Hands-On Machine Learning/11_Training_Deep_Neural_Networks.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "11.Training Deep Neural Networks.ipynb",
      "provenance": [],
      "collapsed_sections": [
        "lbLEe82PkjLP"
      ],
      "authorship_tag": "ABX9TyNrcZp7IBjld+/PueE8yo+7"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HAbk4AI1Ci5d"
      },
      "source": [
        "# Vanishing/Exploding Gradients Problems"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hz7K1p5BEY_N"
      },
      "source": [
        "As we discussed in Chapter 10, the backpropagation algorithm works by going from the output layer to the input layer, propagating the error gradient on the way. Once the algorithm has computed the gradient of the cost function with regards to each parameter in the network, it uses these gradients to update each parameter with a Gradient Descent step\n",
        "\n",
        "**Unfortunately, gradients often get smaller and smaller as the algorithm progresses down to the lower layers. As a result, the Gradient Descent update leaves the lower layer connection weights virtually unchanged, and training never converges to a good solution**. This is called the vanishing gradients problem. **In some cases, the opposite can happen: the gradients can grow bigger and bigger, so many layers get insanely large weight updates and the algorithm diverges. This is the exploding gradients problem**, which is mostly encountered in recurrent neural networks. More gener ally, deep neural networks suffer from unstable gradients; different layers may learn at widely different speeds\n",
        "\n",
        "A paper titled “Understanding the Difficulty of Training Deep Feedforward Neural Networks” by Xavier Glorot and Yoshua Bengio1 found a few suspects, including the combination of the popular logistic sigmoid activation function and the weight initialization technique that was most popular at the time, namely random initialization using a normal distribution with a mean of 0 and a standard deviation of 1. In short, they showed that with this activation function and this initialization scheme, the variance of the outputs of each layer is much greater than the variance of its inputs. Going forward in the network, the variance keeps increasing after each layer until the activation function saturates at the top layers. This is actually made worse by the fact that the logistic function has a mean of 0.5, not 0 (the hyperbolic tangent function has a mean of 0 and behaves slightly better than the logistic function in deep networks)\n",
        "\n",
        "Looking at the logistic activation function, you can see that when inputs become large (negative or positive), the function saturates at 0 or 1, with a derivative extremely close to 0. Thus when backpropagation kicks in, it has virtually no gradient to propagate back through the network, and what little gradient exists keeps getting diluted as backpropagation progresses down through the top layers, so there is really nothing left for the lower layers"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JQUmCcheKOvd"
      },
      "source": [
        "<p align='center'>\n",
        "<img src='data:image/png;base64,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'>\n",
        "</p>\n",
        "Figure 11-1. Logistic activation function saturation"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yyVBDPM0F_AP"
      },
      "source": [
        "## Glorot and He Initialization"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "k9eVfMPxGDX3"
      },
      "source": [
        "In their paper, Glorot and Bengio propose a way to significantly alleviate this problem. We need the signal to flow properly in both directions: in the forward direction when making predictions, and in the reverse direction when backpropagating gradients\n",
        "\n",
        "We don’t want the signal to die out, nor do we want it to explode and saturate. For the signal to flow properly, the authors argue that we need the variance of the outputs of each layer to be equal to the variance of its inputs, and we also need the gradients to have equal variance before and after flowing through a layer in the reverse direction\n",
        "\n",
        "It is actually not possible to guarantee both unless the layer has an equal\n",
        "number of inputs and neurons (these numbers are called the fan-in and fan-out of the\n",
        "layer), but they proposed a good compromise that has proven to work very well in\n",
        "practice: **the connection weights of each layer must be initialized randomly as described in Equation below, where**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PMHx1JxnGr7J"
      },
      "source": [
        "$$fan_{avg}= \\frac {(fan_{in} + fan_{out})} {2}$$"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WScz4X4MHKyF"
      },
      "source": [
        "This initialization strategy is called Xavier initialization (after the author’s first name) or Glorot initialization (after his last name)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pIeHPoYdOZtY"
      },
      "source": [
        "Equation 11-1. Glorot initialization (when using the logistic activation function)\n",
        "\n",
        "Normal distribution with mean 0 and variance \n",
        "$\\sigma^2 = \\frac {1} {fan_{avg}}$ \n",
        "\n",
        "Or a uniform distribution between $-r$ and $+r$, with $r = \\sqrt{ \\frac {3} {fan_{avg}}}$"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "naKJudaJPBqi"
      },
      "source": [
        "If you replace $fan_{avg}$ with $fan_{in}$ in Equation 11-1, you get an initialization strategy that Yann LeCun proposed in the 1990s. He called it LeCun initialization. Using Glorot initialization can speed up training considerably, and it is one of the tricks that led to the success of Deep Learning\n",
        "\n",
        "Some papers have provided similar strategies for different activation functions. These strategies differ only by the scale of the variance and whether they use $fan_{avg}$ or $fan_{in}$, as shown in Table 11-1\n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UxoKJ-wWRoKE"
      },
      "source": [
        "Table 11-1. Initialization parameters for each type of activation function\n",
        "\n",
        "<p align='center'>\n",
        "<img src='data:image/png;base64,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'>\n",
        "</p>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xLGdnP6ESNXh"
      },
      "source": [
        "By default, Keras uses Glorot initialization with a uniform distribution. When creating a layer, you can change this to He initialization by setting\n",
        "\n",
        "kernel_initializer=\"he_uniform\" or\n",
        "kernel_initializer=\"he_normal\" like this:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kXmBDPzIS_9p",
        "outputId": "3d8bac35-6a85-4fac-a067-0e7b39f6d251"
      },
      "source": [
        "from tensorflow import keras\n",
        "keras.layers.Dense(10, activation='relu', kernel_initializer='he_normal')"
      ],
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.layers.core.Dense at 0x7efd2cfdd150>"
            ]
          },
          "metadata": {},
          "execution_count": 51
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TrKtZ_LPTTl7"
      },
      "source": [
        "If you want He initialization with a uniform distribution but based on $fan_{avg}$ rather than $fan_{in}$, you can use the VarianceScaling initializer like this:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ybJTaZe7TggE",
        "outputId": "22ffae6c-e8ea-4198-c9c1-97453ccfa703"
      },
      "source": [
        "he_avg_init = keras.initializers.VarianceScaling(scale=2., \n",
        "              mode='fan_avg', distribution='uniform')\n",
        "keras.layers.Dense(10, activation='sigmoid',\n",
        "                   kernel_initializer=he_avg_init)"
      ],
      "execution_count": 52,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.layers.core.Dense at 0x7efd2cfd7cd0>"
            ]
          },
          "metadata": {},
          "execution_count": 52
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cjmrY3LKT-J9"
      },
      "source": [
        "## Nonsaturating Activation Functions"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dZmtisqNU8BG"
      },
      "source": [
        "One of the insights in the 2010 paper by Glorot and Bengio was that the problems with unstable gradients were in part due to a poor choice of activation function\n",
        "\n",
        "Other activation functions behave much better in deep neural networks—in particular, the ReLU activation function, mostly because it does not saturate for positive values (and because it is fast to compute)\n",
        "\n",
        "Unfortunately, the ReLU activation function is not perfect. It suffers from a problem known as the dying ReLUs: during training, **some neurons effectively “die,” meaning they stop outputting anything other than 0**. In some cases, you may find that half of your network’s neurons are dead, especially if you used a large learning rate\n",
        "\n",
        "A neuron dies when its weights get tweaked in such a way that the weighted sum of its inputs are negative for all instances in the training set. When this happens, it just keeps outputting zeros, and Gradient Descent does not affect it anymore because the gradient of the ReLU function is zero when its input is negative\n",
        "\n",
        "To solve this problem, you may want to use a variant of the ReLU function, such as the leaky ReLU"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kUnkkHrEWI4B"
      },
      "source": [
        "<p align='center'>\n",
        "<img src='data:image/png;base64,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'>\n",
        "</p>\n",
        "Figure 11-2. Leaky ReLU: like ReLU, but with a small slope for negative values"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "S8JIqV1AbRtJ"
      },
      "source": [
        "Last but not least, a 2015 paper by Djork-Clevert et al. proposed a new activation function called the **exponential linear unit (ELU) that outperformed all the ReLU variants in the authors’ experiments: training time was reduced**, and the neural network performed better on the test set\n",
        "\n",
        "Figure 11-3 graphs the function, and Equation 11-2 shows its definition"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hN95gSPibZ4u"
      },
      "source": [
        "Equation 11-2. ELU activation function\n",
        "\n",
        "$$ELU_{\\alpha}(z) = \\begin{cases}  \\alpha(exp(z) - 1) & \\text{if } z < 0 \\\\ z & \\text{if } z \\geq 0 \\end{cases}$$"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mNMELwFvck4k"
      },
      "source": [
        "<p align='center'>\n",
        "<img src='data:image/png;base64,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'>\n",
        "</p>\n",
        "Figure 11-3. ELU activation function"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "I9yVYp_Nczas"
      },
      "source": [
        "The ELU activation function looks a lot like the ReLU function, with a few major differences:\n",
        "\n",
        "* It takes on negative values when z < 0, which allows the unit to have an average output closer to 0 and helps alleviate the vanishing gradients problem\n",
        "\n",
        "* The hyperparameter α defines the value that the ELU function approaches when z is a large negative number. It is usually set to 1, but you can tweak it like any other hyperparameter\n",
        "\n",
        "* It has a nonzero gradient for z < 0, which avoids the dead neurons problem\n",
        "\n",
        "* If α is equal to 1 then the function is smooth everywhere, including around z = 0, which helps speed up Gradient Descent since it does not bounce as much to the left and right of z = 0\n",
        "\n",
        "The main drawback of the ELU activation function is that it is **slower to compute than the ReLU function and its variants (due to the use of the exponential function)**\n",
        "\n",
        "Its faster convergence rate during training compensates for that slow computation, but still, at test time an ELU network will be slower than a ReLU network\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-MjmzhMdHrY7"
      },
      "source": [
        "Then, a <a href='https://arxiv.org/abs/1706.02515'>2017 paper</a> by Guter Klambauer et al. introduced the Scaled ELU (SELU) activation function: as its name suggests, it is a scaled variant of the ELU activation function\n",
        "\n",
        "The authors showed that if you build a neural network composed exclusively of a stack of dense layers, and if all hidden layers use the SELU activation function, then the network will self-normalize: the output of each layer will tend to preserve a mean of 0 and standard deviation of 1 during training, which solves the vanishing/exploding gradients problem\n",
        "\n",
        "As a result, the SELU activation function often significantly outperforms other activation functions for such neural nets (especially deep ones). There are, however, a few conditions for self-normalization to happen\n",
        "\n",
        "* The input features must be standardized (mean 0 and standard deviation 1)\n",
        "\n",
        "* Every hidden layer’s weights must be initialized with LeCun normal initialization. In Keras, this means setting kernel_initializer=\"lecun_normal\"\n",
        "\n",
        "* The network’s architecture must be sequential. Unfortunately, if you try to use SELU in nonsequential architectures, such as recurrent networks or networks with skip connections (i.e., connections that skip layers, such as in Wide & Deep nets), self-normalization will not be guaranteed, so SELU will not necessarily outperform other activation functions\n",
        "\n",
        "* The paper only guarantees self-normalization if all layers are dense, but some researchers have noted that the SELU activation function can improve performance in convolutional neural nets as well"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "v5_kTBHkJFUQ"
      },
      "source": [
        "To use the leaky ReLU activation function, create a LeakyReLu layer and add it to your model just after the layer you want to apply it to, \n",
        "Let's train a neural network on Fashion MNIST using the Leaky ReLU:\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4kJ0jxaPJM_q"
      },
      "source": [
        "(X_train_full, y_train_full), (X_test, y_test) = keras.datasets.fashion_mnist.load_data()\n",
        "X_train_full = X_train_full / 255.0\n",
        "X_test = X_test / 255.0\n",
        "X_valid, X_train = X_train_full[:5000], X_train_full[5000:]\n",
        "y_valid, y_train = y_train_full[:5000], y_train_full[5000:]"
      ],
      "execution_count": 53,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eeCmP9LwKVji"
      },
      "source": [
        "import tensorflow as tf\n",
        "import numpy as np\n",
        "\n",
        "tf.random.set_seed(42)\n",
        "np.random.seed(42)\n",
        "\n",
        "model = keras.models.Sequential([\n",
        "    keras.layers.Flatten(input_shape=[28, 28]),\n",
        "    keras.layers.Dense(300, kernel_initializer='he_normal'),\n",
        "    keras.layers.LeakyReLU(alpha=0.2),\n",
        "    keras.layers.Dense(100, kernel_initializer='he_normal'),\n",
        "    keras.layers.LeakyReLU(alpha=0.2),\n",
        "    keras.layers.Dense(10, activation='softmax')         \n",
        "])"
      ],
      "execution_count": 54,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cBDDJc6xLImS"
      },
      "source": [
        "model.compile(loss='sparse_categorical_crossentropy',\n",
        "              optimizer=keras.optimizers.SGD(learning_rate=1e-3),\n",
        "              metrics=['accuracy']\n",
        "              )"
      ],
      "execution_count": 55,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "m5OHbn2XLWQU"
      },
      "source": [
        "#history = model.fit(X_train, y_train, epochs=10,\n",
        "#                    validation_data=(X_valid, y_valid))"
      ],
      "execution_count": 56,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2TvODCrBLuda"
      },
      "source": [
        "For PreLU, replace LeakyRelu(alpha=0.2) with PReLU()"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jsv5ohrGN7Un"
      },
      "source": [
        "tf.random.set_seed(42)\n",
        "np.random.seed(42)\n",
        "\n",
        "model = keras.models.Sequential([\n",
        "    keras.layers.Flatten(input_shape=[28, 28]),\n",
        "    keras.layers.Dense(300, kernel_initializer='he_normal'),\n",
        "    keras.layers.PReLU(),\n",
        "    keras.layers.Dense(100, kernel_initializer='he_normal'),\n",
        "    keras.layers.PReLU(),\n",
        "    keras.layers.Dense(10, activation='softmax')\n",
        "])"
      ],
      "execution_count": 57,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UvLauS-qPoR9"
      },
      "source": [
        "model.compile(loss='sparse_categorical_crossentropy',\n",
        "              optimizer=keras.optimizers.SGD(learning_rate=1e-3),\n",
        "              metrics=['accuracy']\n",
        "              )"
      ],
      "execution_count": 58,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rQvgZ_tjQt7f"
      },
      "source": [
        "#history = model.fit(X_train, y_train, epochs=10,\n",
        "#                    validation_data=(X_valid, y_valid))"
      ],
      "execution_count": 59,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6skUZWgJRLtp"
      },
      "source": [
        "## Batch Normalization"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8YiUd23VROz3"
      },
      "source": [
        "Although using He initialization along with ELU (or any variant of ReLU) can significantly reduce the danger of the vanishing/exploding gradients problems at the beginning of training, it doesn’t guarantee that they won’t come back during training\n",
        "\n",
        "In a <a href='https://arxiv.org/abs/1502.03167'>2015 paper</a>, Sergey Ioffe and Christian Szegedy proposed a technique called Batch Normalization (BN) that addresses these problems\n",
        "\n",
        "The technique consists of adding an operation in the model just before or after the activation function of each hidden layer. This operation simply **zero centers and normalizes each input, then scales and shifts the result using two new parameter vectors per layer: one for scaling, the other for shifting**\n",
        "\n",
        "In other words, the operation lets the model learn the optimal scale and mean of each of the layer’s inputs. In many cases, if you add a BN layer as the very first layer of your neural network, **you do not need to standardize your training set (e.g., using a StandardScaler)**; the BN layer will do it for you\n",
        "\n",
        "In order to zero-center and normalize the inputs, the algorithm needs to estimate each input’s mean and standard deviation. It does so by evaluating the **mean and standard deviation of the input over the current mini-batch** (hence the name “Batch Normalization”\n",
        "\n",
        "The whole operation is summarized step by step in Equation 11-3"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TPLmmPg1ZG8J"
      },
      "source": [
        "Equation 11-3. Batch Normalization algorithm\n",
        "\n",
        "$$\\mu_B = \\frac{1}{m_B}\\sum^{m_B}_{i=1}x^{(i)} \\\\ \n",
        "{\\sigma_B}^2 = \\frac{1}{m_B} \\sum^{m_B}_{i=1}(x^{(i)} - \\mu_B)^2 \\\\ \n",
        "\\hat x^{(i)} = \\frac{x^{(i)} - \\mu_B}{\\sqrt{{\\sigma_B}^2 + \\epsilon}} \\\\\n",
        "z^{(i)} = \\gamma \\space \\otimes \\hat x^{(i)} + \\beta $$"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3A-11qoObOVM"
      },
      "source": [
        "In this algorithm:\n",
        "\n",
        "* $\\mu_B$ is the vector of input means, evaluted over the whole mini-batch B (it contains one mean per input)\n",
        "\n",
        "* $\\sigma_B$ is the vector of input standard deviations, also evaluated over the whole mini-batch (it contains one standard deviation per input)\n",
        "\n",
        "* $m_B$ is the number of instances in the mini-batch\n",
        "\n",
        "* $\\hat x^{(i)}$ is the vector of zero-centered and normalized inputs for instance i\n",
        "\n",
        "* $\\gamma$ is the output scale parameter vector for the layer (it contains one scale parameter per input)\n",
        "\n",
        "* $\\otimes$ represents element-wise multiplication (each input is multiplied by its corresponding output scale parameter)\n",
        "\n",
        "* $\\beta$ is the output shift (offset) parameter vector for the layer (it contains one offset parameter per input). Each input is offset by its corresponding shift parameter\n",
        "\n",
        "* $\\epsilon$ is a tiny number that avoids division by zero (typically $10^{-5}$). This is called a smoothing term\n",
        "\n",
        "* $z^{(i)}$ is the output of the BN operation. It is a rescaled and shifted version of the inputs"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tPlZ0JatfsA1"
      },
      "source": [
        "So during training, BN standardizes its inputs, then rescales and offsets them. What about at test time? Well, it’s not that simple. Indeed, we may need to make predictions for individual instances rather than for batches of instances: in this case, we will have no way to compute each input’s mean and standard deviation\n",
        "\n",
        "Moreover, even if we do have a batch of instances, it may be too small, or the instances may not be independent and identically distributed, so computing statistics over the batch instances would be unreliable\n",
        "\n",
        "One solution could be to wait until the end of training, then run the whole training set through the neural network and compute the mean and standard deviation of each input of the BN layer\n",
        "\n",
        "These “final” input means and standard deviations could then be used instead of the batch input means and standard deviations when making predictions. However, most implementations of Batch Normalization estimate these final statistics during training by using a moving average of the layer’s input means and standard deviations. This is what Keras does automatically when you use the BatchNormalization layer\n",
        "\n",
        "To sum up, four parameter vectors are learned in each batch-normalized layer: $\\gamma$ (the output scale vector) and $\\beta$ (the output offset vector) are learned through regular backpropagaion, and $\\mu$ (the final input mean vector) and $\\sigma$ (the final input standard deviation vector) are estimated using an exponential moving average\n",
        "\n",
        "Note that $\\mu$ and $\\sigma$ are estimated during training, but they are used only after training (to replace the batch input means and standard deviations in Equation 11-3)\n",
        "\n",
        "Batch Normalization acts like a regularizer, reducing the need for other regularization techniques (such as dropout)\n",
        "\n",
        "Batch Normalization does, however, add some complexity to the model. Moreover, there is a runtime penalty: network makes slower predictions due to the extra computations required at each layer\n",
        "\n",
        "Fortunately, it’s often possible to fuse the BN layer with the previous layer, after training, thereby avoiding the runtime penalty. This is done by updating the previous layer’s weights and biases so that it directly produces outputs of the appropriate scale and offset\n",
        "\n",
        "For example, if the previous layer computes $XW + b$, then the BN layer will compute $\\gamma \\space \\otimes (XW + b - \\mu)/ \\sigma + \\beta $ (ignoring the smoothing term ε in the denominator)\n",
        "\n",
        "If we define $W' = \\gamma \\space \\otimes W/\\sigma$ and $ b' = \\gamma \\space \\otimes (b-\\mu)/\\sigma + \\beta$, the equation simplifies to $XW' + b'$. So if we replace the previous layer's weights and biases (W and b) with the updated weights and biases (W' and b'), we can get rid of the BN layer"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lbLEe82PkjLP"
      },
      "source": [
        "### Implementing Batch Normalization with Keras"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0PqDwsBXkqFv"
      },
      "source": [
        "Just add a BatchNormalization layer before or after each hidden layer's activation function, and optionally add a BN layer as well as the first layer in your model\n",
        "\n",
        "For example, this model applies BN after every hidden layer and as the first layer in the model (after flattening the input images):"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zxCDc2C4k6kq"
      },
      "source": [
        "model = keras.models.Sequential([\n",
        "    keras.layers.Flatten(input_shape=[28, 28]),\n",
        "    keras.layers.BatchNormalization(),\n",
        "    keras.layers.Dense(300, activation='elu',\n",
        "                       kernel_initializer='he_normal'),\n",
        "    keras.layers.BatchNormalization(),\n",
        "    keras.layers.Dense(100, activation='elu',\n",
        "                       kernel_initializer='he_normal'),\n",
        "    keras.layers.BatchNormalization(),\n",
        "    keras.layers.Dense(10, activation='softmax')\n",
        "])"
      ],
      "execution_count": 60,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cWnmdUN8lro6"
      },
      "source": [
        "In this tiny example with just two hidden layers, it’s unlikely that Batch Normalization will have a very positive impact; but for deeper networks it can make a tremendous difference\n",
        "\n",
        "Let's display the model summary:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Tn0QvBPyluL1",
        "outputId": "b3bf0d97-2433-4228-e2a8-2256184a07d1"
      },
      "source": [
        "model.summary()"
      ],
      "execution_count": 61,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_11\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "flatten_10 (Flatten)         (None, 784)               0         \n",
            "_________________________________________________________________\n",
            "batch_normalization_6 (Batch (None, 784)               3136      \n",
            "_________________________________________________________________\n",
            "dense_42 (Dense)             (None, 300)               235500    \n",
            "_________________________________________________________________\n",
            "batch_normalization_7 (Batch (None, 300)               1200      \n",
            "_________________________________________________________________\n",
            "dense_43 (Dense)             (None, 100)               30100     \n",
            "_________________________________________________________________\n",
            "batch_normalization_8 (Batch (None, 100)               400       \n",
            "_________________________________________________________________\n",
            "dense_44 (Dense)             (None, 10)                1010      \n",
            "=================================================================\n",
            "Total params: 271,346\n",
            "Trainable params: 268,978\n",
            "Non-trainable params: 2,368\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kLYQuLiXmBUn"
      },
      "source": [
        "As you can see, each BN layer adds four parameters per input: $\\gamma , \\beta , \\mu , \\sigma$ (for example, the first BN layer adds 3,136 parameters, which is 4 x 784)\n",
        "\n",
