{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "listwise-movie-recommendations-using-rl-methods.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "mount_file_id": "1Guok-lDpcc5mJhz6W4eaFNERAvU0FgRI",
      "authorship_tag": "ABX9TyOTdHQQ3hPyWAyakOXAfBNw"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WlyEOQzwccnK"
      },
      "source": [
        "# Movie List Recommender using Actor-critic based RL method\n",
        "> Training a list-wise movie recommender using actor-critic policy and evaluating offline using experience replay method\n",
        "\n",
        "- toc: true\n",
        "- badges: true\n",
        "- comments: true\n",
        "- categories: [RL, Movie, Tensorflow 1x]\n",
        "- image:"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cvYLG7qBvB6W"
      },
      "source": [
        "### Introduction"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "URFAhP2vvD4T"
      },
      "source": [
        "We will model the sequential interactions between users and a recommender system as a Markov Decision Process (MDP) and leverage Reinforcement Learning (RL) to automatically learn the optimal strategies via recommending trial-and-error items and receiving reinforcements of these items from users’ feedbacks.\n",
        "\n",
        "Efforts have been made on utilizing reinforcement learning for recommender systems, such as POMDP and Q-learning. However, these methods may become inflexible with the increasing number of items for recommendations. This prevents them to be adopted by practical recommender systems."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TkzYH9X2vlOp"
      },
      "source": [
        "![image.png](data:image/png;base64,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)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qepaUtlovTsw"
      },
      "source": [
        "\n",
        "Generally, there exist two Deep Q-learning architectures, shown in the above figure. Traditional deep Q-learning adopts the first architecture as shown in (a), which inputs only the state space and outputs Q-values of all actions. This architecture is suitable for the scenario with high state space and small action space, like playing Atari. However, one drawback is that it cannot handle large and dynamic action space scenario, like recommender systems. The second Q-learning architecture, shown (b), treats the state and the action as the input of Neural Networks and outputs the Q-value corresponding to this action. This architecture does not need to store each Q-value in memory and thus can deal with large action space or even continuous action space. A challenging problem of leveraging the second architecture is temporal complexity, i.e., this architecture computes Q-value for all potential actions, separately.\n",
        "\n",
        "Recommending a list of items is more desirable (especially for sellers) than recommending a single item. To achieve this, one option is to score the items seperately and select the top ones. For example, DQN can calculate Q-values of all recalled items separately, and recommend a list of items with highest Q-values. But this strategy do not consider the relation between items. e.g. If the next best item is egg, all kind of different eggs will get high scores, like white eggs, brown eggs, farm eggs etc. But these are similar items, not complimentary. And the whole purpose of list-wise recommendation is to recommend complimentary items. That's where DQN fails. \n",
        "\n",
        "One option to resolve this issue is by adding a simple rule - select only 1 top-scoring item from each category. This is a good rule and will improve the list quality but we have to compromise with some missed opportunities here because let's say we recommend a 12-brown-eggs 🥚 tray and a small brown bread 🍞. Now it is possible that if a 24-brown-eggs tray is scored higher than 12-brown-eggs tray but in bread category, small-brown-bread is still the highest score item. As per business sense, we should recommend a large brown bread with 24-brown-eggs tray. This is what we missed - either customer will manually select the large bread (lost opportunity for higher customer satisfaction) or just buy the small one (lost opportunity for higher revenue). \n",
        "\n",
        "In this tutorial, our goal is to fill this gap. We will evaluate the RL agent offline using experience replay method. Also, it is possible that productionizing this model cost more than the benefit, especially for small businesses, because if we are getting 1% revenue gain, on $1M, it might not be sufficient, and on $1B, the same model will become one of the highest priority model to productionize 💵."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xO7879psvqw6"
      },
      "source": [
        "![image.png](data:image/png;base64,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)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GnQebSD9vYXL"
      },
      "source": [
        "To tackle this problem, in this paper, our recommending policy builds upon the Actor-Critic framework. We model this problem as a Markov Decision Process (MDP), which includes a sequence of states, actions and rewards. More formally, MDP consists of a tuple of five elements $(\\mathcal{S}, \\mathcal{A}, \\mathcal{P}, \\mathcal{R}, \\gamma)$ as follows:\n",
        "\n",
        "- State space $\\mathcal{S}$: A state $s_t \\in S$ is defined as the browsing history of a user, i.e., previous $N$ items that a user browsed before time $t$. The items in $s_t$ are sorted in chronological order.\n",
        "- Action space $\\mathcal{A}$: An action $a_t \\in \\mathcal{A}$ is to recommend a list of items to a user at time $t$ based on current state $s_t$.\n",
        "- Reward $\\mathcal{R}$: After the recommender agent takes an action $a_t$ at the state $s_t$ , i.e., recommending a list of items to a user, the user browses these items and provides her feedback. She can skip (not click), click, or order these items, and the agent receives immediate reward $r(s_t,a_t)$ according to the user’s feedback.\n",
        "- Transition probability $\\mathcal{P}$: Transition probability defines the probability of state transition from $s_t$ to $s_{t+1}$ when RA takes action $a_t$. If user skips all the recommended items, then the next state $s_{t+1}$ = $s_t$; while if the user clicks/orders part of items, then the next state $s_{t+1}$ updates.\n",
        "- Discount factor $\\gamma$ : $\\gamma \\in [0,1]$ defines the discount factor when we measure the present value of future reward. In particular, when $\\gamma$ = 0, RA only considers the immediate reward. In other words, when $\\gamma$ = 1, all future rewards can be counted fully into that of the current action.\n",
        "\n",
        "With the notations and definitions above, the problem of listwise item recommendation can be formally de!ned as follows: Given the historical MDP, i.e., $(\\mathcal{S}, \\mathcal{A}, \\mathcal{P}, \\mathcal{R}, \\gamma)$, the goal is to find a recommendation policy $\\pi : \\mathcal{S} \\to A$, which can maximize the cumulative reward for the recommender system.\n",
        "\n",
        "According to collaborative filtering techniques, users with similar interests will make similar decisions on the same item. With this intuition, we match the current state and action to existing historical state-action pairs, and stochastically generate a simulated reward. To be more specific, we first build a memory $M = {m_1,m_2, ···}$ to store users’ historical browsing history, where $m_i$ is a user-agent interaction triplet $((s_i, a_i) \\to r_i)$. The procedure to build the online simulator memory is illustrated in the following figure. Given a historical recommendation session ${a_1, ··· , a_L}$, we can observe the initial state $s_0 = {s_0^1, ··· ,s_0^N}$ from the previous sessions (line 2). Each time we observe $K$ items in temporal order (line 3), which means that each iteration we will move forward a window of K. We can observe the current state (line 4), current $K$ items (line 5), and the user’s feedbacks for these items (line 6). Then we store triplet $((s_i, a_i) \\to r_i)$ in memory (line 7). Finally we update the state (lines 8-13), and move to the next $K$ items. Since we keep a fixed length state $s = {s_1, ··· ,s_N }$, each time a user clicked/ordered some items in the recommended list, we add these items to the end of state and remove the same number of items in the top of the state. For example, the RA recommends a list of !ve items ${a_1, ··· , a_5}$ to a user, if the user clicks $a_1$ and orders $a_5$, then update $s = {s_3, ··· ,s_N , a_1, a_5}$."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FCYiZWs4vvPT"
      },
      "source": [
        "![image.png](data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAggAAAICCAYAAACqZzUFAAAgAElEQVR4Aey9B1QUV///v7O7A4igRGMllseCiRqjEntFY68YxF6isaFoNBpBDWrUxNhjjL0n0RBR7HossUWi6LEi6okKcizoX5Gj8IX97c55/c9snaVjyYNPhnN0Z2dn7tz7uu19P/dz72hQ/1QCKgGVgEpAJaASUAlkIKDJ8F39qhJQCagEVAIqAZWASgBVIKiFQCWgElAJqARUAiqBTARUgZAJyf/YCdMD/vx5Bb9EPcCUa9KMPIs7z8EtOzmfnOvF6gVKAqYUHlw7yfatfxCXO2jlnerxv5iA8Vkc5w9uYUd0EtKrcDA+I+78QbbsPM+br7omUh5c4+T2rfzxNhX2f5RRxsw08OjqYbauWMSi5eGcTkjLeEGB/O4QCOm3OBERwfYdkezavYc9e/awe9dOIndEELHnPImvVHoLZNqzjpSUxNWDEURYOVgYbCdirzODF9ePsj1iBzt37WbP7l1Ebt/O4Zhnr1bJs47RK51N3/853joNQpE2/BSXfSYaLy7B/4PiiIIGjUsLFsVnf61ThAwPuXjiAg+MTmcL1pc3GkeJxIjR+JYthKDRoCs3gkPpBSv5eYrNG2WUpxjk4yKJxPN7idge6ah/OyLY9vvvbN99mKjrjyjwWWC8zNJPa/CuKKDRiDT+/u88CPisEBm5uMSfD4qL5vLn0mIRea26WYWW2zkpMYLRvmUpJLcTunKMeCsK+z/LKBNDYxyRE/yo7/81y5dPpb23Dm2R+sw8W+BLqWKK4flZNkwbTtvKloZOo9GgEXSUrNePL+ft4fbbNiqSjBhfJs7SQ06unsbgBiXQygy0XtQKmMDMnw4Rbw9P4ulf6wkbG0CdYlo02neo3fNLlp9ILHACQUo6w6qxXalZvAQDInNRrYYzTK6uR+PyCT89zItASOfo6IrohUI0m3/zJRu4TNXpNZ/4Z+IoPVxLJ08BfeWxHDe85iS88eD+GUavLxkmEk7+zE9zBlDLXUBwrUS7oKnMnD2T0BFdqfGOGyU+Hs6vfxf0jDByYdpHiBoXWi5JeKW2w3BmMtX1Glw++Yk8Vd1XyQzpIWs7eSLoKzP2LSrs/ygjO18jMfObUbRkH35/Jp+USNw+hMpePoze99x+VUE9cFgQrDF8sWswZbQazAJBbMaCO/ZesaCmwSleaTd28u2ITtQuW4lh+19eodkKk+694RzMtl81cmV6XdxqhHC2QLdFT/klsDGT/8olktJ9fmzlgsa1PaueOGHN5ouJ+PAgmtVpw/Tj5tKfzXVZnDbc4eCi0XT7bCW33mgRe4U4ZhHtbE+lH2JEOR16n/GcygVztmH81374hxi97vQl/4y/p4BLyyXcVejZZ3+M5X1RS/He4f+Auf3VEvV4ZTtcNS60+vF+7gIhhzoj3f+RVi4aXNuvIk9V95Winc6hEeXQ6X0YX9AKe4FhZAVsOM7YynoKd92IsoU0md5oo/dKuau8OZNAMF4Mo7ZoEQhC0UDCU5WXF/TjdI6MqoBOHvm/oro13VlAM1GDWGsaF3Iwnz9Z1R7PFovfqFnvlainJXD0u09pPWpH7iMLKZHlbWSB0IE1Sa/01Fxvlu4txc9Fg9hsITnMfOQaToG5IP0Ioyq8rQKhwFDMV0QMpydSTa+nyrgTOGmytN0MLqXFpSDXS2tKk9Z0sAqEB7kKhJzqjJS4nDayQOiwhjdcdQFrO1sABULBYWTJYNONb2kg6vAefrDgT3tlUfsyC4Qr06lrEwhefdiW7eg5i9D+66dSCQ8sYp6L0+hrEHI2h549l7hK8YtoLndgtcO4mEMwcgX39PuBBMUIJpeg/8GfJR4fXcGC8Is8zUv87AKhE+uUcvdNxPjxStq6anBp9SP38xK3NxGH1xmmKhBeJ808hCXxcHkb3AQPum507hKlBz/Rxk1P5bHHC3yjnB+BQA51xi4QOq1zGqnmAeRLXFKABILhEdfPXyEhxZqMAsPIEh/DqfH46HVUCj5W4MtiVgUhXwLhRcJFju/azNIZ4+jfoTk9Fl3GiMSzK+HM/KwdDep+TKs+X7Pjb4VpX0om7vwf7Fi/iK+DetOm+SDWxxlI/HM5ozs3on7zjvQaNYvfr9h8b41cXh1E37796D9gAAP696N/yO/m+X/T3W2E9OtD3379GTBgAP36jWSltfeWHkexOmQoLStanHU02uLU6Wa5btC49VzNoZPPCsyrCYQXXN85n3ED/encJYABwyYwY+lebiiwyM+Unl4gfEEIw/sG0m/YZBbvisU2KyUl3yZq3xaWTR9Jz3FbuG96ztUt0xjo/ymj11zEVh94cZ2d88cx0L8zXQIGMGzCDJbuvfFyhdEuEDqzPhlMj6PZOj+UkX398e8XzLz9cTgwviA++hDb1s5nyucBTAjP4H+RQ7xenP+ZbyZ1papeg75yZyZMm8a0aWEs2hun8GMwkHByLWEjetOtYyf8+49h1qYoJ2fI3BnlEkfA9PA0G2ePZ7B/Rzr1GMSE+ZHE2IqitWAYE8/y86xgBvboTNfAoUxdF8VDBwjLVfkVCDnwcZRHiacXwlkQMpy+gf0YNnkxu2JtJcRx1YvrO5k/biD+nbsQMGAYE2YsZa9TYcupPObOyJBwkrVhI+jdrSOd/PszZtYmopReqaYnXD+xi82LpzCkZxgH5cL5IpbdS79mzAB/uvcexZydf2cuk3li4Ein81E6B4Z5oxPrEHZJmRnP+Wvax3iW68UvynkH55tz/JZbfpueXOfErs0snjKEnmEHzXXxRexuln49hgH+3ek9ag47lW2g/WkSSRd+5ZugvnTv1IXAzycR1u9Dsw9Cqx9ztiDkVmfsAqHzepIx8Th6K/NDR9LX359+wfPYH6dkZI2Q9JQL4QsIGd6XwH7DmLx4F1kUL3vsHQc5CYScy+xLszMlErV+OkED/OncuSdDQ5ezY8sMetT/D146HeVHHeZJLu3KSzFyJNrpKNc6IT3l9NqZTBnckHe0Wt5pOJgpcjsXtppTeRqtOT3uv/YlHwLBQNSCT/nY280yQtcIeHT8hg0TP8HnvYqUL+ZqP+/yYSh/2TrDF0eY2ekjSpq9dWWnvxI069GOGj71adm4Cl462YtXg7aIL5P/sIwE0h5eYseEehSWPWU1cididfwyPuH6wbl0LKOz+EgIhenxq2UOxBS3h/lTJuH/vlUg6MrScthkQkJCmbZov8XBMOUYk+sUpXDlwfyeiyfPywuEVKLC6uFVuh1zT9zl6f1o1gyshpubpdO15LTEk5Nz6exTg4AFB7gad5OTywOpLHpSd/Ix87ypMXYrIb18KaET0PsEs3ZBF+pUt/DSV5vIadmmmhpFWD0vSreby4m7T7kfvYaB1dxwMzcSL1GmFAJh8d7pfFK9Fm36BjMpqB2VXAUEz1YstfmkmOLYNbM/DcvoETSutF352PHAXOKVfCmSVYsG8aE8hVO9P/NXrGDFyjXsuPDUYmaVEjkU0gjv6v1YdvQat6+dYtu8vlT3ECnlN5vTVutGroz+L4c4IvHwUBit/+NDjwWHuZ74mNgtg/ERtRRt8h22PifpeBiNy9Rm1LYrPEx+wrXwYVR3daHqyH3Optz8CIRc+JhBSk84ObczPjUCWHDgKnE3T7I8sDKiZ10mH3MomNSoMOp5labd3BPcfXqf6DUDqebmRmdZ4Zn/cimPOeWj7FB1KIRG3tXpt+wo125f49S2efSt7oFYyo/ZtoxIi2Z1UBeqF9WiERswc996BvjWoHnPUUwM7kK1wgKCexO+j1XMu+aFgTUFWX6YbjK3kYjWy5+1fydw904s54+FM29oU2o0HsUvsS9n+sxLfqdFryaoS3WKajWIDWayb/0AfGs0p+eoiQR3qUZhQcC9yfcok4uUxJ/zulC1Sjumb7/AnbgYjq8fSwPZydnsg5CzQMitzjg6v8Xsnf4J1Wu1oW/wJILaVcJVEPBstRRb1ZV5Sk9OMrezDzUCFnDgahw3Ty4nsLKIZ93JKIpXluiznWLIQ5l9KXbPT/NN85J4d1zEmcepPI87xDetS6Fzq8fMSy+4dXALB/9O5XUzyjrxeawT0mPOhq9i2aRPKKnVUvKTSSxbsYKV6w9w4y2ats+HQLDgSjswjPd0lo5b61WHwT+c4L7cWb2I4qta1s5ZX42J5h7MhjiZ3wLfsawKEFyoM/6odST4nDPTPrYsmdEIiB+G2p39TLfm0cQ61SF3iFH2ScZUtgZ4WsSIUJReTk4SqWzubhUwYl1mXHFWzVLiRrp5aRHcGvLtNUVjZYum4vOlBUL6AYZ56yjUdaPDQSppCz0r9WGbtWBID7bQq6wrtULOYC8r0gOWtymE4N6KpbZ1SsZoQmuKCKInlXqu47Yxhet7NrL9oqUjTT8wDG9dIbputHUGkLSlJ5X6bHOEq0hTroc2gaD1oErXGRxMsPFL4+QXPugFD3r8arddyONvYmfXR9S40mWDY2Sbp3g9XU17eYqhdUava4kHW3pRVqxE0BHls4zcWOSHpyDiE3yUF7bE5MIouzia4tfRtYRIheEHHGEZ/mLSB3qEwp1YK3t6Sff56ZNCCJ4BbLVFRUpkTcfCaD06sfaRYm4kHwIhdz5WBq61CDljLyFID5bTppCAe6ulVp8X6yi6UFccRSCJLT0r0cdW2PJQHrNjZCmnIpWCjjgsVoDxxiL8PAVEn2CO2jMijb1DSqMV3CjVYBibrtp+MFg89QU32q92uM/lzsCWwdl8pvxO73e06Er70tnfn+6dP6HRByVx04oUr/Up3+xPUFi7sgkj4+n85HfaXoaUltuSUjQYtglHci8w7SMRwa09juRKPNrWn/dEbwZGKqdDTNz4tkGeLAjmqGZbZ8AmELQeVeg64yCOqnuSL3z0CB49sFdd6QFbepXFtVYIjuIl8WB5GwoJ7rRaGp+RTIbvWVkQ8lpmgXyxMxE7txFuOm+GHbCNOsF4YRq1RC3v9g53nlJ5XYwypNj2NX91AgzHg6mkk6cYjjv7ydgCLOCf+RYIxuhQalo7bpc2yxX7I6Sxb2gZqwjwcjRQZgBygSpvcR5UjPrln6TEdXQparEiaPTvM8nmaf9klbkDMVsQ3p+E7TSksXtwKftzev/uaEDhOes7u1qsC7Lp8bKtg7PlgsTTmKMcPP8g18YjfwJhKfdsfYXhNF/66NGWaMfCC7aO28jTh4+xjGmMXJpeF1FXgT7L9nNg/z7LPgo7tvBVc08EbQkG7bKOfkyxzKkvoskyLWA4/SU+ei0l2i3E8ainPHz8cqMnbAJBbJpp9Ury+s64Ci74LVV6W0sk/OCHi8aVbpsd+ZCneGVXkU03mNvYFW2JAUQ6grRkYNKvBLyjRVusF7/ZfCRyYSQvK8ocRyNnQ2qg11Ui2GmZlsSjC3vZc/a+tTKnE7NlBmGr/uSxLX8xcDy4Ejq5rDpUK+RDIOTKx3iJ6XVFdBX6sGz/Afbvk/ckiWTHlq9o7imgLTEISxExcPpLH/TaErRbeMEuSI1PH2IvArmWR3MtzIKRiRtzG+OqlZfHZsoIfg14B622GL3sGWHloi3NZ3ucy19qeC+KCiINv7thn0LKlYGtymbzaTw3hZqy89ewA4qpCwMJB6fSrJgWwbUGE47bREo2gWQ6nY/8NhwnuJIObenPcE5uKuG9iiKIDfnuhnUQYrrGnAYu6MoOZZ8zGvLlg5BdnTG3oxYnRbHpAidLASSb20TBxY+lVmcf4yXZz0xHhT7L2H9gP/v27GZX5A62fNUcT0FLiUG7MpFxPpGFQMhzmUXuNfPOjlR+6+mJIH7IlHOK9jxlKwGySK07Hadm/jUxck6v7Vt+68S/USCYK6bFguAsEOSlL+9ZRYAnPX9TNirWAmXeW8ExLWDGbrxsbgzNyyqFwvj/Yh2qWTPaLBA+cBYIez7LTiAksaaDTSDUJiwn70JbnmfzaRcItaZyXlEunS+XuLfUD08nz2EjMYv8KKbVILiW4ePAUNaftnU4ck1O4Ac/F/Ma4vajxzP+iy8YN3YswWPGMDooiFFBofwSY32gtfMTPHvihNMWCWMMi/yKodUIuJb5mMDQ9Zw2m3NsF+Tz0yYQXNuxSjFjIIeS+os/7oJIk3m37I181p2vPMTMQ7yyq8jPNtKtsID+/UkKq5E1HVazskZfjS9tFqrcGGUlEKQ4FjZ3MW8ItdhmrckFleHRZXavmMGY3h3wLVcIQVeR0UcdI5r8CITc+EgJP+DnIu+p0J7R48fzxRfjGDs2mDFjRhMUNIqg0F+wFRFjzCL8ZDO14EqZjwMJXX/aYtGzpyeX8mi+LisR9YyN3QojZBRC5utN3JzbCFGjp9qXp61PsgqETKIL0iIH8K5WpE7YJYcwz0sZsach88GTtZ1wF9xou+JhBu9/I1e/8UUUtLzTe5uT5SNzKNmfyTW/rZ2crlJwhn0v0ogc8C5ahW+ElLiSdm4Cou/MTL5Qr1sguLZbhXPVTeUXf3cEsQnzzOuJrXkt6KncfjTjx3/BF+PGMjZ4DGNGBxE0KojQX2KyB2P+JbNAyE+ZtQmEvLADAyfGVUGvK8tQpbpK3mRuJzw6rVEMUoHs2hWFiMqdUXbJz2+dUAWCInNeUiDYGn3zpkTF6LfdKrFfSiA8ZmU7m0D4iGk5rU/MrgzYzietpaNrbhvfGIn5ph7vDdlrtQ7Ybn5BbEQYPeuWxEXQIOiK0yjkEGaLtPE802rJVgFbhbXdk8WnrfPLabnpi1giwnpSt6QLgkZAV7wRIYceZWg0swg7q1OvSyDIYecWr2wqsnR3MS1cLH4nJ+zTSrbIPmdDZ1cEXXlGHbF2zrkyyqLzM17Icx5Iz86x6vPGVKnRkfE/RHD6zhNOTayG/lUEQi58jOdl86kGscm8PO0R8SI2grCedSnpIpg3NyveKIRDyukPciiPZrRZMJLusriFi2W5cOaM4PkG2aIkO4cdsWZOPgVCLgysgWbzkc4fo/9jseI4TIv2a9O29+UdrUBh/5/tTr/2H3M5yHN+50MgmK7N4mM5Pz+eRcaZzX9eIBg5P60Woiaj2M8FjNPPmQVCvspsPtjJjzVe/5G274qUH7jd2t8YubW8A8ULVWP0AeWUzRsWCPmuE6pAeHWBYDjFeB+9eVpAcGnKfNuWjU9X08HV6qSYYYohWwuC9IgV8jp+WWxkNEk5FfA8fEk/zMhyOgT39qzKdp/p54T39qbVD3cdHXLaYxKf2fwbUvh7zwzaeesR9JUY80camG4zv6mIRluMgF8dc7JZxiiXzi/tcSKOR/3Nnhnt8NYL6CuNQX5Uvv9ek0DIU7yyEQi8CKeX7Cfi0YUNTzOmwGohEhsx96bNfGuZhsl+z46sOr8ElshWHNkcvltp6crwPFMsP3xSDH2pHmx2zCER9YoCITc+ptvzaSpq0BYLIOciksbjxGd2i07K33uY0c4bvaCn0pg/LKI1t/JoTnIWjHhBeC8vtIIHXTJnhNU0LtJo7k0rtPwJhNwYZMgJ569SHIuauyAU/ZQtDtcX6zUm7ixohovgQv05sXY2zgFk8y0/+Z2PTs6yNl9A917mbbj/eYFg4vb8pogaLcUCfn3JTZUyC4S8l1nHFEPeLAhAyp+ENqhFs9YfU6dRG7p0bUvLdsNYcuyewyJly9Ls2pXXYkHIb534XxQICh8DJ8cWawYY/vqKD/TWKQanNexWJyXzNEIhum9Wzv9lP8VguvEdjeSRj0bLO53XOPYTSP2NnkWsKxxKD1HM86Wzb2hZqw+Ce4bnPGZlW6sFQV+FcRlHPqZbbB3XhQ6fLedCDv2CJakv2DPEG53gSeult7NsaKS7a+lSsS3LFO7B6QdGUG/4foVFQeLR6g64Cy60XvZArh2c/rIaenkVSLPviVFYqc3PlSSH2DBeZaaviFD4U7Zkim86B0bUY/h+hRKQHrG6gzuCS2uWPbBPmtuqTu6f0j2W+smm909YnkEUvdjUDTd5HvlbxzyyPMUQt7C52Qeh83pbS53HeNkqcoZd8JAe8XOP4ub5bf9NGZZOpp9gXBU9bvVnO0y1OTKSk5xVHI1cDKtjNkOXCvglwwZSEk+uXSY+HYwXv+YjUcC922bFSNTAnxN80OveY/hBRealH2S4t86x4iZb2nngI/sNVNOjETxo9n2MYo7dEqgkWfM2/QAj6g3HuQispoO7gEvrZchFIPfymB0jiUc/96C4Vksx/02KgYB8fbrF7OtWn9n29cPpHAmqgE5XkSDl1IvskxYhj+hFak07b23Q88AgW37yfNdOBpbQItab7bxSQL7HGMOchm7oSgfws2KDDSntBSnZThVaHpa//D5CUAUduopBOCc3jYi+76AVazHNNjdp28JcV5ERh2zervIzJeIXtcAlr6P57OqMHJJ14zGXT5R+YfIzXrCpm5vZJ+Jbq0+E7P9RTa9B8GjG95kbIGzFy0Ilq//TOTjcG51yM7q8llk5uPR8sAOebPKnRMs8bkb3mhhllWo5v/JXJyD98CjK63RUHH00Uz3O+hkF62wGJ0WJhNUd8bJttayvzld/KTog4Fl4L961/i7vMmirA0jx/OBnWwIp8mFotMJrUykQXGny7TVLB5p+g7X+3ug0Ai4VAtlosx7IjEx/s6CZNTxtUWr1CmHOdzOZOMgPHy/raglZBTf5gg2n4q3w03BYF3R4t/mKJT99y9cr/jR7qpsrkatAXjdRMt1aS/eyerRFfRn923WHt7vcRt3ZTUjzqrT49pxCDIB5Y4ySXVhjn9s2EPtdI1xdahJ6xmIzN91ZS9dSOgTBhcp9VnHukeV8ys0IJnXqwULb6ouUvQz11qER6zPHac2UDMjAqfE+lOyyxrGLoyGW7xq54lIzFOuj8lfaDFFMfF+PRqydYW25ietzZG9rHRVGHVak19pRaER8Z1yxNv55jFeKRQDqyn1G5IPH3Di8kkURlhfWGGKW0Lq4FrHyILYl2Kwx6dxc1p7i7jUZf0RhVsyRkZz8rOIoL1D4jb7l9Aja4jQJ2UOcLMDSHnJu/ec06zSfS+lguj6HBqIGfbVgjpr9TU08PPE9XSu6IAiedN/0lLTkZHPZkxLX0NFDQOvVg18yWT6U2ZAXPiburO1KKZ2A4FKZPqvOYSkiKdyMmESnHgsxFxGz9a0kXdbE20WlIfY7Grm6UDP0jLn+5aU8ZscIQwxLWhdHK1Zm0LYEu0hOv7mM9sXdqTn+iGKp52M2dCuCIBTFf7MSgMU64SpoKT1op7UO5YWBkpnzsfH8VD4UBYoGhjv5GBgeRrFyUE2KvNuIyQcd4tIUv4auJfS4/Gcg4QrR4Bxq/vKbxxvoVkRAKOqPc3JlHyNXs3Vq0E7bIMnEzaWf4KUV8Kgzlj33jCA949LGIFpWKIJOEHCt6s/Mjce5k2laTRHLnOpM1ETe11s2drMt0TXfabrOnAYiGl0FRh22tuWmO6ztWsr8XJfKfVh17pGlrU65ScSkTvRYeEXx0CwOzSt5PBC0XvSwF/Y8llk5uHyyk1dKuZbvy89/vzCXc8OLpzx+/JQXhiwGQa+LURbJNp/KV52QeLi6Ax6CgJf/5pe01mQXkX/mvF0gSI938EXz9y3zmLIVwPxPQFe0Ih/7LyD6/26x4bOPKeehte53IL/MyZ1yrb/jTMpFFnWqiIdNWGg0CPp3qNrsC3aY3b+VAkGLm2cZfBo0pU45D/SFvWnYfy6H7OtyHAlPPvUNLUtZxICgdaOMbwBTNp7mz1kNEQU9HmWr06hdAEPnHbF7madGL6RLpcLmOMr3vFujM9P23LU0btJT/lo9ha8WHnKsOnA8LsujlOs7mNm3HqVdRYp4f0CDln409X2fKjXbMGbdeZIylFHj5Zk08HShSNU2DA4ew6AOtalYuSnD1l52asySo5czuF4pXGUfBbEIpcqWoWL9gSw5/RgJE7c3DaNpVS/05r0gBNxK16D50HXYLOtg5PLMBni6FKFqm8EEjxlEh9oVqdx0GGsvK0cpWSYr08n0YzNpW9PiM6HRCLiXa0BwRCKS9IiI4I/xLmSx5ghiMd5vN5tTibuZ1MKH4tb9LWSHzFrtZ3IsPY/xkh7wa29v89sjBdGLmr2XclYB81n0SoY28qZIydp0GjCE/h3rUqVWd2bYl6/lgVFSdnG0JD/l8nqGN/bGTRDQFvLCy6MIlTvP5KhtjwzpPjvH1qOEqMWjvC+NP/6Ixv0Xs2dVL0rrBHSFS1C5+TQ2LOlHg4qeaOW8EnQUrdyYIeuUlhYl7jzyIZno5YOpV0reX0RALFKKsmUqUn/gEk7bllQYLzOzgScuRarSZnAwYwZ1oHbFyjQdthZbEci1PObCiGfRrBzaCO8iJandaQBD+nekbpVadJ+x376UznR7E583qICHVi4jAjqvKrSdcxoDz9k3qSHlPeS1/rIvjhdVW37NkbS8MlByk40DGxnbuytNK3maLYiCSwneb+RH61YtaOz7IdVrNyNgwk/8cde5l5XubWXoB3JdKkTHtUrx4hy+vKw1L/n92+rPaVDBw57fXlXaMkd2mn2+j0kNy1vbQZlDVVp+fcQiqKX77Av14z03WfR5Uea9arQNjeT6ycnUcC9FnW4jmLowkmuZLIWKOGZZZ/6PYzPbUtPsg2RtkxsEE5EoIT2KIPhjb8cy8mLv0272KYsYSI5m+eB6lJIHTIJIkVJlKVOxPgOXnLa3pYon2w+NV5bTr0FFPG15XbQyjYesw2KcyK3MynU2/+yMVxbiV1yHoHWhcGEXdNb9cQRdYcp81IWvtik2hnudjOypznCQhzohTyVvHFrfXPblt7wKuqJUqvcJY7a+2ku5MsTkjX+1CwS5w0lLSSXdYMRk7fQkyYTRkEZqqgHZ8G1ISyXN/ruEyWggLTUdk/m3NNKN8pFs73L8ZrHqKQWCvIohiUex5zh76SaJOVUIOazUB8ScOcOleHl/MMuf9Ow2MbeTMs8/WX/HkMSdyxe4/sjZ+jChWq8AACAASURBVGH7+WU+TSmJ3LkaTdSZS1xPeJb9s82BG3l2N4boqCiiYxJItkU804ONPIu/QvRf0VyNT1JYXECSuafb8kLmmU5qmpwPGf6Mz7gbE01UVDQxCQ5GGa7K/aspnTTb8+T8M8j5abnNmJ6GwZq3kslIujnPLZ9Gc2GRMJ9Pk8uC9S8v8TIlcfvSFeKzBWQiOSGGc1FnuHjriRMf+Sm5M8oljuaomkiOv8LZqLNcuZs1v9SH1zl/9gI3ElOt/A08unGeS7cteSbHw14vrHUmLavRjY2N/JkXPtbr4q9E81f0VeKTnDs+W3DGZ3eJiY4iKjqGhCxZ5lQe88IITMkJxJyL4szFWzzJGA3JQFqaAVtZMJfVdEtJMKan2tuFlyojtkTKnya5DqRbn2Myt0cvnj/neUp65nqhvM98/JyjwbXosOJhpl8ynsh3fstpNCfXmaXMIc3KwfYM49PbXIi+yv0Ua002PODG7Xy8Jj6LOmNKd7S9cpttkBmZH2gkPZt8scfnWTxXov8i+mo82RQv26WWz0x5bcl7p3bJ+IzsymymupIHds+jlzJ4yEy2bv+NreERRO7cwbbfNrNi9ijaVHJH61KDr04rpvteMyNnAI5vOdYJa39p6xPNfWl6Kmm5THM5Qi8YRwqB8CYjlFEg5KYK3mRc1LBVAiqBfxsBQ/w2hvoNYEsOUwz/NiZvRXqT9zG8UnF6brX5ODnHWnq8nYHvudF8YVweRKLzveq33An8YwLBsUeCu2Ovg9zjp16hElAJqAReiUD6X0sY1HcCGy9n3cm8UuDqzW+UQPq+oZTVv0v/HdkMKlMPM6pqjQw7977RKP2rAv+HBEISm/2LWn0XRJqpau9fVcjUxKoEVAIqgZchIDukylsqv9M0jJMZXnKUdu8EP/RrRodZp5y3W36ZB6n3ZEngjQsE0/WfGdfNl9K2lzVpBPTFa9Fx6HKiM85lZhlF9aRKQCWgElAJ/CsJSE84NqsTVYvqEUvUon2/oQwd0JMubfxo3WMcPx3PYi+EfyWoN5PoNy4QSLlP7OUrXI25Ruz1G1yPvUbM1StcvhpPspNny5tJoBqqSkAloBJQCbzdBKQX97gS9QeH//iLy7ce8uItc/Z7W+m/eYHwtpJR460SUAmoBFQCKoF/MQFVIPyLM19NukpAJaASUAmoBLIjoAqE7Mio51UCKgGVgEpAJfAvJqAKhH9x5qtJVwmoBFQCKgGVQHYEVIGQHRn1vEpAJaASUAmoBP7FBFSB8BZkvvFZHOcPbmFHdNL/7m5hphQeXDvJ9q1/EGffr/ktyBw1iioBlYBK4H+UgCoQCnLGGi+z9NMavGveQ0Kk8feWtx2+1igbHnLxxAUe/NeWDUkkRozGt2wh80ZaunIjOKTYVl1Oq+HhRU5ceJDL+y9eDxVT/Gl2bN9B5M5d7N6zm107I9m+fT8XbS9IMtzmZEQEO8y/72FX5HZ2/nXv9TxcDUUloBJQCRQgAqpAKECZkXVUjFyY9hGixoWWS173m8DSOTq6ovktd83m33S8aCnriLzRs9LDtXTyFNBXHstx5QZa6UcZXVGPUKgZ8x2vsnxjcTHcPMiq+ZPoWFFEEPSUaTqMGT9s5cwj66YdqbHsnuNPJb0WzyqtGfzV9/x69vEbi48asEpAJaAS+G8RUAXCf4t8Pp77eGU7XDUutPrx/ktMMRi4c3ARo7t9xspbGW33JuLDg2hWpw3Tjz/LR4zewKXphxhRTofeZzynlALBFE94UDPqtJlOfqNouHOQRaO78dnKW/kUPy/YElAEwaUZC+9kYGa4zZZBvjQYuZUbr+9loW8AqBqkSkAloBJ4NQKqQHg1fv/I3UlrOlgFwoP8CwTpHkv9XNCIzVgYV4C3rkw/wqgKWQiElyYscW+pHy6al3j3hzGa0BpiZmvG8wss9f+YNjNP8KQAo3xpZOqNKgGVgEpAQUAVCAoYBfXwlQQCj1nZ1hWNSyt+LMivun3tAgEer2z7cpaXR6vpUEigcNcNPLUWCinxMFP96uD/40VeFNSCosZLJaASUAm8RgIOgSAlcztqH1uWTWdkz3FsuW/i+dUtTBvoz6ej13AxxfFU6ekFwheEMLxvIP2GTWbxrliyfJGq6SGnN85m/GB/OnbqwaAJ84mMSXYEZD4ykHByLWEjetOtYyf8+49h1qYoZ6c54yNijkWyYcFkBgV+w7F0kJIu8/ucIHp360rgmIUcvWcxBafePsya2V8yNKAL3QeFEWG3Axt5FHOMyA0LmDwokG8sgXD59zkE9e5G18AxLDx6z2KKTr3N4TWz+XJoAF26DyIs4gaZrckSTy+EsyBkOH0D+zFs8mJ2xToomJ5c58SuzSyeMoSeYQeR8b2I3c3Sr8cwwL87vUfNYeffGbzxzDwkki78yjdBfeneqQuBn08irN+HZh+EVj9msCAYEzn78yyCB/agc9dAhk5dR9RDhbfhi/P8/M0kulbVo9FXpvOEaUybNo2wRXuJM70gPvoQ29bOZ8rnAUwIT8xknTAknGRt2Ah6d+tIJ//+jJm1iSgnb0YTT66fYNfmxUwZ0pOwg+ZUErt7KV+PGYB/996MmrOTLJOZoRSQlUB4EU/0oW2snT+FzwMmEJ7oPGx/cX0n88cNxL9zFwIGDGPCjKXsvSEzfcH5n79hUteq6DV6KneeYE73tLBF7M3DEon0w6MorxOpHXbR7BiZ/vcWhvr6Mujnm2SVYxmTon5XCagEVAL/CwQcAsEYy9aQXviW0CHofQheu4AudapTxUv+Xs36vm2JJyfn0tmnBgELDnA17iYnlwdSWfSk7uRjKLt+6eEhwlr/B58eCzh8PZHHsVsY7COiLdqE7y5ZOzEpkUMhjfCu3o9lR69x+9opts3rS3UPkVJ+szltmxZP/pNlIzvg46FF4+LH/OM/M6JVU7oPn8TEfr4U0wp4NJtF5OpBNPJtQ9/RIYR+3owyegH3FouwTL0n8+eykXTw8UCrccFv/nF+HtGKpt2HM2liP3yLaRE8mjErcjWDGvnSpu9oQkI/p1kZPYJ7CxYp5++lJ5yc2xmfGgEsOHCVuJsnWR5YGdGzLpOPWSikRa8mqEt1imo1iA1msm/9AHxrNKfnqIkEd6lGYUHAvcn3xCqnuKUk/pzXhapV2jF9+wXuxMVwfP1YGhTTojH7ICgEQtJxwhqXofaobVx5mMyTa+EMq+6KS9WR7EuyFs3kS0SuWsSgD0U0YnX6z1/BihUrWbPjAk//Xxy7ZvanoZw+jSttVyod7SQSD4XQyLs6/ZYd5drta5zaNo++1T0QS/kx254xaUSvDqJL9aJoNSINZu5j/QBfajTvyaiJwXSpVhhBcKfJ97G5+wBkIRBMcbuY2b+hOR81rm1RRjE1Kox6XqVpN/cEd5/eJ3rNQKq5udF5vcw/mUuRq1g0SBZWItX7z2fFihWsXLODCxleGZu5Epu4Pb8JorY4/ba/4NGJ7+hYzZegXVbxmPkG9YxKQCWgEvifJOAQCObkGYkOrYkoiHhW6sm620ZSru9h4/aLyO2q9GALvcq6UivkDKk2HNIDlrcphODeiqXx1hGeKZ51XUsgVhjOAbs91sBfkz5ALxSm09ongMSDLb0oK1Yi6IjCPIGRG4v88BREfIKPKsy5KYQHepk7HO+WE4iMs3qymW4yt5GIRluEusE7ibc5uJluMa+JiODemXU2OzGQEh6Il9w5e7dkQmQclstN3JzbCFGjpUjdYHY6AuHWvCaIgjud7YFY4+1ai5AzdgpID5bTppCAe6ul2DCQtpchpbUIbqVoMGwTV20sDBeY9pGI4Nae1TIK85/Eo239eU/0ZmCkrYeXfzBx49sGGSwIEvd/+oRCgicBW23sJBLXdKSw1oNOax8prAFPWd1enmJozU8PnUfgctixs+sjalzpssFh/bDks0iloCNmy4c1ghhvLMLPU0D0CeaoLS1A2t4hlNYKuJVqwLBNV+15ZrgwjY9EAbf2q7En0xZYxs8sBIL5ElMss+uLaFy74IhiOgeGeaMr1JWNdlWaxJaeleizzZEnT1e3N08xtP7poYJHxgdn/J5CRN9iaLVFqfSRD8VFAX3VsRx3BJvxBvW7SkAloBL4nySQQSCYiJ0jdxgidcIuZ1h3buTS9LqIugr0WbafA/v3sWf3LiJ3bOGr5p4I2hIM2mUxxBvPhlBDr6NS8HFrB2xhJz26wN49Z7kv98qmG8xt7Iq2xAAiMza+Sb8S8I4WbbFe/GazIpDG3iFl0Or+wxh5esD+l8b2fsXQinWZfkVhXieVrQGeCGJjvleM/uXOrIxWx3/GHHMyF6dt70cxrUjd6Vec0p26NcAsVhp/b/WEN15iel0RXYU+LNt/gP379rB7VyQ7tnxFc08BbYlBWDGA4TjBlXRoS3/GHqc5ilTCexVFEBvy3Q2rCcF0jTkNXNCVHco+p2shKx+E9JgtzAhbxZ+29fnyfgHHg6mk0/P+pCgF95wEgkTCD7IjnyvdNtsywcSNuY1x1ZZgQOaM4deAd9Bqi9HLkTHW52op/dke56mY1HB6FRUQG36HLZn2bMt4kJ1AkBL4QXaydO2GPYoYOP2lD3ptCdotvGC3XBmfPuSxgt1LCQTjeabVEtFV7M/y7YsZUN0dQfCg8bdXFEwzRl79rhJQCagE/vcIZC0QBE96/mbrMKyJtjbUgr4y7UePZ/wXXzBu7FiCx4xhdFAQo4JC+SVG7qAl4hY2x0XjQovF8dmP3J5tpFthAf37k4iyjfptfG1WAX01vjxt+9EqEMSaTDmnFAJp7BxYAq1Ym7CLyvOphAfKnXB95ijs+BaBIFJzyjknIZC2cyAltI55Z1tUUsMDKSqI1J9jMZNLCT/g5yKv12/P6PHj+eKLcYwdG8yYMaMJChpFUOgvmDHIAVgFgq5SsPPaftKIHPAuWrEOYdbpFilxJe3cBETfmVxVJoOsBYItfhgecXn3CmaM6U0H33IUEnRUHH1UIX7yKxCesbFbYQT9+0zKnDFWS4ueal+edkTBLEwyC0LSIhnwrhaxThi2WSX7TRkP8iUQwBizCD956kVwpczHgYSuP20RnopwX0ogPFlHJ3eBwp3Xmx0UDZdm07CwgNarDT/+rZwPUjxIPVQJqARUAv+DBLIRCEUJDM8gEKwjK43YhHmKEXlmJvLGPrXMVogm87Jffy7dXUwLF415KdkJmwawB/acDZ1dEXTlGXXEZi0oGALBeH4atUQNYpN5Vt8Ge6QzH+RDIJiuzeJjOdyPZ3EtQz+UlQUB6RnnVn1O4yo16Dj+ByJO3+HJqYlU07+iQJDusriFi9mpcWzmjOH5hs64CjrKjzpiT6/FcvHPCgT54S9iIwjrWZeSLgIaQUfxRiEcsm1oBLyMQEj/YzT/0YnUmnreKiBTOPlldURBS0n/DSRknKWxU1APVAIqAZXA/xaBvAsE023mN5Xn+osR8GtOM8oSCUv8cBFkk/Nuh69CRm4vwunlJTsGdmGDwkfAclkSazq4ohEbMde+e17BEAim2/NpKmrQFgsgRwxyQvIhEKR7S82WCd17mbcaziwQTMT+8AnF9KXosfme3UpjiHoNAoEXhPfyQit40CVzxlinO0Qazb1pz9F/XiCk8Tjxmd3xMeXvPcxo541e0FNpzB/2aY78CwSr9Uv7Dr1/t/l2gPR0F0Mq6NDo3qN/hNK/w45APVAJqARUAv9zBDIIBCNXZ/oiCoX5dEsGC4J53rcaeo28YuB7YmwDexsSSbJ3VMaLYdQRBbSlAvglg2Oc9OQal+PldYqP+LlHcfN8tv+mDEvs0k8wrooet/qzFeZ261SC/gO+OqM0OaSxw+yD8CFTzytt86ls+dQDQfTlG7vNHyxTCXo++OqM05xy2g6LD8KH9pGjJWGpWz7FQxDx/SbGMqI0nObLano0ggfNvo9RmPIt10uSYoiZfoSgCjp0FYM46sQrjYi+76AVazHNFmfDGSZX16PRVWTEIUfnJE/ZxC9qYd7wx26RMV7ka9nJ0b0bmx2+hRj+nICPXsd7ww8q4mWbYmjJkruKuJmja5sOcqXzeltAEo9+7kFxrZZi/ptwXlmYzolxVdC71We2Yh4k/UgQFXQ6KgYppzZk78UI+r6jRaw1DVsybcUl02f6QYZ76zJvTiTFsbC57IPQGXsU0w8wot5w9iv8DSR57wJ3AZfWy3hgTaZNILRcctdeNjM91+lEqnXqpyHf2YWpfIFEwkZ/SmoFxCojOWD3i3G6GdKTeJysLIMZfle/qgRUAiqBt4hABoGQwt6h3ug0jjl3ZVpMd9bStZQOQXChcp9VnHtk6ahTbkYwqVMPFtqcBKX7/Na3HHpBS/EmIeyJk8VGGg/PrefzZp2Yf8nSWxpiltC6uBax8iC2Jdjs6uncXNae4u41GX9E4c0vJbC0lZvZGbJvhK0zk9vu+/z0iRuCrixD9ylEjc3ioSvPiEO2nkQiYWkr3AQtJfpGKPZusKwKcBN0lB26T2H1kJe8NUXU6Cg/4pB1ZGriztqulNIJCC6V6bPqHBYMKdyMmESnHguxYeDxBroVERCK+rNZaSUx+3O4ImhLM2inbTmAiZtLP8FLXrJZZyx77hmRpxEubQyiZYUi6AQB16r+zNx4nDv/d505DUQ0+moEH7W48ZsenuD7rhVxEQQ8u2/iaVoyyWbMKfzWswiCrhyfRT7g8Y3DrFwUgWU63drha0R8ZyicMw0xLGldHK1YmUHbEuwj9fSby2hf3J2a44+gyBkeb+hGEUGgqP9m+8ZCcrkx+2u4CmhLD8KeTGWBUhxLiWvo6CHP9ffgFyUrq1jUiL7MsIE1nGK8T0m6rHH4uBhiv6ORqws1Qx3CL+W3nhQRdJT7LJIHj29weOUiInLyI0g7xYRqerRefVAshrDE0nSNbxu6mcv++yP3ZRBOcvE+xpcfuKF7pzU/XLeVZUUC1UOVgEpAJfCWEbALBNPtTQxrWhUvvYBGo0FwK02N5kNZ5zSSguTo5QyuVwpXQYMgFqFU2TJUrD+QJacfO4/SUi6zfnhjvN0EBG0hvLw8KFK5MzOPOi85exa9kqGNvClSsjadBgyhf8e6VKnVnRn7E+xOhKbbGxlarxyFBQ0ajYDOy4dO885ieHGA0GaVKKKV4ywglqzFoPV/Y3y6iwkNy1NYa7leLFkT//mRbBxaj3KFrenTeeHTaR5nDS84ENqMSkW05rcJCmJJag1az9/Gp+ya0JDyheU9COS0lqSm/0Jr9iYTvXww9Uq5IsjPLVKKsmUqUn/gEk6bVxWYuL3pcxpU8EArx1nQ4VWlLXNkh8vn+5jUsDwe1rjpvKrS8usjFvEh3WdfqB/vycxcvCjzXjXahkZy/eRkariXok63EUxdGMm1VIn7O8dSr4SI1qM8vo0/5qPG/Vm8ZxW9SusQdIUpUbk5YSdlwSTx4NfeeMtvhBREvGr2ZunZJKSk3UxqYVnGZ06faxlqtZ9p3oTKnMhn0awc2gjvIiWp3WkAQ/p3pG6VWnSfsZ8E2yDZdJtNnzeggoeVnc6LKm3nYEnmJBqWl/ecsKa/aku+PmITaspaYuTK8n40qOhpZ1W0cmOGrLvB/7d7Ei18iiNa8921TC3azzxGuvEyMxt44lKkKm0GBzNmUAdqV6xM02FruawwvkgPfqW3t2jJI6+a9F56lqSMRhQ5KsZYNo7uRvNqxSzPEtwpV68DfRb8qbDEpLGjfzFregpTvkkQW5WbLqXs4/Ny8kul6jErV49MZfrVY5WASkAlUDAJ2AUCkoG01HSMJksLKpmMpKemYciqQcXIs/grRP8VzdX4JCdTfcZkmpLjuXI2irNX7pKc7cDKRHJCDOeiznDx1pPM4clxSzNY4yZhMqaTmi4HZiQ9w/l0oxzhzOdT0/8fhrQ0DEaTWchY0pduHh0b0zOcTzear8l0PtVpngCMz4i/Es1f0VeJT1JOe4BkSCPNYMSMUzJhTE/FHGU5bnbOlrSkWX6wozM+vc2F6KvcT7HCNzzgxu1nzgJMvjr1IdfPn+XCjURSbZc+usH5S7dxjo6JpNuXuBKfbLcGmBkp4mHmkWbhYY+IvBo1OYGYc1GcuXiLJ85JNIuP7JhiziNl+tOs6VeGbjnOxMrMTsI5jMysjM/uEhMdRVR0DAnZFC5T0m0uXYnPoezJcTBhSE0j3Vo2MOdXmrWMOeJrNKRb89NSN2w6yXaF4dF1Lt1OzpxPtgvUT5WASkAl8BYRcAiEtyjSalRVAioBlYBKQCWgEnizBFSB8Gb5qqGrBFQCKgGVgErgrSSgCoS3MtvUSKsEVAIqAZWASuDNElAFwpvlq4auElAJqARUAiqBt5KAKhDeymxTI60SUAmoBFQCKoE3S0AVCG+Wrxq6SkAloBJQCagE3koCqkB4K7NNjbRKQCWgElAJqATeLAFVILxZvmroKgGVgEpAJaASeCsJqALhrcw2NdIqAZWASkAloBJ4swRUgfBm+aqhqwRUAioBlYBK4K0koAqEtzLbXkOkTSk8uHaS7Vv/QPlKgdcQshrE/wqBF7f5Y+NaDt7Ndo/0LFNqeHyVw8djFS89y/Iy9aRKQCVQwAnkLBDS7nN2xyrmzQhh/BcTmTprCZsPX+VRhlcSFPA0/gPRM/AwOpLV88P4MngsE6Z+x6qdZ0mwvUvBHgMDDy+e4MKDjLv42y/4Bw4kEiNG41u2kPnlVLpyIzj01uWnROL5PWzbtp0dkTvZc/qO9U2bOeFL5fafe9gZuZ2IbbuJzvAa8pzu/Df+ZrzyE93/UwhB8CIw3PbG0ZxJSI+OM79/fcq4itSdftn+srWc71J/VQmoBAoqgawFgukhJ5cMwvddEbfyTen7xUzmTB2Bf+MKFNYKFCrfkqA153iW5YucXiWpEkZj/kYrr/K0nO6VjEbFi41yuNJ4m63DPqJ0lS5MXbebo0d3sX56IDWKaBE/nMI5hRZIPzqainqBQs3mk+ElmTk84M38JD1cSydPAX3lsRzP9BKmN/PM1xeqibiDS/l6SCNKaDVo3/Fn06OcC6N0fw2digpoi35IwKRF7LtTMMrZ62PyekNK2zuUMuY3juqpPPZ45heoZXqcRNwSPwrJb94UP2LaBUXBz3StekIloBJ4GwhkFgiGW2wZXI1CgkChD4ayLV7ZkCZzbn5bSsoNh9aTD0fu4F7O7XLeGKTdYOe3I+hUuyyVhu1XvGI3b7e/vqvSuLHzW0Z0qk3ZSsPYn+vI2sT1hS0pUrQFC64pG0SJB7/14b2iPfhVMfgyxYcT1KwObaYf59nri/TLhZR+iBHldOh9xnPqZQSC4Q4HF42m22cruaUsIi8Xm5e7K+UX/N3l11i747fkdg6CzsS1bxuaX1Hu2nk9z1/uaf+qu56u64y7+TXbGvQ1QjirLN5ZkTBcZLqv/PpzDdqSg9mV1Zu9s7pPPacSUAkUWAIZBEIa579piIfcMOgqMuKgonezJUG6x4bu76LVyCLBixYLrr6yKTH9yCgq6DRoNHkdrdgi85o/048wqoIOjUaTt5G16QbfNhRxabGYuxmFkjGGWS3a8UNBneC3pvVlBYJ0byl+LhrEZguJy5j215wt2QUnJS6n3XvelBEFxDphXMquE0s/xcRmH1K9pA7Xtit5nF2A6nkrAQMnx/ng5VXEUs/11Zl8JicVKXFvUw/KubuZBYLY+Hv+/m+JRjUPVQIqgddGwEkgSAmr6FBUMHeQukpjOJbNCDrtSBAVzR26Bm2JAH5JfLUeIjU8kCLm0YqeGiFnX1lwvDSd1HACi1jSn7dR019M+kCPruIIstJSLx2Pf+LGVxQIPF5JW1cNLq1+5P6rZf9Lp9Z4KYwG7Wcwp60ngr4ywceyHrY+2zGE1pNW89WHImL9OVxXO6+cmZtimd24NhMWj6SSVbh/8NWZbKcZpPtb6fNROyYN/xhRo6X4gMg8+ITkHAX1V5WASuC/T0AhEEz8Pa+J2Qyr0Qh4dN+UvRk8dScDS2rNQkIjeNJl3SMk42VWB/Wlb7/+DBgwgP79+hPyezwmTNzdFkK/Pn3p138AAwb0o9/IlVyUR3vSY6JWhzC0ZUVEs0DQUrxON/oPGMCAQeNYf9VActx5/tixnkVfB9G7TXMGrY/DkPgny0d3plH95nTsNYpZv18h2cbyZeKBxOOo1YQMbUlF0SIQtMXr0M0c30GMW5+dlSSJ3wLfRSu4UrXPWi7nYLt+ER/NoW1rmT/lcwImhGPRVEYexRwjcsMCJg8K5BtZkUlJXP59DkG9u9E1cAwLj96zmM5Tb3N4zWy+HBpAl+6DCIu4YW+EpXv7WfDVV4SETmXq1J849sTaYz87yfLJk5gcOoWpU+exW2nNyFYgGEk8+zOzggfSo3NXAodOZV3UQyfR9uL8z3wzqStV9bKlpTMTpk1j2rQwFu2NU5j5DSScXEvYiN5069gJ//5jmLUpCqV/ppR8m6h9W1g2fSQ9x23hvuk5V7dMY6D/p4xec5EUW55m85l+cAQfDdzO/a2BvKvVUrr/9sxlVrrH6k+7sOjGDb5vLKKv9iWnsxgMS08vEL4ghOF9A+k3bDKLd8U6piKMj4g5FsmGBZMZFPiNWThLSZf5fU4Qvbt1JXDMQo7es6iO1NuHWTP7S4YGdKH7oDAibmQhWgwJnFwbxoje3ejYyZ/+Y2axKeqBE+Nc2aREsTo0hClTpzE1NISvJk1mVZSVmCmePd9/xVchcr6HEjp3VzYEsz5turMQv+rjOJESTWgN0VzPxZrO/jT2Ow3X+LFrU4bvjOfQqPLoNHp8xp/KVkzY71MPVAIqgQJPQCEQnrKxm4fZRKjR6PjPmD+y9wUw/W1ubGVTvHyt97ADpJPGw0s7mFCvsDUMx3SB8cl1Ds7tSBmr1UEo3INfUwFTHHvmT2GS//tWgaCjbMthTA4JIXTaIvbHJ3NkZic+Kilaw9RSolkPa34a5gAAIABJREFU2tXwoX7LxlTx0lnOa4vgO/kPksy4XyIemIjbM58pk/x53yoQdGVbMmxyCCGh01i0XxY6KRybXIeihSsz+PeH9ow1xCylXUk5HgJu7zXn8++3ceFRRlu3ibhdM+nfsAx6QaMwcyfz57KRdPDxQKtxwW/+cX4e0Yqm3YczaWI/fItpETyaMStyNYMa+dKm72hCQj+nWRk9gnsLFtkm/9Pi+HNDEL4eAhqXFiyKtwoE4wPOh0+lVQktGrEmodGKeGUpEJI4HtaYMrVHse3KQ5KfXCN8WHVcXaoycp+Frpzw5EuRrFo0iA9FDWL1/sxfsYIVK9ew48JTzE+WEjkU0gjv6v1YdvQat6+dYtu8vlT3ECnlN5vTVgcMY+xWQnr5UkInoPcJZu2CLtSpXgUv+Xu1iVl25HbwwPPNPagz8TSGlP0MK6dDKNqJNRnMGaZr39Kx3y88kp6wqr0r2tJD2OvUZ0s8OTmXzj41CFhwgKtxNzm5PJDKoid1Jx+zCM/kP1k2sgM+Hlo0Ln7MP/4zI1o1pfvwSUzs50sxrYBHs1lErh5EI9829B0dQujnzSijF3BvscjJR0NKPERII2+q91vG0Wu3uXZqG/P6VsdDLIXf7NN2gZMrm7REzv06mZZy3mpEPhr5GxftlrwU7pxcx9CarhSp+zk/Hb6pxJbLsUTCsnZ8EHSENIxcnelrqZtiXaZfVpQfORTpPjtHtaLHsquk84wNXeSVMSL1Zl9TCMVcHqf+rBJQCRRYAg6BYLrGrI8towXZF6D65OxNipDE2o6uFguCRqBQ981YvBVM3JrXBNEsHPRUmxjlGEmkbiXA0zI6F4r2IlwWCNa/1M3dcTNbEETqzrjiNJKSL0n+LZB3zB7VAi51xnPUOgx9fmYaHxeyhil+SOhZ29Dw5eJB6ma6u1nCE+vO4IqyPZQS2djNC63gRsNvr9mibv5MuxVJaLuKuGvlewW0HpVoHbyBi3azhuVyU+xs6osaXLtscIxOgZTwQLwEAXfvlkyIjLMyM3FzbiOzybZI3WB2xmdIm+BO53VPHfEw3WFBMxGNSxuW2zsK+efnbOpaCEH0ZeZVRYKyEAjS/Z/4pJCAZ8BW++hdSlxDx8JaPDqtxWmhwNPVtJenGFr/hPOKQYkHW3pRVqxE0BGlDcDIjUV+eAoiPsFHreUFMEYTWlNEED2p1HMdt40pXN+zke0XrWLDkcIMR3Iet8RvvuycaOCvrz5AL7jRZN4NReeUzqmJ7Qg+LBe2VH4LLILg3p2fFa410oMt9CrrSq2QM451+9IDlrcphODeiqU2sUUK4YFeCII73i0nEBlnzQ/TTeY2EtFoi1A3eCeObLrFvCYigntn7NkkPWBLr7KIlYJwRnODRX6eCKIPwUcVkcuVjYlbS/zwEPRUCT5mtyiZQaWfYkLd5nx31VZuMuDL7qvpJvOa+zDysEVFmW58T2NX2RFUpP6cWAdb6RGHJvrRJuyURdTI/jgN5PbDBb8fEixCMbtnqOdVAiqBt4KAQyAYLzO9rkMgVBl3wtG5Z0pKCpu7WRyS5A7Rq/c2e+P6ZFV7XK0C4f1JfznCSNvN4FKWaQnBqze/KwTC8/WdrfeI1AnLvH5admIsb7Y+CBTu8av9WUiJrOtS1G6xkJ9n+3uZePB8PZ1dZauIxuz0lmnA9DSGowfPO5nJbc8DAw/+2kxYv4Z4yw2qPE1TM4jdit5TSvjB7Njn2m2zIw1A2t4hlNHKVptjTlabtO39KKaV15Q7i6bUrQHmjrbx97cUDXYiy9u4oHFtxyonL7xUwgOLIoj1mROrmHzPQiCQHsOWGWGs+vOxo4E3HCe4kg79+5OIUvY12QkE0w3mNnZFW2IAkYo8NnNK+pWAd7Roi/XiN9syDlMsc+qLaMQ6hGUE7oCbxZGBPyfU5lNrb2+6NocGrgJizVDshpJn2xnScabVeTGdA8O80bn48UOCzWnCyKXpdRF1FeizbD8H9u9jz+5dRO7YwlfNPRG0JRhkd8dPY++QMmh1/2GMk3NOGtv7FUMrj7CdFGUqWwPkTr8x31stPaYbc2nsqqXEgEin/JcTl/RrAO9otRTr9ZsjrXlhkxRB/9JatCX78LtCLyZt60+dwZF2i4Qj0JyPjBfD8K08kkM2K4sUxxI/2TIg4Ob3g8UZV0riRJgfjUbs5IENpeE0X1bTmwXCJ8sTHeUn58epv6oEVAIFmIBDIPCY1e0VnX5guGOUlzEB0gN+bOVitSBYHAttfcfT1Q6B8IGTQNjDZ9kIhKQ1HewCoXbYxUwWBMcqhwwCASOX5QbeLEgECvv/Yo/py8SDpDV0sAmE2mEWPwl7iHk9kHh2YSW9qrggCHqqjjth7/RzFggiNaecc0p72s6BlNCKZGQiO3UWzTSiew0CwZ5EA48u72bFjDH07uBLuUICuoqjOap0Ws1OIDzbSLfCQmZBIYdtG23rq/GlzRHA2gkKnj35LaOgsMcnq4M0tg+oxUjbLk/SXX5q44Ggq8CIg7LlQiJhVSABK+5aOysDUZPeR68UIlICP/i5mB0c248ez/gvvmDc2LEEjxnD6KAgRgWF8kuMzepiFQhiTaYoN7cgjZ0DS6AVaxNmdqyxxTWzMHu2sRuFBT3vT1JY1qyXm27OpZGoMftI2EIgT2zS+fPLaugFD1r/aPUBMd1haftGTIm21Up7iLkcvODQyCpUGnVYYY0wcWdhc7NvklCkGxsfP+Xk9OZUaTWfSzYRIYeafpQg8wogV9o5K9Rcnqn+rBJQCRRUAgqBYOLGd41wMZv6NeYG/q/s2hfDSb6oIo8W5OWQ5Rh+0NFSvEzH/HhlO7tA+GjaBadOUgaXvUCwmeHlUb+WYv222zm/TDxkz/x2NoHw0TReZa+XZzsGUlanQaw3m2vWgXvBFwgSz86t4vPGVajRcTw/RJzmzpNTTKwmr9TIm0CQ7i6mhYtlmeiJTOXnORs6uyLoyjPqiFVt2DrBooFO0072jMzuQHrITx0aOM2LP9nSk+JaLe8GbuWJMYY53Yay3WapwETs7HqISguA8TzTasnWiybMs/lzZPe8/5+98wCK6vri/77dfQsIKGqsqD+7scSGBXuLHVTssQtBY48xdkVN1Ng1GntLYo1ij5rRxFiw+7eC6N/u2EeRURh2/7tvPv/Z/li6omJ8zDD7yn3nnvO99737veeeey9vSxAk7s1vgM4ylTcZ79yrtQS4CWiKDHBqkE5sTNdmUddDQKw0gXNGMJydQN2AJThGR5wSUz0y3fmFpjnLMcrlxTdemYKfaJ7WnAv/Fv58li+AFbdk3iizVP1BBhQxTxHW0XjRI8WDkCrSyk0FgY8DARlBAOneKgJz22YnaMvwXYSz4ZebYx1Lt7rRdVUmck7WELxYYfcGmHtK8iGGlDwIEk+XNkVn8QKIfOHSizbnmzJBMHBseGm05mcFHXVn33KomXE9QHq6lKY62xCDyyqIDsGyg5g/w/huvTxy33lTeriIxjoVuvrOdQLeF0FY9oZDDKarP/NlLi352v/uXADLcCJDBIHXf9DFxxxcGchamcvbikwMK1u6oRJrMcO+lGQ6G0EnsrYj4wXCarRiqTww4vVeQgppELybMX/zcAJHRTi8N2aPwqNfmqBT56L7Nlu9Nt1idl1z/EAuOm54niSLxBfeliDA6z+64GMOaAxcS1JorN4rsdYMZ7bpxUZ6wu/tc6PWFCF032P2hPoTstslAMYpNYWj1xwaUpps1X/E4TSxp9T/bfMOmN8zLxrMlcd52BIZjjG8lLnToCyzbIdN+VUQ+NgRSEQQwMjVJa3IpzE3/mp8Gs/nqqzxtxobx+FvzS5NFYJHRUb86+iiWW7Hb+5kW9NATf7gPU5XpX4vIQVtMQjZ2vG7LBbr2bJmjriF5GIfUiQIpmv8VEtniUFQ5wxgpWNsGd5ED/vcfrNnRFtyGIl7wCZubhpGYMu+LDlv9YW/XheET7XJXEqCERgvT8ZP1FHtB2f8gHRnLvV1KlxX87MOJWhxnWuesN0ag/DF+HOJvCrxGzvgJYj4/RApux5rjSKXjXlbyyuGNQHZEFyXv9X/RT9fjWxBKCMXJlZCFLLR9nfZfE1DBN+V1qIp1I+/khtiaLgg8SJR0lPWmRsrdS6CfnMZi9YfYVhJLe41puKIlzReYYqfiODZgY0ZGWJI+JOQ0l8limUxx4Gc+P5ztIIGj1yNmemy4EHc7+3wENxoudLePBs4bnbPm+NF6s0kUm6fGTxJkvWEbUMJ2rKMSrRoUALbLTEIXzDe3H13/MWzsYOXJTj0B1uLKz1dR/vc5hiMIH5LFEgK+iPDKKl1p8bUKw4JZACbuIMDKKZR412+OpVqjsMRr+uUlupRwukJVPXwovHCZAiv2VvzpXVIUVM4hD9l1cMh1D58ZP5udNyY4dgHhxzlQEFAQSDLIOBCEMx6veTEnECKWiKXPSjTbSmnntpbwHiure9NaTcBtXdF+v1xyxmEaDPJdGMO9WwzAdQ5KtJlzDR+mvI9vRuVxsc2hdDcY6vz7VqO3bV+kRP29CWfbZaCxrcpoxYsZvrEpUTYSIScILjVmU6UpQOo59qqIHw1KgTd/+j8a+Kldt9EDxJkXg6NL01HLWDx9IksNSsiPWBhI/NSstaYC7O5hqPfUlKbjUoDwrkpb9wM9wjvWwqfqmM4IuvIWRsBFaKffIaExP2FjXEX1OTpFi6b3SDxcPGXuAsaCobslQW1mbg1uy6iSkOR/gecBAwjp8eURxQ8qDrqCDESSLGXWf/tl5T0EREEkSKtxrPy4HXLDAXL7AQvAbVPe9Zb2ksT0dNqIqq0lBnyj3V6n+kxR2a2oahOQPBux28vEoiNtbWicZvplF1AU7gvOx4949rBZcwLv2EJmjRELqBJbjViid5svW93Reu5/ksLcmerwPC/nVMmifuTEF8NKtcgyjRekVeHh/G5VwPmueypYIz8keo6NTnbrXWZXQEv17Yhm6BFHhtjur2KNvk0CIKOEl8t56ytrsddD2dk6/bMtQceSvdZ2NjdErjYLVzWQkoPWfylO4KmICF7ZZXA7p3QFKG/I+LPQOSCJuRWi5TovRUnNNf5pUVuslUYjhyaDGFjvEhYFXM5e9Jovix4NQ0cLbfjzjC5phfawsHsTsz3bU/bAi5VGgr3+0tWF+XC9RzoXxiNeanlzwJZGhVDvL3o5cmUYwUBBYGPBoFkCIJZd4nn59Yxvks9yvnmQBSzU7hybWqVz0/2z8rQqM80dl+XT2GT2xvLsR8aks9CBgTU7gXw6ziOX49H8KO/iKD1omC5WjTvGMKsv22+8PgzzA0sjqc5/kFQ4/5ZeQIm7MG+y6ycIKjdvSlQuiZ1qxTGS+uJr38PZhy4L+tJ23V5Az2I58zcQIp72qYrun9G+YAJ7LEoIvHi5ArGjZrLAfsGFLF/8n2DypTK54mnbxWatO1Kj65taVC5Ig2CF3DsmT3EO4bdIxtQOrdtPQfBjQIVWzDl76v8GlKdwpb8VAgaH0q3nsVpw2v2j61H8exqi3dEEPNSsfcabhhfsOs7f4p42jwxYl4qBM21G4z0+E++q54LrSCSw7c4JSu1Zmx4FOen1sQjdwVahoxh1pazHF/SnZpFvVFb8NaQo0Rtgldf4/893MnQ6nkQ1V4U8atNtUq16TF/D8u75EcjaPDMU4L6YUet+UmP2NDV1zJHXhB9qNB1IafNrMT29/LMMkJq+ZI9b2Va9wymR6uqlKzYjsn77GVl4tZvodQt5WPxRplnfbjnL0/9kNVpbGRlIGJqE4p5m7ERcMtfmT5rZeRQussvzcvRb5/MRWW6w/oBjaiQ3+ptEjx8qdR4GFtsIfixZ5bQp3o+ayCemJ18BQtQtEYvFhy3zuYw3fqVkOqFrfVTJaDxKU3rWacxvN7P2HrFyW6b3irmrUjvNTcwvtjFd/5F8LSRXjFvBYLmnrEjw5llIdTyzU7eyq3pGdyDVlVLUrHdZPbdt3sg3gQbIxfCKiPm7sgGR72zl0Yqv4brrO30P0S3YvTYmML0RNNNZjXyxa/XIk7aF+FKRuTrPSEWwm6JB2o0z+klSiatcklBQEEg6yOQAkFIrLj++U0unDrLlZuPeZ3OXkH8o0hOnbrI3Vj7AxIvb0VyK8b+EUych9k9HHP7Euejn8p6xdY0coJgnuYY8/QqZ09f5PoTWY/NVZztPON6gCHmNpfOR/M0+RCMZHIy8upBFGePR3Dy4g0eJwOSUR+P3miyuKwlyYRRn4DeJGFISMBgv24yoo/XW3rh5vuJruuNlmeTXI938YtLcTy4cpqz1545xt+lp9e5/txeDiAZEkgwGDGZ23OzLpZze+Mez+Poc5w+f40n8bZrhqdcO3eRWzF2T5INAlMMty5elpWxKzQmYu9HcvbEKS7cfJ7E22TWI15v0wMJk1FPfIJB5tZ3lWc9NxkSbFhKmI8TDHbdbfdfxbrMwDHjLMPfgrN5GSD5n5GXdy9z5uQZrtyNSayrZCAhwYDRCphVT70ZTyN6l+t6o1mXpNfjLell+ZliuR95lhOnLnDzuQuu5mLJMDYS9xY0pvQ3BxxrWMhyS/bQ+PAfZnarT61mXzP/6BPnlFnX1NIrXriWvWsa87npOr9/25NB07cRJXOyJJdUuaYgoCCQ9RFIF0H40Ga4EoS0acGH1ljJX0HgPSNg9py0rEPY20y9ec8qK9kpCCgIZG0EPg6CcKA/hWwLJWULWp/uHlLWhl7RTkEg8xBIODWWOm8wtTHzNFAkKQgoCPzXEPgoCELM70HksK3P8CG3F/6vFb5iz0eOgPScI/MH0rd3F+qW8mPkMcW39pGXqKK+gkCWQiBrEwRTNOuGtcUvv32zJhWCNjcVW4WwJMOrxGUp3BVlFATeHgH9Ab4pokVwK0LreWcVz9rbI6pIUBBQEJAhkLUJAnE8vHqJy1ciiboazbXoq0RFXuHypSvcjU0cmCazSTlUEPhkEDC9eszD9AQQfjKIKIYqCCgIZBYCWZwgZJaZihwFAQUBBQEFAQUBBYGMIKAQhIygpaRVEFAQUBBQEFAQ+EQQUAjCJ1LQipkKAgoCCgIKAgoCGUFAIQgZQUtJqyCgIKAgoCCgIPCJIKAQhE+koBUzFQQUBBQEFAQUBDKCgEIQMoKWklZBQEFAQUBBQEHgE0FAIQifSEErZioIKAgoCCgIKAhkBAGFIGQELSWtgoCCgIKAgoCCwCeCgEIQPpGCVsxUEFAQUBBQEFAQyAgCCkHICFpKWgUBBQEFAQUBBYFPBIGPkCDE8+DCQbas/Jl5C1ezLeKOsgb9J1JZs7SZkp6XD69x+sBm1uyJQp+llVWUUxBQEFAQSBuBj4ogSE/+YVKzwnh4l6dt9/oUMG8BLbhTMjicR8rWDGmX9qecQnrCuT/D2bZjJ7t272H3rh1sD9/Kli3b2H3wBNFP37BJN91kaVBRcogCgkqFSqXCLWANrz5lrBXbFQQUBP4TCHw8BMEYyfzGuVCrRCqMPYPBeI4JlXSWj7Km+BAOG/4T5fHGRuijNjFu4GjWXf5vbvlrenaWY5dj3xgfTPc5um4x03pWJJsg4Fa8OQPHT2HqlLH0b1OenO55qNZvAzfeqB4ZubOoKV6WLckFvDps5L9ZCm8Ov/KkgoCCwMeHwEdDEF5s7kIetQqVphD9/rL29uIiNxE2aBA/7LqN6ePDPhM1NnJ+QiVElZbPR57gjdq4TNQmM0VJsVFs//Erqn6WjUYL7vO2jqLYdUF4CzoaLrgnk/WSQ0M/R1Tnpusfb0ZC9PtD8TV7tFQC3p02KwQhMyuBIktBQEHggyDwkRCEeLZ1z43a7MIVKzHhvPGDgJWVM5VeXGL7r+Gcf/62TWgWsdJ0i11hnalZpiRFcosIKh1NFj+WNepvoqeB49+XQastybAjiWlUwu4+5FPraDD/7hvloT/Qn0IKQXiTQlGeURBQEMiiCLwDgpDAnX+PEp2Zbbh0j/kNdJbxXZVYm5k3P21/QRatS5mrlvSMyyeu8Mwo8eiXJuhUOpouffpGjbdDMekxS5q6I3i14dcYx1VA4tHiprhrSzD08JvFIigEQY6ncqwgoCDwX0AgEwmCngfHVjKiZUm8s3dic6YMwko8O7KYUQO6UD2vxkoQNPmo3r4nPXv2Injafp7JO8zxN9m/cCQ9WzegRpVKVPVvTFDoJH49+ZhEfOX1Ay5H7GXDLz8wrHsL6nf8mctGI3d2T6Rj3Zo0G7yR64keSKao46+ze+4Igju2pUPweNaeeZ5s4xV3+xArwgbQs0MgAV2+ZdlJ13TxXN89lxHBHWnbIZjxa8/g6gSIv76buSOC6di2A8Hj13JGliAhaifzp4xhaGhveo/dyp0k3EnixaXtzP0+hC5tWhHYKZSwDRd4acfNeJ0/f/6Rcd/2o0+vYayNMoL0gvPrwggJCiBo4EouZEpZJoNhOi/Fr2uHu6Cj2dKn6XwihWT6/YT6ahCrhHFRXr6vTjKhmjeFu6znnh2XFETYLxvuH2bVlGH0DKiLXxV/GtUu5YhBSG6IIf7mfhaO7EnrBjWoUqkq/o2DCJ30KycfyxWxS1d+FQQUBBQEPjwCmUYQYg8vYOTkhUztWAzR5yu2ZkqjIvH038VMGN2fRr52glCYLweOZszY8UzdeNExxdF4cwN9y3kiCJ74Dd3CqdP7WRJSxfLRFrR5qR92yNHwGk7Np3NNX9wtQWUqBJ/OLNn8NaXdBJuXomLqwxjGaBa38KVCv61EP77N1uDiiNn9mXZB/rE3cGNDCJUK12DolqvEGmPYHVIE0TeYPXH2gjcSvbgFvhX6sTX6Mbe3BlNczI7/tAsOQmOMXkwL3wr02xrN49tbCS4ukt1/GvasjPdOsHVhD8poVYi1Z5LIuSLFcGx6c4qXCmDqnkgeP4tmXY9iiOqcfLkw2qqE6SGnty8npKKIIFZgbMRl1vb0o1ydFtQv4YVa8KHLHw6F7Yon/jVeYO2oIQwePDgd/9+yOCINeYmlE78uCI9MIAim6zOoJarxCVrFjfv3uH31HP/+MYuQuuWpPWA9VxNcMk721MTdHUOp5qNB9G3J5PAILpzZxdTWhRFtQYqJCYKRmxv6Us5TQPD0Y+iWU5zev4SQKl4IKgFt3vqEHXIljclmrFxUEFAQUBB4rwhkGkGwai3xZElT3HN2Y1u6PrbptNV0lWk1RFvjXYdZiVpBwBjJrHre1hkNhUPZbycnphvMbeBpnX6mKUjXzU+cvfyE/YQWspEOtQ+lO4SxfGYv/PJ48lmdKZywy0hGRUPEd5R2K8uok9ZxbGPUCno2DGTGaee4dtyxMVT29KHR/GhbYx9LePe8aLyDWPfSJtQQwXel3Sg76qQ1sNAYxYqeDQmccdoWaGgg4rvSuJUdhTUrI1EretIwcAayrDBenkxVUUup4cdkAYombq9qQz738ow46myQTTdnU1cnWHrRDtNMt5lTV0Sdtzl929ei7dzTvETi/s+N0AmetPs9jUl7hmNMb9+cZs2apeO/NeP2ZiwQMLMIQtyWruRUa8jvF0BQUDsCvqxF2bzuqMXcVOzwA/vuywmeA51EBwmnJ1PDW0Clzk+PcHtBQkpDDMbIWdQzp1dpKBy63xG8aLoxlwaeVkKqKdiVzU/S6bpIpI1yoiCgIKAg8O4QyGSCAE+XNktKEKRnbA0pRYHKY3ijId40CILh6DBKaq0R5O6tV+McXpZ4uLS5w/XrVncWN+wueOMZxlawk46KjD+XduNgLwbDkaGU0HpQdcShxEMc9gSmG8xv6IUmb3e2WdrCOKLX9aZMtlzUnnTMOUfecIShJbR4VB3BoURjJXZBBo4MLYHWoyojDj1zkhv7bdvvs5WtyKbOTbfw1847sTvp46vBs+li7svbHv1eQgpqUOfr40z7ajOdfdQI2pzU++GkTT89B78pgkasQlgif7zzsfd1lDkEwcjZcRUQNb6E7pfFGRju89f4euRSC7iV/47DMgiT2CfdY1nLHFbC6daC5c+dKZInCAaODiuJ1hxcK7jTerWzZiI9ZGlzsxfBfM+NurNufOIzcZxYKkcKAgoCWQOB90MQeM3xOV0ICF5FZPrbYSdCqRIEEzdn1UG0LFKjJneP7cidF8bzE6gkWhewEbJ3YpPdM2A8yzg7QdB9yZKM9OAMF/mpTg7UgjvF2szkmEvjbor6keqigKf/AOZMHUr7GqUp27Av03ffcPQgrcYZuPhTHXKoBdyLtWHmsaQkwHDxJ+rkUCO4F6PNzGPJEJIE9oYUROPWmEUPnEzg5caO+KhFas90aXj0B+hfSIO6QIgDX8Ox4ZTWaijY4XfuOgjUZSb7iWh8Q9kvB9Tx1Ps7yByC8JxVrbMhuDdj6WMnThYrjFf4wU9EUOek61ant8XVQunpSlrZev3qnN0Il+GSLEEw3WRWHRsJVeemx3bZA9inplqJbfZOm1zqhmvuyrmCgIKAgsD7ReA9EYS3NCpVgmDk/ETzGgDWD613x02OuARzrtK9BTTU2QiCewBr7N7tRAShacYIAmB6uI8x9fMhCmqy+43mX1nn8Pmq1mQTtBRt8T2z1/7JuYd2VpIMDqaH7BtTn3yigDq7H6PlgizJTTzcN4b6+cwNWHb8Rv8r85CYh1cuElZZRPxiHGcd5MvA0W9LoVXnpucOeaMEvFpHkKeArt5cmzImbs+ph06dl147nXpKT5bTIpuAd5u1vEhG7cSXjMS/ekVsbGw6/l8R79AzsZSUzjKFIOgPMaiYBu3nI23DNfLcEtjWLSdqwZOgdSkPpxgvTExENuWBuMkSBON5JlayEQTBm46b5ORD4t6Chuhs9dY9YA32qinXTDlWEFAQUBD4UAj8BwiCxJPlLchmCzgU/aZwRdYAmW7PoZ7Ng6ApOohDdu/yWxIES4EZbrG+WzFEQYfflMu2WAMjV6b4IarcaLlefNdGAAAgAElEQVQy7abVWvAGbq3vRjFRQOc3hcsy/e0Vw3BrPd2KiQg6P6bIEkhPltHcQ0P+vrtlPdAE9vTNh1qdjz67ExME63CMB3VnXbOJfsWmTj6oPZqxRLZedfyu3uRTi/j/dC1t13fCVr7ysQV4Who8KyEzLzuc9N+N5suf2c1K129mEATpzjzq6wRydNjoHOKx526OwainQ9DVYNpVuwvFftP5K69LKl195t5xeiKSJQjSE5a3yGYdRlCJ+E254gg+BSsxsxJbDUUHHVL2b3BCrRwpCCgIZAEEMpkgSI4YBLn71WKn8SUxr1L++KaKhfESk6ra4wWqMzUqsRzp/koCcqotjZGg82d6tPN+/P5Q2wI2GooP/NvpXTCcZFRZrbUB0zVi4UPnxz41XQyPLhAR5Yw6l+7Op4FOwOerrbahDYm78xqgU4lUmuicjQBG7u5ZzGb7GIvhERciohwzK5DuWtZ6EMwzQCxtuoFHFyKIckxplLg7vwE6wYevrAksaibs6Ut+jQfNlz3BFPOIxxYngIFTo8qiFdxoukQWmMlrDg4ojnvJ/uyzezwMR/m2pBaxcphjZgQYOD2mvG1BIT0JTx4R44Q0KTxSDDfPnebUqVPp+D/DtaepCUsqPmZVK9wEHY0WPkwah2F8QfTJ09xOueNvERi/sxd51CLVp15NQniMkdPwd9eQv+M6Uq0GhlOMLmdetMkcN+BJk0XORZX0B7+hsG2hJPfAtTZvgMT9lQHkNK8AqhLQ+U/HWTXj2R9aCI1ZlqY4A/+WexeSYqBcURBQEFAQeN8IZDJBMBH1YzVEjwDWOAO8QXrE6jY+iL4h/Jm4Q5s+e19spkseKwEwR493D7e3bvbHjUQvb4Ov1hot7tt+LTfNEwqMd1jVNg9q88e5VDDb5F//l3/Q5TObTO3nfH/COQPBLjXpbzxbuvigztGS5bbIP/2p0VQQvWi04Kaj4TEc/ZZSWhWawkHMP3iJq5eOsH58ILXaLybS5sGI39IFH3UOWi63LR+sP8XoCiJejRZYpyrGb6GLj5ocLZfbggz1nBpdAdGrEQscsziMXAirbAm86zpvGf1a92LtdWvjG39oMCW0At6Nf7ZNfTTyYPdAKuavzeSTzkg80w3r1L+CIXudsRvSAxY2ckMQ6zD2t0l0aDeJI/LyTArMO7zymoMDS6BVaSkx+JDMS2LO8hU7exewlK/754PZn6LDxsi58V8gCjnonGjKpoHHJ5bRu0J2Pqs1mr/SjEORuPdbBwporN4Sde46DF22md9nfUPjEt5obV4stY8/g1ce5ra5rhujWd7G13pP40v7tTctM02Md1bR1lynBR2lgrelTkzeIbqKaAUBBQEFgZQQyDyCEPMv80LbUim3BkHwoGjDXkzZdc/WaOqJXD+KgZN2O4PgUtIo0XUjlxZ1wa+Q3U1rdVmrPQtTrccyriZyxeu5vWcKnf3y467W4FOiOjXL5sHNw5ca3aZz4IE9sYmba/tSrbB5jr/dBS4g5i5N/dF7cTadiRRxnMSemEVgiVyUajmA4QO7Urt4EWoN2Mh1Ob+QHrF7aBV81OaGRECTozQtx2znln14wywt9gSzAkuQq1RLBgwfSNfaxSlSawAbHYJiOTErkBK5StFywHAGdq1N8SK1GLDxumwqYwJ/BhdAbY6DKNudlZEy9iU95cC4+uR3y0ax+m1o6V+WMnX7s/ZS4q523OZOZFf70PZX2SqF9hkegjuFmv7AYYcXwwHDezgwErW8Dw0r+ZLNgqMKQcxFKf8WTPzLXkqv2BNaFJ25HNV56LVTZr9FQyORvw6la5u6FPc2k0EBXZ7PqdWoCY0b1Mbvi3JUrteR7xYf4p68/FK17hXnV4ZQ/TOtxZMg6PJRrcdM9h2ZTl1RQOtVgM/9m9Eh5Ccn4dDfZs+Uzvjld0et8aFE9ZqUzeOGh28Nuk0/gKNqppqvclNBQEFAQeD9IpB5BAETBr0Bo8nsqpcwGQ0YMuZJTtZyyRBPvN6ISbINAUgmjIYE4uMNSd3NFgkScY+iOXfiOKcuXOeJa5th1s78fIJdV/MFm8wEO4lIVhXnRWMs9y6f5NiJC9yKSekZEy9vn+fU2as8fJ3S8IWR2HuXOXnsBBduxcjGp+VZ3ePyyWOcuHCL5LIyxdzgXAr3zFLin0Rz9sQpLtx6kax8DE+IOn8jyRDC67sXORf91OlVcKr03o4sZS8rJ8lkxJAQT+JiMhL76BqrOvvzfUTSVt6kjyfBUn/MkaUGEuJf8+rVK+L0KZVJOs2Lf0TU6dNcsQegSrHcjrzBi6QqOAVKcTyKPseJ46e4cP3JB8XWqZRypCCgIKAgkDwCmUgQks9Auaog8K4RMEQtpEfwWm5nAiF917oq8hUEFAQUBD4WBBSC8LGUlKJnsggYr65lUOhsjr14S49AstKViwoCCgIKAp8uAgpB+HTL/r9huXkoS+EG/42yVKxQEFAQyFIIKAQhSxWHooyCgIKAgoCCgIJA1kBAIQhZoxwULRQEFAQUBBQEFASyFAIKQchSxaEooyCgIKAgoCCgIJA1EFAIQtYoB0ULBQEFAQWB94eA8RJrhvWhV8+e9JmwjXuuM4Ckp/w9fxj9Q0L4+uuv+br/MGbvf/j+9FNyyhIIKAQhSxSDooSCgILAu0TA8PAovwzsy5zTqS1U8S41yHqyTQ83062QBpWmML13OJePd2pq5Nac+nhX+o5Dz1wZhDOVcvTfRUAhCP/dslUsUxD45BEwPjjML0NaUjaXiJDuJdU/EdhM0UyrYd7jRkAsN4KjSbYDMW88V5NqYReTX2TtE4HpUzZTIQifcukrtisI/McRMN07zZGoF7ze1Zu8YllGnVI8CI4ij91A5y/aEdKyABrBk9o/yXcbNad6wdp2xenrsiOs43nl4D+PgEIQ3riITbx+cJl/t/7B0QfKRPw3hvFDPWiK41HUUbZtOsQdxXv6oUoh/fnGXWfvwvF8N3ETUSmtbp6KNP0/gyiqK8dohSA4UDJEjKBW7228uruKNp+pUX/WhlX3ZN8y/T8MLN2MX+Sb3NmfNt1j78wRjJ65lUsvZc/Y7yu//wkEFILgKEaJ17ePsmnRNCaO/JZvR01k+i+bOXw96R4Jpnvr+dqvIO7mTYK0Zfj++KfbKzE8vsCR84/e2gX5JnLe5BnzPiFPwgfhV9DDstmSpnB/Dsg30XLUB+UgqyDw6swC2pcuTvOZJ3nTBTMVguBamhL3f25O03nmLcuNXJ3TgOyChiLBu7Bvimq6PoN6lUaTMqeKI2rNV5Qt1oSwg/Jt5V3zUs4/VgQUgmDeq+lZBD/3rEI+31r0GD2L1X9sYc3scfQL8iO/Ww7KBI5j+w2XVkS6y/wGOgSxEhPOv0GX5mOtMXK99f8wqKgWwaMes23bTMtvp/v4TeS8yTMyhaTHq2jtLaAtMZTDny6/kyGSRQ9f/cU3xUWyN138Vp4ehSC4lm88O4Nr8e0RW+VPOM2Equ4IugqMOh5vSRy7sRPFum0lSWhCIlExbO/lizZPEL8l52lIlFY5+dgQ+OQJgvTsL0ZUzY5H2VB2PHD1NUs8/TeMBp9p0OZrwqwz8lclnq1f+SCIVZl06RMlCKa7/DGwHlWaTuLwy7eo+inJMdzmr3mDaNt3GTddiyalZ9Krhv4A/Qtr0JYezrHMIAip6ZpenTBw+695DGrbl2VJDE63kP9UQsOpUZTVivj/dN22dfybmacQBBfcjBcIq9uR32Xv7au/B1NaFPCqN4soo4Fjw8vTaO6dFHbNdcp7tS4IT3Uuum9PZutcZzLl6CNE4NMmCNJTtvcuglZbhmGHX6dQfBJPt/WiiEZALDWAv2LtyRLY3iMXajNBuPyJEgQ7FO/oV3qwkEY6FWK9udzJ7GFO/d8M+F/mEYRM0VV6wMJGOlRiPeZmusHvqJDesVjD8RGU1upoOP9emg1VaqooBMEFneeradfgByLlxFt6SnjPQmjUeWi36gRzGpVj2NG02XP8lq74qH3ousXqeXDJSTn9iBHIVIIQf/som5fO5Mfpi9h88hFpV60Pi5zxyo9UdxPQ+U8nWv6iuKplusm8htkQBA9qz7hq68koBMEVpkw/f7aMZm4qdI0Xkeney0wmCGSKrs9Y1swNla4xizLd4EwvnfciMPMIwkD+p1NmMdgLTX9wEDVD9+La5zfdXUWgOWDRJx/58rZnvczDYH/W9VchCK6I/HfOM4kgSDzYPoBGbSay9fglzu/9gS/z58I/LIKU+uUfHkIjlyZVRVRpKfP9iTTIjMSDRY1xEwR0tWZgHW6XEYRzt/l70ShCOgXROXQMiw7eTUbea6J3zmZYryACAjvSM/Q7Ji/8k2uuoQ0vzvPHnDH069aZ7qGjmb/rKq/sYEmx3Dqxl42/TOKbTsPY+NDEqysbmdAriA6DVvJ/buxn7uhRjBk3nvHjxjBqZBgbI63eDdOtnfw0Zgzjxk9g/NixzP7znl0qSC84/8ccxvTrRufuoYyev4urjkydyRIdvb7LmQNbWTV7HF93/I4/nli7+Kbn0RzZ9TvzxwXTKewvy/jl66u7WThxMD2D2tF1wDR2yuM5UpDz+tw6fhjZhlJaFdoSAXw3YQITJoQx7887mFJ4xq6f8clp1v04hF7tA2jTOYTxq0/w2NXJk1GC8DqanbOH0SsogMCOPQn9bjIL/7yGufhS1dWmVJo6vT7Huh9G0qaUFpW2BAHfme2dQNi8PxONvUup1Q87ACn8xl/fzdwRwXRs24Hg8Ws58zyz3TIpZPwWl9NDENJj18sNHfDR+vL1ftcm8S2Uy7KPxnN991xGBHekbYdgxq89Q+KiNnJ5sj9Nf76fjFfGSNTs+ngLKsSa07mWWsfJZr9CELJsRXhrxTKHIBiOM6K0O9V/jLL1rvX8PaAIWp8u/JFlGcIzVrYyR7K70WLF8zSBNJwYyedaFUL2Tmy2eNJsBEGdg6J+TQmespz1v81naKOCaLWfUWvsAZ45vr/xnAirjk/+5sw4co8XD8+wslcZ3N0DWOMYspB4fnQGAaXL03HOfq7cuc7RJZ0pIXpTdfS/WJIZr7JpTBf88mgQtKUZsmoOgVXKUdLHfF6G7w/9X46tHU7d3GpUQjb8R+wiKsamxOvbHF3eg1JiHuqPWMvRO1Z3oPT8KDMCSlO+4xz2X7nD9aNL6FxCxLvqaP516JYUHtOdXUzp4U8BrYDKrRnLnlnTJJxZwcDAcuRQmz8wU9i7pid+5evTacD3DAksg6cgkK3OTK7aPjwpyYm9uIPl83rzhahCLNeD2UuXsnTZSraff8H/SyFvswYxh8OoXaAyA7Ze5nHsc6L+CKWcm45S3+wlRm5GRghC/AnCqvuQv/kMjtx7wcMzK+lVxh33gDWWcklNVzP66dIp9iI7ls+j9xciKrEcPWYvZenSZazcft4WuZ+O+iG3z+XYGL2YFr4V6Lc1mse3txJsDvzzn8YFV+Lk8lzap0YurB3FkMGDGZyO/28XR6QR9JY4R8PhoZQwDzEsSK4xgzTt0p9kUUg7qhUQEVRqvEs1oefUfdj4bOLM/hNnRqIXt8C3Qj+2Rj/m9tZgiovZ8Z92wTLT6NWp1YwJbUkZbw3ZKwQxbGkESZwECacJ8/Om+OB/LQQ4LVjit5qHGLzosFEZYkgLq4/tfuYQBNM99kwbw5IT9qpm5Nz4ioiebfnVfimrIWOM5Ac/EZWQnU7WFj9VDU2351BPVFndv4/Mn30bQdD8j35/yVhQ3EnGVNIhaIrSb7+tG67fT6ivBo82v1obenNOMRvpVPwrttreKenRRroUdKPimFM4XjPpEUuaeiBka8zCu3a2YeTM2AqIgoh38U6svmUkLnoPv267YGtIjET+WB03QcRv8uVE0w8T/hlEpaY/c8PeK5AesbFLQdwqjuGUM1MeLWmKh5CNxgvvpooJpqtMNa/E5hbIWrnHIeFPgvOrEdzzUTP0N67Y4TGcZ0IlEcG9BYk4WUpyXqyghXmIocliHtvNt2uU3DPSQxZ/6YHg3ZFN9nhS6QkrW3mi9mrNqqcyIRkgCPr9ofhqPGjzq5MxxWzsRPGvtjrLKiVdM6ITL1jRwjzE0ITFLganv37YAZL/Goj4rjRuZUdx0jLuZyRqRU8aBs7AvvKw9PQffurTnqA2TWgQMJzN111cW3JxiY4NHJvenubNmtEsHf+tx+11vgOJ5CR/8vqPzpbx7S7J9jTStss8rdWoN2A0WcteMhnQG+wvQPJ5ftRXDRF8V9qNsqNOWr2YxihW9GxI4IzT1nOb/WY0UsPCFPuYJ/b3Ng1ADMeGW+NEUiBxaTyu3M7CCGQOQXA1ULrPkmbe5ApYxX1LTXzG1pBSFKg8hsPp/e64yszsc+MlJlU1EwR32vwqb92Sz0i6/7MlYE7l1pIVlonCsiGGREGKEvd/aYKHoCZvr51WYRYPixZ1nubMPW9vZIy8ePzMNgZo5KJ5uEPzP776ZR/79+1lz+5d7Ni+kVH1vRHUeei9y+4aNXF1Wg1ElUiVsEuJCIBdc+n+Upqbp/CVH+NoAOAl23r7EbLHaavx4iSqihr+99Uv7Nu/j717drNrx3Y2jjK7GNXk6b3LLjL5X+k+P5uD6tza8ruDYACGwwwprkGdvy977GpbJMTzR5ccCKI/P8l9lynJSanRNctK9hk9kRsnE7Y8Qua9MXB4SHE02s8ZeUIWFZMBgmB1c6vJ03wuzuJ7weNnMuNS1DUDOqVIEDJSP5IrKgNHhpZA61GVEYeeJXUrx59gkn9Rmv58FQMSd1cF4lthBEfS2UAkl+NbX5PiuR+xhiG181Gw8U+cshO+RILTsCtR2rc90XN+bitK+n5B8EbZ8JyrWP155rYqie8XwWxMsgOSM7H+/FxalfTli+CNSTdKcibL+JHhCENLaPGoOoJDThdmxuVk5AnTNVZ1KkHuL3qy8MANXv2H+VdGYPkvpH0HBEHi0bbelCnTi8137TXlNcfndCEgeBW2IfEPj530hGXN3BFUIn5TXJcYTaqe8WIYVUQVmiLfcNBCclIiCGBtUFToGi2wCTISOa8RudQqBLcCVOs8ljXHHzrjFGyNnaAtQYtBwxn+7bcMGzrU4rYdNHAgAwaOZb0DOBtBELxT8Xy8YHPXPKi1JRlqY2TSgxW0rT1WRhjMC6U0QidoKdFiEMOHf8u3w4YydMhgBg8ayMABAxm7PjIpEPIryTbSToKgKT7EZY2BBHb0/Ay1WIWwizLfdkpyUmx0UyIITuUMTy+xe+lkBndtiV9hD4tHZ9A/MnaaAYKAMZJ5jXKhVgm4FahG57FrOP5QRjbM2aamq02tNHVKiSBkqH44MZAfGS7+RJ0cZq9OMdrMPCYjUBL3l7Ugx2fd2Grnjq/D6ZYnBy2XP0xKJuRC39Wx6SZbRneh0ee5ye33LXssHrvkM0vZruTTv/FViyfI/L1Qk6/vnhTFSA8X86W7gEqdj76J2bHsGYmHi7+0LLSmzudKomXJ3ujQwMWf6pBDLeBerA0zjyVDCN9IbhoPvYxgWpOC+BSrTdA3K7ng8nqk8bRyO4sikMkEQeLZP+NoWKMPG25m9Rpi3ojED1ElkO3LxaS1WvLLdUFkF9Tk7rzJNpadCkE4NZpyWgGPwLWyYn/N1fAwOlXNi05QIWhyU2vMASxeb+M5JlQ0jz3XYVaa89/tBCEHnf+Qd9tlWQHxB76hqEZDoa/3EYeJqOmNaTbvpmwuuZFzEypaPBF1ZsmvJ5aT6llKDbvNg/AhCIL08izLv65NyfKtGP5zOMdvP+fY92XQaoryxgTBDMLrq4SHdaJqXh2CSkCTuxZjDjx1NqCpEIR065QSQchQ/UipxEw83DeG+vlEBHV2/Eb/a63H0hNWtMyGru5sbtn5vOkGM2rp8Gr7a+K4jRREG+Nf8So2lth0/L+KlxHDFOQ5LhtvsyaoAF7lB7Mvxd5wCnY5hGTeQdy1PSyau4aIh3agkpMdx7U9i5i7JoLUk11jz6K5rIl4KHsnk5P3BtdMD9k3pj75RAF1dj9G/5so+uYNBKbxSPwpfqyTE5/6M7goc6ql8ZRy+yNAIBMJgsTzQ+NoVO8btqXiWstKmJhuLKCRt4Dg5sfki6kRmhds7JQbtaYo/R3xBikThLjwbuRSWz0TFnsTnvHkpf2jEseNPZNp7qtF0BZn8KEEMN1idl0RlToXHTekFTCZPoKA8SzjvhBR5+tB+NMIvq/ThQ2JPrImbs2ui6hSk6vjBtLKNdlyy2oEwXSVn7/MhTZfe353MD4DJ96SICQ8e4Kz+G6wZ3JzfLUC2uKDMRef5S8lgpARnVIiCBmqH8mWlOOi4dZ6uhUTEXR+TDEPjen/ZUhxLV7tN8iCB2NY1coNbdlRqSyzaxeZYF0wTKVClY5/t+bLscWz2gWk+puwNwRfTTZar0q9hiaxK1Wpn8JNA7fWd6OYKKDzm0KiUdBMNl+/N4SCmmwErLYv0pzJGSjiPhgCmUYQDJE/07xiB9betjeERi4vnc12+3ttfElMlhuciufU5JpkV6vJUW8655Md54S4E2Op7C7yv+6bZZ6GlAjCa/Z+XRitTzMW27DQ7+9P9X77ZHOOJZ6uaEk2QUeTXx6ZffIcH1EGrcq8itlMImWecEvNkCRnTxWb50PwTCNq2MT12XXxUGeneosGVB900BlQZ6tu5qGQMuaZGV71mJk0UyRZTF+yNVS6w9z65hiEANbY3dPmhPq/Gfg/DZqiA5F79c2BneHdcqIWKzLhnKwnmZIce6PbcAHyPWSsmCTN23hhIpVEgWxtf3dODcUayKbVFKLfXzJg9X/Rz1eTjqWW9ezvX51+++xMwDy88dTS6xZ0TfjF7v5OQdcM6eQgCA1ZkMjgjNQP15Iy8OhCBFGOeW4Sd+c3QCf48NXWBEjYTo/cGnL32C6rn/GsC/KQDae5ypSfS8TcPMfpU6c4lY7/M9eeZqjHbIj4LoWFktKwy6Zi+oIvjbyIPsnp2/JKLLcxixwbXxB98jQpqml4xIWIKOeURvty8D5fYS5q65/E039+ok/7INo0aUDA8M2kOx7VLsLl1z7NsYuyUJILMh//aeYQBNMtFjbJRZWRB4m+fp3r168TfWoNPar2sEbpS49Y3cYH0TeEPx0VNYuAZ7rHjqHVyKnWUqDxeHbfkiso8eL0z7Qr6sX/2vzsQiAS2NY9F2pNIb7a/MjWgEs8/+d7qvoUImjFNUcAoSXKN28gKx0zEQxc/akWbroKjLXthGK6vYo2+TQIgo4SXy3n7FOrRyPuejgjW7dnrqMLEMefIb5oVCI1ptkXbUoeS+nRagJ91Ai6qkySj/nbk5tus6pNPjTm9R1KfMXys0+tcRFx1wkf2Zr2cy/bUyb/qz/CsJJaVKIfkx36Ac/W0ja7gJAjiN/lnQqLx8ENQZ2f3jtlEXApyYnbTKfsAprCfdnx6BnXDi5jXvgNawOTzDOm6GnUFFVoywzhH0ssqInHR2bSpqgOQfCm3W8vSIiNtUzdkp6spJWXgNqnPevlOiax1LzkbGnyBq7EWXxX+amWG7oKY5097BR0NWRAJ4hjc6fsCJrC9N3xiGfXDrJsXrhl1kn664eLAfFb6OKjJkfL5daAYfScGl0B0asRC8zDWRaipCVnt3AZQYhldWs3tGW+50PvQ5biOghp2WWGIZ3Bl6929qaAOT7I/XMG70+1MriA+z5PX7GzdwFLHIz754NJTs34LV3wUZtjR2xTQvWnGF1BxKvRAsdS5fEnJuFftCk/XzWAdJdVgb5UGHHkrdarsRMEZSXF91kf3k9emUIQpEeL+dJNSOJiFP3tC23oiVw/ioGTduOIW3w/9qUzF7PbfwZ9G5Qip0d2ClVsSLvO7WhYsQSla7Rj+MrTTlbukBjHxo75KNO4BVX+V5J6XULoG1SPL/w68NM/jxL1koyXplDTW0f2Uk3pM2QwvVtWpmiJuoSuuiRz60LsmSX0qZ4PN3OMgpidfAULULRGLxYctwYamW79RmjdUviY1x5QmT9o+SlfP4TVKW6UFMuOXr74NF+atAdutyP2DEv6VCefufwEkez5ClKgaA16LTguC2SzJ3b+xuweSYPSuRHNO1pagvcq0mLK31z97Wtq/s8Ltfm6oMGnZDOmmVuZV3sZ6V8EL7U1vcanFA0n/s2jZOXY5l9Lj9jQ1deShyD6UKHrQk7HSCSf97/opYfsHFqdPKIaryJ+1K5Wido95rNneRfyawQ0nnkoUX8Caxd0p2ZRb4eOOUrUJnj1tURl5rTUyKUpNfHWZadU0z4MGdyblpWLUqJuKKsuyVxOKehKunQK46glnETi0Yau+IrWsvCp0JWFp2Mc3qO06odTZ/lRLCdmBVIiVylaDhjOwK61KV6kFgM2XrdNezPHG4i4t17l2MUP6R4LGupwb7oEu4NELvF9HqdIEEjDLtIffPlqTyhFdeZ3Sk0e+8yj92lkuvJ6xZ7Qopb4JZU6D712yjsyNgGxJ5gVWIJcpVoyYPhAutYuTpFaA9h43TZ8Kt1nWYscfNZtq8PD9jq8G3lytGT5W6zcqRCEdBXgR5koUwgCmDAaXf3RyV3L+hgZX97n6vmTnDx3lbvP41NoNMx2SLy4ecPSiEqvHxJ19gyRD+McH/Oklhp5eS+SMydOcCbyPrH2kZgkCY28vHuZMyfPcOVujHOmgyVLAwnxetmcbiP6+AQMrtDLZErPo4lyjbiX3bcfGl/e5fKZk5y5cpeY1MIxHA/oidcbsUwvlyRMRj0JehOSIYEEg/26CaM+Hr3FVrOudt2d6TEmL8eeDaYYbl28zF05YGk8E/84mnOnz3PtSbytPAw8vXaOi7eseCbR0aJzKiCalTG+5F7kGU6cOEPk/djk60VyutoMSUsnh72YiLl1kct3U8iDVOqHU0iSI2PsPS6fPMaJC7eIkY3ugJGz477A/fORtjUSzCl4BKcAACAASURBVENE5l06Pag5zb7wWRJx7+2CfYihQQp7MaRoVwaDL42xj7i2qjP+30e8N9synpGR2EfXWNXZn+8jUnpJjcTeu8zJYye4cCvxVvXSkxW0zKaj7uxbjvprujGDWjov2v765oGMVs+FD8oQQ8ZLNKs/kUkEIaubqeinIKAgkBICpttLaZm3MuPP2oa1/h1KmcJd2fSh3Qfm6JxjwymlFakxNfXhtCS2ZTT40hDFwh7BshiqJBKzxAVD1EJ6BK/FEeqVAa30/w6huNaL9htknq+YVbRy01J21KnEnZEMyH2xJoBs6uypzqrKgDglaRZCQCEIWagwFFUUBD4MAibubhtC/ZqBDBrZn4DG3fnlnH1Brw+jkT1X6cFSmnupca/xY8Yi8TMSfGm8ytpBocw+9iIVD6Bdow/3a7y6lkGhszn2Ig1vVwoqJmzvQW5NbnrIt2WOX0eQh4Yi3xxM17LKSURL91jR2hx4XJmw84lcU0mSKhc+PgQUgvDxlZmisYLAO0FASnjG7VtPZMGK7ySbDArVc/2P4TQumpvSHWewJ/J5+nq6GQq+NGJIbZwugxq/s+RGQ6rDiWnlq/+rH77anHQLl8UvxK6mtZt5w7rj6cPVkYmJ2Fv/sjTUj3wFahK8/JwjrsGRRDn46BFQCMJHX4SKAQoCnwACphiuHdnBut//di7olJrZlsWesm7wZWqqv6t7lngD0Z3Wq5wzNaR7C2ioc6fpEvtMrHTmLj3i+Obf2Pr3ZZ7IZg+n82kl2UeCgEIQPpKCUtRUEFAQyAgCWTv4MiOWZFpay+Jp7nw+0raRkyUedRBFPWoyLSrFqOlMy14R9PEhoBCEj6/MFI0VBBQE0oFAVg6+TIf67yCJidtLW5K38nis8ahx/Du0DIW7bvrg01nfgbGKyExAQCEImQCiIkJBQEEgKyKQdYMvPxhaprtsG1KfmoGDGNk/gMbdfyGLxKN+MEiUjFNGQCEIKWOj3FEQUBD4DyCQNYMvPySwEgnPbnPriSxY8UOqo+SdZRFQCEKWLRpFMQUBBQEFAQUBBYEPh4BCED4c9krOCgIKAgoCCgIKAlkWAYUgyIom/sEFDm5Zyc/zFrJ6WwR3ZAuOyZIphx8CAUnP85un2bt+J+ff16Z7xljuXfybP/44ylssVf8h0PqP5Smhf/mQa6cPsHnNHqKUaXX/sfJVzMmqCCgEwVwy0hP+mdSMwh7elG/bnfoFNJYNiNxLBhOeBZabzaqV533pZTg5k8Dyn1k2bVLn680uy8ZG7zJ3E9dW96GarweCSsCj2VIev9nide9SyQzKNnH3+Ha2bd/Bzl272bN7Fzt3bGPbvgupbsqVWiamu8fZvm07O3buYvee3ezauYNt2/Zx4ZkNLMMtjoaHs91yfw+7dmxj58kHqYl0uWfi5tIgiuYQESybgqmSbi3u8oRyqiCgIJB5CCgEASOR8xuTS61CrDCWMwYj5yZUsu6apinOkMMpbYqSeYWQRJI+ik3jBjJ63WXeeVuYJPOsecF0bTo1RYFsLZbz5D011nEb2uMliFSdfNmxdfcHQ8f0jLPHLvPmCyAbuP7XcmaPbEVRUUDQFqBu6GR+3nSKp2+Ip+H6XyyfPZJWRc0NuJYCdUOZ/PMmTtkFxl9l97QgimvVeJdsQp9RM9lw+lmGITTeWURTL9sOpl4d2Ki8FBnGUHlAQeBNEFAIwovNdMmjRqXSUKjfX9b1yOMi2RQ2iEE/7HqjTVHepCDkzxjPT6CSqEL7+UhOuPKT2Cvs+nU7F1++4VddntFHdKz/ZyBFNSJVJ72vxtrE9Rm1EDWF6PfXB/RpS7FEbf+Rr6p+RrZGC7j/tsX+ehMdswvo6s3NpLr9mo0dsyPo6jHXdQchwy029vaj5jebuPY2AfP6/YT6mr16KgTvTmxWCMJH9OYqqn7MCHzyBCF+W3dyq1WoVCKVJpz/8D1Fc22SXnBp+6+En3/usnmMxP1FTXDXlmTYUVfm8DFXw7R0l7gztx6ixvc9NtaxbGifA7VnAGuep6Xfu7hv4tauMDrXLEPJIrktwyu6JovfeqjDeHYcFUQtJYYezuDa+ynYaDzD2PIi2hJDSeRse3WehUHVaDrlCM/fltToD9C/kEIQUigB5bKCwDtDIBMJQgJ3j21k4bRJhE1bzI7LMS6N2zuz4S0ES9yb3wCdykoQas+86dgn/S2EvsNHXxPeLTeaHO1Z//IdZpPlRMdZ7FZ7tWGtcxn5d6ulZbtgDWK1H4j8IKvQSjy7fIIrz4xIj36hiU6FrunSNx4OsIP1bFUrsgmeBK7JJCCfrqClh4Bnm7XYJUpPDjK+URWCFl3gtT3jt/lVCMLboKc8qyDwxghkDkGQHrFrkD91+63m8MWLHJwdiK93FSac/ICu2TQgkZ4dYfGoAXSpnheNhSBoyFe9PT179qRX8DT22wOtLHLiubl/ISN7tqZBjSpUqupP46BQJv16ksfyHU5fP+ByxF42/PIDw7q3oH7Hn7lsNHJn90Q61q1Js8EbuS5P76pjQhQ7509hzNBQevcey9Y79pbJxP1/VzLrh8E0KqBBU6AR/cdPYOLk3znn6rqNu82hFWEM6NmBwIAufLvspLMHJ8VwZuNswkYOIqRXX2b8Y2YZRp6cXEtY/650DP6Rffdseb6OInzaQL4K6s7kvQ/TJHvSi0tsn/s9IV3a0CqwE6FhG7hgHwaJv8K2uZMZNSSUPl/P5nAsmB79w7yBnekwcDWXk1QTAw+OrWB8SAfadhrC0oh/GFtBRKw5jWg7JDLs4m4fYkXYAHp2CCSgy7csO2n3vMRzZdtcJo8aQmifr5ltzZh/5g2kc4eBrE6asUOqdWMb2bCT484HOIhfRzt3AV2zpTx9q+z1/DOoKBqxEhMzaWte/cEBFNGIVA67YPG+6W9sJMTPj97rrr/Z9sFm+wz3ObxqCsN6BlDXrwr+jWpTyh6DkNwQQ/xN9i8cSc/WDahRpRJV/RsTFDqJX08+zhoewbcqM+VhBYEPh0CmEATT1WnU8P6SxfaI/4Q/CSngRq0ZN7Jsj1x6+i+LJ4ymfyNfB0Eo/OVARo8Zy/ipG7lon+JovMmGvuXwFAQ8/Yay5dRp9i8JoYr5gyVoyVs/jEM2H6rh1Hw61/TF3RZxLfh0ZsnmryntZg2wUokVmZDah9l4jxNbF9KjjBaVWJuZN+2tYTyRu1ewOCyQwhoNhQImsnjpUpZvjEi0hrrhxgZCKhWmxtAtXI01ErM7hCKiL8F77MbEcePfjfwQ4ItGyEHH9Tf5e2JDipWoTat6xXEXRCpOOI/+8R6G1KpA3RY18RUFxKqTuJQisZGIOTad5sVLETB1D5GPnxG9rgfFRDU5v1xobdANd4nYMp+upbQI7i1YdiWckPK58dIJqMSaTJO3+tJLTs0JoFTZIKZvP8OViCV0qVaYAqKWYoMPuTQ6Bm5sCKFS4RoM3XKVWGMMu0OKIPoGYzXZwN2ILczvWgqt4E6LZVcIDylPbi8dgkqk5rToFOvny/VBZFebe9rpG18wXljLqCGDGTw4Hf/fLibCXiTpeffj1xHkkQkEQbrD3Po61Lm6EZ5a/nH/MrpKDjxL9GFLqtM3TNyaXQdRnZvu217z9MhPtCrjx8BdD1LENS1zTXd3MLSaDxrRl5aTw4m4cIZdU1tTWLQFKboQBOPNDfQt54kgeOI3dAunTu9nSUgVvAQVgjYv9cMOOQlyWpkr9xUEFAQSIZApBAHjfc4ci3ZEWMefGkflbKUY/HdqX6FEenygExNXp9VAtA0x1JnlOsRgJHJWPbzNDb6mMKH77dFRJm7MbYCnhQhoKNh1szOyPmE/obbxUpXah9Idwlg+sxd+eTz5rM4UTthFpGSx8TKTq4poSw3nmEuYQcxv7fBW+9BpUzILAcQdY0xlT3wazSfa1pjHhncnr8aboHXy8Qg9B78pgkZXjR4hdanafTWRr8E6Nq3hf72mMzYwiBknX5ovMu4LEU3RQfyTpJdvNcB0exVt8rlTfsRRHKVtusnsujoEsQphdmZhjOQHPxGxfAijOjViyL4nPD66hMmLDsnG1V9zfJI/OQu0ZdUtOyMxcXVqDUQhBx02JI7hjzs2hsqePjSaH23rKcYS3j0vGu8gnCYbifzBD1EsT8ioTjQaso8nj4+yZPIiDqXY+Ok5NLgYGrEaP6ZzlzvDsem0b96MZs3S8d96HHsTm5JSbbBezyyCEL+TXnnUiHVm4eCeyeQsPfmVtj5qBHd/pqdqv3noJxdqdQ6KVypNblFAW2ooh9Oq48nkabmUcJrJNbwRVGry9wjHUWtTGmIwRjKrnjm9Ck3hUJyv5w3mNvC0XFdpCtJ185M0PWApqaRcVxD4lBHIHIJgR9DwlItbxtOsbDVCfo9yTtGTnrE1pBQFKo/hcAoNjV3E+/1NgyAYjjKspNYaPe3emtUxTu2kh0tpbnd7utVl1g1bb98ctFVBtDxj9hiMP2dv6JzPpnr0bCWtsqnJ3S3cZfxWz8EB5oa9HnNco8UxcWN+Q7w0eem+zdryxEWvo3eZbOSqPYljcj5husb0miKCRqRQ0Gpu2tTT7w/FVyOgy+HH8IMvrB9UW/S4WHc2t+zOjETKx7Kzjy8az6YsThRer2dvSEE06nz02W0bA4n5jXbeajyKl6P5mGNOMiGTl3BiNF+4Z6fJwluyHqiJaz/5I+rqMluuhOkG8xt6ocnbHavJcUSv602ZbLmoPekYTpNj+K2dN2qP4pRrPoZjDhYjy9j10HSNn/xFNIX7cyAr1NdMIgjGS5OoKmoo8s1BF0+MKwASLyL/4a9zj1J30RvPMaGimUD2YMm2+fQslw1B8KL29MtvEAApcW9ZS3JYSLcbLZbLPDcpEATD0WGU1JrjhwTcW6/G+XpKPFza3OJFMN9zqzsL++vpaqlyriCgIJAyAplIEExc3zCS3t2CaFgmFzlKBvDTUXug4muOz+lCQPAqIjPYXqasembcSZ0gmG7Ooo5o/gCpUOfuwXb5eL/xPBMq2YiAkJ1Om2zdJnOv204QdF+yJIOT9hP2hlBQ40bjRQ8S93pMUUytLqIt/R0RLp4FTFH8WF1E8PRnwJypDG1fg9JlG9J3+m5uuPbmXvxKW28BsfQ37HXEWZiInlYTUZ2DutMvOBoPU9RUqotaPh95IvkP/suNdPRRI9ae6fIB1nOgfyE06gKE7LWCpj80iGIaAW3REPY4v+TOQpSe8ntQLjSfdWGz/L4pkqk13JLYbYr6keqigKf/AOZMHUr7GqUp27Av03ffcBJTs3T9IQYV0yBoixKSbMZOFRxHMb8TlF2NV6Az8M5x70McZBJBePl7EN6CB82XP01ct97UpueraZ1NwDNgjSVA0XBxKv6eAmqfpizKaIssPWVlK1uvX52TbuGyly1ZgmDi5qw6Nu+fmtw9tiN7AvtUYcvUyOydsL+eb2qq8pyCwKeIQCYSBBl8+ktMr+ONWGIIh+RvrSxJ1jhMnSAYz0+0rEdgnX/dkU3y3qd0jwUNdVZPgeBOwBqbzzgRQWiaQYJg5GJYZUTxC8addWFSL9YS6KUmZ9ctLp4F4Pkqy4daW7QF389ey5/nHiZuJGVg6w/0p5BGxC/R4j/PWRPghdqnIxsdrm+JJ8tbkE3j62jkZWIsh4aj31JKqyZ3zx2JPs7winVBnrK58bY1BdR5af97CgGPMetp76PGo/ky53ANELMrmMIaNbm+2prI7uerWpNN0FK0xffMXvsn5x66MiGrtqbrM6glqsnb/vd0L5dsXnPhfxqRaj9GyTwZrta7nBvjefUqltjYdPy/ik+9Z+4imkwhCAaODS+NVluO0adcGaZrhuk7t5I+kYrjz9nsiePoiHKIgpq8QWsztmaD8QIT5YRbvthBsgTByPmJlWwEQcC746ZEXinp3gIa6qzkXnAPwP56ps8yJZWCgIKAGYF3QxCwuYXF5NzhWQn41AmC9GQ5LbLZAwz9mHJF1mibbjOnns2DoCnKoEM2X/TbEATpCcuae6DJ35fdLu2d/q9+FNLoaDj/bpLen/HKFPxEFW4tVzqmmiWPspHLk/0QNcWc+poTmoNKC2pwa7SAe44563Fs75kHjU8HNjgGgxNLTdjTl3xqNfn67E5MEGxDMx51Z3HNMjQRy4YOOdDk6sIfTt9/ImHGC2YypqZgyF6HB4O4CEZ9oUMQ3Gj0832Z3UauTPFDVLnRcqV9cl0icY6T2A0dyKHJRZeUMnaktB/YvCmawvTPwPhCwtav8LEFp5oJZar/bs1ZnpEFBTODIEgP+aWJG4JPJ5ILYZGebGf8lP2yoRk7Hin9mtemqI9OnZOuW5zMWXqxi+D/aVBpCtEjPAOeCvn7pNJRf+4dZ3knSxBsBNaGueg3hcSv5xzq2bx/5hga++uZkjXKdQUBBYGkCLwjgmDk/IRKiDk7s9neIzW+JOZVsgPZSbV6b1eMXJpU1RGkWH2qS49Rus/KgP/P3nsARXW9//97d/cuoKLEElSS2DGWmCixR40ae0XFFkssGLuxoyhqEnvvXWNiiEZjb6PxY4miooMNUSeK6AgqP0G+lIH97955/efusstdOgaNmsvMDree8zzvc+457/M8zznnPbRygy8YqDdHEfWelBaMqCs/HHs8pvECk6pY4xY0hqasyMsuP8kHGVBSZx1Fm2OJemJjCamBdvrKjD9vxBx5mgNBtql8IEUsoYlBg/jpdK4qOIwp4iCrd4SmjValZ2xoUwBtuo7adMWfT0SRT/wvpz1rPMc4Tz3OzVfxOP4KO3dfz+BmMF6cRBW9gFOLNQ6j/oQTwyjvXJFvj6T6CoxnGF3BmlZWcBgv+VFNr6VEv32pZCOO05Pq4VnGHZ2+CpMuJPP0fnhqIKxExBJ5/QqRT6dbp9dZq4yJiIOr2WH3Yxk5M7oCeufmrMoq4wx1LTVmwbUTP8WYiQ45R2gWpEb5qhR7jyuXLnLxYi5+wXd4lpdPIXYTbZ0EDE1XZGIFMRFz+wKXwnMQMuWwhQSKdTKZKipFc3ioF502Kqw7xgQSszU0JLG3b3G0Yj3m3lUqI/HoJ2/e1wqIFYdyNAtyqcTOemzk4mTZ+iCTK4GCzVcSYSOrKScY+mHqQknOHdia2qZIjzbS/j15FVQNgqEecxSzYZKO+vKBTk5LR/nhfzpYFzLmrV5REVARyAyBfCEI0rMzrJ6xktO2NdjjTvFd1UJUGXPSOiKRotjc0Q3RYxCH3iiXQww7epSwEgA5cvrr3YpAJytcptvr6eiht0ZKe3Rh6z251TTxYFMnSmjlhqkSA/9QNKwvdtKjuLXR0ug/ZkKGtZIzK4bUvK4G8Jmow6PnEtYNaUe/rXdTTdwJBHYtjCB+QcCBTQyq35QZF9NGbRjP8l0lvWWmhffSE1wPu86Z7f50qN+F1aGKKLukAxYCInf6af2lxMPlTXHSlWaQonCkB4tpbNBRqutsFnVvwNeBT9JGdDYVkv7HyAp6BNdmLE8Nizc9PsDwGiVpMPOC3SVgvjuXeqJIzYDraQTElobtf8x2usrxDJaRYDyXl3WiVtv5zOn2HlpDI/w2jKRR64XYguqt7g05et2bpSeuE3b9DNv9O1C/y2rsKpvvWoINxZoB2UzTtAmQ+j8VS72nLxs3fcsXzQIIypfVftLlk4fThBPDqaDXoK8wkv/ZOGPq+/H7+lNKrofOHzPyaNbWFOPlqdQQBZy+Wu0wNVZKvMf+KY1xL+rNL7a4wNi9DCwjondvw2pFp+sgcvJfjKusR+vWi13pZJJjYubUc0YQDHw89LADeXRII92J9HAbXUvprLMPtMVoOHodO35ewNBmFXDVp1rytG7UG7mR0+FyQ2Li9vqOeFju6fDoshXr5/mATZ3k71rAUGkgf6RV9nQ5qqcqAioC2SGQPwTh0R7GNPyIMg0GMG3mRPo0a0zXgAPct49AUgjdPonhMw4QoRxsZCfZK75nur6SHl4fUEBpFtYW5MPP+7AuTDEMl+Pcwg8yq7sXJZ216NwqULtuFUo4ueBRpzdzjj9O7fTM3Ns6gM8/LITWnqaAWMyTxpMP2zvL7NRKPjSQUloBbeEqfL0xVGG2j+Vn78KWnQULlm/HrD/TT9uSiDowmpry1DR5NKUrgmcbP/bcV5ADuTm1uCIMeM1UdtTJ7P/mfXRunfjJRvDktWrOWCPEBZcytJp1Kou55BLPjk+lcUknCpRrTMc29ahS+Qu+3XrdwVSdGNgVV30FRjusxZseiXjO+H9OEZ0TRd09+Nh7MRdexFj1FnS4fTKQX/5WlIsUxYHRNXHTyh2HgK6IJ2389uCgcmIgXV3zuKxwyjGGyNNUBWc8mvpx6JEiz/Qiv+Jz0631fPPlp3gUsJarRhApWqkeracfs9en+IO+lJXXk9BYrS/pRTLf3c6Yzl9Spbho7XgFJ97zKEu5cuUoV+4DijrLaQu4dd6G3esRd4yJn5dAFKxrRTikaQrjpxGdaFy5qHW0LxTgw9pt6LXoXJpriGT29CmaankryEcNh/ObfdEvh9QynMSHbGRQ7eLo5W9IMOD+eR/mHznDnC9EBH0hSn1cj5ZdBzH32NPUd1MIPziL7l4lcdbqcKtQm7pVSuDk4kGd3nM4/vjfK78MyqkXVATeMgTyhSDYdDbHP+bv+09JspkGbTfexP+SkaSkFExmKXVkLGE2GUlOSsKYhfxSYhS3rwRx/uJV7j7NaAqRjMkkJRstaVpUlsyYLNdy2UiZY/n7ylXux2Z83iybsEPuEWMnXRlBNb8IJ+TiZcIiEzKO9uXHpTjCr93iiSNvwPgsjKt/x6YLyEsg4sY17j3PJkObCElPuX05iItX7xOTUXRIjiT0xsNcmHlNPL97hZC70fbOxhzzNyHXI3iRWbqYeREewsXLYUQmZFZoyUSG3uChwthiEzm7/wkR17hy+6mCoGX39Cu8Z6mjyRhNZmt5ZlGfTHFR3NnUnXoTzmUUxmwkKdlazzPeBMlswpicRFKGSp/MtRn1qTvzRrrXzBiTkklRypSSTFKKI/M3GVMwy0UimUhJSs7acpQudetpElG3LnHppi3YViIuPJS/s6v8SCRG3eZK0HkuXr1LJp9npjmpF1UEVASyRiBfCULW2ah3VARUBF4ZAsZbrOgzkK0Z1sd4+Rzl/RQmNO/EiqxcDC+ftPqmioCKwFuCgEoQ3pKCUsVUEcgUAVMYW0f4svCv1MWtMn0obxdNoVsY1nMYKxSBsHlLQX1aRUBF4F1AQCUI70Ipqjr8hxEwYczgHvgPw6GqriKgIpBvCKgEId+gVBNSEVARUBFQEVAReHcQUAnCu1OWqiYqAioCKgIqAioC+YaAShDyDUo1IRUBFQEVARUBFYF3BwGVILw7ZalqoiKgIqAi8MYhYLq+hTHf9KNv32+Y9sfDdNOpQXr2J0vHfMugQYMZPHgw345ZyFF1cas3ohxVgvBGFIMqhIqAisC7gYCRyLOrGD5gEZdysYTIu6FzTlqYidzR27L0te7D/ux9nsmaJab7LGrsyqfj/ke045IaOSWu3n+FCKgE4RWCqyatIqAi8F9BwMTj06sY1aYKRUUB/ccTyMMq6+88SObbs6kjb54liFQdfzbjommmm8yq+zkB1zJdEe2dx+dNVVAlCG9qyahyqQioCLxFCJh5eOkMt2IS2N//fcQqk8inXbXfIgyyFjXu1+580nkQbeS9Ngo2YK5y6035tZitdC4/gAMZF6jNOlH1zitHQCUIrxzivGdgTnjMjVO72Hn2ceZLJuc9SfUNFQEVgVwikHj3MCv8xzH9t1u5fEP5WAonR5TFUHWyShDssBg5N74+/f+IJ2JTR4prtRTvuEmxtTyknByOZ0vlJnL2lzE/PMz88ZOZv+s6LzLxTqQ9qR7lNwIqQchvRDOkJ5EQfpbfVs5m+sTv+G7SdOas2sHpu7EZ16c3P2T7YC9KO8ub7+ipPOF8hi2WMySvXlARUBHIJwTiCV7WBc/yrZh/4WVXplQJQobCkB6xvFULlsj7d5vCWNSkMILuIwbut+0+aubuvEZ8Ovli1u1d4i229KpCueYBnHiqsoQMGL+iCypBeEXAyslK0edY3rcm7h716TN5AZt3/s6WhVMZ4u1FSaciVO4wlT1/p9s5CYmIpU0wCCKfTgvJSCJeobxq0ioC/2UE4o8NpbxYmBarH2SItM89LipByIBV0j4G1v+OM6lBm8mXplHLWcBQfRLnLVuFxxHoU47eu3LYVS12D/089JTw3qbYrj5DbuqFfERAJQj5CKYyKSn6GONrFcalii97H6cPy5V4diqAJsV16N2bsyDY8cNI2tULN0Gk1gzltszK1NXjdwkBY/gxlozoxIB19/5Bx+SISL6kaQzn2JIRdBqwjnvpq7Bjdu/AmZGLk6qgF+sx9+4/UVYlCOkrg+lqAF90+5kX9hvx/DnSE1EoRKMFtzAZ/2JstaYsfpCTZSCeX7wLoi36NXvUWAU7mq/yQCUIrwJd6Rl7+n+EXl+ZMacTsshB4tkf/fhIJyBWGsaxuLTHkvf0oahWJgg3VAtCGizv6JHE4xVNMWhEGi1+kE8xJ/mTpvR4BU0NGsRGi8mx7X7rS8fI+fGe6A1fsvRhTh1VdsqqBCE9Os83d6bJ96EO5Fd6tpu+H+jQlujMpqBFNK06hrM5TgtN4veebmjdevK7xfKQPif1PL8ReDUEwRTOwWXbuZpjgee3Om9GeqabP1DbScBQbw7Z7pZrvseSLwsgCC40mBdm/4BUgvBmlOPrkiJ6XUucNAaarYzMJ4IA+ZJm9DpaOmkwNFv5HzDp5iNBGF4GgzqLIfXzSeHEiLr4Hk4/5DcTsamDJWDRzd2d97tsV1gYsvryVIKQFTKv6vorIAgJXAiog2sRH3b8J1meieszaiFaggyDsg66sZSoxOOVzXASBAz152GzbKYRhCuE/7mSSYN88O7uGAH3MQAAIABJREFUi9/KE0RkIF0J3N63kDH9vGnfoRt9fccxc8Uh7qQLbZBiQti5yI8hvbvzte9klu4PI95Wq6Q47gcdJnDVDIb6jCEw0kz8zUCm9fOm64iNXI2P5OjiyUzym4q//1T8Jk0kIDDUat0w32ffXD/8pvozzX8KUxYe4qHNQivFELJzEX5DetP9a18mL91PmD1TW+ZZ/Tfz5PxP/Dj2G7zbtqNL/3Es3BuKxdAiPebIoklM8puCv78/q089T+1YX3B2zWQmTp7CVH9/FhxI9SXnpF+iRNz9IA4HrmLGUB/GBEZijr9J4LR+eHcdwcarChdQtjqZeX77DPt/XsrUgT4EHJPfSyDswAqmj+yLd+eeDJu9j7SwkwSu/PI9EztWQq/RU6H9OKZNm8a0gCUcemADMXN8Em7vY+GYfni370C3vr6Mm7mCQ5ZCz02aJp5e+oUfRvWjS/uOdB/kz+agJw7WqoQrv/D9xI5U0mvQV2jPOFmuaQEsOaTwz2eLReZyp11N4u6BxYwf2I1OXQfivzWYzNbPSXv+VR/lhiDkRuYX/NrVDb3HYI6m7xNftQr/Svo5YGK6wcx6LVj+KBOrjOkWCxu7ImhE6s65Yx8gZa2GShCyxubV3MlngiARc3wC3rUqIhbpzs7/JEGIZmNbFwSNE603PM+x1IxBE/lYr0EonEaorARBS5GyXrQYOIv127exdHRTSuv1FK8/hePRto8tiaCA2riVbMW8Mw+JiQxmY7/KODu3Z4vdZSHx/Ow82ntWo9uio9x8cJeza7pTQXSl1uRT1g7XFMZvfj3wKqFD0HsyatMiOtSsSkU3+bwyE87/HxF/bWXsF8XQagQK1BvP/luxqZ1yAuFn19OnkkiJxuPZevYBcrFLz88yr70n1bot4ujNB9w9u4buFURca03mlF22LOCRnnA8oDnlPLuw6MRtnkaHEfiNJ6K2CA3nXsNEMg/ObWW4VyEEjYEmSyJSZTERdWUn/s1KoNWIVJ8SbO30ctQvibDf/OjhVQKdoMdz1CYWdahJ1YpulnN5Non8l7NOyQRvGE6HqkUs+deddZgtfb2o1tiHYRNG0aFyQQShAA3n26xFcVzbu54l/T9B1IhU7bOQtWvXsm7jHkJibGWcEaOkoABqu5Wk1bwzPIyJJHhjPyo7O9PeUug5pRnL6YAGlPpsGLtuPCHu+S12+lbFyVCJoYdj7ZnFXdvL+iX9+UTUIFbtw8K1a1m7biN7QqzR/TljYU8qkwMTt1e3xqP6EHbdfkL4roGW4MB6s686kJSML5q4unUSo0aOZGQuft+tPpdxQZ6MiaZeMXJ6dAWLi2FZZp0ZOcuccmElgzp/TilRQKN1pVLzvvx45GmWOb79N7LDJJ6Lm/3wbVMZV11hqnuPYe25tCgEm+7JlwLwci3PyFPpRjS2Bxz+J7FLdjEU6krgf7JvcQDjtZzkK0GQnu5hWJcZnNrYGZf/KkEwhfK9l4hGKIxPLkwo5vBFNJJXGDM0Y2WUtVOwEgQdZYYcIy2CIZELfp9iEHSUHXLUOvpPOYqvhw6Xjj9ZO3q5ysQG4lO+F7tSPyApKpAepZ2o4XfR0nFbapUUxZoWLggFmrFCnnpk+TMRPKU6oiDiWt6HzfdNJN4+yE9/XMXWV5lCra4T0WsmN5QLniWfZMSnLVj+d+qoV4oisEdpnGr4cdH+IUtErWmBi1CAZisiUvPM7J+ZiM0dKSGWYcjRNO2NFyZSRS9QsN0mrLTLTPiiRogaAy3WPHUwzcdv64iLIOI166aiw8mFfsFTqC4KiK7l8dl8H1PibQ7+9AdXZQDyoFPyoYGU1Ao4u9fFd9tNexkaQ6bxqSjg3HpDqg5W/WM2tLa4GJqvfuKgR2boQApHfT3QuXTkJzvRiiXQpzy9bIUurzuTRZpS5Gq+chFw7fabvfOUnm6kbUEthdpt4pmtOsiZx2ygtexiaL6aJ8rrecAiUx2M5xjn6USVSResFjbTLTb0/ZIO8y7lYHEz8tecLrRq2ZKWufi1m3o47bvIVBDlxQR2dpf92z3YmVbt0h7IjcySiRSjCbMFKwmzMQVj9oagtPTfxqMcMLHqL4ORHRZm4p48tX8j2cNg5K+x1jiRzElc9m+rd/OOQP4RBOkJu7/txo+Xk0n8xduRIEjR7BpUiVKf+XE6N0Qx73q8OW+YrjOjlkwQnOn4U872dOnRcksgmMapDRtSpwWnuRgcgxSlR6to7iKgfb8f++SO13ie8Z56tCVasTjE1luYiHkSjdW6aeKa7O7QlaHXqiMcPXKYgwf2s3dPIJNk0562BP332+ygZsJm17GMZGsGZDF7QnrE2lauCPpq+CkWmn/xR3+8Bh20uyxM12ZQS9RRptcqjhw9wuGDB9i/dw+BkxrjKmgp0X9/1uVluoRfNT268qM4rXSnSM8IOXSQS5G2ixJP17TAoHGi1fpoh/SSdnaniCBSZ7ZtpC7fzlk/c5h1OVixZgDXlQQIyItOxtOjKK/TUnLAwdRySBUvaSc9igiI9eZyR9FxZNWZOyhlP0k1hWtL0GpxiL0DNMU8IdpWlNkQBFJCCZwZwPpz0WlkxHiaUeV16D+e6Lg8cBYEIS9Y2MVWHhjPMLqCHpda4/mf3RqmfOD1HktJjzi3ZRQN3EvTbO5FO3FykOK1ypxCyOK2VPT4hIGBDx3EcDhJCWFx24p4fDKQQLtfz+EJy0lKyGLaVvTgk4GBae6/jI/l/cprxcQqnvnOJnwqFOOTvis4/nd8LtwSeVdLfSMNgXwiCBJRvw/G+/vLlgYxKT1BIIHzi3rQfuAmQtM1vGmivCNH0lPWtXS2+NUcR7CZ62e6FkBNUYPuo6GcSCVPWREEGyHQGJpiZdAmQpc0pahWg+BUis+7T2HL+ci0UZi8QElTA4K+Aq1HjGXsd98xZvRoi4l2xPDhDBs+he32AkntQAXXbC0fMTt6UkKrp+Lo01jElR6zoVMDptgJg8Sj5U0xCHoqtB7B2LHf8d2Y0YweNZKRI4YzfNhwpmwPzRwMeazxYDGNDRoMTZZiN25k+vRLEoRs9LMSBAFXnx1p1hZL3nnTyUoQdJQfdTqtLOR0kvfSt7gWmYAol5zPG0EAU+gSmhbVohGcKPV5d6ZsOY+dN6VilXOaRp5dP8DamSPp2caLD10EdGVHcFJJ4DMlCHnDItOiw8i1uQ0pIltZynVk/l8KspL5C6/sqvne70zu0ZSPixXD67uDpBrxMsnvNcosRbL6K7kN0eI+4GAmslgvWaxB8qJqWncGHFSwQ4c3JCJXf4WzoEHrPoAsH3N4J7cnrxEThUgvzs2meWk3yjXwZujGq4o76mF+I5AvBEGK3ME3baemLnoBGQlCfov9Jqdn4uYsL0TZV//Vah4rTbOZiP3iF28KC1qKdf8Nmwc4a4JwkclV9QguHdhqd+clELY7AJ9a72MQNAi6YtT3O241FZuuMK2GiEZsyIIcJ7LbCEIRumcXPJJ0nKFldeg+GMyRRDDfmkOzlksU8+RNXJlWw2KJaLgg7/P6TSHTqCH7vRsuUKSZCXC8LEHIWj8bQSjSfWc6gpA3nV41QZDRSAjbTYBPLd43CGgEHcXq+3Fc4R/ImiBIvLi8nsENKlKt7ViW7z5P+PO/mFBZn0uCkDcsMis5yzVzJEf8GuMuCmgLezH5lK32Z/mG5YYpKZ74uDjicvGLT8r9aMQUvgXvUoWoNvIIWRo1XlLm7DXK/G7inYOsXLyFc5EKU1OGRxO5c3Ali7ecI/vH7nBw5WK2nIvM/xH3a8REVj/p4g80fM+NxvOuOVrnMmCjXsgPBPKBIEg8Wt2SkmU+o27dupZfHc+i6PTF8azTAO9FVxR+4PwQ+c1Pw/z3Mpq6CghOXsy8ZjOJZyZ3DIE+xdDqyvLtsTTHZ5YEIXE3vYtqEb1mYdnrJDmapy9sDUgifx+cSSsPPYK+PCP/lwzm+yz8QkSjLUq3X3MKmMwlQcDE5amfIGrd6bP7GecmNKTHr8oRoJn7C79A1Ggp2u1XB197ZgikvyY9WkZTg4C25AAO2OMX0j8lnysIwro8uBiElyEIedPp1RKEZKKfvrA39Il/H2RmKw/0gp7yI/9nbzSzIgjmsOV8VVSPe5ef08irMSgPBCFvWGRWcmnXjNzf3ptyooDBa5ZjXEvaQ4qjZHb1ckPQaNDk4ufUaj2ONUORVIbDZA4P8kBXoB2bsv1U8ipzhozewQuvC5MUDg8qja5AezbbVml+B9F8k1TKB4IgqyNhTE4mKSnJ8ovZ0hGXwt3YHpuCyTaCNr0gNt7Wmb1JELwKWZK4OLMuhbVaijSaQ4hilpwyt8SgKXzmLFLm6x1pjbVsic5ioaSEw4P5UO9Gy9Xhlg4i5ei31B5yxN4pyOXwbEMbCggGmq+KkoMUOD++MnqNQKFG8wlVmo8txSal+aFJtXwIBemaQ4iw+e5CvnDRUrh2a5rUHsGJdB258fx4KsszMwo1Yn7GTJFsdUIJhu3YdJWAmiKC1p1u29MF7UnPuXU9wuraAOK2drAEIzaY72ipiN3SngIZlqrOWT/TzVl4iQIFuwamsyDI4R651ynlz+GU0ekoO/ykXVaLesm76f2eFrHGNK4oBre2zvzLZQ8V5WEDJN3/lKN8W3sIRxQWZenZBtoUEDA0X2U3kWeepomr0z9FFArQ6WdFfIwl2EyP7oMhHFPWEZuL4ctlDhvr5AWLdNKDMYqr526lTWmUIljaxIDg1otdCp0yvGe5IBF77wqXLl7kYi5+wXee2YlU5ukprxo5Ny6LhZL+kczKPORglhhuX7hEuAL+dE+8Gac5yZmfmORaY9s0xx7qQkm5xuyfPZhPBMFRCLnhdnbtxm+2jkOKYnNHN0SPQRzKsRFwTOutPTM/ZO/oz3lPq6dUM38O3FcqLhFzaTmdyxaiTMflGQhE8h9fU1Sr44NeO+wNvvT8JBNqufGB9wbupHYuxr/G4vl+BzbanfVGwubWx8lQnSmpe82awzfR0V2HIBio0Gs9l59ZLRqJd3czsV0XFtunIyRySB5BadIH92VSAnJ5dnBDKxioNUOedpjuzxzOpo7u6OT1HSr0Yv3lZ1ZffOJddk9sR5fFN9K9oDyViNzRmw/1AtpiDfE7aJ02mfzkMlsGN6Ldwmv2Ttd0yY9qooBLrUmciZVnGsRxY/t3fFXRDVEQED9qi//GE9y1ELSc9Us8NAgPnQaxzmzC0nPZPOgUvbUThQWBIt4/oxzoWAJSnWTrSH/2pRmMSNzhQ2FBx4cD9hIVfYcT65aw2zYjRAmNfCwvS+v5Ph022qZ2gjFsLvWdDFSfkrbZTeZpGrk9u651jY5RJ60BjuYnnJnfkbIGAcG1M9tikomLS2UJiTvwKSyg+3AAe6OiuXNiHUt2/405D1ikFz/p9x64aYvQZv0jKxlKucjk6iKFmi7LwaWUPqX8Ps96HYT8kzmeff1LWaYKO388kqPKypHf6vyj9HKWM/8wyYugNoKgrqSYF9T+ybP5SxCkaE7MH0iLSoUQtEWp2dWPnZaGLoXQ7ZMYPuMAEekb3n8i/Rv/rmz2n8eAJpV4z6UwH9T4ks7dO/NljQp41unM2I2X0kZSCl0SA7vhXrkZrWuWoWKjHgwa4E2jT7zoOvckUQr8TNdnUdfVQOFKLfhm1Ej6t/mMshW+wHfTdYdI7LjgNXxT2x0nOUZBLIx76VKUrdOPZeetrgHz/W34flEJN728i6QGwbkk1RoPYrNt5SaFbLbDuL398HBrxdqslqWNC2bNN7Vxd5J95CKF3UtTqmwd+i07n7WP15Y4iVzfMoQGHs4IghYXNzcKFa5A+1kn0023e8KhcbUpqhcQi3hQvuKntJuym1shP1LXpRjV2wzCb8HvXLmWg37m+2zz/YJKbnqr+VpwpmS1xgzafNdxBJqTTnI6g+tSppDWko6gc6Niy9mcN0L84YnU+6gQWtk0Luhwq/Ql0/+0kkYp6ld6eogIGgHRrTo9V1xC5juZ/pmuM6uuK4bClWjxzShG9m/DZ2Ur8IXvJq4rLFVZpSlF7mN07RKI2kJ85NWAzz9tQJ+lB1nfoyQ6QUfBEhVoHHDWmrUUxa89PRAt9caN6j1XcMkmWE5YZCq8bPYJYkGHChSt1IZhY4fTs0F5Pqo/jMC72bniskosP69nTRDyT+Z4DvqWtcQKabQl6LdPOWjIT13+aVq5kPNfKUeVIPzTks3r+/lLEOTJZCYjJnkisGTGZDSmuRjyKtk79rzpxSPCQi5w4UoYEc+THDuedLpKMff4W46UkhKIvHWZ4NBIErPqMDDx4mEowUFBBIc+Ik5BIByTNfEi4gbBF4K5GRHrGF0vGUlOSrGWm+wwMptISUrGmGWe8kPPuX1LMWPCMTP7melFBDeCLxB8M4LYvPYB5jgiblwi6NINHmapmETi45tcunyHaJt5XHrG3bvP0zDOUT/ZRZZEislsHdXKdTclieQsAMhaJ6urzZiajhXHFKscphSSUlLnyFvSTyZFUVbm2PtcuxGRTfnZIbUcmF48JDQ4iKDgUB5lgU3WaSbx5PYVLoXc4WlSaiEbn3HnyjXupy8kcyz3r90gIos8ssbCUV7HMxNxD29w4a8grt7PZNtzx4df05nNxdAki70Y8ktmE3FRd9jUvR4TzuX1g3hNUFiyyY2c+YVJbvVK4vce1rUq1L0YcovZP3su3wnCPxNHfVtFQEVAReDfQEBehKcSerEOP2bwL+WvPMZbK+gzcCvhCoKYvznkT2pvnpwxbGlfAG3h/+oqvflTrnlJRSUIeUFLfVZFQEXgHUVA4vHaVhTSOlPnB8cFyvJTYVPYVkb4LuQv2/Kk+Zl4Pqb1JsopPdxAOznI97MAQjIEPuWj8mpSdgRUgmCHQj1QEVAR+E8jkHKXnWObUbaYJ93mHST0+StwAZiM2bvu3pQCeJPkNMdx/9RafL3cKVV3IOuvvOlTQN6UQvzncqgE4Z9jqKagIqAi8A4hYI69w5m9v/Dzn/ffIa3eXlWkqPPs2LaLP288tc9genu1ebskVwnC21VeqrQqAioCKgIqAioCrwUBlSC8FpjVTFQEVARUBFQEVATeLgRUgvB2lZcqrYqAioCKgIqAisBrQUAlCK8FZjUTFQEVARUBFQEVgbcLAZUgvF3lpUqrIqAioCKgIqAi8FoQUAnCa4FZzURFQEVARUBFQEXg7UJAJQhvYHmZEx5z49Qudp59nPPufm+g/KpIKgL5j4CZxKhbnP3jN/734A1fgjD/lc9biol3ObzCn3HTf+OWuqBQ3rD7zz1t5uHh+YyfPJ9d119k6G9UgvDKK4REQvhZfls5m+kTv+O7SdOZs2oHp+9msga9+SHbB3tR2lneNElP5QnnHfdMeOWyqhmoCKRHwMiTq2cIifr3ehrp6W5GeJXGRdCg0X3It8dtm26kymp8wtUzIfyLIqYH7V87jw9eRhfP8rSaf4E3fLHGfw0jNeOMCCTe2kKvKuVoHnCCp4o9eFSCkBGrfLsiRZ9jed+auHvUp8/kBWze+TtbFk5liLcXJZ2KULnDVPb8na6xQyJiaRMMgsin00IybqWcb9KpCakI5IxAyskRlNULuDRaSDabe+ac0D99QnrCpnauCPoKjD6tXOEwhZMjyqIXXGi0MN3um/80z7ft/fhjDC0vUrjFalQjy9tWeP++vLF7+uGhL4H3tki7JUElCK+oXKToY4yvVRiXKr7sfZzeJCrx7FQATYrr0Ls3Z0GwYp9eIGlXL9wEkVozrqsE4RWVz5uUrDH8GEtGdGLAuntpO1C+RgGzy98csZPhjWrSYsZpXrxGmTJmlcLxbz9Ep/dk7F9KgmAmYudwGtVswYzTeZTQGM6xJSPoNGAd99J/ohkFeOOvGC9OoopepN7c/zhReuNL6g0VMP4XvAtqKfr1HmwbkasE4VWUlfSMPf0/Qq+vzJjTCVnkIPHsj358pBMQKw3jWFzaY8l7+lBUKxOEV7dpTFpu6tG/i4DE4xVNMWhEGi1+YGfur0+mfzv/3Gqawp/DymRCEHL7fsbnpMcraGrQIDZazAOFWTXjk2/HFeP58XjqDXy59OG/UI/eDoxUKbNBIOl3erppcev5O0mpj+UbQZCeX2Lnug1s/uknNq1dzqI5s/j+12v/SR+66eYP1HYSMNSbw+3sRibmeyz5sgCC4EKDeWH20aNKELKpxO/greh1LXHSGGi2Ms209zrV/Lfzz52u+U8QiF5HSycNhmYriVQJQu6KQX3q3UXgVRKElP+NpEo5LxrVrUWNalWpXrM+vdfe+g+ayE1cn1EL0RJkGJQDQZJ4vLIZToKAof48u483jSBcIfzPlUwa5IN3d1/8Vp4gQmldtVTVBG7vW8iYft6079CNvr7jmLniEHfShTZIMSHsXOTHkN7d+dp3Mkv3h2HfE02K437QYQJXzWCozxgCI83E3wxkWj9vuo7YyNX4SI4unswkv6n4+0/Fb9JEAgJDrWVrvs++uX74TfVnmv8Upiw8xEMbKZJiCNm5CL8hven+tS+Tl+4nzJ5pTt+ZmSfnf+LHsd/g3bYdXfqPY+HeUCyGFukxRxZNYpLfFPz9/Vl96nnqiOkFZ9dMZuLkKUz192fBgQdW0pWTfokScfeDOBy4ihlDfRgTGIk5/iaB0/rh3XUEG68qXEDZ6mTm+e0z7P95KVMH+hBwTH4vgbADK5g+si/enXsybPY+0sJOErjyy/dM7FgJvUZPhfbjmDZtGtMClnAoByey6eklfvlhFP26tKdj90H4bw7iSRZxhOYn5/npx7F8492Wdl36M27hXkItQGaff0JEMMd3bWLh1MF0G7fTIXjJUnrGR5zdFMC3PTvRtp03fUb+wLagKIdv3vz8Nmf2/8zSqQPxCTiGBZGwA6yYPpK+3p3pOWw2+9IAyaZSZEYQEogIPs6uTQuZOrgb43Y+dRw5J9xm38Ix9PNuT4duffEdN5MVh+5YNvxJuPIL30/sSCW9Bn2F9oyTcZ8WwJJDqXUmG0kst6QX3Ng1lxE9O9K6TQd8+g1h9Pjv2XXHVvlzSiD/7+fegiDx4sYu5o7oScfWbejg048ho8fz/a479kFK/kv3FqT4Bpbpa0XtVRKE+J+7UmfqZYfG4bUq98ZkFs3Gti4IGidab3ieo1TGoIl8rNcgFPZhR6pdx0oQtBQp60WLgbNYv30bS0c3pbReT/H6UzgebRvuJBEUUBu3kq2Yd+YhMZHBbOxXGWfn9myxuywknp+dR3vPanRbdJSbD+5ydk13Koiu1Jp8ytrhmsL4za8HXiV0CHpPRm1aRIeaVanoJp9XZsL5/yPir62M/aIYWo1AgXrj2X8rNrUxTiD87Hr6VBIp0Xg8W88+sJinpOdnmdfek2rdFnH05gPunl1D9woirrUmc8ouWxbwSE84HtCccp5dWHTiNk+jwwj8xhNRW4SGc69hIpkH57Yy3KsQgsZAkyURqbKYiLqyE/9mJdBqRKpPCbbWxxz1SyLsNz96eJVAJ+jxHLWJRR1qUrWim+Vcnk0i/+WsUzLBG4bToWoRS/51Zx1mS18vqjX2YdiEUXSoXBBBKEDD+TZrURzX9q5nSf9PEDUiVfssZO3atazbuIeQbELQY08H0KDUZwzbdYMncc+5tdOXqk4GKg09TKwDpBJPjgfQvJwnXRad4PbTaMICv8FT1FKk4VyumbLL///jwf5Z9KlXCr2gwanlOqIVaUtPj+NX34OqX6/i5K373PprFwt6V6WQ6E7TH8/b4xWSgzcwvENVimg1iHVncXhLX7yqNcZn2ARGdahMQUGgQMP5hOXYr2ZCEMwP2D+rD/VK6S3fW8t1CgmTggio7UbJVvM48zCGyOCN9KvsjHP7LZY6H3dtL+uX9OcTUYNYtQ8L165l7bqN7AmJcSQZCp3thyl3CRxUk6qd5nMiPI7YeyeZ3rgQglNTVjyyfZv2px0OTFe3MmnUSEaOzMXvu9WcU3BTh4QyOTGeHk0F2cWw7FE2OqRwN3AQNat2Yv6JcOJi73FyemMKCU40XZHde5lk+C5d+gdl+s7AkLTL4mIo1DUwv10MZsJ+bES7tc+yqZjvDIzZK2IK5XsvEY1QGB9bj5/NG+bwRTQSNWgMzVgZZW1crARBR5khx0iLYEjkgt+nGAQdZYcctY7+U47i66HDpeNP1o5ezic2EJ/yvdiVSjakqEB6lHaiht9Fe6EjRbGmhQtCgWasiLA1aCaCp1RHFERcy/uw+b6JxNsH+emPq/bpUqZQq+tE9JrJDeVoNfkkIz5twfK/U1t5KYrAHqVxquHHRZszC4moNS1wEQrQbEVEdogQsbkjJcQyDDmapr3xwkSq6AUKttuElXaZCV/UCFFjoMUax5Fj/LaOuAgiXrNuKghrLvQLnkJ1UUB0LY/P5vuYEm9z8Kc/uCp31nnQKfnQQEpqBZzd6+K77aa9DI0h0/hUFHBuvSFVBysMMRtaW1wMzVc/yfn7kSJZ/ZULgms3frN1HtJTNrYtiLZQOzY9s5UnmCM207GESJkhR+0yYLzAxCp6hILt2JTKX7PL3xz2I3VEDU4dtiosTtbyFcsP50+bDLIqpjssaeqKIHoy6mRa2ZF8iIEltQjO7tT13cZN2y1jCNM+FRGcW5Mzl86EIFjgk9ueOogaJzpsTTNPpRz1xUPnQsef0thobKAP5XvtSvsOYjbQWnYxNF/NkzTYsqmbgPEWazp4UKzh91y06SE9YtmXBvRVJnExg4XPMTnjX3Po0qolLVvm4tduKofTxHdMKJOzhJ3dcdO60WOnTbD0Dxm5taYDHsUa8n2a8Dxa9iUGfRUmpRPeFB/J7cv/Y8+mxWw6E5M+sXfnPE9laiI+8jaX/7eHTYs3cSYbIv/WAWT8i7GeegxfLsPGc/MpBiGF40M9+aRNH7q0bU3Hr0cy+/ebigYlml2DKlHqMz9OpzN9v3UMH/zRAAAgAElEQVQg5iSw6TozaskEwZmOP6U1WFm9Jj1abgmU0ji1YUPqN5jmYnAMUpQeraK5i4D2/X7skzte43nGe+rRlmjF4hBbS2Ii5kl0ahSqiWuyu0NXhl6rjnD0yGEOHtjP3j2BTGrsiqAtQf/9tnhVM2Gz5YZWpGZAFrMnpEesbSVPNauG36W0lvDFH/3xGnTQXt6mazOoJeoo02sVR44e4fDBA+zfu4fASY1xFbSU6L8/KzjAdAm/anp05UfhMJtNekbIoYNcirTlK/F0TQsMGidarVeMHOVZIDu7U0QQqTPbNlKXs8tZP3PYbEtnKNYM4LqSAMl9Xx50Mp4eRXmdlpIDDtqjgS0KJ+2kRxEBsd5clJbo7DrojEClEBo4k4D157AbkjByelR5dPqPmRhkw8fEJb9q6HXlGeUIJM9CDnHwUqTd/ZVd/rb66dTpZ3vHar4zjwZOWkr03Wu/ZpMz9tduvKfVUrTHDtslMJ5mVHkd2pIDOGirblZA2NmjCIJYj7lKQNLeVBxlRRAkHi2Xgzyd6PSznY1iNbdrKdFqMWmfRgxPohUC5JkgmLg+pwGuBRsw95bC5JF8iEGl9BT/+g+LC0Uh9Os5lJJ4dG4Loxq4U7rZXC4qSZtCAtP1OTRwLUiDubcUroRkDg0qhb741/zh8J7ZYkHqWr0QgrYUAw8pcFOkmd1hSshi2lb04JOBgWluxwwvpBCyuC0VPT5hYODDDHftF1JCWNy2Ih6fDCTQ7sO03/0HB3ksU4vVqivVCwloSw3kJWBJlTWF4PWjGTxoEIMHD7b8HzTse/ZFyPXKTPjemQwbnHpv8EgWnXz2D3TM7atm7mzyoUKxT+i74jh/x5vJJ4Jg5Pofa9lx4iLXrp5n18xWeIjF+Gp5WOoILoHzi3rQfuAmQtM1vLkV/a15TnrKupbOCJr0I9jMNTBdC6CmqEH30VBOpJKnrAiCjRBoDE1ZZqF4JkKXNKWoVoPgVIrPu09hy/m0hh/pEcubGixzx1uPGMvY775jzOjRjBo5khHDhzNs+BS22wsktQMVXLO1fMTs6EkJrZ6Ko09bfLlIj9nQqQFT7IQhtcEW9FRoPYKxY7/juzGjGS2bVUcMZ/iw4UzZHpo5GLIZ/8FiGhs0GJosxW7cyPTplyQI2ehnJQgCrj470nV8edPJShB0lB912t4JW1RI3kvf4lpkAnJN8R1k10FnqnrqReOz6xxYO5ORPdvg9aELgq4sI06mViLpAYsbG9AYmrA0eyDJLv/MCMKLnzpRUNDz8cSMMTbmu/OoL2rQVx6fJnoqQchA+khmb9/iaMWaBCgBSXtTcZQ3goAplCVNi1pcYk6lPqf7lC2ct5PL1GTzSBCk6B30cNfxXpdfFOQMjBcnU1Uvm+j/hZVPzff4fXIPmn5cjGJe33Ew1QqpAM56KEWzo4c7uve68Esas5SFZ3JVPU5NV/A4gxUlhWNDPkDn2pGtOXtL02UpEbn6K5wFDVr39MRQ8ajFIia3l1rcBxxU3HA8lCJX85W8gJzWnQGOLNPxwTyevVSZphxjyAc6XDtudbAE5jFrubUjJtCHYlqR6qNP8FTRJpB8kwVtvOg2Zz83o//hqNoYydlVwxmw6JJje5SpwC84N7s5pd3K0cB7aH4RhHQ5SQ9Z2awA+nIj+d8/1C1dym/BqYmbs7wQZV/9V6sz+egcVXjxizeFBS3Fuv9m9x9nTRCsH7Pg0oGt9infCYTtDsCn1vsYBA2Crhj1/Y5jsTSbrjCthohGbMiCHCd62whCEbrvTBuJOUorD8+PM7SsDt0HgzmSCOZbc2jWcoliHrmJK9NqWCwRDRfkfV6/KWQaNWS/cMMFijQzSGH5uF7KgiBkrZ+NIBTpvjMdQcibTq+aIEgvLrN+cAMqVmvL2OW7OR/+nL8mVEavJAimkFyXfd4IgsRDeSEvOahy9JmMDU78Vto7Ceg+GpZWaP8GQZBzTwhjd4APtd43IGgEdMXq43dc4QbNI0F4vs0bN21B2m5UpEEMhwaVQ6evyuR0Jvo0ABRHpiTi4+OIi8vFLz5J4SJTpJHZoSmcLd6lKFRtJEeUBMD27PNteLtpKdh2o7VtSL0ec2gQ5XR6qk6+mLEszXeYU1dErD8Pm/fQllyu/ife4eDKxWw5F6mwWGR8M/HOQVYu3sK5SIVVJsNjidw5uJLFW86R7WMZ3sv+wsuUqfnOHOqKIvXn/Z2FXsk8uR3Ko6w8PXaRJB5a3DsVGXPGZvkD6dkpZvfpy7yztuBr+wt5OzA95vSqUbSpUhRRJvQTMhJ6xwSTuPhDQ95za8y8a1aLUT5ZEByzIdXkqXfugMItmP6hd/bc/PcymroKCE5ezLyWVvAZFY4h0KcYWl1Zvj2WVpuyJAiJu+ldVIvoNYubMttMjubpC9tHlcjfB2fSykOPoC/PyP8lg/k+C78Q0WiL0u3XnIYAuSQImLg89RNErTt9dj/j3ISG9Pg1WuE7N3N/4ReIGi1Fu/2aZ4YtPVpGU4NgMUcfyIanyOzbThCUwWk5uRheiiDkTadXShDMYSz/qih69y78bB/yGQlKTxBkv7hsPdKWZED2QObZgpCwswduWoFCHbaSwTMdu5E2ThpLp2Kv7/8CQUiOfkrap/E3B2Wrpl5AX34k8qdh+csTQTByZkxF9HrHxjz2+AgqiwLa4n3Y42Cit2vvcJBsWQRNg0aTi59TK9J5zxzSSn+SfHgQHroCtLMFlygeMJ4ZQ0W9nopjFKQu9jgjKosI2uL0yUR46dkm2hXU4zn+XV3y/WXKVOLZpnYU1Hsy/nwWbbvpJrNqV0qz5inKwfEwiT19iqMr7M0vqQO+pNCtfNtlGL/ctlVSxzfydGZ+yKUzt4hJ2E//90WqTMqEBCoTTDnMoNI6CrTfbP+uXxFBSGZfvxKI5UZyymZBML0gNt7WmSmlehePk7g4sy6FtVqKNJpDSBYNR2LQFD5zFinz9Q4HS0NWBCHh8GA+1LvRcnW4hbmmHP2W2kOOKPzcEs82tKGAYKD5qig5SIHz4yuj1wgUajSfUFtZ2CCXJEXHnmr5EArSNTDbnhnz3YV84aKlcO3WNKk9ghPpHpf9v5XlmRmFGjE/Y6ZIGUyZNoFkZ/9VAmrKjZY73banC9qTnnPreoTVtQHEbe1gCUZsMN/RUhG7pT0FMixVnbN+ppuz8BIFCiqieG2S5UWnlD+HU0ano+zwk3ZZLekk76b3e1rEGtO4ojAn2kbwXy7LeYEb09XplkDHAp1+tsd8yOV8bpwnet0HDDlmK2QTVwNqIgpa3LttTxeEJ/H81nUiUh/NLn+by8ep/RZ7ftKzX+hSTIu2qDfblAu3AymWjsiZOj/etEEHKX8yvIwOXdnh2Dwg1pvJ7O79HlqxBtOUgKS9qTiSzd0e6DIstSzxYHFjSwxC+y22mJ8Ujn5bmyFHFI2s9IwNbQogGJqzymaGtxGEL5fxMLs6aZEimUMDS6JV5G+6/xM9GjSi/kd6DE2X8yjxIfce2/BXiK44lGLvceXSRS5ezMUv+A7P8tBkGs+Ny3KhJGvgrGz1SXV7me7zU48GNKr/EXpDU5Y/SuThvccO9TVpf3/c9SXouzdjA2aKuc2FS+H2OqFQ8Q06NBFz+wKXwm31Ir1oL1OmSezv746+RF8ygcWaQW4JgukK/jVExDqzuW02E3VsGl26zuCkItA4vcQvdZ5ykhFlDZlbiZQJ2qY59sjvhZISz7Gg30AWnbVOEZKe76aPhyt1vg+xmq2kKDZ3dEP0GPQPgjqUmrwFx+aH7B39Oe9p9ZRq5s+B+4rGSvY9XVpO57KFKNNxeQYCkfzH1xTV6vig1w5sbZn0/CQTarnxgfcG7qR2Lsa/xuL5fgc22n3MRsLm1sfJUJ0pqeZOc/gmOrrrEAQDFXqt5/IzK+tNvLubie26sNg+HSGRQ4M80GnSB/dlgrVcnh3c0AoGas2Qpx2m+zOHs6mjOzp5fYcKvVh/+Zm1HiTeZffEdnRZfCPdC8pTicgdvflQL6At1hC/g9Zpk8lPLrNlcCPaLbxmb8RMl/yoJgq41JrEmVh5pkEcN7Z/x1cV3RAFAfGjtvhvPMFdS/uWs36Jh+QRmMbywWaYdpcHnaK3dqKwIFDE+2c7E5c1tPjznWTrSH/2pRmMSNzhQ2FBx4cD9hIVfYcT65awOwubrvn2bOpafPyjOGmJSzXz5Mx8OpY1IAiudN4WQ3JcnAUjKXIHvT/UI2iL0dDvIA9kIpf8hMtbBtOo3UKupfZl2eVv7fA1OM5cMRK6rLnFd1qh/y4e2TqxlLusal2MAtXH8qdyvmX0VjoVFhCKePOz0uRgiZFxslg5+isBUVYH27FlpoYcMOdGl+3KRFKsI3s55memLajXyF9jPXm/w8a0OBZjGHPrO2GoPiVtpkHiDnwKC+g+HMDeqGjunFjHkt1ZmY1TZ0sIBrymnuNe0Cp6ejVmym8/0MSgpbj3XJb3rEfvX5/YJH7t/7NbB8E6G0XA4DWVc/eCWNXTi8ZTfuOHJga0xb2Zu7wn9Xr/qiCSslXqY8QCrVmffopH/D76l9JaArE/Hnn0teuZ2wzj9/WnlByf5fwxI48q64wthZcoU2MQEz4WKdB6vQIrW3qp/3NJEKRnG2jjouW9ryay+LuvKFdzHGcV7UK6VF/+NK8EId9XUjTeYss3n+NRtglDA/zo17gWzSfsIdzec6QQun0Sw2ccwBKk+fKqvmVvymb/eQxoUon3XArzQY0v6dy9M1/WqIBnnc6M3XiJ55mMXBIDu+FeuRmta5ahYqMeDBrgTaNPvOg69yRRtsZYHmxfn0VdVwOFK7Xgm1Ej6d/mM8pW+ALfTdcdoqnjgtfwTW13nOQYBbEw7qVLUbZOP5adt7oGzPe34ftFJdz08i6S8gdVkmqNB7E5m9154vb2w8OtFWuzGnrFBbPmm9q4OwloBJHC7qUpVbYO/ZaddwjwyrxAE7m+ZQgNPJwRBC0ubm4UKlyB9rNOOn6U0hMOjatNUb2AWMSD8hU/pd2U3dwK+ZG6LsWo3mYQfgt+58q1HPQz32eb7xdUcpPn02ssDV/Jao0ZtDndmvY56SSnM7guZQppLekIOjcqtpyNbImMPzyReh8VQmtJX4dbpS+Z/qeVNEpRv9LTQ7T4yUW36vRccQmZ72T6J0Wyb3RtSohaCn3kRYPPP6VBn6UcXN+DkjoBXcESVGgcwNlUq07i9S0MaeCBsyCgdXHDrVBhKrSfxUlFo595/rEcmNgEz2KyXDImTpSq0ZpZdpPgC4LXDaK+R2He/6wdfQf2oW2titToPJMjj2wfvpn72wZTt0whtPJOjIIOt4otmW0FhIn1PqKQVja1C+jcKvHl9D8V1rA07U031vB13bK4auX6KaArUoEGAzdz5/8dYGITT4qJqfXWqRQ1Ws/iVIqJ67Pq4mooTKUW3zBqZH/afFaWCl/4sum6YjQsRfFrTw9Ey3fhRvWeK7iUJfBgfrANnw+t5WTwaE7A8ScYb8+hnkFAEN1pOPmIY/1MU+G1HGVHEDA/YJvPh1ZdDR40DzjOE+Nt5tSTiaWIe8PJHFHUCcz3WNBQRPz8B5QTNiyKxB/EVyakGg3aEv1ei24vk0n8QV/KGuS6oaVEv32ZJpHXMjXfW0BDUeTzHxxngjx/9IDw8HDr795xxn9Wlv6/3U279iCCZ4qqJwuTcmwIHjoDVdr1o2lZZwTnT/E7n+4hB6mTccjHlp/yfyb58K8TBJsSKdHcu/OQOFv7YLuu/sf04hFhIRe4cCWMiOdJWQS3WIGSYu7xtxxoJCUQeesywaGRJGbVYWDixcNQgoOCCA59RJyCQDjCbuJFxA2CLwRzMyLWMSBJMpKclILJbM1EMptISUrGmGWe8nD4ObdvKWZMOGZmPzO9iOBG8AWCb0YQm4XLzv5w+gNzHBE3LhF06QYPs1RMIvHxTS5dvoM92Fd6xt27z9MwzlE/CWNyEikms9XlIpkxpSSRnAUAWeskp5OMMTUdK44pVjlMKSSlmLBAbEk/mRRFWZlj73PtRkQ25ecITtKT21y5FMKdp0mpbiIjz+5c4dr9dGVrec1MXMQNLgVd4sbDuDRcFElmlr+MgQ0TKROZrUnH8Sj0MkEXr3LvecYClozJJBuVeiel6i3XMVudkzCbUkhWAqKQDbn8ko2p9VN+1nou4ZiGBe/kVLzl900veBgaTFBQMKGPMtcbcyz3r90gIsv6pRRE/iQfcv1KGE/sBsEUntwKITQyUeGuc3zndZ3ZXAxNstqLQUrg4fUrhKUJT8qTW4Rk1r4830pHVz3lRp6yW+wc9DDFEXVnE93rTXC4/KadmOKiuLOpO/UmnMtStLyU6fOtHXHVl2OknSjLntyzTKn9AaVLl079vU8Rg0ih4rbz0pT2qITvXnulsUxlvC1PKxc/wf+KCdOd5XzlpsXFa7pi7Zh0ImfIR5G+Le8M+chMJPcuhh7yXgz57mJIp4d6qiKgIqAioCLwehGQXY6V9CJ1flSu//FyMiQfGkRp/Xv02JmV/97IrRV9GLg1/OUyeF1vGW+xos9AtoYr2PhL5y2vGVEa/Xs9yBIWOe1cuRji2N6lCLpitvUnTIQuaExhbQFqz7ySOSl7WblzSxBittC+gJbCillcryhI8WU1Ud9TEVARUBFQEXgZBKTHa2lVSItznR8cVzrNc2JGyyJbonMzVtqW1HNIw0TY1hH4LvzLvsqqw+035cQUxtYRviz8KxfLZ+dGZqO8iJuIc7OV9pUGM30tNwRBXrWwkh5D4yVpO4mmhPB9HXlF1Pr8eDX7YNdM883qYspJhpcx5DCLQeLhhna8pxX5LCDEHlemEoSsQFWvqwioCKgIvFUIpHB351ialS2GZ7d5HAx97uhKzK0u5nDLIltiDX9O/bWYb6en3+MDTEbjv+5SyVkdE8Ys3IQ5v5vxCXO4vIibSA3/U/y1+FumH1ZG4iqezwVBsM4O0mVw4SRd8KOGQYtrw9mEpJsdpsghb4cvfqWrmx6PwUczjfExx93n1FpfvNxLUXfgeq4ojEYqQcgb1OrTKgIqAioCbzYC5ljunNnLLz//yf2XsawbzzFOXsK9oAef91rEuXdpv4F/UHLWGA8tBT0+p9eic1lbT7IjCKYwdkwfSpdaxS2zvApUasWQgF1pS6+n/I8R5XSWwO7S9b9h9SVl3EJehU/hwspBdP68lCU4VetaieZ9f+SIw9RkiajzO9i2609uPM1otVAJQl4xV59XEVARUBF4xxFIfhrO41wGbr7jUCjUS+Zp+ONcBxIrXlQcSphSHANujUaTwhojB+HKwdISZmMKxpcheMrcTCn2oGkkM8YUY6ZByopXHA5VguAAh3qiIqAioCKgIqAioCIgI6ASBLUeqAioCKgIqAioCKgIZEBAJQgZIFEvqAioCKgIqAioCKgIqARBrQMqAioCKgIqAioCKgIZEFAJQgZI1AsqAioCKgIqAioCKgIqQXgD64A54TE3Tu1i59nHiujWN1BQVSQVgVeKgIkXD65wLHAPwdnskWAVwUxi1C3O/vEb/3vwD0O/X6lO/27iiXcPs8J/HNN/u/XvCqLm/gYhYObh4fmMnzyfXddfOPQ5KkF45cUkkRB+lt9Wzmb6xO/4btJ05qzawem7sfbVquwimB+yfbAXpZ3lDUb0VJ7wru7DbtdYPVARyBQB0/UVdK1W3DJ/WyM2YH4Wu1vKL0tPdzPCqzQu8oZQug/59ni6+dzGJ1w9E0LUf3qPmHiCl3XBs3wr5l/Ip5UFMy059eLbikDirS30qlKO5gEnsC2VoBKEV1iaUvQ5lvetibtHffpMXsDmnb+zZeFUhnh7UdKpCJU7TGXP3+kaMyQiljbBIIh8Oi1tyctXKKaatIrAm4mAKYRpn4poDF+yLNMlfxViS0/Y1M4VQV+B0aeVm0alcHJEWfSCC40WptudU/H6u34Yf2wo5cXCtFj9IE/z4N91XFT9HBGI3dMPD30JvLdFWiwJKkFwxCffzqToY4yvVRiXKr7sfZze5Cnx7FQATYrr0Ls3Z0Gw4xafSbt64SaI1JpxPaOVId8kVBN6UxAwhh9jyYhODFh3T228HQolmnWtnNAYmrEyMrutReWXUjj+7Yfo9J6M/UtJEMxE7BxOo5otmHH6hUPqOZ28O+Vi5OKkKujFeszNZgv3nPBQ7/8HEIj/Be+CWop+vceyLLNKEF5FmUvP2NP/I/T6yow5nZBFDhLP/ujHRzoBsdIwjsWlPZa8pw9FtTJBuKEShDRY3tEjiccrmmLQiDRa/MDB//eOKpwHtWLZ2CaVIETlTBD+HFYmE4KQh+wcHn2XysXI+fGe6A1fsvRhTjg6gKCe/NcQSPqdnvKWzz1/R94K4tUQBOMzQvZu4eDt/6bTz3TzB2o7CRjqzeF2euOBssKZ77HkywIIggsN5qVt0aoSBCVI7/5x9LqWOGkMNFtpNeu9+xrnVsN/kyDAu1MuKkHIbY37zz/3agmCmcjjP+DTtAPjt10m+j/JD0xcn1EL0RJkGJTDbmoSj1c2w0kQMNSfh836l0YQrhD+50omDfLBu7svfitPEKG0nlpqcwK39y1kTD9v2nfoRl/fccxccYg76UIbpJgQdi7yY0jv7nztO5ml+8Owb9olxXE/6DCBq2Yw1GcMgZFm4m8GMq2fN11HbORqfCRHF09mkt9U/P2n4jdpIgGBoVbrhvk+++b64TfVn2n+U5iy8BAPbaRIiiFk5yL8hvSm+9e+TF66nzB7pjl9imaenP+JH8d+g3fbdnTpP46Fe0OxGFqkxxxZNIlJflPw9/dn9annqSPvF5xdM5mJk6cw1d+fBQdS/a056ZcoEXc/iMOBq5gx1IcxgZGY428SOK0f3l1HsPGqwgWUrU5mnt8+w/6flzJ1oA8Bx+T3Egg7sILpI/vi3bknw2bvIy3sJIErv3zPxI6V0Gv0VGg/jmnTpjEtYAmHsojEl+LuE3Q4kFUzhuIzJpBIczw3A6fRz7srIzZeJU1SiZiQnSzyG0Lv7l/jO3kp+23gS3J52spsKlMmT2b2nnsYn5xgmd9kpvhPw3/qFCZPXsyRx/KI08TVbf4WvGVcp/r589OV1IpoesqlX35gVL8utO/YnUH+mwl6ovzw84JtLCG/fs/w3p1p16E7gycG8PUncgxCM1a+lAUhgYjg4+zatJCpg7sxbudTRwtNwm32LRxDP+/2dOjWF99xM1lx6A4p5L1c5NqcdPMPFs+cxCjfbxi88DRxmIk6uYTh3bsyfPMN0n2SOX0A+Xg/e4Lw5sqdjxDkMqn/PBavjiCYuL/9aypX7sXPf2foxXJZPO/CY9FsbOuCoHGi9YbnOSpkDJrIx3oNQmEfdqRu72klCFqKlPWixcBZrN++jaWjm1Jar6d4/Skcj7aZCZMICqiNW8lWzDvzkJjIYDb2q4yzc3u22F0WEs/PzqO9ZzW6LTrKzQd3ObumOxVEV2pNPmXtcE1h/ObXA68SOgS9J6M2LaJDzapUdJPPKzPh/P8R8ddWxn5RDK1GoEC98ey/FZva2CYQfnY9fSqJlGg8nq1nH1hMU9Lzs8xr70m1bos4evMBd8+uoXsFEddakzllly0LeKQnHA9oTjnPLiw6cZun0WEEfuOJqC1Cw7nXMJHMg3NbGe5VCEFjoMmSiFRZTERd2Yl/sxJoNSL/P3vnAdzE8cV/naSTCzY4dGNaKIYAoZneTQgl9A6hBUw1LeRHTQgtAULvLXSSUELHtIFA6C2MabaBARcY+h+MB+y/pb908/nPqdiSLXfTzzOyTnd7u+99d+/2u++93a0w4ZKFxKSpXxxhW8bT1S8fGkGL7/A1zG1dhXKlvMy/5dkk8l/aOsVz6fdAWpfLZS6/5tQDrOvlR/kGnRkyejity+RAENypO8tmLYrh6u5VzO/zJaJKpFzPOaxYsYKVq3cRnMIOesawLYzv6kc+jYDWdzhr5ramSrlSeMm/y4zmrPzoSc859VsrfMt3Yu6hG0TePsXyLiURPasyzgx+PFGnNzKmkYyTQI7a4zh05zWSPpJ/l/SinJuASluanr+f4555MzmJF2H/sG5QFdzcfOkyaydX5TYYfYJJdbypPGQ71x/H8Dx0GwPKuaArPZjEnXCN6cM2+gyzW5emVLPJ7AyOIDLkBOtG1CS3WpV5gmCKZO/UntTy1pqfx6YrnyU2uLhzTKruRcFmv3Hy3gseXlpN7zKuuLZaRwwZrxc5Y0PUGf5e0I3SWgHX5iu5sSOA8nk80AkqxJrTU7cmyiRs/ViGDxvGsHR8vl92xo4MJqrl/MjAiRElzS4GZ8GeWZPbeYkf6tlPHou47WYXg0fHzdnrYtBfnUHd3L4MOZzCHtkfaovJqNzGEKb5iaiEnHS29fip5GGKmEt90fElaCEIGooNPExiBEMs58dXQidoKD7wkGX0rz/EAB8Nbm02WDp6uZzozXQu0Z3tVrIhPdpM10IuVBx/wVzhZlGkRyz/2g3BvTGLo2xkw8ilCRUQBRHPEp1ZG24k9mYQG3ZeSdjW1BhicZ2IflO4bj9IjD/G0Epfs8g2FU16xOauhXCpOJ4LCXuaSzxa/jVugjuNF0elhghRa9uQTyzGwEOJ2hvOj+ELrUCOlmuw0C4TEXPrI6p0fL3ccWT4amMb3AQRv6k37GI40qHfpQlUEAVEzxJ0XhuOMfYmQRt2ckXurDOgU/z+fhRUC7gWqMmAjTcS6tAQPJFKotx5/G7VwQLDi9+bm10MXy177DjCTQkl4yUmVBARRE9KdF5LuDGWm0Eb2HlFnr4m8WhzVwq5VGR8IvhIj5bztZuAe+PF2KrcdGsO9d2TusJiODigOBp1AXrssGdyEveWNKXiyBPWdiTxcFkT3ARPOm2x2S0knqz+hhxqD1queeqgizFVbJ+yvWdhRJ/e7LZ/fZhuMaNmViwIMoAmwnoPnYYAACAASURBVH6tgahyofX6RPOV/tAAfDRutNmQqGP05s6U6L494TnJcL3ItpaQafiJIuUDxtLZfzgHnzzm1PIpLDmeVt0aOD2jA82aNqVpOj4tfzyQ+Myn1E4Szr9mWxcv1F5d2Zb4SCVclQ/SltvIq4c3+e/4LtbMW8PJFAisQ6Yf6I9PGgvDaUb5atE1Wog8cSh7YhCkx2xsnxdP/xmcOPcPB4+cIeSJ3T7W0jO2B5TGu/J4Trw7O9vbaa7Ga0yuKhMEV9psSHwhpVS4dH8R/joVKpcW/P7CkirRxeAYpCjdX8pXbgLq/L3ZI3e8hrP8T963PV8z5gXbXnRGXjx+Zo5AlU3DV2V3h6YY3Zce5NDBAwTt28vuXZsZ28ATQZ2PPntt9WQibLr8IhWpMimF2RPSfVY0k6eSlWf8xUQr0cudffALCEpwWRivTqaqqKFY96UcPHSQA0H72Lt7F5vHNsBTUJOvz96U4ADjRcaX16IpMRyH2WrSU4L3B3Hxoa1ciSfLv0ancqHZKruRoWzq3daFXIJIjem2kbpcXNr6mcKmU0NUIVaZxDV7AiS/QDOgk+HEcEpo1BTsG2StB6u6cdvomktArDUzcf93IMMdkSmM6TVEVGIVJiUX1Nz+NMW6s/TgIQ4eCGLf3t3s2jyWBp4C6nx9SKhy6Qkb2n+GWqzEz8E2hSWerm2Fh6DGq92GhPnQmEKZ0bAB00Nt/iPQh2xmyqRVnEmwaIFFdy1lxzi611LD1hQ6nZo6DYUCDjjiRXbEIEjcXyQHgbrQdlMCW8Vw9n/4atXkazaPxEfnBY+f2Z6HTNSLzM83tsNT7UaJcs0Yf9pGnFJu7m/6ihR3nzPrhlOnQCEaz7yQotUhTbnN1piOVPAQUHv3Y38iTOlUQU/wvG8o5fMl/TbfS/kefTDzvimFz5f92Jzgq0yeXB88j29K+fBlv82JLs3kyTJ15s1jYRNLz6VVI+gfEED//v3N3wFDprEnSn7GTETsnsKQ/tZr/Ycx99hT241v8NvErTWdKZnnS3otPpI9BEF6uoZWniKfNx/D/LXrWDyhLb5eRWj22zlrp/Gas3O70qrfGkJs76E3qOI7zVp6wsqmrgiqpCNY51IZr06iiqhCU3QwR63kKSWCYCMEKp2/dV64kZD5/mYzrODiTbUuE1h39mFi3IN0n0X+OvPc8OZDRzHq++8ZOWKE2Yw5NDCQIYET+DOhQqwdqOCZquXjxdZu5FNrKTXihMWnKj3g97Z1mJBAGKwvZEFLyeZDGTXqe74fOYIRw4cxbGgggUMCmfBniHMwZOt45Dwa6FToGi5IGOk6T5xJgpCKfpZOTMCz89aEUaSl7IzpZOkkNZQYfiKxLuSM4nfTK6/aTECu2j0HmSUIgmeiW8qGkYVwCmhLNmfoqFF8//1IRowYzrBhQwkMHELghD8dnsFX+/rio9FSeuRJa31GseybSlTwdUVw92dhhIUQyK6wGq1XYQ5JsBVm+zY85dq+FUwZ1o0WfkVwk61cQ485+NxTw/bJyma4JrP4yJm/OYKAMYT5/rnNLhYX72p0mbCOswnk06JYhusFPceHfm52SxUPCMLeGGKD6m1+m+7+zbiu/pTNkwe/74NIOYwjnXLrDzOwsAbPNusdLGDp0kl6yLIm8ntRTYG+QSneIj1cRhN5oTh1AfoGpcRCLNYrV0GFukBfUkyWYimpXXgLWDgUL/Fic2fyqEUqjDjKE7v3AvE3mN3Cj04z9nLjWRZH1oaHnFoaSN+5Fx3fSQ6y2H685Mz0ryjk9Xn2EATDqZGUEosx5B+bEnoujCuP6NaIBeGJIw5b8R/3t5EbU/0QZV99k2XOX6h2ALz8oz05BTV5umxJeKGkTBAuMK6cFsGtNesTpnS/JmzHJDpXzW/2dQqaPNQef4Sn5tiyy0ysKI806zL7blr1YCMIueiyLXGkZSeq5TDuCIOLa9AU7s/BWDCFzqBx0/kkZm/k8sSKZktE3dkZn9dvDJ5IRXkUX3e2XZ7JpJCpROYsCELK+tk6sVxdtiUhCBnT6a0RhFxdSFpVxsvpxc+Kqf4MP/hq0fgEEPRars/pfNVmOVc2tCe3Wke1abKb5jX7B/jRa0eSLk96yX+r+lOnVHm+GbWIHWcjeH56NGW0KROE5NiaCP2lmrm9VPslNMk6EG+QIMjqvw5jx6TOVM2vQ1AJaPLUZvyRRNdIhgmC6Ta/1RZR5+/ApjTXbUjepo1xr3gVE0NMOj6v4ux7kuR52Z8xRqyjvbcH5YcdxM7Yk5gknXKbbs2gpihS+7c7SeopMavUjmJvBbFk3jrOPEztXRTLraAlzFt3htST3SJoyTzWnXmYKVlSlDMbsYh/fJOQ+yn4dBIEkLi3sBE6bSlGnrRZR0F6+i/Te/bit1O2AOyEGzJ2YHzAiaXDafFFbkRBS9nRjpa95JnFceGXunzm1YDfrsZnD0HQHx/G56LjCmb6o4MpKhZhsG1YnFySj/aM6c5C/D0FBBc/plxNrPTkCr9gc+c8qDXFGXQ4sSGlSBBid/BtbjWi31RuyO+H+Gc8eWl72GK5EzSFZj5aBG0Jhh2PB1M4c+qJqNS56fRXWgGT6SQIGPnvxy8R1QXoueMpZ0bXpetfz+z8zSbC59RDVKnJ3emvDI80pPsL8dcJqAv2ZV8qPMWBINgHn6XlYsgUQciYTu+SIJjC51BPVKHO3Yk0q9zcII3cmFYNnToPnTc/5tz4RvTc9gLp9UH6F9GYAx/PPPyDzjVGcsJhQGcibFETcmsL0GFT4p4hhnMZJQjW9QYEDYUHHXGwOrxJC0L8syckPjp3CJrSDB+tgLbEMORHR/7LMEGI+YuOuTTk7rotwd1mySk9/+PZ3t0LQaVClY6PS7NVODrWUisjngMBPmjcW7LG2WsgXXJLPF3TkhxaX/5njoRNrbwP+Fq2YSEPFKtTOoklLTkycezqmRdNzvb8YR30xYWsZ1CHIfxx0+GBS35res6Y7nHxZCgvXu+lT36RL8ZeSN2CoD9AQCEN7q3WInu8syUGwRQxj/ouXnTZatfJyYFaLjWZblsIwPiS6Fe2ziw9mn3IaeK4MKUmOdVqctWfQXAKrsjYcxOo7CpSrMdWB0tDSgTh9YH+FNF60XRZhJk16w8NovrAg3Z+W4mnv7fAXdDx1dJHcpACZ/9XBq1KwKP+LEJsBh4btJIc0mb7s1o+hBx03Jxqz4zp9hzquanJWb05DasP5WiS5LJ/t4w8M8OjPrOSF4qUWKit8MRv4xUmVRER1AXo9GeSwC7pOaHXohI6kZj1rc3BiHVmOVoqote1wj3ZUtVp62e8MRU/USCHNYI3USg53CP9Oun/CaSYRkPxQEczO/E7+PYzNWLFiVy2GwDaOqJGC+/Z1Yd96UmOjTeY6ici5OhIsqqS41LKaFEJHtSfFZKAlS0HyQn4UtRSmnio8ajZjhb1hvOPuT4NXJSDVjXeNGpag/pTk8SlGK/wcyURwb0tm+xCbQxnfsBXq6HwwMMOZaeK7YVxlNOq0BQfxBH7Z0WKYn5DXTotYHoOD/RBk2ypZYnIeQ3MMQit1tkE1XNoUHUGHrR7AUtP+b2FO4LuK5ZabfEZrRfDyRGU1Lry1dLMrGchEX33MhcvXOBCOj6Xbj3NwMjZwJkfUl4oKX1yx7G3TwG0+Xqx276OzA3LyIub57kYYcPX1tres2/jC26ev0hqYmYdC5vO6SQIxsv8VFFErCHPcjHx6PBEOnSczDGzCdiWVzZ8648xtLiOcuPSIAi2aY5ds3OhJNMdFvjnokC3bWbWIQfHXZ5YidyNF2EObJcesbaNF6JPQCaCW7IBnHeRhekeu0dU4zO1Fu/GP7Ev3O5lhMSLi4toV9yDYm0WJSMQ8Tt7kFutoXD3rQl+Q+n5MUZX9aJw+9+5Ze1cDKdH4Zu/NattYekYCJtZGxddBSZcsFguTBFraFNAgyDoKNl9Ff89tZyPvb2DMS07MC9hOkIs++VRhippcJ8T8OT6bO2FWtBRdbI87TDJnymCNW0KoJHXdyjZnVX/PbWw1tjb7BjTkg7zrie5wf6nxMOt31JEK6DOU5fxQZZpk/GP/2Nd//q0nHM1oeMxXhxPeVHArepYTsq7/UkxXP/ze5qU8kIUBMSi3/DT6qPcNr/Q0tYvdn8APhqV+WENS8plM6DTs/VtySkI5Gq/yfo8WPQzxwe4yNaRPuxJ5NLEbu1MTkFDkb67efTsFkdXzmeHbUaIPTS249j9BPhoUIk1mJ5cUCLWtKGARkDQlaT7qv+wVHkst3eMoWWHeY4zUMx5Pmdz57yoBZHKPyfu/2G6M5cGbgKCRxOWJrQxqxCmm0yXZxhoyzD8mCVA1vT4JLPaFEcnCHi228iL+BhirKQ0dWxvs7iJ3J48qDIiiAdGkF5eZUNgI4rllNuuC6XbT2XDiQgbAsm/pSes/sYDQe1Fhz+t0b7mVHpOjiyFVo4JmmIL+jVwepQv+VuvToxzMYQxs7YLugoTsD46GawXE7dn1kJ0FjiaXNq3fCa1dRDSKbfhHKPLirg3X8XjJAT/1Z4+eKtVCK5lGXbIHvu3rGaqxb1iTx9vc8yJa9lhOBcz61gkipA+giA9/Z0Wbmo+azKGed834fMqP3DK7t2QmF8WjzJKELJ7JUXDnW2MaPgl9XpNYMq4rtRvOIRt4bauQ0/In2MJnLwPc4BmFnX9cG6Xzf6/0bdhaT5zy0nhio1o16UdjSqWxLdGO0atvsjzJA+brFvs5k4UKNOY5lWKUap+VwL6tqf+l350nHmMR3Ydl/HaVGp66shZ+mu+Gz6MPi0qU7xkPQasueYQrRxzaTnfVS+Ai6BCEHNSoJA3xWv0ZuFZi2vAFL6RAfVK46WVd5GUH/SClG8QwFrbyk1OAI/Z3Rsfr2asSGnp1phLLP+uOgVcBFSCSM4ChfAuXoPeC88694M6lBHLtXUDqePjiiCocfPywiNnSVpNPeb4cpIes/+H6uTWCoi5fChRqhItJ+wgNPhXarrloUKLAMbP/pvLV9PQzxTOxgH1KO0lz5dXmWegFCzfgIC1STb3SUsnOZ/+NSnmoTbnI2i8KNV0unltglcHxlCrqAdqc/4avEo34ud/LKRRevQX3XxEsx9c9KpAt8UXcb67sYnwjQOoV9oLrbxzoUrAtWB5GgSsTVhkywJjDJeWf0f1Ai6WPHMWoJB3cWr0XshZp05oiD08mM9zNWaRfcyQ9IQ/OuSjQLetTlxFEg/3jKB6PhG1R1H86lSjUp2eLAhaRdeCGgRNDvKVbMCkf0PSha308AAT/AvjKpNKL28Kl2nKhN03OTWuPO4FqtB20E/M2+18i2Lj9eX0qFkcT7XcfgU0uUpSp99abv2ffYxp6Ese0dquXbyp2Hwq/+qNXJtaE09dTkp//R3Dh/WhReXilKw3gDXXEofH6a8X81PL5o6eaEuOcJx949Cu39WP1AhCbLrkNt2dTV1RJHmcCLwKGkBxnYyxmny997wrJdMo9xVBA2TyqkKlzkfvPfYDNtutWcMi/vl9IiMiiDB/7nLkf5Up3mcLtxPORRL1NLF9yaXqDw/ER6Pji5a98S/uiuBaifFnHdPYpLN8x/P8fqS1DFtZSb4jo0hSDLxrgmARXk90ZCjXbz1JEuTlqOKn+Mv48j5hwec5fzmMqOdxqZoHpRd3uSO/yKXXPAz9j0shD4l1QiQsOBp5eS+ES+fOcSnkPjF2BMIRZyMvo65z6fwlbkRFO/qhJAPxcXqMJkshksmIPi4eQ4plWhbjuRlqN2PCsbCEX8aXUVy/dJ5LN6KITi0cI+EOuwNTDFHXL3Lu4nXupaiYROyDG1z87xYJgb7SU27ffp6IcZr6SRji49AbTRYTv2TCqI8jPgUAUtZJzicegzUfC456ixxGPXF6I2aIzfnHo7erK1N0OFevR6VSfxZcJEN8Yj5ImOR84w3OXRPGl0Rdv8T5SzeIShP8eO5HPEywzthqQYqJJCIFUmFOE/eYm5cvEiw/89b2Ynh6i8tXw631nRFsjbwID+bSjcT2bnh0i/CXqTVEuS0aiI83WNuvjInlt4Tcjm3tWsJcH/HW+pCFN77kXsglzp27RMh9ed3D5H/prRf5zviHIVy/l9rLPXn+b+eMzcXQ0OleDOmR+/n6NnhqP2fYv0n9lBYNjDGPuLWmC7VGn3k7KmWqFCMxj26xpkstRp9x/jLKPBYGTk2oTuFChShk/eTPpUP0yJvwu1AhH0oP2G0nuYmb8tRy8Ut+umzEeGsRTbzUuPn9bLd+jF1y+dBwignVC9vlmVierdxCPqUZsDsJAcoAQegq78WQrS6GJDooPxUEFAQUBBQE3hcEZJdKabRiDX5N5pJKj4zx7A8ohPazrmxLKczAEMrinv1Yb50Wm55c30UaQ+hievZbT+bFTAcWZsXS42KI4c8OudDk6cFOM680EjK7ATnV7lSfcjkZWc8SXuklCC/W0cpdTU7rTK5sCVLMkuDKzQoCCgIKAgoCbxABiQcrmuGhdqXGL7Y4jAwUZ5AXLxNxbbzEvLpesjuNYawfOoA5p+WVPN/fP2PYeoYOmMPprKwCmRYWCeqngyDIqxaW1qJrMJ9IG3D6YKbVyIHaoza/XnFurUkoIiMH+mMEFtOlMYtB4t7vLflMLVJ5kiUWSSEIGQFZSasgoCCgIPAhIqC/zbZRjSmex5dOvwUR8ty5id2ZavIstQY6kYo//cvpeYP4OXGjDWtyI4YU3HHO8ntn54yG1N2m6RAsbSxsmaRNECyLwmn4fNi/DtaCuPPjqahT41l3OsFJZojZcs/w98u/6Oilxaf/IbtZb4m5mGLC+XfFAPwKeFOz3youWy1FCkFIxEg5UhBQEFAQ+KgRMEXf4uTuP9j0T3i69bRMXVWTw6ca3eeeSdibJd0ZfEQJ049FKgTBGMbWnwfToWpe80wv99LNGDhpe+Ly6/rjDP1cYw7uLlT7O5ZdTBJPkCE89ZxfEkC7at6I8sqTnqX5qtevHHxiM1nImUk8OruVjdv/4foTR6uFQhAyBLaSWEFAQUBB4FNDIJ4nEQ/SDKD9NFDJDiwkjHrHoFqDwWjnnpEDbeWAaQmTQY/BWfRsBsCWjPqEwGkkEwa9wWlArrMsFYLgDBXlnIKAgoCCgIKAgsAnjoBCED7xBqCoryCgIKAgoCCgIOAMAYUgOENFOacgoCCgIKAgoCDwiSOgEIRPvAEo6isIKAgoCCgIKAg4Q0AhCM5QUc4pCCgIKAgoCCgIfOIIKAThQ2kAxpdEXj7M5l2XUlin314RIy8jL3N48y4uOV/U3z5xKse2fPZw2bIfTyppP4ZLJmIfhXJq5xaOR2YxdDhFOF5x7Y/f2HDF+Tz0V9f+4LcNVxyXwk4xr+y8YCRk82zWX35pF02dnfm/7bw+PH1ehx9nw5rD3MtQ0zPw7MZRToRl14T5t11PSnnvMwIKQXiLtSPF3OHYxr+5kNJypU5lMXJtcUfK55U381Eh1pll2SHTaVowXltMx/J5zXNeVWIdZqW2K2AKecinjVcW0v6LPJZ8dA2Zn3Q3v1TufVeXDI+vcDL4UfLdJdMhkPRkB0P9CuEmb+aiKcKgI47zgdORRdpJpCccmdiNwZvuOCEAEk+OTKTb4E3ccc4d0s4/qymMkfw9ojtjgx6kexpUVot8o/dnVh/DY66cDOaRba+5NyqknLmR68va8bm8c6ZXF7alazc/iacn5tCzhjcuYlUmX3trwr5xNJQC3h8EFILw1uoinpPfl0ar/oz2fzzLcKnG4IlUElXoGi10vtypfY7GYCZWElHpGrHwvv2CGPaJ0nFsuMC4clpUuiYsS7rHazpuf6tJzGuNaxHc6jMnlV0oU5VJesyalp4I2pKMOJHdvfRrLvzShMY/n3fYadMmz+sLv9Ck8c+cf9d7/cRfYUYzfyae+khMRhnWR8+xocXRCm7Un5NkN09bZWX7t7y+v7wVsQpVetueFMlCfzfLoKHSRIIVfpDttaJkCApBeFut4MXf9CjhRS6tmpyt1jhuW5weGZ6tpJmLCl3jJTxMs89/xspmLqh0jVmSZmIDEYfnM7RtX1beTWLblB6ypLEOlUtzVj1Pj5BZS2OIOMz8oW3pu/Juxkewpii2BdanyteTOfEys3LoOTKoCBqtL6NOZy9BeHl0GOUrjOSEMwLw8ijDyldgpNOLmdUl8/fpL06gapn+BDnbizzz2b6zOzOmj4mobYHUr/I1kzPfkDKo6wvWtnK3bDWu0lJ+/MU0rWCGK5Pxk7dSV6nJ/91ep8vnZlAIJbmCQDIEFIKQDJI3cUIiamkzvhyynpkNXRFyfM2yjI7so1fTwkoQHqVJEKJZ3cJKENJKLD1gsb8OlVifeQk7hlgxkJ6w/GuZILRgdfSbwMU+T4kHi/3RqUTqz4t8R35wPf8MKZb9BEF/kR8r5qTu7DtOiI+eiz9WJGfd2am6juyReuPH0j2WfpWLsqNOfRzbtr/v+hhOMdLXC6+calQqFdpy47iQGj+VHrCxQxHcXWWCIFJnlrN29cZbiVLAJ4BA9hAE6RHH1i9n/R9/8EfCZyNrFi9nf3iSUeknAGoyFY1XmFStKj9e0hO1yB83wY1GC8KddBbJ7kw88aYIAs9Y2TQFa0MCQWjJ2kyPyhNVSOvo2cqmuKh0NF7y8CMiCBLPt3Qlv6660612pedb6JpfR/VfwzLWHtICM0vXJSLnN8Dls7asT4tgZqkc+eY4Yl69afv429Qn44CYwn6lTuUfWDC4BBqzm+ELxqbIECQebulOpWZjGFhNRKXOQ6/dWVmrP+PyKnd8OghkD0HQ/8vwElpc85Xiy8qVqSx/viyGV74ubHnx6YCZkqaxhwfhW+83bppAureUJjkEXGr/lrg5h5Mbpehg/poWyLftWtK6S3/GTOrBl3IMQuMlJHtnS9EE/zWNwG/b0bJ1F/qPmUSPL+UYhMYsSZbYrrDXl/lj2hjalNaafZ+tfpjIxIkTmTR/P+Yg/gSC0Ip1MWB6doktcyYw+Nv2tO8xnNkHI52aQqUXwWybO56B33ahx4BxLNgbRupxma+5/Mc0xrQpjValpWSrH8xyTJw0n/2R/4+Y8HMc2LyUyYM7M3LzQ0yvbrB5Ym/adxzK6iux8DqKS0e2s2bOj/Tv9APbzBuRmHgWdoI9G+Yzvm9nJh2OBeMDzmyczvBvO9H7++lsPPs4ifypWxAyrpeM9TM2tvNC69OfQ07iHp9tbIeX1of+zi5aqyru9j7m/a8fndp2pN9P67mUBdO/9PI622cOpVub5rRo3ZneA0fwv2l2G8VYyzScHEFJrQfNVj60azDZf6g/N5ZKvt3YlMHIzLgbO5k3ZSzDB3xH/zkniMHEo2PzCezSkcC11x12x5OlTp8+r4m6dITta+bwY/9O/LDtiYWomp5z8+ReNi34kX6dJyE3JV6HsW/xzwzr1Z523YYwfc+dZGWmDy0TEfP8KTfyJLGXJlBeVJmtAhV+/C9J27TkZghdQpt6A9kTdYQhRTWo3oA7LH1yK6k+BQSyhyBEr6dL41+5ZvcCjNnTlxoDDzoNyPoUgE3QUXrKpg7FaLHivuVlIz1gRTMPBNGPqTecjZwkos/MpnXpUjSbvJPgiEhCTqxjRM3c5iCmpARBij7D7NalKdVsMjuDI4gMOcG6ETXJrValTRBirrJ71Xz6yGRCLEfPOStYsWIlq3cFW3ZssyMIC/ZPpkm5inz97XDGBDajhIuA4NmYxRH2FiKJ56d+o5VveTrNPcSNyNucWt6FkqInVcf9S8phbzFc3b2K+X2+RFSJlOs5hxUrVrBy9S6CXxgI2zKern750AhafIevYW7rKpQr5WX+XWb0WUyRe5nasxbeWgGVS1NWmmNA47n0eyAtv8iJWiVSc8KfzOrRkp4/L2XV3FE0L+GCIBam3aowuxkFKRGEzOolD5B30yuvGrHWTCeEMI7dvfKiFmsx85Y9jgmtB+PNZTT3qcDA7Td5HLGdfiVEctaazhVnTSfxNqdH+tubCahSjrazjhIRE83dYz/TwEPAxX9xssBX6f4i/HUCuTpudppXtp2UHrE7wJdcZfqyLco5Bs7KMkSd4e8F3SitFXBtvpIbOwIon8cDnaBCrDndTMbt70uXPqZI9k7tSS1vLYLKhaaWhgTxl/g9sDXlcqlRiTWZemAdvfzK06DzEEYPb02ZHAKCe11mhaVf/gTZpPssbfYFgf/Eg/EGU/2ss5WqTibpxATp4R6GNO7A0ht6eLme1m4CKrE6v4ZmotwEAZQDBYGUEcgeghAXwfVbdvOnpUesbV+LMedSc6SlLNTHdMV0ey4NC3dhc0KQn8Sj1d/gKYh8+dPlZKME6el2ehYW8em9G3u3v+nWDGomtSBIT9neszCiT292OyZmRs10WBDMQL/g9+ayi+Gr5DMVbARB7UGpNlM4fN/WK8Vz6ntftIIHHf5KjLqTHm2mayEXKo6/kOi7lh6x/Gs3BPfGLE5jquSL35ubXQxfLXuczMVgvDSBCqKA6FmCzmvDMcbeJGjDTq68sAZkmML4tYaIyqU16+3MFXG7epJHLaAp0IwFVxLnjxnvLqP5Z2qEnF+xOCE40zlByIpexhtT8RNVuHyzhmReGmuHoHL5hjXJLsqVY+DMD764fDGW8+ZHyUjo771o1Po3Lmbw0TKELqe1Tx7qTruADQXp/kIa6bR8MfaCHUmyPn1xm+noIZj94W/8eYwLZpZ/bjwqDGZvmkG1idIYQ6bhJ4qUDxhLZ//hHHzymFPLp7DkePL2Q7r1MRH2aw1ElQut7RsS8ezvVxC14EqBmgPYeMOGooHgiZUQBVea/57wkCcKmcaR6fZsGvgO5qjZS2Di1qw6uAgqBLEG0+0Ih/T0CKP9v2bSaUtDsb0PVDp/FmU0nikNmZTLCgI2BLKHINhys36bbs/Bv+6URAYsPWN7X4LtrQAAIABJREFUQGm8K4/nhJ2VIcltH+FPA+fHlsOzzmi2BgURZP3s2xRIFZ2AtuxoHDmUidDpNdFpChFwIIlf0UkMgil0OjV1GgoFHEgSxZyBIEXSQRDEesx1sBRAzLpWuAg6/Bfb4gWMXJ1cFVFTjO5LD3Lo4AGC9u1l967NjG3giaDOR5+9SXRKUuOpEQRT2HRqiCrEKpMS25X9/dJ9FsnBli5t2WS3Zoz+yCAKazR8PvR4EhNwLEcGF0cjiNRI8P87IwhZ00t/LJDiGgH39n8lkiab3PpjBBbXILi35y87mW2XZYJwckRJtG5V+d/xZ8lIU2K6NI6M15hRx5McdWZiP9iM3x+AtzYvPXYmkryEnOJ30TO3GnWBvgmnUjowPbjA3h3b2b49859tywdSOYcaz6qjOOSwV31KpUL0xnZ4qt0oUa4Z40870cH+1nTrI3F/kRws60Jb+4aEgRPDS6BRF6RvkGM7jtvWlVyCSK2ZtzIYR2LkyiQ/Sg4+kvD8SpEL8ZctA4Ir/ovumetcij7JJP/aDNrzKKENGM7+jzJa2UrYhOXpxMseDuVYQSA9CLwBgmDk0o9+NJptP4f4NWfndqVVvzWE2Aah6ZHuQ08Ts5fvfLzwbdiSNm3a2H1a09DXE7W2JMP/tWNM0hNWNnN17n5IRhAknqxshqsg4jf1RhJLRDYTBJdmrEqydEPcn+1xF0TqzrZOSbR20PIaAs2HjmLU998zcsQIhg8bxtDAQIYETuDPNCo/bYIg4Nl5a/KOVm4nqRIELWVGn0s2Sn71Z3s8BDV5eu22vqCdEIQs6mXuhNUCnp22JJc7fj8B3moEz05scUoQwHB1JnVzqRFcP6fNrNM8S3MGS9KHRuLZ1q4U0HxGhz/sSYaBC+PKoXXxZ/EDJ5nG76VPPjWCV/ekGSb7Hb8ngKJeuciVKysfT1xlF5E6H61XRyQrI/kJPceHfm52MxUPCHKwtiVPC6RbnzQIgqYEw5OskRG/uxd51SJVJl1N8hw6lSTx5OsjDC5VgiEW84HlvCmCeQ1cEFQCOdtu4NmLU0xuUIrGc64mkAg5oUw8i2lUqJw8m4kFKEcKAllDIPsJQuxhBn3RlCX3nLx0sibrB3a3xIPVLclbdRJXnZCimO3fkk+toejAw4kdhymUX+TIZLEav9gP9WTNkxEEE6G/VDP77Kv9Eppk5PIOCILxMhMryrLXZXaCyT5jVZYegpCry7ZEvOyzzwRBsLzYNfj0P2i1LjghCFnUy3DmB3xlP3m7TcnjcQxn+MFXi+Dajk0pDoBNPDw4ngYFRAR1TvzG/Zt2Z2iPC8/Z2N4LdY5vWP3U7pl8sZ+AzzUpT6mL2053LwFN0cEOub2ZHxJPDwRSzs2L2hNPWOJf0irIdJvfaouo83dgU3rcEunW520RBBORS7/ms3I295FNYSM3pvohqlSoc9eiea28FGj1O0kng+mPDqGoTBDSCkS2Zat8KwhkAoFsJggST//oSLFvVqVjMZ9MSPsh3WIM4dcauag7KwWz46td9CygRlOoL/ts7kzrmgSCpnDypX6TEQTrugGChsKDjiQxn78DgmAKZ049edpVbjr9lXFfrFy1b5sgPF/VHBfBhYbzbesuOCEIWdRLerKSZm4Cuq+WOpl9IluM3BB0X7E0tdkmcjRC+J98+7mIoPNj6nUnjDOlZ8NwkpGltGhLjeRkQtxCNEeGlkEU1OTtuSs5cZHzermOVq4CuoYLUso5287rQ5bQrEAuqv5wGHsOk2oBMX/RMZeG3F23pTFDxppLuvV5SwTh9XGG+7pT/ZeQZFYH/T9W64BKheDRkHlOAlgNp0dRWnYxKMssp9pMlItZQyB7CYLpDnMbFqT1Wuv0IHvZjC+JfvXpRNvG/Tuc0jmbsDRFS8pr9vQpiFqdj2+32yLUrGZflYbig444vLilqPk01KkQ687GNkA3XBhHOa0KTfFBHLEfgUpRzG8oL36UntG8LQahEQuTympbRMmJn/P1xrZm90atGTYCZODs/8qgVQl41J9FiJ3nxNwMJCnBf2rfLOyPbQSh0UKL79X+miXYTyBHx80pWBAimddAjkFoxTq7IEVLDIIW3x/OJHExvGRnz4JocjVlWUJ8hZ7DA33QOCx3m0W9jFeZVEVErDiRy8n6dSNXJ1VBFCsyMdlFA4+unCE0YUqjRNSChugEL7pvT/SBG1/c5PzFiJQ7yfj99CuoRltyBBbLuJHwDV2pU782RbU6/BfdJ/beXR4kqS/TTTnmQ0vZMefsqyHbj6VnhxheITflB+3J0KDCMm3Rla+W2mJgUhct/fpIRM5rYI5BaGXfkNDzT2AxNJriBB5zBCt+x7d8phapODF50LFzqeK5OLEqbh6NWexkUzDp8TKa6OTpjhqKBOx3Wrem279RW54Sqfai02bb+8N5acpZBYHMIpCtBMFwfizl87RjQ1JHqTyroY0Xok8A+xPfbZmV+f2/73Uw0+vnRFdzRrLpVvbCv/yzAzkFgZyNZnPDiovp9mKaeKkRPKowIugBRiReXt1AYKNi5NQICC6laT91AyciDGC6zeImXqgFD6qMCOKBEaSXV9kQ2IhiOTUIggul209lw4nUfLqxbO2cE0FThL67H/Hs1lFWzt9hWdXPcI7RZbWoxMpMcvCTmLg5vSaiSkOxIUcTfKOmiDW0KSCXq6Nk91X899QyZI29vYMxLTswL42Rb+zWzuQUNBTpu5tHz25xdOV8dlg3m4rdH4CPRoVYYzp2wd2JcOotI2WV6McUu3IsBEEgV4tVdlP5JJ4eHkmlHHnwt/ftSk9Y/Y0HgtqLDn8mLuCRNb2MBP9cCTFnR/6yIy42wY3BP1NJzEnHpBfj/qarl9pObj0XxlVA9PBnoY0hvtpDH2+1OaCt7LBDtiwdv62zO2TLw49n7nJuaTf8Gkxgyy8N0anz0n7mIrrV+pa/kuy1ESf71UVffjiTYHZwzDebfhnDfmfY6G1EJiNPqRVg4vbMWohiFSYlnQuYwm3p10fPyZGl0KpE/KZctxvdP2N925wIQi7ab0psG2CxOLgIagr22ZMwQyQFMcynYy9NoaaHliL99iWf2SKniNtCJ0/BvGnYwMMpBKfojzCoiMa81HLe1isIjY5L4mZMTQLlmoJA+hDIRoIQw97vfMjdfiNJ4tnkkBpC/hxL4OR9ZGCqc/o0eK9SGbm8oANVfNxRy7sCCu4UqtiYH80rq9gJqv+XqS2qU+ozy5xnlaDG3bsCbedfkaPteHhgAv6FXc0drZd3Yco0ncDum6cYV96dAlXaMuineewOtbw4pIcHmOBfGFdBQOflTeEyTZmw+yanxpXHvUAV2g76iXm7Q+0KT3oo8eivbviIcuS0iFeFbiy+GM3//XcqTSvkN88rV6kE3IvUZPiOJ0jSU3YMr4aPHGktm0DF3JRt9iu2rQtiLi3nu+oFrFO1clKgkDfFa/Rm4Vn7ALmkMlh+S4/+opuPjImA6FWBbosvEv3/wtk4oB6lveS56TKmrhQs34CAtYlBsNH7xtDQ17rzpErAxbsizaf+a3a72AiC6JmP0rU7EBA4mD6tq1HCtwmjt4YmvNCN15fTo2ZxPNWyXgKaXCWp029twtoFWdHLFL6IxrmKMfio48jTrLUpnEWNc1Fs8NEkbqIYzs1uTcncpWkxZBSB3epQomhthmy+nWgJeRXEgOI6My7qfL2dg4qJyI2dKWKuXx0+X03iyGMDN2fUQicIiAXqMu5g0mmBlumVHrUdZz2kUMA7OB3L5o6edlaRtERIpz7R+xjT0Jc8MlZy23bxpmLzqfwTtpH+NYvhYWsbXqVoOv0sBl5xYEwtinpYlkgWNF6UbvQz8pIGKf0Zbq+nczERl897sjmF6Ymmu7Px9/Gj95LzJBiQkmX4mqAAH8vKi+rc+M9PGqic7AblhIJAhhHIRoIQS9SV89x4mMrTkWHxPrwbTIZ49EaTZXqSyYghPo74ZKMjE/q4eAzWdEgmjIZ44gx2LhjjC8KDL3HjYazVNG/g0a1wXtrFmSWiY+RFeDCXbjwk1nrd8OgW4c4TJ96WcGQiOvwq16Pk9eisfyY98XojJjk/ScKil+WaUZ8ou2Qyoo/TJ95nTmLkZdR1Lp2/xI2o6MQOzZZ3Kt+m6HCuXo8iJkEQyYyhDVMzVvo44g12QBj1xNnLapRlt2RgczHICyrFPLlN8IWL3LgfYzcytAojGYiPN2C0KIzJaPltV4p5W97M6fWKf4aXpUT/AwmExB6CV/8Mp2yJ/hywxaLYXTTG3OP6+dOcuxJOdLJ2JIsUw6Nba+hSa7TdXUkPJV7fu8blsMcJ1h70jwkNDuGhrcHY3xJ/ghFlShIQZD9Stk/w7o/jH4Zw/Z69Xy0VmdKtj6Ut29qAuW3H6zElaxtye7O0L6M+zuF518vpnYpi5OGxWXzboDZN+y/g1BPnqeRbpVcviE6H4cZ0exPf9xrKjJ2hTt0QTsVQTioIZACBbCQIGShVSaog8JYQSCQIyac5viURzMVIz4MYUKUJ88Kc9PLSc4IGVKHJvLDkxCVNIQ2ELu5Jv/WpuZHSzMQugYnwFS2p1Gtb8qBKu1QfzuHHps+Hg7wi6YePgEIQPvw6VDRIBQH9of74aLT4jjqdIUtGKllm+tLrc9P4ps1M/nM28H19jmnftGGm04spFWkkbP1QBsw5nb6pgSllY3def2MRHZpN4N+PJO7tY