tutorial-BceGAN_GP-LHCb_RICH.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KcVGQpi5i5xG"
      },
      "source": [
        "# Parameterizing the LHCb RICH system using the pidgan `BceGAN_GP` algorithm\n",
        "\n",
        "**Author:** [mbarbetti](https://github.com/mbarbetti)\n",
        "\n",
        "**Date created:** 16/10/2023\n",
        "\n",
        "**Last modified:** 07/11/2023"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yzrL35_aMf4X"
      },
      "source": [
        "[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/mbarbetti/pidgan-notebooks/blob/main/tutorial-BceGAN_GP-LHCb_RICH.ipynb)\n",
        "[![Open In Kaggle](https://kaggle.com/static/images/open-in-kaggle.svg)](https://kaggle.com/kernels/welcome?src=https://github.com/mbarbetti/pidgan-notebooks/blob/main/tutorial-BceGAN_GP-LHCb_RICH.ipynb)\n",
        "[![Open in GitHub](https://img.shields.io/badge/Open_in_GitHub-blue?style=flat&logo=github&logoColor=white&labelColor=555)](https://github.com/mbarbetti/pidgan-notebooks/blob/main/tutorial-BceGAN_GP-LHCb_RICH.ipynb)\n",
        "\n",
        "**Description:** This tutorial demonstrates how to parameterize the high-level response of the LHCb RICH system using a Generative Adversarial Network (GAN) [[1][1]]. The code is written using the [pidgan](https://github.com/mbarbetti/pidgan) package that relies on TensorFlow and Keras as backends.\n",
        "\n",
        "[1]: https://arxiv.org/abs/1406.2661"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HvTVEAa7i5xJ"
      },
      "source": [
        "## Introduction"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "opu773xWi5xK"
      },
      "source": [
        "### What is LHCb?\n",
        "\n",
        "The [**LHCb experiment**](https://lhcb-outreach.web.cern.ch) has been originally designed to study rare decays of particles containing $b$ and $c$ quarks produced at the Large Hadron Collider (LHC). The LHCb detector, shown in the background of the following photo, is a single-arm forward spectrometer covering the pseudorapidity range of $2 < \\eta < 5$. The detector includes:\n",
        "\n",
        "- **Tracking system** - used for high-precision measurements of the momentum of charged particles and the position of the primary vertices\n",
        "- **Particle Identification (PID) system** - used to distiguish different species of traversing particles (i.e., muons, pions, kaons, protons)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5If4oPxki5xK"
      },
      "source": [
        "<div align=\"center\">\n",
        "  <img src=\"https://raw.githubusercontent.com/mbarbetti/pidgan-notebooks/main/.github/images/lhcb.jpeg\" width=\"800\"/>\n",
        "</div>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZTEAA_GpgXXM"
      },
      "source": [
        "#### The RICH detectors\n",
        "\n",
        "The LHCb PID system counts two ring-imaging Cherenkov ([RICH](https://lhcb-outreach.web.cern.ch/detector/rich-detectors/)) detectors whose response allows to separate different types of charged hadrons (e.g., pions, kaons, protons) using the [Cherenkov radiation](https://en.wikipedia.org/wiki/Cherenkov_radiation) of the traversing particles. The size of the circles produced by the Cherenkov radiation (shown on the left) within the RICH detector is related to the mass of the emitting particle. Reconstructing such circles allows to distinguish particles accordingly to their species (shown on the right)."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UvTtCmaBhqQ_"
      },
      "source": [
        "<div align=\"center\">\n",
        "  <img src=\"https://raw.githubusercontent.com/mbarbetti/pidgan-notebooks/main/.github/images/rich-response.png\" width=\"300\"/>\n",
        "  <img src=\"https://raw.githubusercontent.com/mbarbetti/pidgan-notebooks/main/.github/images/rich-performance.png\" width=\"400\"/>\n",
        "</div>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LPnfIVMoi5xK"
      },
      "source": [
        "### What are GANs?\n",
        "\n",
        "Generative Adversarial Networks [[1][1]] are a powerful class of _generative models_ based on the simultaneous training of two neural networks:\n",
        "\n",
        "*  **Discriminator network** ($D$) - trained by a classification task to separate the generator output from the reference dataset\n",
        "* **Generator network** ($G$) - trained by a simulation task to reproduce the reference dataset trying to fake the discriminator\n",
        "\n",
        "The goal is that $D$ optimally discriminates on the origin of the two samples, and simultaneously the training procedure for $G$ is to maximize the _probability_ of $D$ making a mistake. This framework corresponds to a **minimax two-player game** [[1][1]].\n",
        "\n",
        "Traditional GAN systems suffer from many issues, particularly during the training phase:\n",
        "\n",
        "* the generator may _collapse_ producing only a single sample or a small family of very similar samples (**mode collapse**)\n",
        "* the two players may _oscillate_ during training rather than converging to the [**Nash equilibrium**](https://en.wikipedia.org/wiki/Nash_equilibrium)\n",
        "* if _imbalance_ between the two players occurs, then the system is incapable of learning at all\n",
        "\n",
        "All these drawbacks result from the [**vanishing gradient problem**](https://en.wikipedia.org/wiki/Vanishing_gradient_problem), namely the lack of information for the update of the $G$ parameters. This is due to the saturation of the $D$ that is so good in distinguishing the origin of the two samples that no errors remain to the $G$ to improve the generated space [[2][2]]. To fix the problem, we need to avoid the discriminator saturation, for example constraining the $D$ response to be a **1-Lipschitz function** as suggested for the WGAN systems [[3][3]] in Ref. [[4][4]].\n",
        "\n",
        "[1]: https://arxiv.org/abs/1406.2661\n",
        "[2]: https://arxiv.org/abs/1701.04862\n",
        "[3]: https://arxiv.org/abs/1701.07875\n",
        "[4]: https://arxiv.org/abs/1704.00028"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "z_Hs8F8pi5xK"
      },
      "source": [
        "<div align=\"center\">\n",
        "  <img src=\"https://raw.githubusercontent.com/mbarbetti/pidgan-notebooks/main/.github/images/gan-scheme.png\" width=\"800\"/>\n",
        "</div>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fmUghh5Ii5xK"
      },
      "source": [
        "#### Mathematical details\n",
        "\n",
        "The generator $G(z)$, fed by elements $z$ sampled according to a known distribution $p_z$ (typically gaussian), maps the **latent space** $\\mathcal{Z}$ to the reference dataset $\\mathcal{X}$, inducing a distribution $p_\\rm{gen}$ trained to match with the target distribution $p_\\rm{ref}$. The discriminator $D(x)$ outputs a single scalar, readable as the **probability** that $x$ comes from the reference dataset rather than $G$. Hence, the optimization problem corresponds to train $D$ to maximize the probability of correct labelling, and simultaneously training $G$ to hinder the discriminator task.\n",
        "\n",
        "To define the discriminator **loss function** $\\mathcal{L}^{D}_{\\rm{BceGAN}}$ we can use the [binary cross-entropy](https://en.wikipedia.org/wiki/Cross-entropy) as suggested in Ref. [[5][5]]:\n",
        "\n",
        "<center>$\\mathcal{L}^{D}_{\\rm{BceGAN}} (\\theta_d, \\theta_g) = \\frac{1}{2} \\left ( {\\rm{BCE}}_{x \\sim p_\\rm{ref}} \\left ( \\mathbb{1}, D_{\\theta_d}(x) \\right ) + {\\rm{BCE}}_{z \\sim p_\\rm{z}} \\left ( \\mathbb{0}, D_{\\theta_d}(G_{\\theta_g}(z)) \\right ) \\right )$</center>\n",
        "\n",
        "where $\\mathbb{1}$ and $\\mathbb{0}$ are the labels for true and generated data respectively. The generator loss function $\\mathcal{L}^{G}_{\\rm{BceGAN}}$ follows a similar design:\n",
        "\n",
        "<center>$\\mathcal{L}^{G}_{\\rm{BceGAN}} (\\theta_d, \\theta_g) = {\\rm{BCE}}_{z \\sim p_\\rm{z}} \\left ( \\mathbb{1}, D_{\\theta_d}(G_{\\theta_g}(z)) \\right )$</center>\n",
        "\n",
        "Hence, the _minimax game_ of the **BceGAN-GP algorithm** can be written in this form:\n",
        "\n",
        "<center>$\\begin{cases}\n",
        "\\displaystyle{\\min_G} \\, \\mathcal{L}^{G}_{\\rm{BceGAN}} (\\theta_d, \\theta_g)\\\\\n",
        "\\displaystyle{\\min_D} \\, \\mathcal{L}^{D}_{\\rm{BceGAN}} (\\theta_d, \\theta_g) + \\lambda \\, \\mathbb{E}_{\\hat{x} \\sim p_{\\hat{x}}} \\left [ \\left ( \\left \\| \\nabla_{\\hat{x}} D_{\\theta_d} (\\hat{x}) \\right \\|_2 - 1 \\right )^2 \\right ]\n",
        "\\end{cases}$</center>\n",
        "\n",
        "where, to avoid _vanishing gradient issues_, the disccriminator is **implicitly** constrained to induce a 1-Lipschitz function through a regularization term called **gradient penalty** (GP), further described in Ref. [[4][4]].\n",
        "\n",
        "[4]: https://arxiv.org/abs/1704.00028\n",
        "[5]: https://arxiv.org/abs/1511.06434"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JnwMxPJUi5xK"
      },
      "source": [
        "#### Using input conditions\n",
        "\n",
        "Feeding the generator with additional information besides the latent space sample allows to **conditionate** its output. In particular, it's sufficient to concatenate the conditional features $x$ to the random noise $z$ passed as input to $G$ to make it able to take into account this information [[6][6]]: $y_{\\rm{gen}}(x) = G(x, z)$. Obviously, to preserve the capability to learn through the minimax game, also the discriminator must be fed by the additional features $x$, simply concatenating them to either elements of the reference sample or of the generated space [[6][6]]: $D(x, y)$ with $y \\in \\{y_{\\rm{ref}}, y_{\\rm{gen}}\\}$.\n",
        "\n",
        "[6]: https://arxiv.org/abs/1606.03498"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TvpIdCnSi5xL"
      },
      "source": [
        "## Tutorial\n",
        "\n",
        "The aim of this tutorial is to build a parameterization of the LHCb RICH detectors. Since we expect that the high-level response of the RICH system only depends on the kinematics of the traversing particles and the detector occupancy, the former can be parameterized training a **BceGAN-GP model** properly conditioned [[6][6], [7][7]]. The pidgan package provides several GAN [`algorithms`](https://github.com/mbarbetti/pidgan/tree/main/src/pidgan/algorithms) implementing some of the most used tricks presented in the literature to stablize their training [[3][3], [4][4], [8][8]]. The pidgan [`players`](https://github.com/mbarbetti/pidgan/tree/main/src/pidgan/players), namely the generator and discriminator networks, are designed to be trained taking conditions as input. In this notebook we will see how to build a model for the RICH system using the pidgan [`BceGAN_GP`](https://github.com/mbarbetti/pidgan/blob/main/src/pidgan/algorithms/BceGAN_GP.py) algorithm.\n",
        "\n",
        "[1]: https://arxiv.org/abs/1406.2661\n",
        "[2]: https://arxiv.org/abs/1701.04862\n",
        "[3]: https://arxiv.org/abs/1701.07875\n",
        "[4]: https://arxiv.org/abs/1704.00028\n",
        "[5]: https://arxiv.org/abs/1511.06434\n",
        "[6]: https://arxiv.org/abs/1411.1784\n",
        "[7]: https://arxiv.org/abs/1905.11825\n",
        "[8]: https://arxiv.org/abs/1606.03498"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4ohrh3dQi5xL"
      },
      "source": [
        "### Install\n",
        "\n",
        "The first step is to install the [pidgan](https://github.com/mbarbetti/pidgan) package and some other ones that are typically needed in machine learning applications to High Energy Physics (i.e., uproot, scipy, scikit-learn, matplotlib)."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "r8glZROOi5xL"
      },
      "outputs": [],
      "source": [
        "%%capture\n",
        "!pip install pidgan[hep]"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Kmb7g7HZi5xL"
      },
      "source": [
        "Now, let's verify the correct installation of pidgan printing its version:"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "id": "CWg67WFhi5xL",
        "outputId": "6d045368-6aef-4449-b11b-f335eb095476"
      },
      "outputs": [
        {
          "data": {
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            },
            "text/plain": [
              "'0.0.7'"
            ]
          },
          "execution_count": 2,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "import pidgan\n",
        "\n",
        "pidgan.__version__"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "C8lt6bjDi5xM"
      },
      "source": [
        "In case you're running this notebook on a machine equipped with a GPU, let's also verify the correct installation of TensorFlow:"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xGoxRYbGi5xM",
        "outputId": "c19fcecf-3089-4ee7-bf36-85e29c29a956"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "[PhysicalDevice(name='/physical_device:GPU:0', device_type='GPU')]"
            ]
          },
          "execution_count": 3,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "import tensorflow as tf\n",
        "\n",
        "tf.config.list_physical_devices(\"GPU\")  # outputs a non-empty list in case of GPU equipped"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Uslhx_PjvOwk"
      },
      "source": [
        "Finally, we just have to import all the modules that we will use in the following code cells:"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "id": "twM1ba1PvgVl"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "from datetime import datetime\n",
        "from tensorflow import keras\n",
        "from sklearn.utils import shuffle\n",
        "from sklearn.preprocessing import QuantileTransformer"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Cz9PhSN8wP2h"
      },
      "source": [
        "### Data loading\n",
        "\n",
        "The data needed for this tutorial is provided through a CSV file downloadable from the [pidgan-notebooks](https://github.com/mbarbetti/pidgan-notebooks) repository.\n",
        "\n",
        "**Disclaimer:** The data results from a toy model inspired to the LHCb RICH system combining its high-level response when traversed by a set of long-lived charged particles, namely muons, pions, kaons and protons.\n",
        "\n",
        "The CSV file contains 250000 instances, 4 input features and 4 output features. Let's import the data as a pandas [`DataFrame`](https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.html):"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 297
        },
        "id": "QfnPXl161n2f",
        "outputId": "a8e66e53-5579-412d-d06c-f57fb35298e7"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-5c3933cb-6aeb-4f51-966f-f9c52b1b35fc\" class=\"colab-df-container\">\n",
              "    <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>P</th>\n",
              "      <th>ETA</th>\n",
              "      <th>nTracks</th>\n",
              "      <th>trackcharge</th>\n",
              "      <th>RichDLLe</th>\n",
              "      <th>RichDLLmu</th>\n",
              "      <th>RichDLLk</th>\n",
              "      <th>RichDLLp</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>250000.000000</td>\n",
              "      <td>250000.000000</td>\n",
              "      <td>250000.000000</td>\n",
              "      <td>250000.000000</td>\n",
              "      <td>250000.000000</td>\n",
              "      <td>250000.000000</td>\n",
              "      <td>250000.000000</td>\n",
              "      <td>250000.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>19112.579988</td>\n",
              "      <td>3.362231</td>\n",
              "      <td>180.468471</td>\n",
              "      <td>0.021152</td>\n",
              "      <td>-8.959810</td>\n",
              "      <td>-1.853031</td>\n",
              "      <td>-3.614234</td>\n",
              "      <td>-3.626358</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>22610.408265</td>\n",
              "      <td>0.770202</td>\n",
              "      <td>91.626085</td>\n",
              "      <td>0.999778</td>\n",
              "      <td>16.090133</td>\n",
              "      <td>11.512878</td>\n",
              "      <td>28.799922</td>\n",
              "      <td>28.006328</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>3000.090800</td>\n",
              "      <td>1.800331</td>\n",
              "      <td>9.000999</td>\n",
              "      <td>-1.000000</td>\n",
              "      <td>-140.804340</td>\n",
              "      <td>-110.494200</td>\n",
              "      <td>-149.289830</td>\n",
              "      <td>-149.591670</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>5970.525000</td>\n",
              "      <td>2.749856</td>\n",
              "      <td>116.000000</td>\n",
              "      <td>-1.000000</td>\n",
              "      <td>-12.259203</td>\n",
              "      <td>-3.784543</td>\n",
              "      <td>-16.768035</td>\n",
              "      <td>-18.402633</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>10983.800000</td>\n",
              "      <td>3.340136</td>\n",
              "      <td>164.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>-3.499332</td>\n",
              "      <td>-0.634627</td>\n",
              "      <td>1.748902</td>\n",
              "      <td>1.705883</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>22392.768000</td>\n",
              "      <td>3.949773</td>\n",
              "      <td>228.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>-0.179516</td>\n",
              "      <td>0.651420</td>\n",
              "      <td>14.155344</td>\n",
              "      <td>15.179315</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>199837.420000</td>\n",
              "      <td>5.434650</td>\n",
              "      <td>739.993000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>58.021923</td>\n",
              "      <td>106.868935</td>\n",
              "      <td>120.245310</td>\n",
              "      <td>110.364006</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-5c3933cb-6aeb-4f51-966f-f9c52b1b35fc')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-5c3933cb-6aeb-4f51-966f-f9c52b1b35fc button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-5c3933cb-6aeb-4f51-966f-f9c52b1b35fc');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-da9bbffe-2f74-44ab-aa48-e459e3217508\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-da9bbffe-2f74-44ab-aa48-e459e3217508')\"\n",
              "            title=\"Suggest charts.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-da9bbffe-2f74-44ab-aa48-e459e3217508 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "text/plain": [
              "                   P            ETA        nTracks    trackcharge  \\\n",
              "count  250000.000000  250000.000000  250000.000000  250000.000000   \n",
              "mean    19112.579988       3.362231     180.468471       0.021152   \n",
              "std     22610.408265       0.770202      91.626085       0.999778   \n",
              "min      3000.090800       1.800331       9.000999      -1.000000   \n",
              "25%      5970.525000       2.749856     116.000000      -1.000000   \n",
              "50%     10983.800000       3.340136     164.000000       1.000000   \n",
              "75%     22392.768000       3.949773     228.000000       1.000000   \n",
              "max    199837.420000       5.434650     739.993000       1.000000   \n",
              "\n",
              "            RichDLLe      RichDLLmu       RichDLLk       RichDLLp  \n",
              "count  250000.000000  250000.000000  250000.000000  250000.000000  \n",
              "mean       -8.959810      -1.853031      -3.614234      -3.626358  \n",
              "std        16.090133      11.512878      28.799922      28.006328  \n",
              "min      -140.804340    -110.494200    -149.289830    -149.591670  \n",
              "25%       -12.259203      -3.784543     -16.768035     -18.402633  \n",
              "50%        -3.499332      -0.634627       1.748902       1.705883  \n",
              "75%        -0.179516       0.651420      14.155344      15.179315  \n",
              "max        58.021923     106.868935     120.245310     110.364006  "
            ]
          },
          "execution_count": 5,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "url = \"https://raw.githubusercontent.com/mbarbetti/pidgan-notebooks/main/data/pidgan-toy-data.csv\"\n",
        "data = pd.read_csv(url)\n",
        "\n",
        "data.describe()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rhaU0lWcZ-6L"
      },
      "source": [
        "We expect that the response of the RICH detectors only depends on the **kinematics** of the traversing particles and the **detector occupancy**. Hence, to build the aimed parameterization, we will conditionate the generator training using the following information:\n",
        "* `data.P` - momentum in GeV/$c$ of the traversing parrticles\n",
        "* `data.ETA` - pseudorapidity of the traversing particles\n",
        "* `data.nTracks` - total number of reconstructed tracks describing the detector occupancy\n",
