHhMTC0VltQwVc7yfFiGjSEjLCypkoi4KiOhh61wmsyLglrGxZNpR3RFQnNUwVCpxS2myCgiTN3pqGeLGy1umixiZWMEWdPBQN6G7fnYXmp3WxLXAYSR53WajJWNEb5f0ozVWnub5wfYNGQwoY2oSrCQRER7LCwSyYipYctKEAQoOn65Fa7oePCDADnLRc5yy8dEVQlROWyF0xURtucjWbSRLE1iUUQRcW10AW8WJhaWAGxVXexGWtysMS1ufTkTDRG2uBEBQJ2uIK7JGCrYSJba3TbbBTRHVTTHwna3V/pyWJLQsSShsd2NFiQmtBFVH16lEdGsE0aKRKoExMILANP1Ydij6yz5QYC85SJvuUAhLCzpioRYaZ2liCLC8X2kin45OU4WhXDGkiYjrkn8oElU44SRFLdSi1vcdFCvK9iSLOJAtrgRAQjfH9viGup1BX05E4btYbBgIWM6WJLQENNk9OZMDBs2291ol9XShKSxCW3b02FCm8aENqKK4qcwqgm19GZHkwmCgIgihWlNCAtLljs6Y8lwPHh+AMN2YdhjCkvyaCpcRBHhAkibTjk5Tiq12CW0MBkuqkq8oCCqMboioSWmYjBvob/U4mbCQ0/WxLL6SKWHR1Q1NFnEqsYosqaD/lzY7rYtXURCk9E+pt2tXlewokFnuxstCEMFa1xCm+mGCW3LmdBGVFEsJBHRvBNKs490RUJTNHzMcn0YtouC46Foh4t3G44Lwwlb4QSEHzhH2uEiqgSIQMZ0yslxoiAgro62wsVYWKpqAXvbqKQ5qiBnuTAdD305E8sbIujLWWjQFcQ0XqoQjVWnK4ipYbtbquggZ7ko2AW0RFU0xRRkTAe5fhdLEhra42x3o9qVt1xsK4V2DDKhjaiq8OqMiKqCJovQZBWNpW27NGOpUFpjyfH80oLeHoYR1h+0MTOWoqXCUnmBb4wu8B3OWAoLS7ygJqo+giBgaZ2GzmQROctFpuigPqKgM1XEAW1xvm6JJpBEAe0JDfURGX1ZC0XHw0DBQtp0S+1uEnqyJoYLYbtbPdvdqMZMTGgbYkIbUVVhIYlqAlvbFh9VFqHKYnmtB8fzy0Wlgh0WlkzXg+l6SJaO0SQJMS1sg4spMiBh3ALfAoDYmDWW4qrMD6hEVUKXR1vc+nIWYupoi9vyBra4VVredis9BJqCLkvYqymKjOlgIGfD9jxsSxuo02S0J8J2t7eHC2jQFaxo4CwOGqt6pwV7foC3mdBGVNVYSCKimqBIIhoio4Ul1/NLbXAeCo4H2/VheR4swysfo0oSYqqIiCohpsiQJQF52w0/EOXCy6fomBlLcU1mv/08YoGYJhrb4tabM7GiIYL+vIWGSJhcRZUzUpCn6lSvK4irMgYLNlJG2AKUtwtoiWloispImw6y/S6WJjS0JzS2fVPVGkloM5nQRlTVeFVGRDVJlkTUSyLq9VJhaWSxbidca8lyfdieB7volVPfVEkst8FFFRmKLKBghzOcgPBuV1QZXWMprkqQJd69JZovIyluW5IG8paLdNFBQ0RBZ9LAge0JziCsAiwAVy9JFLAkoaFhbLtb3kSmKKG91O7WnTUxbDhY0aCjTufMjsWqml/HTGgjqg0sJBHRgiCLAup0BXU6AGhw/QDFUiKc4XgwHQ+258P2/HLqmzKusCRBlcXy/iMJIRFZQkIPi0pxTYbCwhLRnNJkcVKKG+CjOxvOUKLKquYPoBQaaXdLFx0M5C1Y5XY3Be0JDSY8vDVUQGNE4aLFi1QQVOcrmQltRLWDhSQiWpBkUUBCl5HQw19znh+gWCoSGbYH0/XheD4ynl9OfZPFMYUlVYImiyi6Hop5DwOl8+qyhIQWFpUSLCztGca20TSaowrylotiqcVtZWMEA/kwxW3kNU3zx/X88tdjvqQq1xBRkNBkDOYtpIoOspaDvO2iNaahMSojVXSQMV101Oloi6uc7bGIVGMZiQltRDNjOR5QwYmlvBojokVBEoWwXa20zoofBDDGLN5tuj5c30fW8pG1nPIxMUVCRJER08LC0sgC34MFGwCgSWK5qJTQZF7sEM2CMMVNR2fSQMEebXHbmmKLWyUE476uxo+gNB1JFLCkTkdDREFfLmx368+bSBclLKkL06+6MkUMFWysbIiwUEsVwYQ2ol33Qk8W37znDWiBi5tOb6vYOPhuQTWHl7A0G0RhtLDUirCwVHQ8FB0fBduF6fjw/ABZy0XWcoF8eExMLRWWVBGaLMLyfFiGjWEjLCypkoi4KiOhh6lwusILIKLdMdLiNjChxa0rU8TKxmilh0dUU3Rlcrvb1pSBel1Bezxsd3tzKI+miIrlDTpn29K8GZvQVrSZ0Ea0I/05C1+67WUUbA9711W2lMNCEtUIlo9oboVFIhkxFWiJqQiCAEXHL7fCFR0PfhAgZ7nl9CJREBBRJMRK7XC6IsL2fCSLNpLh7Gwoooi4NrqAd4SFJaJd1lRKcSs6HnpyJlY1RjBYsNEYUTlzgmg3jCQgDuYtpIsOMqaDnOWiLa6hMaogWbSRMR0sZbsbzYOJCW1dGSa0Ee3Io53JUkhQ5fEqjIhoCoIglNdKQiy82LFcHwV7dJ0lPwhQsF0UbBcohIUlfUxhKaKIcHwfqaJfTo6TRSGcsaTJiGsSIoq0aC/UWR6mnRmb4mbYLlKGg8aogs5SixsXXyWaOVkMW0dH2t1Mx0NfzkS66GBJQkOk1O42XLCxsjFSbgknmm1MaCOamf944O1KD6GM7wxUM0Y+dPJthSpBKBWJdEVCc+kx0w0LSkapuOT5AQzbhVEqLAkAIopcSoUTEVEkuADSplNOjpNKLXYJLUyGi6qLt7BENBVVFtEaV9GfszCQtxArrZfRlS5iVRNb3Ih2V0SRsFdjpNTuZsN0PXSmDDToCtriGorw8MZgHs1RFcvq2e5Gs2u6hLYVDUxoI6oFLCRRTRibUspZDFQtdFmCLksY+SxruX5YSHJ8GLYH1/dhOC4MJ2yFExCuUxFVJURlCRFVAkQgYzrl5DhREBBXR1vhYiwsEaEpqiJnuTBK62esaopgyLDRGFVQp3MNDaLdJQgCGqMqErpSbndLmw5y5XQ3BcOGjXTRQUedjla2uy0MFb6YzpnTJ7SxYElUG1hIqjKu6+Kee+7BfffdhxdffBHJZBKe56GlpQUHH3wwTj75ZJx66qlQlJlfOPf39+Pmm2/Go48+ii1btqBQKKCxsRFLlizBCSecgDPOOAMrVqyYg2e151xWj6gGaLIITVbRWNq23dE1lgzHg+P5pQW9PQwjLCxpcqmwVCowQcToAt8YXeA7nLEUFpYWSmJVwNc1zcDSRKnFzXGRLDhoiinoTBZx0BKZd6+J9tBIu1u9rqAvZ8JyffTlwrajJQkNuiJhe6aIYSNMd4ux3Y12k+V62JwME9oyTGgjqll8F6gizz//PL761a9i69atk/6up6cHPT09uPfee/H9738fV155JY488shdPve1116L733ve7Bte9zjAwMDGBgYwIsvvogf//jH+MIXvoDPf/7zkKQq+0U+5hOnzw+fVCNUWYQqi+XUEccL11gqOh4KdlhYMl0PpushWTpGkyTEtHB9pZgiAxLGLfAtAIiNWWMprsoLprBEtCOqLKItrqIvZ2GwYCGuSYAMbE8XsRdb3IhmRVSVsLopilTRwWDeRtHxsCVpoDGioDWmwYCH1wfzaCm1u8mcPUIzMDGhrY8JbUQ1i4WkKvHAAw/gi1/8IlzX3em+XV1dOO+883DppZfi7LPP3un+3/rWt3D99dfvdD/P87Bhwwa88cYb+MEPflBVU5fH1o4CTmOgGqVIIhoio4Ul1/NRcDwUbQ8Fx4Pt+rA8D5YxmsagShJiqoiIKiGmyJAlAXnbRd52gVxYWIqOmbEU1zg7gxauxqiKnOWhYLvlFrdhw0ZjREE9P4QQzQpBENAUVVGnyRjIhyluqaKDbCndrSGiYMiwkSo6WFavoyXGdjfauSAIsHm4ANP14LhMaCOqdSwkVYE333wTF1988bgi0nvf+158+tOfxiGHHAJFUbBlyxbceuutuP766+F5HjzPwxVXXIG9994bxxxzzLTnvuGGG8YVkWKxGD7/+c/jAx/4ANrb29Hf34+7774bP/nJT5DP5wEA9957LzZs2IAvfvGLc/ekZ2hs7YgXK7RQyJKIeklEfWmNF3dksW4nXGvJcn3Ynge76JVT31RJLLfBRRUZiiygYHulKNDwzl5UGV1jKa5KVXvHmCVh2h1L6zRsHvZgOC6GCw6aYwq2poo4sIp/1olqkSyJ6KgfSXcL2916s6PpbroiYVu6iKEC291o57oyJrKWC98P0JVhQhtRreNv/Cpw2WWXjWs5u/DCC3HRRReN22f//ffHpZdeimOOOQYXXnhhuZh0ySWX4J577oGqTq7kDw0N4eqrry5vNzQ04IYbbsCaNWvKj61YsQLnn38+Tj75ZJxzzjkYHBwEAPz0pz/FGWecgZUrV872090tYz9wajI/KNDCJIsC6nQFdToAaOWp34YT/jEdD7bnw/b8cuqbMq6wJEGVxfL+I2koEVlCQg+LSnFN5kKWVNMUSUR7QkNv1sTQuBY3E6ub2eJGNNtG2t2ShoOhQtju1pk00BhV0RJTYThsd6MdGyqEqZsAE9qIFgr+pq+w5557Ds8991x5+5hjjplURBpr/fr149rZenp68MADD0y577XXXgvDMMrbV1xxxbgi0lh77bUXNm7cCFEMfyQcx8GGDRtm9FzmA9fgo8VEFgUkdBntCQ2rm6JY2xrHioYImmMqIkqY5uZ4PjKmg96siU3DBbw1WEB32kTKcGC5PgCg6HoYyFvYnDTwYm8Wr/TlsDVpIGnYcDx/zsbPLlSaKw0RBTFVhh8E6M2aCIIAyWKYLEVEs08QBDTHVKxpjqJOVxAASBo2Ng8byJRed0OGjVf6c+XFk4kAJrQRLVR89VbYHXfcMW77ggsu2OkxH/nIR8ZtP/7445P2sW0bv/3tb8vba9euxUknnbTD865btw6nnHJKefvuu+8ut7tVi4TCH1lavCRRKK0loGGvpijWtsawoiGCljGFJdf3kbUc9OVMbB4u4M3BPLrTRSQLo4Ul0/UwZNjYUiosvdybRWfSwHDBhu3OXWGJaDYtrdMgCgKKjodkIfwguzVlwJ3D4ijRYidLIpbV61jZGIEqi3B9Hz1ZE1uTRViuD9cPsDVVxOsDORj2ztf9pMqZj5s9TGgjWrj4qbzCXnjhhfLX8Xgc73znO3d6zOrVq8dtb9++fdI+zz77LNLpdHn79NNP36XxfPCDHyx/bdv2tLOdiKjyRCEsLLWOKSytbIygNR5eoImCAM8PkLVc9OfDwtIbA3lsTxcxXHBgOh6CIIDl+Rg2bHSmDLzUl8VLvVlsGTYwVLBgOt7OBzKNnS15EHCVJNoDIy1uADBYsMofYrenzQqPjGjhi6ky9m6KojUeFnQNx8WW4QL6cxY8P0DB9vDaQB7bUgY8xu0uStMltDXHmNBGtBBwjaQKu/nmm9HZ2YnXX38d+XwekrTz6rxp7vwi+a9//eu47V0pUAHA0UcfDUEQyslof/nLX/DhD394l44losoSBQExVUZMBVpiKoIggOn6KNgeDNtD0fHgBwHylou85ZaPiSgSYqV1lnRFhO35SBZtJMOZ6FBEEXFtdAHviMK7iFQdGiIKcqWf556Mib2aIkgWbTQYMhqjTAEimkuCIKAlpqJel9Gft5AzXSQNG1nTRVtCRb2uYLAQprstrw/bsmlxmC6hLVGaVU1EtY+FpAoTRRF777039t57710+5tVXXx23vWrVqkn7vPHGG+WvBUHA/vvvv0vnrqurQ0dHB7q7uwEAL7/88i6Pi4iqi1AqEkUUCYiFF3ZWqbBUdMKkNz8IULBdFGwXKISFJb1UWIooEqKqCMf3kSr65eQ4WRQQV+VSYUkqt9URVcLShIZNtgfT9TBccNASV7EtXURCk7no7yzjLEKaiiKJWF4fQV530Zez4Hg+ejIm0kUXSxIaIIvoTBlhultjhDcjFoHpEtqWhmkiRDRDnh/gz5uHKz2McVhIqkHXXnvtuO0TTzxx0j6bN28uf93W1oZIJLLL51+xYkW5kNTV1QXP83ZpphQRVTehVCTSx1zEm244W8lwwv96fgDDdstrWwgAIopcSoUTEVEkuADSplNOjpNKLXYJLUyGi6pSaWbjTgbEz6Q0C2RJxJI6DT2ZMMUtocuALGJbuoi9m2OVHh7RohHXZKxRJQwbDoYLNgzbxZZhF41RFa0xFXnbxav9ObTFNXTU6UzrWqDGJrT1MKGNaI8Ytoc/vNqHXz/Xg+5MdbXus5BUQyzLwhVXXIGHH364/NgRRxyB448/ftK+g4OD5a/b2tpm9H1aWlrKX7uui6GhIbS3t+/GiImo2umyBF2W0FTatlw/LCQ5Pgzbg+v7MBwXhjNaWNJLbXBRWUJElQAxXEQzUyosiYKAuCrB5qLHNE/qdQU500XOctGdNrG6OYJU0UHSsNHEFjeieTOu3S1nIWeF7W45y0VbXEWdrmAgbyFlOFjeoPP1ucBMTGjLMaGNaLf05yzc/HwPfvtSL/L27q9XOpdYSKpinufBcRxs27YNDz/8MK6//vryTCEAWLZsGTZs2DDpOMuyUCgUytv19fUz+r4T90+n01VVSLJsu9JDIFrQoiIQ1QRAk2F7fml9JR+G48H2A1i2g8zIrxgB0CURUUVCRBERVSVIgoBiEbB3slC3ZVoQPM52pNnRqADpvIuc7aAn6aElpuKtPhsHtMX4AWYPOZ4Py7RgWePff23HhmUy6p2m1qqLiAgiBgo2CnaALUULMVVCW0yFJot43SgirklYXq+z3W2eWKYFw/awLZmHPmbasGmae9yibrk+3hgswPMDZEwXvbnwd8PShArJd2GZTPEj2pnXBwu46YU+/GlzsuqDClhIqmJ/93d/N2nR7BHve9/78M1vfnPc7KERY4tIABCLzWxqfzQa3eH5Kq27u6vSQyBalAQAoh+g6AawvACmG8CZYtKRKoXFpcJOggG2dHZCkznNnWaP6fgYMHz0AOiIS1AlAX3bRCxP8HJnT7h+gM1pZ1K7al9/H7zs4NQHEZX4QQDDDpC2fAQB8BaAOk1EgyZCFICXADTqElp0ka1Pc8jzA/zroyn8pduEFwBNmogvH1GHVXUyNm/evEeFJM8P0JlzYXvh9UFf3oMPoF4TMJyXUF0ruxBVFz8I8Pygjbs7i3g9VTsFV15ZVbHe3t4pH4/H4zjggAPKyWoTOY4zbltRZhaxKcvjfyxct3Z+oIlobsmigIQqIFHadv2woGR64R/HA2wPsD0fjlfdd1Jo4YkpImJKgIITYLDooSMmI+f4SFs+GjTOSiKqBFEQ0KAJiCkCkkUfhhsgY/koOD6adQlRRUDS9JC1fLTHJNSpfK3OhV++lseDXaM3eJKWjyufzuD7723awVE7FwQBugsebC+A6wH9hbCIFFUENOmcaUY0HcsN8HCPiXu2FtFv1N5yEPxNXaWCIJi2kJTP5/GjH/0I73//+7Fx48ZJBSXPG99OMtM7DKI4/sfC92vvB5uI5keY4CaiJSJheVzGyoSEtqiIOlUAg9yoEpp1EZIQFjRTVvj+1W94cKt8init2Oki+kTTUEQB7bHwPUIWAdcPX5v9BQ+OB7hBgO68i61ZFxZvRMy62zcbkx7LOwFeGXam2HvX9RthUdAPgIGiCy8IZya3Rvgxk2gqKdPDLW8VcPFfkvjla4WaLCIBnJFUtVzXxY9//GPst99+aGhoQDabxfPPP48bbrgBjzzyCIBw5tGGDRuQSqXwr//6r+VjJyaszbQQNLEQparVtRDismXLoVXZmIhosro3XgFy07fGrtprL0S5LgbNgSWWi+6sBQjA0oYIIrKImC5jTXN05wfTJI7nw+vLw7QsDL29rfz4kvYlWNWc2MGRRFPzgwBJw8Gw4SAAEAhAXURBU1SBKAiAADTEVSyJaxDZ7jYrun7dM+Xj924t4pIPHDKjhOcRgwUbVtpEI4CurImWhAdJELBXo8616YgmeGuogJte7MP9b6dnfHMroUnoqNPwxuDkgnClsJBUpRRFwXHHHVfebm5uxoknnogTTzwRv/rVr/Dtb3+7/He/+tWv8J73vAcnnHACAEx6I7CsmS2EOXH/aiskaaoKTdcqPQwi2glB3PFFpKZp0FQWkmj2aboGCxKypoNhM8DqJhUWBBi+hOZYdb2n1QLZ86HpDgKMv/BVFb4f0+5bFtHRWu+jL2ehYLvIuoBZ8NAe1xDXZKQdwMi6WF6vo5HpbnMmAKDr+owLSTnTxZBpQdM1DOQt2IEITZOwqjHCxdOJSvwgwGOdKVz/bBee3p6Z8fErGnR84vBl+NCB7fjtS714Y3DLHIxy97CQVIPOOecc9PT04Nprry0/9rOf/axcSEokEpAkqTyzKJ/Pz+j8E/dvaGjYswET0aIkgHeRqXLaExoM24PteRgs2GhPaNieLiKhyVBl3iknqgaqLGJlYwQ500V/3oLt+dieKSJuylgS1wH42Jw0UFewsbIxAk1mgWK27U67quV62DRcQAAgYzoYLoSJjkvrNBaRiACYroc/vjaAXz/bjc5UccbHH76sDmcfsRzHrW4qhxBU23U1C0k16gtf+AKuv/56mKVUpGeeeQaGYSAajUIQBDQ1NWFwMExSSSaTMzr30NDQuO3m5ubZGTQRLSoTZy8QzSdZFLC0LiweJQ0bdZqMiCpha8rAvq3xSg+PiMZI6DJimoShgo2k4SBvudhsF9AcVdEcU5C1XLzSl8OShI4lCba7zaaZvlN7foC3hwx4QYCi7aE3E34WaY6pqNdnFvBDtNAMFWz85sUe3PpiL9LFmQVWSQJw0tpWfPKIZTiwfXLbeLWtPcpCUo2Kx+NYt24d/vrXvwII1zXaunUrDjjgAADA8uXLy4Wk/v7+GZ177P7Nzc271TNNRLQzLDTRXItrMup1BRnTQU/WwuqmCLKWi6GChZYYW7KIqokoCGiLa6jXFfTlTBi2h8GChYzpYElCQ0yT0ZszMWzYWNEQQUOERYtZMYO34iAIsHm4ANP14LgBujImAgAJTUZbnL9TafF6e6iAG57txt1vDMw4tTiuSjjzkKX42LoOLElM/zqqsjoSC0nVwvM8uK4LTdv1X8ItLS3jtkdmJwHA2rVr8dxzzwEAMpkMhoaGJu0/Fd/30dnZWd7ed999d3k8RERE1aY9oaEwqcXNRJ2msMWNqAppsohVjVFkTQf9ORu252NbqS21vdTutmm4gHpdwYoGne1ue2gmN3W6MiaylgvfD7A9XYTr+9BkER31+hyOkKg6BUGAv25N44Znu/DXbekZH99Rp+OswzvwoYPaEVN3XpYRq2xKEq+gKqizsxOf+tSncOKJJ+LQQw/Fhg0bZnR8Op0etz22Be3QQw8d93cvvfTSLp3zrbfegmGMrga/bt26GY2JiIiomkilFjcASBo2DNuDHwToTFVP8gkRTVanK9i7OYqmqApBEJCzXGxOFjCctxEEATKmg1f6cujNmvBnmIBEo3b1X24wb2EgHwby9GRNWJ4HWRSxoiFSdR9wieaS5fr4/ct9+MSvnsUXb3t5xkWkwzrqcOUHD8BvP30kPnH4sl0qIgGouilJnJFUQQ0NDeXWNAB48MEH8ZWvfGWXjnUcBy+88EJ5OxqNYsmSJeXtd7/73eP2f+ihh/C+971vp+d96KGHxm1PPA8R0a7a2aKAu7PAJ9HuiGsyGiIK0kUHvVkTq5uiyFkuBvMWWtmOQVS1JFFAe0JDfURGf86CYXsYKFhIm26p3U1CT9bEcCFsd6tnu9uM7cp7cc50sT0dLhg8kLeQs1wIgoDlDToUifMSaHFIGjZufbEXv3mxF0nDmdGxogCs36cFZx+xDAcvrdut719tr7RqG8+i0tDQgIMOOqi8/fbbb+Oxxx7bpWNvvvlm5HK58vYJJ5wAVR2NRl26dCkOP/zw8vYdd9yBVCq1w3Pato2bbrqpvN3R0YEjjzxyl8ZDRDQR10CiatIW16BIImzPL99V78qYsFyvwiMjop3RZQmrGqPoqNchiyJsz8O2tIHudBGuF8DyfLw9XMCmoQJs16/0cGvKzt6pmdBGi93m4QKuuP8tfOjaJ/HTv26bUREppkr45BHL8LtPH4X/OP2A3S4iAYBQZTP/WEiqsE984hPjtr/1rW+NKxBN5aWXXsJ//ud/lrcFQcBnP/vZSfudd9555a/z+TwuueQS+P70b65XXHEFuru7y9vnnHMORJE/IkREVPskUSgvYpkqOihYpRa35MxjeYmoMupL7W6N0fDmadZysWm4gOGCgyAIkDYdvNKfQ1/WRMBpr3uMCW20WAVBgCe2pfCl217Gx697Fre93Ad7BotoL01o+PIJe+OOvz8aXz5h71lZR6y6ykgsJFXcRz7ykXGzkrZs2YLzzjsPb7311qR9Pc/DzTffjPPOO2/cOkbnnHMODjnkkEn7f+ADHxj3+J/+9CdcfPHFyGaz4/azLAuXX345brzxxvJjK1euxLnnnrtHz42IaEdy1sxiUYn2VFyT0VhqfenNmfD8AHnbxUDOqvDIiGhXjRSFVzdHEVEk+EGAgbyJLcPFcoG4O2vilb4csubM2k9oFBPaaDGyXR9/eKUfn7z+WVz425fxWOeOO3omOqg9ge+ctj9++5mj8MkjliGuzd5KQlU2IYlrJFWaJEn4/ve/j7PPPhv9/f0AgFdffRUf+tCHcOSRR+Kggw6Cpmno6+vD448/joGBgXHHr1+/Hl//+tenPLcoirj66qtx1llnldva7rnnHjz88MNYv349Ojo6MDg4iAcffHDcwt2apuGaa64Z1ypXaU/1W9AkAc169YyJiPbM2dc/hy8cuwpnHrIUklhl7460YLXGNeRtD47nYyBnYWm9ju6siTpdhs42DaKaocsS9mqKIl10MJC3YJXa3eo0Be0JDYCPt4YKaIwoWF4fYUrjDE2V0KYrEhPaaEFKFx389qVe3Px8D4Z3Y/2j965pxiePWI5DlybmrAWt2lrbWEiqAitWrMB1112Hiy++GK+++iqA8C7AU089haeeemrKYwRBwN///d/jy1/+MiRp+gvf1atX4xe/+AU++9nPYnBwEABgGAbuuOOOKfePx+PYsGHDlDOcKunPXRY2Z1387T5RrNu/0qMhol2xs8W2gP3S6QAAojFJREFUc5aLKx/chNtf6ce/rN8HBy1JzNPIaDEbSXHblioibTqo02XENBlbU0WsbY1V3YUaEe1YQ0RBQpMxmLeQKjrIWg7ytovWmIbGqIxU0UHGdLE0oaE9ofE1PsFUzTrTJbQtr9eZ0EYLSmfSwI3PdeOO1wZgzXB9tYgi4v8ctASfOLwDy+sjczTCUdV2z5Wl+SqxatUq3HTTTbj00kuxfPnyafdTFAXvf//7ceutt+Kf//mfIcs7rwXuv//++OMf/4izzz4bsVhs2vOefvrp+MMf/oBjjz12t5/HXPvN2wZSRU5TJqoFu7rY9msDeXzmxudxxf1vIc3XN82DmCqX11jpzVmjLW55u8IjI6LdIYkCltTpWN002u7WX2p3M+zRdrdX+/PImWyr3pGs6TChjRa0IAjwzPY0vvz7V/DRXz6DW1/qm1ERqS2u4ovHrcadf/9OfOW9a+aliFSNOCOpiqiqivPOOw/nnXceNm3ahJdeegnJZBKO46ChoQHLly/HEUccgUhk5j+sdXV1uOyyy/C1r30NTz75JLq6upBOp6HrOlauXIkjjjgCDQ0Ns/+k5sAdrw3is8fGKz0MIppFAYDbXu7Dg28P4R/evRc+fPAS3vWkOdUWV5G33PEtbpki6tniRlSzdGVyu9vWlIF6XUF7XIMJD28O5dEUUVkYmYLpeNg8bIQJbUUmtNHC4ng+7ntzCDc824U3BgszPv6AtjjOPmIZTty3BXIFfndU23UxC0lVas2aNVizZs2sn1dVVRx33HGzft759Ex3FpMz6ohoIciYLr7zwNv4/cv9+Jf1a3BAO9vdaG6IgoCOOh1bUwbSpoOEJiOuy+hMGdivNc72F6Ia1hBREC+1u6WLDjKmg5w12u6WLNpImw466nS0xdVF/XofCbfz/ACbhscktGXDhLYWJrRRjcuaDn77Uh9ufr4Hg4WZzTwWAJywdzM+ecQyHL6srqK/K6rttxQLSVRzGOZKtPC90p/Dp379PD5y6FJ8/thVqONFLM2BqCqhKaoiadjozVnYW5VQsD305ywsqeOCskS1TBYFLK3T0RBR0JezYDoe+vMmMqaEJQkNEVVCV6aI4YKNlY2RWU1XqjVBgHEJbdszxXJCWysT2qhGbU8XceNz3bj9lX6YM1z/SJdFfOigdnxi3TKsbKyO1rVqK3gv3t+YRERU1QIAv3mxFw+8NYSLjluN0w9sq7ppvVT7WuMq8rYL2/XRn7PQUa+jJ2uiPqKwlYNoAYgoEvZqjJTa3WyYrofOlIEGXUFbXEMRHt4YzKM5qmJZ/eJsd+vKmsh7YjmhzfMDJrRRTQqCAM/3ZHHDs93486bhGU9AaI2p+Ni6Dpx5yJKqm4lXbZfALCQREdG8+9D+rThxvzb850Ob0JUxd7hvqujgW/e9idte7sNX37cG+7VxjTSaPSMtbp1JA5lSi1tCl9GZNLB/G1vciBYCQRDQGFWR0JVyu1vadJArp7spGDZspIthu1vrImt3Gy440HSNCW1Us1zPx5/eHsKvnu3Ga/35GR+/tjWGs49YhpPXtlZtMbnaRsVCEhERzT8BePfqJhy5ogG/fHo7/vepLljejqcdv9ibxXm/fg4fPawDnztm1aJuQ6DZFVEkNMdUDBds9OUsRBUJBjz05SwsZYsb0YIxVbtbX85ExnSwJKFBVyRszxQxbNhY0bC42t2Y0Ea1KGe6uO2VPtz0fA/6c9aMjz9udRPOPmIZ3rG8vuqLx9U2vMXz25GIiKqOJos4/12rcOoBbfivhzbjkS3JHe7vB8BNz/fgvjcH8aXjV+PU/duq/o2fakNLTEXOClvc+nIWljXo6M2aqNdlRFVeLhEtJCPtbqmig8G8jaLjYUvSQGNEQWtMgzGm3W15vV6RhKbZ1LODmb+WFyBtOkhaYRMQE9qoFnRnzPL6R4bjzehYTRJx+oFtOOvwZdirKTpHI5x91Xa9yysjIiKquOX1Efy/Dx+EP28axn89tAm9O7mrlDQcfPOekXa3fbBPS2yeRkoL1WiKWxFZy0HClFCnK+hMFnFA++JucQvGLDLBwAtaKARBQFNURZ0mYyBvI2M6SBUdZC0XbXENDZHRdrdl9TpaYrXb7nbna/3T/l1vwUNf3oaqKExoo6r3Yk8WNzzXjQffHoI/wzekpqiCjx3WgY8cuhQNkdr7OV9dZUUvFpKIiGj+TfPm/541zXjnygb84unt+OXTXXC8HV8lPNedxTnXP4uPH74M579z5aJqQ6DZF1EkNEWVcotbTJFRhIferMVFZ0sCVpJogZElER31I+1uJizXR2/WRLo42u62LV3EUMHGyoYIYgvsfaYtIgEBkNCZ0EbVyfUDPPT2EG54thsv9eVmfPw+LVF88vDlOGW/Vqhy7c4uXNPMQhLRHgl4FUu0oOmKhM8dsxdO278dVz+0CX/dmtrh/l4A3PBsN+59YxBfPmE1Tl7bWrN3janyWmMq8pYLy/XRlzOxrCGCvpyJhghb3ABgZgHKRLUjqkpY3RRF0nAwVAjb3TqTBhoiClrjGgzHw+uDebSU0t1qvd1tRIBw7agOrgdHVSZvubj9lX7c+Fz3TmeqT+WYVY04+4hlOHplw4K4Lqy258ArIiIimn+78F64sjGCH5xxEB7cNIxr/rx5p4soDhVsXHrXG/hdqd2t2qYAU20QRlLcUkVkLRcJ02GL2xi8lUMLmSAIaI6pqNdl9OdtZMe0u7XHNdRHFAwZNtLmSLtbbczg2dnrtj2uMqGNqkZf1sSNz/fgtpf7ULBntv6RKgk49YB2nHV4B9Y0c9mDucRCEhERVS1BELB+n5b/n737DnOjutoA/o406mV733UvuBfABkyMAVNNCySAbXAKEHpChySQ5KMktBCCgRB6swFTjEMJ3dhgcAD3ju0t9vbe1Mt8f0irVVuVbZJ239/z8FgzunN1bXZ2RmfuORfHjszA898dxqubKuGMkhT/w+E2LH51M5bOKsJlc0dAq2TRUIqPWiFHllaBRt8qbp4Ut+p2K4rSNIkeHhENMFEuQ1GaGukaEbUdNtidblS3W9FqcSLfqAJEGSpautPdkn22YrTZ/AoZg0iUeLtqO7BicyW+2N+IKJUNQmRoFPjZ9AL8bEYBMrXKgRkgBUju33pEYfBpKNHwo1HIce28UVg0KRcPrj2I7w+3Rmzvckt4eVMlPtpXj5tOGIOTxmUP+5kkFJ9snRKddpdvefDidA1qO2xIVyuGXI0UIgpPpxQxJlPuS3czO5woa3IiQ6tEtk4Jk92FPfWdyNEpUZSmgZwBGaK4uNwS1pc2YeXmKmytbo/7+NGZGiyZXYzTj8iBWuSDw8HEOyEiIhp8vYwIj8rU4onzp+Kz/Y34x7pSNJjsEdvXd9pxxwd7ccyIdNxy4liMzGC6G8VGEAQUGFUob7agw+ZEm9WBNLUC5S0WTMrVQ8YvjETDQle6m1Etoq7Thg6rE81mO9qtTuQaPKucNZjsaLE4UJymQZYu+WZDsLwoJRuz3YX3dtfitS3VqGqzxn383BHpWDK7CMeOzOCDwgRhIImIiFKKIAg4ZUIOjhuVgWf/dwivbamGK0q628ZDrbj4lc249Mhi/HpOCdQKPrWi6NSiHNk6JRo6baht96ziZvWmuBWnD9cUN34jpeFJIZehOE2DTrUTdZ3edLc2b7qbwZPuVt5i9qS7ZWigSaLrDM9aShZ1HTas2lqN1Ttr0WFzxnWsQi7g9Im5WDK7COOyWf8o0RhIIiKiwdcPD490ShG/+8kYnDU5Dw9+cRCbq9oitne6Jbzw/WH8d289bj5hDE4Ym8WnWBRVllaBDpsTVocLNR1WlKRrUNdpQ7pGAf0wSXHrutnnrAYiQK8SoVPK0WR2oMlkh9nene6Wo1Oi0+7E7roO5OpVKDSqkyLdjecuJdre+k6s2FyJT39sjPrwL1iaWsQF0wvw8xmFyE7CGX/D1fC4A6IhhRdDIvI3NkuHp342DR/va8Cj60vRZHZEbF/bYcOt7+/BvFEZuGXB2GE8s4Ri4Z/i1mlzos3iQJpGgfJmMybnGYZFipvD5QYAuHkBJgLg+b2Q3bW6W4cNHTZPuluHzYlcvRJGtQL1nTa0mB0oTlez+C8NS25JwlelzVi5uSrqw75wRmZosGRWEc6clMuZ5EmIgSQiIhoYg/idUxAEnH5ELo4fnYmnN1Zg1dbqqCt+bChvwfevbMKyo0rwi6OLWaSRehSQ4uZdxQ1wo6rdM0NpqOPMPaLwFHIZitM16LQ5Udthg8PlRpU33S3PoAJEoKzZjIbOxKa7SUxuo0Fkcbjwwe46vLalGodaLXEff1RxGpYeWYzjRmVAxutP0mIgiYiIBt8A3dPqVSJuOmEszp6cjwfXHoi6AojdJeHZ/x3Ch3vqccuCMfjJmKyBGRilvCytAp02JyzeFLcRGRrUd9qQMQxS3NRyAd+UN2PToWY4TXZMzFSgX/JTiYYIvUrEGKVndbcmkx0mb7pbpk6JbK0n3W2PN92tIEnS3Yj6W0OnDW9uq8E7O2rQZo2v/pFcJuC0iTlYMqsIE3P1AzRC6k9D+86HiIgSJ4H3yeNzdHj659Px4Z56PPZ1GZqjpLtVt1tx0392Y/6YTNx0wlgUpakHaaSUKjwpbmqUN5thsjvRanEgfRikuDldblz25ja8s6PWt69IJ8elk1jolMifzJvuZlR5VnfrtDnRZPKs7panV8HgXfWt2exASboaGUmT7sbZStQ3++o78dqWKny8rwHOOOsfGVUizp9egJ/PKECuXjVAI6SBwEASpRyr053oIRBRChAEAYsm52H+mCw89W053tpeg2j3N+tLm/G/ilb8ck4JLj2yGCpRNjiDpZSgEmXI1ilR32lDXYdnFTfAjco2C0ZkaBM9vAHx3u66gCASAFSZXPi2xoYjJyRoUERJTCnKUBKU7lbZZoHeKiJP70l3K202w2CyY0S6ZlBqv0S+9g3NIDgNLLck4ZvyFqzYXIkfDsdf/6gkXY2LZxXh7Ml5SbXCIcWOgSRKOXsbTDjnue8wMlOLURkajMrUYlSmBqMytMjUKljLgSgVDOJpalCLuPXEcThnSj4e+OIAdtR2RGxvc7nx728r8OGeOtyyYCyOG5U5SCOlVJDpXcXN4nChusOKkRkaNJjsyNAoYVAPvduqt7bXhN2/qynyLD+i4a4r3a3JZEeT2YFOmxMmmxNZOpVvNcjddR3IM6hQYFAP6KxGKUKhfN41UzysThc+3FOP1zZXobwl/vpHs4qMWDq7GMePzmSKZ4obenc8NCzUdNhQ02HDxoqWgP0GlegLKo3K1GBkhhajM7UoTFND5C8rosGVZLPlJ+bq8exFM/D+7jos/7oMrZbI+fuHW6343bu7cOK4LNw0fwzyjUx3I89Mt0KjGmXNZpjtTrSYHcjQKlDe4klxG2o3xq9tqQq7v9HK2cFE0cgEATl6FdLUCtR22GCyO9FosqHd6kCeXgW9WkRthyfdrTgtMeluSXappiTVaLLjre3VeHt7TdT7p2ByAVg4IQdLZhdhcp5hgEZIg42BJBpSOmxO7KjpwI6awBkHokxASboGozI1GJ2pxciM7llMWiWnUxINFzJBwDlT8nHC2Cz8a0M53tlRG/Umeu2BJnxb3oLL5o7A0tlFUMiZ7jbcKUUZcvRK1HXYUN9pg857HakawiluRNR7SlGGERkadFidqOu0we5y47A33S1frwbgRmmzGUaTZ3U3VT+vIhr5OsdQEvXsQKMJKzdX4aN99XBEWw43iF4px0+nFeDCmYXIN7D+0VDDQBINC063hLJmM8qazViLpoD3cvVKT3qcL03O8zpbp2SaHNFASfB9a5pagTtOHo+zp+TjgbUHsKeuM2J7q9ONJzaU4/3ddbj1xLGYOyJjkEZKySpTq0SHzQmz3YWaDpsvxS1do4BRrUj08IgoCRnUInQqORpNdjR7091K7SZkaZXI0inQbnNiV20H8g1q5BtUg1LEn2EkCiZJEjZWtGLl5kpsPNQa9/GFRjUWzyrE2VPyoFMy3DBU8f8sDXv1nXbUd9rxXdAvSp1SHpQi5wk0FaepIXJGAlF0KRCHnZJvwAsXzcSaXbV44utytNsiT9euaLHgund2YuH4bNwwfwzy+IRtWCswdKe4NZscyNQpUNFiweQ8cciluBFR/5AJAnJ96W5WmO0uNJhsaLM6kG9QQacSUdNhRZPZjpJ0DdI1fQ9MRyiRRORjc7rx0d56rNxShdImc9zHTy8wYunsIpwwNovXwGGAgSSiHpjsLuyq68CuusA0OblMQHGa2psipwmYzaRX8ZQiikkS3V/IZQLOn1aAk8Zl4/Gvy7BmV13UYz7b34gN5c244piRWDyzkMHlYUopypCrV6K2w4YGkw16lScdpbLVgpGZTHEjop6pRBlGZmjRbnWgrsMOu8uNQ60WGFQi8rzpbgebTEhTK1CSru73dLcuDDJRi9mOt7bX4K3tNWg2x7eQgkwAThqXjSWzizCtwDhAI6RkxG+9lJJ++5PRqGg2o7zFgvJmM9qs8RV96wuXW0JFiwUVYVYqyNYpQ1aSG5mpRZ6eaXI0DKXYzWm6RoE7T5mAc6d6Vnfb12CK2N7icOOxr8rw3q463H7iWBxZkj44A6WkkqFVor0rxa3dhpGZGjSaPSluaf0wk4CIhjajWgGdUkSjyY4WiwMdNidMdhOytUpk6hRoszrQXutAgVGNPH3v0t2kCBdkkyPFLtbUb8qazVi5uQof7qmDPc76RzqlHOdOzcdFMwpRmMbFSIYjBpIoJV16ZLHvtSRJaLU4fEGl8mYLyls8f9a0Wwf1u2yjyY5Gkx0/VLYF7NcoZN6gkie4NDrT87okTQOlyJkMRMlkWoERLy2ehbd31OBfG8rRaXdFbF/WbMZVb+/AaRNzcMP8McjWDf6qO5RYhUY1SpvMMDucaDI5kOVNcZuiYoobEUUnlwnIM6iQphFR12GD2e5CvcmGVqsT+QYldCoR1e1WNJk86W7xBqkjzToyORlIGk4kScL3h1uxYnMVvilviX5AkHyDChfPKsS5U/KZiTHM8f8+pTxBEJChVSJDq8SsorSA96wOFw61WlDebEFFi6fYdnmzBYdaLLC5Bm/pYovDjT31ndhTH1jQVy4ARWkaX4BplN+fLNZKQ1qS37fKZQIunFGIheOzsfxrT5HtaD7e14Cvy5rxm2NG4sKZhRAZQBg2FHIZ8gwq1LRb0diV4iYCh1stGMUUNyKKkVqUY2SGFm1WB+o77LC7PPexRpWIPIMn3e1AkwnpagVK0mN/GBktfa2uw4Y0A39XDWV2pxuf/NiAlZursL8x8ozrcKbkGbD0yCKcOC6b9zcEgIEkGuLUCjkm5OgxIUcfsN8tSahpt6G82YyKrplM3llMLZb4coP7wiUBh1otONRqwVdoDngvU6sIKPY9KlOD0Zla5BlUkDFNjlLBEPgxzdQq8edTJ+DcKXl4cO3BqDdfJrsL/1hfivd21+L2E8dhZlBwm4audI0C7VYnTHYnatqtGJmhQZM3xa0/iuUS0fCRplZArxTRYLKj1eJAu82JTrsJ2ToVMrUiWq0OtNc5kW9QId+gilo+Idqzm88PNmNCIVcjHYpaLQ68s6MGq7ZWoynO+kcCgAXjsrB0djGmFxhYpoMCMJBEw5JMEFCUpkZRmhrzRge+12pxhASXKlrMqGqzwj2IsyiazQ40m9uwuSowTc5TnNFvBlOmFqMytCjJUEM9QIUYiXol0vmSYvciM4vS8PKSWXhzWzX+/W0FTFHS3Q40mnHFm9uxaFIurj9+NLKY7jYsFBhVKG1yweJwodnkQJZeiUMtFuiVchZkJ6K4yGUC8g0qpGtE1LbbYHG4UN9pRatF7l3dTe5LdxuRwZnsFKiixYzXtlTj/d11sDnjy8LQKGQ4Z0o+Lp5ZiOJ0zQCNkHpDp5RHvQcdLAwkEQXpeno8ozBw5QGb043DrV0pcpaA2UzWOH9B94XN6caPDSb8GFQIWABQmKbGqAwNRmZ66jB1Ff7m03CivhNlAhbPKsIpE3Lw2Fdl+O/e+qjHfLCnHusONuHq40bh/OkFnA4+xPmnuDWYbNCrRW+KmxWjs5g2QkTxU4tyjMrUotXiQH2nzZvuZoZRpUCeQQXAjf2NJmRoFCjuofamFCW3LcmzzSlGkiRhc2UbVmypwlelzdEPCJKrV+LimUU4b2o+DGqGCZLRjEJjr2pbDQT+hBDFSCXKMC5bh3HZuoD9bklCfYfNr9h392py8U4h7QsJQFWbFVVtVmwI+gWTpha9Bb49waWuYt8FBjULwRLFKVunxN2nT8S5Uz3pbqVN5ojtO+0uPPTlQazZ5Ul3m17I5XGHsnSNAh02JzptTlS3WTEqU4Nmix3pZhEZ2qE1M+3JjYcwPteAAqMaBQYVCoxq5OqVnH1FNADSNQoYVCIaOm1osTjQbnOg0+5Ejk6FDK2IFosDbVYnCgwq5AWlu+1t6IzQM6U6h8uNT39sxMrNlVFXnA1nUq4eS2cX4eTx2fz9neSS6VsbA0lEfSQTBOQb1cg3qnHMyMD88g6r05ceV95iRkWzBWUtZlS1WhDnKpt90mZ1Ymt1O7ZWtwfsV8oFjPClyXXPYBqZoYFawTQ5GkBD4PHnkcXpWLFkFl7fWo1nNh6C2RF5qvGPDSZctmobzpmSh+vmjRpyQQXqVmBQ4aDdBavThSaTA9l6JQ61WmBQiUPqJn1taQvWlgY+uJAJQI5ehUKjCvkGNQqMnhouhUY18r37VFytlKhX5DLPPWe6RoHaDk+6W11XuptRBa1Sjqp2K5rMDoxI1/hmlRwzIgMvfHe4x36jFeOm5NRudWD1jlqs2laN+k57XMcKAOaPycKS2UWYVWRk/SOKGwNJRAPIoBYxrcCIaQWBMxAcLjcq26woazajwm8GU3mzJeqX0f5kd0k40GjGgcbQGRUFBpU3Rc4TaOpaWS5Tq+DFhmIzDH5MRLkMlxxZjFMn5uDR9aX49MfGqMf8Z1cdvjzQhGvmjcJ5U/M5K3AIEuUy5BtVqG7zrOJmUIuAKMOhVgvGZOmid5DC3JJnBai6DhuA9rBtsrQKFBjVAQGmru0Cowo6JW9PiSJRKwLT3WwuFypazEhTK5CnV8EKF35s7ESmRonidDV0Sj4cHEoOt1rw+pYq/GdXXdzlNdSiDGdPycPFM4swIoP1j6j3eKUmSgCFXIbR3jpG/iRJQoPJ7gsq+Rf7jvdJQ1/VdNhQ02HDxorAp81GlegLKvkX+y5MU7P+CwUaQsW2o8nVq/DXMyfhvKkteGjtQZS3WCK2b7c5cf8XB3zpblPyDYM0UhosaWoFOqxOdPiluLVYHGgx24f9bLQmswNNZgd21naEfT9NLXqDSl0zmtQo7Ao2GVUwqkQ+0CCCJ91NrxLR6E13a7M60GHrTndrttjRanWg1eKM2I/FmRzFe6lnkiRhW3U7VmyuwrqDTXFP7M7RKXHhzEL8dFo+0liYPWUl07WPgSSiJCIIAnL1KuTqVZgzIjBNrtPmREWLJ6hU3mzxzGZqMeNQqxWuQVxOrt3mxI6aDuyoCfwCoJALKEnX+KXHeWYzjczQQssnYTRMzBmRgZWXzMbKzVV49n+Hoj4p3FPXiV+9vhU/nZaPq48bxcL4Q0y+QQWzw5Pi1mhyIMeb4qZXiVAMoRS3/tZmdaLN6uyx1odOKfcFmgJmNXn3ceYsDSeiN90tzZvuZvWmu7VZPeeJRilHk9kWsY8fKsPPHqTEc7rc+OJAI1ZsrsLuuvhrXU3I0WHpbM9CIbzuUH9iIIkoRehVIqbkG0JmLji9aXJdK8j5F/3uHMTlIR0uCaVNZm/h4aaA93L1Ss8qct76S6O8s7GyeLNPQ5BCLsMvji7BaRNz8I/1pfjiQFPE9hKAd3bU4vP9jbju+NE4Z0oeZDwvhgRRLkO+QYWqNiuaTDYYVHKoFXIcarFgbPbQTnEbSCa7CwebzDjYQ6F7lVyGPKMKhQYV8r2zmroKgucb1cjRKZlSSkOORiHHqAyNN93NDqvThfIWM9LVCriiFObkNSf5dNqcWL2zFm9srfamCsfn+NGZWDq7CEcWp/FemwYEA0lEKU6UyzzpZZlanDA2y7dfkiQ0mR2oaDajzBtgqmg2o6zF0qsLUl/Ud9pR32nH/w61BuzXKeVBKXKeP4vT1EOqIC0NT/lGNR44azK+LW/Bw18exKHWyOlubVYn7vtsP9bsrMXtJ43DEbn6QRopDSSjWoF2mxMdVieq22wYnaVBq9WBZrMdmUMkxc2oEtFui5w6M5hsLjcOtVhwqIcUU7lMQJ5eFVIIvCuVLk+v4pN7SkmCICBDq4RBrUBDpw2tFgdarQ7UmSLf96lZAD9pVLVZ8cbWKqzZWRd33VSVXIZFk3OxeFYRRgWVz6ChIZliggwkEQ1RgiAgW6dEtk6JI0vSA94z21041OqZueRJkfO8PtRqgWMQl5Mz2V3YVdeBXXWBaXJymYCSNLVvBpP/bCa9ir+2hoRhtELMsaMy8Nols/Hq5ko8/91h2KKku+2s7cAvXtuCC6YX4KpjR8LIWgYpL9+ghtlugs3lQoPJjly9CodaPKu4pXrA4k+nTMDZU/JgsjtR025DTbsVNR021Ha9brehpsOKZrMj0UP1cbklVLdbUd1uDfu+ACBHr/StOtc1m6m7ZpOKK5tSUhNlAgr8VndzD2IJBOqd7dXtWLmlCmsPNCLe/12ZWgUunFGIC6YXMEWeBg2/kRENQ1qlHEfk6kNmPDjdEmrarUEpcp4/B/Nps8steT4/zNPkbJ3SM4MpqNh3rl7JqbuUtJSiDL+eMwKnT8zFI+tKsa40crqbWwLe3FaDz35sxPXHj8aiyblMPUhhokzwpbg1m+wwqkSoFXJUtFgwboikuOmUIsZliz3+faxOF2rbbajt8ASYajtsqG63egJOHTY0dNri/vI0UCR0z6TdXhO+TaZW0UOdJk+wiQ89KBl0pbsZ+EAiKTndEtYdbMSKTVXY0cPiA5GMy9ZiyaxinDYxB0rOKhsWkulOkFc5IvIRZZ6C2SXpGvzEb78kSWi1OFDmXUmuwreinBk17bZBnVzSaLKj0WTHD4fbAvZrFXJf/SX/QFNJuibln/gPScl0JRxEhWlqPHzOZHxd1oyH1h7scUZElxaLA3d/+qNvdbfxOUMj6DAcGdUKdNhcaLc6UNVmw5gsDdqsDjSZ7MjSpW6KW3lz+DpFwdSi3JeGHY7T5UZdpx217VZUt9tQ22ENmOFU12GDM1kiTQCazQ40mx09Fr81qMSgtLnuIFOBUY00NVeeo8EhCAJ0URY94Y/i4Oq0OfGfXXV4Y2sVqtvjLzdx7MgMLJ1dhDkj0vl7ZJgRkugGmoEkIoqqK+c+Q6vE7OK0gPesju40uXJfgMlTm8LmipzC05/MDhf21HdiT33gTb1cAIrSNBiZqcHoDC1G+gWZmDJEiXL86EwcVZKGV36oxIvfH4Y9Skrptup2XLpyM34+oxBXHjuSsx1SVJ5BBbPdBbvLhfpOO/IMKhxu9aS4perT5BZL/6SsiXIZitLUKEpT48gw77vcEprM9u50uYA0Os++wbzmRNNhc6LD5sT+xvArz2kUsu7UuYC0OTUK0zwrz3EWIvUXKXlisMNabbsVr2+txrs7a2GKc0EcpVzAGZPysHhWIcZm8aESJR7vRImoT9QKOSbk6DEhJzBNzuWWUNth86bJeYJLFd4/++uLRyxcEnCo1YJDrRZ8heaA9zK1ipAUuVGZGuQZVLyBpwGnFuW44piROOOIXDz85UFsKG+J2N4lAa9vrcanPzbgd/PH4PSJOXwSmWK6Utwq2yxoNntS3DRKOSpazBifw+LqkchlAnL1KuTqVZhRGPq+JElosThQ453NVN3uDTB1dNdqiveL20CyONwo89YpDEch9/ys5BvUKDR6V5/zCzjl6FUQufIcxYhxpMTaVduBlZur8Pn+BsRbijRDo8DPphfgZzMKhswCDdR7yXTbx0ASEQ0IuUzwPV2eNzoz4L1Wi8MXVPIv9l3dbh3UGhme1IQ2bK4KTJNTizJfmlzXn6MztShJ10CVorMGkg7van2K0zX4x7lTsL60GX//8iBqoqyq2GR24E8f7cPqHTW47cRxQ6bGznBhUItIsynQZnWgut2G0ZkatNucaDTZkK1TJXp4cUuWm1pBEJCpVSJTq8SUfEPYNh1Wp6cuU0f3bKYaX50mK1otybPynMMl4XCrFYdbw6e/ygUgV68KWG2uwNBdpynfoErZWW7U/yROSRp0LreE9aVNWLm5Clur2+M+fnSmBktmF+P0I3KgFlncn5IPA0lENOjSNQqka9IwozAwTc7mdONwV5qcr9i3J9BkjbLSVX+yOt3Y12DCvobAlAQBnho3o/0CTKO8f3KVDOoLQRBwwtgszB2Rjhe+P4xXNlVGXUFxS1U7LlmxGRfPKsIVx4yATslLeqrIM6hg8qa4NZg8KW6VrVYYVYqU+/KfJHGkmBjUIiaq9ZiYG372l9nuCqnNVOv3utFkH+QR98wlwRMI67BhS1X4L6nZOqVvFlN+UApdgVENDVeeGzYYRho8ZrsL7+2uxWtbqlHVFrkOYjhzR6RjyewiHDMyg7PjKanxrpOIkoZKlGFcti5khoVbklDfYUN5i3cGkzfQVNZsHtQlpSUAVW1WVLVZ8XVZ4HvpGhGjMrQYmanFaG8dppGZGhQY1JAz/SAU/0nCUivkuPq4UThzUi4eXnsQGw+1RmzvkoAVm6vwyb4G3DB/DE6ZkM10txQglwkoMHrqIzWb7TCoRGiVcpS3mEPShGnwaJVyjMnSYUwP9UfsTjfqOv1Xm+sOOtW221DfaYs7bWUgdS1O0dNqUGlqMWS1Of9V6Axqfk0YMqL8XPKy0Xd1HTas2lqN1Ttr0RHnSseiTMDpR+RiyawiLqpBESXTucorBBElPZkgIN+oRr5RjWNGZgS81251eFPjLN31mFosqGq1DOoNfavFia2W9pDpyyq5DCMyNH4pchqMzPDMaFLzaTD1YGSGFo/9dCrWHmjC39cdRH1n5JkQDSY7/vjfvXh3ZzpuO3FsjytjUfLQq0SkaxRotThQ027F6EwtOmxONHTakKNPoRS3ZLqrHWBKUeZb2TQcp1tCQ6etxzpNtR22qDMNB1Ob1Yk2a+giFV10Snl3AfAwdZoyNAoGrlNEtJ+6ZFoJKtXsre/Eis2V+PTHRrjirM+QphZxwfQC/HxGIbJTePVOGp4YSCKilGZUKzCtQIFpBcaA/XanG5VtFpS3WFDRbEaZd0W5imYLzI7BK7hqc7mxv9EUsnKPAKDAqMLIrmLfGVrv0tga3pwTAE+620njs3HMyAw8990hrNhcFfUm9fvDrVj86mYsmV2Ey+eOYOpKksvVd6W4ubtT3NqsMKpFqFgTI+WIMsEbZFEDSAt53y1JaDY7vMW/vbOZ/IqB13ZYYXEkz8pzJrsLBxrNONAYviC4SpShwOAfYApcfS5bp+SM3CTBGkn9yy1J+Kq0GSs3V4XU2YzFiAwNlswqwqJJuXyoSCmLgSQiGpKUoixsioIkSWgw2b0pcp7gUtfrhkGsfyEBqG63obrdhm8rAlfrMqpEjMzUYLR3JbmR3kBTYZp66KzSw3vamGmVclx//GicNTkPD649gB8OR75pdbolvPxDJT7e14Cb5o/BieOyGJhMUnLvKm5dKW56pQidSo6KFgtT3IYgmSAgW6dEtk4Z8vAD8Fyf2qzOgELgNe22gDS6eFNmBpLN6fbUM2yxhH1flAnIMwTWZvKf3ZSnV0KUp1ZNsFTFS27/sDhc+GB3HV7bUo1DreF/7iM5qjgNS2YXYd7oTNY/ol5Jpp8aBpKIaFgRhO4lpOeOCEyT67Q5fSvIdRX7rmgx41CrNe7pyn3RbnNiR00HdtQE1rVQyAWUpGt8Bb67in2PzNBCq+QTraFudKYWT54/DZ/+2IhH15dGDXzWddhw+wd7cMzIDNy6YCxGZIRPx6HE0qtEZGgUaLE4UNthxWiFJ8WtvsOGXEPyp7gl001tqhMEwbsYhQKT8sKvPNdpc6K2o4c6TR22Qa0bGI3TLfnqCgKhAXCZ4CkI3lOdpnyjiqtV9RMGkvqmodOGN7fV4J0dNWizxhfMlcsEnDYxB0tmFfVY6J8oVsn0YJCBJCIiL71KxJR8Q8jS0U6XG5VtVlR4C3x3rSRX1myGyT54aXIOl4TSJjNKm8wAmgLeyzOougNMfoGmLG3i0uQixt6S5zqYUgRBwKkTczBvdAae2XgIr2+piloLbGNFCy5+dRMuPbIYvzq6hNPok1COXoVOb4pbfacN+UY1qtqtSNMwxY0C6VUixqnEkEUpulgdLtR22Lxpc4FBppp2Kxo67UkTVHBLQH2n3VMDrjp8m0ytIjDAFLT6HFerjFG0YtuDM4qU82NDJ1ZursLH+xrgjPOBokEl4vxp+bhwZiFyU6nuHSW1ZDpX+duXiCgKUS7zBWZOGJvl2y9JEprMDs8MJr9ZTOUtFtR12AZ1jHUdNtR12PC/oFW+9Eo5RmVq/Yp9e14Xp6kHPKVgV134lYKo73RKETfMH+NLd+tp+e8uDpeE5787jP/uqcfNC8Zi/pjMpHqqNdx1reJ2qMWCFosDBpUInUpEebMFE3J0Sf3/KnlHNjypFXLf9Soch8uz8pz/anP+aXR1nbZBnYEbTbPZgWazo8friVElBsxi6gowdW2nqcWkPn8Gi8RIUszckoRvy1uwYnMVvj/cGvfxJelqXDyrCGdNyuNscRrSGEgiIuolwa/exVEl6QHvme0uVLT4p8h5ZjAdbrUM6qo9nXYXdtZ2YGfQ8s9ymYCSNHVAilxXwEmv6vulId4nd9Q747J1+PfPpuOjfQ14dH1p1LSWmg4bbnlvN+aNysAtC8aiuIfVp2jw6ZTdKW41HTaMVsjRaXeivtNThJuoPyjkMhSnaVCcFv7cd7k9dQS7VpurbQ9cfa623QabK3kKgrfbnGhvcGJfgyns+1qFPGQWk3+dpkTO2h1MrLUdndXpwod76vHa5qoe635FMqvIiKWzi3H86EwWmadhgYEkIqIBoFXKMSnPEFLnwumWUNNuDSj2Xd7sCTa1D2IRVZdb6i6SejAwTS5Hp/QV+B6d2V3sO1evjPmGuzJaEUre1PYbQRBwxhG5+MnoTPz72wqs2lYdOa0QwIbyFnz/yib84qgSLDu6mHVIkkSuwZPi5nC5Ud9hQ0GaGtXtVqSpxaRNSRwG38GHla4C8PkGFWaGeV/yrjwXqU7TYKZ8R2N2uPxSwkMp5QLyvalz3avPeWc1GVTI0auGRFCAl9yeNZnseGt7Dd7aXo1WS5z1jwRg4YQcLJldhMk91DUjGqoYSCIiGkSizFMwuyRdA4zp3i9JElosDk9QqcWbKtdsQUWLGdXtg5sm12Cyo8FkD1kdTKuQ+4JKozK76zGVpGugCEqT403r4NOrRNy8YCzOnpKHB744iO01kdPd7C4Jz/zvED7cW49bFozF8aMzB2mk1BOZIKDQqEZFixmtVgeMam+KW4sZE3P0STlzIvlGRANJEARk6ZTI0ilD6gkCnmtZh83pCywFrD7nfR1vseKBZHdJONRq6XEFLrlMQJ5e6Qs2dc9qUqHQqEaeQRVy/UtGnJEU6kCjCa9tqcJ/99bHPVNcr5Tjp9MKcOHMQuRzxigNUwwkERElAUEQkKlVIlOrxOzitID3rA4XKlo8QaXyZk+KXNe2fRDT5MwOF/bUdWJPXWfAfrkAFHWtJpfhmcEUFb99DpgJOXo8c+F0fLinHo99VYYWS+R0t6o2K25cswsnjMnCTSeMQWGaepBGSuFolXJkaJVoMdtR3W7DmCw5THYX6jo8RbiJkpkgCDCqFTCqFT2uUGW2u1DT4ZnNFG5WU1MSrTznckuobrd5HuhUhb4vwLPyXHCAKb8r6GRQJcVswmg1kobLJVmSJGysaMXKzZXYGFRTMhaFRjUWzyrE2VPyWOidEiKZnifxDCAiSnJqhRwTc/UhN+Uut4TaDpu3wHd3ilx5iznu6dl94ZKAQy0WHGqxYD2aB+1zqWcyQcBZk/Mwf0wmnvq2Am9vr4ma7rautAkbD7Xg13NKcMnsYijF5H/KPlTl6pXotDk9hZE7bCjsSnHTKKBJgi+lRH2hVcoxNkuHsVnhV56zOd2o7QgtBF7rDTbVd9qi/j4bLBK6Z/FurwnfJkOj6KFOk+d1f9QljDrOYV5r2+Z04+N99VixuarHNMdIphcYsXR2EU4YmzUkUh0pdQlJdLYykERElKLkMgFFaWoUpakxLygtqdXiQEWLGWXNgcW+q9usTDsbRoxqBW47cRzOmZKPB744EFJ0PZjN6ca/vqnAB7s96W7HjsoYpJGSP/8UtzarA0aVCL1aRHmzGUfkJmeKG1F/UYkyjMzQYmRG+JXnnC436jvtIbWZ/F8n04IPLRYHWiyOkNm8XfRKechqc75ZTQYV0jV9LwiePP8ag6vF3FX/qCbqYhTBZAJw0rhsLJldhGkFxgEaIVF8jh6Rjk9+bEj0MAAwkERENCSlaxRI16RhRmFgmpzN6cbhVk9wqStFzjOLyQKbM3lW4qH+dUSuHs9dNAPv7arD8q/LotYoOdRqwW/f3YmTxmXhxhPGsgZEAmiVcmRqlWg221HTYcMYhRxmuFDbYUNBMqW4MahFg0yUy1CYpu4xDdctSWgy2QNWmwue1WRNoutdp92F/Y0m7G8Mv/KcWpSFrDZXYPD8WWhUIUunhCzaeTjMIkllzWas3FyF/+6pj3uVQZ1SjnOn5uOiGYVM9aaks3B8Nu77bH+ihwGAgSQiomFFJcowLluHcdmBKQVuSUJdhy2k2Hd5iznup3iUnGSCgHOn5uOEsVl4ckM53t1ZG/W7xRcHmvBNeQsunzsCS2YXpURR2aEkR69Ep90Ju9ON2g4bitLVqGm3Ip0pbkQ9kgkCcvSeFddmFIbOJJEkCa0WB2o6bN11mvyCTTXtVnQm0cpzVqcbZc0WlDWHLwiukAvI04cWAvfMaFIj16CCO0pum2oIpDJLkoTvD7di5eYqbChvifv4fIMKF88qxLlT8gcl3ZCoN9RJdK7yLCEiIsgEwXsTqg5JZ2q3dq8mV+ELNFlQ2WZJmjoVFLt0jQJ/WDge507NxwNrD/SYbtHF6nTj8Q3leG93HW47cSzmjGC622DpSnErbzaj3eaA0SrCkGQpbokfAVF8BEFAhlaJDK2yxyXbO6xOX+pcrV8KXVfwKdoiBoPJ4ZJQ2WZFZZs17PsyAVGv1Vla5QCMbHDYnW588mMDVm6u6nFWVyRT8gxYemQRThyXDZH1jyjZJcF1vwsDSUREFJFRrcD0QgWmBz3ZtTvdqGyz+M1i6g42mR3J8zSXwpuSb8ALF83Euztr8eSGcrTbIqe7VbRYcO07O3HKhGzcMH8McvVMdxsMGoUcWTolmkyemjBahQ5muFDTbkuKtIskuqcl6jcGtQiDWo8JOeFXnrM4XCGzmPzrNDWa7EmTTRbLA59oq7olo1aLA6t31GDVtho0muxxHSsAWDAuC0tnF2N6gSEpgvJEqYaBJCIi6hWlKMOYLB3GBK28I0kS/u+TH/HBnvoEjYxiJZcJuGB6AU4al4XHN5TjP7vqoh7z6Y+N2FDWgiuOGYGLZxZCZLrbgMvReVZx61rNqihdg9oOK9I1IrRcgppo0GkUcozO1GJ0ZviC4HanG/Wd/mlzgXWa6jpscKVe7CYpVLSY8dqWary/uy7u2o4ahQznTMnHxTMLUZyuGaAREg0PvPsgIqJ+JQgCtErWb0klGVol7jplAs6d4kl3+7EhcnqA2eHCP78q86W7HVmcPjgDHaaErhS3FgvabU4YrA4Y1QqUN1swKS85UtyIqJtSlKE4XdNjsMLpltDYafMWAvdPm+sONtkZafKRJAmbq9qwYnMVvi5tjnv+VK5eiYtnFuG8qfkwqPn1l1JXMl3teSYREdHg4/1xUppeaMTLi2fh7e01+Nc35VELzpY2mXHVWztw+hE5+N1PxiBbl7p1NpKdWiFHllaBRpMdtR026BQiLEmQ4pZMN7VEqUKUCcg3qpFvVGNWUej7bklCs9mB2narZ/W5MHWahkMKudPlxqf7G7FycxX21keu5xfOEbl6LJ1dhIXjszl7lqifMZBEREREPnKZgAtnFuLk8dlY/nVZTCmKH+1twFelzbjy2JH4+YxCFiwdINk6JTq8KW41HVYUe1Pc0tQidAlaZYizoYj6n0wQkK1TIlunxNSC0PclSUKb1dlDnSbPdrS6d8ms3erA6h21WLWtGvWd8dc/mj8mC0tmF2FWkZG/o2hISaYfZwaSiIio30VZaZjTGFJAlk6Jv5w2EedOzceDaw/gQKM5YnuT3YVH1pXivV11uP2ksZhRmDZIIx0+BEFAYZoa5c0WdNicaLM6kKZWoLzFgkm5esgYwCMaFgRBQLpGgXSNAkfkhi8IbrI7fQGmm/6zO2J/Ua/Zg6Sy1YLXt1bjP7tqYXHEV/9IJcpw9uQ8LJ5VhBEZrH9ENNAYSCIiIqIezSpKwytLZmPV1mo8vbECpijpbvsbTbh81XYsmpSL3/5kNDJTeFnpZKQW5cjWKdHQaUOdN8XNChdqOqwoSuOXJyLy0ClFjMsWMS5bF71xAkmShG3V7VixuQrrDjbFnfmerVPiwhmFOH96PtLUigEZIxGFYiCJiIiIIhJlApbMLsKpE3Pw6PpSfLyvIeoxH+ypx7qDTbh63ihcMK0Acs6W6TdZWgU6bE5YHZ4AUkm6BrUdNqSrFQlLcSOi1JWICUlOt4Qv9jdgxeYq7K6Lv/7R+GwdLjmyCKdMyIGC9Y9omEimOynebRAR0eBLkmn0FJ9snRL3nnEEfjo1Hw+sPYiy5sjpbp12Fx5aexD/2VmL208ah2kFxkEa6dAmCAIKjCqUN1vQaXOizeJAmiYxKW4MEBJRPDptTry7sxZvbK1GbYct7uOPH52JJbOLcFRxGusfESUQA0lEREQUlyNL0rFy6Sy8trUaz2ysiFrLYl+DCb9+YxvOnZqP6+aNQrqG6Qd95Z/iVtthg07pSXGrbrf2uOT4QDh/Wv6gfRYRpa6qNive2FqFNTvr4l5xTiWXYdHkXCyeVYRRmdoBGiFR8kum4CkDSURENPiS5zpIvSTKZbj0yGKcOsGT7vbZ/saox6zZWYu1+xtx7bxROHdqPmez9FGWVoFOmxMWhyeANCJDg7pOG9I1CugHKcWtwKAelM8hooEzkMW2t1e3Y+WWKqw90Ah3nJ+TqVXgwhmFuGB6AR9AECUZBpKIiKjfMXNt+MgzqPC3RZNw3qEWPLT2ICpaLBHbt9uc+NsXB7BmVy1uO3EcpuQbBmmkQ48nxU2NsmYzTHYnWi0OpGsUKG82Y3Kegau4EVFCON0S1h1sxIpNVdhR2xH38WOztFgyuwinTcyFSmT9I6JkxEBSEnG73fjyyy+xdu1abN26FfX19ejs7IROp0NmZiZmzpyJ+fPn47TTToNcLo+r79bWVrz11lv46quvsH//frS3t8NoNCI/Px9z587FT3/6U0yYMGGA/mZERDTUzR2RgZVLZ2PF5io8990h2JyR091213XiV69vxU+n5eOaeaO42k4vqUQZcnRK1Put4ga4UdXuKcI90JJolj0R9Vr/PP7ptDnxn111eGNrFarb469/dOzIDCydXYQ5I9KTKoWHiEIxkJQkfvjhB/zxj39EeXl5yHttbW1oa2tDWVkZVq9ejZKSEtx999047rjjYup7zZo1uOeee9DREfhEoKmpCU1NTdi1axeef/55LF68GHfccQfUak5TJyKi+ClFGX41pwRnHJGDR9aXYu2BpojtJQDv7KjF5/sbcf3xo3H2lDzI+OUhbpneVdws3lXcRmRoUN/pWcXNoOatHtFwd+WxI/HvbysGrP/adite31qNd3fWwmSPr/6RUi7gjEl5WDyrEGOzdAM0QiLqb5wrmATefvttLFu2LGwQKZzDhw/j8ssvx8qVK6O2ffbZZ3HbbbeFBJHCee211/CLX/wCdrs9pnEQERGFk29U48GzJuOf501BcVr0hxNtVifu/Ww/Ln9jG/bVx78M9HDXleImEwSY7E60mB0AgPIWM9zxFiWJ97MHtHci6g+nTcwZkJp0u2o78McP9+K8F77His1VcQWRMjQKXDF3BN67bA7uXDieQSSiFMNAUoJt3LgRf/rTn+Bydf/inTt3Lp544gls2LABO3fuxDfffIMnnngCxxxzjK+Ny+XCPffcg3Xr1vXY97p16/Dwww/7thUKBa666ip8+OGH2L59O9avX4+//OUvyMnJ8bXZunUr7rrrrn7+WxIR0XB03KhMvH7pkbjy2JFQyaPfcuyo7cCy17bgobUH0GF1DsIIhw6VKEOOXgkAqO+0we50w+5yo7Itcs0qIhr6StI1+OuZR/T4fjzhZpdbwtoDjbhi1Tb88vWt+OTHBrji6GB0pgZ/XDge/7nsaPzm2JHI1Crj+HQiShYMJCWQ0+nEPffcA6ez+2b5pptuwssvv4yFCxciOzsbCoUCWVlZWLhwIV566SX89re/9bV1u934v//7v7AziGw2G/785z9D8i7DoFKp8MILL+DGG2/E2LFjoVKpkJeXh8WLF+Pdd9/FuHHjfMe+++67+OGHHwbwb05ERMOFSpTh8rkj8MayIzF/TGbU9m4JWLWtBj97+Qd8sLvOdx2j6DK1SmgUcrglCTUdnvokDSY72q2OBI+MiBLtpHHZOH5URq+PN9tdeGNrFS546Qfc9v4ebK1uj+v4uSPS8c/zpuD1S4/EeVPzoRbjq/dKRMmFgaQEWrduHQ4cOODbPuuss3DllVdGPObaa6/FokWLfNtVVVX473//G9LuzTffRE1NjW/75ptvxtFHHx22z+zsbDz11FPQarW+fY888kjMfw8iohD88k9BitLU+Ps5U/DIOZNRaIye7tZsduAvn/yI37y5HfsbTIMwwqGh0JviZrY70WzyBJAqWixwDVSKG2taEaUMsYeZoZEu2XUdNiz/ugxnPfcdHv6yFFVt1tg/TybgrMl5WLl0Nh4/fxqOG5XJOnhEQwQDSQn08ccfB2xfe+21MR139dVXB2yHS2974403fK8zMjKwePHiiH2WlJRgyZIlvu1NmzahtLQ0pvEQERHF6idjsvDGstm4fO4IKOXRv1BsrW7HpSs345F1B9FpY7pbNEpRhlxviluDyS/FrZUpbkTDkdPlRqvFgeo2KxyuyKtp+ttb34m7PtqLc1/4Hi//UImOOH7/pqlF/HpOCd67bA7+fOoEjM9h/SOioYaBpATatm2b73VRURHGjBkT03Hjx4+HXq/3bQcX6T58+DB+/PFH3/Ypp5wCpTJ6/vHZZ58dsP3RRx/FNB4iIqJ4qEU5rjx2JF6/9EgcF0OqhUsCXttSjZ+99AM+2lvPdLcoMrRKaJXeFDfvEtyN5oFJcePcAqLk4nJL6LA6Udthw8EmE/Y3mlDTbkVbDOe/W5KwvrQJV721HZeu3IKP9jbENZtxRIYGd5w0Du9fNgdXHzcK2TrWPyIaqrgmbIJIkoS0tDTk5OSgqakJRUVFcR2vUqnQ2elZ2aa9PTBHeePGjQHbc+fOjanPiRMnIi0tDW1tbQCA9evX45prrolrXERERLEqSdfg0XOnYN3BJvx9XSlqvXV9etJkduCuj/Zh9Y5a3HbSWK7yE0GhUY3SJjPMDieaTA5k6RQob7ZgSr44IKs3EVFiuCUJZrsLZocLZrsLFkfoymlqUQ6dSg69MvxXP6vTjbe2VeO1LdU41IvZi0cVp2HJ7CLMG83UNaLhgoGkBBEEAatWrQIAOBwOmEyx139obW1Fc3OzbzszM7B46b59+wK2J02aFPOYJk6ciO+++w4AsHv3brhcLsjlLIZHREQDQxAELBiXjbkjM/DCd4fxyqZKOKM8Ad9c1YalK7Zg8axCXD53BHQ9fDkazhRyGfIMKtS0W9FoskGvkgMicLjVglGZ2ugdxIjfGYkGlyRJsDrdMNldMNmdsDjcIbM0lXI5dEoZdEoRWqXcFzwWe0gnXlfWgnVlLXGNQy4TcNrEHCyZVYSJufroBxDRkMI7rySgUCiQnp4ec/tPP/004IIRnBJXVlbmey0IAkpKSmLuu6SkxBdIstlsqK2tjXu2FBERUbw0CjmumTcKiybn4qG1B/G/Q60R27vcEl7dVIVP9jXghvljsHB8NgRGNQKkaxRotzphsjtR027FyAwNmsx2ZGgUSNMoEj08IoqRzemGye6E2e6Cye6COyhwJMpk0Cnl3v/EkICRXBBgUPXP1z6DSsT50/Jx4cxC5OpV/dInEaUeBpJSjN1ux7PPPhuw76STTgrYrq+v973OyMiIqT5Sl5ycnIBtBpKIqDdYwYZ6a2SGFst/OhVfHGjEI+tKUd9pj9i+vtOOP3y4F++OSMetC8b262yboaDAqEJpkyfdpdnkQJZeiYoWCyYr5T2u4EREieVwdc048qSrOd2BRbJlguALGmmVcqjEwHNZAKBXiTCqRBhUnjZ9DbQXp6mxeHYRzpqUB62S2QpEwx0DSSnmiSeeCCiuXVhYiAULFgS08U97S0tLi6t/o9EYsN1VLynZ2KyR62gQUWK5XKE1Gvy5XW6exxTR8SVGzL5oGl7cVI03ttdGLfj63aFWLH51My6ekY9fzC6ERsEvOl0ylEBNhwOVDgcU0EAlyrC/zolRGZo+92232jgTjKiPnG7JV+PI7HDB7gr8fScIgFYhh0Yhg14pQu0LHLkBpxs2J6BVymFQyWFQidAp5JDJBAAS4HbA6i20vb+hM+6xzSgw4OLp+ThuZLonRc7thM3KFTSJhjsGklLIp59+in//+98B+373u9+FzDgym82+1zpdfIVItdrAJ7nx1G4aTKVlpYkeAhFF0N4e+WbVZO7keUwxOT0PmHZsGl7eY8Ke5sirDjndEl7dUoMP99Rh6RE6HJWrZJDDq8XkgtkpobEGKNCJEASgTi/CoOzbrKTSslL+GxPFyS1JsLkkWJye/+xBz14EAEoR0MhlUIsC1HIBDgFwAOhaYkcpF6ATZdApBGhFAS6ZgFYArRE+98dGc4R3u8kEYE6eCqePUmNMmgJACyoq4quhRERDGwNJKeLbb7/FzTffHFAbaeHChTjvvPNC2trt3WkACkV8NRBEMfBHwunkEwciIkqsIr2IO44y4n+1dqzcZ0KrzR2xfbPVjeVbOzAtS4FLJ+mRr+PspCyNDNZOF2wuoM3uRrpKhhqTCxpRgNiHVdwYRCKKTpIk2FyA1SXB4nTD5gKCyhxBKQfUchk03sCRLCjGq5B5AkY6hSd41JfztidaUcCCYjVOGaFGloa/N4moZwwkpYBvvvkG11xzDWy27jSQkSNH4v777w/b3j+lJN4bPFnQVcvtjnyzTkRENBgEQcAxBSrMyFFg9QEzPjlkRZRsN+xocuAPG1pw5mgNzh6jhaqHFYuGA1EmIFMtQ6PFjVarG1pRBqVcQq3ZhWI9bweJ+pvdO+PI6pRgcUkhgSNRBm/QyBM8Ci5ZJhcEaBUCdN7gkXKAf38tPUKH+UUqaETWTiOi6HjnkOQ++ugj3HLLLXA4uqfz5+Tk4Nlnn4XBYAh7jCiKvvbxBoKC65rEU6h7MI0ZPSZ6IyJKGOPhMgDWHt/XafU8j6nXpowHFjeZ8cjXFdhW0xGxrVMC/lNqwfcNLvxu3kgcPypjkEaZnKrarOiwu6AWZRiZoYYAAQWZGmT0uIpbdcT+eB4Tedhdbs+qag4XzHY3IEnQANAAyIA3MKT0rK6mVcihDIocCYKnQLZBJYfBW0S7//V8Pl+9YMoAfB4R9b/GRA8AAANJSe2ll17C/fffHxAMysnJwUsvvYQRI0b0eJxarfYFkvxnMcUiuH2yBpJUai43SpTM1MrIabUymYznMfXJ5CIVnrkwHf/d24B/flWKZnPk+kk1HXbc8dF+HD86EzcvGIPitL4Xmk5FIxQKlDab4XJL6HTKkK1XosEqISdNBUUvVnHjeUzDldMtwWR3eoJHdhccrq77dRnkChkUggCtUg6tQoROKYM6aAEAAYBO6VlVzagWoeuHldX6gucyEcWDgaQk5HK5cN9992HFihUB+4uKivD8889j1KhREY9PS0tDR4fnCW1nZ3yrMwS3T09Pj+v4wTApJ74C4kQ0+H4+owBvbI08k4GorwRBwJmTcvGT0Zn498YKvLmtOmq629dlzfj+UCt+cXQxlh1VErJs9lAnymXIM6hQ3WZFo8kGvUoOtUKOQy0WjM3m9ZWoJy63BIvDhU67C2a7EzZn4Kx/AYBG4ZlJpFN6VlgLDgxpRDmMak/wSK8SPaugERGlIAaSkkxnZyduuOEGfPXVVwH7x48fj2effRb5+flR+8jJyUFlZSUAoLm5Oa7Pb2pqCtjOysqK6/jBMH/M8E5LIEoFI9KjzPbgvTP1I4NaxC0LxuLsyXl4YO0B7IiS7mZzufH0xkP4YE89bl0wFvNGZw7SSJNDmlqBDpsTHVYnqttsGJ2lQavVgWazHZna5JyJTDTYJEmC2eHyzTiyOt0Bi94AgEouh04lh04hh1YphywoMKSSy2BQiTCoRRhVIsRezPojIkpGDCQlkfr6elx++eXYt29fwP45c+bgiSeegNFojKmfoqIibNmyBQBgMpnQ2dkJvV4f07G1tbW+14IgoLCwMMbRD57RGcMzHYEolXAlJ0qEibl6PHvhDLy/uw7Lvy5DqyXyyqNVbVbcsGYXFozNwk0njEGBUT1II028fIMaZrsJNpcLjSYHcvRKHGqxwKASe5XiRjQUWB2eoJHJ7oLF4YI7KHCklMugVXQFj0TIgwpgizLBl6pmUIlQiVz5jIiGJgaSkkRlZSWWLVuGqqqqgP1nnXUW/va3v8VVq2jChAkB2wcPHsSMGTNiOra0tNT3uri4GFqtNubPJSIiSjSZIOCcKflYMDYLT35TgXe21yBKthu+PNiEbytacNmcEiydXQzlMEh3E2UC8g0qVLVZ0WSyweBNcatosWAcU9xomLA73TDZnd4C2S64gnJjRVlg4EghBgaOZIIncNQVPNIoGDgiouGBgaQkUFNTEzaIdOWVV+LGG2+M+8n+9OnTA7Z37NgRUyCpo6MDZWVlvu1Yg09EROHMKDRiW3V7oodBw5RRrcAdJ43DuVPy8MAXB7GrLkq6m9ONJ7+pwPu763HriWNxzMihn0ZtVCvQYXOh3erwpbi1WR1oMtmRpWOKGw09TpcbJocLJpsLZod/gWwPmbdAtk4hQqeSh9RQ6yqQ3TXjKNEFsomIEoWBpAQzmUy4/PLLA4JIMpkMd955J5YuXdqrPo888khoNBpYLBYAwJdffolLLrkk6nHr1q0LWCHu+OOP79XnExEBwDlT8hhIooSblGfA8xfPwJqdtXhiQznarJHT3Q61WnD96p04eXw2bpg/BvmGob2SUZ5BBbPdBZvLhQaTHbl6FQ63WmBUM8WNUp/LLXlqHDlcMNmdsIcrkK0UoVN4CmSrwxTI1irkvhlHeqUYUgeJiGg4YiApwe655x4cOHDAty2Xy3H//ffjnHPO6XWfSqUSJ598Mt5//30AwIYNG1BaWooxY8ZEPO7VV1/1vdZoNDjllFN6PQYiomkFsdV1IxpoMkHAT6cV4MRx2XhiQznW7KyNmu72+f5GbChrxuVzR2DJ7KIhG1TpSnGrbLOg2WSHQSlCo5SjvNmM8Tmx1VckShZuybOymsnuSVWzOFwhbdRid4FsjSJ8geyuGUcGFsgmIgqLvxkT6LPPPsPq1asD9v3+97/vUxCpy6WXXup77Xa7cdttt8FqtfbY/umnn/YV6AaA888/P+YC3URERKkgXaPAHxeOx/MXzcCk3OjXOKvTjcc3lGPpis34/nDrwA8wQQxqEUa1AhKA6nYb3G4J7TYnGk22RA+NKCLJGzhqNNlR0WLG/gYTDrVY0GSy+4JISrkcGRoFitM0mJCjx+gsLXL1KuhUntlFCpkMmRolRmVoMS3fiKkFRozI0CJDq2QQiYioB/ztmCCSJGH58uUB+0499dSAAFBfzJw5E6eddppve8eOHbj88stRV1cX0M7lcuGJJ57AI4884tuXlpaGa6+9tl/GQURElGymFhjxwsUzcfuJY2FQRZ+cXdZswTVv78AfP9yL+s6hGVzJN6ggymSwe1PcAKCy1QqWf6FkY3O60Wy2o7LVgh8bTChvNqOh0waz3bPKmiiTIU2tQKFRjXFZeozN1iLfqIZBLUIuEyAXBKSrFShJ02BKngHTC40YnaVFlk45LArtExH1B6a2JciGDRuwd+/egH2ffPIJJk6c2Kv+9u3bF7LvL3/5C/bs2YNDhw4BAL7//nssXLgQCxYswMiRI9HW1ob169ejtrbWd4xMJsPf/vY3ZGVl9WocRESxMNlD0w2IBpNcJuBnMwpx8vhsLP+6HO/trot6zCc/NuDrsmb85tgRuGhG4ZCarSCXCSgweuojNZvtMKhEaJVyRM0BJBpgDpcbJnt3uprTHVogW6eUQ6f0/MyGK5CtV4kwelPVtCyQTUTUZwwkJciXX3454J+RmZmJl19+Gb/85S9RXl4OALDb7fjkk0/CtlcoFLj33ntx8sknD/jY+oLXfqLUoInwZPdwa8+ptkSDKUOrxJ9OnYBzp+bjgS8OYH+jKWJ7s8OFR9eX4b1ddbjtxHGYXZw2SCMdeHqViDS1Am1WB2rarRidqU30kGgYcrolmO1OmO0udNrDr6ymUXhqHGl7KJCtU3oLZKsU0ClD6yAREVHfMJCUIIcPHx6UzykoKMB//vMfPPXUU1i5ciVaW1tD2giCgPnz5+OWW27BhAkTBmVcRDT05UVY7UrOm3pKMjMKjXh5ySy8vb0a//qmIuqsuYNNZlz51naccUQufvuT0cjWKQdppAMrz6CC2eGC3eX2pbgRDSS35F1Zze6C2eGCNahAtgBA7Q0a6RQitMrQwJFalAcUyOY1hohoYDGQlCD//ve/B+2zVCoVfve73+G6667Dpk2bUFFRgebmZiiVShQUFGD27NnIzc0dtPH0lU4hT/QQiCgGkVIHJObLUBISZQIumlmEk8fnYPnXZfhwT33UY/67tx7rS5tw1bEj8bMZhRBT/Aus3LuKW1eKG1F/8xTIdsPscMFkd8LicEOSAq8JKrlnZTWtN4AUHBhSymW+oJFRLQ7ZVRWJiJIVA0nDiFwux5w5czBnzpxED6VPJudxNTkiIho42Tol/u+0iTh3Sj4eXHsAB5vMEdub7C78fV0p/rOrDrefNBYzClM73U2vEpGuUaDV4vBMB2Hcl/rI6nR1zzryFsX2p5DLoFV4gkc6hQhRHhg4kguCL2hkUIlQ86EiEVFCMZBEKUfJp05EKU/iF1NKAbOL0/DqkllYta0G//62AmZH5HS3/Y0mXL5qO86anIfrjx+FTG3qprvl6lUsik+95l8g22R3wuUO/KUvlwneGkcidEp5yGppMkGAvqvOkVqERsEC2UREyYSBJCIiIqIeiHIZlswuwikTsvHPr8rw8b6GqMe8v7sO6w424erjRuL8aQUpWa+laxU3olh0FcjuCh71WCBbKUKnlIXMKBIAaJVyGFUKGFQi9CoGjoiIkhkDSURERERR5OhVuPeMI3DeVE+6W1mzJWL7DpsTD6496El3O3EsphYYB2mk/UenFJnZRmH5F8g22Z2wOQMDRwIAjUL0FMhWyqEJs7KaRpTDoBZhVInQs0A2EVFKYSCJiIiIKEZHlaRjxdLZeG1LFZ793yFYHO6I7ffWd+JXb2zDuVPzcd28UUjXKAZppP2DX+0J6C6QbfLOOrI6ey6QrfMWyJaFKZBtVIkweOscsUA2EVHqYiCJiIiIKA4KuQzLjirBqRNz8c/1pfhsf2PUY9bsrMWXBxpxzbxROG9qPmRM26EkZ3W4fKlqFkdogWxlUIFseVCBbFEWWCBbJbJANhHRUMFAEhERhSVJEtyS359++9ySBEkCJPTcJvhpNdFQk29Q4W+LJuHcihY89OVBHGqJnO7WZnXib58fwH921uG2k8Zicp5hkEbaB4LA6vjDhN3pnXHk8KysFr5AtugLHCnEwMCRzLuymn+BbCIiGpoYSCIiShHukMCOJ5jTFdRxS5I3kBPYVgLiOq7rz4HE76U0lBwzMgOvLZ2NFZsr8dx3h0PqxQTbVdeBX762FedPy8fV80YhTZ1a6W40NDhdbpgcLphsLpgd4Qtka5VyX/BIFbSymgBPHa2uGUc6JQtkExENFwwkERH1QvBsHTc803H8Z+u4vQEb/9k6vpk8fsd5trsDOt1t/IM8iY28yAQBguD9E4AAATIZIHi3ZYJ3nwBfXQym7tBwohRl+NWcETj9iFw8sq4UXx5sitheAvD2jlp8fqAR1x8/GmdNzuM5QwPK5fYWyHZ4CmTbwxXIVoq+GkfhCmRrFXLfjCO9Ugypg0RERMMDA0lElPJ6SqvyD8p0zdABws3sCZfGJQXN7Ancl0gCPAEcmeAJ3KArgOML9vi38QZ7BEAGwbdP5okGQYbuNr62COqLX26JYlZgVOOhsydjQ1kzHv7yICrbrBHbt1qcuOfT/Xh3Zy1uP3EcJubqB2mkseHZn7rckgSLt86R2VvnKJha9ASN9Eo5NIrQAtkqucw348igEiGyQDYREYGBJCIaAL6gTdCsm4CZNj3M1ol4HILSsfyCQIkUOFvHG4iRdQdnPLfd3bN1fEEdeNvIPGVIBAQFe7pm9viCPUj409+uIJP/31nm96cQtN0TZrbRUDdvdCaOKknHyz8cxkvfV8LmipzutqOmA8te24KfzyjElceMhEGdHLdoPFdThyRJsDrdAYGj0ALZcuiUMl/KWnCBbIVM5gkaqeUwqhRQigwcERFRqOS4SyGiARM6QyewYDIQfYZO9OMC9yVS8GwdAX5BHgFB+/zaAoDfbJ3uGTzJPVvHP5DjC/LI/II9/m1k3f8mvmCPLOhYv7ZCcP9J8PclSiUqUYYrjhmJMybl4u9fluLrsuaI7d0S8MbWanz6YwN+95PROOOIXJ5zFJHN6YbZ7vStrhZ8DRZlnpXV9Co5tGEKZMsFAXqVCKNKhIEFsomIKEYMJBENsoAiyFFm60ROx0r+2TpdQR3/2TpdQQl4AzEy+AVoZIEpWN0ze7pn5QTM7IFfzZ4km60TmGrW83asM3sC23W3SVUS5znQMFKcpsE/zp2C9aVN+PuXB1HdbovYvtnswJ8//hGrd9Ti9pPGYVy2bpBGGip1f8sMTQ5X94wjk90Fpzu0QLZO6Qka9VQgW9+1sppKhJYFsoeVa44biSe/qQjZf/aknASMhohSGQNJNKz1x/LmYQstI7kLJncFdGItmOwL9oTM1gma2dMV7EmC2Tr+NYICgy9hgj3onq3T+5k9vBEnosjmj8nCnJJ0vPjDYbz8QyUcrsjXhK3V7bhkxWZcNLMIVxwzAnpVct22jcvSJnoIQ57TLcFi9xTHNjlcIQWyZYIAtUIOvbdAtrqHAtlddY5YIHt4mzc6E099W+F7SNll/qiMxAyIiFJWct2REMWg2eyARpJFXt48SqHlZCuYLAQFJQRfsCb+gsm+mT3wDxwlfrZOSDCmqzYQYgz2xDmzh09YiSgZqRVyXHXsKCyalIeH1h7EtxUtEdu7JGDllip88mMDbvjJaJw6MSdpfr9dOac40UMYctySZ2U1s7dItjWoQLYAz89QV40jjUIWcn1Xi3IYVHIY1QoYVCLkDByR14QcPf7vtIm4//MDMDlcEAXgpBI1jh2ZnuihEVGKYSCJUk69yQ6lo/+DQNEKJntiNoEFkwV0z9aJp2ByomvN+AeZogVsQtOsejOzhzexRET+StI1+Od5U/DlwSb8fV0p6joip7s1muy486N9eHdnLW49cSzGZA1OulukS5XI3+195l8g22R3wuJwh8xeVsq7ahx5AkjBgSGFrHtlNaNahIIrq1EEpx+Ri2OL9Phk634o5QIUMiHhM+aJKPUwkEQpR6eUQ60SwxZMDkxnQsBMn1QtmBxSdydMEeR4Z/Yk+u9LxBJJRJ7fxSeOy8YxIzPw/HeH8eqmSjiDc06C/FDZhiUrtmDJrCJcPncEtEoWR041Vmd3jSNzmALZCrmnQLZO5Zl1JAatrCYXBF/QyKASoWaBbIqTXCagUC+izRZ5NUkiSj5XHTsST30bWutssDGQRCmnJE0NlVo1YP0HB24iFz3uj5k9DOoQEQ1nGoUc184bhUWTcvHQlwfx3aHWiO1dbgmvbKrEx/vqceP8MTh5fHZCriWMB8emq0B2V+AouEC2XCZAp5BDqxShU8qhDCqQLRME6JVyX/BIo2CBbOo7TigkSk0aRXLMOmUgiVKOIHhmJQUEY/xSzvoys4ezdYiIKFFGZWrx+E+n4rP9jfjHulI0mOwR29d32vH7D/di7oh03LJgLEZlsvh1MnC6JZjtTl/wyOEKLZCtUcihU4rQKWUhM4oEAFqlHEaVp8aRTilnijj1q646nEREvcVAEqWcSbl6pBsStxQyERHRQBEEAadMyMFxozLw3P8OY+WWKriipLv971ArFr+6GZccWYxfzymBZpBSnaKl4Q0XXQWyu+oc2YJWVhMAaBQitMquOkehK6tpRDkMahFGlQg9C2TTADGqRLTbnJAJAmucEaWoZDlzGUgiIqJBx6+fRJHplCJ++5PRWDQ5Fw+tPYhNlW0R2zvdEl78/jA+2luPm04YgwVjs/plhm2kHjptzj73n4okSYLF4YbJO+vI6gwtkK2Sd9U4kkMTpkC2Ui6DUSXC4K1zxALZNBiydUq0d523yfJtlIji0mhyJHoIABhIIiIiIkpaY7N0+NcF0/DxvgY8ur4UTebIN5C1HTbc9v4eHDcqA7csGIuSdM2Aja19GAWSrA5vjSNH+ALZSm+BbK1KDr1ChDyoQLYo8xTI7qpzpBJZIJsGn8wvuCyylANRSrI6XYkeAgAGkoiIKAGipeoQUTdBEHD6Ebk4fnQmntl4CG9srYIryin0TXkLLnplE5YdVYxfHl0C9QAELobyWWx3emcceQNHwb+zugpk61QitIrwBbK7AkcGlaeQNlGyYAyJKHUlyy00r2pERDToWFuFKH56lYgbTxiDsybn4cG1B7C1uj1ie4dLwnP/O4z/7qnHzQvGYv6YrLg/07MsRfjzVRpCp7HT5fYFjXoqkO2pcSRCr5JDFRQ4EuBJR+yacaRTcmU1Sj56leern+A9s7tUtFqhtrohFwTfCsOCIEAedfXi0FWJiWhgBc+ITRQGkoiIiIhSyPgcHZ7++XR8uKcej31dhuYo6W7V7Tbc/J/d+MmYTNx8wlgUpakHaaTJy+WWYHG40Gl3wRyhQLZOKYdWKYdGEb5AtlEt+mYecWU1SnY9FXG3Ot1w2/snXaYrwOQJNvUccAq33XVcTwEtIgLS1MkRwkmOURARERFRzARBwKLJeZg/JgtPfVuOt7bXRJ3u/lVpM76raMUvji7GsqNKQmbVDGWSJPnqG/VUIFsteoJGeqUcGoU8JDCkkst8M44MKhEiC2RTCpMJQKFeDrcEFBlVUKiUkCTA7fbMeHBD8m5L3m1PSo0kSXC5AQme/ZIUOEPCLUkDMmNCEDyzqHzBJln3drgZVOFnVPUc0CJKFfNGZ+KF7ysTPQwGkoiIaPAlx6RcotRnUIu49cRxOGdKPh744gB21HZEbG9zufH0xkP4cE89blkwFvNGZ0b+gAjfr5L5PJYkCTan2zvjyAWLI3yB7K4ZR7oeCmQbVQoY1HIYVYqQOkhEqU7l/Zk3qESo1Io+9eVydweVJN+2BLcEX1DK5facm93b3jbwBKk8QavuIJXbLfl+z0jefrvOY0c/1hv2D1LJZN5glf82YplR1XNAi6g/TczRY3KeHlaLNaHjYCCJiIiIKMVNzNXj2Ytm4P3ddVj+dRlaLZFXVKtss+KGNbuwYGwWbjphDAqMqZ/uZnO6YbY7YfLOOgoOHIkyz8pqepWn1pFCDPyCJxcE6FUijCoRBrUIjYIrq9HQJxcElKSroVKp4fLOJuoK6ri6Xnv3u9zdr91BM5G60+b6N3DSFYySpMDPlgLG1j1byg2pOyDVFaRyd7eTIPkCWGGDVO4Ig+mFriCVXNZdWypg2xukkskEvwBWV0DKvx5VaACLQarhSS4TcOP8MfiutC6h42AgiYiIiGgIkAkCzpmSjxPGZuFfG8rxzo7aqLOGvjzYhG8rWnDZnBIsnV0c16ybRNf7dLq6ZxyZ7C443aEFsnXeAtm6Hgpk6731jYwqEVoWyKZhyKiUIVunhEajivtYKSjQFGsAyuUODEa53MHtu4M8XTN8vFv99dcOGL8kAS5J8tuWvNvwBam6xhbYxj/AFdgmJEgVbanNXvCfERUccPIFqWR++xAYwPIdE2bWFX8XJjeVKMPCCTkAmhM2BgaSiIiIiIaQNLUCd5w8HudM9aS77a7rjNje5nTjyW8q8P6eetx24ljMHZHhey/yV4nBjSS53JI3aOSEyeGCPVyBbKUIvcKTrqYOUyBbq+gukK1XskA2UV90BXnk/RzgAUKDVIHBptBgVCxturbDBanEAQpSud1dqX3dgTZf/SnfNnwBqO76U6GzsMIFqVwS4BqA38X+wanuAujhC6KHC2D56liFmXVFQwMDSURENPiSubgK0RAxOc+A5y+aiTW7avHkhnK0WSOnux1qseC6d3Zi4fhs3DB/DPIM8c9Q6E9uyRM4Mju8BbLDFEXRKLpqHIUvkK0W5TCo5DCqFdAr5SyQTZQiBitIFXl2lH8wKrY2wUEqz6+cAQhSueFXf0oKU3/KL7VPkoLqT3XtC9oOKpqOQQhShc6oimVFv55nXQ0HtR22RA8BAANJREREREOWXCbg/GkFOGlcNh7fUI41O2ujHvPZ/kZsKG/GFXNHwhFtKbh+JEkSrE63t8aRExZH6MpqSnlXjSNPACl4OXOFTOabcWRQiSyQTUQh/INU/V0KTeohABU9nS909lFwm4AglRzwDL1/gydut99sqZB6U92zrHoKUrmk7kCXZzv8yn6DHaTyn0EVfkW/6AGtZNFp8zwUEgYgyBoPBpKIiGjQcUIS0eBK1yhw58LxOHdKHh5cexB76yOnu1kcbjz2dVnENv1xHludnhpH5h4KZCvkngLZOqUcOqUIMWhlNbkg+IJGRrUINQtkE8Ukeb4WDy3dM5EGZiZVtABU1ICVbzaTt33QTKSuWZ0DGaRyddWN8gtSubrS/dzB290r/7n9glRudNWjGtwgVXewKfIMqsgzqkIDWr0bDxJ6Q81AEhEREdEwMa3AiBcvnol3dtTgX99UoMMWOd0tkrGZ2riPcbjcvlXVzBEKZOuUnuLXwQWyZYIAvVLuCx6xQDYRDReCIIQE0/uLuzcBKL82vhX/3P7b6CFI1f9/B/+xdAepuoNQXUEqt1/6n68Nulf9665RFT5IFfywoz90reznCzbJurfDzaDqIhMEBpKIiIiIaHDIZQJ+PqMQJ4/PxvKvy/H+7t4tIWxUR7+NdLolmO1OmO0udNpdcLhCA0cahafGkU4lDzujSOcNHBlVCuiUoXWQiIiob2QyAbIBmqfWFbDxBZfcgbOhwgagemjTvR0apPI8duj/v4PL3R1UknzbfjOl3N2BqO4aVZJf+h+CalR5gnAhK/t5/z5hygGGlehnKAwkEREREQ1DmVol/nzqBJw3NQ8PfHEQ+xtNMR87J08Zdn9XgWyTt0h2cIFsAYBa4Z1xpJBDqwxdWU0tygPqHA1EiggREQ2OriDVQAQe3G4pMLjkDpwNFbb+VCxt/IJU3deggVnZT5ICPztwxUK/2VKS5HeMBMHl6NfxxIuBJEo5vJ0kSn0SqyQRJY0ZhWl4ecksvLWtGk99WwGTPfrjUJ3C8+xXkiRYHG6YHS502pywOkMLZKvkntlGOoUcmjAFspVyma/GkUElQsGV1YiIKAYDNZNKCgo0xVYwPTgAFL5N8Mp+3q24x1jXzEASERENN4wjESUVUSbg4llFWDghB499VYb/7q2P2L7d7klR+7HBFLZAtk4hh1Ylh04RWiBblAUWyFaJLJBNRETJw39lv/4WHKQKDkCFD1h1b7dYugJIXLWNiIiIiJJAtk6Ju0+fiHOnelZ3K20yh23ndHuCR25JglwmeGscedLVlD0UyDaqFTCo5NAqeftJRETDU1+DVJsqW/t3QL3EKzkREQ26Zktip+MSUWRHFqfjXxdMw2lP/6/HNjl6BbIMupCV1QQAOmX3jCMdV1YjIiIaUhhIIiIiIqIQ6ggpZ6JMQJZG6QsiafwKZOtZIJuIiGhIYyCJiIgG3egMTaKHQERRBKeo+SvSe4JMYzK1MKhEiCyQTURENGwwkERERIPOKUk43GqBKBN8/8llAkSZzLMtFyBjKgxRQokyAbOL0rC5qi3kvdm5SgBAhlY52MMiov7ASyxRSkv08xsGkoiIaNAJENBpc0ZsIxM8ASXRP8DkDTLJfe959hPRwLh23ij89t2dMNldvn3H5CsxyshbSKJUxisnEfUF7wKIiGjQKeUCCoxqOF0SnG43nG4JLjd8r7uWRLU7JdgBAK4e+xIEIWBmkyfYJAua6eT5jwV/ieIzvdCI5y+agXe216DJZEOBaMXReSrOGCRKURpRDhuArlM4XcW0VKKUlODrMANJlFKUvNYRDQkl6WqkaxQ9vu9yS3C6JU+gSXLD6fIGmVwSXFL3fpdbgiRJcLgkOHqONfl0B5W6U+jCpdaxUDBRt5J0DRZNzoPd4UBjbRWDSEQprCRdgw6zBQCQoZJBLfJ8JqL4MZBEKeW0USzQS5RKfja9AG9trwnZv3RmAQqNajhcnhlInkCQGw7vbCS5N5ijEgGg55WjJEnqDjJ1BZ/cEpwuN5yS5BeQckOCJ0DlckuwwR1x3DLBbyaTPCi1LqiuE2c50XAiZ0IMUUozqEVMzdMDTQqmhhNRrzGQRCnDqBQwv1Cd6GEQURyuPm4U3t9dB6uzO3AzK0eJ0RkaFBjDn88utzeo5JLgcLu7g0zeNLiu/U63BEEQoBABRYRgk69flwSn5A0y+WY7dW3DG4Ry+9Lq3F2znBw99ykI/vWagv+TBdRz4iwOGgoEJHw2PRH1kcx7nSIi6i0GkiglzMhW4OgCPfJ00b8sElHyOGZkBv51wTS8ubUKP9a2YmaOEmeM1kScxeOZjSSHuufMNwDwS2nzzGTyBZ+8206/IJQEQC4XIIcAVYQlzQHA7T+zye1Nq5Pc3npOfrOc3G7PjChJgjPyBCcAkYuHd+2T8+aeUgBTP4mIiBIr0Q8oGUiilHDSCDUMSgaRiFLR1AIjRqcp8N3ug8jTyvttNoMgCFCKApRRAkMA4Owh2BScWueSJMhkApQyAZ5FzSOn1fkKhPtmN0kBaXZdQaf+LB7uX0CcaXWUCHL+2BERESVUoi/FDCQREdGAcbslAJ6nJomslS/KZRDlgEYROSDtdoem0/lv+4JObrcn4CMHRLkciDJ7KlLx8K4AlKufi4f7Fw7nDBLqL4xdEhEREQNJREQ0YLoCGIIAQEj+L6EymQCVTO4t8t0zSZK8QaXwtZwc3iBRfxcP706v6+fi4X77iCKRCULCn4ISERENd4m+FjOQREREg06X4qmqgiBAIRegkEefZxWpeHjXtnNAioe7A9Lq+qt4eNcsp0Tn5lNiMNZIREREDCQREdGgKzSqEj2EQdNfxcO7U+v6t3h4V7pdV1pdfxYPF5lWN+QwfkhERJR4ib4cM5BEREQDLjgZRh1Dgezhpr+Kh/vXc+rv4uFdBcT7u3h4V8CJxcOTnygIIeczERERDS5Zgif3M5BEKUdK9ACIiBKsv4qH++o5edPq+rN4uNPthlvq/+LhopxpdYkkypjeRkRElGgqeWIjSQwkERHRoOB3z8HXn8XDu7YlgMXDhzn+qxIRESWWPMEP1RhIopQhBP1JRET9o7+LhztcbrgkFg8fimTeFRiJiIgocRJ928JAEhERDYpEX/Cof/Rn8fCuFetYPDx1yFghiYiIKOFkCb4fYSCJiIgGBWd8DC/9XTzc4Z2N1N/Fw53edL1Yi4cHpNUNs+LhAgBRDkgsVkhERJQQWoUc5lgKTw4wBpKIiGhQDK2v1NSfUql4eEBaXRRDpXi40x0YOUry4RIREQ1ZE3L02FrdluhhMJBEqUEGwXfjGsvTbSJKPvzySX2V6OLhbrffLKdhVDy80+YEgJCktnyDKhHDISIiGraSJcWegSQiIiIaUgaqeLhMJkAmS/7i4eIApdX537saFDLk6KJM9SIiIqIhiYEkSgld964aMTkisEQUG//CyZxLSMko1uLhXUW+U6V4eNcsq/4oHq5Timg02QP2FRtEiDEE6oiIiGjoYSCJUgwDSUSpxH9WBFPbKJX5inynSPHwrrS6/ige7mZ1bSIiIvLDQBKlBu8XUEssj2GJKCkpkiSnm2igDdXi4V14JhMRESWOWpTD6kzsGBhIIiKiQcE4ElGgVCse7htPP/zdiYiIqHeSIbOcgSQiIhoUInPbiHolGYqHO10S6to60d7eH38jIiIiSmUMJFFK4NdPotRVkqbBAauNJzLRIBio4uFdRbqJiIiIGEgiIqIBpVV6Zjv4fwVlagxRYsVaPFySJOys7YDdxRqFREREySAZ1sBIguw6ouiYEUNERDT4BIEzkYiIiCgQA0mUEngLS5S6lN66Lv4BYTmjw0REREREKYmBJEoR/NJJlKqUogxGdXcmtVEpg4wzHIhShtnhAgAIvBYTERElXDI8j2WNpBRQUVGBRYsWweFw4LrrrsP1118fdx91dXVYtWoVNmzYgLKyMphMJmRkZCA/Px/z58/Heeedh5KSkgEYff+QMeRJlNJGZ2jQoBchANArkuDqR0RxY/yXiIgo8ZKhRhIDSUnO5XLhD3/4AxwOR6/7eO655/Doo4/CbrcH7K+vr0d9fT22b9+Of/3rX7jmmmtw9dVXQy6PvgzwYGMciSi1yWQCjEqeyUSpTJTzHCYiIiJ+P096d999N3744Yc+Hf/ggw+GBJGCuVwuLF++HDfccAOkZAhxEhERUVJR8K6RiIiIwEBSUvvb3/6G119/vdfHr1y5EitWrPBt63Q63HLLLfjss8+wY8cOfPbZZ7jlllug1+t9bT755BMsX768T+MeCMmQB0pERDQc5elVAACReeZEREQEBpKSUmdnJ66//nq8+OKLve6jsbERDz30kG87PT0db775Jq644gqUlJRAqVSipKQEV1xxBd5++23k5OT42j799NM4dOhQX/4KRERENEQUp2sg+hVImpihSOBoiIiIhje1mPhSNAwkJZk9e/bg/PPPxyeffNKnfp577jmYzWbf9n333YexY8eGbTtq1Cg8/vjjkHmfNDocjiSclcQpSURERIkyozANs4qMmJSphIzThImIiBImW6dM9BAYSEoWzc3NuPvuu/Gzn/0MFRUVvv2ZmZlx92W32/HOO+/4tidMmICFCxdGPGbmzJk47bTTfNsfffQROjs74/7sgSLnPSsRERERERENcwa1iEKjOqFjYCApCWzYsAGnnnoqVqxYAafT6dt/zDHH4Nlnn427v82bN6O1tdW3vWjRopiOO+uss3yv7XY7Pv/887g/e6BwyWEiIiIiIiIiIF2T2DRzBpKSwP79+9HR0eHb1mg0+P3vf48XXngBaWlpcfe3cePGgO25c+fGdNycOXMg+E1XX79+fdyfPXA84yrUiQkeBxEREREREdHwxW/lSUQQBCxatAg33ngjiouLe93Pvn37Avo84ogjYjrOaDSisLAQVVVVAICdO3f2egz9TSEH7C5Ap+DUJCIiIiIiIqJEYSApCSgUCpx99tn49a9/jcmTJ/e5v9LSUt/r3NxcaDSamI8tKSnxBZIqKyvhcrkglye+KrxCALJ08oBVY4iIiIiIiIhocDGQlASWLl2KpUuX9lt/DQ0Nvte5ublxHZudne177XQ60djYiLy8vH4bW2+NMHKpYSIiIiIiIqJEY42kIcZms8FkMvm2462xFNzev2g3EREREREREQ1vnJE0xPgHkQBAp9PFdbxWq43Y32Bxu909vtfe3g6r1TqIoyGivrLb7QHbPI+JUg/PY6KhgecyUeoLPo8jfX8eCAwkDTEOhyNgW6GILyVMFAN/JJxOZ5/H1BvBJ4a/ysrKQRwJEQ0EnsdEqY/nMdHQwHOZKPXZ7fa4J5H0BVPbhhiXyxWwLQjxFaeWyQJ/JAY7sklEREREREREyYuBpCEmeIW1eANBwYEopVLZ5zERERERERER0dDA1LYhRqPRBGzbbLa4jg9un6hAksFgwJgxYwLGETxbioiIiIiIiGi4cbvdAeVgDAbDoH4+A0lDjMFggFwu980s6uzsjOv44Pbp6en9NbS4iKKIjIyMhHw2ERERERERUTIbzJpIwTjFY4gRBAGZmZm+7ebm5riOb2xsDNjOysrql3ERERERERERUepjIGkIKi4u9r2uq6uL61j/9llZWSGpckREREREREQ0fDGQNARNmDDB97qtrS1kllFP3G43ysvLfdvjx4/v76ERERERERERUQpjIGkImj59esD2jh07Yjpu//79MJvNvu2ZM2f257CIiIiIiIiIKMUxkDQEzZs3L2D7yy+/jOm44HbB/RARERERERHR8MZA0hBUUFCAWbNm+bbff/99tLS0RDzGbrfjjTfe8G0XFhbiqKOOGrAxEhEREREREVHqYSBpiFq2bJnvdWdnJ/7whz/A7Xb32P6+++5DVVWVb/uSSy6BTMYfDyIiIiIiIiLqxkjBEHX66adj2rRpvu0vvvgCN9xwA9rb2wPa2Ww2/OUvf8Hrr7/u2zdixAhceumlgzZWIiIiIiIiIkoNYqIHQANDJpPhoYcewuLFi31pbR9//DG++uornHTSSSgsLERDQwPWrl2L1tZW33EqlQqPPPIIlEplgkZORERERERERMmKgaQhbPTo0XjxxRdx+eWXo6GhAQBgNpvx/vvvh22v1+uxfPnygJlMRERERERERERdmNo2xB1xxBH48MMPsXTpUuh0urBtFAoFFi1ahPfeew/HHXfcII+QiIiIiIiIiFKFIEmSlOhB0OCw2+347rvvUFlZidbWVqjVaowYMQKzZ89Genp6oodHREREREREREmOgSQiIiIiIiIiIooJU9uIiIiIiIiIiCgmDCQREREREREREVFMGEgiIiIiIiIiIqKYMJBEREREREREREQxYSCJiIiIiIiIiIhiwkASERERERERERHFhIEkIiIiIiIiIiKKCQNJREREREREREQUEwaSiIiIiIiIiIgoJgwkERERERERERFRTBhIIiIiIiIiIiKimDCQREREREREREREMWEgiYiIiIiIiIiIYiImegA0+CoqKrBo0SI4HA5cd911uP766+Puo66uDqtWrcKGDRtQVlYGk8mEjIwM5OfnY/78+TjvvPNQUlLSq/G1trbirbfewldffYX9+/ejvb0dRqMR+fn5mDt3Ln76059iwoQJverbbDZjzZo1+OKLL7Bnzx60trZCp9MhLy8PM2fOxPnnn4+ZM2f2qm+73Y6PPvoIn3zyCXbu3Inm5mYolUrk5+dj8uTJOO+883DMMcdAJmP8lmLndrvx5ZdfYu3atdi6dSvq6+vR2dkJnU6HzMxMzJw5E/Pnz8dpp50GuVweV98810J1/Xu///772LZtG5qamiAIAvLy8jBhwgScc845OOGEE6BQKHo1dhq+nE4nPv74Y3z66afYvn07mpub4XK5kJ2djalTp+KUU07BGWec0aufLV6TQ/GaTINtz549+PnPfw6HwwEA+Pzzz1FcXBzz8TyPQ/E8JkpegiRJUqIHQYPH5XJh2bJl+OGHHwCgV4Gk5557Do8++ijsdnuPbeRyOa655hpcffXVcX25XbNmDe655x50dHREbLd48WLccccdUKvVMff91Vdf4Q9/+APq6+sjtjv11FNx7733Ii0tLea+d+zYgVtvvRVlZWUR2x199NF46KGHUFBQEHPfNHz98MMP+OMf/4jy8vKobUtKSnD33XfjuOOOi6lvnmuhKioqcOutt2Lbtm0R202YMAF///vfe33TTcPP1q1bcdttt6GioiJiu+LiYjzwwAM46qijYu6b1+RQvCbTYLPb7bjgggvw448/+vbFE0jieRyK5zFRcmMgaZj585//jNdff923HW8g6e6778aKFStibn/qqafisccegyAIUds+++yzeOihh2Lue+bMmXjllVegVCqjtv3ggw9w6623wuVyxdT3yJEj8cYbbyAjIyNq2//973+46qqrYDabY+o7IyMDq1atwogRI2JqT8PT22+/jbvuuivmn1nAc5N55513YsmSJRHb8VwLtX//fvziF79AU1NTTH2r1Wq89NJLvX7KSsPH559/jt/+9rdwOp0xtZfL5fjjH/+IpUuXRm3La3IoXpMpER566CE8++yzAftiDSTxPA7F85go+XEe4DDyt7/9LSCIFK+VK1cGXOh0Oh1uueUWfPbZZ9ixYwc+++wz3HLLLdDr9b42n3zyCZYvXx6173Xr1uHhhx/2bSsUClx11VX48MMPsX37dqxfvx5/+ctfkJOT42uzdetW3HXXXVH73rNnD+644w7fhU4QBCxevBirV6/Gtm3bsGHDBjz00EMYNWqU75iKigr87ne/g9vtjth3bW0tfvvb3wZc6M4880y89tpr2Lx5MzZu3IgnnngC06ZN873f0tKCq6++GiaTKerYaXjauHEj/vSnPwXcnM2dOxdPPPEENmzYgJ07d+Kbb77BE088gWOOOcbXxuVy4Z577sG6det67JvnWqjOzk5ce+21AUGk4447Di+88AK+//57/PDDD3jxxRcxb9483/tWqxXXXXcd6urqov670PD1448/4oYbbggIIi1YsAAvvvgiNm3ahO3bt2PNmjVYtmyZb4aBy+XCfffdh2+//TZi37wmh+I1mRJh06ZNeP7553t1LM/jUDyPiVKERENeR0eHdN1110kTJkwI+e+xxx6LqY+GhgZp5syZvuPmzJkjHThwIGzbsrIyad68eb62U6ZMkSoqKnrs22q1SieccIKv/bRp06Tvvvuux3GceeaZAX+H77//vse+3W639LOf/czXduLEidIHH3wQtm1HR4d0ySWXBPS9evXqnv9RJEm6/vrrA9o/88wzYdvZ7XbppptuCmi7fPnyiH3T8ORwOEJ+xp966qmIxzz++OMB7U888UTJZrOFtOO5Ft79998f0P7//u//JLfbHfbvGNz2jjvuiNg3DW8XXXRRzNfczz//XJo0aZKv7YIFC8Kex5LEa3JPeE2mwWYymaSFCxeGvcc+fPhwxGN5HofH85goNXBG0hC3Z88enH/++fjkk0/61M9zzz0X8GTgvvvuw9ixY8O2HTVqFB5//HFf4TuHwxHxycmbb76Jmpoa3/bNN9+Mo48+Omzb7OxsPPXUU9Bqtb59jzzySI99f/nll9i+fbtve9myZTjzzDPDttXr9Xj88ceRn5/v27d8+fIe0xH27duHjz/+2Le9cOFCXH755WHbKhQK/O1vf8PkyZN9+1544QW0trb2OHYantatW4cDBw74ts866yxceeWVEY+59tprsWjRIt92VVUV/vvf/4a047kWqrm5Ga+++qpve8qUKbjzzjvDpgwIgoDbbrsNCxYs8O1bs2YNSktLw/ZNw9uWLVuwZcsW3/axxx4bMZX8pJNOCkhnq66uxueffx62La/JoXhNpkR48MEHcejQoV4dy/M4FM9jotTBQNIQ1dzcjLvvvhs/+9nPAop7ZmZmxt2X3W7HO++849ueMGECFi5cGPGYmTNn4rTTTvNtf/TRR+js7Azb9o033vC9zsjIwOLFiyP2XVJSElADZtOmTT1+kfPvW6FQ4IorrojYd1paWsCX9srKyh7TC958882A7WuuuSZi30qlEjfccINvu7OzEx999FHEY2j48b+BAjxBolhcffXVAdvh0tt4roVas2ZNQHHTq666KuLqL4Ig4NZbb/Vtu1wurF69OuJ4aHh6//33A7Z/85vfRD3mggsuCNgOd07wmsxrMiWHr7/+Gq+99hoAQCaT4fjjj4/5WJ7HPI+JUh0DSUPQhg0bcOqpp2LFihUBEf9jjjkmpBBgLDZv3hwQ3fef+RDJWWed5Xttt9vDPlk9fPhwwAoXp5xySkwF/s4+++yA7XAXDavVig0bNvi258yZE5AH3pMzzjgjYDWMni5I/n+fUaNGYcqUKVH7Pv7445Genh61bxq+/FcMKyoqwpgxY2I6bvz48QE1FIJXeuO5Fr1vrVYbMNuoJ+PGjcOkSZN828HBPyIg8FzW6/WYO3du1GNGjx4dsH348OGQNrwm85pMidfe3o4//vGPvu1f/epXcS2+wPOY5zFRqmMgaQjav39/wBKfGo0Gv//97/HCCy/Etexml40bNwZsx3IzDHguLv7pIevXr++3vidOnBjwdwnX95YtWwJmGsTad0ZGRsCy3uFmdhw6dAjV1dVx9y2XywOWdf7+++9hsVhiOpaGPkmSkJaWhpycHMhkMhQVFcV1vEql8r1ub28PeI/nWui5ZrPZsHXrVt/27NmzY7rZBjy/37pUVFREXZ6Yhp9Vq1bhv//9L/7xj3/g9ttvj2m5bqvVGrUNr8m8JlPi3XPPPaitrQUAjB07NmBWTCx4HvM8Jkp1DCQNYYIg4KyzzsL777+PX/7ylxHTNSLZt29fQJ9HHHFETMcZjUYUFhb6tnfu3BmxbwABT/kjEQQBEydO9G3v3r07ZPnR4L5jHXfwOBoaGkJWZuqvvp1OJ/bs2RPzsTS0CYKAVatW4euvv8b27dtjWpWlS2trK5qbm33bwWmsPNdCz7WDBw/C4XD0uW8g/O83Gt5kMhnGjBmDM888ExdeeGFMx+zevTtge+TIkSFteE3mNZkS65NPPsF//vMfAJ4gxv333x/zQ4guPI95HhOlOgaShiCFQoGzzz4b77zzDv7+97+juLi4T/3550jn5uZCo9HEfGxJSYnvdWVlZcgFyf8pviAIAe3j6dtms/meDIUbN4CAJUijCf43868zNdB9EwGe89h/qnY0n376KSRJ8m0Hp8TxXOvfvoP//XgeU3947rnnArZPPvnkkDa8JvOaTInT1NSEP//5z77tK664AtOnT4+7H57HPI+JUh0DSUPQ0qVL8fDDDwesYtAXDQ0Nvte5ublxHZudne177XQ60djYGPB+fX2973VGRkZcT3SCc7KDL3b+fQPxjT1a3/7/Jv3dN1G87HZ7SP2zk046KWCb51r/9u3/uy1c30TxsNls+NOf/oSvvvrKt2/27Nn4yU9+EtKW12Rekylx7rrrLt/s3wkTJsS8IEYwnsc8j4lSnZjoAVBys9lsMJlMvu14aywFt29tbUVeXp5v2z8VJ96+jUZjwHZbW1vAtn/fSqUyrqc90fpuamoK2I5n7NH6JorXE088EVBcu7CwMKRoNM81nseUPFwuFxwOBw4dOoSvvvoKK1asQFVVle/9oqKisKmtvCaH75vnMg2G1atX+4pBKxQKPPDAA3GntAE8j3vqm+cxUWphIIki8r/QAYBOp4vreK1WG7E/s9mc8n3H23+0voni8emnn+Lf//53wL7f/e53ITe3PNf6t+/gtjyPKR6//vWvQwridjnxxBPx5z//OWTWG8Brcix9x9s/r8kUi5qaGtx3332+7SuvvLLXM/95HkfvO97+eR4TDT6mtlFE/oVoAc8TmHiIYmCs0ul0Bmz7r/wwkH0Ht40meCyR+o63/2h9E8Xq22+/xc033xxQG2nhwoU477zzQtryXOvfvoPbBv+uJIqkpqYm7H69Xo9JkyYFnNP+eE2O3ne8/fOaTNFIkoTf//73vhWRJ0+ejKuuuqrX/fE8jt53vP3zPCYafAwkUUTBBfz8lxyNRfBKcW63u8f+B7LveFesCx5LpL7j7T9a30Sx+Oabb3D11VfDZrP59o0cORL3339/2PY81/q37+C2PX3xJwomSVKPgaTOzk48+eSTOPXUU/H444+H/Fzxmhy973j75zWZolmxYgW+/fZbAJ6Axf333x93gMYfz+PofcfbP89josHH1DaKSC6XB2zH+4s5+KIQnG4jiqLvycxA9N1lIPuOt/9ofRNF89FHH+GWW24JeKqZk5ODZ599FgaDIewxPNei9x3cPpLgp508jylWTqcT//rXvzBx4kSkp6ejvb0dW7duxcqVK/H1118D8MxYWL58OVpaWnDXXXf5juU1OXrf8fbPazJFUl5ejocffti3fe2112LixIl96pPncfS+4+2f5zHR4OOMJIoouIie/+yHWAS3D/7FrlarU77vePuP1jdRJC+99BJuvPHGkCDSSy+9hBEjRvR4HM+16H0HT6vvS99EPVEoFDj++OORk5MDhUKBrKwsnHzyyXjuuecCgkYA8Oqrr2L9+vW+bV6To/cdb/88l6knLpcLd9xxBywWCwBg2rRp+M1vftPnfnkeR+873v55HhMNPgaSKCKDwRDw5KSzszOu44Pbp6enB2z7r8gwkH2bTKa4Uk/i6Ttc+770TRSOy+XC3Xffjb/+9a8BT+mKiorw6quvYuzYsRGP57nWv30HF/LkeUz94ZJLLsFll10WsO+ZZ57xveY1OXrf4dr3pW8avp599lls2bIFgCcw8cADD4TMJuoNnsfR+w7Xvi99E1H/YyCJIhIEAZmZmb5t/2U/Y9HY2BiwnZWVFbCdk5PT676DlwmN1LfL5YprKdB4+gbiG3tw3/7/vkThdHZ24sorr8SKFSsC9o8fPx4rV67EqFGjovbBcy30XOtL38G/23geU3+55pprAp7Mb9q0ybeaEa/J0fsGeE2mvtu7dy+WL1/u2/7tb38b9YFNrHgeR+8b4HlMlOwYSKKoiouLfa/r6uriOta/fVZWVsh03qKiIt9rk8kU19OH2tpa32tBEFBYWBjwvv+4g9vH03e4vgaybyJ/9fX1WLJkCb766quA/XPmzMHKlSuRn58fUz8813geU2rQ6/WYOXOmb9vlcqGiosK3zWsyz2UaeJ9++mlACvnDDz+MiRMnRvzv8ccfD+jj5JNPDni/srLS9x7PY57HRKmOgSSKasKECb7XbW1tIU9CeuJ2u1FeXu7bHj9+fMS+AeDgwYMxj6u0tNT3uri4GFqtNuD94M/zbx9P3wqFAiNHjozYd2/HDYT+GxB1qaysxMUXX4x9+/YF7D/rrLPw3HPPwWg0xtwXz7XQfwOexzQYXC5X3LVEsrOzA7atVqvvNa/JvCZT6uN5zPOYKNUxkERRTZ8+PWB7x44dMR23f/9+33R8AAFPWPvad0dHB8rKynzbM2bMCGkTvC/WviVJCmg7derUkJUkJk2aFLD0686dO2PqGwC2b9/ue52bm4uCgoKYj6Xho6amBsuWLUNVVVXA/iuvvBIPP/xw3IUkea6FnmvB+2Idd3DfCoUCkydPjvlYGvrKy8vxi1/8AieffDKmT58ekCITi9bW1oBt/xQQXpN5TabUx/OY5zFRqmMgiaKaN29ewPaXX34Z03HB7YL7AYAjjzwyYEpurH2vW7cuoODw8ccfH9KmpKQkYBWrWPveunUrWlpaIvatVqtx1FFH+ba/+eabmFZ8qqurw+7duyP2TWQymXD55ZcHBJFkMhn+9Kc/4aabboIgCHH3yXMt/Lnm/3tp9+7dqK+vj9q3zWbDt99+69ueNWtWyFNbGt7S09OxceNGVFZWwul0Yu3atTEf63A4sG3bNt+2VqsNSGHlNZnXZBp4119/Pfbt2xfXf9ddd11AH59//nnA+/7pVjyPeR4TpToGkiiqgoICzJo1y7f9/vvvB1wMwrHb7XjjjTd824WFhQEXhy5KpRInn3yyb3vDhg0xTZN99dVXfa81Gg1OOeWUsO3OPPNM3+vS0lJ88803Uft+5ZVXfK8FQcDZZ58dte+Ojg6sWbMmpnH7r3BxzjnnRD2Ghp977rkHBw4c8G3L5XI88MADWLp0aa/75LkW/lzz71uSpJCC5uGsXr0aHR0dUfum4Ss9PR1TpkzxbR84cCCmcwIAVq1aFfDzNX/+/IAZiLwm85pMqY/nMc9jolTHQBLFZNmyZb7XnZ2d+MMf/hDw1CLYfffdFzCb4pJLLoFMFv7H7dJLL/W9drvduO222wLqQQR7+umnfcuxAsD5558PvV4ftu1FF10UME32rrvuirgKxAcffIAPPvjAt33CCSeE5HB3WbRoETIyMnzbDz/8cMCU4GA//PADXnjhBd/2hAkTcMwxx/TYnoanzz77DKtXrw7Y9/vf/75fbox4roU67rjjMG7cON/2888/j02bNvXYd2lpKf7+97/7trOzs7Fo0aIe29PwdfHFFwds33333QEBonB27NiBhx9+2LctCAIuv/zykHa8JofiNZlSDc/jUDyPiVIHA0kUk9NPPx3Tpk3zbX/xxRe44YYb0N7eHtDOZrPhL3/5C15//XXfvhEjRgRc0ILNnDkTp512mm97x44duPzyy0NWsXC5XHjiiSfwyCOP+PalpaXh2muv7bHvwsLCgM+urKzEL37xi7AF/FatWoU77rjDt61QKHDLLbf02LdOpwuYxtza2opf/epX2Lx5c0jbzz77DFdffXXACiB33HFHr1KUaOiSJCmklsqpp54a8fyJB8+1UIIg4Oabb/Zt2+12XH311fj8889D2m7atAm/+tWvAn7v3XDDDUxro7AuuOCCgFlJZWVlWLZsGfbv3x/S1uVyYdWqVVi2bFlA/ZNLLrkk4NrbhdfkULwmU6rheRyK5zFR6hAk/7mANORVVlYGTHe97rrrcP3118d0bFlZGRYvXhww9Var1eKkk05CYWEhGhoasHbt2oAioSqVCitWrAh7I+yvubkZF110EQ4dOuTbp1QqsWDBAowcORJtbW1Yv359wPKeMpkMjz/+eMDfJxybzYZLLrkkoBCfXC7HvHnzMH78eJjNZmzcuDHkicddd92FSy65JGLfkiTh2muvDfnSefTRR2Pq1KlwOp3YvHkzdu3aFfD+r3/9a9x+++0R+6bh5+uvv8Zll13Wb/0Fr/YG8Fzryb333hsw9R4ApkyZgtmzZ0Mul2PXrl34/vvvA94/7bTT8Nhjj0Xtm4avw4cPY+nSpQFf3gRBwFFHHYUpU6ZApVKhtrYW3377bUh9rpNOOgmPP/445HJ52L55TQ7FazIl0vLly/H444/7tj///POoy9DzPA7F85goNTCQNMz0JZAEAHv37sXll1+OhoaGqG31ej2WL1+O4447Lqa+a2pq8Mtf/jJgWdOeKBQK3HvvvTjvvPNi6rujowNXXHFFwLTdnshkMtx44434zW9+E1PfdrsdN9xwQ9gZDOEsXboUd911F5+YUIhwwYy+CBdIAniuhSNJEu6++26sXLkypr4XLlyIf/zjH3GvnkfDT0VFBW644YaAYrCRCIKAyy67DDfeeGPIqkbBeE0OxWsyJUpvAkkAz+NweB4TJT8GkoaZvgaSAKC9vR2PPvoo3n33XZhMppD3FQoFTj31VNxyyy0oLCyMq2+bzYannnoKK1euDFn+GPDcYM+fPx+33HILJkyYEFffLpcLr776Kp577rmQqb1djjzySNx0001hixdGs2bNGjz55JM9XqyPOOIIXH/99Vi4cGHcfdPwcOWVV8a8AkosegokATzXerJ+/Xo88sgj2LNnT9j3R4wYgd/85jf4+c9/HnffNHzZ7Xa8/vrreOmll1BZWRm2jUKhwIknnoirrroqICUuGl6Tw+M1mQZbbwNJAM/jnvA8JkpeDCRRr9ntdnz33XeorKxEa2sr1Go1RowYgdmzZyM9Pb1PfbtcLmzatAkVFRVobm6GUqlEQUEBZs+ejdzc3D71LUkSduzYgf3796OpqQlyuRy5ubmYOXMmSkpK+tQ34PnyvmfPHjQ2NkKSJGRnZ2P69OkYO3Zsn/sm6m8818I7dOgQtm3bhoaGBjidTmRkZGDq1Kk44ogj+MST+uTgwYPYsWMHmpub4XA4kJ6ejuLiYsyePTtgye548ZocHq/JlEp4HofH85go+TCQREREREREREREMeGqbUREREREREREFBMGkoiIiIiIiIiIKCYMJBERERERERERUUwYSCIiIiIiIiIiopgwkERERERERERERDFhIImIiIiIiIiIiGLCQBIREREREREREcWEgSQiIiIiIiIiIooJA0lERERERERERBQTBpKIiIiIiIiIiCgmDCQREREREREREVFMGEgiIiIiIiIiIqKYMJBEREREREREREQxYSCJiIiIiIiIiIhiwkASERERERERERHFhIEkIiIiIiIiIiKKCQNJREREREREREQUEwaSiIiIiIiIiIgoJgwkERERERERERFRTBhIIiIiIiIiIiKimDCQREREREREREREMRETPQAiIiKioaq6uhr79+9HVVUVOjs74XQ6odFoYDQaUVxcjAkTJiAjIyPRwyQiIiKKGQNJRERERP1o7969ePvtt/HZZ5+huro6avuxY8filFNOwU9/+lOMGjWqT59tsVig0Wj61AcRERFRJIIkSVKiB0FERESU6qqrq3Hvvffi888/79XxMpkMZ5xxBm6//Xbk5eXFffzq1auxYsUKvPXWW736fCIiIqJYMJBERERE1EfffPMNrr/+enR2dva5L6PRiH/+85847rjjYmp/4MAB/OUvf8H333+PoqIifPHFF30eAxEREVFPGEgiIiIi6oMtW7Zg2bJlsNvtAft1Oh1OOukkTJ8+HaNHj4bBYIBarUZnZyeam5uxZ88efPXVV9ixY0dInwqFAs888wyOPfbYiJ/94Ycf4rbbboPD4QAABpKIiIhowDGQRERERNRLFosFZ5xxBmpqanz7ZDIZLr/8clx99dXQarVR+9i+fTvuuusu7N27N2B/eno6PvzwQ2RlZfV47PLly/H444/7thlIIiIiooEmS/QAiIiIiFLVyy+/HBBEAoB7770XN998c0xBJACYPn063njjDRx99NEB+1tbW/HYY4/121iJiIiI+gMDSURERES99MYbbwRsn3LKKbjgggvi7ketVuPRRx+FwWAI2L9mzZp+qbtERERE1F8YSCIiIiLqhbKyMlRVVQXsO//883vdX3Z2Ni699NKAfRaLBRs3bux1n0RERET9jYEkIiIiol44ePBgyL4RI0b0qc9FixaF7NuzZ0+f+iQiIiLqT2KiB0BERESUisxmc8i+1tbWPvU5duxYaLVaX986nQ4ul6tPfcbD7XZj586dOHz4MBobG2Gz2ZCZmYmcnBzMmjULRqOx3z9TkiTs2LEDe/fuRXNzM0RRRFZWFiZNmoQJEyZAJuv7c8+6ujps374d9fX16OjogEajQUZGBnJzczFjxgxoNJp++JsQERENDwwkEREREfVCcD0jAPjkk09w1FFH9bpPQRCwevVqX6BDqVSGtLn00kvx3XffhT2+qqoKEydODNj38ssvY+7cuRE/t6ysDE8//TS++OKLHoNhoihi9uzZWLZsGU455ZTY/kIA7rjjDqxevdq3/eCDD+Lcc88FALz55pt48sknUV1dHfbYwsJCXHjhhfjlL38Zd7DHbDZjxYoVePfdd3HgwIEe2ymVSsyaNQtnn302zj333LD/5kRERNSNqW1EREREvRAcsAGAlStX9rmm0ahRo5CXlzcoAQ273Y57770XZ511Ft55552IM6qcTie+++47XHfddVi8eDEOHz7c68+1Wq247rrrcOedd/YYRAKA6upqPProozj99NOxefPmmPvfsmULTjvtNDz88MMRg0iA59/gf//7H+68806cffbZ2L59e8yfQ0RENBwxkERERETUC4WFhZg0aVLAPofDgSuuuAKPPPIIGhsbEzSy2LS3t+Oyyy7DK6+8AqfTGdexmzdvxkUXXdSroIvb7cZvf/tbfPrppzEfU1tbi2XLluHLL7+M2nbTpk345S9/ifr6+rjHVl5ejmXLlmHHjh1xH0tERDRcCJIkSYkeBBEREVEqev/993HzzTeHfU+hUGDevHlYuHAh5s+fj7y8vH75zMcee8w3y+bAgQMBRb81Gg3mz58f0P7666/H+PHjA/Y5HA4sXboU27ZtC9gviiJOPfVUzJ8/HyUlJVCr1WhoaMD333+PDz74ALW1tQHtDQYDVq9ejZKSkh7HG5zaNn78eOzfv9+3bTQacd555+H4449HVlYWmpqa8NVXX+Hdd99FR0dHQF9qtRpvvvkmJkyYEPazzGYzzjrrrJDV9ObNm4eTTz4ZY8eOhdFohMvlQl1dHb7//nusXr0abW1tAe3HjRuH9957r1/qMxEREQ01DCQRERER9cFVV12FtWvXRm03evRoHH300TjyyCNx1FFHobi4uM+fvXz5cjz++OO+7aKiInzxxRdRj7v//vvxwgsvBOybOXMmHnrooR5XnrNarVi+fDmeffbZgP1Tp07Fa6+91mMqXnAgyd+xxx6LBx98ELm5uSHv1dXV4ZZbbgmpBzVlyhS89dZbYYM8b775Ju68807ftlwux8MPP4wzzzwz7OcDQFNTE6666qqQ2VVPPvkkTj755B6PIyIiGq74mIWIiIioDx555BHMmTMnaruysjKsWrUKt99+O04++WQsWLAAf/jDH/Dhhx+is7NzEEbqsW/fPrz44osB+4499li88sorPQaRAM9soFtvvRW///3vA/bv3Lmzx0BRJPPmzcPTTz8dNogEAHl5eXjmmWdCipfv2rULH330UdhjPv/884Dt888/P2IQCQCysrLwxBNPQKfTBeyPJ/WOiIhoOGEgiYiIiKgPtFotnn/+eVx22WUQxdgXxK2pqcHbb7+NG2+8EccccwyuuuqqmGoA9dWLL74I/wnpaWlp+Oc//xlzce9f/vKXOPHEEwP2vfDCC4hnkntGRgYefvjhqJ+pVqvxwAMPQKvVBux/5ZVXwrYPTmmbPHlyTOPJzc3FWWed5dtWKBRoaGiI6VgiIqLhhoEkIiIioj5SKBS47bbbsGbNGpxzzjlxr7jmcDiwdu1aXHnllbjwwguxc+fOARlnR0cH3n///YB9S5cuRVpaWlz9XHHFFQHbZWVl2LRpU8zHX3311cjMzIypbXFxMS644IKAfVu2bAm72ptCoQhpF6tf/epXePHFF/H5559j27ZteO6552I+loiIaDhhIImIiIion4wbNw4PPfQQvvnmG/z1r3/FwoULQ2bTRLNt2zZcfPHFeOmll/p9fJs3b4bdbg/YFzy7KBazZs1CRkZGwL4ffvghpmOVSiXOPffcuD7vvPPOC9iWJAnffvttSLuRI0cGbL/33nt4+umnY1qVbvTo0Tj22GNRXFwMuVwe1/iIiIiGEwaSiIiIiPqZwWDABRdcgCeeeALfffcdVqxYgeuvvx5z586FWq2OerzD4cBf//pXPPPMM/06rs2bN4fsCw6+xEImk+GII46I2nc406ZNQ3p6elyfN3nyZGg0moB9wSvOAQhITwM8Aae///3vOPXUU/Hggw/iu+++g8PhiOuziYiIKFDsifxEREREFDeFQoGjjjrKVzTabrdj69at+Pbbb7F+/Xrs2rWrx/pCjzzyCKZPn465c+f2y1j2798fsu/OO++EIAhx91VRURGwXVlZGdNxU6dOjfuzZDIZJkyYEBA8Cv58wDO7aubMmdi6dWvA/qqqKjz33HN47rnnoNPpMHfuXBx//PGYP38+SkpK4h4PERHRcMZAEhEREdEgUiqVmDNnDubMmYPf/e53qK2txTvvvIOXX34ZLS0tAW3dbjf++te/Ys2aNf3y2W1tbSH7PvnkkwHrO5zCwsJe9R+cStfU1BTSRiaT4cknn8SSJUtQXl4eth+TyYQvvvgCX3zxBQBgzJgxOOmkk3Daaadh+vTpvRobERHRcMLUNiIiIqIEys/PxzXXXINPP/0Up59+esj7e/fujauQdSSxBnsGsm+9Xt+r/oOPM5vNYdtlZWVh9erVWLZsWUjx7XBKS0vx7LPP4uc//znOOOOMkGLkREREFIiBJCIiIqIkYDAY8I9//APHH398yHsbNmzol88YyPpAsfYdS3AnHJfLFbAtij1PrNdqtfjjH/+ItWvX4o477sDMmTNjSt8rLS3FzTffjMsuuwxWq7VX4yQiIhrqmNpGREREFKdVq1bh66+/RlNTExobG6FUKvHee+/1uV+ZTIbbb78dX3/9dcD+AwcO/H979xrT8/vHcfz1k/kqSVTMOSpMjIVyamZmHbBlzjYsh1mb4Qabccts3GE2hzndiGLmWMKwGKaMWFY0x5oOztJRKPS70f7Wp8+3ur6V/f9/no971/W9Tt1rr13X+9PqtaW6sKq+ESNG6OTJk22ytqnPnz+3aF5FRYWl3alTp2bn+Pn5KTY2VrGxsSouLtatW7d0+/Zt3blzR+/evWt0XlpamtasWaMDBw606KwAAPzJCJIAAABc9ODBA125csXSV1xcLB8fn1avPWjQIPn5+enDhw+/+hqGKC3VpUsXS/t3PnVrTEv3bFgTqWfPni7N9/HxUUxMjGJiYiRJubm5Sk9P17Vr15SRkaGfP39axt+4cUOpqamaOnVqi84LAMCfiqdtAAAALnJWMLqt6hhJdXWT6jO5fWOid+/elvbr169VXV3dJmubcvbluObU1NTYbmUNHDiwVecICAjQ4sWLdeTIEaWmpmr8+PG2MRcvXmzVHgAA/IkIkgAAAFzk7OteiYmJbbb+mzdvLO22+kT9qFGjLO3q6mplZ2e3aK2kpCRduHBBmZmZevfunWpra43mZWVlubxXdna2rQbTmDFjbONqamqUm5ur1NRUJScnG6/fp08f7d27V35+fpb+tnpSCADAn4SnbQAAAC4aO3asvLy8VF5e/qsvIyNDd+/eVVhYWKvWvnfvnj5+/GjpCw0NdTrWpIB0fc7ClzNnzmj06NEurVNYWKiNGzdanoNNmjRJBw8ebHZuUVGRMjMzFRISYrxfUlKSpe3h4WE7c3x8vLZv367v379LqrvFNX369CaLcjtb89KlS7/6vnz5YnxGAAD+FtxIAgAAcJHD4dCsWbNs/evWrVN+fn6L1/327Zu2bdtm6fP19VV4eLjT8e3aufavXK9evTRu3DhL3/nz5/Xs2TOX1tm5c6etplBkZKTx/P379xuPzcvLU0pKiqUvIiLC9twvICDgV4gk1RX1vnr1qvE+kvTp0ydL29fX16X5AAD8DQiSAAAAWmDlypW24tXv37/XvHnzlJqa6vJ6nz59UlxcnHJycmz7NHarpkOHDpa2yQ2aZcuWWdo1NTVas2aNrZh1Y5KTk221g7p3767o6Gij+ZJ08+ZNHTt2rNlxlZWVWr9+vb59+/arr127dra/QZLCwsLk7e1t6duxY4cqKyuNzvT8+XPdv3/f0tfwKSAAACBIAgAAaJGuXbtq69attv6SkhKtWrVKCxcu1KVLl/T169cm18nPz9e+ffsUFRWl9PR0y28jRozQggULGp3bMMgqKSlpNhAKDw/X5MmTLX15eXmaO3dukwXDa2trdfToUW3atMn22+rVq9WxY8cm921oy5YtOnjwoH78+OH098LCQsXGxurRo0eW/jlz5igoKMg23uFwaN68eZa+goICLV++3FZzqqHc3FzFxcVZzuLm5qbZs2eb/jkAAPw1/qk1rYwIAAAAm8OHD9ueo9XXoUMHBQcHa8CAAerWrZvat2+v0tJSlZSUKCcnR0VFRU7n9e7dW8eOHWvyM/dpaWm22zmjRo1SXFycfH19VV5eroCAANsTrbKyMs2cOVOvXr2yrTlhwgRNmTJFgYGB8vLyUllZmR4+fKiUlBSnT+AiIiK0a9euRs+4YcMGW42j+oKCgjRr1iwNHTpUnTp10tu3b3Xjxg2dP3/eFsIFBATo9OnT8vDwcLpWZWWlZsyYodevX1v63d3dFRkZqbFjx6pfv37y8PBQVVWVCgoKlJ6ersuXL9u+XrdkyRJt3Lix0XMDAPC3IkgCAABopeTkZG3evFlVVVVtst6wYcO0Z8+eJkMkSSovL9f48eNtXzSrb/PmzZo/f76tv7CwUCtXrlRubm6Lzzlx4kTt3r270WBHsgdJgwYN0vfv35WXl+fSXv7+/kpISFCPHj2aHJeVlaWlS5caP2lzZuLEidq3b5/t6SAAAOBpGwAAQKvFxMTo3LlzioqKcvlLavV5enpq7dq1On78eLMhkiR5eXlp1apVTY55+vSp0/6+ffvqxIkTmj59ustnbt++vZYtW6b9+/c3GSI54+3trcOHD2v48OHGc6Kjo3Xq1KlmQySp7jlgQkKCBg4c6NK5pLr6S0uWLCFEAgCgCdxIAgAAaEMvX75USkqKrl+/rqdPnzZaA+g/3NzcFBwcrGnTpikmJsZWMNpEfHy8du3a5fRGVEhIiI4fP97k/CdPnujQoUNKS0tTaWlpo+M8PT01bdo0LVq0yGmdImca3kgKDQ1VYmKifvz4ofj4eB09etRpDSM3NzeFh4drxYoVGj16tNFe9VVXV+v06dM6e/asHj582OTYzp07KyIiQrGxsQoMDHR5LwAA/iYESQAAAL9JVVWVXrx4oVevXqm4uFhVVVWqqamRw+FQly5d5O/vr8GDB8vLy6vVe1VWViojI0MFBQX6/Pmz3N3d5ePjo8DAQAUHBxut8fPnT+Xk5KigoEAlJSWqqKiQu7u7vL29NWTIEAUFBcnNzc2lczUWJNXfMzMzU/n5+fr48aMcDof69++vkSNHqmvXri7t1ZjS0lI9fvxYRUVFqqio0JcvX+RwOOTn56fAwEANGTLE5b8LAIC/FUESAAAAfpvmgiQAAPD/hRpJAAAAAAAAMEKQBAAAAAAAACMESQAAAAAAADBCkAQAAAAAAAAjBEkAAAAAAAAwQpAEAAAAAAAAIwRJAAAAAAAAMEKQBAAAAAAAACP/1NbW1v63DwEAAAAAAID/fdxIAgAAAAAAgBGCJAAAAAAAABghSAIAAAAAAIARgiQAAAAAAAAYIUgCAAAAAACAEYIkAAAAAAAAGCFIAgAAAAAAgBGCJAAAAAAAABghSAIAAAAAAIARgiQAAAAAAAAYIUgCAAAAAACAEYIkAAAAAAAAGCFIAgAAAAAAgBGCJAAAAAAAABghSAIAAAAAAIARgiQAAAAAAAAYIUgCAAAAAACAEYIkAAAAAAAAGCFIAgAAAAAAgBGCJAAAAAAAABghSAIAAAAAAIARgiQAAAAAAAAYIUgCAAAAAACAEYIkAAAAAAAAGCFIAgAAAAAAgBGCJAAAAAAAABghSAIAAAAAAIARgiQAAAAAAAAYIUgCAAAAAACAkX8BZP5CwJGmGVYAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 1200x1200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "from mighty.mighty_utils.plotting import plot_performance_over_time\n",
    "\n",
    "# Make episode reward a scalar instead of list\n",
    "ppo_cartpole['episode_reward'] = [np.array(x).mean() for x in ppo_cartpole.episode_reward.values]\n",
    "# Just take eval points for plotting\n",
    "plot_data = ppo_cartpole[ppo_cartpole.episode_reward.isna() == False]\n",
    "\n",
    "plot_performance_over_time(plot_data, x=\"total_steps\", y=\"episode_reward\", hue=\"algorithm\", aggregation=\"mean\", logx=False, xlabel=\"Steps\", ylabel=\"Mean Reward\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Score history:\n",
    "\n",
    "PPO: max 30 - ..."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "mighty",
   "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": "3.10.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}