9PHrqqUQwWBN46AQhDeOMRKAe8SgehN7ckpT+f7dsd7YIaVeP7vL/Tqv4zLTgIWpef/8kuv/ixzdjEFEI0GQ5qzQ1K4Ndnp2BtrGdJzEoeTBCwmS/iBnPjY9PlAYFfE/IgQUAjCR1SZiip2CJhus2VcF2r7yKvSqRBci1K30wDmn0h5yyi7u9/ooSHqNGfuOHE1yKUaojh95k4mXA1ZFdlExLlThH80IUQfmz5ZrV/lfgWBjCOgEISMY6bc8UEgEMvDsGtcDwnl5s2bhIWGcOPaNcJtmzt9EDooQioIKAgoCLw7BBSC8O6wV0pWEFAQUBBQEFAQeG8RUAjCe1s1imAKAgoCCgIKAgoC7w4BhSC8O+yVkhUEFAQUBBQEFATeWwQUgvDeVo0imIKAgoCCgIKAgsC7Q0AhCO8Oe6VkBQEFAQUBBQEFgfcWAYUgvIWqiXtwhaN/r2bR/MWs3XmGSGcL5bwFOZDieXr7PEF/BHH9o5nO9jaAe3/LkPQveXjrIke2riMo1MleD++v6IpkCgIKAu85AgpBeJMVJD3h2OSmFHHzpHzbHjTwlndfE3At1Y8djxxX+X+TYsh5xx+fTJMyn6EV5O2hh5Jkx9o3Xfzbyd8UxdldO9m1ew979wWxb+8edu/cycErto2iDISf2sGOXZbrQXt3s3PPeR6kvCz+25E7E6WY7q6gffFciIJlYyGVygXH7Ykzkalyi4KAgoCCgB0CCkGwAyN7D42ELGhMbrUKscIELhmMXJ5YybI7oqYEw0+kYzeW7BUI49VJVBEFvDpt5t0vF5SonCkmnNPbFjNj5nJ2XXmWwmY3ielTPDLc5vCqOYz5pri549R612PAlEVsufDUutpgHGH7ptO+hBa1Zym++m4ss/66SNLdyVPMPxsvmJ79x+nrWawFYyRLvvaw7nLpQcfNKWwNnI1yK1kpCCgIfDoIKAThTdX1i610zSdvA6uh8MDDlq18Y0PYMmkoQ6ftJeIdjFrj9/ejoFpHg/mR2bY8b1bhi/1vPi2K+NBg+EJWzBpMvcKFaTb7YpaWRX69uRM5BR3150UkIRsGwjf3wa/mYLbcejc+FikmlF2/dKdqXnf8F97PYj3oOTTAuuWv4EnnrQpByGp7VO5XEFAQSERAIQiJWGTrUdzOHuRRq1CpRCpNDH4HS+cmVcfI1UlVELVlGXPu7Vsvkkpj/v3qKMN8XSncYztPzB4XiaebO5PftQJjTmc2UMPIpQnlEbUlGeFgpXlF8OL2VPt6Kiefv133jllXUzh7J3WhZplSFM0jIqh0fLXscZYJwpFBhdGoVKgUguC0iSknFQQUBDKPQLYShNi7x9i0aDrTZi5l24XH70GnmHlgsnanxL0FDdHJL26VSJ1Zd5OMZLOWe6bulp6wqrkb6rw92ZXZvjdTBad0k4nwhf7k0BZjyFG74LrXO+mRT4NXuw08zVQ//pTfW7gh5GjD+hfWsqUnHP3Jnyrtl3DldUryvOHz0jOun7vBM6PEo6VfoVPp+HqFzfWR2bL1KAQhs9gp9ykIKAikhUA2EQSJx3sGUbFSd5b+c4WrJ1bR64uitF0T/u47xrQQyObr0rOTLBs7hK7V81tGdioNBap3oFevXvTuN51DDg7vOO4eWsyYXi1pWKMKlarWonH7AUzecJ7HDnv5vObB9TMc+Gsp00b2oHmDTiy6bsQYuY+fO9WjZtNhbL7tcENyreL3089bg2uTZTxM6HjjiTyygOFd2tDtl394qwNrUwTzGrgguLdmXbSduKZw5tQTEXK2Z1NmthzWH2VIUQ1i5UlckSHR32FzgB9+ff7gth0PsSvxrR/G/dEOV0FH0xVPM1i2gfsn1jB1ZC9a1fOjSi1/6pS2xSA4dzHE3T3E4jG9aNmwBlUqVaVW4/YMmLyB844NLINyKMkVBBQEPgUEsocgSA9Z+pUHFX+6bLUaGAn+uRLu/ot5kNAZfQpwgvT0X5ZNHMcgf6tvWKWhSJNAxo2fwE+/buaqbfRuvMtffcuRQxDI4TeCvy9c5NDyAKp4CKgELfkbTOK4rcc2XGBBl5r4uFoj1gUvuizfSn9fy06FspWi4sTgVAE2XvmZSqIlnZlKxN5gQ//qFM7nhataTd5269OoKyNX1o9l+LBhDEvH5/tlZ7Cp6lSw2O10+0yNplhgkhkVcfzVIQeCtjSjTmfcFWIKn0NdUU2eHjt5/fQkM78pg1/g3vdqpkLcH+1xyyhBMEWxe0Q1vDQiPi2msOPMFS7t/ZWWRWR3hTMXg5G7f/WlXA4BIYcfI/6+wMVDywmoIhMKAW3+Bkw6/jyLLg6nNaucVBBQEPhIEMgegsBzNrbLRc76v3FdHqWZ7rDoK2+qT/mPjL/iPwZkTYRNr4FodTHUnZ3UxWAkZHZ9PAUVKk0RBhyyBZeZuDOvITnk8yoNhbpttfrmZUziOTTA6m9WqfHy7cikVbPo7ZePHHnrMvWcLQ9n+Ek8XtEUN403/fbHY3pwgPFN6tB9yQWem4y8iAznSZqjawOnZ3SgWdOmNE3Hp+WPB1KdKSFFzqOBToVY8ScuOxg/4tn7XX7Ughfdt6emkzM9IXbHt+RWq8lVohK+sq9fW5oRJzKej/Pcs+dsxglCPBen1DC3F3XBnuxIsKyk7GIwhsymvqdMKDUUGXAIGwKmO/NomMNCNDWFurHVEvyRPYopuSgIKAh8VAhkE0GA2Eu/4Z/PhYL1BjG6T3M6TDnEfduLX3rG9oDSeFcez4k0O6KPAd80CILhFCNLaVGpVAiuLVlrZ2KXHq6gmWxFkK+51GP2Hdt0Bzn4rkIC6Ui01qQHrzj2fVcAjZs/k7fOokP1xvx4JKsBcukpN+U0ppBpVBNViNWmEWJT0Zw8nn1mguBO+z9TtUE4yVyeSloRUVOcnst3sqBXOdwFAY86M7ieYaZq4sGFvezYvp3t6fjsOh2RbjKcUYIg3VtJi1yWNuHSfBXPEzRPiSAYODWyFFqzZcGVlo4NjBXNbG4JF+rNvvPJuQET4FMOFAQUBFJFINsIglxK9D+jqFbQC/dcX9B5xpFEgsBrzs7tSqt+awixkYZUxfrQL6ZOEEx3Z1NXlK0EKtR5erLLfsadMZiJlUTzNZWQk85bbGM/I//9aCMIOposf5J+87DxMj9VFFF7FqZ4Xhd8ms/jQhan4Ge1hqQHi/CXLQi2WIGEDOPY3t0LQZ2f7/baA5OQIJWD56xt6Y6QoxXr5ABFw1V+rZUDQe3F10sy2hHGsyegKF65cpErHZ987dfZddypiAhkjCBIPF39jdWqpOazb3eQiEoKBMF0l9l1rW1InYeejg2M4ImVrERTIGfnLQnWhdSlVq4qCCgIfGoIZBNBMHFvRyD1m4zn+LNXhP45GD8vF4r33JqGX/tjhTt1gmAMluMBLARB8OzEFvuBsnSPhY10VoLgSqt1tp7ckSB8nQGCID1azteuGrz7BfHy7GjKii5U+ulSuke8tloyxr3iVUwMMen4vIpLgwnG7aF3PjXa0t9zymF0/4qNbd0QxCpMuppGHjbBbN/64wz9XOPgtog99T/KiQLq/O1Zf//9CIjJGEEwcuVn+w59q12HngJBMAbzcwLJ9KSTYwPj3sJG1hk2Aq6t1qXqCrJBq3wrCCgIfHoIZA9BeLGVbt5l+eG0zX8gcX9Na3K71GLmLQf78SeCcOoEQXqyiubuFpOxSvRj6g27jtAUwdz61tGfpjhDj9swzTxBiNvTm/xqN5qveoJkCmV6TR0a757stHNtpF0x8ZaRvTmuwkJuZAtISh+XZqt4llqm5mmX7qg92/OHjQPJ6Y1XmVRZROPTn4OJQ+XUckq4ZolrUPNZt78TAySlF+ztVwyNvGBVzx2ZnDqZUES2HGSMIJiImFvfOuJXoWswj8gEnpMCQbBiaw5eVIn4Tb1hN+XYPj8NxYcetyzilS2aKZkoCCgIfEwIZAtBMJwbQ1mvTiRYwwEpaj4N3L5g7AXr8ND4kuhXnwpZMHJtctWEeIHqv4Y6+nml+6xu9Rlqs49YR60ZNxOvxx1iQGF5zwYVmhKB/JNgXTBwfuwXFr+ySof/4ofpdDFYiYVYmUnmeX8mwhc0wk3tSePFSVcaTK1pS0TfvczFCxe4kI7PpVtPE3Vymq3Ekz86kEcsTuA/iUxAerqetrlESo88YWdKN/Li5nkuRrxympPtZNzuXuRVi9SaeduhbOn+BtrnVyOIpRh8KCHCz3bbW/+OXvMNLkL669BwYZzZCiK3CSHHVyyJsjEEPUcHF7EulORK6/U2piVxf3UrPjMv1CWgqzWDmwmPXlxisKumBIGJDeyt46AUqCCgIPB+I5AtBEF6vI42uX0ZdizxBfV893eUrDCa0/K7X3rE2jZeiD4B7E/sC95vZLIk3Qu2ds1nIQAqNQV77CDpYN14cxVtfLTmKWoanw6svysTKSORa9qST61C0JWm3057EvCSbV3zWvPUUnb0ufS5CKT7LGnsiqZgvwTspWd/8623Bm2pgRy4F8zCgLHsSypglvRP5836y/xSOxcFO26yuqJiuTylJrmK9+Jvu82sXu3pg7eMiWtZhh2yrX6UtIx4Tv9QBq3ay8nsBxOhM2rhKgjoyg7mwDuN3H/N0cCSaFVaSg47bucuSKqP3W/pHhs7eqMxz25Rk6fuCFZu3cTswY0p6WlpQyp5ZkutYaw+EWG50XiTVW18zJtzqTQ+dFh/19xejJFraCsvAS7oKN1vp92aGHblKYcKAgoCCgJAthAE0HPzj0HU/bIuPcZOZcoPPWnaNIA1123DXz0hf44lcPI+ohJGMh8n/sZrS+jqVxh388vcaoJX56BItZ6sDLNzJchr+EQEMbWLHwVd1Wi8SlK95hfkc3HDp8a3zDjyIMEsbLq7nr7ViuChtrol5JGkmAffBuM4kNbKgPFB9C2gIVerNTy2DTwxEbmxC0VFAV3eaowMykDAYzZXmzFiJ6Pql8DX/1v6tK9Jufoj2RnhEJTAq6ABFNfJuqvJ13tPEgmMhG0YStsGZcgtymkE3ItUp0X3uZyxeWfkSaK7epo3zpKv5yhal8AtkQ5WhiSZvoGfRkJXfUejSj64W+tREHNTulZzfj6cViUCr4JZHVCdvFqLjroC1eg56yAnZ9RDFLR4eJelVtOOBMw8nCi7PoKgqV3wK+iKWuNFyeo1+SKfC24+Nfh2xhEeODbHxPuUIwUBBQEFgewjCFYsja95fCeUsKjo9I1uP8YqkAzExekxmiSrC0DCZDQQHxeHIaGDdlRcin3EzcvnOHvhCrefODGxSPL98RiMpiR5xieQCMcc7X8ZeBp2hTvRSZmZiZcRoUTGpCCUfRZv/NjAi/BrXI+MSbHTNsY84taaLtQafSaZNCZDHPF6IyazKjLeeuLj9I55GQ3ozQkkjPo44t9B5ygZ4oiLN5jbhqyEZDJiiM+YLHGPQrl48QYPrZNbpJgIQu68SP15k2J5dPMy585e4MrtJ3aum2RQKicUBBQEFAQSEMgmC0JCfsqBgsCbQcAQyuKe/Vj/LrbBfDMaKbkqCCgIKAi81wgoBOG9rh5FODMCxjDWDx3AnNMv0hmYqeCmIKAgoCCgIJBVBBSCkFUElfvfAgJGDCn5Z95C6UoRCgIKAgoCnyICCkH4FGtd0VlBQEFAQUBBQEEgDQQUgpAGQMplBQEFAQUBBQEFgU8RAYUgfIq1ruisIKAgoCCgIKAgkAYCCkFIAyDlsoKAgoCCgIKAgsCniIBCED7FWld0VhBQEFAQUBBQEEgDAYUgpAGQcllBQEFAQUBBQEHgU0RAIQifYq0rOisIKAgoCCgIKAikgYBCENIA6IO9HPeAK0f/ZvWi+Sxeu5MzkbZ9MT5YjT4SwY28fhLOlX93sWH7RaLfh5WuPxJkFTUUBBQEshcBhSBkL55vPrfYMHbNHs+EufsJt9uMKLFgiSfHJtO0iBue5dvSo4G3eTtgwbUU/XY8SkymHL1lBGLYN6IKPjk05h085a2bxWrTCEm6RcZblkopTkFAQUBBICUEFIKQEjLv5XkTN3+rjYu8U6TgSYtV9ttBWwQ2hiygcW41KrECEy4ZMF6eSCXzTogaSgw/8V5q9T4KZYoJ5/S2xcyYuZxdV545bvyULoEljMbkvf/rQwMpprHs8inW/o07yZOkK3clkYKAgoCCwJtGQCEIbxrhbM1f4unW7nhrBARdKYYcjEmS+wu2ds2HWqVCU3ggh80WhlhT/NckAAAgAElEQVRCtkxi6NBp7FU2OkqCl/Ofsf/Np0URHxoMX8iKWYOpV7gwzWZf5JXz5I5n42+xZ8YgWlYuRIkBB0lq5DGF/kI10UoQ6s7mrkIQHPFTfikIKAi8NwgoBOG9qYr0CqLn0dVTnLsTnXzjorid9Mijxmy+rjSR4HewpXF6tch6OomYF9Hp2O46gyW9OsowX1cK99jOE+v20U83dya/awXGnE47jkP/zxCrhUBLyREnkm3DbAqbTg2FIGSwUpTkCgIKAu8CgWwlCPFRJ9g4bxrTZv/O/tCko9t3od6nVaZ0bwENddbRaZ1ZH/noNJ4dPYrg22wYS/6JIC5bqtpE+EJ/cmiLMeSo3dj/9U565NPg1W4DT9MIKozb1oWcsgtIpaX8+IvJCIxCELKlopRMFAQUBN4CAtlGEF6dm0qDci2Zceg/Lu6dQatSJWi/MizZCOot6PRxFiFFE3Z8BxuXz2HKmMH0aNmILgtvWDog6Rknl41lSNfq5Lf6tzUFqtOhVy969e7H9EPPklsbkqAUd2Mn86aMZfiA7+g/5wQxmHh0bD6BXToSuPZ6MlN5ktvfyc/4eydYOaolZb1y4FO7N7/uuJG1WQGmCOY1cEFwb826aDuVTOHMqSci5GzPppd25+0PpWec+308AY2KI5oJgpo8VdrSU66DPiNZd8NizklOEEw8vbiJSd81p1a1ujTv8SNbQp3THen5Zf76dTg9vqmLX9WaNO44jHkHw4m3yfH6PldO7GXT4imM7NmCBh3mc80I0svrbJval2Y1q1KtcXd+3nXnvaxPmxrKt4KAgsD7gUD2EARTGDNre/Llj/9ZR0wmIhb545GvC1uevx+KfvBSmCLYPakrfvm11ih4LV+MPW8hYNJT/l02kXGD/PGxEYQiTQgcN54JP/3K5qtpm8YNUWf4e0E3SmsFXJuv5MaOAMrn8UAnqBBrTufme+wrNz2/wtYp31Ld253cFdozdu1ZHhoyUeOx2+n2mRpNsUCO2RkQII6/OuRA0JZm1OkUMjZFEjTnR8a0L2slCBoKNRrAuPHjmTBxPgejLAA6EISyHRjWxY/iRUtS9DOdtV4FRN8R/JuEI8ReXkTrojo0Bb5i8r4LHJvX2lzXgrYQ7daGm4MoDefm0rGaD65mgqJC8PiGaetH08S3MMWL5rYEt6pUCLovmXDeQcFMgKXcoiCgIPCxI5AtBEG6vwh/F3fabkwM45Kj5yvqChFwUHkRZV8jMhEyrRqiSjZhi1SaGOxgwnbofDIRAGcMmYafKFI+YCyd/Ydz8MljTi2fwpLjjy0WCOMrHt78j+O71jBvzcksqCUR++QOV07uY+PCNRy3OPuzkJ/11ti7HF44hCYlc+L5eWMGzT/Indfpz1aKnEcDnQqx4k9cdojfiGfvd/lRC150356k506SfdymdtYOWqTqlOsO9SMnta8jTaGmTD1wFzN9e3WeCVVdLCRBU5TB9i6O6AMMLCkiqET8bHkaLjCunNYcb6ItNZJTNt4Sf4gBhTXm8yq1F1W+W8RJM1t6zbmxFa3kRUuZ0WcV616SulN+KggoCDgikC0EwXRrBjVFHV8tfZRgyjaFz6Gu6Eqbja9Besb2gNJ4Vx7PCYUvONZAhn5J3F/oj85KEKpMuurQAdl3PmImCEL0xnZ4qt0oUa4Z450E5Jki9zK1YwU8BDXe/fZnSHKHxNIzji8MoE4BDeq8PdmZtoHD4fY0fxgec2HDBDpWyot7AT+6T95C8PO0TSCmkGnmGQbJ1yeIZ5+ZILjT/s/UhX21rhUuCfVzzaF+ZLlTriMDJ0aURCvfK7jSZoON2Zi4PbsebuaprTlpvyaKx48f8/jRdWY20FmIgK4RC+9ZgyOMl5hQQbSe/5rlduQr/kAA3mqZXAp4dd+eTXEbadaGkkBBQEHgA0UgWwgChnOMLafDvc6vXLM6RI3BE6kk5qX3HvnEa87O7UqrfmsIcRiZfaCovTOxJe4velMEQc/xoZ+jEbQUDwjC3gVvr67+8EAKazxpsz6LviPTTabXFHFvvorHaQT+2Zef/mMjT84v49sybqh1Bag18bhlpJ5KBtKDRfjLFoTKk7ji0E7j2N7dC0Gdn+/2Jnj8neYUvbpFAkGoPOlKhgjC2f/5WgiCyoWWa63BDtJjln/tarEsCG54l6lM5cr2nypUbTCKfbYKM/7HjykQBP2RQRQ2u6AEPDtvVQiC0xpUTioIKAjYEMgeggDEXt/I8GaV+bJ2W/oMGUlAw0JocjRj5cM38va3yf+Jfb9BgmC6zW+1RdT5O7ApxTozcWtGTUSxNr+luMJPPI9vhnDfNgBOqYZittDZS0f1X0IzsQhRSpnK52MJ///snQdwlMX7x++9u/cSQhICSA0o0lRAem8SOtKM9P4DBCQ0FaliAAWVFmnSBVFBkCZIGxQpUiNDrwMpZCBABgID5H93c/fO5z/v9UtyKRogwGYmc2/dffaz+7773d1n390zj4gWZQgOKkXY4Ch2ZHWcIeU3+hbSoi/3kbvL3hbVQ1Z3zIMkVyPytJdySGNI0pJWLoGQeghIvTijHoR0BYLlFJ9XcfQI6Csw7phzLCFN1PYDQiD4ACMOCwKCQHYJ5JhA8IrYeo2oJoEU67mB/9jO9ApW7DxBgfBgDZ3y6SjQbb3vDwIpd1jRNi/68qM57Kuespxjaq1yDPP28kuTdaa/hlFKLsuoA74CSnNLxges9zi9fiq9ahcjoEBF3hv7PYez66mo3GZp6wC0QeH85DlL13KayKoyutAP2JlhB4LCncUtXENAbqddt+nZFghKHHOcQwlSMB1WZTIjRQgEN2yxJQgIAv+JwBMQCAo3fupMaOj7/ODw3LZZaLlP8sPMx4H/U2pe+JsV4uY0dlVAlSed8OrCtpyZTHXnR3hqTeNCNnCbD4ykjN6fZgvTfr7ZhTVlK/2K6CnUZ4vv7vosCQQL56bWwFCwJxsdPQ3mhD+ZM7A5Vd9uxrSj2RANpgQOLP2E9m+FEBBal77TNnLuX6+ApHD7p/cpKJci4k+3ElDurKJjPplyo/a7pxS6oHhvJC1p6ehB0FN21IE0joCWM5FUc+ZRbc/ZISb2jShtWzdDozHQYtFthz+PhVOR1RzOhRL+FT9gwzW3bZab+5g94EOWOXt0zEcZ+5bdeVFjaMoCj94g4/YBFHX4IOR570cy6+TxTpnYEwQEgZeNQA4LBCuJe8bTuEIrvjnq0QRTEvm+Qwhy6EC2u99tLxvrHEiv0e3IptFStO8Wr9b+vXXdKGSrADRoi/Zio3NcOtOYrVz5ui6y2oWuTpz38Wc+8ilvygG0XuqY1ZDedVkRCMotlrQKIKDVYm4pJmI2jyW8wyAivxjD4GFz2J+UlWEpCxd/+pDGJfMSXLYlwxf8SWzGEwzSszbtMdMJvqyXj6KdfuSGzYzHnJhSh3yl+vBrYuZ2GX/vTxFHJawLbcHYud/x1eeLOeSoje/+0tWVR2qPxA6nz6Nym+/bBbumsJYdsc/lI6Dc3kz/0uosBruDoTYwlEr1wmjWqColg/wo1f1n4pxi8P56ur1i/5qmRq7MJNd0DIX4eWGuKZDy2xOIzoYOSwtKHBEEBIEXnUCOCQQl6QiLP+pGp0Gz+TMhdSVj4vzPY4mYvA3PToUXHW6Opi/5Nz5uUIYQvWT3UFfns+vyUbreEH6OOcWCbjUoEeA+p9FoyVuyJr2XXPTqZUjfpses7RSEvsxI9vusNKxcm9kAWa7Jl15dE0buJsQRGxtr/7+2h9FVS9HvlyvuY3Hx3HFWhKoBj7fQp5CBmpF72Pp5J8LHbiU+dZFJ31CPoyb2fjWQST8e53a27/UIJp1NS+wmPm5UmvJhPekXXocKjUaxKdYnGO8QUqKZ0740eW2zDrT4v1KRdpN+57rpTNo8kmQKVvqQXxMvsqRLWYK17vyz5+2HrHHMTjBeXs+nrcqRz5X/EnKBioRP3sF1m2lWrq3qT82SgWgd30FQZysElGzG18cecyqqLaUCHcLBNlNCT/5yjfhocyZDFt6pE3uCgCDwEhHIMYGAKYnbyc5mzEtE8Kkl1YIpxYjZYnV0PStYLWaMKUYsKJhTUjCp5xyNXMVqwWxMIcWceatXTYLx5nnOXvesxVMn7C6rOgShf304+zynqpoPMqFWCYoXL+74L0w+g0zgK8794hQPLcegLe6uI/Ph0ZSX8xD6WlGCKw5hq0c3eOpYn92+mXsxZzgbp35TMrt/ZpJjz3Dy0h2PIYn080jNU1v+GR35Z4vKM2+94zbfi+HM8SMcPXWVJDdS+11mIylGMxarPc+dZcBkBcVsxGiyYD+l4DqXw+LK21qxJwgIAs8zgZwTCM8zBWF75gSM2xlYXE/+jJwY1VAyHWKwcnVGfQzB77H61mHGvGWgWI9fydKoQuZWiisEAUFAEBAEcoiAEAg5BPJFD8Z8fDwVZX+aLkhwfQwr3TRnKhDu8UPHIAwNZxFjtXL5m/r456nLdPGBjHRxioOCgCAgCDwrAkIgPCvyz1W8VmLV2RNyZT7b9zdzhnzODl8OkJkJBOMOBobKrnUklJureK+gntDeG0l6eJJfN59J4/n/XKESxgoCgoAg8IIQEALhBcnIJ5sMM4c+KY9em5fQmj2Yfeie716ETASC5dTnVDEUo//vzikHZi4sCue1wPyUazyYH66KQfEnm5cidEFAEBAEskZACISscRJXGW8Te+PfOOylQqc8JOHqzTTfUTDfu0Oy0AapYIldQUAQEASeHQEhEJ4dexGzICAICAKCgCCQawkIgZBrs0YYJggIAoKAICAIPDsCQiA8O/YiZkFAEBAEBAFBINcSEAIh12aNMEwQEAQEAUFAEHh2BIRAeHbsRcyCgCAgCAgCgkCuJSAEQq7NGmHY803Awv24E+xe+xsnPNYte/JpUngQe5TtPy1i/uIf2XzwKg+z9rXtJ2+aiEEQEASeKwJCIDxX2fUEjLXGc2jzRjZt+Y2t27ay5dc1/LBiMQvmzmPRip9Y//thYkQNky3wllNzCX+roH2JZsM7RMU/pRraeou/prfjzUrtGBO1nOWzR9GshD8hFToTdfR+ttLwxC423+LUgZMkPm9TWp9Xu59YRoqAXwYCQiC8DLmcURrN59k0YwID6hdFp64CqM1HlZ6RLFixnIVff8R7FUOQC9Rg8M+X8VyjKaMgxTnAfIxxFfRoDM357tbTEAhmTk+vR2BAXaaddeaUQuKyNgRJEnKFcRzL4oKUTy7/TOwdVgq9lIdGs678i0WwnpxlGYf8vNqdcarEWUEgMwJCIGRG6CU5bzk7meqyBl2Jwez2XCXw4d98+rYByb8ak447K56XBEqmyTQTuzuKYR37s+RaqjUflZssaGpA49eapXczDei/X5Cykw9K6PBvOh/HCtG2MJV7e5gQVpEafVZyJZWJ/zlScyy7o4bRsf8SUic//bCtxK+PoFG1Fkze79mjkQHH9AN6ykd92f2UzRDRCQJPmYAQCE8ZeG6NTrmzmBYGDfryozns1dK0cGZydWSNjteH7xO9CJ4ZqNxgfpgBjdyIOXGpegmU2yxqoQqENiz3tW6FZ1j/cdtyOpJqso6SH/7x1PJIuTGfMIMGudEcUic/W8nJiGO2AhIXCwKCQE4S+BcCwUjcvoNcSmcMUbl/iT2rZjN97jr+uZPTzZWcTLYIKw2BpKW08ktPICjc+q4ZBo1EUJd1OFdQSHP/S3kgiSUt/dAYmrLgpi+B0JbvPRvLT4iTac8QSupkKk6IJp1H88nEmrSEln4aDE0XkDr52YswA47ZC0hcLQgIAjlIIBsCwcSNv5czuk1ZgoK7sC5VTaHc3sXH9WoxcNUB9v84lDpv9+EXz77OHDRaBPUECPgUCBZOfPY2skamyuen0lY+yj1Orp/N+ME96dprEOO+3crFh077LNw5v48tq2Yzrl9XvthnAiWZM79OJ6J7Rzp0Hc6cvTfsY9EpMfyxfBqjB3am/Xv9iNx4Gc+RDnuIZhIOriBySHc6vtuW8N7D+XL1EbfDm3KTXXPGMXb8RD77bCLjx44hcu15u83WGH77ejzjJ37GpM8mMGHWdq47NWyGaXCmJdXvoxP89MUYOpTTo9GXod0nk5g0aRKRUduJU8N19SC0Y+UDsCZF88usCXzYM5zwXiOYuTMuLUsU7p1cz+zxg+nZtReDxn3LVjfMVAY4dpV7HF4xlYm9ahAk6SjSeBATJ9ltUe2ZNOkzJowZQf/3B7Piqj3ByoMYjuxYy8LJH9Jl1FpuWh9ybu0k+oZ3YtjyU+51Mh5d4rdZo+gb3o72nfsw6JMpzN9u90V5dOInvhjTgXJ6Dfoy7fjEFlckUdvjfPgWPCI+eg8bVsxi4ged+WT9bfuCX5lxdCXzJOtnj2dwz670GjSOb7dexF3M7nB+3xZWzR5Hv65fYC9mZ/h1egTdO3ag6/A57L1hT3tKzB8snzaagZ3b816/SDZeTlvKvEH7sNt6l0sHtvLjtxMZ0CWS3Y+BRxfZNv9zhvcJ573uQ5n+29Ws9+akXGHbnNEM6NyRTgM+Y1X0Xd8LonkbKPYEgSdGIMsC4cH+uYyZMp9pnV9HDunBBi+BkMzW/iUp8P5P3LOZep8NPYvy6sDf3Q/xE0uCCDhHCPgSCPf3MuJNGV1oF35K8G4lK3cP8k278lTsPJtd5+K4cnARXcvIBFUfxz7b1L4HHFr4IW3KB6LVGAibtZ+fhjSl4XuDGfNpL2oU0CIFNuLLLcvoV68GLXoOY/yED2hUTI8U8A5RngPbym32jK9HaIVeLNx7gZgLf7NhZk8qBMoUCZvGYVsr3Uj836v4uGFBtBqJgLqj2Xoh2fGifUTswaX0LidTqPFoVh2Ms/WGZJ4GH3QfnGbL0ij6vS2jkSvQe9ZiFi9ewvLNJ7mnYvIQCN9un0zzCpVp0XMEYyJaUdpPQgpqyvxYp0JRr7/LwW/aUb5iZ2bvOkfclYMs6loGOag64+ww0zdESeL4+qUsHNWIEK2O0LaT+G6xaov9f9GsD6mXX4tGrsNXl+3xWS7+wvhuNSikk9CXH8GK2e2pVqEsIer+G5/ah5hSjhBZK4Sirb7hwPV73IxeTt83/PFvtxI1ax+c3sLSqH68LWuQK/RmlhrfkuVsPuljpU9rHFun9qaumrcaP1ouSbKnJzOOKNw9+A3tylek8+xdnIu7wsFFXSkjB1F93D6bLTw4xMIP21A+UIvGEMas/T8xpGlD3hs8hk971aCAViKw0ZdsWdaPejVa0HPYeCZ80IhieomAd6Iy9p/wZbcxmmUR7amQz8526o6V9KlRkcZdhvLpiPa8kVdCCmjAjIseeZx+DoLlEt+1DqXS4A1cuhXLhgGlkYPrMv3UU+sL8mWZOP6SE8iyQLBzUri9qAX++XuyyVN43/uZTiH+NP8u0fEyVri5oCn+BXvw66OXnPDzknyHQNC91oeVh45wYM9Wfln0OT2qvkKBqgP4/kyqjFQSWdutOH6Vx3PMJRYVEhe1II8UQNP58a6UP17flRBJIiC0CZ9sicPu4mDlyjf1kDVagquP4Ld4p+ODlWszGyBLAbT73i43QSFxbTeKy6WJ+FNtqjn/LFyOCiNIkik/Yi9OCy3nv6SWn4RcYwpnPd+xxr0Mq9KCeY6WNNlIgzNG7997LGutDjE0SztTwSkQtIGU7TCF3QlOQ4wc/Kg8eimQ99c40+JIn19lxrthoiQuokUeiYCm88lspqRx+wCKaWUqTfwnVc+Es/u+MVGejgKWaCZUkpHkIEp3+Z4Yy2Mu/f4Dm07ZK3jTrkGE6vLQ4Qf3RxyS13ahdI8N7mGme8torQ4xNPuOrE3UsHJxWm1kjR/tV7na/4BvjkriWroV96Py+GPueJVEFrXIgxTQlPkuMI9Z3zUESQogtMknbIlzlCfrFb6pJ6PRBlN9xG+4i9k1ZjaQkQLa4Spm3pnrsefLbiPbBxRFK/lTpM4gVp9zlkAzJydVQZb8ab0scw9V86FPKO/3FmOP2m22XFhGnybt+ea485nwMEVsCgJPkUA2BQLcWdwyjUAw7x9JGTmUD3a5vdyNm3pRQK7IhGgLKElsGFiOYlXHs999yVNMpogqUwIOgaANqUb4/3rSsf7rBGkltEW6sy7Ju+dADctyWp31oOO1HgvZuWsnO37fxtYtm1k7tjFBkpZC/ba6orRXXmmdHG1lRCtTffJZr0ot5ZfOtkq//oxr9u5q62W+qe+HtlAftrjEiCP45DV0zq9FW6Ab65xj/UoCi1sFIekrMt7jJXt/Uz9qePRqZScNrsR4bfiu2Fw9CHJDZnv2FKit75Xt8JMMhM2/aRfUltNMri6je60HC3fuYueO39m2dQub146lcZCEtlA/tnoKci8b7Du+BcJDVrVXRUwTvvUc8rNeZHpttfejGpFnnOLFHbD58GjK67UUajWHk06NYLnHrSQPQ7ItEBQS5oVh0PjR8UfPjPTF0cJp1UFW9xo9Fu5k184d/K5+q2PzWsY2DkLSFqKfC4xaWRdDq3ud4eoYg+vPyKZeBdDK1ZnspRZT+KVzEJJcnxmePVWu+zw3fNltZv+I0ui0Ren/uwcXIGV9N/JJMnW/vuxjyMUdvvnASMro81B99F+k86i5LxRbgsBTJpAjAiFlfVfyyeX55JBb8Zp2fkCoKhp2qg/rIw7P7ka7ASs4n/Zd9JSTLKJLl0CaIYYH7P+kEgbJQKUxf7vHpW03O16Ykp4yrYfx8ccf8dGokYwcMZzhwyKIGBrBhJ/Pu6LxVXkZf+tLIa1M1Uhv3wZbeZJkak+/aH+53v+Bjnkl9G+O4Yi7iNnDd7YQ9W8w2mP6xb113Smk1VN25H77OLByg2Ud6zPBJRiylwZXYrw2fFVsnkMMrVjq6E133pryczgBkkyDmXYBpCTMI8wgoS/TmmEff8xHH41i5MgRDB8+jIiIoURM+DnT58YXY8hYIEhBaf2JbHZazhMVVsA2VONXrCZdJ6zk8M1U8J+0QFASmBdmQNKXofWwj/n4o48YNXIkI4YPZ1hEBEMjJvCz64XiEAhyJSb+4/mSMfJb30Jo5apEenXZp7C+az4kuTbTMx0GyEQg6EozYr83G+OWPryilakWedpL/DrLgNev+TRfN8hn64l4vcMM/hYqwQuP2Hl2BHJEIBg39SS/XN7rBW3c9j+KyKUYttdTzT+7hIqYMyGQRiAA9/cwtJyMlLcu0057vgAtnJhU2ea46KzkMgrdV+WVVYGgXP+Wd9QpmGVGcsDTDFukD1nVzg9J9ypD//Qoayl7+LCUDl2JD9j5GKwXvqJpS8/x5uylIf305YxAsJyYRGV1LL/BzIzHw9M3wnbUF+NMBUK+rqz3bMx7xvHoIhsju1C9sAFJI6ErWI/xe+64neeetECwnGBSZbWXowEzM23lP8cCAbDe3Mn4xkWQJS3BNcax7ylMjfXMarEtCKRHIEcEgvn4OCoYQhnkMcSQsq4r+Qx1+drhGJVe5OJYLiKQnkBAIWlzP0rqtIS0XOhReVmJmdXQ5j9QoPMaMhtl9VV5ZVUg8Gg93UJUh8b2rHK6JbjQJbO8jR8auR7feH0JyMI/E99G1hah98Y7HPq0Ad3WJLkrN7KXBld0Xhs5IxCsMbNoKGvQFujMmsxgesXv3vHF+N8KBGPSbe47/eseX+X3Ka0I1UvoSw/nL2dv+pMWCNYYZjVU/QcK0DlTMM+3QLDlpDmGn3u+jiwZqDHVe9jNndNiSxB4egSyKRAUlw/CRudLQrVV/YpbaB6aLnCMqWLl8td18ff86I7lPskPnW+cp5dAEVPWCCiJC2nmaKV79ZYqN/ixU1G0uqJ0+dmZv6COUb+h19hmIcw479Fyt0WnoHi4LdiFgJ63xh5zOCjabTJu7kUBrczbn53w6oZNWduJQEmmxheOKYrKHX56vyBabQHCVzumxzmTZTrAqLJ6/GtP45xnz7LaKrsyi4Z5tATXas07tYbxR6qWcnbS4IzO+9cpEJow13N8X73I+fEfQ3MW3faAoQ64re6Ivzo+/ZVjfNp8mNFv6NFIgTSacT7N1DjFE6a3Aa4949Z+FNamZQzJrHhXFVCNvH0hLOeYWkPtHerE2lRcwMSuIbUYvNPjIVfusKxNAJKhGQsTHelxCoQmc72+3ugyKs2GQtycxjYfhHYr03NSTM3RzOHRb6DXqLMQZpC2mCkegs8xlKB/i7Fe35Q2stnmg/A2n53wLCAprO0UiCTX4AvXMEUagx0HfNlt4s+I19DpShGRqqfUuLEn+bUylSd5l+3UMZgTT3HogntKoxKv9pZJhPTYkM4039R3i31B4MkSyKZAsHLhy5rIedqx0ukQZrPPzD+RNSjWYSW2d4f1GnPDihP27SX7GLKSyPcdQpBDB7Ld453zZJMmQs8OAeO+EZRVK/zgd1nmrAAcAVhjv6djES26oh35PsYh8qyxrOhQBJ0kYSjTg6X/3LFX/o+vsHFMW96fc9Zxt0LC/Kb4q46LPTd6THtVuPldc/wlHcUH7nB7qLta9jpeHbLH9ZI0n59Ls4Ja5DL92JDgFJomrixsTcGASnz8Zzp9smq5ax+CVjJQfXI6Y8FZToMvko9Z1yUYSVeS/lsSSbr8B0uiNmKbJGE+wqdv6tGoY9+nPSsmK5em17F9mfK1oX840mcldkUHiugkJEMZeiz9hzu2oZTHXNk4hrbvz/GejZHGHOfsEZXxBvvUP+c1ltNEVpPR6Msy6oCHkHu8nYGhOjTpjsGb+fvj8hRuv9w9e8J8ka/r+WGoNMG9psPjdXQJltCV7M+WxCQu/7GEqI1XM3DKM3FgVFn0GpkaUzxbyL45WmNX0KGIDkkyUKbHUv6xg+HxlY2Mafs+c5yOh0oC85v62xwXe270EB/KTb5r7o+kK87AHR5KyNk7oXuVIXsyeyn5sjuJVR2DkaR8hP/o2bVl91nwk99k5wAAACAASURBVLQU7feba3aNM0vcvyn82i0Ebb42LHVMITYdG0clOZCwuQ4HXffFYksQeOoEsi4QkvcRNagjVQqqD2seSjXpy9St190vA+NpFnVtQPOBnzGuTys6TtpNovM9jonzP48lYvI24l3HnnpaRYTpEUj5g8jWdSlXQEbSaNBoJPKWqEbYoFX2is52j8K9g1/RtlQAAa83Y/Cyf+whPYhm0f9qUcRPQiPJBBcpTrFStek797DdG9saww8Da1Eyr4RGo0HShVC+7UyOmx+xa0IjSgdrbXFKcmEq91vJVcs9tn5Sl1fzau3Xy4WpFD7HZfX96CUMrBdKcOGqtO0zgN7vVqds5feYsjPBqwfCdYM6Y2BLX0JDWrE4dQvfeVFmaXBel+6vQuKa7oTK9vSHVOrO/OPJ/N++qbSsVBiDuviV+j2GknUYsfE2inKHjSNqEprHwUMuwJutpvG3TQw8IHrR/6hVxM823i8HF6F4sVLU7juXwxk5rSl32PRRA14PcrDUhVC2fj+WXbBgObOQ7tWL4++wQy70FmFjthCzehANy4Wgdxz3L1qRxgO/91irwcKZqXUIMgRTrsX/GDG8H22qlqJMw0GsOOOcmqn2kiSypnuobdVKSQ6hUvf5HE/27i1xYUvexph3ylNQZaWWBb9iVG491fZRI9s01nQ4OkN6EL2I/9Uqgp+kQZKDKVK8GKVq92XuYfuQkTXmBwbWKkleR3p0IeVpO/M45ke7mNCoNMFaNU4JuXBl+q28iuXeVj6p+yp5tfb8kQtXInxOtMtUrw0fdv95cTUf1HmNQEfYupCytJx+GDMP2TGmLq+q32RwlPlyTT7nTx8a5MGRmbQvU4BybYbycUR36pd+lXpD13Ilja+Nl1ViRxB4KgSyLhCwYjaZsVjVx1bBajFjTlPZW3l4I4aEB2lOPJXEiEj+DQELxsdGTBarfVhAsWIxG0lJMXt03zrCNd3l6sljnIp3zve2H7fcj+ds9FGiz8WT7PViUzAbjZjVsNVSY7VgSjHZRKXFlOq4yWK7Js3xFI9Wry06Kw8SzvPPkWOcunbXa8gi3dQrd7l04Wam1/lOQ7qhehy0khxzmrPxD9xi2WrCaLJgf1QUrGaVr/2W9NLn9bRY7hN/Npqj0eeI94bpEaf3psX4mBQ1PhfjFMwqcKuJFFveqkM+VszGFB4bLShq/jrtsz3LJlKM6eS35T7Xz0dz5Eg05xM80ucZvTWZmNNnic/0mbfnvfP9YSsLRntZsAeXDkfPeLBwP/4s0UejORef7J2fihmj0fPdZCLFpNKwYEp13GRRwaQ9br/eK0LHjg+7fcYJFlOK/Xly5odXOtOJw/KA62eP8veRU8Qke/Y2pXOtOCQIPEUC2RAIT9EqEZUgIAgIAoKAICAIPFMCQiA8U/wickFAEBAEBAFBIHcSEAIhd+aLsEoQEAQEAUFAEHimBIRAeKb4ReSCgCAgCAgCgkDuJCAEQu7MF2GVICAICAKCgCDwTAkIgfBM8YvIBQFBQBAQBASB3ElACITcmS/CKkFAEBAEBAFB4JkSEALhmeIXkQsCgoAgIAgIArmTgBAIuTNfhFWCgCAgCAgCgsAzJSAEwjPFLyIXBAQBQUAQEARyJwEhEHJnvrwAVlm4H3eC3Ws3E+3r+/weqbTcj+PE7rVsjk5O+4lnj+sy23SG89uJB5ldKs4LAtkg4CzPv/Hki5aVRzfOsm/Deg7eUD8NLf7+MwHzHc798QuLo6JYtP4wCT7WxvjP8bxgAQiB8IJlaI4mR3nA1b2r+fWYx+p4WYjAcmY+nSq+YlvERyPXZ4ZteUMfN1rOML9TRV6xLeIjU39GRqsB+ghDPWw5xdzwtxyLARl4Jyr+PwmNDGLyOmW+dYoDJxN9LhbldbHYyQECZm6dOsDJxKe3ZoHl1FzC3ypoL8+Gd4iKf3KVtvX6z3xQw7HAlv4NPj3stbhJDvB7+YKwxG3hk7DahH++iEWftSZUpyW49lSOp17m5eVDk2mKhUDIFNHLe4HxwEeU02vJH/4TSdnFYDnJpCoyGkMT5jqWsvUdhIWTk6ogaww0mZvwHyp2M8fGVUCvMdD8u1v/IRzflnqdMe1lWCk9Up5GzLriteSS12ViJ+cImPYOo5ReIk+jWR6rT+Zc+D5DMh9jXAU9GkNzvrv15ASCLX4lnm/fMSDJVZh08ukJIZ9pf55PWM4zq1E+Cvf4lftqOpTbbBpQhpDyw9iRvXbP80zhX9suBMK/Rvei33iPX3uVJiSfHm1wO1Zk+6WYxJJWfmgMTVlwM/MXatKSVvhpDDRdcDPTit0cu5uoYR3pv+SaewVFW3Yo3FzQFIPGj9ZL7z75DLLGsz6iEdVaTGa/7e3jiNIcy+6oYXTsv4RrQjfkaD5Y49cT0agaLSbvt7/wczR0M7G7oxjWsT9LUmeccpMFTQ1o/Frz5ItWCht6hCDJ1Zl8JocFQkZlM6NzOcr56QVm3j+SMvq8dPjB8wG1YhXPZZYyQQiELGF6+S5S4hfS6u2hrPr6HfylvLT4Lrst+2SWt3EIhMTMBULy8jYOgZCYiUBQuDE/DINGptGcuFTXKtxe1MImENosT35mmabcmE+YQYPcaA5xmSf9mdkpIk5FQLnB/DADGrkRc1JnnHKbRS1UgdCGJ1+0jGzuXQCtKhDO5qxAyKhsZnQuFannZNfK5a/qIOtCGbxbjCf8m0z7FwLBSNy+g1zyWW4zO/9vzBT3PF0CFk5F1qT6xGhM8fMIyyORp8m3xGRLdT8pgQBJS1r66G1wC4S233u2GJ4uPZKW0NJPg6HpArLQefKUjRPR+SaQxJKWPnq9XAKhLU++aD05gZBh2Xzhyq2Zvz8uj15XmhH7hEDwXe59n8mGQDBx4+/ljG5TlqDgLqxLSR1oZudTXy/2cy2Bx7sZUr4h31yygnKdhc3zIvnV45vLGSgEJZmTa74goud7tG3flQ/GRNLrbdUHoSkL0vQgKCSfXMMXET15r217un4whsheb9t8EJouyKgH4REnfvqCMR3KodfoKdPuEyZNmsSkyCi2x6m2uQVCu5UPwJpE9C+zmPBhT8LDezFi5k7i0hO2yj1Orp/N+ME96dprEOO+3crFzMYnH8UTvWcDK2ZN5IPOn7D+tr2r4NGJn/hiTAfK6TXoy7TjE9W+SZFEbY/zGA5RuHdyPbPHD6Zn114MGvctWz0itNw5z74tq5g9rh9dv9iHCYXkM78yPaI7HTt0ZficvdywZUUKMX8sZ9rogXRu/x79IjdyOZve2cr9s2z4ehjdO7SmTfsu9B08ktFfbCCjrH5W5fZRfDR7Nqxg1sQP6PzJeuzIrdy9dICtP37LxAFdiNz9GHjExW3z+Xx4H8Lf687Q6b9xNbP64dEJfvpiDB3K6dHoy9DuEzXfJhEZtR170XL2ILTDXrSi+WXWBD7sGU54rxHM3BmXjqNqxvnsm6OHQDgRy58LxjKwSzhdB41nwR/xpOu2aE7g4IpIhnTvyLttw+k9/EtWH/F2ns2obD7ISrnNQhxPp+yaSTi4gsgh3en4blvCew/ny9VH8PRbVe4dZsXUifyvbn602vzU/d9Ee34u+5t7mfTqpVzZxpzRA+jcsRMDPltF9N1MbvCdkc/9mSwLhAf75zJmynymdX4dOaQHG1IJhMzOP/ekXpoEKNz58X1ea7MYu2+hwo3FrQiUZGpMPZfOSxCU5EPMbF+Osq0ms+lkLHHn97NyZB0KaDVpBYKSzKGZ7SlXthWTN50kNu48+1eOpE4BLRqbD0JGAuEBp7csJaqfKiZkKvSexeLFi1myfDMnbU+9h0D4djuTm1egcouejBgTQavSfkhSEE3nx3pU1KDcPcg37cpTsfNsdp2L48rBRXQtIxNUfRz7MpgpaY3bytTedSmml9D4tWSJw4vzwektLI3qx9uyBrlCb2YtXsziJcvZfPKefThEucvBb9pRvmJnZu86R9yVgyzqWgY5qDrjHBE+OLSQD9uUJ1CrwRA2i/0/DaFpw/cYPOZTetUogFYKpNGXW1jWrx41WvRk2PgJfNCoGHopgHeiUvtl+C64pitrGVitAh1n/EHsg2Su7f2cxoESfmHzHXnv+96nf8ZK3Nap9K6rplODX8slDsdZI9HLImhfIR9ajUydqTtY2acGFRt3YeinI2j/Rl4kKYAGMy565Xsa+x+cZsvSKPqpolauQO9Zi1m8eAnLN5+0VyiuHoR2fLt9Ms0rVKZFzxGMiWhFaT8JKagp82M9BHQW8jmNDa4DDoGgzUepGi0YMHUpP6/+lpFhxdHrX6HehD0kedRZyu09jK8XSoVeC9l7IYYLf29gZs8KBMpFCJt22OWrkVHZTM6k3GY5jidddpXb7Blfj9AKvVi49wIxF/5mw8yeVAiUKRI2jcOOjkMl6Tjrly5kTPPCaLWFaT5moS0/V+66TKqqy0Vd3bBc+o7WoZUYvOESt2I3MKC0THDd6ZxKr2HhdeeLuZNlgWBPvv0F7J+/J5vSbalkdv7FhPhCpcp6hdnvlKDrWreTn5K4nHeDJOS3P+NE6gdFucOG3iWQQ/uyxXPY33qZr+qk7kFQuLOhNyXkUPp6X2wfK8xUINhJ31vW2jbE0CzNTAWnQNASWLYDU3YnuASN8eBHlNdLBL6/BrWNaftTElnbrTh+lcdzzPXWUEhc1II8UgBN58c7r0z/13qRabVlNH7tWeXZ43BvGa3VIYZm3+Ht26mQuLYbxf0qM94dIUriIlrkkQhoOh/XDLrH6+kaIiEFhNLkky3EOZqN1ivfUE/WoA2uzojf3K1J67WZNJAlAtp9z730rfU6ar6wiPahBWnwxTEeOc4oCXNpYtDz1thjXq1Uy8ObXPrnLzavmMOKA1kJ3Ssq147y+DZXTx1g2+q5rPjrdir/EddlGW5YL06jtqzBr/0qPJEbtw+gqFbCv0gdBq0+50qT+eQkqsgS/q2X4S7RvqK4x7LW6hBDs7QzFZwCQRtI2Q5T2J3gfBCMHPyoPHopkPfXOEtWNvI5XVMcAkH3GoN3O3MHeHyU8VUMSLpSDN7lSL2jDMulI/jTGb0apuUyUWFBSHJ5Ruz1CMNn2QR8nctuHE+s7Dq4yqWJ8E4sl6PCCJJkyo/Y68p7MLN/RGl06hDD/nT7XVLRN3Pok/L4vTWWo7bLLVxY1ocm7b/heFZuTxXai7CbTYEAdxa3xLdA8HFeSWLDwHIUqzqe/Zl19b0IVJ/jNJiPjqVCUH0+Xfc7v//u+N/2IxHVDEj6N/n0iPeTYr0wnToGHcUH7sBbM6bjg2C9wPQ6BnTFB7LD+2Ky7qSovscyEwgyDWd79xTwYCXt/CQMYfNdfgGW05OpLut4rcdCdu7ayY7ft7F1y2bWjm1MkKSlUL+tGeekksA81anNryM/ugRGBi9ay2kmV5fRvdaDhTt3sXPH72zbuoXNa8fSOEhCW6gfW51cjNsZUEyL7vXheA2fGjfRq4AWufpkvPzXUn6hsyri6s/IfOaE5Qxf1Q8ib/2vueDR6DVuH0gx/Sv02uRZ09hb7p0qBSJpizFgu9PAjNGkPauQ9NdcBtYvgk77Cr294kh7ta8jSsI8mwOoX8cfvVqC5v0jKK3TUrT/797lMGU93fJJyHW/zsKwSRYEgtyQ2Z49BcCDle3wkwyEzXfMwMlOPqebUI8hBq9MVkhY2Iw8kpbCfX+z3Wm9/A31/bQU6rPFi4d6MnlNZ/JrtRTots4diy8RoF7h41y243hSZdd6mW/q+6Et1Ictns+bPbF0zq9FW6Ab61zuR9kXCAdGlkGfpzqj/0r6VwLWDfrF2Ho6AoFHHJ7djXYDVnDeKbxfDH4vWCoesPV/oYSUf4e2HTrQweO//TvlCdLqKTNCHRN3/incXtIK/3SHH9IKBOX2Elr5S8g1pnIuVTnIWYHgR6ulqb7ckPIz4QEScoOZjgpUIWFeGAZJT5nWw/j444/4aNRIRo4YzvBhEUQMjWDCz+edCU3/N5sCwV65SejLtGbYxx/z0UejGDlyBMOHDyMiYigRE352Px+Ol6xcaSL/eLIy/kbfQlrkqpHe3Z4p6+mqVoS1p3PRo9JPa7hC0rpuFNHl5/2fPF+Cjm9I+IUxP83X+0zsHlwCXVAHVmXeDE8bpeuIlUvT6yAHtGapd9eK64rMNjIWCDpKj9jv1fuBcQt9XtEiV4vktCfHdCPKgkDwa0XaohVOgCTTYKZ9eCdb+ZyuHb4EApgPj6a8Xh16mmu78/4PHckr6XlzzBHvdAPO3ib9G6PdsfgQAbYLfJzLdhxPquze/4GOeSX0b44hVTtFTSzf1JPR6N9gtOvjUtkVCGA+/TUN8mmR/F+nw4y/vYZy3BBfnq2nJBBeHqDPc0qVG8tp+0p1ItN7kz7YQM9CWnSvDma3S71bufBlTZs/QM0vL6Qa400rEKwXvqSmOjZf80uvlqvK7OkLBAsnJlW22e58sWc777IpECwnJlFZTb9LpGQQ45N6yXKX1eEhaPO+y/I7HgPZ97Yz8HUd+grjOObdSQSO4SK53jf4/Cim8RaXzid4dO+ml7YH/NIlBEOttPmf3tXpHXseBEK28jm9RJKBQDg2jgrqh6Lar7I55V7/9h0MqsPuyANpBAIPV9l6zXSvDnXH4kME2C5I95xCtuN4QmVXuf4t7xhU59+RHEhdRnnIqnZ+SLpXGfqnswmTfYEAVm7uHE/jIjKSNpga4/bhOXLqBvlybAmB8HLkcxZSaeH8tNrkazDDR1fsQzb3LoJWV5z+25xjmo5vEkg6SgzZ49GzoEaXViDY51lL6EoMYY/zGXZY9vQFgpWYWQ2RNVoKdF6ThfHpdBBmUyBYY2bRUPUfKNCZNZm1xJ/QSxbzAUaV1aMvO8rjJZvMnmFvIEtaXum92e2j4UiycmcFbfPqKT/6cNpKyHGN5dxUapUbxt5U+epFzfQXw0rJlB2VTmXmdaHvnedBIGQrn9NNqm+B8HhjTwpo7Q7D6q2P1ncjRCsR2H5VWt+T5OW08dOgCjvXX7oiwHHWx7lsx/Gkyu6j9XQL0SIFtmdVGlcYx/tGrsc3rq+a/huBYGdhjvmZnq/LSIYaTPUa5nGRfCk2sikQFJcPwsZ0hyIzOG+5T/LDDPs+XwrguTaRKfsYUS6Y5guv+xx7e/RbP4pqtRTqucHlGW22tWg06EoNYY/n0LUST9Q76kdnGjDT+VU65+dqdaUY4n0x8VFqS8jdTZsRJ6cPQpO5qW11fkTJQPNFqZzgHq2mozq8UfcrlwBSu2vf0GuQAhsx43zqmk1B8Whgp2uPEsecxqoPQjtWenrMOV+0TeZy3TMM82FGv6FHo85CmHE+laACxTNCx1CC/q2x3i1642a7D0Jqh9GUtXQKVIdvvnAPU6RntPryLqq1tcLsflsWYn7oRv1G9XhVbyBsXgKPr1/jhgeOlK39KKIvRJ8tnhnsHXhWBIJ6TQ1DQXpudAhMcwJ/zhlI86pv02zaUe8AfewpcXNobNDg126ll5Oi6c8IXtPpKBWx15urcSM982uRK09K62CbJg7nEEMT5nplnNpYd3xEydCcRY4prc7bH63uaBtmq/vVZXsvWnby2RmI168vgfCIHR+URB/Sku8cfhDKnZ94v6A69h7O6lR2mQ6Moqzen9rTzrlD91U21St8nMt2HE+q7Cp3+On9gmi1BQhfner5NtmFr3/taR7Dlyb+GPoqOl0phmWoXNXEm0k8dYgLrimNCvFq74wUQo8N6VZ2bqYv8FY2BYKjSzlPO1a6HEE86fg4ryTyfYcQ5NCB/GsfJ89oxHYOE3jEyemNCDbU4Sv12we+/u7/zPvBElJwE2aeczw01ivMbx5im3pXbeTv3LCAcv80P0Q04bVgHZLkR7nwqfywP1YdFeXK/Ob2Fk+1kfxuv5jTP0TQ5LVgdJKEX7lwpv6wn9g0XYhuox6v60KwpKNk/y0kJl3mjyVRbLT1fZs58umb6DUyVSNPu2YwqHdaL02njqxB99pQ/nA+79ZYVnQoYovXUKYHS/+5Y28hP77CxjFteX/OWXek6W05XkoauQZTPFsZj9fRJVhCV7I/WxKTuPzHEqI2qotQWYld0YEiOgnJUIYeS//hji2dj7mycQxt35/jcjxUEubT1F91XOzJRg/xodz8jub+kt3R0zXUA85Wq+7VIexxpi89mx0zL9SW0cRD1ziysDs1Gk/gly/fwaB9hfCv59G9bk/WuHwE7Ewz8xvIXCAo3FrSioCAViy+pWCK2czY8A4MivyCMYOHMWd/Kp+R9GwH7JWeBrnGFBcr9dKkVR0JliTyhf/o1ZK29Tj4SWiL9uM3Z8eXj7DhMeu6BCPpStJ/SyJJl/9gSdRG+7CK+QifvqlHI1dNNQTn8KvQ6Hht6B8OB8ms53P6phjZ1KsAWl0JeqxzTvtVuLv3U6qHlCB82WWPsm3m/NxmFNTKlOm3gQTn42u6wsLWBQmo9DF/evaR+yyb6iwJX+U2e3E8sbKrVuPn59KsoBa5TD82uBPLlYWtKRhQiY89E6vcYlmbQCQphPAfM+myS/nV1juRr81SxxRfE8fGVUIODGOus4GTfma90EezLhCS9xE1qCNVCqov/TyUatKXqVuvu8edMzxv4vzPY4mYvI14ZwF+obE+P4mznPiW96uFEqCV0GgkAopXpunE3am6mU3sm9qGWmXz21e000hoA4pRqWOULaHKzR1MCCuBvyRhCClGiTdaMmHLJQ6Oq0hAkWp0HPIZc7ZcsENRbrJjQhgl/NVKMoRiJd6g5YQtXDo4jooBRajWcQifzdnCBY/KLzVNJXEN3UNlJI2EHFKJ7vOPk/x/+5jashKFDWo6NEgBJakzYiO3FYU7G0dQMzQPkkaDRpIp8GYrpv3tUCAPoln0v1oU8ZNs54KLFKdYqdr0nXs4Qwel5G1jeKe8Y4U/jYRfscq0nupw4FQSWdM91MZKkkOo1H0+x11LXj8getH/qFXEz25/cBGKFytF7b5zOWyb3G4l5oeB1CqZ12GvjpDybZl53MyjXRNoVDoYrWRPR+HK/Vh51cK9rZ9Q99W8aB3pK1wpnDnRvhSWlbjVXSiprp4pGQhtFsmeW2YufVUXgyQhF2nAuJ0eC11ZrzGzgZzWb8R4l4S4WGJj7f/X9oymaql+/HLFfSwu/o5H1j1mS59CGGpGsmfr53QKH8vW+Ey9Bj3uT2bbmHcoX1DNdzX9fhSr3Jqpf15k9Qd1eC1Qazsu6UIo23I6qp/awx1jqPtqoIOLjpByTfj8z4zUk0Limu6E2tjIhFTqzvzjyfzfvqm0rFQYg8pdfUZK1mHExtsoyh02jqhJaB5HmZML8GaradiLVmb57JG0NJuPWdu5CG80bU2118rSqNtA+oc34u0anfh6b6L7neu67z7RSwZSLzSYwlXb0mdAb96tXpbK701hp2s6puPijMpmRufIShxPuuza03A/egkD64USXLgqbfsMoPe71Slb+T2m7HRPa7bG/MDA2q8S6Hiv6fKVplbz4fzic+G4BxyZ2Z4yBcrRZujHRHSvT+lX6zF07RWfw2ou/C/wRtYFAlbMJjMWq9pnqmC1mDF7VfaZnX+BKT7PSbOaMZos2LPVisVsJMWY9sVtNaVgNDuuc+S/McWzErJwL+Yk0edu8tjRrW5OvEzMfc8+djcoy70YTkaf46b7Yi7H3Pc5vOG+075lTY7h9Nl4HrjKoBWT0eQqn4ojXbarLSa37YoViykFk+s+e3iW+/GcjT5K9Ll4kj2TlTpi577FRIqLm/o8mDB6BmpNJub0WeLdBjrvdEZI/NlojkafIz5VhIrZmL69adJhwqLiTXM8bfq8I1d4dP0MJy7eck8JNN3iwsnz7vxw3nB3FR2C9Lw+3HP2CpgPTqBWieIUL+74L5wPgxzIK8794sUJLTfIGYrqfs/o8jJ5Ql+jaHBFhmzNfFEu9832LVu+Way2MqLY8tGIyapgNhoxO49bLZhSTPZK1CuP1HxXr08daup9K8kxpzkb/8BdEVvVvHU+IwpWsxGT4xFRw0w3bmewlvs+89l5SdpfhXvXrtoEqvLoJhf+ieb8zceZPxvWBySc/4cjx05x7W4GhTijspnROdXQTOJ48mXXScvKg4Tz/HPkGKeu3U1biStmjCme7wO1XBg9el6c4Xj/Wh5c5+zRvzlyKobktK9B74tfgr1sCISXgIZIoiAgCHgRUL+NUFyfn27rPcY5vK6w72Q2xGC9OoP6hmDeW32Lw2PewlCsB796fg4wnTDFIUFAEHi2BIRAeLb8ReyCQC4mYOb4+IrI/k1Z4LNr1m5+ZgLh3g8dCTI0ZFaMFduHd/zzUHf6+UxbdLkYjjBNEHjhCQiB8MJnsUigIPAvCVhjbbM05Mqfse/vOQz5fIfPOeEZCwQjOwaGIjs/YavcZNV7BdGH9mZj0kNO/rqZMxn0iP9L68VtgoAg8B8JCIHwHwGK2wWBF5aA+RCflNejzRtKzR6zOZTBMngZCgTLKT6vYqBY/99dnwNW14IIfy2Q/OUaM/iHq6In4YUtRCJhzzMBIRCe59wTtgsCT5iA8XYsN3w5WmY5boWHCVe5mfozCuZ73BGeYFmmKC4UBJ42ASEQnjZxEZ8gIAgIAoKAIPAcEBAC4TnIJGGiICAICAKCgCDwtAkIgfC0iYv4BAFBQBAQBASB54CAEAjPQSYJEwUBQUAQEAQEgadNQAiEp01cxCcICAKCgCAgCDwHBIRAeA4yKVeYqJi4f/Myx/esY+XvF7xXzPtPBqZw49Qf/Lp8HlHzv2fTobhU60D8p8DFzYKAICAICAL/koAQCP8S3Etzm/Uai8NLkU+W7IvkaNIuteuLxeOLm5k5fgKzt8ekKyiU23uZ3LIkeYIq0rFXY4rp1EV4/Ck7YCOJ6S/h4CsqcVwQEAQEAUEghwkIgZDDQF/M4CzELWhBoGM1u8BOpAm9PQAAIABJREFUa10fvPGZXuslvqmnrlioQQpqw9KbqWp8y3m+bVoArUam0oRozJYTTKpisF2vKz2C/eLLej7RihOCgCAgCDwNAkIgPA3KL0Acpl2DCFVb+BqJoC7rMhcIyh3W9SiGTl0CuuxQdj7whnBvXTcKaTVodCUYvNtkO/n4/C9EDhvGF1tj3Svped8m9gQBQUAQEASeEgEhEJ4S6Oc9GtOeIZTIjkBQE2xK5PTBI1xNTtV7QAqbehVEq9Ggkasw6aRYV/V5Lx/CfkFAEHjxCPwLgWAkbt9BLqXzTk+JPci6xTP48qsFrDuamHaN7heP30uTon8lEHzRUa7z7TsGNDaBUJ8Z16y+rhTHBQFBQBAQBJ4RgWwIBBM3/l7O6DZlCQruwroUT4sVbmweSliHz9lw+Awnd3xB86IFqBt5iEeel4ntHCGg3D3Bmmkj6PVuA2pUr0PTTsOZszMGoyv0RySc2s/WH+czZVRv2jR+n6gzFlDuc3b9VPq3qkP1mk3p8flmrtp79113OjfMCftZMXUUfdo1pEa1uoTVL+fyQchsiEFJvshfG1ezaNYUxnzYi7ZNujL3nKooFZIOfMfYod2oVVhnFwi6ItR6vw99+vRlwPRdJKXubHAaJH4FAUFAEBAEniqBLAuEB/vnMmbKfKZ1fh05pAcbPAWC+TCjy/tT68sLjrFjE38OfRV9SDfWC4WQoxn6+MQ82r9qQFekGZO3HWPvnPY23wBJX5z3vo+x8zcfYXanmoT6S/ZKWArk3S9W8Wnz8pQo9SoF/JzHDbw94WiqGQZW4reMpGaIDjm0DVM2HuJU9FamtS2J7HBSzEwgWGO3ENmtBoX1jnj0bzH2qOp1qHBn33dMGjeEsFCnQChJ84hxjJ/wGdPWnhZTHHO0tIjABAFBQBD49wSyLBDsUSjcXtQC//w92eRuroL1Or9PH8+iI/cdllg48Vll5Lwd+cF56N/bKO50EkjeweAyMpJGpsaUs/Ylcs3HGFdBbxMC+rKjOOjy/jeya1AJdGo3vkZLSLX/Me/ATduwz6MjY6ks2ytv/Rufcth1DxiPT6F2kIRGW5TeG92Zl+0hBut5vqgpO4YRUvkZWC8yvbbzXANmiiEGZw6LX0FAEBAEcg2BbAoEuLO4ZVqBkDo5SgKLWgZRoN0KEtQuYyWJDQPLUazqePb76NJOHYTYT03AypWZDcmjtuKlYMJXxHPr1i1uJZ7l68aO8XxDE+Zed/bRW4ieUAnZJhAMtFh0G+cZjDsYWExrq7wlz94g5TpL2uSzf+/ArzVL77ptyLZAUBKYG+b0M6hG5GkPpxUhENxgxZYgIAgIArmUwBMQCAqJm/rxxht9WRfvdD57xOHZ3Wg3YAXnPeqJXMokd5ql3GJRC3975S3lodgbVala1fO/GtUbf8y2ZKf5Fv6Z6EMgmPYwpIS9i18KcvuTKHeW825ee8+CNn9PNnr0Ev0bgTBPCARnZohfQUAQEASeOwI5LBAUkvZOpEnt/7Hmmke/9XOHJRcabDnF51Uc3fL6Cow7lhnf7AsEy6nPqSKrQxIapFSOqEIg5MIyIUwSBAQBQeAJEshBgaBw96+JhDX6kE3XnT0HT9Dyly1oJY45zqEEKZgOq5LcQwbpssi+QLDGzqaRQyBoDI2ZE+calEAIhHQhi4OCgCAgCLywBLIpEBSXD4Jn97NKx3x+Hq0qd2JVrFMcWDi7eBabnePYlvskP3See2F5PsGEWTgVWc01k8C/4gdsuOYeA7Dc3MfsAR+y7KqTsZmjY99C7/BBaLrgpltQGLczoKjDByHPe/zonGlic3hUnSBVP4e8NFsQ77rH9MeHlHR8KMm//SpSfRgxbbo9BY1cmUknPMaWLGeYXN3ppFiLaRecNqcNRhwRBAQBQUAQeDYEsikQrFz4siZynnasdDu4gzWG+c0KUG3MH1y6coUrV65w6dhKelfvbZ8OqSTyfYcQ5NCBbHfXac8mxc9xrMrtzfQv7ajANRLawFAq1QujWaOqlAzyo1T3n4lz1bX3Wd/tFfvXCjUylSedsM96UH1G4+cR5pwCKb/NhGjncIXC9dWdKKZz+CEUbMDIJev4ceaHNC0ThN42zVGDNqQuw5fvJzajvDTuZ2QZ++wKdUZE3y0P3eTvraNbIbtAUc/12uhynHBfI7YEAUFAEBAEnimBrAuE5H1EDepIlYI6JCkPpZr0ZerW67Z590ridzR3zq23tVjt49hy3a+4bKuwTJz/eSwRk7fh8lt8psl+fiM3Xl7Pp63Kkc/5jQGNhFygIuGTd3DdUc9br62if82SBGod3yFQfQoCStLs62M8PhVF21KBDuFgX1tBn78cjT7a7PhI0UNOLh9IrVf0tp4EyVCEmr1nsPPAVzSUJfSBxXizbks6Dfya3bfdQxCeRJN/+5gGZUJcgkJdv0GXrzT1hqxm97xu1CgRYO+lcJQVbd6S1Oy9hIsenQye4YltQUAQEAQEgadPIOsCAStmkxmLVa0UFKwWM2ZXa9WKxZK6skjv2NNP4Isao/leDGeOH+HoqaskpW7JK2aMKUbMFqt9iECxYjEbSTFZQT1nNDnyERTXuVS1c0oiF44f59xNxxexlAfEnr/KPWdnQ0ZgLSZSjM6yohYXR/xGC4o5hRSTBaviKC/Ocylm13BGRkGLc4KAICAICAJPh0A2BMLTMUjEIggIAoKAICAICALPnoAQCM8+D4QFgoAgIAgIAoJAriMgBEKuyxJhkCAgCAgCgoAg8OwJCIHw7PNAWCAICAKCgCAgCOQ6AkIg5LosEQYJAoKAICAICALPnoAQCM8+D4QFgoAgIAgIAoJAriMgBEKuyxJhkCAgCAgCgoAg8OwJCIHw7PNAWCAICAKCgCAgCOQ6AkIg5LosEQYJAoKAICAICALPnoAQCM8+D4QFgoAgIAgIAoJAriMgBEKuy5L/ZpBius/Ny8fZs24lv18wZS+wlBuc+uNXls+LYv73mzgU9zh794urBQFBQBAQBF4YAkIgvCBZab22mPBS+ZAl5wJNfrRb6bGCYobpVLi9dzItS+YhqGJHejUuhk5d4Mm/LAM2JmZ4pzgpCAgCgoAg8GISEALhRcpXSxwLWgTaV0qUAum01rHQUiZptJz/lqYFtGjkSralny0nJlHFoAoNHaVH7M/kbnFaEBAEBAFB4EUkIATCC5WrJnYNCrW1/jVSEF3WZUUg3GNdt0K25Z91JQaz2zYq8Zjzv0QybNgXbI11Ldn5QpESiREEBAFBQBDImIAQCBnzec7OmtgzpET2BELKJnoV1KLRaJCrTOJkqlWfnzMAwlxBQBAQBASBHCLwLwSCkbh9B7mUpiIxEv/3WuZPn0zk9O/YcjYZJYeMFMFklUD2BYJy/VveMWjsAqH+DK6JDoOswhbXCQKCgCDwQhPIhkAwcePv5YxuU5ag4C549V4riWwdVpeGg79n/+nT/DGrPaFB1Zh0NJte9C8YauXuCdZMG0GvdxtQo3odmnYazpydMRid6XyUwKn9W/lx/hRG9W5D4/ejOGMB5f5Z1k/tT6s61anZtAefb75K+iTNJOxfwdRRfWjXsAbV6oZRv5zTByGTIQYliQPfjWVot1oU1tkFgq5ILd7v04c+fQcwfVeSl8BLubaL+WP60Pad2lSrUp26TcMZNPkHjt7yVoqPbpzl0I41LPxiFL1aN6bzvLNYLHFs+7wzDeu0ZPjaK87Ui19BQBAQBASBXEwgywLhwf65jJkyn2mdX0cO6cEGj+Ft68Xp1A5qzneJjj4D43YGFvOj3jdXeVkbpI9PzKP9qwZ0RZoxedsx9s5pT6hOg6Qvznvfx9i4mI/MplPNUPwlewUtBb7LF6s+pXn5EpR6tQB+zuOGt5mQWmxZ49kysiYhOpnQNlPYeOgU0Vun0bak7HBSzEwg3GHfd5MYNyTMZpc6xKAr2ZyIceOZ8Nk01p52TnG0cG1NfyrklZDy1mDkr8c4vmsRA6upQkRCX7gxkX/ddYgJM8e+7UqdUH+7DRqJkK6LWPdBeVda5MqTcvHjIEwTBAQBQUAQcBLIskCw36Bwe1EL/PP3ZJOrGQxYEoj++xIPHKGmHJtI1YByDP/TWck4o3tJfpN3MLiMWlHL1JhyFlsb23yMcRX0tq58fdlRHDQ7WBh3MaiEznZcow2h2v/mceCmevIRR8ZWRraJBD1vfHoY5y1g5PiU2gRJGrRFe7PxvpNr9ocYbOJOdgwxNJiZZojBcn4mjYLsMxpKDtqFUxdar87hnbz2KZW64t1Zd9s9oGTcNYgSjl4JbUh5OkUuZUbfGhTK+woNph5xGit+BQFBQBAQBHIxgWwKBLizuGVageBMoPkOp3/9jJZv1WTgjxdclQlKEhsGlqNY1fHsT7+v3BnCC/Br5crMhuRRK3YpmPAV8dy6dYtbiWf5urHBLgQMTZh73VGhWqKZUEl2HG/BIs+KdsdAimnVylsipMcGF0/l+hLa5LNXzn6tl3LXRS2nBYKZg6PKoteoafGn7ffJrphQbrK4lXM4w4+GM929RZboCVRyio7Kn3HCexTCHYbYEgQEAUFAEMi1BHJQIFi5smYM/XqG0+SNAuQr246vDzodFR9xeHY32g1YwfkXvbJQbrGohaOLXcpDsTeqUrWq5381qjf+mG3OutbyDxN9CATTniGOlrhEUJd1DoGgcGf5u+S19Sxoyd9zo9ungRwWCNZrzGzgEC/agvTe7NVtxMlJVZBV8aCRCO7yi0vAWP6Z6BIIhuaL8NA8ufZBEIYJAoKAICAIeBPIQYHgEbDpDF81CEIuM4K/POsUj0te2E3LKT6v4qhU9RUYd8w9MJBumrMtECyc+tyzYnYKBzX0HBYIlpPutEhBdP7Fc8hI4frcJhgcAsG/3UrXEJOXQGghBEK6+S4OCgKCgCCQywk8GYGAlctf10WWGzH7ZfvQjhLHHOdQghRMh1XeswHSlIdsCwQrsbMbOVruGgyN5xDnGv7PYYGg3GZp6wCHw6FMjann7P4UtkR42qGj1LC/XDMthEBIk8vigCAgCAgCzx2BbAoExeWDsDHDngGLvfs5f1fWOT0XLfdJfvgyzGmwcCqymsO5UMK/4gdsuOaGZbm5j9kDPmTZVQcL81HGvmV3XtQYmrLgpqu2x7h9AEUdPgh53vuRR47iZT42jgqy3QdBytuMBfHOe0z88WFJx4eS/Gm/ygnfd7m0nJlMdae/QK1pXPDKIoWE5e3I77DBUPcrLrnOp7BrkOOjTLrSRHg4pJqPjuUtvd3x0RA2H48k+TZEnBEEBAFBQBDIVQSyKRCsXPiyJnKedqx0ec6DcucA301ewP47jorqwT4+qhDIW6P2YlsuSEnk+w4hyKED2e6uK3MViJw0Rrm9mf6lHdMNNRLawFAq1QujWaOqlAzyo1T3n4lzVrT319PtFfuXDDVyZSa5PPoU4ueFuaZAym9PINo5WqFcZ3WnYugcfggFG4xkybofmflhU8oE6R0tfi0hdYezfH9shkm7t64bhWwCQJ0R0YuNTt8I512WSyztEIpejUsXyvurrtlmU1jiVtCxkBaNZKDcgE1eIuD++m684ghT/+anHHHa7QxT/AoCgoAgIAjkegJZFwjJ+4ga1JEqBXVIUh5KNenL1K3XbfP5lYTNjGrwKq/V78+kKWPo3bQxnSK3EeOqGEyc/3ksEZO3Ee+sGHM9mv9moPHyej5tVY58entLX3XkkwtUJHzyDq7buFi5tqo/NUsGorVV9HZnv4CSzfj62GNORbWlVKBDONhmEejJX64RH212DFk8PMnygbV4xRa+hKFITXrP2MmBrxoiS3oCi71J3ZadGPj17vQTYjnDgm41KBHgtE+NX0vekjXpveSix1CC6toQy+9Tu1KjqD9aXQhlatXhrUJ+5AmtTc+v9nDD6Xhqvcaq/jUpGah1iBR19oNMwfKNGbfD2f+RvjniqCAgCAgCgkDuIpB1gYAVs8mMxar2EihYLWbMqSp768MbXI25TYqzxzt3pfWZWGO+F8OZ40c4euoqSal6TxSzkRSjkykoVgtmYwomK6jnjCYLNtwo7nPOytiRmpTECxw/fo6bjg8UKA9iOX/1nsc3E3wlW8GckoLJYkVx5Jcz/hSzjwxUHpN46QRHDh/j1P+3dybAURVr/541CyQQNgFDvIgQZZEtgaBAJOy7skbWT0AEFMX7XWSTXcQLlx0FkSVePtFElnsBEQtUEFnCUkQQBP4EAhYhIcVWQmpm6syp518zc85kspLoiAReqlKZnHO6++2nh3p//fbbp89l+uye0NtQ3fbbHLrdOX2y5bFbLyG/hYAQEAJC4MEkUAKB8GB2QKwSAkJACAgBISAE/E9ABIL/mUqNQkAICAEhIARKPQERCKV+CKUDQkAICAEhIAT8T0AEgv+ZSo1CQAgIASEgBEo9AREIpX4IpQNCQAgIASEgBPxPQASC/5lKjUJACAgBISAESj0BEQilfgilA0JACAgBISAE/E9ABIL/mUqNQkAICAEhIARKPQERCKV+CKUDQkAICAEhIAT8T0AEgv+ZSo1CQAgIASEgBEo9AREIf9EQqvZbpJ89zK7EdWw/bS+ZFdlXSNn9JauXLmLZ2s3sT7tbsvLytBAQAkJACAiBexAQgXAPQP6+7UxdSa+a5bEa9UOSAum+zn3mZTGaUsn8bgYdI4IJrf8ig2Kru492NgbVZvimq8UoL48IASEgBISAECgeAREIxePk36eUNJZ3CPGceGgMoc/n2klL92hFObWYthVNGKwNmHzEgXJsKo0CXELDTK03996jtNwWAkJACAgBIVB8AiIQis/Kj0/a2Tky3D37NxhD6ZdYHIFwg8T4KpgMBsw1XuMb96rEXU59MZ033pjN1ot5jtb0o7VSlRAQAkJACDx6BEQg/CVjbmfXqBolEwjZmxlUyYTBYMDaaCrH5fjkv2TkpFEhIASEwKNC4HcIBBtpe/ZxpigHpVxk+5LPSHE8KhhL2s+SCwT18mJeCDB4BMLz80iVgEFJocvzQkAICAEhUAICJRAIdq78uJp/dKlNaLl+FB4Vv8Oh6c0JLV/UMyWw8D48ql4/xoY5bzKoa0uimsbQts9YFn59AZve9p1fSdm7lfXLZjJucBdiey/ihALqrZMkzRpGp5imRLcdwLQt5yl4P4KDX/euYda4IXRvFUWTFnE8X0fPQbjHEoOaxQ8fTWBMfDMeM3sEgrlqM3oPGcKQocN5f2cWqm4nkJ26k2XvDKHbC81p0qgpLdr2YuSMTzmU4avo7nDl5H52bPiQ2eMG0Tm2L0tPKihp25jWtxUxHcfy+Tnf530akI9CQAgIASHwSBAotkC4vXcJ78xcxpy+T2ING8DGApfNVW7sGk+vprWxlu9PUoHPPFhc7x5bSo8nAjBXbceMbcl8t7AH4WYDRsvjvLT2Aq6JuuPgAvpEhxNk9DhoY0hXZieMp31kDWo+UZFA/XrAs0w+lEciOC/xn7eiCTNbCe8yk037UziydQ7dIqxakuK9BMI19nw0lYmj4tx2uZYYzBHteX3iJCa/O4fPf9K3OCqkbhhGvbJGjGWjeOvLZA7vXMGIJi4hYsTyWCzTv7/uEROOZBb3jyE8SNtJYQyj/4pEXo0M9NhksNJw6vEHa6DEGiEgBISAELivBIotEDxWqWSu6EBQhYFs9k6vc+xVM7cwpvcM9qx+ieDSIBBu7uC1p1yO2krUzJO458yOZCbWs7hD+Zba49inL5PYdjKyhtl93WAKo8krS/kh3XXzDgcnNMTqFgkWnh5/AL0I2Dg8szmhRgOmaoPZdEtnVfIlBucv79Pcqi0xtJyfb4lBOTWf1qGeHQ0RI3eiazPn+YW8UNYjBMyPv0xiph5vsLFzpJYHYTARFtmH6avmMTSqCmUrt2TWQb0G3Wb5LQSEgBAQAo8SgRIKBLi2smPBAkHNYNOovsw5auPu//XKLRDULDaOqEP1xpPYm2eC/dfBdnJufiuCXY7dWI5eay6RkZFBxtWTfBAb4BECAW1YcllzqMoRJjewatc7sMLraMG2YwTVTS7nbSRswEavc1Yvf0yX8h7nHNh5Fde9nfW3QHCwb1xtLAZXX4LotvamtyXUdFZ20pczAmk1/7w7KgIKRyY3wOoq44oYvHvMI5BySsonISAEhIAQeIQJ+EkgqFz98lV6zT7qXrfPzisQuMOBBfF0H76GUw/K0raawYoOQVqYP5jqTzemcWPfnyY0jf0723RfqxxlSiECwb5rFDXc+QFGQvslagJB5drqrpR1RxZMVBi4KSenAT8LBGcq81tq4sVUicFbfMM7CsenNtKEgJFy/b7Q7FM4OkUXCAG0X5GZK5fhEf4/IV0XAkJACAgBwC8CQU1P5JWuUzigRaXzC4QHkLWSwrRGmlO11GNics7CQIHWllggKKRM83XMunBw1e5ngaAcz+mLMZS+X+h5Ca62VC4vaUOAO1JgJKj7Om67O5hbIHQQgVDgsMtFISAEhMCjSsAPAkHl1486Uu1vjYmJiXH/NI+siNlSmcjmz9NrwQMaulbTWKgvJRjL0TMh926AfF+IEgsEJxcXtNZm7gYCYheSpi//+1sgqJms6lzGm2AYNetnn+UCXzvM1Hzje22nhQiEfGMsF4SAEBACQsBLoIQCQfXmIGzyjWKj4rDZyM7Odv/cWNeT4HJ9+eymHUV3isotbv72IG3eV0iZ3kRLLjQSVP9VNqbmdEpJ38OC4aP55Lxms+MQE+p6khcNAW1Znq53DGxfDaealoMQ/NJ67mh4HckTqWf15CAYy7Zj+SW9jJ3doyO0FyUF0SPBM6f3jkoBH5QTM2iqJyk2m8PpXChVfl3dnQqaDQEt5nLGez87JxnRXIvXv9WjCw4OTajryVswBBC3LF2WGArgLpeEgBAQAo8qgRIKBCen34vGGtyddd6M/Pzobif0ICi0L1/oifDqVdb2DMMaPoKvcnxw/oL3+Ypr18WwWtp2Q4MRU0g4DZ6Lo13rxkSEBlLz5c9I0x3trSTiK3veZGiwNmTqMT2ZQuXS0jjvFkjrs5M5oq9WqJf5d5/qmLU8hEot3+LjxPXMH92Wp0It2ozfRFiLsazee7HI3t9IjKeKWwC4dkQMYpOeG6GXUs6wqmc4Fldb5nB6J6S6d1MoaWt4sYoJgzGAOsM3k6NrbpEUX9n96maDwcIz4w/67L7QK5XfQkAICAEh8KgSKL5AuLmHRSNfpFElM0ZjMDXbDGXW1staRryGT81i97zhdHC9BMhUkSZ9JpHknoHbOfXZBF6fsY1LusN9QIjbziYxvlMdyls8M33XTgRrxfr0mrGDy25H7yQ1YRjRESGY3I7es1uhTEQ7Pki+S8qibtQM0YSDexeBhQp1WvP2Fm3J4rfjrB7RjMru+o0EVI1m8Lyv+WFuK6xGCyHVn6FFxz6M+OCbgokoJ1geH0WNMrp9rvZNlI2IZvDHv/gsJbhSGy6yfVZ/oqoFYTKH8VSzGOpWCSQ4vDkD5+7iiqZpnKkJDIuOIMSUU6fRWonI2Ins0MMfBVsjV4WAEBACQuARIVB8gYATh92B4nSFyVWcigNHAc7edd39jOpEcThylhgecKCOGxc4cfggh1LOk5UnyqE6bGTb9L6D6lRw2LKxO8F1z2ZXcGNBzbmnBxi0fmdfPc3hwz+TrkVV1NsXOXX+RjFm7SqO7GzsihNVW6HQ28926EsWeeCqd7l65hgHDySTci7TZ/eE9pzqwJZtw+Gq033JM56ua3nMzlOx/CkEhIAQEAKPCoESCIRHBYn0UwgIASEgBISAEBCBIN8BISAEhIAQEAJCIB8BEQj5kMgFISAEhIAQEAJCQASCfAeEgBAQAkJACAiBfAREIORDIheEgBAQAkJACAgBEQjyHRACQkAICAEhIATyERCBkA+JXBACQkAICAEhIAREIMh3QAgIASEgBISAEMhHQARCPiRyQQgIASEgBISAEBCBIN8BISAEhIAQEAJCIB8BEQj5kOS+oNpvkX72MLsS17H9tD33zXv8lX0lhd1frmbpomWs3byfNP0gxXuUk9tCQAgIASEgBP5qAiIQChkBZ+pKetUsj9WoH2gUSPd1vxXydJ7LaibfzehIRHAo9V8cRGx1M65DoIJqD2fT1ULOT8hThfwpBISAEBACQuCvJCACoSj6ShrLO4R4jmU2htDnc/386iILcWpxWyqaDFgbuI5+Vjg2tREB7mOYa/HmXv0s6KLqkHtCQAgIASEgBP5aAiIQiuRvZ+fIcMzuY5xD6ZdYDIFwI5H4Kq7jn83UeO0b3IsSd0/xxfQ3eGP2Vi4WcAJmkSbITSEgBISAEBACfwEBEQhFQreza1SNEgmE7M2DqGQyYDBYaTT1uByfXCRfuSkEhIAQEAIPKoHfIRBspO3Zxxkld5fU64dJ+vgT1n76KWtWLmXB3FnM3vATpTugXlKBoHJ58QsEuCIOBivPz0tFAga5vyfylxAQAkJACJQOAiUQCHau/Liaf3SpTWi5fuSNttu/H0vdJ6NoHdOUhvXr0aDJcwxcedqPM2iV68c2MOfNQXRtGUXTmLb0GbuQry/Yckjf+ZWUvVtZv2wm4wZ3Ibb3Ik4ooN46SdKsYXSKaUp02wFM23LeE/rPKal9cvDr3jXMGjeE7q2iaNIijufr6DkIRS8xqFk/8NGEMcQ3e8wTcTCYqdqsN0OGDGHo8PfZmeWbnJhN6s5lvDOkGy80b0Kjpi1o22skMz49REYu4XWHKyf3s2PDh8weN4jOsX1ZelJBSdvGtL6tiOk4ls/P5SqQr0dyQQgIASEgBITA7yFQbIFwe+8S3pm5jDl9n8QaNoCNeZbjf1vfh+ZTjvpREPh25y7HlvbgiQAzVdvNYFvydyzs4coNMGJ5/CXWXvDM0x0HF9AnOpwgV0KgwYAxpCuzE8bTPrIGNZ+oSKB+PeBZJh/Ks2XReYm0STQJAAATPklEQVT/vBVNmNlKeJeZbNqfwpGtc+gWYdWSFO8hEK7t4aOpExkVp+UsGMxEtH+diZMm8+6cz/lJ3+KopLJhWD3KGo2UjXqLL5MPs3PFCJqEGDEYLTwWO53vr2tiwpHM4v4xhAdpOymMYfRfkcirkYEemwxWGk497gtKPgsBISAEhIAQ8AuBYgsET2sqmSs6EFRhIJt9Ju7g5Jc5rem28hq+82S/WAjc3PEaT1mNGKxRzDzpmTE7kidSz+ISAhZqj9uXs5Rh28nIGq5thQYMpjCavLKUH9JdCx13ODihIVa3SLDw9PgDOWWwcXhmc0KNBkzVBrPplm55SZcYnPzyfnOs2hJDy/l5lxgUTs1v7W7HYI5g5E5dZTk5v/AFyrptM/P4y4lkekHa2DlSy4MwmAiL7MP0VfMYGlWFspVbMuugXodus/wWAkJACAgBIfDHCZRQIMC1lR0LEAh2do2O5Nkug+ndtTM9B43l/S9/xvvWADWLjSPqUL3xJPbmmbjfswvOc8xvFeyeMRvL9WLNpQwyMjK4evIDYgM8kYKANku4rDtU5QiTG1g9AiGgAytyPC22HSOo7k4gNBI2YCO6a1Uvf0yX8p5ZemDnVVz3GuVngeDYx7jaFk90I6gba296G0JNX0knVxTBFfkIbMX883r2gsKRyQ28oqPhu8f+pChNji3ySQgIASEgBISAnwSCgxObV5K4O5mfUg6wcWYnwq2VaL/0F82Z3eHAgni6D1/DqRIumasZK+ighdiNwdV5unFjGvv+NGlK7N+34fW1ylGmFCIQ7LtGUcPsEhVGQvslagJB5drqrtrs3USFgZvICY74VyA4U+fT0uoRNaZKg9mS0xAox5naSBM2xnL0+0KXLwpHp+gCIYD2KzL/lCiN/FcQAkJACAgBIeBLwE8CwbdKQL3M8rZlsDw5lu9LGjHIU5WSMo1GmlO11JtI8r22RZRYICikTGukzdCNlPMKB5ch/hUIyvGcvhhD+/KFnpfgakq9zJI2AZ7IhzGI7utuayRyC4QOIhDyfEPkTyEgBISAEPgzCPw5AgEHe9+shSWoBwnedYbfZ76attC7lGAs15OEXLsBCqizxALBycUFrTWBYCAgdiFp+nKFnwWCmrmKzmW0hENrFLN+9gmnOC+yoLUWQTDX5A2vshKBUMAoyyUhIASEgBD4kwmUUCCo3hyETb7h8XxG2vjv0CpYnxzLHj2CoNzi5m/6unq+AoVfUFKY3kTfSRBE/Vc3kuptWyF9zwKGj/4E75K94xAT6nrW+Q0BbVme7vX22L4aTjUtByH4pfXc0Vp1Jzy6kiBd6/9l27H8kl7Gzu7REdqLkoLokaDP6gszV+HEjKbefIFmc07nfg+C+iuru1fA5H4zYwAt5p7JuZ+dk1xprvU633qjCw4OTaiLxZ34GEDcsnRZYigMv1wXAkJACAgBvxEooUBwcvq9aKzB3VnnzfQH7u5n/tDhLNh3w+281OubGBweSvPZxz07BdSrrO0ZhjV8BF95nXtx+6CSuWUYtTQHbjCaCAlvwHNx7WjdOILQwJq8/FlajqO9lUR8Zderjg0YrA2ZekyfpatcWhrn3QJpfdZ1ToJmg3qZf/epjtm9i8BEpZZv8XHieuaPbstToRZtS6GJsBZjWb33YhGG3yAxvopHABhMVBu0KSc3QiulnFlFz3BPnebw3iSkuoxQSFvzIlVMBowBdRi+2VcE3CIpvrJWp4Vnxh/02X1RhClySwgIASEgBITAHyBQfIFwcw+LRr5Io0pmjMZgarYZyqytlz2O2XGada9EE17zBUZPn8TQ2Ka0G7+Fi7pvxs6pzybw+oxtXPodQQSwcTZpPJ3qlMeivcvAYLRSsX4vZuy4rDlMJ6kJw4iOCMGkP2MwUiaiHR8k3yVlUTdqhmjCwT2Dt1ChTmve3pLlmZH/dpzVI5pR2eKKJBgJqBrN4Hlf88PcVliNFkKqP0OLjn0Y8cE3BeJWTiwnPqoGZbxtu7ZZliUiejAf/+IF4S5rv7idWf2jqBZkwhz2FM1i6lIlMJjw5gOZu+uKd5eCMzWBYdERhJi0ZQlXhMNaicjYiezQwx8FWiMXhYAQEAJCQAj8MQLFFwg4cdgdKE5X+F3FqThw5HX29ixSz17mdm5/+McszFXawY0LJzh88BAp57N8dht4HlIdNrJtuo2gOhUctmzsTnDds9kV3Oaj5tzLY2v21dMcPvwz6domAvX2RU6dv3HvWbvqIDvb7ubjWaDwMLJlZ+PQVyxy9QXUu1c5c+wgB5JTOJdZQGhFdWDLtuFQnNqygl6nzSsi8lQpfwoBISAEhIAQ8AuBEggEv7QnlQgBISAEhIAQEAKlgIAIhFIwSGKiEBACQkAICIH7TUAEwv0mLu0JASEgBISAECgFBEQglIJBEhOFgBAQAkJACNxvAiIQ7jdxaU8ICAEhIASEQCkgIAKhFAySmCgEhIAQEAJC4H4TEIFwv4lLe0JACAgBISAESgEBEQilYJDERCEgBISAEBAC95uACIT7TVzaEwJCQAgIASFQCgiIQCgFgyQmCgEhIASEgBC43wREINxv4n9pe3e48P2nrPlGO0OjuLY4svh5915+0V4/Xdxi8pwQEAJCQAiUXgIiEErv2JXMcuUkH730JMFGI2H9k7xHXRdZiXqNvf8aTPPqgVibzuBEnnMriiwrN4WAEBACQqBUExCBoCooeQ+dKtVDWojxtq8YUd1zmqXlqbfYqx91Xcjjrstq2hLigl0nSVppNPW4HBBVBCu5JQSEgBB42Ag8sgLBdva/zB3VjcaP12Lk1/aHbVzz9+fGWrqX0Y6NttRn0uF7hQMcpMyIItB1fLXpMV7ZWsBpk/lbkStCQAgIASHwkBB4RAWCnW/H/A2zwYDB8hRvFWc6XcoH3LFvHJFhYZQzGTAYLNSbmFzkEdbqlX/TO6IMQS6BYH2eeecfhTBLKR9kMV8ICAEh4EcCv0Mg2Ejbs48zRU1AHdc4/p91bC/yIT/2osRVZZPUvxxGt0Aozmy6xA08YAWc/DLneRr/72JG1zJjMBiw1J1AcmHLDGo6XwxoRKd3XiPaasBUaQj/kQDCAzamYo4QEAJC4M8lUAKBYOfKj6v5R5fahJbrR2KBGe1O0ne9R7+4Hvzj30fJKkpEFLdfd65wcv8ONnw4m3GDOhPbdyknFYW0bdPo2yqGjmM/55xPO+r1Y2yY8yaDurYkqmkMbfuMZeHXF9D9m5p1kE8mjaBNTatHIJgq0eTFwQwZMoT/GbeaTStfZ+DAQQweMoQhgwcxeNKXXHKC8/JGJg0awMBBnmcHDRrNxyl6w3e4cnI/OzZ8yOxxg+gc25elJxWUtG1M69uKmI5j+fycwp1fU9i7dT3LZo5jcJdYei86gYLKrZNJzBrWiZim0bQdMI0t5/285OG8yMK4eoz74S5HJtfH6hJG1gZMOarb7zsYDk4v70mr1/7LpV1jeMJswBL5d34sTEz4FpXPQkAICAEh8NAQKLZAuL13Ce/MXMacvk9iDRvAxnwCQeHCZ4N4+ukBrD/vP2/iSF5M/5hwT6jbYMAY1p8Via8SGaitp1sbMvW4x9HdPbaUHk8EYK7ajhnbkvluYQ/CzQaMlsd5ae0FXEFyZ9p2/jXlHXo9owkE8+O0GTmRSZMmM3XR15y98hNb/rcZZV2hdddMW0/oU65z5pt/0rW6ZwZuMJal9wYNgiOZxf1jCA/SbDKG0X9FIq9GBnpEiMFKw6mHObigD9HhQdo1IyFdZ5Mwvj2RNWryREX9WSMBz07mkB81gvrrh3Sq+zrf2kD5eRZRVk/iYdMZLoHi+08l/b9jaNv7Q362w62EHgQbDVibzeG0rDD4gpLPQkAICIGHnkCxBYKHhErmig4EVRjIZn1KriGy/zSXlhUjGfPNTf9Ds+1kZA3NMZvCiOwznVXzhhJVpSyVW87ioMtP39zBa0+5nL6VqJknPY7PkczEehaPo689jn1e3ZLN+pc0R21tysyTud2kM3U+La2aQHh6PAd9yn3RN9Tj4I3liU/yVUk2do6s4clrMJgIi+zD9FXzGBpVhbKVWzLLbSTYdo6khtlTtymsCa8s/YF0V/13DjKhoSZaLE8z/oC30T/I08m5+bFEjt7tiaI4zzLveZcYMWJt/j6/eB2/yrVd44nrMJ0fb7madHJ2bow72hAQt5Rf1T9ohhQXAkJACAiBUkWghAIBrq3smF8gqBn8u1dlQuPmsvfgt3y9az+nMn0UhJrFxhF1qN54Ent/z8xYOcLkBla3ozdYG/LusdwO3eXMzs1v5Z7tGozl6LXmEhkZGWRcPckHsQGecgFtWHJZ93K/sa57oFZfE6bn3eB/fRWdAzWB8Mw7HPL6ahvbXqmKyRWiN4bx8pe+AkHhyOQGnvC9K2Lw7rE8s3PP90I5MpkGmvgI6LCCTN0kbOwYUd1b94D8IZrf98VSUpge9RSjd+njoZK2JM7NyhgUx1I3E5WbP0wn7rlR/PeqbpCDA/94GovBQEB7Xzt/nxlSSggIASEgBEoXAb8IBPXaGrqHWnmy8zssWruOZZNfJDIsgk7/PMhvbh53OLAgnu7D13Aqr28vDi/lKFN0gRDQnhU5XtVTWs1gRQctImAMpvrTjWnc2PenCU1j/842b3DjJqu76AKhMdO9uQSaMTc+yREIdXMLhO3DChcIR6foAiGA9isy0V2tbxeVo1MKEQh2do3SIhDGUPoVnOThW1WxPt/ZNZratcawW9cHrtjAxYXEupZojOV48dMsbuybQWzttvzrJ5+HsPPd656dHoGdVpFVrNbkISEgBISAEHhYCPhFILi20NW2/o0x3+rhATvJE+tjDW7D4gveGPbvZ5ZLIHTILxCUFKY10iIMlnpMLDQ9Xzchi4876QKhkTeHQb+LHwRChwdBIDjT+LBDBepNOJR7S6PyM7OiXLxMVGzRmRaVq9L9E0+OhpcBdnaPecK9ZBLQdjnewELOA/JJCAgBISAEHmICfhEI9u/H8qQ19/sE7LtH84Q1gtG7ddHwByjeSyCoaSzUlxKM5eiZkFXg7N1rgXqNlR20pQfrs/mz+W98QpdClhiKG0F4EATCne/fJLJMM97LF7ax860WHTAYjIS8sJCz+XScgx//Xse9xCCvWfZ+c+SDEBACQuCRIVBCgaB6cxA2+USjXSHr1oFh9E+84wVn+2o41QJjeP+M5nmUW9z8LZ8X8j5f5AfHISbU9SQbGgLiWJaeN3ivkDK9CVb3zgMjQfVfZWNqjoFK+h4WDB/NJ96X/WTxcUctgmCpzbgfvEkGHjOyE+lXzrMjwVRtONu9VdnZMeJxLU+gDC+tz+kvODg0oa7boRoMAcQtSy9QpDgOTaCuxZPf4JqZ53TFxlfDq2l1B+epu0g6Bd+0HWZq02BC2i4jLR92lYyP2hPgyqUwRzDiK89CUO6KnJz753PunApTWF8+dycu5n5C/hICQkAICIGHl0AJBYKT0+9FYw3uzjpfh+E8z+K48lR9OYkbblYKx6Y2omLbpbh9snqVtT3DsIaP4Cuvsy0B1FtJxFf2nCNgsDzD+JxtBd5K1MwtDKul7QIwGDGFhNPguTjatW5MRGggNV/+zMdR2siJBJgJ7zCBJR/NZdrK/Z5DjJznWdBay2kwladh/CTe/2AW4/8njsgwvQ0TFVu+TcKPl/DESG6RFF/Z4+ANFp4ZfzB3WF+z9FZSPJXdbzM0YG04FW++pXqJpXH6Fkgrz04+UmB5b4eL/HCXIzNjCLFEMHyb70DlFMr+oi+hRgPmiNf4xjfXMucR7LtGEeHacWGqTI+Vp7mZnU9p+DwtH4WAEBACQuBhIlB8gXBzD4tGvkijSmaMxmBqthnKrK05xwY7zifx1gvP0mrIZGZOjKf1C2NIuqBnJNo59dkEXp+xzf3SoeIDdJKaMIzoiBBM2nsJXCFxa6VIYifuyHcioe1sEuM71aG8RXsfgevZivXpNWMHl/MGCY4spEetsu4ti0ZTEJXrd2fq9pz+3P5xNm2qesSA6371qL5M+fQA+99rgdVoIeTxejzXqS8j5n9Lxv9LYFh0BCEmvV0DRmslImMnskMPMjhTSRgWTUSISXsPgmsnRBki2n1A8t0UFnWrSYgmHFzvXzBaKlCn9dtsycobLbkXPQfnEvrxN2sgTw7+vJDtiU5S58cRHjWU5YeuFxjpcLdyZzsjwj3bS00V41j0sz6e97JB7gsBISAEhEBpJ1B8gYATh92B4nQ5LBWn4sCRb0Jp52baaU6ezaSQSWmJeakOG9k2vV1X004U97XCnZXjxgVOHD7IoZTzZBUVsXDc5OKJ45y5VshD2Vc5lZzMT5duu1+y5DJevXWBUxdu5t7CqDqwZdtwKE7N2Xr4uK7lWKnisGVjcyi4EWoMbdl2XKUcNht2vbyql7f7lC8GOiWd7+YNJPa5jry6eB+Z+cZHr0Pltxs3ixGhcHJu/dsMeWMum08XtAyh1ye/hYAQEAJC4GEjUAKB8LB1XfojBISAEBACQkAIFEZABEJhZOS6EBACQkAICIFHmIAIhEd48KXrQkAICAEhIAQKI/D/Afp44DWWU9rxAAAAAElFTkSuQmCC)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "f8NJ1rPavbi3"
      },
      "source": [
        "Then we calculated the similarity of the current state-action pair, say $p_t(s_t,a_t)$, to each existing historical state-action pair in the memory. In this work, we adopt cosine similarity as:"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LlkT0ah2vyYQ"
      },
      "source": [
        "![image.png](data:image/png;base64,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)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "eded1f-OvdnJ"
      },
      "source": [
        "where the first term measures the state similarity and the second term evaluates the action similarity. Parameter $\\alpha$ controls the balance of two similarities.\n",
        "\n",
        "The proposed framework is as follows:"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yDst-u-dv1sk"
      },
      "source": [
        "![image.png](data:image/png;base64,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)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CvHJ6duyvgSA"
      },
      "source": [
        "The framework works like this:\n",
        "\n",
        "**Input**: Current state $s_t$ , Item space $\\mathcal{I}$, the length of recommendation list $K$.\n",
        "**Output**: Recommendation list $a_t$.\n",
        "\n",
        "1. Generate $w_t = {w_t^1 , ··· , w_t^K}$ according to $f_\\theta\\pi : s_t \\to w_t$ where $f_\\theta\\pi$ is a function parametrized by $\\theta^\\pi$, mapping from the state space to the weight representation space\n",
        "2. For $k = 1:K$ do\n",
        "    1. Score items in $\\mathcal{I}$ according to $score_i = w_t^ke_i^T$\n",
        "    2. Select an item with highest score $a_t^k$\n",
        "    3. Add item $a_t^k$ in the bottom of $a_t$\n",
        "    4. Remove item $a_t^k$ from $\\mathcal{I}$\n",
        "3. end for\n",
        "4. return $a_t$"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "n5mSyGfby5iE"
      },
      "source": [
        "### Setup"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_7FcQ8w7tIVf",
        "outputId": "a3bd33ed-ef59-403c-8db0-e3989ef68d89"
      },
      "source": [
        "%tensorflow_version 1.x"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "TensorFlow 1.x selected.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6DvK7jdgtLpr",
        "outputId": "ab5fddba-53ed-433a-874a-42535e2483af"
      },
      "source": [
        "import itertools\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import random\n",
        "import csv\n",
        "import time\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "import tensorflow as tf\n",
        "\n",
        "import keras.backend as K\n",
        "from keras import Sequential\n",
        "from keras.layers import Dense, Dropout"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Using TensorFlow backend.\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NPnJFRj8y8Mm"
      },
      "source": [
        "### Download data"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8c2XxhlZy_ZD"
      },
      "source": [
        "Downloading Movielens dataset from official source"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SlRThma7tMNq"
      },
      "source": [
        "!wget http://files.grouplens.org/datasets/movielens/ml-100k.zip\n",
        "!unzip -q ml-100k.zip"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-YBKO19ZzFmo"
      },
      "source": [
        "### DataGenerator `class`\n",
        "1. Load the data into pandas dataframe\n",
        "2. List down user's rating history in chronological order\n",
        "3. Generate a sample of state-action pair\n",
        "4. Split the data into train/test\n",
        "5. Store the data back into csv file format"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BvX6KQirtgJV"
      },
      "source": [
        "#collapse-hide\n",
        "class DataGenerator():\n",
        "  def __init__(self, datapath, itempath):\n",
        "    '''\n",
        "    Load data from the DB MovieLens\n",
        "    List the users and the items\n",
        "    List all the users histories\n",
        "    '''\n",
        "    self.data  = self.load_data(datapath, itempath)\n",
        "    self.users = self.data['userId'].unique()   #list of all users\n",
        "    self.items = self.data['itemId'].unique()   #list of all items\n",
        "    self.histo = self.generate_history()\n",
        "    self.train = []\n",
        "    self.test  = []\n",
        "\n",
        "  def load_data(self, datapath, itempath):\n",
        "    '''\n",
        "    Load the data and merge the name of each movie. \n",
        "    A row corresponds to a rate given by a user to a movie.\n",
        "\n",
        "     Parameters\n",
        "    ----------\n",
        "    datapath :  string\n",
        "                path to the data 100k MovieLens\n",
        "                contains usersId;itemId;rating \n",
        "    itempath :  string\n",
        "                path to the data 100k MovieLens\n",
        "                contains itemId;itemName\n",
        "     Returns\n",
        "    -------\n",
        "    result :    DataFrame\n",
        "                Contains all the ratings \n",
        "    '''\n",
        "    data = pd.read_csv(datapath, sep='\\t', \n",
        "                       names=['userId', 'itemId', 'rating', 'timestamp'])\n",
        "    movie_titles = pd.read_csv(itempath, sep='|', names=['itemId', 'itemName'],\n",
        "                           usecols=range(2), encoding='latin-1')\n",
        "    return data.merge(movie_titles,on='itemId', how='left')\n",
        "\n",
        "\n",
        "  def generate_history(self):\n",
        "    '''\n",
        "    Group all rates given by users and store them from older to most recent.\n",
        "    \n",
        "    Returns\n",
        "    -------\n",
        "    result :    List(DataFrame)\n",
        "                List of the historic for each user\n",
        "    '''\n",
        "    historic_users = []\n",
        "    for i, u in enumerate(self.users):\n",
        "      temp = self.data[self.data['userId'] == u]\n",
        "      temp = temp.sort_values('timestamp').reset_index()\n",
        "      temp.drop('index', axis=1, inplace=True)\n",
        "      historic_users.append(temp)\n",
        "    return historic_users\n",
        "\n",
        "  def sample_history(self, user_histo, action_ratio=0.8, max_samp_by_user=5,  max_state=100, max_action=50, nb_states=[], nb_actions=[]):\n",
        "    '''\n",
        "    For a given history, make one or multiple sampling.\n",
        "    If no optional argument given for nb_states and nb_actions, then the sampling\n",
        "    is random and each sample can have differents size for action and state.\n",
        "    To normalize sampling we need to give list of the numbers of states and actions\n",
        "    to be sampled.\n",
        "\n",
        "    Parameters\n",
        "    ----------\n",
        "    user_histo :  DataFrame\n",
        "                      historic of user\n",
        "    delimiter :       string, optional\n",
        "                      delimiter for the csv\n",
        "    action_ratio :    float, optional\n",
        "                      ratio form which movies in history will be selected\n",
        "    max_samp_by_user: int, optional\n",
        "                      Nulber max of sample to make by user\n",
        "    max_state :       int, optional\n",
        "                      Number max of movies to take for the 'state' column\n",
        "    max_action :      int, optional\n",
        "                      Number max of movies to take for the 'action' action\n",
        "    nb_states :       array(int), optional\n",
        "                      Numbers of movies to be taken for each sample made on user's historic\n",
        "    nb_actions :      array(int), optional\n",
        "                      Numbers of rating to be taken for each sample made on user's historic\n",
        "    \n",
        "    Returns\n",
        "    -------\n",
        "    states :         List(String)\n",
        "                     All the states sampled, format of a sample: itemId&rating\n",
        "    actions :        List(String)\n",
        "                     All the actions sampled, format of a sample: itemId&rating\n",
        "  \n",
        "\n",
        "    Notes\n",
        "    -----\n",
        "    States must be before(timestamp<) the actions.\n",
        "    If given, size of nb_states is the numbller of sample by user\n",
        "    sizes of nb_states and nb_actions must be equals\n",
        "    '''\n",
        "\n",
        "    n = len(user_histo)\n",
        "    sep = int(action_ratio * n)\n",
        "    nb_sample = random.randint(1, max_samp_by_user)\n",
        "    if not nb_states:\n",
        "      nb_states = [min(random.randint(1, sep), max_state) for i in range(nb_sample)]\n",
        "    if not nb_actions:\n",
        "      nb_actions = [min(random.randint(1, n - sep), max_action) for i in range(nb_sample)]\n",
        "    assert len(nb_states) == len(nb_actions), 'Given array must have the same size'\n",
        "    \n",
        "    states  = []\n",
        "    actions = []\n",
        "    # SELECT SAMPLES IN HISTORY\n",
        "    for i in range(len(nb_states)):\n",
        "      sample_states = user_histo.iloc[0:sep].sample(nb_states[i])\n",
        "      sample_actions = user_histo.iloc[-(n - sep):].sample(nb_actions[i])\n",
        "      \n",
        "      sample_state =  []\n",
        "      sample_action = []\n",
        "      for j in range(nb_states[i]):\n",
        "        row   = sample_states.iloc[j]\n",
        "        # FORMAT STATE\n",
        "        state = str(row.loc['itemId']) + '&' + str(row.loc['rating'])\n",
        "        sample_state.append(state)\n",
        "      \n",
        "      for j in range(nb_actions[i]):\n",
        "        row    = sample_actions.iloc[j]\n",
        "        # FORMAT ACTION\n",
        "        action = str(row.loc['itemId']) + '&' + str(row.loc['rating'])\n",
        "        sample_action.append(action)\n",
        "\n",
        "      states.append(sample_state)\n",
        "      actions.append(sample_action)\n",
        "    return states, actions\n",
        "\n",
        "  def gen_train_test(self, test_ratio, seed=None):\n",
        "    '''\n",
        "    Shuffle the historic of users and separate it in a train and a test set.\n",
        "    Store the ids for each set.\n",
        "    An user can't be in both set.\n",
        "\n",
        "     Parameters\n",
        "    ----------\n",
        "    test_ratio :  float\n",
        "                  Ratio to control the sizes of the sets\n",
        "    seed       :  float\n",
        "                  Seed on the shuffle\n",
        "    '''\n",
        "    n = len(self.histo)\n",
        "\n",
        "    if seed is not None:\n",
        "      random.Random(seed).shuffle(self.histo)\n",
        "    else:\n",
        "      random.shuffle(self.histo)\n",
        "\n",
        "    self.train = self.histo[:int((test_ratio * n))]\n",
        "    self.test  = self.histo[int((test_ratio * n)):]\n",
        "    self.user_train = [h.iloc[0,0] for h in self.train]\n",
        "    self.user_test  = [h.iloc[0,0] for h in self.test]\n",
        "    \n",
        "\n",
        "  def write_csv(self, filename, histo_to_write, delimiter=';', action_ratio=0.8, max_samp_by_user=5, max_state=100, max_action=50, nb_states=[], nb_actions=[]):\n",
        "    '''\n",
        "    From  a given historic, create a csv file with the format:\n",
        "    columns : state;action_reward;n_state\n",
        "    rows    : itemid&rating1 | itemid&rating2 | ... ; itemid&rating3 | ... | itemid&rating4; itemid&rating1 | itemid&rating2 | itemid&rating3 | ... | item&rating4\n",
        "    at filename location.\n",
        "\n",
        "    Parameters\n",
        "    ----------\n",
        "    filename :        string\n",
        "                      path to the file to be produced\n",
        "    histo_to_write :  List(DataFrame)\n",
        "                      List of the historic for each user\n",
        "    delimiter :       string, optional\n",
        "                      delimiter for the csv\n",
        "    action_ratio :    float, optional\n",
        "                      ratio form which movies in history will be selected\n",
        "    max_samp_by_user: int, optional\n",
        "                      Nulber max of sample to make by user\n",
        "    max_state :       int, optional\n",
        "                      Number max of movies to take for the 'state' column\n",
        "    max_action :      int, optional\n",
        "                      Number max of movies to take for the 'action' action\n",
        "    nb_states :       array(int), optional\n",
        "                      Numbers of movies to be taken for each sample made on user's historic\n",
        "    nb_actions :      array(int), optional\n",
        "                      Numbers of rating to be taken for each sample made on user's historic\n",
        "\n",
        "    Notes\n",
        "    -----\n",
        "    if given, size of nb_states is the numbller of sample by user\n",
        "    sizes of nb_states and nb_actions must be equals\n",
        "\n",
        "    '''\n",
        "    with open(filename, mode='w') as file:\n",
        "      f_writer = csv.writer(file, delimiter=delimiter)\n",
        "      f_writer.writerow(['state', 'action_reward', 'n_state'])\n",
        "      for user_histo in histo_to_write:\n",
        "        states, actions = self.sample_history(user_histo, action_ratio, max_samp_by_user, max_state, max_action, nb_states, nb_actions)\n",
        "        for i in range(len(states)):\n",
        "          # FORMAT STATE\n",
        "          state_str   = '|'.join(states[i])\n",
        "          # FORMAT ACTION\n",
        "          action_str  = '|'.join(actions[i])\n",
        "          # FORMAT N_STATE\n",
        "          n_state_str = state_str + '|' + action_str\n",
        "          f_writer.writerow([state_str, action_str, n_state_str])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VQbGG-QC0ZeJ"
      },
      "source": [
        "### EmbeddingsGenerator `class`\n",
        "1. Load the data\n",
        "2. Build a keras sequential model\n",
        "3. Convert train and test set into required format\n",
        "4. Train and evaluate the model\n",
        "5. Generate item embeddings for each movie id\n",
        "6. Save the embeddings into a csv file"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "O45OFa8ytgGG"
      },
      "source": [
        "#collapse-hide\n",
        "class EmbeddingsGenerator:\n",
        "  def  __init__(self, train_users, data):\n",
        "    self.train_users = train_users\n",
        "\n",
        "    #preprocess\n",
        "    self.data = data.sort_values(by=['timestamp'])\n",
        "    #make them start at 0\n",
        "    self.data['userId'] = self.data['userId'] - 1\n",
        "    self.data['itemId'] = self.data['itemId'] - 1\n",
        "    self.user_count = self.data['userId'].max() + 1\n",
        "    self.movie_count = self.data['itemId'].max() + 1\n",
        "    self.user_movies = {} #list of rated movies by each user\n",
        "    for userId in range(self.user_count):\n",
        "      self.user_movies[userId] = self.data[self.data.userId == userId]['itemId'].tolist()\n",
        "    self.m = self.model()\n",
        "\n",
        "  def model(self, hidden_layer_size=100):\n",
        "    m = Sequential()\n",
        "    m.add(Dense(hidden_layer_size, input_shape=(1, self.movie_count)))\n",
        "    m.add(Dropout(0.2))\n",
        "    m.add(Dense(self.movie_count, activation='softmax'))\n",
        "    m.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])\n",
        "    return m\n",
        "  \n",
        "  def generate_input(self, user_id):\n",
        "    '''\n",
        "    Returns a context and a target for the user_id\n",
        "    context: user's history with one random movie removed\n",
        "    target: id of random removed movie\n",
        "    '''\n",
        "    user_movies_count = len(self.user_movies[user_id])\n",
        "    #picking random movie\n",
        "    random_index = np.random.randint(0, user_movies_count-1) # -1 avoids taking the last movie\n",
        "    #setting target\n",
        "    target = np.zeros((1, self.movie_count))\n",
        "    target[0][self.user_movies[user_id][random_index]] = 1\n",
        "    #setting context\n",
        "    context = np.zeros((1, self.movie_count))\n",
        "    context[0][self.user_movies[user_id][:random_index] + self.user_movies[user_id][random_index+1:]] = 1\n",
        "    return context, target\n",
        "\n",
        "  def train(self, nb_epochs = 300, batch_size = 10000):\n",
        "    '''\n",
        "    Trains the model from train_users's history\n",
        "    '''\n",
        "    for i in range(nb_epochs):\n",
        "      print('%d/%d' % (i+1, nb_epochs))\n",
        "      batch = [self.generate_input(user_id=np.random.choice(self.train_users) - 1) for _ in range(batch_size)]\n",
        "      X_train = np.array([b[0] for b in batch])\n",
        "      y_train = np.array([b[1] for b in batch])\n",
        "      self.m.fit(X_train, y_train, epochs=1, validation_split=0.5)\n",
        "\n",
        "  def test(self, test_users, batch_size = 100000):\n",
        "    '''\n",
        "    Returns [loss, accuracy] on the test set\n",
        "    '''\n",
        "    batch_test = [self.generate_input(user_id=np.random.choice(test_users) - 1) for _ in range(batch_size)]\n",
        "    X_test = np.array([b[0] for b in batch_test])\n",
        "    y_test = np.array([b[1] for b in batch_test])\n",
        "    return self.m.evaluate(X_test, y_test)\n",
        "\n",
        "  def save_embeddings(self, file_name):\n",
        "    '''\n",
        "    Generates a csv file containg the vector embedding for each movie.\n",
        "    '''\n",
        "    inp = self.m.input                                           # input placeholder\n",
        "    outputs = [layer.output for layer in self.m.layers]          # all layer outputs\n",
        "    functor = K.function([inp, K.learning_phase()], outputs )   # evaluation function\n",
        "\n",
        "    #append embeddings to vectors\n",
        "    vectors = []\n",
        "    for movie_id in range(self.movie_count):\n",
        "      movie = np.zeros((1, 1, self.movie_count))\n",
        "      movie[0][0][movie_id] = 1\n",
        "      layer_outs = functor([movie])\n",
        "      vector = [str(v) for v in layer_outs[0][0][0]]\n",
        "      vector = '|'.join(vector)\n",
        "      vectors.append([movie_id, vector])\n",
        "\n",
        "    #saves as a csv file\n",
        "    embeddings = pd.DataFrame(vectors, columns=['item_id', 'vectors']).astype({'item_id': 'int32'})\n",
        "    embeddings.to_csv(file_name, sep=';', index=False)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GPDw4iPw1N3U"
      },
      "source": [
        "### Embeddings `helper class`"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Pkl50IeYtgCE"
      },
      "source": [
        "#collapse-hide\n",
        "class Embeddings:\n",
        "  def __init__(self, item_embeddings):\n",
        "    self.item_embeddings = item_embeddings\n",
        "  \n",
        "  def size(self):\n",
        "    return self.item_embeddings.shape[1]\n",
        "  \n",
        "  def get_embedding_vector(self):\n",
        "    return self.item_embeddings\n",
        "  \n",
        "  def get_embedding(self, item_index):\n",
        "    return self.item_embeddings[item_index]\n",
        "\n",
        "  def embed(self, item_list):\n",
        "    return np.array([self.get_embedding(item) for item in item_list])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5e1gKdTW1ezk"
      },
      "source": [
        "### read_file `helper function`\n",
        "\n",
        "This function will read the stored data csv files into pandas dataframe"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eQMNYF2R1XmX"
      },
      "source": [
        "#collapse-hide\n",
        "def read_file(data_path):\n",
        "  ''' Load data from train.csv or test.csv. '''\n",
        "\n",
        "  data = pd.read_csv(data_path, sep=';')\n",
        "  for col in ['state', 'n_state', 'action_reward']:\n",
        "    data[col] = [np.array([[np.int(k) for k in ee.split('&')] for ee in e.split('|')]) for e in data[col]]\n",
        "  for col in ['state', 'n_state']:\n",
        "    data[col] = [np.array([e[0] for e in l]) for l in data[col]]\n",
        "\n",
        "  data['action'] = [[e[0] for e in l] for l in data['action_reward']]\n",
        "  data['reward'] = [tuple(e[1] for e in l) for l in data['action_reward']]\n",
        "  data.drop(columns=['action_reward'], inplace=True)\n",
        "\n",
        "  return data"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jG9pMXDf1sYS"
      },
      "source": [
        "### read_embeddings `helper function`\n",
        "\n",
        "This function will read the stored embedding csv file into pandas dataframe and return as multi-dimensional array"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jIa8hSX0uZWx"
      },
      "source": [
