Leshinskaya_Chakroff_NeurIPS_2023.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# Setup"
      ],
      "metadata": {
        "id": "w30X8ATifJ9K"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#import packages\n",
        "import sys, os, random, requests, json\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import scipy.stats as st\n",
        "import matplotlib.pyplot as plt\n",
        "from google.colab import drive, data_table"
      ],
      "metadata": {
        "id": "Z2CkGcw5hLA8"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#set up paths for local modules and data sources\n",
        "drive.mount('/content/drive',force_remount=True)\n",
        "sys.path.append('*** working directory ***')\n",
        "stim_path = \"./Stimuli/\"\n",
        "data_path = \"./Results/\""
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "J3uYqfKwhPu0",
        "outputId": "dda3c7a1-b758-4b16-9c14-ae99bc70f810"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#set environment variable with API key\n",
        "%env OPENAI_API_KEY = Bearer ***your-key***"
      ],
      "metadata": {
        "id": "nZEyWxJQhQgq"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# read in stimuli\n",
        "# action_data = pd.read_csv(stim_path + \"action_neurips.csv\", header=None)\n",
        "# action_list_all = action_data[0].tolist()\n",
        "\n",
        "action_data = pd.read_csv(stim_path + \"actions_subset_uncor.csv\", header=0)\n",
        "action_list_all = action_data[\"Item\"][0:15].tolist()"
      ],
      "metadata": {
        "id": "pkXNCDZRge_j"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "action_list_all"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SRLjRDEg_t2l",
        "outputId": "de547ae5-4951-46b4-a389-957d91c972c0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['sitting in a chair',\n",
              " 'running into a burning building to save a precious antique',\n",
              " 'watching a rival sports team lose badly',\n",
              " 'running a marathon for charity',\n",
              " 'running a marathon for fitness',\n",
              " 'mowing the lawn',\n",
              " 'donating money to charity',\n",
              " 'writing letters to the local government to support a homeless shelter',\n",
              " 'helping someone change their flat tire on the road',\n",
              " 'volunteering at a voting location',\n",
              " 'painting a wall',\n",
              " 'taking a day off to go to the spa',\n",
              " 'winning millions in the lottery',\n",
              " 'watching your favorite movie',\n",
              " 'cleaning up litter in a rough neighborhood']"
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Embedding Projection"
      ],
      "metadata": {
        "id": "wz2HUbWbfKaT"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#@title define all functions\n",
        "def getEmbeddings(text):\n",
        "    url = 'https://api.openai.com/v1/embeddings'\n",
        "    headers = {\n",
        "      \"Content-Type\": \"application/json\",\n",
        "      \"Authorization\": os.getenv(\"OPENAI_API_KEY\")\n",
        "      }\n",
        "    data = {\n",
        "        \"input\": text,\n",
        "        \"model\": \"text-embedding-ada-002\"\n",
        "    }\n",
        "    response = requests.post(url, headers=headers, json=data)\n",
        "\n",
        "    return response.json()\n",
        "\n",
        "def return_embeddings_diff(attributes_a, attributes_b):\n",
        "  '''\n",
        "  Pass two sets of attributes and get all vector differences of all a-b combinations\n",
        "  '''\n",
        "  emb_high = pd.DataFrame()\n",
        "  emb_low = pd.DataFrame()\n",
        "\n",
        "  #loop through each attribute set and save the embeddings in two dataframes\n",
        "  for a in range(len(attributes_a)):\n",
        "    #get embedding of the high, low, and compute difference\n",
        "    this_emb = getEmbeddings(attributes_a[a])\n",
        "    emb_high.insert(loc=0, column=a, value=this_emb[\"data\"][0][\"embedding\"])\n",
        "\n",
        "  for b in range(len(attributes_b)):\n",
        "    #get embedding of the high, low, and compute difference\n",
        "    this_emb = getEmbeddings(attributes_b[b])\n",
        "    emb_low.insert(loc=0, column=b, value=this_emb[\"data\"][0][\"embedding\"])\n",
        "\n",
        "  #get all differences\n",
        "  vector_diff = pd.DataFrame()\n",
        "\n",
        "  #report correlations\n",
        "  print(emb_low.corr())\n",
        "  print(emb_high.corr())\n",
        "\n",
        "  for a in range(len(attributes_a)):\n",
        "    for b in range(len(attributes_b)):\n",
        "      this_col = str(a)+'_'+str(b)\n",
        "      vector_diff.insert(loc=0, column=this_col, value=emb_high[a] - emb_low[b])\n",
        "\n",
        "  return(vector_diff.mean(axis=1))\n",
        "\n",
        "\n",
        "def return_list_embeddings(alist):\n",
        "  '''\n",
        "  get embeddings for a list of items\n",
        "  '''\n",
        "  n_df = pd.DataFrame()\n",
        "  for i in alist:\n",
        "    this_em = getEmbeddings(i)[\"data\"][0][\"embedding\"]\n",
        "    n_df.insert(loc=0, column=i, value=this_em, allow_duplicates=True)\n",
        "\n",
        "  return (n_df)\n",
        "\n",
        "\n",
        "def get_projections(stim_list, moral_v, hedonic_v, movement_v):\n",
        "  '''\n",
        "  get projections for list of items\n",
        "  '''\n",
        "  #construct dataframe to save items\n",
        "  projection_df = pd.DataFrame(index=stim_list, columns=['moral_v','hedonic_v','movement_v'], data=0)\n",
        "\n",
        "  #loop through items to get embeddings in moral, hedonic, and movement vector directions\n",
        "\n",
        "  for a in stim_list:\n",
        "    this_em = getEmbeddings(a)[\"data\"][0][\"embedding\"]\n",
        "    projection_moral = np.inner(np.array(this_em),np.array(moral_v))\n",
        "    projection_hedonic = np.inner(np.array(this_em),np.array(hedonic_v))\n",
        "    projection_movement = np.inner(np.array(this_em),np.array(movement_v))\n",
        "    projection_df.loc[a, \"moral_v\"] = projection_moral\n",
        "    projection_df.loc[a, \"hedonic_v\"] = projection_hedonic\n",
        "    projection_df.loc[a, \"movement_v\"] = projection_movement\n",
        "\n",
        "  return projection_df"
      ],
      "metadata": {
        "id": "jrAksmmqfzHv"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# get moral direction in GPT embeddings\n",
        "attributes_morality_high = ['morally virtuous','ethical', 'high moral value']\n",
        "attributes_morality_low = ['morally wrong','unethical', 'low moral value']\n",
        "moral_v = return_embeddings_diff(attributes_morality_high, attributes_morality_low)\n",
        "\n",
        "# get hedonic direction in GPT embeddings\n",
        "attributes_hedonic_high = ['personally rewarding','pleasurable for me', 'high hedonic value for me']\n",
        "attributes_hedonic_low = ['personally costly','unpleasurable for me', 'low hedonic value for me']\n",
        "hedonic_v = return_embeddings_diff(attributes_hedonic_high, attributes_hedonic_low)\n",
        "\n",
        "# get physicality direction in GPT embeddings\n",
        "attributes_movement_high = ['physical','bodily action', 'high movement']\n",
        "attributes_movement_low = ['mental','minimally active', 'low movement']\n",
        "movement_v = return_embeddings_diff(attributes_movement_high, attributes_movement_low)\n"
      ],
      "metadata": {
        "id": "W3f-PQlYgaZx"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Report correlation among attribute embeddings\n",
        "print('Embedding Correlations:')\n",
        "print('r(moral, hedonic) = '+ str(round(np.corrcoef(moral_v, hedonic_v)[0, 1], 2)))\n",
        "print('r(moral, movement) = '+ str(round(np.corrcoef(moral_v, movement_v)[0, 1], 2)))\n",
        "print('r(hedonic, movement) = '+ str(round(np.corrcoef(hedonic_v, movement_v)[0, 1], 2)))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8E_4yIo2gohz",
        "outputId": "25851670-9321-4225-b803-792134915d0c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Embedding Correlations:\n",
            "r(moral, hedonic) = 0.44\n",
            "r(moral, movement) = 0.09\n",
            "r(hedonic, movement) = 0.16\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# get projections of stimuli onto each vector direction\n",
        "item_projections = get_projections(action_list_all, moral_v, hedonic_v, movement_v)"
      ],
      "metadata": {
        "id": "3aUsrDQggmQA"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print('Item Embedding Projection Correlations:')\n",
        "print('r(moral, hedonic) = '+ str(round(np.corrcoef(item_projections[\"moral_v\"], item_projections[\"hedonic_v\"])[0, 1], 3)))\n",
        "print('r(moral, movement) = '+ str(round(np.corrcoef(item_projections[\"moral_v\"], item_projections[\"movement_v\"])[0, 1], 3)))\n",
        "print('r(hedonic, movement) = '+ str(round(np.corrcoef(item_projections[\"hedonic_v\"], item_projections[\"movement_v\"])[0, 1], 3)))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3hrkSWXKulW8",
        "outputId": "5c744b15-935e-43ed-8a78-a25aa9562a8b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Item Embedding Projection Correlations:\n",
            "r(moral, hedonic) = 0.799\n",
            "r(moral, movement) = 0.112\n",
            "r(hedonic, movement) = 0.041\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Plot batch 1, Moral x Hedonic\n",
        "\n",
        "# sort proections by morality before plotting\n",
        "item_projections_1 = item_projections[:18].sort_values('moral_v', ascending=False)\n",
        "\n",
        "\n",
        "fig, ax = plt.subplots(figsize=(10, 10))\n",
        "cmap = plt.get_cmap(\"tab20\")\n",
        "\n",
        "plot_i = 1\n",
        "for i in range(len(item_projections_1)):\n",
        "    x = item_projections_1['hedonic_v'].iloc[i]\n",
        "    y = item_projections_1['moral_v'].iloc[i]\n",
        "    colors = cmap(i/len(item_projections_1))\n",
        "    l = str(plot_i) + '. ' + item_projections_1.index[i].lstrip('[').rstrip(']')\n",
        "    ax.scatter([x],[y], label=l)\n",
        "    ax.annotate(plot_i, (item_projections_1['hedonic_v'][i]+.0005, item_projections_1['moral_v'][i]-.0001))\n",
        "    plot_i += 1\n",
        "\n",
        "ax.set_xlabel('Hedonic')\n",
        "ax.set_ylabel('Moral')\n",
        "ax.legend()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 872
        },
        "id": "rIs5x8LUhElR",
        "outputId": "650c24bc-3b83-4e72-df76-6eaad7ad4039"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7858a1b79f30>"
            ]
          },
          "metadata": {},
          "execution_count": 11
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x1000 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Prompting"
      ],
      "metadata": {
        "id": "GbgxbjpR4xQo"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# @title promptGPT\n",
        "def promptGPT(prompt_message_list, gpt_temperature=0):\n",
        "  gpt_url = \"https://api.openai.com/v1/chat/completions\"\n",
        "  gpt_headers = {\n",
        "    \"Content-Type\": \"application/json\",\n",
        "    \"Authorization\": os.getenv(\"OPENAI_API_KEY\")\n",
        "    }\n",
        "  gpt_data = {\n",
        "        \"model\": \"gpt-3.5-turbo-1106\",\n",
        "        \"response_format\": {\"type\": \"json_object\"}, # only works on 3.5-turbo-1106, 4 and above\n",
        "        \"temperature\": gpt_temperature,\n",
        "        \"messages\": prompt_message_list,\n",
        "    }\n",
        "  response = requests.post(gpt_url, headers=gpt_headers, json=gpt_data)\n",
        "  output = response.json()['choices'][0]['message']['content']\n",
        "\n",
        "  return output"
      ],
      "metadata": {
        "id": "otLfWGOYbdcZ",
        "cellView": "form"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# @title prompt_sort\n",
        "def prompt_sort(action_list, dimension, context=\"\"):\n",
        "  \"\"\"\n",
        "  Given a list of actions to sort, and a dimension to sort them on, craft a prompt\n",
        "  for GPT, with example calls and responses.\n",
        "\n",
        "  Optionally, include context regarding the hypothetical rater (Ziv),\n",
        "  which may influence the ranking results.\n",
        "\n",
        "  Returns a list of dicts to pass on to GPT as prompt messages\n",
        "  \"\"\"\n",
        "\n",
        "  #randomize the order of the list\n",
        "  random.shuffle(action_list)\n",
        "\n",
        "  #turn list into formatted string\n",
        "  action_string = (', ').join(action_list)\n",
        "\n",
        "  # Switch by sorting dimension\n",
        "  if dimension == 'moral':\n",
        "\n",
        "    system_prompt_content = \"\"\"You are an expert judge of the relative moral virtues of different actions.\n",
        "Moral actions are those which humans consider virtuous, that consider others' wellbeing and happiness, and that are guided by principles of ethics.\n",
        "You will be given a list of actions that an average human person, Ziv, is considering.\n",
        "You must sort them in terms of their relative moral virtue according to Ziv.\n",
        "You will recieve a JSON object of items to rate. This object will have the following key / value pairs:\n",
        "1. actions - Array<string>: an unsorted list of actions that the user has been faced with\n",
        "2. context - string: information that you can use to help you make your choice\n",
        "You must respond with a JSON packet with a single key / value pair:\n",
        "1. ranking - Array<string>: A sorted list of actions, starting with the most morally wrong, and ending with the most morally virtuous from Ziv's perpsective.\"\"\"\n",
        "    system_prompt= {\n",
        "        \"role\": \"system\",\n",
        "        \"content\": system_prompt_content\n",
        "    }\n",
        "    user_example = {\n",
        "        \"role\": \"user\",\n",
        "        \"content\": json.dumps({\"actions\": [\"walk a dog\", \"rob a bank\", \"rescue a cat\", \"steal food\"],\n",
        "                               \"context\": \"\"}) # Initial example has no context\n",
        "    }\n",
        "    assistant_example = {\n",
        "        \"role\": \"assistant\",\n",
        "        \"content\": json.dumps({\"ranking\": [\"rob a bank\", \"steal food\", \"walk a dog\", \"rescue a cat\"]})\n",
        "    }\n",
        "\n",
        "  elif dimension == 'hedonic':\n",
        "\n",
        "    system_prompt_content = \"\"\"You are an expert judge of the relative hedonic reward of different actions.\n",
        "Hedonically rewarding actions are those which humans consider pleasant, make them feel happy, and benefit their own wellbeing.\n",
        "You will be given a list of actions that an average human person, Ziv, is considering.\n",
        "You must sort them in terms of their relative hedonic reward for Ziv.\n",
        "You will recieve a JSON object of items to rate. This object will have the following key / value pairs:\n",
        "1. actions - Array<string>: an unsorted list of actions that the user has been faced with\n",
        "2. context - string: information that you can use to help you make your choice\n",
        "You must respond with a JSON packet with a single key / value pair:\n",
        "1. ranking - Array<string>: A sorted list of actions, starting with the\n",
        "worst and least hedonically rewarding, and ending with the best and most hedonically rewarding from Ziv's perpsective.\"\"\"\n",
        "    system_prompt= {\n",
        "        \"role\": \"system\",\n",
        "        \"content\": system_prompt_content\n",
        "    }\n",
        "    user_example = {\n",
        "        \"role\": \"user\",\n",
        "        \"content\": json.dumps({\"actions\": [\"miss the bus\", \"win a sweepstakes\", \"eat a meal\", \"lose my wallet\"],\n",
        "                               \"context\": \"\"}) # Initial example has no context\n",
        "    }\n",
        "    assistant_example = {\n",
        "        \"role\": \"assistant\",\n",
        "        \"content\": json.dumps({\"ranking\": [\"lose my wallet\", \"miss the bus\", \"eat a meal\", \"win a sweepstakes\"]})\n",
        "    }\n",
        "\n",
        "  elif dimension == 'movement':\n",
        "\n",
        "    system_prompt_content = \"\"\"You are an expert judge of the physical body movement\n",
        "of different actions. Physically active actions are those which humans consider to involve substantial movement of the body.\n",
        "You will be given a list of actions that an average human person, Ziv, is considering.\n",
        "You must sort them in terms of how much physical movement they involve.\n",
        "You will recieve a JSON object of items to rate. This object will have the following key / value pairs:\n",
        "1. actions - Array<string>: an unsorted list of actions that the user has been faced with\n",
        "2. context - string: information that you can use to help you make your choice\n",
        "You must respond with a JSON packet with a single key / value pair:\n",
        "\n",
        "1. ranking - Array<string>: A sorted list of actions, starting with the\n",
        "least physical body movement, and ending with the most physical body movement.\"\"\"\n",
        "    system_prompt= {\n",
        "        \"role\": \"system\",\n",
        "        \"content\": system_prompt_content\n",
        "    }\n",
        "    user_example = {\n",
        "        \"role\": \"user\",\n",
        "        \"content\": json.dumps({\"actions\": [\"play soccer\", \"listen to music\", \"pack boxes\", \"order coffee at a cafe\"],\n",
        "                               \"context\": \"\"}) # Initial example has no context\n",
        "    }\n",
        "    assistant_example = {\n",
        "        \"role\": \"assistant\",\n",
        "        \"content\": json.dumps({\"ranking\": [\"listen to music\", \"order coffee at a cafe\", \"pack boxes\", \"play soccer\"]})\n",
        "    }\n",
        "\n",
        "  else:\n",
        "    raise Exception(\"Specify a rating dimension: moral, hedonic, movement\")\n",
        "\n",
        "\n",
        "  # List actions to sort\n",
        "  user_prompt = {\n",
        "      \"role\": \"user\",\n",
        "      \"content\": json.dumps({\"actions\": action_list,\n",
        "                             \"context\": context}) # Optional context, defaults to empty string\n",
        "  }\n",
        "\n",
        "  return [system_prompt, user_example, assistant_example, user_prompt]"
      ],
      "metadata": {
        "id": "TmBV-cJv43Rf",
        "cellView": "form"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# @title prompt_rate_sorted\n",
        "def prompt_rate_sorted(action_list, dimension, context=\"\"):\n",
        "  \"\"\"\n",
        "  Given a sorted list of actions to rate, and a dimension to sort them on, craft a prompt\n",
        "  for GPT, with example calls and responses.\n",
        "\n",
        "  Optionally, include context regarding the hypothetical rater (Ziv),\n",
        "  which may influence the rating results.\n",
        "\n",
        "  Returns a list of dicts to pass on to GPT as prompt messages\n",
        "  \"\"\"\n",
        "\n",
        "  #turn list into formatted string\n",
        "  action_string = (', ').join(action_list)\n",
        "\n",
        "  # Switch by sorting dimension\n",
        "  if dimension == 'moral':\n",
        "\n",
        "    system_prompt_content = \"\"\"You are an expert judge of the relative moral virtues of different actions.\n",
        "Moral actions are those which humans consider virtuous, that consider others' wellbeing and happiness, and that are guided by principles of ethics.\n",
        "You will be given a list of actions that an average human person, Ziv, is considering.\n",
        "They are already sorted in terms of their relative moral virtue according to Ziv.\n",
        "You will assign a number to each adjacent pair, rating 0 if the actions are very similar in terms of moral virtue, and 10 if the actions are very different in terms of moral virtue.\n",