        "* `data.trackcharge` - charge of the traversing particles (boolean value)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 262
        },
        "id": "j99DTVou5KdZ",
        "outputId": "4a8da4c1-b59a-4375-c568-157f3590ed20"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 2400x500 with 3 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.figure(figsize=(24, 5))\n",
        "\n",
        "plt.subplot(1, 3, 1)\n",
        "plt.xlabel(\"Momentum [GeV/$c$]\", fontsize=12)\n",
        "plt.ylabel(\"Candidates\", fontsize=12)\n",
        "plt.hist(data[\"P\"]/1e3, bins=np.linspace(0.0, 200.0, 101), color=\"#3288bd\", label=\"toy data\")\n",
        "plt.legend(loc=\"upper right\", fontsize=10)\n",
        "plt.yscale(\"log\")\n",
        "\n",
        "plt.subplot(1, 3, 2)\n",
        "plt.xlabel(\"Pseudorapidity\", fontsize=12)\n",
        "plt.ylabel(\"Candidates\", fontsize=12)\n",
        "plt.hist(data[\"ETA\"], bins=np.linspace(1.5, 5.5, 101), color=\"#3288bd\", label=\"toy data\")\n",
        "plt.legend(loc=\"upper right\", fontsize=10)\n",
        "\n",
        "plt.subplot(1, 3, 3)\n",
        "plt.xlabel(\"$\\mathtt{nTracks}$\", fontsize=12)\n",
        "plt.ylabel(\"Candidates\", fontsize=12)\n",
        "plt.hist(data[\"nTracks\"], bins=np.linspace(0.0, 750.0, 101), color=\"#3288bd\", label=\"toy data\")\n",
        "plt.legend(loc=\"upper right\", fontsize=10)\n",
        "\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yV9PdrDtb5gK"
      },
      "source": [
        "The high-level response of the RICH detectors describes how good the reconstructed circles match with a specific particle hypothesis in terms of **differential log-likelihood** (DLL). We aim to build a model able to reproduce the following output features:\n",
        "* `data.RichDLLe` - RICH DLL between the electron and pion tracks\n",
        "* `data.RichDLLmu` - RICH DLL between the muon and pion tracks\n",
        "* `data.RichDLLk` - RICH DLL between the kaon and pion tracks\n",
        "* `data.RichDLLp` - RICH DLL between the proton and pion tracks"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 689
        },
        "id": "2r8zWzbI7A2V",
        "outputId": "5259d7a6-ad76-40ef-ce0c-3475d0021f75"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1600x1000 with 4 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.figure(figsize=(16, 10))\n",
        "\n",
        "plt.subplot(2, 2, 1)\n",
        "plt.xlabel(\"RICH differential log-likelihood $(e - \\pi)$\", fontsize=12)\n",
        "plt.ylabel(\"Candidates\", fontsize=12)\n",
        "plt.hist(data[\"RichDLLe\"], bins=np.linspace(-150.0, 150.0, 101), color=\"#3288bd\", label=\"toy data\")\n",
        "plt.legend(loc=\"upper right\", fontsize=10)\n",
        "plt.yscale(\"log\")\n",
        "\n",
        "plt.subplot(2, 2, 2)\n",
        "plt.xlabel(\"RICH differential log-likelihood $(\\mu - \\pi)$\", fontsize=12)\n",
        "plt.ylabel(\"Candidates\", fontsize=12)\n",
        "plt.hist(data[\"RichDLLmu\"], bins=np.linspace(-150.0, 150.0, 101), color=\"#3288bd\", label=\"toy data\")\n",
        "plt.legend(loc=\"upper right\", fontsize=10)\n",
        "plt.yscale(\"log\")\n",
        "\n",
        "plt.subplot(2, 2, 3)\n",
        "plt.xlabel(\"RICH differential log-likelihood $(K - \\pi)$\", fontsize=12)\n",
        "plt.ylabel(\"Candidates\", fontsize=12)\n",
        "plt.hist(data[\"RichDLLk\"], bins=np.linspace(-150.0, 150.0, 101), color=\"#3288bd\", label=\"toy data\")\n",
        "plt.legend(loc=\"upper right\", fontsize=10)\n",
        "plt.yscale(\"log\")\n",
        "\n",
        "plt.subplot(2, 2, 4)\n",
        "plt.xlabel(\"RICH differential log-likelihood $(p - \\pi)$\", fontsize=12)\n",
        "plt.ylabel(\"Candidates\", fontsize=12)\n",
        "plt.hist(data[\"RichDLLp\"], bins=np.linspace(-150.0, 150.0, 101), color=\"#3288bd\", label=\"toy data\")\n",
        "plt.legend(loc=\"upper right\", fontsize=10)\n",
        "plt.yscale(\"log\")\n",
        "\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vDfCdUkIi5xM"
      },
      "source": [
        "### Data preprocessing\n",
        "\n",
        "As usual in a machine learning application, the data sample is split into **training** and **test set**."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xinGo4AJ9OFu",
        "outputId": "f2c1e8f9-59e4-4699-f3a8-1782dd437d5b"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "x_train shape: (150000, 4) \t x_test shape: (100000, 4)\n",
            "y_train shape: (150000, 4) \t y_test shape: (100000, 4)\n"
          ]
        }
      ],
      "source": [
        "data = shuffle(data)\n",
        "\n",
        "x_vars = [\"P\", \"ETA\", \"nTracks\", \"trackcharge\"]\n",
        "x = data[x_vars].values\n",
        "\n",
        "y_vars = [\"RichDLLe\", \"RichDLLmu\", \"RichDLLk\", \"RichDLLp\"]\n",
        "y = data[y_vars].values\n",
        "\n",
        "x_train, x_test = x[:150000], x[150000:]\n",
        "y_train, y_test = y[:150000], y[150000:]\n",
        "\n",
        "print(f\"x_train shape: {x_train.shape} \\t x_test shape: {x_test.shape}\")\n",
        "print(f\"y_train shape: {y_train.shape} \\t y_test shape: {y_test.shape}\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HcGuX9oZ_HHN"
      },
      "source": [
        "To normalize the different ranges and orders of magnitude of input $x$ and output $y$ features highlighted by the previous histograms, we will use the scikit-learn [`QuantileTransformer`](https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.QuantileTransformer.html) using a Gaussian as output distribution.\n",
        "\n",
        "Since one of the $x$ features is boolean (`data.trackcharge`), the latter should be kept unchanged. Let's preprocess the rest of the conditional information and verify the results with the [`DataFrame`](https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.html) `describe()` method:"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 297
        },
        "id": "BElR-YUR_0df",
        "outputId": "650c0639-c947-4db5-844a-64d9d728de5f"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-7f5fd6da-066a-4a7a-850d-71c6784e1403\" class=\"colab-df-container\">\n",
              "    <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>P</th>\n",
              "      <th>ETA</th>\n",
              "      <th>nTracks</th>\n",
              "      <th>trackcharge</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>150000.000000</td>\n",
              "      <td>150000.000000</td>\n",
              "      <td>150000.000000</td>\n",
              "      <td>150000.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>-0.009431</td>\n",
              "      <td>-0.006100</td>\n",
              "      <td>-0.004597</td>\n",
              "      <td>0.020347</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>1.008775</td>\n",
              "      <td>1.015461</td>\n",
              "      <td>1.010669</td>\n",
              "      <td>0.999796</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>-5.199338</td>\n",
              "      <td>-5.199338</td>\n",
              "      <td>-5.199338</td>\n",
              "      <td>-1.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>-0.678436</td>\n",
              "      <td>-0.681672</td>\n",
              "      <td>-0.667419</td>\n",
              "      <td>-1.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>-0.008836</td>\n",
              "      <td>-0.000121</td>\n",
              "      <td>-0.010037</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>0.674326</td>\n",
              "      <td>0.675394</td>\n",
              "      <td>0.686350</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>5.199338</td>\n",
              "      <td>5.199338</td>\n",
              "      <td>5.199338</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-7f5fd6da-066a-4a7a-850d-71c6784e1403')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-7f5fd6da-066a-4a7a-850d-71c6784e1403 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-7f5fd6da-066a-4a7a-850d-71c6784e1403');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-5e31f9b6-fbcb-491d-842b-027838875bbd\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-5e31f9b6-fbcb-491d-842b-027838875bbd')\"\n",
              "            title=\"Suggest charts.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-5e31f9b6-fbcb-491d-842b-027838875bbd button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "text/plain": [
              "                   P            ETA        nTracks    trackcharge\n",
              "count  150000.000000  150000.000000  150000.000000  150000.000000\n",
              "mean       -0.009431      -0.006100      -0.004597       0.020347\n",
              "std         1.008775       1.015461       1.010669       0.999796\n",
              "min        -5.199338      -5.199338      -5.199338      -1.000000\n",
              "25%        -0.678436      -0.681672      -0.667419      -1.000000\n",
              "50%        -0.008836      -0.000121      -0.010037       1.000000\n",
              "75%         0.674326       0.675394       0.686350       1.000000\n",
              "max         5.199338       5.199338       5.199338       1.000000"
            ]
          },
          "execution_count": 9,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "x_scaler = QuantileTransformer(output_distribution=\"normal\")\n",
        "x_train_prep = x_scaler.fit_transform(x_train[:, :3])\n",
        "x_train_prep = np.c_[x_train_prep, x_train[:, 3]]\n",
        "\n",
        "pd.DataFrame(x_train_prep, columns=x_vars).describe()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PY6OOsx6BhQP"
      },
      "source": [
        "Let's preprocess also the $y$ features checking what obtained:"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 297
        },
        "id": "VvfBZUjvB4oT",
        "outputId": "f3309c09-ee05-41e6-816a-5be3c82e6b16"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-ff5689dd-118b-41f4-bc2d-23c7a97b05f4\" class=\"colab-df-container\">\n",
              "    <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>RichDLLe</th>\n",
              "      <th>RichDLLmu</th>\n",
              "      <th>RichDLLk</th>\n",
              "      <th>RichDLLp</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>150000.000000</td>\n",
              "      <td>150000.000000</td>\n",
              "      <td>150000.000000</td>\n",
              "      <td>150000.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>0.022794</td>\n",
              "      <td>0.006636</td>\n",
              "      <td>-0.007148</td>\n",
              "      <td>-0.011786</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>1.003191</td>\n",
              "      <td>1.001094</td>\n",
              "      <td>0.991690</td>\n",
              "      <td>1.012406</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>-5.199338</td>\n",
              "      <td>-5.199338</td>\n",
              "      <td>-5.199338</td>\n",
              "      <td>-5.199338</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>-0.662363</td>\n",
              "      <td>-0.671916</td>\n",
              "      <td>-0.672839</td>\n",
              "      <td>-0.692312</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>0.038326</td>\n",
              "      <td>0.021147</td>\n",
              "      <td>-0.005824</td>\n",
              "      <td>-0.013393</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>0.688597</td>\n",
              "      <td>0.672258</td>\n",
              "      <td>0.666479</td>\n",
              "      <td>0.685960</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>5.199338</td>\n",
              "      <td>5.199338</td>\n",
              "      <td>5.199338</td>\n",
              "      <td>5.199338</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-ff5689dd-118b-41f4-bc2d-23c7a97b05f4')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-ff5689dd-118b-41f4-bc2d-23c7a97b05f4 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-ff5689dd-118b-41f4-bc2d-23c7a97b05f4');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-154bca55-4eeb-4c5e-81f4-897768c26918\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-154bca55-4eeb-4c5e-81f4-897768c26918')\"\n",
              "            title=\"Suggest charts.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-154bca55-4eeb-4c5e-81f4-897768c26918 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "text/plain": [
              "            RichDLLe      RichDLLmu       RichDLLk       RichDLLp\n",
              "count  150000.000000  150000.000000  150000.000000  150000.000000\n",
              "mean        0.022794       0.006636      -0.007148      -0.011786\n",
              "std         1.003191       1.001094       0.991690       1.012406\n",
              "min        -5.199338      -5.199338      -5.199338      -5.199338\n",
              "25%        -0.662363      -0.671916      -0.672839      -0.692312\n",
              "50%         0.038326       0.021147      -0.005824      -0.013393\n",
              "75%         0.688597       0.672258       0.666479       0.685960\n",
              "max         5.199338       5.199338       5.199338       5.199338"
            ]
          },
          "execution_count": 10,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "y_scaler = QuantileTransformer(output_distribution=\"normal\")\n",
        "y_train_prep = y_scaler.fit_transform(y_train)\n",
        "\n",
        "pd.DataFrame(y_train_prep, columns=y_vars).describe()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GVNTdWLPi5xM"
      },
      "source": [
        "### Model definition\n",
        "\n",
        "We are finally ready to define the **BceGAN-GP model** to be used to parameterize the RICH response. GAN algorithms rely on the competition (namely the minimax game) between the generator and discriminator networks. The latters are provided by the [pidgan](https://github.com/mbarbetti/pidgan) package through the [`players`](https://github.com/mbarbetti/pidgan/tree/main/src/pidgan/players) module that implements the two networks customizing the TensorFlow [`Model`](https://www.tensorflow.org/api_docs/python/tf/keras/Model) class."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gvx8Dkbz5o9Q"
      },
      "source": [
        "#### The generator\n",
        "\n",
        "The **generator player** can be implemented with the high-level [`Generator`](https://github.com/mbarbetti/pidgan/blob/main/src/pidgan/players/generators/Generator.py) class that provides the neural network model wrapping the TensorFlow [Sequential API](https://www.tensorflow.org/guide/keras/sequential_model). Before to initialize a `Generator` instance, let's have a look at its arguments:\n",
        "\n",
        "* **`output_dim`** (_int_) - Dimensionality of the generator output space.\n",
        "* **`latent_dim`** (_int_) - Dimensionality of the latent space.\n",
        "* **`num_hidden_layers`** (_int_, default=5) - Number of the hidden layers passed to the [`tf.keras.Sequential`](https://www.tensorflow.org/api_docs/python/tf/keras/Sequential) model. Each hidden layer has a LeakyReLU as activation function (see [`tf.keras.layers.LeakyReLU`](https://www.tensorflow.org/api_docs/python/tf/keras/layers/LeakyReLU)).\n",
        "* **`mlp_hidden_units`** (_int_ or _array_like_, default=128) - Number of neuron units per each hidden layer. It can also be passed as an array of unit numbers having length equal to the number of hidden layers.\n",
        "* **`mlp_dropout_rates`** (_float_ or _array_like_, default=0.0) - Fraction of the input neurons to drop per each hidden layer. It can also be passed as an array of float numbers having length equal to the number of hidden layers.\n",
        "* **`output_activation`** (_activation_like_ or _None_, default=None) - Activation function to use in the generator output layer (see [`tf.keras.activations`](https://www.tensorflow.org/api_docs/python/tf/keras/activations)). If None, no activation is applied (i.e., \"linear\" activation: $a(x) = x$).\n",
        "* **`name`** (_str_ of _None_, default=None) - The name of the generator model.\n",
        "* **`dtype`** (_dtype_ or _None_, default=None) - The dtype of the generator layer computations and weights."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "id": "0th7xpVKCGN0"
      },
      "outputs": [],
      "source": [
        "from pidgan.players.generators import Generator\n",
        "\n",
        "generator = Generator(\n",
        "    output_dim=y.shape[-1],\n",
        "    latent_dim=64,\n",
        "    num_hidden_layers=5,\n",
        "    mlp_hidden_units=128,\n",
        "    mlp_dropout_rates=0.0,\n",
        "    output_activation=None,\n",
        "    name=\"generator\"\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "luH3v1XVEBlc",
        "outputId": "76aa5245-2013-4bc0-c9f3-3e853661bb15"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model: \"generator_seq\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " dense_0 (Dense)             (None, 128)               8832      \n",
            "                                                                 \n",
            " leaky_relu_0 (LeakyReLU)    (None, 128)               0         \n",
            "                                                                 \n",
            " dropout_0 (Dropout)         (None, 128)               0         \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 128)               16512     \n",
            "                                                                 \n",
            " leaky_relu_1 (LeakyReLU)    (None, 128)               0         \n",
            "                                                                 \n",
            " dropout_1 (Dropout)         (None, 128)               0         \n",
            "                                                                 \n",
            " dense_2 (Dense)             (None, 128)               16512     \n",
            "                                                                 \n",
            " leaky_relu_2 (LeakyReLU)    (None, 128)               0         \n",
            "                                                                 \n",
            " dropout_2 (Dropout)         (None, 128)               0         \n",
            "                                                                 \n",
            " dense_3 (Dense)             (None, 128)               16512     \n",
            "                                                                 \n",
            " leaky_relu_3 (LeakyReLU)    (None, 128)               0         \n",
            "                                                                 \n",
            " dropout_3 (Dropout)         (None, 128)               0         \n",
            "                                                                 \n",
            " dense_4 (Dense)             (None, 128)               16512     \n",
            "                                                                 \n",
            " leaky_relu_4 (LeakyReLU)    (None, 128)               0         \n",
            "                                                                 \n",
            " dropout_4 (Dropout)         (None, 128)               0         \n",
            "                                                                 \n",
            " dense_out (Dense)           (None, 4)                 516       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 75396 (294.52 KB)\n",
            "Trainable params: 75396 (294.52 KB)\n",
            "Non-trainable params: 0 (0.00 Byte)\n",
            "_________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "_ = generator(keras.Input(shape=x.shape[1:]))\n",
        "generator.summary()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-MkE6GA14-hz"
      },
      "source": [
        "#### The discriminator\n",
        "\n",
        "The **discriminator player** can be implemented with the high-level [`Discriminator`](https://github.com/mbarbetti/pidgan/blob/main/src/pidgan/players/discriminators/Discriminator.py) class that provides the neural network model wrapping the TensorFlow [Sequential API](https://www.tensorflow.org/guide/keras/sequential_model). Again, before to initialize a `Discriminator` instance, let's have a look to its arguments:\n",
        "\n",
        "* **`output_dim`** (_int_) - Dimensionality of the discriminator output space.\n",
        "* **`num_hidden_layers`** (_int_, default=5) - Number of the hidden layers passed to the [`tf.keras.Sequential`](https://www.tensorflow.org/api_docs/python/tf/keras/Sequential) model. Each hidden layer has a LeakyReLU as activation function (see [`tf.keras.layers.LeakyReLU`](https://www.tensorflow.org/api_docs/python/tf/keras/layers/LeakyReLU)).\n",
        "* **`mlp_hidden_units`** (_int_ or _array_like_, default=128) - Number of neuron units per each hidden layer. It can also be passed as an array of unit numbers having length equal to the number of hidden layers.\n",
        "* **`mlp_dropout_rates`** (_float_ or _array_like_, default=0.0) - Fraction of the input neurons to drop per each hidden layer. It can also be passed as an array of float numbers having length equal to the number of hidden layers.\n",