        "def read_embeddings(embeddings_path):\n",
        "  ''' Load embeddings (a vector for each item). '''\n",
        "  \n",
        "  embeddings = pd.read_csv(embeddings_path, sep=';')\n",
        "\n",
        "  return np.array([[np.float64(k) for k in e.split('|')]\n",
        "                   for e in embeddings['vectors']])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VrPAuw7W2QYx"
      },
      "source": [
        "### Environment `class`\n",
        "\n",
        "This is the simulator. It will help orchestrating the whole process of learning list recommendations by our actor-critic based MDP agent. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ImgTOp-kujq7"
      },
      "source": [
        "#collapse-hide\n",
        "class Environment():\n",
        "  def __init__(self, data, embeddings, alpha, gamma, fixed_length):\n",
        "    self.embeddings = embeddings\n",
        "\n",
        "    self.embedded_data = pd.DataFrame()\n",
        "    self.embedded_data['state'] = [np.array([embeddings.get_embedding(item_id) \n",
        "      for item_id in row['state']]) for _, row in data.iterrows()]\n",
        "    self.embedded_data['action'] = [np.array([embeddings.get_embedding(item_id) \n",
        "      for item_id in row['action']]) for _, row in data.iterrows()]\n",
        "    self.embedded_data['reward'] = data['reward']\n",
        "\n",
        "    self.alpha = alpha # α (alpha) in Equation (1)\n",
        "    self.gamma = gamma # Γ (Gamma) in Equation (4)\n",
        "    self.fixed_length = fixed_length\n",
        "    self.current_state = self.reset()\n",
        "    self.groups = self.get_groups()\n",
        "\n",
        "  def reset(self):\n",
        "    self.init_state = self.embedded_data['state'].sample(1).values[0]\n",
        "    return self.init_state\n",
        "\n",
        "  def step(self, actions):\n",
        "    '''\n",
        "    Compute reward and update state.\n",
        "    Args:\n",
        "      actions: embedded chosen items.\n",
        "    Returns:\n",
        "      cumulated_reward: overall reward.\n",
        "      current_state: updated state.\n",
        "    '''\n",
        "\n",
        "    # '18: Compute overall reward r_t according to Equation (4)'\n",
        "    simulated_rewards, cumulated_reward = self.simulate_rewards(self.current_state.reshape((1, -1)), actions.reshape((1, -1)))\n",
        "\n",
        "    # '11: Set s_t+1 = s_t' <=> self.current_state = self.current_state\n",
        "\n",
        "    for k in range(len(simulated_rewards)): # '12: for k = 1, K do'\n",
        "      if simulated_rewards[k] > 0: # '13: if r_t^k > 0 then'\n",
        "        # '14: Add a_t^k to the end of s_t+1'\n",
        "        self.current_state = np.append(self.current_state, [actions[k]], axis=0)\n",
        "        if self.fixed_length: # '15: Remove the first item of s_t+1'\n",
        "          self.current_state = np.delete(self.current_state, 0, axis=0)\n",
        "\n",
        "    return cumulated_reward, self.current_state\n",
        "\n",
        "  def get_groups(self):\n",
        "    ''' Calculate average state/action value for each group. Equation (3). '''\n",
        "\n",
        "    groups = []\n",
        "    for rewards, group in self.embedded_data.groupby(['reward']):\n",
        "      size = group.shape[0]\n",
        "      states = np.array(list(group['state'].values))\n",
        "      actions = np.array(list(group['action'].values))\n",
        "      groups.append({\n",
        "        'size': size, # N_x in article\n",
        "        'rewards': rewards, # U_x in article (combination of rewards)\n",
        "        'average state': (np.sum(states / np.linalg.norm(states, 2, axis=1)[:, np.newaxis], axis=0) / size).reshape((1, -1)), # s_x^-\n",
        "        'average action': (np.sum(actions / np.linalg.norm(actions, 2, axis=1)[:, np.newaxis], axis=0) / size).reshape((1, -1)) # a_x^-\n",
        "      })\n",
        "    return groups\n",
        "\n",
        "  def simulate_rewards(self, current_state, chosen_actions, reward_type='grouped cosine'):\n",
        "    '''\n",
        "    Calculate simulated rewards.\n",
        "    Args:\n",
        "      current_state: history, list of embedded items.\n",
        "      chosen_actions: embedded chosen items.\n",
        "      reward_type: from ['normal', 'grouped average', 'grouped cosine'].\n",
        "    Returns:\n",
        "      returned_rewards: most probable rewards.\n",
        "      cumulated_reward: probability weighted rewards.\n",
        "    '''\n",
        "\n",
        "    # Equation (1)\n",
        "    def cosine_state_action(s_t, a_t, s_i, a_i):\n",
        "      cosine_state = np.dot(s_t, s_i.T) / (np.linalg.norm(s_t, 2) * np.linalg.norm(s_i, 2))\n",
        "      cosine_action = np.dot(a_t, a_i.T) / (np.linalg.norm(a_t, 2) * np.linalg.norm(a_i, 2))\n",
        "      return (self.alpha * cosine_state + (1 - self.alpha) * cosine_action).reshape((1,))\n",
        "\n",
        "    if reward_type == 'normal':\n",
        "      # Calculate simulated reward in normal way: Equation (2)\n",
        "      probabilities = [cosine_state_action(current_state, chosen_actions, row['state'], row['action'])\n",
        "        for _, row in self.embedded_data.iterrows()]\n",
        "    elif reward_type == 'grouped average':\n",
        "      # Calculate simulated reward by grouped average: Equation (3)\n",
        "      probabilities = np.array([g['size'] for g in self.groups]) *\\\n",
        "        [(self.alpha * (np.dot(current_state, g['average state'].T) / np.linalg.norm(current_state, 2))\\\n",
        "        + (1 - self.alpha) * (np.dot(chosen_actions, g['average action'].T) / np.linalg.norm(chosen_actions, 2)))\n",
        "        for g in self.groups]\n",
        "    elif reward_type == 'grouped cosine':\n",
        "      # Calculate simulated reward by grouped cosine: Equations (1) and (3)\n",
        "      probabilities = [cosine_state_action(current_state, chosen_actions, g['average state'], g['average action'])\n",
        "        for g in self.groups]\n",
        "\n",
        "    # Normalize (sum to 1)\n",
        "    probabilities = np.array(probabilities) / sum(probabilities)\n",
        "\n",
        "    # Get most probable rewards\n",
        "    if reward_type == 'normal':\n",
        "      returned_rewards = self.embedded_data.iloc[np.argmax(probabilities)]['reward']\n",
        "    elif reward_type in ['grouped average', 'grouped cosine']:\n",
        "      returned_rewards = self.groups[np.argmax(probabilities)]['rewards']\n",
        "\n",
        "    # Equation (4)\n",
        "    def overall_reward(rewards, gamma):\n",
        "      return np.sum([gamma**k * reward for k, reward in enumerate(rewards)])\n",
        "\n",
        "    if reward_type in ['normal', 'grouped average']:\n",
        "      # Get cumulated reward: Equation (4)\n",
        "      cumulated_reward = overall_reward(returned_rewards, self.gamma)\n",
        "    elif reward_type == 'grouped cosine':\n",
        "      # Get probability weighted cumulated reward\n",
        "      cumulated_reward = np.sum([p * overall_reward(g['rewards'], self.gamma)\n",
        "        for p, g in zip(probabilities, self.groups)])\n",
        "\n",
        "    return returned_rewards, cumulated_reward"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sqaLWAyQ3Eep"
      },
      "source": [
        "### Actor `class`\n",
        "\n",
        "This is the policy approximator actor "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "o2oyBorPul4U"
      },
      "source": [
        "#collapse-hide\n",
        "class Actor():\n",
        "  ''' Policy function approximator. '''\n",
        "  \n",
        "  def __init__(self, sess, state_space_size, action_space_size, batch_size, ra_length, history_length, embedding_size, tau, learning_rate, scope='actor'):\n",
        "    self.sess = sess\n",
        "    self.state_space_size = state_space_size\n",
        "    self.action_space_size = action_space_size\n",
        "    self.batch_size = batch_size\n",
        "    self.ra_length = ra_length\n",
        "    self.history_length = history_length\n",
        "    self.embedding_size = embedding_size\n",
        "    self.tau = tau\n",
        "    self.learning_rate = learning_rate\n",
        "    self.scope = scope\n",
        "\n",
        "    with tf.variable_scope(self.scope):\n",
        "      # Build Actor network\n",
        "      self.action_weights, self.state, self.sequence_length = self._build_net('estimator_actor')\n",
        "      self.network_params = tf.trainable_variables()\n",
        "\n",
        "      # Build target Actor network\n",
        "      self.target_action_weights, self.target_state, self.target_sequence_length = self._build_net('target_actor')\n",
        "      self.target_network_params = tf.trainable_variables()[len(self.network_params):] # TODO: why sublist [len(x):]? Maybe because its equal to network_params + target_network_params\n",
        "\n",
        "      # Initialize target network weights with network weights (θ^π′ ← θ^π)\n",
        "      self.init_target_network_params = [self.target_network_params[i].assign(self.network_params[i])\n",
        "        for i in range(len(self.target_network_params))]\n",
        "        \n",
        "      # Update target network weights (θ^π′ ← τθ^π + (1 − τ)θ^π′)\n",
        "      self.update_target_network_params = [self.target_network_params[i].assign(\n",
        "        tf.multiply(self.tau, self.network_params[i]) +\n",
        "        tf.multiply(1 - self.tau, self.target_network_params[i]))\n",
        "        for i in range(len(self.target_network_params))]\n",
        "\n",
        "      # Gradient computation from Critic's action_gradients\n",
        "      self.action_gradients = tf.placeholder(tf.float32, [None, self.action_space_size])\n",
        "      gradients = tf.gradients(tf.reshape(self.action_weights, [self.batch_size, self.action_space_size], name='42222222222'),\n",
        "                               self.network_params,\n",
        "                               self.action_gradients)\n",
        "      params_gradients = list(map(lambda x: tf.div(x, self.batch_size * self.action_space_size), gradients))\n",
        "      \n",
        "      # Compute ∇_a.Q(s, a|θ^µ).∇_θ^π.f_θ^π(s)\n",
        "      self.optimizer = tf.train.AdamOptimizer(self.learning_rate).apply_gradients(\n",
        "          zip(params_gradients, self.network_params))\n",
        "\n",
        "  def _build_net(self, scope):\n",
        "    ''' Build the (target) Actor network. '''\n",
        "\n",
        "    def gather_last_output(data, seq_lens):\n",
        "      def cli_value(x, v):\n",
        "        y = tf.constant(v, shape=x.get_shape(), dtype=tf.int64)\n",
        "        x = tf.cast(x, tf.int64)\n",
        "        return tf.where(tf.greater(x, y), x, y)\n",
        "\n",
        "      batch_range = tf.range(tf.cast(tf.shape(data)[0], dtype=tf.int64), dtype=tf.int64)\n",
        "      tmp_end = tf.map_fn(lambda x: cli_value(x, 0), seq_lens - 1, dtype=tf.int64)\n",
        "      indices = tf.stack([batch_range, tmp_end], axis=1)\n",
        "      return tf.gather_nd(data, indices)\n",
        "\n",
        "    with tf.variable_scope(scope):\n",
        "      # Inputs: current state, sequence_length\n",
        "      # Outputs: action weights to compute the score Equation (6)\n",
        "      state = tf.placeholder(tf.float32, [None, self.state_space_size], 'state')\n",
        "      state_ = tf.reshape(state, [-1, self.history_length, self.embedding_size])\n",
        "      sequence_length = tf.placeholder(tf.int32, [None], 'sequence_length')\n",
        "      cell = tf.nn.rnn_cell.GRUCell(self.embedding_size,\n",
        "                                    activation=tf.nn.relu,\n",
        "                                    kernel_initializer=tf.initializers.random_normal(),\n",
        "                                    bias_initializer=tf.zeros_initializer())\n",
        "      outputs, _ = tf.nn.dynamic_rnn(cell, state_, dtype=tf.float32, sequence_length=sequence_length)\n",
        "      last_output = gather_last_output(outputs, sequence_length) # TODO: replace by h\n",
        "      x = tf.keras.layers.Dense(self.ra_length * self.embedding_size)(last_output)\n",
        "      action_weights = tf.reshape(x, [-1, self.ra_length, self.embedding_size])\n",
        "\n",
        "    return action_weights, state, sequence_length\n",
        "\n",
        "  def train(self, state, sequence_length, action_gradients):\n",
        "    '''  Compute ∇_a.Q(s, a|θ^µ).∇_θ^π.f_θ^π(s). '''\n",
        "    self.sess.run(self.optimizer,\n",
        "                  feed_dict={\n",
        "                      self.state: state,\n",
        "                      self.sequence_length: sequence_length,\n",
        "                      self.action_gradients: action_gradients})\n",
        "\n",
        "  def predict(self, state, sequence_length):\n",
        "    return self.sess.run(self.action_weights,\n",
        "                         feed_dict={\n",
        "                             self.state: state,\n",
        "                             self.sequence_length: sequence_length})\n",
        "\n",
        "  def predict_target(self, state, sequence_length):\n",
        "    return self.sess.run(self.target_action_weights,\n",
        "                         feed_dict={\n",
        "                             self.target_state: state,\n",
        "                             self.target_sequence_length: sequence_length})\n",
        "\n",
        "  def init_target_network(self):\n",
        "    self.sess.run(self.init_target_network_params)\n",
        "\n",
        "  def update_target_network(self):\n",
        "    self.sess.run(self.update_target_network_params)\n",
        "    \n",
        "  def get_recommendation_list(self, ra_length, noisy_state, embeddings, target=False):\n",
        "    '''\n",
        "    Algorithm 2\n",
        "    Args:\n",
        "      ra_length: length of the recommendation list.\n",
        "      noisy_state: current/remembered environment state with noise.\n",
        "      embeddings: Embeddings object.\n",
        "      target: boolean to use Actor's network or target network.\n",
        "    Returns:\n",
        "      Recommendation List: list of embedded items as future actions.\n",
        "    '''\n",
        "\n",
        "    def get_score(weights, embedding, batch_size):\n",
        "      '''\n",
        "      Equation (6)\n",
        "      Args:\n",
        "        weights: w_t^k shape=(embedding_size,).\n",
        "        embedding: e_i shape=(embedding_size,).\n",
        "      Returns:\n",
        "        score of the item i: score_i=w_t^k.e_i^T shape=(1,).\n",
        "      '''\n",
        "      ret = np.dot(weights, embedding.T)\n",
        "      return ret\n",
        "\n",
        "    batch_size = noisy_state.shape[0]\n",
        "\n",
        "    # '1: Generate w_t = {w_t^1, ..., w_t^K} according to Equation (5)'\n",
        "    method = self.predict_target if target else self.predict\n",
        "    weights = method(noisy_state, [ra_length] * batch_size)\n",
        "\n",
        "    # '3: Score items in I according to Equation (6)'\n",
        "    scores = np.array([[[get_score(weights[i][k], embedding, batch_size)\n",
        "      for embedding in embeddings.get_embedding_vector()]\n",
        "      for k in range(ra_length)]\n",
        "      for i in range(batch_size)])\n",
        "\n",
        "    # '8: return a_t'\n",
        "    return np.array([[embeddings.get_embedding(np.argmax(scores[i][k]))\n",
        "      for k in range(ra_length)]\n",
        "      for i in range(batch_size)])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7O8wgnzS3PIU"
      },
      "source": [
        "### Critic `class`\n",
        "\n",
        "This is the value approximator critic"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wjl-YYbOup2V"
      },
      "source": [
        "#collapse-hide\n",
        "class Critic():\n",
        "  ''' Value function approximator. '''\n",
        "  \n",
        "  def __init__(self, sess, state_space_size, action_space_size, history_length, embedding_size, tau, learning_rate, scope='critic'):\n",
        "    self.sess = sess\n",
        "    self.state_space_size = state_space_size\n",
        "    self.action_space_size = action_space_size\n",
        "    self.history_length = history_length\n",
        "    self.embedding_size = embedding_size\n",
        "    self.tau = tau\n",
        "    self.learning_rate = learning_rate\n",
        "    self.scope = scope\n",
        "\n",
        "    with tf.variable_scope(self.scope):\n",
        "      # Build Critic network\n",
        "      self.critic_Q_value, self.state, self.action, self.sequence_length = self._build_net('estimator_critic')\n",
        "      self.network_params = tf.get_collection(tf.GraphKeys.TRAINABLE_VARIABLES, scope='estimator_critic')\n",
        "\n",
        "      # Build target Critic network\n",
        "      self.target_Q_value, self.target_state, self.target_action, self.target_sequence_length = self._build_net('target_critic')\n",
        "      self.target_network_params = tf.get_collection(tf.GraphKeys.TRAINABLE_VARIABLES, scope='target_critic')\n",
        "\n",
        "      # Initialize target network weights with network weights (θ^µ′ ← θ^µ)\n",
        "      self.init_target_network_params = [self.target_network_params[i].assign(self.network_params[i])\n",
        "        for i in range(len(self.target_network_params))]\n",
        "\n",
        "      # Update target network weights (θ^µ′ ← τθ^µ + (1 − τ)θ^µ′)\n",
        "      self.update_target_network_params = [self.target_network_params[i].assign(\n",
        "        tf.multiply(self.tau, self.network_params[i]) +\n",
        "        tf.multiply(1 - self.tau, self.target_network_params[i]))\n",
        "        for i in range(len(self.target_network_params))]\n",
        "\n",
        "      # Minimize MSE between Critic's and target Critic's outputed Q-values\n",
        "      self.expected_reward = tf.placeholder(tf.float32, [None, 1])\n",
        "      self.loss = tf.reduce_mean(tf.squared_difference(self.expected_reward, self.critic_Q_value))\n",
        "      self.optimizer = tf.train.AdamOptimizer(self.learning_rate).minimize(self.loss)\n",
        "\n",
        "      # Compute ∇_a.Q(s, a|θ^µ)\n",
        "      self.action_gradients = tf.gradients(self.critic_Q_value, self.action)\n",
        "\n",
        "  def _build_net(self, scope):\n",
        "    ''' Build the (target) Critic network. '''\n",
        "\n",
        "    def gather_last_output(data, seq_lens):\n",
        "      def cli_value(x, v):\n",
        "        y = tf.constant(v, shape=x.get_shape(), dtype=tf.int64)\n",
        "        return tf.where(tf.greater(x, y), x, y)\n",
        "\n",
        "      this_range = tf.range(tf.cast(tf.shape(seq_lens)[0], dtype=tf.int64), dtype=tf.int64)\n",
        "      tmp_end = tf.map_fn(lambda x: cli_value(x, 0), seq_lens - 1, dtype=tf.int64)\n",
        "      indices = tf.stack([this_range, tmp_end], axis=1)\n",
        "      return tf.gather_nd(data, indices)\n",
        "\n",
        "    with tf.variable_scope(scope):\n",
        "      # Inputs: current state, current action\n",
        "      # Outputs: predicted Q-value\n",
        "      state = tf.placeholder(tf.float32, [None, self.state_space_size], 'state')\n",
        "      state_ = tf.reshape(state, [-1, self.history_length, self.embedding_size])\n",
        "      action = tf.placeholder(tf.float32, [None, self.action_space_size], 'action')\n",
        "      sequence_length = tf.placeholder(tf.int64, [None], name='critic_sequence_length')\n",
        "      cell = tf.nn.rnn_cell.GRUCell(self.history_length,\n",
        "                                    activation=tf.nn.relu,\n",
        "                                    kernel_initializer=tf.initializers.random_normal(),\n",
        "                                    bias_initializer=tf.zeros_initializer())\n",
        "      predicted_state, _ = tf.nn.dynamic_rnn(cell, state_, dtype=tf.float32, sequence_length=sequence_length)\n",
        "      predicted_state = gather_last_output(predicted_state, sequence_length)\n",
        "\n",
        "      inputs = tf.concat([predicted_state, action], axis=-1)\n",
        "      layer1 = tf.layers.Dense(32, activation=tf.nn.relu)(inputs)\n",
        "      layer2 = tf.layers.Dense(16, activation=tf.nn.relu)(layer1)\n",
        "      critic_Q_value = tf.layers.Dense(1)(layer2)\n",
        "      return critic_Q_value, state, action, sequence_length\n",
        "\n",
        "  def train(self, state, action, sequence_length, expected_reward):\n",
        "    ''' Minimize MSE between expected reward and target Critic's Q-value. '''\n",
        "    return self.sess.run([self.critic_Q_value, self.loss, self.optimizer],\n",
        "                         feed_dict={\n",
        "                             self.state: state,\n",
        "                             self.action: action,\n",
        "                             self.sequence_length: sequence_length,\n",
        "                             self.expected_reward: expected_reward})\n",
        "\n",
        "  def predict(self, state, action, sequence_length):\n",
        "    ''' Returns Critic's predicted Q-value. '''\n",
        "    return self.sess.run(self.critic_Q_value,\n",
        "                         feed_dict={\n",
        "                             self.state: state,\n",
        "                             self.action: action,\n",
        "                             self.sequence_length: sequence_length})\n",
        "\n",
        "  def predict_target(self, state, action, sequence_length):\n",
        "    ''' Returns target Critic's predicted Q-value. '''\n",
        "    return self.sess.run(self.target_Q_value,\n",
        "                         feed_dict={\n",
        "                             self.target_state: state,\n",
        "                             self.target_action: action,\n",
        "                             self.target_sequence_length: sequence_length})\n",
        "\n",
        "  def get_action_gradients(self, state, action, sequence_length):\n",
        "    ''' Returns ∇_a.Q(s, a|θ^µ). '''\n",
        "    return np.array(self.sess.run(self.action_gradients,\n",
        "                         feed_dict={\n",
        "                             self.state: state,\n",
        "                             self.action: action,\n",
        "                             self.sequence_length: sequence_length})[0])\n",
        "\n",
        "  def init_target_network(self):\n",
        "    self.sess.run(self.init_target_network_params)\n",
        "\n",
        "  def update_target_network(self):\n",
        "    self.sess.run(self.update_target_network_params)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EpWire7m3cnB"
      },
      "source": [
        "### ReplayMemory `class`"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "N9b_is_Su_C-"
      },
      "source": [
        "#collapse-hide\n",
        "class ReplayMemory():\n",
        "  ''' Replay memory D. '''\n",
        "  \n",
        "  def __init__(self, buffer_size):\n",
        "    self.buffer_size = buffer_size\n",
        "    # self.buffer = [[row['state'], row['action'], row['reward'], row['n_state']] for _, row in data.iterrows()][-self.buffer_size:] TODO: empty or not?\n",
        "    self.buffer = []\n",
        "\n",
        "  def add(self, state, action, reward, n_state):\n",
        "    self.buffer.append([state, action, reward, n_state])\n",
        "    if len(self.buffer) > self.buffer_size:\n",
        "      self.buffer.pop(0)\n",
        "\n",
        "  def size(self):\n",
        "    return len(self.buffer)\n",
        "\n",
        "  def sample_batch(self, batch_size):\n",
        "    return random.sample(self.buffer, batch_size)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5cOXOxdV3lde"
      },
      "source": [
        "### experience_replay `function`"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9foofY8zvHYE"
      },
      "source": [
        "#collapse-hide\n",
        "def experience_replay(replay_memory, batch_size, actor, critic, embeddings, ra_length, state_space_size, action_space_size, discount_factor):\n",
        "  '''\n",
        "  Experience replay.\n",
        "  Args:\n",
        "    replay_memory: replay memory D in article.\n",
        "    batch_size: sample size.\n",
        "    actor: Actor network.\n",
        "    critic: Critic network.\n",
        "    embeddings: Embeddings object.\n",
        "    state_space_size: dimension of states.\n",
        "    action_space_size: dimensions of actions.\n",
        "  Returns:\n",
        "    Best Q-value, loss of Critic network for printing/recording purpose.\n",
        "  '''\n",
        "\n",
        "  # '22: Sample minibatch of N transitions (s, a, r, s′) from D'\n",
        "  samples = replay_memory.sample_batch(batch_size)\n",
        "  states = np.array([s[0] for s in samples])\n",
        "  actions = np.array([s[1] for s in samples])\n",
        "  rewards = np.array([s[2] for s in samples])\n",
        "  n_states = np.array([s[3] for s in samples]).reshape(-1, state_space_size)\n",
        "\n",
        "  # '23: Generate a′ by target Actor network according to Algorithm 2'\n",
        "  n_actions = actor.get_recommendation_list(ra_length, states, embeddings, target=True).reshape(-1, action_space_size)\n",
        "\n",
        "  # Calculate predicted Q′(s′, a′|θ^µ′) value\n",
        "  target_Q_value = critic.predict_target(n_states, n_actions, [ra_length] * batch_size)\n",
        "\n",
        "  # '24: Set y = r + γQ′(s′, a′|θ^µ′)'\n",
        "  expected_rewards = rewards + discount_factor * target_Q_value\n",
        "  \n",
        "  # '25: Update Critic by minimizing (y − Q(s, a|θ^µ))²'\n",
        "  critic_Q_value, critic_loss, _ = critic.train(states, actions, [ra_length] * batch_size, expected_rewards)\n",
        "  \n",
        "  # '26: Update the Actor using the sampled policy gradient'\n",
        "  action_gradients = critic.get_action_gradients(states, n_actions, [ra_length] * batch_size)\n",
        "  actor.train(states, [ra_length] * batch_size, action_gradients)\n",
        "\n",
        "  # '27: Update the Critic target networks'\n",
        "  critic.update_target_network()\n",
        "\n",
        "  # '28: Update the Actor target network'\n",
        "  actor.update_target_network()\n",
        "\n",
        "  return np.amax(critic_Q_value), critic_loss"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8uRr4bJg36El"
      },
      "source": [
        "### OrnsteinUhlenbeckNoise `class`"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-VYMfzSJvNHI"
      },
      "source": [
        "#collapse-hide\n",
        "class OrnsteinUhlenbeckNoise:\n",
        "  ''' Noise for Actor predictions. '''\n",
        "  def __init__(self, action_space_size, mu=0, theta=0.5, sigma=0.2):\n",
        "    self.action_space_size = action_space_size\n",
        "    self.mu = mu\n",
        "    self.theta = theta\n",
        "    self.sigma = sigma\n",
        "    self.state = np.ones(self.action_space_size) * self.mu\n",
        "\n",
        "  def get(self):\n",
        "    self.state += self.theta * (self.mu - self.state) + self.sigma * np.random.rand(self.action_space_size)\n",
        "    return self.state\n",
        "\n",
        "def train(sess, environment, actor, critic, embeddings, history_length, ra_length, buffer_size, batch_size, discount_factor, nb_episodes, filename_summary):\n",
        "  ''' Algorithm 3 in article. '''\n",
        "\n",
        "  # Set up summary operators\n",
        "  def build_summaries():\n",
        "    episode_reward = tf.Variable(0.)\n",
        "    tf.summary.scalar('reward', episode_reward)\n",
        "    episode_max_Q = tf.Variable(0.)\n",
        "    tf.summary.scalar('max_Q_value', episode_max_Q)\n",
        "    critic_loss = tf.Variable(0.)\n",
        "    tf.summary.scalar('critic_loss', critic_loss)\n",
        "\n",
        "    summary_vars = [episode_reward, episode_max_Q, critic_loss]\n",
        "    summary_ops = tf.summary.merge_all()\n",
        "    return summary_ops, summary_vars\n",
        "\n",
        "  summary_ops, summary_vars = build_summaries()\n",
        "  sess.run(tf.global_variables_initializer())\n",
        "  writer = tf.summary.FileWriter(filename_summary, sess.graph)\n",
        "\n",
        "  # '2: Initialize target network f′ and Q′'\n",
        "  actor.init_target_network()\n",
        "  critic.init_target_network()\n",
        "\n",
        "  # '3: Initialize the capacity of replay memory D'\n",
        "  replay_memory = ReplayMemory(buffer_size) # Memory D in article\n",
        "  replay = False\n",
        "\n",
        "\n",
        "  start_time = time.time()\n",
        "  for i_session in range(nb_episodes): # '4: for session = 1, M do'\n",
        "    session_reward = 0\n",
        "    session_Q_value = 0\n",
        "    session_critic_loss = 0\n",
        "\n",
        "    # '5: Reset the item space I' is useless because unchanged.\n",
        "\n",
        "    states = environment.reset() # '6: Initialize state s_0 from previous sessions'\n",
        "    \n",
        "    if (i_session + 1) % 10 == 0: # Update average parameters every 10 episodes\n",
        "      environment.groups = environment.get_groups()\n",
        "      \n",
        "    exploration_noise = OrnsteinUhlenbeckNoise(history_length * embeddings.size())\n",
        "\n",
        "    for t in range(nb_rounds): # '7: for t = 1, T do'\n",
        "      # '8: Stage 1: Transition Generating Stage'\n",
        "\n",
        "      # '9: Select an action a_t = {a_t^1, ..., a_t^K} according to Algorithm 2'\n",
        "      actions = actor.get_recommendation_list(\n",
        "          ra_length,\n",
        "          states.reshape(1, -1), # TODO + exploration_noise.get().reshape(1, -1),\n",
        "          embeddings).reshape(ra_length, embeddings.size())\n",
        "\n",
        "      # '10: Execute action a_t and observe the reward list {r_t^1, ..., r_t^K} for each item in a_t'\n",
        "      rewards, next_states = environment.step(actions)\n",
        "\n",
        "      # '19: Store transition (s_t, a_t, r_t, s_t+1) in D'\n",
        "      replay_memory.add(states.reshape(history_length * embeddings.size()),\n",
        "                        actions.reshape(ra_length * embeddings.size()),\n",
        "                        [rewards],\n",
        "                        next_states.reshape(history_length * embeddings.size()))\n",
        "\n",
        "      states = next_states # '20: Set s_t = s_t+1'\n",
        "\n",
        "      session_reward += rewards\n",
        "      \n",
        "      # '21: Stage 2: Parameter Updating Stage'\n",
        "      if replay_memory.size() >= batch_size: # Experience replay\n",
        "        replay = True\n",
        "        replay_Q_value, critic_loss = experience_replay(replay_memory, batch_size,\n",
        "          actor, critic, embeddings, ra_length, history_length * embeddings.size(),\n",
        "          ra_length * embeddings.size(), discount_factor)\n",
        "        session_Q_value += replay_Q_value\n",
        "        session_critic_loss += critic_loss\n",
        "\n",
        "      summary_str = sess.run(summary_ops,\n",
        "                             feed_dict={summary_vars[0]: session_reward,\n",
        "                                        summary_vars[1]: session_Q_value,\n",
        "                                        summary_vars[2]: session_critic_loss})\n",
        "      \n",
        "      writer.add_summary(summary_str, i_session)\n",
        "\n",
        "      '''\n",
        "      print(state_to_items(embeddings.embed(data['state'][0]), actor, ra_length, embeddings),\n",
        "            state_to_items(embeddings.embed(data['state'][0]), actor, ra_length, embeddings, True))\n",
        "      '''\n",
        "\n",
        "    str_loss = str('Loss=%0.4f' % session_critic_loss)\n",
        "    print(('Episode %d/%d Reward=%d Time=%ds ' + (str_loss if replay else 'No replay')) % (i_session + 1, nb_episodes, session_reward, time.time() - start_time))\n",
        "    start_time = time.time()\n",
        "\n",
        "  writer.close()\n",
        "  tf.train.Saver().save(sess, 'models.h5', write_meta_graph=False)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "J8v1uNNs3-eU"
      },
      "source": [
        "### Hyperparameters"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2YWoY788vTvE"
      },
      "source": [
        "# Hyperparameters\n",
        "history_length = 12 # N in article\n",
        "ra_length = 4 # K in article\n",
        "discount_factor = 0.99 # Gamma in Bellman equation\n",
        "actor_lr = 0.0001\n",
        "critic_lr = 0.001\n",
        "tau = 0.001 # τ in Algorithm 3\n",
        "batch_size = 64\n",