        "You will give the first action a rating of 0. You will give the second action a rating in comparison to the first action, and so on.\n",
        "You will recieve a JSON object of items to rate. This object will have the following key / value pairs:\n",
        "1. actions - Array<string>: a sorted list of actions that the user has been faced with\n",
        "2. context - string: information that you can use to help you make your choice\n",
        "You must respond with a JSON packet with a single key / value pair:\n",
        "1. ranking - Array: A list of lists, each list contains 2 elements:\n",
        "1. action from the original list\n",
        "2. similarity rating from 0 to 10 between this action and the previous action in the list\"\"\"\n",
        "\n",
        "    system_prompt= {\n",
        "        \"role\": \"system\",\n",
        "        \"content\": system_prompt_content\n",
        "    }\n",
        "    user_example = {\n",
        "        \"role\": \"user\",\n",
        "        \"content\": json.dumps({\"actions\": [\"rob a bank\", \"steal food\", \"walk a dog\", \"rescue a cat\"],\n",
        "                               \"context\": \"\"}) # Initial example has no context\n",
        "    }\n",
        "    assistant_example = {\n",
        "        \"role\": \"assistant\",\n",
        "        \"content\": json.dumps({\"ranking\": [[\"rob a bank\", 0], [\"steal food\", 3], [\"walk a dog\", 10], [\"rescue a cat\", 6]]})\n",
        "    }\n",
        "\n",
        "  elif dimension == 'hedonic':\n",
        "\n",
        "    system_prompt_content = \"\"\"You are an expert judge of the relative hedonic reward\n",
        "of different actions. Hedonically rewarding actions are those which humans consider pleasant, make them feel happy, and benefit their own wellbeing.\n",
        "You will be given a list of actions that an average human person, Ziv, is considering.\n",
        "They are already sorted in terms of their relative hedonic reward for Ziv.\n",
        "You will assign a number to each adjacent pair, rating 0 if the actions are very similar\n",
        "in terms of hedonic reward, and 10 if the actions are very different in terms of hedonic reward.\n",
        "You will give the first action a rating of 0. You will give the second action a rating in comparison to the first action, and so on.\n",
        "You will recieve a JSON object of items to rate. This object will have the following key / value pairs:\n",
        "1. actions - Array<string>: a sorted list of actions that the user has been faced with\n",
        "2. context - string: information that you can use to help you make your choice\n",
        "You must respond with a JSON packet with a single key / value pair:\n",
        "1. ranking - Array: A list of lists, each list contains 2 elements:\n",
        "1. action from the original list\n",
        "2. similarity rating from 0 to 10 between this action and the previous action in the list\"\"\"\n",
        "\n",
        "    system_prompt= {\n",
        "        \"role\": \"system\",\n",
        "        \"content\": system_prompt_content\n",
        "    }\n",
        "    user_example = {\n",
        "        \"role\": \"user\",\n",
        "        \"content\": json.dumps({\"actions\": [\"lose my wallet\", \"miss the bus\", \"eat a meal\", \"win a sweepstakes\"],\n",
        "                               \"context\": \"\"}) # Initial example has no context\n",
        "    }\n",
        "    assistant_example = {\n",
        "        \"role\": \"assistant\",\n",
        "        \"content\": json.dumps({\"ranking\": [[\"lose my wallet\", 0], [\"miss the bus\", 2], [\"eat a meal\", 9], [\"win a sweepstakes\", 8]]})\n",
        "    }\n",
        "\n",
        "  elif dimension == 'movement':\n",
        "\n",
        "    system_prompt_content = \"\"\"You are an expert judge of the physical body movement of different actions.\n",
        "Physically active actions are those which humans consider to involve substantial movement of the body.\n",
        "You will be given a list of actions that an average human person, Ziv, is considering.\n",
        "They are already sorted in terms of how much physical movement they involve.\n",
        "You will assign a number to each adjacent pair, rating 0 if the actions are very similar\n",
        "in terms of amount of physical movement, and 10 if the actions are very different in terms of amount of physical movement.\n",
        "You will give the first action a rating of 0. You will give the second action a rating in comparison to the first action, and so on.\n",
        "You will recieve a JSON object of items to rate. This object will have the following key / value pairs:\n",
        "1. actions - Array<string>: a sorted list of actions that the user has been faced with\n",
        "2. context - string: information that you can use to help you make your choice\n",
        "You must respond with a JSON packet with a single key / value pair:\n",
        "1. ranking - Array: A list of lists, each list contains 2 elements:\n",
        "1. action from the original list\n",
        "2. similarity rating from 0 to 10 between this action and the previous action in the list\"\"\"\n",
        "\n",
        "    system_prompt= {\n",
        "        \"role\": \"system\",\n",
        "        \"content\": system_prompt_content\n",
        "    }\n",
        "    user_example = {\n",
        "        \"role\": \"user\",\n",
        "        \"content\": json.dumps({\"actions\": [\"listen to music\", \"order coffee at a cafe\", \"pack boxes\", \"play soccer\"],\n",
        "                               \"context\": \"\"}) # Initial example has no context\n",
        "    }\n",
        "    assistant_example = {\n",
        "        \"role\": \"assistant\",\n",
        "        \"content\": json.dumps({\"ranking\": [[\"listen to music\", 0], [\"order coffee at a cafe\", 2], [\"pack boxes\", 6], [\"play soccer\", 9]]})\n",
        "    }\n",
        "\n",
        "  else:\n",
        "    raise Exception(\"Specify a rating dimension: moral, hedonic, movement\")\n",
        "\n",
        "\n",
        "  # List actions to sort\n",
        "  user_prompt = {\n",
        "      \"role\": \"user\",\n",
        "      \"content\": json.dumps({\"actions\": action_list,\n",
        "                             \"context\": context}) # Optional context, defaults to empty string\n",
        "  }\n",
        "\n",
        "  return [system_prompt, user_example, assistant_example, user_prompt]"
      ],
      "metadata": {
        "id": "4hmGAX7ruGkI",
        "cellView": "form"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# @title sort_rate wrapper function\n",
        "def sort_rate(action_list, dimension):\n",
        "\n",
        "  # Sort actions\n",
        "  resp = promptGPT(prompt_sort(action_list, dimension))\n",
        "  resp_sorted_list = json.loads(resp)['ranking']\n",
        "\n",
        "  # Rate distance between sorted actions\n",
        "  resp = promptGPT(prompt_rate_sorted(resp_sorted_list, dimension))\n",
        "  resp_rated_list = json.loads(resp)['ranking']\n",
        "\n",
        "  # parse the ratings into a dataframe\n",
        "  resp_df = pd.DataFrame(resp_rated_list, columns=['item', 'rating'])\n",
        "  resp_df['rating'] = pd.to_numeric(resp_df['rating'])\n",
        "  resp_df['rating_sum'] = resp_df['rating'].cumsum()\n",
        "  resp_df['rescored_' + dimension] = resp_df['rating_sum'] / resp_df['rating_sum'].max() * 100\n",
        "  resp_df = resp_df.drop(columns=['rating', 'rating_sum'])\n",
        "\n",
        "  return resp_df"
      ],
      "metadata": {
        "id": "TMbocXyeqXpx",
        "cellView": "form"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# run the pipeline and get data!\n",
        "moral_df = sort_rate(action_list_all, 'moral')\n",
        "hedonic_df = sort_rate(action_list_all, 'hedonic')\n",
        "movement_df = sort_rate(action_list_all, 'movement')\n",
        "\n",
        "full_df = pd.merge(moral_df, hedonic_df, on='item', how='inner')\n",
        "full_df = pd.merge(full_df, movement_df, on='item', how='inner')"
      ],
      "metadata": {
        "id": "2VZ5YH1D5_57"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print('Prompting Correlations:')\n",
        "print('r(moral, hedonic) = '+ str(round(np.corrcoef(full_df[\"rescored_moral\"], full_df[\"rescored_hedonic\"])[0, 1], 2)))\n",
        "print('r(moral, movement) = '+ str(round(np.corrcoef(full_df[\"rescored_moral\"], full_df[\"rescored_movement\"])[0, 1], 2)))\n",
        "print('r(hedonic, movement) = '+ str(round(np.corrcoef(full_df[\"rescored_hedonic\"], full_df[\"rescored_movement\"])[0, 1], 2)))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "w-2XjLuSxDi1",
        "outputId": "b2098dc5-9574-4989-f941-0e441a9de0af"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Prompting Correlations:\n",
            "r(moral, hedonic) = 0.17\n",
            "r(moral, movement) = 0.5\n",
            "r(hedonic, movement) = 0.04\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Prompting Statistics"
      ],
      "metadata": {
        "id": "mFzy_Bmz_Th8"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Create list of lists with all items\n",
        "niter = 100\n",
        "dimension = 'hedonic'\n",
        "\n",
        "result_lists = [[action, []] for action in action_list_all]\n",
        "\n",
        "for i in range(niter):\n",
        "  result_df = []\n",
        "\n",
        "  while(len(result_df)<15):\n",
        "    result_df = sort_rate(action_list_all, dimension)\n",
        "\n",
        "  for action in action_list_all:\n",
        "    act_val = result_df[result_df['item'] == action]['rescored_'+ dimension].values[0]\n",
        "    cur_vals = []\n",
        "    for row in result_lists:\n",
        "      if row[0] == action:\n",
        "        row[1].append(act_val)\n",
        "        break"
      ],
      "metadata": {
        "id": "fc216hnMLbYW"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# summary statistics for each item\n",
        "new_result_lists = []\n",
        "\n",
        "for row in result_lists:\n",
        "  if len(row[1]) > 0:\n",
        "    CI = st.t.interval(0.95, len(row[1])-1, loc=np.mean(row[1]), scale=st.sem(row[1]))\n",
        "    new_row = [\n",
        "        row[0],\n",
        "        max(0, CI[0]), # 0 or low CI\n",
        "        np.mean(row[1]),\n",
        "        min(100, CI[1]), # 100 or high CI\n",
        "        ]\n",
        "    new_result_lists.append(new_row)\n",
        "  else:\n",
        "    print('Need more data for ' + row[0])\n",
        "\n",
        "new_results_df = pd.DataFrame(new_result_lists, columns=['item', '95CI_Low', 'mean', '95CI_High'])"
      ],
      "metadata": {
        "id": "GAY1YM4GzlrI"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "new_results_df.to_csv(data_path + 'hedonic_100_iter_results_uncorr14set.csv')"
      ],
      "metadata": {
        "id": "UEvDl7Bb_547"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# # Load and merge dataframes\n",
        "moral_df_new = pd.read_csv(data_path + 'moral_100_iter_results.csv')\n",
        "hedonic_df_new =  pd.read_csv(data_path + 'hedonic_100_iter_results.csv')\n",
        "movement_df_new =  pd.read_csv(data_path + 'movement_100_iter_results.csv')\n",
        "\n",
        "# moral_df_new = pd.read_csv(data_path + 'moral_100_iter_results_uncorr15set.csv')\n",
        "# hedonic_df_new =  pd.read_csv(data_path + 'hedonic_100_iter_results_uncorr15set.csv')\n",
        "# movement_df_new =  pd.read_csv(data_path + 'movement_100_iter_results_uncorr15set.csv')\n",
        "\n",
        "moral_df_new.columns = ['null', 'item', 'low_moral', 'mean_moral', 'high_moral']\n",
        "hedonic_df_new.columns = ['null', 'item', 'low_hedonic', 'mean_hedonic', 'high_hedonic']\n",
        "movement_df_new.columns = ['null', 'item', 'low_movement', 'mean_movement', 'high_movement']\n",
        "\n",
        "moral_df_new = moral_df_new.drop(columns=['null'])\n",
        "hedonic_df_new = hedonic_df_new.drop(columns=['null'])\n",
        "movement_df_new = movement_df_new.drop(columns=['null'])\n",
        "\n",
        "full_df = pd.merge(moral_df_new, hedonic_df_new, on='item', how='inner')\n",
        "full_df = pd.merge(full_df, movement_df_new, on='item', how='inner')"
      ],
      "metadata": {
        "id": "EymW-HuiGykh"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print('Prompting Correlations:')\n",
        "print('r(moral, hedonic) = '+ str(round(np.corrcoef(full_df[\"mean_moral\"], full_df[\"mean_hedonic\"])[0, 1], 2)))\n",
        "print('r(moral, movement) = '+ str(round(np.corrcoef(full_df[\"mean_moral\"], full_df[\"mean_movement\"])[0, 1], 2)))\n",
        "print('r(hedonic, movement) = '+ str(round(np.corrcoef(full_df[\"mean_hedonic\"], full_df[\"mean_movement\"])[0, 1], 2)))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GmX0B30XSLwG",
        "outputId": "57966185-d2fd-462d-89ea-a7edefb905ed"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Prompting Correlations:\n",
            "r(moral, hedonic) = -0.06\n",
            "r(moral, movement) = 0.49\n",
            "r(hedonic, movement) = 0.03\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print('Prompting vs Embedding Correlations:')\n",
        "print('moral r = '+ str(round(np.corrcoef(full_df[\"mean_moral\"], item_projections[\"moral_v\"])[0, 1], 2)))\n",
        "print('hedonic r = '+ str(round(np.corrcoef(full_df[\"mean_hedonic\"], item_projections[\"hedonic_v\"])[0, 1], 2)))\n",
        "print('movement r = '+ str(round(np.corrcoef(full_df[\"mean_movement\"], item_projections[\"movement_v\"])[0, 1], 2)))"
      ],
      "metadata": {
        "id": "s8hOy2z5CHgm"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Prompting Plots"
      ],
      "metadata": {
        "id": "oIcDbgB2_jhH"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "full_df = full_df[full_df['item'].isin(action_list_all)] # filter to the 15 less correlated items\n",
        "full_df = full_df.sort_values('mean_moral', ascending=False)\n",
        "\n",
        "fig, ax = plt.subplots(figsize=(10, 10))\n",
        "\n",
        "plot_i = 1\n",
        "for i in range(len(full_df)):\n",
        "\n",
        "    # X axis, means with 95% CI low and high values\n",
        "    x = full_df['mean_hedonic'].iloc[i]\n",
        "    x_l = full_df['mean_hedonic'].iloc[i] - full_df['low_hedonic'].iloc[i]\n",
        "    x_h = full_df['high_hedonic'].iloc[i] - full_df['mean_hedonic'].iloc[i]\n",
        "\n",
        "    # Y axis, means with 95% CI low and high values\n",
        "    y = full_df['mean_moral'].iloc[i]\n",
        "    y_l = full_df['mean_moral'].iloc[i] - full_df['low_moral'].iloc[i]\n",
        "    y_h = full_df['high_moral'].iloc[i] - full_df['mean_moral'].iloc[i]\n",
        "\n",
        "    l = str(plot_i) + '. ' + full_df['item'].iloc[i].lstrip('[').rstrip(']')\n",
        "    ax.errorbar(x, y, xerr=[[x_l], [x_h]], yerr=[[y_l], [y_h]], label=l, fmt='o', ecolor='lightgray', elinewidth=2)\n",
        "    ax.annotate(plot_i, (x+.3, y+.5))\n",
        "\n",
        "    plot_i += 1\n",
        "\n",
        "ax.set_xlabel('Hedonic')\n",
        "ax.set_ylabel('Moral')\n",
        "ax.legend()"
      ],
      "metadata": {
        "id": "zYZj8outAinr",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 867
        },
        "outputId": "cc987b87-7823-498b-d10e-86885b444a6b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7b9d40a02260>"
            ]
          },
          "metadata": {},
          "execution_count": 10
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x1000 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1IAAANBCAYAAADqZI8yAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOzdd3gU1f7H8c9mF9ITDAspdCGQUKWpgAoqUkQEr1cUUEGxXEQRAUG8FhQQsNC83Fh/gBSxUMQGAldQEWkapIRqKEoILCUhlMBuzu8PLntZCCEbkmzK+/U8+8DMnDnzndnZyX73nDljMcYYAQAAAAByzc/XAQAAAABAcUMiBQAAAABeIpECAAAAAC+RSAEAAACAl0ikAAAAAMBLJFIAAAAA4CUSKQAAAADwEokUAAAAAHjJ5usAioKsrCzt27dPoaGhslgsvg4HAAAAgI8YY3Ts2DHFxMTIz+/S7U4kUpL27dunKlWq+DoMAAAAAEXE3r17Vbly5UsuJ5GSFBoaKunswQoLC/NxNAAAAAB8JT09XVWqVHHnCJdCIiW5u/OFhYWRSAEAAAC47C0/DDYBAAAAAF4ikQIAAAAAL5FIAQAAAICXSKQAAAAAwEskUgAAAADgJRIpAAAAAPASiRQAAAAAeIlECgAAAAC8RCIFAAAAAF4ikQIAAAAAL5FIAQAAAICXSKQAAAAAwEskUgAAAADgJRIpAAAAAPASiRQAAAAAeIlECgAAAAC8RCIFAAAAAF4ikQIAAAAAL5FIAQAAAICXSKQAAAAAwEskUgAAAADgJRIpAAAAAPCSTxOpH374QZ07d1ZMTIwsFovmz5/vsdwYo5deeknR0dEKDAxU27ZttX37do8yhw8fVs+ePRUWFqZy5cqpT58+ysjIKMS9AAAAAFDa+DSROn78uBo1aqTJkydnu/z111/XpEmT9M4772jVqlUKDg5W+/btderUKXeZnj17atOmTVq8eLG++uor/fDDD3rssccKaxcAAAAAlEIWY4zxdRCSZLFYNG/ePHXt2lXS2daomJgYDRo0SIMHD5YkpaWlKTIyUlOnTtV9992npKQk1a1bV2vWrFGzZs0kSQsXLtTtt9+uP//8UzExMbnadnp6usLDw5WWlqawsLAC2T8AAAAARV9uc4Mie49UcnKy9u/fr7Zt27rnhYeH67rrrtPKlSslSStXrlS5cuXcSZQktW3bVn5+flq1atUl687MzFR6errHCwAAAAByq8gmUvv375ckRUZGesyPjIx0L9u/f78qVqzosdxmsykiIsJdJjujR49WeHi4+1WlSpV8jh4AAABASVZkE6mCNGzYMKWlpblfe/fu9XVIAAAAAIqRIptIRUVFSZJSU1M95qemprqXRUVF6cCBAx7LnU6nDh8+7C6THX9/f4WFhXm8AAAAACC3imwiVaNGDUVFRWnp0qXueenp6Vq1apVatGghSWrRooWOHj2qdevWucv85z//UVZWlq677rpCjxkAAABA6WDz5cYzMjK0Y8cO93RycrISExMVERGhqlWrasCAARo5cqRiY2NVo0YNvfjii4qJiXGP7BcfH68OHTro0Ucf1TvvvKMzZ87oySef1H333ZfrEfsAAAAAwFs+TaTWrl2rm2++2T09cOBASVKvXr00depUDRkyRMePH9djjz2mo0eP6oYbbtDChQsVEBDgXmfmzJl68skndeutt8rPz0933323Jk2aVOj7AgAAAKD0KDLPkfIlniMFAADwPw6HQy6XS1arVXa73dfhAIUqt7mBT1ukAAAAUPQ4HA45nU7ZbDYSKeASiuxgEwAAAABQVJFIAQAAAICXSKQAAAAAwEskUgAAAADgJRIpAAAAAPASiRQAAAAAeIlECgAAAAC8RCIFAAAAAF4ikQIAAAAAL5FIAQAAAICXSKQAAAAAwEskUgAAAADgJRIpAAAAAPASiRQAAABy5YcfflDnzp0VExMji8Wi+fPn+zokwGdIpAAAAJArx48fV6NGjTR58mRfhwL4nM3XAQAAAKB46Nixozp27OjrMIAigRYpAAAAAPASiRQAAAAAeImufQAAAChQDodDLpdLVqtVdrvd1+EA+YJECgAAAAXK4XDI6XTKZrORSKHEoGsfAAAAAHiJFikAAADkSkZGhnbs2OGeTk5OVmJioiIiIlS1alUfRgYUPhIpAAAA5MratWt18803u6cHDhwoSerVq5emTp3qo6gA3yCRAgAAgJsry2h9ykk5Mk7LHlJWsbWNrH4WSVKbNm1kjPFxhEDRQCIFAAAASdLCjSl65cvNSkk75Z434ZfDerlzXXWoH+3DyICih8EmAAAAoIUbU9R3xq8eSZQk7U87pb4zftXCjSk+igwomkikAAAASjlXltErX25Wdp32zs175cvNcmXRrQ84h659AAAAPuBwOORwOHwdhiRpfcrJi1qizmckpaSd0ufLE9UoOtDr+p1Op/vfLVu25DVMN7vdzvOo4HMkUgAAAD7gcrncCYavOTJO57qc01nmiraVH/vscrmuuA7gSpFIAQAA+IDVapXNVjS+itlDyua6XF5iPj95yo99tlqtV1wHcKWKxqcXAACglClK3dNiaxtN+OWw9qedyvY+KYukqPAA/b31Ne6h0L2xZcsWOZ1O2Ww2xcXFXXG8QFHAYBMAAAClnNXPopc715V0Nmk637nplzvXzVMSBZRUJFIAAABQh/rRSri/iaLCAzzmR4UHKOH+JjxHCrgAXfsAAAAg6WwydVvdKH2+PFGOjNOyh5TNc3c+oKQjkQIAAICb1c+iRtGBcjrLyGazkUQBl0DXPgAAAADwEokUAAAAAHiJRAoAAAAAvEQiBQAAAABeIpECAAAAAC+RSAEAAACAl0ikAAAAAMBLJFIAAAAA4CUSKQAAAADwEokUAAAAAHiJRAoAAAAAvEQiBQAAAABeIpECAAAAAC+RSAEAAACAl0ikAAAAAMBLJFIAAAAA4CUSKQAAAADwEokUAAAAAHiJRAoAAAAAvEQiBQAAAABeIpECAAAAAC+RSAEAAACAl2y+DgAAAABFi91ul8vlktVq9XUoQJFFIgUAAAAPdrvd1yEARR5d+wAAAADASyRSAAAAAOAlEikAAAAA8BKJFAAAAAB4iUQKAAAAALxEIgUAAAAAXiKRAgAAAAAvkUgBAAAAgJdIpAAAAADASyRSAAAAAOAlEikAAAAA8BKJFAAAAAB4iUQKAAAAALxEIgUAAAAAXiKRAgAAAAAvkUgBAADAJ0aPHq3mzZsrNDRUFStWVNeuXbV161ZfhwXkCokUAAAAfGL58uXq16+ffvnlFy1evFhnzpxRu3btdPz4cV+HBlyWzdcBAAAAoHRauHChx/TUqVNVsWJFrVu3TjfddJOPogJyhxYpAAAAFAlpaWmSpIiICB9HAlweiRQAAAB8LisrSwMGDFCrVq1Uv359X4cDXFaRT6SOHTumAQMGqFq1agoMDFTLli21Zs0a93JjjF566SVFR0crMDBQbdu21fbt230YMQAAALzVr18/bdy4UbNnz/Z1KECuFPlE6pFHHtHixYs1ffp0bdiwQe3atVPbtm31119/SZJef/11TZo0Se+8845WrVql4OBgtW/fXqdOnfJx5AAAAMiNJ598Ul999ZW+//57Va5c2dfhALliMcYYXwdxKSdPnlRoaKi++OILderUyT2/adOm6tixo0aMGKGYmBgNGjRIgwcPlnS2b21kZKSmTp2q++67L1fbSU9PV3h4uNLS0hQWFlYg+wIAAFBabdmyRU6nUzabTXFxce75xhg99dRTmjdvnpYtW6bY2FgfRgmcldvcoEi3SDmdTrlcLgUEBHjMDwwM1E8//aTk5GTt379fbdu2dS8LDw/Xddddp5UrV16y3szMTKWnp3u8AAAAULj69eunGTNmaNasWQoNDdX+/fu1f/9+nTx50tehAZdVpBOp0NBQtWjRQiNGjNC+ffvkcrk0Y8YMrVy5UikpKdq/f78kKTIy0mO9yMhI97LsjB49WuHh4e5XlSpVCnQ/AAAAcLGEhASlpaWpTZs2io6Odr8++eQTX4cGXFaRTqQkafr06TLGqFKlSvL399ekSZPUvXt3+fnlPfRhw4YpLS3N/dq7d28+RgwAAAC3LJeCUtcpfM9iBaWuk7Jc7kXGmGxfvXv39l28QC4V+Qfy1qxZU8uXL9fx48eVnp6u6Oho3Xvvvbr66qsVFRUlSUpNTVV0dLR7ndTUVF1zzTWXrNPf31/+/v4FHToAAEDptnmBtHCoqqbv+9+8dSOlDmOlunf6Li4gHxT5FqlzgoODFR0drSNHjmjRokXq0qWLatSooaioKC1dutRdLj09XatWrVKLFi18GC0AAEApt3mB9OmD0vlJlCSlp5ydv3mBb+IC8kmRb5FatGiRjDGqU6eOduzYoWeffVZxcXF66KGHZLFYNGDAAI0cOVKxsbGqUaOGXnzxRcXExKhr166+Dh0AAKB0ynJJC4dKym5waCPJIi18TorrJPlZCzk4IH8U+UQqLS1Nw4YN059//qmIiAjdfffdGjVqlMqUKSNJGjJkiI4fP67HHntMR48e1Q033KCFCxdeNNIfAABAceRwOORwOHwdhleCUtd5due7iJHS/9KeH2frRGTTQosrP9jtdtntdl+HgSKgSD9HqrDwHCkAAFBUpaam6uDBg74Owyvhexaryurhly2399rhSqt6W8EHlI8qVKhw0YjRKFlymxsU+RYpAACA0sxqtcpmK15f2UxwxVyXK277ZrXSFRFnFa8zFwAAoJQpll3JaseeHZ0vPUXZ3ydlkcJiVPXG+7hHCsVWsRm1DwAAAMWEn/XsEOeSJMsFC/873WEMSRSKNRIpAAAA5L+6d0rdPpLCoj3nh8Wcnc9zpFDM0bUPAAAABaPunVJcJ+35cbYsxw/IBFekOx9KDBIpAAAAFBw/q05ENpXT6Tw7sARJFEoIuvYBAAAAgJdIpAAAAADASyRSAAAAAOAlEikAAAAA8BKJFAAAAAB4iUQKAAAAALxEIgUAAAAAXiKRAgAAAAAv8UBeAAB8zOFwyOVyyWq1ym63+zocAEAukEgBAOBjDodDTqdTNpuNRAoAigm69gEAAACAl0ikAAAAAMBLJFIAAAAA4CUSKQAAAADwEokUAAAAAHiJRAoAAAAAvEQiBQAAAABeIpECAAAAAC+RSAEAAACAl0ikAAAAAMBLNl8HAABASeJwOORyuWS1WmW3230dDgCggJBIAQCQjxwOh5xOp2w2G4kUAJRgdO0DAAAAAC+RSAEAAACAl0ikAAAAAMBLJFIAAAAA4CUSKQAAAADwEokUAAAAAHiJRAoAgEKQkJCghg0bKiwsTGFhYWrRooW+/fZbX4cFAMgjniMFAKUED4r1rcqVK2vMmDGKjY2VMUbTpk1Tly5d9Ntvv8lqtfo6PACAl0ikAKCU4EGxvtW5c2eP6VGjRikhIUG//PKLWrVq5aOoAAB5RSIFAEAhc7lc+uyzz3T8+HG1aNHC1+EAAPKARAoAgEKyYcMGtWjRQqdOnVJISIjmzZununXrasuWLb4ODQDgJRIpAAAKSZ06dZSYmKi0tDR9/vnn6tWrl5YvXy4/P8Z+AoDihkQKAIBCUrZsWdWqVUuS1LRpU61Zs0YTJ07UM8884+PIAADe4icwAAB8JCsrS5mZmb4OAwCQB7RIAQBQCIYNG6aOHTuqatWqOnbsmGbNmqVly5Zp0aJFvg4NAJAHJFIAAOQTV5ZLm9I3yXHKIXuAXbFZsbL6nX1G1IEDB/Tggw8qJSVF4eHhatiwoRYtWqTbbruNwSYAoBgikQIAIB8s2b1EY1aPUeqJVPe8hN0Jeu7a59S2Wlt9+OGHPowOAJDfuEcKAIArtGT3Eg1cNtAjiZKkAycOaOCygVqye4mPIgMAFBQSKQAAroAry6Uxq8fIyFy07Ny8savHypXlKuzQAAAFiK59AIB84XA45HA4fB1GoduUvumilqjzGRntP7Ff89fOV72wetmWcTqd7n9L2v1Sdrtddrvd12EAQL4jkQIA5AuXy+VOCEoTx6ncJY+OUw45gy5/fEraMXS5aIkDUDKRSAEA8oXVapXNVvr+rNgDctfaYg+wX/L4nJ885XQM33//fY0bN04PPPCAnn/+ee8C9RGr1errEACgQJS+v3gAgAJRWrtwxWbFKmF3gg6cOJDtfVIWWRQZFKmuzbq6h0K/0JYtW+R0OmWz2RQXF5dtmTVr1mju3Llq2LChIiIiLlkOAFA4GGwCAIArYPWz6rlrn5N0Nmk637npodcOvWQSlRsZGRnq2bOn3n//fV111VV5DxYAkG9IpAAAuEJtq7XVuDbjVDGoosf8yKBIjWszTm2rtb2i+vv166dOnTqpbVvPehwOh1JTU0vlIB8A4Gt07QMAIB+0rdZWN1e5WfPXzpfjlEP2AHuO3flya/bs2fr111+1Zs2ai5Y5HA53l8DS2K0SAHyJRAoAgHxi9bOqXlg9OYPOJjdXmkTt3btXTz/9tBYvXqyAgIB8ihIAkB/o2gcAQBG1bt06HThwQE2aNJHNZpPNZtPy5cs1adIk2Ww2hhYHAB+iRQoAgCLq1ltv1YYNGzzmPfTQQ4qLi9PQoUNltVpL3HOnAKC4IJECAKCICg0NVf369T3mBQcHq3z58qpfv762bNnio8gAACRSAAD4kHG5ZDZslOWQQ6a8XSY2VhYeYgsARR6JFAAAPpL+3XdKfW20tH+/+6blHZMnK/L5YQpr1y7bdZYtW1Zo8QEALo3BJgAA8IH0777TX08PkHP/fo/5ztRU/fX0AKV/912+bIdnTQFAwSCRAgCgkBmX62xLlDHZLDw7L/W10TL5MCqfw+HQwYMHSaQAIJ/RtQ8AUOQ4HI5i+8X/3Ch6TqfzkoNBmA0bpQtaojwLGDn379eWufNkaVD/ksVys63clPElu93Ow4QBFEskUgCAIsflcpWIYb0vtQ+WQ45cdQnJOuSQyeVxyM3xKorHlGdhASiuSKQAAEWO1WqVzVY8/0Sdn6xcah9M+dy1wPiVt8uSw3HIzbZyU8aXrIxQCKCYKnpXVABAqVecu3tt2bJFTqdTNptNcXFx2ZYxsbHaMXmynKmp2d8nZbHIFhmpWn+7K8eh0HOzrdyUAQB4j8EmAAAoZBarVZHPD/vvhOWChWenI58fxvOkAKAII5ECAMAHwtq1U6WJE2SLjPSYb4uMVKWJEy75HCkAQNFA1z4AAHwkrF07hd56q7bMnaesQw75lbdftjsfAKBoIJECAMCHLFarLA3qyzidsthsJFEokex2u1wuF4OLoEQhkQIAAECBKq6DxwA5IZECACAf8cs7AJQOJFIAAOQjfnkHgNKBUfsAAAAAwEskUgAAAADgJRIpAAAAAPASiRQAAAAAeIlECgAAAAC8RCIFAAAAAF4ikQIAAAAALxXpRMrlcunFF19UjRo1FBgYqJo1a2rEiBEyxrjLGGP00ksvKTo6WoGBgWrbtq22b9/uw6gBAAAAlHRFOpEaO3asEhIS9K9//UtJSUkaO3asXn/9db399tvuMq+//romTZqkd955R6tWrVJwcLDat2+vU6dO+TByAAAkh8Oh1NRUORwOX4cCAMhnNl8HkJOff/5ZXbp0UadOnSRJ1atX18cff6zVq1dLOtsaNWHCBL3wwgvq0qWLJOmjjz5SZGSk5s+fr/vuu89nsQMA4HA45HQ6ZbPZZLfbfR0OACAfFekWqZYtW2rp0qXatm2bJGn9+vX66aef1LFjR0lScnKy9u/fr7Zt27rXCQ8P13XXXaeVK1dest7MzEylp6d7vAAAAAAgt4p0i9Rzzz2n9PR0xcXFyWq1yuVyadSoUerZs6ckaf/+/ZKkyMhIj/UiIyPdy7IzevRovfLKKwUXOAAAAIASrUi3SH366aeaOXOmZs2apV9//VXTpk3Tm2++qWnTpl1RvcOGDVNaWpr7tXfv3nyKGAAAAEBpUKRbpJ599lk999xz7nudGjRooN27d2v06NHq1auXoqKiJEmpqamKjo52r5eamqprrrnmkvX6+/vL39+/QGMHAAAAUHIV6RapEydOyM/PM0Sr1aqsrCxJUo0aNRQVFaWlS5e6l6enp2vVqlVq0aJFocYKAAAAoPQo0i1SnTt31qhRo1S1alXVq1dPv/32m8aNG6eHH35YkmSxWDRgwACNHDlSsbGxqlGjhl588UXFxMSoa9euvg0eAAAAQIlVpFuk3n77bf3973/XE088ofj4eA0ePFiPP/64RowY4S4zZMgQPfXUU3rsscfUvHlzZWRkaOHChQoICPBh5ABQvPz111+6//77Vb58eQUGBqpBgwZau3atr8NCARo+fLgsFovHKy4uztdhAUCxUaRbpEJDQzVhwgRNmDDhkmUsFoteffVVvfrqq4UXGACUIEeOHFGrVq10880369tvv1WFChW0fft2XXXVVb4ODQWsXr16WrJkiXvaZivSXwsAoEjhigkApdzYsWNVpUoVTZkyxT2vRo0aPowIhcVms7kHbgIAeKdId+0DABS8BQsWqFmzZrrnnntUsWJFNW7cWO+//76vw0Ih2L59u2JiYnT11VerZ8+e2rNnj69DAoBig0QKAEq5P/74QwkJCYqNjdWiRYvUt29f9e/f/4qf2Yei7brrrtPUqVO1cOFCJSQkKDk5WTfeeKOOHTvm69AAoFigax8AlHJZWVlq1qyZXnvtNUlS48aNtXHjRr3zzjvq1auXj6NDQenYsaP7/w0bNtR1112natWq6dNPP1WfPn18GBkAFA+0SAFAKRcdHa26det6zIuPj6ebVylTrlw51a5dWzt27PB1KABQLJBIAUAp16pVK23dutVj3rZt21StWjUfRQRfyMjI0M6dOxUdHe3rUACgWCCRAoBS7plnntEvv/yi1157TTt27NCsWbP03nvvqV+/fr4ODQVo8ODBWr58uXbt2qWff/5Zd911l6xWq7p37+7r0ACgWOAeKQAoBbKyXDqUvEMn0o4qKLycsmrHys/PKklq3ry55s2bp2HDhunVV19VjRo1NGHCBPXs2dPHUeNKZWUZHdmbqZPHzigwtIyyahv5+VkkSX/++ae6d++uQ4cOqUKFCrrhhhv0yy+/qEKFCj6OGgCKBxIpACjhtq/6Wf+Z+p4yDjvc8zYu+Fy39H5Msde1lCTdcccduuOOO3wVIgrAzt8O6MdPtuv40Uz3vK1Lf9aN98aqZuOKmj17tg+jA4Dij659AFCCbV/1sxaMe80jiZKkjMMOLRj3mrav+tlHkaEg7fztgBa+u9EjiZKk40cztfDdjdr52wEfRQYAJQeJFACUUFlZLv1n6ns5lvl+2nvKynIVUkQoDFlZRj9+sj3HMj99ul1ZWaaQIgKAkomufQBKPYfDIYfDcfmCxcyh5B0XtURd6Nghh1Yu+lbla9QqpKgKl91ul91u92qd/DwfnE6n+98tW7ZccbncrHdkb+ZFLVEXyjiSqVVLf9dVVfxzva2iJi/vLQDkJxIpAKWey+VyfyEtSU6kHc11ufASuP/S2fc2L+sUxPmQ2zrzuu1z6508diZX5U8eO6NQpzVP2yoK8vLeAkB+IpECUOpZrVbZbCXvchgUXi7X5Uri/ktn39u8rJNfx+P8pCinOnNbLjfrBYaWydW6gaFlivX7npf3FgDyU/G9ggJAPimpXYSyasdq44LPc+zeF1rerhbtO7qHQkf+ng9btmyR0+mUzWZTXFzcFZfLzXpZtY22Lv05x+59IVf567pbG7qHQgcAeI/BJgCghPLzs+qW3o/lWObmXo+RRJUwfn4W3XhvbI5lbugWSxIFAFeIRAoASrDY61rqzoHPKyTCs4UltLxddw583v0cKZQsNRtXVIfH6yu4nOdgEiFX+avD4/VVs3FFH0UGACUHXfsAoISLva6laja/TisXfasTaUcVFF6O7nylQM3GFVWjUQWtWvq7Th47o8DQMnTnA4B8RCIFAKWAn59V5WvUUvh/76chiSod/PwsuqqKv0Kd1v++7yRRAJBfSKQAlFoOh0Mul0tWq7VEDjYBAAAKDokUgFLL4XC4RzwjkQIAAN4gkQIAwMfsdru7dRQAUDyQSAEA4GO0iAJA8cPw5wAAAADgJRIpAAAAAPASiRQAAAAAeIlECgAAAAC8RCIFAAAAAF4ikQIAAAAAL5FIAQAAAICXeI4UUMQ4HA73gzl5tgwAAEDRRCIFFDEOh0NOp1M2m41ECgAAoIiiax8AAIWsevXqslgsF7369evn69AAALlEixQAAIVszZo1crlc7umNGzfqtttu0z333OPDqAAA3iCRAgCgkFWoUMFjesyYMapZs6Zat27to4gAAN6iax8AAD50+vRpzZgxQw8//LAsFouvwwEA5BKJFAAAPjR//nwdPXpUvXv39nUoAAAvkEgBAOBDH374oTp27KiYmBhfhwIA8AL3SAEA4CO7d+/WkiVLNHfuXF+HAgDwEi1SAAD4yJQpU1SxYkV16tTJ16EAALxEIgUAgA9kZWVpypQp6tWrl2w2OogAQHHDlRsAgAJgsoysKadlyXDKLyRLpraRxe9/o/ItWbJEe/bs0cMPP+zDKAEAeUUiBQBAPju50aGjX+5UYNpp97z9v6xWuc41FVjfLklq166djDG+ChEAcIXo2gcAQD46udGhQzOS5DoviZIkV9ppHZqRpJMbHT6KDACQn0ikAADIJybL6OiXO3Msc/TLP2SyaIkCgOKOrn0AUIo5HA45HLSQ5BdrymmP7nzZcaVlasfyDXJFl73i7TmdTve/W7ZsyXOZksBut8tut/s6DAClCIkUAJRiLpfL/UUbV86SkbtjmZXhlNOZv51CcvM+luT32uVy+ToEAKUMiRQAlGJWq5Wht/ORX0hWLsvZ8uW4n58YXaq+3JQpCaxWq69DAFDKlNwrKgDgsugOlb9MbaP9v6y+aKCJ81nD/VWrdQOPodDzasuWLXI6nbLZbIqLi8tzGQCA9xhsAgCAfGLxs6hc55o5linX+ep8SaIAAL5FIgUAQD4KrG9X+fvjZQ33HEzCGu6v8vfHu58jBQAo3ujaBwBAPgusb1dA3fLasXyDsjKc8gux5Vt3PgBA0UAiBQBAAbD4WeSKLiun0082m40kCgBKGLr2AQAAAICXaJECAKAEs9vtcrlcDA8OAPmMRAoAgBKM4e0BoGDQtQ8AAAAAvEQiBQAAAABeIpECAAAAAC+RSAEAAACAl0ikAAAAAMBLJFIAAAAA4CUSKQAAAADwEokUAAAAAHiJRAoAAAAAvEQiBQAAAABeIpECAAAAAC+RSAEAAACAl0ikAOA8LpdLL774omrUqKHAwEDVrFlTI0aMkDHG16EBAIAixObrAACgKBk7dqwSEhI0bdo01atXT2vXrtVDDz2k8PBw9e/f39fhAQCAIoJECgDO8/PPP6tLly7q1KmTJKl69er6+OOPtXr1ah9HBgAAihK69gHAeVq2bKmlS5dq27ZtkqT169frp59+UseOHX0cGQAAKEpokQKA8zz33HNKT09XXFycrFarXC6XRo0apZ49e/o6NAAAUISQSAHAeT799FPNnDlTs2bNUr169ZSYmKgBAwYoJiZGvXr18nV4AACgiCCRAoDzPPvss3ruued03333SZIaNGig3bt3a/To0SRSAADAjXukAOA8J06ckJ+f56XRarUqKyvLRxEBAICiiBYpADhP586dNWrUKFWtWlX16tXTb7/9pnHjxunhhx/2dWgAAKAIIZECgPO8/fbbevHFF/XEE0/owIEDiomJ0eOPP66XXnrJ16EBAIAihEQKQKmUlZWl1NRUHT9+XMHBwapdu7b8/PwUGhqqCRMmaMKECb4OEQAAFGEkUgBKnc2bN2vhwoVKT093z1u3bp06dOigunXr+jAyAABQXBT5wSaqV68ui8Vy0atfv36SpFOnTqlfv34qX768QkJCdPfddys1NdXHUQMoqjZv3qxPP/3UI4mSpPT0dH366afavHmzjyIDAADFSZFPpNasWaOUlBT3a/HixZKke+65R5L0zDPP6Msvv9Rnn32m5cuXa9++ffrb3/7my5ABFFFZWVlauHBhjmUWLlxYYkfos9vtqlChgux2u69DAQCg2CvyXfsqVKjgMT1mzBjVrFlTrVu3Vlpamj788EPNmjVLt9xyiyRpypQpio+P1y+//KLrr7/eFyED+crhcMjhcGjNmjX6v//7P23atEkHDx7U22+/rbZt27rLGWP09ttv67PPPtOxY8fUuHFjvfzyy6pevbrvgi9iUlNTL2qJulB6erp+/PFHRUZGFlJUvuFwOAplO3a7ncTNB44dO6YXX3xR8+bN04EDB9S4cWNNnDhRzZs393VoAFBiFPlE6nynT5/WjBkzNHDgQFksFq1bt05nzpzx+DIZFxenqlWrauXKlSRSKBFcLpecTqcyMjIUGxurrl27asCAAe7553z44YeaMWOGRo4cqUqVKulf//qXHnnkEX3xxRfy9/f34R4UHcePH891ufOPLfLO5XL5OoRS6ZFHHtHGjRs1ffp0xcTEaMaMGWrbtq02b96sSpUq+To8ACgRilUiNX/+fB09elS9e/eWJO3fv19ly5ZVuXLlPMpFRkZq//79l6wnMzNTmZmZ7unL/UIN+JLVapXNZtPNN9+sm2+++aL50tnWqBkzZugf//iH2rVrJ0l6/fXXdcMNN2jZsmXq1KmTT2IvaoKDg3Nd7tyxxZWxWq2+DqHUOXnypObMmaMvvvhCN910kyRp+PDh+vLLL5WQkKCRI0f6OEIAKBmK1TeFDz/8UB07dlRMTMwV1TN69Gi98sor+RQVULAu1TWqcuXKiouLkyT98ccfcjgc6tGjh3ueJF1//fXas2ePx7zSrHbt2lq3bl2OP56EhYXpxhtvlJ9fkb+FFMiW0+mUy+VSQECAx/zAwED99NNPPooKAEqeYvNNYffu3VqyZIkeeeQR97yoqCidPn1aR48e9SibmpqqqKioS9Y1bNgwpaWluV979+4tqLCBQnGuBfbC+3ou1zpb2vj5+alDhw45lunQoQNJFIq10NBQtWjRQiNGjNC+ffvkcrk0Y8YMrVy5UikpKb4ODwBKjGLzbWHKlCmqWLGiRxelpk2bqkyZMlq6dKl73tatW7Vnzx61aNHiknX5+/srLCzM4wWgdKhbt666det20ec+LCxM3bp14zlSKBGmT58uY4wqVaokf39/TZo0Sd27d+dHAgDIR8Wia19WVpamTJmiXr16edy3EB4erj59+mjgwIGKiIhQWFiYnnrqKbVo0YKBJlCqnGuBTU1NVXR0tHt+amqqrrnmGh9FVXTVrVtXcXFx+vHHH3X8+HEFBwfTnQ8lSs2aNbV8+XIdP35c6enpio6O1r333qurr77a16EBQIlRLL41LFmyRHv27NHDDz980bLx48frjjvu0N13362bbrpJUVFRmjt3rg+iBHynRo0aioqK8midTU9P16pVq3JsnS3N/Pz8FBkZqapVqyoyMpIkCiVScHCwoqOjdeTIES1atEhdunTxdUgAUGIUixapdu3ayRiT7bKAgABNnjxZkydPLuSogMKVkZGhHTt2uKeTk5OVmJioiIgIVa1aVQMGDNDIkSMVGxurGjVq6MUXX1RMTIy6du3qu6AB+MSiRYtkjFGdOnW0Y8cOPfvss4qLi9NDDz3k69AAoMQoFokUUFq4jNFvmS4dPJOlCmVcijVGVotFkrR27VqP4c8HDhwoSerVq5emTp2qIUOG6Pjx43rsscd09OhR3XDDDVq4cOFFI3cBKBmMcSkz83edPuNQ2TJ2GRMri+XscPNpaWkaNmyY/vzzT0VEROjuu+/WqFGjVKZMGR9HDQAlh8VcqqmnFElPT1d4eLjS0tIYeAI+8/XBo3ph+19KyTzjnhftX0YjYyupU4VyvgusBNuyZYucTqdsNhtDxKNAFNQ5duDAIm3b/qoyM/83Kqe/f5Rqx76kihXb59t2AKA0ym1uwE0BQBHw9cGjemTjLo8kSpL2Z57RIxt36euDR30TGIAi58CBRdqwsZ9HEiVJmZmp2rCxnw4cWOSjyACgdCGRAnzMZYxe2P6XsmsaPjfvxe1/yUXjMVDqGePStu2vSjlcMbZtHyFjXIUaFwCURtwjBZ9xOBxyOBy+DsPnfst0XdQSdT4jaV/mGX36+2Y19rcWXmBFlN1ul91u93UYpRafW+84nU73v1u2bLni+jIzf7+oJcqTUWZmin7/fY78/Rte8fbgPa5RQOlBIgWfcblc7i8ZpdnBM1m5LOeS00qrlMvFL+2+xOc27/LjuJ0+k7sk9vQZh6xW3idf4BoFlB4kUvAZq9Xq8YDl0qpCGZekyydTFcpYZbPRImW1cgx8ic+td85PnvLjuJUtk7uWjrJl7LxPPsI1Cig9uMrCZ+j+cFasMXp95WbtzzyT7V0PFp0dva9bw7ruodABX+Fz6538HrXPmFit+HmiMjNTlf19Uhb5+0epYcO73UOhAwAKBoNNAD5mtVg0MraSpLNJ0/nOTY+IrUQSBUAWi1W1Y186N3XhUklS7dgXSaIAoBCQSAFFQKcK5fRB/eqK8vd8WGa0fxl9UL86z5EC4FaxYns1qD9Z/v6RHvP9/aPUoP5kniMFAIWErn1AEdGpQjl1sIfr09836+AZlyqUsdKdD0C2KlZsrwoV2ur33+fo9BmHypax050PAAoZiRRQhFgtFjX2t8ppNbLZrCRRQDFnt9vlcrkKZAACi8Uqf/+GslrP3oNFEgUAhYtECgCAAsLAHABQcnGPFAAAAAB4iUQKAAAAALxEIgUAAAAAXiKRAgAAAAAvkUgBAAAAgJdIpAAAAADASyRSAAAAAOAlEikAAAAA8BKJFAAAAAB4iUQKAAAAALxEIgUAAAAAXiKRAgAAAAAvkUgBAAAAgJdIpAAAAADASyRSAAAAAOAlEikAAAAA8BKJFAAAAAB4iUQKAAAAALxEIgUAAAAAXiKRAgAAAAAvkUgBAAAAgJdsvg4AwKU5HA65XC5ZrVbZ7XZfhwMAAID/IpECijCHwyGn0ymbzUYiBQAAUITQtQ8AAAAAvEQiBQAAAABeIpECAAAAAC+RSAEAAACAl0ikAAAAAMBLJFIAAAAA4CWGPwcAAACKKJ4pWXSRSAEAAABFFM+ULLro2gcAAAAAXiKRAgAAAAAvkUgBAAAAgJdIpAAAAADASyRSAAAAAOAlEikAAAAA8BKJFAAAAAB4iUQKAAAAALxEIgUAAAAAXiKRAgAAAAAv2XwdAAAApZXD4ZDL5ZLVapXdbvd1OAAAL5BIASi17Ha7+0ss4AsOh0NOp1M2m41ECgCKGRIpAKUWX1wBAEBecY8UAAAAAHiJFikAAFBouC8MQElBIgUAAAoN94UBKClynUhNmjQp15X2798/T8EAAAAAQHGQ60Rq/PjxuSpnsVhIpAAAAACUaLlOpJKTkwsyDgAAAAAoNhi1DwAAAAC8lOfBJv78808tWLBAe/bs0enTpz2WjRs37ooDAwAAAICiKk+J1NKlS3XnnXfq6quv1pYtW1S/fn3t2rVLxhg1adIkv2MEAAAAgCIlT137hg0bpsGDB2vDhg0KCAjQnDlztHfvXrVu3Vr33HNPfscIAAAAAEVKnhKppKQkPfjgg5Ikm82mkydPKiQkRK+++qrGjh2brwECAAAAQFGTp0QqODjYfV9UdHS0du7c6V7mcDjyJzIAAAAAKKLydI/U9ddfr59++knx8fG6/fbbNWjQIG3YsEFz587V9ddfn98xAgAAAECRkqdEaty4ccrIyJAkvfLKK8rIyNAnn3yi2NhYRuwDAAAAUOJ5nUi5XC79+eefatiwoaSz3fzeeeedfA8MAAAAAIoqr++RslqtateunY4cOVIQ8QClnt1uV4UKFWS3230dCgAAAC4hT4NN1K9fX3/88Ud+xwJAZxOpyMhIEikAAIAiLE+J1MiRIzV48GB99dVXSklJUXp6uscLAAAAAEqyPA02cfvtt0uS7rzzTlksFvd8Y4wsFotcLlf+RAcAAAAARVCeEqnvv/8+v+MAAAAAgGIjT4lU69at8zsOAAAAACg28pRISdLRo0f14YcfKikpSZJUr149PfzwwwoPD8+34AAAAACgKMrTYBNr165VzZo1NX78eB0+fFiHDx/WuHHjVLNmTf3666/5HSMAAAAAFCl5SqSeeeYZ3Xnnndq1a5fmzp2ruXPnKjk5WXfccYcGDBiQrwH+9ddfuv/++1W+fHkFBgaqQYMGWrt2rXu5MUYvvfSSoqOjFRgYqLZt22r79u35GgMAAAAAnC/PLVJDhw6Vzfa/noE2m01DhgzxSHKu1JEjR9SqVSuVKVNG3377rTZv3qy33npLV111lbvM66+/rkmTJumdd97RqlWrFBwcrPbt2+vUqVP5FgcAAAAAnC9P90iFhYVpz549iouL85i/d+9ehYaG5ktgkjR27FhVqVJFU6ZMcc+rUaOG+//GGE2YMEEvvPCCunTpIkn66KOPFBkZqfnz5+u+++7Lt1gAAAAA4Jw8tUjde++96tOnjz755BPt3btXe/fu1ezZs/XII4+oe/fu+RbcggUL1KxZM91zzz2qWLGiGjdurPfff9+9PDk5Wfv371fbtm3d88LDw3Xddddp5cqV+RYHAAAAAJwvTy1Sb775piwWix588EE5nU4ZY1S2bFn17dtXY8aMybfg/vjjDyUkJGjgwIF6/vnntWbNGvXv319ly5ZVr169tH//fklSZGSkx3qRkZHuZdnJzMxUZmamezo9PT3fYgYAAABQ8uUpkSpbtqwmTpyo0aNHa+fOnZKkmjVrKigoKF+Dy8rKUrNmzfTaa69Jkho3bqyNGzfqnXfeUa9evfJc7+jRo/XKK6/kV5gAAAAAShmvEqmHH344V+X+7//+L0/BXCg6Olp169b1mBcfH685c+ZIkqKioiRJqampio6OdpdJTU3VNddcc8l6hw0bpoEDB7qn09PTVaVKlXyJGQAAAEDJ51UiNXXqVFWrVk2NGzeWMaagYnJr1aqVtm7d6jFv27ZtqlatmqSzA09ERUVp6dKl7sQpPT1dq1atUt++fS9Zr7+/v/z9/QssbgAAAAAlm1eJVN++ffXxxx8rOTlZDz30kO6//35FREQUVGx65pln1LJlS7322mvq1q2bVq9erffee0/vvfeeJMlisWjAgAEaOXKkYmNjVaNGDb344ouKiYlR165dCywuAAAAAKWbV6P2TZ48WSkpKRoyZIi+/PJLValSRd26ddOiRYsKpIWqefPmmjdvnj7++GPVr19fI0aM0IQJE9SzZ093mSFDhuipp57SY489pubNmysjI0MLFy5UQEBAvscDAAAAAJJkMVeQAe3evVtTp07VRx99JKfTqU2bNikkJCQ/4ysU6enpCg8PV1pamsLCwnwdDuC2ZcsWOZ1O2Wy2i57bBqD4u9LPeHG8RhTHmAFf4jNT+HKbG+TpOVLulf38ZLFYZIyRy+W6kqoAAAAAoNjwOpHKzMzUxx9/rNtuu021a9fWhg0b9K9//Ut79uwplq1RAAAAAOAtrwabeOKJJzR79mxVqVJFDz/8sD7++GPZ7faCig0AAAAAiiSvEql33nlHVatW1dVXX63ly5dr+fLl2ZabO3duvgQHAAAAAEWRV4nUgw8+KIvFUlCxAAAAAECx4PUDeQEAAACgtLuiUfsAAAAAoDQikQIAAAAAL5FIAQAAAICXSKQAAAAAwEskUgAAAADgJRIpAAAAAPASiRQAAAAAeIlECgAAAAC8RCIFAAAAAF4ikQIAAAAAL5FIAQAAAICXSKQAAAAAwEskUkAx9MMPP6hz586KiYmRxWLR/PnzPZbPnTtX7dq1U/ny5WWxWJSYmOiTOAEAAEoqEimgGDp+/LgaNWqkyZMnX3L5DTfcoLFjxxZyZAAAAKWDzdcBAPBex44d1bFjx0suf+CBByRJu3btKqSIAAAAShdapAAAAADASyRSAAAUMTndB3nmzBkNHTpUDRo0UJMmTXTLLbdo6NCh2rdvn+8CBoBSiEQKAIAiJqf7IE+cOKFff/1VL774oubMmaPx48dr165duvPOO30QKQCUXtwjBQBAEZPTfZDh4eFavHixJGnLli2qUqWKXnjhBXXr1k179uxR1apVCzNUACi1aJECAKCYO3bsmCwWi8qVK+frUACg1KBFCiiGMjIytGPHDvd0cnKyEhMTFRERoapVq+rw4cPas2eP+56JrVu3SpKioqIUFRXlk5gBFIzMzEy99dZb6t69u8LCwnwdDgCUGrRIAUWUK8tofcpJLU8+rvUpJ+XKMu5la9euVePGjdW4cWNJ0sCBA9W4cWO99NJLkqQFCxaocePG6tSpkyTpvvvuU+PGjfXOO+8U/o4AKDBnzpzR4MGDZYxRQkKCr8MBgFKFFimgCFq4MUWvfLlZKWmn3PMm/HJYL3euqw71o9WmTRsZYy65fu/evdW7d+9CiBSAr5w5c0bPPPOM9u3bp6lTp9IaBQCFjBYpoIhZuDFFfWf86pFESdL+tFPqO+NXLdyY4qPIABQVZ86cUbdu3bR79269//77uuqqq3wdEgCUOrRIAUWIK8volS83K7u2JiPJIumVLzfrtrpRsvpZCjk6AIUlp/sgo6Oj9fe//12//vqr3n77bWVlZengwYMqV66cIiIiVLZsWR9GDgClB4lUKeJwOORwOHwdBnKwPuXkRS1R5zOSUtJO6fPliWoUHVh4gcFrdrtddrvd12GgKMtyKSh1nSzHD8gEV5Rqx0p+Vkln74O8+eab3UUHDhwoSerVq5eGDx+uBQsWSJLuuusujyq///57tWnTpnDiB4BSjkSqFHG5XHI6nb4OAzlwZJzOdTmns0wBR4Mr4XK5fB0CirLNC6SFQ1U1fd//5q0bKXUYK9W987L3QZ5btmXLFjmdTtlsNsXFxRV01ACA85BIlSJWq1U2G295UWYPyV2XHHtIWd7LIs5qtfo6BBRVmxdInz4oXdiJNz3l7PxuH0l17/RJaACA3OObWClCV6OiL7a20YRfDmt/2qls75OySIoKD9DfW1/DPVJAcZTlkhYO1UVJlCT3nZALn5PiOrm7+QEAiiYSKaAIsfpZ9HLnuuo741dZ5PlV61za9HLnuiRR8DnuucyboNR1nt35LmKk9L+058fZOhHZ9LL1neuu7XQ6tWXLlnyKsmAVZMz8YAigMJFIAUVMh/rRSri/yUXPkYoKD3A/RwrwNe65zBvL8QO5Luft8S2O70d+x8y9iQAKE4kUUAR1qB+t2+pG6fPliXJknJY9pCzd+VCkcM9l3pjgirkul5vje34iUlzej4KMmXsTARSm4nHVBUohq59FjaID5XSWkc1mI4lCkUIXqjyqHXt2dL70FGV/n5RFCotR1Rvvy9U9UsVx1L7iGDMAZMfP1wEAAFBq+FnPDnEu6X93PspzusMYBpoAgGKARAoAgMJU986zQ5yHXXC/Y1gMQ58DQDFC1z4AAApb3TuluE7a8+NsWY4fkAmumOvufACAooFECgAAX/Cz6kRkU/f9QiRRAFC8kEihxHG5XDpz5oyvw8gXWVlZ7tepU6cuvwKAYuVKP+PF8RpRHGMGcqNMmTKMHFnKkEihxDDGaP/+/Tp69KivQ8k3TqdTxhg5nU4lJyf7OhwA+exKP+PF8RpRHGMGcqtcuXKKioqSxcJIu6UBiRRKjHNJVMWKFRUUFFQiLmLn/1obEBDgw0gAFIQr/YwXx2tEcYwZuBxjjE6cOKEDB84+dDs6Ovoya6AkIJFCieByudxJVPny5X0dTr4x5n/PmeELB1DyXOlnvDheI4pjzEBuBAYGSpIOHDigihUr0s2vFGD4c5QI5+6JCgoK8nEkAACgtDr3PaSk3KuNnNEihRKlJHTnA4Dcstn4Mw4UJXwPKV24AgPZcDgccrlcslqtstvtvg4HALJVpkwZX4eQb44dO6b9+/frxIkTOnPmjGrWrKmrrrrKvXzfvn06fPiwTp8+LYvFoqCgIFWqVEkhISE+jBpAaUbXPiAbDodDBw8elMPh8HUoebZr1y5ZLBYlJib6OhRJUvXq1TVhwgRfh4FsFPS5YrFYNH/+/AKpGyVHVlaWgoKCVLVq1WyX+/v7q2rVqqpXr57i4uLk7++v7du304UKgM+QSAEXcGUZrU85qeXJx7U+5aRcWebyK12BH374QZ07d1ZMTEyJ+MI5depUlStX7qL5a9as0WOPPVb4ARVjRS0ZzquUlBR17NhRUsnZJ+S/8PBwVapUyaMV6nzly5dXWFiY/P39FRgYqCpVqsjlcunkyZOFHCkAnEXXPuA8Czem6JUvNysl7X/D80745bBe7lxXHeoXzFCmx48fV6NGjfTwww/rb3/7W4FsoyioUKGCr0NAITt9+rTKli2rqKgoX4eCEiYrK0sHDx6U1Wp1j5QGAIWNFingvxZuTFHfGb96JFGStD/tlPrO+FULN6YUyHY7duyokSNH6q677rqielavXq3GjRsrICBAzZo102+//XZRmeXLl+vaa6+Vv7+/oqOj9dxzz8npdLqXt2nTRv3799eQIUMUERGhqKgoDR8+3KOOcePGqUGDBgoODlaVKlX0xBNPKCMjQ5K0bNkyPfTQQ0pLS5PFYpHFYnGvf2HXPovFog8++EB33XWXgoKCFBsbqwULFnhsa8GCBYqNjVVAQIBuvvlmTZs2TRaLJceHLlssFr377ru64447FBQUpPj4eK1cuVI7duxQmzZtFBwcrJYtW2rnzp0e6yUkJKhmzZoqW7as6tSpo+nTp19U7+Xi3bhxozp27KiQkBBFRkbqgQcecHcP/eijj1S+fHllZmZ6rNO1a1c98MAD2e5LjRo1JEmNGzeWxWJRmzZtJJ39Evnqq6+qcuXK8vf31zXXXKOFCxde8picW+f1119XrVq13F2kRo0a5VHmjz/+0M0336ygoCA1atRIK1eudC87dOiQunfvrkqVKikoKEgNGjTQxx9/7LF+mzZt9OSTT2rAgAGy2+1q3769+9ida2nNbp9++OEHlSlTRvv37/eob8CAAbrxxhtz3C+ULkePHtWvv/6qX3/9Vampqapdu3aJuk8MQPFCIgXobHe+V77crOw68Z2b98qXmwu8m19eZWRk6I477lDdunW1bt06DR8+XIMHD/Yo89dff+n2229X8+bNtX79eiUkJOjDDz/UyJEjPcpNmzZNwcHBWrVqlV5//XW9+uqrWrx4sXu5n5+fJk2apE2bNmnatGn6z3/+oyFDhkiSWrZsqQkTJigsLEwpKSlKSUm5KI7zvfLKK+rWrZt+//133X777erZs6cOHz4sSUpOTtbf//53de3aVevXr9fjjz+uf/7zn7k6HiNGjNCDDz6oxMRExcXFqUePHnr88cc1bNgwrV27VsYYPfnkk+7y8+bN09NPP61BgwZp48aNevzxx/XQQw/p+++/z3W8R48e1S233KLGjRtr7dq1WrhwoVJTU9WtWzdJ0j333COXy+WRfB04cEBff/21Hn744Wz3Y/Xq1ZKkJUuWKCUlRXPnzpUkTZw4UW+99ZbefPNN/f7772rfvr3uvPNObd++/ZLHZNiwYRozZoxefPFFbd68WbNmzVJkZKRHmX/+858aPHiwEhMTVbt2bXXv3t2daJ86dUpNmzbV119/rY0bN+qxxx7TAw884I7xnGnTpqls2bJasWKF3nnnnVzt00033aSrr77aI3k9c+aMZs6cecljg9IpNDRUdevWVVxcnMLDw7Vz507ukQLgOwYmLS3NSDJpaWm+DgV5dPLkSbN582Zz8uRJr9Y7ePCgSUpKMrP/86upNvSry75m/+dXk5SUZA4ePFgg+yHJzJs3zz194sQJ9ysn7777rilfvrzH/ickJBhJ5rfffjPGGPP888+bOnXqmKysLHeZyZMnm5CQEONyuYwxxrRu3drccMMNHnU3b97cDB069JLb/uyzz0z58uXd01OmTDHh4eEXlatWrZoZP368x76+8MIL7umMjAwjyXz77bfGGGOGDh1q6tev71HHP//5TyPJHDly5JLxXFjvypUrjSTz4Ycfuud9/PHHJiAgwD3dsmVL8+ijj3rUc88995jbb7891/GOGDHCtGvXzqOOvXv3Gklm69atxhhj+vbtazp27Ohe/tZbb5mrr77a4z05X3Jyssd7eE5MTIwZNWqUx7zmzZubJ554Itt60tPTjb+/v3n//fdz3M4HH3zgnrdp0yYjySQlJWW7jjHGdOrUyQwaNMg93bp1a9O4ceOLyp1/Xl9qn8aOHWvi4+Pd03PmzDEhISEmIyPjktsvCXL7GS9JcrPPa9asMYcPH75sXb///rvZt29ffoYHXJG8fh/JSVJSktmwYUOO12Pkr9zmBrRIoVRzuVxyOp1yZJzOVXlHxmk5nU65XK4Cjsw7SUlJatiwoQICAtzzWrRocVGZFi1aeDzjolWrVsrIyNCff/7pntewYUOP9aKjo3XgwAH39JIlS3TrrbeqUqVKCg0N1QMPPKBDhw7pxIkTXsd9/raCg4MVFhbm3tbWrVvVvHlzj/LXXnut1/Wea3Vp0KCBx7xTp04pPT1d0tlj06pVK486WrVqpaSkpFzHu379en3//fcKCQlxv+Li4iTJ3Y3w0Ucf1Xfffae//vpL0tmBOXr37u3Vc0fS09O1b9++XMV7TlJSkjIzM3XrrbfmWPf5+xcdffaewHP753K5NGLECDVo0EAREREKCQnRokWLtGfPHo86mjZtmut9OV/v3r21Y8cO/fLLL5LOHptu3bopODg4T/Wh9DCmaPYUAFDyMdgESjWr1SqbzSZ7SNlclbeHlJXNZpPVai3gyHznwvsNLBaLsrKyJJ0dce2OO+5Q3759NWrUKEVEROinn35Snz59dPr0afcT3fNjW1fi/HrPJSnZzfN2WznFm5GRoc6dO2vs2LEXrXcuKWncuLEaNWqkjz76SO3atdOmTZv09ddfexVDXuT2ZvycjtEbb7yhiRMnasKECe575AYMGKDTpz1/hMhr4lOxYkV17txZU6ZMUY0aNfTtt99q2bJleaoLxZPL5fK4h/D06dM6ceKE+zqdkpKicuXKqUyZMnI6nTp48KBOnz59yVH+AKCgkUihVLPb7bLb7YqtbTThl8Pan3Yq2/ukLJKiwgP099bXyOpX9J5aHh8fr+nTp+vUqVPuVqlzv+yfX2bOnDkyxri/JK9YsUKhoaGqXLlyrrazbt06ZWVl6a233pKf39kG7U8//dSjTNmyZfOlxa5OnTr65ptvPOatWbPmiuvNTnx8vFasWKFevXq5561YsUJ169bNdR1NmjTRnDlzVL16ddlsl760PvLII5owYYL++usvtW3bVlWqVLlk2bJlzyb45x/PsLAwxcTEaMWKFWrdurVHvJdqsYuNjVVgYKCWLl2qRx55JNf7dL4VK1aoS5cuuv/++yWdTbC2bdvm1TGSst+ncx555BF1795dlStXVs2aNS9qdUPxZ4zRqaxTcmY5ZfOzKcAEuK9HJ06c0NatW91l9+7dK+nssOfVqlXTqVOntHPnTjmdTtlsNgUHBysuLo5R+wD4DF37AElWP4te7nz2C+GFadK56Zc71y2QJCojI0OJiYnu5+okJycrMTHRo8vUSy+9pAcffPCSdfTo0UMWi0WPPvqoNm/erG+++UZvvvmmR5knnnhCe/fu1VNPPaUtW7boiy++0Msvv6yBAwe6k6LLqVWrls6cOaO3335bf/zxh6ZPn37RgALVq1dXRkaGli5dKofDkacuf5L0+OOPa8uWLRo6dKi2bdumTz/9VFOnTpUkr7rC5cazzz6rqVOnKiEhQdu3b9e4ceM0d+7cHAfKuFC/fv10+PBhde/eXWvWrNHOnTu1aNEiPfTQQx5JQ48ePfTnn3/q/fffv+xAChUrVlRgYKB74Iq0tDR3vGPHjtUnn3yirVu36rnnnlNiYqKefvrpbOsJCAjQ0KFDNWTIEH300UfauXOnfvnlF3344Ye53r/Y2FgtXrxYP//8s5KSkvT4448rNTU11+tfbp8kqX379goLC9PIkSP10EMPeV03irb0zHRtO7JN+07u04HMA9p3cp+2Hdmm9MyzXWxDQ0PVrFmzi141atSQn5+fatWqpUaNGqlp06Zq1KiRatWqRddPAD5FIgX8V4f60Uq4v4miwgM85keFByjh/iYF9hyptWvXqnHjxmrcuLEkaeDAgWrcuLFeeukld5n9+/dfdC/K+UJCQvTll19qw4YNaty4sf75z39e1MWsUqVK+uabb7R69Wo1atRI//jHP9SnTx+98MILuY61UaNGGjdunMaOHav69etr5syZGj16tEeZli1b6h//+IfuvfdeVahQQa+//nqu6z9fjRo19Pnnn2vu3Llq2LChEhIS3KP2+fv756nOS+natasmTpyoN998U/Xq1dO7776rKVOmuIcbz41zrUQul0vt2rVTgwYNNGDAAJUrV84jUQ0PD9fdd9+tkJAQde3aNcc6bTabJk2apHfffVcxMTHq0qWLJKl///4aOHCgBg0apAYNGmjhwoXuoeIv5cUXX9SgQYP00ksvKT4+Xvfee6/HvW+X88ILL6hJkyZq37692rRpo6ioqMvG780+SWdHhOzdu7dcLleOPxyg+EnPTNfeY3vlzHJ6zHdmObX32F53MgUAxYnFcJem0tPTFR4errS0NIWFhfk6HOTBqVOnlJycrBo1angMuJAXriyjz5cnypFxWvaQsj7tznfy5En3/+m+Io0aNUrvvPOOu8tPcXXrrbeqXr16mjRpkq9DKXL69OmjgwcPXvSMrpKqNHzGjTHadmTbRUnU+cr4lVHsVbH53toMFLb8/D5yzpYtW9xdWs8NYoSCldvcgHukgAtY/SxqFB0op7PM2YEliuA9UaXFv//9bzVv3lzly5fXihUr9MYbb3g8/6m4OXLkiJYtW6Zly5bp3//+t6/DKVLS0tK0YcMGzZo1y6dJ1JkzZzweUl2Yzk+qSpJz90Tl5EzWGR09cVQBfvnzxRNnW395WDFQsEikgGzY7Xa5XK4SPTpfcbB9+3aNHDlShw8fVtWqVTVo0CANGzbM12HlWePGjXXkyBGNHTtWderU8XU4RUqXLl20evVq/eMf/9Btt93m63CQjy6XRHmU44YDAMUIiRSQDbvd7usQIGn8+PEaP368r8PIN7t27fJ1CEUWQ52XXDa/3H3VyG05ACgquGoBAPBfZcqUKdTuUKXhHqkAE6ADmQcue49UuaBy3CMFeOmHH37QG2+8oXXr1iklJUXz5s275EBA//jHP/Tuu+9q/PjxGjBgQKHGWVLRiA4AAAqMxWJRdHDOo55GBUeRRAF5cPz4cTVq1EiTJ0/Osdy8efP0yy+/KCYmppAiKx1okQIAAAUqzD9MVVRFKcdTPFqmyviVUVRwlML8GTEXyIuOHTuqY8eOOZb566+/9NRTT2nRokXq1KlTIUVWOpBIAQCAAhfmH6bQsqE6euKonFlO2fxsdOcDClhWVpYeeOABPfvss6pXr56vwylxSKQAAEChsFgsZ4c49/vfNICCM3bsWNlsNvXv39/XoZRIJFJANhwOh3v4c0bwAwAAxc26des0ceJE/frrr/xoUUAYbALIhsPh0MGDB+VwOHwdSrGwbNkyWSwWHT161NehFFtTp05VuXLlfLLt/fv367bbblNwcHChx9CmTZsCGz2qd+/elxy9CgBKuh9//FEHDhxQ1apVZbPZZLPZtHv3bg0aNEjVq1f3dXglAokUcKEsl4JS1yl8z2IFpa6TslwFurnRo0erefPmCg0NVcWKFdW1a1dt3bq1QLeZ31q2bKmUlBSFh4f7OpRioXr16powYYKvw3AbP368UlJSlJiYqG3btvk6nHwzceJETZ061T1dkEkbABQ1DzzwgH7//XclJia6XzExMXr22We1aNEiX4dXItC1Dzjf5gXSwqGqmr7vf/PWjZQ6jJXq3lkgm1y+fLn69eun5s2by+l06vnnn1e7du20efNm+fnl7bcOY4xcLpdstsL5iJctW1ZRUVGFsq2iqrCPeX7auXOnmjZtqtjY2DzXcfr0aZUtWzYfo8o7l8sli8VCYg+gxMvIyNCOHTvc08nJyUpMTFRERISqVq2q8uXLe5QvU6aMoqKiVKdOncIOtUSiRQo4Z/MC6dMHpfOTKElKTzk7f/OCAtnswoUL1bt3b9WrV0+NGjXS1KlTtWfPHq1bty7XdZzrWvftt9+qadOm8vf3108//ZRt16YBAwaoTZs27uk2bdqof//+GjJkiCIiIhQVFaXhw4d7rGOxWPTBBx/orrvuUlBQkGJjY7Vgwf+Ox4Vd+851U1u0aJHi4+MVEhKiDh06KCUlxb2O0+lU//79Va5cOZUvX15Dhw5Vr169cuyKdejQIXXv3l2VKlVSUFCQGjRooI8//jjHY3Mulq+++kp16tRRUFCQ/v73v+vEiROaNm2aqlevrquuukr9+/eXy/W/1sfp06erWbNmCg0NVVRUlHr06KEDBw5c9pjv3LlTXbp0UWRkpEJCQtS8eXMtWbLE43jv3r1bzzzzjCwWy0X91nM6ZllZWXr11VdVuXJl+fv765prrtHChQvdy3ft2iWLxaK5c+fq5ptvVlBQkBo1aqSVK1de8vhUr15dc+bM0UcffSSLxaLevXtLkvbs2aMuXbooJCREYWFh6tatm1JTU93rDR8+XNdcc40++OAD1ahRQwEBAZfcxooVK9SmTRsFBQXpqquuUvv27XXkyBGP/crp/Bs3bpwaNGig4OBgValSRU888YQyMjLcy8+9xwsWLFDdunXl7++vPXv2eJz/vXv31vLlyzVx4kT3cU9OTlatWrX05ptvemwvMTFRFovF48sJABRFa9euVePGjdW4cWNJ0sCBA9W4cWO99NJLPo6sdCCRAqSz3fcWDpVksln433kLnyvwbn6SlJaWJkmKiIjwet3nnntOY8aMUVJSkho2bJjr9aZNm6bg4GCtWrVKr7/+ul599VUtXrzYo8wrr7yibt266ffff9ftt9+unj176vDhw5es88SJE3rzzTc1ffp0/fDDD9qzZ48GDx7sXj527FjNnDlTU6ZM0YoVK5Senq758+fnGOepU6fUtGlTff3119q4caMee+wxPfDAA1q9enWO6504cUKTJk3S7NmztXDhQi1btkx33XWXvvnmG33zzTeaPn263n33XX3++efudc6cOaMRI0Zo/fr1mj9/vnbt2uVOMs534THPyMjQ7bffrqVLl+q3335Thw4d1LlzZ+3Zs0eSNHfuXFWuXFmvvvqqUlJSPBKlyx2ziRMn6q233tKbb76p33//Xe3bt9edd96p7du3e8T0z3/+U4MHD1ZiYqJq166t7t27y+l0Kjtr1qxRhw4d1K1bN6WkpGjixInKyspSly5ddPjwYS1fvlyLFy/WH3/8oXvvvddj3R07dmjOnDmaO3euEhMTs60/MTFRt956q+rWrauVK1fqp59+UufOnT2S1sudf35+fpo0aZI2bdqkadOm6T//+Y+GDBly0Xs8duxYffDBB9q0aZMqVqzosXzixIlq0aKFHn30Ufdxr1q1qh5++GFNmTLFo+yUKVN00003qVatWtnuEwAUFuNyyWzYKMtPP8ls2Cjj8vwe0qZNGxljLnqd3635fLt27aKLc34yMGlpaUaSSUtL83UoyKOTJ0+azZs3m5MnT3q13sGDB01SUpLZvWyGMS+HXfa1e9kMk5SUZA4ePFgg++FyuUynTp1Mq1atjDHGnDhxwv3Kyffff28kmfnz53vM79Wrl+nSpYvHvKefftq0bt3aPd26dWtzww03eJRp3ry5GTp0qHtaknnhhRfc0xkZGUaS+fbbbz22f+TIEWOMMVOmTDGSzI4dO9zrTJ482URGRrqnIyMjzRtvvOGedjqdpmrVqhfFezmdOnUygwYNuuTy7GJ5/PHHTVBQkDl27Jh7Xvv27c3jjz9+yXrWrFljJLnXudQxz069evXM22+/7Z6uVq2aGT9+/GXjvPCYxcTEmFGjRnms17x5c/PEE08YY4xJTk42kswHH3zgXr5p0yYjySQlJV0yvi5duphevXq5p7/77jtjtVrNnj17Lqpn9erVxhhjXn75ZVOmTBlz4MCBHPe9e/fu7vM5O7k5/y702WefmfLly7unzx27xMREj3IXnv+tW7c2Tz/9tEeZv/76y1itVrNq1SpjjDGnT582drvdTJ06Ncf9yi+5/YyXJKVxn1F65PX7SHbSFi0y21q3MZvrxLlf21q3MWmLFuVDpMhJbnMDWqRQqrlcLjmdTlmOH7h8YUmW4wfkdDo9fk3PT/369dPGjRs1e/bsPK3frFmzPK13YetVdHS0Rze2C8sEBwcrLCzsojLnCwoKUs2aNbOtMy0tTampqbr22mvdy61Wq5o2bZpjnC6XSyNGjFCDBg0UERGhkJAQLVq0yN3ak9tYIiMjVb16dYWEhHjMO39/1q1bp86dO6tq1aoKDQ1V69atJemibV14zDMyMjR48GDFx8erXLlyCgkJUVJS0mVjzC7O849Zenq69u3bp1atWnms06pVKyUlJXnMO/+9io6OlqQc36sLJSUlqUqVKqpSpYp7Xt26dVWuXDmPbVWrVk0VKlTIsa5zLVI5udz5t2TJEt16662qVKmSQkND9cADD+jQoUM6ceKEu0zZsmW9aoU9JyYmRp06ddL//d//SZK+/PJLZWZm6p577vG6LgDIL+nffae/nh4g5/79HvOdqan66+kBSv/uOx9FhvORSKFUs1qtstlsMsEVL19YkgmuKJvNJqvVmu+xPPnkk/rqq6/0/fffq3LlynmqIzg42GPaz89Pxnh2Vzxz5sxF65UpU8Zj2mKxKCsry+sylyt/YSzeeuONNzRx4kQNHTpU33//vRITE9W+fXudPn06x/WyiyWn/Tl+/Ljat2+vsLAwzZw5U2vWrNG8efMk6aJtXXjMBw8erHnz5um1117Tjz/+qMTERDVo0OCyMV4qzrwcs/PrOXcPVk7vVV5duO/ZCQwMvGyZnN6LXbt26Y477lDDhg01Z84crVu3TpMnT5bk+V4EBgbm+TkpjzzyiGbPnq2TJ09qypQpuvfeexUUFJSnugDgShmXS6mvjZayu/7/d17qa6Mv6uaHwlf8hpcC8pHdbj/7wN3asWdH50tPUfb3SVmksBhVvfE+yS9/kyhjjJ566inNmzdPy5YtU40aNfKt7goVKmjjxo0e8xITEy/64lrYwsPDFRkZqTVr1uimm26SdLa16ddff9U111xzyfVWrFihLl266P7775d0NjnYtm2b6tatm6/xbdmyRYcOHdKYMWPcrTJr167N1borVqxQ7969ddddd0k620K1a9cujzJly5b1ulUzLCxMMTExWrFihbt17Nz2zm/Zyw/x8fHau3ev9u7d697/zZs36+jRo14f64YNG2rp0qV65ZVX8hTLunXrlJWVpbfeess9iuWnn36ap7ouddxvv/12BQcHKyEhQQsXLtQPP/yQp/oB+MaZM2cueR9oYcvMzNSZM2f0xx9/5H3k3Q0bpQtaojwLGDn379eWufNkaVA/j5EWPe7vZMVIkW6RGj58uHt0pXOvuLg49/JTp06pX79+Kl++vEJCQnT33Xd7jCoF5Jqf9ewQ55KkC3/V/u90hzH5nkRJZ7vzzZgxQ7NmzVJoaKj279+v/fv36+TJk+4yjzzyiIYNG+Z13bfccovWrl2rjz76SNu3b9fLL798UWLlK0899ZRGjx6tL774Qlu3btXTTz+tI0eO5NiqEBsbq8WLF+vnn39WUlKSHn/88QL5zFetWlVly5bV22+/rT/++EMLFizQiBEjcrVubGyse/CF9evXq0ePHhe1BlWvXl0//PCD/vrrL68e+vzss89q7Nix+uSTT7R161Y999xzSkxM1NNPP+3V/l1O27Zt1aBBA/Xs2VO//vqrVq9erQcffFCtW7f2uvvosGHDtGbNGj3xxBP6/ffftWXLFiUkJOR6v2vVqqUzZ86434vp06frnXfeyctuqXr16lq1apV27dolh8Phfl+sVqt69+6tYcOGKTY2Vi1atMhT/QAgnf2B1Ol05vmVdSh318esQ44r2k5RexXUbRMFqUgnUpJUr1499whLKSkp+umnn9zLnnnmGX355Zf67LPPtHz5cu3bt09/+9vffBgtirW6d0rdPpLCoj3nh8WcnV9Az5FKSEhQWlqa2rRpo+joaPfrk08+cZfZu3evx+huudW+fXu9+OKLGjJkiJo3b65jx47pwQcfzM/w82zo0KHq3r27HnzwQbVo0UIhISFq3759jsNov/DCC2rSpInat2+vNm3aKCoqKsfh0vOqQoUKmjp1qj777DPVrVtXY8aMuWiI7EsZN26crrrqKrVs2VKdO3dW+/bt1aRJE48yr776qnbt2qWaNWte9h6j8/Xv318DBw7UoEGD1KBBAy1cuFALFiy4ouc/ZcdiseiLL77QVVddpZtuuklt27bV1Vdf7XFO5lbt2rX13Xffaf369br22mvVokULffHFF7l+3lajRo00btw4jR07VvXr19fMmTM1evRor+OQzna7tFqtqlu3ripUqOBx31qfPn10+vRpPfTQQ3mqGwDOsVgsstlseX75lc9dq4xfefsVbaeovQritomCZjFXetNCARo+fLjmz5+f7bC6aWlpqlChgmbNmqW///3vks52x4mPj9fKlSt1/fXX53o76enpCg8PV1pamsLCwvIrfBSiU6dOKTk5+bLPs8mVLJf2/DhbluMHZIIrFkh3vtw6v1UqN/eaFGdZWVmKj49Xt27dct36A+SXH3/8Ubfeeqv27t2ryMjIQttuafqMn1Ma9xmlR358HzEul3bc2lbO1NTs75OyWGSLjFStpUtkKYbJR3GQ29ygyLdIbd++XTExMbr66qvVs2dP9y+I69at05kzZ9S2bVt32bi4OFWtWjXHh08Cl+Vn1YnIpkqreptORDb1WRJV0u3evVvvv/++tm3bpg0bNqhv375KTk5Wjx49fB0aSpHMzEz9+eefGj58uO65555CTaIAIDsWq1WRz/+3O/+F3d3/Ox35/DCSqCKgSCdS1113naZOnaqFCxcqISFBycnJuvHGG3Xs2DHt379fZcuWVbly5TzWiYyM1P6cbtDT2T+c6enpHi/gfHa7XRUqVCh2Nz0WJ35+fpo6daqaN2+uVq1aacOGDVqyZIni4+N9HRpKkY8//ljVqlXT0aNH9frrr/s6HACQJIW1a6dKEyfIdsGPO7bISFWaOEFh7dr5KDKcr0iP2texY0f3/xs2bKjrrrtO1apV06effnpF3QFGjx6d5xGkUDqQQBW8KlWqaMWKFb4OA6Vc79691bt3b1+HAQAXCWvXTqG33qotc+cp65BDfuXtqvW3u2iJKkKKdIvUhcqVK6fatWtrx44dioqK0unTp3X06FGPMqmpqYqKisqxnmHDhiktLc392rt3bwFGDQAAAHjPYrXK0qC+zA03yNKgPklUEVOsEqmMjAzt3LlT0dHRatq0qcqUKaOlS5e6l2/dulV79uy57NC1/v7+CgsL83gBAAAAQG4V6a59gwcPVufOnVWtWjXt27dPL7/8sqxWq7p3767w8HD16dNHAwcOVEREhMLCwvTUU0+pRYsWXo3YBwAAAADeKtKJ1J9//qnu3bvr0KFDqlChgm644Qb98ssv7ueujB8/Xn5+frr77ruVmZmp9u3b69///rePowYAAABQ0hXpRGr27Nk5Lg8ICNDkyZM1efLkQooIAAAAAIp4IgX4isPhkMvlktVqZQQ/AAAAXKRYDTYBFBaHw6GDBw/K4XD4OpRCM3z4cF1zzTW+DuOKVa9eXRMmTPB1GMXSsmXLZLFYLhoNNb9NnTr1omcA+kJh7S8AoGQikQIu4MpyaVP6Jq04vEKb0jfJleUq0O0lJCSoYcOG7hEkW7RooW+//bZAt1lYdu3aJYvFosTExELb5po1a/TYY48V2vYupagkC5fSpk0bDRgwwGNey5YtlZKSovDwcN8EVYBK2/4CAAoeXfuA8yzZvURjVo9R6olU97yE3Ql67trn1LZa2wLZZuXKlTVmzBjFxsbKGKNp06apS5cu+u2333T11VcXyDZLotOnT6ts2bLuwWjgvbJly172OXwlSWnbXwBA/qJFCvivJbuXaOCygR5JlCQdOHFAA5cN1JLdSwpku507d9btt9+u2NhY1a5dW6NGjVJISIh++eWXXK2flZWlypUrKyEhwWP+b7/9Jj8/P+3evVuStGfPHnXp0kUhISEKCwtTt27dlJqaml2VkrL/Bb9r167q3bu3e7p69ep67bXX9PDDDys0NFRVq1bVe++9515eo0YNSVLjxo1lsVjUpk0b97IPPvhA8fHxCggIUFxc3EUjbu7du1fdunVTuXLlFBERoS5dumjXrl3u5b1791bXrl01atQoxcTEqE6dOu6Yzu/aZ7FY9MEHH+iuu+5SUFCQYmNjtWDBAo9tLViwQLGxsQoICNDNN9+sadOmXbbL17hx49SgQQMFBwerSpUqeuKJJ5SRkSHpbJexhx56SGlpabJYLLJYLBo+fHi29ezcuVNdunRRZGSkQkJC1Lx5cy1Zculzbdu2bbJYLNqyZYvH/PHjx6tmzZru6eXLl+vaa6+Vv7+/oqOj9dxzz8npdLqP3fLlyzVx4kR3fLt27bqoq9u5VrVFixYpPj5eISEh6tChg1JSUtzbcTqd6t+/v8qVK6fy5ctr6NCh6tWrl7p27XrJfchOQkKCatasqbJly6pOnTqaPn26x/KjR4/q8ccfV2RkpAICAlS/fn199dVXkqRDhw6pe/fuqlSpkoKCgtSgQQN9/PHH7nVzu7+SNGfOHNWrV0/+/v6qXr263nrrLY84LnfOAwBKDxIpQGe7841ZPUZG5qJl5+aNXT22wLv5uVwuzZ49W8ePH7/sg6XP8fPzU/fu3TVr1iyP+TNnzlSrVq1UrVo1ZWVlqUuXLjp8+LCWL1+uxYsX648//tC99957xTG/9dZbatasmX777Tc98cQT6tu3r7Zu3SpJWr16tSRpyZIlSklJ0dy5c92xvfTSSxo1apSSkpL02muv6cUXX9S0adMkSWfOnFH79u0VGhqqH3/8UStWrHB/iT99+rR720uXLtXWrVu1ePFi95fq7Lzyyivq1q2bfv/9d91+++3q2bOnDh8+LElKTk7W3//+d3Xt2lXr16/X448/rn/+85+X3W8/Pz9NmjRJmzZt0rRp0/Sf//xHQ4YMkXS2y9iECRMUFhamlJQUpaSkaPDgwdnWk5GRodtvv11Lly7Vb7/9pg4dOqhz587as2dPtuVr166tZs2aaebMmR7zZ86cqR49ekiS/vrrL91+++1q3ry51q9fr4SEBH344YcaOXKkJGnixIlq0aKFHn30UXd8VapUyXZ7J06c0Jtvvqnp06frhx9+0J49ezz2ZezYsZo5c6amTJmiFStWKD09XfPnz7/s8TvfvHnz9PTTT2vQoEHauHGjHn/8cT300EP6/vvvJZ39saBjx45asWKFZsyYoc2bN2vMmDGyWq2SpFOnTqlp06b6+uuvtXHjRj322GN64IEH3Odfbvd33bp16tatm+677z5t2LBBw4cP14svvqipU6d6lMvpnAcAlCIGJi0tzUgyaWlpvg4FeXTy5EmzefNmc/LkSa/WO3jwoElKSjKfr/rc1J9a/7Kvz1d9bpKSkszBgwfzNf7ff//dBAcHG6vVasLDw83XX39tjDHmxIkT7ldOfvvtN2OxWMzu3buNMca4XC5TqVIlk5CQYIwx5rvvvjNWq9Xs2bPHvc6mTZuMJLN69WpjjDEvv/yyadSokXt569atzdNPP+2xnS5duphevXq5p6tVq2buv/9+93RWVpapWLGie7vJyclGkvntt9886qlZs6aZNWuWx7wRI0aYFi1aGGOMmT59uqlTp47JyspyL8/MzDSBgYFm0aJFxhhjevXqZSIjI01mZqZHPdWqVTPjx493T0syL7zwgns6IyPDSDLffvutMcaYoUOHmvr163vU8c9//tNIMkeOHDG59dlnn5ny5cu7p6dMmWLCw8Nzvf756tWrZ95+++1LLh8/frypWbOme3rr1q1GkklKSjLGGPP8889fdPwmT55sQkJCjMvlMsZk//5+//33Hvs9ZcoUI8ns2LHDo57IyEj3dGRkpHnjjTfc006n01StWtV06dLlkvFfeGxatmxpHn30UY8y99xzj7n99tuNMcYsWrTI+Pn5ma1bt16yzgt16tTJDBo0yD2dm/3t0aOHue222zzKPPvss6Zu3bru6cud897K7We8JCmN+4zSI6/fR3KSlJRkNmzY4L7Go+DlNjegRQqlmsvlktPplONU7kbnc5xyyOl0yuXK35apOnXqKDExUatWrVLfvn3Vq1cvbd68OdfrX3PNNYqPj3e3Si1fvlwHDhzQPffcI0lKSkpSlSpVPH6Fr1u3rsqVK6ekpKQrir1hw4bu/1ssFkVFRenAgQOXLH/8+HHt3LlTffr0UUhIiPs1cuRI7dy5U5K0fv167dixQ6Ghoe7lEREROnXqlLuMJDVo0EBly5b1Ksbg4GCFhYW5Y9y6dauaN2/uUf7aa6+9bJ1LlizRrbfeqkqVKik0NFQPPPCADh06pBMnTlx23fNlZGRo8ODBio+PV7ly5RQSEqKkpKRLtkhJ0n333addu3a5u3/OnDlTTZo0UVxcnKSz73eLFi1ksVjc67Rq1UoZGRn6888/vYovKCjIo8tgdHS0+9ilpaUpNTXV43hZrVY1bdrUq20kJSWpVatWHvNatWrlPjcTExNVuXJl1a5dO9v1XS6XRowYoQYNGigiIkIhISFatGhRjsfQmzi2b9/u8Zn39pwHAJRMDDaBUs1qtcpms8kekLtnRdkD7LLZbO4uRfmlbNmyqlWrliSpadOmWrNmjSZOnOjVMN49e/bUrFmz9Nxzz2nWrFnq0KGDypcvn+eY/Pz8ZIxnV8czZ85cVK5MmTIe0xaLRVlZWZes99x9RO+//76uu+46j2XnjmtGRoaaNm16Ufc1SR6DSQQHB19mL/IW4+Xs2rVLd9xxh/r27atRo0YpIiJCP/30k/r06aPTp08rKCgo13UNHjxYixcv1ptvvqlatWopMDBQf//73z26MF4oKipKt9xyi2bNmqXrr79es2bNUt++ffO8PznJ7thdeF4UtMDAwByXv/HGG+7Py7n71gYMGJDjMbwS+X0+AQCKJ1qkUKrZ7XbFxcWpa7OuigyKlEWWbMtZZFFUUJS6NuuquLi4An9Ib1ZWljIzM71ap0ePHtq4caPWrVunzz//XD179nQvi4+P1969e7V37173vM2bN+vo0aOqW7dutvVVqFDBY1ABl8uljRs3ehXTudai83/Nj4yMVExMjP744w/VqlXL43VucIomTZpo+/btqlix4kVl8nuo6jp16mjt2rUe89asWZPjOuvWrVNWVpbeeustXX/99apdu7b27dvnUaZs2bK5arlcsWKFevfurbvuuksNGjRQVFSUx6Aal9KzZ0998sknWrlypf744w/dd9997mXx8fFauXKlR8KzYsUKhYaGqnLlyl7Fl5Pw8HBFRkZ6HC+Xy6Vff/3Vq3ri4+O1YsUKj3krVqxwn5sNGzbUn3/+qW3btmW7/ooVK9SlSxfdf//9atSoka6++uqLyuZmfy8VR+3atfP9xxMAQPFHIgVIsvpZ9dy1z0nSRcnUuemh1w6V1S//v0wNGzZMP/zwg3bt2qUNGzZo2LBhWrZsmUci9Mgjj2jYsGE51lO9enW1bNlSffr0kcvl0p133ule1rZtWzVo0EA9e/bUr7/+qtWrV+vBBx9U69at1axZs2zru+WWW/T111/r66+/1pYtW9S3b1+vH1xasWJFBQYGauHChUpNTVVaWpqks4M/jB49WpMmTdK2bdu0YcMGTZkyRePGjZN0Nkmw2+3q0qWLfvzxRyUnJ2vZsmXq37+/113TLufxxx/Xli1bNHToUG3btk2ffvqpe3CB87vGna9WrVo6c+aM3n77bf3xxx+aPn263nnnHY8y1atXV0ZGhpYuXSqHw3HJLn+xsbGaO3euEhMTtX79evXo0SNXrRt/+9vfdOzYMfXt21c333yzYmJi3MueeOIJ7d27V0899ZS2bNmiL774Qi+//LIGDhwoPz8/d3yrVq3Srl275HA48tyi8tRTT2n06NH64osvtHXrVj399NM6cuTIJY9ddp599llNnTpVCQkJ2r59u8aNG6e5c+e6B7Vo3bq1brrpJt19991avHixkpOT9e2332rhwoWSzh7DxYsX6+eff1ZSUpIef/zxi0akzM3+Dho0SEuXLtWIESO0bds2TZs2Tf/6178uOVAIAKB0I5EC/qtttbYa12acKgZV9JgfGRSpcW3GFdhzpA4cOKAHH3xQderU0a233qo1a9Zo0aJFuu2229xl9u7d69E6dCk9e/bU+vXrddddd3l0h7JYLPriiy901VVX6aabblLbtm119dVX65NPPrlkXQ8//LB69erlTriuvvpq3XzzzV7tm81m06RJk/Tuu+8qJiZGXbp0kXQ2Mfzggw80ZcoUNWjQQK1bt9bUqVPdLVJBQUH64YcfVLVqVf3tb39TfHy8+vTpo1OnTiksLMyrGC6nRo0a+vzzzzV37lw1bNhQCQkJ7lH7/P39s12nUaNGGjdunMaOHav69etr5syZGj16tEeZli1b6h//+IfuvfdeVahQQa+//nq2dY0bN05XXXWVWrZsqc6dO6t9+/Zq0qTJZeMODQ1V586dtX79eo+kW5IqVaqkb775RqtXr1ajRo30j3/8Q3369NELL7zgLjN48GBZrVbVrVtXFSpU8Pp+onOGDh2q7t2768EHH1SLFi0UEhKi9u3bKyAgINd1dO3aVRMnTtSbb76pevXq6d1339WUKVM8hsufM2eOmjdvru7du6tu3boaMmSIu4XphRdeUJMmTdS+fXu1adNGUVFRFw2/npv9bdKkiT799FPNnj1b9evX10svvaRXX33VY8h/AADOsZjC7uxeBKWnpys8PFxpaWn5/iUNhePUqVNKTk5WjRo1vPoClx1Xlkvz186X45RD9gC7ujbrWiAtUblx8uRJ9/8vd58I8s+oUaP0zjvveHSFRO5kZWUpPj5e3bp104gRI3wdTpFXGj/jpXGfUXrk5/eRc7Zs2SKn0ymbzeYeVAgFK7e5AYNNABew+llVL6yenEFnL1q+SqJQeP7973+refPmKl++vFasWKE33nhDTz75pK/DKhZ2796t7777Tq1bt1ZmZqb+9a9/KTk52f1MKwAASioSKSAbdrtdLpeLG8xLie3bt2vkyJE6fPiwqlatqkGDBl32njSc5efnp6lTp2rw4MEyxqh+/fpasmSJ4uPjfR0aAAAFikQKyEZBj8qHomX8+PEaP368r8MolqpUqXLRSHcAAJQGDDYBAAAAAF4ikQIAAAAAL5FIAQAAAICXuEcKAAAfsdn4MwwAxRVXcAAAfKRMmTK+DgEAkEckUkA2HA6He/hzRvADAADAhbhHCsiGw+HQwYMH5XA4fB1Kodm1a5csFosSExNzLNemTRsNGDCgwOOpXr26JkyYUODbKc4K4hgtW7ZMFotFR48ezdd6c9K7d2917dq10LYH+IIvPlvesFgsmj9/vq/DuEh+XOe4xqCgkEgBFzAul8yGjbL89JPMho0yLlehbXvMmDGyWCyFkqhcqEqVKkpJSVH9+vUlXfqP/ty5czVixIhCj+9yhg8frmuuueai+b78cpBfSefUqVNVrly5K64HpU9uv7xzjp2V2x+UULwV1g+CKPno2gecJ/2775T62mhp/373rww7Jk9W5PPDFNauXYFue82aNXr33XfVsGHDAt1Odk6fPq2yZcsqKirqsmUjIiIKIaKi58yZM9zPAg/GGLlcLgaMKCFOnz7t6xBQzJz724nSixYp4L/Sv/tOfz09QM79+z3mO1NT9dfTA5T+3XcFtu2MjAz17NlT77//vq666iqv12/WrJnefPNN93TXrl1VpkwZZWRkSJL+/PNPWSwW7dixQ9LZrhIjRozQgw8+qLCwMD322GMev8Tu2rVLN998syTpqquuksViUe/evSVd/Ete9erV9dprr+nhhx9WaGioqlatqvfee88jvp9//lnXXHONAgIC1KxZM82fP9/rX32PHj2qRx55RBUqVFBYWJhuueUWrV+/XtLZX9NfeeUVrV+/XhaLRRaLRVOnTlX16tUlSXfddZcsFot7WpK++OILNWnSRAEBAbr66qv1yiuvyOl0updbLBYlJCTozjvvVHBwsEaNGqUjR46oZ8+eqlChggIDAxUbG6spU6ZkG2/v3r21fPlyTZw40R3Trl27JEnLly/XtddeK39/f0VHR+u5557z2Pb5li1bpoceekhpaWnueoYPH+5efuLEiRyP/d69e9WtWzeVK1dOERER6tKlizuO3JozZ47q1asnf39/Va9eXW+99ZbH8szMTA0dOlRVqlSRv7+/atWqpQ8//FCS5HK51KdPH9WoUUOBgYGqU6eOJk6c6NX2Jen9999XlSpVFBQUpLvuukvjxo27qAUlISFBNWvWVNmyZVWnTh1Nnz7dvaxHjx669957PcqfOXNGdrtdH330kSQpKytLo0ePdsfaqFEjff755+7y51p3vv32WzVt2lT+/v766aef1KZNG/Xv319DhgxRRESEoqKiPN4j6ez59O677+qOO+5QUFCQ4uPjtXLlSu3YsUNt2rRRcHCwWrZsqZ07d3qsl5vz9IMPPtBdd92loKAgxcbGasGCBZKU4+f4fDmdY0eOHNGDDz6oq666SkFBQerYsaO2b99+yffJGKPhw4eratWq8vf3V0xMjPr37+8R77n4zilXrpymTp3qjtlisWj27Nlq2bKlAgICVL9+fS1fvvyi9+Hrr79Ww4YNFRAQoOuvv14bN270qPdy521218EaNWpIkho3biyLxaI2bdpku59Xcl6vW7dOzZo1U1BQkFq2bKmtW7d6LM/pPD53DAviXLrQ5a4dy5Yt07XXXqvg4GCVK1dOrVq10u7duyVJ69ev180336zQ0FCFhYWpadOmWrt2bbbbudw5I+XvdS6na/PGjRvVsWNHhYSEKDIyUg888IBHF/82bdroySef1IABA2S329W+fftLHj+UEgYmLS3NSDJpaWm+DgV5dPLkSbN582Zz8uTJPK2f5XSaba3bmM114rJ/xcWbba3bmCynM58jP+vBBx80AwYMMMYY07p1a/P0008bY4w5ceKE+5WTgQMHmk6dOp3dl6wsExERYex2u/n222+NMcbMmDHDVKpUyV2+WrVqJiwszLz55ptmx44dZseOHSY5OdlIMr/99ptxOp1mzpw5RpLZunWrSUlJMUePHr0ovnN1RUREmMmTJ5vt27eb0aNHGz8/P7NlyxZjzNnPV0REhLn//vvNpk2bzDfffGNq167t3talVKtWzYwfP9493bZtW9O5c2ezZs0as23bNjNo0CBTvnx5c+jQIXPixAkzaNAgU69ePZOSkmJSUlLMiRMnzIEDB4wkM2XKFJOSkmIOHDhgjDHmhx9+MGFhYWbq1Klm586d5rvvvjPVq1c3w4cPd29PkqlYsaL5v//7P7Nz506ze/du069fP3PNNdeYNWvWmOTkZLN48WKzYMGCbOM/evSoadGihXn00UfdMTmdTvPnn3+aoKAg88QTT5ikpCQzb948Y7fbzcsvv5xtPZmZmWbChAkmLCzMXc+xY8dydexPnz5t4uPjzcMPP2x+//13s3nzZtOjRw9Tp04dk5mZme32vv/+eyPJHDlyxBhjzNq1a42fn5959dVXzdatW82UKVNMYGCgmTJlinudbt26mSpVqpi5c+eanTt3miVLlpjZs2e7Y3jppZfMmjVrzB9//GFmzJhhgoKCzCeffOJev1evXqZLly6XPBd++ukn4+fnZ9544w2zdetWM3nyZBMREWHCw8PdZebOnWvKlCljJk+ebLZu3WreeustY7VazX/+8x9jjDFfffWVCQwMdB87Y4z58ssvTWBgoElPTzfGGDNy5EgTFxdnFi5caHbu3GmmTJli/P39zbJlyzyOTcOGDc13331nduzYYQ4dOmRat25twsLCzPDhw822bdvMtGnTjMViMd999517W5JMpUqVzCeffGK2bt1qunbtaqpXr25uueUWs3DhQrN582Zz/fXXmw4dOrjXye15WrlyZTNr1iyzfft2079/fxMSEmIOHTqU4+f4fDmdY3feeaeJj483P/zwg0lMTDTt27c3tWrVMqdPn872vfrss89MWFiY+eabb8zu3bvNqlWrzHvvvecR7+zZsz2ua+Hh4e7z6dx1qHLlyubzzz83mzdvNo888ogJDQ01DofD432Ij4833333nfn999/NHXfcYapXr+6OKzfnbXbXwdWrVxtJZsmSJSYlJcUcOnQo2/3MzXl9oXNxX3fddWbZsmVm06ZN5sYbbzQtW7Z0l7nceXzuGBbUuTRv3jz3/uV07Thz5owJDw83gwcPNjt27DCbN282U6dONbt37zbGGFOvXj1z//33m6SkJLNt2zbz6aefmsTExDydM/lxnTv/GnOpa/ORI0dMhQoVzLBhw0xSUpL59ddfzW233WZuvvlmdyytW7c2ISEh5tlnnzVbtmxxx3C+K/0+kp2kpCSzYcMGk5SUlG91Ime5zQ1IpAyJVEmQ1wvXwYMHTVJSktn86WeXTqLOf336mUlKSjIHDx7Mt9g//vhjU79+fXfseUmkFixYYMLDw43T6TSJiYkmKirKPP3002bo0KHGGGMeeeQR06NHD3f5atWqma5du3rUcX4iZczFX6jPyS6Ruv/++93TWVlZpmLFiiYhIcEYY0xCQoIpX768x3vz/vvve5VI/fjjjyYsLMycOnXKo0zNmjXNu+++a4wx5uWXXzaNGjW6qJ7zvxycc+utt5rXXnvNY9706dNNdHS0x3rnkttzOnfubB566KFLxnyhC4+VMcY8//zzpk6dOiYrK8s9b/LkySYkJMS4XK5s65kyZYpH0nDO5Y799OnTL9pWZmamCQwMNIsWLcp2Wxe+7z169DC33XabR5lnn33W1K1b1xhjzNatW40ks3jx4uwPQjb69etn7r77bvf05RKpe++91/1DwTk9e/b0OCYtW7Y0jz76qEeZe+65x9x+++3GGGPOnDlj7Ha7+eijj9zLu3fvbu69915jjDGnTp0yQUFB5ueff/aoo0+fPqZ79+7GmP8dm/nz53uUad26tbnhhhs85jVv3tz9+TPm7Pn0wgsvuKdXrlxpJJkPP/zQPe/jjz82AQEB7uncnqfn15uRkWEkuX9EudTn+ELZnWPbtm0zksyKFSvc8xwOhwkMDDSffvpptvW89dZbpnbt2pdMtHKbSI0ZM8a9zpkzZ0zlypXN2LFjPfbpXLJujDGHDh0ygYGB7kTmcuetMbm7DnrjwvP6QufiXrJkiXve119/bSS5r4+XO4+NKdhz6dy18nLXjkOHDhlJ7h8ZLhQaGmqmTp16yWNxvsudM/lxnbvwGpPdtXnEiBGmXbt2HvP27t3r/iHi3HqNGzfOcX9IpEqG3OYGdO1DqeZyueR0OpV1KHej82UdcsjpdMqVTwNQ7N27V08//bRmzpypgICAPNdz44036tixY/rtt9+0fPlytW7dWm3atNGyZcskne1KdmEXlWbNml1B5J7Ov6/LYrEoKipKBw4ckCRt3brV3f3mnGuvvdar+tevX6+MjAyVL19eISEh7ldycvJF3VdyW9+rr77qUdejjz6qlJQUnThxwl3uwmPUt29fzZ49W9dcc42GDBmin3/+2ettJyUlqUWLFrJYLO55rVq1UkZGhv7880+v68vp2K9fv147duxQaGioez8jIiJ06tSpXB+3pKQktWrVymNeq1attH37drlcLiUmJspqtap169aXrGPy5Mlq2rSpKlSooJCQEL333nvas2dPrvdx69atF50zF05fKs6kpCRJZx98261bN82cOVOSdPz4cX3xxRfq2bOnJGnHjh06ceKEbrvtNo/z4qOPPrroWGX32bnw3sbo6Gj3+5BdmcjISElSgwYNPOadOnVK6enpknJ/np5fb3BwsMLCwi7adl4kJSXJZrPpuuuuc88rX7686tSp4z6uF7rnnnt08uRJXX311Xr00Uc1b968HLuPXUqLFi3c/7fZbGrWrNlF2zy/TEREhEdclztvz7mS62Bez+vz36/o6GhJcr9flzuPs6sjP8+lcy537YiIiFDv3r3Vvn17de7cWRMnTlRKSop7/YEDB+qRRx5R27ZtNWbMmByvN7k5Zwr6Oneunu+//97jGMXFxUmSRz1NmzbNdZ0o+bhDFqWa1WqVzWaTKZ+7Z0X5lbfLYrPJarXmy/bXrVunAwcOqEmTJu55LpdLP/zwg/71r3/p6NGjudpWuXLl1KhRIy1btkwrV67Ubbfdpptuukn33nuvtm3bpu3bt1/0RTc4ODhf9kG6+KGiFotFWVlZ+VZ/RkaGoqOj3Ynh+fIy0lhGRoZeeeUV/e1vf7to2fkJ34XHqGPHjtq9e7e++eYbLV68WLfeeqv69evncX9aYcvp2GdkZKhp06bu5OF8FSpUyJftBwYG5rh89uzZGjx4sN566y21aNFCoaGheuONN7Rq1ap82b43evbsqdatW+vAgQNavHixAgMD1aFDB0ly30/49ddfq1KlSh7r+fv7e0xn99nJzWfg/DLnEuns5p3//uXmPC3oz583qlSpoq1bt2rJkiVavHixnnjiCb3xxhtavny5ypQpI4vFImOMxzpnzpzxSaxS3q+DV3Je5/Se51ZBnUvn5ObaMWXKFPXv318LFy7UJ598ohdeeEGLFy/W9ddfr+HDh6tHjx76+uuv9e233+rll1/W7Nmzddddd11U3+XOmQv37dz+5fd1LiMjQ507d9bYsWMvWnYu4ZXy928nij8SKZRqdrtddrtdJjZWOyZPljM1Vbrgj7wkyWKRLTJStf52lyz5lERJ0q233qoNGzZ4zHvooYcUFxenoUOHXvQFLietW7fW999/r9WrV2vUqFGKiIhQfHy8Ro0apejoaNWuXdur2M6NRHSlrW916tTRjBkzlJmZ6d6fNWvWeFVHkyZNtH//ftlsNo8BIy6MN7tYy5Qpc9H8Jk2aaOvWrapVq5ZXcUhn/zD36tVLvXr10o033qhnn332kolUdjHFx8drzpw5Msa4v+ysWLFCoaGhqly5slf7djlNmjTRJ598oooVKyosLMzr9c/Fu2LFCo95K1asUO3atWW1WtWgQQNlZWVp+fLlatu27UXrr1ixQi1bttQTTzzhnudtK2KdOnUuOmcunD4XZ69evTy2XbduXfd0y5YtVaVKFX3yySf69ttvdc8997i/oNWtW1f+/v7as2dPjq1rhelKztNzcvs5vtS56nQ6tWrVKrVs2VKSdOjQIW3dutXjuF4oMDBQnTt3VufOndWvXz/FxcVpw4YNatKkiSpUqKD95w3os3379mxbRH755RfddNNNkiSn06l169bpySefvKhM1apVJZ0dFGPbtm2Kj493x57TeZvTcZAuf7zy47zOTm7O47zw9lzK7bWjcePGaty4sYYNG6YWLVpo1qxZuv766yVJtWvXVu3atfXMM8+oe/fumjJlSraJlJTzOZNfsZ4vu/O9SZMmmjNnjqpXr85InMg1uvYBkixWqyKfH/bfCcsFC89ORz4/LF+TKEkKDQ1V/fr1PV7BwcEqX7686tevrzJlyqhMmTLq06ePhg0blmNdbdq00aJFi2Sz2dzdEdq0aaOZM2fm6YthtWrVZLFY9NVXX+ngwYPuX+y91aNHD2VlZemxxx5TUlKSFi1a5E48LBce60to27atWrRooa5du+q7777Trl279PPPP+uf//yneySo6tWrKzk5WYmJiXI4HMrMzHTPX7p0qfbv368jR45Ikl566SV99NFHeuWVV7Rp0yYlJSVp9uzZeuGFF3KM46WXXtIXX3yhHTt2aNOmTfrqq6/cX9yyU716da1atUq7du2Sw+FQVlaWnnjiCe3du1dPPfWUtmzZoi+++EIvv/yyBg4cKD+/7C/J1atXV0ZGhpYuXSqHw5HtF8/s9OzZU3a7XV26dNGPP/6o5ORkLVu2TP379891N8JBgwZp6dKlGjFihLZt26Zp06bpX//6lwYPHuyOrVevXnr44Yc1f/7/s3ffYVEc/x/A38fRjo6AFEVAqSogRQ0YBVuwEaxYiGDX2E1QY4y9obFHYyz5ARqMJdhiQyGAikYBAVEORAQxEUQUQaTIHfP7wy8bD44qiOXzeh6ex92dnfns7N65czM7e4Ir48iRIwAAMzMzxMTEICQkBHfv3sWSJUvq3ZCeNWsWzp49i82bNyM1NRW7d+/GuXPnJK6f+fPnIyAgALt27UJqaio2b96MY8eOcXFWGDNmDH755RdcvHiRG9YHvP4s+vr6Yt68eQgMDERaWhpu3ryJn376CYGBgfWKt7E09Dp9U10/x9KuMTMzM3h4eGDy5Mm4cuUKEhIS8NVXX6FVq1bw8PCQmk9AQAB+/fVX3L59G/fv38dvv/0GgUAAIyMjAECvXr2we/duxMfHIzY2FtOmTZP6WoGdO3fi+PHjSE5OxowZM5CXl4cJEyZIpFm5ciXCwsJw+/ZtjBs3Dtra2txLV2u7bqvTsmVLCAQCnD9/Ho8fP0Z+fr7UdI1xXUtT1+u4vup7LdX23ZGeno5Fixbh2rVrePDgAS5cuIDU1FRYWVmhuLgYM2fOREREBB48eICoqChER0dX+11Z2zVTm4Z8z0n7bp4xYwaePXuG0aNHIzo6GmlpaQgJCcH48eMbbTg/+Qi9kye23nM02cSHr7Ee7swPCakye99dF1eWX82D+U1B2kOwLi4uzMfHp8b9nj59yng8HvfwPGOMHT9+nAFgv/zyi0TayjPiMSb9IeuVK1cyPT09xuPxuPKlTTZROS9bW1uJWeiioqKYjY0Nk5eXZw4ODuzgwYMMgNQZj6rLt6CggM2aNYsZGBgwOTk5ZmhoyLy8vFhmZiZj7PVkAcOGDWMaGhrcTH2MvZ6Iw9TUlMnKyjIjIyMuv/PnzzNnZ2cmEAiYmpoa69KlS5XZxSpPUrFq1SpmZWXFBAIBa9GiBfPw8GD379+v9hhSUlLYZ599xgQCAQPA0tPTGWOMRUREsM6dOzN5eXmmp6fHFi5cyMrKyqrNhzHGpk2bxrS0tBgArm7rUvdZWVnM29ubaWtrMwUFBda2bVs2efLkar/vpE1O8Mcff7D27dszOTk51qZNG/bjjz9K7FNcXMzmzZvH9PX1mby8PDM1NWX/93//xxh7fV7GjRvH1NXVmYaGBvv666/Zd999JzExSG2TTTDG2J49e1irVq2YQCBggwcPZqtXr2Z6enoSaX7++WfWtm1bJicnx8zNzSUmlqiQlJTEADAjIyOJh9MZe/0Q+9atW5mFhQWTk5NjOjo6zM3NjUVGRlZbN4xJ/8x6eHhIfGYrX0/SPm/S8m/Idfrm5A2MSf8cSyPtGnv27BkbO3YsU1dXZwKBgLm5ubG7d+9Wm8fx48dZ165dmZqaGlNWVmafffaZxOQK//77L+vTpw9TVlZmpqam7OzZs1Inmzh48CDr0qULk5eXZ+3bt5eYta6inv7880/WoUMHJi8vz7p06cISEhIkYqntupX2+WHs9WQ4hoaGTEZGhrm4uEg9zrpc15VJO79xcXES3w2M1X4dv6trqabvjuzsbDZ48GDuM29kZMSWLl3KxGIxKy0tZaNGjWKGhoZMXl6eGRgYsJkzZ1b7/3Nt10xjfM9V/o6p7rv57t27bMiQIUxDQ4MJBAJmaWnJ5s6dy31XSPusV0aTTXwc6to24DEmbRzTp6WgoADq6urIz89v8PAX0rxKSkqQnp4OExOTt5q0AQCYWIzkY8dR/jQXMlrasGzk4XwECAoK4t5bU9szNoRIM3nyZCQnJ+Py5cvNHQqpp+LiYu7flT//GRkZMDExQVxcHDp16iR1/4iICPTs2RN5eXkNekaSkKbUmPcjFZKTkyESiSRGnJCmVde2AQ0CJaQSHp8PnnVHMJEIPFlZakQ1gv3796Nt27Zo1aoVEhISsHDhQnh6elIjitTZxo0b0bdvXygrK+PcuXMIDAzEzz//3NxhEUII+YRRQ4oQKbS1tSEWixttdr5PXXZ2NpYuXYrs7Gzo6+tjxIgRWLNmTXOHRT4gN27cwIYNG/DixQu0bdsW27dvx6RJk5o7LEIIIZ8wGtoHGtr3MWiKrnRCCCGNr6ahfYR86Gho38ehrm0DmrWPEEIIIYQQQuqJGlKEEEIIIYQQUk/UkCKEEEIIIYSQeqKGFCGEEEIIIYTUEzWkCCGEEEIIIaSeaPpzQqTIzc3lpj/X1tZu7nAIIYQQQsh7hnqkCJEiNzcXT548QW5ubnOHUicRERHg8Xh4/vz5W+XD4/Fw4sSJd14uaVz1PY8fgnHjxmHw4MFvlUdAQAA0NDQatO+JEydgamoKPp+PuXPnvlVeDbF8+XJ06tTpnZXXFDIyMsDj8ZCQkNDcobwVY2NjbN26tbnDIIS8B6ghRUgl5eUMeQ9LkXO3BHkPS1Fe3rSvWlu+fDl4PJ7E34fynghnZ2dkZWVBXV29uUMhH4mKm+34+PhGz3vkyJG4e/dug/adOnUqhg8fjocPH2LVqlUNyqOuN+DSGsK+vr4ICwtrULnNoTEavoQQ8r6joX2EvCEtLgeXD6fi5fNSbl1K2FV0H2mGdnYtm6zcDh06IDQ0lFuWlf0wPpry8vLQ09Nr7jAIqROBQFDjC2BfvXoFeXn5KusLCwuRk5MDNzc3GBgYNGWI1VJRUYGKikq126uL/VNEdUEIeVeoR4qQ/0mLy8H53bclGlEA8PJ5Kc7vvo20uJwmK1tWVhZ6enrcX0Ofy4qNjYWjoyOUlJTg7OyMlJQUie0nT56Evb09FBUV0bZtW6xYsQIikUhqXhU9A4cOHYKzszMUFRXRsWNHREZGcmkqD+2rGO4UEhICKysrqKiooF+/fsjKyuL2EYlEmD17NjQ0NKClpYWFCxfCx8enxl+vHzx4AHd3d2hqakJZWRkdOnTA2bNnue2RkZHo0qULFBQUoK+vj++++07iuFxdXTFr1izMnTsXmpqa0NXVxd69e/Hy5UuMHz8eqqqqMDU1xblz5yTKvX37Nvr37w8VFRXo6upi7NixEsM9S0tLMXv2bLRs2RKKior4/PPPER0dXaV+wsLCGu28VPi///s/dOjQgTvmmTNnSmzPzc3FkCFDoKSkBDMzM5w6dYrbJhaLMXHiRJiYmEAgEMDCwgLbtm2T2L+iR2Hjxo3Q19eHlpYWZsyYgbKyMi5NVlYWBg4cCIFAABMTExw8eLBKr8vz588xadIk6OjoQE1NDb169apxaJeJiQkAwM7ODjweD66urhLba4qntLQUvr6+aNWqFZSVldG1a1dERERw2ysPx6sYLrdv3z6YmJhAUVGxSjwRERFQVVUFAPTq1Qs8Hk8izwppaWnw8PCArq4uVFRU0LlzZ4kfR1xdXfHgwQPMmzeP63mWxtjYGAAwZMgQ8Hg8brny0L6K87NmzRoYGBjAwsICAPDw4UN4enpCQ0MDLVq0gIeHBzIyMqSWVaEun5/Zs2djwYIFaNGiBfT09LB8+fJq81u+fDkCAwNx8uRJ7ljfrLP09HT069cPWlpasLW1xbVr1yT2v3LlCrp37w6BQABDQ0PMnj0bL1++rLE8aecxMzMTHh4eUFFRgZqaGjw9PfH48WNuv9rOGQDk5OTA3d2du8aDgoJqrEtCyKeFGlKE4PVwvsuHU2tMc+VIapMN80tNTYWBgQHatm0LLy8vZGZmNiifxYsXY9OmTYiJiYGsrCwmTJjAbbt8+TK8vb0xZ84cJCUlYffu3QgICMCaNWtqzHP+/Pn49ttvERcXBycnJ7i7u+Pp06fVpi8qKsLGjRtx4MABXLp0CZmZmfD19eW2r1+/HkFBQfD390dUVBQKCgpqfZ5nxowZKC0txaVLl5CYmIj169dzv87/+++/GDBgADp37oyEhATs2rULv/76K1avXi2RR2BgILS1tXHjxg3MmjULX3/9NUaMGAFnZ2fcvHkTX3zxBcaOHYuioiIArxsAvXr1gp2dHWJiYnD+/Hk8fvwYnp6eXJ4LFixAcHAwAgMDcfPmTZiamsLNzQ3Pnj1r0vOya9cuzJgxA1OmTEFiYiJOnToFU1NTiTQrVqyAp6cnbt26hQEDBsDLy4uLq7y8HK1bt8bRo0eRlJSEpUuX4vvvv8eRI0ck8ggPD0daWhrCw8MRGBiIgIAABAQEcNu9vb3x6NEjREREIDg4GHv27EFOjuQPDiNGjEBOTg7OnTuH2NhY2Nvbo3fv3lXqqMKNGzcAAKGhocjKysKxY8fqHM/MmTNx7do1HDp0CLdu3cKIESPQr18/pKZW/9m+d+8egoODcezYManDCd9s+AYHByMrKwvOzs5V0hUWFmLAgAEICwtDXFwc+vXrB3d3d+6zfOzYMbRu3RorV65EVlaWxI8Lb6poiPv7+yMrK0uiYV5ZWFgYUlJScPHiRZw+fRplZWVwc3ODqqoqLl++jKioKO7HjFevXknNoz6fH2VlZVy/fh0bNmzAypUrcfHiRal5+vr6wtPTk/sRpXKdrVixAnPmzMHff/8Nc3NzjB49mmu4paWloV+/fhg2bBhu3bqFw4cP48qVK1V+KKis8nksLy+Hh4cHnj17hsjISFy8eBH379/HyJEjuX1qO2fA6wbrw4cPER4ejj/++AM///xzlWucEPIJY4Tl5+czACw/P7+5QyENVFxczJKSklhxcXG99nvy5AkTCoXs6oV4tmNqWK1/Vy/EM6FQyJ48edJosZ89e5YdOXKEJSQksPPnzzMnJyfWpk0bVlBQUOc8wsPDGQAWGhrKrTtz5gwDwNVJ79692dq1ayX2O3DgANPX1+eWAbDjx48zxhhLT09nAJifnx+3vaysjLVu3ZqtX79eoty8vDzGGGP+/v4MALt37x63z86dO5muri63rKury3788UduWSQSsTZt2jAPD49qj8/a2potX75c6rbvv/+eWVhYsPLycokyVVRUmFgsZowx5uLiwj7//HOJMpWVldnYsWO5dVlZWQwAu3btGmOMsVWrVrEvvvhCoqyHDx8yACwlJYUVFhYyOTk5FhQUxG1/9eoVMzAwYBs2bJCon7c9L5UZGBiwxYsXV7sdAPvhhx+45cLCQgaAnTt3rtp9ZsyYwYYNG8Yt+/j4MCMjIyYSibh1I0aMYCNHjmSMMSYUChkAFh0dzW1PTU1lANiWLVsYY4xdvnyZqampsZKSEomy2rVrx3bv3i01jorrLi4uTmJ9bfE8ePCA8fl89u+//0rs17t3b7Zo0SLG2OvrU11dndu2bNkyJicnx3JycqqtF8YYy8vLYwBYeHg4t65yXtJ06NCB/fTTT9yykZERVzc1efNz+Gastra23LKPjw/T1dVlpaWl3LoDBw5U+SyUlpYygUDAQkJCpJbVkM8PY4x17tyZLVy4sNpj8PHxqfKZrji3P//8MysqKmJFRUXszp07DAATCoWMMcYmTpzIpkyZIrHf5cuXmYyMTLXf79LO44ULFxifz2eZmZncuoqybty4UW3cb56zlJSUKukrrvu6nEfyaWro/UhNhEIhS0xM5D4npOnVtW3wYTyIQUgTEYvFEIlEKH5RVntiAMUvyqAq4kMsFjdaDP379+f+bWNjg65du8LIyAhHjhzBxIkT65WXjY0N9299fX0Ar4emtGnTBgkJCYiKipLo6RCLxSgpKUFRURGUlJSk5unk5MT9W1ZWFo6OjhAKhdXGoKSkhHbt2knEUfELbn5+Ph4/fowuXbpw2/l8PhwcHFBeXl5tnrNnz8bXX3+NCxcuoE+fPhg2bBh3rEKhEE5OThJDpbp164bCwkL8888/aNOmTZW64fP50NLSgrW1NbdOV1eXqy8ASEhIQHh4uNTnUtLS0lBSUoKysjJ069aNWy8nJ4cuXbpUqZ/GPC85OTl49OgRevfuXW19VS5TWVkZampqEr+k79y5E//3f/+HzMxMFBcX49WrV1VmhevQoQP4fL5E7ImJiQCAlJQUyMrKwt7enttuamoKTU1NbjkhIQGFhYXQ0tKSyLe4uBhpaWk1xi9NTfEkJiZCLBbD3NxcYp/S0tIq5b/JyMgIOjo69Y6lssLCQixfvhxnzpxBVlbW6++V4uIG9y7XhbW1tcSzQAkJCbh37x43FLFCSUlJtfXdkM8PIPm5rq+OHTtK5AO8vq4tLS2RkJCAW7duSQyhY4yhvLwc6enpsLKykppn5fMoFAphaGgIQ0NDbl379u2hoaEBoVCIzp0713rOhEIhZGVl4eDgwOVhaWn5TmdrJIS836ghRT5pfD4fsrKyEKjK1Sm9QFUOsrKyEjdzjU1DQwPm5ua4d+9evfeVk/vvOCpujCoaKIWFhVixYgWGDh1aZT9pz4Y01JsxVMTB2NsNiZw0aRLc3Nxw5swZXLhwAevWrcOmTZswa9ast4qrtvpyd3fH+vXrq+Slr69fr4ZAY56XmiZLqK7MinIryjx06BB8fX2xadMmODk5QVVVFT/++COuX79e5zzqorCwEPr6+lKfKWrIzWhN8RQWFoLP5yM2NrbK57OmSRqUlZXrHYc0vr6+uHjxIjZu3AhTU1MIBAIMHz682iF1jaFy7IWFhXBwcJD6HM/bNhbf9lqoLi9pn4epU6di9uzZVfaraNRJ05Dz2BznjBDycaGGFPmkaWtrQ1tbG+XmDClhV6tMNPEmFU0FdO1tAxkZ6Q+JN5bCwkKkpaVh7NixjZqvvb09UlJSqjxLU5u///4bPXr0APB6oojY2Nhan1eojrq6OnR1dREdHc3lKRaLcfPmzVrfkWNoaIhp06Zh2rRpWLRoEfbu3YtZs2bBysoKwcHBYIxxN2VRUVFQVVVF69atGxQn8Lq+goODYWxsLHUWxXbt2kFeXh5RUVEwMjICAJSVlSE6Ohpz586tVzn1OS+qqqowNjZGWFgYevbsWedy3hQVFQVnZ2dMnz6dW1ffHiILCwuIRCLExcVxv9jfu3cPeXl5XBp7e3tkZ2dDVlaWmzShNhU9LPXt9bWzs4NYLEZOTg66d+9er30bQ1RUFMaNG4chQ4YAeP05rjzJg7y8fJ2OS05OrkG93vb29jh8+DBatmwJNTW1Ou3TVJ+fuh5rZfb29khKSqr391RlVlZWePjwIR4+fMj1SiUlJeH58+do3749gNrPmaWlJfed17lzZwCve2LpvXmEkAo02QQhAGRkeOg+0qzGNJ97mjVJI8rX1xeRkZHIyMjA1atXMWTIEPD5fIwePZpL4+3tjUWLFr1VOUuXLsX+/fuxYsUK3LlzB0KhEIcOHcIPP/xQ4347d+7E8ePHkZycjBkzZiAvL09isoT6mjVrFtatW4eTJ08iJSUFc+bMQV5eXrWzmAHA3LlzERISgvT0dNy8eRPh4eHcEJ/p06fj4cOHmDVrFpKTk3Hy5EksW7YM33zzDWRkGv4VN2PGDDx79gyjR49GdHQ00tLSEBISgvHjx0MsFkNZWRlff/015s+fj/PnzyMpKQmTJ09GUVFRvYZkNuS8LF++HJs2bcL27duRmpqKmzdv4qeffqpzmWZmZoiJiUFISAju3r2LJUuW1DipgTSWlpbo06cPpkyZghs3biAuLg5TpkyBQCDgzmWfPn3g5OSEwYMH48KFC9w1vnjxYsTExEjNt2XLlhAIBNzkHvn5+XWKx9zcHF5eXvD29saxY8eQnp6OGzduYN26dThz5ky9jq0hzMzMuIkOEhISMGbMmCo9NsbGxrh06RL+/fffGl/2XdFQzs7OlmiY1sbLywva2trw8PDA5cuXkZ6ejoiICMyePRv//POP1H2a6vNjbGyMW7duISUlBbm5uRKzK9Zk4cKFuHr1KmbOnIn4+Hikpqbi5MmT9f7xpk+fPrC2toaXlxdu3ryJGzduwNvbGy4uLnB0dARQ+zmzsLBAv379MHXqVFy/fh2xsbGYNGlSnXuFCSEfP2pIEfI/7exaot/UjlDWUJBYr6KpgH5TOzbZe6T++ecfjB49GhYWFvD09ISWlhb+/vtviaE4mZmZ1c7yVVdubm44ffo0Lly4gM6dO+Ozzz7Dli1buN6U6vj5+cHPzw+2tra4cuUKTp061eDp2YHXN0qjR4+Gt7c3nJycoKKiAjc3txqHF4rFYsyYMQNWVlbo168fzM3N8fPPPwMAWrVqhbNnz+LGjRuwtbXFtGnTMHHixFobiLUxMDBAVFQUxGIxvvjiC1hbW2Pu3LnQ0NDgbjD9/PwwbNgwjB07Fvb29rh37x5CQkIknhOqTUPOi4+PD7Zu3Yqff/4ZHTp0wKBBg2qcma6yqVOnYujQoRg5ciS6du2Kp0+fSvRO1dX+/fuhq6uLHj16YMiQIZg8eTJUVVW5c8nj8XD27Fn06NED48ePh7m5OUaNGoUHDx5wz6RVJisri+3bt2P37t0wMDCAh4dHnePx9/eHt7c3vv32W1hYWGDw4MGIjo6ucUhYY9m8eTM0NTXh7OwMd3d3uLm5STw/BgArV65ERkYG2rVrV+NQu02bNuHixYswNDSEnZ1dnWNQUlLCpUuX0KZNGwwdOhRWVlaYOHEiSkpKqu2haqrPz+TJk2FhYQFHR0fo6OggKiqqTvvZ2NggMjISd+/eRffu3WFnZ4elS5fW+/1dPB4PJ0+ehKamJnr06IE+ffqgbdu2OHz4MJemLufM398fBgYGcHFxwdChQzFlyhS0bNl07xQkhHxYeOxtH174CBQUFEBdXR35+fl1Hg5B3i8lJSVIT0+v9l0w9VFeznA97BaKX5RBoCr3TobzvY8yMjJgYmKCuLi4WofdvY3y8nJYWVnB09MTq1atarJySNP7559/YGhoiNDQ0FonwyCfruLiYu7f1LtDPjaNeT9SITk5GSKRCLKysrC0tGyUPEnN6to2oGekCKlERoYHTUMFqIpeT0TxKTaimtKDBw9w4cIFuLi4oLS0FDt27EB6ejrGjBnT3KGRevrrr79QWFgIa2trZGVlYcGCBTA2NuaefyOEEEI+ZtSQIkQKbW1tiMXiJp2d71MlIyODgIAA+Pr6gjGGjh07IjQ0tNppjcn7q6ysDN9//z3u378PVVVVODs7IygoqMoMb4QQQsjHiBpShEjxNs8AfSyMjY3fetpyaQwNDev8vAR5v7m5ucHNza25wyCEEEKaBU02QQghhBBCCCH1RA0pQgghhBBCCKknakgRQgghhBBCSD1RQ4oQQgghhBBC6okaUoQQQgghhBBSTzRrHyFS5ObmctOf0wx+hBBCCCGkMuqRIkSK3NxcPHnyBLm5uc0dSqNYvnw5OnXq9M7LdXV1xdy5c995uY3F2NgYW7du5ZZ5PB5OnDjRbPEAzXcum1vlc/GuNKS+63LdN9W1lJGRAR6Ph/j4+EbPuzYBAQHQ0NB45+USQkhzoYYUIZWUl4vxNP0esm7H42n6PZSXi5u8zH///RdfffUVtLS0IBAIYG1tjZiYmCYvt6kdO3YMq1atau4wGk1WVhb69+8PoPob1nHjxmHw4MFNFoOvry/CwsKaLH8iieqbEEJIdWhoHyFvSL1+FX8F7EHhs/96om6f+gO9xk2BWVfnJikzLy8P3bp1Q8+ePXHu3Dno6OggNTUVmpqaTVLeu9SiRYvmDqFR6enpvbOyXr16BXl5+SrrVVRUoKKi8s7iqE518X1s3pf6rotXr141dwiEEPJJoR4pQv4n9fpVnNq8VqIRBQCFz3JxavNapF6/2iTlrl+/HoaGhvD390eXLl1gYmKCL774Au3atatXPv/88w9Gjx6NFi1aQFlZGY6Ojrh+/Xq16fft2wcrKysoKirC0tISP//8s8T2hQsXwtzcHEpKSmjbti2WLFmCsrIybnvFkKcDBw7A2NgY6urqGDVqFF68eMGlqTzEydjYGGvXrsWECROgqqqKNm3aYM+ePRLlXr16FZ06dYKioiIcHR1x4sSJWocqSRsmpaGhgYCAAAD/9R4dOnQIzs7OUFRURMeOHREZGVltnrWVY2JiAgCws7MDj8eDq6srli9fjsDAQJw8eRI8Hg88Hg8REREAgIcPH8LT0xMaGhpo0aIFPDw8kJGRweVd0ZO1Zs0aGBgYwMLCQmoMlYeaVey3ceNG6OvrQ0tLCzNmzJA4V5WlpaXBw8MDurq6UFFRQefOnREaGlrjsVeUu2/fPpiYmEBRUREAkJmZCQ8PD6ioqEBNTQ2enp54/PhxlfjeNHfuXLi6unLLL168gJeXF5SVlaGvr48tW7ZIHR5XVFRU47VTmaurK2bPno0FCxagRYsW0NPTw/LlyyXSPH/+HJMmTYKOjg7U1NTQq1cvJCQkVDnuCiKRCLNnz4aGhga0tLSwcOFC+Pj4VDnG8vLyGssF/uvhFAgEaNu2Lf744w+J7YmJiejVqxcEAgG0tLQwZcoUFBYWcttrumbu37+Pnj17QklJCba2trh27ZpE3sHBwejQoQMUFBRgbGyMTZs2SWzPy8uDt7c3NDU1oaSkhP79+yM1NVUiTUBAANq0aQMlJSUMGTIET58+rXKMhBDyMaOGFCF4PZzvr4Cab8rCA/c0yTC/U6dOwdHRESNGjEDLli1hZ2eHvXv31iuPwsJCuLi44N9//8WpU6eQkJCABQsWoLy8XGr6oKAgLF26FGvWrIFQKMTatWuxZMkSBAYGcmlUVVUREBCApKQkbNu2DXv37sWWLVsk8klLS8OJEydw+vRpnD59GpGRkfDz86sx1k2bNsHR0RFxcXGYPn06vv76a6SkpAAACgoK4O7uDmtra9y8eROrVq3CwoUL61UXNZk/fz6+/fZbxMXFwcnJCe7u7g2++btx4wYAIDQ0FFlZWTh27Bh8fX3h6emJfv36ISsrC1lZWXB2dkZZWRnc3NygqqqKy5cvIyoqCioqKujXr59EL0JYWBhSUlJw8eJFnD59us6xhIeHIy0tDeHh4QgMDERAQADXiJSmsLAQAwYMQFhYGOLi4tCvXz+4u7sjMzOzxnLu3buH4OBgHDt2DPHx8SgvL4eHhweePXuGyMhIXLx4Effv38fIkSPrHDsAfPPNN4iKisKpU6dw8eJFXL58GTdv3qySrqZrpzqBgYFQVlbG9evXsWHDBqxcuRIXL17kto8YMQI5OTk4d+4cYmNjYW9vj969e+PZs2dS81u/fj2CgoLg7++PqKgoFBQUSH3WqbZyAWDJkiUYNmwYEhIS4OXlhVGjRkEoFAIAXr58CTc3N2hqaiI6OhpHjx5FaGgoZs6cKZFHddfM4sWL4evri/j4eJibm2P06NEQiUQAgNjYWHh6emLUqFFITEzE8uXLsWTJEolrZty4cYiJicGpU6dw7do1MMYwYMAAroF+/fp1TJw4ETNnzkR8fDx69uyJ1atX13guKsjKynJ/hBDyQWOE5efnMwAsPz+/uUMhDVRcXMySkpJYcXFxvfZ78uQJEwqF7MrZP9lGz4G1/l05+ycTCoXsyZMnjRa7goICU1BQYIsWLWI3b95ku3fvZoqKiiwgIKDOeezevZupqqqyp0+fSt2+bNkyZmtryy23a9eOHTx4UCLNqlWrmJOTU7Vl/Pjjj8zBwUEiTyUlJVZQUMCtmz9/PuvatSu37OLiwubMmcMtGxkZsa+++opbLi8vZy1btmS7du1ijDG2a9cupqWlJXEe9+7dywCwuLi4amMDwI4fPy6xTl1dnfn7+zPGGEtPT2cAmJ+fH7e9rKyMtW7dmq1fv77afI2MjNiWLVukllORZ+W4fHx8mIeHh8S6AwcOMAsLC1ZeXs6tKy0tZQKBgIWEhHD76erqstLS0mrjYazqufTx8WFGRkZMJBJx60aMGMFGjhxZYz6VdejQgf300081lisnJ8dycnK4dRcuXGB8Pp9lZmZy6+7cucMAsBs3bnDxVa6POXPmMBcXF8YYYwUFBUxOTo4dPXqU2/78+XOmpKRUr2tHGhcXF/b5559LrOvcuTNbuHAhY4yxy5cvMzU1NVZSUiKRpl27dmz37t3ccb9Z37q6uuzHH3/klkUiEWvTpo3EMdZWLmOvr6Vp06ZJpOnatSv7+uuvGWOM7dmzh2lqarLCwkJu+5kzZ5iMjAzLzs5mjEm/Ziquy3379nHrKs6JUChkjDE2ZswY1rdvX4my58+fz9q3b88YY+zu3bsMAIuKiuK25+bmMoFAwI4cOcIYY2z06NFswIABEnmMHDmSqaurM0I+ZQ29H6mJUChkiYmJ3GeYNL26tg2oR4p80sRiMUQiEYryn9cpfVH+c4hEIojFjdczVV5eDnt7e6xduxZ2dnaYMmUKJk+ejF9++aXOecTHx8POzq5OzyS9fPkSaWlpmDhxIvf8h4qKClavXo20tDQu3eHDh9GtWzfo6elBRUUFP/zwQ5UeC2NjY6iqqnLL+vr6yMnJqbF8Gxsb7t88Hg96enrcPikpKbCxseGGjQFAly5daj2munJycuL+LSsrC0dHR64HoCklJCTg3r17UFVV5eq7RYsWKCkpkahza2vrBj131KFDB/D5fG65tvNQWFgIX19fWFlZQUNDAyoqKhAKhbX2SBkZGUFHR4dbFgqFMDQ0hKGhIbeuffv20NDQqHO93r9/H2VlZRLnWV1dXerQxpquneq8uQ8gWTcJCQkoLCyElpaWxGchPT1d4rxUyM/Px+PHjyVi5fP5cHBwqFe5Fd68HiuWK+pNKBTC1tYWysrK3PZu3bqhvLxcoheuumvmzfL19fUBgCtfKBSiW7duEum7deuG1NRUiMViCIVCyMrKomvXrtx2LS0tWFhYSMT35nZpx0MIIR876lcnnzQ+nw9ZWVkoqWvUKb2SugZkZWUlblrflr6+Ptq3by+xzsrKCsHBwXXOQyAQ1DltxTMWe/furXIjVHFc165dg5eXF1asWAE3Nzeoq6vj0KFDVZ6jkJOTk1jm8XjVDid8m31qw+PxwBiTWFfTM0LvWmFhIRwcHBAUFFRl25sNkzdvmuujvnXq6+uLixcvYuPGjTA1NYVAIMDw4cNrnaygIfHJyMg02rlp7OutsLAQ+vr63HNsb3rbabyb4jqXprpz8mb5PB4PAJqkfEII+ZRRjxT5pGlra8PS0hJObv2h0qLmF++qamnDya0/LC0tG/Ulvd26davynMfdu3dhZGRU5zxsbGwQHx9f7XMdb9LV1YWBgQHu378PU1NTib+KCRSuXr0KIyMjLF68GI6OjjAzM8ODBw/qd2ANYGFhgcTERJSWlnLroqOja91PR0cHWVlZ3HJqaiqKioqqpPv777+5f4tEIsTGxsLKyqpBsVb0AlTunZSXl6+yzt7eHqmpqWjZsmWVOldXV29Q+W8jKioK48aNw5AhQ2BtbQ09PT2JiS/qysrKCg8fPsTDhw+5dUlJSXj+/Dn340DlcwNAYuKQtm3bQk5OTuI85+fn4+7du/WOp77s7e2RnZ0NWVnZKudF2mdcXV0durq6ErGKxWKpz3PVxZvXY8VyxfVoZWWFhIQEvHz5ktseFRUFGRmZaiciqSsrKytERUVJrIuKioK5uTn4fD6srKwgEokkJqt5+vQpUlJSuPNqZWVVZTKbysdDCCEfO2pIEQJARoaPXuOm1Jimp88UyMg0Xk9UhXnz5uHvv//G2rVrce/ePRw8eBB79uzBjBkzuDSLFi2Ct7d3tXmMHj0aenp6GDx4MKKionD//n0EBwdXmamrwooVK7Bu3Tps374dd+/eRWJiIvz9/bF582YAgJmZGTIzM3Ho0CGkpaVh+/btOH78eOMeuBRjxoxBeXk5pkyZAqFQiJCQEGzcuBHAf7+qS9OrVy/s2LEDcXFxiImJwbRp06r0CADAzp07cfz4cSQnJ2PGjBnIy8vDhAkTGhRry5YtIRAIcP78eTx+/Bj5+fkAXg93vHXrFlJSUpCbm4uysjJ4eXlBW1sbHh4euHz5MtLT0xEREYHZs2fjn3/+aVD5b8PMzIybMCIhIYGr9/rq06cPrK2t4eXlhZs3b+LGjRvw9vaGi4sLHB0dAbw+NzExMdi/fz9SU1OxbNky3L59m8tDVVUVPj4+mD9/PsLDw3Hnzh1MnDgRMjIyNZ7zxtCnTx84OTlh8ODBuHDhAjIyMnD16lUsXry42ve4zZo1C+vWrcPJkyeRkpKCOXPmIC8vr0GxHj16FP/3f/+Hu3fvYtmyZbhx4wY3mYSXlxcUFRXh4+OD27dvIzw8HLNmzcLYsWOhq6v7Vsf97bffIiwsDKtWrcLdu3cRGBiIHTt2wNfXF8Dr68PDwwOTJ0/GlStXkJCQgK+++gqtWrWCh4cHAGD27Nk4f/48Nm7ciNTUVOzYsQPnz59/q7gIIeRDQw0pQv7HrKszvvzm+yo9U6pa2vjym++b7D1SnTt3xvHjx/H777+jY8eOWLVqFbZu3QovLy8uTVZWVo3Pr8jLy+PChQto2bIlBgwYAGtra/j5+VU7BHHSpEnYt28f/P39YW1tDRcXFwQEBHA9Ul9++SXmzZuHmTNnolOnTrh69SqWLFnSuAcuhZqaGv7880/Ex8ejU6dOWLx4MZYuXQoAEs9NVbZp0yYYGhqie/fuGDNmDHx9faGkpFQlnZ+fH/z8/GBra4srV67g1KlTDe5dlJWVxfbt27F7924YGBhwN5iTJ0+GhYUFHB0doaOjg6ioKCgpKeHSpUto06YNhg4dCisrK0ycOBElJSVQU1NrUPlvY/PmzdDU1ISzszPc3d3h5uYGe3v7eufD4/Fw8uRJaGpqokePHujTpw/atm2Lw4cPc2nc3NywZMkSLFiwAJ07d8aLFy+q/CiwefNmODk5YdCgQejTpw+6devGTc3flHg8Hs6ePYsePXpg/PjxMDc3x6hRo/DgwYNqGysLFy7E6NGj4e3tDScnJ6ioqMDNza1Bsa5YsQKHDh2CjY0N9u/fj99//53r8VFSUkJISAiePXuGzp07Y/jw4ejduzd27NjxVscMvO6JO3LkCA4dOoSOHTti6dKlWLlyJcaNG8el8ff3h4ODAwYNGgQnJycwxnD27FnuB4rPPvsMe/fuxbZt22Bra4sLFy7ghx9+eOvYCCHkQ8JjlQevf4IKCgqgrq6O/Pz8ZrmpIW+vpKQE6enpEu+3aajycjGuhZxDUf5zKKlrwMmtf5P0RJG6CQoKwvjx45Gfn1+vZ8HelJGRARMTE8TFxUm8E4i8n16+fIlWrVph06ZNmDhxYnOHU6Py8nJYWVnB09MTq1atau5wCCHNrDHvRyokJydDJBJBVlYWlpaWjZInqVld2wY02QQhlcjI8KFlYgr1/31pUSPq3dq/fz/atm2LVq1aISEhAQsXLoSnp2eDG1Hk/RcXF4fk5GR06dIF+fn5WLlyJQBwvXzvkwcPHuDChQtwcXFBaWkpduzYgfT0dIwZM6a5QyOEEPKOUUOKECm0tbUhFosbdXY+UjfZ2dlYunQpsrOzoa+vjxEjRmDNmjXNHRZpYhs3bkRKSgrk5eXh4OCAy5cvN+qkLo1FRkYGAQEB8PX1BWMMHTt2RGhoaIMnLSGEEPLhoqF9oKF9H4Om6EonhBBCCKkPGtr3cahr24AmmyCEEEIIIYSQevqgGlJ+fn7g8XiYO3cut66kpAQzZszg3kw/bNgwPH78uPmCJIQQQgghhHz0PpiGVHR0NHbv3g0bGxuJ9fPmzcOff/6Jo0ePIjIyEo8ePcLQoUObKUpCCCGEEELIp+CDaEgVFhbCy8sLe/fuhaamJrc+Pz8fv/76KzZv3oxevXrBwcEB/v7+uHr1Kr1hnRBCCCGEENJkPoiG1IwZMzBw4ED06dNHYn1sbCzKysok1ltaWqJNmza4du3auw6TEEIIIYQQ8ol476c/P3ToEG7evIno6Ogq27KzsyEvLw8NDQ2J9bq6usjOzq42z9LSUpSWlnLLBQUFjRYv+Tjk5uZy05+/j1MwE0IIIYSQ5vVe90g9fPgQc+bMQVBQUKNOab1u3Tqoq6tzf4aGho2WN/k45Obm4smTJ8jNzW3uUD4IERER4PF4eP78eXOH8sEKCAio8qPQu5KdnY2+fftCWVm5SWI4ceIETE1NwefzMXfu3GY9VkIIIaSxvNcNqdjYWOTk5MDe3h6ysrKQlZVFZGQktm/fDllZWejq6uLVq1dVbt4eP34MPT29avNdtGgR8vPzub+HDx828ZGQDwkrZ+BnvYJc+ivws16BlTftq9aMjY3B4/Gq/M2YMaNJy21Mzs7OyMrKgrq6enOH8kEwNjbG1q1bmzsMzpYtW5CVlYX4+HjcvXu30fOfOnUqhg8fjocPH2LVqlUYOXKkRDnLly9Hp06dGr1cQgghpCm910P7evfujcTERIl148ePh6WlJRYuXAhDQ0PIyckhLCwMw4YNAwCkpKQgMzMTTk5O1earoKAABQWFJo2dfJiKb+fi+Z9pEOS/4tZl/30DGu7tIOjYNEP8oqOjIRaLueXbt2+jb9++GDFiRIPzZIxBLBZDVvbdfMTl5eVr/PHiU/Cu67wxpaWlwcHBAWZmZg3O49WrV5CXl6+yvrCwEDk5OXBzc4OBgQG3XiAQNLgsQggh5H3wXvdIqaqqomPHjhJ/ysrK0NLSQseOHaGuro6JEyfim2++QXh4OGJjYzF+/Hg4OTnhs88+a+7wyQem+HYunv4mhPiNRhQAiPNf4elvQhTfbpphfjo6OtDT0+P+Tp8+jXbt2sHFxaXOeVQMrTt37hwcHBygoKCAK1euYNy4cRg8eLBE2rlz58LV1ZVbdnV1xezZs7FgwQK0aNECenp6WL58ucQ+PB4P+/btw5AhQ6CkpAQzMzOcOnWqSvkVvcMVQ7dCQkJgZWUFFRUV9OvXD1lZWdw+IpEIs2fPhoaGBrS0tLBw4UL4+PhUifdNT58+xejRo9GqVSsoKSnB2toav//+e411UxHL6dOnYWFhASUlJQwfPhxFRUUIDAyEsbExNDU1MXv2bIkG7YEDB+Do6AhVVVXo6elhzJgxyMnJqbXO09LS4OHhAV1dXaioqKBz584IDQ2VqO8HDx5g3rx5XO/jm2qqs/LycqxcuRKtW7eGgoICOnXqhPPnz3PbMzIywOPxcOzYMfTs2RNKSkqwtbWtcfIdY2NjBAcHY//+/eDxeBg3bhwAIDMzEx4eHlBRUYGamho8PT0l3tFX0Yu0b98+mJiYSB1+HRERAVVVVQBAr169wOPxEBERITG0LyAgACtWrEBCQgJXHwEBAQBqv+6A1z889O/fHyoqKtDV1cXYsWMlhuT+8ccfsLa2hkAggJaWFvr06YOXL19y8XXp0oUb0titWzc8ePCg2roihBBC3vReN6TqYsuWLRg0aBCGDRuGHj16QE9PD8eOHWvusMgHhpUzPP8zrcY0z/+83+TD/F69eoXffvsNEyZMqHKDXRffffcd/Pz8IBQKq7xzrSaBgYFQVlbG9evXsWHDBqxcuRIXL16USLNixQp4enri1q1bGDBgALy8vPDs2bNq8ywqKsLGjRtx4MABXLp0CZmZmfD19eW2r1+/HkFBQfD390dUVBQKCgpw4sSJGuMsKSmBg4MDzpw5g9u3b2PKlCkYO3Ysbty4UeN+RUVF2L59Ow4dOoTz588jIiICQ4YMwdmzZ3H27FkcOHAAu3fvxh9//MHtU1ZWhlWrViEhIQEnTpxARkYG18h4U+U6LywsxIABAxAWFoa4uDj069cP7u7uyMzMBAAcO3YMrVu3xsqVK5GVlSXRUKqtzrZt24ZNmzZh48aNuHXrFtzc3PDll18iNTVVIqbFixfD19cX8fHxMDc3x+jRoyESiaTWTXR0NPr16wdPT09kZWVh27ZtKC8vh4eHB549e4bIyEhcvHgR9+/fx8iRIyX2vXfvHoKDg3Hs2DHEx8dXydvZ2RkpKSkAgODgYGRlZcHZ2VkizciRI/Htt9+iQ4cOXH28WU5N193z58/Rq1cv2NnZISYmBufPn8fjx4/h6ekJAMjKysLo0aMxYcIECIVCREREYOjQoWCMQSQSYfDgwXBxccGtW7dw7do1TJkypUGfO0IIIZ8oRlh+fj4DwPLz85s7FNJAxcXFLCkpiRUXF9drvydPnjChUMju/pXAHi68VOvf3b8SmFAoZE+ePGmS4zh8+DDj8/ns33//rdd+4eHhDAA7ceKExHofHx/m4eEhsW7OnDnMxcWFW3ZxcWGff/65RJrOnTuzhQsXcssA2A8//MAtFxYWMgDs3LlzEuXn5eUxxhjz9/dnANi9e/e4fXbu3Ml0dXW5ZV1dXfbjjz9yyyKRiLVp06ZKvLUZOHAg+/bbb6vdLi2WqVOnMiUlJfbixQtunZubG5s6dWq1+URHRzMA3D7V1bk0HTp0YD/99BO3bGRkxLZs2VJrnJXrzMDAgK1Zs0Ziv86dO7Pp06czxhhLT09nANi+ffu47Xfu3GEAmFAorDY+Dw8P5uPjwy1fuHCB8fl8lpmZWSWfGzduMMYYW7ZsGZOTk2M5OTk1HnteXh4DwMLDwyWOVV1dnVtetmwZs7W1rbJvbdfdqlWr2BdffCGxz8OHDxkAlpKSwmJjYxkAlpGRUSXvp0+fMgAsIiKixvgJIaQ+Gno/UhOhUMgSExNr/B4njauubYMPvkeKkLchFoshEolQXij91/rKygtFEIlEEkPAGtOvv/6K/v37SzxLUh+Ojo4N2q9y75W+vr7EMLbKaZSVlaGmplYlzZuUlJTQrl07qXnm5+fj8ePH6NKlC7edz+fDwcGhxjjFYjFWrVoFa2trtGjRAioqKggJCeF6e+oai66uLoyNjaGioiKx7s3jiY2Nhbu7O9q0aQNVVVVuqGXlsirXeWFhIXx9fWFlZQUNDQ2oqKhAKBTWGqO0ON+ss4KCAjx69AjdunWT2Kdbt24QCoUS6948V/r6+gBQ47mqTCgUwtDQUGJG0/bt20NDQ0OiLCMjI+jo6NQ534ao6bpLSEhAeHg4VFRUuD9LS0sAr5/7srW1Re/evWFtbY0RI0Zg7969yMvLAwC0aNEC48aNg5ubG9zd3bFt2zaJ3kFCCCGkNtSQIp80Pp8PWVlZyKjUbYIAGZXXs0fy+fxGj+XBgwcIDQ3FpEmTGpyHsrKyxLKMjAwYkxyOWFZWVmU/OTk5iWUej4fy8vJ6p6ktfeVY6uvHH3/Etm3bsHDhQoSHhyM+Ph5ubm549epVjftJi6Wm43n58iXc3NygpqaGoKAgREdH4/jx4wBQpazKde7r64vjx49j7dq1uHz5MuLj42FtbV1rjNXF2ZA6ezOfiqFqNZ2rhqp87E2hpvNUWFgId3d3xMfHS/ylpqaiR48e4PP5uHjxIs6dO4f27dvjp59+goWFBdLT0wEA/v7+uHbtGpydnXH48GGYm5vj77//bvJjIoQQ8nGghhT5pGlra8PS0hKmLtbgq1edcexNfHUFmLpYw9LSskle0uvv74+WLVti4MCBjZanjo5OlV/ZpT3L8q6pq6tDV1dX4kXbYrEYN2/erHG/qKgoeHh44KuvvoKtrS3atm3bJNN1Jycn4+nTp/Dz80P37t1haWlZ5x6dqKgojBs3DkOGDIG1tTX09PSQkZEhkUZeXr7evZpqamowMDBAVFRUlfLat29fr7xqY2VlhYcPH0q8GiIpKQnPnz9v9LKAhtUHANjb2+POnTswNjaGqampxF9FI4/H46Fbt25YsWIF4uLiIC8vzzWKAcDOzg6LFi3C1atX0bFjRxw8eLDRjosQQsjHjRpShADgyfCg4d6uxjQa7m3Bk2maB9HLy8vh7+8PHx8fqdNne3t7Y9GiRfXOt1evXoiJicH+/fuRmpqKZcuW4fbt240R8lubNWsW1q1bh5MnTyIlJQVz5sxBXl5ejQ/7m5mZ4eLFi7h69SqEQiGmTp0qMZNcY2nTpg3k5eXx008/4f79+zh16hRWrVpVp33NzMy4yRcSEhIwZsyYKr1BxsbGuHTpEv799996vfR5/vz5WL9+PQ4fPoyUlBR89913iI+Px5w5c+p1fLXp06cPrK2t4eXlhZs3b+LGjRvw9vaGi4tLg4eP1sTY2Bjp6emIj49Hbm4uSktL67TfjBkz8OzZM4wePRrR0dFIS0tDSEgIxo8fD7FYjOvXr2Pt2rWIiYlBZmYmjh07hidPnsDKygrp6elYtGgRrl27hgcPHuDChQtITU2FlZVVox8fIYSQjxM1pAj5H0FHbWh9ZVWlZ4qvrgCtr6ya7D1SABAaGorMzExMmDBB6vbMzMwGPb/h5uaGJUuWYMGCBejcuTNevHgBb2/vtw23USxcuBCjR4+Gt7c3nJycoKKiAjc3N6nTaFf44YcfYG9vDzc3N7i6ukJPT6/G6dIbSkdHBwEBATh69Cjat28PPz8/bNy4sU77bt68GZqamnB2doa7uzvc3Nxgb28vkWblypXIyMhAu3bt6vWM0ezZs/HNN9/g22+/hbW1Nc6fP49Tp0691fufpOHxeDh58iQ0NTXRo0cP9OnTB23btsXhw4cbtZwKw4YNQ79+/dCzZ0/o6OjUOqV9hYoeOrFYjC+++ALW1taYO3cuNDQ0ICMjAzU1NVy6dAkDBgyAubk5fvjhB2zatAn9+/eHkpISkpOTMWzYMJibm2PKlCmYMWMGpk6d2iTHSAgh5OPDY2/70MJHoKCgAOrq6sjPz4eamlpzh/PJyc3NhVgsBp/Pb/CQuZKSEqSnp1f7Ppv6YOUM9yITUV4ogoyKLExdrJusJ4r8p7y8HFZWVvD09Kxz7w8hhBDyPmnM+5EKycnJEIlEkJWV5SbUIU2rrm2Duj1hT0gTys3N5b4gmuLZo/riyfAg1peHSCQDWVlZakQ1kYrhVC4uLigtLcWOHTuQnp6OMWPGNHdohBBCCCG1ooYUIVJoa2tzvWSkacjIyCAgIAC+vr5gjKFjx44IDQ2lZ1QIIYQQ8kGghhQhUrwPPWMfO0NDwyoz0BFCCCGEfChosglCCCGEEEIIqSdqSBFCCCGEEEJIPVFDihBCCCGEEELqiRpShBBCCCGEEFJP1JAihBBCCCGEkHqiWfsIkaIxXhJMCCGEEEI+XtQjRYgUubm5ePLkCXJzc5s7lHdq3LhxGDx48Fvns3z5cnTq1Omt8/kQVK4zV1dXzJ07t9niIYQQQsi7QQ0pQiopLy/H48ePkZmZicePH6O8vLxJyxOLxViyZAlMTEwgEAjQrl07rFq1CoyxJi1Xmm3btiEgIKBe+/B4PJw4cUJina+vL8LCwhovMEIIIYSQ9wwN7SPkDUlJSTh//jwKCgq4dbGxsejXrx/at2/fJGWuX78eu3btQmBgIDp06ICYmBiMHz8e6urqmD17dpOUWR11dfVGyUdFRQUqKiqNkhchhBBCyPuIeqQI+Z+kpCQcOXJEohEFAAUFBThy5AiSkpKapNyrV6/Cw8MDAwcOhLGxMYYPH44vvvgCN27cqHMeGRkZ4PF4OHToEJydnaGoqIiOHTsiMjKSSyMWizFx4kSu58vCwgLbtm2TyEfaMLXZs2djwYIFaNGiBfT09LB8+XJuu7GxMQBgyJAh4PF43HLloX0V+W7cuBH6+vrQ0tLCjBkzUFZWxqXJysrCwIEDIRAIYGJigoMHD8LY2Bhbt26t9rijo6PRt29faGtrQ11dHS4uLrh582a16W/fvg0ZGRk8efIEAPDs2TPIyMhg1KhRXJrVq1fj888/r3OdEUIIIeTTRA0pQvB6ON/58+drTHP+/PkmGebn7OyMsLAw3L17FwCQkJCAK1euoH///vXOa/78+fj2228RFxcHJycnuLu74+nTpwBeH2Pr1q1x9OhRJCUlYenSpfj+++9x5MiRGvMMDAyEsrIyrl+/jg0bNmDlypW4ePEigNcNGQDw9/dHVlYWtyxNeHg40tLSEB4ejsDAQAQEBEgMI/T29sajR48QERGB4OBg7NmzBzk5OTXG9uLFC/j4+ODKlSv4+++/YWZmhgEDBuDFixdS03fo0AFaWlpcA/Py5csSywAQGRkJV1fXt6ozQgghhHz8aGgf+aTl5uYiNzcXjx8/rtITVVlBQQEuX74MXV1daGtrN9psft999x0KCgpgaWkJPp8PsViMNWvWwMvLq955zZw5E8OGDQMA7Nq1C+fPn8evv/6KBQsWQE5ODitWrODSmpiY4Nq1azhy5Ag8PT2rzdPGxgbLli0DAJiZmWHHjh0ICwtD3759oaOjAwDQ0NCAnp5ejbFpampix44d4PP5sLS0xMCBAxEWFobJkycjOTkZoaGhiI6OhqOjIwBg3759MDMzqzHPXr16SSzv2bMHGhoaiIyMxKBBg6qk5/F46NGjByIiIjB8+HBERERg/Pjx2LdvH5KTk9GuXTtcvXoVCxYsAIAG1xkhhBBCPn7UI0U+aWKxGCKRCC9fvqxT+pcvX0IkEkEsFjdaDEeOHEFQUBAOHjyImzdvIjAwEBs3bkRgYGC983JycuL+LSsrC0dHRwiFQm7dzp074eDgAB0dHaioqGDPnj3IzMysMU8bGxuJZX19/Vp7iqTp0KED+Hy+1HxSUlIgKysLe3t7brupqSk0NTVrzPPx48eYPHkyzMzMoK6uDjU1NRQWFtZ4TC4uLoiIiADwuvepV69eXOMqOjoaZWVl6NatG5e+IXVGCCGEkI8f9UiRTxqfz4esrCyUlZXrlF5ZWRmysrISDYK3NX/+fHz33XfcczrW1tZ48OAB1q1bBx8fn0Yr59ChQ/D19cWmTZvg5OQEVVVV/Pjjj7h+/XqN+8nJyUks83i8Bg1xbKx83uTj44OnT59i27ZtMDIygoKCApycnPDq1atq96mYnjw1NRVJSUn4/PPPkZycjIiICOTl5cHR0RFKSkoAGl5nhBBCCPn4UUOKfNIqhuiZm5sjNja2xuF9ampq6N69O2RkGrcjt6ioqEqefD6/QY2Mv//+Gz169AAAiEQixMbGYubMmQCAqKgoODs7Y/r06Vz6tLS0t4j8NTk5ubfuobOwsIBIJEJcXBwcHBwAAPfu3UNeXl6N+0VFReHnn3/GgAEDAAAPHz6s9d1f1tbW0NTUxOrVq9GpUyeoqKjA1dUV69evR15eHvd8VEX+TVFnhBBCCPnw0dA+QgDIyMigX79+Nabp169fozeiAMDd3R1r1qzBmTNnkJGRgePHj2Pz5s0YMmQIl2bRokXw9vauNa+dO3fi+PHjSE5OxowZM5CXl4cJEyYAeP18U0xMDEJCQnD37l0sWbKkxskh6srY2BhhYWHIzs6uteFTHUtLS/Tp0wdTpkzBjRs3EBcXhylTpkAgEIDH41W7n5mZGQ4cOAChUIjr16/Dy8sLAoGgxrIqnpMKCgriGk02NjYoLS1FWFgYXFxcJPJvijojhBBCyIePGlKE/E/79u3h6ekJNTU1ifVqamrw9PRssvdI/fTTTxg+fDimT58OKysr+Pr6YurUqVi1ahWXJisrq07P5fj5+cHPzw+2tra4cuUKTp06xU2KMXXqVAwdOhQjR45E165d8fTpU4melobatGkTLl68CENDQ9jZ2TU4n/3790NXVxc9evTAkCFDMHnyZKiqqkJRUbHafX799Vfk5eXB3t4eY8eOxezZs9GyZctay3JxcYFYLOYaUjIyMujRowd4PJ7E81FNVWeEEEII+fDxGGOsuYNobgUFBVBXV0d+fn6Vm2jS9JKTkyESiSArKwtLS8sG5VFSUoL09HSYmJjUeONdF+Xl5bh8+TJevnwJZWXlJhnO19gyMjJgYmKCuLg4ifc3fcj++ecfGBoaIjQ0FL17927ucAghhJBaNeb9SIXc3FyIxWLw+fxGmzGY1KyubQN6RoqQSmRkZKCrq8s17t73RtTH4q+//kJhYSGsra2RlZWFBQsWwNjYmHvmixBCCPkUUePp/UUNKUKk0NbW5n79Ie9GWVkZvv/+e9y/fx+qqqpwdnZGUFBQldn+CCGEEELeB9SQIkSKD+3XH2NjY3zoo3Td3Nzg5ubW3GEQQgghhNQJjVkihBBCCCGEkHqihhQhhBBCCCGE1BM1pAghhBBCCCGknqghRQghhBBCCCH1RA0pQgghhBBCCKknmrWPECno5XeEEEIIIaQm1CNFiBS5ubl48uQJcnNzmzuUD0JGRgZ4PB7i4+ObtJyIiAjweDw8f/68SctpLM0db0BAADQ0NJq8nOXLl6NTp05vnY+xsTG2bt3KLfN4PJw4caLa9JWvu+aub/LujRs3DoMHD27uMJrEu/peJYQ0HDWkCKmEMTFKS2+huOQSSktvgTFxk5b34sULzJ07F0ZGRhAIBHB2dkZ0dHSTltnYDA0NkZWVhY4dOzZpOc7OzsjKyoK6unqd9/mYb7RqM3LkSNy9e7e5w6iz6OhoTJkypcH7N+T6aIh31UAltdu2bRsCAgKaO4y3Ju176l19rxJCGo6G9hHyhpycENxNXYnS0mxuXdTVbTA3W4qWLZvmZbGTJk3C7du3ceDAARgYGOC3335Dnz59kJSUhFatWjU437KyMsjJyTVipNXj8/nQ09Nr8nLk5eXfSTnvu1evXkFeXr7WdAKBAAKB4B1E1Dh0dHTean+6Pj4cjDGIxWLIyr7dbUhTN5qb07v6XiWENBz1SBHyPzk5IUi8PUOiEQUApaWPkXh7BnJyQhq9zOLiYgQHB2PDhg3o0aMHTE1NsXz5cpiammLXrl11zqdiCMjhw4fh4uICRUVFBAUFSR1ytXXrVhgbG3PLFb+Ebty4Efr6+tDS0sKMGTNQVlbGpTE2NsbatWsxYcIEqKqqok2bNtizZ0+V8isPsQoLC4OjoyOUlJTg7OyMlJQUiVhWr16Nli1bQlVVFZMmTcJ3331X4xCxykO3KnoGQkJCYGVlBRUVFfTr1w9ZWVkAXg85CwwMxMmTJ8Hj8cDj8RAREQEASExMRK9evSAQCKClpYUpU6agsLCw2rLFYjEmTpwIExMTCAQCWFhYYNu2bdWmf1NUVBRsbGygqKiIzz77DLdv3+a21eccrVmzBgYGBrCwsODq/NixY+jZsyeUlJRga2uLa9eucftV7jmpKOvAgQMwNjaGuro6Ro0ahRcvXnBpXrx4AS8vLygrK0NfXx9btmyBq6sr5s6dW+tx7t69G4aGhlBSUoKnpyfy8/O5bdLyGDx4MMaNG8ctVx7aV9mNGzdgZ2cHRUVFODo6Ii4uTmJ7fa8PABCJRJg9ezY0NDSgpaWFhQsXwsfHp9pezIiICIwfPx75+fncNbV8+XIAQF5eHry9vaGpqQklJSX0798fqamp1R4PYwzLly9HmzZtoKCgAAMDA8yePZvbfuDAATg6OkJVVRV6enoYM2YMcnJyAADl5eVo3bp1le+JuLg4yMjI4MGDBwCA58+fY9KkSdDR0YGamhp69eqFhISEamMCgIULF8Lc3BxKSkpo27YtlixZIvF9UFnFtXjo0CE4OztDUVERHTt2RGRkpES98Xg8nDt3Dg4ODlBQUMCVK1dQXl6OdevWcZ8rW1tb/PHHHxL537lzB4MGDYKamhpUVVXRvXt3pKWlAajak1NaWorZs2ejZcuWUFRUxOeffy7Rwy+tN/HEiRPg8XjcckJCAnr27AlVVVWoqanBwcEBMTEx1R7/5s2bYW1tDWVlZRgaGmL69OkS3yUN/Z6SNrTv7NmzMDc3h0AgQM+ePREQECBxzdfl+wQA9u3bBysrKygqKsLS0hI///xztcdHCKkeNaQIwevhfHdTVwJg0rYCAO6mrmr0YX4ikQhisRiKiooS6wUCAa5cuVLv/L777jvMmTMHQqEQbm5170ELDw9HWloawsPDERgYiICAgCrDZTZt2sTdvE6fPh1ff/11lYZRZYsXL8amTZsQExMDWVlZTJgwgdsWFBSENWvWYP369YiNjUWbNm3q1XisUFRUhI0bN+LAgQO4dOkSMjMz4evrCwDw9fWFp6cnd9OSlZUFZ2dnvHz5Em5ubtDU1ER0dDSOHj2K0NBQzJw5s9pyKm5cjx49iqSkJCxduhTff/89jhw5UmuM8+fPx6ZNmxAdHQ0dHR24u7vXeGMqTVhYGFJSUnDx4kWcPn2aW7948WL4+voiPj4e5ubmGD16NEQiUbX5pKWl4cSJEzh9+jROnz6NyMhI+Pn5cdu/+eYbREVF4dSpU7h48SIuX76Mmzdv1hrfvXv3cOTIEfz55584f/48d500lsLCQgwaNAjt27dHbGwsli9fzp3nmtR0fQDA+vXrERQUBH9/f0RFRaGgoKDG57KcnZ2xdetWqKmpcddURX7jxo1DTEwMTp06hWvXroExhgEDBlR7roODg7Flyxbs3r0bqampOHHiBKytrbntZWVlWLVqFRISEnDixAlkZGRwDU8ZGRmMHj0aBw8elMgzKCgI3bp1g5GREQBgxIgRyMnJwblz5xAbGwt7e3v07t0bz549q/YYVVVVERAQgKSkJGzbtg179+7Fli1baqxn4PV1/u233yIuLg5OTk5wd3fH06dPJdJ899138PPzg1AohI2NDdatW4f9+/fjl19+wZ07dzBv3jx89dVXXCPs33//RY8ePaCgoIC//voLsbGxmDBhQrXX+IIFCxAcHIzAwEDcvHkTpqamcHNzq/F4K/Py8kLr1q0RHR2N2NhYfPfddzX27svIyGD79u24c+cOAgMD8ddff2HBggUSaRryPVXZw4cPMXToULi7uyM+Pp778am+goKCsHTpUqxZswZCoRBr167FkiVLEBgYWO+8CPnkMcLy8/MZAJafn9/coXyShEIhS0xMZEKhsMF5FBcXs6SkJFZcXFyv/Z48ecKEQiGLjz/MQsPa1voXH3+YCYVC9uTJkwbHWpmTkxNzcXFh//77LxOJROzAgQNMRkaGmZub1zmP9PR0BoBt3bpVYv2yZcuYra2txLotW7YwIyMjbtnHx4cZGRkxkUjErRsxYgQbOXIkt2xkZMS++uorbrm8vJy1bNmS7dq1S6L8uLg4xhhj4eHhDAALDQ3l9jlz5gwDwJ2jrl27shkzZkjE1q1btyrxvqki37y8PMYYY/7+/gwAu3fvHpdm586dTFdXV+L4PDw8JPLZs2cP09TUZIWFhRLxycjIsOzs7GrLr2zGjBls2LBhtcZ76NAhbt3Tp0+ZQCBghw8fZozV/Rzp6uqy0tJSbl1Fne/bt49bd+fOHQaA+yz5+/szdXV1bvuyZcuYkpISKygo4NbNnz+fde3alTHGWEFBAZOTk2NHjx7ltj9//pwpKSmxOXPmVHucy5YtY3w+n/3zzz/cunPnzjEZGRmWlZXFGGPMxcWlSh4eHh7Mx8eHWzYyMmJbtmzhlgGw48ePM8YY2717N9PS0pL4jO/atUvqdVef60NXV5f9+OOP3LJIJGJt2rSpcs28qXK9MsbY3bt3GQAWFRXFrcvNzWUCgYAdOXJEaj6bNm1i5ubm7NWrV9WW9abo6GgGgL148YIxxlhcXBzj8XjswYMHjDHGxGIxa9WqFfe5vHz5MlNTU2MlJSUS+bRr147t3r27TmUyxtiPP/7IHBwcqt1ecS36+flx68rKyljr1q3Z+vXrGWP/nZsTJ05waUpKSpiSkhK7evWqRH4TJ05ko0ePZowxtmjRImZiYlJtHb35+S4sLGRycnIsKCiI2/7q1StmYGDANmzYwBiTfu6OHz/O3rwdUlVVZQEBAdUeb22OHj3KtLS0uOWGfk9V/l5dtGgRa9++vUSahQsXSlzzdfk+adeuHTt48KBEmlWrVjEnJ6d6HimRpqH3I+T9Ute2AfVIkU+aWCyGSCTCq7K6zc73qiyX60VqLAcOHABjDK1atYKCggK2b9+O0aNHQ0am/h9PR0fHBsXQoUMH8Pl8bllfX58bQlTBxsaG+zePx4Oenl6VNJW9uY++vj4AcPukpKSgS5cuEukrL9eFkpIS2rVrV2PslQmFQtja2kJZWZlb161bN5SXl9fYy7Zz5044ODhAR0cHKioq2LNnDzIzM2uN0cnJift3ixYtYGFhAaFQWOt+b7K2tpb6XFRNdSyNsbExVFVVJfapSH///n2UlZVJnAd1dXVYWFjUGl+bNm0knulzcnKqtT7ro6IH483e2zfrtTo1XR/5+fl4/PixxPHy+Xw4ODg0KD5ZWVl07dqVW6elpVXjuR4xYgSKi4vRtm1bTJ48GcePH5foaYmNjYW7uzvatGkDVVVVuLi4AAB3zXXq1AlWVlZcr1RkZCRycnIwYsQIAK+HqBUWFkJLSwsqKircX3p6Ojc0TprDhw+jW7du0NPTg4qKCn744Yd6X+eysrJwdHSscuxvfkfdu3cPRUVF6Nu3r0R8+/fv5+KLj49H9+7d6/S8Z1paGsrKytCtWzdunZycHLp06VKvz9s333yDSZMmoU+fPvDz86uxrgAgNDQUvXv3RqtWraCqqoqxY8fi6dOnKCoq4tI05HuqMqFQKHF9AXX7DLzp5cuXSEtLw8SJEyXqfPXq1bUeJyGkKmpIkU8an8+HrKws5OXq9q4oeTltyMrKSjQ63la7du0QGRmJwsJCPHz4EDdu3EBZWRnatm1b77zebBgAr4ecMCY5XFHaMKPKNyk8Hg/l5eX1TlNTvhXPINS2T31Ji6vyMTeGQ4cOwdfXFxMnTsSFCxcQHx+P8ePH49WrV2+Vb13PUeVzW6G+ddyQ89gY6nqcje1dXR8NYWhoiJSUFPz8888QCASYPn06evTogbKyMm74qZqaGoKCghAdHY3jx48DgMQ15+XlxTWkDh48iH79+kFLSwvA6+GQ+vr6iI+Pl/hLSUnB/PnzpcZ07do1eHl5YcCAATh9+jTi4uKwePHit77OK7x5HVc8R3TmzBmJ+JKSkrjnpBp7spS6XIfLly/HnTt3MHDgQPz1119o3749V/eVZWRkYNCgQbCxsUFwcDBiY2Oxc+dOAJLn6V1dh7UdX0Wd7927V6LOb9++jb///rvR4yHkY0cNKfJJ09bWhqWlJWxshkFBQQ8Ar5qUPCgo6MPGZhgsLS2b5CW9FQ/35+XlISQkBB4eHm+dp46ODrKzsyX+Y31f3kliYWFRZZr3ppj2XV5evkoPopWVFRISEvDy5UtuXVRUFGRkZKrtfYmKioKzszOmT58OOzs7mJqa1vkX3DdvUPLy8nD37l1YWVkBeL/OUdu2bSEnJydxHvLz8+s0hXpmZiYePXrELf/9998S9amjoyMxyYNYLJaYdKM2VlZWuHXrFkpKSiTKeBvq6urQ1dWVOF6xWFzrM2HVXVMikQjXr1/n1j19+hQpKSlo3759tXkJBAK4u7tj+/btiIiIwLVr15CYmIjk5GQ8ffoUfn5+6N69OywtLaX2YIwZMwa3b99GbGws/vjjD3h5eXHb7O3tkZ2dDVlZWZiamkr8VfcddvXqVRgZGWHx4sVwdHSEmZkZN3FFbd48HyKRCLGxsdx1Lk379u2hoKCAzMzMKvEZGhoCeN3jevny5To1utu1awd5eXlERUVx68rKyhAdHc2dAx0dHbx48ULisy/t82Zubo558+bhwoULGDp0KPz9/aWWGRsbi/LycmzatAmfffYZzM3NJT4HdSXtmqrMysoKN27ckFhX+TNQ2/eJrq4uDAwMcP/+/Sp1bmJiUu+4CfnUUUOKEAA8Hh/mZksrlipvBQCYmy0Bj9d4PVEVQkJCcP78eaSnp+PixYvo2bMnLC0tMX78eC7NokWL4O3tXe+8XV1d8eTJE2zYsAFpaWnYuXMnzp0715jhN9isWbPw66+/IjAwEKmpqVi9ejVu3bolMXtWYzA2NsatW7eQkpKC3NxclJWVwcvLC4qKivDx8cHt27cRHh6OWbNmYezYsdDV1ZWaj5mZGWJiYhASEoK7d+9iyZIldW74rVy5EmFhYbh9+zbGjRsHbW1tbqax9+kcqaqqwsfHB/Pnz0d4eDju3LmDiRMnQkZGptbzUlGfCQkJuHz5MmbPng1PT09u+uZevXrhzJkzOHPmDJKTk/H111/X68W5Y8aMAY/Hw+TJk5GUlISzZ89i48aNb3O4AF5fh+vWrcPJkyeRkpKCOXPmIC8vr8bjNTY2RmFhIcLCwpCbm4uioiKYmZnBw8MDkydPxpUrV5CQkICvvvoKrVq1qvZHkYCAAPz666+4ffs27t+/j99++w0CgQBGRkZo06YN5OXl8dNPP+H+/fs4deoUVq1aJTUWZ2dnTJw4EWKxGF9++SW3rU+fPnBycsLgwYNx4cIFZGRk4OrVq1i8eHG1s9CZmZkhMzMThw4dQlpaGrZv315tb0xlO3fuxPHjx5GcnIwZM2YgLy9PYoKZylRVVeHr64t58+YhMDAQaWlpuHnzJn766Sdu4oOZM2eioKAAo0aNQkxMDFJTU3HgwAGpQ0aVlZXx9ddfY/78+Th//jySkpIwefJkFBUVYeLEiQCArl27QklJCd9//z3S0tJw8OBBiYl1iouLMXPmTERERODBgweIiopCdHR0tQ1CU1NTlJWVcefpwIED+OWXX+pUX2+S9j1V2bRp05Camor58+cjJSWlSuxA3b5PVqxYgXXr1mH79u24e/cuEhMT4e/vj82bN9c7bkI+ddSQIuR/WrZ0g3XHnVBQkLyRVlDQg3XHnU32Hqn8/HzMmDEDlpaW8Pb2xueff46QkBCJoSBZWVl1ekahMisrK/z888/YuXMnbG1tcePGjTrNdPYueHl5YdGiRfD19YW9vT3S09Mxbty4KjMYvq3JkyfDwsICjo6O0NHRQVRUFJSUlBASEoJnz56hc+fOGD58OHr37o0dO3ZUm8/UqVMxdOhQjBw5El27dsXTp0/rPCudn58f5syZAwcHB2RnZ+PPP//knnd6387R5s2b4eTkhEGDBqFPnz7o1q0bN01yTUxNTTF06FAMGDAAX3zxBWxsbCSmVJ4wYQJ8fHzg7e0NFxcXtG3bFj179qxzXCoqKvjzzz+RmJgIOzs7LF68GOvXr2/wcVZYuHAhRo8eDW9vbzg5OUFFRQVubm41Hq+zszOmTZuGkSNHQkdHBxs2bAAA+Pv7w8HBAYMGDYKTkxMYYzh79my1z/doaGhg79696NatG2xsbBAaGoo///wTWlpa0NHRQUBAAI4ePYr27dvDz8+v2oajl5cXEhISMGTIEImhcDweD2fPnkWPHj0wfvx4mJubY9SoUXjw4EG1Pxh8+eWXmDdvHmbOnIlOnTrh6tWrWLJkSZ3q0s/PD35+frC1tcWVK1dw6tSpWnvvV61ahSVLlmDdunWwsrJCv379cObMGa53REtLC3/99RcKCwvh4uICBwcH7N27t9o69fPzw7BhwzB27FjY29vj3r17CAkJgaamJoDXzyj+9ttvOHv2LKytrfH7779z09cDr4d7P336FN7e3jA3N4enpyf69++PFStWSC3P1tYWmzdvxvr169GxY0cEBQVh3bp1daqvN0n7nqqsTZs2CA4OxokTJ2Bra4tffvkFa9eulUhTl++TSZMmYd++ffD394e1tTVcXFwQEBBAPVKENACPvS+DxZtRQUEB1NXVkZ+fDzU1teYO55OTnJwMkUgEWVlZWFpaNiiPkpISpKenw8TE5K1vxBkT49atYLwqy4W8nDZsbIY1SU8Uqapv377Q09PDgQMHmjsU8j8vX75Eq1atsGnTJu5X/Y9ZeXk5rKys4OnpKbUHiFSVkZEBExMTxMXF1fgeONL4IiIi0LNnT+Tl5VV5PxZpHo15P0KaT13bBm/3SnFCPkI8Hh8KCjbg81837qgR1TSKiorwyy+/wM3NDXw+H7///jtCQ0Nx8eLF5g7tkxYXF4fk5GR06dIF+fn5WLlyJQA0yjN776MHDx7gwoULcHFxQWlpKXbs2IH09HSMGTOmuUMjhBDynqOGFCFSaGtrQywWN+rsfERSxbCjNWvWoKSkBBYWFggODkafPn2aO7RP3saNG5GSkgJ5eXk4ODjg8uXLTTLByvtARkYGAQEB8PX1BWMMHTt2RGhoaI2TJBBCCCEADe0DQEP7mtv7NrSPEEIIIaQh6H7k41DXtgFNNkEIIYQQQggh9UQNKUIIIYQQQgipJ2pIEUIIIYQQQkg9UUOKEEIIIYQQQuqJGlKEEEIIIYQQUk80/TkhUuTm5nLTn3+s0z4TQgghhJCGox4pQqTIzc3FkydPkJub29yhfHR4PB5OnDjRLGWPGzcOgwcPbvR8m/OYCCGEENI8qCFFSCVixhBXKsZfJeWIKxVD3MSvWrt06RLc3d1hYGBQ7Q05YwxLly6Fvr4+BAIB+vTpg9TU1CaNq6lkZWWhf//+TVpGRkYGeDwe4uPjm7QcQgghhHy6qCFFyBvOPHkOx2tJmPesDGtflGPeszI4XkvCmSfPm6zMly9fwtbWFjt37qw2zYYNG7B9+3b88ssvuH79OpSVleHm5oaSkpImi6up6OnpQUFBobnDIIQQQgh5K9SQIuR/zjx5jkm3M5BVWiaxPru0DJNuZzRZY6p///5YvXo1hgwZInU7Ywxbt27FDz/8AA8PD9jY2GD//v149OhRvYaTBQQEQENDA6dPn4aFhQWUlJQwfPhwFBUVITAwEMbGxtDU1MTs2bMhFou5/fLy8uDt7Q1NTU0oKSmhf//+XG8YYww6Ojr4448/uPSdOnWCvr4+t3zlyhUoKCigqKgIgOQwuIqeo2PHjqFnz55QUlKCra0trl27JhH73r17YWhoCCUlJQwZMgSbN2+GhoZGtcdqYmICALCzswOPx4Orq6vE9o0bN0JfXx9aWlqYMWMGysr+O+elpaXw9fVFq1atoKysjK5duyIiIqLO9QwACxcuhLm5OZSUlNC2bVssWbKEKyM/Px98Ph8xMTEAgPLycrRo0QKfffYZt/9vv/0GQ0PDetURIYQQQt4takgRgtfD+X5I/RfSBvFVrFuS+m+TD/OTJj09HdnZ2ejTpw+3Tl1dHV27dq33zXRRURG2b9+OQ4cO4fz584iIiMCQIUNw9uxZnD17FgcOHMDu3bslGkbjxo1DTEwMTp06hWvXroExhgEDBqCsrAw8Hg89evTgGhp5eXkQCoUoLi5GcnIyACAyMhKdO3eGkpJStXEtXrwYvr6+iI+Ph7m5OUaPHg2RSAQAiIqKwrRp0zBnzhzEx8ejb9++WLNmTY3HeePGDQBAaGgosrKycOzYMW5beHg40tLSEB4ejsDAQAQEBCAgIIDbPnPmTFy7dg2HDh3CrVu3MGLECPTr169eQylVVVUREBCApKQkbNu2DXv37sWWLVsAvD53nTp14uosMTERPB4PcXFxKCws5OrMxcWlznVECCGEkHePGlLkk5abm4vk5GQcuZVUpSfqTQzAo9IyHLmVhOTk5Hc6CUV2djYAQFdXV2K9rq4ut62uysrKsGvXLtjZ2aFHjx4YPnw4rly5gl9//RXt27fHoEGD0LNnT4SHhwMAUlNTcerUKezbtw/du3eHra0tgoKC8O+//3K9Sq6urlyj4NKlS7Czs5NYFxERUaVRUJmvry8GDhwIc3NzrFixAg8ePMC9e/cAAD/99BP69+8PX19fmJubY/r06bU+Y6WjowMA0NLSgp6eHlq0aMFt09TUxI4dO2BpaYlBgwZh4MCBCAsLAwBkZmbC398fR48eRffu3dGuXTv4+vri888/h7+/f53r+YcffoCzszOMjY3h7u4OX19fHDlyhNteuX769u0LKysrXLlypdo6q6mOCCGEEPLuUUOKfNLEYjFEIhGelIlrTwzgSdnr9G8OffuQKCkpoV27dtyyrq4ujI2NoaKiIrEuJycHACAUCiErK4uuXbty27W0tGBhYQGhUAgAcHFxQVJSEp48eYLIyEi4urpyDYWysjJcvXq1ytC6ymxsbLh/VwwLrIghJSUFXbp0kUhfebk+OnToAD6fL1FeRVmJiYkQi8UwNzeHiooK9xcZGYm0tLQ6l3H48GF069YNenp6UFFRwQ8//IDMzExuu4uLC65cuQKxWFylzh49eoR79+5VqbOa6ogQQggh7x69R4p80vh8PmRlZaEjJwZQXmt6HTk+ZGX5EjfiTU1PTw8A8PjxY4lnjx4/foxOnTrVKy85OTmJZR6PJ3VdeXntdVHB2toaLVq0QGRkJCIjI7FmzRro6elh/fr1iI6ORllZGZydnescF4/HA4B6xVAfNR1vYWEh+Hw+YmNjq5zjNxubNbl27Rq8vLywYsUKuLm5QV1dHYcOHcKmTZu4ND169MCLFy9w8+ZNXLp0CWvXroWenh78/Pxga2sLAwMDmJmZVRt3U9cRIYQQQmpHPVLkvVbb1ODjxo0Dj8eDpaUlHjx4gNu3b+Pu3bt1zl9bWxuWlpbwtGkPfQU58KpJxwNgoCAHT5v2sLS0fKcv6TUxMYGenh43/AwACgoKcP36dTg5OTVp2VZWVhCJRLh+/Tq37unTp0hJSUH79u0BvL6p7969O06ePIk7d+7g888/h42NDUpLS7F79244OjpCWVm5wTFYWFggOjpaYl3l5crk5eUBoN49h3Z2dhCLxcjJyYGpqanEX0WDtjZXr16FkZERFi9eDEdHR5iZmeHBgwcSaTQ0NGBjY4MdO3ZATk4OlpaW6NGjB+Li4nD69Olah0ISQgghpPlRQ4q81+oyNXi/fv1w+fJltG7dGhYWFmjbtm29y+HzeFht1goAqjSmKpZXmbUCn1ddU6vhCgsLER8fz73zKD09HfHx8dxQMB6Ph7lz52L16tU4deoUEhMT4e3tDQMDA4mXy/bu3Rs7duxo1NjMzMzg4eGByZMn48qVK0hISMBXX32FVq1awcPDg0vn6uqK33//HZ06dYKKigpkZGTQo0cPBAUFvXWjYNasWTh79iw2b96M1NRU7N69G+fOneN6ZaRp2bIlBAIBzp8/j8ePHyM/P79OZZmbm8PLywve3t44duwY0tPTcePGDaxbtw5nzpypUx5mZmbIzMzEoUOHkJaWhu3bt+P48eNV0rm6ukrUT4sWLWBlZYXDhw9TQ4oQQgj5AFBDirzXapsaHAAUFBSgo6MDPp8POTk5yMo2bMTqQB0N7OtoDD0FyaFf+gpy2NfRGAN1NBqUb21iYmJgZ2cHOzs7AMA333wDOzs7LF26lEuzYMECzJo1C1OmTEHnzp1RWFiI8+fPQ1FRkUuTlpbWJJNg+Pv7w8HBAYMGDYKTkxMYYzh79qzEUDMXFxeIxWKJ53pcXV2rrGuIbt264ZdffsHmzZtha2uL8+fPY968eRLHXpmsrCy2b9+O3bt3w8DAQKLRVxt/f394e3vj22+/hYWFBQYPHozo6Gi0adOmTvt/+eWXmDdvHmbOnIlOnTrh6tWrWLJkSZV0TVlnhBBCCGl6PMaaYT7n90xBQQHU1dWRn58PNTW15g7nk5OcnAyRSARZWVlYWlpWm47H4+H48eMSvTDjxo3DiRMn0LZtW2zYsAEmJiYwMjJqcGMKeD0V+pFbSXhSJoaOHB+eNu2bpCeKNNzkyZORnJyMy5cvN3cohBBCCKekpATp6ekwMTGp8Qc/8n6ra9uAJpsgH7R+/fph6NChMDY2RllZGV6+fInU1FRYWlrWOPSrJnweD3YKfIj47PXEEtSIanYbN25E3759oaysjHPnziEwMBA///xzc4dFCCGEkE8YNaTIB23UqFEA/vsFqFWrVrh79y5evHjxVr2L2traEIvF73R2PlK9GzduYMOGDXjx4gXatm2L7du3Y9KkSc0dFiGEEEI+YdSQIh8VeXl5yMrKorS09K3yeZez8pHavfkyW0IIIYSQ9wFNNkE+Kq9evYJIJKryriBCCCGEEEIaE/VIkfdaYWEh7t27xy1XTA3eokULtGjRAitWrMCwYcOgo6ODkpISZGZmQkFBgSYNIYQQQgghTYp6pEizKi9nyHtYipy7Jch7WIrycslJJGuaGpzP5+PWrVv48ssv0a9fP+Tm5kIgEMDS0hIyMnRpE0IIIYSQpkM9UqTZpMXl4PLhVLx8/t/zTClhV9F9pBna2bUE8Pq9OjXN0B8SEgJAcrIJGtZHCCGEEEKaGv1sT5pFWlwOzu++LdGIAoCXz0txfvdtpMXlNFNkhBBCCCGE1I4aUuSdKy9nuHw4tcY0V46kVhnm9y7l5ubi8ePHyM3NbbYYCCGEEELI+4uG9r2HcnNzP+ob+LyHpVV6oiorzCvF9bBb0DRUqFOe5eXlEIlEKCkpqXEoYF09efKEe4+UsrLyW+dXHVlZ2fdiKGJGRgZMTEwQFxeHTp061WvfiIgI9OzZE3l5edDQ0GiS+AAgICAAc+fOxfPnz5usjIZwdXVFp06dsHXr1gbn8b4eGyGEEEKqRz1S7yGxWAyRSPTR/hW/KKtTPRS/KKtXvo3RgAIAcTnDrexiRKa/xK3sYoibuGfs0qVLcHd3h4GBAXg8Hk6cOFElzbFjx/DFF19AS0sLPB4P8fHxjRqDoaEhsrKy0LFjRwCvG0c8Hq/Kjb2rqyvmzp0rsc7Z2RlZWVlQV1dv1JgqGzlyJO7evdukZTSXj/nYCCGEkI8V9Ui9h/h8PmRlP95TI1CtWw+MQFWuzvVQ0SP1ti4Ic7DufCqyC/7rMdty7RkW9TPDF1Yt3zp/aV6+fAlbW1tMmDABQ4cOrTbN559/Dk9PT0yePLnRY+Dz+dDT02vQvvLy8g3etz4EAgEEAkGTl9Mcaju2V69eQV5e/h1GRAghhJBasffYzz//zKytrZmqqipTVVVln332GTt79iy3vbi4mE2fPp21aNGCKSsrs6FDh7Ls7Ox6l5Ofn88AsPz8/MYMn1RDLC5n/guvsB1Tw6r9C/juChOLy+ucZ3FxMUtKSmLFxcUNjutc4iNmvPA0M6r0Z/y/v3OJjxqcd10BYMePH692e3p6OgPA4uLi6p33s2fP2JgxY5i2tjZTVFRkpqam7P/+7/+q5Fvx7zf/fHx8mI+PT5X16enpLDw8nAFgeXl5jDHG/P39mbq6Ojt//jyztLRkysrKzM3NjT169F/9lZWVsVmzZjF1dXXWokULtmDBAubt7c08PDyqjb8i3wrLli1jtra2bP/+/czIyIipqamxkSNHsoKCgmrzyM3NZaNGjWIGBgZMIBCwjh07soMHD9Zad1euXGEuLi5MIBAwDQ0N9sUXX7Bnz54xxhhzcXFhs2bNYvPnz2eamppMV1eXLVu2TGL/TZs2sY4dOzIlJSXWunVr9vXXX7MXL17Uemx79+5lxsbGjMfj1RojIYSQ5tcY9yOk+dW1bfBeD+1r3bo1/Pz8EBsbi5iYGPTq1QseHh64c+cOAGDevHn4888/cfToUURGRuLRo0fV/qJP3h8yMjx0H2lWY5rPPc0gI8N7RxG9Hs634s8kSBvEV7FuxZ9JTT7MryktWbIESUlJOHfuHIRCIXbt2gVtbe0q6QwNDREcHAwASElJQVZWFrZt24Zt27bByckJkydPRlZWFrKysmBoaCi1rKKiImzcuBEHDhzApUuXkJmZCV9fX277+vXrERQUBH9/f0RFRaGgoEDqkMbapKWl4cSJEzh9+jROnz6NyMhI+Pn5VZu+pKQEDg4OOHPmDG7fvo0pU6Zg7NixuHHjRrX7xMfHo3fv3mjfvj2uXbuGK1euwN3dHWKxmEsTGBgIZWVlXL9+HRs2bMDKlStx8eJFbruMjAy2b9+OO3fuIDAwEH/99RcWLFhQ47Hdu3cPwcHBOHbsWKMP5SSEEELI23uvx4+5u7tLLK9Zswa7du3C33//jdatW+PXX3/FwYMH0atXLwCAv78/rKys8Pfff+Ozzz5rjpBJHbWza4l+UztWeY+UiqYCPvf87z1STa1iYo+ErGJk5ZdUm44ByMovwR+R8bDVF0BbW1tqI+R9lpmZCTs7Ozg6OgIAjI2Npabj8/lo0aIFAKBly5YSE0jIy8tDSUmp1qF8ZWVl+OWXX9CuXTsAwMyZM7Fy5Upu+08//YRFixZhyJAhAIAdO3bg7Nmz9T6m8vJyBAQEQFVVFQAwduxYhIWFYc2aNVLTt2rVSqJBN2vWLISEhODIkSPo0qWL1H02bNgAR0dH/Pzzz9y6Dh06SKSxsbHBsmXLAABmZmbYsWMHwsLC0LdvXwCQeK7M2NgYq1evxrRp0yTyrOzVq1fYv38/dHR0aqgBQgghhDSX97oh9SaxWIyjR4/i5cuXcHJyQmxsLMrKytCnTx8ujaWlJdq0aYNr167V2JAqLS1Fael/N+8FBQVNGjuRrp1dS5jY6uB62C0UvyiDQFUOXXvbvNueqP9N7JFb+KpO6XMLX0EkkpPojfhQfP311xg2bBhu3ryJL774AoMHD4azs3OTlKWkpMQ1ogBAX18fOTmv3w2Wn5+Px48fSzRc+Hw+HBwcUF5eXq9yjI2NuUZU5XKkEYvFWLt2LY4cOYJ///0Xr169QmlpKZSUlKrdJz4+HiNGjKgxDhsbG4nlynGEhoZi3bp1SE5ORkFBATfDZFFRUbVlGxkZUSOKEEIIeY+910P7ACAxMREqKipQUFDAtGnTcPz4cbRv3x7Z2dmQl5evMt2yrq4usrOza8xz3bp1UFdX5/6qG55Emp6MDA+ahgpoaa4ITUOFd9qIAv6b2ENbpW4P8muryENWVhZ8Pr+JI2t8/fv3x4MHDzBv3jw8evQIvXv3luidaUyVp3Tn8XiNNqtibeXU1Bj78ccfsW3bNixcuBDh4eGIj4+Hm5sbXr2qviFdlwkuaoojIyMDgwYNgo2NDYKDgxEbG4udO3cCQI3lNuW0+4QQQgh5e+99Q8rCwgLx8fG4fv06vv76a/j4+CApKemt8ly0aBHy8/O5v4cPHzZStORDo62tDUtLSwx36QR9dUVU14zjAdBXV8Rwl06wtLT84Ib1VdDR0YGPjw9+++03bN26FXv27JGarmKGuMo9b/Ly8m/dG6eurg5dXV1ER0dz68RiMW7evPlW+dZFVFQUPDw88NVXX8HW1hZt27atddpxGxsbhIWFNbjM2NhYlJeXY9OmTfjss89gbm6OR48eNTg/QgghhLwf3vuGlLy8PExNTeHg4IB169bB1tYW27Ztg56eHl69elXlPTePHz+u9fkNBQUFqKmpSfyRTxtfhodl7u0BoEpjqmJ5mXt78Jugx6ywsBDx8fHchALp6emIj49HZmYml+bZs2eIj4/nfkRISUlBfHy8RO+rt7c3Fi1aVG05S5cuxcmTJ3Hv3j3cuXMHp0+fhpWVldS0RkZG4PF4OH36NJ48eYLCwkIAr4fSXb9+HRkZGcjNza33ULwKs2bNwrp163Dy5EmkpKRgzpw5yMvLA4/XtD2SZmZmuHjxIq5evQqhUIipU6fi8ePHNe6zaNEiREdHY/r06bh16xaSk5Oxa9euOr8029TUFGVlZfjpp59w//59HDhwAL/88ktjHA4hhBBCmtF735CqrLy8HKWlpXBwcICcnJzEL8UpKSnIzMyEk5NTM0ZIPlT9Oupj11f20FNXlFivp66IXV/Zo19H/SYpNyYmBnZ2drCzswMAfPPNN7Czs8PSpUu5NKdOnYKdnR0GDhwIABg1ahTs7OwkbsgzMzORlZVVbTny8vJYtGgRbGxs0KNHD/D5fBw6dEhq2latWmHFihX47rvvoKuri5kzZwIAfH19wefz0b59e+jo6Eg09upj4cKFGD16NLy9veHk5AQVFRW4ublBUVGx9p3fwg8//AB7e3u4ubnB1dUVenp6GDx4cI37mJub48KFC0hISECXLl3g5OSEkydP1vkdZ7a2tti8eTPWr1+Pjh07IigoCOvWrWuEoyGEEEJIc+KxpnhwoZEsWrQI/fv3R5s2bfDixQscPHgQ69evR0hICPr27Yuvv/4aZ8+eRUBAANTU1DBr1iwAwNWrV+tVTkFBAdTV1ZGfn0+9U80gOTkZIpEIsrKysLS0bFAeJSUlSE9Ph4mJyVvfjIvLGf6IjEdu4Stoq8hjuEunJumJIv8pLy+HlZUVPD09sWrVquYOhxBCCGmQxrwfIc2nrm2D93rWvpycHHh7eyMrKwvq6uqwsbHhGlEAsGXLFsjIyGDYsGEoLS2Fm5tbjdMJE1IXfBkebPUFEInkXk8sQY2oRvfgwQNcuHABLi4uKC0txY4dO5Ceno4xY8Y0d2iEEEIIIXXyXjekfv311xq3KyoqYufOndwMWIQ0Fm1tbYjF4g9ydr4PgYyMDAICAuDr6wvGGDp27IjQ0NBqn9kihBBCCHnfvNcNKUKay4c6K9+HwtDQEFFRUc0dBiGEEEJIg31wk00QQgghhBBCSHOjhhQhhBBCCCGE1BM1pAghhBBCCCGknqghRQghhBBCCCH1RA0pQgghhBBCCKknmrWPEClyc3O56c9pBj9CCCGEEFIZ9UgRIkVubi6ePHmC3Nzc5g7loxIREQEej4fnz59Xm2b58uXo1KnTO4vpfZGdnY2+fftCWVkZGhoazR0OjI2NsXXr1uYOo9EEBAS8F/VKCCHk40ENKUIqKxdD6XEs1DMvQulxLFAubtLiLl26BHd3dxgYGIDH4+HEiRMS28vKyrBw4UJYW1tDWVkZBgYG8Pb2xqNHj5o0rto01Y2pr68vwsLCGj3f992WLVuQlZWF+Ph43L17t7nDQXR0NKZMmcItS7s2PyQjR458L+qVEELIx4OG9hHypqRTwPmFaFPwRiMldjXQbz3Q/ssmKfLly5ewtbXFhAkTMHTo0Crbi4qKcPPmTSxZsgS2trbIy8vDnDlz8OWXXyImJqZJYmpOKioqUFFRae4wmkxZWRnk5OSqrE9LS4ODgwPMzMyaIar/vHr1CvLy8tDR0WnWOBqbQCCAQCBo7jAIIYR8RKhHipAKSaeAI95AQaWenoKs1+uTTjVJsf3798fq1asxZMgQqdvV1dVx8eJFeHp6wsLCAp999hl27NiB2NhYZGZm1qkMxhh0dHTwxx9/cOs6deoEfX19bvnKlStQUFBAUVERAGDz5s1cL5ihoSGmT5+OwsJCAK+H6I0fPx75+fng8Xjg8XhYvnw5AKC0tBQLFy6EoaEhFBQUYGpqil9//VUintjYWDg6OkJJSQnOzs5ISUnhtlUe2jdu3DgMHjwYGzduhL6+PrS0tDBjxgyUlZVxabKysjBw4EAIBAKYmJjg4MGDNQ5Nu3TpEuTk5JCdnS2xfu7cuejevTu3HBwcjA4dOkBBQQHGxsbYtGmTRHppvTQaGhoICAgAAGRkZIDH4+Hw4cNwcXGBoqIigoKCqsRjbGyM4OBg7N+/HzweD+PGjav1HBQUFEAgEODcuXMSeR0/fhyqqqrceUxMTESvXr0gEAigpaWFKVOmcHm8Wb9r1qyBgYEBLCwsuJgq6s/Y2BgAMGTIEPB4PG4ZAE6ePAl7e3soKiqibdu2WLFiBUQikdR6f7O8tWvXQldXFxoaGli5ciVEIhHmz5+PFi1aoHXr1vD395fYr6bjuHDhAhQVFasMGZ0zZw569eoFQHoPan1jJ4QQQt5EDSlCgNfD984vBMCkbPzfuvPfNfkwv7qqaMDUdWgdj8dDjx49EBERAQDIy8uDUChEcXExkpOTAQCRkZHo3LkzlJSUAAAyMjLYvn077ty5g8DAQPz1119YsGABAMDZ2Rlbt26FmpoasrKykJWVBV9fXwCAt7c3fv/9d2zfvh1CoRC7d++u0sO0ePFibNq0CTExMZCVlcWECRNqjD88PBxpaWkIDw9HYGAgAgICuMZKRZmPHj1CREQEgoODsWfPHuTk5FSbX48ePdC2bVscOHCAW1dWVoagoCAultjYWHh6emLUqFFITEzE8uXLsWTJEoly6+q7777DnDlzIBQK4ebmVmV7dHQ0+vXrB09PT2RlZWHbtm0Aaj4HampqGDRoEA4ePCiRV1BQEAYPHgwlJSW8fPkSbm5u0NTURHR0NI4ePYrQ0FDMnDlTYp+wsDCkpKTg4sWLOH36tNT4AMDf3x9ZWVnc8uXLl+Ht7Y05c+YgKSkJu3fvRkBAANasWVNjffz111949OgRLl26hM2bN2PZsmUYNGgQNDU1cf36dUybNg1Tp07FP//8AwC1Hkfv3r2hoaGB4OBgrgyxWIzDhw/Dy8tLagwNjZ0QQgjhMMLy8/MZAJafn9/coXyShEIhS0xMZEKhsMF5FBcXs6SkJFZcXFyv/Z48ecKEQiF7EPEbY8vUav17EPEbEwqF7MmTJw2OtSYA2PHjx2tMU1xczOzt7dmYMWPqlff27dtZhw4dGGOMnThxgnXt2pV5eHiwXbt2McYY69OnD/v++++r3f/o0aNMS0uLW/b392fq6uoSaVJSUhgAdvHiRal5hIeHMwAsNDSUW3fmzBkGgDt3y5YtY7a2ttx2Hx8fZmRkxEQiEbduxIgRbOTIkYyx19cPABYdHc1tT01NZQDYli1bqj2e9evXMysrK245ODiYqaiosMLCQsYYY2PGjGF9+/aV2Gf+/Pmsffv23LK086Wurs78/f0ZY4ylp6czAGzr1q3VxlHBw8OD+fj41Jim8jk4fvw4U1FRYS9fvmSMvf4uU1RUZOfOnWOMMbZnzx6mqanJHRNjr+tbRkaGZWdnM8Ze16+uri4rLS2VKMvIyEii/qQda+/evdnatWsl1h04cIDp6+tXewwV51MsFnPrLCwsWPfu3bllkUjElJWV2e+//17n45gzZw7r1asXtz0kJIQpKCiwvLw8xljV67UhsRNCSG0aej9C3i91bRtQjxT5pInFYohEIvBeVt978SbeyxyIRCKIxc3TM1VWVgZPT08wxrBr16567evi4oKkpCQ8efIEkZGRcHV1haurKyIiIlBWVoarV6/C1dWVSx8aGorevXujVatWUFVVxdixY/H06VNuyJg08fHx4PP5cHFxqTEWGxsb7t8Vwwtr6kHq0KED+Hy+xD4V6VNSUiArKwt7e3tuu6mpKTQ1NWuMYdy4cbh37x7+/vtvAK+Hfnl6ekJZWRkAIBQK0a1bN4l9unXrhtTU1Hqff0dHx3qlr1DbORgwYADk5ORw6tTrYafBwcFQU1NDnz59uGOwtbXljqniGMrLyyWGU1pbW0NeXr7e8SUkJGDlypXcc20qKiqYPHkysrKyarxOOnToABmZ//770dXVhbW1NbfM5/OhpaXFneO6HIeXlxciIiK4SViCgoIwcODAanttGxo7IYQQUoEaUuSTxufzISsrC6bcsk7pmXJLyMrKStzUvysVjagHDx7g4sWLUFNTq9f+1tbWaNGiBSIjIyUaUpGRkYiOjkZZWRmcnZ0BvH62Z9CgQbCxsUFwcDBiY2Oxc+dOAK8nI6hOXR/mf3OyBR6PBwAoLy+vU/qKfWpKXxctW7aEu7s7/P398fjxY5w7d67WIYaV8Xg8MCY5HPTNZ7cqvNkAqKu6nAN5eXkMHz6cG9538OBBjBw5ErKy9ZtHqCHxAUBhYSFWrFiB+Ph47i8xMRGpqalQVFSsdj9p5/Ntz3Hnzp3Rrl07HDp0CMXFxTh+/Hi1w/reJnZCCCGkstLLBgAASFxJREFUAs3aRz5p2trar1+4a272ena+gixIf06KB6gZoE33UYBM8zWiUlNTER4eDi0trXrnwePx0L17d5w8eRJ37tzB559/DiUlJZSWlmL37t1wdHTkbqhjY2NRXl6OTZs2cT0HR44ckchPXl6+Ss+MtbU1ysvLERkZyfWKNDULCwuIRCLExcXBwcEBAHDv3j3k5eXVuu+kSZMwevRotG7dGu3atZPogbKyskJUVJRE+qioKJibm3MNaR0dHWRlZXHbU1NTG603oy7nAHjdE9O3b1/cuXMHf/31F1avXi1xDAEBAXj58iV3bqOioiAjI8NNKlFXcnJyVc63vb09UlJSYGpqWt/Dq5e6HoeXlxeCgoLQunVryMjIYODAgdXm+a5iJ4QQ8vGiHilCgNeNo37r/7fAq7Txf8v9/JqkEVVYWMj9Ig4A6enpiI+P52bkKysrw/DhwxETE4OgoCCIxWJkZ2cjOztboneod+/e2LFjR41lubq64vfff0enTp2goqICGRkZ9OjRA0FBQRLD8UxNTVFWVoaffvoJ9+/fx4EDB/DLL79I5GVsbIzCwkKEhYUhNzcXRUVFMDY2ho+PDyZMmIATJ04gPT0dERERUhsAjcXS0hJ9+vTBlClTcOPGDcTFxWHKlCkQCARcb1d13NzcoKamhtWrV2P8+PES27799luEhYVh1apVuHv3LgIDA7Fjxw5uUg0A6NWrF3bs2IG4uDjExMRg2rRpUqc2b4i6nAPg9cQZenp68PLygomJCbp27cpt8/LygqKiInx8fHD79m2Eh4dj1qxZGDt2LHR1desVj7GxMcLCwpCdnc01UpcuXYr9+/djxYoVuHPnDoRCIQ4dOoQffvjh7Q6+kroeh5eXF27evIk1a9Zg+PDhUFBQqDbPdxU7IYSQjxc1pAip0P5LwHM/oKYvuV7N4PX6JnqPVExMDOzs7GBnZwcA+Oabb2BnZ4elS5cCAP7991+cOnUK//zzDzdlecXf1atXuXzS0tKQm5tbY1kuLi4Qi8USz0K5urpWWWdra4vNmzdj/fr16NixI4KCgrBu3TqJvJydnTFt2jSMHDkSOjo62LBhAwBg165dGD58OKZPnw5LS0tMnjwZL1++fJsqqtX+/fuhq6uLHj16YMiQIZg8eTJUVVVrHaIlIyODcePGQSwWw9vbW2Kbvb09jhw5gkOHDqFjx45YunQpVq5cyU1NDgCbNm2CoaEhunfvjjFjxsDX15eb9fBt1eUcAK97GkePHo2EhIQqQ9mUlJQQEhKCZ8+eoXPnzhg+fHidGtzSbNq0CRcvXoShoSF3rbq5ueH06dO4cOECOnfujM8++wxbtmyBkZFRww66GnU9DlNTU3Tp0gW3bt2qcVjfu4ydEELIx4vHKg/w/wQVFBRAXV0d+fn59X7uhLy95ORkiEQiyMrKwtLSskF5lJSUID09HSYmJm//fEO5GJmXD4H3MgdMuWWzDecjDffPP//A0NCQm6yhJhMnTsSTJ0+4CRsIIYSQhmrU+xHSbOraNqBnpAipTIYPpQ5uEIvFr5+FoUbUe++vv/5CYWEhrK2tkZWVhQULFsDY2Bg9evSodp/8/HwkJibi4MGD1IgihBBCSL1RQ4oQKbS1tZs7BFIPZWVl+P7773H//n2oqqrC2dkZQUFBNT6v5OHhgRs3bmDatGno27fvO4yWEEIIIR8DakgRQj54bm5ucHNzq9c+ERERTRMMIYQQQj4JNNkEIYQQQgghhNQTNaQIIYQQQgghpJ6oIUUIIYQQQggh9UQNKUIIIYQQQgipJ2pIEUIIIYQQQkg90ax9hEiRm5vLvUeKpkInhBBCCCGVUY8UIZWIy8WIvB+JEyknEHk/EuJycXOH1KwiIiLA4/Hw/PlzqdszMjLA4/EQHx//TuNqqNqO511avnw5dHV1wePxcOLEiWrXkf+8T+ePEELIp40aUoS8IfRBKNyC3bBcuBw70ndguXA53ILdEPogtMnKvHTpEtzd3WFgYFDtzfPy5cthaWkJZWVlaGpqok+fPrh+/Xq9ygkICICGhkbjBP0GQ0NDZGVloWPHjo2e98dMKBRixYoV2L17N7KystC/f3+p6yp73xsSjdmwdnV1xdy5c986H0IIIaQpUEOKkP8JfRCKbyK+weOixxLrc4py8E3EN03WmHr58iVsbW2xc+fOatOYm5tjx44dSExMxJUrV2BsbIwvvvgCT548aZKY6oPP50NPTw+ysjRSuD7S0tIAAB4eHtDT04OCgoLUdYQQQgh5P1FDihC8Hs7nd8MPDKzKtop162+sb5Jhfv3798fq1asxZMiQatOMGTMGffr0Qdu2bdGhQwds3rwZBQUFuHXrVp3KiIiIwPjx45Gfnw8ejwcej4fly5cDAA4cOABHR0eoqqpCT08PY8aMQU5OTrV5FRUVoX///ujWrRueP39epQeiosckLCwMjo6OUFJSgrOzM1JSUiTyWb16NVq2bAlVVVVMmjQJ3333HTp16lRtuWKxGBMnToSJiQkEAgEsLCywbdu2Wo/97NmzMDc3h0AgQM+ePZGRkSGx/enTpxg9ejRatWoFJSUlWFtb4/fff+e2/3979x0WxbX+Afy7sLAsvUkzVEGKQRSxgAUMGCwxWKJGCSW2WNEYu7EF+41GTWK9BsQLRI0RFX8WRMFOk0WUIigKKkhQURBFYM/vDy5zWWmLgii8n+fZ53HPnDnzzpzdZF7OmbNBQUHQ0tJCaWmpxH7Dhg2Dl5dXncdNTk7GZ599BqFQCC0tLUyePBnFxcUAKkcYhw4dCgCQkZHh+uPNsjfdvXsX/fv3BwBoaGiAx+PB19cXAFBaWgo/Pz/o6OhAQUEBffr0QVxcXL3XJjc3F0OGDIFQKISpqSlCQkJgYmKCzZs3c3Wys7Ph4eEBZWVlqKqqYvTo0Xj06FGdbZqamgIAunbtCh6PBxcXFwCAWCzGTz/9hE8++QQCgQBdunTByZMn62zH19cX0dHR2LJlC/eZrd53CQkJ9X6+jhw5Ant7eygoKMDMzAwrV65EeXl5nceLiopCjx49oKSkBHV1dfTu3Rv37t0DUNlfXbp0wc6dO2FoaAhFRUWMHj0az5494/aPi4vDgAEDoK2tDTU1NTg7O+PatWt1Ho8QQsjHjxIp0qYVFBQgLS0NYfFhNUaiqmNgyCvJQ1h8GNLS0lBQUPAeo5T0+vVr7Nq1C2pqarCzs5NqHycnJ2zevBmqqqrIzc1Fbm4u5s6dCwAoKyuDv78/kpKSEBYWhrt373I3528qLCzEgAEDIBaLERERUe9UwSVLlmDjxo2Ij48Hn8/H+PHjuW3BwcFYvXo11q9fj4SEBBgZGWH79u31noNYLMYnn3yCgwcPIiUlBcuWLcPixYtx4MCBOvfJycnBiBEjMHToUIhEIi5hq+7Vq1fo1q0bjh8/jhs3bmDy5Mnw8vJCbGwsAGDUqFGoqKjA0aNHuX3y8/Nx/PhxiXOq7sWLF3B3d4eGhgbi4uJw8OBBnDlzBjNmzAAAzJ07FwEBAQAg0R9vlr3J0NAQhw4dAgCkp6cjNzeXSybnz5+PQ4cOYe/evbh27RrMzc3h7u6OJ0+e1Hl9vL298fDhQ0RFReHQoUPYtWuXRBItFovh4eGBJ0+eIDo6GhEREbhz5w7GjBlTZ5tV1+3MmTPIzc3F33//DQDYsmULNm7ciJ9//hnXr1+Hu7s7vvzyS2RkZNTazpYtW+Do6IhJkyZx18PQ0JDbXt/n68KFC/D29sasWbOQkpKCnTt3IjAwEKtXr671WOXl5Rg2bBicnZ1x/fp1XLlyBZMnT5ZIZjMzM3HgwAEcO3YMJ0+eRGJiIqZNm8ZtLyoqgo+PDy5evIirV6/CwsICgwcPRlFRUZ3XihBCyEeOEfbs2TMGgD179qylQ2mTUlNTWXJyMktNTX3rNl6+fMlSUlLYy5cvG7VfXl4eS05OZjuid7BPAz9t8LUjegdLTk5meXl5bx1rfQCww4cP17rt2LFjTElJifF4PGZgYMBiY2Mb1XZAQABTU1NrsF5cXBwDwIqKihhjjJ07d44BYKmpqaxz585s5MiRrLS0lKuflZXFALDExESJ+mfOnOHqHD9+nAHg+qdnz55s+vTpEsft3bs3s7Oza9Q5TZ8+nY0cObLO7YsWLWI2NjYSZQsWLGAA2NOnT+vcb8iQIeyHH37g3k+dOpUNGjSIe79x40ZmZmbGxGJxrfvv2rWLaWhosOLiYq7s+PHjTEZGhvvsHD58mL35n+Dayt5UdX2rx19cXMzk5ORYcHAwV/b69WtmYGDANmzYUGs7qampDACLi4vjyjIyMhgA9ssvvzDGGDt9+jSTlZVl2dnZXJ2bN28yAHV+/t78PFQxMDBgq1evlijr3r07mzZtWp3n6uzszGbNmiVRJs3ny9XVla1Zs0Ziv3379jF9ff1aj/P48WMGgEVFRdW6ffny5UxWVpbdv3+fKztx4gSTkZFhubm5te5TUVHBVFRU2LFjx+o8P0JI6/O29yPkwyJtbkAjUqRNk5WVBZ/Ph7aCdEucaytog8/nQ1ZWtpkjq6l///4QiUS4fPkyBg4ciNGjR9c7BU9aCQkJGDp0KIyMjKCiogJnZ2cAlVO6qhswYADMzc2xf/9+yMvLN9hu586duX/r6+sDABdveno6evToIVH/zfe1+f3339GtWze0a9cOysrK2LVrV404q0tNTUXPnj0lyhwdHSXeV1RUwN/fH7a2ttDU1ISysjJOnTol0e6kSZNw+vRpPHjwAEDlwh2+vr61Tr+rOq6dnR2UlJS4st69e0MsFteYgtYUbt++jbKyMvTu3Zsrk5OTQ48ePZCamlrrPunp6eDz+bC3t+fKzM3NoaGhIXEehoaGEiNBNjY2UFdXr7Pd2jx//hwPHz6UiA+ovCaNaae6+j5fSUlJ+Omnn6CsrMy9qka2SkpKarSlqakJX19fuLu7Y+jQodiyZUuNEUEjIyO0b9+ee+/o6CjRn48ePcKkSZNgYWEBNTU1qKqqori4uN7PJyGEkI8bJVKkTdPW1oaVlRWGOQyDrqIueKj9xpgHHvQU9TDMYRisrKxa5LellJSUYG5ujl69emHPnj3g8/nYs2fPO7VZNQVNVVUVwcHBiIuLw+HDhwFUTiGsbsiQITh//jxSUlKkaltOTo77d1XCIRaL3zrWP//8E3PnzsWECRNw+vRpiEQifPvttzXibKx//etf2LJlCxYsWIBz585BJBLB3d1dot2uXbvCzs4OQUFBSEhIwM2bN+uc/kjej/o+X8XFxVi5ciVEIhH3Sk5ORkZGBhQUFGptLyAgAFeuXIGTkxP279+Pjh074urVq1LH4+PjA5FIhC1btuDy5csQiUTQ0tJ6588nIYSQDxclUoQAkJWRxcIelc/OvJlMVb1f0GMBZGXe/0hUXcRicY0FEOojLy+PigrJxTLS0tLw+PFjrFu3Dn379oWVlVWdo1zr1q2Dj48PXF1dpU6m6mJpaVljIYSGFka4dOkSnJycMG3aNHTt2hXm5ubcKnd1sba25p7ZqfLmzfGlS5fg4eGBb775BnZ2djAzM8OtW7dqtDVx4kQEBgYiICAAbm5uEqM0tR03KSkJL168kDiOjIwMLC0t6425IVWjgdX7skOHDpCXl8elS5e4srKyMsTFxcHGxqbWdiwtLVFeXo7ExESuLDMzE0+fPpU4j5ycHOTk5HBlKSkpKCwsrLPd2uJTVVWFgYGBRHxA5TWpq52qtt78zErD3t4e6enpMDc3r/GSkan7f3tdu3bFokWLcPnyZXz66acICQnhtmVnZ+Phw4fc+6tXr0r056VLl+Dn54fBgwejU6dOEAgELfosJSGEkOZHiRQh/+Vm7IZNLpugo6gjUa6rqItNLpvgZuzWLMctLi7m/moOAFlZWRCJRNyUoBcvXmDx4sW4evUq7t27h4SEBIwfPx4PHjzAqFGjuHZcXV3x22+/1XkcExMTFBcXIzIyEgUFBSgpKYGRkRHk5eXx66+/4s6dOzh69Cj8/f3rbOPnn3+Gp6cnPvvsM6Slpb31Oc+cORN79uzB3r17kZGRgVWrVuH69et1TpUDAAsLC8THx+PUqVO4desWli5d2mDyNWXKFGRkZGDevHlIT09HSEgIAgMDa7QbERGBy5cvIzU1Fd99912tq9KNGzcO9+/fx+7du+tcZKKKp6cnFBQU4OPjgxs3buDcuXOYOXMmvLy8oKurW+++DTE2NgaPx0N4eDj++ecfFBcXQ0lJCVOnTsW8efNw8uRJpKSkYNKkSSgpKcGECRNqbcfKygpubm6YPHkyYmNjkZiYiMmTJ0MoFHL94ObmBltbW3h6euLatWuIjY2Ft7c3nJ2d4eDgUGu7Ojo6EAqFOHnyJB49esStbDdv3jysX78e+/fvR3p6OhYuXAiRSIRZs2bVea4mJiaIiYnB3bt3UVBQIPWI5rJlyxAUFISVK1fi5s2bSE1NxZ9//okff/yx1vpZWVlYtGgRrly5gnv37uH06dPIyMiAtbU1V6eqP5OSknDhwgX4+flh9OjR0NPTA1D5Odq3bx9SU1MRExMDT09PCIVCqeIlhBDykXpPz2x90GixiZbVkotN1Ka8opz9FfMX2xG9g/0V8xcrryh/5zbrU/Xw/JsvHx8fxljluQ0fPpwZGBgweXl5pq+vz7788ssaD/sbGxuz5cuX13usKVOmMC0tLQaAqxsSEsJMTEyYQCBgjo6O7OjRo7UuHlF9cYOZM2cyfX19lp6eXudiE9XrJyYmMgAsKyuLK/vpp5+YtrY2U1ZWZuPHj2d+fn6sV69edcb+6tUr5uvry9TU1Ji6ujqbOnUqW7hwYYMLVBw7doyZm5szgUDA+vbty/744w+J+B4/fsw8PDyYsrIy09HRYT/++CPz9vZmHh4eNdry8vJimpqa7NWrV/UekzHGrl+/zvr3788UFBSYpqYmmzRpEreAB2Nvv9gEY5XXTk9Pj/F4PInPycyZM5m2tjYTCASsd+/eDS5I8vDhQzZo0CAmEAiYsbExCwkJYTo6OmzHjh1cnXv37rEvv/ySKSkpMRUVFTZq1KgGF1vZvXs3MzQ0ZDIyMszZ2ZkxVrn4wooVK1j79u2ZnJwcs7OzYydOnKi3nfT0dNarVy8mFAq5z4+0n6+TJ08yJycnJhQKmaqqKuvRowfbtWtXrcfJy8tjw4YNY/r6+kxeXp4ZGxuzZcuWsYqKCsZY5WITdnZ2bNu2bczAwIApKCiwr776ij158oRr49q1a8zBwYEpKCgwCwsLdvDgQWZsbMwt3EEIaRtosYnWQdrcgMcYq/nDOW3M8+fPoaamhmfPnkFVVbWlw2lz0tLSUF5eDj6fDysrq7dq49WrV8jKyoKpqWmdz0C875hI4wwYMAB6enrYt29fS4dSJ1dXV3Tq1Albt25t6VCaxf3792FoaIgzZ87A1dW1pcP5YKxYsQJhYWHcqDEhhNSlqe9HSMuQNjfgv8eYCPloaGtro6KiokVW52sLSkpKsGPHDri7u0NWVhahoaE4c+YMIiIiWjq0Wj19+hRRUVGIiorCtm3bWjqcJnP27FkUFxfD1tYWubm5mD9/PkxMTNCvX7+WDo0QQgj54FEiRUgtWmJVvraEx+Ph//7v/7B69Wq8evUKlpaWOHToENzcmuc5tHfVtWtXPH36FOvXr3/nxSI+JGVlZVi8eDHu3LkDFRUVODk5ITg4WGJFPEIIIYTUjqb2gab2tbQPcWofIYQQQkhj0f1I6yBtbkCr9hFCCCGEEEJII1EiRQghhBBCCCGNRIkUIYQQQgghhDQSJVKEEEIIIYQQ0ki0ah8htSgoKOCWP6cV/AghhBBCyJsokSKkFgUFBdxKgpRIEUIIIYSQN9HUPkLewCoqwJJvgHfxIljyDbCKipYOqUXweDyEhYU1aZu+vr4YNmxYk7bZVKKiosDj8VBYWAgACAwMhLq6Ord9xYoV6NKlC/f+QziX5ugjoOa5NhUTExNs3ry5ydslhBBCWgIlUoRU8/z0aWS6ugFLl0Jm8xZg6VJkurrh+enTzXbM8+fPY+jQoTAwMJDqxnjKlCng8XjNfkOam5uLQYMGNWmbW7ZsQWBgYJO22VScnJyQm5sLNTU1qep/COfSFH3UXMlYU6grofuQYyaEENJ2UCJFyH89P30aD2bNRnlenkR5+aNHeDBrdrMlUy9evICdnR1+//33BusePnwYV69ehYGBQbPEUp2enh4EAkGTtqmmpiYxyvMhkZeXh56eHng8nlT1P4RzaY4+akvKyspaOgRCCCEfMUqkCEHldL5Ha9YCjNWysbLs0Zq1zTLNb9CgQVi1ahWGDx9eb70HDx5g5syZCA4OhpycXKOOwRhDu3bt8Ndff3FlXbp0gb6+Pvf+4sWLEAgEKCkpASD5V/+7d++Cx+Ph77//Rv/+/aGoqAg7OztcuXKF279qKtypU6dgbW0NZWVlDBw4ELm5uVydN6fDubi4wM/PD/Pnz4empib09PSwYsUKidjT0tLQp08fKCgowMbGBmfOnGlwRMLFxQUzZ87E7NmzoaGhAV1dXezevRsvXrzAt99+CxUVFZibm+PEiRPcPm9O7WvIm+dSWloKPz8/6OjoQEFBAX369EFcXFyN9iMjI+Hg4ABFRUU4OTkhPT2dq5OUlIT+/ftDRUUFqqqq6NatG+Lj4+uMobF99CYTExMAwPDhw8Hj8bj3Vfbt2wcTExOoqanh66+/RlFREbdNLBZj7dq1MDU1hVAohJ2dncTnSxrZ2dnw8PCAsrIyVFVVMXr0aDx69AhA5edp5cqVSEpKAo/HA4/HQ2BgYL0xHzlyBPb29lBQUICZmRlWrlyJ8vJyieu1fft2fPnll1BSUsKqVatgbm6On3/+WSIukUgEHo+HzMzMRp0PIYSQtoUSKdKmFRQUIC0tDWl/H64xEiWBMZTn5SHt78NIS0tDQUHB+wsSlTetXl5emDdvHjp16tTo/Xk8Hvr164eoqCgAwNOnT5GamoqXL18iLS0NABAdHY3u3btDUVGxznaWLFmCuXPnQiQSoWPHjhg7dqzEjWpJSQl+/vln7Nu3D+fPn0d2djbmzp1bb2x79+6FkpISYmJisGHDBvz000+IiIgAAFRUVGDYsGFQVFRETEwMdu3ahSVLlkh1znv37oW2tjZiY2Mxc+ZMTJ06FaNGjYKTkxOuXbuGzz//HF5eXlzi+K7mz5+PQ4cOYe/evbh27RrMzc3h7u6OJ0+eSNRbsmQJNm7ciPj4ePD5fIwfP57b5unpiU8++QRxcXFISEjAwoULG500N9RH1VUlegEBAcjNzZVI/G7fvo2wsDCEh4cjPDwc0dHRWLduHbd97dq1CAoKwo4dO3Dz5k18//33+OabbxAdHS1VnGKxGB4eHnjy5Amio6MRERGBO3fuYMyYMQCAMWPG4IcffkCnTp2Qm5uL3NxcjBkzps6YL1y4AG9vb8yaNQspKSnYuXMnAgMDsXr1aonjrlixAsOHD0dycjImTJiA8ePHIyAgQKJOQEAA+vXrB3Nzc6nOhRBCSNtEiRRp0yoqKlBeXg7xY+kSI/HjytX8Kt7zAhTr168Hn8+Hn5/fW7fh4uLCJVLnz59H165dJcqioqLg7Oxcbxtz587FkCFD0LFjR6xcuRL37t2T+Kt9WVkZduzYAQcHB9jb22PGjBmIjIyst83OnTtj+fLlsLCwgLe3NxwcHLh9IiIicPv2bQQFBcHOzg59+vSpcWNcFzs7O/z444+wsLDAokWLoKCgAG1tbUyaNAkWFhZYtmwZHj9+jOvXr0vVXn1evHiB7du341//+hcGDRoEGxsb7N69G0KhEHv27JGou3r1ajg7O8PGxgYLFy7E5cuX8erVKwCVIzRubm6wsrKChYUFRo0aBTs7u0bF0lAfVdeuXTsAgLq6OvT09Lj3QGWiExgYiE8//RR9+/aFl5cX1y+lpaVYs2YN/vjjD7i7u8PMzAy+vr745ptvsHPnTqnijIyMRHJyMkJCQtCtWzf07NkTQUFBiI6ORlxcHIRCIZSVlcHn86Gnpwc9PT0IhcI6Y165ciUWLlwIHx8fmJmZYcCAAfD3968Rz7hx4/Dtt9/CzMwMRkZG8PX1RXp6OmJjYwFUfoZDQkIkElxCCCGkNpRIkTZNVlYWfD4fMlrSLXEuo6UNPp8PWVnZZo7sfxISEriFDaR9fqc2zs7OSElJwT///IPo6Gi4uLhwiVRZWRkuX74MFxeXetvo3Lkz9++qaYH5+flcmaKiIjp06CBRp/r2htp8c5/09HQYGhpCT0+P296jR4/6T7SWdmVlZaGlpQVbW1uuTFdXt0b8b+v27dsoKytD7969uTI5OTn06NEDqampdcb15jWcM2cOJk6cCDc3N6xbtw63b99udCwN9ZG0TExMoKKiItFWVTuZmZkoKSnBgAEDoKyszL2CgoKkjjk1NRWGhoYwNDTkymxsbKCurl7jmkkjKSkJP/30k0Q8kyZNQm5ursSoo4ODg8R+BgYGGDJkCP744w8AwLFjx1BaWopRo0Y1OgZCCCFtCyVSpE3T1taGlZUVrEYMB19PD6grUeHxwNfTg9WI4bCysnqvvy114cIF5Ofnw8jICHw+H3w+H/fu3cMPP/xQ45mW+tja2kJTUxPR0dESiVTVCEBZWRmcnJzqbaP6NLOqpE4sFte6vaoOq+25szrarNqneptvq7Z2G4r/fagvhhUrVuDmzZsYMmQIzp49CxsbGxw+fLjJ2n/bdqraqmqnuLgYAHD8+HGIRCLulZKS0ujnpJpKcXExVq5cKRFPcnIyMjIyoKCgwNVTUlKqse/EiRPx559/4uXLlwgICMCYMWPqneJKCCGEAPSDvIQAAHiystBdvAgPZs2uTKaq3/z/92ZUd/Ei8N7jSFQVLy8vuLm5SZS5u7vDy8sL3377rdTt8Hg89O3bF0eOHMHNmzfRp08fKCoqorS0FDt37oSDg0OtN5ktydLSEjk5OXj06BE3glT9OZ4PRYcOHSAvL49Lly7B2NgYQOUUsbi4OMyePbtRbXXs2BEdO3bE999/j7FjxyIgIKDBhUjehZycXKOnqtrY2EAgECA7O7vB6aB1sba2Rk5ODnJycrhRqZSUFBQWFsLGxgZA5UqKtcVWW8z29vZIT09/q+eaBg8eDCUlJWzfvh0nT57E+fPn3+KMCCGEtDWUSBHyX6qffw5s2YxHa9ZKLDzB19WF7uJFldubQXFxscQzLFlZWRCJRNDU1ISRkRG0tLSgpaUlsY+cnBz09PRgaWnJlbm6umL48OGYMWNGncdycXHBDz/8AAcHBygrKwMA+vXrh+DgYMybN6+Jz+zdDRgwAB06dICPjw82bNiAoqIi/PjjjwDwTtMcm5qSkhKmTp2KefPmcf22YcMGlJSUYMKECVK18fLlS8ybNw9fffUVTE1Ncf/+fcTFxWHkyJHNGruJiQkiIyPRu3dvCAQCaGhoNLiPiooK5s6di++//x5isRh9+vTBs2fPcOnSJaiqqsLHx6fBNtzc3GBrawtPT09s3rwZ5eXlmDZtGpydnbnpdyYmJtz34ZNPPoGKigoEAkGtMS9btgxffPEFjIyM8NVXX0FGRgZJSUm4ceMGVq1aVW8ssrKy8PX1xaJFi2BhYQFHR0fpLh4hhJA2jab2EVKN6uefwzzyDODvD/HsWYC/P8wjzzRbEgUA8fHx6Nq1K7p27Qqg8jmZrl27YtmyZY1q5/bt2w2uJujs7IyKigqJZ6FcXFxqlH0oZGVlERYWhuLiYnTv3h0TJ07kVu2rPl3rQ7Bu3TqMHDkSXl5esLe3R2ZmJk6dOiVVYgJUnuvjx4/h7e2Njh07YvTo0Rg0aBBWrlzZrHFv3LgRERERMDQ05D6D0vD398fSpUuxdu1aWFtbY+DAgTh+/DhMTU2l2p/H4+HIkSPQ0NBAv3794ObmBjMzM+zfv5+rM3LkSAwcOBD9+/dHu3btEBoaWmfM7u7uCA8Px+nTp9G9e3f06tULv/zyCzdC2JAJEybg9evXjRrlJYQQ0rbxWEMPMLQBz58/h5qaGp49ewZVVdWWDqfNSUtLQ3l5Ofh8PqysrN6qjVevXiErKwumpqZNcoPdFDGR5nHp0iX06dMHmZmZEgtbEPIuLly4AFdXV+Tk5HDTSAkhpLGa+n6EtAxpcwOa2kdanLa2NioqKt7rSngN+RBjaqsOHz4MZWVlWFhYIDMzE7NmzULv3r0piSJNorS0FP/88w9WrFiBUaNGURJFCCFEapRIkRb3PlfAk9aHGFNbVVRUhAULFiA7Oxva2tpwc3PDxo0bWzos0kqEhoZiwoQJ6NKlC4KCglo6HEIIIR8RmtoHmtrXGtBQOiGEEEJaGt2PtA7S5ga02AQhhBBCCCGENBIlUoQQQgghhBDSSJRIEUIIIYQQQkgjUSJFCCGEEEIIIY1Eq/YRUouCggJu+XNawY8QQgghhLyJEilCalFQUMD9IC8lUoQQQggh5E00tY+QN4jFDE9zSpF/6xWe5pRCLG7zvxDQZKKiosDj8VBYWFhnnRUrVqBLly7vLaamYGJigs2bNzdZe3fv3gWPx4NIJGqyNon0PoTr7+LigtmzZzd5u29+BwMDA6Gurt7kxyGEkLaAEilCqrmdmI+gxZdx7dBjpJ5+jmuHHiNo8WXcTsxvtmOeP38eQ4cOhYGBAXg8HsLCwmrU8fX1BY/Hk3gNHDiw2WKSRnPdgM2dOxeRkZFN3m5ziouLw+TJk1s6jPfC19cXw4YNa/E2CCGEkJZGiRQh/3U7MR8nd97Ai8JSifIXhaU4ufNGsyVTL168gJ2dHX7//fd66w0cOBC5ubncKzQ0tFniaWnKysrQ0tJq6TCk8vr1awBAu3btoKio2MLRNK+KigqIxeKWDoMQQgj5YFAiRQgqp/Nd2J9Rb52LBzKaZZrfoEGDsGrVKgwfPrzeegKBAHp6etxLQ0ND6mMwxtCuXTv89ddfXFmXLl2gr6/Pvb948SIEAgFKSkoAAJs2bYKtrS2UlJRgaGiIadOmobi4GEDl9KBvv/0Wz54940bIVqxYAQAoLS3FggULYGhoCIFAAHNzc+zZs0cinoSEBDg4OEBRURFOTk5IT0/ntr05ta9q9OLnn3+Gvr4+tLS0MH36dJSVlXF1cnNzMWTIEAiFQpiamiIkJKTB6XZxcXEYMGAAtLW1oaamBmdnZ1y7dq3e61gVy+rVq2FgYABLS0sAklP7xo0bhzFjxkjsV1ZWBm1tbQQFBQEATp48iT59+kBdXR1aWlr44osvcPv27XqP/aZt27bBwsICCgoK0NXVxVdffcVtc3FxwYwZMzBjxgyoqalBW1sbS5cuBWP/+/w+ffoU3t7e0NDQgKKiIgYNGoSMjP99B6pGHI8ePQobGxsIBAKMHz8ee/fuxZEjR7h+j4qKwuvXrzFjxgzo6+tDQUEBxsbGWLt2ba1xr1ixotY2ACAnJwejR4+Guro6NDU14eHhgbt373L7StNnPB4PO3fuxBdffAFFRUVYW1vjypUryMzMhIuLC5SUlODk5NTo6x0dHY0ePXpAIBBAX18fCxcuRHl5Obf9r7/+gq2tLYRCIbS0tODm5oYXL15w2//973/D2toaCgoKsLKywrZt2xo8Znl5eb19uG/fPjg4OEBFRQV6enoYN24c8vMl/+Dzf//3f+jYsSOEQiH69+8vcT3fdPfuXcjIyCA+Pl6ifPPmzTA2NqZEmhBC3kCJFGnTCgoKkJaWhpjI6zVGot5U/LQUMZHXkZaWhoKCgvcU4f9ERUVBR0cHlpaWmDp1Kh4/fiz1vjweD/369eNuWJ8+fYrU1FS8fPkSaWlpACpvFLt3786NrMjIyGDr1q24efMm9u7di7Nnz2L+/PkAACcnJ2zevBmqqqrcCNncuXMBAN7e3ggNDcXWrVuRmpqKnTt3QllZWSKeJUuWYOPGjYiPjwefz8f48ePrjf/cuXO4ffs2zp07h7179yIwMBCBgYHcdm9vbzx8+BBRUVE4dOgQdu3aVeOG8k1FRUXw8fHBxYsXcfXqVVhYWGDw4MEoKiqqd7/IyEikp6cjIiIC4eHhNbZ7enri2LFjXNIJAKdOnUJJSQmXLL948QJz5sxBfHw8IiMjISMjg+HDh0t9oxofHw8/Pz/89NNPSE9Px8mTJ9GvXz+JOnv37gWfz0dsbCy2bNmCTZs24d///je33dfXF/Hx8Th69CiuXLkCxhgGDx4skaCWlJRg/fr1+Pe//42bN29i69atGD16tMToqJOTE7Zu3YqjR4/iwIEDSE9PR3BwMExMTGqNfe7cubW2UVZWBnd3d6ioqODChQu4dOkSlJWVMXDgQG7kT9o+8/f3h7e3N0QiEaysrDBu3Dh89913WLRoEeLj48EYw4wZM6S61gDw4MEDDB48GN27d0dSUhK2b9+OPXv2YNWqVQAqE/mxY8di/PjxSE1NRVRUFEaMGMElPcHBwVi2bBlWr16N1NRUrFmzBkuXLsXevXvrPW5DfVhWVgZ/f38kJSUhLCwMd+/eha+vL7c9JycHI0aMwNChQyESiTBx4kQsXLiwzuOZmJjAzc0NAQEBEuUBAQHw9fWFjAzdMhBCiARG2LNnzxgA9uzZs5YOhbylly9fspSUFPby5ctG7ZeXl8eSk5NZ5KE49tt3kQ2+Ig/FseTkZJaXl9cs5wGAHT58uEZ5aGgoO3LkCLt+/To7fPgws7a2Zt27d2fl5eVSt71161bWqVMnxhhjYWFhrGfPnszDw4Nt376dMcaYm5sbW7x4cZ37Hzx4kGlpaXHvAwICmJqamkSd9PR0BoBFRETU2sa5c+cYAHbmzBmu7Pjx4wwA13fLly9ndnZ23HYfHx9mbGwsca6jRo1iY8aMYYwxlpqaygCwuLg4bntGRgYDwH755Zd6roikiooKpqKiwo4dO1ZnHR8fH6arq8tKS0slyo2NjbljlZWVMW1tbRYUFMRtHzt2LBdvbf755x8GgCUnJzPGGMvKymIAWGJiYq31Dx06xFRVVdnz589r3e7s7Mysra2ZWCzmyhYsWMCsra0ZY4zdunWLAWCXLl3ithcUFDChUMgOHDjAGKvsXwBMJBLVuAYeHh4SZTNnzmSfffaZxPHqU1sb+/btY5aWlhJtlJaWMqFQyE6dOlVrO7X1GQD2448/cu+vXLnCALA9e/ZwZaGhoUxBQaHO+N68/osXL64R2++//86UlZVZRUUFS0hIYADY3bt3a22vQ4cOLCQkRKLM39+fOTo61hlDQ31Ym7i4OAaAFRUVMcYYW7RoEbOxsZGos2DBAgaAPX36lDFW83u8f/9+pqGhwV69esUYYywhIYHxeDyWlZVV53EJIf/ztvcj5MMibW5Af14ibZqsrCz4fD6EKnJS1ReqyIHP50NWVraZI5P09ddf48svv4StrS2GDRuG8PBwxMXFcSNM0nB2dkZKSgr++ecfREdHw8XFBS4uLoiKikJZWRkuX74MFxcXrv6ZM2fg6uqK9u3bQ0VFBV5eXnj8+DE39a82IpEIsrKycHZ2rjeWzp07c/+uml5Y3whSp06dJK65vr4+Vz89PR18Ph/29vbcdnNz8wanPj569AiTJk2ChYUF1NTUoKqqiuLiYmRnZ9e7n62tLeTl5evczufzMXr0aAQHBwOoHH06cuQIPD09uToZGRkYO3YszMzMoKqqyo3eNHTsKgMGDICxsTHMzMzg5eWF4ODgGv3Sq1cv8Hg87r2joyMyMjJQUVGB1NRU8Pl89OzZk9uupaUFS0tLpKamcmXy8vISfVUXX19fiEQiWFpaws/PD6dPn5bqPKpLSkpCZmYmVFRUoKysDGVlZWhqauLVq1fcNDxp+6x6zLq6ugAq+6162atXr/D8+XOpYktNTYWjo6PE9ezduzeKi4tx//592NnZwdXVFba2thg1ahR2796Np0+fAqjs/9u3b2PChAnceSkrK2PVqlUNTi+srw+ByimyQ4cOhZGREVRUVLjvXdX1SE1NlejjqjbqM2zYMMjKyuLw4cMAKqd49u/fv84RRkIIacsokSJtmra2NqysrNDTtTOU1AX11lXWEKCna2dYWVm1+G9LmZmZQVtbG5mZmVLvY2trC01NTURHR0skUtHR0YiLi0NZWRmcnJwAVD4r8cUXX6Bz5844dOgQEhISuMUwqqZZ1UYoFEoVi5zc/xLXqhvF+qa1Va9ftc+7Pq/h4+MDkUiELVu24PLlyxCJRNDS0qr3/ABASUmpwbY9PT0RGRmJ/Px8hIWFQSgUSqyyOHToUDx58gS7d+9GTEwMYmJiANR/batTUVHBtWvXEBoaCn19fSxbtgx2dnb1Liv/NoRCocSNfF3s7e2RlZUFf39/vHz5EqNHj5Z4ZksaxcXF6NatG0QikcTr1q1bGDduHADp+6y2z1djP3ONISsri4iICJw4cQI2Njb49ddfYWlpiaysLG6K5+7duyXO68aNG7h69epbH/PFixdwd3eHqqoqgoODERcXxyU/0n6OaiMvLw9vb28EBATg9evXCAkJaXDqLSGEtFWUSBECQEaGh75jLOqt02e0BWRkGr6pfB/u37+Px48fSywW0RAej4e+ffviyJEjuHnzJvr06YPOnTujtLQUO3fuhIODA5ckJCQkQCwWY+PGjejVqxc6duyIhw8fSrQnLy/P/WW8iq2tLcRiMaKjo9/9JKVkaWmJ8vJyJCYmcmWZmZnciEBdLl26BD8/PwwePBidOnWCQCBosmffnJycYGhoiP379yM4OBijRo3ibuQfP36M9PR0/Pjjj3B1dYW1tXWDsdaGz+fDzc0NGzZswPXr13H37l2cPXuW216VnFWpeqZIVlYW1tbWKC8vl6hTFZeNjU29x62t3wFAVVUVY8aMwe7du7F//34cOnQIT548kboNe3t7ZGRkQEdHB+bm5hIvNTU1AM3bZ/WpWrCCVVvo4dKlS1BRUcEnn3wCoPL71bt3b6xcuRKJiYmQl5fH4cOHoaurCwMDA9y5c6fGeZmamtZ73Pr6MC0tDY8fP8a6devQt29fWFlZ1RjVtba2RmxsbI02GjJx4kScOXMG27ZtQ3l5OUaMGNHgPoQQ0hZRIkXIf3XoqoOB331aY2RKWUOAgd99ig5ddZrluMXFxdxfqQEgKysLIpGIm55TXFyMefPm4erVq7h79y4iIyPh4eEBc3NzuLu7c+24urrit99+q/dYLi4uCA0NRZcuXaCsrAwZGRn069cPwcHBEtPxzM3NUVZWhl9//RV37tzBvn37sGPHDom2TExMUFxcjMjISBQUFKCkpAQmJibw8fHB+PHjERYWhqysLERFReHAgQNNdLVqsrKygpubGyZPnozY2FgkJiZi8uTJDY6mWFhYYN++fUhNTUVMTAw8PT2lHlGTxrhx47Bjxw5ERERITOvT0NCAlpYWdu3ahczMTJw9exZz5sxpVNvh4eHYunUrRCIR7t27h6CgIIjFYm4VQaByetecOXOQnp6O0NBQ/Prrr5g1axaAynP38PDApEmTcPHiRSQlJeGbb75B+/bt4eHhUe+xTUxMcP36daSnp6OgoABlZWXYtGkTQkNDkZaWhlu3buHgwYPQ09Or83fGamvD09MT2tra8PDwwIULF7jPjp+fH+7fv8/F3Zx9Vpdp06YhJycHM2fORFpaGo4cOYLly5djzpw5kJGRQUxMDNasWYP4+HhkZ2fj77//xj///ANra2sAwMqVK7F27Vps3boVt27dQnJyMgICArBp06Z6j1tfHxoZGUFeXp77jh49ehT+/v4S+0+ZMgUZGRmYN28e0tPTERISIrFIS12sra3Rq1cvLFiwAGPHjn0v15gQQj5K7+eRrbezZs0a5uDgwJSVlVm7du2Yh4cHS0tLk6jz8uVLNm3aNKapqcmUlJTYiBEjGr0QAC028fFryoc7KyrE7PJpEYs8FMcunxaxigrpHqB/W1ULMLz58vHxYYwxVlJSwj7//HPWrl07Jicnx4yNjdmkSZNqfM6NjY3Z8uXL6z1WYmIiA8AWLFjAlf3yyy8MADt58qRE3U2bNjF9fX0mFAqZu7s7CwoKknhInTHGpkyZwrS0tBgA7tgvX75k33//PdPX12fy8vLM3Nyc/fHHHxLnWr2NqpiqHmavbbGJNxcmmDVrFnN2dubeP3z4kA0aNIgJBAJmbGzMQkJCmI6ODtuxY0ed1+LatWvMwcGBKSgoMAsLC3bw4EGJRSNqU1ssjLFa90tJSWEAmLGxcY1FGCIiIpi1tTUTCASsc+fOLCoqSmKhkYYWm7hw4QJzdnZmGhoaTCgUss6dO7P9+/dz252dndm0adPYlClTmKqqKtPQ0GCLFy+WiOPJkyfMy8uLqampcX1869Ytbntti4kwxlh+fj4bMGAAU1ZWZgDYuXPn2K5du1iXLl2YkpISU1VVZa6uruzatWt1Xsfa2mCMsdzcXObt7c20tbWZQCBgZmZmbNKkSdx/m6Xps+rXsa5rWdvnsLra9omKimLdu3dn8vLyTE9Pjy1YsICVlZUxxir72t3dnbVr144JBALWsWNH9uuvv0q0GRwczLp06cLk5eWZhoYG69evH/v777/rvEbS9GFISAgzMTFhAoGAOTo6sqNHj9aI+9ixY8zc3JwJBALWt29f9scff9S72ESVPXv2MAAsNja2zhgJITXRYhOtg7S5AY+xanMVPjADBw7E119/je7du6O8vByLFy/GjRs3kJKSwk1Bmjp1Ko4fP47AwECoqalhxowZkJGRwaVLl6Q+zvPnz6GmpoZnz55BVVW1uU6HNKNXr14hKysLpqamUFBQeOf20tLSUF5eDj6fDysrqyaIkLxP9+/fh6GhIbdgRlvj4uKCLl261Ps7WoTUx9/fHwcPHsT169dbOhRCPipNfT9CWoa0uQH/PcbUaCdPnpR4HxgYCB0dHSQkJKBfv3549uwZ9uzZg5CQEHz22WcAKn/vwtraGlevXkWvXr1aImzSCmhra6OiouK9r85H3s7Zs2dRXFwMW1tb5ObmYv78+TAxManx20qEkPoVFxfj7t27+O2337jfySKEEFK7j+oZqWfPngEANDU1AVQ+EF9WVgY3NzeujpWVFYyMjHDlypU62yktLcXz588lXoRUp62tDV1d3RZfnY9Ip6ysDIsXL0anTp0wfPhwtGvXDlFRUTVW+yOE1G/GjBno1q0bXFxcaLU+QghpwAc9IlWdWCzG7Nmz0bt3b3z66acAgLy8PMjLy9d4oFlXVxd5eXl1trV27VqsXLmyOcMlhLxH7u7uEgtvtHWN+X0xQqoLDAyUakEKQgghH9GI1PTp03Hjxg38+eef79zWokWL8OzZM+6Vk5PTBBESQgghhBBC2oqPYkRqxowZCA8Px/nz57nf7AAAPT09vH79GoWFhRKjUo8ePYKenl6d7QkEAggE9f/4Kvk4fcBrpxBCCCGklaP7kLblgx6RYoxhxowZOHz4MM6ePVvjxwu7desGOTk5REZGcmXp6enIzs6Go6Pj+w6XtKCqZ2FKSkpaOBJCCCGEtFVV9yH0jG7b8EGPSE2fPh0hISE4cuQIVFRUuOee1NTUIBQKoaamhgkTJmDOnDnQ1NSEqqoqZs6cCUdHR1qxr42RlZWFuro68vPzAQCKior1/hgrIYQQQkhTYYyhpKQE+fn5UFdXp1V/24gP+nek6roRDggIgK+vL4DK9fp/+OEHhIaGorS0FO7u7ti2bVu9U/veRL8j1TowxpCXl4fCwsKWDoUQQgghbZC6ujr09PToj7kfOWlzgw86kXpfKJFqXSoqKlBWVtbSYRBCCCGkDZGTk6ORqFaiVfwgLyFvQ1ZWlv5DRgghhBBCmtUHvdgEIYQQQgghhHyIKJEihBBCCCGEkEaiRIoQQgghhBBCGomekcL/fjzt+fPnLRwJIYQQQgghpCVV5QQNrclHiRSAoqIiAIChoWELR0IIIYQQQgj5EBQVFUFNTa3O7bT8OQCxWIyHDx9CRUXlrdb9f/78OQwNDZGTk0PLp7dS1MetH/Vx20D93PpRH7d+1MdtQ0v2M2MMRUVFMDAwgIxM3U9C0YgUABkZGXzyySfv3I6qqip9oVs56uPWj/q4baB+bv2oj1s/6uO2oaX6ub6RqCq02AQhhBBCCCGENBIlUoQQQgghhBDSSJRINQGBQIDly5dDIBC0dCikmVAft37Ux20D9XPrR33c+lEftw0fQz/TYhOEEEIIIYQQ0kg0IkUIIYQQQgghjUSJFCGEEEIIIYQ0EiVShBBCCCGEENJIlEgRQgghhBBCSCNRIvWW1q1bBx6Ph9mzZ3Nlr169wvTp06GlpQVlZWWMHDkSjx49arkgSaOsWLECPB5P4mVlZcVtp/5tHR48eIBvvvkGWlpaEAqFsLW1RXx8PLedMYZly