        "The last two parameters, $\\mu$ and $\\sigma$, are the moving averages; they are not affected by backpropagation, so Keras calls them \"non-trainable\"\n",
        "\n",
        "If you count the total number of BN parameters, 3,136 + 1,200 + 400, and divide by 2, you get 2,368, which is the total number of non-trainable parameters in this model\n",
        "\n",
        "Let’s look at the parameters of the first BN layer. Two are trainable (by backpropagation), and two are not\n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "i0VBQeN4iSqZ",
        "outputId": "47332341-a208-4293-f55a-8b591684b028"
      },
      "source": [
        "[(var.name, var.trainable) for var in model.layers[1].variables]"
      ],
      "execution_count": 62,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('batch_normalization_6/gamma:0', True),\n",
              " ('batch_normalization_6/beta:0', True),\n",
              " ('batch_normalization_6/moving_mean:0', False),\n",
              " ('batch_normalization_6/moving_variance:0', False)]"
            ]
          },
          "metadata": {},
          "execution_count": 62
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wCK9bp-FicYW"
      },
      "source": [
        "Now when you create a BN layer in Keras, it also creates two operations that will be called by Keras at each iteration during training. These operations will update the moving averages. Since we are using the TensorFlow backend, these operations are TensorFlow operations"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "980oV9wVi-8R"
      },
      "source": [
        "The authors of the BN paper argued in favor of adding the BN layers before the activation functions, rather than after (as we just did). There is some debate about this, as which is preferable seems to depend on the task\n",
        "\n",
        "To add the BN layers before the activation functions, you must remove the activation function from the hidden layers and add them as separate layers after the BN layers\n",
        "\n",
        "Moreover, since a Batch Normalization layer includes one offset parameter per input, you can remove the bias term from the previous layer (just pass use_bias=False when creating it):"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dIX5sUkximho"
      },
      "source": [
        "model = keras.models.Sequential([\n",
        "    keras.layers.Flatten(input_shape=[28, 28]),\n",
        "    keras.layers.BatchNormalization(),\n",
        "    keras.layers.Dense(300, kernel_initializer='he_normal', use_bias=False),\n",
        "    keras.layers.BatchNormalization(),\n",
        "    keras.layers.Activation('elu'),\n",
        "    keras.layers.Dense(100, kernel_initializer='he_normal', use_bias=False),\n",
        "    keras.layers.BatchNormalization(),\n",
        "    keras.layers.Activation('elu'),\n",
        "    keras.layers.Dense(10, activation='softmax')\n",
        "])"
      ],
      "execution_count": 63,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "j50X0ZDvkExe"
      },
      "source": [
        "The BatchNormalization class has quite a few hyperparameters you can tweak. The defaults will usually be fine, but you may occasionally need to tweak the **momentum**\n",
        "\n",
        "This hyperparameter is used by the BatchNormalization layer when it updates the exponential moving averages\n",
        "\n",
        "Given a new value v (i.e., a new vector of input means or standard deviations computed over the current batch), the layer updates the running average $\\hat v$ using the following equation: "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PjLg8XzckzJ_"
      },
      "source": [
        "$$\\hat v \\gets \\hat v \\times \\text{momentum} + v \\times (1 - \\text{momentum}) $$"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "w04W6fXZlGxG"
      },
      "source": [
        "A good momentum value is typically close to 1; for example, 0.9, 0.99, or 0.999 (you want more 9s for larger datasets and smaller mini-batches)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_LD90qHClP1d"
      },
      "source": [
        "**Another important hyperparameter is axis: it determines which axis should be normalized**\n",
        "\n",
        "It defaults to –1, meaning that by default it will normalize the last axis (using the means and standard deviations computed across the other axes)\n",
        "\n",
        "When the input batch is 2D (i.e., the batch shape is [batch size, features]), this means that each input feature will be normalized based on the mean and standard deviation computed across all the instances in the batch\n",
        "\n",
        "For example, the first BN layer in the previous code example will independently normalize (and rescale and shift) each of the 784 input features\n",
        "\n",
        "If we move the first BN layer before the Flatten layer, then the input batches will be 3D, with shape [batch size, height, width], therefore, the BN layer will compute 28 means and 28 standard deviations (1 per column of pixels, computed across all instances in the batch and across all rows in the column), and it will normalize all pixels in a given column using the same mean and standard deviation\n",
        "\n",
        "There will also be just 28 scale parameters and 28 shift parameters. If instead you still want to treat each of the 784 pixels independently, then you should set axis=[1, 2]\n",
        "\n",
        "Notice that the BN layer does not perform the same computation during training and after training: it uses batch statistics during training and the “final” statistics after training (i.e., the final values of the moving averages). Let’s take a peek at the source code of this class to see how this is handled"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jdAn-V88mWsx"
      },
      "source": [
        "class BatchNormalization(keras.layers.Layer):\n",
        "  #...\n",
        "  def call(self, inputs, training=None):\n",
        "    #... \n",
        "    return\n"
      ],
      "execution_count": 64,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kVv1gGyem4NS"
      },
      "source": [
        "The call() method is the one that performs the computations: it has an extra training argument, which is set to None by default, but the fit() method sets to it to 1 during training\n",
        "\n",
        "If you ever need to write a custom layer, and it must behave differently during training and testing, add a training argument to the call() method and use this argument in the method to decide what to compute\n",
        "\n",
        "BatchNormalization has become one of the most-used layers in deep neural networks, to the point that it is often omitted in the diagrams, as it is assumed that BN is added after every layer\n",
        "\n",
        "But a recent <a href='https://arxiv.org/abs/1901.09321'>paper</a> by Hongyi Zhang et al. may change this assumption: by using a novel fixed update (fixup) weight initialization technique, the authors managed to train a very deep neural network (10,000 layers!) without BN, achieving state-of-the-art performance on complex image classification tasks. As this is bleeding-edge research, however, you may want to wait for additional research to confirm this finding before you drop Batch Normalization"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GtlhER62qYhS"
      },
      "source": [
        "## Gradient Clipping"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qHrFQOfOqe6s"
      },
      "source": [
        "Another popular technique to mitigate the exploding gradients problem is to **clip the gradients during backpropagation so that they never exceed some threshold**. This is called <a href='https://arxiv.org/abs/1211.5063'>Gradient Clipping</a>\n",
        "\n",
        "This technique is most **often used in recurrent neural networks**, as Batch Normalization is tricky to use in RNNs. For other types of networks, BN is usually sufficient\n",
        "\n",
        "In Keras, implementing Gradient Clipping is just a matter of setting the clipvalue or clipnorm argument when creating an optimizer, like this:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "T8A8wCQBq4Jw"
      },
      "source": [
        "optimizer = keras.optimizers.SGD(clipvalue=1.0)\n",
        "model.compile(loss='mse', optimizer=optimizer)"
      ],
      "execution_count": 65,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rrbNWgzyrE2q"
      },
      "source": [
        "**This optimizer will clip every component of the gradient vector to a value between –1.0 and 1.0**. This means that all the partial derivatives of the loss (with regard to each and every trainable parameter) will be clipped between –1.0 and 1.0\n",
        "\n",
        "The threshold is a hyperparameter you can tune. Note that it may change the orientation of the gradient vector. For instance, if the original gradient vector is [0.9, 100.0], it points mostly in the direction of the second axis; but once you clip it by value, you get [0.9, 1.0], which points roughly in the diagonal between the two axes\n",
        "\n",
        "In practice, this approach works well. If you want to ensure that Gradient Clipping does not change the direction of the gradient vector, you should clip by norm by setting clipnorm instead of clipvalue. This will clip the whole gradient if its $l_2$ norm is greater than the threshold you picked\n",
        "\n",
        "For example, if you set clipnorm=1.0, then the vector [0.9, 100.0] will be clipped to [0.00899964, 0.9999595], preserving its orientation but almost eliminating the first component\n",
        "\n",
        "If you observe that the gradients explode during training (you can track the size of the gradients using TensorBoard), you may want to try both clipping by value and clipping by norm, with different thresholds, and see which option performs best on the validation set"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qH3w_ngTr-DJ"
      },
      "source": [
        "# Reusing Pretrained Layers"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FY3ynqbqsAlr"
      },
      "source": [
        "It is generally not a good idea to train a very large DNN from scratch: instead, you should always try to **find an existing neural network that accomplishes a similar task to the one you are trying to tackle, then reuse the lower layers of this network. This technique is called transfer learning**. It will not only speed up training considerably, but also require significantly less training data\n",
        "\n",
        "Suppose you have access to a DNN that was trained to classify pictures into 100 different categories, including animals, plants, vehicles, and everyday objects \n",
        "\n",
        "You now want to train a DNN to classify specific types of vehicles. These tasks are very similar, even partly overlapping, so you should try to reuse parts of the first network"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6y9Z8kg5tEAE"
      },
      "source": [
        "<p align='center'>\n",
        "<img src='data:image/png;base64,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'>\n",
        "</p>\n",
        "Figure 11-4. Reusing pretrained layers"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mPAThLuRtTqA"
      },
      "source": [
        "If the input pictures of your new task don’t have the same size as the ones used in the original task, you will usually have to add a preprocessing step to resize them to the size expected by the original model. More generally, **transfer learning will work best when the inputs have similar low-level features**\n",
        "\n",
        "The output layer of the original model should usually be replaced because it is most likely not useful at all for the new task, and it may not even have the right number of outputs for the new task\n",
        "\n",
        "Similarly, the upper hidden layers of the original model are less likely to be as useful as the lower layers, since the high-level features that are most useful for the new task may differ significantly from the ones that were most useful for the original task. You want to find the right number of layers to reuse\n",
        "\n",
        "The more similar the tasks are, the more layers you want to reuse (starting with the lower layers). **For very similar tasks, try keeping all the hidden layers and just replacing the output layer**\n",
        "\n",
        "**Try freezing all the reused layers first** (i.e., make their weights non trainable so that Gradient Descent won’t modify them), then train your model and see how it performs. Then try unfreezing one or two of the top hidden layers to let backpropagation tweak them and see if performance improves\n",
        "\n",
        "**The more training data you have, the more layers you can unfreeze**. It is also useful to reduce the learning rate when you unfreeze reused layers: this will avoid wrecking their fine-tuned weights\n",
        "\n",
        "If you still cannot get good performance, and you have little training data, try dropping the top hidden layer(s) and freezing all the remaining hidden layers again. You can iterate until you find the right number of layers to reuse. If you have plenty of training data, you may try replacing the top If you still cannot get good performance, and you have little training data, try dropping the top hidden layer(s) and freezing all the remaining hidden layers again. You can iterate until you find the right number of layers to reuse. If you have plenty of training data, you may try replacing the top\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "72M1ZVpNuLxT"
      },
      "source": [
        "## Transfer Learning with Keras"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ruovN28J0q64"
      },
      "source": [
        "Let’s look at an example. Suppose the Fashion MNIST dataset only contained eight classes—for example, all the classes except for sandal and shirt. Someone built and trained a Keras model on that set and got reasonably good performance (>90% accuracy). Let’s call this model A\n",
        "\n",
        "You now want to tackle a different task: you have images of sandals and shirts, and you want to train a binary classifier (positive=shirt, negative=sandal). Your dataset is quite small; you only have 200 labeled images\n",
        "\n",
        "When you train a new model for this task (let’s call it model B) with the same architecture as model A, it performs reasonably well (97.2% accuracy). But since it’s a much easier task (there are just two classes), you were hoping for more, you realize that your task is quite similar to task A, so perhaps transfer learning can help\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8YdxAyJo1H5S"
      },
      "source": [
        "Let's split the fashion MNIST training set in two:\n",
        "\n",
        "* X_train_A: all images of all items except for sandals and shirts (classes 5 and 6)\n",
        "\n",
        "* X_train_B: a much smaller training set of just the first 200 images of sandals or shirts\n",
        "\n",
        "The validation set and the test set are also split this way, but without restricting the number of images\n",
        "\n",
        "We will train a model on set A (classification task with 8 classes), and try to reuse it to tackle set B (binary classification). We hope to transfer a little bit of knowledge from task A to task B, since classes in set A (sneakers, ankle boots, coats, t-shirts, etc.) are somewhat similar to classes in set B (sandals and shirts)\n",
        "\n",
        "However, since we are using Dense layers, only patterns that occur at the same location can be reused (in contrast, convolutional layers will transfer much better, since learned patterns can be detected anywhere on the image, as we will see in the CNN chapter)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "l9xsjvfj1j2N"
      },
      "source": [
        "def split_dataset(X, y):\n",
        "  y_5_or_6 = (y == 5 ) | (y == 6) # sandals or shirts\n",
        "  y_A = y[~y_5_or_6]\n",
        "  y_A[y_A > 6] -= 2 # class indices 7, 8, 9 should be moved to 5, 6, 7\n",
        "  y_B = (y[y_5_or_6] == 6).astype(np.float32) # binary classification task: is it a shirt (class 6)?\n",
        "\n",
        "  return ((X[~y_5_or_6], y_A),\n",
        "          (X[y_5_or_6], y_B))\n",
        "\n",
        "(X_train_A, y_train_A), (X_train_B, y_train_B) = split_dataset(X_train, y_train)\n",
        "(X_valid_A, y_valid_A), (X_valid_B, y_valid_B) = split_dataset(X_valid, y_valid)\n",
        "(X_test_A, y_test_A), (X_test_B, y_test_B) = split_dataset(X_test, y_test)\n",
        "X_train_B = X_train_B[:200]\n",
        "y_train_B = y_train_B[:200]"
      ],
      "execution_count": 66,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hOsaDFTc3Nei",
        "outputId": "366b7396-7571-4c94-ae1a-2bb3d62220af"
      },
      "source": [
        "X_train_A.shape"
      ],
      "execution_count": 67,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(43986, 28, 28)"
            ]
          },
          "metadata": {},
          "execution_count": 67
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "m2szi1Ir3Pob",
        "outputId": "6d12805d-616f-4b9a-8f9a-d3e8c23d966f"
      },
      "source": [
        "X_train_B.shape"
      ],
      "execution_count": 68,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(200, 28, 28)"
            ]
          },
          "metadata": {},
          "execution_count": 68
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "faCPYNVU3RwM",
        "outputId": "23ba1733-6963-4cdd-e17a-daa625279c54"
      },
      "source": [
        "y_train_A[:30]"
      ],
      "execution_count": 69,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([4, 0, 5, 7, 7, 7, 4, 4, 3, 4, 0, 1, 6, 3, 4, 3, 2, 6, 5, 3, 4, 5,\n",
              "       1, 3, 4, 2, 0, 6, 7, 1], dtype=uint8)"
            ]
          },
          "metadata": {},
          "execution_count": 69
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hsJZBxw_3TpY",
        "outputId": "634e8d77-9ab0-46bb-eb02-ea9684b0cb35"
      },
      "source": [
        "y_train_B[:30]"
      ],
      "execution_count": 70,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([1., 1., 0., 0., 0., 0., 1., 1., 1., 0., 0., 1., 1., 0., 0., 0., 0.,\n",
              "       0., 0., 1., 1., 0., 0., 1., 1., 0., 1., 1., 1., 1.], dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 70
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BjWi96Om3Wg6"
      },
      "source": [
        "tf.random.set_seed(42)\n",
        "np.random.seed(42)\n",
        "\n",
        "model_A = keras.models.Sequential()\n",
        "model_A.add(keras.layers.Flatten(input_shape=[28, 28]))\n",
        "for n_hidden in (300, 100, 50, 50, 50):\n",
        "    model_A.add(keras.layers.Dense(n_hidden, activation=\"selu\"))\n",
        "model_A.add(keras.layers.Dense(8, activation=\"softmax\"))"
      ],
      "execution_count": 71,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3yzagk2p30yv"
      },
      "source": [
        "model_A.compile(loss='sparse_categorical_crossentropy',\n",
        "                optimizer=keras.optimizers.SGD(learning_rate=1e-3),\n",
        "                metrics=['accuracy'])"
      ],
      "execution_count": 72,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RlptiluWKQUH"
      },
      "source": [
        "pixel_means = X_train.mean(axis=0, keepdims=True)\n",
        "pixel_stds = X_train.std(axis=0, keepdims=True)\n",
        "X_train_scaled = (X_train - pixel_means) / pixel_stds\n",
        "X_valid_scaled = (X_valid - pixel_means) / pixel_stds\n",
        "X_test_scaled = (X_test - pixel_means) / pixel_stds"
      ],
      "execution_count": 73,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "iaspcXDS39jp",
        "outputId": "f264b13d-4d0b-4023-8b3f-08928d9288e9"
      },
      "source": [
        "history = model_A.fit(X_train_A, y_train_A, epochs=20,\n",
        "                    validation_data=(X_valid_A, y_valid_A))"
      ],
      "execution_count": 74,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/20\n",
            "1375/1375 [==============================] - 7s 5ms/step - loss: 0.5927 - accuracy: 0.8104 - val_loss: 0.3896 - val_accuracy: 0.8667\n",
            "Epoch 2/20\n",
            "1375/1375 [==============================] - 7s 5ms/step - loss: 0.3523 - accuracy: 0.8786 - val_loss: 0.3289 - val_accuracy: 0.8824\n",
            "Epoch 3/20\n",
            "1375/1375 [==============================] - 7s 5ms/step - loss: 0.3170 - accuracy: 0.8896 - val_loss: 0.3014 - val_accuracy: 0.8989\n",
            "Epoch 4/20\n",
            "1375/1375 [==============================] - 7s 5ms/step - loss: 0.2973 - accuracy: 0.8975 - val_loss: 0.2894 - val_accuracy: 0.9021\n",
            "Epoch 5/20\n",
            "1375/1375 [==============================] - 6s 4ms/step - loss: 0.2835 - accuracy: 0.9022 - val_loss: 0.2776 - val_accuracy: 0.9066\n",
            "Epoch 6/20\n",
            "1375/1375 [==============================] - 7s 5ms/step - loss: 0.2729 - accuracy: 0.9061 - val_loss: 0.2735 - val_accuracy: 0.9071\n",
            "Epoch 7/20\n",
            "1375/1375 [==============================] - 7s 5ms/step - loss: 0.2641 - accuracy: 0.9093 - val_loss: 0.2720 - val_accuracy: 0.9086\n",
            "Epoch 8/20\n",
            "1375/1375 [==============================] - 7s 5ms/step - loss: 0.2573 - accuracy: 0.9126 - val_loss: 0.2588 - val_accuracy: 0.9145\n",
            "Epoch 9/20\n",
            "1375/1375 [==============================] - 7s 5ms/step - loss: 0.2518 - accuracy: 0.9136 - val_loss: 0.2562 - val_accuracy: 0.9143\n",
            "Epoch 10/20\n",
            "1375/1375 [==============================] - 7s 5ms/step - loss: 0.2469 - accuracy: 0.9154 - val_loss: 0.2541 - val_accuracy: 0.9160\n",
            "Epoch 11/20\n",
            "1375/1375 [==============================] - 7s 5ms/step - loss: 0.2423 - accuracy: 0.9177 - val_loss: 0.2496 - val_accuracy: 0.9153\n",
            "Epoch 12/20\n",
            "1375/1375 [==============================] - 7s 5ms/step - loss: 0.2382 - accuracy: 0.9189 - val_loss: 0.2514 - val_accuracy: 0.9126\n",
            "Epoch 13/20\n",
            "1375/1375 [==============================] - 7s 5ms/step - loss: 0.2351 - accuracy: 0.9198 - val_loss: 0.2446 - val_accuracy: 0.9165\n",
            "Epoch 14/20\n",
            "1375/1375 [==============================] - 7s 5ms/step - loss: 0.2316 - accuracy: 0.9214 - val_loss: 0.2416 - val_accuracy: 0.9175\n",
            "Epoch 15/20\n",
            "1375/1375 [==============================] - 6s 5ms/step - loss: 0.2288 - accuracy: 0.9212 - val_loss: 0.2447 - val_accuracy: 0.9188\n",
            "Epoch 16/20\n",
            "1375/1375 [==============================] - 7s 5ms/step - loss: 0.2255 - accuracy: 0.9225 - val_loss: 0.2386 - val_accuracy: 0.9203\n",
            "Epoch 17/20\n",
            "1375/1375 [==============================] - 7s 5ms/step - loss: 0.2231 - accuracy: 0.9233 - val_loss: 0.2410 - val_accuracy: 0.9180\n",
            "Epoch 18/20\n",
            "1375/1375 [==============================] - 7s 5ms/step - loss: 0.2201 - accuracy: 0.9245 - val_loss: 0.2427 - val_accuracy: 0.9153\n",
            "Epoch 19/20\n",
            "1375/1375 [==============================] - 7s 5ms/step - loss: 0.2178 - accuracy: 0.9252 - val_loss: 0.2330 - val_accuracy: 0.9203\n",
            "Epoch 20/20\n",
            "1375/1375 [==============================] - 7s 5ms/step - loss: 0.2156 - accuracy: 0.9261 - val_loss: 0.2333 - val_accuracy: 0.9203\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eVMLxANj4XSS"
      },
      "source": [
        "model_A.save('my_model_A.h5')"
      ],
      "execution_count": 75,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "J1dNg_jlM6MX"
      },
      "source": [
        "model_B = keras.models.Sequential()\n",
        "model_B.add(keras.layers.Flatten(input_shape=[28, 28]))\n",
        "for n_hidden in (300, 100, 50, 50, 50):\n",
        "  model_B.add(keras.layers.Dense(n_hidden, activation='selu'))\n",
        "model_B.add(keras.layers.Dense(1, activation='sigmoid'))"
      ],
      "execution_count": 76,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SC2zynpENlCI"
      },
      "source": [
        "model_B.compile(loss='binary_crossentropy',\n",
        "                optimizer=keras.optimizers.SGD(learning_rate=1e-3),\n",
        "                metrics=['accuracy'])"
      ],
      "execution_count": 77,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tKtZ1QbeNv5s",
        "outputId": "1d8cad8b-d32e-49dd-ba9e-a3c2b444ab2e"
      },
      "source": [
        "history = model_B.fit(X_train_B, y_train_B, epochs=20,\n",
        "                      validation_data=(X_valid_B, y_valid_B))"
      ],
      "execution_count": 78,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/20\n",
            "7/7 [==============================] - 1s 54ms/step - loss: 0.9573 - accuracy: 0.4650 - val_loss: 0.6314 - val_accuracy: 0.6004\n",
            "Epoch 2/20\n",
            "7/7 [==============================] - 0s 32ms/step - loss: 0.5692 - accuracy: 0.7450 - val_loss: 0.4784 - val_accuracy: 0.8529\n",
            "Epoch 3/20\n",
            "7/7 [==============================] - 0s 18ms/step - loss: 0.4503 - accuracy: 0.8650 - val_loss: 0.4102 - val_accuracy: 0.8945\n",
            "Epoch 4/20\n",
            "7/7 [==============================] - 0s 17ms/step - loss: 0.3879 - accuracy: 0.8950 - val_loss: 0.3647 - val_accuracy: 0.9178\n",
            "Epoch 5/20\n",
            "7/7 [==============================] - 0s 18ms/step - loss: 0.3435 - accuracy: 0.9250 - val_loss: 0.3300 - val_accuracy: 0.9320\n",
            "Epoch 6/20\n",
            "7/7 [==============================] - 0s 18ms/step - loss: 0.3081 - accuracy: 0.9300 - val_loss: 0.3019 - val_accuracy: 0.9402\n",
            "Epoch 7/20\n",
            "7/7 [==============================] - 0s 17ms/step - loss: 0.2800 - accuracy: 0.9350 - val_loss: 0.2804 - val_accuracy: 0.9422\n",
            "Epoch 8/20\n",
            "7/7 [==============================] - 0s 18ms/step - loss: 0.2564 - accuracy: 0.9450 - val_loss: 0.2606 - val_accuracy: 0.9473\n",
            "Epoch 9/20\n",
            "7/7 [==============================] - 0s 16ms/step - loss: 0.2362 - accuracy: 0.9550 - val_loss: 0.2428 - val_accuracy: 0.9523\n",
            "Epoch 10/20\n",
            "7/7 [==============================] - 0s 17ms/step - loss: 0.2188 - accuracy: 0.9600 - val_loss: 0.2281 - val_accuracy: 0.9544\n",
            "Epoch 11/20\n",
            "7/7 [==============================] - 0s 16ms/step - loss: 0.2036 - accuracy: 0.9700 - val_loss: 0.2150 - val_accuracy: 0.9584\n",
            "Epoch 12/20\n",
            "7/7 [==============================] - 0s 20ms/step - loss: 0.1898 - accuracy: 0.9700 - val_loss: 0.2036 - val_accuracy: 0.9584\n",
            "Epoch 13/20\n",
            "7/7 [==============================] - 0s 22ms/step - loss: 0.1773 - accuracy: 0.9750 - val_loss: 0.1931 - val_accuracy: 0.9615\n",
            "Epoch 14/20\n",
            "7/7 [==============================] - 0s 21ms/step - loss: 0.1668 - accuracy: 0.9800 - val_loss: 0.1838 - val_accuracy: 0.9635\n",
            "Epoch 15/20\n",
            "7/7 [==============================] - 0s 32ms/step - loss: 0.1570 - accuracy: 0.9900 - val_loss: 0.1746 - val_accuracy: 0.9686\n",
            "Epoch 16/20\n",
            "7/7 [==============================] - 0s 22ms/step - loss: 0.1481 - accuracy: 0.9900 - val_loss: 0.1674 - val_accuracy: 0.9686\n",
            "Epoch 17/20\n",
            "7/7 [==============================] - 0s 22ms/step - loss: 0.1406 - accuracy: 0.9900 - val_loss: 0.1604 - val_accuracy: 0.9706\n",