        "nb_episodes = 100\n",
        "nb_rounds = 50\n",
        "filename_summary = 'summary.txt'\n",
        "alpha = 0.5 # α (alpha) in Equation (1)\n",
        "gamma = 0.9 # Γ (Gamma) in Equation (4)\n",
        "buffer_size = 1000000 # Size of replay memory D in article\n",
        "fixed_length = True # Fixed memory length"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UfjpsYgE4Ait"
      },
      "source": [
        "### Data generation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZboYOTAmyPBg"
      },
      "source": [
        "dg = DataGenerator('ml-100k/u.data', 'ml-100k/u.item')\n",
        "dg.gen_train_test(0.8, seed=42)\n",
        "\n",
        "dg.write_csv('train.csv', dg.train, nb_states=[history_length], nb_actions=[ra_length])\n",
        "dg.write_csv('test.csv', dg.test, nb_states=[history_length], nb_actions=[ra_length])\n",
        "\n",
        "data = read_file('train.csv')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "id": "vLfbBCivyTCf",
        "outputId": "57164d2a-3d21-44f1-e419-106619690c4c"
      },
      "source": [
        "data.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>state</th>\n",
              "      <th>n_state</th>\n",
              "      <th>action</th>\n",
              "      <th>reward</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>[732, 257, 507, 602, 481, 568, 1286, 50, 501, ...</td>\n",
              "      <td>[732, 257, 507, 602, 481, 568, 1286, 50, 501, ...</td>\n",
              "      <td>[731, 525, 80, 88]</td>\n",
              "      <td>(3, 4, 3, 3)</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>[1226, 855, 339, 124, 16, 147, 59, 827, 323, 2...</td>\n",
              "      <td>[1226, 855, 339, 124, 16, 147, 59, 827, 323, 2...</td>\n",
              "      <td>[52, 1005, 347, 70]</td>\n",
              "      <td>(4, 5, 4, 3)</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>[316, 286, 313, 748, 258, 272, 300, 302, 347, ...</td>\n",
              "      <td>[316, 286, 313, 748, 258, 272, 300, 302, 347, ...</td>\n",
              "      <td>[751, 271, 689, 289]</td>\n",
              "      <td>(4, 4, 4, 5)</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>[235, 433, 96, 117, 429, 7, 471, 201, 276, 55,...</td>\n",
              "      <td>[235, 433, 96, 117, 429, 7, 471, 201, 276, 55,...</td>\n",
              "      <td>[31, 198, 724, 654]</td>\n",
              "      <td>(3, 5, 3, 4)</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>[77, 241, 98, 423, 71, 157, 955, 186, 121, 421...</td>\n",
              "      <td>[77, 241, 98, 423, 71, 157, 955, 186, 121, 421...</td>\n",
              "      <td>[316, 427, 313, 959]</td>\n",
              "      <td>(4, 5, 4, 5)</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                               state  ...        reward\n",
              "0  [732, 257, 507, 602, 481, 568, 1286, 50, 501, ...  ...  (3, 4, 3, 3)\n",
              "1  [1226, 855, 339, 124, 16, 147, 59, 827, 323, 2...  ...  (4, 5, 4, 3)\n",
              "2  [316, 286, 313, 748, 258, 272, 300, 302, 347, ...  ...  (4, 4, 4, 5)\n",
              "3  [235, 433, 96, 117, 429, 7, 471, 201, 276, 55,...  ...  (3, 5, 3, 4)\n",
              "4  [77, 241, 98, 423, 71, 157, 955, 186, 121, 421...  ...  (4, 5, 4, 5)\n",
              "\n",
              "[5 rows x 4 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IVEyLl844DdK"
      },
      "source": [
        "### Embedding generation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "r8DHUEutvXw-",
        "outputId": "6b8d810e-2b7e-4d27-c503-61a513f29ac4"
      },
      "source": [
        "#collapse-output\n",
        "if True: # Generate embeddings?\n",
        "  eg = EmbeddingsGenerator(dg.user_train, pd.read_csv('ml-100k/u.data', sep='\\t', names=['userId', 'itemId', 'rating', 'timestamp']))\n",
        "  eg.train(nb_epochs=300)\n",
        "  train_loss, train_accuracy = eg.test(dg.user_train)\n",
        "  print('Train set: Loss=%.4f ; Accuracy=%.1f%%' % (train_loss, train_accuracy * 100))\n",
        "  test_loss, test_accuracy = eg.test(dg.user_test)\n",
        "  print('Test set: Loss=%.4f ; Accuracy=%.1f%%' % (test_loss, test_accuracy * 100))\n",
        "  eg.save_embeddings('embeddings.csv')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /tensorflow-1.15.2/python3.7/tensorflow_core/python/ops/resource_variable_ops.py:1630: calling BaseResourceVariable.__init__ (from tensorflow.python.ops.resource_variable_ops) with constraint is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "If using Keras pass *_constraint arguments to layers.\n",
            "1/300\n",
            "WARNING:tensorflow:From /tensorflow-1.15.2/python3.7/tensorflow_core/python/ops/math_grad.py:1424: where (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Use tf.where in 2.0, which has the same broadcast rule as np.where\n",
            "WARNING:tensorflow:From /tensorflow-1.15.2/python3.7/keras/backend/tensorflow_backend.py:422: The name tf.global_variables is deprecated. Please use tf.compat.v1.global_variables instead.\n",
            "\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 493us/step - loss: 6.9202 - accuracy: 0.0100 - val_loss: 6.5489 - val_accuracy: 0.0160\n",
            "2/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 392us/step - loss: 6.4452 - accuracy: 0.0150 - val_loss: 6.3391 - val_accuracy: 0.0144\n",
            "3/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 397us/step - loss: 6.2737 - accuracy: 0.0172 - val_loss: 6.2418 - val_accuracy: 0.0142\n",
            "4/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 395us/step - loss: 6.2539 - accuracy: 0.0156 - val_loss: 6.1208 - val_accuracy: 0.0152\n",
            "5/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 387us/step - loss: 6.1386 - accuracy: 0.0196 - val_loss: 6.1390 - val_accuracy: 0.0230\n",
            "6/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 399us/step - loss: 6.0841 - accuracy: 0.0170 - val_loss: 6.0209 - val_accuracy: 0.0210\n",
            "7/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 396us/step - loss: 6.0844 - accuracy: 0.0204 - val_loss: 5.9844 - val_accuracy: 0.0240\n",
            "8/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 397us/step - loss: 6.0329 - accuracy: 0.0222 - val_loss: 5.9625 - val_accuracy: 0.0244\n",
            "9/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 396us/step - loss: 6.0147 - accuracy: 0.0204 - val_loss: 5.9273 - val_accuracy: 0.0246\n",
            "10/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 387us/step - loss: 5.9878 - accuracy: 0.0250 - val_loss: 5.9182 - val_accuracy: 0.0264\n",
            "11/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 388us/step - loss: 5.9183 - accuracy: 0.0254 - val_loss: 5.8513 - val_accuracy: 0.0324\n",
            "12/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 390us/step - loss: 5.9244 - accuracy: 0.0266 - val_loss: 5.8591 - val_accuracy: 0.0334\n",
            "13/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 386us/step - loss: 5.8851 - accuracy: 0.0312 - val_loss: 5.8540 - val_accuracy: 0.0326\n",
            "14/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 392us/step - loss: 5.8685 - accuracy: 0.0316 - val_loss: 5.8123 - val_accuracy: 0.0378\n",
            "15/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 399us/step - loss: 5.8641 - accuracy: 0.0334 - val_loss: 5.8084 - val_accuracy: 0.0338\n",
            "16/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 400us/step - loss: 5.8641 - accuracy: 0.0322 - val_loss: 5.7705 - val_accuracy: 0.0390\n",
            "17/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 399us/step - loss: 5.8240 - accuracy: 0.0434 - val_loss: 5.7161 - val_accuracy: 0.0510\n",
            "18/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 387us/step - loss: 5.7978 - accuracy: 0.0442 - val_loss: 5.7052 - val_accuracy: 0.0492\n",
            "19/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 384us/step - loss: 5.7698 - accuracy: 0.0396 - val_loss: 5.7139 - val_accuracy: 0.0484\n",
            "20/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 392us/step - loss: 5.7513 - accuracy: 0.0440 - val_loss: 5.6890 - val_accuracy: 0.0524\n",
            "21/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 401us/step - loss: 5.7090 - accuracy: 0.0488 - val_loss: 5.6584 - val_accuracy: 0.0514\n",
            "22/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 401us/step - loss: 5.7031 - accuracy: 0.0432 - val_loss: 5.5928 - val_accuracy: 0.0622\n",
            "23/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 396us/step - loss: 5.6509 - accuracy: 0.0554 - val_loss: 5.5568 - val_accuracy: 0.0670\n",
            "24/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 381us/step - loss: 5.6746 - accuracy: 0.0600 - val_loss: 5.6020 - val_accuracy: 0.0604\n",
            "25/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 390us/step - loss: 5.5880 - accuracy: 0.0618 - val_loss: 5.5072 - val_accuracy: 0.0756\n",
            "26/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 406us/step - loss: 5.6083 - accuracy: 0.0634 - val_loss: 5.5330 - val_accuracy: 0.0712\n",
            "27/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 391us/step - loss: 5.5992 - accuracy: 0.0658 - val_loss: 5.5303 - val_accuracy: 0.0782\n",
            "28/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 390us/step - loss: 5.5620 - accuracy: 0.0730 - val_loss: 5.4303 - val_accuracy: 0.0820\n",
            "29/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 386us/step - loss: 5.5142 - accuracy: 0.0720 - val_loss: 5.3807 - val_accuracy: 0.0934\n",
            "30/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 387us/step - loss: 5.5019 - accuracy: 0.0752 - val_loss: 5.3523 - val_accuracy: 0.0960\n",
            "31/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 391us/step - loss: 5.4424 - accuracy: 0.0876 - val_loss: 5.3459 - val_accuracy: 0.0970\n",
            "32/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 391us/step - loss: 5.4085 - accuracy: 0.0878 - val_loss: 5.3406 - val_accuracy: 0.1014\n",
            "33/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 398us/step - loss: 5.3886 - accuracy: 0.0888 - val_loss: 5.3043 - val_accuracy: 0.0976\n",
            "34/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 393us/step - loss: 5.3653 - accuracy: 0.0924 - val_loss: 5.2621 - val_accuracy: 0.1140\n",
            "35/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 388us/step - loss: 5.4007 - accuracy: 0.0890 - val_loss: 5.2699 - val_accuracy: 0.1140\n",
            "36/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 404us/step - loss: 5.2848 - accuracy: 0.1040 - val_loss: 5.2311 - val_accuracy: 0.1204\n",
            "37/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 395us/step - loss: 5.3042 - accuracy: 0.1008 - val_loss: 5.1873 - val_accuracy: 0.1230\n",
            "38/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 396us/step - loss: 5.2921 - accuracy: 0.1066 - val_loss: 5.1309 - val_accuracy: 0.1314\n",
            "39/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 398us/step - loss: 5.2618 - accuracy: 0.1070 - val_loss: 5.0953 - val_accuracy: 0.1448\n",
            "40/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 390us/step - loss: 5.1676 - accuracy: 0.1136 - val_loss: 5.0245 - val_accuracy: 0.1492\n",
            "41/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 389us/step - loss: 5.1354 - accuracy: 0.1242 - val_loss: 5.0307 - val_accuracy: 0.1498\n",
            "42/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 402us/step - loss: 5.1219 - accuracy: 0.1276 - val_loss: 4.9901 - val_accuracy: 0.1650\n",
            "43/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 400us/step - loss: 5.1534 - accuracy: 0.1336 - val_loss: 4.9714 - val_accuracy: 0.1694\n",
            "44/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 399us/step - loss: 5.1184 - accuracy: 0.1376 - val_loss: 4.9489 - val_accuracy: 0.1592\n",
            "45/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 401us/step - loss: 5.0298 - accuracy: 0.1376 - val_loss: 4.9274 - val_accuracy: 0.1720\n",
            "46/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 395us/step - loss: 5.0099 - accuracy: 0.1504 - val_loss: 4.8445 - val_accuracy: 0.1786\n",
            "47/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 395us/step - loss: 4.9666 - accuracy: 0.1526 - val_loss: 4.7906 - val_accuracy: 0.1948\n",
            "48/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 399us/step - loss: 4.9227 - accuracy: 0.1608 - val_loss: 4.7955 - val_accuracy: 0.1882\n",
            "49/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 392us/step - loss: 4.9143 - accuracy: 0.1596 - val_loss: 4.7671 - val_accuracy: 0.1988\n",
            "50/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 396us/step - loss: 4.8729 - accuracy: 0.1662 - val_loss: 4.7239 - val_accuracy: 0.2054\n",
            "51/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 394us/step - loss: 4.7912 - accuracy: 0.1836 - val_loss: 4.6844 - val_accuracy: 0.2136\n",
            "52/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 393us/step - loss: 4.8110 - accuracy: 0.1784 - val_loss: 4.6213 - val_accuracy: 0.2212\n",
            "53/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 408us/step - loss: 4.6814 - accuracy: 0.1954 - val_loss: 4.5962 - val_accuracy: 0.2324\n",
            "54/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 394us/step - loss: 4.7148 - accuracy: 0.1876 - val_loss: 4.5031 - val_accuracy: 0.2446\n",
            "55/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 395us/step - loss: 4.6427 - accuracy: 0.2044 - val_loss: 4.5063 - val_accuracy: 0.2494\n",
            "56/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 394us/step - loss: 4.6296 - accuracy: 0.2088 - val_loss: 4.4856 - val_accuracy: 0.2628\n",
            "57/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 389us/step - loss: 4.5943 - accuracy: 0.2136 - val_loss: 4.4382 - val_accuracy: 0.2612\n",
            "58/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 394us/step - loss: 4.5594 - accuracy: 0.2218 - val_loss: 4.3101 - val_accuracy: 0.2852\n",
            "59/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 395us/step - loss: 4.5126 - accuracy: 0.2246 - val_loss: 4.3327 - val_accuracy: 0.2772\n",
            "60/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 397us/step - loss: 4.4381 - accuracy: 0.2378 - val_loss: 4.2424 - val_accuracy: 0.2890\n",
            "61/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 388us/step - loss: 4.4603 - accuracy: 0.2266 - val_loss: 4.2749 - val_accuracy: 0.2970\n",
            "62/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 392us/step - loss: 4.4146 - accuracy: 0.2394 - val_loss: 4.1974 - val_accuracy: 0.3112\n",
            "63/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 405us/step - loss: 4.3489 - accuracy: 0.2488 - val_loss: 4.1782 - val_accuracy: 0.3094\n",
            "64/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 395us/step - loss: 4.3680 - accuracy: 0.2514 - val_loss: 4.1138 - val_accuracy: 0.3308\n",
            "65/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 395us/step - loss: 4.3094 - accuracy: 0.2558 - val_loss: 4.0360 - val_accuracy: 0.3318\n",
            "66/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 392us/step - loss: 4.2179 - accuracy: 0.2738 - val_loss: 4.0146 - val_accuracy: 0.3466\n",
            "67/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 400us/step - loss: 4.1800 - accuracy: 0.2802 - val_loss: 3.9621 - val_accuracy: 0.3408\n",
            "68/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 399us/step - loss: 4.1356 - accuracy: 0.2862 - val_loss: 3.9211 - val_accuracy: 0.3626\n",
            "69/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 390us/step - loss: 4.0549 - accuracy: 0.3104 - val_loss: 3.8791 - val_accuracy: 0.3770\n",
            "70/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 401us/step - loss: 4.0348 - accuracy: 0.3038 - val_loss: 3.8400 - val_accuracy: 0.3868\n",
            "71/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 397us/step - loss: 3.9944 - accuracy: 0.3064 - val_loss: 3.7690 - val_accuracy: 0.3856\n",
            "72/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 414us/step - loss: 4.0106 - accuracy: 0.3132 - val_loss: 3.7704 - val_accuracy: 0.3956\n",
            "73/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 404us/step - loss: 3.9379 - accuracy: 0.3204 - val_loss: 3.6701 - val_accuracy: 0.3974\n",
            "74/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 403us/step - loss: 3.8949 - accuracy: 0.3356 - val_loss: 3.6144 - val_accuracy: 0.4296\n",
            "75/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 390us/step - loss: 3.8187 - accuracy: 0.3368 - val_loss: 3.5836 - val_accuracy: 0.4128\n",
            "76/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 405us/step - loss: 3.8102 - accuracy: 0.3428 - val_loss: 3.5028 - val_accuracy: 0.4346\n",
            "77/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 404us/step - loss: 3.7603 - accuracy: 0.3534 - val_loss: 3.4695 - val_accuracy: 0.4408\n",
            "78/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 402us/step - loss: 3.7149 - accuracy: 0.3650 - val_loss: 3.4650 - val_accuracy: 0.4510\n",
            "79/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 392us/step - loss: 3.6468 - accuracy: 0.3808 - val_loss: 3.4893 - val_accuracy: 0.4436\n",
            "80/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 414us/step - loss: 3.6445 - accuracy: 0.3664 - val_loss: 3.3478 - val_accuracy: 0.4686\n",
            "81/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 398us/step - loss: 3.5365 - accuracy: 0.3988 - val_loss: 3.3008 - val_accuracy: 0.4854\n",
            "82/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 402us/step - loss: 3.5254 - accuracy: 0.3944 - val_loss: 3.3290 - val_accuracy: 0.4742\n",
            "83/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 414us/step - loss: 3.4863 - accuracy: 0.4056 - val_loss: 3.3257 - val_accuracy: 0.4792\n",
            "84/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 398us/step - loss: 3.4112 - accuracy: 0.4164 - val_loss: 3.1776 - val_accuracy: 0.5012\n",
            "85/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 402us/step - loss: 3.4042 - accuracy: 0.4216 - val_loss: 3.1592 - val_accuracy: 0.5088\n",
            "86/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 402us/step - loss: 3.2852 - accuracy: 0.4386 - val_loss: 3.1144 - val_accuracy: 0.5164\n",
            "87/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 403us/step - loss: 3.2709 - accuracy: 0.4408 - val_loss: 3.0742 - val_accuracy: 0.5240\n",
            "88/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 403us/step - loss: 3.2722 - accuracy: 0.4414 - val_loss: 3.0320 - val_accuracy: 0.5292\n",
            "89/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 402us/step - loss: 3.1894 - accuracy: 0.4532 - val_loss: 2.9413 - val_accuracy: 0.5480\n",
            "90/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 404us/step - loss: 3.1212 - accuracy: 0.4698 - val_loss: 2.8748 - val_accuracy: 0.5718\n",
            "91/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 394us/step - loss: 3.0990 - accuracy: 0.4772 - val_loss: 2.9096 - val_accuracy: 0.5594\n",
            "92/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 397us/step - loss: 2.9949 - accuracy: 0.4906 - val_loss: 2.7876 - val_accuracy: 0.5796\n",
            "93/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 405us/step - loss: 2.9939 - accuracy: 0.4926 - val_loss: 2.7424 - val_accuracy: 0.5816\n",
            "94/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 405us/step - loss: 2.9415 - accuracy: 0.4980 - val_loss: 2.6546 - val_accuracy: 0.5992\n",
            "95/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 397us/step - loss: 2.9397 - accuracy: 0.5084 - val_loss: 2.6482 - val_accuracy: 0.5952\n",
            "96/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 387us/step - loss: 2.8384 - accuracy: 0.5260 - val_loss: 2.6494 - val_accuracy: 0.6058\n",
            "97/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 396us/step - loss: 2.8045 - accuracy: 0.5276 - val_loss: 2.6678 - val_accuracy: 0.6152\n",
            "98/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 406us/step - loss: 2.7874 - accuracy: 0.5284 - val_loss: 2.5639 - val_accuracy: 0.6278\n",
            "99/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 408us/step - loss: 2.7609 - accuracy: 0.5396 - val_loss: 2.5057 - val_accuracy: 0.6398\n",
            "100/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 404us/step - loss: 2.6811 - accuracy: 0.5578 - val_loss: 2.4427 - val_accuracy: 0.6516\n",
            "101/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 410us/step - loss: 2.6871 - accuracy: 0.5456 - val_loss: 2.4066 - val_accuracy: 0.6654\n",
            "102/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 408us/step - loss: 2.5663 - accuracy: 0.5738 - val_loss: 2.3830 - val_accuracy: 0.6528\n",
            "103/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 393us/step - loss: 2.5405 - accuracy: 0.5778 - val_loss: 2.2917 - val_accuracy: 0.6684\n",
            "104/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 398us/step - loss: 2.5373 - accuracy: 0.5870 - val_loss: 2.2606 - val_accuracy: 0.6852\n",
            "105/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 402us/step - loss: 2.4934 - accuracy: 0.5946 - val_loss: 2.2190 - val_accuracy: 0.6852\n",
            "106/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 409us/step - loss: 2.4745 - accuracy: 0.5920 - val_loss: 2.1826 - val_accuracy: 0.6910\n",
            "107/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 400us/step - loss: 2.3676 - accuracy: 0.6158 - val_loss: 2.1226 - val_accuracy: 0.7058\n",
            "108/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 413us/step - loss: 2.3591 - accuracy: 0.6148 - val_loss: 2.0710 - val_accuracy: 0.7092\n",
            "109/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 413us/step - loss: 2.3524 - accuracy: 0.6132 - val_loss: 2.0538 - val_accuracy: 0.7120\n",
            "110/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 412us/step - loss: 2.2083 - accuracy: 0.6440 - val_loss: 2.0058 - val_accuracy: 0.7246\n",
            "111/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 402us/step - loss: 2.2500 - accuracy: 0.6318 - val_loss: 1.9156 - val_accuracy: 0.7410\n",
            "112/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 398us/step - loss: 2.1819 - accuracy: 0.6530 - val_loss: 1.8126 - val_accuracy: 0.7542\n",
            "113/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 402us/step - loss: 2.0875 - accuracy: 0.6602 - val_loss: 1.8725 - val_accuracy: 0.7480\n",
            "114/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 403us/step - loss: 2.0695 - accuracy: 0.6694 - val_loss: 1.7876 - val_accuracy: 0.7560\n",
            "115/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 402us/step - loss: 1.9872 - accuracy: 0.6808 - val_loss: 1.7615 - val_accuracy: 0.7638\n",
            "116/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 396us/step - loss: 2.0265 - accuracy: 0.6764 - val_loss: 1.8029 - val_accuracy: 0.7500\n",
            "117/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 395us/step - loss: 1.9593 - accuracy: 0.6968 - val_loss: 1.7222 - val_accuracy: 0.7694\n",
            "118/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 405us/step - loss: 1.9793 - accuracy: 0.6876 - val_loss: 1.7054 - val_accuracy: 0.7854\n",
            "119/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 406us/step - loss: 1.8934 - accuracy: 0.6984 - val_loss: 1.6764 - val_accuracy: 0.7760\n",
            "120/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 411us/step - loss: 1.8731 - accuracy: 0.7044 - val_loss: 1.6600 - val_accuracy: 0.7762\n",
            "121/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 413us/step - loss: 1.8633 - accuracy: 0.7042 - val_loss: 1.5655 - val_accuracy: 0.7950\n",
            "122/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 400us/step - loss: 1.8155 - accuracy: 0.7212 - val_loss: 1.5577 - val_accuracy: 0.8050\n",
            "123/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 406us/step - loss: 1.7936 - accuracy: 0.7226 - val_loss: 1.5797 - val_accuracy: 0.7924\n",
            "124/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 400us/step - loss: 1.6727 - accuracy: 0.7428 - val_loss: 1.4695 - val_accuracy: 0.8170\n",
            "125/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 401us/step - loss: 1.6898 - accuracy: 0.7436 - val_loss: 1.4540 - val_accuracy: 0.8192\n",
            "126/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 405us/step - loss: 1.6531 - accuracy: 0.7418 - val_loss: 1.4008 - val_accuracy: 0.8206\n",
            "127/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 399us/step - loss: 1.6137 - accuracy: 0.7506 - val_loss: 1.3895 - val_accuracy: 0.8118\n",
            "128/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 406us/step - loss: 1.6290 - accuracy: 0.7560 - val_loss: 1.3585 - val_accuracy: 0.8338\n",
            "129/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 401us/step - loss: 1.6117 - accuracy: 0.7524 - val_loss: 1.2903 - val_accuracy: 0.8374\n",
            "130/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 411us/step - loss: 1.5511 - accuracy: 0.7594 - val_loss: 1.2891 - val_accuracy: 0.8316\n",
            "131/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 396us/step - loss: 1.5014 - accuracy: 0.7786 - val_loss: 1.3013 - val_accuracy: 0.8308\n",
            "132/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 416us/step - loss: 1.4592 - accuracy: 0.7834 - val_loss: 1.2238 - val_accuracy: 0.8452\n",
            "133/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 415us/step - loss: 1.4994 - accuracy: 0.7738 - val_loss: 1.1824 - val_accuracy: 0.8542\n",
            "134/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 402us/step - loss: 1.4425 - accuracy: 0.7844 - val_loss: 1.1497 - val_accuracy: 0.8548\n",
            "135/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 422us/step - loss: 1.4319 - accuracy: 0.7870 - val_loss: 1.1830 - val_accuracy: 0.8530\n",
            "136/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 405us/step - loss: 1.3450 - accuracy: 0.8006 - val_loss: 1.1152 - val_accuracy: 0.8616\n",
            "137/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 408us/step - loss: 1.4073 - accuracy: 0.7928 - val_loss: 1.1236 - val_accuracy: 0.8584\n",
            "138/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 413us/step - loss: 1.3359 - accuracy: 0.8014 - val_loss: 1.1054 - val_accuracy: 0.8554\n",
            "139/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 420us/step - loss: 1.3105 - accuracy: 0.8080 - val_loss: 1.0732 - val_accuracy: 0.8714\n",
            "140/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 406us/step - loss: 1.2528 - accuracy: 0.8166 - val_loss: 1.1127 - val_accuracy: 0.8648\n",
            "141/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 412us/step - loss: 1.2472 - accuracy: 0.8178 - val_loss: 1.0218 - val_accuracy: 0.8784\n",
            "142/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 420us/step - loss: 1.2278 - accuracy: 0.8228 - val_loss: 0.9639 - val_accuracy: 0.8844\n",
            "143/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 402us/step - loss: 1.2285 - accuracy: 0.8170 - val_loss: 1.0322 - val_accuracy: 0.8720\n",
            "144/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 407us/step - loss: 1.1853 - accuracy: 0.8240 - val_loss: 0.8959 - val_accuracy: 0.8954\n",
            "145/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 414us/step - loss: 1.1545 - accuracy: 0.8328 - val_loss: 0.9459 - val_accuracy: 0.8820\n",
            "146/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 409us/step - loss: 1.1736 - accuracy: 0.8298 - val_loss: 0.9650 - val_accuracy: 0.8752\n",
            "147/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 408us/step - loss: 1.0828 - accuracy: 0.8468 - val_loss: 0.8727 - val_accuracy: 0.8946\n",
            "148/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 418us/step - loss: 1.0743 - accuracy: 0.8454 - val_loss: 0.8732 - val_accuracy: 0.8952\n",
            "149/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 397us/step - loss: 1.1223 - accuracy: 0.8380 - val_loss: 0.8399 - val_accuracy: 0.8968\n",
            "150/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 409us/step - loss: 1.0736 - accuracy: 0.8504 - val_loss: 0.8629 - val_accuracy: 0.8986\n",
            "151/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 404us/step - loss: 1.0527 - accuracy: 0.8480 - val_loss: 0.7800 - val_accuracy: 0.9100\n",
            "152/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 406us/step - loss: 1.0271 - accuracy: 0.8550 - val_loss: 0.8736 - val_accuracy: 0.8946\n",
            "153/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 401us/step - loss: 0.9801 - accuracy: 0.8648 - val_loss: 0.7773 - val_accuracy: 0.9082\n",
            "154/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 398us/step - loss: 0.9624 - accuracy: 0.8606 - val_loss: 0.7587 - val_accuracy: 0.9122\n",
            "155/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 412us/step - loss: 0.9653 - accuracy: 0.8668 - val_loss: 0.7569 - val_accuracy: 0.9102\n",
            "156/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 409us/step - loss: 0.9171 - accuracy: 0.8768 - val_loss: 0.7783 - val_accuracy: 0.9008\n",
            "157/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 410us/step - loss: 0.9824 - accuracy: 0.8598 - val_loss: 0.7716 - val_accuracy: 0.9114\n",
            "158/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 412us/step - loss: 0.9050 - accuracy: 0.8720 - val_loss: 0.6798 - val_accuracy: 0.9246\n",
            "159/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 419us/step - loss: 0.8868 - accuracy: 0.8782 - val_loss: 0.7305 - val_accuracy: 0.9134\n",
            "160/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 414us/step - loss: 0.8656 - accuracy: 0.8774 - val_loss: 0.6773 - val_accuracy: 0.9174\n",
            "161/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 412us/step - loss: 0.9094 - accuracy: 0.8732 - val_loss: 0.7563 - val_accuracy: 0.9078\n",
            "162/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 408us/step - loss: 0.8625 - accuracy: 0.8852 - val_loss: 0.6772 - val_accuracy: 0.9216\n",
            "163/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 411us/step - loss: 0.9143 - accuracy: 0.8728 - val_loss: 0.7034 - val_accuracy: 0.9154\n",
            "164/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 417us/step - loss: 0.8791 - accuracy: 0.8746 - val_loss: 0.7079 - val_accuracy: 0.9188\n",
            "165/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 410us/step - loss: 0.8265 - accuracy: 0.8880 - val_loss: 0.6240 - val_accuracy: 0.9240\n",
            "166/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 414us/step - loss: 0.7777 - accuracy: 0.8974 - val_loss: 0.6988 - val_accuracy: 0.9150\n",
            "167/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 403us/step - loss: 0.8305 - accuracy: 0.8876 - val_loss: 0.6100 - val_accuracy: 0.9274\n",
            "168/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 407us/step - loss: 0.7893 - accuracy: 0.8900 - val_loss: 0.6668 - val_accuracy: 0.9140\n",
            "169/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 414us/step - loss: 0.8148 - accuracy: 0.8926 - val_loss: 0.6679 - val_accuracy: 0.9230\n",
            "170/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 419us/step - loss: 0.7757 - accuracy: 0.8962 - val_loss: 0.6510 - val_accuracy: 0.9218\n",
            "171/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 403us/step - loss: 0.7666 - accuracy: 0.8958 - val_loss: 0.5848 - val_accuracy: 0.9266\n",
            "172/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 417us/step - loss: 0.7406 - accuracy: 0.8994 - val_loss: 0.5751 - val_accuracy: 0.9296\n",
            "173/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 413us/step - loss: 0.7387 - accuracy: 0.8992 - val_loss: 0.5893 - val_accuracy: 0.9270\n",
            "174/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 402us/step - loss: 0.7045 - accuracy: 0.9026 - val_loss: 0.5447 - val_accuracy: 0.9280\n",
            "175/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 417us/step - loss: 0.7532 - accuracy: 0.8990 - val_loss: 0.5440 - val_accuracy: 0.9338\n",
            "176/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 410us/step - loss: 0.7437 - accuracy: 0.9028 - val_loss: 0.5865 - val_accuracy: 0.9252\n",
            "177/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 416us/step - loss: 0.6795 - accuracy: 0.9090 - val_loss: 0.5411 - val_accuracy: 0.9344\n",