5ZBX18fQqEQbm5uyMjIaMGISWOZmJjU+C7zeDxMnz4dAH2XW4OKigosXboUpqamEAqF6NChA/z9/VF9/Sz6Ln/8ioqKMHv2bBgbG0MoFMLJyQlxcXHcdurjj8/58+cxdOhQGBgYgMfjISwsTGK7NH365MkTeHp6QlVVFerq6pgwYQKKi4vf41n8DyVSbyEuLg47d+5E586dJcq///57HDt2DAcPHkR0dDQePnyIESNGtFCU5G106tQJubm53OvixYvcNurfj9/Tp0/Ru3dvyMnJ4cSJE0hJScHGjRuhoaHB1dmwYQO2bt2KHTt2ICYmBkpKSnB3d8erV69aMHLSGHFxcRLf44iICADAqFGjANB3uTVYv349tm/fjt9++w2pqalYv349NmzYgF9//ZWrQ9/lj9/EiRMRERGBffv2ITk5GZ9//jnc3Nzw4MEDANTHH6MXL17Azs4Ov//+e63bpelTT09P3Lx5ExEREQgPD8f58+cxefLk93UKkhhplKKiImZhYcEiIiKYs7MzmzVrFmOMscLCQiYnJ8cOHjzI1U1NTWUA2JUrV1ooWtIYy5cvZ3Z2drVuo/5tHRYsWMD69OlT53axWMz09PTYv/71L66ssLCQCQQCFhoa+j5CJM1g1qxZrEOHDkwsFtN3uZUYMmQIGz9+vETZiBEjmKenJ2OMvsutQUlJCZOVlWXh4eES5fb29mzJkiXUx60AAHb48GHuvTR9mpKSwgCwuLg4rs6JEycYj8djDx48eG+xV6ERqUaaPn06hgwZAjc3N4nyhIQElJWVSZRbWVnByMgIV65ced9hkreUkZEBAwMDmJmZwdPTE9nZ2QCof1uLo0ePwsHBAaNGjYKOjg66du2K3bt3c9uzsrKQl5cn0c9qamro2bMn9fNH6vXr1/jPf/6D8ePHg8fj0Xe5lXByckJkZCRu3boFAEhKSsLFixcxaNAgAPRdbg3Ky8tRUVEBBQUFiXKhUIiLFy9SH7dC0vTplStXoK6uDgcHB66Om5sbZGRkEBMT895j5r/3I37E/vzzT1y7dk1ifm6VvLw8yMvLQ11dXaJcV1cXeXl57ylC8i569uyJwMBAWFpaIjc3FytXrkTfvn1x48YN6t9W4s6dO9i+fTvmzJmDxYsXIy4uDn5+fpCXl4ePjw/Xl7q6uhL7UT9/vMLCwlBYWAhfX18A9N/q1mLhwoV4/vw5rKysICsri4qKCqxevRqenp4AQN/lVkBFRQWOjo7w9/eHtbU1dHV1ERoaiitXrsDc3Jz6uBWSpk/z8vKgo6MjsZ3P50NTU7NF+p0SKSnl5ORg1qxZiIiIqPHXEdI6VP0lEwA6d+6Mnj17wtjYGAcOHIBQKGzByEhTEYvFcHBwwJo1awAAXbt2xY0bN7Bjxw74+Pi0cHSkOezZsweDBg2CgYFBS4dCmtCBAwcQHByMkJAQdOrUCSKRCLNnz4aBgQF9l1uRffv2Yfz48Wjfvj1kZWVhb2+PsWPHIiEhoaVDIwQALTYhtYSEBOTn58Pe3h58Ph98Ph/R0dHYunUr+Hw+dHV18fr1axQWFkrs9+jRI+jp6bVM0OSdqKuro2PHjsjMzISenh71byugr68PGxsbiTJra2tuCmdVX765ghv188fp3r17OHPmDCZOnMiV0Xe5dZg3bx4WLlyIr7/+Gra2tvDy8sL333+PtWvXAqDvcmvRoUMHREdHo7i4GDk5OYiNjUVZWRnMzMyoj1shafpUT08P+fn5EtvLy8vx5MmTFul3SqSk5OrqiuTkZIhEIu7l4OAAT09P7t9ycnKIjIzk9klPT0d2djYcHR1bMHLytoqLi3H79m3o6+ujW7du1L+tQO/evZGeni5RduvWLRgbGwMATE1NoaenJ9HPz58/R0xMDPXzRyggIAA6OjoYMmQIV0bf5dahpKQEMjKStzCysrIQi8UA6Lvc2igpKUFfXx9Pnz7FqVOn4OHhQX3cCknTp46OjigsLJQYlTx79izEYjF69uz53mOmVfveQfVV+xhjbMqUKczIyIidPXuWxcfHM0dHR+bo6NhyAZJG+eGHH1hUVBTLyspily5dYm5ubkxbW5vl5+czxqh/W4PY2FjG5/PZ6tWrWUZGBgsODmaKiorsP//5D1dn3bp1TF1dnR05coRdv36deXh4MFNTU/by5csWjJw0VkVFBTMyMmILFiyosY2+yx8/Hx8f1r59exYeHs6ysrLY33//zbS1tdn8+fO5OvRd/vidPHmSnThxgt25c4edPn2a2dnZsZ49e7LXr18zxqiPP0ZFRUUsMTGRJSYmMgBs06ZNLDExkd27d48xJl2fDhw4kHXt2pXFxMSwixcvMgsLCzZ27NgWOR9KpN7Bm4nUy5cv2bRp05iGhgZTVFRkw4cPZ7m5uS0XIGmUMWPGMH19fSYvL8/at2/PxowZwzIzM7nt1L+tw7Fjx9inn37KBAIBs7KyYrt27ZLYLhaL2dKlS5muri4TCATM1dWVpaent1C05G2dOnWKAai17+i7/PF7/vw5mzVrFjMyMmIKCgrMzMyMLVmyhJWWlnJ16Lv88du/fz8zMzNj8vLyTE9Pj02fPp0VFhZy26mPPz7nzp1jAGq8fHx8GGPS9enjx4/Z2LFjmbKyMlNVVWXffvstKyoqaoGzYYzHWLWfASeEEEIIIYQQ0iB6RooQQgghhBBCGokSKUIIIYQQQghpJEqkCCGEEEIIIaSRKJEihBBCCCGEkEaiRIoQQgghhBBCGokSKUIIIYQQQghpJEqkCCGEEEIIIaSRKJEihBDSZpmYmGDz5s1N2qaLiwtmz57dpG0SQgj58PBbOgBCCCGkPr6+vigsLERYWJhEeVRUFPr374+nT59CXV29RWKrzd9//w05ObmWDoMQQkgzo0SKEEIIaUKampotHQIhhJD3gKb2EUIIaRUuXryIvn37QigUwtDQEH5+fnjx4gW3PT8/H0OHDoVQKISpqSmCg4NrtJGdnQ0PDw8oKytDVVUVo0ePxqNHj7jtK1asQJcuXbBv3z6YmJhATU0NX3/9NYqKirg6b07tKy0txYIFC2BoaAiBQABzc3Ps2bOneS4CIYSQ94YSKUIIIR+927dvY+DAgRg5ciSuX7+O/fv34+LFi5gxYwZXx9fXFzk5OTh37hz++usvbNu2Dfn5+dx2sVgMDw8PPHnyBNHR0YiIiMCdO3cwZsyYGscKCwtDeHg4wsPDER0djXXr1tUZm7e3N0JDQ7F161akpqZi586dUFZWbvqLQAgh5L2iqX2EEEI+eOHh4TWSj4qKCu7fa9euhaenJzcSZGFhga1bt8LZ2Rnbt29HdnY2Tpw4gdjYWHTv3h0AsGfPHlhbW3NtREZGIjk5GVlZWTA0NAQABAUFoVOnToiLi+P2E4vFCAwMhIqKCgDAy8sLkZGRWL16dY24b926hQMHDiAiIgJubm4AADMzsya6KoQQQloSJVKEEEI+eP3798f27dslymJiYvDNN98AAJKSknD9+nWJ6XqMMYjFYmRlZeHWrVvg8/no1q0bt93KykpikYrU1FQYGhpySRQA2NjYQF1dHampqVwiZWJiwiVRAKCvry8xslWdSCSCrKwsnJ2d3/7kCSGEfJAokSKEEPLBU1JSgrm5uUTZ/fv3uX8XFxfju+++g5+fX419jYyMcOvWrSaL5c0V+Xg8HsRica11hUJhkx2XEELIh4WekSKEEPLRs7e3R0pKCszNzWu85OXlYWVlhfLyciQkJHD7pKeno7CwkHtvbW2NnJwc5OTkcGUpKSkoLCyEjY3NW8Vla2sLsViM6Ojotz43QgghHyZKpAghhHz0FixYgMuXL2PGjBkQiUTIyMjAkSNHuMUmLC0tMXDgQHz33XeIiYlBQkICJk6cKDFi5ObmBltbW3h6euLatWuIjY2Ft7c3nJ2d4eDg8FZxmZiYwMfHB+PHj0dYWBiysrIQFRWFAwcONMl5E0IIaTmUSBFCCPnode7cGdHR0bh16xb69u2Lrl27YtmyZTAwMODqBAQEwMDAAM7OzhgxYgQmT54MHR0dbjuPx8ORI0egoaGBfv36wc3NDWZmZti/f/87xbZ9+3Z89dVXmDZtGqysrDBp0iSJZdkJIYR8nHiMMdbSQRBCCCGEEELIx4RGpAghhBBCCCGkkSiRIoQQQgghhJBGokSKEEIIIYQQQhqJEilCCCGEEEIIaSRKpAghhBBCCCGkkSiRIoQQQgghhJBGokSKEEIIIYQQQhqJEilCCCGEEEIIaSRKpAghhBBCCCGkkSiRIoQQQgghhJBGokSKEEIIIYQQQhqJEilCCCGEEEIIaaT/Bx+M80hFA2dxAAAAAElFTkSuQmCC\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "data_table.DataTable(full_df, include_index = False, num_rows_per_page = 20)"
      ],
      "metadata": {
        "id": "O3pPwxOsMiA1"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Judgment Rating Approach"
      ],
      "metadata": {
        "id": "4iFWK2E0cQ0V"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# @title prompt_rate\n",
        "def prompt_rate(action_list, dimension, context=\"\"):\n",
        "  \"\"\"\n",
        "  Given a list of actions to sort, and a dimension to sort them on, craft a prompt\n",
        "  for GPT, with example calls and responses.\n",
        "\n",
        "  Optionally, include context regarding the hypothetical rater (Ziv),\n",
        "  which may influence the ranking results.\n",
        "\n",
        "  Returns a list of dicts to pass on to GPT as prompt messages\n",
        "  \"\"\"\n",
        "\n",
        "  #randomize the order of the list\n",
        "  random.shuffle(action_list)\n",
        "\n",
        "  #turn list into formatted string\n",
        "  action_string = (', ').join(action_list)\n",
        "\n",
        "  # Switch by sorting dimension\n",
        "  if dimension == 'moral':\n",
        "\n",
        "    system_prompt_content = \"\"\"You are an expert judge of the moral virtues of different actions.\n",
        "Moral actions are those which humans consider virtuous, that consider others' wellbeing and happiness, and that are guided by principles of ethics.\n",
        "You will be given a list of actions that an average human person, Ziv, is considering.\n",
        "You must rate them in terms of their moral virtue according to Ziv.\n",
        "You will recieve a JSON object of items to rate. This object will have the following key / value pairs:\n",
        "1. actions - Array<string>: a sorted list of actions that the user has been faced with\n",
        "2. context - string: information that you can use to help you make your choice\n",
        "You must respond with a JSON packet with a single key / value pair:\n",
        "1. rating - Array: A list of lists, each list contains 2 elements:\n",
        "1. action from the original list\n",
        "2. rating from 0 to 10 for this action, where 0 is the most morally wrong, and 10 is the most morally virtuous\"\"\"\n",
        "\n",
        "    system_prompt= {\n",
        "        \"role\": \"system\",\n",
        "        \"content\": system_prompt_content\n",
        "    }\n",
        "    user_example = {\n",
        "        \"role\": \"user\",\n",
        "        \"content\": json.dumps({\"actions\": [\"walk a dog\", \"rob a bank\", \"rescue a cat\", \"steal food\"],\n",
        "                               \"context\": \"\"}) # Initial example has no context\n",
        "    }\n",
        "    assistant_example = {\n",
        "        \"role\": \"assistant\",\n",
        "        \"content\": json.dumps({\"rating\": [[\"walk a dog\", 5], [\"rob a bank\", 1],[ \"rescue a cat\",8],[\"steal food\", 2]]})\n",
        "    }\n",
        "\n",
        "  elif dimension == 'hedonic':\n",
        "\n",
        "    system_prompt_content = \"\"\"You are an expert judge of the hedonic reward of different actions.\n",
        "Hedonically rewarding actions are those which humans consider pleasant, make them feel happy, and benefit their own wellbeing.\n",
        "You will be given a list of actions that an average human person, Ziv, is considering.\n",
        "You must rate them in terms of their hedonic reward according to Ziv.\n",
        "You will recieve a JSON object of items to rate. This object will have the following key / value pairs:\n",
        "1. actions - Array<string>: a sorted list of actions that the user has been faced with\n",
        "2. context - string: information that you can use to help you make your choice\n",
        "You must respond with a JSON packet with a single key / value pair:\n",
        "1. rating - Array: A list of lists, each list contains 2 elements:\n",
        "1. action from the original list\n",
        "2. rating from 0 to 10 for this action, where 0 is the least hedonically rewarding, and 10 is the most hedonically rewarding\"\"\"\n",
        "\n",
        "    system_prompt= {\n",
        "        \"role\": \"system\",\n",
        "        \"content\": system_prompt_content\n",
        "    }\n",
        "    user_example = {\n",
        "        \"role\": \"user\",\n",
        "        \"content\": json.dumps({\"actions\": [\"miss the bus\", \"win a sweepstakes\", \"eat a meal\", \"lose my wallet\"],\n",
        "                               \"context\": \"\"}) # Initial example has no context\n",
        "    }\n",
        "    assistant_example = {\n",
        "        \"role\": \"assistant\",\n",
        "        \"content\": json.dumps({\"rating\": [[\"miss the bus\", 3], [\"win a sweepstakes\", 9], [\"eat a meal\", 7], [\"lose my wallet\", 2]]})\n",
        "    }\n",
        "\n",
        "  elif dimension == 'movement':\n",
        "\n",
        "    system_prompt_content = \"\"\"You are an expert judge of the physical body movement of different actions.\n",
        "Physically active actions are those which humans consider to involve substantial movement of the body.\n",
        "You will be given a list of actions that an average human person, Ziv, is considering.\n",
        "You must rate them in terms of how much physical movement they involve.\n",
        "You will recieve a JSON object of items to rate. This object will have the following key / value pairs:\n",
        "1. actions - Array<string>: a sorted list of actions that the user has been faced with\n",
        "2. context - string: information that you can use to help you make your choice\n",
        "You must respond with a JSON packet with a single key / value pair:\n",
        "1. rating - Array: A list of lists, each list contains 2 elements:\n",
        "1. action from the original list\n",
        "2. rating from 0 to 10 for this action, where 0 indicates the least physical body movement, and 10 indicates the most physical body movement\"\"\"\n",
        "\n",
        "    system_prompt= {\n",
        "        \"role\": \"system\",\n",
        "        \"content\": system_prompt_content\n",
        "    }\n",
        "    user_example = {\n",
        "        \"role\": \"user\",\n",
        "        \"content\": json.dumps({\"actions\": [\"play soccer\", \"listen to music\", \"pack boxes\", \"order coffee at a cafe\"],\n",
        "                               \"context\": \"\"}) # Initial example has no context\n",
        "    }\n",
        "    assistant_example = {\n",
        "        \"role\": \"assistant\",\n",
        "        \"content\": json.dumps({\"rating\": [[\"play soccer\", 9], [\"listen to music\", 0], [\"pack boxes\", 6], [\"order coffee at a cafe\", 2]]})\n",
        "    }\n",
        "\n",
        "  else:\n",
        "    raise Exception(\"Specify a rating dimension: moral, hedonic, movement\")\n",
        "\n",
        "\n",
        "  # List actions to sort\n",
        "  user_prompt = {\n",
        "      \"role\": \"user\",\n",
        "      \"content\": json.dumps({\"actions\": action_list,\n",
        "                             \"context\": context}) # Optional context, defaults to empty string\n",
        "  }\n",
        "\n",
        "  return [system_prompt, user_example, assistant_example, user_prompt]"
      ],
      "metadata": {
        "id": "CTfRp_VTcxtn",
        "cellView": "form"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Create list of lists with all items\n",
        "resample_size = 20\n",
        "dimension = 'movement'\n",
        "\n",
        "result_lists = [[action, []] for action in action_list_all]\n",
        "\n",
        "for i in range(resample_size):\n",
        "\n",
        "  # Rate actions\n",
        "  resp = promptGPT(prompt_rate(action_list_all, dimension))\n",
        "  resp_list = json.loads(resp)['rating']\n",
        "  resp_df = pd.DataFrame(resp_list, columns=['item', 'rating'])\n",
        "  resp_df['rating'] = pd.to_numeric(resp_df['rating'])\n",
        "\n",
        "  for action in action_list_all:\n",
        "    act_val = resp_df[resp_df['item'] == action]['rating'].values[0]\n",
        "    cur_vals = []\n",
        "    for row in result_lists:\n",
        "      if row[0] == action:\n",
        "        row[1].append(act_val)\n",
        "        break"
      ],
      "metadata": {
        "id": "tXm-gs3nsH78"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# summary statistics for each item\n",
        "new_result_lists = []\n",
        "\n",
        "for row in result_lists:\n",
        "  if len(row[1]) > 0:\n",
        "    CI = st.t.interval(0.95, len(row[1])-1, loc=np.mean(row[1]), scale=st.sem(row[1]))\n",
        "    new_row = [\n",
        "        row[0],\n",
        "        max(0, CI[0]), # 0 or low CI\n",
        "        np.mean(row[1]),\n",
        "        min(10, CI[1]), # 10 or high CI\n",
        "        ]\n",
        "    new_result_lists.append(new_row)\n",
        "  else:\n",
        "    print('Need more data for ' + row[0])\n",
        "\n",
        "new_results_df = pd.DataFrame(new_result_lists, columns=['item', '95CI_Low', 'mean', '95CI_High'])"
      ],
      "metadata": {
        "id": "fyhUp8uesMpD"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "new_results_df.to_csv(data_path + 'movement_20_iter_results_likert_subset15.csv')"
      ],
      "metadata": {
        "id": "0WLXGwX8sQmC"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Load and merge dataframes\n",
        "\n",
        "# 15 subset\n",
        "moral_df_new = pd.read_csv(data_path + 'moral_20_iter_results_likert_subset15.csv')\n",
        "hedonic_df_new =  pd.read_csv(data_path + 'hedonic_20_iter_results_likert_subset15.csv')\n",
        "movement_df_new =  pd.read_csv(data_path + 'movement_20_iter_results_likert_subset15.csv')\n",
        "\n",
        "moral_df_new.columns = ['null', 'item', 'low_moral', 'mean_moral', 'high_moral']\n",
        "hedonic_df_new.columns = ['null', 'item', 'low_hedonic', 'mean_hedonic', 'high_hedonic']\n",
        "movement_df_new.columns = ['null', 'item', 'low_movement', 'mean_movement', 'high_movement']\n",
        "\n",
        "moral_df_new = moral_df_new.drop(columns=['null'])\n",
        "hedonic_df_new = hedonic_df_new.drop(columns=['null'])\n",
        "movement_df_new = movement_df_new.drop(columns=['null'])\n",
        "\n",
        "full_df = pd.merge(moral_df_new, hedonic_df_new, on='item', how='inner')\n",
        "full_df = pd.merge(full_df, movement_df_new, on='item', how='inner')"
      ],
      "metadata": {
        "id": "1ZxZFHNDx0vk"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print('Prompting Likert Correlations:')\n",
        "print('r(moral, hedonic) = '+ str(round(np.corrcoef(full_df[\"mean_moral\"], full_df[\"mean_hedonic\"])[0, 1], 2)))\n",
        "print('r(moral, movement) = '+ str(round(np.corrcoef(full_df[\"mean_moral\"], full_df[\"mean_movement\"])[0, 1], 2)))\n",
        "print('r(hedonic, movement) = '+ str(round(np.corrcoef(full_df[\"mean_hedonic\"], full_df[\"mean_movement\"])[0, 1], 2)))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "u2_Cr_5wxs5x",
        "outputId": "92fcecb7-d494-4ddc-8059-ef551f640ad8"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Prompting Likert Correlations:\n",
            "r(moral, hedonic) = 0.31\n",
            "r(moral, movement) = 0.55\n",
            "r(hedonic, movement) = -0.09\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Human Rating"
      ],
      "metadata": {
        "id": "tK3IrSsAMb__"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "rating_df = pd.read_csv(data_path + \"Neurips_Prolific_20_R.csv\")\n",
        "rating_df = rating_df[rating_df['item'].isin(action_list_all)] # filter to the 15 less-correlated items"
      ],
      "metadata": {
        "id": "Ls4HftxqMeuT"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# summary statistics for each item\n",
        "new_result_lists = []\n",
        "\n",
        "for row in rating_df.iterrows():\n",
        "  item = row[1][1]\n",
        "  vals = row[1][2:]\n",
        "  CI = st.t.interval(0.95, len(vals)-1, loc=np.mean(vals), scale=st.sem(vals))\n",
        "  new_row = [\n",
        "      item,\n",
        "      max(0, CI[0]), # 0 or low CI\n",
        "      np.mean(vals),\n",
        "      CI[1]\n",
        "      ]\n",
        "  new_result_lists.append(new_row)\n",
        "\n",
        "new_results_df = pd.DataFrame(new_result_lists, columns=['item', '95CI_Low', 'mean', '95CI_High'])"
      ],
      "metadata": {
        "id": "VOKpaJZ_Osi9"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# 15 subset\n",
        "moral_df_new = new_results_df[:15]\n",
        "hedonic_df_new =  new_results_df[15:30]\n",
        "movement_df_new =  new_results_df[30:]\n",
        "\n",
        "moral_df_new.columns = ['item', 'low_moral', 'mean_moral', 'high_moral']\n",
        "hedonic_df_new.columns = ['item', 'low_hedonic', 'mean_hedonic', 'high_hedonic']\n",
        "movement_df_new.columns = ['item', 'low_movement', 'mean_movement', 'high_movement']\n",
        "\n",
        "full_df = pd.merge(moral_df_new, hedonic_df_new, on='item', how='inner')\n",
        "full_df = pd.merge(full_df, movement_df_new, on='item', how='inner')"
      ],
      "metadata": {
        "id": "VTcsEZzJPBiF"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "fig, ax = plt.subplots(figsize=(10, 10))\n",
        "\n",
        "full_df = full_df.sort_values('mean_moral', ascending=False)\n",
        "\n",
        "plot_i = 1\n",
        "for i in range(len(full_df)):\n",
        "\n",
        "    # X axis, means with 95% CI low and high values\n",
        "    x = full_df['mean_hedonic'].iloc[i]\n",
        "    x_l = full_df['mean_hedonic'].iloc[i] - full_df['low_hedonic'].iloc[i]\n",
        "    x_h = full_df['high_hedonic'].iloc[i] - full_df['mean_hedonic'].iloc[i]\n",
        "\n",
        "    # Y axis, means with 95% CI low and high values\n",
        "    y = full_df['mean_moral'].iloc[i]\n",
        "    y_l = full_df['mean_moral'].iloc[i] - full_df['low_moral'].iloc[i]\n",
        "    y_h = full_df['high_moral'].iloc[i] - full_df['mean_moral'].iloc[i]\n",
        "\n",
        "    l = str(plot_i) + '. ' + full_df['item'].iloc[i].lstrip('[').rstrip(']')\n",
        "    ax.errorbar(x, y, xerr=[[x_l], [x_h]], yerr=[[y_l], [y_h]], label=l, fmt='o', ecolor='lightgray', elinewidth=2)\n",
        "    ax.annotate(plot_i, (x+.3, y+.5))\n",
        "\n",
        "    plot_i += 1\n",
        "\n",
        "ax.set_xlabel('Hedonic')\n",
        "ax.set_ylabel('Moral')\n",
        "ax.legend()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 867
        },
        "id": "86IuQyHDQylI",
        "outputId": "2b433146-978b-4e49-a5e9-0326c5457edd"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7b9d42a30cd0>"
            ]
          },
          "metadata": {},
          "execution_count": 16
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x1000 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Archive"
      ],
      "metadata": {
        "id": "9Wr9uCCSzrHn"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Create list of lists with all items\n",
        "\n",
        "rank_list_size = 15\n",
        "resample_size = 100\n",
        "dimension = 'hedonic'\n",
        "result_lists = [[action, []] for action in action_list_all]\n",
        "\n",
        "for i in range(resample_size):\n",
        "  action_list = random.sample(action_list_all, rank_list_size)\n",
        "  result_df = sort_rate(action_list, dimension)\n",
        "\n",
        "  for action in action_list:\n",
        "    act_val = result_df[result_df['item'] == action]['rescored_'+ dimension].values[0]\n",
        "    cur_vals = []\n",
        "    for row in result_lists:\n",
        "      if row[0] == action:\n",
        "        row[1].append(act_val)\n",
        "        break"
      ],
      "metadata": {
        "id": "dwSRbtrjw42k"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}