        "* **`output_activation`** (_activation_like_ or _None_, default=\"sigmoid\") - Activation function to use in the dicriminator output layer (see [`tf.keras.activations`](https://www.tensorflow.org/api_docs/python/tf/keras/activations)). If None, no activation is applied (i.e., \"linear\" activation: $a(x) = x$).\n",
        "* **`name`** (_str_ of _None_, default=None) - The name of the discriminator model.\n",
        "* **`dtype`** (_dtype_ or _None_, default=None) - The dtype of the discriminator layer computations and weights."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "id": "ufj8Xi4fC_Ot"
      },
      "outputs": [],
      "source": [
        "from pidgan.players.discriminators import Discriminator\n",
        "\n",
        "discriminator = Discriminator(\n",
        "    output_dim=1,\n",
        "    num_hidden_layers=5,\n",
        "    mlp_hidden_units=128,\n",
        "    mlp_dropout_rates=0.0,\n",
        "    output_activation=None,\n",
        "    name=\"discriminator\"\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nfqi9DHvETlY",
        "outputId": "8f710b00-4ff0-454c-effe-47b6f5fb7424"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model: \"discriminator_seq\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " dense_0 (Dense)             (None, 128)               1152      \n",
            "                                                                 \n",
            " leaky_relu_0 (LeakyReLU)    (None, 128)               0         \n",
            "                                                                 \n",
            " dropout_0 (Dropout)         (None, 128)               0         \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 128)               16512     \n",
            "                                                                 \n",
            " leaky_relu_1 (LeakyReLU)    (None, 128)               0         \n",
            "                                                                 \n",
            " dropout_1 (Dropout)         (None, 128)               0         \n",
            "                                                                 \n",
            " dense_2 (Dense)             (None, 128)               16512     \n",
            "                                                                 \n",
            " leaky_relu_2 (LeakyReLU)    (None, 128)               0         \n",
            "                                                                 \n",
            " dropout_2 (Dropout)         (None, 128)               0         \n",
            "                                                                 \n",
            " dense_3 (Dense)             (None, 128)               16512     \n",
            "                                                                 \n",
            " leaky_relu_3 (LeakyReLU)    (None, 128)               0         \n",
            "                                                                 \n",
            " dropout_3 (Dropout)         (None, 128)               0         \n",
            "                                                                 \n",
            " dense_4 (Dense)             (None, 128)               16512     \n",
            "                                                                 \n",
            " leaky_relu_4 (LeakyReLU)    (None, 128)               0         \n",
            "                                                                 \n",
            " dropout_4 (Dropout)         (None, 128)               0         \n",
            "                                                                 \n",
            " dense_out (Dense)           (None, 1)                 129       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 67329 (263.00 KB)\n",
            "Trainable params: 67329 (263.00 KB)\n",
            "Non-trainable params: 0 (0.00 Byte)\n",
            "_________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "_ = discriminator((keras.Input(shape=x.shape[1:]), keras.Input(shape=y.shape[1:])))\n",
        "discriminator.summary()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IpQQWkPS5htN"
      },
      "source": [
        "#### The BceGAN-GP algorithm\n",
        "\n",
        "The **BceGAN-GP algorithm** [[4][4], [5][5]] can be implemented with the high-level [`BceGAN_GP`](https://github.com/mbarbetti/pidgan/blob/main/src/pidgan/algorithms/BceGAN_GP.py) class that defines the training procedure of the two players customizing the TensorFlow [`Model`](https://www.tensorflow.org/api_docs/python/tf/keras/Model) class and, in particular, its `fit()` method. Once again, before to initialize a `BceGAN_GP` instance, let's have a look at its arguments:\n",
        "\n",
        "* **`generator`** (_generator_like_) - The generator network. It must be passed as a pidgan [`Generator`](https://github.com/mbarbetti/pidgan/blob/main/src/pidgan/players/generators/Generator.py) instance.\n",
        "* **`discriminator`** (_discriminator_like_) - The discriminator network. It must be passed as a pidgan [`Discriminator`](https://github.com/mbarbetti/pidgan/blob/main/src/pidgan/players/discriminators/Discriminator.py) instance.\n",
        "* **`lipschitz_penalty`** (_float_, default=1.0) - Importance of the regularization term GP that implicitly constrains the discriminator to induce a 1-Lipschitz function as proposed in Ref. [[4][4]].\n",
        "* **`lipschitz_penalty_strategy`** (_str_, default=\"two-sided\") - If \"two-sided\" is passed, the norm of the discriminator gradient is forced to be equal to 1. When \"one-sided\" is preferred, only gradient norm values greater than 1 are penalized.\n",
        "* **`feature_matching_penalty`** (_float_, default=0.0) - Importance of the regularization term added to the generator training to force the hidden state of the discriminator to match between reference and generated examples. Its design follows from what proposed in Ref. [[8][8]].\n",
        "* **`referee`** (_classifier_like_ or _None_, default=None) - When passed it enables the training of a third independent player: the referee network. It can be used to assess the quality of the generator output during the training. If passed, it must be a pidgan [`Classifier`](https://github.com/mbarbetti/pidgan/blob/main/src/pidgan/players/classifiers/Classifier.py) instance.\n",
        "* **`name`** (_str_ or _None_, default=\"BceGAN_GP\") - The name of the BceGAN-GP model.\n",
        "* **`dtype`** (_dtype_ or _None_, default=None) - The dtype of the BceGAN-GP model computations.\n",
        "\n",
        "[4]: https://arxiv.org/abs/1704.00028\n",
        "[5]: https://arxiv.org/abs/1511.06434\n",
        "[8]: https://arxiv.org/abs/1606.03498"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qgW6y0Ru1vPp"
      },
      "source": [
        "To not further complicate this tutorial, we avoid to use the referee network in this case, even if its initialization is straightforward:\n",
        "\n",
        "```python\n",
        "from pidgan.players.classifiers import Classifier\n",
        "\n",
        "referee = Classifier(\n",
        "    num_hidden_layers=5,\n",
        "    mlp_hidden_units=128,\n",
        "    mlp_dropout_rates=0.0,\n",
        "    name=\"referee\"\n",
        ")\n",
        "```\n",
        "\n",
        "It's time to initialize a [`BceGAN_GP`](https://github.com/mbarbetti/pidgan/blob/main/src/pidgan/algorithms/BceGAN_GP.py) instance!"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "id": "SBEOKmeEDWC4"
      },
      "outputs": [],
      "source": [
        "from pidgan.algorithms import BceGAN_GP\n",
        "\n",
        "gan = BceGAN_GP(\n",
        "    generator=generator,\n",
        "    discriminator=discriminator,\n",
        "    lipschitz_penalty=1.0,\n",
        "    lipschitz_penalty_strategy=\"two-sided\",\n",
        "    feature_matching_penalty=0.0,\n",
        "    referee=None,\n",
        "    name=\"BceGAN_GP\",\n",
        ")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZMrE8bs952KB"
      },
      "source": [
        "#### Model compilation\n",
        "\n",
        "To finalize the [`BceGAN_GP`](https://github.com/mbarbetti/pidgan/blob/main/src/pidgan/algorithms/BceGAN_GP.py) model preparation before to run the training procedure, we need to execute the `compile()` method. Differently from the TensorFlow `compile()` method, the [pidgan](https://github.com/mbarbetti/pidgan) one doesn't require to pass the loss function since it is encoded in the customized `gan.fit()` method. The arguments of the `gan.compile()` method follows:\n",
        "\n",
        "* **`metrics`** (_list_ or _None_, default=None) - If not None, list of metrics to be evaluated by the BceGAN-GP model during training and testing. Each of this can be a string (name of a built-in pidgan function) or a [`pidgan.metrics`](https://github.com/mbarbetti/pidgan/tree/main/src/pidgan/metrics) instance.\n",
        "* **`generator_optimizer`** (_optimizer_like_, default=\"rmsprop\") - String (name of optimizer) or optimizer instance for the generator network (see [`tf.keras.optimizers`](https://www.tensorflow.org/api_docs/python/tf/keras/optimizers)).\n",
        "* **`discriminator_optimizer`** (_optimizer_like_, default=\"rmsprop\") - String (name of optimizer) or optimizer instance for the discriminator network. (see [`tf.keras.optimizers`](https://www.tensorflow.org/api_docs/python/tf/keras/optimizers)).\n",
        "* **`generator_upds_per_batch`** (_int_, default=1) - Number of the generator weights updates per batch of data.\n",
        "* **`discriminator_upds_per_batch`** (_int_, default=1) - Number of the discriminator weights updates per batch of data.\n",
        "* **`referee_optimizer`** (_optimizer_like_ or _None_, default=None) - If not None, string (name of optimizer) or optimizer instance for the referee network (see [`tf.keras.optimizers`](https://www.tensorflow.org/api_docs/python/tf/keras/optimizers)).\n",
        "* **`referee_upds_per_batch`** (_int_ or _None_, default=None) - If not None, number of the referee weights updates per batch of data."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "id": "kHL20AxZFNOq"
      },
      "outputs": [],
      "source": [
        "gan.compile(\n",
        "    metrics=[\"wass_dist\"],\n",
        "    generator_optimizer=keras.optimizers.RMSprop(learning_rate=0.002),\n",
        "    discriminator_optimizer=keras.optimizers.RMSprop(learning_rate=0.001),\n",
        "    generator_upds_per_batch=1,\n",
        "    discriminator_upds_per_batch=1,\n",
        "    referee_optimizer=None,\n",
        "    referee_upds_per_batch=None,\n",
        ")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0htLzsSri5xM"
      },
      "source": [
        "### Training procedure\n",
        "\n",
        "The [pidgan](https://github.com/mbarbetti/pidgan) package provides several learning rate scheduling strategies through the [`callbacks.schedulers`](https://github.com/mbarbetti/pidgan/tree/main/src/pidgan/callbacks/schedulers) module. The pidgan schedulers are designed as [custom TensorFlow callbacks](https://www.tensorflow.org/guide/keras/writing_your_own_callbacks) that also allow to trace the learning rate value of each passed optimizer during the training. In this tutorial, we will use an exponential decay schedule both for the generator and discriminator networks:\n",
        "\n",
        "<center>$\\eta(t) = \\eta_0 \\cdot e^{- \\alpha \\cdot t / \\tau}$</center>\n",
        "\n",
        "where $\\eta(t)$ indicates the learning rate value at a specific training step $t$, $\\eta_0$ the initial learning rate, $\\alpha$ the decay rate and $\\tau$ the decay step."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "metadata": {
        "id": "C0-PqvTDGd1J"
      },
      "outputs": [],
      "source": [
        "from pidgan.callbacks.schedulers import LearnRateExpDecay\n",
        "\n",
        "schedules = list()\n",
        "\n",
        "g_lr_sched = LearnRateExpDecay(\n",
        "    gan.generator_optimizer,\n",
        "    decay_rate=0.1,\n",
        "    decay_steps=50000,\n",
        "    min_learning_rate=1e-6,\n",
        "    verbose=True,\n",
        "    key=\"g_lr\",\n",
        ")\n",
        "schedules.append(g_lr_sched)\n",
        "\n",
        "d_lr_sched = LearnRateExpDecay(\n",
        "    gan.discriminator_optimizer,\n",
        "    decay_rate=0.1,\n",
        "    decay_steps=75000,\n",
        "    min_learning_rate=1e-6,\n",
        "    verbose=True,\n",
        "    key=\"d_lr\",\n",
        ")\n",
        "schedules.append(d_lr_sched)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0hDNGvV_FI0S"
      },
      "source": [
        "Calling `gan.fit()` allows to perform the **minimax two-player game**, training the generator to reproduce the reference space $y$ on the basis of the conditions $x$ thanks to the discriminator feedbacks. The arguments taken by the pidgan `fit()` method are the same ones traditionally passed to a TensorFlow [`Model`](https://www.tensorflow.org/api_docs/python/tf/keras/Model) instance. The schedulers are passed through the `callbacks` argument and, since their verbosity has been enabled (`verbose` and `key` passed), the learning rates of both the generator and discriminator are printed during the training procedure.\n",
        "\n",
        "**Note:** Training the [`BceGAN_GP`](https://github.com/mbarbetti/pidgan/blob/main/src/pidgan/algorithms/BceGAN_GP.py) model on Colab using the hyperparameters chosen for this tutorial (i.e., batch-size, epochs, updates per batch) takes about 20 minutes with no-GPU equipped or 10 minutes enabling the T4-powered runtime."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "N1mGZ3_FHMNg",
        "outputId": "7fe60156-4866-4440-cb70-370b77ed2a36"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/100\n",
            "  6/411 [..............................] - ETA: 8s - g_loss: 0.7421 - d_loss: 1.6122 - wass_dist: -0.8731 - g_lr: 0.0020 - d_lr: 9.9992e-04"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:Callback method `on_train_batch_begin` is slow compared to the batch time (batch time: 0.0146s vs `on_train_batch_begin` time: 0.2988s). Check your callbacks.\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "411/411 [==============================] - 28s 36ms/step - g_loss: 0.7772 - d_loss: 0.9192 - wass_dist: 0.0047 - g_lr: 0.0020 - d_lr: 9.9372e-04 - val_g_loss: 0.7377 - val_d_loss: 0.8342 - val_wass_dist: 0.0359\n",
            "Epoch 2/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.7578 - d_loss: 0.8048 - wass_dist: 0.0411 - g_lr: 0.0019 - d_lr: 9.8126e-04 - val_g_loss: 0.6651 - val_d_loss: 0.7653 - val_wass_dist: 0.0161\n",
            "Epoch 3/100\n",
            "411/411 [==============================] - 6s 16ms/step - g_loss: 0.7289 - d_loss: 0.7542 - wass_dist: 0.0089 - g_lr: 0.0019 - d_lr: 9.6895e-04 - val_g_loss: 0.6754 - val_d_loss: 0.7410 - val_wass_dist: 0.0094\n",
            "Epoch 4/100\n",
            "411/411 [==============================] - 6s 13ms/step - g_loss: 0.7139 - d_loss: 0.7548 - wass_dist: 0.0091 - g_lr: 0.0019 - d_lr: 9.5680e-04 - val_g_loss: 0.7542 - val_d_loss: 0.7503 - val_wass_dist: 0.0099\n",
            "Epoch 5/100\n",
            "411/411 [==============================] - 7s 17ms/step - g_loss: 0.7064 - d_loss: 0.7231 - wass_dist: 0.0163 - g_lr: 0.0018 - d_lr: 9.4481e-04 - val_g_loss: 0.7408 - val_d_loss: 0.7163 - val_wass_dist: 0.0151\n",
            "Epoch 6/100\n",
            "411/411 [==============================] - 6s 16ms/step - g_loss: 0.7048 - d_loss: 0.7160 - wass_dist: 0.0019 - g_lr: 0.0018 - d_lr: 9.3296e-04 - val_g_loss: 0.7142 - val_d_loss: 0.7218 - val_wass_dist: 0.0178\n",
            "Epoch 7/100\n",
            "411/411 [==============================] - 7s 17ms/step - g_loss: 0.7016 - d_loss: 0.7110 - wass_dist: 0.0086 - g_lr: 0.0018 - d_lr: 9.2126e-04 - val_g_loss: 0.6520 - val_d_loss: 0.7133 - val_wass_dist: -0.0044\n",
            "Epoch 8/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.7014 - d_loss: 0.7029 - wass_dist: -0.0070 - g_lr: 0.0017 - d_lr: 9.0971e-04 - val_g_loss: 0.7178 - val_d_loss: 0.7015 - val_wass_dist: 0.0024\n",
            "Epoch 9/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6996 - d_loss: 0.7011 - wass_dist: 0.0036 - g_lr: 0.0017 - d_lr: 8.9830e-04 - val_g_loss: 0.7264 - val_d_loss: 0.7003 - val_wass_dist: 0.0024\n",
            "Epoch 10/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6992 - d_loss: 0.6992 - wass_dist: 5.2065e-04 - g_lr: 0.0017 - d_lr: 8.8704e-04 - val_g_loss: 0.7001 - val_d_loss: 0.6986 - val_wass_dist: 0.0039\n",
            "Epoch 11/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6986 - d_loss: 0.6986 - wass_dist: 0.0071 - g_lr: 0.0016 - d_lr: 8.7592e-04 - val_g_loss: 0.6783 - val_d_loss: 0.7080 - val_wass_dist: 0.0058\n",
            "Epoch 12/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6983 - d_loss: 0.6981 - wass_dist: 0.0074 - g_lr: 0.0016 - d_lr: 8.6493e-04 - val_g_loss: 0.6786 - val_d_loss: 0.7022 - val_wass_dist: 0.0063\n",
            "Epoch 13/100\n",
            "411/411 [==============================] - 7s 17ms/step - g_loss: 0.6979 - d_loss: 0.6977 - wass_dist: 0.0094 - g_lr: 0.0016 - d_lr: 8.5409e-04 - val_g_loss: 0.6985 - val_d_loss: 0.7008 - val_wass_dist: 0.0137\n",
            "Epoch 14/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6973 - d_loss: 0.6974 - wass_dist: 5.0442e-04 - g_lr: 0.0015 - d_lr: 8.4338e-04 - val_g_loss: 0.6453 - val_d_loss: 0.7057 - val_wass_dist: 0.0122\n",
            "Epoch 15/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6972 - d_loss: 0.6973 - wass_dist: -0.0036 - g_lr: 0.0015 - d_lr: 8.3280e-04 - val_g_loss: 0.7130 - val_d_loss: 0.6992 - val_wass_dist: 0.0042\n",
            "Epoch 16/100\n",
            "411/411 [==============================] - 6s 16ms/step - g_loss: 0.6971 - d_loss: 0.6969 - wass_dist: -0.0030 - g_lr: 0.0015 - d_lr: 8.2236e-04 - val_g_loss: 0.6667 - val_d_loss: 0.6982 - val_wass_dist: 0.0020\n",
            "Epoch 17/100\n",
            "411/411 [==============================] - 7s 17ms/step - g_loss: 0.6967 - d_loss: 0.6969 - wass_dist: -0.0103 - g_lr: 0.0015 - d_lr: 8.1205e-04 - val_g_loss: 0.7107 - val_d_loss: 0.6962 - val_wass_dist: -0.0025\n",
            "Epoch 18/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6964 - d_loss: 0.6966 - wass_dist: -0.0012 - g_lr: 0.0014 - d_lr: 8.0187e-04 - val_g_loss: 0.7031 - val_d_loss: 0.7029 - val_wass_dist: 8.1235e-04\n",
            "Epoch 19/100\n",
            "411/411 [==============================] - 7s 17ms/step - g_loss: 0.6962 - d_loss: 0.6967 - wass_dist: 0.0243 - g_lr: 0.0014 - d_lr: 7.9181e-04 - val_g_loss: 0.6885 - val_d_loss: 0.7014 - val_wass_dist: 0.0238\n",
            "Epoch 20/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6964 - d_loss: 0.6965 - wass_dist: -0.0064 - g_lr: 0.0014 - d_lr: 7.8189e-04 - val_g_loss: 0.6883 - val_d_loss: 0.6965 - val_wass_dist: -6.7904e-04\n",
            "Epoch 21/100\n",
            "411/411 [==============================] - 7s 16ms/step - g_loss: 0.6963 - d_loss: 0.6964 - wass_dist: -0.0014 - g_lr: 0.0014 - d_lr: 7.7208e-04 - val_g_loss: 0.6884 - val_d_loss: 0.6981 - val_wass_dist: 0.0013\n",
            "Epoch 22/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6957 - d_loss: 0.6963 - wass_dist: -0.0097 - g_lr: 0.0013 - d_lr: 7.6240e-04 - val_g_loss: 0.7072 - val_d_loss: 0.6967 - val_wass_dist: 9.6772e-04\n",
            "Epoch 23/100\n",
            "411/411 [==============================] - 7s 17ms/step - g_loss: 0.6962 - d_loss: 0.6963 - wass_dist: 0.0031 - g_lr: 0.0013 - d_lr: 7.5284e-04 - val_g_loss: 0.6913 - val_d_loss: 0.6986 - val_wass_dist: 0.0087\n",
            "Epoch 24/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6959 - d_loss: 0.6963 - wass_dist: -0.0032 - g_lr: 0.0013 - d_lr: 7.4340e-04 - val_g_loss: 0.6836 - val_d_loss: 0.6975 - val_wass_dist: 5.9471e-04\n",
            "Epoch 25/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6955 - d_loss: 0.6964 - wass_dist: -0.0081 - g_lr: 0.0013 - d_lr: 7.3408e-04 - val_g_loss: 0.7144 - val_d_loss: 0.6974 - val_wass_dist: 0.0036\n",
            "Epoch 26/100\n",
            "411/411 [==============================] - 6s 13ms/step - g_loss: 0.6956 - d_loss: 0.6964 - wass_dist: 0.0063 - g_lr: 0.0012 - d_lr: 7.2487e-04 - val_g_loss: 0.7030 - val_d_loss: 0.6975 - val_wass_dist: 0.0037\n",
            "Epoch 27/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6960 - d_loss: 0.6963 - wass_dist: -0.0024 - g_lr: 0.0012 - d_lr: 7.1579e-04 - val_g_loss: 0.6907 - val_d_loss: 0.6977 - val_wass_dist: 3.8899e-04\n",
            "Epoch 28/100\n",
            "411/411 [==============================] - 5s 13ms/step - g_loss: 0.6955 - d_loss: 0.6962 - wass_dist: -0.0017 - g_lr: 0.0012 - d_lr: 7.0681e-04 - val_g_loss: 0.6789 - val_d_loss: 0.6984 - val_wass_dist: -7.3912e-05\n",
            "Epoch 29/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6957 - d_loss: 0.6960 - wass_dist: 0.0037 - g_lr: 0.0012 - d_lr: 6.9795e-04 - val_g_loss: 0.6966 - val_d_loss: 0.6975 - val_wass_dist: 0.0028\n",
            "Epoch 30/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6954 - d_loss: 0.6960 - wass_dist: -0.0058 - g_lr: 0.0011 - d_lr: 6.8920e-04 - val_g_loss: 0.6786 - val_d_loss: 0.6986 - val_wass_dist: 0.0028\n",