            "178/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 411us/step - loss: 0.7007 - accuracy: 0.9012 - val_loss: 0.5581 - val_accuracy: 0.9260\n",
            "179/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 421us/step - loss: 0.6832 - accuracy: 0.9086 - val_loss: 0.5115 - val_accuracy: 0.9390\n",
            "180/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 413us/step - loss: 0.6835 - accuracy: 0.9100 - val_loss: 0.5173 - val_accuracy: 0.9446\n",
            "181/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 415us/step - loss: 0.6935 - accuracy: 0.9046 - val_loss: 0.5112 - val_accuracy: 0.9412\n",
            "182/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 410us/step - loss: 0.7066 - accuracy: 0.9000 - val_loss: 0.5668 - val_accuracy: 0.9314\n",
            "183/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 429us/step - loss: 0.6225 - accuracy: 0.9148 - val_loss: 0.5051 - val_accuracy: 0.9388\n",
            "184/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 414us/step - loss: 0.6505 - accuracy: 0.9174 - val_loss: 0.5356 - val_accuracy: 0.9334\n",
            "185/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 425us/step - loss: 0.6816 - accuracy: 0.9102 - val_loss: 0.4791 - val_accuracy: 0.9428\n",
            "186/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 407us/step - loss: 0.6619 - accuracy: 0.9106 - val_loss: 0.5131 - val_accuracy: 0.9410\n",
            "187/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 407us/step - loss: 0.6706 - accuracy: 0.9084 - val_loss: 0.5034 - val_accuracy: 0.9348\n",
            "188/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 413us/step - loss: 0.6367 - accuracy: 0.9156 - val_loss: 0.4722 - val_accuracy: 0.9390\n",
            "189/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 412us/step - loss: 0.6209 - accuracy: 0.9154 - val_loss: 0.4924 - val_accuracy: 0.9394\n",
            "190/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 417us/step - loss: 0.5862 - accuracy: 0.9240 - val_loss: 0.4789 - val_accuracy: 0.9396\n",
            "191/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 419us/step - loss: 0.6070 - accuracy: 0.9210 - val_loss: 0.4566 - val_accuracy: 0.9392\n",
            "192/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 428us/step - loss: 0.5869 - accuracy: 0.9196 - val_loss: 0.4740 - val_accuracy: 0.9422\n",
            "193/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 428us/step - loss: 0.6011 - accuracy: 0.9222 - val_loss: 0.4707 - val_accuracy: 0.9468\n",
            "194/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 412us/step - loss: 0.5858 - accuracy: 0.9198 - val_loss: 0.4336 - val_accuracy: 0.9468\n",
            "195/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 422us/step - loss: 0.5947 - accuracy: 0.9202 - val_loss: 0.4398 - val_accuracy: 0.9484\n",
            "196/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 427us/step - loss: 0.5615 - accuracy: 0.9256 - val_loss: 0.4687 - val_accuracy: 0.9408\n",
            "197/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 420us/step - loss: 0.5673 - accuracy: 0.9236 - val_loss: 0.4215 - val_accuracy: 0.9478\n",
            "198/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 415us/step - loss: 0.5637 - accuracy: 0.9294 - val_loss: 0.4343 - val_accuracy: 0.9456\n",
            "199/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 417us/step - loss: 0.6137 - accuracy: 0.9172 - val_loss: 0.4341 - val_accuracy: 0.9462\n",
            "200/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 414us/step - loss: 0.6006 - accuracy: 0.9218 - val_loss: 0.3884 - val_accuracy: 0.9512\n",
            "201/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 417us/step - loss: 0.5635 - accuracy: 0.9268 - val_loss: 0.4230 - val_accuracy: 0.9480\n",
            "202/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 418us/step - loss: 0.5658 - accuracy: 0.9256 - val_loss: 0.4512 - val_accuracy: 0.9440\n",
            "203/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 416us/step - loss: 0.6056 - accuracy: 0.9200 - val_loss: 0.4215 - val_accuracy: 0.9438\n",
            "204/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 414us/step - loss: 0.5344 - accuracy: 0.9278 - val_loss: 0.4380 - val_accuracy: 0.9458\n",
            "205/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 426us/step - loss: 0.5138 - accuracy: 0.9304 - val_loss: 0.3961 - val_accuracy: 0.9506\n",
            "206/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 418us/step - loss: 0.5704 - accuracy: 0.9264 - val_loss: 0.3948 - val_accuracy: 0.9486\n",
            "207/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 436us/step - loss: 0.5551 - accuracy: 0.9248 - val_loss: 0.3943 - val_accuracy: 0.9526\n",
            "208/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 427us/step - loss: 0.4828 - accuracy: 0.9366 - val_loss: 0.4855 - val_accuracy: 0.9334\n",
            "209/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 416us/step - loss: 0.4814 - accuracy: 0.9376 - val_loss: 0.3574 - val_accuracy: 0.9580\n",
            "210/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 422us/step - loss: 0.4560 - accuracy: 0.9418 - val_loss: 0.4189 - val_accuracy: 0.9474\n",
            "211/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 417us/step - loss: 0.5182 - accuracy: 0.9278 - val_loss: 0.3576 - val_accuracy: 0.9526\n",
            "212/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 416us/step - loss: 0.4829 - accuracy: 0.9360 - val_loss: 0.3724 - val_accuracy: 0.9542\n",
            "213/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 422us/step - loss: 0.5509 - accuracy: 0.9252 - val_loss: 0.4110 - val_accuracy: 0.9492\n",
            "214/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 431us/step - loss: 0.5758 - accuracy: 0.9202 - val_loss: 0.4106 - val_accuracy: 0.9498\n",
            "215/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 440us/step - loss: 0.4821 - accuracy: 0.9340 - val_loss: 0.3331 - val_accuracy: 0.9600\n",
            "216/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 430us/step - loss: 0.4816 - accuracy: 0.9352 - val_loss: 0.3872 - val_accuracy: 0.9562\n",
            "217/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 418us/step - loss: 0.4919 - accuracy: 0.9354 - val_loss: 0.3316 - val_accuracy: 0.9600\n",
            "218/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 424us/step - loss: 0.4545 - accuracy: 0.9382 - val_loss: 0.3393 - val_accuracy: 0.9538\n",
            "219/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 426us/step - loss: 0.4772 - accuracy: 0.9376 - val_loss: 0.3637 - val_accuracy: 0.9542\n",
            "220/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 423us/step - loss: 0.4726 - accuracy: 0.9400 - val_loss: 0.3490 - val_accuracy: 0.9598\n",
            "221/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 440us/step - loss: 0.4793 - accuracy: 0.9378 - val_loss: 0.3734 - val_accuracy: 0.9488\n",
            "222/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 415us/step - loss: 0.5026 - accuracy: 0.9352 - val_loss: 0.3776 - val_accuracy: 0.9526\n",
            "223/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 427us/step - loss: 0.4759 - accuracy: 0.9306 - val_loss: 0.3640 - val_accuracy: 0.9504\n",
            "224/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 424us/step - loss: 0.4789 - accuracy: 0.9378 - val_loss: 0.3393 - val_accuracy: 0.9588\n",
            "225/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 431us/step - loss: 0.4675 - accuracy: 0.9372 - val_loss: 0.3774 - val_accuracy: 0.9558\n",
            "226/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 422us/step - loss: 0.5579 - accuracy: 0.9288 - val_loss: 0.3467 - val_accuracy: 0.9576\n",
            "227/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 446us/step - loss: 0.4209 - accuracy: 0.9410 - val_loss: 0.3965 - val_accuracy: 0.9468\n",
            "228/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 447us/step - loss: 0.4648 - accuracy: 0.9406 - val_loss: 0.3432 - val_accuracy: 0.9578\n",
            "229/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 438us/step - loss: 0.5176 - accuracy: 0.9314 - val_loss: 0.3913 - val_accuracy: 0.9500\n",
            "230/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 435us/step - loss: 0.4967 - accuracy: 0.9360 - val_loss: 0.3768 - val_accuracy: 0.9560\n",
            "231/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 424us/step - loss: 0.4823 - accuracy: 0.9396 - val_loss: 0.3141 - val_accuracy: 0.9628\n",
            "232/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 424us/step - loss: 0.4552 - accuracy: 0.9438 - val_loss: 0.3027 - val_accuracy: 0.9600\n",
            "233/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 423us/step - loss: 0.4230 - accuracy: 0.9444 - val_loss: 0.3282 - val_accuracy: 0.9578\n",
            "234/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 427us/step - loss: 0.4708 - accuracy: 0.9340 - val_loss: 0.3755 - val_accuracy: 0.9472\n",
            "235/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 414us/step - loss: 0.4087 - accuracy: 0.9416 - val_loss: 0.3489 - val_accuracy: 0.9550\n",
            "236/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 433us/step - loss: 0.4523 - accuracy: 0.9386 - val_loss: 0.3294 - val_accuracy: 0.9594\n",
            "237/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 431us/step - loss: 0.4482 - accuracy: 0.9392 - val_loss: 0.3893 - val_accuracy: 0.9538\n",
            "238/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 434us/step - loss: 0.4506 - accuracy: 0.9400 - val_loss: 0.3563 - val_accuracy: 0.9530\n",
            "239/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 406us/step - loss: 0.4631 - accuracy: 0.9388 - val_loss: 0.3517 - val_accuracy: 0.9570\n",
            "240/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 421us/step - loss: 0.4394 - accuracy: 0.9492 - val_loss: 0.2732 - val_accuracy: 0.9688\n",
            "241/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 428us/step - loss: 0.4131 - accuracy: 0.9440 - val_loss: 0.3108 - val_accuracy: 0.9628\n",
            "242/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 420us/step - loss: 0.4081 - accuracy: 0.9440 - val_loss: 0.3488 - val_accuracy: 0.9532\n",
            "243/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 415us/step - loss: 0.4218 - accuracy: 0.9440 - val_loss: 0.3199 - val_accuracy: 0.9612\n",
            "244/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 414us/step - loss: 0.4196 - accuracy: 0.9452 - val_loss: 0.3118 - val_accuracy: 0.9610\n",
            "245/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 427us/step - loss: 0.3984 - accuracy: 0.9474 - val_loss: 0.3152 - val_accuracy: 0.9612\n",
            "246/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 422us/step - loss: 0.4311 - accuracy: 0.9400 - val_loss: 0.2962 - val_accuracy: 0.9634\n",
            "247/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 424us/step - loss: 0.4427 - accuracy: 0.9402 - val_loss: 0.3643 - val_accuracy: 0.9556\n",
            "248/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 420us/step - loss: 0.4436 - accuracy: 0.9410 - val_loss: 0.3296 - val_accuracy: 0.9612\n",
            "249/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 423us/step - loss: 0.4356 - accuracy: 0.9440 - val_loss: 0.3044 - val_accuracy: 0.9630\n",
            "250/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 428us/step - loss: 0.3828 - accuracy: 0.9490 - val_loss: 0.2771 - val_accuracy: 0.9684\n",
            "251/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 435us/step - loss: 0.4115 - accuracy: 0.9416 - val_loss: 0.3557 - val_accuracy: 0.9580\n",
            "252/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 432us/step - loss: 0.3686 - accuracy: 0.9490 - val_loss: 0.3319 - val_accuracy: 0.9634\n",
            "253/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 435us/step - loss: 0.4639 - accuracy: 0.9432 - val_loss: 0.2853 - val_accuracy: 0.9638\n",
            "254/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 437us/step - loss: 0.4792 - accuracy: 0.9362 - val_loss: 0.3423 - val_accuracy: 0.9564\n",
            "255/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 426us/step - loss: 0.4066 - accuracy: 0.9480 - val_loss: 0.3347 - val_accuracy: 0.9576\n",
            "256/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 428us/step - loss: 0.4724 - accuracy: 0.9376 - val_loss: 0.2919 - val_accuracy: 0.9658\n",
            "257/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 432us/step - loss: 0.4215 - accuracy: 0.9410 - val_loss: 0.2725 - val_accuracy: 0.9642\n",
            "258/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 425us/step - loss: 0.4419 - accuracy: 0.9464 - val_loss: 0.3282 - val_accuracy: 0.9636\n",
            "259/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 420us/step - loss: 0.4133 - accuracy: 0.9474 - val_loss: 0.2633 - val_accuracy: 0.9680\n",
            "260/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 438us/step - loss: 0.4547 - accuracy: 0.9410 - val_loss: 0.3277 - val_accuracy: 0.9632\n",
            "261/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 432us/step - loss: 0.3550 - accuracy: 0.9518 - val_loss: 0.2824 - val_accuracy: 0.9662\n",
            "262/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 432us/step - loss: 0.4278 - accuracy: 0.9406 - val_loss: 0.2624 - val_accuracy: 0.9686\n",
            "263/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 437us/step - loss: 0.4622 - accuracy: 0.9398 - val_loss: 0.3406 - val_accuracy: 0.9556\n",
            "264/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 426us/step - loss: 0.3704 - accuracy: 0.9546 - val_loss: 0.3197 - val_accuracy: 0.9664\n",
            "265/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 421us/step - loss: 0.3736 - accuracy: 0.9492 - val_loss: 0.3204 - val_accuracy: 0.9618\n",
            "266/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 427us/step - loss: 0.3926 - accuracy: 0.9480 - val_loss: 0.3420 - val_accuracy: 0.9558\n",
            "267/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 426us/step - loss: 0.3492 - accuracy: 0.9552 - val_loss: 0.3409 - val_accuracy: 0.9594\n",
            "268/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 424us/step - loss: 0.4315 - accuracy: 0.9456 - val_loss: 0.3871 - val_accuracy: 0.9564\n",
            "269/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 438us/step - loss: 0.4241 - accuracy: 0.9416 - val_loss: 0.3569 - val_accuracy: 0.9562\n",
            "270/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 433us/step - loss: 0.4078 - accuracy: 0.9438 - val_loss: 0.2925 - val_accuracy: 0.9646\n",
            "271/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 419us/step - loss: 0.3924 - accuracy: 0.9468 - val_loss: 0.3646 - val_accuracy: 0.9536\n",
            "272/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 433us/step - loss: 0.3643 - accuracy: 0.9520 - val_loss: 0.3494 - val_accuracy: 0.9608\n",
            "273/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 418us/step - loss: 0.3252 - accuracy: 0.9564 - val_loss: 0.2771 - val_accuracy: 0.9666\n",
            "274/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 422us/step - loss: 0.4002 - accuracy: 0.9480 - val_loss: 0.3212 - val_accuracy: 0.9644\n",
            "275/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 422us/step - loss: 0.4312 - accuracy: 0.9450 - val_loss: 0.3275 - val_accuracy: 0.9604\n",
            "276/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 431us/step - loss: 0.4204 - accuracy: 0.9418 - val_loss: 0.2861 - val_accuracy: 0.9620\n",
            "277/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 434us/step - loss: 0.4327 - accuracy: 0.9462 - val_loss: 0.2808 - val_accuracy: 0.9636\n",
            "278/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 429us/step - loss: 0.4367 - accuracy: 0.9428 - val_loss: 0.3191 - val_accuracy: 0.9568\n",
            "279/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 434us/step - loss: 0.3983 - accuracy: 0.9492 - val_loss: 0.3552 - val_accuracy: 0.9592\n",
            "280/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 432us/step - loss: 0.3887 - accuracy: 0.9472 - val_loss: 0.2665 - val_accuracy: 0.9658\n",
            "281/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 435us/step - loss: 0.3997 - accuracy: 0.9486 - val_loss: 0.2733 - val_accuracy: 0.9666\n",
            "282/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 435us/step - loss: 0.3756 - accuracy: 0.9484 - val_loss: 0.2950 - val_accuracy: 0.9626\n",
            "283/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 430us/step - loss: 0.3351 - accuracy: 0.9502 - val_loss: 0.2685 - val_accuracy: 0.9652\n",
            "284/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 431us/step - loss: 0.3385 - accuracy: 0.9566 - val_loss: 0.2542 - val_accuracy: 0.9664\n",
            "285/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 446us/step - loss: 0.3217 - accuracy: 0.9572 - val_loss: 0.3173 - val_accuracy: 0.9628\n",
            "286/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 421us/step - loss: 0.3968 - accuracy: 0.9494 - val_loss: 0.3565 - val_accuracy: 0.9536\n",
            "287/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 434us/step - loss: 0.4626 - accuracy: 0.9386 - val_loss: 0.3295 - val_accuracy: 0.9590\n",
            "288/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 436us/step - loss: 0.3845 - accuracy: 0.9450 - val_loss: 0.3039 - val_accuracy: 0.9630\n",
            "289/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 427us/step - loss: 0.4553 - accuracy: 0.9394 - val_loss: 0.2742 - val_accuracy: 0.9644\n",
            "290/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 423us/step - loss: 0.4083 - accuracy: 0.9486 - val_loss: 0.2771 - val_accuracy: 0.9692\n",
            "291/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 447us/step - loss: 0.3854 - accuracy: 0.9468 - val_loss: 0.3103 - val_accuracy: 0.9640\n",
            "292/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 429us/step - loss: 0.3969 - accuracy: 0.9484 - val_loss: 0.2863 - val_accuracy: 0.9642\n",
            "293/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 435us/step - loss: 0.3743 - accuracy: 0.9508 - val_loss: 0.2994 - val_accuracy: 0.9654\n",
            "294/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 431us/step - loss: 0.3768 - accuracy: 0.9484 - val_loss: 0.3063 - val_accuracy: 0.9620\n",
            "295/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 422us/step - loss: 0.3746 - accuracy: 0.9496 - val_loss: 0.3148 - val_accuracy: 0.9662\n",
            "296/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 421us/step - loss: 0.3738 - accuracy: 0.9516 - val_loss: 0.2518 - val_accuracy: 0.9680\n",
            "297/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 427us/step - loss: 0.3634 - accuracy: 0.9562 - val_loss: 0.3113 - val_accuracy: 0.9608\n",
            "298/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 433us/step - loss: 0.3427 - accuracy: 0.9530 - val_loss: 0.3012 - val_accuracy: 0.9556\n",
            "299/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 437us/step - loss: 0.3641 - accuracy: 0.9508 - val_loss: 0.2865 - val_accuracy: 0.9550\n",
            "300/300\n",
            "Train on 5000 samples, validate on 5000 samples\n",
            "Epoch 1/1\n",
            "5000/5000 [==============================] - 2s 439us/step - loss: 0.3436 - accuracy: 0.9544 - val_loss: 0.2552 - val_accuracy: 0.9686\n",
            "100000/100000 [==============================] - 10s 101us/step\n",
            "Train set: Loss=0.2559 ; Accuracy=96.9%\n",
            "100000/100000 [==============================] - 10s 101us/step\n",
            "Test set: Loss=14.9620 ; Accuracy=1.8%\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DMq-auNX4TJ0"
      },
      "source": [
        "### Load embeddings"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "o2SB5xqsvZq3"
      },
      "source": [
        "embeddings = Embeddings(read_embeddings('embeddings.csv'))\n",
        "\n",
        "state_space_size = embeddings.size() * history_length\n",
        "action_space_size = embeddings.size() * ra_length"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UFGOLABA4XJr"
      },
      "source": [
        "### Start Agent training"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_LEg4umm4qZc"
      },
      "source": [
        "environment = Environment(data, embeddings, alpha, gamma, fixed_length)\n",
        "\n",
        "tf.reset_default_graph() # For multiple consecutive executions\n",
        "\n",
        "sess = tf.Session()\n",
        "# '1: Initialize actor network f_θ^π and critic network Q(s, a|θ^µ) with random weights'\n",
        "actor = Actor(sess, state_space_size, action_space_size, batch_size, ra_length, history_length, embeddings.size(), tau, actor_lr)\n",
        "critic = Critic(sess, state_space_size, action_space_size, history_length, embeddings.size(), tau, critic_lr)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BpUHlaigvZnu",
        "outputId": "528eaaec-2aa9-43e9-c5a2-1b7b9a0bd1d9"
      },
      "source": [
        "#collapse-output\n",
        "train(sess, environment, actor, critic, embeddings, history_length, ra_length, buffer_size, batch_size, discount_factor, nb_episodes, filename_summary)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From <ipython-input-19-1a5cd2de5f02>:69: GRUCell.__init__ (from tensorflow.python.ops.rnn_cell_impl) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "This class is equivalent as tf.keras.layers.GRUCell, and will be replaced by that in Tensorflow 2.0.\n",
            "WARNING:tensorflow:From <ipython-input-19-1a5cd2de5f02>:70: dynamic_rnn (from tensorflow.python.ops.rnn) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Please use `keras.layers.RNN(cell)`, which is equivalent to this API\n",
            "WARNING:tensorflow:From /tensorflow-1.15.2/python3.7/tensorflow_core/python/ops/rnn_cell_impl.py:559: Layer.add_variable (from tensorflow.python.keras.engine.base_layer) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Please use `layer.add_weight` method instead.\n",
            "WARNING:tensorflow:From <ipython-input-19-1a5cd2de5f02>:40: div (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Deprecated in favor of operator or tf.math.divide.\n",
            "Episode 1/100 Reward=552 Time=4s No replay\n",
            "Episode 2/100 Reward=551 Time=52s Loss=2082.7095\n",
            "Episode 3/100 Reward=551 Time=68s Loss=123.3409\n",
            "Episode 4/100 Reward=551 Time=67s Loss=67.4848\n",
            "Episode 5/100 Reward=551 Time=67s Loss=44.0779\n",
            "Episode 6/100 Reward=552 Time=67s Loss=40.3507\n",
            "Episode 7/100 Reward=552 Time=67s Loss=29.4237\n",
            "Episode 8/100 Reward=552 Time=68s Loss=26.5713\n",
            "Episode 9/100 Reward=552 Time=68s Loss=27.1646\n",
            "Episode 10/100 Reward=552 Time=68s Loss=25.9928\n",
            "Episode 11/100 Reward=552 Time=67s Loss=22.3598\n",
            "Episode 12/100 Reward=552 Time=68s Loss=18.7026\n",
            "Episode 13/100 Reward=552 Time=67s Loss=17.8249\n",
            "Episode 14/100 Reward=551 Time=67s Loss=19.8687\n",
            "Episode 15/100 Reward=552 Time=68s Loss=20.9238\n",
            "Episode 16/100 Reward=551 Time=67s Loss=19.9583\n",
            "Episode 17/100 Reward=551 Time=68s Loss=20.2933\n",
            "Episode 18/100 Reward=551 Time=67s Loss=19.9462\n",
            "Episode 19/100 Reward=551 Time=67s Loss=24.3696\n",
            "Episode 20/100 Reward=551 Time=67s Loss=25.6828\n",
            "Episode 21/100 Reward=551 Time=67s Loss=28.5111\n",
            "Episode 22/100 Reward=551 Time=67s Loss=29.4505\n",
            "Episode 23/100 Reward=551 Time=67s Loss=27.2863\n",
            "Episode 24/100 Reward=551 Time=68s Loss=28.4667\n",
            "Episode 25/100 Reward=551 Time=67s Loss=26.7666\n",
            "Episode 26/100 Reward=551 Time=67s Loss=26.2378\n",
            "Episode 27/100 Reward=551 Time=67s Loss=25.0391\n",
            "Episode 28/100 Reward=551 Time=67s Loss=25.0170\n",
            "Episode 29/100 Reward=551 Time=68s Loss=22.9655\n",
            "Episode 30/100 Reward=551 Time=68s Loss=23.8730\n",
            "Episode 31/100 Reward=551 Time=67s Loss=20.4020\n",
            "Episode 32/100 Reward=551 Time=67s Loss=22.4662\n",
            "Episode 33/100 Reward=551 Time=67s Loss=23.8943\n",
            "Episode 34/100 Reward=551 Time=67s Loss=21.1020\n",
            "Episode 35/100 Reward=551 Time=67s Loss=22.8528\n",
            "Episode 36/100 Reward=551 Time=67s Loss=21.2307\n",
            "Episode 37/100 Reward=551 Time=67s Loss=19.7094\n",
            "Episode 38/100 Reward=551 Time=67s Loss=21.4233\n",
            "Episode 39/100 Reward=551 Time=67s Loss=23.6903\n",
            "Episode 40/100 Reward=551 Time=67s Loss=24.4852\n",
            "Episode 41/100 Reward=551 Time=67s Loss=25.7120\n",
            "Episode 42/100 Reward=551 Time=67s Loss=21.7722\n",
            "Episode 43/100 Reward=551 Time=67s Loss=20.9898\n",
            "Episode 44/100 Reward=551 Time=67s Loss=20.6604\n",
            "Episode 45/100 Reward=551 Time=68s Loss=20.8646\n",
            "Episode 46/100 Reward=551 Time=67s Loss=19.4622\n",
            "Episode 47/100 Reward=551 Time=67s Loss=20.4751\n",
            "Episode 48/100 Reward=551 Time=67s Loss=18.9989\n",
            "Episode 49/100 Reward=551 Time=67s Loss=17.7407\n",
            "Episode 50/100 Reward=551 Time=67s Loss=17.3576\n",
            "Episode 51/100 Reward=551 Time=68s Loss=17.2397\n",
            "Episode 52/100 Reward=552 Time=67s Loss=16.5722\n",
            "Episode 53/100 Reward=552 Time=67s Loss=15.5511\n",
            "Episode 54/100 Reward=552 Time=67s Loss=15.7651\n",
            "Episode 55/100 Reward=552 Time=67s Loss=14.0308\n",
            "Episode 56/100 Reward=552 Time=67s Loss=14.4518\n",
            "Episode 57/100 Reward=551 Time=67s Loss=15.9018\n",
            "Episode 58/100 Reward=552 Time=67s Loss=14.2520\n",
            "Episode 59/100 Reward=551 Time=67s Loss=14.2282\n",
            "Episode 60/100 Reward=551 Time=67s Loss=14.1576\n",
            "Episode 61/100 Reward=551 Time=67s Loss=13.1366\n",
            "Episode 62/100 Reward=551 Time=67s Loss=13.7383\n",
            "Episode 63/100 Reward=551 Time=67s Loss=12.3095\n",
            "Episode 64/100 Reward=551 Time=68s Loss=11.7993\n",
            "Episode 65/100 Reward=551 Time=67s Loss=12.1072\n",
            "Episode 66/100 Reward=551 Time=67s Loss=12.8614\n",
            "Episode 67/100 Reward=552 Time=67s Loss=11.4739\n",
            "Episode 68/100 Reward=552 Time=67s Loss=12.6560\n",
            "Episode 69/100 Reward=552 Time=67s Loss=12.8773\n",
            "Episode 70/100 Reward=552 Time=67s Loss=11.7954\n",
            "Episode 71/100 Reward=552 Time=67s Loss=11.2212\n",
            "Episode 72/100 Reward=552 Time=67s Loss=12.3400\n",
            "Episode 73/100 Reward=552 Time=67s Loss=12.5248\n",
            "Episode 74/100 Reward=552 Time=67s Loss=11.2045\n",
            "Episode 75/100 Reward=552 Time=67s Loss=11.1089\n",
            "Episode 76/100 Reward=552 Time=67s Loss=11.6253\n",
            "Episode 77/100 Reward=552 Time=67s Loss=10.9183\n",
            "Episode 78/100 Reward=552 Time=67s Loss=9.2532\n",
            "Episode 79/100 Reward=552 Time=67s Loss=10.4258\n",
            "Episode 80/100 Reward=552 Time=67s Loss=10.0044\n",
            "Episode 81/100 Reward=552 Time=67s Loss=10.4150\n",
            "Episode 82/100 Reward=552 Time=67s Loss=10.9766\n",
            "Episode 83/100 Reward=551 Time=67s Loss=8.8571\n",
            "Episode 84/100 Reward=551 Time=67s Loss=10.5467\n",
            "Episode 85/100 Reward=551 Time=67s Loss=8.8356\n",
            "Episode 86/100 Reward=551 Time=67s Loss=11.0192\n",
            "Episode 87/100 Reward=551 Time=67s Loss=9.3348\n",
            "Episode 88/100 Reward=551 Time=67s Loss=11.0042\n",
            "Episode 89/100 Reward=551 Time=67s Loss=8.5757\n",
            "Episode 90/100 Reward=551 Time=67s Loss=8.8545\n",
            "Episode 91/100 Reward=551 Time=67s Loss=10.0286\n",
            "Episode 92/100 Reward=551 Time=67s Loss=10.2471\n",
            "Episode 93/100 Reward=551 Time=67s Loss=9.3180\n",
            "Episode 94/100 Reward=551 Time=68s Loss=8.2303\n",
            "Episode 95/100 Reward=551 Time=68s Loss=9.1910\n",
            "Episode 96/100 Reward=551 Time=67s Loss=8.2708\n",
            "Episode 97/100 Reward=551 Time=67s Loss=8.1502\n",
            "Episode 98/100 Reward=551 Time=67s Loss=8.1103\n",
            "Episode 99/100 Reward=551 Time=67s Loss=8.2638\n",
            "Episode 100/100 Reward=551 Time=67s Loss=8.2967\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xidr_jCI5A_Z"
      },
      "source": [
        "## Testing"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PUpOpP-kvZk1"
      },
      "source": [
        "dict_embeddings = {}\n",
        "for i, item in enumerate(embeddings.get_embedding_vector()):\n",
        "  str_item = str(item)\n",
        "  assert(str_item not in dict_embeddings)\n",
        "  dict_embeddings[str_item] = i"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cE21wfPrvd1B"
      },
      "source": [
        "def state_to_items(state, actor, ra_length, embeddings, dict_embeddings, target=False):\n",
        "  return [dict_embeddings[str(action)]\n",
        "          for action in actor.get_recommendation_list(ra_length, np.array(state).reshape(1, -1), embeddings, target).reshape(ra_length, embeddings.size())]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sr3TRcdWvdxi"
      },
      "source": [
        "def test_actor(actor, test_df, embeddings, dict_embeddings, ra_length, history_length, target=False, nb_rounds=1):\n",
        "  ratings = []\n",
        "  unknown = 0\n",
        "  random_seen = []\n",
        "  for _ in range(nb_rounds):\n",
        "    for i in range(len(test_df)):\n",
        "      history_sample = list(test_df[i].sample(history_length)['itemId'])\n",
        "      recommendation = state_to_items(embeddings.embed(history_sample), actor, ra_length, embeddings, dict_embeddings, target)\n",
        "      for item in recommendation:\n",
        "        l = list(test_df[i].loc[test_df[i]['itemId'] == item]['rating'])\n",
        "        assert(len(l) < 2)\n",
        "        if len(l) == 0:\n",
        "          unknown += 1\n",
        "        else:\n",
        "          ratings.append(l[0])\n",
        "      for item in history_sample:\n",
        "        random_seen.append(list(test_df[i].loc[test_df[i]['itemId'] == item]['rating'])[0])\n",
        "\n",
        "  return ratings, unknown, random_seen"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tLc9t6ME5I5O"
      },
      "source": [
        "### Test 1 - Trainset and target=False"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8yJVKrr7vduz",
        "outputId": "f212f8d8-17db-4602-949f-0490214bfdd5"
      },
      "source": [
        "ratings, unknown, random_seen = test_actor(actor, dg.train, embeddings, dict_embeddings, ra_length, history_length, target=False, nb_rounds=10)\n",
        "print('%0.1f%% unknown' % (100 * unknown / (len(ratings) + unknown)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "91.5% unknown\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 390
        },
        "id": "mAC1k2LnvdsB",
        "outputId": "d9a94c9d-205c-4aaa-a6ac-4fc5e0b7dd21"
      },
      "source": [
        "plt.figure(figsize=(12,6))\n",
        "plt.subplot(1, 2, 1)\n",
        "plt.hist(ratings)\n",
        "plt.title('Predictions ; Mean = %.4f' % (np.mean(ratings)))\n",
        "plt.subplot(1, 2, 2)\n",
        "plt.hist(random_seen)\n",
        "plt.title('Random ; Mean = %.4f' % (np.mean(random_seen)))\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x432 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dAujqOh-5O3h"
      },
      "source": [
        "### Test 2 - Trainset and target=True"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xBvnWS3tvdnr",
        "outputId": "79ecc940-f652-4fdb-f61e-53a43e4fd78d"
      },
      "source": [
        "ratings, unknown, random_seen = test_actor(actor, dg.train, embeddings, dict_embeddings, ra_length, history_length, target=True, nb_rounds=10)\n",
        "print('%0.1f%% unknown' % (100 * unknown / (len(ratings) + unknown)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "91.5% unknown\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 390
        },
        "id": "vvlWDAuLvyJw",
        "outputId": "5d2db4a1-212f-46cb-936e-080d22420b7f"
      },
      "source": [
        "plt.figure(figsize=(12,6))\n",
        "plt.subplot(1, 2, 1)\n",
        "plt.hist(ratings)\n",
        "plt.title('Predictions ; Mean = %.4f' % (np.mean(ratings)))\n",
        "plt.subplot(1, 2, 2)\n",
        "plt.hist(random_seen)\n",
        "plt.title('Random ; Mean = %.4f' % (np.mean(random_seen)))\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x432 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}