            "Epoch 18/20\n",
            "7/7 [==============================] - 0s 21ms/step - loss: 0.1334 - accuracy: 0.9900 - val_loss: 0.1539 - val_accuracy: 0.9706\n",
            "Epoch 19/20\n",
            "7/7 [==============================] - 0s 14ms/step - loss: 0.1268 - accuracy: 0.9900 - val_loss: 0.1482 - val_accuracy: 0.9716\n",
            "Epoch 20/20\n",
            "7/7 [==============================] - 0s 17ms/step - loss: 0.1208 - accuracy: 0.9900 - val_loss: 0.1431 - val_accuracy: 0.9716\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "j0aOrtztN69y",
        "outputId": "ad906c1a-ed59-46ed-ed95-88338ae19754"
      },
      "source": [
        "model_B.summary()"
      ],
      "execution_count": 79,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_14\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "flatten_13 (Flatten)         (None, 784)               0         \n",
            "_________________________________________________________________\n",
            "dense_54 (Dense)             (None, 300)               235500    \n",
            "_________________________________________________________________\n",
            "dense_55 (Dense)             (None, 100)               30100     \n",
            "_________________________________________________________________\n",
            "dense_56 (Dense)             (None, 50)                5050      \n",
            "_________________________________________________________________\n",
            "dense_57 (Dense)             (None, 50)                2550      \n",
            "_________________________________________________________________\n",
            "dense_58 (Dense)             (None, 50)                2550      \n",
            "_________________________________________________________________\n",
            "dense_59 (Dense)             (None, 1)                 51        \n",
            "=================================================================\n",
            "Total params: 275,801\n",
            "Trainable params: 275,801\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4dpyWZDyN9LB"
      },
      "source": [
        "model_A = keras.models.load_model('my_model_A.h5')\n",
        "model_B_on_A = keras.models.Sequential(model_A.layers[:-1])\n",
        "model_B_on_A.add(keras. layers.Dense(1, activation='sigmoid'))"
      ],
      "execution_count": 80,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "31KVIX_AOQRw"
      },
      "source": [
        "Note that model_A and model_B_on_A now share some layers. When you train model_B_on_A, it will also affect model_A\n",
        "\n",
        "If you want to avoid that, you need to clone model_A before you reuse its layers. To do this, you clone model A's architecture with clone.model(), then copy its weights (since clone_model() does not clone the weights):"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Yt_U1TNHO3AT"
      },
      "source": [
        "model_A_clone = keras.models.clone_model(model_A)\n",
        "model_A_clone.set_weights(model_A.get_weights())"
      ],
      "execution_count": 81,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9_LTMG5iPCAq"
      },
      "source": [
        "Now you could train model_B_on_A for task B, but since the new output layer was initialized randomly it will make large errors (at least during the first few epochs), so there will be large error gradients that may wreck the reused weights\n",
        "\n",
        "To avoid this, one approach is to freeze the reused layers during the first few epochs, giving the new layer some time to learn reasonable weights\n",
        "\n",
        "To do this, set every layer’s trainable attribute to False and compile the model:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DbtT3_14Pb45"
      },
      "source": [
        "for layer in model_B_on_A.layers[:-1]:\n",
        "  layer.trainable = False\n",
        "\n",
        "model_B_on_A.compile(loss='binary_crossentropy', optimizer='sgd',\n",
        "                     metrics=['accuracy'])"
      ],
      "execution_count": 82,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tvaCbfdgPvGN"
      },
      "source": [
        "Now you can train the model for a few epochs, then unfreeze the reused layers (which requires compiling the model again) and continue training to fine-tune the reused layers for task B\n",
        "\n",
        "After unfreezing the reused layers, it is usually a good idea to reduce the learning rate, once again to avoid damaging the reused weights"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "aiGtiDghP2QU",
        "outputId": "80931dcf-248b-468a-ad81-8561c86f1e51"
      },
      "source": [
        "history = model_B_on_A.fit(X_train_B, y_train_B, epochs=4,\n",
        "                           validation_data=(X_valid_B, y_valid_B))"
      ],
      "execution_count": 83,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/4\n",
            "7/7 [==============================] - 1s 61ms/step - loss: 0.4678 - accuracy: 0.7450 - val_loss: 0.3580 - val_accuracy: 0.8458\n",
            "Epoch 2/4\n",
            "7/7 [==============================] - 0s 17ms/step - loss: 0.2769 - accuracy: 0.9400 - val_loss: 0.2359 - val_accuracy: 0.9432\n",
            "Epoch 3/4\n",
            "7/7 [==============================] - 0s 17ms/step - loss: 0.1834 - accuracy: 0.9650 - val_loss: 0.1779 - val_accuracy: 0.9655\n",
            "Epoch 4/4\n",
            "7/7 [==============================] - 0s 30ms/step - loss: 0.1373 - accuracy: 0.9800 - val_loss: 0.1451 - val_accuracy: 0.9817\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QHhM9dmAQh0s",
        "outputId": "f2073dec-8c35-4085-e699-f7c00294a1d9"
      },
      "source": [
        "for layer in model_B_on_A.layers[:-1]:\n",
        "  layer.trainable = True\n",
        "\n",
        "optimizer = keras.optimizers.SGD(learning_rate=1e-4) # the default lr is 1e-2\n",
        "model_B_on_A.compile(loss='binary_crossentropy', optimizer=optimizer,\n",
        "                     metrics=['accuracy'])\n",
        "history = model_B_on_A.fit(X_train_B, y_train_B, epochs=16,\n",
        "                           validation_data=(X_valid_B, y_valid_B))"
      ],
      "execution_count": 84,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/16\n",
            "7/7 [==============================] - 1s 39ms/step - loss: 0.1180 - accuracy: 0.9900 - val_loss: 0.1433 - val_accuracy: 0.9828\n",
            "Epoch 2/16\n",
            "7/7 [==============================] - 0s 14ms/step - loss: 0.1162 - accuracy: 0.9900 - val_loss: 0.1413 - val_accuracy: 0.9828\n",
            "Epoch 3/16\n",
            "7/7 [==============================] - 0s 17ms/step - loss: 0.1143 - accuracy: 0.9900 - val_loss: 0.1394 - val_accuracy: 0.9828\n",
            "Epoch 4/16\n",
            "7/7 [==============================] - 0s 17ms/step - loss: 0.1124 - accuracy: 0.9900 - val_loss: 0.1376 - val_accuracy: 0.9828\n",
            "Epoch 5/16\n",
            "7/7 [==============================] - 0s 14ms/step - loss: 0.1107 - accuracy: 0.9900 - val_loss: 0.1358 - val_accuracy: 0.9828\n",
            "Epoch 6/16\n",
            "7/7 [==============================] - 0s 17ms/step - loss: 0.1090 - accuracy: 0.9900 - val_loss: 0.1341 - val_accuracy: 0.9838\n",
            "Epoch 7/16\n",
            "7/7 [==============================] - 0s 16ms/step - loss: 0.1074 - accuracy: 0.9900 - val_loss: 0.1326 - val_accuracy: 0.9848\n",
            "Epoch 8/16\n",
            "7/7 [==============================] - 0s 17ms/step - loss: 0.1059 - accuracy: 0.9900 - val_loss: 0.1311 - val_accuracy: 0.9838\n",
            "Epoch 9/16\n",
            "7/7 [==============================] - 0s 18ms/step - loss: 0.1044 - accuracy: 0.9900 - val_loss: 0.1293 - val_accuracy: 0.9838\n",
            "Epoch 10/16\n",
            "7/7 [==============================] - 0s 34ms/step - loss: 0.1027 - accuracy: 0.9900 - val_loss: 0.1280 - val_accuracy: 0.9838\n",
            "Epoch 11/16\n",
            "7/7 [==============================] - 0s 31ms/step - loss: 0.1015 - accuracy: 0.9950 - val_loss: 0.1266 - val_accuracy: 0.9838\n",
            "Epoch 12/16\n",
            "7/7 [==============================] - 0s 21ms/step - loss: 0.1001 - accuracy: 0.9950 - val_loss: 0.1253 - val_accuracy: 0.9838\n",
            "Epoch 13/16\n",
            "7/7 [==============================] - 0s 18ms/step - loss: 0.0988 - accuracy: 0.9950 - val_loss: 0.1238 - val_accuracy: 0.9848\n",
            "Epoch 14/16\n",
            "7/7 [==============================] - 0s 22ms/step - loss: 0.0974 - accuracy: 0.9950 - val_loss: 0.1227 - val_accuracy: 0.9848\n",
            "Epoch 15/16\n",
            "7/7 [==============================] - 0s 33ms/step - loss: 0.0963 - accuracy: 0.9950 - val_loss: 0.1214 - val_accuracy: 0.9848\n",
            "Epoch 16/16\n",
            "7/7 [==============================] - 0s 34ms/step - loss: 0.0951 - accuracy: 0.9950 - val_loss: 0.1202 - val_accuracy: 0.9858\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yVozGQmWQOFl"
      },
      "source": [
        "So, what’s the final verdict?\n",
        "\n",
        "> Indented block\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8R6ZGiaWQxzZ",
        "outputId": "14e56a8f-8ad1-4d90-b0ad-f3817d0048a7"
      },
      "source": [
        "model_B.evaluate(X_test_B, y_test_B)"
      ],
      "execution_count": 85,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "63/63 [==============================] - 0s 3ms/step - loss: 0.1408 - accuracy: 0.9705\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.1408407837152481, 0.9704999923706055]"
            ]
          },
          "metadata": {},
          "execution_count": 85
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hapcNiZmQ91i",
        "outputId": "b76abb93-b2bc-475f-9178-8394005cf8f8"
      },
      "source": [
        "model_B_on_A.evaluate(X_test_B, y_test_B)"
      ],
      "execution_count": 86,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "63/63 [==============================] - 0s 3ms/step - loss: 0.1170 - accuracy: 0.9875\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.11698455363512039, 0.987500011920929]"
            ]
          },
          "metadata": {},
          "execution_count": 86
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dQk3refpRVnz"
      },
      "source": [
        "The model's test accuracy is 99.7%, transfer learning reduced the error rate from 1.4% to 0.8%"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "j-SCRdDbYgFW"
      },
      "source": [
        "Are you convinced? You shouldn’t be: I cheated! I tried many configurations until I found one that demonstrated a strong improvement. If you try to change the classes or the random seed, you will see that the improvement generally drops, or even vanishes or reverses. What I did is called “torturing the data until it confesses.” When a paper just looks too positive, you should be suspicious: perhaps the flashy new technique does not actually help much (in fact, it may even degrade performance), but the authors tried many variants and reported only the best results (which may be due to sheer luck), without mentioning how many failures they encountered on the way. Most of the time, this is not malicious at all, but it is part of the reason so many results in science can never be reproduced\n",
        "\n",
        "Why did I cheat? It turns out that **transfer learning does not work very well with small dense networks**, presumably because small networks learn few patterns, and dense networks learn very specific patterns, which are unlikely to be useful in other tasks\n",
        "\n",
        "**Transfer learning works best with deep convolutional neural networks**, which tend to learn feature detectors that are much more general (especially in the lower layers)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PKDNrHtSYwyw"
      },
      "source": [
        "## Unsupervised Pretraining"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9taQVGIGY2mw"
      },
      "source": [
        "Suppose you want to tackle a complex task for which you don’t have much labeled training data, but unfortunately you cannot find a model trained on a similar task\n",
        "\n",
        "Don’t lose hope! First, you should try to gather more labeled training data, but if you can’t, you may still be able to perform unsupervised pretraining (see Figure 11-5)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wyp6c2ZaZEvJ"
      },
      "source": [
        "<p align='center'>\n",
        "<img src='data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAooAAAGtCAYAAACRNVWfAAAgAElEQVR4AexdBXhUR9dO6ffXaL+696sbLfUWiru7u2tpcSlenCLFpZTi7u7uFiwkIRAICYFABEtCEiR5/+edZZbNshvdTXZvzn2e+9x7556xd87eeffMzBkPyCEICAKCgCAgCAgCgkA6ETh06FA6U5DoroiAhysWSsokCAgCgoAgIAgIAu6FQMWKFREbG+tehZbSJouAEMVkIRIBQUAQEAQEAUFAEEgKgWPHjiFfvnzYtGlTUmLyzg0REKLoho0mRRYEBAFBQBAQBFwJgaFDh6JChQpo1aqVKxVLyuIABIQoOgBESUIQEAQEAUFAEMiqCMTFxaFKlSqoVasWihYtisjIyKwKhSHrLUTRkM0qlRIEBAFBQBAQBDIGgW3btqF06dKoU6cOKlWqhFmzZmVMxpJLhiAgRDFDYJZMBAFBQBAQBAQBYyLQqVMnVKtWTRHFmjVrokGDBsasaBatlRDFLNrwUm1BQBAQBAQBQSC9CHCYuVSpUqhdu7YiiryWLVsWfn5+6U1a4rsIAkIUXaQhpBiCgCAgCAgCgoC7IbB48WKUK1dOkcSqVauah5/HjBnjblWR8tpBQIiiHWAkWBAQBAQBQUAQEASSRqB+/frgcDNJ4tixY9UQNJ9pWbx//37SkeWtWyAgRNEtmkkKKQgIAoKAICAIuBYCQUFBKFmypCKFDRs2xMCBA9G8eXO1+rlMmTLYv3+/axVYSpMmBIQopgk2iSQICAKCgCAgCGRtBMaNG6dWOdeoUQNdunRB9erVFVnklRbGfv36ZW2ADFJ7IYoGaUiphiAgCAgCgoAgkJEIcHiZw8w8+/Tpo3wptm/fXlkU6VORrnJiYmIyskiSlxMQEKLoBFAlSUFAEBAEBAFBwMgIHD9+XK1uJiFs3bo1unbtCloWGzVqhF9//dU8/Lx+/Xojw5Al6iZEMUs0s1RSEBAEBAFBQBBwHAL9+/dXFkRaEzt37qxIIh1uc8h5wIAB6kri+MsvvzguU0kpUxAQopgpsEumgoAgIAgIAoKAeyIQGxurVjfTmsjh5549eypiyGfOT+zWrRu4GprvSpQogYiICPesqJRaISBEURRBEBAEBAFBQBAQBFKMwNatW83Dzk2bNlXEsGXLlmoIulWrVurK4WgSx/Lly2PmzJkpTlsEXQ8BIYqu1yZSIkFAEBAEBAFBwGUR4ApnvWWftirWrVsXlietiXqxC69yuC8CQhTdt+2k5IKAICAICAKCQIYiEB4ejuLFi6uhZpJFfWpiyKsO09ciRYrAx8cnQ8spmTkOASGKjsNSUhIEBAFBQBAQBAyNAIed6QKH8xL12atXL2U9JEnkAhYdrq9c7DJ//nxD42LkyglRNHLrSt0EAUFAEBAEBIEMQIA7s3AFdNmyZTMgN8kiIxEQopiRaEtegoAgIAgIAoKAwRCIj49Xq5yFKBqsYR9UR4iiMdtVaiUICAKCgCAgCGQIAkIUMwTmTMtEiGKmQS8ZCwKCgCAgCAgC7o+AEEX3b8OkaiBEMSl05J0gIAgIAoKAICAIJImAEMUk4XH7l0IU3b4JpQKCgCAgCAgCgkDmISBEMfOwz4ichShmBMqShyAgCAgCgoAgYFAELIlimTJlDFrLrFstIYpZt+2l5oKAICAICAKCgEMQ0O5xypUr55D0JBHXQUCIouu0hZREEBAEBAFBIAsgcOHCBQwfPhzVq1dH5cqV3f7kDiz0n8h9nWlRrFq1KqpUqeL29WLbDBo0CN7e3llAK+1XUYiifWzkjSAgCAgCgoAg4FAE/vnnHxQsWBBTpkyBl5cXfH19DXH6+flBn0ap0+nTpzF9+nSULl1a7TbDIfaseAhRzIqtLnUWBAQBQUAQyHAESBJpabt161aG5y0Zph2BhIQEcBvCTp06pT0RN44pRNGNG0+KLggIAoKAIOAeCHC4mZbEmzdvukeBpZSPIFCqVCkcPHjwkXCjBwhRNHoLS/0EAUFAEBAEMh0BzkmcOHFippdDCpB2BNatW4fffvst7Qm4aUwhim7acFJsQUAQEAQEAfdBoFGjRvDx8XGfAktJH0EgPDwcRYsWfSTc6AFCFI3ewlI/QUAQEAQEgUxHgEQxKCgo08shBUg7ApxbWrhw4bQn4KYxhSi6acNJsQUBQUAQEATcBwESxcDAQPcpsJT0EQQ4v7RIkSKPhBs9QIii0VtY6icICAKCgCCQ6QgIUcz0Jkh3AYQophtCSUAQEAQEAUFAEBAEbCEgRNEWKu4VJkTRvdpLSisICAKCgCAgCLgNAq5CFM+cOYPBgwcrf47cUWXo0KE4d+6cQ3C8c+eOQ9JJLpG4uLjkRJzyXoiiU2CVRAUBQUAQEAQEAUHAFYhipUqV4OHhoc5nnnkGzz77rPmZpDE9R40aNTBgwID0JJGiuJwjmFluhoQopqiJREgQEAQEAUFAEBAEUotAZhPFnDlzKlJYp04d3Lhxw1z8iIgI1KtXT7378ccfzeGpubl48aKK/+eff6YmWqpluTUgie6MGTNSHdcREYQoOgJFSUMQEAQEAUFAEBAEHkEgM4niyJEjFcHq1avXI+XSAc2aNVMygwYN0kEpvt6+fVvFnTBhQorjpEUwNDRU5bNkyZK0RE93HCGK6YZQEhAEBAFBQBAQBAQBWwhkJlGkFe7555+3VaxEYZTLnj27OYzDyQyznnvYsWNHFU7BrVu34vHHH1fPHMrOli0boqOjlXNxpuXt7a22LmQ6PHPlyoXg4GBzHrzJnz+/OT3LF7Vq1TKHz58/H0888YR6fu655/DKK69k+HaIQhQtW0fuBQFBQBAQBAQBQcBhCGQWUbx8+bIiVxxyTu6oWrWqkuWCFx6ct0hyd+/evURRO3TooMIZyD2s27Rpo565F3LPnj2VrKenpwpj/C+//FIRSg4Za8IYGxtrTjNfvnzm9MyBAGrXrm0OP3XqFFq1aqWeK1euDFo+M3pRixBFy9aRe0FAEBAEBAFBQBBwGAKZRRQ3b96syFWPHj2SrQvnGJLILV++XMlqi6I1UezUqZOS0+GRkZHqedy4ceY8NFF87733zGG88ff3V7KWi2cKFCigwhIJAiC5ZXn0ERISop4XLVqkgzL0KkQxQ+GWzAQBQUAQEAQEgayDQGYRxaVLlypyldT8RN0KY8aMUbKLFy9WQSklitpqOWzYMJ0UNFFcuHChOUzffPzxx/jPf/6jH5FSokhLJ4njzJkzzXEz8kaIYkaiLXkJAoKAICAICAJZCIHMIooHDx5U5Kpr167Jot2+fXsle+TIESWbWqI4fPhwcx5Mg6SOcxStD52uHn62RxTr1q2r0tDxhShqJDL2+tCmm7H5Sm6CgCAgCAgCgkCWQSCziCIBJmH79ttvk8X6gw8+ULIxMTFKVhO6+/fvJ4rbuXNnJaeHnrVF0RZRPH36dKK4fNBzD/UiGSGKj0DkUgFCFF2qOaQwgoAgIAgIAkZEIDOJYsmSJRWx27Ztm11o9+/fr2TKly9vltGLWayJol6NnBRR1EPPtvIkIbUcera3mIUrpC3nKIpF0dw0GXojRDFD4ZbMBAFBQBAQBLIiAplJFDm3joSL57Fjxx6Bn2H6/ZUrV8zvmzdvrsK9vLzMYXR9o2U1UaS7G4ZxjqM+NFEk2bM8Tp48qWRbtGhhDi5XrpwKu3XrljmM5WCadLejD1onGWZr3qOWceZV5ig6E11JWxAQBAQBQUAQyMIIZCZRJOyauJFo/fTTT8qNTffu3ZEjRw5Fvhi+b9++RC10/Phx8zuudNbuaXScu3fvKvnr168ruTfeeANdunQBwy3jFixYENu3bwdXRTMfnpbHunXrzOFcnU2sKPPNN98kkr169ap6/uSTT9C7d2/oOY6WaTnzXoiiM9GVtAUBQUAQEAQEgSyMQGYTRQ09F6w8/fTTZmLGe0vrnpbT1507d+Kzzz5T8nSgvWDBAkU6Sda0RZGyXCyjSeCBAwfUIhY+09VO7ty5ze+KFi36iANvxl+zZg3eeecdJffSSy9h/fr1oGuf77//HgkJCbo4aNiwoTktvejG/NLJN0IUnQywJC8ICAKCgCAgCGRVBFyFKGr8aY1ztMNqEkcOTfM4evSoInRr165Vz9zmLz4+XmefrivLzfQy+hCimNGIS36CgCAgCAgCgkAWQcDViKKzYddD3StXrnR2VhmWvhDFDINaMhIEBAFBQBAQBLIWAlmNKGr/jUuWLDFMQwtRNExTSkUEAUFAEBAEBAHXQoBEMSgoyLUK5cTScNVyu3bt4Ofn58RcMjZprsouXLhwxmbqArklXnrkAgWSIggCgoAgIAgIAkZDoGnTpjZd0xitnkauD8lviRIljFxFm3UTomgTFgkUBAQBQUAQEAQch8CECRNguXOJ41KWlDIKAfpv5K40We0QopjVWlzqKwgIAoKAIJDhCHCbO/oTDA0NzfC8JcP0I8CV1oUKFTLUUHpKURGimFKkRE4QEAQEAUFAEEgHAvPnz0eRIkVAx9FyuA8CdPlTs2ZNjBw50n0K7cCSClF0IJiSlCAgCAgCgoAgkBQCdCRNB9TcWYTOrLk6mA6q3f2kOxz6TuTV3euiy8/9rwcPHoyff/4Z//77b1LNauh3QhQN3bxSOUFAEBAEBAFXQ4CrZ2fPno02bdqAi1yaNWvmtif3g+bOLqVKlULJkiVRvHhx9cxwd64Xy862IUG03P/a1XQpI8ojRDEjUJY8BAFBQBAQBAQBAyPArfU4PFu2bFkD1zJrVk2IYtZsd6m1ICAICAKCgCDgEAS4NV/9+vWFKDoETddLRIii67WJlEgQEAQEAUFAEHAbBIQouk1TpamgQhTTBJtEEgQEAUFAEBAEBAEiIETR2HogRNHY7Su1EwQEAUFAEBAEnIqAEEWnwpvpiQtRzPQmkAIIAoKAICAICALui4AQRfdtu5SUXIhiSlASGUFAEBAEBAFBQBCwiYAQRZuwGCZQiKJhmlIqIggIAoKAICAIZDwCQhQzHvOMzFGIYkaiLXkJAoKAICAICAIGQ0CIosEa1Ko6QhStAJFHQUAQEAQEAUFAEEg5AkIUU46VO0oKUXTHVpMyCwKCgCAgCAgCLoKAEEUXaQgnFUOIopOAlWQFAUFAEBAEBIGsgIAQRWO3shBFY7ev1E4QEAQEAUFAEHAqAkIUnQpvpicuRDHTm0AKIAgIAoKAICAIuDcCDRo0kL2e3bsJ7ZZeiKJdaOSFICAICAL2EQgPD0dYWJicLooB28fVj2vXriE8LAzh4e57XouIQGhoKOrXq4caNaqjXNmyqj4Md+d6seyslxyAEEXRAkFAEBAEUoHA7NmzUbZsWZQuXRrVqlVD1apV5XRBDNhGJUuWxPjx41PRuhkjOmfObFSuWB5ly5ZB1arVUKVqVbc+q1Wvjtq1a6Nu3bqoU6cOqlWrrn4T7lwvtku5cuVQoVwZjBszJmMUw0VzEaLoog0jxRIEBAHXQiAuLg61atVCvXr1EBQU5FqFk9LYRIAWobZt26Jo0aK4dOmSTZmMDLxz5w7q1K6JajVqYbPnWfiG34ffNRjmPHMd8ONpkDqxPjtOXEDr3zqgRPGiWdbCKEQxI78SkpcgIAi4LQKtWrXC8OHD3bb8WbngCxcuRJUqVTIdgvbtfkPX3v0RFAOcuhKH4xej5HRxDLwux+JCDDBp1hJUKFc203UoMwogRDEzUJc8BQFBwK0Q2LFjB6pXr+5WZZbCJkagRYsWWLx4ceLADHzas2c3ylesBN+w+zgRHI1jQZFCEl2cJFoSef+bQKNmLbFw/vwM1BrXyEqIomu0g5RCEBAEXBiBzp07Y9myZS5cQilacgjs2rULzZs3T07Mae/79u6F8dMWwC8CQhLdiCCSLJLU+4Tex8L1u9Cgfn2n6YirJixE0VVbRsolCAgCLoNApUqV4O/v7zLlkYKkHoGIiAhUrFgx9REdFKNJ40bYcMAHXiGxOC7WRDezpkbi5KXbOHzuGooXKewgjXCfZIQouk9bSUkFAUEgkxCoXLkyzp49m0m5S7aOQICuaNiOmXU0b9oYmw76gnPeLIc05d495mmeuHQbh/wjUKJYkcxSoUzLV4hipkEvGQsCgoC7IMCFEEIU3aW1bJeTFkUhiu5BylyRPJuIYrgQRds/LwkVBAQBQSBrIyBE0f3b34hEkXPn9GmPXFm/18+82oujw7WsfrZ31XIpSdNeGukJt8zfWWUQi6L7fwOkBoKAICAIOA0BIYpOgzbDEjYaUSQhOh0ej+BYwPvqHZsLZChzIQoIum1aQMNnDn2H3AP8IuJtxtGEjbKX7wIBkUkvvqGcb9g9lSZd/vBZp+Hsq87rTARU/txHJTAa8Loc4/ByCFHMsJ+qZCQICAKCgPsh4AyiuGnTJpQqVQonT560C0jPnj0T+f8LCQlB7ty5sXLlSrtx9AvuHNO9e3f9aPe6b98+fPvtt5k2tE4fh/nz58enn36qViXfunXLblnT88JIRJEEiURv4pwVyFu4OJZv97Tpl/HMNaBBi99Qrkot8J4kcadXIHLk/BbjZy9Vrno02bIkdQwjufwuV1607tQDAbdsk0XK+YTexayVW/Hx519i46HTpsU6GbCq2VTuaEVmh4yfju9+yoP3PvoEVes2geeFmzh73XaZLeuZmnshiun59UlcQUAQEAQMjoAziOLgwYPh4eGBVatW2UXv3XffVTJa4PTp0+p50KBBOsjulWnnyZPH7nv9gkSNsocOHdJBGXblanLmzXpyK0Te8zx69KjDy2A0ohgQBbTp1FvhNXneakX6rIkPrWvPv/Cikgm8TSffd7D+4Cn13LHPYLtkiiSMO2WzLXLlL4zLd2yTLsqRgP41ZbaSXbbtsLJuWpfDWc/BcUCegsVV3m++8z8UK2PSJ5Z78aYDyuJqIpTpn5spRNHhP0lJUBAQBAQB4yDgDKI4YsQI1cGtW7fOLlCff/65ktECd+/exeHDh8EVvMkd7CwLF07elcfSpUtVHkeOHEkuSYe+nzdvnsqXPiotj+eeew5PPPGEZZBD7o1IFDv0GKAwnLponU2iGBQNvPHW20qGRNHk6PsWlm49iL2+l3HyUoxy1aPJFF33cBibrmDCHhDF/MVKJSKKlGU63lfuKCsmieLof01tuWKHp5komtO8HKPCTtEt0AOfhLaIo06XZVDlSkbW/zowaMxUVbeW7X8HfxGX70GR38ceewwvvPgyzkfCYcPgQhQd8jOURAQBQUAQMCYCrkIUk0KXexmTRMbGxiqxpIgih7t9fX2V3OrVq1Vna4sochjY09MT58+fTypr87vAwEBER0ebn5O64ZaILKP10adPHxXu6L2ZhSiaSNOpkDg1n4/z+o5fNC2GOXk5BrTO7TgRgKVbDuLohZu4a4Moksz53wD8whOwctcxrN17CpwXOHH2YtVmmihSjvMEmeaBs2GKmG45ek4NZ3M+oyaRJIxnbwDnb0Gle+4WsPXoOeWvkHMv7c01ZHxaVCvWrK/yZXwSOYYzzwYt26nwLZ7+8AqJc4jfSiGK1r9UeRYEBAFBQBAwI+BMosi5ivaOHDlyqA5Pvz937px6HjdunA5S17x586pwEi+ec+fOVVa5IkUS+3xjXlqG11deeQXasmdNFDlv0FL2xRdfxJkzZxLlW6NGDfz444/grieWst9//z1iYmISydp6SEhIeCS4du3aKq2UEs5HErATYGSiOH3pJmVJo9Xu4RmnLIFvvvWOwlMNPYfEYfMRX/Xc+89x5nmLZ64l4PMvv0nUhuNnLlPP+YuWNFsUA6OACbNM4bq9abEcPO5fJauJoveVOEXyvv3x50Rp0sq3/oAv/CISFKkj2StcqjxyfvMDZq3ankg2x1ffgqSW1k1bFkiGcTHPieAos29KynKGa/4ipVRaB85eBUmwIxycC1G088OSYEFAEBAEBAGoBSWO9qOoh57nz5+P69ev4/Lly+aTi1YiIyNhPfTM3WHYQf/111/mZsmZM6c5jHGWLFli7nCLFy9ulvPy8lLhr7/+Oo4dOwYSJz1HkGlaEsWnn35ayU6cOBE3b95UVkUdZjnsTQKtCcOAAQPUPMdffvlFhX333XfmvFNyc//+fVUvplffCdukGZkojvpnPvb7XcXOE4EPz5NBOBZ4C2+8+ZAocrh4s6dpnmv3wX8pKx4Xrbz1v/dUm/UY9BcO+kdg5JS55nYtWLw0Lt3hyuq7WL7NU4W/8fY7WLb1sMozX5GSZlkSRd/Qezh3E3j2uedV+B/DxytH1XPX7MJTT5n0iotNaC1k3gWKlTbH79xnCBZs2Ifq9ZqoMJJF5m1pgbQmjSRwyjJ5HYpUdu49VMWtUrsxSI6TimudVlLPQhRT8isWGUFAEBAEsigCzrQoaqKV1FXDri2Ko0aNUkEHDhxQnaL14pbt203WmWLFiumoKFCggJI1Bzy4oUWQeWuiOGPGDPVsvbKa1j/KNWzY0JwELYoM27JlizmMN4UKFVLhiQKTeGjWrJmSZ1qpJZhJJJvolZGJInFL7iRpIlHc4umnZHsMGY2LscC0JRvVc5e+Q9UCFs4RvBQHzFhmsj5rokiXOp/n/FbJ+ly9a7JgXonDDQAkdMyfRPFCNNB3+AT1zEU2VxMAn9B7aqjYM+CGCq9ap7FyY8MyFSlVXoXNXbsDIfehiCbT/OKr71U4Sefx4Gi7VkUSQQ49l69qskSzHF9++6MaEueKbCGKiX4GaXp4dIJImpKRSIKAICAIGBcBZxLFJk2aYPjw4eAqaMuTVsPXXntNdZYaWWui2KlTJ/WeFkjrgx2mXsxy584dJZcrVy5rMSxYsEC900RRDznTOsk5irR23rhxA/Hx8fj666+RPXt2cxp6pTLTtzw6duyYqNyW72zdb9u2Dbt37watkiz3s88+a55raUs+LWFGJookXp16D0G77v0Tnb8PGInnHlj2bBFFrmyuUb+5qf3PRygrn7aqXblvWvXMxSwkiSRdbBuuguYzCRhPuqEZOn66erd8h6cim5/n/AaPZcuG40Hh8LwQgcPnQ3H4fBjO3byP9z/6BP/5v/9TcwlpUSxUoqyKS3c/J4IfzjOs3dg0h9X/mmkRji6XrSutnVMXbcCMZZvxa9e+Kr0XXnpZzXWkSyBHkEWxKKblVydxBAFBQBDIIgg4kyju2LHDLop6WFkLWBNFPfQbFxenRczXt956S1kRGRAWFqY6z1q1apnf6xu6oiEB0ERRk1OG2Ts1MdT5c9jY8mjfvr2KaxmW0ns9j7Jx48YpjZIiOSMTRQ7rcoiWljXL8zqAt98xuViyRRRpuStQzDSfzzTfzzSMS2IVmgA8k/1Z5C1cDCSNnucjVJtWrftwSJdyp8PvY+6anerd8u1H1BzBZ5/7r13d0TrlFx6PizFAweJllCxd+Ry/GK1I3cU4oGajFiqcq5u5ytoWQdQEkFdaLU+H3VckdsqCNSouySbnQWo5W2mkNEyIYop+ZiIkCAgCgkDWRMCZRDE17nGsiWL16tVVh0i3OdbHhx9+qBxZMzw8PFzJ1alTx1oMPj4+6p0mii+88AKef/55RRxp6du6das6eb9//37s2bPHnIYmivfu3TOH8SY9RJHxSSbee++9RGmm98HIRHHqQrrHMa0mJinSxIjk6403H7rHsR56JlHURO1MRIKy6JE4Mf7VeOCFl15CnkIPiGLANdUu1eo2QWDMQ4uib9h9LNq4X70jUYwE8OSTT+HFl17B/PV7MG3pRvy7ZMODcyNIamcu36IW3dCiqPPnDjCpIYrHgqKUOxztAkfXm6SSw9VKhz74GOc5dO0AB+BCFNP7C5T4goAgIAgYGAFXJYr9+vVTHaKthTbsKPXQM5uGzx999NEjrbR4scm1iSaKegX1I4KAWqximVd6iOJPP/2EN99801Y2qqxcyOPIw9BEMYV+FK2JIn0lNvmlg8KbO7bwvSaZZj+KRUsh5C5w9ppJhz778mtFIjUxo7uc4ZNmqjT00PNnX3ylns9cT1DxSCZ50qo5d+0urN3nDa8Q02KWtBJFDnl/8nlOcIiZ8yB1eTh87X/LVFYOgXMLQyGK6fslyRzF9OEnsQUBQSALIOCqRDEoKEh1yGXKlEnUCtrljeVilhYtTEN5XP1sedDySBKpieKsWbPUc9euXS3FEBoaqsI5pK2P9BDFRo0aqfR27typk1PXCRNMCyGGDh2aKDy9D0IUH13MQuK24YDJXQ4XlXComg60Oew8bKKJ/JHIcVibw8Tlq9VVbbZ8x1EVRpLI+YrvvPu+CteLWfqNnKSeW7Xvrhx3M036Rdx69LwKf+nlV9VCGuafFqJIUsi4pSuZLOoLNuxV5WA+nHf5+wCTM/tuA0bC/6YMPaf3tyNEMb0ISnxBQBAwPALOIIpDhgxRnSYdXts73n/f1AHr97TmkdRx8Ys+9IKWypUrg8PYPXr0UDKU40pnfXAxCsN4jhkzBitWrFArjHWY5RZ+XHnM8AYNGmDNmjUYP348smXLpsK4jaA+ypUrp8Ksh761ixwtZ+vKxTL/+c9/VHxiwXyaNm2qnrnvs6MPIxLF37qanJP/Pd/2Fn7cmeWll15WmJJY0WK46YEfxW4DRpjc48QCtRuZFo6UKF8V42ctQ/O2XVQc6kCeQkWVH0X6KOSKaK0vvw8cibEzFuPjz74why3bfljNFeTQ72dffK3CK9dqgPGzlqJb/+F4zOMxFbbp8Bm1WwuHnvMULKbCrIeeq9RtrMJpybSeo0iiyEUqehU1y9S+R3+Mm7kUVWo3VPE+z/m1qh/jUj69VkUZenb0L1LSEwQEAUHAQAg4gyguWrQIXKxiSdCsIeOKaMuVyvS1SBK3bNmyRKL0d6g7cF45r7Bs2bKwXhBy+/ZtFC1a1CybO3dunDhxApyXeOrUqURp9urVC48//rhZ9ocffoDlsDOF+/fvjxIlSsB6jiLL8+WXXyZKz9YDnWqXL29yj6LLz4g5Q7IAACAASURBVB1bnHEYjSie4WrjCdPx0ac5sHDDPkUCrckQh2dLV6yO73PnVTugkFztOHlBkbsx0xerxR+MQxJJIqfbgNf563bjh9z5UL1+M+XKhmSLPhK508qPeQqaZX/4Ob+SfSZ7dqzd66WcZNNHIq17zX/rgsceM/3BYJpffv09Nhw8DZad6dEiye33vs+dH5wjybIwnDu0dO0/Ap/kyInTYffU3EXruqnyhN2D54UbyJ2/iLk8zKd2k9ZqnqKjVjwzbyGKzvhVSpqCgCAgCBgEAWcQRYNA4zbVMBJR5E4jJErcSo9b1vlcfTi3MBGhCopSc/SCYkzz9I5djMTJy7Fq2JernJkGT8ahFZA+ENX1NsAt/mjlI5nTcrxytxQSSy4iIaGjVZArn3nlu2MPyB4teVyBrdM8FwmwHPTBqPPklfnp8rEcOi+SXPp5VD4U7VgEKUsrJ/M+H/Wg7NGmPB1lSdR4ClF0m5+6FFQQEAQEgYxHQIhixmPu6BwNRRS5ivcBydPEShMa66v1e/3M60PZh4RRD/Navn/0nnGjH3GETZKo09T58FmnqfeW1jK8WsrpcB2m8k1UzofpW8vyWeej42sZR1yFKDr6FynpCQKCgCBgIASEKLp/YxqOKFqQMkcQIUnjURJqiYmJKIajRLHE+6e7/y8j+RrIYpbkMRIJQUAQyOIIcE9k+jCUw30R4A4zbMfMOpo2aYRNh/3UIgxLAiL3SRM0V8GHRPFIwA0UK1Ios1Qo0/IVophp0EvGgoAg4C4IcFGJpaNpdym3lPMhAt7e3qhZs+bDgAy+a9f2N8xesQU+offNc/RchQRJOZImqxzKPnUlDhsOeKNShfIZrDmZn50QxcxvAymBICAIuDgC8+fPV7uNuHgxpXhJIMB9pEePHp2EhHNfLVywAM1b/6b8+gkxS5qYuRo+JIpcjNOj32D8NXKEcxXFBVMXouiCjSJFEgQEAddCgD4I6dR6+/btrlUwKU2KEDh+/LhyCxQTE5MieWcIcT/sShXKYcaS9WrFMFfz6kUXcjUtpnFFHLhoh067V+86hsIF84MuprLaIUQxq7W41FcQEATShMDJkyfV3sncuUQO90Fg+fLlKFiwoEtMHaCzcpKN4eOngu5pSEDO3jDAeR2gOxv6R+TJeyPUi+3jFwFMnrMMBQsUwPbt29xH8R1YUiGKDgRTkhIEBAFjIxAWFqZ2KyldujS6d++OYcOGgVvNufs5adIk6NMIdWIdevfuDe4cw3mJgYGBLqOY169fQ9MmjVGiREn80qEruv8xBN36DnLr8/c/BmPg8LEYPHKcOvlshDq179oL/K3XqV0rSy9mE6LoMp8PKYggIAi4CwLBwcFYuXIllixZgqVLl7rtyR1eWA92hlwRzOH1xYsXq51f3LleLDstiQEBAS6rUtShtWtWY/mypVi+bJnbniuWL8PKFctRvVpVVK5UEWXLlMaqlSvAcHeuF9uFdQgMvOCyOpRRBROimFFISz6CgCAgCLgoAtzur1atWqhYsaKLllCK5eoIcF9wWm+pS3IYCwEhisZqT6mNICAICAKpRsCSKMbFxaU6vkTI2ghwsVf9+vWFKBpUDYQoGrRhpVqCgCAgCKQUASGKKUVK5GwhIETRFirGCROiaJy2lJoIAoKAIJAmBIQopgk2ifQAASGKxlYFIYrGbl+pnSAgCAgCySIgRDFZiEQgCQSEKCYBjgFeCVE0QCNKFQQBQUAQSA8CQhTTg57EFaJobB0Qomjs9pXaCQKCgCCQLAJCFJOFSASSQECIYhLgGOCVEEUDNKJUQRAQBASB9CAgRDE96ElcIYrG1gEhisZuX6mdICAICALJIiBEMVmIRCAJBIQoJgGOAV4JUTRAI0oVBAFBQBBIDwJCFNODnsQVomhsHRCiaOz2ldoJAoKAIJAsAkIUk4VIBJJAQIhiEuAY4JUQRQM0olRBEBAEBIH0ICBEMT3oSVwhisbWASGKxm5fqZ0gIAgIAskiIEQxWYhEIAkEhCgmAY4BXglRNEAjShUEAUFAEEgPApZEMT3pSNysi0CDBg1kr2eDNr8QRYM2rFTLPgLBwcFYsWIFli1bJqeLYrBy5UoEBgbab8RMfhMcfBFrVq/CyuXLsHLFcrc9V61cgXVr16BM6VKoXKkiypUtg6VLFmP1yhVuWyfdHqxbQMB5m5py69Yt9dsfP348pk2bhhs3btiUc2bgpUvBWLN6tSF0iJhXq1oFlStVQJlSJUHseeq2cOerPR1ypm64WtpCFF2tRaQ8TkPg0qVLaNKkCWg96dGjB4YMGYLBgwfL6WIYsF169uyJSpUqKQvF2bNnnaYTqU04NPQqWjRvijJlyqBNh674/Y/B6NZnoFufv/cdhCF/jcdgdY5Dd0PUaTDademB8hUqok7tmvD19TE39caNGzF06FDMmTMHI0eOxB9//IHGjRujdu3amDVrllnOWTfh4eEPdKgs2nToYhAdGoxBI8aa9GjkePze1wi/C61DFVCvbm14eXk5SyVcPl0hii7fRFJARyDg7e2N/PnzY+7cuY5ITtLIIATYqefJkwfbtm3LoBztZ8OOokihghg5cRp8Qu/h7A0Y57wOnNWnUep1HfANj8e/C1ejSOFCOLB/P6ZMmYJr166pRr58+TIOHTqEHTt2YPPmzVi9ejXat2+PcuXKISYmxr4ipOONn58fChXMjxETp8H76j2cIeYGwtuIOuQTdh/TFq1BoUIFsX79unS0vvtGFaLovm0nJU8FAkWLFlUdQiqiiKiLIECLYsGCBZ3Weae0mhXKlsasZZtw7hZwIjgax4OicCwoUk4XxuD4xWhFxrZ4nsFXOb/E7t27kJCQgIsXL+LIkSM4fPgw9u3bhzVr1mDevHmYMWOGOj///POUqkWq5CpXLI/pi9biXOQDHbooOuTqv6HjwdHqT9SmQ6dRIF9exMbGpqrNjSAsRNEIrSh1SBKB6dOno0OHDknKyEvXRmDAgAEYPXp0phVy7pzZaNGmvSKJxx907rzK6R4Y0GpXq14jRISHIyAgQJFEzk8sUKAAvvzyS1StWhXjxo1TIw6cH7tlyxY19cGRCrdw/jw0ad5akUST3kSK/rjJb4hkluS+Z7+hGD7sT0eqhVukJUTRLZpJCpkeBOrVqwdPT8/0JCFxMxmBkydPon79+plWilYtW2Duqq3wDbuvLIhCEN2DILKdTl6Owdbj5/Hbr78qSyJ16c8//8Q333wDDj/zpIWxZcuWqFixInbt2gVfX19UqVIFYWFhDtO59u3aYvaKzfAJuyc65CYEUf/OSRRPXbmD9ftPqUVfDlMKN0lIiKKbNJQUM+0IVK5cGVevXk17AhIz0xEIDQ0F25H+2jLjqFq5EjYe9IVXSKxYgdyok2cHfzo8HgPHTMWcWTPh4+Oj5iV+9NFHNtVo7dq16NWrFy5cuIC9e/cqQmlTMA2BTZs0wqZDfvC6LDqkCZg7XU9cuo0j56+jeNHCaWh9944iRNG9209KnwIEOKx05cqVFEiKiKsiwPYjUeT8ssw4alSrgvX7TglRdCOSSBJCougbcR+j/52LI4cOqpGFDRs24LfffrOrRr///rv6Y8mFJ/ny5bMrl9oXzZs2xib+2RCi6JZ/tkgUD/mHo0SxIqltereXF6Lo9k0oFUgOASGKySHk+u9JFDkUmLlE0VuIohsSRb+IBAwZ+w/27tmFAwcOKLc4nPNq7+Aq6IMHD4L+VidOnGhPLNXhQhTdZ7qCLUuniShGCFFMteZLBEHADRAQougGjZRMEY1GFGnp4srpk5duJ2ldYedkLeN1Oca06joZ0sa5eYzLvGx1fDqM5WCa+jkjr7psJy/F4FRIrLkcOjzdZeEihFtAibKVcOjgAezevRuTJ08GrYb2jp07d4IL4OjvsH///vbEUh3uaKLobB1i+tSL5HRIl8OwOvTgdyZEMdUqLxEEAfdBwBlEkXPlIiMjk5wzFxcXh+jo6ERAca5dStwr0NfbzZs3E8W19XDv3j01THb//n1brzMkjPUJCgpCRESE0/IzElHUHeuxi1E4dO4a6H7DFiEigfO8cBOHz19PRBZ3eQfDM+BGkmSRnfte3xAcOBOqFnMwT3t5HA28hV2nLiZLKG3FT08Yy3TqShwCbwOe529g69FzqsznI+GwRUPMIzAayF+kNGbPWYz1G7Zg1KjR6Natm11dpcsckkn+dulTUftdtBshhS8cSRQf6lCkSYfs/Gk4cSla6cojOnQqGEcu2Nchpk/it8fnEg6cDQOJPMOs21