            "Epoch 31/100\n",
            "411/411 [==============================] - 7s 16ms/step - g_loss: 0.6952 - d_loss: 0.6958 - wass_dist: 0.0032 - g_lr: 0.0011 - d_lr: 6.8055e-04 - val_g_loss: 0.6971 - val_d_loss: 0.6954 - val_wass_dist: 3.0259e-04\n",
            "Epoch 32/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6952 - d_loss: 0.6958 - wass_dist: -0.0031 - g_lr: 0.0011 - d_lr: 6.7202e-04 - val_g_loss: 0.6821 - val_d_loss: 0.6963 - val_wass_dist: 0.0014\n",
            "Epoch 33/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6954 - d_loss: 0.6954 - wass_dist: -0.0018 - g_lr: 0.0011 - d_lr: 6.6359e-04 - val_g_loss: 0.7057 - val_d_loss: 0.6960 - val_wass_dist: 0.0047\n",
            "Epoch 34/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6952 - d_loss: 0.6954 - wass_dist: -0.0031 - g_lr: 0.0011 - d_lr: 6.5527e-04 - val_g_loss: 0.6846 - val_d_loss: 0.6948 - val_wass_dist: -2.1523e-04\n",
            "Epoch 35/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6953 - d_loss: 0.6954 - wass_dist: -0.0018 - g_lr: 0.0010 - d_lr: 6.4706e-04 - val_g_loss: 0.6956 - val_d_loss: 0.6955 - val_wass_dist: 0.0023\n",
            "Epoch 36/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6954 - d_loss: 0.6952 - wass_dist: 0.0016 - g_lr: 0.0010 - d_lr: 6.3894e-04 - val_g_loss: 0.7027 - val_d_loss: 0.6960 - val_wass_dist: 5.7101e-04\n",
            "Epoch 37/100\n",
            "411/411 [==============================] - 6s 13ms/step - g_loss: 0.6951 - d_loss: 0.6953 - wass_dist: -9.7611e-04 - g_lr: 0.0010 - d_lr: 6.3093e-04 - val_g_loss: 0.7019 - val_d_loss: 0.6945 - val_wass_dist: 0.0026\n",
            "Epoch 38/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6952 - d_loss: 0.6953 - wass_dist: -8.5395e-04 - g_lr: 9.8353e-04 - d_lr: 6.2302e-04 - val_g_loss: 0.6860 - val_d_loss: 0.6949 - val_wass_dist: 1.1399e-04\n",
            "Epoch 39/100\n",
            "411/411 [==============================] - 9s 22ms/step - g_loss: 0.6953 - d_loss: 0.6951 - wass_dist: -0.0024 - g_lr: 9.6509e-04 - d_lr: 6.1521e-04 - val_g_loss: 0.6940 - val_d_loss: 0.6954 - val_wass_dist: -5.1760e-04\n",
            "Epoch 40/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6952 - d_loss: 0.6952 - wass_dist: 0.0023 - g_lr: 9.4699e-04 - d_lr: 6.0750e-04 - val_g_loss: 0.7075 - val_d_loss: 0.6955 - val_wass_dist: 0.0034\n",
            "Epoch 41/100\n",
            "411/411 [==============================] - 6s 16ms/step - g_loss: 0.6948 - d_loss: 0.6951 - wass_dist: -0.0063 - g_lr: 9.2924e-04 - d_lr: 5.9988e-04 - val_g_loss: 0.6981 - val_d_loss: 0.6944 - val_wass_dist: 0.0023\n",
            "Epoch 42/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6950 - d_loss: 0.6951 - wass_dist: 0.0073 - g_lr: 9.1181e-04 - d_lr: 5.9236e-04 - val_g_loss: 0.6978 - val_d_loss: 0.6947 - val_wass_dist: -0.0029\n",
            "Epoch 43/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6949 - d_loss: 0.6951 - wass_dist: -2.8068e-04 - g_lr: 8.9472e-04 - d_lr: 5.8493e-04 - val_g_loss: 0.6854 - val_d_loss: 0.6961 - val_wass_dist: 0.0041\n",
            "Epoch 44/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6948 - d_loss: 0.6951 - wass_dist: -0.0060 - g_lr: 8.7794e-04 - d_lr: 5.7759e-04 - val_g_loss: 0.6862 - val_d_loss: 0.6952 - val_wass_dist: -7.6473e-04\n",
            "Epoch 45/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6952 - d_loss: 0.6950 - wass_dist: 0.0036 - g_lr: 8.6148e-04 - d_lr: 5.7035e-04 - val_g_loss: 0.6993 - val_d_loss: 0.6993 - val_wass_dist: 0.0016\n",
            "Epoch 46/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6950 - d_loss: 0.6949 - wass_dist: 0.0092 - g_lr: 8.4533e-04 - d_lr: 5.6320e-04 - val_g_loss: 0.6924 - val_d_loss: 0.6970 - val_wass_dist: -9.4081e-04\n",
            "Epoch 47/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6949 - d_loss: 0.6949 - wass_dist: 0.0028 - g_lr: 8.2948e-04 - d_lr: 5.5614e-04 - val_g_loss: 0.6881 - val_d_loss: 0.6957 - val_wass_dist: 0.0040\n",
            "Epoch 48/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6946 - d_loss: 0.6949 - wass_dist: 0.0014 - g_lr: 8.1393e-04 - d_lr: 5.4917e-04 - val_g_loss: 0.6876 - val_d_loss: 0.6951 - val_wass_dist: 4.2886e-04\n",
            "Epoch 49/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6948 - d_loss: 0.6949 - wass_dist: 2.9241e-04 - g_lr: 7.9867e-04 - d_lr: 5.4228e-04 - val_g_loss: 0.6913 - val_d_loss: 0.6976 - val_wass_dist: 8.4995e-04\n",
            "Epoch 50/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6949 - d_loss: 0.6947 - wass_dist: -0.0050 - g_lr: 7.8369e-04 - d_lr: 5.3548e-04 - val_g_loss: 0.7038 - val_d_loss: 0.7016 - val_wass_dist: 0.0024\n",
            "Epoch 51/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6946 - d_loss: 0.6947 - wass_dist: -0.0078 - g_lr: 7.6900e-04 - d_lr: 5.2877e-04 - val_g_loss: 0.6912 - val_d_loss: 0.6946 - val_wass_dist: 0.0031\n",
            "Epoch 52/100\n",
            "411/411 [==============================] - 7s 17ms/step - g_loss: 0.6946 - d_loss: 0.6947 - wass_dist: -0.0089 - g_lr: 7.5458e-04 - d_lr: 5.2214e-04 - val_g_loss: 0.6854 - val_d_loss: 0.6947 - val_wass_dist: -0.0011\n",
            "Epoch 53/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6944 - d_loss: 0.6947 - wass_dist: -0.0013 - g_lr: 7.4043e-04 - d_lr: 5.1559e-04 - val_g_loss: 0.7026 - val_d_loss: 0.6945 - val_wass_dist: -0.0034\n",
            "Epoch 54/100\n",
            "411/411 [==============================] - 7s 17ms/step - g_loss: 0.6946 - d_loss: 0.6947 - wass_dist: -6.5982e-04 - g_lr: 7.2655e-04 - d_lr: 5.0912e-04 - val_g_loss: 0.6848 - val_d_loss: 0.6972 - val_wass_dist: 0.0014\n",
            "Epoch 55/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6945 - d_loss: 0.6948 - wass_dist: -0.0016 - g_lr: 7.1293e-04 - d_lr: 5.0274e-04 - val_g_loss: 0.6918 - val_d_loss: 0.6965 - val_wass_dist: 0.0043\n",
            "Epoch 56/100\n",
            "411/411 [==============================] - 7s 17ms/step - g_loss: 0.6944 - d_loss: 0.6947 - wass_dist: 4.0736e-04 - g_lr: 6.9956e-04 - d_lr: 4.9644e-04 - val_g_loss: 0.6822 - val_d_loss: 0.6946 - val_wass_dist: 5.6405e-04\n",
            "Epoch 57/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6946 - d_loss: 0.6946 - wass_dist: -0.0014 - g_lr: 6.8645e-04 - d_lr: 4.9021e-04 - val_g_loss: 0.6958 - val_d_loss: 0.6962 - val_wass_dist: -0.0014\n",
            "Epoch 58/100\n",
            "411/411 [==============================] - 6s 16ms/step - g_loss: 0.6947 - d_loss: 0.6945 - wass_dist: -0.0013 - g_lr: 6.7358e-04 - d_lr: 4.8406e-04 - val_g_loss: 0.7003 - val_d_loss: 0.6949 - val_wass_dist: 0.0036\n",
            "Epoch 59/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6945 - d_loss: 0.6945 - wass_dist: -0.0054 - g_lr: 6.6095e-04 - d_lr: 4.7799e-04 - val_g_loss: 0.6879 - val_d_loss: 0.6944 - val_wass_dist: -0.0023\n",
            "Epoch 60/100\n",
            "411/411 [==============================] - 7s 17ms/step - g_loss: 0.6945 - d_loss: 0.6945 - wass_dist: 0.0017 - g_lr: 6.4855e-04 - d_lr: 4.7200e-04 - val_g_loss: 0.6899 - val_d_loss: 0.6973 - val_wass_dist: 0.0012\n",
            "Epoch 61/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6946 - d_loss: 0.6944 - wass_dist: -0.0022 - g_lr: 6.3639e-04 - d_lr: 4.6608e-04 - val_g_loss: 0.6904 - val_d_loss: 0.6946 - val_wass_dist: 2.7576e-04\n",
            "Epoch 62/100\n",
            "411/411 [==============================] - 7s 17ms/step - g_loss: 0.6945 - d_loss: 0.6946 - wass_dist: 0.0013 - g_lr: 6.2446e-04 - d_lr: 4.6024e-04 - val_g_loss: 0.7032 - val_d_loss: 0.6952 - val_wass_dist: 0.0026\n",
            "Epoch 63/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6944 - d_loss: 0.6945 - wass_dist: -0.0019 - g_lr: 6.1275e-04 - d_lr: 4.5447e-04 - val_g_loss: 0.6909 - val_d_loss: 0.6972 - val_wass_dist: 0.0020\n",
            "Epoch 64/100\n",
            "411/411 [==============================] - 7s 17ms/step - g_loss: 0.6944 - d_loss: 0.6945 - wass_dist: 9.2950e-04 - g_lr: 6.0127e-04 - d_lr: 4.4877e-04 - val_g_loss: 0.6903 - val_d_loss: 0.6944 - val_wass_dist: -0.0013\n",
            "Epoch 65/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6945 - d_loss: 0.6944 - wass_dist: 0.0027 - g_lr: 5.8999e-04 - d_lr: 4.4314e-04 - val_g_loss: 0.6930 - val_d_loss: 0.6949 - val_wass_dist: 0.0023\n",
            "Epoch 66/100\n",
            "411/411 [==============================] - 7s 17ms/step - g_loss: 0.6945 - d_loss: 0.6945 - wass_dist: -7.7023e-04 - g_lr: 5.7893e-04 - d_lr: 4.3759e-04 - val_g_loss: 0.6900 - val_d_loss: 0.6967 - val_wass_dist: 0.0037\n",
            "Epoch 67/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6946 - d_loss: 0.6944 - wass_dist: -0.0050 - g_lr: 5.6808e-04 - d_lr: 4.3210e-04 - val_g_loss: 0.6977 - val_d_loss: 0.6964 - val_wass_dist: 0.0017\n",
            "Epoch 68/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6943 - d_loss: 0.6945 - wass_dist: -5.6149e-04 - g_lr: 5.5743e-04 - d_lr: 4.2668e-04 - val_g_loss: 0.6857 - val_d_loss: 0.6948 - val_wass_dist: 2.9901e-04\n",
            "Epoch 69/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6944 - d_loss: 0.6944 - wass_dist: 0.0036 - g_lr: 5.4697e-04 - d_lr: 4.2133e-04 - val_g_loss: 0.6962 - val_d_loss: 0.6963 - val_wass_dist: 0.0012\n",
            "Epoch 70/100\n",
            "411/411 [==============================] - 7s 17ms/step - g_loss: 0.6945 - d_loss: 0.6944 - wass_dist: -0.0036 - g_lr: 5.3672e-04 - d_lr: 4.1605e-04 - val_g_loss: 0.7010 - val_d_loss: 0.6942 - val_wass_dist: 7.0266e-04\n",
            "Epoch 71/100\n",
            "411/411 [==============================] - 6s 16ms/step - g_loss: 0.6943 - d_loss: 0.6945 - wass_dist: 0.0024 - g_lr: 5.2666e-04 - d_lr: 4.1083e-04 - val_g_loss: 0.6993 - val_d_loss: 0.6950 - val_wass_dist: 0.0028\n",
            "Epoch 72/100\n",
            "411/411 [==============================] - 7s 16ms/step - g_loss: 0.6943 - d_loss: 0.6944 - wass_dist: -7.7450e-04 - g_lr: 5.1678e-04 - d_lr: 4.0568e-04 - val_g_loss: 0.6937 - val_d_loss: 0.6941 - val_wass_dist: 0.0024\n",
            "Epoch 73/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6943 - d_loss: 0.6943 - wass_dist: -0.0030 - g_lr: 5.0709e-04 - d_lr: 4.0059e-04 - val_g_loss: 0.6848 - val_d_loss: 0.6984 - val_wass_dist: 0.0033\n",
            "Epoch 74/100\n",
            "411/411 [==============================] - 6s 16ms/step - g_loss: 0.6945 - d_loss: 0.6943 - wass_dist: 0.0018 - g_lr: 4.9758e-04 - d_lr: 3.9557e-04 - val_g_loss: 0.6946 - val_d_loss: 0.6941 - val_wass_dist: -0.0013\n",
            "Epoch 75/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6942 - d_loss: 0.6943 - wass_dist: -7.9608e-04 - g_lr: 4.8825e-04 - d_lr: 3.9061e-04 - val_g_loss: 0.6946 - val_d_loss: 0.6944 - val_wass_dist: 9.0539e-04\n",
            "Epoch 76/100\n",
            "411/411 [==============================] - 6s 16ms/step - g_loss: 0.6943 - d_loss: 0.6942 - wass_dist: -8.5010e-04 - g_lr: 4.7910e-04 - d_lr: 3.8571e-04 - val_g_loss: 0.6986 - val_d_loss: 0.6949 - val_wass_dist: -3.3904e-05\n",
            "Epoch 77/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6942 - d_loss: 0.6944 - wass_dist: -7.6207e-04 - g_lr: 4.7012e-04 - d_lr: 3.8088e-04 - val_g_loss: 0.6965 - val_d_loss: 0.6952 - val_wass_dist: 4.2082e-05\n",
            "Epoch 78/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6943 - d_loss: 0.6943 - wass_dist: -0.0030 - g_lr: 4.6130e-04 - d_lr: 3.7610e-04 - val_g_loss: 0.6963 - val_d_loss: 0.6933 - val_wass_dist: 1.6889e-04\n",
            "Epoch 79/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6941 - d_loss: 0.6943 - wass_dist: -4.6970e-04 - g_lr: 4.5265e-04 - d_lr: 3.7138e-04 - val_g_loss: 0.6910 - val_d_loss: 0.6952 - val_wass_dist: 0.0034\n",
            "Epoch 80/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6944 - d_loss: 0.6942 - wass_dist: -0.0034 - g_lr: 4.4417e-04 - d_lr: 3.6673e-04 - val_g_loss: 0.6945 - val_d_loss: 0.6944 - val_wass_dist: -5.3706e-04\n",
            "Epoch 81/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6944 - d_loss: 0.6943 - wass_dist: 0.0031 - g_lr: 4.3584e-04 - d_lr: 3.6213e-04 - val_g_loss: 0.6936 - val_d_loss: 0.6966 - val_wass_dist: 9.7554e-04\n",
            "Epoch 82/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6940 - d_loss: 0.6942 - wass_dist: -0.0028 - g_lr: 4.2767e-04 - d_lr: 3.5759e-04 - val_g_loss: 0.6972 - val_d_loss: 0.6943 - val_wass_dist: 0.0022\n",
            "Epoch 83/100\n",
            "411/411 [==============================] - 7s 16ms/step - g_loss: 0.6942 - d_loss: 0.6941 - wass_dist: -0.0044 - g_lr: 4.1965e-04 - d_lr: 3.5310e-04 - val_g_loss: 0.6939 - val_d_loss: 0.6941 - val_wass_dist: -0.0013\n",
            "Epoch 84/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6942 - d_loss: 0.6941 - wass_dist: -0.0018 - g_lr: 4.1178e-04 - d_lr: 3.4868e-04 - val_g_loss: 0.6985 - val_d_loss: 0.6943 - val_wass_dist: 6.3465e-04\n",
            "Epoch 85/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6942 - d_loss: 0.6942 - wass_dist: -0.0052 - g_lr: 4.0406e-04 - d_lr: 3.4430e-04 - val_g_loss: 0.6937 - val_d_loss: 0.6940 - val_wass_dist: -0.0042\n",
            "Epoch 86/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6944 - d_loss: 0.6941 - wass_dist: 0.0023 - g_lr: 3.9648e-04 - d_lr: 3.3999e-04 - val_g_loss: 0.7039 - val_d_loss: 0.6953 - val_wass_dist: 0.0033\n",
            "Epoch 87/100\n",
            "411/411 [==============================] - 7s 17ms/step - g_loss: 0.6939 - d_loss: 0.6942 - wass_dist: 0.0019 - g_lr: 3.8905e-04 - d_lr: 3.3572e-04 - val_g_loss: 0.6859 - val_d_loss: 0.6949 - val_wass_dist: 0.0055\n",
            "Epoch 88/100\n",
            "411/411 [==============================] - 6s 16ms/step - g_loss: 0.6941 - d_loss: 0.6941 - wass_dist: 0.0030 - g_lr: 3.8176e-04 - d_lr: 3.3151e-04 - val_g_loss: 0.7025 - val_d_loss: 0.6941 - val_wass_dist: -0.0014\n",
            "Epoch 89/100\n",
            "411/411 [==============================] - 7s 17ms/step - g_loss: 0.6941 - d_loss: 0.6941 - wass_dist: 0.0012 - g_lr: 3.7460e-04 - d_lr: 3.2736e-04 - val_g_loss: 0.6896 - val_d_loss: 0.6939 - val_wass_dist: 0.0015\n",
            "Epoch 90/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6941 - d_loss: 0.6941 - wass_dist: 0.0022 - g_lr: 3.6758e-04 - d_lr: 3.2325e-04 - val_g_loss: 0.6908 - val_d_loss: 0.6950 - val_wass_dist: 0.0012\n",
            "Epoch 91/100\n",
            "411/411 [==============================] - 7s 17ms/step - g_loss: 0.6939 - d_loss: 0.6941 - wass_dist: -8.6893e-04 - g_lr: 3.6068e-04 - d_lr: 3.1920e-04 - val_g_loss: 0.6957 - val_d_loss: 0.6944 - val_wass_dist: -0.0013\n",
            "Epoch 92/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6940 - d_loss: 0.6940 - wass_dist: -0.0017 - g_lr: 3.5392e-04 - d_lr: 3.1520e-04 - val_g_loss: 0.6887 - val_d_loss: 0.6945 - val_wass_dist: 6.3896e-04\n",
            "Epoch 93/100\n",
            "411/411 [==============================] - 7s 17ms/step - g_loss: 0.6941 - d_loss: 0.6940 - wass_dist: 3.3929e-04 - g_lr: 3.4729e-04 - d_lr: 3.1125e-04 - val_g_loss: 0.7002 - val_d_loss: 0.6940 - val_wass_dist: 2.2042e-04\n",
            "Epoch 94/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6939 - d_loss: 0.6940 - wass_dist: -7.7633e-05 - g_lr: 3.4077e-04 - d_lr: 3.0734e-04 - val_g_loss: 0.6907 - val_d_loss: 0.6986 - val_wass_dist: 0.0020\n",
            "Epoch 95/100\n",
            "411/411 [==============================] - 6s 16ms/step - g_loss: 0.6941 - d_loss: 0.6940 - wass_dist: -0.0043 - g_lr: 3.3438e-04 - d_lr: 3.0349e-04 - val_g_loss: 0.6950 - val_d_loss: 0.6941 - val_wass_dist: 0.0014\n",
            "Epoch 96/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6940 - d_loss: 0.6940 - wass_dist: -0.0047 - g_lr: 3.2812e-04 - d_lr: 2.9968e-04 - val_g_loss: 0.6953 - val_d_loss: 0.6941 - val_wass_dist: 0.0026\n",
            "Epoch 97/100\n",
            "411/411 [==============================] - 7s 17ms/step - g_loss: 0.6941 - d_loss: 0.6940 - wass_dist: -0.0014 - g_lr: 3.2196e-04 - d_lr: 2.9593e-04 - val_g_loss: 0.6952 - val_d_loss: 0.6940 - val_wass_dist: 0.0012\n",
            "Epoch 98/100\n",
            "411/411 [==============================] - 6s 15ms/step - g_loss: 0.6938 - d_loss: 0.6940 - wass_dist: -0.0050 - g_lr: 3.1593e-04 - d_lr: 2.9222e-04 - val_g_loss: 0.7002 - val_d_loss: 0.6950 - val_wass_dist: 0.0013\n",
            "Epoch 99/100\n",
            "411/411 [==============================] - 6s 16ms/step - g_loss: 0.6939 - d_loss: 0.6940 - wass_dist: -0.0026 - g_lr: 3.1000e-04 - d_lr: 2.8855e-04 - val_g_loss: 0.6855 - val_d_loss: 0.6942 - val_wass_dist: -6.2330e-04\n",
            "Epoch 100/100\n",
            "411/411 [==============================] - 6s 14ms/step - g_loss: 0.6940 - d_loss: 0.6940 - wass_dist: -0.0017 - g_lr: 3.0419e-04 - d_lr: 2.8493e-04 - val_g_loss: 0.6917 - val_d_loss: 0.6938 - val_wass_dist: 6.0081e-04\n",
            "Training procedure completed in 0:11:26.684098\n"
          ]
        }
      ],
      "source": [
        "start = datetime.now()\n",
        "\n",
        "train = gan.fit(\n",
        "    x=x_train_prep,\n",
        "    y=y_train_prep,\n",
        "    batch_size=256,\n",
        "    epochs=100,\n",
        "    validation_split=0.3,\n",
        "    callbacks=schedules,\n",
        "    verbose=1,\n",
        ")\n",
        "\n",
        "stop = datetime.now()\n",
        "\n",
        "print(f\"Training procedure completed in {stop - start}\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GomKgX8QLeQM"
      },
      "source": [
        "The following code cells report the **learning** and **metric curves** for the training just finished."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Z97AGJog5-jN"
      },
      "source": [
        "#### Learning curves"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 393
        },
        "id": "-jdGN4aFIhdn",
        "outputId": "8484f446-9b1f-467f-8b58-c5a39c7a0530"
      },
      "outputs": [
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABSwAAAHbCAYAAADBM9RWAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAADd2UlEQVR4nOzdd3gU5drH8e/spkMKJQQCgdAJxVATBSkCSvFwwAaKShUsgAqigkoRX8GjiIhiQ6pIVawUpUjHAKGXhBY6JNRAEkhIdt4/MHvMIWACIbsbfp/r2gt2ZnbmnuyKd+69n+cxTNM0EREREREREREREXECFkcHICIiIiIiIiIiIpJJBUsRERERERERERFxGipYioiIiIiIiIiIiNNQwVJERERERERERESchgqWIiIiIiIiIiIi4jRUsBQRERERERERERGnoYKliIiIiIiIiIiIOA0VLEVERERERERERMRpqGApIiIiIiIiIiIiTkMFSxGRXGrWrBmGYTg6DBERERHJQ8rxnJcj35spU6ZgGAZTpky5pfNkdw/Lly/HMAyGDx9+S+cWKYhUsBSR2+LgwYMYhkHr1q0dHYqIiIiI5BHleK5r+PDhGIbB8uXLHR2KiMg/cnN0ACIirmbatGmkpKQ4OgwRERERyUPK8SS/RUREsHv3booXL+7oUEScjgqWIiK5VLZsWUeHICIiIiJ5TDme5DcfHx+qVavm6DBEnJKGhIuIU7h48SLDhg2jRo0aeHt7ExAQQKtWrVi9evU1x0ZHR9O3b19q1qyJv78/3t7e1KpVi/fee48rV65cc3xoaCihoaGcP3+evn37EhISgpubG1OmTLEPa+rWrRv79u3joYceokiRIhQqVIiWLVuydevWa86X3fwzf5/b5vfff6dhw4b4+PhQrFgxunbtypkzZ7K97y+//JIaNWrg5eVFSEgIr732GpcvX8YwDJo1a5bjn59pmkyePJnGjRsTEBCAj48PlStX5tlnn+Xw4cPX/Cyyk919/X3o0JQpU6hbty4+Pj40a9aMb775BsMwGDFiRLbn27RpE4Zh8OSTT2bZnpCQQP/+/alUqRKenp4UL16cRx55hB07dlxzjr1799K9e3fKly+Pp6cnRYsWJTw8nJdffhnTNHP88xERERHHUI538znejfIwgMTERP7zn//QtGlTgoOD8fDwIDg4mC5durB///5r7u3tt98G4L777sMwDAzDuCYvzE2edj2JiYkMHTqU6tWrU7hwYfz8/KhUqRJdu3bl0KFDWY7NaQ6b6cqVKwwfPpzQ0FA8PT2pUqUKn332WbZxmKbJpEmTaNSoEX5+fvj4+FC/fn0mTZqU7fFnz57lueeeIygoCB8fHxo0aMAPP/yQ7bE3mnvy75+9f3K982R+tpOSknjppZcIDg7G09OTu+66i++++y7bcx08eJBOnTpRtGhRChcuTNOmTVm5cqWmAhCXpQ5LEXG4s2fP0qRJE3bu3EmjRo147rnnuHDhAj/99BP33Xcfc+fOpUOHDvbjJ0yYwC+//EKTJk1o27YtKSkpLF++nMGDB7Nhwwa+//77a66RmppK8+bNSUpK4t///jdubm4EBQXZ9x88eJC7776bGjVq0KNHD/bv32+//u7du7MceyM///wz8+fPp127djRs2JCVK1cybdo09u/ff01iPnToUN555x2CgoLo1asX7u7uzJkzh5iYmFz9/Gw2G506deK7776jdOnSPPHEE/j5+XHw4EHmzJlDmzZtbrlj4IMPPuCPP/6gffv2PPDAA1itVh5++GGef/55vv32W4YOHXrNa7755hsAnn76afu2/fv306xZM44ePcoDDzxAhw4dSEhI4Pvvv+e3335j6dKlREZGAnD8+HEiIiJITk7mwQcfpFOnTiQnJ7N3714+++wzRo8ejZub/jcmIiLirJTj3VqOlym7PAxg9+7dDB06lPvuu4+HHnqIQoUKERMTw4wZM5g/fz6bNm2iXLlyAPbi2YoVK+jatau9UBkQEGC/Tm7ytOsxTZNWrVoRFRVFo0aNaN26NRaLhUOHDvHzzz/z9NNP22O6mRz2iSeeYP369bRp0war1cqcOXPo06cP7u7u9OrVK0scTz75JDNnzqRy5cp07twZDw8PFi9eTM+ePdm1axejR4+2H5+SkkKzZs3Yvn0799xzD02bNuXIkSN06tSJBx544Kbet1t15coVHnjgAc6dO8cjjzxCSkoKs2bNomPHjixatChLXMeOHaNhw4acOHGC1q1bU6dOHWJjY7n//vtp3ry5Q+IXuWWmiMhtEBcXZwJmq1at/vHYzp07m4A5YcKELNvj4+PNkJAQMzAw0Lx06ZJ9+6FDh8z09PQsx9psNrNHjx4mYK5evTrLvnLlytljSUlJyTZOwHzvvfey7HvrrbdMwBw1alSW7U2bNjX/95/PyZMnm4Dp5uaW5frp6elms2bNTMBct26dfXtsbKxptVrN0qVLm/Hx8fbtFy5cMKtXr24CZtOmTa/3I8vik08+MQGzRYsW19xfSkqKeebMmSw/i3LlymV7nuzua9iwYSZgFipUyNy2bds1r3nqqadMwIyKisqyPT093QwKCjJLliyZ5b1q2LChabVazUWLFmU5PjY21vT19TVr1apl3zZu3DgTMMeOHXvNdf9+TyIiIpJ/lONdlR853j/lYefPn882J1q2bJlpsVjMZ555Jtvz/fHHH9leLzd52vVs27bNBMwOHTpcs+/y5cvmxYsX7c9zk8NmvjeRkZFmYmKifXtMTIzp5uZmVq1aNcvrv/rqKxMwu3fvbqalpdm3p6ammu3atTMBc+PGjfbtmT+bXr16ZTnPokWL7J+jyZMn27f/8ccfJmAOGzbsmvvM/Ox17do1y/bsPl/XO0/mZ7t9+/ZmamqqffuSJUuy/e8vMyd/9913s2yfOHGiPf7rve8izkpDwkXEoU6fPs3s2bNp3rw5zzzzTJZ9JUqU4NVXX+XUqVMsWbLEvr1s2bL2b5YzGYZBnz59ALIc+3fvv/8+3t7e2e4rX748r776apZtPXv2BGDDhg05vp/OnTvTqFEj+3Or1UrXrl2vOc/MmTPJyMjglVdeoUSJEvbtvr6+vPXWWzm+HsBnn32G1Wrl888/v+b+vL29KVq0aK7Ol53evXtTq1ata7Zndk9Onz49y/bff/+d+Ph4Hn/8cft7tXnzZtauXUvXrl1p1apVluOrVKlCr1692L59+zVDjrJ7z/LinkREROT2UY536zlepuvlYf7+/tnmRPfddx81atS47s8rOzebp11Pdu+Hp6cnhQsXtj+/mRx21KhR+Pn52Z9XrVqVRo0aERsby8WLF+3bP/30UwoVKsT48eNxd3e3b/fw8ODdd98Frr5XmaZNm4aHh8c1Ux21atWKFi1a5Oieb4ePPvoIDw8P+/MWLVpQrly5LJ+51NRU5s6dS4kSJXjllVeyvL579+5UrVo13+IVyUsaSyciDrVhwwYyMjJITU3Ndg6YvXv3AhATE8O//vUvANLS0vj000+ZNWsWMTExJCUlZZnP8Pjx49ecx8vLK9tEL1Pt2rWxWLJ+h1OmTBkAzp8/n+P7qVev3jXbsjtP5rxJ99577zXH/z0Z/idJSUns3r2bSpUqUbly5Ry/LrciIiKy3d6iRQtKlSrFrFmzGDNmjH2IdmYB8+/Dwf/8808A4uPjs32vM4dJxcTEULNmTdq1a8fgwYPp06cPS5cupXXr1jRt2pQKFSrk5a2JiIjIbaAc79ZyvL+7Xh4GV+dAHDt2LFFRUZw+fZr09HT7vr8Xuv5JbvO06wkLC+Ouu+5i5syZHD16lA4dOtCsWbNr3oebzWH/6X3w9fUlJSWF7du3ExwczH/+859rjs+cDzXzni5cuEBcXBzVq1enZMmS1xzfuHFjli5dmuMY80pAQADly5e/ZnuZMmVYt26d/XlsbCypqanUr18fT0/PLMcahkHDhg2JjY297fGK5DUVLEXEoc6ePQvAmjVrWLNmzXWPS05Otv/90Ucf5ZdffqFKlSp06tSJEiVK4O7uzvnz5/n4449JTU295vUlSpS4ZhL1v/v7N7WZMotvGRkZOb6fnJ7nwoUL9rj+V07nUoKrk5oDlC5dOsevuRnXi8lqtdK5c2c+/PBDfvvtNx588EGSkpL48ccfqV69OnXr1rUfm/lez58/n/nz51/3WpnvdWhoKH/++SfDhw9nwYIFzJkzB4Bq1aoxYsQIHnvssby6PREREcljyvFuLcfLyevmzp1Lp06dKFy4MK1atSI0NBQfHx/7IkH/u8DNjeQ2T7seNzc3li1bxvDhw/n+++/tHX+BgYH07duXN998E6vVetM5bE7eh3PnzmGaJseOHbMvNHSje7nRewY3/77dKn9//2y3u7m5YbPZ7M+dNX6RW6WCpYg4VGbS8corr2SZ+Pp6NmzYwC+//EKrVq2YP39+lmFDf/75Jx9//HG2r7tRIusImfedkJBgn3g8U3x8fI7Pk5nIHDt2LEfHWywW0tLSst2XmThm50Y/v6effpoPP/yQ6dOn8+CDD/L999+TkpKSpbsS/nvPn3zyCX379s1RvDVr1uS7777jypUrREdHs3DhQsaNG0enTp0IDg6+6U4FERERub2U491ajvd317vH4cOH4+XlRXR09DVdirNmzcrVNW4mT7ueYsWK8cknnzBu3DhiYmJYtmwZn3zyCcOGDcPd3Z3BgwfnOofNjcx7qVevHhs3bszx8QkJCdnuz+59y+wW/XtHa6Yb5dS3w83EL+IKNIeliDhUgwYNMAwjy7CGG9m/fz8ADz744DVzHK1atSrP47tdwsPDAbLtOFi7dm2Oz1O4cGGqV69OXFycfWjVjRQpUoSEhIRrkqvM1bdvRnh4OLVq1eKnn37i4sWLTJ8+HcMwePLJJ7Mcl7mqZE7f679zd3fn7rvv5u2332bcuHGYpsmvv/56U/GKiIjI7acc79ZyvJzYv38/YWFh1xQrT5w4wYEDB645PvPnml1n6a3kaddjGAZhYWH06dOHxYsXA1dXW4fc57C54evrS1hYGLt3787RsH8/Pz/Kly/Pvn37OHny5DX7s/v8FSlSBMi+4Lp58+bcB30LqlatiqenJ9HR0dd0IZummafvqUh+UsFSRByqZMmSdOzYkbVr1/LBBx9kmacoU1RUFCkpKQD2b6pXr16d5ZidO3cyatSo2x9wHnn88cexWCx8+OGHnD592r49OTnZPhF4TvXp04eMjAxeeOEFLl26lGXf5cuX7UN84OovD1euXOHbb7+1bzNNk8GDB//jEJ8befrpp7l06RLjxo1j2bJlNG3alJCQkCzHREREEBkZycyZM5k9e/Y157DZbKxYscL+PDo62j7E5e8yvyX28vK66XhFRETk9lKOd+s53j8pV64c+/bty9JBd/nyZZ5//nn7PI1/l7mIzZEjR67Zl9s87XoOHjzIwYMHr9meXf6Wmxw2t1588UVSUlLo1atXtjluXFxcljiffvpp0tLSGDp0aJbjfv/992znr6xatSq+vr78/PPPWeKMj4/n//7v/2467pvh6enJo48+Snx8PGPHjs2yb9q0afa5OkVcjYaEi8httX37drp165btvmrVqjFo0CA+++wzYmNjee211/jmm2+45557CAgI4MiRI2zcuJG9e/dy4sQJfHx8iIiIICIigjlz5nDixAnuvvtuDh8+zM8//8yDDz7Id999l783eJOqVq3KoEGDGDlyJLVq1aJjx464ubkxb948atWqxY4dO66ZIP56nn/+eVasWMGcOXOoXLky//73v/Hz8+Pw4cP89ttvTJw4kQ4dOgDQt29fJk+ezDPPPMPixYsJDAxk1apVnD9/nvDwcPtE8bnVuXNnBg0axNtvv43NZrtmOHimmTNnct999/H4448zduxY6tati7e3N4cPH2bdunWcOnWKy5cvA/DNN9/w5Zdf0qRJEypWrIifnx+7du1iwYIFFC1alO7du99UrCIiInLrlONlLy9zvH/Sr18/+vXrR506dXj00UdJT09n8eLFmKaZbV533333YRgGb7zxBjt37sTf35+AgAD7EPDc5GnXs2XLFh5++GEiIiLsi9gcO3aMH3/8EYvFQv/+/e3H5iaHza1nn32WP//8k6lTp7JmzRpatmxJcHAw8fHxxMTEEBUVxYwZMwgNDQXgtddeY968eUyYMIGdO3fSpEkTjhw5wpw5c3jwwQevmdfTw8ODfv36MXLkSOrWrUv79u25ePEiv/zyC02bNrV3DOeXUaNGsWTJEgYNGsSKFSuoU6cOsbGx/Prrr7Ru3ZpFixbl2edOJN+YIiK3QVxcnAnc8NG0aVP78SkpKeb7779v1qtXzyxUqJDp7e1tli9f3uzQoYM5bdo088qVK/ZjExISzB49epjBwcGml5eXWatWLXP8+PHmgQMHTMDs2rVrlljKlStnlitX7oZx/u9rMv1vnKZpmk2bNjX/95/PyZMnm4A5efLka87xxx9/mIA5bNiwa/Z99tlnZlhYmOnh4WGWKVPGHDhwoHnkyBETMNu3b59tTNmx2Wzm119/bd59991moUKFTB8fH7Ny5crmc889Zx4+fDjLscuWLTMjIyNNT09Ps1ixYubTTz9txsfHZ3tfw4YNMwHzjz/++McYWrZsaQKml5eXmZiYeN3jzp49a7711ltmzZo1TW9vb7Nw4cJm5cqVzc6dO5vz5s2zH/fnn3+azz77rFmzZk0zICDA9Pb2NitXrmz27dvXPHToUI5/NiIiIpJ3lOP91+3O8f4pD7PZbOYXX3xh1qhRw/Ty8jJLlixp9uzZ00xISMj2XkzTNKdMmWLWqlXL9PT0NIFrfn45zdOu58iRI+agQYPMu+++2yxRooTp4eFhli1b1nz44YfNdevWZXsPOclhr3c/pmmaXbt2NQEzLi7umn2zZ882W7ZsaRYpUsR0d3c3S5cubTZr1sz88MMPzVOnTmU59syZM2bv3r3NwMBA08vLy6xXr545b968634GMjIyzOHDh5shISGmh4eHWaVKFfPjjz++7uc1u3u43mfoRp/t6/0sDhw4YD722GOmv7+/6ePjYzZu3NhcsWKF2bdvXxMwN2/enO35RJyVYZrZ9OaLiIjDLFmyhPvvv5/XXnuN//znP44OR0RERETygHI8cYR7772XdevWkZiYSOHChR0djkiOqSdYRMRBTp06dc2k5+fPn2fw4MEANz0ERkREREQcRzmeOMKJEyeu2TZ9+nT7kHgVK8XVaA5LEREH+fbbbxk9ejTNmzcnODiYEydOsGjRIhISEujWrRv33HOPo0MUERERkVxSjieOULNmTerUqUP16tWxWq1s2bKF5cuX4+vry+jRox0dnkiuqWApIuIgDRs2pF69eixZsoSzZ89itVoJCwtjyJAhvPDCC44OT0RERERugnI8cYTnnnuOX375hY0bN5KcnExgYCCdO3dmyJAhVKtWzdHhieSa5rAUERERERERERERp6E5LEVERERERERERMRpqGApIiIiIiIiIiIiTkNzWOaAzWbj+PHj+Pr6YhiGo8MRERERyTXTNLl48SLBwcFYLPrO2tUoHxURERFXl5t8VAXLHDh+/DghISGODkNERETklh05coQyZco4OgzJJeWjIiIiUlDkJB9VwTIHfH19gas/UD8/PwdHIyIiIpJ7Fy5cICQkxJ7XiGtRPioiIiKuLjf5qAqWOZA57MbPz08JooiIiLg0DSd2TcpHRUREpKDIST6qCYxERERERERERETEaahgKSIiIiIiIiIiIk5DBUsRERERERERERFxGprDUkRExImZpkl6ejoZGRmODkWcnNVqxc3NTXNUioiISJ5SPio5lZf5qAqWIiIiTiotLY0TJ06QkpLi6FDERfj4+FCqVCk8PDwcHYqIiIgUAMpHJbfyKh9VwVJERMQJ2Ww24uLisFqtBAcH4+Hhoc45uS7TNElLS+PUqVPExcVRuXJlLBbN/OOszp8/T8uWLUlPTyc9PZ2XXnqJXr16OTosERGRLJSPSm7kdT6qgqWIiIgTSktLw2azERISgo+Pj6PDERfg7e2Nu7s7hw4dIi0tDS8vL0eHJNfh6+vLypUr8fHxITk5mZo1a/Lwww9TrFgxR4cmIiJip3xUcisv81F99S4iIuLE1CUnuaHPi2uwWq32X/xSU1MxTRPTNB0clYiISPaUX0hu5NXnRZ86EREREXEJo0aNokGDBvj6+lKiRAk6dOhAbGxsnl5j5cqVtGvXjuDgYAzD4Mcff8z2uPHjxxMaGoqXlxeRkZGsX78+V9c5f/484eHhlClThldffZXixYvnQfQiIiIiBYMKliIiIiLiElasWEGfPn34888/Wbx4MVeuXOGBBx4gOTk52+PXrFnDlStXrtm+a9cu4uPjs31NcnIy4eHhjB8//rpxzJ49mwEDBjBs2DA2bdpEeHg4rVq1IiEhwX5M7dq1qVmz5jWP48ePAxAQEMDWrVuJi4tjxowZ141HRERE5E6kOSxFRERExCUsWrQoy/MpU6ZQokQJoqOjadKkSZZ9NpuNPn36ULlyZWbNmoXVagUgNjaW5s2bM2DAAF577bVrrtGmTRvatGlzwzjGjBlDr1696N69OwBffPEF8+fPZ9KkSQwaNAiALVu25OiegoKCCA8PZ9WqVTz66KM5eo2IiIhIQacOSxEREZG/adasGS+//LKjw5AcSExMBKBo0aLX7LNYLCxYsIDNmzfTpUsXbDYb+/fvp3nz5nTo0CHbYmVOpKWlER0dTcuWLbNcq2XLlqxbty5H54iPj+fixYv2e1i5ciVVq1bN9tjx48dTvXp1GjRocFPxioiIiGu603NSFSxFRETkjpDd0ODbKS0tLV+vd6ex2Wy8/PLLNGrUiJo1a2Z7THBwMMuWLWP16tV07tyZ5s2b07JlSz7//PObvu7p06fJyMggKCgoy/agoCBOnjyZo3McOnSIxo0bEx4eTuPGjenXrx+1atXK9tg+ffqwa9cuNmzYcNMx50ZSaroWABIREbmNlJPmjAqWTsLc8ycZ017H9sdUR4ciIiJOyjRNLqVl5Psjt8WLixcv8uSTT1KoUCFKlSrFRx99lOUb4tTUVAYOHEjp0qUpVKgQkZGRLF++3P76KVOmEBAQwG+//UZYWBiFCxemdevWnDhxIst1vv76a8LCwvDy8qJatWp89tln9n0HDx7EMAxmz55N06ZN8fLy4ttvv+XMmTM88cQTlC5dGh8fH2rVqsXMmTPtr+vWrRsrVqzg448/xjAMDMPg4MGDwNX5EyMiIvD09KRUqVIMGjSI9PR0+2ubNWtG3759efnllylevDitWrXK1c9NcqdPnz7s2LGDWbNm3fC4smXL8s033zB79mzc3NyYOHEihmHkU5TZi4iIYMuWLWzdupVt27bx7LPPOjSeTDbT5OXvtvLSd9tIuJjq6HBERMQJOSofVU565+WkmsPSSZipl+D0EUw/rRApIiLZu3zFxr0frcj3667u3xRvD2uOjx8wYABr1qzh559/JigoiKFDh7Jp0yZq164NQN++fdm1axezZs0iODiYH374gdatW7N9+3YqV64MQEpKCqNHj+abb77BYrHw1FNPMXDgQL799lsAvv32W4YOHcqnn35KnTp12Lx5M7169aJQoUJ07drVHsugQYP48MMPqVOnDl5eXly+fJl69erx+uuv4+fnx/z583n66aepWLEiERERfPzxx+zZs4eaNWsyYsQIAAIDAzl27Bht27alW7duTJs2jZiYGHr16oWXlxfDhw+3X2/q1Kk8//zzrFmz5hZ/6nIjffv25ddff2XlypWUKVPmhsfGx8fTu3dv2rVrx4YNG+jfvz+ffPLJTV+7ePHiWK3WaxbJiY+Pp2TJkjd9XmcQG3+RHScucCXD5NGv/2RA88q0v6uUwwu8IiLiPByVj4Jy0jstJ1XB0llY/3orMtJvfJyIiIgTu3jxIlOnTmXGjBm0aNECgMmTJxMcHAzA4cOHmTx5MocPH7ZvGzhwIIsWLWLy5MmMHDkSuDpU5osvvqBixYrA1YQyM1kDGDZsGB9++CEPP/wwAOXLl2fXrl18+eWXWZLDl19+2X5MpoEDB9r/3q9fP3777TfmzJlDREQE/v7+eHh44OPjk6X49NlnnxESEsKnn36KYRhUq1aN48eP8/rrrzN06FAslquDVipXrsz777+fNz9MuYZpmvTr148ffviB5cuXU758+Rsef/r0aVq0aEFYWBhz585lz549NGvWDE9PT0aPHn1TMXh4eFCvXj2WLl1Khw4dgKvD05cuXUrfvn1v6pzOIqykHzO6RfD2gt3sOHGBdxbFsDgmgbfaVKOUn5ejwxMREckx5aSun5OqYOkkDDd3TID0/J3LQEREXIeXu4XV/Zs65Lo5deDAAa5cuUJERIR9m7+/v31Bke3bt5ORkUGVKlWyvC41NZVixYrZn/v4+NgTQ4BSpUqRkJAAQHJyMvv376dnz5706tXLfkx6ejr+/v5Zzlu/fv0szzMyMhg5ciRz5szh2LFjpKWlkZqaio+Pzw3va/fu3dxzzz1ZOs0aNWpEUlISR48epWzZsgDUq1fvhueRW9OnTx9mzJjBTz/9hK+vr33OSH9/f7y9vbMca7PZaNOmDeXKlbMPB69evTqLFy+mefPmlC5dmv79+19zjaSkJPbt22d/HhcXx5YtWyhatKj9fR4wYABdu3alfv36REREMHbsWJKTk+2rhruyCsULMempeny74QifrzrAnwfP0mliFC/dV4mHw4PVbSkicodzVD6aee2cUk7q+jmpCpbOwup+9c8MFSxFRCR7hmHkahiMM0pKSsJqtRIdHY3VmvVeChcubP+7u7t7ln2GYdjnLUpKSgJgwoQJREZGZjnuf89ZqFChLM8/+OADPv74Y8aOHUutWrUoVKgQL7/8cp5NRv6/15O8lblYTrNmzbJsnzx5Mt26dcuyzWKxMHLkSBo3boyHh4d9e3h4OEuWLCEwMDDba2zcuJH77rvP/nzAgAEAdO3alSlTpgDQqVMnTp06xdChQzl58iS1a9dm0aJF1yzE46qsFoMukWVpUqkYIxbGsPVYIiN/i2VJTAJD2lQj2N/7n08iIiIFUkHIR0E5qStQwdJZZBYs1WEpIiIurEKFCri7u7Nhwwb7N7yJiYns2bOHJk2aUKdOHTIyMkhISKBx48Y3dY2goCCCg4M5cOAATz75ZK5eu2bNGtq3b89TTz0FXO3C27NnD9WrV7cf4+HhQUZGRpbXhYWF8f3332Oapv0b7TVr1uDr6/uPcyhK3sntZPv3339/ttvr1Klz3dc0a9YsR9fp27evyw8B/yehxQoxoXNdZkcf5dOV+1l/6BwdJ67nxWYVebROaSzqthQRESelnNT1aZVwZ6EOSxERKQB8fX3p2rUrr776Kn/88Qc7d+6kZ8+eWCwWDMOgSpUqPPnkk3Tp0oV58+YRFxfH+vXrGTVqFPPnz8/xdd5++21GjRrFuHHj2LNnD9u3b2fy5MmMGTPmhq+rXLkyixcvZu3atezevZtnn332msVTQkNDiYqK4uDBg5w+fRqbzcYLL7zAkSNH6NevHzExMfz0008MGzaMAQMG2OcKEimIrBaDzg1CmN0jgrohAVy6ksF/Fu/huZmbOXIuxdHhiYiIZEs5qesrWHfjytxUsBQRkYJhzJgx3HPPPfzrX/+iZcuWNGrUiLCwMLy8ri7aMXnyZLp06cIrr7xC1apV6dChQ5Zvv3PimWee4euvv2by5MnUqlWLpk2bMmXKlH9chOWtt96ibt26tGrVimbNmlGyZEn7wimZBg4ciNVqpXr16gQGBnL48GFKly7NggULWL9+PeHh4Tz33HP07NmTt956K9c/HxFXFFLEhy+fqMNrLavg7W4l+sh5Ok1az4yNR7DlsvNVREQkPygndW2GmduxNXegCxcu4O/vT2JiIn5+frflGmbCQWzT34BCAVif/ey2XENERFzH5cuXiYuLo3z58vakylUlJydTunRpPvzwQ3r27OnocAq0G31u8iOfkdvHmd6/Y+cv8c6iGDYcOgdAeGl/hrUNo1zRGy8UICIirqUg5aOgnDS/5FU+qg5LZ6E5LEVEpIDYvHkzM2fOZP/+/WzatMk+p0/79u0dHJmI5IXSAd581qk2gx+oio+Hla3HEnli8nqmRR0mw6ZeCBERcQ7KSV2bCpbOQkPCRUSkABk9ejTh4eG0bNmS5ORkVq1aRfHixR0dlojkEYth8Gid0szpEUFkaBFS0218vHwfPaZHE3c62dHhiYiIAMpJXZlWCXcWf1t05++rPYmIiLiaOnXqEB0d7egwRCQflPL3ZnzH2vy07QRjlu1lx4kLdJ6ygd73hvJ0RFncCtgCACIi4jqUk7o2ZRDOIrNgaZpgy7jxsSIiIiIiTsIwDDqEBzO3ZySNKhQjLcPGpysO0P2baPadSnJ0eCIiIuKCVLB0Fm5/a3bNSHdcHCIiIiIiNyHIz4uPH72L4W3D8PV0Y9fJizw5ZQNfr43jSobN0eGJiIiIC1HB0llkdlgCZKQ5Lg4RERERkZtkGAbtapVibs9ImlQqTrrN5PNVcXT9ZiN7Ei46OjwRERFxESpYOgnDYgXjr7cjXR2WIiIiIuK6An09GfNwLf6vXXX8vdyIjU/iqakb+XL1AXVbioiIyD9SwdKZaKVwERERESkgDMOgTfWSzOkZSfMqgWTYTL5ac5Cnp25k90l1W4qIiMj1qWDpTKwqWIqIiIhIwVK8sCfvd6jJqH/XIMDbnb2nkug6bSPjV+4nLV3dliIiInItFSydSWbBMl0FSxERKViaNWvGyy+/DEBoaChjx469bdcaPnw4tWvXvqVzHDx4EMMw2LJlS57EJHKnMwyDB8KC+O6ZSO6vVoIM02TSukM8NXUDO09ccHR4IiJyh1BO6jpUsHQmmSuFq8NSREQKsA0bNtC7d+/bdv6BAweydOnSWzpHSEgIJ06coGbNmnkU1VW3OzEWcXZFfDx4r31NPuhQk6I+7uw/nUy3bzYybvk+UtMzHB2eiIjcQZSTjs3Tc+Y1FSydidXj6p8ZWnRHREQKrsDAQHx8fPL8vKZpkp6eTuHChSlWrNgtnctqtVKyZEncMr9MdDJpaWmODkHkljSvWoK5z9xN6+pB2EyYGnWYJyZvYNuxREeHJiIidwjlpLfuduakKlg6E+tfH0ANCRcRkWyYpol55XL+P0wzV3EmJyfTpUsXChcuTKlSpfjwww+z7P/7N7qmaTJ8+HDKli2Lp6cnwcHBvPjii/ZjU1NTef311wkJCcHT05NKlSoxceJEAJYvX45hGCxcuJB69erh6enJ6tWrrxl+061bNzp06MDIkSMJCgoiICCAESNGkJ6ezquvvkrRokUpU6YMkydPtr/mf4ffZF5r6dKl1K9fHx8fHxo2bEhsbKz9Nfv376d9+/YEBQVRuHBhGjRowJIlS+z7mzVrxqFDh+jfvz+GYWAYhn3f999/T40aNfD09CQ0NDTbn9k777xDly5d8PPzu63dACL5JcDbnXfb1WDMw7UoVsiDQ2dT6DE9mjHL9nLpirotRUSckcPyUeWkWa51J+SkzlmivVNp0R0REbmR9FRsn/TI98ta+k0Cd68cH//qq6+yYsUKfvrpJ0qUKMEbb7zBpk2bsp3D5/vvv+ejjz5i1qxZ1KhRg5MnT7J161b7/i5durBu3TrGjRtHeHg4cXFxnD59Oss5Bg0axOjRo6lQoQJFihRh+fLl11xn2bJllClThpUrV7JmzRp69uzJ2rVradKkCVFRUcyePZtnn32W+++/nzJlylz33t58800+/PBDAgMDee655+jRowdr1qwBICkpibZt2/Luu+/i6enJtGnTaNeuHbGxsZQtW5Z58+YRHh5O79696dWrl/2c0dHRdOzYkeHDh9OpUyfWrl3LCy+8QLFixejWrZv9uNGjRzN06FCGDRuWw3dCxDU0rRxInZAARi/dy/wdJ/l2wxFW7jvNsDZh1AkJcHR4IiLydw7KR0E56d/dCTmpCpbOxO1qwdLMuILxD4eKiIg4o6SkJCZOnMj06dNp0aIFAFOnTr1uwnX48GFKlixJy5YtcXd3p2zZskRERACwZ88e5syZw+LFi2nZsiUAFSpUuOYcI0aM4P77779hXEWLFmXcuHFYLBaqVq3K+++/T0pKCm+88QYAgwcP5r333mP16tU8/vjj1z3Pu+++S9OmTYGrSemDDz7I5cuX8fLyIjw8nPDwcPux77zzDj/88AM///wzffv2pWjRolitVnx9fSlZsqT9uDFjxtCiRQuGDBkCQJUqVdi1axcffPBBluSwefPmvPLKKze8TxFX5eflzogHq3N/tRKMXBTLkXOX6DVjE53qlaFvk4p4e1gdHaKIiLgQ5aSun5OqYOlMrFp0R0REbsDN8+o3yw64bk7t37+ftLQ0IiMj7duKFi1K1apVsz3+scceY+zYsVSoUIHWrVvTtm1b2rVrh5ubG1u2bMFqtdqTseupX7/+P8ZVo0YNLJb/zoQTFBSUZfJyq9VKsWLFSEhIuOF57rrrLvvfS5UqBUBCQgJly5YlKSmJ4cOHM3/+fE6cOEF6ejqXLl3i8OHDNzzn7t27ad++fZZtjRo1YuzYsWRkZGC1WnN8nyKurnHF4szp6c9Hf+zjp20nmBV9lFX7TzO0TRj1yxZxdHgiIuKofPSva+eUclLXz0lVsHQmbn8tuqM5LEVEJBuGYeRqGIwrCAkJITY2liVLlrB48WJeeOEFPvjgA1asWIG3t3eOzlGoUKF/PMbd3T3Lc8Mwst1ms9lyfJ7M+X4yXzNw4EAWL17M6NGjqVSpEt7e3jz66KN5Nhl5Tu5TpCDw9XJnaJswWlYtwf8tiuHY+cs8O3Mzj9YpzYtNK1LIU7/CiIg4SkHMR0E5aW7kV06qRXeciGHvsNQq4SIi4poqVqyIu7s7UVFR9m3nzp1jz549132Nt7c37dq1Y9y4cSxfvpx169axfft2atWqhc1mY8WKFfkR+i1bs2YN3bp146GHHqJWrVqULFmSgwcPZjnGw8ODjIysi4mEhYXZ5xz6+7mqVKli/yZb5E7UsEIx5vSM5JHawQB8t/kYnSatJ+rgWQdHJiIizk45qevnpE5XsBw/fjyhoaF4eXkRGRnJ+vXrr3vslStXGDFiBBUrVrSP01+0aNEtndOhMhfdSb99y8KLiIjcToULF6Znz568+uqrLFu2jB07dtCtW7csQ1/+bsqUKUycOJEdO3Zw4MABpk+fjre3N+XKlSM0NJSuXbvSo0cPfvzxR+Li4li+fDlz5szJ57vKmcqVKzNv3jy2bNnC1q1b6dy58zXfjoeGhrJy5UqOHTtmn6j9lVdeYenSpbzzzjvs2bOHqVOn8umnnzJw4EBH3IaIUyns6cYbrarxeafaBPt7ceLCZV6YvYX/WxRDUqq+5BcRkewpJ3X9nNSpCpazZ89mwIABDBs2jE2bNhEeHk6rVq2uO3b/rbfe4ssvv+STTz5h165dPPfcczz00ENs3rz5ps/pUPZVwpV8iYiI6/rggw9o3Lgx7dq1o2XLltx7773Uq1cv22MDAgKYMGECjRo14q677mLJkiX88ssvFCtWDIDPP/+cRx99lBdeeIFq1arRq1cvkpOT8/N2cmzMmDEUKVKEhg0b0q5dO1q1akXdunWzHDNixAgOHjxIxYoVCQwMBKBu3brMmTOHWbNmUbNmTYYOHcqIESOyTG4ucqeLCC3K7B4RdKxbGoAfth6n48Qo1h444+DIRETEWSknde2c1DBN03TIlbMRGRlJgwYN+PTTT4Gr4+9DQkLo168fgwYNuub44OBg3nzzTfr06WPf9sgjj+Dt7c306dNv6pzZuXDhAv7+/iQmJuLn53ert3ldtmVTMLf8jhHZAUujjrftOiIi4vwuX75MXFwc5cuXx8ur4M0TJLfHjT43+ZXPyO2h9++/og+fY8TCGI6evwTAv2uVYkDzSvh6uf/DK0VEJDeUj8rNyKt81Gk6LNPS0oiOjrYvEQ9gsVho2bIl69aty/Y1qamp19y8t7c3q1evvulzZp73woULWR75wj4kXIvuiIiIiIhkp17ZIszqHkHn+iEYwM/bT/DYxChW7jvt6NBEREQkjzhNwfL06dNkZGQQFBSUZXtQUBAnT57M9jWtWrVizJgx7N27F5vNxuLFi5k3bx4nTpy46XMCjBo1Cn9/f/sjJCTkFu8uh9w0JFxERERE5J94e1h5pUVlvn6yLmWLeHMqKY3+329j6K+7SLykL/9FRERcndMULG/Gxx9/TOXKlalWrRoeHh707duX7t27X3cS1ZwaPHgwiYmJ9seRI0fyKOJ/YF8lXIvuiIiIiIj8k9plApjZPYKnI8piMWD+zpM8NjGKP/accnRoIiIicgucpmBZvHhxrFYr8fHxWbbHx8dTsmTJbF8TGBjIjz/+SHJyMocOHSImJobChQtToUKFmz4ngKenJ35+flke+UKL7oiIiIiI5IqXu5WX76vEpCfrEVrUhzPJaQz8YTtv/LyDcylqBBAREXFFTlOw9PDwoF69eixdutS+zWazsXTpUu65554bvtbLy4vSpUuTnp7O999/T/v27W/5nA6hOSxFROR/ONHaeOIC9HmRO1mt0v7M6N6AbneXw2LAb7sTeGxiFEtiEhwdmoiIS1N+IbmRV58XpylYAgwYMIAJEyYwdepUdu/ezfPPP09ycjLdu3cHoEuXLgwePNh+fFRUFPPmzePAgQOsWrWK1q1bY7PZeO2113J8Tqfy1xyWZoYKliIidzp396v/T0hJSXFwJOJKMj8vmZ8fkTuNp5uVfk0rMvXp+lQsXohzKVd4/acdvPbjds4mq9tSRCQ3lI/KzcirfNQtL4LJK506deLUqVMMHTqUkydPUrt2bRYtWmRfNOfw4cNZ5qe8fPkyb731FgcOHKBw4cK0bduWb775hoCAgByf06mow1JERP5itVoJCAggIeFqZ5CPjw+GYTg4KnFWpmmSkpJCQkICAQEBWK1WR4ck4lDVS/kxvWsDvl57kCl/HmJp7Ck2Hj7P6y2r8EBYCf17KiKSA8pHJTfyOh81TPX2/qMLFy7g7+9PYmLibZ3P0hazFnPBp1CmOtaOb92264iIiGswTZOTJ09y/vx5R4ciLiIgIICSJUtm+8tEfuUzcnvo/bt5MfEXeXvBbvYkJAHQrHJxBj9QleKFPR0cmYiI81M+KrmVV/moU3VY3ukMN3dM0CrhIiICgGEYlCpVihIlSnDlirrv5cbc3d3VWSmSjWpBvkzrUp/J6w7x9bqDLN97mk1HzjOwRWXa1sj+FyoREblK+ajkRl7moypYOhOtEi4iItmwWq0qRImI3AJ3q4Xe95bnviqBDF+wm5j4iwydv5vFMQm80aoaJXzVbSkiciPKRyW/OdWiO3c8e8FS31qIiIiIiOS1yiUKM+XpevRpUgF3q8Gq/Wd4bGIUP287rlVwRUREnIgKls7ETYvuiIiIiIjcTu5WCz3uCeXbrg2oUcqXpNR03l4Yw4tzt3LywmVHhyciIiKoYOlcNCRcRERERCRfVAwszKSn6vFis4p4WC2sjTtLx4lRzNtyTN2WIiIiDqaCpTPJLFima9EdEREREZHbzc1ioWtkOWZ0b8BdwX4kp2Xw7m+x9Jm9heOJlxwdnoiIyB1LBUtnYv1rDSR1WIqIiIiI5JvyxQrx9ZP16H9fJTzdLEQdOkenSeuZs+koNnVbioiI5DsVLJ2JmxbdERERERFxBKvF4KmIsszqHkHtMv6kpGXwn8V7eG7WZo6eV7eliIhIflLB0plkDgm3ZWDabI6NRURERETkDlS2qA8TOtdlYIvKeLlbiD58nk6TopgVfUTdliIiIvlEBUtnktlhCeqyFBERERFxEIth8ET9EGZ1j6Be2QAuX7HxwZK99J6xicNnUxwdnoiISIGngqUzsapgKSIiIiLiLEKK+PDF43UY9EAVfDysbD6ayOOT1zN9/WEybOq2FBERuV1UsHQmFitgXP17ugqWIiIiIiKOZjEMHqtThtk9IogsV4TUdBsf/bGPZ76NJu5MsqPDExERKZBUsHQihmFopXAREREREScU7O/N+E61ebNVVQp5WNl2/AKdJ29gatQh0jX/vIiISJ5SwdLZaKVwERERERGnZBgGD9cuzeyekdxTvihpGTbGLd9Pj+nR7D+V5OjwRERECgwVLJ2NVQVLERERERFnVsrPi08eC2dYm2oU9nRj54mLPDl1AxPXHuRKhrotRUREbpUKls4ms8NSc1iKiIiIiDgtwzD4913BzO0ZSeOKxbiSYfLZqgN0+yaavQnqthQREbkVKlg6G3VYioiIiIi4jBK+nnz0yF2886/q+Hm5ERN/kaembuCr1XHqthQREblJKlg6Gy26IyIiIiLiUgzDoG2NkszpGUmzysVJt5l8uSaOLtM2EhN/0dHhiYiIuBwVLJ1NZodleppj4xARERERkVwJLOzJ6IdqMbJdDfy93dmTkESXaRv5bOUB0tLVbSkiIpJTKlg6G/sq4eqwFBERERFxNYZh0Kp6EN/1jKRl1RJk2EwmrjvIU1M3sOvEBUeHJyIi4hJUsHQ2f3VYmlp0R0RERETEZRUt5MF/OtTkP+1rUsTHnf2nk+n6zUY+WbGf1PQMR4cnIiLi1FSwdDZadEdEREREpMBoWa0Ec3tG0iosCJsJU/48xJNTNrD9WKKjQxMREXFaKlg6Gw0JFxEREREpUIr4eDDy3zX48OFaFCvkQdyZFHp8G83YP/Zx+Yq6LUVERP6XCpZOxrCvEq5Fd0RERERECpJmlQOZ2zOSB2uWxGbCN+sP03nKBrYcPe/o0ERERJyKCpbOxupx9U91WIqIiIiIFDj+3u6MeLA6Yx+9i8DCHhw6m8Iz325i9JI9XEpTt6WIiAioYOl83P7qsNSiOyIiIiIiBVbjisWZ2zOS9rVKYQIzo4/y+OT1RB8+5+jQREREHE4FS2ejRXdERERERO4Ivl7uDG0bxiePhRPk68nR85foPXMz7/0eS0qaRlyJiMidSwVLZ5NZsFSHpYiIiIjIHaFhhWLM6RnJw+HBAMzdfIyOE9ez/uBZB0cmIiLiGCpYOhv7ojsqWIqIiIiI3CkKe7rxZutqfN6pNqX8vDhx4TLPz97Cu4tiSEpVt6WIiNxZVLB0Nm6Zi+6oYCkiIiIicqeJCC3KnJ4RdKxbGoB5W4/TcWIUaw+ccXBkIiIi+UcFS2eT2WGZrm9RRURERETuRD4ebrx+f1W+fKIOpQO8iL+YSr+5WxmxYDcXL6uxQURECj4VLJ2N29U5LE11WIqIiIiI3NHqly3C7O6RPFGvDAbw0/YTPDYxilX7Tzs6NBERkdtKBUtno1XCRURERETkL94eVga2rMKEJ+tStog3p5LSePm7bQz9dReJl/Q7g4iIFEwqWDobFSxFREREROR/1CkTwIzuETzVIAQDmL/zJI9NjGL53lOODk1ERCTPqWDpZIzMgmW6CpYiIiIiIvJf3u5W+jevzKSn6hFa1IczyWm8Mm87b/y8k3MpaY4OT0REJM+oYOls3DI7LLXojoiIiIiIXOuu0v7M6N6ArpFlsRjw2+54Ok6MYklMgqNDExERyRMqWDob+yrh6rAUEREREZHsebpZebFZJaY8XZ+KxQtxNuUKr/+0g9d/3MHZZHVbioiIa1PB0tm4aQ5LERERERHJmRql/JjetQE97wnFahgsiU3gsYlR/LYrHtM0HR2eiIjITVHB0tlo0R0REREREckFDzcLLzSpwNQu9akcWJjzl67wxi87efXHHZxOSnV0eCIiIrmmgqWzUcFSRERERERuQlhJX77pWp/ejUKxWgz+2HOKjhOjWLDzpLotRUTEpahg6WzctEq4iIiIiIjcHHerhWfvrcD0rvWpGlSYxMvpDPl1F/2/38api+q2FBER16CCpbPJXHRHq4SLiIiIiMhNqlLCl6lP1+f5xuVxsxis2n+GRydG8fP2E+q2FBERp6eCpbOxelz9M+OKEgkREREREblp7lYLzzQsz7fdGlC9pC9Jqem8vWA3L323jZMXLjs6PBERketSwdLZZHZYgrosRURERETkllUKLMzkp+vRt2kF3K0Gaw6coePEKH7YelxNEiIi4pRUsHQ2mXNYghbeERERERGRPOFmsdD97lBmdougVrAfyWkZ/N+iGPrM2cKJxEuODk9ERCQLFSydTZYOSxUsRUREREQk75QvXoiJT9bj5fsq4elmIergOTpOWs/czUexqdtSRESchAqWTsYwLFp4R0REREREbhurxeDpiLLM7B5BeGl/UtIyeO/3PTw/azNHz6vbUkREHE8FS2dk/WtYeHqaY+MQEREREZECq1xRHyZ0rssrLSrj6WZh4+HzdJoUxazoI+q2FBERh1LB0hllFizVYSkiIiIiIreR1WLQuX4Is3tEUDckgMtXbHywZC+9Z2zi8NkUR4cnIiJ3KBUsnVHmkPB0zWEpIiIiIiK3X0gRH758og6v318Fb3crm48m8vjk9Xy74TAZNnVbiohI/lLB0hllrhSuRXdERERERCSfWAyDjnXLMLtHBA3KFSE13caYZft45ttoDp5JdnR4IiJyB1HB0hlZVbAUERERERHHKB3gzeedavNmq6oU8rCy7fgFnpi8gWlRh9RtKSIi+UIFS2ekIeEiIiIiIuJAhmHwcO3SzO4Zyd2hRUnLsPHx8v10nx7NgdPqthQRkdtLBUtn5OZx9U91WIqIiIiIiAOV8vPi047hDG1TjcKebuw8cYHOU9Yzad1B0m02R4cnIiIFlAqWzuivDktTq4SLiIiIiIiDGYZB+7uCmdszknsrFuNKhsn4lQfoOi2avaeSHB2eiIgUQCpYOiPNYSkiIiIiIk6mhK8nYx+5i7cfDMPX042Y+Is8NWUDX62J40qGui1FRCTvqGDpjDILlprDUkREREREnIhhGPyrZinmPhNJ00rFSbeZfLk6ji7TNhIbf9HR4YmISAGhgqUTMtzUYSkiIiIiIs4rsLAnHz5ci/9rVx1/Lzf2JCTx9LSNfL7qgLotRUTklqlg6YzUYSkiIiIiIk7OMAzaVC/J3GfupkWVQDJsJl+vPciTUzaw68QFR4cnIiIuTAVLZ2TvsNSiOyIiIiIi4tyKFfLg/Ydq8V77mhTxcWf/6WS6fRPNpyv2k5qe4ejwRETEBalg6Yz+WiWcjDTHxiEiIiIiIpJD91crwdyekTxQrQQZpsnkPw/x5JQNbD+e6OjQRETExahg6YysHlf/VIeliIiIiIi4kCI+HoxqX5MPOtSkqI87cWdS6DE9mrF/7OPyFXVbiohIzqhg6YwyOyw1h6WIiIiIiLig5lVLMPeZu2lTPQibCd+sP0znKRvYelTdliIi8s9UsHRGWiVcRERERERcXIC3O//XrgYfPXIXxQt7cOhsCj2/jebDpXu5pG5LERG5ARUsnZFVi+6IiIiIiEjB0KRSceb2jKRdzZKYwIyNR3h80no2HTnn6NBERMRJqWDpjP4aEm6ma9EdERERERFxfX5e7gx/sDrjHg2nRGFPjp6/RK8Zm3l/8R5S0tSoISIiWalg6YzctOiOiIiIiIgUPI0qFmPOM5F0uKsUALM3HeXxSetZf+isgyMTERFn4nQFy/HjxxMaGoqXlxeRkZGsX7/+hsePHTuWqlWr4u3tTUhICP379+fy5cv2/cOHD8cwjCyPatWq3e7buDWZi+5oDksRERERESlgfD3dGNImjE87hlPSz5NjiZd5ftYWRv4WS3KqmjZERMTJCpazZ89mwIABDBs2jE2bNhEeHk6rVq1ISEjI9vgZM2YwaNAghg0bxu7du5k4cSKzZ8/mjTfeyHJcjRo1OHHihP2xevXq/Lidm2ZkLrqjVcJFRERERKSAuqd8Meb0iOSR2qUB+H7LMTpOiuLPOHVbiojc6ZyqYDlmzBh69epF9+7dqV69Ol988QU+Pj5MmjQp2+PXrl1Lo0aN6Ny5M6GhoTzwwAM88cQT13Rlurm5UbJkSfujePHi+XE7N8+qVcJFRERERKTgK+TpxhutqvL547Up7e/FyQup9JmzhXcW7uaiui1FRO5YTlOwTEtLIzo6mpYtW9q3WSwWWrZsybp167J9TcOGDYmOjrYXKA8cOMCCBQto27ZtluP27t1LcHAwFSpU4Mknn+Tw4cM3jCU1NZULFy5keeQrrRIuIiIiIiJ3kIhyRZnVI4JO9coA8OO2E3SaGMWa/WccHJmIiDiC0xQsT58+TUZGBkFBQVm2BwUFcfLkyWxf07lzZ0aMGMG9996Lu7s7FStWpFmzZlmGhEdGRjJlyhQWLVrE559/TlxcHI0bN+bixYvXjWXUqFH4+/vbHyEhIXlzkzmVWbDUKuEiIiIiInKH8PFw47WWVZjQuQ5lAryJv5jKi99tZfj8XVy4rNFnIiJ3EqcpWN6M5cuXM3LkSD777DM2bdrEvHnzmD9/Pu+88479mDZt2vDYY49x11130apVKxYsWMD58+eZM2fOdc87ePBgEhMT7Y8jR47kx+38l5s6LEVERERE5M5UN6QIs3pE8GT9EAzglx0neWxiFCv3nXZ0aCIikk/cHB1ApuLFi2O1WomPj8+yPT4+npIlS2b7miFDhvD000/zzDPPAFCrVi2Sk5Pp3bs3b775JhbLtfXYgIAAqlSpwr59+64bi6enJ56enrdwN7dIq4SLiIiIiMgdzNvdyoAWlWlRtQRvL9zNobMp9P9+G22qBzGwZRUCvN0dHaKIiNxGTtNh6eHhQb169Vi6dKl9m81mY+nSpdxzzz3ZviYlJeWaoqTVagXANM1sX5OUlMT+/fspVapUHkV+G1i1SriIiIiIiEh4GX9mdGtAl4iyWAxYuCuejhOjWLbnlKNDExGR28hpCpYAAwYMYMKECUydOpXdu3fz/PPPk5ycTPfu3QHo0qULgwcPth/frl07Pv/8c2bNmkVcXByLFy9myJAhtGvXzl64HDhwICtWrODgwYOsXbuWhx56CKvVyhNPPOGQe8wRDQkXEREREREBwMvdykv3VWLSU/UoX8yHM8lpvPrDdgb/tINzKZr3X0SkIHKaIeEAnTp14tSpUwwdOpSTJ09Su3ZtFi1aZF+I5/Dhw1k6Kt966y0Mw+Ctt97i2LFjBAYG0q5dO9599137MUePHuWJJ57gzJkzBAYGcu+99/Lnn38SGBiY7/eXY/ZVwvU/XxEREREREYBawf58260BE9YcZFrUYX6PSWDD4XO8fn9V7q9WwtHhiYhIHjLM642dFrsLFy7g7+9PYmIifn5+t/165qUkbJ/3BsDy8jcYFuttv6aIiIgUbPmdz0je0vsnktXukxcYNn83+08nA9CiSiCvP1CVYoU8HByZiIhcT27yGacaEi5/cftb46vmsRQREREREckirKQf33ZrQK+GoVgtBkv3nOKxiVEs2nXyuusZiIiI61DB0hlZ/7binVYKFxERERERuYa71cJzjSvwTZf6VClRmMRLV3jzl1288sN2TiWlOjo8ERG5BSpYOiHDYgXjr7dGHZYiIiIiLu/8+fPUr1+f2rVrU7NmTSZMmODokEQKjKpBvkzrUp9n7y2Pm8Vgxd7TdJwYxfwdJ9RtKSLiolSwdFb2lcJVsBQRERFxdb6+vqxcuZItW7YQFRXFyJEjOXPmjKPDEikw3K0Wejcqz/RuDQgr6cuFy+kMnb+bl7/fRsJFdVuKiLgaFSydlVUFSxEREZGCwmq14uPjA0BqaiqmaarzS+Q2qBxYmClP16NPkwq4Ww1W7z/DYxOj+Gnbcf03JyLiQlSwdFb2gmW6Y+MQERERuQOsXLmSdu3aERwcjGEY/Pjjj9ccM378eEJDQ/Hy8iIyMpL169fn6hrnz58nPDycMmXK8Oqrr1K8ePE8il5E/s7NYqHHPaHM6BZBjVJ+JKWmM2JhDP3mbuXEhcuODk9ERHJABUtnlblSeHqaY+MQERERuUknTpxg69atJCcnOzqUf5ScnEx4eDjjx4/Pdv/s2bMZMGAAw4YNY9OmTYSHh9OqVSsSEhLsx2TOT/m/j+PHjwMQEBDA1q1biYuLY8aMGcTHx183ntTUVC5cuJDlISK5U6F4ISY9VZeXmlXEw2phXdxZOk2MYt6WY+q2FBFxcipYOit1WIqIiIiL+umnn6hWrRplypShbt26REVFAXD69Gnq1KmTbfeio7Vp04b/+7//46GHHsp2/5gxY+jVqxfdu3enevXqfPHFF/j4+DBp0iT7MVu2bGHHjh3XPIKDg7OcKygoiPDwcFatWnXdeEaNGoW/v7/9ERISkjc3KnKHcbNY6BJZjpndGxBe2p/ktAze/S2WF2Zv4XjiJUeHJyIi16GCpbPSHJYiIiLign755RcefvhhihcvzrBhw7J0MRUvXpzSpUszefJkB0aYe2lpaURHR9OyZUv7NovFQsuWLVm3bl2OzhEfH8/FixcBSExMZOXKlVStWvW6xw8ePJjExET748iRI7d2EzmU8ctYbKtmYl7UgkBSsIQWK8SEznV5pXllPN0srD90jo4T1zNn01Fs6rYUEXE6Klg6q8yCZboKliIiIuI6RowYQZMmTVi9ejV9+vS5Zv8999zD5s2bHRDZzTt9+jQZGRkEBQVl2R4UFMTJkydzdI5Dhw7RuHFjwsPDady4Mf369aNWrVrXPd7T0xM/P78sj9vNPH0E9q7H3PALtq9fwjb/E8wT+277dUXyi9Vi0LlBCLN7RFA3JIBLVzL4z+I9PDdzM0fOpTg6PBER+Rs3Rwcg1+F2tWBpZqRjODgUERERkZzasWMHY8aMue7+oKCgLPM+3ikiIiLYsmWLo8O4saKlsbR/BVv0Qji6CzN2HWbsOgiugqVua6jUAMNidXSUIrcspIgPXz5Rh7mbjvHJiv1EHznP45PX06dJRR6vVwaLod/AREQcTQVLZ2X9663J0KI7IiIi4jp8fHxuuMjOgQMHKFasWD5GdOuKFy+O1Wq9ZpGc+Ph4SpYs6aCo8p5hsUDFelgr1sNMOIi5aSFmzFo4vgfb8T3gVxyjdiuMWvdhePo4OlyRW2IxDDrVK8O9FYvxzqIYNhw6x4dL97IkJoFhbcMoV1SfcRERR9KQcGfl5nH1z3QtuiMiIiKu47777mPq1KmkZ5PDnDx5kgkTJvDAAw84ILKb5+HhQb169Vi6dKl9m81mY+nSpdxzzz0OjOz2MUqEYmn9PJZen2BEPgTevnDhNObKb7F91RfbH1Mxz19/lXMRV1E6wJvPO9XmjVZV8fGwsvVYIk9MXs836w+TYdPcliIijqIOSydlWN0wQYvuiIiIiEt59913ufvuu2nQoAGPPfYYhmHw22+/sWzZMr788ktM02TYsGGODvMaSUlJ7Nv33/ka4+Li2LJlC0WLFqVs2bIMGDCArl27Ur9+fSIiIhg7dizJycl0797dgVHffkahAIxGj2FGtMeMWYO5aSGcOYq5+TfMzb9DxbpY6raFMtUwNIxWXJRhGDxSuzQNyxflnUUxRB08x9g/9rE0NoFhbcIoX7yQo0MUEbnjGKapJdH+yYULF/D39ycxMTFfJjwHsC38DHP3aowmnbHU/1e+XFNEREQKrvzMZ3bu3MlLL73EH3/8kWWV8GbNmjF+/HjCwsJu6/VvxvLly7nvvvuu2d61a1emTJkCwKeffsoHH3zAyZMnqV27NuPGjSMyMjJf4nNEPpod0zTh8A5s0Qvg4Nb/7igRilG3DUbVezCs6okQ12WaJj9tO8GYZXtJTsvA3WrQu1F5ukSWxc2iAYoiIrciN/mMCpY54JCC5e8TMHf8gdHwMSx3P5Qv1xQREZGCyxH5zLlz59i3bx82m40KFSoQGBiYL9ctiJylYPl35pljmJsXYe5aBel/zbteKAAj/H6M8BYY3s4Rp8jNiL9wmXd/i2XNgTMAVC/py7C2YVQKLOzgyEREXFdu8hl9ReSs/lolXEPCRURExJWMGDGCHTt2AFCkSBEaNGhAZGSkvVi5c+dORowY4cgQJY8YxUpjadnz6jyXjTpCoQBIPo+5di62r/phWzwB88xRR4cpclOC/Lz4+NG7GN42DF9PN3advMiTUzYwYU0cVzJsjg5PRKTAU8HSWVn/Klimq2ApIiIirmP48OFs27btuvt37NjB22+/nY8Rye1mePtiieyA5ZlxGG1egKDykHEFc/sf2Ka+Rsb372Ee3IoGdomrMQyDdrVKMbdnJE0rFSfdZvLF6ji6frORPQkXHR2eiEiBpglmnFXm3D8ZWiVcRERECo6zZ8/i4eHh6DDkNjCsbhhh92JWawTHYrFtWgj7NsKhbdgObYOipTHqtsYIa4zhrs+AuI5AX08+fLgWi3bH88HiPcTGJ/HU1I30vKccPe4Jxd2qPiARkbymgqWz0pBwERERcRErV65k+fLl9ufz5s3LsuJ2pvPnzzN79mxq1aqVj9FJfjMMA8pUw1qmGub5+Ksriu9cAWePYS6ZiLl6NkZ4y6tzXRYu4uhwRXLEMAzaVC9Jg7JF+M/iPSzbc4qv1hzkjz2nGda2GmElNWeriEhe0qI7OeCQRXc2/IK5aiZG9cZYWj+fL9cUERGRgut25jNvv/22fZi3YRg3HPpbvXp1Jk6cmG+raxcUzrjoTm6YqSmYO5Zjbv4NLpy6utFivbqqeL22GCVCHRqfSG6YpsnimAT+s3gP5y9dwWoYdL27LL0alsfDTd2WIiLXo1XC85hDCpabFmEun4ZR9W4sD76YL9cUERGRgut25jOXLl0iJSUF0zQpUaIEX3zxBY888kiWYwzDwMfHBy8vrzy99p3C1QuWmUxbBuyLxrZpARzf898dpathqdcGKtTDsKjgI67hXEoa/1m8h8UxCQBUKFaIYW3DqBnsuv+NiojcTrnJZzQk3Fn9NSTc1KI7IiIi4uS8vb3x9vYGIC4ujsDAQHx8fBwclTgjw2KFKhFYq0RgntyPuWkh5p4oOBaD7VgM+Adh1GmFUbMphoe3o8MVuaEiPh68174mD1RLYNTvsRw4k0z36Rt5OqIsz95bHk83q6NDFBFxWfr60lllrhKuRXdERETEhZQrV07FSskRo2RFLG37Yuk5FqPBv8GzECTGYy6fhu2rvthWTMdMPOXoMEX+UfOqJZj7zN20rh6EzYSpUYd5YvIGth1LdHRoIiIuSx2WzipzlXB1WIqIiIiL2bZtG5988gmbNm0iMTERm82WZb9hGOzfv99B0YmzMXyLYTR+HPPuDpi7VmFuWgTnTmBGL8DctBAqNcBSry2Uqnx1QR8RJxTg7c677WrwQLUSjPw9lkNnU+gxPZrO9UN4vkkFvN3VbSkikhvqsHRShlYJFxERERe0fPlyIiIi+PXXXwkODubAgQNUqFCB4OBgDh06ROHChWnSpImjwxQnZLh7YQm/H0u3D7B0eBXK1gTThL3rsc0ajm3mUGwxazE1AkmcWNPKgcztGcm/apbEBL7deIQnJq9n85Hzjg5NRMSlqGDprKwqWIqIiIjrGTp0KBUqVCA2NpbJkycD8MYbb7B69WrWrl3L0aNH6dixo4OjdB3jx4+nevXqNGjQwNGh5BvDsGBUqIP10TewPP0eRo1mV3Pjk/sxF3yKbeLL2Nb/jHk5ydGhimTLz8udtx+szseP3kWJwp4cOXeJXjM28f6SPVxKy3B0eCIiLkEFS2elgqWIiIi4oE2bNtGzZ0/8/PywWq8OgczIuPoLemRkJM8++yxDhgxxZIgupU+fPuzatYsNGzY4OhSHMALLYmnVG0uvcRj3PAo+fpB0FnP1LGxf9cO2dDLmuROODlMkW/dWLM6cZyJpf1cpTGB29FE6TY5i4+Fzjg5NRMTpqWDprDKHhKdryIuIiIi4Djc3N3x9fQEICAjA3d2dhIQE+/4KFSqwa9cuR4UnLsrw8cdyz8NYnvkEo9WzULwspKdibl2MbfIrZPz4AebhHZim6ehQRbLw9XRjaJswPnksnCBfT46dv8yzMzcz6vdYklP1u56IyPWoYOms1GEpIiIiLqhSpUrs3bsXuLq4TrVq1fjhhx/s++fPn0/JkiUdFZ64OMPNHUuNplieHoXl0TehQp2rOw5sxvbdSGzfDMa2YzmmFq4UJ9OwQjHm9IzkkdrBAHy3+RidJq0n6uBZB0cmIuKcVLB0VipYioiIiAtq27YtM2fOJP2vUSIDBgxg3rx5VK5cmcqVK/Pzzz/z7LPPOjhKcXWGYWCUrYG1w6tYun+IEX4/uHnC6cOYv3+F7et+2NZ9j5mS6OhQRewKe7rxRqtqfP54bYL9vThx4TIvzN7C/y2K4aK6LUVEsjBMjZv4RxcuXMDf35/ExET8/Pzy5ZrmuZPYJg8Ady+s/SblyzVFRESk4MqvfObKlStcuHCBokWLYhgGANOnT+f777/HarXyr3/9i27dut226xdUjshHXY15OQlz+x+Ym3+DpL+61qzuGNUaYtRtgxFY1rEBivxNSlo6n644wOxNRwEI8vXkrdbVaFihmIMjExG5fXKTz6hgmQMOKVhePINtQj+wWLG+/E2+XFNEREQKrvzIZ65cucLu3bspWrQoZcqUuS3XuFOpYJlzZkY65t71mJsWwsn9/91RtgaWum2hfDiGoYFm4hyiD59jxMIYjp6/BMC/a5ViQPNK+Hq5OzgyEZG8l5t8Rv+ndlaZQ8JtGZimzbGxiIiIiOSAxWKhXr16zJs3z9GhyB3MsLphqdYQa+d3sDw+HKNKJBgGHN6J7ccPsE15FdvWxZhXLjs6VBHqlS3CrO4RdK4fggH8vP0Ej02MYuW+044OTUTEoVSwdFZuf/tGTSuFi4iIiAuwWq2UK1eO1NRUR4ciAoARXAXLv17C0nMsRr0HwcMbzp3AXDoZ21f9sK2aiXnxjKPDlDuct4eVV1pU5usn61KuqA+nktLo//02hvy6k8RLWtNARO5MKlg6K+vfCpYZaY6LQ0RERCQX+vXrx1dffcXZs1r5VpyH4ReIpemTWHp/itGsC/gHQWoy5oZfsE18Gdv8TzH/PnxcxAFqlwlgRrcGPB1RFosBC3bG89jEKP7Yc8rRoYmI5Ds3Rwcg12GxAgZgQoY6LEVERMQ1ZGRk4OnpScWKFXn00UcJDQ3F29s7yzGGYdC/f38HRSh3MsPDG6Nua8zaD8CBTdg2LYSjuzFj12LGroXgKlfnuaxUH8Oi3g7Jf17uVl6+rxItqgby9oLdxJ1JYeAP22kVVoJXW1ahiI+Ho0MUEckXWnQnBxw1yXnGx10h4wqWnh9j+Afm23VFRESk4MmvfMaSgyKPYRhkZGTcthgKIi26c/uY8XGYmxZixq4D21+fS79AjDoPYNS8D8PTx7EByh0rNT2Dr9YcZFrUIWwmFPFxZ9D9VWlZrYSjQxMRuSm5yWfUYenM3Nwh48rVh4iIiIgLiIuLc3QIIrliBJXHaPMCZuMnMLcuxty6BC6cwlzxLeba7zFqNsOo0wojIMjRocodxtPNSr+mFWlRJZDhC3az/3Qyr/+0gxYxgQy6vypFC6nbUkQKLhUsnVnmPJYaEi4iIiIuoly5co4OQeSmGIWLYDTqiBnRAXP3aszNC+HMMczNizA3/wYV62Gp1wZKV8MwDEeHK3eQ6qX8mN61ARPXHWTyn4dYGnuKjYfP81rLyrQKC9LnUUQKJBUsnVnmSuHpWnRHRERERCQ/GO4eGHc1x6x1Hxzahi16IRzaBvs3Ytu/EYLKY9Rpg1H1bgyrfp2S/OHhZuH5xhW4r8rVuS33JCTx5i+7WByTwOAHqlK8sKejQxQRyVOaSdqZqcNSRERERMQhDMPACA3H+sggLF3fx6jV/Gp+Hh+HuegzbF+/iC3qR8xLFx0dqtxBqgX5Mq1LfZ69tzxWi8Hyvad5bGIU83ecQMtTiEhBooKlM8v8xlZzWIqIiIiIOIxRrAyW+5/B0vsTjEYdoVAAJJ/HXDMH24R+2BZ/jXnmmKPDlDuEu9VC70bl+bZrA6oF+XLhcjpD5++m//fbSLiY6ujwRETyhAqWzszeYamCpYiIiIiIoxneflgiO2B5ZhxG6xegRCikp2FuX4Zt6qtkzPsP5sFt6nSTfFG5RGGmPF2PPk0q4G41WLX/DI9NjOLnbcf1GRQRl6dJV5yZfQ5LFSxFRERERJyFYXXDqH4vZlgjOBZzdZ7L/dFwcCu2g1uhWOmr81yG3YvhrpWc5fZxt1rocU8oTSsV5+2FMew8cYG3F8bwe0wCb7WuRkk/L0eHKCJyU9Rh6cz+6rA01WEpIiIiIuJ0DMPAKBOGtf0ALD3GYNRpBe5eV1cXX/I1tq/7YVszBzPpnKNDlQKuYmBhJj1VlxebVcTDamFd3Fk6Toxi3pZj6rYUEZd0Sx2Whw8f5vDhw9x77732bVu3buXDDz8kNTWVJ554gg4dOtxqjHcuqzosRURExLX06NHjhvsNw8DLy4syZcrQrFkz7rnnnnyKTOT2MgKCMO7ritnwMcwdf2Bu/g0unMaM+hFzwy8Y1Rpi1G2DUSLU0aFKAeVmsdA1shxNKhVnxILdbDt+gXd/i2VJTAJvtalGsL+3o0MUEckxw7yFr1s6dOhAUlISS5YsASA+Pp6wsDDS0tLw9fUlISGBuXPn8vDDD+dZwI5w4cIF/P39SUxMxM/PL9+um/HLWNi7HqN5dyy178+364qIiEjBk1/5TGhoKJcuXeLUqVMAFClSBIBz5652mAUGBmKz2Thz5gyGYdCqVSu+++47fHx8bltMrmz8+PGMHz+ejIwM9uzZk+/5qNw805YB+zZi27QQju/5744y1bHUbQ0V6mJYNOBNbo8Mm8ms6COMX3mA1HQbPh5W+jWtyKN1SmMxDEeHJyJ3qNzko7f0f8j169dz//3/LaRNmzaNS5cusXXrVo4dO0aLFi0YPXr0rVzijmbYVwlPc2wgIiIiIjm0cOFCPD09GT58OGfOnLE/Tp8+zbBhw/D29mbNmjWcO3eOIUOGsGjRIoYMGeLosJ1Wnz592LVrFxs2bHB0KJJLhsWKUSUS6+PDsTwxAqNqQzAscHQXtp/HYJv8CrbNv2GmXXZ0qFIAWS0GTzYoy6zuEdQp409KWgb/WbyH52Zt5uj5S44OT0TkH91SwfLs2bOUKFHC/vzXX3+ladOmVKxYEYvFwsMPP0xMTMwtB3nHsv41QXdGumPjEBEREcmhvn370rZtW4YOHWrvrgQoWrQow4YNo3Xr1vTt2xd/f3+GDx/O448/znfffefAiEVuP6NUJSwP9sXyzMcYDdqBZyFIjMf8Yyq2CX2xrfgW88IpR4cpBVDZoj581bkur7asjJe7hejD5+k0KYqZG49g09yWIuLEbqlgGRgYyKFDhwA4f/48f/75J61atbLvT09PJz1dxbab5vZXh6XmsBQREREX8eeffxIeHn7d/eHh4axdu9b+vHHjxsTHx+dHaCIOZ/gWw9L4CSy9P8Fo3h2KlILUFMzo+dgm9sf268eYfx8+LpIHLIbB4/VCmN0jknplA7h8xcbopXvpNWMTh8+mODo8EZFs3dKiOy1btmTcuHH4+fmxfPlybDZblkV2du3aRUhIyK3GeOfKXHRHq4SLiIiIiwgICOD333/n+eefz3b/okWL8Pf3tz9PSkrSnIxyxzHcvTBq348Z3gLitmLbtAAO78TcE4W5JwpKVsKo1wajcgSGxerocKWAKBPgzReP1+H7LccYt3w/W44m8vjk9bzQuAJP1A/BatHcliLiPG6pw/K9994jLCyMgQMH8vvvvzN69GjKly8PQGpqKnPmzKFFixZ5EugdyV6wVJeqiIiIuIZevXrx008/8eijj7J06VIOHTrEoUOHWLp0KY8++ii//vorvXr1sh+/YMECateu7biARRzIMCwYFepgffRNLE+/h1GjKVjd4OQ+zPmfYJv4MrYNv2BeTnJ0qFJAWAyDx+qUYXaPCCLLFSE13cZHf+yj57fRxJ1JdnR4IiJ2t7RKeKbExES8vb3x8PCwb7t06RJ79uwhJCSEokWL3uolHMpRq4Tb1n6H+ec8jLtaYGnZM9+uKyIiIgVPfuUzpmkyaNAgPvroIzIyMrLss1qtvPTSS7z//vsYhsHly5eZPXs2d911F3Xq1LltMRUEjspHJf+ZyYmYWxdjblsCKReubnT3xKjeBKNua4wipRwboBQYpmny47YTfLRsL8lpGXhYLTx7b3meigjBTSvYi8htkJt8Jk8KlgWdwwqW63/GXD0Lo0ZTLK2ezbfrioiISMGT3/lMQkKCvcMSoFy5crRo0SLLgo2ScypY3nnM9DTMmLWYmxbC6SN/bTWgQh0sddtASHUMQ0N45dadvHCZdxfFsDbuLAA1SvkyrE0YFQMLOzgyESlo8q1guXTpUjZt2sSrr75q3zZp0iSGDx9OamoqnTt3ZvTo0Vitrj3visMKltELMFdMx6jaEMuDffPtuiIiIlLwqODl2vT+3blM04QjO7FFL4S4zf/dEVjuasdl1YYYbu6OC1AKBNM0+WXHST5cupek1HTcrQa9GpanS2RZ3K3qthSRvJGbfOaWFt0ZPnw45cqVsz/fvn07zz77LHfddReVKlVi3LhxlCxZktdff/1WLnPn+ivxMLXojoiIiLiYixcvcujQIc6dO0d23483adLEAVGJuB7DMKBsTaxla2KePY65eRHmzlVw6hDmb19irpqFEX4/RngLDB//fz6hSDYMw+DftUpxd2hRRv4Ww6r9Z/hs1QGW7TnF8LZhVC6hbksRyV+3VLDcvXs3jzzyiP35N998g5+fH6tWrcLHx4fnnnuOadOmqWB5s7RKuIiIiLiYM2fO0LdvX77//nv7HJamadqHrmb+/X/ntxSRf2YUDcZo0QOzYUfM7cswt/wOSWcx132Huf4njLBGGHXbYBQPcXSo4qJK+Hry0SN3sXBXPB8s2UNM/EWenLqBZ+4Jpfs95dRtKSL55pYKlsnJyVlaOBctWkTr1q3x8fEBoEGDBkyfPv3WIryTqWApIiIiLqZXr1788ssvvPjiizRu3JgiRYo4OiSRAsfwLowR8W/Mem0x967HjF4A8QcwdyzH3LEcytXCUqc1lA/HMFRgktwxDIO2NUoSUa4Io36PZfne03y5Ju5qt+WDYVQL8nV0iCJyB7ilgmVISAgbNmygR48e7Nu3jx07dvDKK6/Y9589exZPT89bDvJOZVjdMQHSVbAUERER1/D777/Tv39/3n//fUeHIlLgGVY3jGoNMaveA8f3YNu0EPZtgEPbsR3aDkVKXZ3nsnpjDHcvR4crLqZ4YU9GP1SL33cn8J8le9h7KokuUzfS7e5yPNMwFA83FcNF5Pa5pYLlk08+yYgRIzh27Bg7d+6kSJEitG/f3r4/OjqaKlWq3HKQd6zMybMz0h0bh4iIiEgO+fj4EBoa6ugwRO4ohmFA6apYS1fFTDyFueU3zO1/wLkTmEsnY66eg3FXC4za92P4FnN0uOJCDMOgVfUgGpQrwnuLY1kae4qJ6w6yfO/VuS2rl9IiYCJye9zSVyJvvvkmgwYN4siRI5QtW5Yff/yRgIAA4Gp35fLly/n3v/+dF3Hemax/1ZM1JFxERERcxFNPPcUPP/zg6DBE7liGfyCWpk9h6f0pRrMu4F8CUpMxN/yMbeLL2OZ/inlyv6PDFBdTtJAH73eoxX/a16SIjzv7TyfT9ZuNjFu+j9R0zUksInnPMLNbtlGyyM2y63nJPBaDbfYICCiJtceYfLuuiIiIFDz5lc+sXbuWfv36ERgYSO/evQkJCcFqtV5zXN26dW9bDAWRo/JRcX2mzQYHorFFL4RjMf/dEVwFS922UKk+hkVDeyXnzqWk8cGSvfy2Ox6A0KI+DG8bRq3SWqVeRG4sN/lMnhUsk5KSOHLkCHB1bsvChQvnxWmdgsMKlgkHsU1/AwoFYH32s3y7roiIiBQ8+ZXPWP5W+MhcGfzvtEr4zVHBUvKCGR+HuWkhZuw6sP3136BfIEadBzBq3ofh6ePYAMWl/LHnFKN+j+VMchoWAzrXD+H5xhXwcr/2SyoREchdPnNLc1gCbNiwgddee43Vq1djs9mAq4lq48aNef/996lfv/6tXuLO5VXo6p+Xkx0bh4iIiEgOTZ482dEhiMh1GEHlMdq8gNn4CcytizG3LoELpzBXfIu5bh5GjaYYdVphBAQ5OlRxAfdVCaRuSAAfLt3L/J0nmb7hCKv2n2Fom2rULhPg6PBExMXdUodlVFQUzZo1w8PDg86dOxMWFgbA7t27mTlzJmlpaSxfvpyIiIg8C9gRHNZhmZqCbfwzAFj6TcFw98i3a4uIiEjBog4916b3T24H80oa5u7VmJsXwpljf201oFJ9LHXbQOmq2XZKi/yvlftOM/K3GE4lpWEAj9crQ58mFfH2ULeliPxXbvKZW150p3Tp0sTGxvL555/z4osv8uKLL/L5558TGxtLcHAwb775Zq7OOX78eEJDQ/Hy8iIyMpL169ff8PixY8dStWpVvL29CQkJoX///ly+fPmWzuk0PLwgM0FIVZeliIiIiIjkHcPdA8tdzbF0eR/Lw69DubsAE/ZtwDZnBLYZb2HbtRozI93RoYqTa1KpOHN7RvLvWqUwgZnRR3l88nqiD59zdGgi4qJuaUh4VFQUQ4cOpWTJktfsCwoKonfv3rzzzjs5Pt/s2bMZMGAAX3zxBZGRkYwdO5ZWrVoRGxtLiRIlrjl+xowZDBo0iEmTJtGwYUP27NlDt27dMAyDMWPG3NQ5nYlhWMCzEFxOujosvHARR4ckIiIikkWPHj0wDIOvvvoKq9VKjx49/vE1hmEwceLEfIhORHLCMAwIDccaGo555ijmpkWYu1ZBfBzmos8wV83EqH0/xl0tMLx9HR2uOClfL3eGtQ3j/mol+L9FMRw9f4neMzfTsW5p+jWtiI/HLc9IJyJ3kFv6F8NisZCefv1v2zIyMrJMvP5PxowZQ69evejevTsAX3zxBfPnz2fSpEkMGjTomuPXrl1Lo0aN6Ny5MwChoaE88cQTREVF3fQ5nU5mwTI1ydGRiIiIiFxj2bJlWCwWbDYbVquVZcuW/eMQUg0xFXFeRrEyGPc/g3lvR8xtyzC3/A7J5zDXzMGM+hGjemOMOq0xipV2dKjipBpWKMacnpF8/Mc+5m09zpxNx1i9/wxDWlcjIrSoo8MTERdxSwXLhg0bMn78eDp37ky5cuWy7Dt8+DCfffYZjRo1ytG50tLSiI6OZvDgwfZtFouFli1bsm7duutef/r06axfv56IiAgOHDjAggULePrpp2/6nACpqamkpqban1+4cCFH93BbeBWCRLTwjoiIiDilgwcP3vC5iLgmw9sPI7IDZv1/Ycb+iblpASQcxNy2FHPbUggNvzrPZbla+hJCrlHY0403W1ejZbUSvLMwhuOJl3l+9hYeqR3Mi80qUdhT3ZYicmO39K/EyJEjadKkCdWqVeOhhx6iSpUqAMTGxvLTTz9htVoZNWpUjs51+vRpMjIyCArKuiJdUFAQMTEx2b6mc+fOnD59mnvvvRfTNElPT+e5557jjTfeuOlzAowaNYq33347R3Hfdn+tFG6mpqA0QERERERE8pNhdcOofi9mWCM4FoMteiHsj4aDW7Ed3ArFymDUbYNRrZEWCZVrRIYWZXaPCD5ZsZ+5m4/x/ZbjrN5/hrdaV6NhhWKODk9EnNgtFSzr1KlDVFQUb775Jj///DMpKSkA+Pj40Lp1a4YPH07x4sXzJNDsLF++nJEjR/LZZ58RGRnJvn37eOmll3jnnXcYMmTITZ938ODBDBgwwP78woULhISE5EXIuWZ4FsIEdViKiIiIy0lKSuLcuXOYpnnNvrJlyzogIhG5WYZhQJkwrGXCMM/HY25ehLljBZw5irl4AubqWRh3tbw612WhAEeHK06kkKcbgx6oSouqJXhn0W6Onb9Mv7lbaV+rFP2bV8LXy93RIYqIE7rlPuzq1avzww8/YLPZOHXqFACBgYFYLBbeffddhg4dSkZGxj+ep3jx4litVuLj47Nsj4+Pz3ZRH4AhQ4bw9NNP88wzzwBQq1YtkpOT6d27N2+++eZNnRPA09MTT0/Pf4w5X/zVYalVwkVERMQVXL58mbfffpuJEydy5syZ6x6Xk/xQYPz48YwfP14/L3EqRkAQxn1dMRs+hrnjD8zNv8GF05hRP2Bu+BmjWsOrXZclQh0dqjiRBuWKMLt7JJ+u3M/s6KP8tP0E6+LO8kbrqjSuePsanUTENeXZxBEWi+Waode54eHhQb169Vi6dCkdOnQAwGazsXTpUvr27Zvta1JSUq5Z1MdqtQJgmuZNndPpeP5VsFSHpYiIiLiAF154galTp9KhQwcaN25MkSJFHB2SS+vTpw99+vThwoUL+Pv7OzockSwMTx+Meg9i1mkN+zZi27QQju/B3LXq6irjZapjqdcGytfByMVirFJweXtYebVlFVpWK8GIBbs5fO4SL3+3jQdrlmRgi8r4qdtSRP7iVDPdDhgwgK5du1K/fn0iIiIYO3YsycnJ9hW+u3TpQunSpe3zYrZr144xY8ZQp04d+5DwIUOG0K5dO3vh8p/O6fTUYSkiIiIuZN68eTzzzDN8+eWXjg5FRPKJYbFClUisVSIxT+zD3LQIc8+fcHQXtqO7wD8Io25rjBpNMTy8HB2uOIE6ZQKY0T2CL1Yd4NsNR5i/4yR/xp3lzVZVaVo50NHhiYgTcKqCZadOnTh16hRDhw7l5MmT1K5dm0WLFtk7Nw8fPpylo/Ktt97CMAzeeustjh07RmBgIO3atePdd9/N8TmdnqcPAKY6LEVERMQFGIZB3bp1HR2GiDiIUaoSxoN9MZs8gbnl96sriifGY/4xFXPtXIya92HUaYXhpyHAdzpvdyv9m1emRdUSvL1gNwfPpjBg3nZaVw/i1ZZVCPBWt6XIncwws5sFPY/kZg5LZ5Y5BCcxMRE/P798vba5509sv46D0lWxdhqWr9cWERGRgiO/8plu3bqRnJzM3Llzb9s17kSOzEdFboV55TLmzlWYmxbC+ZNXNxoWjMoNrs5zGVzFsQGKU0hNz+DL1XF8s/4wNhOK+rjbF+oRkYIjN/lMrjssN23alONjjx8/ntvTy//SHJYiIiLiQoYMGULHjh3p3bs3zz77LGXLlrVP1fN3RYsWdUB0IpLfDHcvjNr3Y4a3gLgtV+e5PLwTc08U5p4oKFkJo14bjMoRV4eWyx3J083Ki80q2bst959O5rUfd9Cyaglev78KRQt5ODpEEclnue6wtFgsGIaRo2NN08QwDHVY3gIz/gC2b9+CQkWwPjs+X68tIiIiBUd+5TN/n77nRjmjq+eH+U0dllKQmKcOY25aiBmzBjLSr270LYZR+wGMWs0xMufxlztSWrqNr9ceZMqfh8gwTQK83XmtZRUeCCuR41qEiDin29phOXny5JsOTG5CZodlaopj4xARERHJgaFDh+oXShG5ISOwLEarZzHvfRxz62LMrUvg4hnMVTMx/5yHUb3J1UV6ipRydKjiAB5uFl5oUoHmVQN5e8Fu9iQk8cYvO/k9Jp7BD1SleGFPR4coIvngts5hWVA4tMPychK2z3oDYHlpGobVqdZJEhEREReRH/nMlStX2L17N0WLFqVMmTK35Rp3KnVYSkFmpqdhxqy9Os/l6SN/bTWgQh0sddtASHV9EXKHupJhY/K6Q3y97iAZNhM/LzdebVmFNtWD9JkQcUG5yWcsN9wrjufpA/z1D7HmsRQREREnZrFYqFevHvPmzXN0KCLiQgw3Dyw1m2F5+j0sj74B5esAJhzYhO27d7FNfwPbzhWY6VccHarkM3erhd73lufbrg2oFuTLhcvpDPl1F/2/38api6mODk9EbiMVLJ2cYVjA0/vqk9QkxwYjIiIicgNWq5Vy5cqRmqpfIkUk9wzDwChbE+tDr2LpNhojvCW4ecKpQ5i/fYnt6xexrZuHmXLB0aFKPqtcojBTnq7HC40r4G41WLX/DI9OjOLnbcfRoFGRgkkFS1fgpZXCRURExDX069ePr776irNnzzo6FBFxYUbRYCwtemDp9QnGvY9D4aKQkoi57jtsE/ph+/0rTPvwcbkTuFst9GwYyrddG1CjlC9Jqem8vTCGF+du5eSFy44OT0TymCZEdAWehYBTKliKiIiI08vIyMDT05OKFSvy6KOPEhoaire3d5ZjDMOgf//+DopQRFyJ4V0YI+LfmPXaYu5djxm9AOIPYO5YjrljOZSrhaVOaygffnV0mhR4FQMLM+mpeny74QhfrIpjbdxZOk6M4uX7KvFQeLDmthQpILToTg44epLzjLnvwpGdGG1ewBJ2b75fX0RERFxffuUzFss/FwwMwyAjI+O2xVAQOTofFXEWpmnCib3YohfAvg2Q+ets0WCMOq0xqjfGcNcq0neKuDPJvL1gN9uPX50mIDK0CENaV6OUv/c/vFJEHCE3+Yw6LF2BhoSLiIiIi4iLi3N0CCJSgBmGAcFVsAZXwUw8hbnlN8ztf8DZ45hLJ2GumYNxV3OM8AcwfIs6Oly5zcoXK8TEJ+sxc+MRPlt1gKiD5+g4aT0vNqvII7VLY1G3pYjLUsHSBRiehTABUlWwFBEREedWrlw5R4cgIncIwz8Qo+lTmHc/jLlzBebm3yAxAXP9z5gb52NUicSo2xajZAVHhyq3kdVi8FREWRpXKs6IhbvZcjSR937fw5KYBIa0CaNMgLotRVyRJvlwBZkdlqkpjo1DRERERETEyRiePljqtsHSfQyWdv2hdDWwZWDGrMU24y0yZr99df5Lm83RocptVK6oDxM612Vgi8p4uVvYePg8nSZFMSv6CDbNhCfictRh6Qo0JFxERERcyLZt2/jkk0/YtGkTiYmJ2P6nSGAYBvv373dQdCJSUBkWC1RugLVyA8z4A5ibFmHGroNjsdiOxYJ/4NV5Lms0xfD0cXS4chtYDIMn6odwb8VivLMohujD5/lgyV6WxCQwtE0YZYvqfRdxFeqwdAWeVwuWpgqWIiIi4uSWL19OREQEv/76K8HBwRw4cIAKFSoQHBzMoUOHKFy4ME2aNHF0mCJSwBlBFbC0eQHLM+MwIjuAV2FIPIW5/BtsE/phW/4NZmKCo8OU2ySkiA9fPF6H1++vgre7lc1HE3l88nq+3XCYDJu6LUVcgQqWrsA+JFwFSxEREXFuQ4cOpUKFCsTGxjJ58mQA3njjDVavXs3atWs5evQoHTt2dHCUInKnMAoXwdKoI5Zen2C07AlFS0PaJcxNC7FN6k/Gzx9hHou5uvq4FCgWw6Bj3TLM6RlBRLkipKbbGLNsH898G03cGf1uLeLsVLB0AYanhoSLiIiIa9i0aRM9e/bEz88Pq9UKQEZGBgCRkZE8++yzDBkyxJEhisgdyHD3xHJXCyxd/4PlodehXC0wTdi3AdvsEdhmDMG2ezVmRrqjQ5U8FuzvzWedavNmq6oU8rCy7fgFOk/ewNSoQ6RrXlMRp6WCpStQh6WIiIi4CDc3N3x9fQEICAjA3d2dhIT/DrusUKECu3btclR4InKHMwwLRvlwrI8MxtLlfYya94HVHeIPYC78DNvXL2Fb/xPmpSRHhyp5yDAMHq5dmtk9I7mnfFHSMmyMW76fHtOj2X9K77WIM1LB0hWow1JERERcRKVKldi7dy9w9RfEatWq8cMPP9j3z58/n5IlSzoqPKeSkpJCuXLlGDhwoKNDEbkjGcXLYHmg19Xh4g0fg0IBkHwOc/VsbBP6YlsyEfPsMUeHKXmolJ8XnzwWzrA21Sjs6cbOExd5cuoGJq07qG5LESejgqUr8PprJbMrlzVEQURERJxa27ZtmTlzJunpV3OWAQMGMG/ePCpXrkzlypX5+eefefbZZx0cpXN49913ufvuux0dhsgdz/Dxw3L3Q1h6fozR+jkILAfpaZjblmKb8ioZ8/6DeWi75rksIAzD4N93BTO3ZyT3VizGlQyT8SsP0HVaNHsT1G0p4ixUsHQFmR2WAKkpjotDRERE5B8MGTKErVu32uev7Nq1K9OmTaNmzZqEh4czadIkXn/9dQdH6Xh79+4lJiaGNm3aODoUEfmL4eaOpXoTLE+NxPLYEKhYDzDg4FZs34/CNm0Qtu1/YKanOTpUyQMlfD0Z+8hdjHgwDD8vN2LiL/LU1A18tSaOKxnqthRxNBUsXYBhsYKH99UnmsdSREREnJi7uzvFihXDMAz7tqeeeooffviB7777jm7dut3S+Y8dO8ZTTz1FsWLF8Pb2platWmzcuPEWo/6vlStX0q5dO4KDgzEMgx9//DHb48aPH09oaCheXl5ERkayfv36XF1n4MCBjBo1Kg8iFpG8ZhgGRkgY1vavYOn+IUbtVuDuCWeOYC6egG3Ci9jWzMVMPu/oUOUWGYbBgzVLMadnJM0qFyfdZvLl6ji6TNtITPxFR4cnckdTwdJVaB5LERERcSGpqamsW7eOn376idOnT+fJOc+dO0ejRo1wd3dn4cKF7Nq1iw8//JAiRYpke/yaNWu4cuXKNdt37dpFfHx8tq9JTk4mPDyc8ePHXzeO2bNnM2DAAIYNG8amTZsIDw+nVatWWRYXql27NjVr1rzmcfz4cX766SeqVKlClSpVcvkTEJH8ZhQpiaV5Vyy9PsVo0hl8i8OlC5hRP2D7+kVsi77APHXI0WHKLQos7Mnoh2rxbrvq+Hu7sychiS7TNvL5qgPqthRxEMPURBz/6MKFC/j7+5OYmIifn59DYsj4ZjCcOoTl4dcxQsMdEoOIiIi4rvzMZ8aNG8fw4cNJTEwEYPHixTRv3pzTp09TrVo13n//fXr06JHr8w4aNIg1a9awatWqfzzWZrNRt25dKleuzKxZs+xD1GNjY2natCkDBgzgtddeu+E5DMPghx9+oEOHDlm2R0ZG0qBBAz799FP7tUJCQujXrx+DBg36x9gGDx7M9OnTsVqtJCUlceXKFV555RWGDh16zbHjx49n/PjxZGRksGfPHofmoyICpi0D9m3AFr0QTuz9746Q6ljqtoEKdTAM9QW5sjPJafxncSxLY08BULF4IYa3DaN6Kf3bK3KrcpOP6l9SV/FXh6WpDksRERFxYpMnT+bll1+mdevWTJw4McsiFcWLF6d58+bMmjXrps79888/U79+fR577DFKlChBnTp1mDBhQrbHWiwWFixYwObNm+nSpQs2m439+/fTvHlzOnTo8I/FyutJS0sjOjqali1bZrlWy5YtWbduXY7OMWrUKI4cOcLBgwcZPXo0vXr1yrZYCdCnTx927drFhg0bbipeEclbhsWKUeVurE+8jeWJERhV7wbDAkd2YfvpQ2yTX8G2+TfMtMuODlVuUrFCHrzfoRbvta9JER939p9Opts30Xy6Yj+p6RmODk/kjqGCpavw+mtIuOawFBERESf24Ycf0r59e2bMmEG7du2u2V+vXj127tx5U+c+cOAAn3/+OZUrV+a3337j+eef58UXX2Tq1KnZHh8cHMyyZctYvXo1nTt3pnnz5rRs2ZLPP//8pq4PcPr0aTIyMggKCsqyPSgoiJMnT970eUXE9RilKmF58EUsPcdi1P8XePrA+XjMP6Zim9AX28oZmBfyZkoMyX/3VyvB3J6RtAorQYZpMvnPQzw5ZQPbjyc6OjSRO4KbowOQnDE8C2GC5rAUERERp7Zv3z5efPHF6+4vWrQoZ86cualz22w26tevz8iRIwGoU6cOO3bs4IsvvqBr167ZvqZs2bJ88803NG3alAoVKjBx4sQsCwI52q0uQiQijmf4Fcdo0hnz7ocxd63E3LQIzp/E3PgrZvQCjCoRGHXaYARXdnSokktFfDwY+e+a3F/tFKN+jyXuTAo9pkfzZIOyPHdvebzcrY4OUaTAUoelq/DSojsiIiLi/AICAm64yM6uXbsoWbLkTZ27VKlSVK9ePcu2sLAwDh8+fN3XxMfH07t3b9q1a0dKSgr9+/e/qWtnKl68OFar9ZpFe+Lj42/6vkSkYDA8vLDUfgBL99FY2r8CIdXBtGHG/olt1jAyZg7FFvvn1XkwxaXcVyWQuT0jaVsjCJsJ36w/TOcpG9hy9LyjQxMpsFSwdBJJqelsOXqemPiL2R+gIeEiIiLiAtq2bctXX33F+fPnr9m3c+dOJkyYwL///e+bOnejRo2IjY3Nsm3Pnj2UK1cu2+NPnz5NixYtCAsLY968eSxdupTZs2czcODAm7o+gIeHB/Xq1WPp0qX2bTabjaVLl3LPPffc9HlFpOAwDAtGxXpYH3sLy1MjMWo0AasbnNiHOX8cton9sW38VesTuBh/b3fe+VcNPnrkLgILe3DobArPfLuJD5fu5dIVFaFF8poKlk7i+y3H6PntJqZFXadDwNMH0KI7IiIi4tz+7//+j4yMDGrWrMlbb72FYRhMnTqVp556ivr161OiRInrLjDzT/r378+ff/7JyJEj2bdvHzNmzOCrr76iT58+1xxrs9lo06YN5cqVY/bs2bi5uVG9enUWL17M5MmT+eijj7K9RlJSElu2bGHLli0AxMXFsWXLlixdnAMGDGDChAlMnTqV3bt38/zzz5OcnEz37t1v6r5EpOAySoRiafUclmfGYdz9MHj7wcXTmCtnXJ3nctlUzHOa/9aVNKlUnLk9I/l3rVKYwIyNR3h80nqiD59zdGgiBYph/n3pRslWbpZdv1mr95/mpe+2USmwELN7RF6z37Z7DebC8RBSA+tjb96WGERERKTgyo98JlNCQgJvvPEG8+bNs3da+vr68sgjj/Dee+9RokSJmz73r7/+yuDBg9m7dy/ly5dnwIAB9OrVK9tjFy9eTOPGjfHy8sqyffPmzQQGBlKmTJlrXrN8+XLuu+++a7Z37dqVKVOm2J9/+umnfPDBB5w8eZLatWszbtw4IiOvzeHySn6+fyJy+5jpaZgxazGjF8KZI39tNaBiXSx120CZMKeaZ1dubM3+M7z7WwzxF1MB6Fi3NP2aVsTHQ8uFiGQnN/mMCpY5kB8J4okLl/nX52txsxisHtAUd2vW5lczbgu2H96HEqFYnxp5W2IQERGRgstRBa9Tp05hs9kIDAzEYrGQnJxMYmIiwcHB+RZDQaCCpUjBYpomHN6BbdNCiNvy3x2B5TDqtcGocg+Gm7vD4pOcu5iazsd/7OOHrccBCPb3YkibakSUK+rgyEScT27yGQ0JdxIlfT0p5GEl3WZy+GzKtQd4atEdERERcT2BgYEEBQVhsVxNO8eOHUtISIiDoxIRcSzDMDDK1cL60GtYun2AcVcLcPOAU4cwF32BbeJL2P6ch5lywdGhyj/w9XTjrdbVGN+xNqX8vDieeJnnZ21h5G8xJKWmOzo8EZelgqWTMAyDisWvFiX3n86mKKlFd0RERERERAoco2hpLC17Yun1Cca9naBQEUg+j7n2O2wT+v1/e/cdH1WV/nH8c+609BBaQm+C9F5EVBRQQH4qaxcLIpbdxYqubVdsq9jWdVVWV0Wxg72toAiCiyK9dwQBKaGnt5l7fn9MiEZaUMhMku/79ZpXMnfOvfeZnACHZ55zDu6XL2J3/hTpMOUwTmhSnQlXdef8TvUAeH/hFi56eRYz1++KcGQiFZMSllGkWa0EANbuOFTCMhfruuUYlYiIiIiIiBxrJjYRp/s5OFf/CzNwBKQ2hVARdunXuK/dTuj90dj1i9CqbtErPuDlrjOO5/mLO1EvOYZtmQVc/84iHpi4gixVW4ocEa0EG0V+rrDM3v/FfVPCAQpyITahnKISERERERGR8mI8XkyrXtiWJ8KWVbjzJsIPc2HDEtwNS6B6XUzngZhWJ2F8gUiHKwfQrVEKE67qwbPf/MD4eT/x8eKtzFy3m78OOJ6TmtWMdHgiFYIqLKNIs1oHnxJuPF7Y94+RpoWLiIiIiIhUasYYTL2WeM6+Beeqf2I6DwR/LOzegv1qbHi6+IwJ2Ow9kQ5VDiDW7+Ev/Vrw0pDONEiJZXt2ATe9t5h7/7uczPyiSIcnEvVUYRlFmhdPCf9pTx55RSFifZ7SDQLxUFSgjXdEREQkqsyfP7/Mbbds2XIMIxERqZxMcm3MqZdje56HXTYdu2ASZOzAzv4YO/czTIsTwruLpzaNdKjyK50aVOPtYd157n/reGvOJj5buo3vf9zN3WccT+/mtSIdnkjUUsIyiqTE+ake52N3bhHrd+bQus6vtniPiYfs3aqwFBERkajStWtXjDFlamutLXNbEREpzQTiMJ0HYjv2hx/m4s6fCJtXYVd+i135LdRridN5IDTrgnE0oTJaxPo8jOzTnL4tanP/xBVs2J3LyA+WMLB1Krf1a0G1WF+kQxSJOkpYRplmtRLYvWEPaw+UsAxop3ARERGJPq+88kqkQxARqVKM40Dz7niad8duW4edPxG7+nvYvBJ380pIro3p1B/TpjcmEBfpcKVYh/rJvHVlN174dj2vz97IxOXpzN6whztPb0Gf42tHOjyRqKKEZZRpVjOeORv28MOOA2y8U7xTuM3PQXUJIiIiEi2GDh0a6RBERKosk9YUc+YI7MmXYBd9iV08FTK2Y6e9jv3uPUy70zAd+2OSNf04GsT4PNx46nH0aVGb+z9fwbpdOfzlo6Wc3rI2d5zegpQ4f6RDFIkKqhGPMj/vFH6AjXf2VVhqDUsRERERERH5BZNYHeeki3GueQbT9yqoXhcK87DzPsd9+WZCnz6F3bwKa22kQxWgbd0k3ryyG1f1bITHGCav3M75L83iyxXp6iMRVGEZdY4r3njnhx0HSErGaEq4iIiIiIiIHJzxBTAd+mHb94EfF4fXudywBNbMxl0zG1KbYrqciWneHeNRSiCS/F6HEac0o0+LWtz3+QrW7sjhrk+WMXnldu4843hqxKvaUqouVVhGmabFFZbbswvIzC8q/aIqLEVERERERKQMjHEwTTriOe8unCsexbQ9DTw+SF+H/fxZ3LE3487+BJt3gOXIpFy1SkvijaHduObExngcw9TVO7jgpe/5fNk2VVtKlaWEZZRJCHhJSwoAB5gWrgpLEREREREROUKmZgOcM64JTxc/8XyIS4bs3dgZ43FfvAF3ysvY3VsiHWaV5vM4/PHkprx+RVeOr51ARn6Qez5bzq0fLGFHVkGkwxMpd0pYRqGDTgsP/LzpjoiIiIiIiMiRMHFJOCeci3P105j+f4RajSBYgF30Fe642wh9+Bh2wxJV9UXQ8amJvHpFV/50chO8jmH62p1cMHYWny7Zqn6RKkUJyyi0b+Odtb/aKdyowlJERERERER+J+P14bQ5Beeyh3HO/ys06wIYWL8Q9/3RuK/fibt0GjZYGOlQqySfx+HqE5vwxpXdaJWWSFZBkPs+X8FN7y0mPTM/0uGJlAslLKNQSYXlwaaEq8JSREREREREfidjDKZhGzzn3Ioz7AlMxzPAF4Cdm7BfvoD70o24372HzcmIdKhVUvNaCYy7vAvX926Kz2P4dt0uLhg7i48WbVG1pVR6SlhGoX0Vlj/syC79l9C+TXcKciMQlYiIiIiIiFRWJqUOTp8rw+tcnnwJJNaA3Ezs9x/gvnQD7hfPY3dsjHSYVY7XcRh2QmPeurI7beskkVMY4sFJK7n+nUVsVbWlVGJKWEahxjXicAxk5AfZmfOLEvxAXPhrQQ7WupEJTkRERERERCotE5OA0+0snOFPYQbdAGnHQSiIXfYN7ut3Enr3Iey6+fo/aTlrWjOely/rwk2nHkfA6/D9j7u5cOws3l+4WdWWUil5Ix2A7C/g9dAgJY4Nu3P5YUcOtRLCu4aXTAm3Fgrzf05gioiIiIiIiBxFxvFgju8Jx/fEblmDnT8Ru2Y2bFqGu2kZVEvDdB6AaX0Kxh8T6XCrBI9juKJHQ3o3r8n9n69g0eYMHv5iFZNXbueeAS2pVy020iGKHDWqsIxSx+2bFr7z5413jNcPHl/4idaxFBERERERkXJg6jbH+b8bw1WXXQaFi2f2bsNOHYf74vW437yFzdoV6TCrjEbV43hxSGdu7dOcgNdhzoY9XPTybN6Z/xOuqi2lklDCMko1q7VvHcuDbLyjncJFRERERESkHJmkmji9L8W55lnMaUOhWioU5GLnfob70k24/30au3VtpMOsEjyOYUi3Bky4qjudG1QjryjEo5NX88e3F7Bpj/a9kIpPCcsotW+n8LW/qLAEtFO4iIiIiIiIRJTxx+B06o9z5T9wzrkVGrQG62JXfY/79ihCb9+Lu+p7rBuKdKiVXoOUOP5zSSfuOL0FsT4P8zbt5aKXZ/PWnE2EXFVbSsWlNSyjVMlO4TtzcK3FMSb8QkAJSxEREREREYk84zjQrAueZl2w23/ELpiEXfkdbF2D/e8abGJNTKczMG1Pw+wrvpGjzjGGCzvXp1fTGjw4aSVzNuzhH1PX8NWq7dx7ZisaVdf+F1LxqMIyStVPicXvccgvctmSkf/zC8V/yVtNCRcREREREZEoYWo3xun/R5yrn8accC7EJkHWTuw3b4XXuZz6KnbPtkiHWanVqxbLcxd15O7+xxPn97BocwaXvDKb12ZtVLWlVDhKWEYpr+PQpEb4U5Afdvxi452SCsvsA50mIiIiIpXImDFjaN26Nd26dYt0KCIiZWLiq+GceD7ONU9jTr8GatSHogLswi9wX7mV0Mf/wG5ajtXmMMeEMYbzOtbjneE96NE4hYKgy7+mreWqN+axfqcKn6TiUMIyipVsvPPLv1RKNt3RIroiIiIild2IESNYvnw5c+bMiXQoIiJHxHj9OO1Ow7niUZzz7oImHQELP8zDfffvuG/cjbv8G2ywKNKhVkp1kmIYc2FH7hnQkni/h6VbMxkybg6vfP8jQdeNdHgih6WEZRRrVrN4451f7hQe0C7hIiIiIiIiUjEYYzCN2uH5w+04Vz6Oad8XvH7YsQE76XncsTfhfv8hNi8z0qFWOsYYBneoy7vDe9CraQ0KQy7PTl/HsNfnsXaHZm1KdFPCMor9XGH5i79ItEu4iIiIiIiIVECmej2cfsNxrnkGc9LFEJ8COXux372L+8INuF++iN35U6TDrHRSk2L41/ntuX9QKxIDXpZvy+LScXN48dv1FIVUbSnRSQnLKNYwJbyG5ZaM/J/X9yiusLRKWIqIiIiIiEgFZGITcbqfjXP1vzADR0BqUwgVYZd+jfva7YTeH41dv0jrXB5Fxhj+r20d3h3eg97H1SToWp6fsZ6hr89l9fasSIcnsh9vpAOQg6udGAAgtzBEdkGQxBgfJiYeC5oSLiIiIiIiIhWa8XgxrXphW54IW1bhzpsIP8yFDUtwNyyB6vUwnQdgWp2E8QUiHW6lUCsxwD/ObcekFek8Pnk1q9KzuezVuVx1QiOGn9gYn0d1bRId9JsYxWJ9HqrF+gDYllkQPhjQlHARERERERGpPIwxmHot8Zx9C85V/8R0Hgj+WNi9GfvVWNwXb8CdMQGbvSfSoVYKxhgGtk7jneE96NOiFiHX8uJ3P3L5q3NZsU1riUp0UMIyyqUmhT9F2paVHz4Q8/OmO9ZabNZu7I+Lced9jrvsmwhFKSIiIiIiIvL7meTaOKdeHl7nsvdlkFQL8rOxsz/GfelG3In/xqavj3SYlULNhACPDW7L6LPbUC3Wx5od2Qx9bR5jvvmBwqDWtpTI0pTwKJeWFMOq9Gy2ZRYnLPdVWOZl4f77GijILdXe1mqIqd24fIMUEREREREROYpMIA7T5UxspwHww1zc+RNh8yrsihnYFTOgXkuczgOhWReMo1qs38oYwxmtUunWKIXHJq/my5XbeXnmBqat3sm9Z7aibd2kSIcoVZQSllEuLTEGgPR9U8JjE8Ol8YV54WSlcaBaKuRmhte13JsOSliKiIiIiIhIJWAcB5p3x9O8O3bbOuz8z7GrZ8HmlbibV0JybUyn/pi2p2L8sZEOt8JKifMz+py2nN5yO6O/XMW6XTkMe2Mul3dvyHUnNSHg9UQ6RKlilLCMciVTwosrLI3Xh3PB37B7tmJq1IeUOhivj9CnT8Ga2djs3ZgIxisiIiIiIiJyLJi0ppgzr8eePAS76Evs4qmQsR077XXsd+9h2p2G6dgfk1wr0qFWWH2Or03nhik88dVqJi5P59VZG5m2Zif3ndmK9vWSIx2eVCGqm45yJRWWWQUlx0xqE5yWJ2JqNcR4w5vymITq4Rez95Z3iCIiIiIiIiLlxiRWxznp4vA6l32vgpQ6UJiHnfc57ss3E/r0KezmVVhrIx1qhVQt1sffz2rDP89rT80EPxt253LVG/N4cuoa8opCkQ5PqoioTFiOGTOGxo0bExMTQ48ePZg9e/ZB25566qnhHcV+9Rg0aFBJmyuvvHK/1wcMGFAeb+V3+3WF5UElpIS/Zu8+xhGJiIiIiIiIRJ7xBXA69MO58nGcwX+Bhm3BWlgzG3fC/bhv3YO78jtsKBjpUCukU46rybvDe3BW2zQs8OacTVzyymwWbNob6dCkCoi6KeETJkxg5MiRPP/88/To0YOnnnqK/v37s2rVKmrXrr1f+w8++IDCwsKS57t27aJDhw5ccMEFpdoNGDCAV155peR5IBA4dm/iKEpLCldYbs8qIORaPM5BJnwXV1ja7D3lFZqIiIiIiIhIxBnjQNNOeJp2wu7YiF0wCbviW0hfh/38WWxCdUzHMzDt+mBiEyIdboWSFOPjvkGtOb1lKg99sZJNe/K45q35XNilPjec0oxYv9a2lGMj6iosn3zySa655hqGDRtG69atef7554mLi+Pll18+YPvq1auTlpZW8pg8eTJxcXH7JSwDgUCpdikpKeXxdn63mgl+HANB17I7p/Cg7X6eEq4KSxEREREREamaTK2GOGdci3PN05ie50NcMmTvxs4Yj/viDbhTXsbu3hLpMCucXs1qMGF4Dwa3r4MFJsz7iYtensWcDSqakmMjqhKWhYWFzJs3j379+pUccxyHfv36MXPmzDJdY+zYsVx88cXEx8eXOj5t2jRq167N8ccfz5/+9Cd27dp10GsUFBSQmZlZ6hEpXsehVkLxtPCsQ0wLL5kSvkfrdIiIiIiIiEiVZuKScXqei3P105j+10GtRhAswC76CnfcbYQ+fBy7YYn+/3wEEgNe7hnYijEXdiQtKcDmjHz+OH4BD3+xipwCTbuXoyuqEpY7d+4kFAqRmppa6nhqairbtm077PmzZ89m6dKlXH311aWODxgwgNdee40pU6bw6KOPMn36dAYOHEgodODFYkePHk1ycnLJo0GDBr/9TR0F+6aFb8ssOHijfQnLYAEU5JZDVCIiIiIiIiLRzXh9OG1641z2MM75f4WmnQED6xfgvj8a9/U7cZdOwwYPPqNRSjuhSXXeuaoH53WsB8D7Czdz4cuz+H69ZnzK0RN1a1j+HmPHjqVdu3Z079691PGLL7645Pt27drRvn17mjVrxrRp0+jbt+9+17nrrrsYOXJkyfPMzMyIJi3TkgIs2gzph9h4x/gCEIgLJyuz90BM/EHbioiIiIiIiFQlxhho2AZPwzbYPVux8ydhl30DOzdhv3wBO2M8pn0/TIfTMfHJkQ436sUHvNzd/3j6tazF3yeuZHNGPiPeWcjg9nW4uU9zEgOVKt0kERBVFZY1a9bE4/GQnp5e6nh6ejppaWmHPDcnJ4fx48czfPjww96nadOm1KxZk7Vr1x7w9UAgQFJSUqlHJKUm7quwPNxO4VrHUkRERERERORQTEodnL7DcK59BnPyJeH/S+dmYr//APelG3C/eB67Y2Okw6wQujeqzvirunNRl/oAfLR4KxeNncW3Pxx8GT6RsoiqhKXf76dLly5MmTKl5JjrukyZMoWePXse8tx3332XgoICLrvsssPe56effmLXrl3UqVPnd8dcHvZNCU/POsSUcCiZFm5ztOitiIiIiIiIyKGYmAScbmfhDH8KM+gGSDsOQkHssm9wX7+T0LsPYdfNx1o30qFGtTi/l9v7teDFIZ2oXy2W9KwCbnxvEff+dzmZ+UWRDk8qqKhKWAKMHDmSF198kVdffZUVK1bwpz/9iZycHIYNGwbAFVdcwV133bXfeWPHjmXw4MHUqFGj1PHs7Gz+8pe/8P333/Pjjz8yZcoUzjnnHI477jj69+9fLu/p90pNKt505zAVlj/vFK6EpYiIiIiIiEhZGI8X5/ieeIY8gHPxfZgWPcAY2LQM96MncMf9BXfhZGzRYWY9VnGdG6Qw/qruXNq1AQb4bOk2Lhg7i2/W7ox0aFIBRd2iAhdddBE7duxg1KhRbNu2jY4dOzJp0qSSjXg2btyI45TOs65atYoZM2bw5Zdf7nc9j8fD4sWLefXVV9m7dy9169bljDPO4MEHHyQQCJTLe/q9jrTCUlPCRURERERERI6cqdsCU7cFNnMHdsGX2KVfw56t2KmvYL99B9O+D6bjGZjEGoe/WBUU6/Mwsm9z+h5fm/snrmDD7lxueX8xA1unclu/FlSL9UU6RKkgjLXWRjqIaJeZmUlycjIZGRkRWc9yT24h/Z6ZAcDMW0/F7z1wYay76CvslJehWRc859xaniGKiIhIlIv0eEZ+H/WfiEhk2MI87LJvsPMnQUbxfhvGwbTogek8EFPnuMgGGMXyi0L8Z8Z63pizEddCjXg/d55xPH1a1Ip0aBIhRzKeibop4bK/arE+AsVJyvSsQ+wUrgpLERERERERkaPG+GNxOvXHGfYPnLNHQv1WYF3sqpm4b48iNP4+7OpZWDcU6VCjTozPw02nHcfLl3WhaY14duUU8pcPl3DXx0vZk1sY6fAkyilhWQEYY8o2LVxrWIqIiIiIiIgcdcZxMMd1xXPhPTiXPYxpfTI4HtiyGvezf+G+fAvu3P9iC3IjHWrUaVc3mTev7MawExrhMYYvV27ngrGzmLxye6RDkyimhGUFkZZYho139iUsczKwoWA5RCUiIiIiIiJStZjajXEG/Ann6qcxPQZDbCJk7sR+8ybuC9fjfv0qdm96pMOMKn6vw/W9m/HqFV1oVjOePblF3PnxUm7/cAm7clRtKftTwrKCSN1XYZl5iArLuMTwJzxYyM0on8BEREREREREqiCTkILT60Kcq5/BnH4N1KgHRfnYBV/gvjyS0MdPYn9agbYO+VmrtCTevLIb15zYGI9jmLJ6Bxe89D2Tlm/Tz0lKUcKygkhLOnyFpTEOxFcLP9G0cBEREREREZFjzvj8OO1Ow7niMZxz74TGHQALP8zFfedB3Df/irv8f5oJWczncfjjyU15/YquHF87gYz8IH/9dDm3friEHdmHKNKSKkUJywoiNbEMa1gCaOMdERERERERkXJnjME0bo/n3Dtwhj6Oad8XPD7Y/iN20nO4L92IO+sjbF5mpEONCsenJvLqFV3540lN8DqG6Wt2csFLs/hs6VZVW4oSlhXFvk13DrmGJZSsY2lVYSkiIiIiIiISEaZGPZx+w3GufQbT68LwbMicvdhv38F94QbcyS9hd/0U6TAjzudxuKZXE964shut0hLJKghy739XcPP7i9l+uIItqdSUsKwgUsswJRzCa2gAqrAUERERERERiTATm4TTY3B4g54Bf4bajSFUhF0yFffV2wm9/wj2x0VVvqKwea0Exl3ehRGnNMXnMcz4YRcXjJ3Fx4u3VPmfTVXljXQAUjZpxVPCcwpDZBUESQwcpOv27RSuCksRERERERGRqGA8Xkzrk7CtesHmlbjzJ8HaubBhMe6GxVCjHqbTQEyrkzA+f6TDjQiv43BVz8ac2rwW932+gmVbM3lg4komr9zOXwe0pE7xzFOpGlRhWUHE+j0kx4STlOmHqrIsrrC0qrAUERERERERiSrGGEz9VnjOvgXnqicxnQaALwZ2bcZ+9RLuSzfgfvtOlV7mrWnNeF6+rDM3ndoMv8dh5vrdXDR2Fu8v3KxqyypECcsKpCzrWBpVWIqIiIiIiIhEPVMtFee0K3CufRbT+1JIqgl5WdhZH4U36Jn4b+z2HyMdZkR4HYcrejTi7WHd6FAvmZzCEA9/sYo/T1jIloy8SIcn5UAJywokNakMO4X/Yg1LffIgIiIiIiIiEt1MIA6nyyCcq/6J8383Q90W4IawK2bgvnE3oXcexK6di3XdSIda7hrXiOfFIZ25tU9zAl6H2Rv2cOHY2bwz/ydc5TwqNSUsK5C0smy8sy9hWVQAhfrUQURERCRa5ebm0qhRI2677bZIhyIiIlHAOB5Mi+54Lr4P55IHMMefCI4HflqB+8mTuK/cijt/EraK/V/f4xiGdGvAhKu606l+MnlFIR6dvJo/vr2ATXtyIx2eHCNKWFYgqcUb76RnHrzC0vhiIBAXfqJp4SIiIiJR66GHHuKEE06IdBgiIhKFTJ3jcAZdjzP8KUy3syAQDxnp2Gmv4b54A+70N7GZOyIdZrlqkBLHC0M6c3u/FsT4HOZt2svFr8zmrbmbVG1ZCSlhWYGUqcISSk0LFxEREZHos2bNGlauXMnAgQMjHYqIiEQxk1gD5+RLcK59BtN3GKTUgYJc7Lz/4o69Bfezf2G3rI50mOXGMYaLutRnwlU96NqwGvlFLv+Ysoar35zPht2qtqxMlLCsQPatYbkt63AJy/DGO1V5VzERERGp3B555BGMMdx8881H9brffPMNZ511FnXr1sUYw0cffXTAdmPGjKFx48bExMTQo0cPZs+efUT3ue222xg9evRRiFhERKoC44vB6XA6zpWP4wz+CzRsA9bFrp6FO/4+Qm+Nwl35HTYUjHSo5aJ+tVieu7gTd51xPHF+D4s2Z3DJK7N5ffZGQq6qLSsDJSwrkH27hKdnFhyy3NmowlJEREQqsTlz5vCf//yH9u3bH7Ldt99+S1FR0X7Hly9fTnp6+gHPycnJoUOHDowZM+ag150wYQIjR47k3nvvZf78+XTo0IH+/fuzffv2kjYdO3akbdu2+z22bNnCxx9/TIsWLWjRokUZ37GIiEiYMQ6maSc85/8V5/JHMG1OBY8Ptq3Ffv4s7tibced8is3PjnSox5xjDOd3qsc7V3WnR+MUCoIuT329luFvzmP9zpxIhye/kxKWFUitBD+OgaBr2Z1TePCGxRWWWsNSREREKpvs7GwuvfRSXnzxRVJSUg7aznVdRowYwZAhQwiFQiXHV61aRZ8+fXj11VcPeN7AgQP5+9//zh/+8IeDXvvJJ5/kmmuuYdiwYbRu3Zrnn3+euLg4Xn755ZI2CxcuZOnSpfs96taty/fff8/48eNp3Lgxt912Gy+++CIPPPDAb/hpiIhIVWZqNcTpfy3ONU9jep4HcUmQvRv7v7dxX7gBd8or2D1bIx3mMVcnOZYxF3bkbwNaEu/3sGRLJkPGzWHc9xsIVsGd1SsLJSwrEK/jUCth3zqWB994Z98allYVliIiIlLJjBgxgkGDBtGvX79DtnMch88//5wFCxZwxRVX4LouP/zwA3369GHw4MHcfvvtv+n+hYWFzJs3r9T9HcehX79+zJw5s0zXGD16NJs2beLHH3/kiSee4JprrmHUqFEHbDtmzBhat25Nt27dflO8IiJS+Zm4ZJye5+Fc/TTmjGuhZgMIFmAXTcZ95TZCHz2O3bgMW4k3pjHG8IcOdXlneA96Na1BYcjlmek/cNUb81i7o/JXm1ZGSlhWMKmJ4YRl+iHWsTSqsBQREZFKaPz48cyfP7/Maz/WrVuXqVOnMmPGDIYMGUKfPn3o168fzz333G+OYefOnYRCIVJTU0sdT01NZdu2bb/5ugczYsQIli9fzpw5c476tUVEpHIxXj9O21NxLn8E5/y/QtNOgIV1C3Dfewj3jbtxl03HBvdfLqWySEuK4V/nt+e+M1uRGPCybGsWl46bw0vfracopGrLisQb6QDkyKQlxbB4S+ahdwovWcNSCUsRERGpHDZt2sRNN93E5MmTiYmJKfN5DRs25PXXX6d37940bdqUsWPHYow5hpEemSuvvDLSIYiISCVjjIGGbfA0bIPdsxU7fxJ22TewYwP2i/9g/zce0+F0TIe+mLjkSId71BljOKtdHU5oXJ2Hv1zFN2t38tz/1jN19Q7uO7MVLWonRjpEKQNVWFYwqb/YeOeg9lVY5u7FuqGDtxMRERGpIObNm8f27dvp3LkzXq8Xr9fL9OnTefrpp/F6vaXWqfyl9PR0rr32Ws466yxyc3O55ZZbflccNWvWxOPx7LdpT3p6Omlpab/r2iIiIkebSamD03cYzjXPYE6+JJwvyM3AznwP98Ubcb94AbtjY6TDPCZqJQZ48tx2PPh/rUmO8bIqPZvLXp3Lf2asU7VlBaCEZQWzb0r4ISssY5PAOGAt5GSUU2QiIiIix07fvn1ZsmQJCxcuLHl07dqVSy+9lIULF+LxePY7Z+fOnfTt25dWrVrxwQcfMGXKFCZMmMBtt932m+Pw+/106dKFKVOmlBxzXZcpU6bQs2fP33xdERGRY8nEJuB0Owtn+FOYM6+HtGYQKsIum4b7+p2E3nsYu24B1lauRJ4xhjPbpPHO8B6c1qIWIdfywrc/cvmrc1mxLSvS4ckhaEp4BZNWXGG5LevgFZbGcSC+GmTvDj8Sq5dTdCIiIiLHRmJiIm3bti11LD4+nho1aux3HMJJxIEDB9KoUSMmTJiA1+uldevWTJ48mT59+lCvXr0DVltmZ2ezdu3akufr169n4cKFVK9enYYNGwIwcuRIhg4dSteuXenevTtPPfUUOTk5DBs27Ci/axERkaPLeLyYlidij+8JW9fgzvsc1s6BjUtxNy6FlDqYzgMwrU/G+Mq+BEu0q5kQ4PHBbZm8cjuPTl7Nmh3ZDH1tLkNPaMg1JzbB71U9X7RRwrKCSUsqQ4UlhMu8s3drHUsRERGpkhzH4eGHH+bkk0/G7/eXHO/QoQNfffUVtWrVOuB5c+fO5bTTTit5PnLkSACGDh3KuHHjALjooovYsWMHo0aNYtu2bXTs2JFJkybttxGPiIhItDLGQN0WeOq2wGbswC78ErtkKuzZip3yCnbGO5j2fTEdz8BUkiIoYwxntEqlW6MUHp28mskrt/PyzA1MX7OTe89sRZs6SZEOUX7B2Mq8r/1RkpmZSXJyMhkZGSQlRfYXeE9uIf2emQHAtyN7E+Pbf/oTQOiTf8LaOZjThuJ06l+eIYqIiEgUiqbxjBw59Z+IiBxrtjAPu3Q6dsEkyNgePuh4MC16YDqfiUlrGtkAj7Ipq7bzyJer2J1bhGPg8u4Nue6kJgS8B86zyO93JOMZ1bxWMNVifVSP8wGwMv3g6y2YfRvvqMJSRERERERERA7D+GNxOg/AGfYkztm3QP1W4IawK7/DfetvhMbfh109G+tWjnUu+x5fm3eG92BA61RcC6/O2siQV+aweLP2AokGSlhWMMYY2tVLBmDx5syDN0xICX/N3l0OUYmIiIiIiIhIZWAcB3NcNzwX3oNz6UOYVieB44Etq3E/ewr35Vtw532OLciNdKi/W0qcn4fOasM/zm1HjXg/P+7O5ao35vHk1DXkFYUiHV6VpoRlBdS+bjhhuWTLIbL+xRWWNkcVliIiIiIiIiJy5ExqE5yBf8a5+mlMj8EQkwCZO7DT38B98Qbcr1/D7k2PdJi/26nNa/Hu8B4MapuGBd6cs4lLXpnNgk17Ix1alaWEZQXUvl54nv/iLRkcbAlSU1JhqYSliIiIiIiIiPx2JiEFp9eFONc8g+k3HKrXg8I87IJJuK+MJPTJP7E/rTxojqIiSI718cCg1jx1fntqJwTYtCePa96az+NfrSavUNWW5U0JywqoVVoSHsewM7vw4LuFJ2oNSxERERERERE5eowvgNO+L87Qx3DOvQMatQdrYe0c3HcewH3rb7grZmBDwUiH+pud3Kwm7wzvzjnt6mCB8fN+4qJXZjF3o/Ir5UkJywoo1uehRe0E4BDrWMYXV1gW5mEL88opMhERERERERGp7IwxmMYd8Jx3J87QxzDtTgOPD9LXYyf+G/elm3BnfYTNO/hmwdEsMcbHqDNb8cwFHUhNDLB5bz7Xvb2AR75cRW5hxU3GViRKWFZQ7fdtvHOQdSyNPwb8seEn2nhHRERERERERI4BU6M+zunX4Fz7DKbXhRBfDXL2YL99J7zO5Vdjsbs3RzrM3+TEpjV4Z3gPzutYF4B3F2zmwrGzmfWj8izHmhKWFVT7uuF1LMuy8Y6mhYuIiIiIiIjIsWRik3B6DA5v0DPgT1C7MQQLsYun4I77C6EPHsX+uLjCrXOZEPByd/+WPHdRR+omx7A1M58/T1jI3yetJLtA1ZbHihKWFdS+CsuV6dnkFx1k8dfijXfsnq3lFZaIiIiIiIiIVGHG48VpfTLOpQ/hXHAPNOsCGPhxEe4Hj+C+dgfukq+xRYWRDvWIdG9cnQlXdefCzvUA+HDRFi4cO4vv1u2KcGSVkxKWFVSdpBhqxPsJuZYV6QdeE8I0bAuAXfgl1rrlGZ6IiIiIiIiIVGHGGEyDVnjOuRXnqicxnfqDLwZ2/YSd/CLuSzfgfvsuNmdvpEMtszi/lztOP54XLulEvWoxpGcVcMO7i7j/8xVk5RdFOrxKRQnLCsoY8/M6lpsPso5lh37hdSx3bYYf5pdneCIiIiIiIiIiAJhqqTinDcW55hnMKZdCYk3Iy8LO+hD3pRtxJz2P3f5jpMMssy4NU5gwrAeXdKmPAT5ZspULxs7im7U7Ix1apaGEZQW2bx3LgyYsA3GYDqcD4M7+pMKtEyEiIiIiIiIilYeJicfpOghn+D9x/u8mqNsCQkHs8m9w37ib0Dt/x66di3Wjf5ZorN/Dbf1a8OKlnWmYEsuO7EJueX8xoz5bTkaeqi1/LyUsK7B2xRWWS7ZkHjQZaToPBI8Ptq2FTcvLMzwRERERERERkf0Yx4Np0QPPxffhXPIA5vieYBz4aTnuJ0/ivnIr7vxJ2MK8SId6WJ3qV+PtYd25vHtDHAP/XbaNC8bO4uvVOyIdWoWmhGUF1io1EY9j2JVTyJaM/AO2MfHJmLanAuDO+aQcoxMREREREREROTRT5zicQTfgXP0vTNezIBAHGenYaa/hvngD7vQ3sZnRnfyL8Xm4+bTjePnSLjSuHseunEJu+3AJd3+yjD25FWtzoWihhGUFFuPz0DI1AYDFWw48LRzAdP2/8CcVG5Zg09eVV3giIiIiIiIiImViEmvgnHIJzrXPYvoMg2ppUJCLnfdf3LG34H72L+yW1ZEO85Da1UvmrWHduPKERjgGvliRzoVjZ/HVyu2RDq3CUcKygmtXt3ha+ObMg7YxybUwLU8EwmtZioiIiIiIiIhEI+OLwel4Os6wJ3AG/wUatgHrYlfPwh1/H6G3RuGu/A4bCkY61AMKeD3c0LsZ4y7vSrOa8ezOLeKOj5dy+0dL2J2jasuyUsKygmtfnLA8VIUlgOl2dvibNXOwuzcf67BERERERERERH4zYxxM0054zv8rzuWPYNr0Bo8Xtq3Ffv4s7tibced8is3PjnSoB9SmThJvDO3G8J6N8BjDlFU7OH/sLL5Ynq5NkctACcsKrn298E7hq7dnk1cUOmg7U7M+NOsCWOycz8opOhERERERERGR38fUaojT/zqcq5/B9DwP4pIgezf2f2+H17mc8gp2z9ZIh7kfv9fhz6c049UrutK8VgIZeUXc/ekybvtwCTuzCyIdXlRTwrKCS0uKoVaCn5BrWbHt4NPCAZzu5wBgV8zAZu4sj/BERERERERERI4KE5+M0/M8nKufxpxxLdRsAEUF2EWTcV+5jdBHT2A3Lou6CsZWaYm8PrQr1/VqgscxTFuzkwvGzuLzZduiLtZooYRlBWeM+Xla+CHWsYTwzls0aA1uCDvzvfIIT0RERERERETkqDJeP07bU3EufwTn/L9C006AhXXzcd97CPeNu3GXTccGiyIdagmfx+Hak5rwxtCuHJ+aQGZ+kHs+W84t7y9mR5aqLX9NCctKoF29sq1jCeD0uhAw2GXfYNcvOsaRiYiIiIiIiIgcG8YYTMM2eAb/BWfYPzAdTgdvAHZswH7xH9yXbsSd+QE29/D5kvLSonYir17elT+f3BSvY/jfD7s4f+wsPlm8RdWWv6CEZSXQvmSn8IzD/nKbui0wnfoD4E5+EVuQe8zjExERERERERE5lkxKHZy+w3CueQZz0sWQUB1yM7Az38N98UbcL17A7twU6TCBcLXl8BMb8+aV3Widlkh2QZD7J67kxncXsS0zP9LhRQUlLCuBlmkJeB3D7twiNu7JO2x7c9KFkJwaXqB2+pvlEKGIiIiIiIiIyLFnYhNwup+NM/wpzJnXQ1ozCBVhl03Dfe0OQu89jF23AGvdSIfKcbUSeOXyLtzQuxl+j8N363dz4dhZfLBwc5WvtlTCshIIeD10bZgCwIeLthy2vfHF4PS/FgC79Gvsj4uPaXwiIiIiIiIiIuXJeLw4LU/EueQBnIvvg+bdwRjYuBT3o8dxx/0Fd9FkbFFkKxq9jsOVJzTirSu70a5uEjmFIR76YhUj3lnI1ozDF6VVVkpYVhIXd60PwAcLN5NTEDxse1O/FabjvqnhL2lquIiIiIiIiIhUOsYYTN0WeM66GeeqpzBdBoE/FvZsxU55BfeFG3D/Nx6btTuicTapGc/YS7twy2nHEfA6zPpxDxe+PJt3F/yEWwWrLZWwrCR6Na1Bo+px5BSG+GTJ1jKdY06+CJJrQdZO7P/ePsYRioiIiIiIiIhEjkmuhdP7Upxrn8WcegUk14aCHOycT3DH3oT7+bPYbT9ELD6PY7ise0PeHtadjvWTyS0M8ciXq/nj+AX8tLdqVVsqYVlJOMYwpGsDAN6et4mQe/jsu/HF4JxePDV88RTshiXHNEYRERERERERkUgz/liczgNwhj2Jc/YtUK8luCHsyu9w37qH0Pj7sKtnY93IrHPZqHocLw7pzG19mxPjc5i3cS8XvTyL8fM2VZlqSyUsK5FBbdNIjvGyeW8+09fuLNM5pmEbTIfTAXC/eavKL+oqIiIiIiIiIlWDcRzMcd3wXDQK59KHMK1OAscDW1bjfvYU7su34M77PCLL6DnGcEnXBowf1p0uDauRX+Ty+FdruPat+WzcXfmX9VPCshKJ9Xk4r1M9AN6as7HM55kTLwCPD3ZsgK1rjlV4IiIiIiIiIiJRyaQ2wRn4Z5yrn8b0GAwxCZC5Azv9DdwXb8D9+jXs3vRyj6tBShzPX9yJO89oQZzfw4KfMrj4ldm8MXtjmWbXVlRKWFYyF3aqj9cxLPgpg+VbM8t0jolNwLTsCYBd9NWxDE9EREREREREJGqZhBScXhfiXPMMpt9wqF4PCvOwCybhvjKS0Cf/xP60slxnqDrGcEGn+ky4qjs9GqVQEHT559drufrNeazflVNucZQnJSwrmVqJAc5oVRuAN+duKvN5+6aF29XfY/PKlugUEREREREREamMjC+A074vztDHcM69Axq1B2th7Rzcdx7AfetvuCtmYEPBcoupbnIsYy7qyF/7H0+838PiLZkMeWUOr87aQDBC620eK0pYVkJDujYEYPLK7WzPKijTOSatGaQ2hVAQu3T6sQxPRERERERERKRCMMZgGnfAc96dOEMfw7Q7LbysXvp67MR/4750E+6sj7B5WeUWz7kd6zFheA96NqlOYcjl6Wk/cNUb8/hhR3a5xFAelLCshFqlJdKlQTVCrmXC/J/KfJ7p0A8Au/iriO2EJSIiIiIiIiISjUyN+jinXxOeLn7iBRBfDXL2YL99J7zO5Vdjsbs3l0ssdZJieOaCDtw7sCUJAS/LtmYxZNwcxn73I0Whip/TUcKykhrSrQEAHyzcTE5B2cqTzfE9IRAHGTvgx0XHMjwRERERERERkQrJxCXhnPAHnOH/wgz4E9RuDMFC7OIpuOP+QuiDR7EblhzzdS6NMZzdvi7vDu/Byc1qEHQt//7fOoa+Ppc12yt2taUSlpXUyc1q0iAllsz8IOe+9D1vzdlEXlHokOcYXwDTpjcA7mJtviMiIiIiIiIicjDG68NpfTLOpQ/hXHAPNOsCGPhxEe77o3FfuwN3ydfYosJjGkftxAD/PK89D/5fa5JivKxKz+bSV+fwwoz1Fbba0tjy3NaogsrMzCQ5OZmMjAySkpIiHU6ZLdi0l799toxtmeF1LKvH+bi8eyMu6FSPWL/ngOfYPVtxX7kVMDjDn8Ik1yrHiEVERORYqajjGQlT/4mIiFQMds827MIvwvuDFOWHD8YmYtr3w3Q8HRNf7Zjef0d2AY98uYppa3YC0LxWAvcNakXL1MRjet+yOJLxjBKWZVCRB4hFIZfPlm7j5Zk/siUj/AelWqyPUQNb0rv5gZORofdHw4YlmG5n45x8cXmGKyIiIsdIRR7PiPpPRESkorH5Odil07ALvoCscPIQjxdz/ImYzgMwtRsfu3tby5crtvPoV6vJyCvCYwxXntCQq09sgt8bucnWSlgeZZVhgFgUcpm4bBtjZ27gp715eIzhgf9rxYDWafu1tWvn4H7yT4hNxLnmWYzXF4GIRURE5GiqDOOZqkz9JyIiUjFZNwRr5+DOmwhb1/z8Qv3WOF0GQtNOGHNskoi7cwp5dPJqvlq1HYBmNeO598xWtKkTmbHEkYxntIZlFeHzOJzdvi7vXd2DQW3TCFnL3z5dzkeLtuzfuGlnSKgOeVnYNbPKP1gRERGRKiA3N5dGjRpx2223RToUEREROUaM48G0OAHPJffjXPJAeMNj48BPy3E//gfuK7fiLvgCW5h/1O9dPd7Po4Pb8ug5bUmJ8/HDzhyufH0uT09bS0Hw0PucRJoSllWMz+Nw35mtOK9jPSzw4KSVvD13U6k2xvFg2vcBwM58H7t5VQQiFREREancHnroIU444YRIhyEiIiLlxNQ5DmfQDThX/wvT9SwIxMHedOzXr+K+eD3u9DexmTuP+n37tazNu8N70L9VbVwLr87ayJBX5rBkc8ZRv9fRooRlFeQYw11ntOCybg0AeGLKGl6e+WOpNqZdX4hNgr3puBPuJ/TJP7F7tkYgWhEREZHKZ82aNaxcuZKBAwdGOhQREREpZyaxBs4pl4SX4etzJVRLg4Jc7Lz/4o69Gfezp7Fb1hz2OkciJc7Pw2e35Yk/tKNGvJ8fd+dy1Zvz+OfUNeQXRV+1ZVQmLMeMGUPjxo2JiYmhR48ezJ49+6BtTz31VIwx+z0GDRpU0sZay6hRo6hTpw6xsbH069ePNWuObsdXNMYYbj7tOK7t1RiAMd+s45z/zGTIuNkMf3MeN/x3PQ/XHMoPad2xxoTXW3j1dtyp47C5mZENXkRERKqk5557jvbt25OUlERSUhI9e/Zk4sSJR/Ue33zzDWeddRZ169bFGMNHH310wHZHMl49kNtuu43Ro0cfhYhFRESkojL+GJyOZ+AMewJn8G3QoA1YF7v6e9zx9xJ6exTuqu/D62AeJae1qMW7w3swqE0aroU35mxiyLg5LPxp71G7x9EQdQnLCRMmMHLkSO69917mz59Phw4d6N+/P9u3bz9g+w8++ICtW7eWPJYuXYrH4+GCCy4oafPYY4/x9NNP8/zzzzNr1izi4+Pp378/+flHf32AisQYw3UnNeWmU5sB8NPePFalZ7Pwpwxmrt/NR+vyuGRTe673X8T2Wq3BDWEXfon78i248/6LDQUj/A5ERESkKqlfvz6PPPII8+bNY+7cufTp04dzzjmHZcuWHbD9t99+S1FR0X7Hly9fTnp6+gHPycnJoUOHDowZM+agcZRlvNqxY0fatm2732PLli18/PHHtGjRghYtWhzhT0BEREQqI2McTNPOeC74K87lozFtTgGPF7auxf736XDV5ZxPsfk5R+V+ybE+Hvi/1vzzvPbUSvCzYXcuV785n5e+W39Urn80RN0u4T169KBbt248++yzALiuS4MGDbjhhhu48847D3v+U089xahRo9i6dSvx8fFYa6lbty633npryYLmGRkZpKamMm7cOC6++OLDXrMq7Mq4NSOPbVkF5BWGyCsKP3bnFDJ+3k+kZxUA8Ifqe7nJM4u4vcVrXqbUwel9GaZppwhGLiIiImVRWccz1atX5/HHH2f48OGljruuS+fOnWnevDnjx4/H4/EAsGrVKnr37s3IkSO5/fbbD3ltYwwffvghgwcPLnX8945X77rrLt544w08Hg/Z2dkUFRVx6623MmrUqIOeU1n7T0RERA7M5mRgF03GLvoK8opnuvoCmDa9MZ0GYFLSjsp9svKLeHLqWj5ZspUnz21H7+a1jsp1D+RIxjNRlbAsLCwkLi6O9957r9TAcOjQoezdu5ePP/74sNdo164dPXv25IUXXgBg3bp1NGvWjAULFtCxY8eSdr1796Zjx47861//2u8aBQUFFBQUlDzPzMykQYMGVXKAmF8U4q25mxj3/QZyCkMYLNelbOLCvG9JCIUz+5lprQn1uoTqDZrgOFFXtCsiIiJUvoRXKBTi3XffZejQoSxYsIDWrVvv12bLli2ccsop9OjRg9dff53169dzyimncNZZZ/H8888f9h4HSlgejfHqL40bN46lS5fyxBNPHPD1MWPGMGbMGEKhEKtXr640/SciIiJlY4OF2JXfYedNhF37Nk020LQTTueB0KA1xpjffZ/V27NoUTvxd1/nUI5kPOo9ppEcoZ07dxIKhUhNTS11PDU1lZUrVx72/NmzZ7N06VLGjh1bcmzbtm0l1/j1Nfe99mujR4/m/vvvP9LwK6UYn4erejZmcPu6vPjdet5fuIXn9zTkddK4ylnAxc4ykrYth/fvYS8xbPPVIicxDVOrIYn1GpOYWp+U6jWIDUTVr5qIiIhUUEuWLKFnz57k5+eTkJDAhx9+eMBkJUDdunWZOnUqJ598MkOGDGHmzJn069eP55577jff//eOV4/UiBEjGDFiRMkAX0RERKoW4/Vj2p6KbdMbNi7FnT8R1i+EdfNx182HWo0wnQdiju+J8fp+832OdbLySFWqLNLYsWNp164d3bt3/13Xueuuuxg5cmTJ830VllVZ9Xg/d5x+PJd3b8jizRlsycjnp4zGjN7VgwE7p9Al+CPJJp/kok2wexPsngOrwufmWi+rTRI7PNXI8KdQGJsCiTXwVatJfI3aVKtRg0Y1EkiJ80f2TYqIiEjUO/7441m4cCEZGRm89957DB06lOnTpx80admwYUNef/11evfuTdOmTRk7duxRqUI4Wq688spIhyAiIiIVgDEGGrXD06gddvcW7IJJ2GXfwI4N2C+ex84Yj+nQD9O+Hyau4s/GiKqEZc2aNfF4PPstgp6enk5a2qHn5ufk5DB+/HgeeOCBUsf3nZeenk6dOnVKXfOXU8R/KRAIEAgEfsM7qPzqJsdSNzn2F0daAn0oKsjnp3U/sGfTOoq2byQ2Yws1C3dQw80mzgRpxm6ahXZDHuHHbmBD+AqF1mEH8WxwEimIScYk1SCuei3ik5Lwebz4vA4+rxe/14M3Ng5f9VScpFoYT1T9+oqIiEg58Pv9HHfccQB06dKFOXPm8K9//Yv//Oc/B2yfnp7Otddey1lnncWcOXO45ZZbeOaZZ37z/X/PeFVERETkaDDV62L6XoU98ULskqnYhV9C9m7sd+9hZ32MadUrXHVZs+IW30VVxsfv99OlSxemTJlSsiaQ67pMmTKF66+//pDnvvvuuxQUFHDZZZeVOt6kSRPS0tKYMmVKSYIyMzOTWbNm8ac//elYvI0qyReIoVGrNjRq1abUcbeokPzd6WRv30L+zm24e9Mhexe+nD3EFewlIZSN37jUI4t6NgvytoQTmgfeuLNEkTVsNwlsM8ns9SQQ43WI9TnEeB1ivIaAz4snPhlfcg1iU2oRm1ITk5ACbgiKCqAoHwrzsUUF4ZLpmASITQh/jUnAOJ5j98MSERGRo8Z13VJrj//Szp076du3L61ateLdd99l9erVnHrqqQQCgYOuGXk4v2e8KiIiInI0mdgETPezsV3OxK6ehZ0/EdLXYZdOwy6dBo3ahde5bNweYyrWniNRlbAEGDlyJEOHDqVr1650796dp556ipycHIYNGwbAFVdcQb169Rg9enSp88aOHcvgwYOpUaNGqePGGG6++Wb+/ve/07x5c5o0acI999xD3bp199vxUY4+x+cnLrUBcakHzurbUBBy9pC3ewfbt24hY0c6+Xt2YrJ34wkWYF0XrIu1FmMtSaaAemQSY0LUIYs6NguChB/5B4/DFj8O9tqvFeCl0HgJGi9FxV+t48Xv8xLj9xLr9+L1esA4YByKcMgpsmQHLXlFFp/HIeA1BDyGGI8h4AGPDUEoWPwoCn+1LpTse1X81Tjg9YPHi+t4ybceQsZDbEwAn88Hjgc8XvD4ICYeYhPDSdbiZCuefXEZwBRfzwf+WPDFgNcXVVPhREREyuquu+5i4MCBNGzYkKysLN566y2mTZvGF198sV9b13UZOHAgjRo1YsKECXi9Xlq3bs3kyZPp06cP9erV45ZbbtnvvOzsbNauXVvyfP369SxcuJDq1avTsGFD4PDjVREREZHyZDxeTKte2JYnwpbV4XUu186BDUtwNyyB6nXDFZetTsL4KsaM4qhLWF500UXs2LGDUaNGsW3bNjp27MikSZNKFjbfuHHjfjtRr1q1ihkzZvDll18e8Jq33347OTk5XHvttezdu5eTTjqJSZMmERMTc8zfjxya8XghqRZxSbVo3PjAa0/tE3RdCopcCoMhsrP2ENq9DTK2E8reS05hiOyCEFmFIbILQ+QVFOHLzyShKIsUN5taJoca5BHEIRcvefjIsz7y8BIgRJIpIJl8kkwhAAGCBGxw/2xmIZCzf2w+oFrx42hygLhfPD9Y0vVwr5VqZxyMPyac8DSmJOmKMbheP6HYahTFJFEQSCTPn0SBL56YgJ/Y4keM34vxeLFA0IX8kCU/6JIfhKDjI+TxE/IGCHr8BB0/8QEPNQIOyX5w3OJkrRsqDiacsHVdS6FrCXljKPLGhq9jw+8p4HEIeB0CPgevdqEXEanStm/fzhVXXMHWrVtJTk6mffv2fPHFF5x++un7tXUch4cffpiTTz4Zv//ndbI7dOjAV199Ra1atQ54j7lz53LaaaeVPN+3rvnQoUMZN24ccPjxqoiIiEgkGGOg3vF46h2PzdiOXfBFuNJy9xbsV2OxMyZg2vfBdDgDk1g90uEekrHWljXPUWUdybbrEn0KgiF25xaxJ6eQrIIg2cWPnIIQ2QVBHAdifR5ifB5ivYYEm4/PLcQUFWJDQUywEEJFZOfms3lPDlv25rIzKx8Hi4OLB4sXl1pxXlITfFSP9VAQdMksCJGZHyKzIEhukaUIhyI8FOChCA+FOIT4OQG37w+iB4ufED5CBAhRzQcxHpf8gkK8uHhx8eESIEgSBSQXJ1uTTQFJFODBYrA4xV89WAKEiDXByHTAbxCyhhzCCWUHin/WxQ9jsSVJVg/GcTCOgzUOLga3+PwQJvwTMM7PX4sTsx4sjgHHWBzCP/tC11DgGvJDkBcy5LvgGg+u48Eaz8+VrcVfjceL4/FivF5ivF5ifU7JI8Yb7tfCYIjCoEtBMERhMESB65BnAuQ6AXJNgGwCFHl8VIvxkRLrJTnGQ7UYH0kBpzg+U/wuwHEMMV4H59fVsYbwzwITTjpjcN0QHuti3OKqXre47x1v+D14fT9X6Xp8pZ5bx0sIQ2HIUhiyFIQshUEXr8cQ8HrCyWOvg9cxZarUDbouRSGLxxg8jsEx4X9EreuCKds1fouQawm5FmPA51GiWwQ0nqno1H8iIiLyW9iCXOyy6dgFkyBjR/ig48G0OAHTZSAmtWm5xXIk45moq7AUOdoCXg91kjzUSTp6FbW5hUHWbM8hPSufBilxNK4RR6zv4Ote5hWF2JFVwLasfLZnFZCeWUB6Vj7BkMXrcfB5DL7ir4kBL/WqxYYfyTEkxvgAyC4Isn5nDut25fDDzhx+yiwgKcZLSpyParF+qsX5SAh42ZtbyLasAtIz80nPKmB7VgEZ+UXkFoSTsLEUEUcRPtxSSUAHSzyF1CSPOt58Up08ajm5JNkCrBsCN4SxLl7j4i2eZB9Ok1k8xuIlXJkaQxGxFOE5QM1nyBqKcAgSTi6GU6pgMXhxiaMIjwlfL4lCkig88A/0l3P8Q7+jIw/HAu4xvP5vUJZw9iVhf+unUQ4QU/wI2XAfhZPsHrLxsAeHQjwYY/AQ/n3wGounOLpCG07IF1qHQhv+cxFjiogjSAzB8O+fcUvej8XgEk4mh3AIGYcQnl98b4rbGFwLIQyODSfvPTac3PcSwmLIt17y8JJPeDmFIjzEOiFijUuMcQmYcHu35Nrh38Wg8YQT3cXXD9rwe/cY8DkWv7H4jMVHONEaMp79H8XXdPEQKl4qotCFAtdQEIL8EBS5EOeFBK8h3meJ8xhiPOCa8Aca+z7UKLQOXkLE2kJibCExbiEBN7xGX6HxUmC8FOAjHy+u8RBwKFmGIuABjzHk4yXb9ZIVcsgMecgMOhAK4riFeN0gjhvE4wbDH9bEBIiLiyExLoak+FiMx0ducfI+O2TICYaT+rZ4qQlrPOGEM+At/vn7bAivDS91kReErCBkBw1ZRZacIqgW46F2vIdacR5qxXpJiXUwGPJcQ17Qkhey5BWBi8HrMXiMwesJJ8cdYwgS/v0I7vuwxxhiPJZYD8R5INZj8TkQNF4K8VJoPBTgpQAvrutiXQtuCOu6WBvC7/UQ8PuJCfiKq8d9WGPIL3LJKwyRWxgkpyhEMOTidZxwTB4PXiecfA8FQwRDQYLBEKFgEOuGiPd7SIrxkRhw8PwyGb9viY7i78N/tYR/31xL8e+dQ0LNVDweraEsIiIiIkePCcRhOg/EduwPP8wLTxffvBK78lvsym+h3vHhdS6bdcVE0axGVViWgT7RlsqiKOSWVJiGXFuSDPA4Bq9j8HscYv2e/av4iuUXhcjKD5JTGE5yxAe8xPk8eJzS7a214anfRfmAIeR42FMAe/JD7MopJORa4gNe4v0e4v1e4vzhClevAcctxCnMh8I8KCrABYpcQ6FrKQxBXpHL3txC9uTksTengIycfDLzCvAAMd5w8ifgdYjxhKtVrbVgQ+BarOsSsuGKvyIXCkOWIjdcwVgt4KFawJAUcEjyO8R7wQ0FCQWDuMEiQsEgoWARbjCEGyzCDRVhg0FsKEhRyKUg6FIQcikIWgqC4aSW1+MUJ6TDjxjjEmsLibUFxLj5xIQK8LqF4WSFhaCFoGsIWVuSzAVKff9Lxamjkq/7EtAupjj5FU7uFNnwPzp+4+IjVKqC14cbfm6iLDMrUgVlDX2aajVqHrPrazxTsan/RERE5Gix6euw8ydhV838ecm25FqYnhfgtD7pmN1XFZYickA+j0NKnJ+UOP/hGx9ATPHU+VocepFeY0x44yBv+D5eoFYAapXp/1ex4Q2CSAHAU/z4ZX3sgbdwqnysteFKLBue3pxTECIjv4iMvCIy84Nk5BXh9zgkxnhJjPGSFOMjPuDF5zEEXUswZAm6LkHXUlT8fWHIkhNyKQy5uC4EfE54nVAPBByIMSH8xVV6Xgcca8NrjYaC2GARRUWFBAsLKSwooMi1FFlD0BqKXIciCxiKKxnDFY1+4+J1DCFvANfjL17j1E8RXnILg+QWFJFbECSvsIi8wiAeGyqu2nTx4OJxXTyOxWMIV3Hum9Lv9eDx+nC8PjxeLx6fP1zp6xbhDRXhcQvxhAqxoRC5IUNWyCE7CBlF4WpBp3i5BO8v7uVzwL/v4QGfgULXkhuEnCJLTtCSXQQh1/05Trc4ZuviIXzM2feadfF7DDGOJeBY/E74PeSFIDcI2UHIKrJkF1m8WPwmnEQOFCeWg8ZDHn7yjI8c6yMXHxZDrOMS54SINSFiKMLjhsgPWvJCLnlFkBt0cUMuSV6XJE+IBE+I+OL2eLxYjw/r8YHXH97YqzBIbn4BefkFFBQUUlRYgA+XGMcS67HhfnTC1aWG8AZo4a/hJHfQeCgyHoJ4CJrwkgIBU/yeTTgh7iVEoRuupswNGXKKLLmh8JIHfsfiNxR/DU/jt9ZiLVgsrgVjf67i3ddvBkvQmuJKXkPQhqtlwz/HIDGECJgggeIy7H3VtG7x8hHY4mU9bOlrljAlNZH7npZ6PVRcGVxy3V9UAe+rHLeU/mCBkg8YKFm2wym+poMlpI+QRURERKQcmNSmmIF/xp58CXbhl9jFU8LTxQsOsGlHhChhKSISpUzxlFvHGLxOeHmD6vG/Ldl8VOIBAsWP+IhFceTigQNvrSGRlF8UwuOYo7bGaGHQJT8Ywle8UdfBKsV/zdrwxmG5heHEZkLAg9/j/Ob1VUOuJSOviD25hezNK8JxDAGPg794/Vd/8fdO8bquHseUrPGacGyWdBUREREROSCTkII56SJsj8HYFTMwLXtFOqQSSliKiIhIuYs5xLq/v4XfG04EHiljDLE+zyHXIT4SHsdQPd4f0Q8XRERERESOhPEFMO37RjqMUqJnNU0RERERERERERGp8pSwFBERERERERERkaihhKWIiIiIiIiIiIhEDSUsRUREREREREREJGooYSkiIiIiIiIiIiJRQwlLERERERERERERiRpKWIqIiIiIiIiIiEjUUMJSREREREREREREooYSliIiIiIiIiIiIhI1lLAUERERERERERGRqKGEpYiIiIiIiIiIiEQNJSxFREREREREREQkaihhKSIiIiIiIiIiIlFDCUsRERERERERERGJGt5IB1ARWGsByMzMjHAkIiIiIr/NvnHMvnGNVCwaj4qIiEhFdyTjUSUsyyArKwuABg0aRDgSERERkd8nKyuL5OTkSIchR0jjUREREaksyjIeNVYfsx+W67ps2bKFxMREjDHH7D6ZmZk0aNCATZs2kZSUdMzuI0dOfRPd1D/RS30T3dQ/0etY9I21lqysLOrWrYvjaFWgikbjUQH1TzRT30Q39U/0Ut9Et6PdP0cyHlWFZRk4jkP9+vXL7X5JSUn6gxql1DfRTf0TvdQ30U39E72Odt+osrLi0nhUfkn9E73UN9FN/RO91DfR7Wj2T1nHo/p4XURERERERERERKKGEpYiIiIiIiIiIiISNZSwjCKBQIB7772XQCAQ6VDkV9Q30U39E73UN9FN/RO91DcSKfrdi27qn+ilvolu6p/opb6JbpHsH226IyIiIiIiIiIiIlFDFZYiIiIiIiIiIiISNZSwFBERERERERERkaihhKWIiIiIiIiIiIhEDSUsRUREREREREREJGooYRklxowZQ+PGjYmJiaFHjx7Mnj070iFVOaNHj6Zbt24kJiZSu3ZtBg8ezKpVq0q1yc/PZ8SIEdSoUYOEhATOO+880tPTIxRx1fbII49gjOHmm28uOab+iZzNmzdz2WWXUaNGDWJjY2nXrh1z584ted1ay6hRo6hTpw6xsbH069ePNWvWRDDiqiMUCnHPPffQpEkTYmNjadasGQ8++CC/3HNP/VN+vvnmG8466yzq1q2LMYaPPvqo1Otl6Yvdu3dz6aWXkpSURLVq1Rg+fDjZ2dnl+C6kMtOYNPI0Jq04NB6NPhqTRieNR6NLRRmPKmEZBSZMmMDIkSO59957mT9/Ph06dKB///5s37490qFVKdOnT2fEiBF8//33TJ48maKiIs444wxycnJK2txyyy18+umnvPvuu0yfPp0tW7Zw7rnnRjDqqmnOnDn85z//oX379qWOq38iY8+ePfTq1Qufz8fEiRNZvnw5//jHP0hJSSlp89hjj/H000/z/PPPM2vWLOLj4+nfvz/5+fkRjLxqePTRR3nuued49tlnWbFiBY8++iiPPfYYzzzzTEkb9U/5ycnJoUOHDowZM+aAr5elLy699FKWLVvG5MmT+eyzz/jmm2+49tpry+stSCWmMWl00Ji0YtB4NPpoTBq9NB6NLhVmPGol4rp3725HjBhR8jwUCtm6deva0aNHRzAq2b59uwXs9OnTrbXW7t271/p8Pvvuu++WtFmxYoUF7MyZMyMVZpWTlZVlmzdvbidPnmx79+5tb7rpJmut+ieS7rjjDnvSSScd9HXXdW1aWpp9/PHHS47t3bvXBgIB+/bbb5dHiFXaoEGD7FVXXVXq2LnnnmsvvfRSa636J5IA++GHH5Y8L0tfLF++3AJ2zpw5JW0mTpxojTF28+bN5Ra7VE4ak0YnjUmjj8aj0Ulj0uil8Wj0iubxqCosI6ywsJB58+bRr1+/kmOO49CvXz9mzpwZwcgkIyMDgOrVqwMwb948ioqKSvVVy5YtadiwofqqHI0YMYJBgwaV6gdQ/0TSJ598QteuXbnggguoXbs2nTp14sUXXyx5ff369Wzbtq1U3yQnJ9OjRw/1TTk48cQTmTJlCqtXrwZg0aJFzJgxg4EDBwLqn2hSlr6YOXMm1apVo2vXriVt+vXrh+M4zJo1q9xjlspDY9LopTFp9NF4NDppTBq9NB6tOKJpPOo9aleS32Tnzp2EQiFSU1NLHU9NTWXlypURikpc1+Xmm2+mV69etG3bFoBt27bh9/upVq1aqbapqals27YtAlFWPePHj2f+/PnMmTNnv9fUP5Gzbt06nnvuOUaOHMndd9/NnDlzuPHGG/H7/QwdOrTk53+gv+fUN8fenXfeSWZmJi1btsTj8RAKhXjooYe49NJLAdQ/UaQsfbFt2zZq165d6nWv10v16tXVX/K7aEwanTQmjT4aj0YvjUmjl8ajFUc0jUeVsBQ5gBEjRrB06VJmzJgR6VCk2KZNm7jpppuYPHkyMTExkQ5HfsF1Xbp27crDDz8MQKdOnVi6dCnPP/88Q4cOjXB08s477/Dmm2/y1ltv0aZNGxYuXMjNN99M3bp11T8iIlFOY9LoovFodNOYNHppPCq/haaER1jNmjXxeDz77RyXnp5OWlpahKKq2q6//no+++wzvv76a+rXr19yPC0tjcLCQvbu3VuqvfqqfMybN4/t27fTuXNnvF4vXq+X6dOn8/TTT+P1eklNTVX/REidOnVo3bp1qWOtWrVi48aNACU/f/09Fxl/+ctfuPPOO7n44otp164dl19+ObfccgujR48G1D/RpCx9kZaWtt8GKMFgkN27d6u/5HfRmDT6aEwafTQejW4ak0YvjUcrjmgajyphGWF+v58uXbowZcqUkmOu6zJlyhR69uwZwciqHmst119/PR9++CFTp06lSZMmpV7v0qULPp+vVF+tWrWKjRs3qq/KQd++fVmyZAkLFy4seXTt2pVLL7205Hv1T2T06tWLVatWlTq2evVqGjVqBECTJk1IS0sr1TeZmZnMmjVLfVMOcnNzcZzS/9x7PB5c1wXUP9GkLH3Rs2dP9u7dy7x580raTJ06Fdd16dGjR7nHLJWHxqTRQ2PS6KXxaHTTmDR6aTxacUTVePSobd8jv9n48eNtIBCw48aNs8uXL7fXXnutrVatmt22bVukQ6tS/vSnP9nk5GQ7bdo0u3Xr1pJHbm5uSZs//vGPtmHDhnbq1Kl27ty5tmfPnrZnz54RjLpq++WujNaqfyJl9uzZ1uv12oceesiuWbPGvvnmmzYuLs6+8cYbJW0eeeQRW61aNfvxxx/bxYsX23POOcc2adLE5uXlRTDyqmHo0KG2Xr169rPPPrPr16+3H3zwga1Zs6a9/fbbS9qof8pPVlaWXbBggV2wYIEF7JNPPmkXLFhgN2zYYK0tW18MGDDAdurUyc6aNcvOmDHDNm/e3F5yySWRektSiWhMGh00Jq1YNB6NHhqTRi+NR6NLRRmPKmEZJZ555hnbsGFD6/f7bffu3e33338f6ZCqHOCAj1deeaWkTV5env3zn/9sU1JSbFxcnP3DH/5gt27dGrmgq7hfDxDVP5Hz6aef2rZt29pAIGBbtmxpX3jhhVKvu65r77nnHpuammoDgYDt27evXbVqVYSirVoyMzPtTTfdZBs2bGhjYmJs06ZN7V//+ldbUFBQ0kb9U36+/vrrA/5bM3ToUGtt2fpi165d9pJLLrEJCQk2KSnJDhs2zGZlZUXg3UhlpDFp5GlMWrFoPBpdNCaNThqPRpeKMh411lp79Oo1RURERERERERERH47rWEpIiIiIiIiIiIiUUMJSxEREREREREREYkaSliKiIiIiIiIiIhI1FDCUkRERERERERERKKGEpYiIiIiIiIiIiISNZSwFBERERERERERkaihhKWIiIiIiIiIiIhEDSUsRUREREREREREJGooYSkiVcqVV15J48aNf9O59913H8aYoxtQJXHqqafStm3bSIchIiIiEvU0Hj02NB4VqVyUsBSRqGCMKdNj2rRpkQ5VRERERCohjUdFRKKHsdbaSAchIvLGG2+Uev7aa68xefJkXn/99VLHTz/9dFJTU3/zfYqKinBdl0AgcMTnBoNBgsEgMTExv/n+ldWpp57Kzp07Wbp0aaRDEREREflNNB6t2DQeFalcvJEOQEQE4LLLLiv1/Pvvv2fy5Mn7Hf+13Nxc4uLiynwfn8/3m+ID8Hq9eL36a1NERESkMtJ4VEQkemhKuIhUGPvWpZk3bx6nnHIKcXFx3H333QB8/PHHDBo0iLp16xIIBGjWrBkPPvggoVCo1DV+vWbQjz/+iDGGJ554ghdeeIFmzZoRCATo1q0bc+bMKXXugdYMMsZw/fXX89FHH9G2bVsCgQBt2rRh0qRJ+8U/bdo0unbtSkxMDM2aNeM///nPEa1DNGvWLAYMGEBycjJxcXH07t2bb7/99oAxrly5kgsvvJCkpCRq1KjBTTfdRH5+fqm2wWCQBx98sOQ9N27cmLvvvpuCgoL97j1x4kR69+5NYmIiSUlJdOvWjbfeemu/dsuXL+e0004jLi6OevXq8dhjj+3X5plnnqFNmzbExcWRkpJC165dD3gtERERkWij8ajGoyJSPvTRjIhUKLt27WLgwIFcfPHFXHbZZSXTccaNG0dCQgIjR44kISGBqVOnMmrUKDIzM3n88ccPe9233nqLrKwsrrvuOowxPPbYY5x77rmsW7fusJ+Cz5gxgw8++IA///nPJCYm8vTTT3PeeeexceNGatSoAcCCBQsYMGAAderU4f777ycUCvHAAw9Qq1atMr3vqVOnMnDgQLp06cK9996L4zi88sor9OnTh//973907969VPsLL7yQxo0bM3r0aL7//nuefvpp9uzZw2uvvVbS5uqrr+bVV1/l/PPP59Zbb2XWrFmMHj2aFStW8OGHH5a0GzduHFdddRVt2rThrrvuolq1aixYsIBJkyYxZMiQknZ79uxhwIABnHvuuVx44YW899573HHHHbRr146BAwcC8OKLL3LjjTdy/vnnlwxaFy9ezKxZs0pdS0RERCRaaTyq8aiIlAMrIhKFRowYYX/9V1Tv3r0tYJ9//vn92ufm5u537LrrrrNxcXE2Pz+/5NjQoUNto0aNSp6vX7/eArZGjRp29+7dJcc//vhjC9hPP/205Ni99967X0yA9fv9du3atSXHFi1aZAH7zDPPlBw766yzbFxcnN28eXPJsTVr1liv17vfNX/NdV3bvHlz279/f+u6bqn33KRJE3v66afvF+PZZ59d6hp//vOfLWAXLVpkrbV24cKFFrBXX311qXa33XabBezUqVOttdbu3bvXJiYm2h49eti8vLz94tpnX9+89tprJccKCgpsWlqaPe+880qOnXPOObZNmzaHfL8iIiIi0UDj0Z9pPCoi5U1TwkWkQgkEAgwbNmy/47GxsSXfZ2VlsXPnTk4++WRyc3NZuXLlYa970UUXkZKSUvL85JNPBmDdunWHPbdfv340a9as5Hn79u1JSkoqOTcUCvHVV18xePBg6tatW9LuuOOOK/mk91AWLlzImjVrGDJkCLt27WLnzp3s3LmTnJwc+vbtyzfffIPruqXOGTFiRKnnN9xwAwCff/55qa8jR44s1e7WW28F4L///S8AkydPJisrizvvvHO/xd1/PXUoISGh1BpPfr+f7t27l/oZVqtWjZ9++mm/6U0iIiIiFYXGoxqPisixpynhIlKh1KtXD7/fv9/xZcuW8be//Y2pU6eSmZlZ6rWMjIzDXrdhw4alnu8bLO7Zs+eIz913/r5zt2/fTl5eHscdd9x+7Q507NfWrFkDwNChQw/aJiMjo9QAt3nz5qVeb9asGY7j8OOPPwKwYcMGHMfZ7/5paWlUq1aNDRs2APDDDz8A0LZt28PGWb9+/f0GjSkpKSxevLjk+R133MFXX31F9+7dOe644zjjjDMYMmQIvXr1Ouz1RURERKKBxqMHpvGoiBxNSliKSIXyy0+u99m7dy+9e/cmKSmJBx54gGbNmhETE8P8+fO544479vu090A8Hs8Bj1trj+m5ZbEv/scff5yOHTsesE1CQsIhr3GwhdTLusB6WZTl59CqVStWrVrFZ599xqRJk3j//ff597//zahRo7j//vuPWiwiIiIix4rGox0P2EbjURE5mpSwFJEKb9q0aezatYsPPviAU045peT4+vXrIxjVz2rXrk1MTAxr167d77UDHfu1fdN7kpKS6NevX5nuuWbNGpo0aVLqPq7rluxI2ahRI1zXZc2aNbRq1aqkXXp6Onv37qVRo0al7r106dIyffpeFvHx8Vx00UVcdNFFFBYWcu655/LQQw9x11137TfNR0RERKQi0Hh0fxqPisjvoTUsRaTC2/dJ6i8/OS0sLOTf//53pEIqxePx0K9fPz766CO2bNlScnzt2rVMnDjxsOd36dKFZs2a8cQTT5Cdnb3f6zt27Njv2JgxY0o9f+aZZwBK1ig688wzAXjqqadKtXvyyScBGDRoEABnnHEGiYmJjB49mvz8/FJtf8sn9rt27Sr13O/307p1a6y1FBUVHfH1RERERKKBxqMaj4rI0aUKSxGp8E488URSUlIYOnQoN954I8YYXn/99aM2BeZouO+++/jyyy/p1asXf/rTnwiFQjz77LO0bduWhQsXHvJcx3F46aWXGDhwIG3atGHYsGHUq1ePzZs38/XXX5OUlMSnn35a6pz169dz9tlnM2DAAGbOnMkbb7zBkCFD6NChAwAdOnRg6NChvPDCCyVTmGbPns2rr77K4MGDOe2004Dwp+j//Oc/ufrqq+nWrRtDhgwhJSWFRYsWkZuby6uvvnpEP4czzjiDtLQ0evXqRWpqKitWrODZZ59l0KBBJCYmHtG1RERERKKFxqMaj4rI0aWEpYhUeDVq1OCzzz7j1ltv5W9/+xspKSlcdtll9O3bl/79+0c6PCD8qfTEiRO57bbbuOeee2jQoAEPPPAAK1asKNOukaeeeiozZ87kwQcf5NlnnyU7O5u0tDR69OjBddddt1/7CRMmMGrUKO688068Xi/XX389jz/+eKk2L730Ek2bNmXcuHF8+OGHpKWlcdddd3HvvfeWajd8+HBq167NI488woMPPojP56Nly5bccsstR/xzuO6663jzzTd58sknyc7Opn79+tx444387W9/O+JriYiIiEQLjUc1HhWRo8vYaPrIR0Skihk8eDDLli0r2Xnx97rvvvu4//772bFjBzVr1jwq1xQRERGRykvjURGJRlrDUkSknOTl5ZV6vmbNGj7//HNOPfXUyAQkIiIiIlWKxqMiUlFoSriISDlp2rQpV155JU2bNmXDhg0899xz+P1+br/99kiHJiIiIiJVgMajIlJRKGEpIlJOBgwYwNtvv822bdsIBAL07NmThx9+mObNm0c6NBERERGpAjQeFZGKQmtYioiIiIiIiIiISNTQGpYiIiIiIiIiIiISNZSwFBERERERERERkaihhKWIiIiIiIiIiIhEDSUsRUREREREREREJGooYSkiIiIiIiIiIiJRQwlLERERERERERERiRpKWIqIiIiIiIiIiEjUUMJSREREREREREREosb/A6pL1Um7B0deAAAAAElFTkSuQmCC",
            "text/plain": [
              "<Figure size 1600x500 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.figure(figsize=(16, 5), dpi=100)\n",
        "\n",
        "plt.subplot(1, 2, 1)\n",
        "plt.title(\"Learning curves\", fontsize=14)\n",
        "plt.xlabel(\"Training epochs\", fontsize=12)\n",
        "plt.ylabel(\"Loss\", fontsize=12)\n",
        "plt.plot(train.history[\"g_loss\"], color=\"#3288bd\", label=\"generator\")\n",
        "plt.plot(train.history[\"d_loss\"], color=\"#fc8d59\", label=\"discriminator\")\n",
        "plt.legend(loc=\"upper right\", fontsize=10)\n",
        "\n",
        "plt.subplot(1, 2, 2)\n",
        "plt.title(\"Learning rate scheduling\", fontsize=14)\n",
        "plt.xlabel(\"Training epochs\", fontsize=12)\n",
        "plt.ylabel(\"Learning rate\", fontsize=12)\n",
        "plt.plot(train.history[\"g_lr\"], color=\"#3288bd\", label=\"generator\")\n",
        "plt.plot(train.history[\"d_lr\"], color=\"#fc8d59\", label=\"discriminator\")\n",
        "plt.legend(loc=\"upper right\", fontsize=10)\n",
        "plt.yscale(\"log\")\n",
        "\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 31,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 393
        },
        "id": "2id4-AK5J-WW",
        "outputId": "e07ade49-7454-439a-e7e1-88201a05e314"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1600x500 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.figure(figsize=(16, 5), dpi=100)\n",
        "\n",
        "plt.subplot(1, 2, 1)\n",
        "plt.title(\"Generator learning curves\", fontsize=14)\n",
        "plt.xlabel(\"Training epochs\", fontsize=12)\n",
        "plt.ylabel(gan.loss_name, fontsize=12)\n",
        "plt.plot(train.history[\"g_loss\"], color=\"#d01c8b\", label=\"training set\")\n",
        "plt.plot(train.history[\"val_g_loss\"], color=\"#4dac26\", label=\"validation set\")\n",
        "plt.legend(loc=\"upper right\", fontsize=10)\n",
        "\n",
        "plt.subplot(1, 2, 2)\n",
        "plt.title(\"Discriminator learning curves\", fontsize=14)\n",
        "plt.xlabel(\"Training epochs\", fontsize=12)\n",
        "plt.ylabel(gan.loss_name, fontsize=12)\n",
        "plt.plot(train.history[\"d_loss\"], color=\"#d01c8b\", label=\"training set\")\n",
        "plt.plot(train.history[\"val_d_loss\"], color=\"#4dac26\", label=\"validation set\")\n",
        "plt.legend(loc=\"upper right\", fontsize=10)\n",
        "\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "e5wv91XH6BDC"
      },
      "source": [
        "#### Metric curves"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 22,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 492
        },
        "id": "P8rgOZgpKjPj",
        "outputId": "6e9a3a9e-3426-461d-fc3f-aa4e9b23b026"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 800x500 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.figure(figsize=(8, 5), dpi=100)\n",
        "\n",
        "plt.title(\"Metric curves\", fontsize=14)\n",
        "plt.xlabel(\"Training epochs\", fontsize=12)\n",
        "plt.ylabel(\"Wasserstein distance\", fontsize=12)\n",
        "plt.plot(train.history[\"wass_dist\"], color=\"#d01c8b\", label=\"training set\")\n",
        "plt.plot(train.history[\"val_wass_dist\"], color=\"#4dac26\", label=\"validation set\")\n",
        "plt.legend(loc=\"upper right\", fontsize=10)\n",
        "\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6hdn_BRai5xM"
      },
      "source": [
        "### Validation plots\n",
        "\n",
        "Once trained the GAN system, the **generator model** can be used to parameterize the high-level response of the LHCb RICH detectors, namely the RICH differential log-likelihoods for different particle hypotheses. Let's have a look at the achieved performance on the **test set**!"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "metadata": {
        "id": "HVid9HKfLzPW"
      },
      "outputs": [],
      "source": [
        "x_test_prep = x_scaler.transform(x_test[:, :3])  # input pre-processing\n",
        "x_test_prep = np.c_[x_test_prep, x_test[:, -1]]\n",
        "\n",
        "out_test = gan.generate(x_test_prep, seed=None)\n",
        "out_test_post = y_scaler.inverse_transform(out_test)  # output post-processing"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CfMNEUEs6EHn"
      },
      "source": [
        "#### RICH differential log-likelihood $(e - \\pi)$"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 24,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 374
        },
        "id": "XM985aPPNPR3",
        "outputId": "09daf7c0-a85f-40fd-b75a-8366122bc33e"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1600x500 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.figure(figsize=(16, 5), dpi=100)\n",
        "\n",
        "bins = np.linspace(-100.0, 50.0, 101)\n",
        "\n",
        "for i, scale in enumerate([\"linear\", \"log\"]):\n",
        "    plt.subplot(1, 2, i + 1)\n",
        "    plt.xlabel(\"RICH differential log-likelihood $(e - \\pi)$\", fontsize=12)\n",
        "    plt.ylabel(\"Candidates\", fontsize=12)\n",
        "    plt.hist(y_test[:, 0], bins=bins, color=\"#3288bd\", label=\"toy data\")\n",
        "    plt.hist(out_test_post[:, 0], bins=bins, histtype=\"step\", lw=2, color=\"#fc8d59\", label=\"BceGAN-GP model\")\n",
        "    plt.legend(loc=\"upper left\", fontsize=10)\n",
        "    plt.yscale(scale)\n",
        "\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "df-bcHAW6lGH"
      },
      "source": [
        "#### RICH differential log-likelihood $(\\mu - \\pi)$"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 25,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 374
        },
        "id": "KeRJbgTMPl6g",
        "outputId": "703d602f-9c92-4193-921b-28d1761a8aee"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1600x500 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.figure(figsize=(16, 5), dpi=100)\n",
        "\n",
        "bins = np.linspace(-75.0, 75.0, 101)\n",
        "\n",
        "for i, scale in enumerate([\"linear\", \"log\"]):\n",
        "    plt.subplot(1, 2, i + 1)\n",
        "    plt.xlabel(\"RICH differential log-likelihood $(\\mu - \\pi)$\", fontsize=12)\n",
        "    plt.ylabel(\"Candidates\", fontsize=12)\n",
        "    plt.hist(y_test[:, 1], bins=bins, color=\"#3288bd\", label=\"toy data\")\n",
        "    plt.hist(out_test_post[:, 1], bins=bins, histtype=\"step\", lw=2, color=\"#fc8d59\", label=\"BceGAN-GP model\")\n",
        "    plt.legend(loc=\"upper left\", fontsize=10)\n",
        "    plt.yscale(scale)\n",
        "\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "36gOX6UI6nsL"
      },
      "source": [
        "#### RICH differential log-likelihood $(K - \\pi)$"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 26,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 377
        },
        "id": "rMh4nOK3PyEI",
        "outputId": "a8540a1d-f0a1-49e5-e67f-1e342063eb1f"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1600x500 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.figure(figsize=(16, 5), dpi=100)\n",
        "\n",
        "bins = np.linspace(-150.0, 100.0, 101)\n",
        "\n",
        "for i, scale in enumerate([\"linear\", \"log\"]):\n",
        "    plt.subplot(1, 2, i + 1)\n",
        "    plt.xlabel(\"RICH differential log-likelihood $(K - \\pi)$\", fontsize=12)\n",
        "    plt.ylabel(\"Candidates\", fontsize=12)\n",
        "    plt.hist(y_test[:, 2], bins=bins, color=\"#3288bd\", label=\"toy data\")\n",
        "    plt.hist(out_test_post[:, 2], bins=bins, histtype=\"step\", lw=2, color=\"#fc8d59\", label=\"BceGAN-GP model\")\n",
        "    plt.legend(loc=\"upper left\", fontsize=10)\n",
        "    plt.yscale(scale)\n",
        "\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7EW3FlfL6qJX"
      },
      "source": [
        "#### RICH differential log-likelihood $(p - \\pi)$"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 27,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 377
        },
        "id": "z3WQRODrQE7H",
        "outputId": "5fa57565-fc8e-446f-9dc2-39e1a06ccdba"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1600x500 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.figure(figsize=(16, 5), dpi=100)\n",
        "\n",
        "bins = np.linspace(-150.0, 100.0, 101)\n",
        "\n",
        "for i, scale in enumerate([\"linear\", \"log\"]):\n",
        "    plt.subplot(1, 2, i + 1)\n",
        "    plt.xlabel(\"RICH differential log-likelihood $(p - \\pi)$\", fontsize=12)\n",
        "    plt.ylabel(\"Candidates\", fontsize=12)\n",
        "    plt.hist(y_test[:, 3], bins=bins, color=\"#3288bd\", label=\"toy data\")\n",
        "    plt.hist(out_test_post[:, 3], bins=bins, histtype=\"step\", lw=2, color=\"#fc8d59\", label=\"BceGAN-GP model\")\n",
        "    plt.legend(loc=\"upper left\", fontsize=10)\n",
        "    plt.yscale(scale)\n",
        "\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jOYaaPWMnJXD"
      },
      "source": [
        "### Model export\n",
        "\n",
        "The [pidgan](https://github.com/mbarbetti/pidgan) package provides the **generator model** with the `export_model()` to export it as a plain [`Sequential`](https://www.tensorflow.org/api_docs/python/tf/keras/Sequential) model to ease the **deployment** of the RICH parameterization through tools like [scikinC](https://github.com/landerlini/scikinC) or [keras2c](https://github.com/f0uriest/keras2c) [[9][9], [10][10]].\n",
        "\n",
        "[9]: https://pos.sissa.it/409/034\n",
        "[10]: https://doi.org/10.1016/j.engappai.2021.104182"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 28,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mxbF1xSaPlhf",
        "outputId": "205f9f62-eee5-42ee-a528-a7d5c3afb23c"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n"
          ]
        }
      ],
      "source": [
        "keras.models.save_model(\n",
        "    gan.generator.export_model,  # TF Sequential model\n",
        "    filepath=\"saved_generator\",\n",
        "    save_format=\"tf\",\n",
        ")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IUflfVY7bm5z"
      },
      "source": [
        "Exporting the generator as a [TensorFlow SavedModel format](https://www.tensorflow.org/guide/saved_model) file allows to use **scikinC** to transpile the [`Sequential`](https://www.tensorflow.org/api_docs/python/tf/keras/Sequential) model to a C file as follows:"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 29,
      "metadata": {
        "id": "MxsMQAkhU2_S"
      },
      "outputs": [],
      "source": [
        "%%capture\n",
        "!scikinC saved_generator > generator.C"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 30,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dZ9f1NZAafov",
        "outputId": "3b4666fa-9ebd-43ec-aa41-6d95ade6821e"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "-rw-r--r-- 1 root root 1.8M Nov  7 17:23 generator.C\n"
          ]
        }
      ],
      "source": [
        "!ls -lrth generator.C"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UKqBzg-Fi5xM"
      },
      "source": [
        "## References\n",
        "\n",
        "1. I.J. Goodfellow _et al._, \"Generative Adversarial Networks\", [arXiv:1406.2661][1]\n",
        "2. M. Arjovsky, L. Bottou, \"Towards Principled Methods for Training Generative Adversarial Networks\", [arXiv:1701.04862][2]\n",
        "3. M. Arjovsky, S. Chintala, L. Bottou, \"Wasserstein GAN\", [arXiv:1701.07875][3]\n",
        "4. I. Gulrajani _et al._, \"Improved Training of Wasserstein GANs\", [arXiv:1704.00028][4]\n",
        "5. A. Radford, L. Metz, S. Chintala, \"Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks\", [arXiv:1511.06434][5]\n",
        "6. M. Mirza, S. Osindero, \"Conditional Generative Adversarial Nets\", [arXiv:1411.1784][6]\n",
        "7. A. Maevskiy _et al._, \"Fast Data-Driven Simulation of Cherenkov Detectors Using Generative Adversarial Networks\", [arXiv:1905.11825][7]\n",
        "8. T. Salimans _et al._, \"Improved Techniques for Training GANs\", [arXiv:1606.03498][8]\n",
        "9. L. Anderlini, M. Barbetti, \"scikinC: a tool for deploying machine learning as binaries\", [PoS **CompTools2021** (2022) 034][9]\n",
        "10. R. Conlin _et al._, \"Keras2c: A library for converting Keras neural networks to real-time compatible C\", [Eng Appl Artif Intell **100** (2021) 104182][10]\n",
        "\n",
        "[1]: https://arxiv.org/abs/1406.2661\n",
        "[2]: https://arxiv.org/abs/1701.04862\n",
        "[3]: https://arxiv.org/abs/1701.07875\n",
        "[4]: https://arxiv.org/abs/1704.00028\n",
        "[5]: https://arxiv.org/abs/1511.06434\n",
        "[6]: https://arxiv.org/abs/1411.1784\n",
        "[7]: https://arxiv.org/abs/1905.11825\n",
        "[8]: https://arxiv.org/abs/1606.03498\n",
        "[9]: https://pos.sissa.it/409/034\n",
        "[10]: https://doi.org/10.1016/j.engappai.2021.104182"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1f26wCo4i5xN"
      },
      "source": [
        "## Credits\n",
        "This tutorial is based on the notebooks provided by the [mbarbetti/tf-gen-models](https://github.com/mbarbetti/tf-gen-models) repository."
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "gpuType": "T4",
      "provenance": [],
      "toc_visible": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python",
      "version": "3.9.16"
    },
    "orig_nbformat": 4
  },
  "nbformat": 4,
  "nbformat_minor": 0
}