uXg3pKHaLLKGsZZz4zf2/qUDRwJOCBDp0OgUmHHLtoSIhiChVdxAQBd0TAGURx0qRJ8PDwUE567WHyww8/4PHHHzcPl5IkMs7AgQPtRTGHP/fcc8rRtDnAzs2qVatUmrSCZPTBbdBKliyp8me9eH7wwQc4evSow4tiNKLofxPo9sdwhdnUxWuVA2/rDtX/BvDZF18pmeA44FRIHDYc9FXP3fqPwJlrCXY75QvRUHI/5imAkHuw2ckzP67iHj9rsZKdt3YXfK7etZumdfnS88wO3jfsHg6eCUOuvIVU/lqH3nznXazadQz+1+2XOyV5Mw8OOx8KvIvSZUth1eI2WLpgBBYuWoRhw8ciNPS6TT3lQha6YuIfIO4O5KiFcI4miudvAZ16D1bY/bt4vU0dCogE3vvgIyXDexLzjYd91HO3gSNw9pptjIldaIJJh/IXLQnqH8OscdcYj/pnrkpz2bYj8L5y5xE563iOeGbe1KEDZ8Px7U8/q/y1Dv3v/Y+wZs9J5UPTVrnTkr8QRZs/FwkUBIyBgDOI4tixY9WHiXtG2zty5MihZPR7Wib++9//4p9//tFBdq/PPPMMfv75Z7vv9YvMJIovvfSSqh9djZw+fRoLFizAY489psJSYg3VdUjJ1WhEkZ18x16mTv7vhasVYbPuvDhk+t4HHys8Sfa4CIJWt+f++zyGjJ+B0+H3zR0yO0N9Mp2gGFMn/0Oe/Lh6/2Enr2X0lSuCHxLF3fAJfUgUtYzl1bqMlu+s761lLZ9p8fQOuYPszz6n6tdj4F9YuesEhk+epZ7Z4XPhgGnhx6MExTIte/dnrgGjZ+/B8CF1gYAWQBitiCeBO2tw/84hXAzajSWL/sbfUybg8BF/3Iq8Z1bF4cOHK4sit/kjcXTE4WiiGBAFtPu9n8Lrn4VrbepQUDTw2utvKBlabjm8v+2YP57J/iyGTpgBv/B4MwG0br9wmHQoX5ESuHzHvg6diUjA6Kkmorh8hye8rz4kitZp8tm6vWzJ6DBrWctnEjcSRf4Zp750HzgSq6x0aL/fVVVnW/lappWSeyGKjvgVSBqCgIsi4AyiSOe8/DjROa+9g458KZOW49lnn0XevHmTjcodJZgHfb9l5EGLC/NdtmxZomzPnTunwrlC2ZGHEYlipz5DFVZTFq2x2cmTKH7w8adKhkSRnRmtNbSDnbuZuOOmpYjytCTSuTQtQGyfn/IVNBNF3VnSynYhCiDRoJVpwuwlSnb+uj1mokhZWjBp1WSaJLbctpBl0OnwyvIochkcrepAOW7jlpRVifFY/vbd+6t8/56/Wlk9ac3kEOLWY/4qvFSFargY87CeKenMlcyDIe2Zaz0xbdD/EHewCEJ3lsbdi78DMZ5A9B4gZicQuwm4vwq4vxheh9th8tgSaN6kIHr07IOOnfoo9d20aRuCgy87RJWdQRQ79BigsJq6aJ1NHSJRfOOtt5UMiaJJh+Jww4YOsc3MOnQd4EQS6lD+YqXMRJFtxykRZh2KhJo6MObf+Up2hQVRPBYUpSyYiXXI9EfEpg5djFLElWVgHOq0lrNue4ZTd1t16K7ynbZ0Iy7fhbKIs56bj5xR4VXqNFJ6bi8d63STehai6JCfgSQiCLgmAq5CFDmfkMQuMDAwEVAhISHo3r07Spcubd59hNY6W0SRft5q1KiBunXr4sSJE4og2iKKdBQ8dOhQlC9fXjkS5mpP62P9+vXK0TDDaeVs164dZs6cibt371qLPvLcuXNnvPjiizbnWz799NN46623HomTngBDE8WFJqLIzkyf7LAUUfzoIVFkB3044DqmzF+lLIucQ0Z5kq4dJwLRqHV75CtSEt36DcfVB8OGlkSRVjySw3EzFqNE+SooU7kmlm3zxJw1W1SnqomiTpNzztr93h/5i5ZCjfpNsXjzAUUwWQ7KkEjSgjR/3W5lwZy6aD1qN2mNzn2HYtMR/wdk9tE5a4xLIlC/+W944smnlJ9DnSbrrYc8OWR6ycKSlVQnbvlOpX8TaNt7NLznPIsLa4vg6o7SuHuhAxC1B7i1AYhcB0RtAKLWm653dgIJGwGsQNSNPti+vjrats6BGtVyYfCgAfA86pMe9VVxnUkU/1lIomiak8f6q/NilCLeb7z5kCieCL6NIwHXMXneikQ6RIK/9dh5NGjRVrU3dYizqxVRLFrSTBSpQ9TLMdMXo3i5ykqH1uzxwvSl65SsJVGk3M6TgWjb7Q+ll7UatcKSLQeVDur2phPrZdsOKx26FAdMmrsSdZq2Qec+Q1X5LP8QWbcx5yHWatgCj2XLpv7M8D3rzW0b9R+ljz77wvzOMn5a7oUopvsnIAkIAq6LgDOJ4tatW+1W/Ntvv1UfTy3AuU788PbpY7JWMHzjxo0qjOFanvP8OPScP39+HVVdSfooRxLJvWh5r+cIWloU160zfbT5nrtP6CHiZs2aJUqP72n50++feuopleaTTz6ZZr+Teh4m92Z25GFkojhv3Q5ciTdZ7mi940mryDUAH39mskrTokgLy4YDpvllHXsPAYdWab2bOGeFWYc++yKnuv/wE5N+5MpfyGRRfEAacuUrot6/8OJLeP+jT9R9weJl1VUTRVpqRv4zT4VRRzhP8tnn/queazVsqSw57GgDo4Cf8hXC8y+8qMgFZZ96+hlzvBGTZ6vymTrvxISRYSSLPEk+dCdPqxYX4DAtDnmmlSgGxQKFS9WG/4I3cWF1XoRsK4FY//ZA5AYk3FgG3FwB3FwK3FyMhFtLgajFpjN6CRC7EkhYD4CjBYtw7Up3tP/VtoPu1Oi4M4ki2+4RHYoGbgF4+3/vKTypU8R3wyFv9dyxzxBzG4yfuczcbp8+0KEvvv5OhRV4YFFUbX4b+Or7XCr8hRdfxvsP/sjwDwrbTBNFErw/Jz6cRvB5zm/wdPbsSqZes1/VHw6mF3Qb+PanPHj5ldeQp2Ax9f7Jp55WV6Y3dvoimzqkdYr5cC4riedDHYrD3tNXVBrFy1ZWZFnLUyatpxDF1Gi7yAoCboaAM4lihQoV0L59e/z666+JTlrcNAHTcHFVMD9+nNOnDz6/8MIL4MIQfRQpYurMCxUqpIPUKkzK9uzZ0xx27Ngx8wdVE0XOg6QcyablSui+ffuqcMs5lZTjyZ0o9ME5WQxr2rSpDkrVlcST8XV5UhU5CWEjE8Wipcsr61rtxq2gzzpNWisL4XPPv6Dw1ERRD6n1GDRaDQd7Bt5U7znn7GSwyZJy9noCvsuVV4XnLlgEV+5DWYG69jMtnvm1a28Ex0KRsOXbjig5thnJBsknV2Lz+YOPP1fzuzjfkRaaJr92VuF/z1uthh5JFGmZpGyOr77D0cAbqlPedPi0CnviiSfNpNJW58xV39bhV+OBvIWKq/izVm7D6TRsmUhSQGtW3wFtcHb2SwhY/jMuby2CuLNtgWvLgGtzgWvzgOtzgGuzkXBjNnBzFnBjJnBrNhIiZyMhahYQPRu4PUtZGUcN/BrHjvsloaHJv3ImUSxausIjOkTrbuNfOpoJmiaKWzz9FL49hoxW7XP0gkmHnv3v8zgVEqN0g9MXvvj6eyVXsHgZpSvEtGOvQSqsfY8BSo7DvQs27FVh1AMSxTMRwI6TQSrs0xxfqVXJ1CHm37BVexU+ayW3w7ynwgqXMv0B/uLrH+AVclv9UdpwwLRo68knn1LTD6z1hM9sZw5vW7/jkLle3DJ79Q41JC9EMXn9TEoibROokkpR3gkCLoaAM4kiP47JnRoOa6K4fPlyFZdDwJYHSSPTtLQocssxWhmtDxJSympiRv9wfKZrHuuD8x6/+OILczDl3nzzTfOzvmE+3333nX5M8ZVOjJlmrVq1UhwnpYJGJIocXiNeKTltEUUuMPhzwkwVf9bK7co6xE7TN/Qe9vmZrNfKohhvmo/46utvgR0v5zhyCJKyHOat1biVSmPeuj1qnlfdpr+o50PnIhRBoNWSQ5O0cJK45sj5rSKEiiiWM/0x2HYsAFzUwGE/lrVK7YYqDS6W0dYe6w5dP6tOPDga3FW5VUfTnLPy1eqosqSlg2cc/0hgYL/G8Jv2X/gv+QEhGwsgIbgXED4DCJsChP2NhPApQMQUJET8DVz/G8ACIHY6Em7+DdyajISoKYiP+luF+RyrifUbdqdUXW3KOZMopkSHbBFFtlXfYeNVWy3auN+sQ5yPSvc5TLdg8dJq6Jl/FkgmX3r5FWW9pIWNbcg0ajdqqWRJFKmX1es3U8+HA66pPxnUIVqqQwE8+dRT+PzLb1Q8lqlwSRNRPHg2/ME8y2jwD0PpSjVUGpYWQ60z1le2OfWMJLFpG9M3sVq9prgYa5qXaS2flmexKNpUawkUBIyBgDOJIucc8qBfRX3q3UO+/vpr9aHTKFoTxfr166v3dOxrfbzyyitm9zhRUVFKjkPP1gcJIj/mmih+/73JCsBV2YMHD8agQYPUOWbMGLzzzjtKVvt2ZLwSJUpYJ6nI41dfffVIeFIBzZs3V2lXq1YtKbE0vzMiUdSLWaYuWq0seVzhq0+uTuUQ9EeffqFwZWfMoWezRXHwaEXydGfK4VrTCmGThSX4LvD44//Bj3nzKwJ2yN80nFugmKnT1wSMpHLmCtMcxXlr94Ca+P6HpiHpHoNGoEOvwejQa5A6ew0ZhdcfzHcj0bx4GyhWpqIqn29onCKfTDcwBmjSpp0K9wu7byaltjpnynOuJS1TlWo2UHFKVaiuymFLPqVhJ64A1Yu9jIA5L+HMwu9waX1+4FJ34Op4JFwdhYSro5EQ+hcSwkcB18cgIXIKvvnqDZzd3w64PREJ10cBN0YDN8cA0RMR4FcXc+YkXriVWmV2JlGcunjDIzrEfclJlN5+J/HQs6VFkeS/WGlTG566EptoGgDf8RvBKQC0Sh8LNBHHMpVqqHTZdjy5sGXK/NVKdvmOI+qPCPXksceyoVv/P9XqfrMODR2DV197U8ly2JiWRlosmQ/1nX80mCZJae0mpj8w/tceDi3ban+lQyGxKk7ZKrVVWuWr1U23DlnnJUQxtRov8oKAGyHgTKK4evVqu0jkzGmaL6YFrIkiF6XwA2nLlQznFubKlUtFpWNvynEBi/Xh729aIaqJ4ttvmyauc54hXdXoM1u2bMiePTtefvllXL9u8h/HNG0Rxddffx0kuSk99FB5WoerU5KPoYni4nXKGse5eg/PGDXp/4OPP1Ntb4so0gJXrJSpk/e8cEM51mbnxuE4drT/e+9DfJc7r+owD5y5rNIpU7kGOH+PnStl6cpk9R4v9W7e2t1qNexzz7+onv/v/54w6Y+HBx7j+dhjePqZZ9R8RS6IIAnRRNEnNDYRUWzY6jeVhl+4faLIMjB/Dml+9d2PSr5mw5bKimnPCbl1523rmZal/QHAvJGf4PSUZ+E37xtcXJsLCYHtkXBpGBIuDQJCBiEhZLA6ETkc48ZWVfl3aPIxcG8sEDHYdN4YCkQOw43LbVClSo2UqKpdGWcSxX8Xb1CE7aH+mHSJUwzetFj1zDmKlkSRq585B5HfAr+Ih9Zftg3J4UuvvIo8hYqpe8/zpqkztNTREkgZnhxCXrTpgEpj+fYjiATw1JOmeYa2dOiZ7NnBcPrw5Kp2TRRpcdRE8WIcULNRC5VmUhZF5s/V8r5h8cj5zQ9Kvk6TX5TF05ZupCdMiKJd1ZYXgoD7I+BMopga9zjWRLFFC9OHkCuUrY93333X7EcxJiZGfQDLli1rLaacW/Mjr4kih5b5nJIjvUSRq7g//PBDlZ/lvMuU5J1aGSMTxZS6x7G2KHJls7bCcfcMS4sih//++98X8EOefMqi6Blg6uSVRfHuQ6LITnbxpv2qDWlR5NDdm2//D1xQwOE/zlkkgVDntQRFGBjO7QE59JxWosgOnjtq0Or12hsmCxNXS5PgkujZmr+Y0k6eZGmL922M+e0pnBz/FHznfokLq74HApoj4eIfSAjqAVzshYTgHupEWF/47jcNt8+YWBuI7g+E9gYiegPhfZFwrT9wvzUGDuqQWrVNJO9MophS9zjWRJFWQ/554LeA+kXsiDPbh9blbNkeR97CxXHlHtvFNLJRulJ11U6U4cnpBbMeWKVJFEk+X3zpZbzw0stKzlKH6CSeBJS6y7y4mCWtRJF5c9U068QFVaxDt/7DwdXTKdWV1MgJUUykzvIgCBgLAVcliosWLVIft6VLlyYCnASMHz3LOYrvv/++ciybSBBQw8uU1USxSZMmKu7ly4/6fiPR5PxBPTTOeOmxKGqLaVJWVevypvVZiKLV0POg0cryNmjMVNXeU+avUfPLTJ2nqcNn++o5ihzme+XV15UsCYAiY0GRCLkL/NK5lwonUeTK2RLlTAtUNh85q4giZdmp0zpVqERZlK5cQ5EK+uhLK1FkmvTlqFfN/rNovSK0tDJxOJyndxp3ifG5egcTFh5En+oe8Br/NLxnfYoLK75H/Nl6SAjojISAjsCFrkgI6oaEoO5ICP4diOqLkf0rID5sABDSGQlXuiAhtCsSwroj/kpb4H5pdGhfL63qq+K5IlGkpbrnYJNP1BnLN5t1iAT+TES80osCD+YocpiYi6boJJ1TBbQOkZi1bG+aW8o5itShQg9W0tNBPMmhSYdilCUyf9HSKFmhqlrNnB6iSOsjrdqvvfmWKifdM3EOzaPPrAAAIABJREFUZGIdeuj8OzWk0JasEMV0qb9EFgRcGwFXIYqci8jOm/4N9fHEE0+oMMvFJ9w2jHKWq57p2Jphbdu21VFBMsgwnpoonj17Vj2/9tprZjneTJ1qIhSWC00Yzx5RTG6O4ogRI1Q+tIoeP34cW7ZswebNm9VJlz/bt29PlH96HwxJFHs/cLj9wI+idefEIdkPLPwoKovi4QcrVgeNBrcB5O4s//dAh7jLBjt+WvzyFymh2keveuZigmGTZ6uwRq07KMshrTo7T11UYdQFLmahuxoSRD6/9c67yhcirZO0NPYaYiIUlWs3VItbaFEs+mB+m09ojO2hZxtzFElm6QOvTVfTSvwaDZpj/f6jmLl8o/mcsWwD5q3dnqTjbmu89DOHs6euPIzVfTxwYOTj8Jr2IQKWfYl43+rA+XaIP1Mf8C2KhOPfA965AO+fAf/82LuyFBBYBgiohIQLVYDA8ki4WAbxgYWB2DKoUf7VdKmxc4iiyWn5VOVH8eFOPRoL5XDbwo8irW+bLVY9c5ES5xiyvTm14Mx1KOshxzh+ylNAhXMxC90UUYf+nGRaPNWifTdlceRiqC2eJif7TINEkSumuUMKn9/78GOlK1qHuvYbpsKr12uqplZwCNuuRbGh/aFn6hDnNDZv102lRy8B6/ZRhzZZ6NB6LFi/G15X4hxiYRSimC71l8iCgGsj4AyiyC2++CGkVdDeoRePaAue9qNIVzX68PEx+cVjWtpC9+qrr6oVztbzBPVQNcnlZ5+Z5q6R6DGupT9HvZqa4dwdhnMTef/jjz/qbNWVYblz504UxgfOb6QFM6mDcZM7k4qf2neGI4qRwK8PiNLEOcserPZM7OpDWQFfM1kBL2s/igdN+sKFMKYhvQQs2XLI3BYffWbaNpIrnP/zf/+HL775ThFHdqxMo2KNekqWc8TefbAHMB1qsy1nr9ph2tkimqupZ5jT/PDTHGqlKmW++ymPIo+0EJEo5ilYVMl5X7mdiCjWaNBUhdMySFlNXHjlM61+yelPtsceOlK2jJ/UvcmiehfzNx3FlgEe2DPsMZz4512cX/IlErxLI2H/e0jY+ywS9r8CHH4H8Z7vASfeB05/ijkjC+C+X0Ek+H2FBL/vkeCfGzifHwlBhYHQYpg+Pl9q1TaRvDOIYusOPRSOk+astKlD3Okme/ZnlQyJFYnixgd+FDv/MVT5IKTFb8nmg+b2+OhTkw5xCoJq89x5zQ63L90FSj1YjfzEk08qIkiZnwuYXHrRoTaJOq3Fg8b8+zDNz74w/6H5IXe+B34Uo5WF8cc8BZVcwK3EcxTLV6+rwrkftS0d8g6JM6dvT5eefvoZ9ecnKZ1J6TshionUWR4EAWMh4AyiSGseVzzbWrGs0Tty5IiytOln+jXkzipBQUE6SF1pTaRrmXLlyqF3794qjE6zeVof+/btQ7169dTuLHxPEsrtBK3nOXI+JNPkKmSurrb0n6jTpAWQvhitj927d4NlT+ogMd25c6eyZNKaaXkyfO/evUlFT/U7oxFFrmredPgMxs5Ygj3el8xzwyw7Le58Mmf1TrVbBd2VsLPkIpJxM5coq5/emYWWHs8LN9G6Uy/QJ2PrTj2VpXHMv4swae4KNb+QBIppcw7gtKWbULFmPZSpVB2T5q5SHfvvA0Zi16lgVQ7KcmHBLq+L+LXrHyhZoQoq1aiH8bOWqQ6e8xNNhCwOS7cdxt/z1+Dk5Ydz2zj3cNXuExg3c2mieZOWdeOCG+4m8u+SjeD2a9YnV/HOXpV6iyLLRYtWm65DsXuAB3YP9cCxyW/i/MIvgd1PAbuzI2HvS0g48DoSDr0FHPkf4o++D5z8EGH78iL+1I+Az9dI8PsR8MsNnM2jyCLCCqNvp/dw9ETa9312NFGkfmw85Kdw3u1z2aYOkRjOWrEN/yxYq0ii1iHqHS2BnNeqMTt87hpatv8dRctUUNMRaPUdNmkWpixYo/SJcmxDDvdOW7JR/ekoXbEaOORLEsW9lvf6hiTSoe0nLqg/RCUqVFEuk+gcniSSUxmYHsvHPzqT561Wc1aZvgq/egcrdhxVOsdhcEvd0fdHg24lo0PrMXftLrWDkI6TnqsQxVR/tiWCIOA+CDiDKLpP7Y1RUiMRRXZW7AxJ9OhQWi8gsNWJmVZ0mvZY5nt29JyDpTt4nRbJG4cR2YnzyjQ5FG3LYTXjc24Y55xxIQLjkmyevGQigDpNklkOfzNNyjM9/U5fSQppNeSzPlk3khjmo8NsXdVcRM5JtHOy7rbiJRXGvDl82n9QF5yf4oGdgzxwdNKr8J//GbDzSSTszI6E3c9bkMW3kXDkbSR4vm+yLHp9CnjlRILP14j3+16RxQT/PMClgpgxLgc2bfFM8w/KkUSRGJh0KFbhl6wOWbTRQx1K3N7UR61DbHe6QOKwNNuJeVniTr0y6xB12KxDJgKoy0c90GlymJn6pt490BemS7JoX4cS65ZlGXjvDB2yzkM/C1FMs+pLREHA9REQouj6bZRcCV2DKJ5Sq3R1x+GIq3UHbJ0m31vL2ApjPB2u5fWzvTT1e8urpaxluL63fG+ZZ0rDLeV0mkldLeVTcs+0/G4AI0d2wKmxHtg52AOHx7+Cc/M+RPzWJ4C9TwF7swP7ngf2vwwcfA049DZw5F3EH/0ACcc/RvyJT5Hg9SUSfL5BwunvgDMcgs6F4JNFsGXrieRU1e57RxNFjQfrrO9tX23rkLUsrbyWbcH3+vlRWVOa+r3l1VLWMlzfW75PKg/9zlre8lmnmdTVUj499yaiGI4SxYrYbWOjvkiZHw2j1l7qlSUQIFHk/EA53BcBth+3B9TzPTO6JtWrVcH6/d4OJ4rp6bgk7kMrpiUWvhFAsTzvKqK4a4gHDo19Gl7jH8OJsR7w++dx+E/PhvAV2XFt9VO4u/tl3Nv3KuD5NnD8A9N54kM1FA2vT5HgnRMJvl8DZ77Bdd+GaN3m4XaXqdVBZxFFy7rLvW2dcAQuJIqH/SOEKKZW8UVeEHAHBLiKOCQkxB2KKmW0gwCJItsx04hi1crYcMDHRBSTteA4r7NyRIdn7DSi4XUVaFblffhO8MCuoR7Yy3OYBw6M8MChvzxwdLQHjo7xwPHx2XByQjZ4Tc6Gs7OegP+c7Ajd+DGu7fkBsd4VEedVEPFniwMBFYCzPwC36uDPEYPsaGjywU0bNxQdspii4G56yOH9Q/4RKF60cPKNbTAJsSgarEGlOo8i0KZNG9BlixzuiwAXy7Ru3TrTKtC9W1dMmr1ErTTmMJe7dXJZp7zROBIMjOvyGnzHeODIUA+cGuehSOOxvzywd4gH9g31wKFRHjg4wgMHR3rg4CgPHBmbDUfHP45j4x7HiUnZ4PX3f+Dz75M4t+RdBK7+CqG7i+H+mUJo2aJBmnWwd6+emDBjkZr3JzrkXn+m2F5cULZowx40alA/zTrgrhGFKLpry0m5U4wAV+g6aw/iFBdCBNOFQKNGjbB48eJ0pZGeyDt37kDFylXVVmFcDCAdvet29OzQ+/81BUWKFEaThmUx8o8KaF/rM3So/Dgmd/LAtPYe+KOMB1b3NC12OTnOAwcHm8ik72QPBEz3wLFRHjgx7jGcmpwNvlMex62Nz6BC0VdwNSw6zWq0a+dOVKhYCafDTQ6oRYdcV4cs/1jpduICnwaNm2H+vHlp1gF3jShE0V1bTsqdKgRINP76669UxRFh10Bg0qRJyh1QZpfml9Yt0bP/n7hwG/AKiXvg240+AuV0JQxI5P2uAQHRwLkogJNORkxZjBq1m+BuAnAfwKZtnli8fANWb9iOJs1/Q6VKVTB8YEMM7FYORb/zwJiuL2NMWw8Ma+yBYXU90Lh+MUTHpl8D27X9FV17D1D7bYsOucHvJjhaeRjgb37EhKmoWrlS+pXADVMQouiGjSZFTj0Cd+7cARe1NGvWDIGBgalPQGJkOAJc6dyyZUtUrFgRUVFRGZ6/dYb0d1m3dk3UrtcA246eUzs+cLcUdvjufLIO1qc714dlN9XnDnZ5BaFx85aoUqkCbtzgLsQpP+IBLF29BwtX7Ex5pGQk6Uu1bp1aqFW3PjYf9lN6Izrkmr8ftgudh+88GYhmrdugYvmyqdahZNTBbV4LUXSbppKCOgKBqVOnqkURJB9VqlQxxNmwYUPwbNCggSLDRqgX26dChQoYM2aMI5rdoWnMmD4d1apUQsUKFVGpUmW3PytXroIGDag/prNKlapuXye2S8WKlVC5YnmMHzfWoe3viMSmT5uG6lUrG0aHqDP16tU3nA5VUjpUAePGup4OOUIPU5qGEMWUIiVyhkKA1qpLly659cndYbgjS/ny5VG9enU1PHv+/Hm1wtvd6+YO7oxCr15V+nP58iW46xkS8kCHypVDzRo1lPX2QkAArlwJcds6sS2o//yNu/oRFhZqAB0KQXDwRdSrVxc1qldXO0wRf3qacNffhdYh8ZZh+gUJUXT1L4mUTxBIBgFu/VezZk3UqlUL9+7dS0ZaXgsCjyJQtmxZpT+05HJ4VA5BILUIcESD3yHqkhzGQkCIorHaU2qTBRGwJIqxsQ6YcZ8FMczqVbYkipyLKYcgkBoE4uPj1Z7yQhRTg5r7yApRdJ+2kpIKAjYREKJoExYJTAUCQhRTAZaIPoKAEMVHIDFUgBBFQzWnVCYrIiBEMSu2umPrLETRsXhmtdSEKBq7xYUoGrt9pXZZAAEhilmgkZ1cRSGKTgbY4MkLUTR2AwtRNHb7Su2yAAJCFLNAIzu5ikIUnQywwZMXomjsBhaiaOz2ldplAQSEKGaBRnZyFYUoOhlggycvRNHYDSxE0djtK7XLAggIUcwCjezkKgpRdDLABk9eiKKxG1iIorHbV2qXBRAQopgFGtnJVRSi6GSADZ68EEVjN7AQRWO3r9QuCyAgRDELNLKTqyhE0ckAGzx5IYrGbmAhisZuX6ldFkBAiGIWaGQnV1GIopMBNnjyQhSN3cBCFI3dvlK7LICAEMUs0MhOrqIQRScDbPDkhSgau4GFKBq7faV2WQCBrEwU2UGdP39e9idOp55nRaKYkJCgdIc6JEf6EMhqRJH1DQgIwL1799IHnJvEFqLoJg0lxRQE7CFQvnx51KxZC3Xq1LEnYrjwU6dOYfLkyRg+fDgKFiyI9u3bo1mzZhg2bBiioqIMV19nV4h/NmrVro1KlSo5O6tMT9/b2xvjx4/HX3/9hQIFCqBDhw5o3rw5hg4dipiYmEwvn7sWoGHDhjD6Xs8nTpxQ3xzqD7877dq1A+vN71BkZKS7Nl2y5XYronjr1i3Mnj1b/ajr16+PBg0ayOliGNSrVw8tWrTArFmzcO3atWQVMKMFoqOiMH/+PPzSqiUaN6yPpo0auu/ZuCGaN22MsqVLonzZ0ihXtjSaNm5oOt24XmyXNq1bYdbMGbh58+YjKrJ27Vps27YNd+7cwdWrV0HSuHfvXmzatAnjxo1D0aJFlf49EtFBAdHR0Zg3dw5+/aWVMXSoyUMdoi4ZQYeaNKyPX1q3xIzp03Dd4ju0Zs0abNy4EbGxsUp3SBqpO5s3b8bYsWNV5z99+nQHaYr9ZEhIF8yfh7a/tjGIDjVS3yB+h6hDzZo0cvvvUJOGDdCqRXP1HYqJuY2dO3eqb0xcXJxZd/bt24etW7diwoQJKFGiBKZMmWK/0d34jdsQxaVLl+Lnn39Gnz59sGvXLhw6dAgHDx6U08UwYLvs3r0bvXv3Rp48efDPP/+4zM9j9epVyJc3D9p1/h1zVm7FgnW7sGDtTvc+1+3Ekk37Hpx7DVKnXZi5fBO69OyHPLl/wr//TjXrEP8s8t88jxs3boD/8Pfs2aM+1itXrsTChQuxaNEi8I9k06ZNzfEcdUOSmj/vz2jfuQdmLd9kELwtdWifQeq0S/3Gu/T6A3lz58K06dOUvuiOnH9ATp48qXSHfzpWrFih9Ia6Q8t8q1atHKUyj6SzadNG/Jw7F9p2+h0zl27AgvVG+A7twuKNe83fImN8W3dh7qqt6Na7P/Lly4cxY0artqTl0MvLy6w7/O5QbxYvXqx0p2PHjo+0ubsHuAVRnDFjBkqXLo0rV664O95ZqvzsyKtVq4aRI0dmer3nzp2DEiVKYvux8wiIBnxC78H76h14X70rp8thcEe1z/lI4ND5a6hSrQZGjfpL6RA78IiICISGhqqOfsuWLRgwYIAaeiYJWLBgAf7991+sX79eEcVly5Y5TPcWLJivdGiL59kHOiS64+q/n/NRwCH/cDRq0gxffvEFLl++jLCwMKU7JIgDBw5Uw4f8Qzt//nxMmzYNGzZsUKNWtDw6+li+bBmKFCmCrUf9Lb5DokeurEch94HWHXvi8qVg9d3hn9Pt27djyJAhSncmTZpk1p1169apb9HcuXMdrTqZmp7LE0X+sHPnzm3o8f9M1QAnZ84J47QsXrhwwck52U+eHUOB/Hmx2+siTofH41hQpJxugsGpK3dw/GIkSpYogaNHj6Jt27a4dOmS+kdPkpgjRw78+eefGDVqlJqfWLx4cfTr109ZGDmsyGdHLFYICQlBsaJFsOP4BdEhN9Ed/Tvnn8Jd3sGoWaO66uhpDSIZ/Oijj/DHH39g9OjRan5ihQoV0L9/fzUMfe7cOeTKlcv+RyUNbzhNoliRwth+/LzokJvo0IngaHiFxKF58xa4ciVE/cGgBfH7779XRFF/d7gYrFevXmpKjJ+fH6hLRjpcnigOGjRITVo3EuhZrS609PCDnFnHiOHD0GfwSJyLhCKIxy9GQU73wICd/ZlrwPhp85A7149qfpmvry+OHTumOnrOT7Q+Wrdujb///htnz55V3w5+2NN7TBg3Fj0HDIP/LdEhd/rtUH98wu5h0rzVmPrPFJw+fVp19iSJnLpkfXTr1k1ZpX18fJSVaPXq1dYiaX6eNHECuv8xBP43RYfcRYd8rt7FhDnL8eeQQSAB5PeGhg9PT89H9ED/6Thz5gzmzZunrNOPCLlpgMsTxerVq4PAy+G+CNCaWLVq1UyrQM3qVbFu3ymcuhInRNHNSDI7+lMhcdh9MhAv/Pc5NZeMH2t24F999ZVdneIqVi4YIBlo0qSJXbmUvqhftxaWbz2kpiuwTO7S0WX1crKtTkfEY9DYKdiyaQOOHz+urD6cc2bv4IK88PBwRSo5TOyoo0Wzpli65SBOXb0j3yE3+Q7RGj3633lYt3qV+oPBRU8vvPCCXZXg1BiOPvCsUaOGXTl3e+HyRLFKlSpqqMndgJXyPkSACxAqVqz4MCCD72rWqIpdp4Jx8lKMdPBu8oG2JDgnL8dgn+9l/PDdN+qfPMkfF7dxLqK9g4taOMTIYWq9+MWebErCq1etjA0HfOEVEis65EY6pIhi+H2Mm74Ahw8ewOHDh5XFkKvj7R1LliyBv7+/8pNHC6OjjmZNGmPzYT94XRYdsvx9u+o9dcfn6j1MnLsMh/bvw8GDh9Sq52+//dauSnAVNL87nEPdsmVLu3Lu9sItiOLFixfdDVcprwUCXNSSmf7Zalavhp1egSDhcNWPkpTL/lA42223dzAKFcgPDgnSTQWJ4syZMy20LPEt56BxIcv169eV7tGVTnqOGtWqYP2+U0IU3Ygk8jfFzp5DvXWa/IK9e3arlaqcCpMUAaRXDS5s4dxmusxx1EFXVpsO+gpRdBMdou5wQV295r9i984dSnf4B/TVV1+1qxK0WHPaC/s8+lY0yiFE0Sgt6cL1MCJR5EdEn/ZInvV7/cyrvTg6XMvqZ3tXLZeSNO2l4ahwXRZHpafTIVHc6x2Mzz75GEePHgf/tXPuclKr6TlExBXQ9HlIp8q8pucwEUVvhxJFjVdSbadlNBb6Oak41rL62d41NWnaS8NR4bosjkqPC6HO3wIKFC2Jw4cOK5+J9HnXvXt3u+pAv4pcBU0XOhyGvn37tl3Z1LxwBlHUeCWlD1pGY6qfk4pjLauf7V1Tk6a9NBwVrsuS3vSYzsU4IF+Rkjh08BD27t2n/Di/+eabdpudRHHEiBGgr8W6desaxvm/EEW7TS4vHIWA0YgiPyCc7+h/A8o6wGdbHyWusD57DeodZU5euq06LU6QthdHp3P2OtQijqTk+I7z99gRcjgrKVmdrrOuzNsvPAG+ofccWg6my7od8r+GV19/D8P+7Ic//xyFXr0GYNhw+26X6CNv4sSJ6oPNHRRcjSiyXnTPxHZOqu04v44ylOf5UIeSnud2LCgK524AZyKSXjjBNKnLJh2KcWjbpVbXWBa/iATlGon3qY1vLa/TuHIfyFukMvr07opBA4dh0MDh6NzF/pAyF0txaJpOuelEmcOIjjgcTRRZv+R0SGPKBWHEh88p16FInLsJ1SYaS2uMdZreV+64jA6duUYdSv4ba6su1mHcMuK7PKXRo0d7DB8+Sn173nvvA7vqQKLYt29ftaUo5+UHBwfblXWnF1mSKHL+CV0icK9GewdXSk6dOtW8hyxdbHTu3Fk5k7YXR4dzyTx3Jknu4NAYHYi7wrF8+XLQB5QzDiMRRX4w2bGu2nUcXfr+iS1H/OFlY0ibfsHGTFuEvsMm4FRIrPo4K39urTtg3tqdynejvY8vP3Idew/GgL/+NpME6w8Ynzlfbt0+H9Rv3hZbj52zWQ5b8RwdxnpwiOavKfPwz4I1DiOLGmu/G0CnPiPgs7cjENMFuLsJUTe2we/0ERw/7oWFC1fB70xi90u0BtHPGYecO3XqpIaC0qPbjrQosl7s4Jdv80SXvsOw9eh5m23nc/UORk1dgJ6DR8MvIl7p0H6/q2j8S0cs3LAP7JzttSX/pLT9vT8GjZmarA6t3eeFBi2oQwE2y2EvD0eGE5OASGDYpFmYtnRTkr+PlOSrdCckDqfCgLY9h+D0/i7A7a7AvS24HroBZ896wuvUaSxZuhZ+ZwITqQZ1h9v7cTEUV7I6qrN3JFHUOrR06xF06zfcrg7xj9vwyXPwx4hJioCTJO7zu4JGrdpj4cb9dhdnMX0O2bds1021CQk8w6yxVzhficPKncfQoMVv2HkyKNPmgrMsF6KBwWP/xcwVW1XdrMub0md+W49eSkDbniNxcH0zIKojEL8NEaEbcD7gOI6f8Mb8BSse+e7QJRddePGoXLmy8tuZSLnc9CFLEkVaGjw8PDBnzhy7zfb2228rGb1vLD8ajMN9QZM7nnzySeXfLTm5MmXKqDSTk3P2+6CgIFWONm3aOCUroxFFfkA79xmiMBs3ayFOh91/5AN6/ibw2htvKpnT4XdVp7716Dn13Lpjd2WNtPXhPX4xWn2gqWuf5MgJWkNsy0XhdPh9jJ2+SKU5a8UW1RGk9EPoCDmWi2dAFLByxzFVjsZt2qlne2VOTb78WC/b6YdOvxRA7PGKuLGvDO5d6QfEnQJidwOxW4DY5cD9Tdi9tTem/d0Mw//sjgkT/sb5gCsYONA0R4grpKmD6TkcTRRJrH/t0kdhNnHOMvja0KFzt4BXXn1dyYTcg/qDsm6fl3r+7fd+igDawzPwNpRczu9+wNX4pHQoHmOmzVeys1ftUJYYe2k6I1zrEDv4hev3qnK0bN9dWbLSo0P887bhSBB+bVYIkYfLmHQnuDdwxxuI3QvEbgZilgNxG7B/Vz9MmdAIo0b2xMiRY3DmbDCGDR+j1GX79h3g99ERh6OJ4oUooFWH7gqzSXNW2tShwGgge/ZnlQx1gn8u1u0/qZ5/697P7neI2IfDpEM/5MmPy3ds6xDlaK0cMXmmSnPp1oPpImhp0TGtJxduA3PX7FDl+KVTT1W3tKTHRY+bj1/GwD61cP1QGdzcXw4x57sBd3yBuN1AHL87y4C7W7BvR19MHt8A48f1xdixk7B8xUY0bNRCqcugwYMRFJT4T4gj9Cgz0siSRJF7ebIj5pY79o4vvvhCyej5KXQcPXnyZPAfQ3LHK6+8gp9++ik5MbVrCcuRmQe33GMZeHbp0sUpRTEiUew5ZLTCbPL85baJ4i3gw09zKBm/iHtqIc3h89fxx/CJyhrJIWP9gVPX4GicuHQbdPBKIvr0M9nx9Q+5ExFFJRcUhRPBJjkS1IlzVqg8+IGkKwf9YdRp6zT1s37PK8P0yWfmrcMt5WzdMx4/qJfuAPMfdPDUoV86d3cYUeQQ/ajZW3B1/Zu4sKYQQneWwr1LPYGovUiI3AxEbQKitwDRm4G72wBsBe7Owjm/1lg4pxxqVH4XQ4e0Q5OmLbBx0/506bbDieItWkmHqrabsmiNzU6eRPGDjz9VMiSKbPeDZ8PRb+QkrN5zUlmqdduwPdh+ur0DY0yd/E/5CuKqxZ8NyjGOWYfC4zFh9hKVx/x1exIRRbPsA73UzzpPXhlmeaZahy7H4PI9YNqSjaoM1KF2/8/eVYBHdXTRxVoqFCrUfihSaEtLaQsUt+LuJLjT4l6slOLu7lbcvUhwJ0FDQiBCDAvumj3/d+7b2WyWTUiAZIW33/e+t2/evJl5M2funLlz574e/V6JKLI8JN4jZq1DwMo0CNlA7JTFk+AOwL0Dz2PnsQk7kfMQ5N8OKxZVRK2qX6PvPy3Qtm1HLF+55ZWwox5+3USRE7SOPftLvc1YutEmhkLuAZ9/qSk9SBSpUTzkH4F/RkzCesFQdDmkMMS2jTARxUIlykQjigoHCkPUNo6duVDKsXqnVzSiaI6bwBiiHJq2eL0ZQx3/GvhSRJHl5U7nqct34eT8TxC6oRCu7KqEe4EdgPsHBTtGyp2724D7WwHBznYgci4C/Npi6bzSqOuWFb3/aoxBg4ZhxYrXgx2FIXud7ctS4vDWdI/zunc9vwxRfFFRabyqfrERRct49BFJwRjT71V3asaULsOZduPGjSV/+qPTiWLMu26tB0YSufgSRQpgLlnfBZ4jUtSmnLtJIX5Xlk5ICqyJIgXY6UuPxLiaS9k+Vx6LIJ+yaJ20nTVRZJrUWp0MvydEloKUxE4Jbp5NOmpUAAAgAElEQVSpraQNEjUCIQ8gX4sIfaiVRcWzfHf1n/eYlmfgdVSuVU/yz5z1Ozm37Px6iCLzIFFq1Lo/wld9jPPri+DijjJ4HNITxttbYby1Bri9FuD51koY76wC7q4C7q0BHtGEgp9fW4XIB2Nw7HA7LJ5fBg3rF42tS8R6zxGIIu0OuWQtGLJyHk8SQFtGtjfbXg3ylkSRdUoMhT2CkEwSccabvGCltJ01UaTGisvBgqGrz+Q5awwxTthDTVNFjHHJm8+E3AeOmyYeCjeWZ8HQhQfwCrqB8tXdJf+MX2ukuF33vq9MFInnomXqwW/RFwjeUAgXt5fFo4COMN71gPG2JXZWmbCzWsPO401m7Dy9PxInj3bC4nml0axxqVjxEZebjkAU2QaxYui6CUN3ITjjuGBJFAVDlx9Lm1MO0cSGdnxjZy+WNlREkfGYl8ih21qaJJTE3nMYugdQ7lA+hj8CzlwzCob4rCVmrP8LhsIfwOv8LRQvW1nyz5TlWzl37DXgpYkineo3ad0ZJ+Z8hvPrCuLCjjJ4dL67kEPjzTUw3loLEEO3VgEid0zYeUTskBiugfHhaJw80h5L/i2L35sVx+MnxrhAxGHjxMxSHKTICUkU6S8rpt8PP/wggFMaxSdPnuDzzz9/zts6P9vDzqQO+nij139rjSJtHj/44ANzPLpooJ8lPmf9447OZMmSmeNy1+bTp0+jReNnDfldUn5z8sMPP5S43LYf23K6ZQKMx7zVB8z5X9covpgsUjhZEsWpS1aLsbfMxKnRCbsns/bzd4CvLTSK1CDu9bmAL9N9hWGT54rNGQfS0AdA83Z/mts6WbLkWLb1ID759PNoGkVuSqAvyKzZNFyyvYqUKIcZyzbIs5ZE8bKRaXZF0qRJzenmzl9Ilp2UkTeFcJZvs2Hw+FmYsXwTUn2QWuKm/fRzjJgyHyQeStjbEtAkl72HjJdnmrbpAhIGlqll5+7PEWHr5+NyzbypFatdpxJubs6AwNX5ccGjGB4G9gBurIbx5mIYby6F8eYiGG8ugPH2IuDOIhhvLQDuLNSOewuB+0uBJ1w52IRubdLj6LFzlt0gzv8TkihOX7bBJoaoMVIDHycPXE7dceI80mfIiNHTF4kGiXgjUaPdIuufR/IUKbBhvzeSJ08OS6LINtvjHY4MmbOa4xJDc9do2jwzUQy7J5pitqslhgr9VkrKoDDkfwNo1qYLyld1w8Gzl/HN9zkk3ZQp30Gdxi1losLy2Wpvti+J7Z/9hskzbbv1gffFe6b//7wyUWTdtWzfDP4L0yJodT5c8CiOB/5tgRvrYLy+xAo7xIsFdu4uBO4uMmFnKYCNmDTiZ1y79iDOeLEVMSGJ4sxl/z2HIWqXKV++sNAo0pxj54kgfJkuPUbPXGReDeH3jOs0aWnGRfIUybHXJ1yuCxUvbdYoEkO0qbXEUOlK1TFjmTZhVUSRmkuSvyatOyGJIYk53SIly8oGPC6BEwOcqDb8oz0qVHeXiWeGrzVsvvPOO6jbtCVCHsZMFvk8NxR27aNp5tv37I9jIVclrw49+780UTx7Exgz+R+ErvoEQavy4sLWkoi80Bu4vhrGG9bYsZI70bBDubMOy+cVxxm/12O+YAtXiRH2PEtJjFzjkUdCEsXVq1fHWJIcOTShp4giNXAUwtzRpH61a9eWMO6M8/DwQM+ePc0dwtLzP31y8dn06dPLlyW4SUYJdZ4tf2XLlpV79OrONLn7TsW1jMewDBkyyD1+sqx///5IlSqVXMdlIw038ly6dMmcJNPTiWL8ieLMZetlQPS98gTmI+KpzJy/yZZd2oNLzxSMu04Gy3WX3tqyCMlaqYrVJKzQb6UxY9km1G/RVq7ZHjnzFsQl2qaZSKbCQff+IzBp/mp88b+vzHEXbNCWnjkjz52vsISXqVRDlmMYXz3LtDijJ/FgWFqTHaVbw9/RplsfvPXW2xJOA3i129Z6oKeApjZh7Z6T2OdzQUgF3ZAwvddJFIPuA/371UXY4o/hv/xXhG8uiqeh/WC8Ng+4Ogu4OhPGazOAazOB6zOAmzOAp/8Cd2bDeHsWjHdmIPLuTBjvzQCezcOaheWxY/fzn94yd4JY/iQkUZy9YrPsGDXjh1gihh4CX3+rmS+wvahx2XL4jNRzt34jpH1I6IuWKi9hxcpUxIylm1DbYsDPW7iYLD2z3ff5XpR4bKce/Udg4vxV4MRAYUMRReaV/adcEl62ck1MX7IBf/bVCJ0hSRLT5qwHCL4LFC1ZDilSpJC4eQsVQ8e/B+P7HL/IdeES5WKccCgMrd55FAfPXhEisu+MVr623V6dKPrfAYYPawO/WR8gYMWvuLClCJ6FdIfx2nwYr86Mjp0bM4EbMWFnOvB0PjatKIV1G3bEgpAX30pIojh75VabGCIB/DKdNk4QK5RDWzx9pX26DRghZIpavnyFf5OwEuWqYMbSDXBv1EKuiQ2SO2qLuTt//5lL5vCeA0Zh0oLVSP3hR+YwEkVOJGgb+eMvv0p4+WrumLHsP1DLJ1hLYpCJMpeuqXnOX7QkUr7zrtyjK5oOvQbi2x+08ZeYZj8gXmzLoUdYtcNLTDIuPAF2nTwv6bw8UbwLvxtAx461cXZOKvgvz40Lm4sg8kIfwY7InQhLuUPsaBjBvTnPyx3jfEwfVwhz5694MUAcOEZ0luKABU1IoqgEZExnCkBFFKlRZDwSMv7oyJfXXD62/NERMMMtiSJn5Z999pllNCFpKl91g9pBhpFwWv74MXmGW3p654YZhvFTQZY/hlnmbXkvtv98TieK8SeKrLcXHYoo7vYOlbjd+g6VAXSbaXNLhep1ZPmGs3UuBQ6dqBmGC1F8pi1VV6iuTUq2eQWI0OZMmnG/yvS1pEmNIonnVJOdTuuuf4PbN5gmhfze0xckXv3mbczL26rc1DBwQOGA4RV8S+LlLfibxLMWzuqagpvaCR7UHhwNvi3PvS6iyHx8rwO1y3+G8/9+DP+lORG+sRAiwwYCVyYDEROAKxOAiHEwXh0PXBsP3JuKyWNr4v7l8cCdiTDeHAvcmgjj7YnA/Snw962DLVuOxNYNYryXEESxi8lGUbVDbGdFFLd6npV67jlwjAzGa/eckOsq7g1wy9Te3ITQb9RUrR0LFxMTA2qW1PLcbu8QaWtqBG8gasMMiSKXjacv1rTUHGx5nxjiZoYth7W8m7XtKpMjEsXSpolO76HjJS4JAnGpTBHoHio2rSLJLwkItaXUuLMOXpUoEjs+14AGVbMhYH4anFuSExf+KwRc6AVcmaLhJhp2JgBPZmL6RHfcDh0K3CZ2xllgZyqC/N2x6b89MeIjLjcSkijGhh11TxHFbV5+GoYGjxGitsLjiFzXqNcEt6W9jbgGiMcGPlukZDmRIZQjJHUM23nivIahm8AlI2QFhOEkijR5mTBXGwe7/jPMLNsuPgHW7Nbw2rJzLyG2LNNvZSpKmv1HTxWZRTnG/L9MrxFcYlLJHeuzNuF4JDbgmrY0UNJ6WaKosPNb/q8Quigtzi3+BeGbCgEhtrAzTpM7z2Zh1tQGuOo/yCR3xgE3TXLnwWQEnmmBjRu5LO28vzeaKPLj3tTccRu75UFHme+9954s/8ZEFOl1nR2Dn+ux/nHJh0vD/PFLEow3bNgw62goVqyY3FM36tevH+1ahfOcO3duJEmSxBxE8pk1a1bztfrD5e3MmTOryzifWUadKMafKP6cOx9KlquC30pXjHaUrlQD7777nrTn80RxCK4YAdpisd7X7DouGiQKPS4X0aUFw3/KnU92rNJNSoq33kbmrNlEE8h4POj7ru/IKRKXRJE2h6UqaBpKClraaVHochnuCYCs32XHe++lktk8d8Iyj++y/wTOxJkehSRn7x99nBaZv/lOiIi1YLa8VuUQm6PXTBRJLo5dBGoWS4HgeR/h7KKfELY+PyLD/obx4ggYLwwFLg6F8RKPYcCdsVg4u6680+LRBYFH44Hrw2C8MQyRN4cCd0fjWkgjtG4TfRIW1w6SkETxl1/z28RQmYrVxVaV7WSLKDKsVRfN9GWrl7/YnrFNWHckd3yOS8/EGjFE+UECR2yotuOg3mfERIm7cMNe0GyhQDGNDJAYcPmQGOJxhxjKll00SJyocKNEyfJV5FnPoBtmf5DEXcM/2kk4N1zFRBSJJVUOksW9vq+HKDK/g+eNGN0zI87NSg0/YmdDPhjDusJ4ceTz2Lk1GiuXNJPyTh1UCHgyGcZrQ6Nh50pwGzRu3CqucLEZLyGJYkwYokZYrRLYIook9Y1adpR3p0aXKwVsE3pfoLNpYqhwiTIgyTt7Q7vO/ktu8056xqUc+mvIWIm7aqeXTBR+yVNArjlBscYQJ7fvp0ol5JMaxaKlKkhcyj3KP6YZ9hhwa9hcwmObbCgM8Uyt+c4Tr04Uj10A/mn3Hc7NTgW/RTkQsq4AjOc7P4+dK8OBGyOxc1s3KefUgbmBxxNhvDYMxpuUO8NE7oQGtMKkCZNsYsJZAt9oorhpU8ws33rp2Vqj2LBhQwEHNYvWv2+++UaIHcPpK5GdzdYyd4cOHeSeep7fkGRctZTD/9bX/Fg9fwwvWvR543x+sDxLliwqyTifmZ5OFONPFOev2yoEjTZ65uMeZBb9zffaJiFbRJEG4IrUcUmQAyWXbynwOChzWZlLNxzkPYOuS3tzmZr3GIcHNyMs36rtWidRJPnLkCmLxE2eXFsOVBii7aP6z/zoQpjXtD8i4dAGh7tCLj/6JC2yfPf9C4miKkdCEcU9AUbM+ictToxLCb+FPyJkfU4gpBOMYQNgDO0DhPWFMbyPHLg1ELOn15J36tfue+DRYBgj+gJX+8J4rT9wcwAQ2RajxvSNc5+wjJgQRFHtel64cadodM34IZZMGMryrWaT+hxRHDRG2pCTFLbjkcDr0Xwg0gQhdZqPkSt/QXFzcvCcRsJKV6wudl+qvYmhZVsOShokisTlp19ou2STJY/CDPOwtJv2DLopBKB4OW0TwZmIR7KTmunStrSRiSj6CVG8b8aswoz1+XUTxQPBwLReaeE98W2cWfAjQtblhvF8axjDBpqxE6mwc3Mg5s+qIXXQuWU+4PEwK+z0B+61Qbv2zSwhEe//CUkUF27cYxND1DL/z6SZs0UUuerApWW275kIjaixbdiOJJGUI5w4cBLqZZJDVd0biiaScXhwAxP9FjKNVTs8RbOdOo22HB0bhri5hUSSGks+S200CSrTJEl1b6TZ8FPzHdtkQ2HpdRDFE2F3ZYI6pvPXOD2J2PkewevyAUG/R5c7XIoO7wNc748Nq7RJxvDeOYCH1tgZANxthw7ta8cbL470wBtNFOPjHicmonj7NpX10X958uRBzpw5JZAbZtgJ1q9fHz0Sl4f69ZN76sZ3330nJJHaSi5x8746hg4dikGDBsluZcZnmtzkYv1LnTq1ThStKuV1fuuZQsxyM0tc3ePQNihq6XmILNOVLF9V2pFOlC2ddlNgfvt9DmT/ObdGFAM1oliuipuQUSUYmeZ/hzSbNUUU0376BZIkTSZOeKltatU56ujQoz/ad+8Han+URrFg0ZLinoTvxXSpDfrwY/sTxZMX7mOL9yNM72jA0TEpcXredwhelwPGoBYwnu8O4/muQPCfQAiPHkB4N+DmX3CvXQL3LgwDLv8JXO4GXOkORPSEMeJv4FlVNK5fxQodcbtMSKIYV/c4XKY1Lz0PGiMDefEylQRD3PmpvmXO3dHUBlJz80veAhpRPKttTChXtVY0osgdsBv3n5Y0Fm7YI7h874PUeCtlSnEo39oSQ13+goahvjgeckfIQgkTUfS58jAaUWz4h2Zr63eVGsXEJYrc2T1r3SkMqGPAyYnv4vT8bDi/5hdEBjSE8XwPK+x017BzpzfKlcqOW6EDgIg/YRTsdDNhpzcQWRUjR8b8NZe4oCghiWKs7nFMxD8mokiNIccTv2tRhIzygG6VPvn0M+QvWkIjioHXJF6Nek3NtqeMR3taNWElUaTm+a0Ub+O991PJhwNad+kdJYeIoZ79wTBij/JGEUVtiTmKKLqZ7CQTkyhy+XrlvhD0qGqA9+R3RO4ErswN+NeLLndCuwIhxE534OHfqFD6W1wJGAZc0bBjNMkdXKXvzuqYOLF9XCDisHF0ohhD01j7UbQmitzUIp3Lz++5FLj7WO169vLyknj8Eoz1r06dOnJPhZcuXTratQrn2foTZDpRtKyd2P87IlHkbL1ZW223M3eoqp2knFFzqZhLM/SjeCUS8L7wEEmslokpoDmTn75ko2BGLT3TrpHYoKaAgpfLQjyoYfI6f1OcdFN7Q5tExitYzDGJIpehBk3bhJXdDDg48i14z/kawat/AM7VhzGwPSID28EY2BxGv9rAOXcgoDYQWh+nvToj8mJ7IOwP4EIr4EILRF5ogMjQcsCT8siT49PYwRLDXYcjigPHyOCtdjtvP35eNrsowk8TArZv7gKFZbJxKlz7YMB32X8W4sh4CkPK/xw1isRltuw/ybPEHrWcCkNMkzasdPROLTJtFB2RKFJLOnnFIUxrZYDnmOSCnfMrcyDSzy0adiKJHX93GImd8AZYP782Ii+2AsJbAjwLdhriWWhZ4Gl5FMzzSQzoiFuw3Yiixa5nTi4tbRS5PuXe+A9pb/pXJFFS2KCJAjGkVh18Ip7Jda68hQR7Kh6XhsfP0xQiXHqmVjpD5iyy+z78iRWGRA7dEPOXE2EPZLLhWETxEZbtDsCOsQYcGpEc3rMzIXD5z4j0rQZjUHtEBmhyR7Bzzg2R/rWBS/WwfGolGMPaAxf+iIadSMFOOeTNnSZuIHHQWDpRjKFhXkQUL1zQlnKaNGkSLQV+J5Sdq0CBAuZwXmfMmNF8rf68+66200tdqy/GjBz5/DdsmQaXftSP17pGUdVG7GfHI4pDZSa9YZ+3YKV2o99lMOcAR5udpVsOSPgv3PXMzSx3aAumaR+5bMyBm3EpkOmyhFggUaQWqYvJVQQ/+0Z7M87aaa9GLSXjcZZPu7FLJhtFRyWKpy8/xeSlOxD8rwG7ByfDqRkZELQiK+BXB8YTJRF5OAOMh9PCePgLGL2+BI5/BZzIjIveVfDsTC4Yz/6MyLO5gYB8MAYVQGRIMeBqccybXil2sMRw1xGJIrGyfMshaVduUiLJY3tzOY/aQba32vVMbVLh4toS45GA6+Kjkhji0mSufIUk7sKNe2WDQoce2krH38MmCP5ILrgD/vQVzfND6g8/hh99b95zTKLofekxFm45hp1DDdg7PKlgJ3BFNhh9a8B4qoyGnUOW2EkPnM4Cj4UFgMDcMPr9gsizuYCAvDAGKuyUxIyxBWNAR9yCHZEoEisL1u+S9m/UsoNMIgRDDwFubiKG1K5nbnbLnb+IhB04e1nD0JUn8kzOvBqGuJmFNonUGPLZgWNmiGkLMUS7VpJLhr+f6gORgcSloxBFlo3lXLjtNNb3MmD/cMqd9PBf+gOeeVeC8VRZRB7KAKMJO5FeX8J47Cvg7NfYPO8H4HwuE3Zyw2iJnbtl0KJ+zN+Hjht67BvrjSSKM2fOFLAuXUofWbZ/336rOe603szSt2+UjRM3wBD0vXv3lt3HO3dqnw9iGDfKqB/d1TCsQoUK8n1pbnBRRJThlj/16cCJEyfKdyKPHj2KX37R3E3QF6P68TlLMqrC33//fZukVN2P6cz0+E3chPi54pdZeg4cJW06eeFKsy8ytSTMM8lcBpMDYWWjuOuU9qnErn0Gy/I17dCKl9WWDtt07YPtx4MwlQ60k2h+x37+Nb+4x6Hw2nlSe/bdd98HNT/c/ezWMMqFxYL1O0TDSG2Psg/qNWgstnn5Y+nmA8jyrfalofFzV8igT5s3tjndYvDLGBSSaun5gzQfInPWb+Nlo3jUtFv6945/vrIfRZaF5Lbn0Gk4PMSAPUMNOD4tHc4v/R44kBbY/TZw4CPgwOeIPPI/GD0zIvJYZuB0VuxeWRaPfUvA6PsjjGd/Bc7mA/wLwBhYBIgoiiF//Yyr17i1J36/hCCKnf4aJG0wdfFam1/VoMNxtavdbKN4RNux2mPAKM09zj2gUHFt6bBT74HiZ3Gy6Ws9bN9fCxU1u8fZfUrbdZ/mw4+xYMMewUbNek2lDIxLcul79ZlgN8Vbb0k4Nyl4eAVgyeb9oiVivKkL14nPPmoUi5XWNiL4XH4Qbem5fovW8ry2mSVuS897TH772vz590v7USR26LB58MSl2N7HgD1DDDg+9X8IWvojcOB/wO6UwH5i5zMzdozHMgE+WbFhSgE89f0NRt8ciDybJwo754md4hjR51tEXIvuzzY+KEoIotjetCGOO9VtfQaSZP5TkwsktfS81VMzV+kxcJTmHuch8KuJAPL79duPBWHS/FVmXCg/irQBpC9XYiDtZ59j4Ybd4lOxipu2CZPhq3Z6Sjm4/K828/F79/S9SEL6VabM8vycVdtEM80yMX0+a730XLO+ZvvnH8vOeUuZy/LtOKF9JpUbBUlMLe+/6D+xww8fNO0wGBu6G7CPcmfqFwik3Dn0mWDHSOwcjMKOyB3fzNg4LT+enSktcidS5E5eIKAAjEFFgMvFsGh6nvhAxeHiRmcpDlc8ICHc4yjixo0mMf24a5mkSy358osqBPOQIUOiPdK0aZSg5X2Sxly5cpmXnlVkOtjmfXVwx3P79u3lWsXhmZ8KVN+AVnF5XrBggWU0cNMKl6qtf+nSpTPbR1rfi+2aefTq1Su2KC99z9WIIrUr1LawzqYtWRft03lKGNGuJvvPOZE8xVvwjXgiO/IUUSTJZBp0K8Nlm8q1ogQt02zVpbc4w6aNotqRzPjr9pwEiSLjqGPEFG0SsmDdTvlKAgU0/fFxeUjFUefRM5Zom2G4q9C0o5GuKagBUESRy95fps8oy95Bt+MmaPkeXNZmPnScTIKj0lP1EZ8zn6XtUrFStXFilDbYH53yGYKWfAvsTg7sTgXsTYPIAx8DBz+H8Ug6wDMdjEczA96ZgZNZAe8ciPT9GUa/XICfRhYRXghrl+TH7r0n443l100Uaefara/m33Lm8k3RPp2n6oqDVrbs2gY3GvdzR6qyUeTgy80AXAImqVbuk1Rbd+49CP/7KiNy5i1g3qxETHKHPTGp4vE8dtYSuV6wYbdoJDkx4eYBZcZgGXfc7GXi/oRl5GSI+aZ89314XyJR1OzLiKFmbbtImtRaxmUjAm1095zW7Cg79R4s/eNlMMRnWLdtO3bAydEG7BlkwNEpnyJgcTZg19vA7vdh3BOFHcGNwo5PVuDUN1HYoVaR2AkoAFwuit6d02Gbh2e8saMeeN1Ekc6qO5u+OT9z2WabGCL54hdv2Obc4U4ytc1Lc3PUZ/gk0JE/+y/TKmHawa7a+69Bo8UDQv5iJc2+DEm+lm05hBQWGOL/oRPnSHuv3O4pWjlqJWk3S68QKj11pv9OloVtRazQQwTLxzIQVwwngazbrI0863fVGEcMac7EmQ/d8rBfqL4UlzPz5c7uLn0G4OhIE3Ymp0XgsizA7rc07FDu7Kfc+Qw4nA5Gr/Sa3PHJEg07kWdzRmHnUiGM7/8Ddu05pqDgdOc3kii+7lZ69uwZzp07hwcPXuy5PyAgABERXCSK/RcZGYnAwECEh4fHHtEJ7roaURSBct0oG0JoyM1rW4KIXxUg0VMDKG0DuZR87rpRnuFz4nT2AcQtxOYjvjgWqglJCjnaICrBybi0YyTB23kyGNuPBSDwrvbZNhIKDgAqLjc1kPzxvM3rnDj6puAl2VTlZHrUUlGryf88tOc1FzkkatwUoeLHeub3p8Pvy7up5SWVXqzPmXZvW8fhswF3gN5/N8WZiQbsHmSA58RPEbQoI7AnhaYV2quRReOBT2AkWTz8JYyeGRBJ7dCJLDCeyAaczo5In58AIYt5geA88NtTHXv2+ca717xuosh3ZHsQD5xIxFTXbAe2E+/zIFnkM2evR5rCNAxxCfFE6D1sOeIrbcZ2PXvdaBNDjMtJCzFEkwQSTg1D3HmvDdQkboKh8PvwOBqFIcvBl+/ALxDRxyJxzGt1cNBnORX2rdvY+prvxjylf9zQ+od1nLhcM3+/m8D4id0RMM2AnQMN8JrwCQIWfA3sIVFMCZix87FgB9Gw842GHe8fZaIh2KFWOig/go4Wx+atx+ONHfXA6yaKfNcoDNmWQ6xXygxiiPX3PIa0NlNyiJuUthw5I+1G+1QSJ65SaM+a7FojngmR23VKk0M0gSAu2Pdp46jiUt4RQ5QNJKf030k5pDCksMJ8KCdV+6rwQBOG5FOQMchY9Yw5T1P/OCcYiqP8Mskh5nvmGjB2ynDzJMNz/McIXp4V2EuiGB07nKQSO/DMgGfHMgInssJ44jvA+0cYldw5mx8IyoPgU5WwbsOr+eFUOLLHWSeK9qj1NyxPVyKKSiBpwkwbVC2FleV/6zhR11HPaWH8dvJ9IXs8M0ylY+s/BbASyBLP6jNpKh+VpoprmRaf46ChHVH5WYarMsTpbE4r6t3i9JwNsshy8ssaQ/s2hOcwA/YMNuDIhNTwmZ4KoXOTIWJ5CtzdlBJPtr+HyN2pgcNpgSOfA0fTAbQZOpoRMBFGeGdDpE8OTbMYmAu+u6qhTbu/490DXydR1OpYkSqtDWKqK60to9onqs2iBkEVR7X368NQFC5jxhDfI3qba+Wxja2Y3lOrE/VM1LvFFj+me2duAI1rF8aJ0bRvNeDwuDTwnfEBguck1bCzMSWe7Hg3Ona80gHHMwBHMwDHtckGsWNU2PHPiYsnqmH8xEXxxo564HUSRa2+LDEUhRHrelHtocKjY0h7LqEwROKpcEm5pcodVZaod1BhKo4ZVxby0DKOrf/qXfmsrfuxhfFZfue5bs0y8BmnYefQuA/gOyM1gmcnRcSKFLj3H+XOu4jckwY4lBY4/AVA7BwjdjKZsMMVDdGcjjIAACAASURBVMqdn2D0ywmcy4UI38oYO3aOgoLTnXWi6HRN5nwFdjWiGJuw0e/FX0A/X2f3cOIyUK/U5zgxVrNRPDTCgMOjDPAcTXc5BhwdlwQnJiSB79TkODcnBS6sSIXLa1Phlkc6PPL8Cc+88wLnigE+PwO0N/P9ETj3Ha6fyIMBQ0bEuxNpRNE7OkG3QXKff5fXUR96GnGv13vwiQBKF/gCvtRGDzHg4IgkODLKgCOjDfAaY8AxE3Z8iJ3ZKRC+LBUur0uFW9vS4dGRn/DUOx9wtijgk8OEnRzAmSzAlYZo17FjvLGjHnjdRDHudaLjJy51RXLpdwvo2CA7/KcasGuwAQcUdihzzHInKYgd/znEznu4vPZ93NxK7OTAs1P5AL8iMCq5c+ZHwC8LHge7o3I15/Wl6PBEkRtGwsLCVF/Tz05YA/Q1WblyZbuV3K1WdVn2oD1XXASGHse+AwuXK/ece4bFwz/D+ekG7B9kwP6hHPAN2D/MgL1DtOsDijyO1YjjsYlJcXJqMpyelhxn536EwMWfInhdFlzeXQ43TtbEg7MNEHn+N4wdOzjeWKxZvRr+O+ijE0WHJ8f34BkGzPgnPcJnGXCA2BnCAT86doglmXiMTQKvsQYcJ3amJMfpaclwbu6HCFj8KYLWZsLl3eVx80Qt3POrD9wqh34DuscbO+qBZk0bY+thP5Nzffv2MV3G2aj/kLvwvQlM/jsLLs/VsLNPJhrRsaPJnSQ4MiYJjo4zQOSOYIdy50MELkyLwDUZcHlPeVw/UQP3ztYDbpdHnz5tFRSc7uwURPH8+fNOV7F6gaNq4NKlS6hS5eUcHUel8vL/alSrgq1e58yfGNOFpA0h6WAE4OTFSLhXLohqOQyY3tmAuR0MGFffgGUdDTKwn5lggA8H+FEGeI834MwkA44MM+DISPrOM8BrnAHHJyTDiclJcWpaUnhPTYab69/DyI5vYdykufEGk1vN6li/90Q0X4U6jhwTR6cjgNrVS6PmzwZMbm/AtD8MGFHTgHU9DPCbZRC7V18b2Dk83IAjMukw4NiEZDg5ORm8iZ1pyXBr/Xv4q4EBCxZtiDd21ANNmzTC2l1H4X3xcTTzEh1HjoEjTlC9gu+jSJ7vUesXAya307AzvJYBS9rSebsBQTMNODfJJHfGaXLHc7gBxI7IHZl0JMOJSRpuOPG4vvpd9G9twOKlGxUUnO7s8ESRO3Gtd/w6XS2/4QVetGgRund/+Zn4q1bfgP59MWTsVLE/oR2KLpgdQzDH1A5sIwrtoPvA6RuAz21g4NSNKFHGHWs27cPKDbvQtcdQtGhUC+OHNkDzWt8jd2YDRnX6EDO7vYWJrQzYRs3jSAO8xmubYUIXJkXHegZs9Dj3UnAaPmwoBo6c8NJuW2J6Vz08YbCosON1Gdjq8xgj5m1Hg1Z9UM29OXr1GY1mjdwwflhD/O6eHTkzGTCqc2rM7P42JrY0YPNgA/YMN+DoBG3nq/8cA/p2+QirNhx9Keyoh8aOHoX+w8fLxiFdDiVMu79KfxK5E34fIY80ueN1CfA4E4n+U9ZixJwtmLxyP4pXbIaCebJh8ohmaFk3B3792oChbVJjZo+3MaGlAVsG03enhp1dgww4O9OAkb0+wrxFHgoGTnl2eKJ45swZFCpUyCkrVy+0VgNsPzoit9cvKCgQhQsXwpGAa+JChgJBF9SOJ6gthbxqIzpP5u7tsmVL48SJE7FC6MFT4OEz4L/tJ7Bx61547DmEZi26wK1WY/T/pyW2bn959xSBAQEoXvw3HDoboWPIwbTPlrhR/y37N7/y43ctEhUrVcbJk7ZdIz18Cjx4BmzZeUqws2OvFxq36Az3Wk3wV8+m2PjfgVixF5ebgqHfiuHguSs6hhwUQ0ruEEcnQu/KRhx6tuDucrrpqVqtOk55+0Rr7kfPgEeRwNZd3tiwZQ927PVEi5Z/oka1Rhg4oA08dnpFi++MFw5PFFmp/PZxw4YN4+R+xhkbwVXLTN+TjRo1Ar9Tbe/f+PHjUKt2XRwJuCp+1vjdXNos6odj1gF32YrD3LB7qFO/EQYPGmhvCGHihAmo6V4Xh/11DDlDvzl18ZFo7zjoN2raAgP6RX0swV5gmj5tGqpWrwF+HYf+HnU55JjyR+Gbcoiuo4ihJi1aok+f+HtMsBfWXme+TkEU+cIjRoxA0aJFwa+q0BfhlStXQNs3Zz4uX74sPhXpV5EHr535fVh2tktQUBBmzZol7eUIJFF1mAnjxqJY0aIYMGKCbEzY63sBe06HOfkRjv1nLpoPOi129ndiu2w+fAZDx08TDHHZ11F+U6ZMRpHChTFo9GRsOuCNfTqGHBJvxNCWw34YMXEWihUr6hATDYXhaVOnolDBAhgwYjw27j+lY8hBZfBenwvih3bY+OkoWaIkBg7sr5rwjTs7DVFky5AgDhgwANwJXalSJdlJy920znpwg0fJkiXlCyulSpWSDR/O+i6q3GwXvlf//v3FCbmj9ajg4GAMGzoE1atWQfWqlU1n/nfOg7txy5UuhQrlyqBcmVKoUb0quHnHWd9HtQvfYeCAfnY1WYgJu6EhIRgyeBBY966AoVrEUBnXwxCx1PefPvA7cyamprRbeGhoKIYOGQy6XXIFDLEvlC9bGhXKlkHZMqVM7+X8cqhq5Uro3/cfh5RDiQlepyKKiVkxiZVXjRo14ObmJkdi5ann41o1ULFiRcFP7drO66fLtVrE+d6G36F3d3e3qxsr56s1vcSWNUDzMI5l5cuXtwzW/7tADehE0Y6NyO9IV69eXTpXrVq1zN+VtmOR9KydsAY4yFNAc6CnXaj+02sgvjXAwV0RRR1D8a09PT4/OVu/fn2dKLooFHSiaMeG1YmiHSvfhbK2JIoPHz50oTfTXyWxakAniolV066Zj04UXbNd1VvpRFHVhB3OOlG0Q6W7YJY6UXTBRk3kV9KJYiJXuItlpxNFF2tQq9fRiaJVhSTmpU4UE7O2XTcvnSi6btsm1pvpRDGxato189GJomu2q3ornSiqmrDDWSeKdqh0F8xSJ4ou2KiJ/Eo6UUzkCnex7HSi6GINavU6OlG0qpDEvNSJYmLWtuvmpRNF123bxHoznSgmVk27Zj46UXTNdlVvpRNFVRN2OOtE0Q6V7oJZ6kTRBRs1kV9JJ4qJXOEulp1OFF2sQa1eRyeKVhWSmJc6UUzM2nbdvHSi6Lptm1hvphPFxKpp18xHJ4qu2a7qrXSiqGrCDmedKNqh0l0wS50oumCjJvIr6UQxkSvcxbLTiaKLNajV6+hE0apCEvNSJ4qJWduum5dOFF23bRPrzXSimFg17Zr56ETRNdtVvZVOFFVN2OGsE0U7VLoLZqkTRRds1ER+JZ0oJnKFu1h2OlF0sQa1eh2dKFpVSGJe6kQxMWvbdfPSiaLrtm1ivZlOFBOrpl0zH50ouma7qrfSiaKqCTucHzx4EO1bz7zWf3oNxLcGLIni48eP4/u4Hl+vAVgSxSdPnug1otdAvGugQYMG+ree411rzvGAUxJFCjJ+uN6Zj6dPnuLGjRtCFGvVqoWaNWvKNcOd+b1YdmcgKyyjq9RzxQoVQQy5u7nj9u3bcIX+8eSx45MVV6hn9gOj0YgK5SvIIF+5UmWXwZAzyCFXwBD76sMHD1GvXj2RQ5x0PHr4yCXkkDNgKDGoplMRxf/++w8tmv+OqlWqCiA5ODrz4e7uDjd3d9SuXRv8z8OZ30eVvUrlqtJO69atSwwMxyuPLVu24I/fW6JypSqo5ebc+GF9u7m5CW6IIYUj1Q7OfK5atSqaNWmG1avXxKt9EyOyh4cHfm/mOnIoGobca8PN3c3p5RDfqUrlKmjcqAlWr16dGLCIVx47tu/AHy3+kDI6cz9VZVfjl5JBroAhvluVKlXRtHEzrFy5Ml7t62qRnYYotmrZGlWqVsGKzYtxKswLZy6fhK8LHH5XTkEdrvA+bBfvsKNYuWUp3NxroUH9hg7TZzp26IQqVapg2aZFOBV6FGeu6BhyRMwpDK32WCEYatigIe7fv293HD179gydO3aWZdrlm5fgVKiXS2LIFWQr+/bp8GNYt2Ml3GrXQsP6joEhgrjbn91Rrlw5LNu4CCdDPF0SQ44oV+JbJvYDnwvHscZjOerWq4N6derh7t27dpdD9iiAUxDFTh07oV2nNrjwMAghd8/B/5oPzl09rR8OWgdsn+A7Z3Hx0Xn0GdgbzZs2twe2o+XZo3sPtG7fCmH3A3QMOShuLPu0wtDlp6H4e+BfaNK4WbT2tMfF373/Rqt2LRF6jxjy1+WQg+NIYeji42AMGNYXTRo3tQdsouXZv98ANG/ZDGEPAnUMOTh+RB5dOy1j2aXHweg/rC/q1qkbrT3flAuHJ4p79+xFyZIlceVpKM5dO23Wvp2N8IZ+OGYdKA0pO9pVhKNc+fLY7rHdbn3q4MGDKFmihFkwq/Lp+HFM/LBdVBsRQ5efhKBqtSrYsnmL3TDk6emJUqVKIejmGfhf9zGXT8eQc2Ao/GEQarnVxH+b/rMbho4cOSJjWdAtPx1DTjR+UxZRDl16EoKatWpg44aNdsOQvTJ2eKLYqWNnzFk6Q2bxbDBdMDuuYLZuG7YXNS/zls9Ch/Yd7IVxdOvaDdP/nQwOFjqGnAc/ijASQwvXzEPH9p3shqHevXpj0uzxsqqhY8gJMXTPHwvXzkPbNm3thqGBAwZi4qxxok3UMeSEGLp7Dss3LUazpvZf3UhsEDs8UaxZrSYO+O5GwHVfnSQ60SxMkUZqX46c248aVWskNrbN+dWoVhP7fXYj4IYvzl5xLgGl6vFNPXNAZbsd9N2DGtXsh6Ga1Wtiu+dmBN48o082nEwOCYau+8LT/4BshDQLhkT+w6VvjyP/CYZ0OeR8cpimDCdCPFHyt5KJjBz7Z+fwRLFWdTcc1Imi05Jkdq7jgYdR3Y6DfK0abjh8dr8s97yphMuZ35uTjUO+e4Uo0pWLPX41qtfEDq+t2mTDyYiSM7f96yq7GuSrVatuD/hIntzFv/uEh670cNL+QwydCjuqE0W79aBYMiZR1DWKzjf7UgKenetYwGG7aoOEKPrpRFG1ibOdA4QoUqNYU3z+xSIuEuwWNYokioHUSjvpQPcml1smrMFHYG+iuOu4hz7ZcNL+E0UUSyWYnHHUhJ1Co/iyRJFLDupwJiGpysyzM5XbVlmdnSg6a1s4a7ltYcjZiaIzt4Uqu612caYwZyeKqh14dqZ6d9Zy26pjnSg6Ko0F8LIaRQKUDcvd0nRn4SgdjOXgDipbQGQY73NgZLm5O85Ryh1TeV8U7sxEUWsLX9BFS+hdf4dpCz/aWV7VsGKr/lW5BUM3nR9DzkwUVVtcfhIqG7scoT/7RUSRjReVh25B6FLqRfFs4dCRwpyZKDokhuJAWBVmLj0OQZgDjcEvi0udKLogUWSj0plpg+Z1MXneOIcQdiSI13BBnHhyoLcFWJbby/8A6jZ2x4Y9qxHk5AM938dZl55Zds+AA2j8RwNMWzDRIQZ6YuiqMRy+F0/IhEMJY0ssiU2f317UaeSGzQc3OP0GDGclimqAP3BmN5q2boTpi6Yg+M45u5IukkTi+sqzMPloAfFkC0PEU8CNM2j3Z2uMnjbMoSZKlliP63++83EnXHpWGNrvswtNWjXErKVT7d4WqkzEkO+lEzbHMdUudCfVutMfGDtjBMIfOLfXCWJIs1HUl54djjK+rEYx8MYZHDi9EwaDAeWrlsUNXIpRICpQJ+SZAplAy50vFwaN6Yew+4E2y0NiuO3wRin3sAmDXKJzOStRZFtsOaS1RVW3SjK4xjSoJiR2VNrEEHf//5wzB4ZOGBgjNqiJ3rRvjWBo4pyxMWJNpevoZ2cmiiSGq3csl7aoXrcarjwNs9nvE6sNKIP4xYnsP2fHuJmjYsQGsUan+ZSfeQv+iqvGC3Yt96vWjzMTxZA757Bi62JpizqN3XE10r5twbr0vnAM32X/FjMWT0Hw7XM2ySIxxLITQ8VKFcYtXHF6DOlE0eEoolagVyGKh3z3aAK6dtXniKIa8Al6HhREKsxSKKkwLZ62ZKzCVDyNBEZfTmYcLVxzzssOs3LbEinPmBmjYh7kb/phh9dmiTd66nBzPJWn/zWNLJAw2HJAzjyt30U9q+5Z3ud70f0ItVCW4erdLM/merJYurK8b+s/n3FmoujhqbVF7Ua1niOKql41bMSt/thmrCf1rPrPtlHtqeqRcTQMac+cv3UWa0ykY/zs0bjw8Hy0dNRzJIpbDq4XDE1dMNFMBlSetjCknuXZEicqXD1reU+FvQyGVLpxOTs7UVy/Z7W0Rd2mtZ8jipZ1aO5fNpb11HIx47wIQ9Z1qrDFcA7qizf+K+WZvmhyjD79+Ay/gpU0aVIUK1XETBRZXh68TxkkssNCfqr3scSJrfKoMBU/NgypOOoZqYNYzHdUPMszn3FWjSLHjtXbtclG45YNcTUyPFq/V/XDd+TB91ZhlnWg/rNtbMXjM7xnq+20ZzTtM025ZiyZIhhasHaukEGVtuWZz9Bkh0SxXOXSZqLIfFReCYIhG3VgXR98f5ZPzHjiuLmGz+hE8Q0higQu7W7oxJdffOD3QAluqtGpJleAokZS7Bvv+iPiWRj8LmudiP/PW9gOamA7Bd9LJ6N1PhXO2Tv/06v7jqNbpNOwk93FNXN8y85FLZYtosjyRESGi3D2CjiAo4EH5T/t55i+SoMaDDXDUx2S9xh2/vZZmAec6z7yXnz28Nm98n1m1kH0d/PG5cchYuNJOxN2kgsPgmRwUOmofGM6s3O5GlFUGOKZ9e8dfgznIk4LCVB+9vwivAVPxBDb5MqzcMGQ/9XTiHgWbrY/VQKTODljwpBqN4UhLu+wHolbuteg4J29bBru4Xq0tldtEBNRZNmYNycFXgEHBUPEFfuBepZnlpdYUWGqT1iHU8hHRIZJPMFQ+DGtH1n1Dw1D/oIl1hU/xclnVfovOrsWUQyV91YY0uRQqMghtjflC3dWqzpnWxJDJAs8Uw4RC5QFyiyFcRlG/Cg5peo0CkMnhWBeeRKKzaZJxKL180QOqbiWZz5niygyXLPZDcDx4MM44r9fNJSUHZZtao0VlbaE3/Iztz2fUcTn8Ll9Ug+sAyWLtb7gLXKHnwS98DAYfpdPgpMmpqXSfdGZ9eNqRJGETY1lPHMsUxgigVcYYt3IWHZPk1e8x/GK/V6tbDEunyWGzpjGOlWnDOd9PnOO2HsWhpVbNaXH2h0rcBsRNtuBzxHn1kSR4cRQyB1/HD8fhSHKEvZ15ss4lEFsZ1UOdWa7s19QxjJMk0MaeT58bi9OXzgm/cMsi03ElHLnwoPzuPgoWN6TaRPjcSWLxJBOFN8Aohj+IBA9+v2JH3J8hz0ntyPLt18LiAnkfIXy4tj5QwJACqlN+9biw4/SYN3ulchXOI853q/5cuGAz27zBhkuJeUvkhfvf/A+ruOiAJed6jouoXLNivLcbVzF6GnD8dZbb8l1qg/exxdffoZTYV7PkUVbRJGdgp20YLEC5nKwzDzadG0l31Rm55m7cqaEDZ80GBdN2iaWhQJ24Ki+cm/n8a0y6DCfitXLR0vvp1w/4vDZfdJ52FGDbvsh3VfpZImqQrVyEjftZ2nFhlJpIFXnjenMzuUyRPFpmGh4u/Xtgp9z58Cu49uQ8euM5josUCSftCnJF4Xcyq2L8UGaD+Dh+R9+zv2TOV6e/LnE+S9JAtvnBi4jV55f8N577+IOrkqYFn4JZSqWwjvvvoNruIjB4/rj7ZRvSzqpPkiFtJ99Ig7E2VaW9W+LKHICcCrU8zkMUWPUoUdbEaDE0MwlUyX98TOjtN4sCycKvQf3lHuCkTvat47LVyljfi/i8ZfcPwkB5cSE5ToX4Y30GdPJElXREoUl7qeffSIDBLFhWe6Y/rsaUWRddurVHjl//Rn7T+/CF//73FyHBYvmF7svDubE0KIN8/He++/JBCHHL9nN8fIWyI1jQYfNdrOceH7/43eCCRIvVZcckH8rUxSpP/xA5NM/w//GW29rcih1mtSCuQuPzpvjq+fYdtZEke1FO7M/OrRAsuTJzGVhuxcpXghnr56W+zR3YNj0hZPMmxiIIRKTLn93knsH/fZI2dm2pSuUjJbWL7/+hFOhXvL+fI7Yr1C1LKq5V8HwyUPMceeumCFxVJljO7PsrkIUOUkg8WvbtSV+ypUDB3z2RMNQ0RKFhExTDrFOqKRI+U5KjJ05Ci3aNTXXX/oM6TB/1SxtAiJj1kVk+/FbZMySUQgon2X9kxgWLl4Q777/Dh7gBvoM6YUUb6WQdNJ8mBppPkyD8PuBNjFkSRRv4orIBH4hq0XbpkiSRBvDiBUeRUsWlvGX49WoKcMkbM7y6WabTA1DoejQow1SpEghbvMoQzmx4rMqHZ5/LZAb3mHHzBi6icsoWb642P73G9nHHPffVbM0DMXhQwzEkE4U3wCiSAHdvG0TM0gooLjRxb1hLQnL+HUG0XZwwFSqfoKOhHLJxn8xbOIg87Oi/bnug2vGC9JZDUkMuIXL0lkIaNpjlK1USuKzg2zYuxpuDWrIdcXq5dDtny5mY3JLAWdNFLm8yE747rvvyLNDJgwEZ3GL1s1Dlm80ojtyylBZhqQdEcv7ddbMsmlGpXvlWSi++N8X+OyLT0UocCBJlyGdxG37Zyts2rsGY2eMlGs+z1kjhUywaTaYKnUquVeqfHF8/c3X0nHPxbAZR+WpzuxcrkQUOdg1ad3YXFe0f500bxxq1KkmYRkzZxANCWf7C9bMNsf7/sdsgqEh4waYwzij5YDMiUS2H76V8Pu4YSaKtxCBIiZyRY3N2h3LUaehm8SrVKMC/uzTSQZn67awJooc3ImjJCaBTPvGtTtXYv7q2ciURSO6k+aOkzgkJ8QAy8sBSbUjd+1+8unHQkTu4JoQgvQZ00vcjr3aycSKOOSzPJinLEtePS3XH378oZyLlymG7D99L9oga4Kr8rI+uxpRZP+r36yOua6quFXGpLljUdW9soR9/U1mMZXhIKiW+FinP/6cHUs2zofCULJkyaR9aErwADdB7JEQ0B5b1SHzylvoVyF2HPBXeSxDrbrVJZ/qblXQsWc7sUVU8dXZmihSzpFwVKpeQZ6lHF2+eRFWeyxHVfcqElaspGaHRg04y8t25gCtVh9I+Eh6KYuIIe6E/V+6LyVuj35dsWHvGoycrGEoSdKkMjlheTh5ypHzR7ydUiO47BMkuZywE2OqzLGdXYooPgvHDVxErXpaO7KuSaInzh2LKrUqmeueMoOygfXOOJxkkuBPmD0G/66Zg48/+UjC/107B2H3AsFxKkPmr0BlAIko61MR9Vx5c0pcrmLQhKpyDQ0H7vVrovNfHRBssQKh2oEYsiSKHBMpP8tU1CYGbbq2FAyt2rYUlGcsY7nKZQQz3KzHaxJhToJYDqZL3DA863dZBOcXHwdDyZbu/bpgowWG3k/1vmCMG0fv4ya++f4bpDRNtH8rVQSpP0wtE3ZLTbgqu62zThQdlCSyWK/TRpFEsVXH3wVonf9qj4e4JcTwMW6jdEWN1HEphYMcB1ICkuSRAzZV9BSWymaN2jgKQRoW58zzs3RAa6KotC0kasx7u9d/kua0hZNFXc+OpDqAAqY1UWQn50yQZRk9fbh0FC5DceDmAMpwtwY1hZwwrhpsDvntkaUpqt/3ee+QeP1G/C35chcjnxs1dbgsX3JA4rvRXyXDWUdcKmU4r3mQPLKjU30f1wGe7+SKRPH3Ds2lTnoN6IYHJgw9wm0UL/ubhHMZmRji0h7rjjP3x7gjQotaZ4/DmySc5JJtxqWbH3J8LzNsa6L4W+miEjf0XqDM8qkRZpozl0yRAZR1/ByGrGwU2bZqIkAsceDlZIjL0CeCj0h69ZvXFSFOPHPjDvOgBkZptrYe2iBhnCwR90PGD5RrTrQoyAVDuGheGm/btRUijOHwv+qDZMmSSlxqt/gs6yY+GHJFosgdrKzjf4b9hYe4LXXyFPdQoGg+CecAT5kza9k0uSZ5JMZIroghfk6Qz9dvWkc0PhwIv86aSYiYNVHkige1iKx3kr2N+9bKswvWzBXZpmEouhcGto+lRpFa75MhGlYKFSsoZSYJYBkpRz/94lNpZ5aDE1ZqAFk+th1lADfGcBBnGFdXiME+Q/+S69lLp5sxxJWY3Se3Szi1rjRbYFyu5PDZ6YsmSX4kHHEd4JUcchmNooko1m1SW+pk4Ji+MlFg+wKP8FPOHyWcS7CsI0UUWX80/WA/pBy/8iQMb72VAh9+mAYkXBznMmfJiM+//Awcs5Rs4fN58ucGlSFsH66icZLA9FZ7LDNPTKzlEDFkSRQpJzgu8Tli6Cnui9wghvifxI5KCWKNpLVY6SISl5NXYohnZTYxZsYIUFZ269dV4nDSSxJJOcQy7jy2TcJ79v9TVtQY90eTRn7OihlSXxqG4raqoTCkaxQdlCy+bqL4h2mQ33tqh9lO7NKjYNlFSgDvO7VdNCuKKE5dOEmuVScgWcqQ6SsYkiQRMHMQjgtRJIDX7loh4B07I+bdhtZEkZ0/6NZZ6dyqk1J4crmIy+csc5VaFWXwICncfVKzYevYs62QSbpSafR7PYl3MsxLiOEPObIhSZIkuG68LAKCeVDjRIGf+ZtM4NI43yvkXoA8xyVVCgt2lvgerkgUW7RvJvXCJVi2F7FB4t53eG+tnkO9pD4XrZ8r1/PXzJYBnvF43MMNfJL2Y7yX6n1pVw6EcSGKJHfr92obI6gVYLspXFq2i7VGkYKYGIp4GiZCnZoGYoiDw6b966SMNepWk3egZoBmF8QVzTQuPQkWzaIalM5d9ZE0smX/VsgHhTrTURgi2eFAw6VxCm62P9MqUDS/2Zjdsqxx+e+KRLFxywZSLz4XjstgrjDU7Z/OEn4y2EvqVBHFJZv+jYYhUqw9twAAIABJREFUksZUqVKJlpd9lxOWuBBF2oUt36LtoJ2xaDJCY/CRaE0USf64iebykzCRBdRUEvOXHoUIBqnNZDtThpAUUkvE696De8gkmc9Xr1NVwi49CRXb1u9++BbUij7CHekvxBAP4KnIoM+++Ewm4sRRrry/mJ4NidckQ+GLOHQ9ougudcLVH74fMURbxfbd2kj40YADsjKkiCLrn2SQ8XiwDX/voMmy/T47ZWyIC1Hkasm8VbMkj4Xr5sqkk+mpulZna6LIsZPlJEHjZIYabsHQ42BRRFBucEmZMoXjpcLQP8N6i10hJ5k0xSGuuMpGV3OZs2aWa+CJkF8lhyLxAKnTpBKzF9okUuP+fY5sMpnhZEttCFNljcuZZdeJ4htGFGkfyNkWAU4gDRrbTwC333uHCK01OzRSt/nAetkIojrXNeNFlDEtKXN2ztm9LaLIGZGlRpHAX6eIIu2/YnGPY72ZhQSBIOXsKG/BPPjItIzHDsOjmntlKQfBTqHw+Zefy7ICtZjsXIyTv3Be6Vic8dMWTj0b05nlDb0fKPG4nEGhYksYvKiDsdyutvTcvJ0mXM9cPiEbQ1gvFHi03WF90r6KgzeFKK/3ntouhJJ1xbgUhIV+Kyj3KOQ5044rUaQJA9MkUeTGIlttYosoEkPc6NBDMPSreblGtb9b/RpCWrlMw3f59PNP8dHHH0lZSTQZj3ZAnEAQ82nSfCBh6nlbZ/arwOu+Eo9LYxwYbJX3RRhyZaJIjYva7c4BtEvvjlJfp0KPRiOKRwMPmT8fqDCUK59GnjggkzjGiSjejU4Uwx7YdqZtiyiyLxPXk+aNlQHb0rZStT+JItuT2uoPUqfCJ2k/EQLCCRXjUENO2UiNlloyVM/aOnPyexfXRcbyvuUy5ItwY3mfZXdVosg+qRzwU+ZzSZd1dSzwYDSi2HNAN+njxA8Pyo8RkzSbz6X/LZTJRvyJ4rwYNxVZE0W2u8LQhNmjUbJc8Wi2lSwzxyaOWWw7rrQwjJMQkkyOSbwuVaGkEF6Wn/bbDIvt4DhPXGbL/h3ee/9d8UNriY24/mfZdaL4hhHFE8GH40QUPY5sMhvEsnORiKmNHRTsMRFFztwqVNU2gFCN/7JEkWTU58IxmWmxM9A2w61hLQwdPxDe4Uelg1Rx0zSKLB8HZ0VY9pz0MC/3UDPBLyxwZpU8RXKx8ZkwZ4xoUmmzpo4x04djxJQhCLzpZ9Yocpmd9mpMP66dSsVj53JVosh24aYe1ktsRJGzdQ54rBPGJVEsWkpbVuHGqSiimERsaRiHBzFEmz62O5eeSfZehigSE7SpVcL0m2xZxS6X9mDcVMDwmiaNorzL0xD82UfTbHHDhNJAcamS2gRi/t333pUJy7hZozB0YhR+iKMx00dg5JQhUt4Ak0aRG7tIHpi+wkZcz65MFINpp3pNM0GJjShy2VfZ47EOqZHJU/BXaTv2sehEURtoWb+c4FGbq5aeKYdUe1KjGFeiSLlHzXGBwnklT5K88tXKgeYXR87uw69cmjRpFDkhYb7turWWMO+wo5i3Sttot3r7MsEFB3kSSW7ImjBntFn+mOXQjBFiLkH8UKPIyTjTZ5+IK24s47GOXJUo0p6QpIy4iI0o9hn2l5geMB4Pat9ozsR6Xb55sQVR/Py5pWf60lRLz9E1inEnimw77jrOU0DDLdu+QrXy6DWgO46dPyxLwylSJBf5yPJRXrTs1ELKRxOZGYsny///9q8TDLH8yZMnx2eff4opCyY8j6HpIzBu9igxf+G4l+3HbEIs1bK6JT7i8l8nig5KElmshFp6fhFRVEvPSzYtMC/5EEyc7XyVMT2SJI1aeuaOVhoJ38YVMxmgcKMhLjvhyxPFYULsuCzIdJZsXCCCkoKBSwzsKAyvVlvTKLJzcfbEWQ/D+438G1VM9maMS2HCZc5MmTOAxuJ8F5II3uNBAkBySTcL1HSF3NNmcDpRNED8KD7lUkkIlEbR52LsRFEtPa/fvUoEmxLQ1Oqm+SiN7JQnNtgm32f/TtqMAk3F43IiN1IlTZbseaI4J44axX8nyNIeN1BpGPpXZufEEN+FrjAYrjSKzJu7CCm4Gd5/ZB/ZwJA8eQqE3j0npIZa9QyZ0uPtt9+StCwxREztPLZVvj6klsSYjk4UDdD8KIYKiVJLz3ElilsPb4y2skFiyA1uHGyV2Qg3JnGziNLIUF5xUpLlm8xI+W5K6ePRieIUhMZRo8jJDDdAsC2btGwkG7BIBrlDlXjOavIgwXIRQ5ZyaMDovihbuQxIAogVbQk7BF9l+koGetrv8h2UHKL2iJ4CNLdQmsxSRDEmVywvGuh1omhA09aNBRtKvrD9mrfRNuYd9N0lmt9MX2cQ0xHl9or1yv6e6euMSJY8qdj/vQxR5MbOJ7gjX7gihrgZihtjqIBg2z/BXXyZ/kvxDML2Z76cGHHlj/FHTR0m2mj+59jK+7S1pPcQhtFcQeGHZ/aB7V6bBUMc92ijqBPFlyd6rGGH/tmbKP6aP5eQNWrjOIvnZ/UITBr/U+XPWU+REoUk7MylE7K8y5mT2kDCuIoorjE5TZ22cKKQAzULtBRy0W0Uh4E2X8oIl0SOS1XcLUtCMXC05vKGG3FYFgoA6di4iIJF8kmHZ/61G7lJ2XmfBFNpi/oN/1s6K+3SSBpIEBn/k08/EYN0nSj6mTcvvRxR1Daz8Gs8tJmhgKUAU06zKbgprBmWp4CmkSEmqL3hgHjojLa5iDtZlUZx3S5tkxXd2HDwZnxL/PC/5dLzlH8nAHiMLN9mkbblwM5yEEMUnn2GaBsKqP2OsNjpSGH9y68/I91X2q5UfsaQAwYxxMG+U692kh412xTcNDQn+aRfNIUh7uam30he60Tx1Ygi7YSf4b60Hdtm7U7NNKZVpxZS71yepQ0W65qutIghTkCU71Zq7ziA0kaRy4yMR2fJjGONH14TV5abWShv+o74W55buG6eTCqpDaft4X7TF7CYptL4ESfEJ+0Q/5f+f/Jcy46/m5c+qYFv1UnbWEgNNOMqDNF0g2llyJxB/HKyjDpRjHK4zTGHu57rNtFsFOOqUUzx1lti8kSbfPZhygLWc/qv0km7cFLxw48ahrgJjW3Luldj3tvvvG0minOWz5Bn12znZpYoDbYlloghtZmFRNGIh2I+xTzpuYO20hzvaEe/75S24ZIu5Pg+Kh1uEKUdIjcE8rn23duI1pT4YvmatNQ2hdElE8vKscxSDhF/fFdOZnSi+PJUz6WJ4kEfbaCtXLOSECWq5Zu30dzjHA8+FG3pecAozbeS9WYWgpNLN/RR2HuI5keOYVxK5CyVg65SiTOcPumqmYy2v8mmDc4kihyYdx7THG5TI9mpVwfZgUbXFgS96hjsOB5HtN3RIycPEfDT/QDTpk0Hv/m59dBGNGihbVBheM48v4j9F9NgWizT3BXaUg/v7/eOWvpk5+VGho9MrhG4PMTNC1MXTsTbb2s++jYfWCdL5co9TtnKpd/YpedtpraoWU/7/BoJtXKPw12FlkvPfw3sLu100ryZRSOKbINipYpi7soZsszCax7ckMT2oA0gCR3DqKmmMKxWW3M5Qrsahofe5YYUP7NApQlCh+5tpb2ZhsIPz4zHNuRzk+eNFxuvtn9qy4BlK5XGAWLo8EbUbaztmmQ8Lh0qdzjEEMs0c6nmU5H36TOSxIDpE/dcBueEgveIT26KmTR/vNlX6O7j28RfpyKK5UlE39Cl57Umck83XBzEODngN+hZd5QLlkvPdFfD8FMh0TezMKxE2WLgjs2e/btJHC7/qx2h7PN0ccR4HGwph5QHhKzffS0rHrSLZvwth7Sv9hBbrTr/LmWwxA//E1P0rsD0ChUrIBOXLaad79yIRRMI4qj/iCifdIxLUxk+a8aQyS8n7+3z3mne/EUMUYYqe2vaX2/ct0ZcTTEuj70nNdteEgDlQ/JN1Siu8tA2BzVoXk/6EYmie4OaUk/WRFEt11pvZmGdss/SjQ7lzTvvpJTnmXbo3QAhkdyRznicnHb6q72YF/CapJ1nbqikj1T1lTG60aK8oqkS7SQtcSQYUnaF5YqLnaByO0fTBcrWfad3oO8wbROg8qsoPlhN4yIxO3n+eMmb+XMiqkww2G+IUfYD3jNjaL7WDxi2++Q2kYckilm+yyJmVxyPLcsZ1//ErG6j+PJkM0GffFmNIpc+DvvtE8DTQSy39HNGTTB98tkn4haEDa8EGm2t6ET0oM8u0DG3WnrmNnw1myXwaNB/KsRTdvbxWYKMHXXi3HHmWQ/TpzNY2gBxdyjTU/GUnyumxSVJCm6VDuNwuY6DLHd/TVs4SUgfCW79ZtrAwud4ZM6SCYfO7AEHfs62uBtVEQae+QzT+DjtR0IUVB48U0CzY1esFt3h9hfpvhB3B9Q6Mg0SDvpbpOaLszKVRlw7FuOxjp3VRpFtsfP4NnHbQH+Tanmse7+uYqQvRNGEIdpckdhzFyeJIrGm3OPMXDxFvouq2o4G/fQTptqe9crZMbH2ZbovpH3TfvoJ5q2eJQSe+GNctgmFXDkLJ9e0oWU5LduE19s9/5MlazrU5XIj8e9WT/PjqcpBsnnQd49MQNJl+J+QPyXsmRc16IybIVMG+a/yYHnpwJtah/JVNDcoKs3/pf8C1HoSL0yDDre/ypQezds2fWkMObONopD2g+vF1KB997aCIfbNLr07iOkB24r1xDolZgaM+kcM+k+FcDNLlHscapC/zZZV2oN1XaLMbzjLdrCQHySgIyYPMWOIhI7mKiQG1FgH3/GXvJhPhWpR7Ua3WJSXqn15ZplYdmKEKxIsMzVQ1P4pp+8sBzcH8FOAIyZp/g+53Cf2uFe0NEgsGI+EguVTeQiGxO3JafNOVoUhLkFyoqNkDldSKlUvL87t2U9UGvE5Uw45q40iJxN0LUTfkV3+7ij1wjGHu5tTpkwpG4NYFwpD3Gn+8Scf41iQ9vUuapdZt1wFU4oSXmfOmgnrdq+SVSY+yzSY7vCJg/DFl5oT+M//97nYtI6aNgy58+YUHBAb9LtYplKUo3ROGmxhiHKCJgaN/qgv7U9t5ojJQ8Utj2rv91O9h1lLp2HIuP5ImiQpOCGhwoRlYrtRs824P/7yg8hJ1e5RGPKRz0yq9HimZxJOhgVDl0/Jqg03wfzw0/fSB1Ua8TmzLDpRTFC69/KJvyxRJAAIXGpJqB1RHYEzb6q82ejKESzvUUPCuHyGwldpARaaPnPFjSOMR60IBaFKj/nwP4kChSnTZWejkFcDtCUYueRClxj8qDo/+6Y9H30mxhkTy8JynDUJXNkxevkkPP33w+ficfNydug9LkVHfRKL6fH6WNAh6Vz8mga1YNbl5XuSPLDT85OANBamqwvWj4rLe3yXMBPRtXyPuP5nfTgrUeQ70l5Pw1CAeYCKhiGTgGWdcVmPcfnOXHJRRHGNx3Kx5VOfi+SASYKg6pn58D8HYsHQdWJIm9gQQ7TFUQSOcdmexACJKtvIVluocisiobVlmNgkKgzRDonlJIZIClR/YHq8PhHsKRjikuPlJ9EnCiyvYEic+p6Gl/9+DUOPQwT36t2YLzHPPmirnHEJc1aiyHdjPbCt2X9pvqLC6JaGWLFsP8Zle7NdiCH+V+5xSBRookA/eEyDhIltrOpZpUsMcXCkplv67n1NDhFXKg7zZBzfF2CI8aghJLHkszz4xSdiijLjZPARGXSZJ+Uny6vi8czB/kSIhqFBY/trg7bF6gnLrmQd/0uaIUcE30yLYfy8Gs+cpL2sJohlYX06I1Fk2fn+HHOkvyp3Rlw5MmHIUjYwPvszxzj2G14rolivaR1Z/uUnECk76EdR1TPjqYPYEAzJ5/o4fgbIuMB+zDgsj2DjcfAL5RDjqTGEz/F5pk0MefofwMkQT2lbti+1g8SRKgfPXEomCSX5Gz1txHOeHpimYMjkUUHGshBPkZlMT/I0YUjkq9XnSi3zetF/Ykgnii/P5RL0yVchirE1vOWgaB2PIFYaRS4XkhgQJDwYVwHe8jkVxo4RUxwVP3paUR1U3bc+M20efI6Ds3U5VJ58jgMyXQGUM2l6OEip+JbpqvKqNJku76twy7iv8p/pOzNRjO3dY6orhmtEUXOPs+y/RSIAWReqLWw9q8JUe6prW2WwTMvWfeswlRafeyGGHgTJBhjlQJwDBZd5XpRmQmHImYmidZ3F/VojjYooym7h25qbrFfFkCUWVFpxLRefJT7Z1tLeV6NkhiVuOfA/xl3kL5pfBnkO5rbysiyLOU0rORTXb/HG9g7M21mJoq33ikudqG8hK6JYp7GbLB2znVRbqPq3zEOFWban5X3+V3GYjkrLOk5M13w2CkM+Ztc+jG+ZJwklN/SpXdIKQypvlb66ZjliwhAVLir+y56Zvk4UE5TuvXziCUUUYwMLiSI/dcVZDL+AYalli+05e95jZ6Ej51Sp35dys+y0MeHMUnUke5SPnctViWJM9cn6JlGcv1pzSrto/Xyx64kpviOFU9vJHdnED4/ufbuK02N7YujNJIreou2h+QnbgS5tKJccCSsxlYWmK6kt/Gz2GdLzpe1TY8ojvuGUQ65EFOPz/ldNS8/0ZUpTAHv25biWm3KIX41RcojeF6jRtGfZdaL48jwuwZ+sWa0mDvntNavR4wq0V4nHWQnV3VTV002D9TLhq6SdUM+yA3GJaMj4AWJzRncCXNpKqPzimq4I6KDDqFa1eoJjJaYMalavJcscXI6La7lfJR7bgpjhBpB6zeqIyyHRvlgs77xK+gn5LCdFg8cNQMnyxcXHGu3D7Cmc+a5sN9obV69aA5GRkTE1c4KGV69WQz4LRq1GQta/ZdrEED8ZSgxxZ7EzYYgeGbj5buzMkWLfancMXfORZc6qVaslKE5iS7xp46byNa3Ebkf2adoITluk2bzbuy0sMR7Tf66M8YssJcv9Bn4mlMQxtlXAmNJ5neHceMol+xLFSsbWzC55z+F3Pdev2wBbD258zi7wdQLAOi12JHZm2hVxp6czdCy+A8vJLx7QlsnSrsj6/RLzmnZU249sRt3a9ezWgerVqY9Ne9fh/G3NQDox3p9twYGeGl1nmGioOtEwdF5cXhBDavlK3U/sM8tDe0mPw8RQXbthqEmjJli9bbnYTLFMiVEP8u43NQyRoCZWvq/6biwnl57pNsXSPvxV033Z57V69MN2zy1wr+VuNwy1ad0GK7csS1Q5pOqMNs/OsDKmyss2o00hxzJ+ck+F2+vM8nAs23tyByqWr2g3DNkrY4cnimNGj0GPPt2iPv/1GmwN4gI2AkMdcYnvKHFUmXm2a5lMBsQ0QP+rX08MGzrcXhjHlMlT0KVHJ/MyfFzse15H3TlMW8RTk+kw5VZG6I+D0X94Xwy3I4amT5uOLj07RX3ZIiJx+pfDtIWzYsg0gSbpGDi8P4YMHmo3OTRn9lx07Npevp9O+ZKYMlrh6HXItcRKQ5U5MesppndjGS49OY9h44ag7z997YYhe2Xs8ETx/v37KJC3APZ770L4Q23npCWA9P9RhNaR6oIdjrul6X4lX54CuHfvnr0wjmfPnqFQgULwOPwfLjzSdnA6Ul3pZbGNYQ1DAdL3ixQqiqtXr9oNQ5RDJX8ria2HNkVzR6K3ne22c5R6IYa4O/jQmb0oWqQYIiIi7Iahhw8folSJ0ti8f734UGXZHKWe9HLEjGORQ/cDxR1dwQKF7CqH7AVehyeKrJidO3Yif74CWLF5CS4+DpLlDH41xGWOu67zLnTNQMHMGfyqrctQIH9+eHh42Avf5nz37t2LAvkLYun6Bbj4yAUx5Er94Z6/uOS4+CQY63asQsECBbF58xZzW9rrz759+2XCsWjdv+LChbvBXUYGuRh+RA7dCxC3Tmt3rELhwkWwZvUae0HHnO/BgweRL09+LF43X76drbmecR3572r9Qcayx+exYddqFCtWDKsdAENmMCXiH6cgiqyP48eOo0rFKqhTvzaGjB6EsVNHYcyUkS5xTJ49AWOmjHCJdxk7dSSGjR0s7VSlclV4eXklIpxjz+rUqVOoVqUaatdxx9Axroah8S6BH/Zp9m22T90GdVC+XAUcPXo09oZNxLu+vr6oWrka6tSrjUGj+ruUHJo0y5UwNFKWCRs0ricYOnz4cCKiJPaszp49i6qVq8KtjjsGjRzgUhiaPNu1MDR8whDUa1AXlSpUwoEDB2JvWBe+6zREkW3w9OlTHDxwEFOnTMO0qTymO/0xf96/4G64f+f/6/TvorWH1jb79++H0Wh0uK7DMnHQ0PDjGhiaO2cu/vj9D9AGyhX6hOrb+/ftF7MBRwMRd14fOnTIpTA0e9ZstG7ZGjNnzMR0F5CrgqEp07Bnz16HxBAxfeTwEcyYPgPTZDxz/rFs3tz5+OP3lpgze46LYGg6Jk+aAsqhx48fO5oYStTyOBVRTNSaScTMSpQokYi56Vm5Yg2UL1/eYQdEV6xvV3ynatWq4cmTJ674avo7JVINVKpUSRQ6iZSdnk0i1YBOFBOpomPKZuPGjShatCjWrVsXUxQ9XK+BWGtg27ZtKFiwIFauXBlrPP2mXgMx1cDOnTsFQytWrIgpih6u10CsNUAMFS5cGEuWLIk1nn7T+WpAJ4p2brO2bduiatWqaN26tZ1LomfvrDVA7OgYctbWc4xyd+nSBVWqVEGLFi0co0B6KZyuBnr16iVyqG5d+/k7dbpKc5IC60TRjg11+fJlVKxYEe7u7nIODw+3Y2n0rJ2xBm7fvo0yZcqgdu3aqFChAsLCwpzxNfQy27EG6LpKYah06dLQ5ZAdG8NJs6Yc4rIz5RBNqXQ55KQNGUOxdaIYQ8UkRvD8+fOlc9WpU0fOs2fPToxs9TxcqAb+/fdfVK5cGTqGXKhRE/lVli1bBtq4EkPUKk6cODGRS6Bn5+w1sHr1ajOGuLoxfvx4Z38lvfwWNaATRYvKSOy/zZo1Q82aNUVA16pVC02aNEnsIuj5OXkNEDPEDgd5Nzc31K9f38nfSC9+YtdAo0aNBDvEEFc3ONDrP70G4lMDv//+ezQ5RFmk/1ynBnSiaKe29PPzQ7ly5URVT3U9j7Jly8LHx8dOJdKzdbYa8Pf3l1m8wg8H+lKlSukYcraGtGN5Q0NDweVmYoc4Uhjy9va2Y6n0rJ2pBmiqULJkSTOGiCNqqI8cOeJMr6GXNZYa0IliLJWTkLfGjh0ryzzsVJx98cxln1GjRiVktnraLlQD48aNew5D1AYNGjTIhd5Sf5WErAEuM1PuWMqhGjVq4J9//knIbPW0XagGpk6dKuYvlhiiq6V+/fq50Fu+2a+iE0U7tT+XC7nMww41bdo0OfOaQlr/6TXwohqg02fuLuQkg5iZMGGCmDFQWHM2r//0GohLDVhiaMyYMYInyiGSR37gQP/pNfCiGqD5lBq7KIfU2MZwfiNd/zl/DehE0Q5tyM/acZchO9cff/whLino4oTXXH7mVx/0n14DsdWAp6enEEJipmnTpqBrilatWpkxxC/j6D+9BmKrgWPHjpkxRDvF3r17o0OHDkIWuYN++/btsT2u39NrAKdPnzbvmCeGOnbsKOMZ5RIxtHXrVr2WXKAGdKJoh0bs2bOnaIE44+rRo4cYj/fp00fCqB3q2rWrHUqlZ+lMNdC/f3/RQlevXh3Dhg0DzwMGDJAzcdW5c2dneh29rHaoAZoo0FSBeKFM4uBOf4rUCFEO8b/+02sgthoYOHCgyCFihnghbmi2QEzx0Mey2GrPee7pRDGR24rfGqbxOJcIuexDoshBnoK6Xr16Ek7DYH3ZJ5Ebxomye/TokbhT4sDOo1OnToIhOm/nrmeG0T/nnTt3nOit9KImZg08fPhQBnVihYP8kCFDBFOUQ2r3M/3iXblyJTGLpeflRDXw7Nkzs8kUxzNqpOmqixNWYkqZwVy7ds2J3kovqq0a0ImirVpJwLBNmzaZl3s4sNNFDoU1B/g2bdqYl33WrFmTgKXQk3bmGtiyZYvsmKd9Ipd66CKHGGrcuDHatWsnGOIgv3z5cmd+Tb3sCVgDHh4eZjlEs5cGDRoIhoglaoGILS4d0sei/tNrwFYN7NixQzBC2UMMcfzif2oSOXlVGFq8eLGtx/UwJ6oBnSgmcmPRjoyzLR5//vmndCbOvNipunXrZr7XvHnzRC6Znp2z1AAFMoUxN0Jx2ZnLPdQCEVOWyz4U3vpPrwFbNUDZw5UMtTtVYYjXI0aMkCVp4ok21PpPrwFbNcDxihgiTji5UBiibKIpFbHEca1hw4a2HtfDnKgGdKKYiI0VEREhm1U466Lhr1oyZGdih2vfvr1ohXifO1f5ib//s3ceYFIUSxxXxJxzfs+EOeec4zPnrAgSFDEnxIhZMSBBSSqgoqgoiqKAIooSJMc77rg7cs451vt+NVt7fXO7F/f2bve6v293Zno6VldX/7u6usc7TwGXAkuXLtVPZDG5QAvE5gMENDyEYGbpEM0iPIRGiHPyvPMUcCmwaNGiqDYRcxeWDOEdBnx4icmsrXSwuS4vL8+N7u89BQQeYpkZOQMP2cTDeAi51LBhQ32PKZX/pF9qM40Hiklsv08++STaudAY0plYfma5kOtTTz2lM3g6H8dTtGvXLoml81mlAgU6depUgIeMf+Ah4yNm8ABJBHn79u1ToVq+jEmkAMvJ2LDCI+yYZ4Ia5iP8eY8catOmTRJL57NKBQpg1sJEFB5B6REey5BF9erVU6DIhimOXvIudSnggWIS2w5ND5ofMxZn9uX+UNkDEm0Z0avsk9g4KZIVtqxF8RD8hPCGhwjnP+mXIg2bxGKiiS6Oh0wOcUUueecp4FIAbSGyBjmDvHHHMbt3eQiNtXepSwEPFJPUdpydeM455+jOQjQ9/OyLCNbZeLZ3bEYgPPG88xSAAnxW7eyzz47yiPEK/BOLh3h/1lln+U9pefYMf7hPAAAgAElEQVSJUiA7O1tOPfXUUvHQ6aefLgMHDoym4W+qNwX4dOiZZ55ZgIcYuwwYIovcsczk0J9//lm9CZfCtfdAMUmNN2DAAOFTRyw/8/v000+lQ4cOOhOjgzHDZ5kQfwvDF1t850pSA6VANhyi7fIQfNKlSxe1b2VWzw+eCvNQ//79U6B2vojJoACAD7liMibMQ9goduzYsQAPYQLTp0+fZBTP55ECFEB58eGHH0Z5yOQNx70xlmFvD1+ZP/fwkD98OwUaN04RPVCMQ5hkeHOmom1EQF3Ps3eeAqWlAPZmTDQAit55CpSFAmyeY5BH++Odp0BZKICpFHLIf0K0LNSr2nE8UKzE9lm2bJnOvuhcAEWevfMUKC0FMCqHhxjoOUjZO0+B0lLABYoc6O6dp0BpKMC3581+2gPF0lAuNcJ6oFiJ7eSBYiUSP42y9kAxjRqzkqrigWIlET5NsvVAMU0aMk41PFCMQ5hkeHugmAwqp38eHiimfxtXdA09UKxoCqd3+h4opnf7eqBYie3rgWIlEj+NsvZAMY0as5Kq4oFiJRE+TbL1QDFNGjJONTxQjEOYZHh7oJgMKqd/Hh4opn8bV3QNPVCsaAqnd/oeKKZ3+3qgWInt64FiJRI/jbL2QDGNGrOSquKBYiURPk2y9UAxTRoyTjU8UIxDmGR4e6CYDCqnfx4eKKZ/G1d0DT1QrGgKp3f6Hiimd/t6oFiJ7euBYiUSP42y9kAxjRqzkqrigWIlET5NsvVAMU0aMk41PFCMQ5hkeHugmAwqp38eHiimfxtXdA09UKxoCqd3+h4opnf7eqBYie3rgWIlEj+NsvZAMY0as5Kq4oFiJRE+TbL1QDFNGjJONTxQjEOYZHh7oJgMKqd/Hh4opn8bV3QNPVCsaAqnd/oeKKZ3+3qgWInt64FiJRI/jbL2QDGNGrOSquKBYiURPk2y9UAxTRoyTjU8UIxDmGR4e6CYDCqnfx4eKKZ/G1d0DT1QrGgKp3f6Hiimd/t6oFiJ7euBYiUSP42y9kAxjRqzkqrigWIlET5NsvVAMU0aMk41PFCMQ5hkeHugmAwqp38eHiimfxtXdA09UKxoCqd3+h4opnf7eqBYie3rgWIlEj+NsvZAMY0as5Kq4oFiJRE+TbL1QDFNGjJONTxQjEOYZHh7oJgMKqd/Hh4opn8bV3QNPVCsaAqnd/oeKKZ3+3qgWIntu2rVKrnuuuvkpptukhtvvFF49s5ToLQUcIHiunXrShvdh/cUEBcoeh7yDFEWCtx55506lsFL3qUXBTxQFJG8vDzJyclJ6o88J0yYINdcc42CRQAjz5VRlmTX3eeXGF7Lzc2V6dOn6yAP/9xwww2SkZHheSjJfTmV+RkemjZtmvLQ9ddfL1dccYVkZmZ6HgrxEHKZvlYRrn///vLHH3+k7O/PP//UsqPsuPbaa+Wyyy4T6mT+qVy30pSd+o4YMaIiWKTS06z2QJHGPe200+Suu+6S22+/Pak/ZmCNGjWSBx54QK88J7sMPr/ktnlF0Bv+MR664447PA8luR9XRJsmO03jIa6ehwrKhNq1a8tVV10ll156acIH7Pfee0+uvPJKue+++6RBgwYp+2vYsKE8+eST8vTTT+uP51SuT2nLTr8BJDNZT0dX7YHieeedJ0OHDtW2xc5iw4YNSf2Rp/2SnbfPL7ltXVH0Nv6pDP6tqDr5dJPLm56HYtPbBn1AwODBg+0xIdepU6cK48+8efMSkp5PpHIpcPXVV0exROWWJPG5V2ug2Lp1a7n//vuVqkuXLhX/8zTwPOB5wPOA5wHjAQaHFi1ayEMPPZTw0bdu3brSsWNHTXfJkiV+/EnBMZh2w7377rvy4IMPJpxHqkqC1RYozpkzR8444wzhumLFCvEd1Q8ONjj4q+cFzwPVmwcYD9hcOHnyZDnrrLNk4cKFCR2z//77b7UHZeMQp1/48Sf1+I02AzvMnj1bTj75ZAX6CWWSKpRYtQWK2BS0bdtWm8J30tTrpH4g923mecDzQEXyAIMDtuuff/55wofss88+WwYNGiQART/+pC4fwxh16tSJaoYTzihVJMFqCRT/+ecf4UiRlStX6myuIoWNTzt1hUBxbYeAL0rI2/uiwhSVh8UvKkysdxavrPkWlWasd2Xxq4gylqUcPk769s+yti28iY1sz5499USKRI/VmDyxiRGXyD5a1vr6eKXvA7QbIP/333/XXd6J5pGqll61BIqARBrYz+ZK30G8UAlohqBgA0A8Yc/75cuX6/u1a9eWaUAwYVGawYSwDHK4RC1pkaa5RLQ/6dmZoWvWrCkTbRJRDp+G7/9hHoA36bf02TPPPFPGjh1rrJ+Q69y5c+X0008XrmUxeaJcuET2G5MXYVqU9TnR8oJyrF69WuuNcqes5UpkPGQrMozjpDAjSHdX7YDiJ598IhgR4xLJOD6t6jXoIKi//PJLeeeddxQwhtufwQbbFY6/+Omnn3TgQYCGw8V7ZhLz1FNPlco8gvQR+l27dpVbbrlFFi9enBCNOYC4VatW8tJLL2k94pW5pP4Ie857ZCcpy28I3JLG9eGqVz9LdnsbyIHXOeol0Y5NMR988IEmWxp5AB3oh+3atZOtttpKfvnll3IrOqyufPBh00031TKVl96W5mGHHaYgikRLW0+3DMRFpn311Vey0UYbKXCvbHlhdWzfvr1wDFB1cNUKKDILwOh0xowZOjiVh4FdZvb31W/wQjjsu+++Kry4d3kAvmIGPGbMGH1/ySWXqCwpDb8RAcGI9gFXkriEwXEmG3EXLFigWgu3bGW5J81DDjlENttsM02/LGm4cQDZnF9KGbH/YgB03/v76tefqkKb03/gzXHjxuk4ocyewD8mRRdffLFOthiLSltnivLaa69pv/nmm2+0ZMXJBd6Hf5Yv/riLLrpI0+Sed254C+te3ffch9+RDn374IMP1vTdMEXFDadjYUnks88+0zRHjRqlstUNm8x7yoQmmE2waJxnzpypdUz3v2oFFJnNoeHBucybTEbzeaXHIAgPHX744QUErLUtvAVQ5Es7CEw0Z2GeM/5D88iPuObHPY64F154YYG4hGGQQVihmXPj8g738MMPa9xFixZpGMIRPpyHW17SjBeONI8//njZaaedNH0b5Ky8blnwM383fe4tHMtnLNdQv27dunmgmILHgljbpssVnoWvcXyhBvvERDuWKTF5YmIU7iMloSPlefPNN7XffPfdd1q8otLhHX2O/MzR98iLd/xwfEmFvoizpW20eMiWcPqWpgYWUW2fG87SZFJ5xBFHaDA3DTM54QWrJlYWq7+FtaVmy4cJJWUcPXp0pQJFyolr3LixtGzZ0oqX9tdqAxSHDRsmF1xwgTK2DXTGnP6aHuAtme2IZCgLUDShbQIZrZ8NUAhRE5SkHwsomkRiJsvn1wCDOAYEE9IGFNGO4AiHTRQOMGh5GL3wp1wcBcLn3HAMFBaO5zBQ5J3Vgdk1n7EjbRz1sbhuONLmkGEch9x7oOj7nfFgZV/hUxyaK3Y6J9p16tRJ6tevr8mWta5ELg1QJDyHeffo0UPefvttadOmjbCRE2fgjnsXKPLJOvJgAkf/du34jUbz58+XL774Qt544w35+uuvFbiRDu8tTCygSBg+D8lpI5SHz/zhTO4R12QhtqHvv/++sPEHvx9//LHSgSLlQ6aCJWwCrxWoBn/VBigyS+zVq1cBxi9rh/Xx/ACHbCgtUCQOAvKAAw7QJYtatWqp8AMwnXLKKSrUzf6GsC5QhOdwTZo0icbhPb/zzz9fBT9gEWdAEaG/zTbbRMMzIBDGBLOl+eqrr0qNGjWi4XbddVfp27evpmVhXKBofiwDHXTQQdF4lIWZNs7AIoCTwcENd+CBB8r333+v8bxG0fcl+KkyfwAA+gWgir7EpCeRjv7Gmb2TJk0q0PdKW2fKVFKgSFgOgTYZ4V5ZMsVZPzageM899xQIv8UWW0h2dnZ0zCROly5dCoSxdAcOHKhpxgKKls/dd99dKO6RRx6pttRMMi0cB1dbunbFFpD7ytYoUkk0wyYftdLV4K9aAEUMYfmOMq60ndOHr1whXlXpDy+VBSg+88wzUSGIvRJLKvAmQnD33XdXHrVZNX7uzPXWW2/VcNggMhvH5olNK4QzO0gSePTRR6N5oMVAuL7wwgvqBzjFWR4AO+JTlgEDBqgA3GOPPdTv33//jWoNXaCIpmHixInRPNDCjBw5UurVq6d+9r1ThD+aTNLnx0YyNIl8tsz8PFD0/asq9HH6xGOPPSZvvfWW9o9E/j3++OOqfSNNgFRZ60v8kgBFtF5MEulj2B/aMi5XzGDw/+GHH6J9m1NA8Ntkk02iNnf0ffzYOINDszhkyBD1Y6OKrWSwYa9mzZrqDyBGU4lzNYo8N23aVMNg/mUrEWzIIY9zzz1X4/DHBkH8+La2lZsNg/jxqyygaAAYWcd32KubS3ugyIDI4abMjFxNSlk7q4/nBzZ4AFcWoPj888+rwAMw4tC44W688Ub1Z3nWlnARjAYUWWrmGV7G2bIQ9wDMzTffXP35M6BouzZtCZrPhZEGy2C46dOn6zO7HnFWFu6xR9xvv/3Unz8XKPKMJoC0TIvJQIJ7+eWX1Z/+hrONNexyxtkgcemll2o4DxR9f6pMmQoAgCf5jjOmSYl2bGhDg0f/MC17WetL2UoCFDEjYZmZSSFAjn5tMoB+Sb8lHXMGFK3P2iSS0xNcecFGUJ7NmRkNX63Bv1mzZvaqEFDkvW3MI5DF/fDDDzUuABDHJHXLLbfUexd4MvkkjcoAivAIMhkzIerAZtjq5vJbPU1rzmDJTjFcWTuoj+cHszAPwE/lAYp5eXnRiQtpcfwMgpAdlzaT5tmAIsLb/JWZI8KWcrGUizbAnC09s5SGgEPQAegIS5rM1nHYGPGMpgBHWAOplobZNrpAkbIQzwCmxWNAsoHoiSee0DR32WUX2WuvvfSe/CkL4Uyb4IGi71vhvpXsZ5gTjbrZzCmzJugPDR4mT/A8vF+eulGkkgBF8nEdx2Sxsa5Pnz7R1Qv6vjlWI+jPOOujyAtsEfHnyyM47jGbGT9+vPTr10+1ltBs+PDh+g4gac7VKGIXSVyWj7Hvs7j4f/zxx/quefPm0TyuvvpqvaceVhfTNFYGUIQmOMx+XLqpZzX5S2ugmJWVJeecc47O5Mx4tzwd1cf1g5rxAPKhOKCIFg0BaYKPOKZRJB2EsQlCdtARFoFugJBnFygSHzMK7G0NHBKGn83CCWM2PtwD6siLK7P4rbfeWjUNvLvttts0rqUR62qDpwsUDQzGCm9+2FzieD711FP13miHpgA7MN5RH8pl7/zV97Fk8YD1vY8++kjNJpRJE/jHJOiOO+4owPvlqRsJlQQokgeuQ4cO0SO8rF/akV4u4GF52p3MER95gSMe2n+7t3RiXf/73/9qOP5coMhRPrHCu35MTC0PO+fYgCLaWDtOK9lAkTKgjSVfdxk/WtFqcpPWQPG6666T7t27a1PS4OXppD6uH8BcHoCpmIQg7AA6TETsPbzGcpbt7MWI25wBRWx8SgMUiY/Wg/zYoHLllVcKm1CmTJmiNn9sRjEXDyjyfocddoieb2bLOS1atNCDfBkw7cdMn2Uh2wXtAkWzTwSwdu7cORqHuBwIzBI3GkMc5cWQH2f0QQNpy94eKPp+ZXyRzCt9lAkLu/CZyPCcSEffxhaXySL5JGL8oXyxgCJpuz/CcRg0fW///ffXe+yZ0SyaCcvrr78erS4AaLfddtNnKyfyDLlGGrZywD2bfQBuLL/ajzSpIzItlo0i9pDEZTmcNrZ4XG3VA3+0roS7K7Lr3OpEOThWiHfJBoqUC3fzzTfr7vEo0arZTdoCRQCinV+XTAHk80r/gQ8BhrNNKGgBAYYm2Oy9CWt2OpsrK1Dk6y4ISnYmhh2HYcdaekZ4GxjlCkAjjWuuuUaTePLJJ/WZpaSwY9mJZWfSwLlAEeHuCvRwXDSOZhO14447Rjfp0DegjdmEkYZfek7//lJVZSJ8ywYs+mmiHfZ6ZrOXqPpTRgOKHHmDs35t5h884wB+rkyw/mibXNxNO7brmXi28gZos4muyS9WI9gJjUMukKetftSuXVu/CGX5uBpFVvZceWFlJR0mncg08sIRzmyjTV7gz/nHvEsmUDQ5zpmVbBqszi4tgSKMiNE/DApD0+CJ6qw+HT+wwU/Mqnv37q3Cy5ZWXUHCLNh2D7sbVMoKFPnsF4ISeyecAThm5fjzCS7KhHvkkUfUzwQ8wAz3yiuvqL8d1stZZcS15W3KbI4NMryz3Y0uUCTM3nvvre/JH2d52A5FdlnjsBEmHZaOcFZG233pgaLvT8mWqfRfeP3XX38tYBaiDJqAPyZKrDaQj03UElFHimZAkXMMSXPWrFkFfmj3cDvvvLMCRbdP42/HVNlSL362maVRo0Ya1/722Wcf7bvUA8eucPryiy++aEH0+uyzz6o/ss2cCxTxs7TYTe062xRnQNE2w3366afRYCyDI9/IO1lAkTqTL3IN+Q6grc4uLYEizGZMm4gO6tPwg1ksHkBwcLQDAgzNGYIU2x9m1xtvvLH6285jE7Z2TAQ7BW0QIR0OlyWd8GYWBhwcRuC8Z+mYewQXSzn4oTngytISzpae8UOoY2zOlWfbeWg2SHa8DsCN43F+++03DUNYd3nqqKOO0iVv0mfwsS+rYBvJobvsGrU8bBnLNAdoIUiPM9jYOMPSEs/8MFInvVj09X6+31UED8D78CamHAZQtOMk6I+lWiZAOPp9oupAegakrP/EuhLOJmzYHtIvsfsjLF9L4UrdzZkMw/+YY44RAKOdv0r/xJm8OOmkkzQ+G1cIxwSSeCanqCsOP7NZROtopib4I2vY2LL99ttrOCawbh6HHnqo+lNGjvcizp577qnXESNGqBYzUTSNl47Ja7TCAOHq7tIOKDKb45gD+85tIjtqPKby/tVvQIOvzB6Ho2622247FWQINX5o3LDVMwFogodNK5xNBlC0ZR6AEudzITjRgjOI4Tjewgy7eQZoMVO3PFgK4hxG+7zVX3/9pfEQbIA1sw2y8CyVM0Mmfcpjwh9h6B64TR7YLeLgbRzH9wAyWZIyP8Che2g4+WD4znvT5JMfWg7iWjmoJ59IQ0vAEhphfB+qfn2oMtrc+iETuqeeekp5O5F/2OaaaUei6wfgoo/Tv7FPBmC5P/qxu6QMyDvxxBOFA/RPO+00tSemrsTnPENzrFZAC/opgI/VBMCje4g2dLNlbSaoAEZWTJhA2qfs3Ikv5cJm2XZ7s5LARJbTEACCxGWcRquLg1bkYbKPk0rYsIccpa5sfsOmmtMiyCfRtHXToxzQmsk42sSwVtboVp2uaQcUGQwZOHE0uMsA/t4PRonkAfjLwBb8BvhjKciWa/EjP8IZL9rSKyDRLQtgCecKQfVwNoHwTHzysGNrzA9/ysIPkGb5YDPEAGDaRgOnVi4rP+H5FB9pm2B0yxcuC/WxPKgveRAeZ0DU8rBlcuweCWeO+IR18/H3vo9WFA/As/Ain6oEOBnvGz8m4srGLVYF6HfW5xNVH9IzG8Ciymr5AtLCzq0z4fhZOLM3tDjIJEuLOnDvxrdwXPG3sFxxBhItrgFNNx4ywKVPvDysbAZG3TgVcU8ZWflgid87kbQCithvuRtYjHErgpF8mn5AMx4w4YYQ42cgMMx/4WeLX9zV4oXzcYUz96RjYe2Za3FlKirdospmeVHf4vIgHaOPW7ai0vfvfB9LFA8YrzLos1ufI1sS7dBSskEMZ/klqvyWDukW93PD2j1XK5PFt3f2bFfX3+7dq6VjfuFn/MNpxfKLFaaocPHCWzkScSUPAC6bBw1LJJpPUjG9tAKKzOb4lFh4JpQIBvJp+EGrKB4wIWbXosKW911p8rCwXIvKt6ThYqVR0rgWLlYa3s/3r4rkAXgPB0CsCADA0igbKNHc20SxIuvj066Y/kLbMZHlyCC+Ze9dQIG0AYrYZrCZAIdQ8B2pYjqSp6unq+cBzwOpxAOMB2izMX3ABg9Ql2jHMTt8GhNtVCrRxpe1YF+GL9gIZF+WSjSfpGp6aQEUOX4EmxMEAjMCDxQLMr8XBp4engc8D1RnHmCAZhMHu4YT7TibkA0srGSZWUV1pnWq1p3JBDbUfJubzbDe5VMgLYAisznbYepBoh8QU1VQ+XJ73vU8kFgeYDxgAwjHqrDL1jaN5Q+B5b9jmZLvFrMxw48/iW2/ZPUH2g3Hd60r4gD28nNZ5aaQ8kCRc9/sW7rJYiqfT2oKg3Rot0QMRIlIIx1o6etQPfoxQ+xVV10lffr0SfhoyzFSjRs31nSrMz+lskyh7Jz0gGYY0O9dYQqkPFDk80PM5vwGluoh9L0wDo6goSuXlRb+SBrfV8rKO6kUDwCAY7UJTVGiHTaPp5xyih4r5R4JlUo0oqx29AxHB5W27NAYc69UNfmi/JgLoHUGJIa/HJNonknV9FIaKLZu3Vq/QgHxS8vgPrwfLFONB2zg4+sKfHYLh19p6kEcTgbAHqc08XxY319SiQfoF/A4Z5vyFRH71KR2mgT93X///fp1pLL0w6pAS2jEcrl9R54vOJUGLBKfzTsAZg7CBixWhXqVpgzUAffhhx/ql2YSxBppl0zKAkUEABtY+DRQKs/mSsPUPmz1G6wRZO4PCcQXTQ4++GAVRrwzvnDD2b2944rjyyl8XxXnvuPe4tg1/N4/Vz/+S9U2h4dx9913n/DlkUS7QYMGqc0j6abaBhbr31wBel27dtWvJvFZTjRr1uZuOLt336GJ5BOIfHGpf//+BeISzuK4V4tfFa6UC+zABpazzjpLNcOJ5pN0SS9lgSLfgORTQjgavCowni+DH0gTyQPwNYMQZhUIdPsqA58ARKtovG/8jzYADQFhieMuB5EObvPNN9fPUnFvA5zlg+DHVsfy4r2lnch6+bR8P6lIHoBn4WPAy3nnnad8n+i/K664QtNPJZMn68uAI+jDD/fVV18p2OOTfQYUCYv84Bl5QFjkA+3GO344Nn4AFAGM9pUVyweNrskjrlVxFYM68DnDVq1aaX38X2wKpCRQxCaRHWw4G+wqUvD4tP3AlmweQNia7RBG1h06dBC+rYzbZpttCgBFBD+O5Z/vvvtOw/bt21cFM0KesiOk+e24447CwfTmzzsGAZ5ZkuZ70nx32mx1vLbe836yeb88+dFvADjw+iWXXKIARjtHAv+weWzYsKGmWJ6yJjOugTcKjQaNz9wCEAG6P//8cyGgSDhAIlrGTz/9VPhuNJ8mxDHm8gP8GVAkHDSH9rzD8UlQvpaG7OK78zzjklnveHlBD+qOTIVPvCuaAikJFNnAwsfRYVTrAPEYwvv7gS7VeACeBiRyltcee+yhQpxZO7977rlHdtttt+jyMYKZsJhhWBj3yqkAOIS468+9zaJZRtt6660LvT/ooINU8JNHqtHQl7f69nv4/b333pNHHnmk6NGvDG/5rjlaytzcXLXnS6Xxh+q+/vrrhfo5dEIemEaRcLYcHZYZpqAhzHXXXVcoLcZkHN9JDsfl+f3339f3VaF/UhBsvU1GasH8X0wKpBxQZHZy7733amWqArP5MlTfAaki2x4G33fffVXYoiVEs5eVlSU77LCD+h177LHRDs3OS4Qws34+IUbYX3/9Vf322msvDYcfy0NoI4mLdmD27NkKRom77bbbSmZmpmoDSINv1uJvg20qDYgV2S4+7ard3+FzvrwCmJs3b160jyTq5tFHH5W3335bk0sVXqDvsmpgmkMOlJ41a5YCXQ4gp5/zAyjizO4QW+a5c+eqTICWHCpOOEAkjv0BL774ovp98cUX0U/eNWnSRP2ee+453UzEJBV5Y5NR0sSvsugHPXCdO3dWQKsP/q9ICqQUUGQA4xNMkydP9htYllZtgV1ZQiDV80WIseQzYMAAFbavvPKKdmAEKw5wh7A+5phj9JnlkwMPPFBuu+02fUa7SHwcs3/CEpcBFLf99tvr0jP3zP6x4wI89uzZU9/bcjcPxDUNggeKvr+lQt+Cb++++25dLlWGTuAfYIfxh76Elj2V+gRkOPHEE7VP0++RGyZTrr/+evU3oNiuXTvZYostdJwlnskEkz1PP/10lKqYqiAnMFsxh5bu8MMP10dkkcU3beb48eNLtbs6kXxHm1FvsAS74VmG9654CqQUUES7wZICLpU6aSIZ3aeV3gM2fI1Dc4EAxu6QTSr42zvXRhF+cB32QXzS8vvvv5c999xT08AP4Ygd4nbbbRfdzEJ6BiotDZa7J0yYIO+++67GvfDCC/UVYT3vpTfvpXL7wp/w948//qiaL+PnRF4BQGjqySeV+gP9H7DmTvxoa+qA+/rrr/UdQNFAnUs3NICjRo3S7x+ThgsUP/74Y407ZMgQjWtpWnxkF2AMO2s0mcQfO3ashq0sfqNsfMvZNMNWVn+NT4GUAYqjR49Wo1MGPG8z5QesyhIyFZ2vCVrMKxCqPMPzrmBnSejQQw/VXo12gx+2i+yGJo79ttxyS9l44421v7hAEXtGHGni0AZwPITF44pGgat9qYByVHTdffq+X5eFB+BNs9Vlo1ZGRobydSL/2PxhWvuylLEy40Ab+j/9+fbbb1eyQDN+TBQBebxjk6hNHFmmRs7gbz8ze2Fp2VwYKNpGlueff1522WWXaFzSYCMd18oCitQXTSpL6+eff75VwV9LQIGUAYocR8DBoKk2m6tMAeHzTr2BF2GG4zBfhCoavjBQPOyww6JAkbD77LOPhr3pppt0yc2WkFiCIw0bKEyjaECRQYFBlTD8sEvEHtIG2po1a0YFKuXy/JR6/FQd2sz6DOCkadOmJRj2ShcEGrLkjO2jafdTia70f8pNH7/xxhu18pQfuiED2MzGO4Airnv37voMsHvttdfULAVzL9NKFqBYCUcAACAASURBVKVRJP4NN9yg8Zl8cpB1v379NB9OUyCfygKKgFhsNW+55Ra119TK+r8SUSAlgCLb+G0mlEod1JfVD6yl5QEb9PiGLEK1d+/e2pENqPGAvx2a/fvvv+uzbTrhPbNm3AknnKDvAJquRvH000/X9/zRr0jPbIwAkwhTyo2/1yh6Hi4tDyczvIEd7AcBJrbrNsrgCbh54YUXxGyFk1m3ROVlWj5WCf7zn/8oRUjbZA02ifR1m2AeddRR+mzL0MgD3JgxY9TfBeOmUeQ4LcKxTE1a2P+ZM3lk5jSVYaNIXZFt33zzjdxxxx1WNH8tIQWqPFBkxnPSSScVsNVKVAfy6fhBsKrxAALNltEQuByP47q2bduqILZdz9giEs4GMgvL+WD480N4szyNIGfnIUbt5v73v/9pmClTppiXXvmiBXEBmzjKVdVo5ctTvfsvPAkIAgCgKeOsvkQ7TgJgB7X1y1TkOcqOa9asmfZpbBJdx3Fb9HUDivvtt58+u2G4tyO4+NiFOQOZI0aMUK+pU6dqXJtgWjjTaJLPn3/+qZPZZMkUazvk6rnnnqunO1i5/LVkFKjyQBE1N+pvXLIYKxWFgS9z+gyaJtjtLDOOrnnmmWf083sI2ho1aghnHOLQoODHr1GjRtKyZUu15eWZJWquEydOjBqpm90Rx+awFMRyE2FYZn7rrbfk1VdfVXDKM6ASuyTf99KHt9JJTlg/gY85t68iHOYc3377bcr3AdMq1qpVS/s7ZyAytmLDzOY4ZAAnLeCaN2+uz8iI1q1bazjkgW2Osw1uKHFMfiCTTIvIiQykd/XVV+vX0+688059Pvroo/XKMV6A+2SN58Ynzz77rB7nUxF8ku5pVmmgiC0DszmWzWD0dBJyvi5+8C2KB0y4IbwRwJtttpl+35ndhyz9PP744yqbENZoDDnjDIEPqORA+uzsbJk0aZLaVnGgLOFYSuIsRnYfIvQBhTh2cqJl5BvSe++9t+64xp9BAs0A5fT9z/NrUfya7Hf0D8aF+fPnR+0HlZkT+NejRw8BKOKSXb9E5we9WFXAARB32mkn3fzG5JKJJJs7sE22MCwpAyoBiFw52gb34IMPCsCP5WTKiGNpniO6OBLHvr7SoEED2XnnnXVTHJpIZAyOr6Bgt8jqBmVKdD3D6ZEH2kzsSykH996VngJVGiiyy4yltWTOPsKM5p/9AFlZPICQC9tcuc+852fC3e3+gEKzMeJqYc2PsCasuYYdyzTmKqv+Pl/f9+LxAPyMe/jhh1UDZryayCsTKiZmTLLIL15ZUsWfOrj92mhlNoSAKKunvXOvppXEz62zG4b0Y+WBjAHY40zuuGlU5D15soGlW7dublH9fSkoUGWBIgCRxgUkViQT+bT9YFSVecAEt5WRZ/uF/cLPFs5Nw713w9s9VzderPBuWH/v+0+yeQCeBNxwzAm7kSvCoUHjFABcOvUB69vWZvZsV9ff7rkaDcLh3HcWPhzGnt2rha3IK/nhOC2FU1O8KzsFqiRQRGty9tlnR7fR0+AVyVA+bT/YeR7wPOB5oOrzAGOBabYwi2BjRKIdh9yzgRJNPdoxP/5Ufb4I913ajLaDV1hWZ1e8d2WnQJUEitg88J1IHA3uf54Gngc8D3ge8DwAD+BatWol9erV0/tE/3H+KN8uxnmeS12eo/3YoOceEJ5oXqku6VU5oMgxHWxhx14CzaLZWvlrYHPm6eDp4HnA80B15QFMkTj8mTMTFy5cmPBxuk+fPrpZI+EJ+wSTTgE0w2gTK4JPkl6ZSs6wygFFjjl4+eWX9QBgdnz6n6eB5wHPA54HPA/AA9gl1qlTR9q3b5/woRO7R45+4ZzAl156SThOxf9SkwYvvviingTx6aefJpxPqmOCVQooLl68WGrXri1169bVWR3b8P3P08DzgOcBzwOeB+ABTsJgp3NFOLSVfOmIryH16tXL/1KcBmiHvUsMBaoUUExMlXwqngKeAp4CngKeAp4CngKeAomggAeKiaCiT8NTwFPAU8BTwFPAU8BTIA0p4IFiGjaqr5KngKeAp4CngKeAp4CnQCIo4IFiIqjo0/AU8BTwFPAU8BTwFPAUSEMKeKCYho3qq+Qp4CngKeAp4CngKeApkAgKeKCYCCr6NDwFPAU8BTwFPAU8BTwF0pACHiimYaP6KnkKeAp4CngKeAp4CngKJIICHigmgoo+DU8BTwFPAU8BTwFPAU+BNKSAB4pp2Ki+Sp4CngKeAp4CngKeAp4CiaCAB4qJoKJPw1PAU8BTwFPAU8BTwFMgDSnggWIaNqqvkqeAp4CngKeAp4CngKdAIijggWIpqciH4/nQeM+ePePGzM7OllatWklubm7cMNX1xa+//ip33XWXrF+/PqkkGDhwoH5HfP78+UnN12fmKeApkDwK/PXXX/LMM8/IDTfcIA0aNJBPPvkkeZknKaeXX35Z3nrrrSTlFmSzYsUKadSokXz//fdJzddnVjUo4IFiKdthwYIFstFGG0mtWrXixmzZsqWGAVB6V5ACjRs3VtqsWrWq4IsKfvrggw8034kTJ1ZwTj55TwFPgcqgwPHHH699HPns/jbZZBNhopgubuONN5atttoqqdVhgg1N69atm9R8fWZVgwIeKJayHRYvXqwd5oQTTogbs3379hrmyy+/jBumur7IysqSb775JunV/+ijj7RNJk2aVKK8TznlFJk+fXqJwqZyoHTSsKLBP/TQQ2XOnDkV2iR//vmnXHvttRWah0+8dBS46aabtH+//fbbBSICEA00IrvTwfXr108GDBiQ1KosXLhQ6fjAAw+UKF9W1S644AIN+9RTT8n1119fonjlCYSMr127dnmSqNC4b775prz++usVmkdFJe6BYikpm0yguHr1alm3bl0pS5iY4MnW+FHqNWvWCEv7xTnClYQubjgD7yUBivfff39UyBVXllR9b3w8dOjQVK1CoXKPHTtWB7MpU6YUepdoD8DHF198kehkfXploACyivY444wzYsZmYsr7p59+Oub7ivBEPiF/ku2gxYYNG4rNduXKlcWGIQBjEM5oXFKgCL27d++ucc877zzZcccd9b4i/+644w7ZbLPNKjKLcqcNXfLy8sqdTrIT8ECxlBS3Aba0GsUbb7xRHnnkERk+fLjstNNOKrg233xzeeihhwqVABsbGMp+e++9t/Tv3z8ajjR499VXX0X9uLFlcZa+zQF6brvtNl162XXXXTXedtttJ6+++qoFiV4XLVokF154YTRf8rjnnnui7+3m/PPPF/J49tlnNex+++0nl112mWpzLIx7vfjii+Wcc85Rr7Zt28ouu+xSABAyE4QWVl+WVWLZFjVr1kxYdrFwhxxyiGRkZLhZ6f0///wje+yxRzTcVVddpXalxCsOKM6cOVPjzZgxo1C66eQxe/ZsrWcyQFWy6AYv0MbTpk2r8Cz79u2reVV4Rj6DYimwfPlybYsjjzwyZlhATtOmTeX333+Pvgc0wivITNe999576o8sxA0ZMkT++9//Cishrmz83//+J2jZwg45veeee2oapI9c69ChQ4FgrVu3lpNPPlnHApbFCXfddddpPrFsAJmQ7LzzzlFZd8UVV2h4N1FAEunY7+ijj5awmQ10uvzyy6NhCHvrrbcWshcHHJKHpcV4hRad55IARcx83KXxSy65ROWxW96KuGdZfIcddqiIpBOW5hNPPCGHHXZYwtJLVkIeKJaS0mUFijDH9ttvr53twQcfFAATy5vhzmcg8bnnnhMMs3/44Yco6Jk8ebKW9t9//9V4n332WYHSG1B85513ov6o/63DY4yMsMTQGz/X3gRNnoEwlm/IgzIQ7sQTT4ymxw1Az0AnAPGII44QykJYm0VaBLNtefTRR9ULgU040xx+/fXX+nz77bdr2X777bcoXdwNQyYICffHH3/o8vU222yjcV1Qh3Akfd4BpEnvgAMOUD/8iwOKgGqEeNgBIOvXry+nnnqqXotLh/hz584VgPpxxx2noBoNA3RYunRpOPlSPSPIX3nlFQGwo0WhXWfNmhVNA75BWLuO5ViA9pIlS7RctD30YKnG1YyxCevss89WYM99LEebXXTRRZr/+++/XyAI9G7Tpo1g/E4elJF8Y7nmzZtr+Rl03TJYWAz2TzvtNM0r1nt4C/rCn88//7wAeqmTCxSp78MPP6ztduedd8ro0aMteb0+9thjMnXqVGnSpIn2izFjxqg/vAdvQ18GR8KEHXmF6x8O45+TQwH4kfaAl+B/wGFRDhlM+DBQZHkQf+Q8ziYE+CHnfv75Z0G+8oy8dB3v8EfefPvttwquAEn4uf3RQCrxsatkot2nTx8Nx5gQdgcffLBsscUWUe/dd99d45jHpZdeqnGR2yy1f/755/pMvlYPwm655Zbq/9prrykARoYQhgm36w488ED1R1YPHjxY5R3h+EG34hzhOnfuHA0GDbDpR2YyHlHeWGZZmPqgODnzzDMFLSSKiLB2lHalPyMXMDdwFSixgCL9mjbFMUaNHDlS607fvvnmm+Xvv/+OlpMbZBKTg44dOyqo/vHHH/W9TTYYGwDRtLXrunXrJoMGDVL5AhhnjGWcCjtkP/QZNmxY+FWVfvZAsZTNU1agaIbWaLtcxwxo0003jXoxc0QguY5BinBdu3ZVb5YLYbbw4Gl2JMyKzdkM0vXj3d13361pjBs3ToPee++9+jxixAiLqlfyIC8AgLn//Oc/6scs23WEO/bYY10vHYDxN3u/F154QePa0jHaRt6HHVpUQDNu1KhRGqZevXrhYOrPIGGOjkx6zJ5dx9IH/sUBPMK4GlnSsNk0GguEDrN1wmErFM9RX8IAqIljwhc/F9jGi1+UP2kwcLz44ovy5JNPaj74Ac5wDRs2VD83DfiOMLQ3QApwxfNJJ52koJOwBx10kPox60X7zXv40XU2KAHQsD0izG677RYNYoMg/gDOq6++WsMwGLrO6MFgAFglPBudcABqNBL48R6Azj2DgzmjL3lDXwYi00obUMROinj77ruvMDiee+65+uz2G97Da1z50aZM4rhHm85gatrpMO/Qh/baay8rkr9WIgWQJ8ccc0y0HWk/7FXh0ZycnEIlA2wQJqwVBGzhbwALfuAZsOY6QAH+9ANz2267rfKgPdsV0ENYAz1Mani+7777LIheOQ0Cf7dM8DJ+TNrNIX/d8vA+LBtRCLByRDlxTIgIF7ZtBAjh/91332k4rjyzyuM6Wz0qDigCvIhvS9ak4WonH3/8cbnllls0DG1gDgBHPBQq9FWbyNJ3zc2bN0/DIBvol6bhZeKKCwNFJuikaZpV5JuNA/DFlVdeqe/btWtnWQggHJlHPH6UF6UG9+TLpNfKD1A1hzyF3oSD1qwgcv/uu+9akOgVWYjio7Qu2SeFuOUrPEK7b/19IQqUFSgCLmLZT9jAaxkxm4LBMHpleTCWKw1QDKdv6THokQ+dEsc9gjaW4x3pmGPgZDkm7Kxzu/YvAJrDDz88GjQMFG1mT1zTmEYDR25MyJkW0n1v2lEGCn6U1QWOFhYwwbvwYG/vuRqwMMFi74gH4HEdghn/eO6ss85SweG+ZwmcOPHa1Q0b7960vOH3pIsWA2cgzw1jWmjs+HCmfTZQhaaTNMJ2VfgZgGPWzHPYxgY/juzAceX5l19+0Wf+TGtsWjl4mzAGbAnDchx+tCEbRbh3nfEr2krc6aefXgCg4mcDktUJ3mOC5jorn9WTfACKrsMPnnMdA0hYw9qjRw8tpzuwu3H8ffIpAH+jOUY+0Y72Y6XAdaUFii4/WzpMUsz2zsyBsIUOO9tQY2DM+nAYwNpqSIsWLaJJoBGjDu6KQRgobr311hoGwGNL5tEEIjeAHCZnsRzjEv0Jh6aN/MKylhUJ/ItbeqbfhDWtBsis/5MPJ4KQnikQ0NyzCuQ607La5jQDhm4Ym4jih7LDJra2WudqljEXI093om6g3fowE07CuGMYYJXld9eZ8qB3797qbSt3jCHm0E6SFqtRrmPMA1SW1iEj69SpU9poCQlfUBonJMn0ToSOSOOHl2PdWqO2JoyruUAbBcAKu7BGjUHYlghIg8EOEOXa4pUGKNK53FmZ5W92PQaAyKuon9tREJAswYadCTrTXqLCJ03XljIMFBFI7tIw4QEKdERztqxUVPkAEqZlMmBj8bnaslBRQJHlEvJgudKcaTN79eqlAsby4ZmwtlRp4e3KOwPh5kdY/MsDFC0troAd+AWASLoGokoCFIlHnAkTJmiSACE0H2EHuIUfcQh8+IB8oQM/wB6DzD777KNhXnrpJU3XTccAuC37MmliSTrsDDhSLuoQdkcddVRU204YwL/rDAwzqBp/Y+vKM2Vladra08waSAcNqutM+0/7GX3c93ZPfYjP8px3VY8CaLUA8/Am7eSCnJICRTRzxI11Jq7JJWqOjCNcUT+0WDgDiqa1dCmHhhrNlzlsGNH6uy4MFNESuvkCluBpF1y672Pdo43HoYXFRCqWI56rQY0VBlkQthVljHM1oBaP9FxQbP7IAVarDMTZ2Ef4cJ9n/DAwyOQdOeaCR0uTKyt1jIeuAxCSLpNgHBOM8PI/7z/88EM3mt4D9mw3NwqecDsRiLhhG1U7Pq+0GkImIfBcvfoNCpWloj08UCwlhWncGjVqFLARCSdhyxc2GPGezuMu0VmcMFA0f0AmGhKza4ThTMUeDygaiDWgRlp0jFgzSQZ70rRjPrhn2fbjjz/WzksH5odtDZ3Elr1Jk3rE6hS8Y6kVQYbDnpB0XRcGivaOmRkqfdcQHBsZnC0ZUg46mZWNK37MtOjwzFjJL9byiM0AAQzxnGm6TPAQDhtR0qxZs6ZeuednRug2o3TTRNARJmyHw6wVf1eAu/FKes9GJFuaJT1bMrYllNIARRPC1CcWwDZhTdkQtOQH/3O1H3GNt2MBRWbwhDVtJmATrU88R1gDvW4Ylmuwj8URpkuXLu5rtf3EH9tQA+WYdeBnP7PDZTJn6bjLevjBS6YhIB71ZRk6rGUxm8iwvVKBQvmHCqcAQB3NW9jcxM0YOYrmzVw8oGhaIANxBhRdbZilYWY9PJuJDukik1wZhSaan9mlGVA0LZalx9XsH9Gi2eassBwJA0XiwZuYTCDvXZ63POFjJlqx5DvLzNaXzJbeLZPd0w+KA4qUjdUU1zEpNIWE60+ZTENG+Tkg3fopVzNPYaK5bNkyfefaPrppcc9SPvFs6di11ec9IDhW+YmDnMNRfncSi6zmvWt6pQFFBA0ly9s4FCdhrTX+xDX7fItnKzPUqTTug5at5dvOLaXp443l3nrJBYsFR/HSlLoah2VQhAHiOVs6s0GYcCUFiiybMdC5DuNs8jONpGlOwgJk/PjxGs4FihgSx1ryZtmDNG3JkPt4WlKWhF3wVBRQtCVEBlEEi2vHQZ0MKNpsCruT8JIzs0mz9yCO2U+yLBl2LKGawLXZYaylZwQkdSwKKLJ5gjCkY86WP9AoYGMEHVxaWLjwlXTC9ikGXsqjUbRleNLOzMyMloX8GChwDFYASdfZ0piBtbBGEfsqjLvDDsFK2jgEo/GI0cG1RSIMdpMW3tIyOyvLm6VebHjCzmxeiW986YZhUmW2joQJLwUbcGOQNf6mr+CsvG563JOOTUjC7wAeGKQDEgmHRtV1lh884l3lUYCJE+3DV5/iOcALkwSTO2yaII67ekBcm2QZUDQbRQNcbvrIQVtCNEAZ67QG4sAr5ooCijaZBGzGM28JA0VkUxh02OZClA046gpQiuUom8k0U1yEJ0WmoWfjXFFu//33L3RMEZMuJvthR5ns9A2zLWdibtpbtPmEQS5QHu7D2jnkjy33mi0z+TDpJbyZoeAXCyiauZK1G7R1y2pjimkc3TqwEmZjDfbT8YCiu5GJ+AYU3XHGTTfefYsPWknXDu+KrM+Txg3rSv0GBW1c48VLhH98tJOI1NM0DZu5MGsMOwYpGNSdvRKmJEDROqMNxm7aLJ/YphcDHCxJu85U7i5QZBcg5QkPZraZxezxsBEhnG1usXTNaBttn7migCJhSMe0gAAU1xlQNMNuNERopMIOlT7p4NDMcs8mjbDDn58NAGg6eQ4DGJY+8C8KKNoSrgvUzZYvPDtlUKI9wvYnVj60fOHdhAbyygMUWQYOz9hNw2z2UWb3aWXhauYQBtYQxtDDBCmzaJ7DjvxMe2xamHAYZtO2BFMUUIRvcYRFQ+s6Kw+TK9rKXX6zcJTPjMDpXwwurrMlHbNBIry73EhY0idv27RFmDBQRPsUPhQ+1qANeCC+1csti79PHgWMd5gQx5pMMhGlndzNULY5A/MY12HqQ9gwUAxPEqzt7XgzAxS2hOumaRMNA7JFAUXisQkM/kcuxpq8uUARWUJ5YwE4gLEto5r20yZOVj7kPfGxTcTZZhaAqusAaISLtVrjhgM4UT7X2Rjk+tnElauBtfBh1DYBsDGKlYiw7DNATNoARddEivK69vGYvIRX11gRIpxNUsNAkXR5j8LFdYwv+FuZOf6NZW/X2diJuYvrbMXR9SvJPUDxs7bNRRZNEFkyQRo1rCt1k7QMXXhkKEmJfZjo8hsbQGAWliyuueYaZR4YKLwjltmMGT675AsP0DYgMcAxqDGw2yYRGMyc7cRkMwd5w+DsjCZv147DBAT+HOWBPYvt2rJBlzQNEBEOTSXC1wyO8XOXdaiHqdytPO7V8mQZOexMQJtAtx2mLJkgtKizHdtg4IM0DMgyU0Rw0AlZJqFsnTp1imZj9cAwGm0QWjcDrYS12Wc0gnMDQCQMGlzX2W5EhCUCwjYxmBAmrB1JYUtUNhsG9DNLdpdVDCiaIHHzY+k+DG7cspjQ5VgIBjMTlJTbwKzZTwLKobOVlzBmJ2j2nLTVTz/9pICX9wymAC0ApG2sMkFtyz8IROiI9sO+iIFGBWeDoFtmMwkwgWnaTNqcOtDmaGZsB7HtnKT9OQIHIW7LUPYlGdsgwKQNbYiBfOpgbQAA5Jn+Cf9aHNNKUkbehw9ipq/iT51I27RFtKHrmJARznjZfefvk0sBtMu0BT/ACjKQdufe/N1JHfxr/pyuYLxiZhy2KcQ0ioRF1tJX3bzcWposY6IDIEMGmGbLzpElvB2BhqyK5Qy8kKf1PTccchXNnTlkDGGpM/KOfmbjCrIAZ0oIwiEvke+YE/HMz1ZlCGu7hRlX0DbaSgvhYgFSKwdX7HoJ5zrbIIMcJh9MBQjjjiEoDAD65IeyxcYJwtmkjjMmeaatMO8xOtk4xukJroLGdlIznuBMa8mZlcgyA6vuUjO0Re65Dm0j+aJtRgNNeSgrk2hTeCDLCMOmT8LY6p5Nst30kLlhJYL7Pt59FCguGCcbZgwXWZYpDzdqIHfVrvgNLgVbNF4JvX8hCgAYDBDBIPZjkIll3I5K3+0YliDq6vAhoS6osHTDZ9HRUdz8SR9H/mZvwjOMy8zUvk5g6YU3WhAWcBA+YgK7NFfAEo562JIGz2FnxtWxNK5mg2O7TomLgLNy2ZVZeNgBoOw9V5a2Y31PGwEBmLGwgGg7BqIojSL5oS2NNWt2j30hXQ6qdZ1thHGXmBAWaNsIzyTCNtSYjaIdWeNOKlgCttm9m757z6BjdUPgkCdLuRyVY86W1iycCUVXY2yacZuls1xrhv/EYzLihidtQC6zdEsXDYrttuY9S+LurB4/s1F0N4aQLrS2dJgIuDwBD5mtEWHgQ9MUWh0BwHYkDlpC7IgQ3raTknA2e7d86A/ushpaYXcCZmnbDmqLF4snGKBt6cni+WvlUYAJj4EcazeuTK5sguGWjskW/YcwyEj4gD6L7LCzTg0oIj9tVYLwgAt38mzpEs60klaGsD0u+QA0XHBm8e3KsmZ4U4i9Q6a4wBN/m0BanvQLJpGuYyJsKy4Wjl2+TNzCzlV6EJZ6IQ9cGROOw7P1ddPQ4YespP+YCRHphccPJncG0nlPv2KMYzxzTaxM42nlNxtH8mEDj6tBxA/TGVu9wHSG+tvpE6SBDHQdtI2lxWW5mDazfNH6ukvHtAd5E9fCuGVz8+B9LJnjhol17wLFtdOGyoaZI0QWT5BHHmggte8puLoYK355/DxQLA/1RHTJE4DFIBu2dylP0gyadOBYndhN12Y0rp97z8AIY5ozbZY9x7oym2bG7S7BxgqXaD8AD3Q0e5lY6aO9YTYctmuMFZb0ShLOjYvWMHy8g/uemWhxNCc8BuLUxXW2MaYoPkETEcvw202He/jDltvD7+yZyUxZ2pABMNYgaOlyRUgiyMvrGJApZzwHvYvLx5YJ46WBP+m4ALGosPaO8PHSNpup8IG9FtdfK48C8CaTBfq+O5jHK1FR8saAItpoHP2puL5BOAATk9Ki0o5XnrL6Uy5kjmnU46UDTyPfixsLALIAvpLIOzcvJnWm5XP9uac9itLAl4S2pIMMLW252Mxj5geMccXJz3DZeUZexWpTJtvU28oWT9aYNtUmIrHyiOfnAsV104ZKFCwumSAPP1C/Qje45COIeKXz/ilNgTBQTOnKJKnwAGs786ysWZom0UAhm3ZYFomnJbB8OMoIrYZ3VZsCLB+i9fEuvSlgQNH9Akh617j8tbNl19ICufLnXHQKyN6wXXPRMUr+luVqtJ/FOY7eCpuwFBfH3oeB4rrpwyJgcbjI0kzd4FK3gnZDe6BorZCmV1TkrkYxTauZ0GrZMnV5EmVmjM0dtLcfZ3TFm2mWJy8fN7kUQCNCm4ZNMpJbCp9bMihgpx7wKT/vSk4BNn+wfF2VHCZIJQFzZSkzJluxbPLdtOxcYdevNPexgGIULGKzuDRDHrqvntSpW/gLZqXJJ1ZYDxRjUSWN/NgJGz7HKY2qV2FVwXAao+ryOpagMC633cXlTc/Hr3wKsMHA3YBU+SXyJagoCrBEi41hcbbNFZV/KqfLBj/v8inAprjSmkLlxxaJuwRS9AAAIABJREFUBxQLgsVMebRxA7nr7sI2/m5apb33QLG0FPPhPQU8BTwFPAU8BTwFPAWSSIGigGIULEY2uDzauKHUrnNvwkrngWLCSOkT8hTwFPAU8BTwFPAU8BRIPAWKA4r5YDFYhn6kUX1JlM2iB4qJb0+foqeAp4CngKeAp4CngKdAwijQqvWHwYHbC8YJu54BhrF+uhsam8UlGfLQ/fWkzr31y12GjTh7rFOnLpHfZ9Kpk/95Gnge8DzgecDzQOJ5gEOe//lnoPw14G//8zTwPFBCHvh36DB5sklT6dLuHZFigGJUs6gbXAKbxbvLec7iRjfddLO89erT8lHLZtKmxUv+52ngecDzgOcBzwMJ54H2bV7Rw47vues2eeKhhvJo4/r+52ngeaCEPPDYg/fJ8N+/FZkzqkiNomkZo5rFpRPk8Qfvkzvvrl1mzeJGt99xl8yd9gdfQRXZMM7/PA08D3ge8DzgeSDxPCCT9MsVf//0ucjaHF0aY3nM/zwNPA+UkAfmjNKzEw0MFncNDuUerl9weezBhnJPGY/O2ei22++SKVm/iiwnsX/9z9PA84DnAc8DngcSywNLhoqsHys33nST9P3mY5H5o3XAYyDzP08DzwMl54HiwGH4vYJF/TZ0hjzUqJ7ULYPN4ka3G1BcMUKEzux/ngaeBzwPeB7wPJBoHigEFGMb44cHOv/s6eR5oHw8EAWLSzMVLNa+p3RH53igmGhh6NPzA6znAc8DngcK84AHijF3qXoQVD4Q5OlXMvoBFterZjFTHn+wodxxV8ltFj1Q9AK9sED3NPE08TzgeSDRPOCBogeKcY508WCvZGCvvHRSzeLMwGbx8YfY4FKyL7iUDyiW1o4nEYLHzTOR6SUircpMw+hSmWUI521lKuk1HL8sz25eZYkfjmPpLfk39YCDld29hutXmc9uuUpyn4iyuvkkMr1EpJXsNFxa2H1FliGJQJEBsaS/ddPjn0lXmoHZ8itNnOoe1tMsOQDR5TNovgGwuCRD2OBSuwQbXMoOFBEsa8eKyIyS/VaNDoyjyyOIyJPd2TJNZNmwBKWXJSJ5qQcCjI7QhHuZFNDGnu19ZV2VP8aUjDfgoVWjEtSe8MfU/M1Z5am/8tvECH8ME0kVsGg8IBNEZEqkDaaJSLbIssimtfLQJRFxKePq0SXnjzWJkh/QBP4YkSB+M/5IIftuaL8UmZEZyFKV4ZNF1o0VWUw9KmhSlCSgyEAoy5DrM0vwmy4bZgyX8oJFzXN1rsi6KbJ2amKApzu4J+KeMpKOzBpZ7voWVR43H7uPF17W5okszyrVTt54aXn/koNO2gW+l6WZwhdcitsNXTagiKBZO0Z6/fihbLTRRgV+m25aUzatWbOAH2GGD/wiAgbKKFDJUzLljVcf1rR1l/ZSBu/ypJcj99e/SY456uAAZJU1rcqKB03IW6bIFf87W15q2kBkw/iy0ySR9Vg9Wvr80LoQHyh/bFqYPwYP/Crgj7KWQfkjQ5q/9ohsvvmmMn/WgAAUlSu9ifJwo1vltJOOkvUrR4mUh9/KWo7SxoMOCgJy5NO2zeTQg/fTNthyyy2kUb0bRajH6gSA8tKWKxx+1Sjp369Lifnj9z4dA2AZTqc0z+vHSYv3m8gWm20qM6f8HoDF0sR3w0bk0ROP3C1HHX6QyPpxVaPfuWWMdU+5mSxIlrz6QiPZeafttQ2OOuIg6de7QzDh1HgVABaTARQBicuzpHHDOwrx1mabbSqbbFKjkP+qWSMjYLHkA60LShh0ZeUkueT8M2T//+wjItOrHPChjOsBiSsnSe6wXyoMLBowlFU5khfJx/xcmmlZlmTKcUceKg/Wu01k2cQqRzO3vOl4T7sYWAzOWYy/DF12oLh6tPzTv4sccMC+cuIJR+rvlJOO1k648cYby6mnHBPxP0LDZIz8TmTlyMiMdWhwRWjxCws083ffcS9Z8tTjdTUP1QigVSxJ3HAYnjW9aXLO2ScE6aGRC4dzy6EzbQeUumWLFS+Wn6UXfhd+Lkk4wnCkkUyR119+UOvw6AO3B1qCcHqV8bxqlAz5o7PUOng/OeH4I5QXTj054A8mDmecfryceELg/9//7C3jRv0Q8Idb1lg0NtrY1cLzLBPlmcfrKC1mT/8zoI+9t2s4nvmHr5perlx0/sma3oZVYwoDRUtLrw5vkBZ+4TTt2eLZs4U3/6LiumHd+HYPn0qeXHbR6Vruo46sJc83qS83XnuRPm+5xWayjn5jx2FZvGRfV46U4f9+I/v9d++o/Dj9tOO0jPDHqaccq/7wTq1a+8nQgV+Wjz+o34bx8uILjTSPqTl9RWKd9GBtUBw9CCeT5H+Xna3pwXsK0N14llas9rR3bni7Lyp8UfHc+LHSsPeSLeecebyW+5yzT5TXX3xADth/H31u0fzJiAwpgn8tndJekwUUl2TK2y8/IYcefICccNyRctLxR8kxRx6q9dthu23l1JOPlROOO0KOP+ZwOfaoQ2V1CCgygMYCN+ZvVwMPPMuaPDnskAMivDCrUHyLw9XilfRaXFx7H04v7C+LJshNV14ke+22i5Y31mfgLA7X4tKz9xbHnmVFtpxx8jEKAmV1biFaEC4ArTlKr/POOlnC4SxNu1raXGP5hd+7z26c4uK6YcNppOMz9AjAYoY8+fD9cvudsTe4lA0oIiAQRLp0NDVYymE5WObITjtsKwfut5eILIwsa0Teo8kgHrNZrgxUa8bkP5tgQ3izBIJmjPdocUw4LhsmaxYPlcXzhuTHc9MkLEtUlKu4JTbSlGly4QWnRjo3B8BGBnveUb61Y4JysMTuDqxobAC98fKIvosAYtLFD60DPzcty5M0DfgSljrE02BRPsomGVKvznVafgbXJx++q+oARcpYiD/myvbbbiM77bitiKyJ8A38MS0fBMTjD6MTtInHH0uHydolxfAHZYJH4rWd5aP8kSuXX3KG1NxkE9mwemx+exTHH9be8fJY4fAOaVl4eIO6hZdFCaP8Eek78fiDcGvHyl99P1GeaNTwZu2TTLBYhuv9U9vAvwH+E4sGs0aHirpSVswNVG6YDFkh226zpeyy8w6F5QemK5TF+INrAfkR6bvQNhZ/EBf+iMoPZ5JpadLfSsUfk+Saq8+PyA8HKFI3k2/IsbD8oCy0YTz+iPUOP+WPcQHAJQ+3bUrCH4RfO1a+/qK5lvnZZxpE+CNbtWCnnHSk+m9ABhtN3DzKe58MoBgBErLEltUZl2bLsilDtG717rhORJZF+G6ayPqpOlAqcJk5IlianTNKZN4Y3SFqwEHfzxktsnB88Js7OgBSEVApM4fL8mlDZcnUf0VmjlAwww5TbMEAZPjJ/LEic4N0GaBLCjxk9qj8fGePLPBVDtLR91aeyGaRwH+k1oN8dLfrkgmy7dZbyTGH1RJZP1nBGu8o+/oZgTZVqOO8MSKzgjpYGTU93s0Nzr8s4E/55o7R+iid1k5WWl9+wVki6ybHrKeGWxUAxUvOPz0fKNryOGkuGCeAW5k3VtuIMvDbAC1nj9I68eyWRUGPviuoHY7SkDRj0JB4/JQe1H9OpH2rweabtdOGRWwWJyhYjLXBpexAEaGBsLIfgmrNGNl2m61k3713y1/KtfeEXzZclswfrOBr+fxB0r9XO1mNcAYgIZwkV2bm9pWfvmkhX3zymgz87dN8oEY6S4fJ+qXDZNUS4rC+PlTWLxkmSxcOEVF7rAwZ3K+T/Nu/s6xFUGODQ7xYAg7/eEBRcmTp3IHS/5d28lmHV3QJdeUi8pioaa2cN1Amjf9Jls8dWHhQXzlScjN7ybzJvwf10nLkSt6En6Rb5zekx5fvytJ5AwN7Madc65YMlWVaj2yZNK6nDP3zswhQjqGpWjVSJo35QTausbF2yCcfDbRojz94R9UBirH4Y91Y2XqrLZVH1HbOeMPaaBlt+a+2+cqFQ6Tfz21l1aLIoEt7S57MyO0rvb79QL74+DX5u+/HAX+sGxe0M/yxbLisWhzRmC0dKhuWRviDAVsy5N/+XWRIv06yBp4rCiwpf8QBijJJls0fJH/+2l66dHhF+v74oSxfMCjKH2sWDJassT1l+dx/YvLH5Im/yNy83wIwEeGPKZm95KtO8Mc7snjOP/n8QTmWDhP4Y+kCeDBbcsb31HoI4MmdTGiZs+Seu65Wvpg9rX8ASDSNwI51l522l//ss0e0rDH7hsOXFfqecrk/yRE0njtsv00c+TFMlswfogc3r1o0RP7s1V5WLBxcgI6zJv8mvbp/IJ93fEX++e0TEXgDgBWhwYZlw2Ul/BGRH/CHpinwR6YMMf5AnhXLH3GAokyS5QuHyF+/to/Kj2Xa57NUfqxdMFiyx/4oy/Gj/d3VitWjJTejl8zO6xu8Uz7Nk7yMn6Vb5zfl+67vyqLZA4IlYupEW0Xk4DL6jmTJ5MxeMqR/Zwn4w5EfhN8wXt5+5UHZecftZcHMv/JBp0yUju2aKd+MHNKtfGYg8fgnSUCRwd5ABVdAQt7Qn7Vut113mcjKbAVbFgbQsnbGMFk6BVmTrUumA3/spMBOAQ0Ab+UkyfznB/m6Q3P5qt1bMur3bgpusPcjHQDh6hkjZCV2XzNHKOhYNW2YLAM4rpokq2YMlz+6d5BR/b5WuzBAI/EM5MS6KuhZN1mmjeoj33Rorr+ZY38X7Pq0zNRtzijJGviDDOvdVTbMCoCuAtPZIyVvaC8Z9HMXBX2Uc/3C8bLHLjvJsUccIrI6T9ZGyrpkyr+yAWC2drJk/NNDBv3UWZZPY6KfG6WTzB0l4//qrvkACo12AKqJ//SQYb2/UACmgHjlJKX1/y46W1iCXjcj0CC6dVS6xgCKlBPt7LTRfaRnlw/kszavycCenUUWZShwhCbzJ/whucN6yboZANzh0bJoG0wbKtlDesqynIH6TjdtrM2TqaN6y1dt35TuHd+RBVkDgrpF+IQ0V0wfJisBzatyZPDPXWTioB+jINstd7reKw9HbBYfa9xQatepV+Bzf+UDiq5AYMBaGwKKCFsLs3yEzJ3xpzJQu1bP6RUtGD8AIlqUE487vIC/vc8a90NgVyVZ8vQT9wZx0LpIhnz9eTA7/qtPx0JxP/vkVRWcMcEiQjMMFHVwyJN777m2UFqU5dN2zXSgXrFgkL4/4djDdOk3mv6aMdKv78f67vNOrweDnWTIySceUSi9Zi/cn28PtGG8vPfuMxqm1TtPRcNedvEZBQcEaBkR9h1aPSeb1awpuRN6yoY14zXOY42rGFC0tucKf6wbVxAoukbzK0bInLzftB4d2zwfpcHG8Afa0/Vj5cTjY/PHuBHdA1DN0vMTgWnCHJaeN4yX7l3f0bQG/PZpNE3jqy4dXi6GP0JAUQftXGl47w2F0iLNdq2fV3CxfsUIfX/8sYcWBHzrxsnvv3TQd10+eS2Y3KwbK+ecESwDWrm4PtekvojkBO0tE+TDSJ/p3P7laN6XnH9qQcAHbywdJhuWDpUZOX0CkGptoGWfoHEPO2S//HLZ+6pwlVzZckuAIhrnjIJLuStGSF5WHy3/l53fiNIAWgWayYy4/DFi8FfR1YEXX3xA407N7av80fOHQMv603ctC6RJuh3a0J6Z+TLMpZHKjzBQZPKRLQ82urVQWqTX6r0m2p9XzBuo74849ICgjUmX9NaOkUH9P9N37Vo+F+QtGXLuWYF5DGnY76lHWSKaEsRbP046tg5kapeOr0TDnHHK0SIyOQjjlh3wDH2RofTLVSNFZJ7UvvNKjbuACUasZXk3jbLcJxEougM6YCpvWC+tWz5QzNc4obnq2OY1fd/27fyx6dQToN8cmTi4p2y/7dZRulob7LXHriILxykoBGQdfuiBovJKZimwvPPmq2TP3XaRrm3fLBB3k002kazBPyoQjQcWFSSuypFzTivc9rfdcLmCTwVbaybLeWecqOmvnx+URYHiykly+81Be8r8cWqTaOW2a/PnHlZ7Tp5fa9pYDj84WDq39/fVvklEZgUa0Q3TpNZBgb0zflGguGGqnHfGSZo/WtycMX0L1JW0Pu/YXEGeu9QdCygqwJ4zSo487KBCaWxas2YA3pZmSp9v2+n7N59/RDctRcuyapI0qhv0vSU5AxVEU6aLzw1WDa1eXB9pcIcCY6XV6lw57cRj5bQTjpI6t+WP/e1bNBNZNL5YQO/yWirfR8Hisgx55IH6cvc9daNgMalAkQ0G1ljXXnWetHmvibz24gO6LHZ4xL6jTy8Mqtm5miXdv3pXw++79+6BUJRsafJUPfULBodM6f7le9E0m7/2qCyeP1j+/aerbmggr1ULIhqHsGBTQR9aepaJ8uhDd2p6r1Au3dWdLZOzekeNoFfMRxs4Ta654jwNt7DArDxPLjwnsGnTZTGZLgcf9F8N17kjoJXlkAly121XqN9nnwIm2YU5QVp+kC+gXniqvjz7ZF3p2/OjAABRVit/5H5DxBYNbcn8Wf9oeqkOFOdP6af1oN2uvvwc+ahFU3n5eQD1bDn8sAP1Xd9fOkZoliW9egSbqfb/z54BoJYsefapYCIxbyZalwny/Vf5/IHx/qJ5A2XssO5SY+NAGzsDWzWW9VwaQ2vlDxcostSfJU2eCLS3zzdpILKGXf85kj3+Z9lqy821fIAANJ+33nCxPk9Hc2jpS47Wi/oFPD5JTj4hAL8ft2USMlFk/Xipc3cgrDp9HAGTkiHtP3whShs2LT37RF3p81PbAAC5Zeeewd8Gep4jdWn+1hOaRvPXHwnyd+MZf1XmtRigOHlS/iB01eVnq/wI+GOuHBHhjx+/axVomWWS/PJDwB//UflB38uQZi81DtoFjb9MkF4/BcCdNnn+mQayaN5gGfnvN7LlFkF7Tsvpnd9+Lm2UpiGgKJnyQtOGmv4zT90bkR+TJCejl+y043bqLyuDlZPaEa3vwpl/Olq9PLn6inODcGoukCOnRex6O3V8JZAf68dLw3tv1DDwhMqU9ePk049eVD/q8czj96j86K3yo4jd4vDJ8uGSk/GTNKgbTH4ebnRbxU0iqipQXDhePnXA3JONastzj9aXXl3bKNiApgDAKWjzlrGqlCHPPBzImfp33qBaKJZYjzjsIJUrgEs0U3XuuD7aJj27ttZlaQONrKzI6py4GkWWxA+rtb/Gf+uFR4Ml2GVZ0uDuoO3r3HaNglEA6mXnny41N6khsmB8sHyKpnFljtS5M8hfgeLskTJ6cE/VJB9aa3+ZOOo3mT/xL1k3f5xssdlmms8RhxwYaEAXjhfSp97PPVIvyGfDNDnqiEMiQHh2PlCUqXLZ+WdoWFk8QbV8WSN76/NZZ5wkE0f2loXZf0fLZUCqEFBclaua0r332E3j9vmuY/AN8JWTpNvHAQ44vNb+QVlWTZKttthC9sDWclmWalcVWC8YJzU22ljOPu14kQ1TRVZMioLYVm89q6CYtnsgAiZbvtE0aM81eXLuGcG4zbjQtvlzcu/t18mCzD9VO+wCXCt/ul4Bi3YoNzaLdih3UoHigtl/KxMcdzS7jOcEA6NMFJZyDz5wX3n4/lsDQMjAqvZti+WkiBYpsGXKlqZP1w+YUu2bMuW7r97X5yaPs2NnTjBLlsnSuV2geenza4fAJs0V8tyroA8DxUy59srz5NBa/1VwogM8ZZF50uz5+zSfrIyfdYAZMeRrfX5HB92JwTLzurHq16Du9QomMcCns739+qNB2bCpwn5K5skRh+4vO+2wXQT0ZEjrVoEW7eWXAM7QJqswCLA62ADPdflwmZdmQPHIw9C0zI3yB0uxRx9xkNS7+5rgOBXs0LT9V8qJxx0WEV5MLrLluaeDiYQBxR7dAv549ME7gzRVE50r30bstH7u/kGwPGk0dWksDlBkM4vkyVWXnSXHHlkr0IJTBuWP2fLum49rW0/NDPiD5TvaPgBlmdpOaxb/q34sDXNEy8h/gzDs5AcM6+Qiwh9oIwPtxGQFOB3avqRxn3uSTy9Bm6zCkwi37NTHfpItY4Z+o/H/u+8ewVIsIMHCV5VrMUBxyqRA4xwAP+zMAH+ZupP7oAP2lnvugK7TgzZhZ7cslDNPO0Y2NRtTyZSXmwUbv2ZEgOIvPwcrEQ/qRrDZAWiTHPmuWzDB+PG71kXwR2GgeMt1F8kxRyLfsgMeVf6YI++9HYD0yZP6iMh4GRGRDa9qf8/SpfD1ywNN9C3XX6yT5xH/dNU2e+m5+1RmKH+obfISOebIWrLZZjUDPtgwXjpF+OMJtVOeX7T8oL3hDcmW++oFwANePXD/fWTlosGBTOJ9ovmiCgPFzu3fUlo/3ghN7fwAVCzPkn97d5XtttlKenT5QIGH2rotwkxhmoY/+7QTAp5bO1mOOuJgqVljk0B+r8qRe+8KaPt3L8yI8lSrh2y/6apgY9nSiD2jCzoYqFmWHvhLoFluVPeWQDZgl8dP5soFZ5+iec8CuMp0BWo1a25SCCjWvcuA4tgg7xVZsuvOO8qxh8Of03WpePXcMbL5ZptqeoBTbBCxT4QGR0U0ewo0ZYYcfeShUmNjJrkFgeL/LjgziA9dtIzT9fmKC89WOgHiwmCrEFBckyfrZg6XPXbdWR5G26flC2xFRRbIUYfXkm0A18uzhc0yTR4J5Hzm4J5qH4lmt++3waSv77ftFUCyjA5fNwXsIjMxIcBGURbJGScHG+e0XdZOlvPPDrSOE4f8pEccYWoAHcLldtsqXe+jYHFlljzz+IPyQOPGUilA8fWXHwpAgA1kDFosPzNYSrZgv5g38Rf5vtv7stsuOwZMCONIfKA4aMAXAbhEuK0eLQP6B0dv9Pi+ZTCzDws9FZQhoEgYtDo6EOfIuqXDZEbeb/JX34/lzNMDxsoY1zOyVDNVatTYWHcL6qC9YYJ06hAs+4wc/LV2EGwGYdRffmonGaN7yOhh3+pvctavUjuiNZqW9asK9VYtA6CYMbx7/vElJRHWaQgUA7Ac2WwBDZQ/ciL8kSUrFgxW/ujZvaXssvOOskmNGhENXXyg2L/3x/nAe9UoGTn8O22b7l3fzbdhc3lE+cMBirqZBf5gYwhAPlftIWfm/Sb9+3SU888JhPfkCT9FQGym7L7LjrLXnrtG4kyQzzsFS1AjWQqVyfJCBNT2/L61ZI75MZ8/svtK3TqBhmfOlH7K9wYUJ46F/yLH25SEPyRP+v3SXuu6zdZbyjI07EzCShLXpUcy7ksIFDEviNoPWj0c+QF/TJnUW3p0e0/22H1n4ViU9auYoMUCisHg8muP1vl8sGaUjBkaTAS7ffl+vr9LA+WPEFA0u0aWHiP8gU0tm4suuuA0bYOczN4R/siSvfbcRbbacguVFWg3u0aW1AH1aKqbRXZo9+zxUQH5wRJ8gwjAmzjqe+WvThGN4vDB2M5F2tdo45bb7nm3cpTaOk/J7Su//vihbBdZXh0z/NugjBY2UdcqDBQ7tQv65qBfPosuCStoY2OLbriaqdqnBdl/S8bAH+Sj5sEK0FmnHB9MXmMAxbp3Bn1YN2PMDGzpsI984fFA6zw/d1AARpwNE5rnykny5AP3KL9MHPiDgjcFJYDIxRnStX1gbtWl7Rs6VqHRKx4ocm7oeNllpx3k6ENriazNFZkxXACKjFF3sJy9JtihrGVYkS1vv/CIvvv3189V5pUEKKpmb3WexgNAWpphUFUIKK7ODXZCr2NiPEftSJflDZbcEb/KNx3fUXOUbbeOAMW5o2XqyMAM5aH6tytwxA70yosjJxAACJdOlOYvPabl6NbpfckY3FNG/dldf7mj+kjj+rfrOzUBkOlRjSIaWsoGDaojSKSdjO9XTh0qt996o7Rp82HlAMVW72OrE9logsBaNVLWLB0mde++RhjMYFz7RWc7xQDF0SO+zxdua0bLH78HNmk/MACwhBwWdiroQ0BRAckkefXFB2TPPXaJloGy7Lj9tvqsQFG1BBP12BHeZY/5QZn74IP2lR1VSwiomSI3XhssP1pdYl3/+qOzLpsaUJyKxpL0w+WN95yGQLH5208V5A92zK8YqbajW2+1RYF22XzzzdRWM1jKjQ8UB/39pTORGCXDhnyr6Xz35XtFAIEQUFQ7vxx59aXGss9ewRKJtemOOwT8MdnaTybKyy8GR7HkqJZxmhx39CEqqIOy5sqtN1xSoC6WlnsdwKYEyREDitPZ5IBGNR4/uP4yRVq1CGxf2e24jo0TaKSKAhBu/GTflxAovv4KE80JAQ2oC1ridWPl3ruvUe2PSz/4Y8stN5cNxQDFAX0+jmho2ZQ3WkYN+krb5uuvWhTBHyGgqPyRpSsI++6ze4G23X67wMYtd2IfCZafs+W1yDJ4nvLHFGHX8WabbhqZNE+SuyLL0259wve9kW+SIwYUx42OTGTjtZ21PWXlHl7iJ9mSl/Wrlvnk47C9npp4PkkBoDiyf/coMNMBc9EEmTT05+jyqtF/442DMSrQKM5Qe7ewRjEKFJdNjG66ACg+G9GGLYgHFFfnysMNAxOo2RP+CHbpAiYBigvGysBegRKkzbuYHswrQqMYAaoAJ7SERQDF1555UMEVddZ6LxwvP3/RSvnhmw5vq4axQoEiS8+zR8naWSPkrpuulG22LCjna9asqXaiaBQD+81cOfqwWmoSxjL/htkjtax177he7TcJd1/tfG25tVv42uuLVmqHeY7ZWcqMagsQXZC4ZvowueqqK6T5O++qnWKlaBQ/eMcBAoCzdWNl14jm8IarL5DPu7wpEzh3UebJ5ZebzU7RGsVRw7qXHyhKjlwXOe7i5OMPlw9bPiuDBjCbmiiffvy6MmIUKK4YobsGYbznmzaQDasCRuVMw0ArmSc3XR8AgbzMXjJnyu8yM6dv5NdHdy4umjVA1qutYYYYUJyCRqqaA8W33nwiHyjCH+vHq/YFWt983UXSpdMbMm7Etyq8rr36Am2X4oDiP3+5GueyAMVg6fnmiO0h5hOtWzaVf/7pFkhAAAAgAElEQVTsooPs558Fm2aiQHHlSJk9+XctW5NH71GwQfmbqU0uWslJcvtNl+n7CSO6y9wp/Qrwx+LZA2Sh8gdLhJlRoBi1mYsHBMxfpsp99QNBGSzlZwX9w4CChatK1xICxVebNc4HipRfcqI2gNddea50+vR1GTMMrdwMuemGS4RBvTig+JeaqACiywEUJVvuufMqbVOWhtt80DQiPzKjR9IEQHGkgltOP4AnHn3gDlm/JjBbeb4py8yBGUWdyNFXmWN+KMQfyI5Fs/+W9YuDEx8MKI4d+UO+HAy3rbW9TJT1AEV2flsYnSRnRg/gjrkJxsKW9ZoCQHFEv2+iR8Cwo3fqqEBzRTs9VO82+e7zVroLmbEJP5Yw9etkRWkUl2SWGig+WD9YjZo17ncFUFGN4vyxMqBnJ837w/dejADF02VTlp4XTlAQpUBPl75LDhTfaPpQQaC4OEO+7RjItJ8/B0zN06XnTWpsrEqRKKCUaXKZs/TsahQvOx+NYk6gnXO0ptSlkEaRpflVOVoP6Mp5j5+1fVPG/43GfLZcdO5pssXmm+nSs8ZdkimftgxW8DKH/ixffxyUddyf3wbH+yzLksZ1blE6Df+9m8zN6C8zxvwW/Eb3VbtJbCfXTB2qR/jkA8Vp1RYoKt/MHqkTk8suvTQKEkGKlQ8U142V7l+30AZtoQByfmATpDZoU9VeEMYJPhUYf+l5FEu2xEGIlUmjOFlmRTZTnH4qu91Wqi1RYAs3XR5ocJOWMXO8A+QkR048/jBh9+I7rwVfjFlF/nqUy0R59YX7Nc7fqjXMCuyf9BigLD1yp23L52Slbo7xQJHNF7aZpQBQXD9evu8W2Iu9pbZ8Cx3+mC0H7r93xIC8aBvF8gJFWTte1i4bru2JGYLI4gCsqC3tZGnyWG19NwXtkAF9yZPTTz1G9thtZ+n4UWBjOG/aHxE7uAxp8VZg19gfbRY2bfCG8sck6f1Da2n9XlNZzfEvBYAiGqkiNIqAAZkSXZq89cZLAnsfyolGkZ8LEMo66FdEvLIAxTVj5NfIppVmzzdS+yM9EiciP4447AD9UpSeg1nE0nP5gWK2rF8UnNV35qnHCHZVqvXUckyJboLKzY60n7bTNOUPVi86tw+Oppkf5Y8JwoQa2dePY34weYjyR7b0+aGNtG/9fHAEk2RENYpjObje5GC4jTTPyXLQ/ntruvqFHj2ex77WMkG/WrOrnmM5o1pqFPOBIkB6ojS4KwBbEwb+GIxBaOfmj5X1s0cpDUukUSwtUFySKe3eCUyRenR+X4/TiQKzVTny2rMPBXzxPXJjtlxy7mmBPTP2dzNHBJsRZJqcenxwLqba5qFRXBQsPauNIp/Oiyw9Ywt9yXmn6RJ7NB+ZLvfcGhyzNWd8f9W6HXn4wQHfyGzdhKOATebJbrvspGfNkr4LFC/HRjGynK1A1wGLhYCizJRvImCv5RusOC6MnlfJcvB+++4l22y1hdqJalzqOSdog6YP3Stnnnys7LPX7rrZRd8vGi8d3g82eP30ZRvd3MP5ifpbOUn6fdteWr/xjCxFq7tuiuQDxenVEigqSJwzStbNHCEXX3yxvPve+9Edz1UDKK4fJ50/DY67aNcGVTrfjmbQnyo9ugeq7wAoYjs4Ke5mlvIDxSmSG1l6uegCjh3hwHDKkStLFgzWDkI5hmELyaYUhO6G8fL1F2/rO3ZJnnTCEYE9C+8cjeMB++0dsVPje8y5snLpsGh6GLADBLxGMQ5Q3DBBunYK+KPtB00j324N+MOOmmGnGgdtwx/xNrOUGyiuGy9L5wfHIl2qxxaxoYRy5MmiOcEmLeWPf7rm24htGB/dNMPSL8ckEV4nMxwHNL2/8oFuMNGNGRH+WDw0yh8KNkoKFOG7dWNlUP/OGv+yi89Ufls6f4gsmjtIfwvnDJSFswfmH+4eBhOV+VwWoLh2rHzbNbDZeueNxyJ2rLTLNPmjT3BUVY0aNWQ153FiDlBoM0tgo5gIoLgqoiG85CIGXewU2WwT8Memkc+aDuMLM/bd6nXjop9BhT9OOeko5WHlj5UjZU5EI43ZC3JDJxMcmbQ2sCuD39ScoDRAccN4ef/tJ5U/gk1UyFs03NOFJX3SfEtXRYo4g7asPJJqGsXl2cJxOtBk0tBeurwcfEN6hjx2fzAxPPbIQ4IdtuumFNrMEl16LgVQBFAB9lbNCMaIPXZnZ+/EYGf12jw965HyMN5wniFnHTaMgNmvP3kvGD/XTZGvIzuFCatAkU0mi8ar5v3gA9msuUSBGDaKW0V2+GcPZSPeNK0PZzcSV+u3bopuxrni4nPU7+9fWGWbrgCr/fvBBFjLEwWKuRru3DNPDiZu8ziMu+CZkYWB4iz5uEUwWerw/ktBP15K/5kh33f5QNPDDM0O39Zl+HV5cuPVF6udL2XFJpElZ9OMoTHEn7Njse1EYwlw5UBx+5SjHi5ezYFiABJHy/IpQ+TKyy+Xd96Fjwq6xGoUV4/WhuHQbR1Ew+cozvxL3zd/g13AEUG0fISwzEaD8nu2SX2dKV90YbAL6YDI7HeNAoEceeyRwMhXtSqSKd0+CwaJYf9+kz+TXjNa+vYOdjN++22LIm0Uzzj9WM2X5SvKjHaQcrCE9HHbl+Tee4Ivn7C7Fv8eXzs2S8uHy9plI/SQYN79/D07JMcGIFJn79nS6r2nNR5fm3ju6fry+EN36zPh//r9k8D+SSbIu+82Uf/csT8WrTEKC+nlw2XOjKBDNKrPuVdxzn0Lx0v2M0tba4PlNequZ8CFzlHkEGrevfIyu4Dz+WOJA8ReeLahtGv9nPzvkmCnHWcCEoelfdrwyUcC+s6Z8ZdqdL6JnLP5Z78uBWwUBw8Mdhx36/J2ETZouXJe5BOP+gk/yZUjI/xx+y2XySftmknDyKaCow4Pju/5/psW+Tygy+bjZPPIt627fY6tz4R8TY1kykctn9XyAwZeeKaBPPbQXbo5R/kDTRKHRUuGfNgqCMdGqCI1ipJbyL6WtNzfjtttE/QJt38mmx9i5SfBlxooayH5sWKE5EbOUXz+Wb4oErFRRDu6PND0Eo+NLu3bPB89ZubwQ4MjRmZP/k2B1rPPBqcX6DmKMkF+7BGco9iPo4Z0AhisSAz7+wul2Refv1MEf0ySiy+OHA2im+Am6mSRctx206XSqcPLUr9OsPP0/+1dC3yOZRvvfJR0+qWvlEqSEn0iEWIY24wQOZdDyKnzp4PORB+ysQ+tkKSDUFIpx5Aa2rybjW1Ow4wZkm1OcX2///W817P7ffa+72YHm7nu3+/1PM99fv7P7Xn+u063vFcQcB+mFPyHJju/WB9VtJmF2Kun3OuD3x9J9NEkSyqCIOTYivHFZ3uw4wLqr4QkGoHkT8ZTpHsdrY82bLW9Ycwq5k30QE1LOtQqsAG9/84QavSIFa8P+5qz6QzqeWtfmLwSJIqboxbw8wwNbOwOuG3EUTyYQJERI7h83eKvclTP++NpxXfWLkcg8mEj/0Pj3nmJqlWpzFJqfOfKl7uSw7pAKnXH7ZW4D3bGOLqVunZ0xzH8e5OH6vmlgdY7KgOx/iRItiFt4w93VjJ9Ptn6A/nq8lfRK0N72yF5+NlzEOoEVkvviM1Rj99T9Q666QbLvr5Da8u7GmSIA09nb+Zg22hf5fZKFDlmOEvgEB6n3JVX8Nw7tQ2kzu2C+Bz19iWutOIRZsRR/Iq5dn6NalXo2grWPuHtQ5pZ930QEkU4pqbSNeUtm+1KN1ekeTPC2NsY9yWSRSaKR6z/7w3r12b179FdOX8gD3+hH00Z+warnDGPuypb2GalRDFmjNH+DbR2kRVZBHUObVnt3sPavcvM4USaPtF6riDDb7zQn14c2NMOnbfo68kWgTy+nbd5RB+Y+7nkxGKRRBdl71hDbUJDaKzbJtGTJhal6hkvkKwYQsDXrz4bbXkGmy+VzD/Zmxk2iMsXRuZI5fBCPBFHqVsXUYsAy3MU4SzaBjdmVXD0b7Oo3oP3kQt/iVMSzZ8dRp3atbDUZ26jc/SZuuUXy6gdY2bH0JaEBYRQFfFQSXtT1fGLOJEmjHmZBoFg4YMM1c7xWOrXq50dMqDOA/ewVyvUWQ3q1qDIiOE5L3oQHUqh4JYNrf8o+FiYL1geI5nW/jaLoI6CfQcWY2DAw5ScsMD6GKLOsVhauWQaOzcc3LnMUg+a2Pk7Z1yjqHf3NjT/q3EWrv7ql2RZdgzNnDqCZnz8nptMGQHZM6Mpa+8qxmDJDx/lWh/w/m3ZzPIchYdfSKuGlLZtCT9frI8oVu8n0o9zwqlH52DKzPiDcY1fM5vtGrcnLrSeL+4/K4Z2bv6FOrQNoDh4IENVh+dgYoPrUwk0JexVK2wTG/3HMMFCwO0rLrNijyFUzkrsIHRkPQU0qsP1eS2hPX60m1oHW3+JM7mR9SG7cVASRf/+JTVq8G+2pYMaqHmThygx7rscsnw8jlYv/ZTne3D3Cv+q46Mueu6ZLvQsfgO70NABnT1+g/p14hh7PtWTJgZn+vx4HKtgP0cwcqf6NDOaDqatop6dQ3gnHPbsxfyA8fE42p+6gloE1GMVHGLKBQbUo7SUpZQQPY/q1r6Xfuf1sYl+WTCJ1wf+OEUfG9d/Rx3bNaOtcEjD/388l6wYSk1aSB0fa0Yxf7h3KcE4Jh64RqDrSW9S/17trfXqfn8MHvCELaWpXasaLf0pkk4diaUWTR9icwOEx+F58/rYSe1CrW0ATzjtBjEeJRO8kOX9gfXRpGFt2ujC+nBHBjjq4t2oEFZnz/al/t8fGBPqZkqkUe8OpYo3XMvvJDhnjRsNiWySVe68X/PeC3peQkQR6tDDW1bTxFGv0K+Iz4ft5nYZRDE9ltYs+oKeaBPINonYHUQIDQI2r5g/nWpUs/4QrHBVORrcpzMh/3+jX6PG9R+kzG2/s33f6OFDaUifLlYYnYMJNHPyKOoOb2KWqLm9SQ/E844j3doH8baCmJuM5XGE40p2Mm1YOZeDbl94/vlsYtOicT1Ki1/G4zOhYQeXeNqXuILgAVyzelV6smMo7UlYzt7Zw5/vS6ewKwsklftiKdW1mOrVrkG4j+HPPc2SQnyTnurcllYumEGVb/kXq5IRbPso2rh3kGFCcWgjpcYtoT5dH6Na91alp3u0p0Pb/6B1v3xBI4YNsre9wzgIUl6z+l1UoXw5mjDiPxwH0iSKuFdIOJ/p2ZEmIcYhtkY8EE+7XIvtAOKwu2wT+ChlJK+iqF9mUZ1a1cm1fI7l9IJ4f+jDHaaoQd1aLAl1kjzKSqb4lfOo6SN1GD/8/3m0Xm2KhS1jZrIlfTy0icJHDqMBPR/n5+jsw+O5GIT+bM8XkoiwRCEhQR42icVLFPECgUTNm5cxytgrMMmS7JgvIpyDzBFUb9ibE3u/brbyoKZBPspRD3/1CyGz2+HlFuP5IsdLG/W8kUR50aE9/roXKRyu3faF1hwwF4ir49wv1y050iIQB9znyXh+0fbsGmLNWfo2j2zHBokl+sNvi/tD6P74YFyug3iMhoG52Ye/c5YSJJZuj1aZP9aHSF0lT46CvajmJN9+zl7WB+PmxhP1eL2441qiPa+rpBySKH2yyt+9PtBO8s0j8t0SPS7HtcyR16ixPrLx8d2cE3MP649j3iXz+uj4WHNL7WyOJWQx1/rY6kle0YZt3fK5PlglDlMNX79N3u/XvPeSOpf14W18xj7R+v/ogaPbe9fb+yPX+nC/P9AX+vC2Ptxk0X5/mGOZ80I+rw+3FB/Xea0P1Ec9rA+e225eHx14fWyzyswxcJ7X+kAdWR8ggc72zmuMz+8MBPvfYakRsYMLpLTIR7mzTVFclxBRxAeRbedARpgk5ki28LHnDybiB2Yle0ikhAiw6vYE8NnNe0Mj9IrlRbyRJXKQ1nEfh4BfIp/zNWwGM5M9iCDnZ2zgfJ8k0SAjrF6GypSjfqRa6lPYkhnSOZxzfEeoqGF7iPtAWB/8IM1098djY6s+OI4gZmJmMh1Lt7SA7YIDLHMJ3OfJHURZuEdre0LBgcdBn+gfKtzMZMvmLwPhmDba98njYF/o4ymWWv6vBJ4v9hS2+3LPibE8aO1+YrVzWVsj8v3usgJsY8774zzuHY5GGN+13IoPu2BWBJNN9JFrDDxbE8Nj2zzWAY+LPaUPJ3niajwHZ59n+zXf814Xe4s3b96CxjpsEoufKOIl4+9F46tM2vk8GtInZx/Oa3mp+cqXchxlPMmTa+fRrHtkPSW6viXEQOz8eEt+0ftVCTr7kmsZ0+zbzDudc299nk77M1U3r3n6Kpd8f0e5B9SRcxyd11LmK1/Kpa1ZD+def8Zayo6hrRvm046khdSjq7ULTxpUnyAlZl8yjtf+HPcgc5E2/o6++jPz/bUvyTKZo685+CqXfH9H9Cnl0r/zWvKlrnnt7dzZXq5zHY2xs9fzfu47k36mXj2tHTA2uebl/gNaxsvVl493LOpJm7yOvvo8nT7yGsNZXkJEER91fBjl5+0jn1cZEzN3bD2piyPy+WiMIf1LPbk2j9LGzPN2LmM4x3fWNevJuHI065r1xJkFEsXHgpoSHbX2dpaxuK6DLNl9GnjaeWZdNzY5feUmiZiXsy2PmautG2PkQ4q410Wu5bMpMep7uuVfFfn7C3Jr3qd57qtPZx2uZ95DGT3HfWLvbuxF3jIwkMaHhTt5Ya7rorNRdL4Uyuo1XqS0mQb3t1zv8Z9sGO+7mpz/l3RZxUbvyyIitJGGub2gsT6GDMC2aD6kRYrZufX/ht8fiTTsZWtPcqyPgf06WevDtNkti+uiBImiSQr03CJtkGgecXtvBzapT3TECrhd2vEBocVmBvi/g9/SuZEs0TxXiF5hno+QxKyUNdQ6BI4rnt7NuRiiO0OJ4um+kPGiz4qhfSlL6ePIdykK8flEdV3WX/Sni9W5WN+9PtJTltDUj0fQH78izmIpdTA6F59PSd8zr49o9mie9slIWrV0uuV1jHmhrKTnV5zjK1H0KfUqzMe/MG1PpkWzbWLi6u9s27/C9Hcm2kLtHL3sa5oaMYK2x/xsq/vPxNhn8xhMEtNddGTnWmodEuLVu1mJYlG+APFCh40RPA5hE8d9l/GXfFHiV9b7cpNFa33Ele2Pf1l/lsVxf97WR1knicBRiWKpI4ogPhwiJt1lq9BLPRli1WksO8jADpPV22VUTVxUz4JJ4t71dGqvi1q1DDwtkgjyeF6Xrj1oB0JugPgUx0uxTPep5FDXjGE7m2ut6/rQ9aHrw14DbqK4ZM40ogNxHDLm5O5oPSoGBVwDfxaw3dm75gpCHEWSeHJPDDUNCKCwfNgkOiWL53Xt1oMyUhF5HcEt4QGsP8VA14CuAV0DugaKeg1spc6dO9PiOdOZKP6D0DBp+lMMdA3kZw2c2B3t1SvfH3m0bBJjOU5iq1Yt8+W44iSJLFHs1OkJ6tm9E/Xt3Y369OqqP8VA14CuAV0DugaKeA10o359u1PLli0pODiI2oS2ptDWIfpTDHQN5HMNtA4NpUVzp1txOd0e4HmTRBdlpkRRKGwS8wiB440gSt55u3btooSEjfpTDHQN6BrQNaBroFjXQFJSMu3LyKC96en6Uwx0DeRzDWRmZtH7o/5LM6eM5cDkedllirr5WOo6CgkOog8/DBPOV6Aj9qzRpAgoAoqAIqAIKAKKgCJQShEInxhBM6eMyZMoWo4rLvonLYZatGhBH44vHEkEHEoUS+mi0GkpAoqAIqAIKAKKgCIABMLCJ+ZJFEWSeDx1HTVv3pzCwycUCXhKFIsERu1EEVAEFAFFQBFQBBSB4kEgL6IojiuZ26MoOCiIwsKKhiTibpQoFs8z1V4VAUVAEVAESgECW7Zsofj4+ALN5Pjx49zuyJEjBWqvjRSBokLAH1EUSaLsuFIU6mZz3koUTTT0XBFQBBQBRaDMIPDQQw/ZW71dfPHFtHfv3nzd26xZs+x2DRs25PPVq1fnq21hKy1atIh+++23wnZTZO3HjBlTZH356uiLL75gjJWQ+0LIt+pZSOKJtGhq1bLgIXB8j6wSRX/YaJkioAgoAorAWYrAs88+y+RDpn/DDTfQAw88IJd+j9hDuEGDBhQXF0dr1qyhTp060fbt2/22KYrCgwcP8pznz59fFN0Vuo+nn36aLr300kL3k1cHM2fOVKKYB0jeJIpCEuG40qRpAIVPmJhHLwUrVoliwXDTVoqAIqAIKAKlGIEmTZrQddddZ88QZO/KK6+0r/2dgCjOnj3bXxUuO3nyJP3zzz9+62VnZ/stNwuPHTvGhGnZsmVm9mmd5yWVw3zzqiMDvvjii3TTTTfJpccRfWC+/tKJEyf8Fdtln3/+Od/30aNH7Tw5AX5iAiB5/jD3VpaVlSVNfR5PnTrlUYa55HV/Hg2K+cJJFMUmEXs3w7t5woSIYpuBEsVig1Y7VgQUAUVAESgpBKZPn87k45NPPmFVLsjf8OHD/U4nOjqarrjiCm532WWXUeXKlSk2NpbKlSvH0kU0hqQRfffo0YProd+goKBc5Oubb76ha6+91q4D6Zy/BEJ0wQUX2GPDaxXp3nvv5fHMtuirS5cunPXzzz9TjRo1aOnSpVS+fHluf8011+QiuiB2bdq0sedz++23U1RUlNmtx/nLL79s173wwgtp3bp1XB4TE0N33323XVa7dm3atGmT3RYEHRLCjh072nUGDRpkl3s78UYUn3vuOcK4wBe/e+65h7Zt28bNJ02axHl//fWXR3evvvoqVapUyc6DCUGFChXsPpzzAAn+9NNPuQ3G+Prrr2nHjh1UpUoVu81tt91GZ8rswJ64lxOTKApJPLz9D46TOD483EuLostSolh0WGpPioAioAgoAqUIgWbNmtkf/Oeffz7Pmf39999MFkAaBg4cSL/++itt2LCB+1i7di23B3lEOXaZAXn66KOP+BokTBLIB+q88cYbBDKzcOFCvhbyJ/WcR2n31ltv2SQO/bzzzjseVevUqUPVq1fnvHnz5nHfIFXTpk1jVXn9+vU5D6RHEqSpkLCuX7+e9u3bRx06dOA6JsmTujhu3ryZWrVqRVdffTV9//33BCKbkpLCbSDB2rlzJ23dupVq1qzJedi8AwkkFXMODg6m/fv3c1tc9+zZ0+ze49xJFEGC0ebHH3+kQ4cO0cqVK+nyyy8nmA8gHThwgMsnT57s0Q/aDBs2jPPkubz77rvcxw8//MBt2rZta7dBffzat29PvXv35vsBjqGhobRnzx6+R8ESZgElmUyiSOkuyt6xlnc5Gl+AvZtP9z6UKJ4uYlpfEVAEFAFFoNQjIOQDRAAk43QS2kBqiASPaVz/+eeffA2ieMcdd3h0B5IBCaQk2PU9+eSTcsnHhIQE7gde2L4S1J0Ya/HixVwFqm1cjxw50qMJpJq1atXiPNgzog5IrSSoWpH35Zdfchakqrh2qmVBNgMCAqRZruMLL7xAN954o53/4IMPsnTVznCfoO/HH3+cr66//nqPNsj87LPPePz09HRnU76WZyWqXuwJPmXKFI+6YscomY8++qjHc1ixYgWPARKJhDkNGDBAqvMR0lDkg+RKHWApCWQY5cuXL5csPr7//vtMej0yz/CFTRQPJ9IJdzDtsCKKk5jXrShRzAshLVcEFAFFQBE4qxAYN24cf/A/+OADtm/Dx18kcPfffz+99tprPu8HKlrUh0QKyUkUb775ZlvtK528+eabdP755/OlSLsgFRsxYgSru1EO6SL6jYiIYIkeiNCMGTNY9RkZGcltIX1EHUi/kPJDFL/99ltuk5GRwW3wD2wD0Q8klEiQDEJSBg9mqN/xGzVqFDv3oJ6vNGTIEFuKhzqoO3jw4FzVQeygnkeC+htqYzOBoKLt3LlzzWz7XIii00YRqvGpU6dS3759WRpqzlWktJBaIkGCKeQ5LS2Nx4N5wHvvvWc/A9w3+hBJJM5feuklex44EbUz1svYsWMJfZWGBKI4e3o40YEN7LgyMeJ/Z2xavlfIGZuCDqQIKAKKgCKgCBQNAkKuunfvbncoUjexm5swwXcw4vwQRZGeyQCQOAlRhAoWBATq4aZNm1KjRo34h/PAwEBasmQJ2/ChjvlDX76IIkidmR555BGbFAlR3L17t10FjhnoW4gi1KeXXHIJj9+4cWN7TiCQXbt2tds5T7wRRRBeZ4LTC/pHAmEEHs6E+TilhFLHSRRFBY82t9xyC3Xr1o2eeuopvidpgyPKxe4U5+gHKTExkcvq1atHsJl0PoOffvqJ66ENpKbONHr0aKpWrRr3gTqwAcW6KMkEYhgZPpLatW1DEycWn+OKt3tUougNFc1TBBQBRUAROCsRgP0dPu6w1zMTJEvIv+iii8zsXOcFIYpQDQtRPHz4MI8D+zpnEucLkFlIHmF/h59IxYQoCpERSZxIHKU/kFAJ9SNE0ZR8OYkibPBMFbL0g6M/AuQkiiCD7dq1M5vzOWwvQeiQrrrqKttOUCrCVhLYI9SQtyREEfcr+IuzjtQHuUcfZoJTDyS8ULujTFTrwBPXThUy2pq2hqhjEkV4V5t2nZBwitr8lVdeMYc+4+eQcNetW48izqAkUW7SE3XJ1aMioAgoAoqAInCWIgACABWimcQLGmW+bOVQX4iKP9WzU6JoEkX0UbFiRZs4yRygQsXYYuso+eZR4ihC6igJbZ555hm5tNXRCCaOlB+iKBJVqGvNBNIMr2VfCePCOUUSpI+YD2z5JIld5dChQzkLDifwHDcTQhOhna8kRBEEGvaDqCvqd2lz66235uoDDjeoC5tR00kFbTCHqlWrSnM+Qu2P+oiPiYRzkyjCuxl5TkKLPNihlmRCUPLwYvZu9nV/vp+crxaarwgoAoqAIqAIlGIE5syZwx98hHGBGlE8c99++22CswU+/L5Cw4gjiKinXS6XR33Y4MEezkxifyh5CKmDMRCaBV63kMLhGqydcN4AAAIDSURBVLZ8/pLYFqIu5owk5Kxfv34sqUMIHUjt7rvvPi5HvEfUF69jZHqTRELqh3qQ1AEH2CziGp7MvhKICergJ1JOCcEDL3JIHFGGUDuSEHIGefCyBoGGyh3X4lgj9cwjpL+oIzaKd955J1/DgxkhbVAG21IcxVlF2kuoHufzRKgj1MfzhjRZQgP16tVLmnK5ScJRcNddd3E+HGHgbS4hjiCpPleTEsVz9cnrfSsCioAiUIYRQBgYePRC2gQ7NYRYQQIZw7Z8vlSuUD+CoMBDFgnb/oEUiWp3/PjxuWIUrlq1iskIN3D/A+kgyAYkfyBpiNGXn5SUlMTtQDAlgezCuQLqZtwHYidKyBzY4yHmIUL7SILqGQ40IKxmguQOxO3hhx9mKWVmZqZZ7PUchLl///4exBoEEHaPsAGEw4eZQI5BRhG2B7EPQdD8eXqjLUIQAWMzQDdiIqJ/EGW5D9hqOvuCfaQvr3aooOEIg2eAkD6Yk5lef/11wpaJzoR7wth169ZlMiwE1lnvXLlWoniuPGm9T0VAEVAEFAFFoJgRQNzFkJCQYh4lp3tIDSEx1FR8CChRLD5stWdFQBFQBBQBReCcQgAxJPMKLF5YQBBvEV7t4plc2P60vX8ElCj6x0dLFQFFQBFQBBQBRSCfCPTp04cQv7K4E3aWgfc3VO+aiheB/wMgVrnzaso5ogAAAABJRU5ErkJggg=='>\n",
        "</p>\n",
        "Figure 11-5. In unsupervised training, a model is trained on the unlabeled data (or on all the data) using an unsupervised learning technique, then it is fine-tuned for the final task on the labeled data using a supervised learning technique; the unsupervised part may train one layer at a time as shown here, or it may train the full model directly"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ze2ArZwBZapM"
      },
      "source": [
        "Indeed, it is often cheap to gather unlabeled training examples, but expensive to label them. If you can gather plenty of unlabeled training data, you can try to use it to train an unsupervised model, such as an autoencoder or a generative adversarial network (see Chapter 17)\n",
        "\n",
        "Then you can reuse the lower layers of the autoencoder or the lower layers of the GAN’s discriminator, add the output layer for your task on top, and fine-tune the final network using supervised learning (i.e., with the labeled training examples)\n",
        "\n",
        "Note that in the early days of Deep Learning it was difficult to train deep models, so people would use a technique called greedy layer-wise pretraining (depicted in Figure 11-5)\n",
        "\n",
        "They would first train an unsupervised model with a single layer, typically an RBM, then they would freeze that layer and add another one on top of it, then train the model again (effectively just training the new layer), then freeze the new layer and add another layer on top of it, train the model again, and so on\n",
        "\n",
        "Nowadays, things are much simpler: people generally train the full unsupervised model in one shot (i.e., in Figure 11-5, just start directly at step three) and use autoencoders or GANs rather than RBMs"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Veqe9byQaRtn"
      },
      "source": [
        "## Pretraining on an Auxiliary Task"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "z8m5wp43aUeY"
      },
      "source": [
        "If you do not have much labeled training data, one last option is to train a first neural network on an auxiliary task for which you can easily **obtain or generate labeled training data, then reuse the lower layers of that network for your actual task**\n",
        "\n",
        "The first neural network’s lower layers will learn feature detectors that will likely be reusable by the second neural network\n",
        "\n",
        "For example, if you want to build a system to recognize faces, you may only have a few pictures of each individual—clearly not enough to train a good classifier. Gathering hundreds of pictures of each person would not be practical\n",
        "\n",
        "You could, however, gather a lot of pictures of random people on the web and train a first neural network to detect whether or not two different pictures feature the same person. Such a network would learn good feature detectors for faces, so reusing its lower layers would allow you to train a good face classifier that uses little training data\n",
        "\n",
        "For natural language processing (NLP) applications, you can download a corpus of millions of text documents and automatically generate labeled data from it\n",
        "\n",
        "For example, you could randomly mask out some words and train a model to predict what the missing words are (e.g., it should predict that the missing word in the sentence “What ___ you saying?” is probably “are” or “were”\n",
        "\n",
        "If you can train a model to reach good performance on this task, then it will already know quite a lot about language, and you can certainly reuse it for your actual task and fine-tune it on your labeled data (we will discuss more pretraining tasks in Chapter 15)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yG-Bw2vQbLfD"
      },
      "source": [
        "**Self-supervised learning is when you automatically generate the labels from the data itself**, then you train a model on the resulting “labeled” dataset using supervised learning techniques. Since this approach requires no human labeling whatsoever, it is best classified as a form of unsupervised learning"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "b9OEz6FwbUjV"
      },
      "source": [
        "# Faster Optimizers"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7N9TjXmPbWOz"
      },
      "source": [
        "Another huge speed boost comes from using a faster optimizer than the regular Gradient Descent optimizer. In this section we will present the most popular algorithms: momentum optimization, Nesterov Accelerated Gradient, AdaGrad, RMSProp, and finally Adam and Nadam optimization"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ddMNUexPcAJr"
      },
      "source": [
        "## Momentum Optimization"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3pKh8QDRcByl"
      },
      "source": [
        "Imagine a bowling ball rolling down a gentle slope on a smooth surface: it will start out slowly, but it will quickly pick up momentum until it eventually reaches terminal velocity\n",
        "\n",
        "This is the very simple idea behind momentum optimization, [proposed by Boris Polyak in 1964](https://www.researchgate.net/publication/243648538_Some_methods_of_speeding_up_the_convergence_of_iteration_methods). In contrast, regular Gradient Descent will simply take small, regular steps down the slope, so the algorithm will take much more time to reach the bottom\n",
        "\n",
        "Recall that Gradient Descent updates the weights $\\theta$ by directly subtracting the gradient of the cost function $J(\\theta)$ with regard to the weights ($\\nabla_\\theta J(\\theta))$ multiplied by the learning rate $\\eta$\n",
        "\n",
        "The equation is : $\\theta \\gets \\theta - \\eta \\nabla_\\theta J(\\theta)$. It does not care about what the earlier gradients were. If the local gradient is tiny, it goes very slowly\n",
        "\n",
        "Momentum optimization cares a great deal about what previous gradients were: at each iteration, it subtracts the local gradient from the momentum vector m (multiplied by the learning rate $\\eta$, and it updates the weights by adding this momentum vector (see Equation 11-4). In other words, **the gradient is used for acceleration, not for speed**\n",
        "\n",
        "To simulate some sort of friction mechanism and prevent the momentum from growing too large, the algorithm introduces a new **hyperparameter $\\beta$, called the momentum, which must be set between 0 (high friction) and 1 (no friction)**. A typical momentum value is 0.9\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ly_9aIk-gV1F"
      },
      "source": [
        "Equation 11-4. Momentum algorithm\n",
        "\n",
        "$$ m \\gets \\beta m - \\eta \\nabla_\\theta J(\\theta) \\\\ \\theta \\gets \\theta + m$$"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qplIH-5vgmxH"
      },
      "source": [
        "You can easily verify that if the gradient remains constant, the terminal velocity (i.e., the maximum size of the weight updates) is equal to that gradient multiplied by the learning rate $\\eta$ mutliplied by $1 \\over {1 - \\beta}$ (ignoring the sign)\n",
        "\n",
        "For example, if $\\beta = 0.9$, then the terminal velocity is equal to 10 times the gradient times the learning rate, so momentum optimization ends up going 10 times faster than Gradient Descent\n",
        "\n",
        "**This allows momentum optimization to escape from plateaus much faster than Gradient Descent**\n",
        "\n",
        "Gradient Descent. We saw in Chapter 4 that when the inputs have very different scales, the cost function will look like an elongated bowl (see Figure 4-7)\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fnXd5xdNhe-U"
      },
      "source": [
        "<p align='center'>\n",
        "<img src='data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAooAAADsCAYAAADzRofiAAAgAElEQVR4AeydB5gURfbAMafzuPOvZzrDiWLOIgiKgAICCgKiEsSsgBExYkLPHMGczoQBRAXJSTJIhiXvkpacNu+STO///Wrv7ZXtzOzMTveEpfr7enu2p6fDq1evfvXeq+oq4hYnASeBtJTAypUrpWXLlnL88cdLvXr1ZMSIEWn5HO6mnQScBJwEnARSVwJVUvfW3J05CTgJhJPAvHnzpEqVKn9a33zzzXA/cfudBJwEnAScBJwEYpZAGSju3LlTfvzxRxk9erTs2LEj5hO5HzgJOAkkTgI1atQwkHj++ecLdbdHjx5l0Pjbb78l7kbclZwEnAScBJwEKrUEDCgOHjy4rJFRL8V3331XqR/cPZyTQLpKYOvWrWX1dfbs2WWPoXV30KBBZfvcBycBJwEnAScBJ4F4JFBl+/btZY3OySefLDVr1iz7Pzs7O55zu986CTgJBCCByZMnl9XR33//vewK1apVM/u7detWts99CFYCq1evlkmTJsmUKVMCW6dOnWquMX36dAlynTFjhjn/zJkzJeh11qxZMnfu3MDWjIwMQW5z5swR0jSCWrnO4sWLZeHChWbL56BW5JWfnx+sQruzOwmEkECVp59+2jQuNDK6VK9e3ey75ZZbdJfbOgk4CaSIBPr27VsGis8//7y8/vrr8tZbb8nhhx9u9l955ZUpcqeV/zaA8pdfftnIn/xQv9d33nlH/v3vf0urVq3k0UcflYceeiiQ9ZFHHpFbb71VrrnmGuncubPcfvvtgaydOnWSjh07SuPGjaV58+ZyxRVXBLJy7oYNG0qdOnUCXWvXri1NmzaVM844Qw499FBTB6mHfq+HHXaYsH700UeVv1K5J0w5CVSpVauWaVzuvPPOspt7+OGHzb5jjz22bJ/74CTgJJAaEvj444/LQFHDzfaWRtgtiZHA448/Lnl5efLLL7+YXFHyRf1c8RgvWrRIunbtajyWY8aMkSBWvKLvvfeePPbYY/LZZ58JOhbE+umnnxqwbtu2rQCNQQJpmzZt5OKLL5b69esHtjLbwA033CAnnnhixDpp18+Kft5zzz2lX79+iVFsdxUnAUsCVQ4++GCj4K+99lrZbgwFyrz33nuX7XMfnAScBFJDAl9++WVZo4Qn6KmnnpJnn31WDjnkELOfKXPckhgJAFZr1qwxIcHc3Fzxey0qKhLCtHfffbeMHTtWhg4dGsjKIEa8oTgJ8FoBjUGsH374oeAFv/baa+W2224zXkw8mX6unJcVUATkLrnkkkDWBg0aGAAFFE855ZSyOllRECzvd/vuu68DxcRUa3cVjwSq2Mq5++67y2677WZW3U9P2S1OAk4CqSMBGnWtn/ZdacrIAw88YO92nwOUAKC4du1aKSgoMJ5FvIt+rqFAcdiwYeLnCnwqKNLxABTff//9QFYbFPEmKtQFsXWgGKDiu1PvUhL4Ayhq42NvHSjuUvrgHjYNJLB58+YyUCRJX5c99tjD7B8wYIDuctuAJeBAMTagdKD457lP7fY20mfnUQy4MrvTh5VAlQMPPNA0LoSvdHnjjTfMvr322kt3ua2TgJNACkngyCOPNHWUpH0WBrNoI/Pzzz+X3SkeKUbmuiUYCThQdKCYqqHn/v37m4FDDLYhNQVPt9/Lpk2b/D6lO18KSqDKSSedZBoYRtPpoiOhjz76aN3ltk4CTgIpJAFe16dgaG+ffPLJsrtkAIT9HdOFuMVfCThQdKCYiqB46qmn/qHuqx1YsGCBLxUgMzNT/v73v8srr7ziy/ncSVJbAlWuv/56o1C84UGXCy64wOy7+uqrdZfbOgk4CaSYBBjcwNynNAIMSuvZs2fZHU6YMMHsv+OOO2TLli1StWpV2X///cu+dx/8kYADRQeKqQaKtNvYhMsuu0yYJ5lFO4377bdfzIof6k1PTNvENb755puYz+d+kH4SqMLkoNrbQOGZskD/t9/6wASs7du3l+uuu85M15B+j+ru2Elg15HAqlWr5O233xbe4sLC3IrUa7f4KwEHig4UUw0UqedMpeNd7r33XuEtbLowzRLgqO193bp19SuzZXS6fsf2uOOOM/uZ+Nvez8hyt1RuCZiWgwld7YLnsx2K/vrrr8335EX99a9/NZ/nz59fuSXjns5JoJJIYOnSpabOaj5jJXmslHgMB4oOFFMJFHHo0H6fffbZEevHkiVLzHEcS/5ihw4dzP+absYbbfiOCdjXr18vr776qvkfbyWdT51h4ZxzzpEXXngh4rXcl+kvgTIXA40JyvDSSy+Z1xHZj8akpYS2dEGBmAfLLcFKgBHnvLJpxYoVZrJdpsT46quvzLQV5IbwpoYHH3xQ7rnnHunSpcsf5iLjrTpMOXHffffJE088YXJJmBft888/l4EDB5rpMPAmM++bG9kebDkm8+zkJFFfmfrKLf5LwIGiA8VUAsVRo0aZ+k76WKSFN+JgFzheF+aFZN/EiRNl2rRp5jPgSFvDqwo1OsHxjF7nWNojt1R+CZSBYnmPqnkKQAsKAqS4xR8J/Prrr2Zk6vjx4w0EIlsmoGUeMPJMeLPARRddJBdeeKF5HRVbVvaxEjIIt+ox+hteaaW/JWTAWzy4DqD5zDPPCG9O4D4o55KSEn8e0J0lKRLIysoydfWYY45JyvV3hYs6UHSgmEqgqB1D3hTjXXhPuC5HHXWUsQ1MtaULOc607bySkuXGG280/7OPlem3cDSw6LEffPCB+d/9qdwSiBoUEQOeJ+ZyQmncUnEJ5OTkyLhx4+TFF180lfHyyy83bxAA9hToADw8ucCcvoKKHp9fq56T83Mdrq0QyWeFSEINNIbktripECpe5sn4pRp4GghCUrymzS3+SsCBogPFVAJFtFvr/caNG8uUnbZb97OzZs2a5n/yDXVhDALHfP/997pLmGqLASuMTbB/36tXL/M/b3FzS+WXQNTEx5sHVFGys7Mrv2R8fELCu7z9AE8hLn+FQbZAmQ2EXhAM6vVTel7v9fhfIVIBEmjlXgFJ7p/nGDJkiHkjhY9icqfyUQK8nUXrq7318RLuVCKmE+XezBI9LLoJt4OfcPv11183dX+fffYxESIm4NdBK3gJWUhjwi787W9/M6+GJLzM/3/5y1/M9wqCdl4zsyZwDIt+f9NNNwkzLLilcksgalDEk8jk3Lyk3i2RJYCMyOkgJ5CKiXcOyAIKWQExG9AU2qLZ2r+L53M019Jj9Drct8KjPg8exx49ehgQLi4ujiwY923CJMD7h8lBJfzMyrxnrG7xVwLOoxg9JPJaQAeKwYMiGv7cc88ZqLM7iS1atPiD8hPRsr8nHI3d0KVVq1Z/+J5jcRCw2LOlwAVuqdwSCAmKXo8hM7ujJIcffrgBHkY8YSDd8kcJkB9Cz4yRYgAhMOX1Fip8RdoqmOkWQAM2WdXbp7B56aWXCmvDhg2lUaNGZstne+V7rsf59Pf2+fQ6uo323jgXz4fHkWdFT9ALQpya0/pHCbn/gpaA68gFLeE/nt+BogPFVAs9q4bScWeuVaJZ4d7ORGiZ8DODWUMtgCM566Sv7Ny58w+HkEI1efJkl5L0B6lUzn9CgiL5CAyw0OXcc88tA4JatWrJmWeeabxI+v2uvGXqAF55iGcNaCJEy1ZBDviKFrwU3tjyO2CPwSzkMNIbbN26tbkOINquXTszr2XHjh3Na5qYOJ0VD6a9so/yJP+E37Rt29aALANYWrZsac7dpEkTcy3u2b4H+xnCPYfCpQ2NQDLz9tGrddMoJa52MEL+448/TtwF3ZVc6Pl9B4qpCoquejoJ+CWBP4Ei4AMYMkTeLeElMHr0aCE/Q8PJbBXwyoMqvlcoYwtIAoTkgwCDTD0EAGKAuIZOdcN0N7fffnvIle/CreF+w35GV998880GMrkmMAlEAnrNmjUzXkr7XhUMQz2jfsfxyENzGznXu+++63Iaw6uTL9+gk5SZWxInAdujSC6y3yvv6p41a5bcfffdZd4hPER+r+jOm2++KY888oh89NFHZvYFQsV+ry70nJjQc+JqgLvSriCBP4HiW2+9JXiYunfvvis8f0zPiKsd+RBi1Ry9aOEQeNIVTyENOlCIhw+vnwKhF+rCwR+AZ6/hjmN/NMd5r8tvAFS8k0zGCjwykIXwNmDLc/M8CoehvKZ8Z0Mj4MjbAcjfdIv/Enj44YeNN3vRokX+n9ydMaQEAMUNGzYIYb7CwkLfV+auY/Jj5kolBMg7voNYeUsHtg1Q/M9//iNMexLECoQyQTOdYWxOJLsVz3ecG5tl22e1VX5usW906PVVmnbOn9+f9957b+nXr19IPXQ7nQSClMCfQFHDm4Q66c26RUz+xlNPPWXCs8COHVoOBUjsU0jCUPE/8E1yMB47wBAvHjCmgOY1igp33v1B/c/1wp1bDbqCI2Hsq666yoAj8zDqs6oB9spE9yML9TTyDmLyW9zijwSYlYDOB95bXr3llsRIALAC5JYtW1Y2cEgHEPmxZT5TJkXm1ar9+/eXPn36BLJ+9913Rm8AUkbNkmsdxPraa6+ZcD1pL0QwSIsJYuXcRGjo0GOzg1pJNSIViHkLAcPddtstkJVzM6D0iy++SIxiu6s4CVgS+AMoktRKCJTeHo3OoEGDrEN3vY+M7MIDBuhgaOzeqReGvHDIb8gtpFeLIVQwBMYUvMKBWbrsBxzpTeMVBVDwNiIjW07IwZYV/6uXEWjk90zV4Jb4JMDcZ3i6qbvUYbckRgJ9+/Y1rzvFsxjU+uSTTwreYmYX4HPQK9cJcqXT/fTTTwuvjg1q5eUBlAfXIVea19QFtfI2M7ykjCIOcqXcXUpYYuq1u8ofJfAHUKQC09Ojh4THCK/Prrgw6vuuu+4ynsPyABHwYQWOACAAG9kpHFYmMAwHsIAvHkmgD1ChJ89Ia5WJyigUMOqoaTzZ9kSvu6LexfPMpAjQKWFQFaDo5jaLR5rut04CTgJOAk4CKoEyUNy+fbuBRECRV/jQ4JNLZ8/urj+qrFtyjZio2Jt/aAOOflbPGDCEJw25RYLDSKHdcACWjvuBRvU00uHAq4rMgGgvMOr/6mEkjINXjBwst0QvAQagIUtSArp162Z0kRCiW5wEnAScBJwEnATilUAZKI4cOdLk0REivfPOO014FO/YJ598Eu81Uv73O3bsMDPN09jao5cVCu0txyj0IB9gCE8aIFhZQsp+ACqyYEWfAGjyMzU0rfILJVedYoh7WL58ecrrTirc4DvvvGPyRclju//++41XEd1kcIVbnAScBJwEnAScBOKRQBko0sAwgIWQK406HglCqKyVeeH1RoRK7RCzDTD6GUDEe8j/jP5lUAojlQEaB4h/nprH9qAiH/4nwRx9igYYAXZkjncbkHdLeAlQXwk708EjBI28AUU3QjK8zNw3TgJOAk4CTgLRScCAIhP10rCQ38TgDRpzgJHRuYBRuFnbo7tEah7F+1mBF4Ak3ChmQMX2IDJgg6liCK3acGhDkR/euIqcg3sItVbkXEH9RmWGXgE25NLZeYwK5WwVzAF4QH7gwIGpqUhJvismNAe8kSlz7ZHrSR0mFQIPo1ucBJwEnAScBJwE4pGAAUVG7jF9C4NXaHAIp5Kj2KVLF+NlZARZZVrefvvtMkAEVIASG1IUVDREiqc1FCAGBVTe8yoAevf78X8yIFeBkZA9uobuhQJGBXUNRxPCXrlyZWVSxbifhZGQdGA6d+5sOnqAIp085Eq+p/d1nHFf0J3AScBJwEnASWCXkoABRUKoOo0LDQzTnQCKhAp1FGtlkEpmZqZpTBk0UR4gAomEmBkgQJ6dwo0fcBbpHKHAjWvb18ejyUq5AVusQBQwy1ZX/qcM+Z6QpE7qbYOnfd5I9xXEd3pt7g2905C0F9z5n/LQEdK8MtEtYkLyeA6JADAIi44e4WfSSOjkob/vvfeeE5WTgJOAk4CTgJNAhSVQBQ/NBRdcYBoVRqjikQAUAUf+p7EhPJvuc93xDlyAI1yYWb2IACSeGGQAIAYBSPY5vWCo8MR+wA7QIx8SkALaAVfAClDEi0S48b777jPzrD3++ONir8wj9uCDD5pQJBBBKJJnAhy1M8A5AUr28R3X5R70Pux7DeqzXgtvNvl2TK2j3lzb00v5UT6Eo7mXXWlEfqgaPnz4cBO+J2UEWKQsWfV/rcuhfuv2OQk4CTgJOAk4CUQjgSrz5s2TTz/9VL7++mv59ttv5dFHHzWhLAazMK8d+3v37m3eDhDNCVPtmLy8PAMVkQarKIAwHRANrg4ICAqM7PMqJAFpQCCADhTiFdRRrMxvyVsSGN3Ku1J5xRYr8Muq/4fb6nH2sbzDlXe7MlHsE088IV27djXwyT0AGqxAqjcf0753vz8jC64HuNJBCReOVu8iOY68o3ZXXair2nGgU6GQSLnxP3oMeFPH3eIk4CTgJOAk4CRQEQmUjXrWH5OvCDC9+uqruittt2PHjo3oRdSQJkBCHiJwBvwovPkFQl6vIecFDLmeQhn78AYCb7wPVeEO+ON/AJF3rwJ49kpoMdrV/h2fOR/n5fxch6mQuC7fMdqYECbQAbwCIXj8ADmvXEI9n/eYWP5H/oAOwMwrAikfysrrXcQ7jLebNy64RYznGG/sjBkznDicBJwEnAScBJwEfJHAn0AR7yG5YryKKJ2XXr16mYmzgQkgIxRo4JkCRIAgwM1vQFQ44rzAlIZ8Cfvyf/fu3aVnz54G0PDqAmw2DIYCQC/sxfN/pPMDkEDjZ599Zu4JfSDEzTMgLzxZyEyfMYgtcuN6hFCBRG84mjLVvEVkunr16nRW2bjvnfxEQHHSpElxn8udwEnAScBJwEnASQAJVDpQLCkpMQn9hJoVEm1PFJ8BDKCD0aKASBCQo+fEMwbEsNKQ815QIAzvHV48BUOFtnjAz+/f2vek4Mh9v/vuuyZcDSji9SNUHGSImnMDpjqdjhf6NRQNJO3K0+g4UHRG3UnAScBJwEnAbwlUKlBkvkdGgRKOBB68gMj/hDHxmAI4QXgR8YIBNgxAIWQLMBIaBeLw0AFcfLYhzG/AC+p83LOeW0PVjEB+5JFHykLUhKoVkv3eEvomRSBcKJr9vH4x3b3hFa3kDhQrKjn3OycBJwEnASeBcBKoNKA4bdo0E3aL5EXkO0Zyay6inyADIAKeDIbh/Iw4xvMGHBJSVsCyYUv3peNWnwOPKNDIShiddw0TKgaU/ZQv51IZk9ep0G97F/kMLNJRIEy+qy0OFHe1EnfP6yTgJOAkELwEKgUofvPNN2bKFADBBgf1KLIPDyMAQygY4PATYhhdCrywZRAQeX12vqFCVToCYXn3zLPp8wGNgDHhaUCZXEaAMdQAmHjkT34n4W6dqNsucy1rdIFwdXFxcfC1KEWu4EAxRQrC3YaTgJOAk0AlkkDagyKDViJNfQMgMoobaPE7j45QKIAI9CggElpWcNJtebBVWb63oRFgxMvYo0cPA3XI309IB/YBc1INKONwsMjr/3aVQS4OFCuRZXaP4iTgJOAkkCISSGtQfO655wwkekFBPYmEIfE64X3yawoXAAVAJP+QSax11LINiJUF/OJ5DoVkDUuTpwnYkRvqF7BrKJpz6iTdWvZsgUfSDchJnTNnTopUueBuw4FicLJ1Z3YScBJwEthVJZC2oMicg3gSQ0GiAgL5iECdX6FmYJMcRCbCfuutt0yI1R6cEg9YVdbfKjASisfLyATfGpL2s1wYVY7n2KsPqguA4+TJkyt1PXegWKmL1z2ck4CTgJNAUiSQlqDIW0QiQSKeRJ36xg8Y4RzkuwE4TG/DnIfk49mh1soKen49lwIjYI38eLUgOaN4Gf0qIwYRhZpCR2ERvajMsOhAMSk21F3UScBJwEmgUksg7UCRt4VEgkSgwK9BKwAM8ywSZgZsdFJsB4h/fDtMLDCpwIh3kVHh5HeS5+lHOJrywoPMu7qBQnRBQ9F8Zh+DXCZOnFgpK7UDxUpZrO6hnAScBJwEkiqBtAJF3kmskKgAoFtAgJXQcLTzIxJKBlBC5S+yXweqvP322ybMDBAp6MQCR+7YP4MlcsQrC3wTjsZjywjpeL2LCvcMYgkXhgYWK6Nn0YFiUm2pu7iTgJOAk0CllEDagOLTTz9dNpG27SkCFBUS8fxF45ki3Fk6nc31csedneS6jh3+MAE3XkS+Z7AMU71omNkB35+Bzw+ZkL/IoBe8xYB+KHCPZTodYJEyJv0glGeRAS4MflmyZEmlqtQOFCtVcbqHcRJwEnASSAkJpAUovv7663LhhRf+yUOkkIjnSMOX5QEFeYZP9uguW/KWSl5OiSyaWyDFxSWybOVsuevuLgYuCF8ChwAMIOS8iMEAokKmyhcoJwcUWIw3dzEaWLziiiskJycnJSqiHzfhQNEPKbpzOAk4CTgJOAnYEkh5UPzyyy9NuNnrGVJIxJvI9CjReKGAxO/695YNq0SanztVTtl/lJx6AOtoubnZHNla9Jv85+N3zWv2nBcxWDhUSLS3ACNwzsqci5RXPKFofst5mGvRqz/oDSFovNDbt2+360TafnagmLZF527cScBJwEkgZSWQ0qA4duxY08B7G3kbEvEkRgOJeAlffe15WbXsdzlp39FS77hpcumJU+WSE8dLvePHSI3DRkr9ajOkuEDk88//YwZa2BDjPicGHNW7yNttHnroIQNy5JyW5ykO9300nkXSFSrD4kCxMpSiewYnAScBJ4HUkkDKguKiRYvMRMnhIJFws+YkhoME3Q9IXnvtNVJUslHqVZssF/1rlOy31z/ltMOfkgbHz5D61cdJq/MzpNmZGdKtw3JZtzFTCHc7OEwMHIaTM6Fo3njDIJd45sNUWCTU7NUn9IgBUuRHpvviQDHdS9Ddv5OAk4CTQOpJICVBcceOHdKsWTPzVo1QA1do7JkCJ9rRzRz3wAP3yoJZhXJG1dFySfWJss+eB0uVKlWkbvX+0rbOCmldc56Bxfr/mi1FRSXyxhs9XW7i+8kFRQ1FkwbAACNGRlc0FM3vgE3e1AMc4pVWzzT/16lTR954443Uq6Ex3JEDxRiE5Q51EnAScBJwEohKAikJijTojEwNB4mMZo3lvcE33XiTvPBSDxn01SY57x8/yqUnTpG61cbLbrvtYWDxqlpLpE2txdK6ZoZcfOxsWZG1Td577y0HikkGRTyNwCKgyDQ6vBEnnrxFYJFJuXmlnxcW6Xygc1OmTImq4qTiQQ4UU7FU3D05CTgJOAmktwRSDhRffPHFkHMlAo007ryWD+9SLJ4lPIoPPtRVMqYXGI/ihceOkuZnz5cWNaYYUNxj9/3l2tor5apaCw0o5uVulTffdB7FcCHhRO/XvMXPP/9c7rvvPpNyoGkFFdny7m+mx7E7InwGFNGvvLy8tKzVDhTTstjcTTsJOAk4CaS0BFIKFAcPHmwaa7w7GhrULZDI69k6duwYEyQCEuQotu/QTkq25kqNQ3+Uy8+cKVfVmi/X1s6WS07/xsDiQX85U9rUWCt3tFomW3JXymuvv+ZyFFPAo+iFUjyLjz32WNywyLuhNfSsOgYsMhIa4ErHxYFiOpaau2cnAScBJ4HUlkDKgOL69etDvqfXbszJUauIB4nfMCCi5xsvyNyffpbaR8yWludlmFBz+ws3yHn/etrA4hEHtpSSXJGv+/SWd955J2VAEY+ad7UBitAsK+9R9q76nR7vPY/+r9+nw5ZBLs8884yZb7Gi+kDngfka6YDYnkX+Z3ALcky3xYFiupWYu18nAScBJ4HUl0DKgCLh4Uh5icyVWFEoYPoTRki/+MILsnT5XJnz03ZpdsY8qXvMbLnk+NnS+IR1Uv2I6w0sdrnjNmHuRgAqmdBkX18hkLeX8I5kvGps2f/WW2/Ja6+9Ziaq5u01PXr0KFuBqVdeeUV69uwpvIYQ+AGyWDkX8xWyz35O+7r2/lT7zPM///zzZo7EWNIQVIf4DXmu3pHQQCMebTyLGRkZqV+DrTt0oGgJw310EnAScBJwEvBFAikBir169Sp784qGAdWTSKPdqlUr82o+beRj2eo7m998800DRkDViFEDZecvuZK1oFgmjcyXdatK5HfZKWeeeaaBxa5duyYcFm3PHgAIxAFzgBygBwQ+8MADZkAH4Xfy7IDrzp07m7y9hx9+2BzDm01YeX/yk08+KQ8++KDcfffd5ncMEsIrS47nXXfdJd27dzevKcR7yrUUHhUKUxkauTeAl+dkBHxFYRFZNm7c+E+DW+i0MOhl586dvlS0RJzEgWIipOyu4STgJOAksGtJIOmgiNeG1/MBhHYIEFAkDMg0ObEOXlGQJLyIJ5G5+IAgBR+8cD17vi6ffvaR9P2mt3z0n/fljTd6Se/eveWQQw4xsAiAMOmz/kbhye+tnh8gxEvGljkcH3nkEQM/wB1Qh6eQ+/76669lyJAhMmbMGBk3bpyMHz/ebJmcnH2jR48uW/mf/Ryjx40aNUp++OEH45UEmh999FGTw4mMAY2nnnrKXEfBEWjlmfU+/X7+eM7HPQGLyIZyrggsoiukJaB7tv7xGVhEJumyOFBMl5Jy9+kk4CTgJJA+Ekg6KOItDBdybtiwYVxz53FugA8A84IO/7/77rtm5TOretSYX5GV70P9Nh640d9yPSCM87My2hvQAQzxAOIV7Nu3r4wcOdKAINAH5I0YMUKGDx8uw4YNk6FDh8a88jt+z3mBSkCSlYFEPD/QxX0wjQyTUOOJBcbwcHLv3Lc+Q6psuT/e4kJ6QkVgEa8zXklvZ0XzFfv06ZMWNdqBYloUk7tJJwEnASeBtJJAUkERzxmQSIMcKuTM6/nUOxjLFlig4f/3v/9tICdauOE4oAMPJKBYtWpV42XEyxftOSLBk56D85FniIcQKMQbxrQveDABOKDwxx9/NFAYCgjZF+9qQ6YNj1ybFa8lIW8gihHChKm5b+STil5G5InnFe9grLDI8c5kKaUAACAASURBVHa+ouqiehXxahcUFKR8xXagmPJF5G7QScBJwEkg7SSQNFDkFX00xF4vju5r0aKFabxjAUSOpdEHvJhCBahROIsEcPZ3HE8Impw/YPH44483nj37mFg/6z0oID777LMGaIATvgMK8eoBiQqACnL6fyK29jXxOuJxJLwNNDJwRCe7BnDxgvI8yEKfL1a5+HW8Xh9YBGzxhlYEFsPlKzIK+oknnkj5yu1AMeWLyN2gk4CTgJNA2kkgaaBIiJWRpYChenDY8j8hZ76PtbEHFBmswWAUoKGiIAJ4fPXVV2bACLBIDmW/fv3KoIgRxMASK+HpcNdRgNGQNvl/eLzwHpInOGHCBBNOBgJtSEsEFJZ3Db0ftkAjMMv9Aoc33nijATJGVAPj7NNnDSeLoPfr9Sl3PKBAeKydDI4nfI2H29ZLOjPowNy5c1O6gjtQTOnicTfnJOAk4CSQlhJICiiS80XDGy7kjOeqIo08AzJY8Xb5ASbcJ/cILAJ4TJvz9ttvysBB38ikKcNlzLjBZkAMIVoFFb0u/wNQ3AsDRgBYwrcAIgNLgC+FsfKgLRW+5165D/IkuX/ektKlSxczAboCI8/ulYPKI1FbwuKAPKkHFdEhDUHbuqlebiAylRcHiqlcOu7enAScBJwE0lMCCQfF7du3S5MmTf4UcsabiOeGee0q6g1iAmU8fMCCX8DCgBLCz8DiK6+8KL/L7zJ6YK6899x6+eq9TbJqWbGs3bDYACTX1uvi2WJUMXmWjz/+uAnfEl5mMIpCVyoAYKz3oPeug2yQD6PLWQFjvKfIQOWQKEC0r8M9MJ1QRUdCM/WQerfV2w040rnB05yqiwPFVC0Zd19OAk4CTgLpK4GEgyL5eZFCzjTSsYacOZ4RzowcVlCxwaGin4EdoBM42P+AvQwsXnHeOGlycqY0OXWOND5ljtQ/bo68+MAa+fX3QjPFDqODWQEnwucDBgww+Yfp5kEsDyC9wAgkUwZ4TRmU4yesV6T8CIkDTuhTrJ5F8hx5Fjt/Fq+izq24bdu2lKzxDhRTsljcTTkJOAk4CaS1BBIKiqtXr5batWubBlg9Neq5oVEmXAhgxdKwA4ma94cXz29PFpNR9/3mS9mwWgwo4llsdf5caVNriXkFYKvzM6RBtdnyXNfVsmHTMjMpNp5Nfjdp0qS0CzGXB4je7xUYdeALYXa8qITj/UoBqAgo8huurxOTx6pTpDCEmoibgS0MlErFxYFiKpaKuycnAScBJ4H0lkBCQfGOO+4IOWciYb3LLrusQhNrM7CCRh0PUkWBItLvmMJnY06m3Hv1Oml46kgDi/vtfai0rbNaWtdcIK1rzpOras2TOkfOko3rS+T++7ua0csM/lCI8sJVZfxfn5VR0gz8AdCYNFzLxW+Aj1RmfMf1CIUz4Kgi+Yp0QMiVtb2KdGr4H32dNWtWytV8B4opVyTuhpwEnAScBNJeAgkDxfnz5xtIpKH1ehNpeCv6Lmd+x+AFQp3lwUOs3wMbvLGlqDhfGlSbI9fWXiUXnfSBgcX/O/As6Vg3X66qNV9anDtbzvzbcPnus42ycNEsk4+o4FQZoTDSM/HcDHjBw8jrAwm/E4pPxshoyg9QZbQ5Xs5YUxrCDWwhdYL5L1NtcaCYaiXi7sdJwEnASSD9JZAwUGSEbKjcRCCRCY3xDMYaHsTjw0ARQoxBeKw4J6Occ7bkm1xEwsztLtwgZxzzkIHFEw6/Tlqds0rqHvejtDh3ifR+M0cWLJpu3nISCaYq+3fAIjmZjI4Gzq+88kozsCeocorUAaAMyZkk77AiOsZo9UsvvbRsuhxyFdFZ8hWZCzSVFgeKqVQa7l68Evjll1+kuLhY8vPzZd26dbJmzRpZunSpWWfPni2sTMFFyg4D/wYNGmRsKW+NYtCcrjgFQq0MHoy0Eh0K9Tvsh56bmS4GDhxoOvtcn5cPcD/cl95jVlaWsHL/PAeT8RcVFQnP5xYngcoogYSA4syZMw0k0sDa89PptCMVeZvGzTffbKZmIS8xEijE+93rr78mW7dvlBbnzJeW52UIsNixbq4cd1gbA4unHvmw3N6kWJqcmiUzJxXLxEk/GiNT2WGwvOcDFlkJwXMsnl9eD6ih6HjLJZbf0zjgdSZ3NJbOCMeSM9uyZcs/hKDRWzo9vOIwlRYHiqlUGul/Lzt27DBQBxDRKSLdQqf2AqwYOEguMm/A4gUF99xzj6lfTHiPB58BYTgByPUlikTnipU8X+oPW9Y6deqYGQXY8j957PqZrf297rd/q8foPrbMUBBq1WP1nPob3c9vvJ+5H47T33CMXkOfQ5+N5+R5eW6eHzmQhtOpUycjH9JxmJEBaEV+gCn2kZQd5Lt48WIhlz8vL0+YIcQtTgKpIIGEgGLnzp3DehOZDgfoi6UBJ4RIo9+rV6/AQ5q863jq9DHS98MtUv+42f/NSZwvzU5dKlX3O9nA4iVnfCU3N8qRX34tlAED+u9SuYnRACNTAgGMlHG3bt2SAot4M3kfNB7CWHWNhs/2KmquIo3FqlWrUqEem3twoJgyRZEyN/Lbb78JwAd4AHykAAF7zOfKXKjYt2eeecakiWCn0XUAhynM0HkFIhugFOQUmDjGXoGnUKvCVKQtqUneFQdDqJUOW7jVTm/Sz+GOZX+o87PPey/h7t37vCoPBUlbjvqdAilblanCKMdw35QDHVVgk3YPG0Z5UW6Un87LO2/ePAOYubm5snPnTqHc3eIk4JcEAgdFDBOVhUqnFZatVs6KeBMxZuSIBTHK2eupInTZs+frsvOXfLm/4wq5+NhZUueY0dL83LlyU8Mc2a3KQQYWJ4zKkox5s3b5sHMocNRQNOGkBx54wIAa4WCvrIP+nzxJQJEwdCywyPG8UhIdRm9Vf9FrPASpsjhQTJWSCP4+8Dbl5OTIihUrTEiUzhjTeDF4C4/Vvffea+ajxZOPd6tRo0YGehRYFEjYsk+Bhq2u4aDIhidvneCtWlyLFc8aa9OmTc09XH755cJKXWIlJYUVEGLAGSseOFbuW1dSjFh54xLtha7UZQAq1EobEWoNdSz79JxsuQ6rXlfvQ+8NJwX3ClBz7zyDPhOOD54RmQN5KgOVCfJR+6FtIPIMJWu7HLR8FDKBdIV3BXbOwfm5PvdKjjVeXiI56AUvjCAlCM8leoP+pOpUX8HXIHeFWCQQOCjSA0LJtYFVWMTAoNCxehMJBVJRcdsHDRZ6fqa6+fDD92X7jgLp+9FaaVtvgVx5znJpf/FaebX7Ktl3v91k9z2qGK8ZE1EDRqGAaVfepzLBm0GoCiOWaFhkUM1zzz1nGqBYB7bQmKDDqsdsMfAYbnKuUmFxoJgKpRD/Pfz666+yadMmyczMNPlxzCLwxhtvmPeNU8boIoAChKCHChEKfwp+NmgoiHihRKFPgQ8vIrDBLBSADpCHnW7evLm5JnCE/VWYU3jjngAzBq8xCwX5wLpi4+ls6YoN1zVUh426Wd7q/V15x4f6viLn8P6G//VZ2OozsuW5WVUOyAX58E55hVMFUuSpAErZAp5Apw2cwKbdfmrZadmypcx1q3qBPihUqm6gN+gP1+FemJGE99mjZ+jbxIkTZcmSJbJ582aXexl/lU77MwQKiigZyopCawOrvSkME5WE3I1QlS/UPio7vTtAg0Y/iAEsCoe61WtwvSuvbCFTp42Vrdu3SElJgRRvzZHc/LXyxRe9jVexatWqpremk2vvymAY6tkVFkkMJ3yCMQ06x1TLUbdcT413KB0Lt4/f0Fiit2qstZEGPlNhcaCYCqUQ3T3gzVm4cKGZHYBOL15Ayg/7Bhyobmlj74VAYEBBIdxWdRXwU88e0AccACMKfNhhvHOAHhAD0NCRY1XwsetFKOjSffZx7vNtYds2lZd3a8tMZa9lAXACmpQT5UW5oS+8gQqAx7MJXAL3lLd6L1U/bKDUz6pfCpPqqWQ/KzrE+bgGA1KZQYI0Hma1QH+3bNkSncK7o9JaAoGCIkCHstmQSCOL8mEMqQB2xSjvM8fTYCdqQIRCInBBj49BEXgMyUP84YcBZssIucmTJxtPFZNxn3XWWTJjxoyyibYVmBSS9P9deYssgEVCIhifRJUnsEgZUq4YVox0eTqn33MsPW9019ZnDC49fXLAkr04UEx2Cfzv+ngF165dK9OmTTNvZ2LQB+FgGnZsIA3zBRdcUDZAgv/V2xMKAtE7Gnxb/2wABBCAPxp0rgFEqIcPwFDbCXzYOq2gQoddP+v30WxtbxqfQ/0mmmO8v/P+xv7fe6yf/9vX8X6O9jrR/M4+JtrzcpyWkV1e7NNzUL60kZS3DZaE1BUq0RM6vbTBeI4VKLFrqmc2XCo0qo4qTKK/7EOfOTdhbkadf//99zJ9+nSTE/vzzz//r1K4T2krgcBAkWRaFA+jpx4YtqqMKJat4Kro4bYcy29eeumlhHkTAQsgBoOLoSdsGgr42Acccu/AIvkrNBDsByQHDvxBBg0aaBKPBw0aHPIcuyI4Is+uXbuavEXkrGCu3r8gtlwD8KfRxnsSTt9C7ccA45HBmKpOo88YTpLLk704UExsCTAdCoMHGEjAFC6MrGckPN4ewrY0tjpIAR2hwbUhED2yG2b9H50iBEwHhPMAgXiL8ACSN0ejj01Szx96GQ4AFSxC6bO9r6Lgouewf89n3V/Rrd53pG1Fzx3qd5Guo9+F+l0s+/yQkX2O8q6t9+3d8jvOo0CpIXH0Fv2iswFM2t5JHcxn6yv6rfqsMIme4/1G7/kOEEVfqRfUD9rDjIwMkx/pphNKrL2K52qBgSK9CnobGD1tVNmiaCgPvZ3yFN3+HqXmN7i9gwCIUOcEXjDAhEnDQaICHlDINEA8I7AIAE2b9pOsWrtQ8os2SF7+ZikoWisLF88w3sghQ4bs0nmMyIsQPQNckDEhDUJwiYBFyprrAPQYUVvPIn3mWDw16LDqNVsMIh7nZC8OFIMrAUJsc+bMMflbdFbxhANuwBx27vTTT5fTTjtV6lxYS2rXrmn+p7FU+FMPjW6xE3hysIV0PmwQpMHWEDB2j1X10tZX+7N+790qWHj3l/e//o5tecfyPffiXfV3ei7qOavacgUUnhXwxWPPswMWrNQ1HcSCk0BX6hqr/m9vdQBKpC3He3/P/+qN1fPZ19d74v64T1bu287J1LLiGfWZVQZe2URTdvxWz6NbPV8022h/o/fGOe37ssuK51SQRO9JX0Bv6cioR9ILkarrQCR1RCGS//kN52Gk/fPPPy/ffPONGZRFSoZbUk8CgYEirm8vKNKookwoSDSKrsegvFRecsESkZsIROh0KrxdhDCpAmGkLeCDJ7FatWoGFseOnSIfv7xdbmi0UtpetFw6t8iWgV8RosyXocOG7PLzLdqwCLQxMi9Rb3ChfAErDKDqWTRb9JqQDXqsHSA+Y/zwKidzcaDoj/QZEYpuEkZDptgrGjY6BDR2rNg2/gcGzz33bPl+wOdSsjVHspfly8b1BbJ+U6bcc29nOeusMw1MojM0rpwLiAGMAA30ScEC/QvXaIfSzWhAQI9hyzl0G+p89j69DwUHfucFB55B4QH7DGhRj3lGPiuMcYwNvoRNGTxBqPK+++4zczAyDyODKVjJ1yRtCXv/wgsvyCuvvCIvv/yy2TIlGisRHuyFd8VrFc1KBMD+rZ6XLeWu1+P63AfOAu6J+3vyySfNbAfcM9N98Rx33XWXybdHRgAwzwtMUtYKochF5YO8+A4AVV3AEQJscg5WysMuBy0Lu5y8n+3y5bP9v/dY/o/mGP2d3ov9O+4XG8qzAtY8I50eIJJOEDZSgZH6op/Z8j/1SPNvsaF0vDgH+kG541ChProluRIIBBSZNBRFsBtTFEa9MFSiaJReFZRKQ2VKxMAHIBFYQUm5JhOhAoc0HJEgUY8x0w/Mnin/qna4gcWWNRZIh3prpd3FS+Xai5ZK09MypVv71fLLb4UyeMigqM5b3nXT+XvkytQezAdGuEM9xkF7FslVpLGgQYtFF9FJfoN+qz6zxeDhFU3m4kAxNukz39yCBQtMThVlB9Bgp2wYpFxpwNSeYdN0pYFrceXl8utvO+T5B5fJCfsMkmp7DZLj9xgsDapPkMyM32XGrHEmH9Zu+DW/TO2bd6uNe7SNeLTHcR3VdX7DPQGqNtgALgCMetX4jFyAAaCAe2dqMqa56t69u4E6YOrVV181APfuu++aiA82lLqMzfauRGpY+Z6VSAIrTgBWfhtqpc4mag11ffbpPeo96zOw5Zm8z8rMDnzH7zQ/mpHFyOvFF1808nv00UfNPJbIFQ8b5WJ78OyyoHzUm0m5eTsaXl2K9L/qWXnH6HG6DXU8emXrNc+gOZIKzOSFE84GIAllU4+oV/YKPFLfqHdaD9liYwFR8tq/++47U2/dhOSx2bt4jg4EFHElU9DakNqeF5Jo1WiGUjjvPhSQipIobyJhSSo1Sg28ADHRQKKCIrmI27YVyIM3LDegSBi64yVrpGODVdK+3lLpUH+pXHFWpnzac4usyJ5vXlOVzqDnx70jX16VRW8e3cDghkoF8HsfBhzDHEsaBPpIRwcPk+o3WwwcXiMmN07W4kAxvOSBwmXLlpk6DdhQhjRWtjfDBkLKVFcFQ7bYMrwepdPGNJXcvM3S5sJpcubfR0i9amOlwQnjpEH1cXLxceOk+j4jZEjfPBk8tJ+BLbVtNLh2o2t/1mO8W+9v7O/RSVbdRx1SAFSvH4BBQ8uqHj7AD/0nX5f5QGmEsd141OhEqS2knigA2ZAHLClQKcB56ygdvl199cpEZaWyQ44KmjZs8lkBE+8n5QJc0qmhvPDGkgKhXj3KVctYy5nyBzoVKFVH2No6Y++PpGvhjgv3m3C6qSCMtxUHAbaTekX90vqmnTMFSdsDSb3lWCD02WefNfWa1yo6eAxvA+P5xndQZLQfrmOMrjakFCifKXAMlq1s5X1GAVF6DJW3wvn9PwaNHiCVi8/AS7SQCDDhJh87bqQsmFMkrc5ZLy1qjTWwWHX/anJzwxzpUH+FAcW2dZdKhwbLJSd3iwwaPDCma/gBZql6Dry35HY+9thjCZknE4PNq7ToiIQzmqH0k4aYnrGCg+o3xos30CRrcaD4P8mTU0heMcBDGIsOKg0NZeT1EioQsqVMtYGiXJlmhDAaHUf0hIYJTw7bPt98Ip+9uUHOPmi0NKg+/k9rvePHyfmHjZWN6wrk9k7/CyeG0indF02DyzEKB9wH9pF7U+8fHR/0mZAo4VEAEDgmXEtYFr0HTrzQB5QALwoytn0NB3v2Me7z+xVqo6KRrZaJejUVLhUwFdrx6NLhJo+WED5AiU3VDrFXX/ifThMgiV6hh164U91kq8fY+yJ9DqXPen7dcgw2Fb2l7UWXCdGr95F6qPVS66b+D2ew4nWkfvMbnpV6T3uycePG/xkF96nCEvAdFMnRwxDbjag2pPQYYvHeoIAYv0TMm0hlpbJRsZgknLzEWCAR+Bo8eIjMnjNZBn2VJ83PypSbG+VJvdM/MLD4z4Mbym2XlUiH+suMZ7FVjaWSvbxAho8YEvN1UhX04rkvZE3YftSoUabhoweNcQy68aGxVA9MJINnf4eBw5ih49oZYovBwigna9mVQZH34zKiEm8LjQ2Ap+EryoVGRMvLBkP1VrAP+4RnAw8H5ct5sFfoB40Z5a560L59O1m+Yok0P/cnqXscXsTxUt94EyeYsDNeRfade8iP8v1nm+TNt14ynh39PdtQja7deCoM0rnmfkh50PviODxKdKrwAtLh4cUACoHYMl1jAcCg65s7f8Vg0pabFyzt7/iMDthAiR4AlGzZD0yiL3gnyblkRDJwBTDS3uLlY0XvNOVAddWuA6rL+l04ndbjdMvx9m90v+o+W7634ZH7orNGp426TX3Vuku91pV9Co7Uf44FhskrJUKYnZ2dLPOc1tf1HRRJ+qWAKEgAUVcKEAOsShHtFsOIofNWBj//p+JRuahAVBBG4lYEevAoTpsxXkYNKJDmZ2UZ7+GtjYvlrOO6GVg89ejb5dbGJWWguCa7QIYNd9PlqKyBRUDxiy++MOEx3mVK2fhZ1va5ODeGE28LehatTmLI8CoBFraeo+MASUlJSVKMwq4Eihs2bDBT0pAnR/I8srfBkP+9YMj/7Gel3MiVwtNInacx0UYRPdBGK5RO0Ih16NBO1qzZKBcePV7wHAKJl5w4RS6pPlHqnzCmDBZrHj5GXn54pfTp94lpiO3zaYMIENJI4xnkXgBCGmkaSmAQjyA503gDsYW2FykSBKLrNlTYuu8+xw9sqSxDu9xD2VAvTKpOoU+0g4S56XTpK1epH+glK3pKJypUONvWb/0cCgr1O3sb7ji7LlIn1KOPzabuM/IaztC6rdCo9R1wxDZoh5Hf4GkHHNevX58UW51uF/UVFAk74/qlsbQbUD5TaORQUOi2ckT6jEKQ4IsSB10p8SxhsKkozDofqzcR2OE3I0YOlXWrC+XKc7L+m5O4TG67bJscd+iVBhYvPKWnXF8/X7q0ypbc/HXGC1KRaylcVbYtsiBkSN4UoRPKI5Sh81Mf0C96rJF00fsdRo1RrOi1dobQcwxS3759k2IHKjMoFhQUmHoJNGHoNbdQGwCFP8pAV7vBIB+RvEK8EjR0drgNm0QifqgyZl+oBqxdu7ayYuUyuey0KXJxtXEGEC+qNsTU8WoHd5ZGJ802uYo1Dv1RPntjvXzwUa8yGMUOom/a4OLNIYpBrhVhQzovmhOIblIHaNhZVe+9IBB0HdHrum3lAMzy9EdBEv1TXcQjSaSH8QJE3kjpoI1Gj9FnzX21w9jeOhWuPkU6LlT9U3jkO/W8cx/Ub6IC5JDb9V9tAnYCPsFusBKyxp48/vjj5mUaqfI61qQ0IBEu6isoMhs7BtxuPGlEKSSoH6UKpRCh9qEI9BiYpgDDGaSB4vyEb0gM1lHOFQEwIIfw1/pNS+WdZzfL5WdmGq8iuYm3NMqX/zvwdNOQXHLaEFkyR2TmrEnGK1KRa9m/4bqRVvvYdPkMrOOBxjgFWfYYTDoJhIzxKIXSxVD70E+8Pug6+q16Tq8VCEnGUplAkTc6MJE1dRMvAvLV/EJvGFllTznQEFAmgCE2Bx2iEaNMaFC0LCk//cw2VGNkf+/9jEdl+Mjv5aWHV8p5h5bmKDY+ea7ss+fBsu+eR0jDE2dL3Wpj5fSqI2VFFjmKpdOeEOYjlYbQH3qteYLouQIh+70NeZB1wJ27csBfEOVo6yHnV4BEX1V32UcHh44OHR4NYytA4o0khUPrWLx1L1R9tc9JPVeApf5HAkdshYIj9gUbgr3hWZk3NZmDE5PRhoS7pq+giLsaSteGEwPOivGONexMYWGMgwYFFIJr0JgMGDAgplHOoaALYBsyZLBs25ErPR/fKA1PXCItz8uUNheslNY11kmVKnsZWJw7d6bpwXB8qPNE2qdQyDGMygaqCJcDuXjj7C0DcnjtIPl/HG//NtI1kvkd98igEJLv9RV/GKwgDKEaP6asYBCWbXC8cOD9H+On+TK2rqPvyeiZpjsobt261QwgI28Ke4EBx3gD3wqA2Ba1L2wx9HzHZ0LJeHnpYCoY0jhRpt5y1UbLW6aR/rd/g326887OUlS4XS46doLUPGK0NDh+mvzroE6mftetNlhqHPKTvP3MGlmePU/QLxpW20Oo+uxtjHW/2zqAS0UdCKWvwCIdHdpS1XFSJRhUQx4ttok6SWoF9RN41I6b33WTOqx1njqrHkfaeOyDPb+jdvRtW4K9we5gfzgee0RKVHFxcTiOqvT7fQVF8hcQrhpy7elTGBSSVyEiGWUgkcmuUbqgIIHzotzkORHSArYqAm5eqCJXEVjMK1gta1Zul95v5kivJzfKiG93yKb1xaYh2WuvvQzUAXHlXdOGO8AQ+NPBNozSxhvKCEcqBBWQ3hSVEvkiQ4CrT58+Br4AScDSBkfv/afK/9wrz8EINgxRUEYTPUCO9IBtGIikn3zHsYQt0G8FRXSfzhKwnuglHUGxsLDQdNBIMcEDqCEh7IhtxJGrruwHDpE53gJGSBL2sr0Wocou1rL1nkPtF/WM6zVrdrk82eMRKSn+Va6tO0tq/mOW1DnmR1O/TzrsAen3wU7ZvnOTgUS1YWz1c1D6nCrn1WeNtI3mXqn73hVPc7gVmx5uDfcb9nuvEa3NifR8+l00z5nux+izsuVZkB9yBR515TuihLS51CegkfpLdEZD1lrP7PoXS921f2d/1vNyHeovvMLgGKIP2BPtcNp2BjuETQIecQrguGAeR+zWrrT4Bookl0PhdqOpoEjSP2E9LSi78EJ9VgWiF46yqeIFUZEABBSVHgOAVB60RQtRnAcP5chRw2TB4hmyPHuuzJo9ScaMGW0qDfMrHnnkkTJ16lQDbfZ1vZ+BOjxsrL179zbwB9Tg1aDHhuLiJl+5cqV5h2ZeXp6ZFoB54yZPnmxyTABhIJJyIPTVv39/A8Z+P3e08invOGQARNNRQEfwxgSlB5wXQ8bEtxiQUDoZah96SgdIDYzqO3kvwHuil3QBRQb7DBo0yIA2Rhh5sfXCocoTw41dYSX3iDxo5I4uUwcoh0g5hqHKLtS+cI0RAIpe4A2hQSO0xihK8rVI/P/ss49l6451sn7Ndpk5XqTav06WY487VNZvXG4mVg5Kb4Owh9Gck+cJtXp/q6AALCi4KTCwpW7r//q9ghznZxQ3MqYdILRJuB57x2hdwpz6thRymbFvulKPw616DFt9A4yeh/NyfiJjdK65LtfnPrgffR69V713+znYp9/rs/A7r2z4PxoZhvpdOu3TZ7TvWfVBcx+RL2XB221oo4BH2jccHtTvUHU1ln2h6rVODs75NUyN9xBW0Y6oAqPaH4VG7czSYaQdLSoqSrSpT/j1fAPFb7/91hh8hIqB1xWhY9hjrsuZ5gAAIABJREFUKXCOhfqBA1vB/PyMAlOh/fYmegEI4Bk8eLBpGJk+h/+nTJliDBmwyCvAeEc0+wcOHCQDB/4gTNpduh1k4JDvaJhoGHGDA38a2uTF6tu2bTPKSsI/+1n5TK8Hdzl5Fr/99ptZGeXFe7hJREbRMYQAKF5G7p1reZ8hmf9zbzTU3Gc4g+uHXnBuQiX0MmPp0ODxVIOiYJOsPMVUBkUmwkW38HwjJwVDhWzbbqg8bThkdLK38bDLSRsD3cbSkOixnI/fKxgCozRcyJURoOggUEADh+1QfdRRoh/9530ZMWqg3HDD9caryGh6GkU/9DOR59DG3d7q9XlmbegV9pCJwhLf8TtkpRNEA2IMFsDWkqOJDmB/aKyx9cgbe4894jP7+Ey54MFB/vyGEeCMwuVchDPZAowKjXTO8FYBe5FWgJB70t/a5yMCY7+Wj/vkPtALvVe9T71X9Ib7xCvObxnowf3RIeeeeAsLURHkgvxswFS5oU/ITnVK5W2XAZ91f7pu9Xm4f30e1Snae+oW9YmyYUAj9pV6SEcNoKMc7HqvZaN1uLyt2gfdcryej/LlGlyL1Bc8iGqD1CbpFruFnaeTyzHoJ+086TOVcfENFBkMgOC8Bh+BxtL4UnAYZ3qEVChVpiAqBpWUUVIKSYkEIuab5DmBRRRz5qyZkrVsjuTkrZWcnM2SX7heVq5aID16PGkqCyNpgT9GluORAQZzc3PLVryI4VY9ju85B40258EDiRHDS0IPmlBvLG+iCVpegAVlA8DRWFBeQeiBnpPz0whoI1We0eF7GjhyXmy9R+fxfG3evDmhNiMVQXHu3LlmlCSdRYwqNgL5YFxtmSlks1/lBxzSSNieBTXq0ZRNNMdwPhoIUl3wZFAXaPTpmOFRogGnAbPB0G7sVHfY0sCxpS5Rr/FOfPnll0Z/0WH93v5NMj/rc+iWe9FGm+elPvDsrNhi7h/PHgCs8+8BRypD6gLwhNzxutImaEdcJ/zGe8R5kcs333xjPDKk6mB3iCAQ4dDoCalA5a16rG6xF/o52i2pPOVdh2M4H/fHfaqdJGrUr18/+frrr42cVEYAK14y2jHgUYETGbGic8gJLxoAzHHAEbqDjGz5Iy/k7wVJLTe2ydQjv65tP4fWO56bTgeyRNdoM/Hs02Zip9G9IGwCIGl7GoFGr90KBY2EsulUzpo1K6G2P+iL+QaKGHUaAfUkquFnS6HGUpgoAhXG27vyUyFRRHqSGDy/chNjBSdGideqVcs0Kp98/JkM+VLk9ubZckPDVdL24nny9nNZklewTrZuLTRwZ8NhOCiMZj/gyHF4HBldymd6vsgdo4c8eJZU8S5ioNEh25D4pQv2edAJGjaMUTSQoccQsgBwVPf5jMeMRiuRS6qAIh5u4IIyI1wDIIYKK6uNwOBihNmSc4gealgZGcdiO7RMIm05Hw0198cAJq7FaE3mKsTm0DCjC2p/tEG2dSXSZ/KB//nPf5p6PXBgfxk+or+MHD1APvv8P6Zjlkhg1Hu3644XBvHiUF48E+FdPGF41gBmoAaIZkVmgA1QA/gBvzTkQBIhOIAPiELvAStdFdjYb4MWsEVajeZLh7Kf2KBkr6Hui31672xt0LVhVWXAVp8fOTGHH51/dA19ACxpj/BMASnIGzsErOB5xRMLgOCl5Xh+p+UWDiIpTy3/SPqait/Z942+Uh+1XsIG2Gl0EycUkQY+B2EnsCPoPeVAR5IOL3YLe6agaNsxDU+zpdPJPVeGQTC+gOKaNWvKGgJtLFV4kDhCjmS4vd9RIGq4glJiChAFo9JGMlThjES8+xXCmALkjLOON43KJacMkRsu2SIXHTdGGp8yVW5umCsPtCuUbdtKJL+g1HsYDQhGe4x6GtVTSU4jxomwDN5F5KL3Ge/zVvT3XB8Di3Gg541RtBs9P/UDg4SHFS+WVyfD/Y8RR18VdFTvAUUa00QuyQZF3rVKY0dyOJ1GVvUcYlTVNqiBVe8hNgLPProXlJeA8uPcGG8aFzxf3CuNrnrM8NioPlVUx/gdtuX22281dXrAt5NkWN+t0uf9zbJ00VbZun2DfPb5xyYUqdfya2s3rpzTC4TcF8+Ip5S58HQqE6BEw/p4bQASjmHCe9JUsAPAHgO06EQq+NnQxzGh7Gh5kFdRu5Bqv4vlOVVWoeAS+SJnZAx047HkBQR41SgzvLRAEXqsHVrC8dhsbCTljo2krIGryuKFtHUbveYZqbd8pmND/aaDie1GPuHsdTz7tYNJfaGDSSRJbZjaNN2yX20g9pAJvjMzMxPZHPh6LV9AkcR0Gka7MaBRQFix5idSyDS+9JYqaqzLM7woF94DesfqPUuG4cEIFG1dJ70e3SZVquxpGhYm7b3x0rXyYLutck/rPLm1ca689kiB/CaFJswcLQTGcpwCo/Z8yOEB7jFWGmJJhnzsawKLNPBB6oXqDZ6maA0KxoM8GgVE3WIkMCiJXJIFipMmTTIy0NCy3du2bYJtRPmM9xDDDrThRfHbI0AZAoeUJzDPZyCI+o8exeM1VF3xbgkb9u79qezYJqY+71ellrQ4Y700OnmOXPTPWXJz0yVSUvSz9B/wjWn8vb+P5X/bPvJMPA8NKCv/EyYGIPTtGugpsiYMSg4d3xMCJrcKO8hMCgopQKB6/LQelgdDfK/Huu2wiLIoT5YqP6CScqA8FNa1nDgG7zUQyeAcIJ/2k3LGC0bHmpCtDZDohq336JsNYbHoX7KO1ftV/ed5tD4TvtfOOzCtqRDR2vNojlM7xbm5BtFU5R21cdoO8D9OBNoDxiMQvcC7nG6LL6CIMtJIIBQEpCsCgryjEb4eg1FH6Sl8VQS/FZLKQqOKMaXXppUykVtjVIcPkS2biqTFWWuleY2fTMOyz95V5cG2v8i9VxXKPa1z5e5WudKxHqHiYsnPLw0ZxwKBsRyrIWmUGONE2Wk4KZmNANfGw4kh1PCg3zqh56Mnji7G0ivlWKZ2oWNk6z49yUS+lD6RoMjgKHQDGNY5x2yvqspBt2owkQmheuq5ylgNr9qAeLacS0N3CoeklwBw2phQ1trYaLn7seWceOIKCtfL/dfmy2EHXmHqdMvzZ8lVtRZK65rzpNkZc+WqmguksLBA3nuvdDRtNNfW+1WbiKcIXWXFjuEJJyQMBALEyJbPPDsgQZgT+wYQUpc0FGrDoBdeEmkP3bViA0uVl0Ik5akeX9o0BpeiK3SMFCDRCdIJ8G4x2Ae9Ux3yetNVz6LRzWQfo3WD+4AbNBqpnkbsAPYGePbT1qid4rzYQaIi5CVjB2kLbB7is3oZYSU8+DiK0mXxBRQha4g5lGAoIBVoeVsKETjBqGP8glJAzk2vi0qmFS7RWxrZCRN/lJmTiqThiRnS8MQZck290jnYDjqwujx2nRiPIrB4fb1cmT+rWAqLggVFhUqAkdHUTLmjsIiskgWLXBfjx9RAVEoMQRCGjHNiOBm1SMUvT1/t7xklZ4MidQGDgCc0UUuiQJFwJGEeHbnsNYo2HCIHDCeAiIyQaxAGG9sBINEoYITxogFtlKd62OwGJQjbQs7eoCH9ZPywArn0+PlyxbkTDSieetTd0rbOGml1foaBxUurz5Gv398iEyaNCOlV1PvULfdKQ47es2If8ZwAAHgo8P4zQpTGn+m+yH/DAwUUAhDqHVQbZwOh7nPbyKCWyvIJVZ54Iil3yh89wA4R+SOMTR4qOY+027SDwCPOGToU1JdQHSpbF4OoO36d075PoJHnob7QicL2aNqJ38DI+eigUh+Zbk8Hv2D/QnERURc62NwPUJ/qS9ygyHBwwss8uDYQbBEOjQOCi6VQMPIoKgXul/LY58HgYmRJ1qYCJRN+xo0fLRkz8uWUA4bKfVcXSvf2Io3P/8A0Lice1Uqe7Ags5sp1dXMlc37iQBFgBBaZVicjI8M08BgcjGWy5MW16TEDKBgAu0z9+ozOoR94LWPNU+S+bGBC/zEEQEqilqBBceLEiQbCNPfGfl5v3ef5FRDpadMgBRFexrYAShhnGj7SJpA5DYR2NoOyJV69A0znL/xJnr5rtTQ7fa5cd+Em+et+1U19vuaCZdK65nxpXTNDmp89V+5qs0yyVy0yMKvnse8TPeQ5AD+eA+8IgI2e8ZyEjjkGKCQkSf0EBuhQ2bmCoSAilaHH3Zt/wOote/QC/UBP0Bf0BocF7S1RQeonHS3abM3f5TvqErpm66mtq7o/1bZ6j9QftQl43rEZapNiYRPbMRDuM+dTYMTJwuwXGmkJB4zYU+ROeaTqEjcoLl++3DSQtjdFQZFkTw0vhROsvR+DT2MXlMcIRUZhuOZXX32VNI8iFRgPXd++fWTt+uVyc8M8uad1vgkzP9ZB5PyT7zeNS61THpVH2onc1iRPdv5cJHl5ifEo2p5FRkVjVMjFABSSBYpcl/tg2hLCKRguNQR+GyjOTYfF1s1InzEOeM4xBGoM2GIACP0lagkKFBnkBOhpDmI4QNR6r9+rBzEIQKQ81EOAZw1vHlClDRq6EZR+hNM3QHFx1my5/7oVcsVZc6VNrSXS8IwBpi6fdvR90rb2KmlVM0NanJsh1zdcItmrlsobb/Qqa4DVA8L5mYWAEbDUOxpu9AibRR0gR41tOCh0sOUfbFU2WYaDR9UrOuN4uNA/PPR0mPE4kuvKwC/aT/TU7oQlup6Fq3/h9tu2gPsGfjmWqAMdL2w3NiqSjY/1OwVQ7cTiNFO76O1Us189jNzP0qVLE9VkRH2duEER4Ik0kCWWAqDAND8xXKHHux9Fp4HBLZ9MI0DvARjJWpYho77fJtddWJqPiAfxiY4ixxx2qWlgGp75pUwZLrLzl4I/zZmoOYUKdkFsuQYLjSBeDAxJsmARXaOxRKeC7Exwbrw3GMpoDARGAZDCGHhBkfMkavEbFHfu3Gmm49AQs/aMbUNnf8bgsTKXWFAhZsqD0Z40YITPdNQyDUCyGywmVZ46fYy8/9wGaXzKHBNmbl9nnRx0wJmmLl99QZa0rrnA5Ck+dttKyVw6u2yqE0CX+oW+4DEkjwxbR13TnEL0H5uljX0y7Ze7duWCUVun0DPaJ/SOz9hDPIzYQ3ST0fIMhMLTSHtqQ2O87XMifq92gvsHeplqhw4Zayy8Em3bQJ2mfeB1r7QP5QEjDgZmzGCasVRZ4gZFDFy4gSy4d6MRph6D+5Wwn3oF/FYaFBoFJ8k7mcBD7w3lpALisftdiqTP+wUmF/GWRrnSqVm+3HWFyJ5VjjMNzNyMmeY4prFhZHJxSVHpWswAl9KJt4OARPucKCzeYfIE8WQkCxYJn1DpMFB+6wfnw4hQLkxDQW9QdbO8LQbUO/E2vURGxCVq8RMUaSQI5wKJoQybDYgYPyCSvBx6xNECdnky9X4PfHJ+0kaoyxpeVsMfhD7Eck6gtfcXn0j2shJpUG22yUm8ulaWNDyjv6nHxx92nbS7cL3U+9dsyZi+Tfp9+4Xcf/8DxmPIs+K1YRJqGmnSYujMou92I+4ArXIBWqqWp61z2HvaLFYGYNBG42WkPjKymlHV2Ey1yalSH8uru3qfCoxEq+iAAoxe2xPv/zgTsIs4hxj4GK7TrbYUWGTQX6qEo+MGRRL/Q3kUEQTgpy7YaARNIwB4ag+lvIKO9Xt6D9wvk5tSAMmAHQw/72YGLHg7SimM5covvxfKutVF8v0nBfL+cwUy+Kvtsi57q+yzb+m0OUyOvSVnnczOGCs/jqXijpS5C8bLps1rzJtabKjz+zNeRa7P+7yBBxqxZBk4jBWgD/AHoScYDwwePWidpywa3aXXyHQvQJVCFJ+pB+qVDRoY/QJFerPcN8YKw8Wqz2Rv9TuOpbdMiDQaWcVyDPaDDgp6x4ANygfDTt1XQx+rHQjy+J69esnizJny/ac5csFhs0yYuV2dDXLE3/8bIThpprz55CaZOuNH0yjR6DJZNfaI1XkNw4Ogwksitsmyb6l4XZU390b7pTmODI6hveYtPAAWXnCcPECj2uZUrKPe+q/3iN2HEfCYMsgk1vEV0do12n7sGe1DqE642lUcDdhg8keTvcQNirhqtUHRRoQHRQA0tLGAIrRNQXkL0o//UQZyEyh8nUQ20aDI9cgvIsS+YMECA3hAhK4Fhbmy4+d8+eW3ArP9+ZcdsmLFCuON2HOvKtLv8yXS9ZotcnuTQrN2a7tOPnlvlMxf9JPxNPoNiPb5uEdGQiNHEuuTBdoYKSaeJUyHrmgl90NH9BwYOzoTGL9oKz/1wPuGFuoB3nZGjydiiRcUKWPyCqnPwB/3r3Xau9U6TrgdOdFbjqWuRytXOpt4dgnragMURJlr2ce75d4YVbpq3UJZOHOb3Ng4U5qcslAanTTL1OMjjzjGqEKfPl+bqAb6TAPMoAJ9F3yi7VIqwInCiG699wSgsALSeLhY8bjqihy9q8I3W+93/K+/Zavn5Px6Lbb2fei92Vv7+13hsz47z4p8kB3RB/4nr5H6SqeZOkCHTqODqVxntc7rPWpHlPkYAcagwtF4LplOJxQsYm+xsdhhHHEMjFm9enUimpGQ14gLFAEHwsuhRjzzoLEQuebmaCFp4fm5BSwoHMLOia7UVDCMEW87ILeKhYZZYUxh0bvluK/7fGEamQP3OkceastI6AK5u1We3HVlvtxy6VZ59anpsmjJTCkqKiw7n57Xzy1hbka5A7rq+Ui0HLkuUzxQidEVrdx+6gm9YaZUwEjEAj9UZhuwqOiMfB44cGDIyuf3znhAcebMmWaEHnU5nOFSWFQDRgiF8FO00BftccicXjcRBsJB1BnKhLIOorz91B29x549e8mw4QNkxao5sip7lezc8ZvJR+Qd0K2vai0zZswwYDhw4A8ydtxImTl7kkyfOUFGjRpmRjJjLxJdt4K+nkKG/WwKfwAH8AbUETVQuKO+MyE4IU9eL0rdx/ODnOk8MIqVke50XslvZ9APOkNEAM+QrnQs2a/fE+bHc45Hl/NQ34EadI3wP9fDY8b17fvSe+N++c6GyVDPF7RMk3l++3mRA2WGI4QyIuqDo0jHHKgDKF3qL/YGxxJTTgG/pDvF0hZEY+s4H+clp9tuN9TOKixij+m88/+UKVP8bjaiOl9coAg46HxB3odjWHgsOV4cy5tSggJFCp4BGeRWoMy2sUpEZeN69LzwvqxatcqEcm1QDAV0yHfj5jXy/VcLpMEZ/zGweMox7cw0Ory1hRVgvK1xsXz9xQjJy98cKChyv+RUYqjpPWJAEyG7UNdAX7S36mdDr+fiGWPxKFLpOd6GLICK3iBh8kQsFQVFQJZ7BRLZ2nXZ+1khkSmxYukIRmM4OQY50pmjfHVet1RvXFRnuE90Es8n9UNHzn/xxecydOhgA4c1a9Y09RhomT9/jhRt3SgL5xTJoK/yZPi3ebJ6RZHk5K2UIUNLPYyhdD9d9ilIcL+ABB1lQIvUFVb+J+wOWFDWTFsG2DGKnY4gK/pAu0DnmrxuAI/3HSNr5MybZYA6PLIKl5MnTxZWGlVddR/HsHI/TC3Ee6p1CiLaB6IVzLnHiFhGnet9sOU+2E8okGP5nUan9LnYApEKkLYM0qXcKnqf2qZSrrR12BVSvajPADpte1BRIK2Dfm7RMe4XdiC8DjBGa8eiPQ79xtaRumO3HeHsLkBJRybRS1ygyLtd1QNhPxiNCd6GWJLa8UygVNpb9LPAORcGHCNDD4fKrEpd0UoR6+8wHPSIacxZyoNEwJG8wEVZM+TRW5bJw9eKnH18F9PIXHT6MwYWeWsLo6S7XF4kz3SbJyuyFwoDXkJBpx/7uGfOz5aeEAY3Vjn4cTzXJRzA4AG/9YTzYSDoTaKT0YYdqPD0oNF9hS229ARpcBKxVAQUaWCjCTVTv3ke6jshdjx+PHO0BjGa4zgfXlwS5IEA6ixlEUQZ+3lOvUd0Bg8XDSNlTn0neoHnCXuDDQBY/vGPf5h63P+70fLYTdvk8jMypcU5WdLirCxpflaWvPzQBvn5t0IZPnyoASA/6kyQ51AY0i3XUiikU059Baior3j2aHRpHIkiUdbYZRpkoPGnn36SRYsWydq1ayUnJ8ek1DCfKyPwWYli0VnlM/u3b99uRocS6dC1pKTEpPWE2+pxjCrl93p+zmufn2twDuzd+vXrZcmSJTJt2jTjaUQ/8UgysIrn4HnQX0AXiKTsNSyrUEz5q4x0G2S5JOvc+mw8L8+OZxb4JkKA5xfAVjvrZz0M4lzUbUCR8sZLTW5hNLYslmPQG1iJc4fzLKr95XuiVHSOErnEBYoYvVAjniFj3KlUoGgFhnHFYATROFDYAGi3bt1MiAEjlshKRMWhwnB9tmp8yoO3oqIimT5zonS9eqPc1TJfuncQOfof9U0j0+qiASYMDSze2SJfHrlxncxfNNXAZXnnjed7jCavbwPqkzEXpcqShpg8GCqx3wYCfcEw4EWItrNDZcfDFmqKHELSiVhiBUUa8VggESOFJzEISMROkMaCMQa40qkhwWbR+CF/OhdAEfVcARGdxd6wxfaMHTNWjjzqIFOPLzzpQ7n9sm3Sof4y6VB/qbSvt1San50lj3daK8Xb1guhaf19Im1WNNfS+wIIFAzxJNH5oF7ifcOrysosAgx8AJ5wMACBgJpCnwIfszpg9+iQElHx2irsT9Cr95rcB/dDx537AzS5X0CS++czxzCnMG0ibQ3ASMcRLxSeSMLbeDDpOKAbyAu5qV6oLKORe7ocwzOxKjDymYge9oOOAavftjvI8+ERJc0BOxWtAyFa/lFYJEc8GljkGHQoUUtcoEjBQ7fqQVGvIqDItCCxCBNjEnTDD7jitsWAJ7qyEaYltw8jE8oAeo0TxhCDmTFvitzXbq3JR7y3dYGZgLvqAaXT5tzSdKHcf83Pck/LEuly9VjJzJqTEFAEdAnp4A1ItHcWneOahO7ondJI+20ctGOBtwD4i7ayYwBJTrbrAx44Ok2JWGIBxTVr1phOXiSjpPWZ56FOM6pbPSfRyiTa44BpyhNIRP6sfpdrEOej8cCLSONBI8Kk9Gpf0FWvncG7kr1qsXz3yU7Zs8q/DCyedVw3ubVxsXRssMqAIrDY7PQsmTutSCZNHlPmVeR8QBjnGDhooNnyf6jreK/r1/9ci5XGH5tGXeQekC3hYew4HSzSLSZNmmRAT3VfPXQAl20Dw4Gf1yYm+/9Q96n3xPMoSOKh/P33381jA5azZ882UERIHYcI9RT5AI7IDzkiT5WtX2WVKufR5+IZqR/oCkBEfQ8q1SyIuq7tAmWH9y8WvonGDgKLzPBAxCaSXdbIDm0L3vdELHGBYp8+fcKCIjF3HjwaAXEMBgaXfZAeIiqpGvFEVSIqCdckFwZvCaENNS6RthilgoJCyV69SF56dI50alpkwsxd25TIvVcVym677W4amXta5Zqw9MlHXy0rsxcbuNTzcg797OcWowjwUmZ4EHjGRMoTw4qu0GsPqlfKeSkvvALR6jCVnJxdGxSp8OTrJmKJBRSpCxga+14VDL1bjuEZkEUsdToauXE+vC7IGi9uugAiDRH3SyiVzgT5WHiKyqsLHMc0V52br5UbLymQow5uaOrx/x14prS/OFNuaphrvIutzs+Sd5/dJEtXzJVBgwYbGOO3U34aJ8tWZsiadZlmwu6xY0eZa4e7Lvt1jaeO6vnxYuAVA1aZGYDyUzCcOrU0ogEk4WkDkrAVNmD5aYdS9Vz28/L8dPgVHvGizps3z7RzgCPOC4CJKdOQKx5XyknlHU+ZpeJveS7sNyspCMiAehQE2AVxToVFdJ/ObTQ2LpZjqE+kdTEnL51zry3W/xUWsZ0wRdBLXKBIQYQKPdM40sOOpVEBOoKcQxHPEwVADy6RFYiKQY4OYXVejYShiAXgCgvzZcSoodLlijy5o3mBgcX7r9kpt16+zDQw++51mPR8SOS8GmdLr149jb4UFOZJUXFpmEQNtZ9GlftnoaLTYNBTTKRMAW+S2Alj0CMNAi7QF8oMoIq2ohOmJjfXruB8JrzLnJlBL9GCIj1inUhbDU+4LQaJZyC/KFo5xHIc8EljoeHmIIy73+dE37hfOio8K40eABVN404oOTdnk1xXf7m0u3i53Na4RGqd9IKpy4yIPr/6v+XWRoXSts56ef6+DbJy1ULjeSJfsWTbRpk5qVB6PbFRetyxTj54eYusWlYs6zdl/QkWuRc8VoMGD5SRo4bJ0GGlI4eZgsdbVzmWekzeEytAipdQj+N7no9OIfWNes9zE27ndWMagvWCoZ82J93PpfDIcwCOyIwGHs8+uZnUATqatIHqZYxGn7SM0mnLc9FmoE/UIToa6QaL1AM6t4BaLJwTjW3kfLQ72F7WcLYZu4wdJ1c26CUuUGToODfqfRhAMVbaJiwbhDdRGwkSqWnIk+EBo+LTiOMmxkjEAorkxmzctEa+++4Hubv1Zrm14Va5s3mR3N9G5Mqak0wD07ARc97Vk0MOOViyshbJD9+PlBEjhkvGgvGSm7sp5muWZ5S5fwwd5cVoRRqRRBoqjAzGlfIMygsFKFIBYxn5jHegadOmfwJF0jMox6CXaECR3DC8ntF6EzFG9G6DyEtEXnQQ0y38hM7RwNFRoW6jj9E26kDYxi3Z0q39armmzlLjPby5Ua5cVXuaVN2/mqnPe+/1N7mg2mcyqp/I6nWZJswsslW637JGmp2RKa3Pz5I2F2RJyxpZcvmZWfLlOzlSVLzO1AnqIdA3YsQwyStcKYVFhbI8s1ByNhfKth3rZeKkMQYg9X6ZembQoB9kSdYsKShZJcVb18j6zUvkxzEjzXGcj1xWwJiBRnhQ8YhR//GSEW7FXtggVJ792NW/t2WFXSDHEWgkx1HTGHjbCWHaWHQrkTbYr2vhRWVENHCUTrAIV3C/gB+2MRoAjOUY2jbywbG/4UAR7oK1WNGpIJe4QJGRTOFAEQ8EPYVohYM+FErCAAAgAElEQVRnIYh8MwVFgIaeMIqpRtIvZS/vPHgUkcPmzZv/kJsTjcFEATDGObmbZNJPQ+WzD8fJI53HycOdRsvIwYvl5ZefL/NG4JG4pfFaua3RTrm1UbHc32GtfNV7sKxYuVCKiopjAtRI98Y9kacIdDMbP8+XSJliPPF4kC8XlEcRCGYaDvQ42h4j4EM+ol25+QwoZmdnB1mPzbmjAUWmbQlVZ0MZIzVEwFy0Moi2vnM+ZMt0JIT5g/AKa933a8s9aucITwI5eNT9WHQfMJszd4qMGVgoTU7LLBvA0rHBarn9su1y6VlfSpUqpW9j2v+AveSWW2420+W89vCv0uq8NdKh/goDl+3rA5mlg18anbhEJg4vlrkZU41H8KepE2T7tu3S44610vS0LGlyapY0OS1Lbm++UtZl75TlK+cbCARap06bIL/+tt3A5p2tV8mNjVfIv+9ZJ2tX7pDNOcvl+/7fm4aQfOSNGzea3Ds6u9gHG3gi2Ytov/P7fNFeN5nH2c+MnSfyADxqaBOvMB7rWHSsvPYoVb7XZ2IAEPMu6uj3dLIFpEDFOt9uNDYS+wgTYYO9jjjbVvMdEStYLMglLlAEEkI1OhBuLC5ZkkLJ8wkq3wzjzkAZRspGk0cUREUCmoGrihgljElpXmCJ5OVvMob+1k7tDSDuv//+svvuu5XB4tnH3yEPtxPR0dC3XLJNPv1gpKxbv9IYIM5VkXvw/gZjtnLlSpNjkwz4xotJKkFQ3igFAjzj0UISoBjq7SyA4sKFC4Osx1GDIrnD0XoTMUI6H2q0MojGCHIMdR5DyLvD06FhUNjEi4BtI40k2nCzbU9oHAE0wsjP3rdemp2eKW3rlgJf+4uXSvuL18vV5xXJs0/0lROqH1tWr/+27zly6VmfS8dL1sjNjfLkxks3ynUNsqVD/eXm94ya3rZziwwe8oMUFW6Tay9aJuQ5MjCG79hee+FSaVBtiSyYvVVmzpkskyaPlc0bdkjLGkulxdlZ5vu2Fy2VNhcslUbVM6Xvh/kyfhI51sONflHnbbDx2oSK/r9u3TqTz8hsCsBSENeo6L0l8nf63Nh6FkaH05lC9xLdGbd1NsjP1Ac6/qQTMUAkKAbQ+uvnFruFpx0HFGkD0dq+aI/jnN5UJhsS9TP2nEGUQS5xgSJ5FdCsTbx8xotCyC7axoUGFqEwrYCfBannotFnRCJgS6hIezJBVgA9N5WAhoGpcQjVxGN4SiGP0XVFRicwsPXr1StrTPAo7rn7vmZktE7Izdtb7mq5RSZMGilFxaWegHjuQX+rSdrJGNCCbOllq0dRy9nvLekKsfQWgR+dOFUrMfUBUGR+uKCX8jyKGRkZJqfY9njqfYbachzhD+pntMYt2uOYRobwbVDzpvqtCzQKRDwYeMBo+HhSWErz/4bK9h1bpP/neXJ9wxVy5TlZ0ur8pYJXb/aUEiks3mg8lgMHfi/t2nSTKlVKB69Rx486pLHUPfVtuarOdLnh0vVyS+N8aXl2vuzcLlJUskFefqBArqq5Sjo2yJbrGqz8rxdyuQHGtnWXSYd6K6SweLNZb2i4Qq6unSV4KIHJUrBkmyWn/mWkzJ+xTX79vTSvWkHGz63mUKNrhCB5nzwLoW39rtTu5cVlO9VupcuWZ8axgPdWU2wq0jHRdiiVt7THtMuE3Wkn08kmALZMFB9LilK0NpL2hLEeON5C2WfdR91hrAhvewpqiQsUcRf7BYpUhqBAEQNPYT711FMGMBJZaQBFJl/lrQIYPz8MlRrqnTu3y/Spc6R9g+ly0IEnlwHj1fVGSLertxuvIm9u6dSkSD59f7zk5G2Q/Lw/z01W0XsCfAHFZHgUAUVgIyidATRiBUUMQDhQxCMQ9FIeKDLyHmi1O3ZqbEJtMVD08qPt8EVrADkfnhJC+3Ti/Ia6IM4HKOJNJDLAa97ibbSBxUGDBsrizFmSX7BZ1q0ukI0b8iUnb63MmDnJhIaHDB4iGfOny4QhIlfVWCeNz+krRx/cqKyeA42sf9nvKDl4v0bS6fa7jUyZm7HpeYOkVe3Jck3deQYYb2y4SW5quMV4I9vUyJMNK0UWTBdpW2urCXnfdtlWKV1LpFPTbdLklEXS6bKd8sYj0WmtTgfjPZr9hFMjrfqbatVKczTxkDAXpS4MkqmojUrn32HneXYmHyeHH5uXyLYrkdeineR6tCd46dIpygBfxOJQiMVOIg/scCSbzXdEdrmPoJa4QBFPQzqBIq9eSnRlowLQQJN/wZQRfvaOi0vypO8Xo6XL5b9I93YiLesOMA3HEQdfYLyKpW9uyZPOzQql59MzZcPGFTHnSEYytIAiPSnC+Yk0Klwr3UCRHnPQS3mgSD5gqFSRUJDIvlhTSGIxgIAi79dNF1AEPumUkCKD/mnDFo/e40lh4AnAOHToIBkyZJCZYJs8Rr5jHTV6uKxZUSyNTlwqN1yytjTs3HCztKo9SWqd+IIc+49m8tf9j/8TPCpEht7uLgfsX1UO2O9g2X/vf8qB+x0tTM9z0IGny6F/qylHHHSRVN33HDnuiMvkhCPbyNVXtzEdhlCNIV4P5k8kR4rOsHdlYn7ep0zOXaSV1CBA/C9/+UvZs+y+++4marBgwQJTdSLZosr6He0FIXmcHMl4wUE8+h3Lb9F1vPR0/tPJJmAXSH8iHSVa+xftcXSoSc/BY1geKOJRxCMf1OJAcdiwQCGHBoX5Jpkg2n9QLJSBA0ZIpyaF5p3PTL7dvb3I+SfdL11arJF7Wueb/czB+O5rU2TzltW+gyLD+JOR95luoJgKHkVSL7wdu3CQ6EDx/T95OgFF8mL9AkVtSBUK2eo+3TJdzY5fN0jXdqvMaOfSN7gsN5Nz33DpBrm1caFcdV6xjO4nsmHTMlm7OkcuO32yND7nO6l3xodSs/ozcsa/7pWTj7pJqh/RTo499Ao59K8N5YKaF8sJ1WrIgfucJH874ATZf5/D5YB9/yl77VlV9thjP9mtyt6y5x77lkFbaOAs9Wgm4js8atHOQVuZoFFBkYgY+byJngdY9TDoLbrP6HrqVxARgSDPSQcniNBzLKCIA4ABZ0EtcYEivUlvwwP5QsCx5igGHXpmtCdTOyQ6KRhQxNjzOqt4cxS9BpD8lZmzJ8gD162RO1sUmFAzuYkPXPOrgUQ+m9Bz0wIZNmyEFBb5+x5o5oSknOPJ16qoAQIUMSqpFHrGwx4u9Ex4PuilPI8iU26Emvc0HCziUYx1mqtoessYQLxTzOeYLt4DQmF4Drj3RHp2aECZ6mbH9h1y82Ur5fIzM820OjropOlpmfLiA+tl284c+e77byWvYJ08cuNGubbOarn+knVm0IuGnG9ulC83NyyS6y4qNKo4b6pImxoFcmvjoj+snZpulcYnz5cuTX+WN7qXaq2GkKnz3pUBKOQThloZAMNoaXIPI63YRryrNnQSgma+VHL1WPyMxnhtaSr+z/OaGS9ycoxniXSHitrLdPgdz0d7AngFCXZ+n5s8xVDe9mhsYXnHxBJ6hnGCWuICRd5hGQoUYw1Z0cAyMCGoEU80RiTKQtzJ8H6Rj4SsmC/LT4OEYd68Za30/nikmbiXgSv3tM4tW4HEWy/dKq8/PU2WrsjwdT5FGgAMP7CWDFDkDQZcO8hRz4RGYxn1TMeJV1fSUVL4ouNUu3ZtmTZtWlB1uOy85YEiHZZYchR5DkZxBzGYBQ8R4UpAPx3ykbhHcoB69OhhOpyJzMsFoCZMHCNbtxZI/9750rXdarm12Up5vPNamTq2WAqK18nQIUOEkDXzJK5euVUuPyPrv/M0lk6jw1Q67S5eKg1PWCKjBxTIosyZUli0UTo1z5aramWVTrVjBrTwzullcu2FS+SMqqNlVdZv8vMvW43dCgWC7MMWRFqBnUgrEEqHGkg89dRTTUoCcMlCFIZz76qQiBeVQaPYIjrHobzO6QCBke6RZ8KBw2TjPCs2PR1sAsCJTWCgbBChZyCSHHF4StuTUFvaGBwAdKqCWuICRcDLC4o8SKygiEDITUDoftM+5wMUeSk7E+Qm0sBr5QBqeEYgEcPqJywyAnrpinny8fsjpUvzPLml4Vbp3LTIgOOtDYvltaemyeyM8RWemifcveLN5B2myZxyKGhQxOMVS08RUPROkkolBs7mzp0bVB0uO295oMhbIKivNsiGMjy6j3tnepwgJpQFPik/BoakS6OALaERw3gDb8BNohpuIJAcxvkLp8umLdmSl79e1m3IkmnTJ5i8xiFDS9+kwgwLi5fMltXLdpjR1I1OypQrzsyUJqdkSouzl8qEYcWSm59tBouMHz9atm7dIdc1WC54Jq+uvdTAJdPqND4pS8YOKZZBQ/qYeg7MYbsANj+hDQhcvHix6cSvWrXK6DJwBFhie/y8Vjhblir7VbbIGUBmS90jt5McZ21PKuOW56NeBcUAQXAFdgvvJ3Y3CBtJlNX7Age1zfYWe07omQnbg1riAkUSlcOBYiwNLBAV9JtZ6K3QkCfD+0VvidGSTLqMYfTbMDEybv2GbJk4ZYh88v5Y6fXsdPnozYkyfMRgWbZiXhkk+mV0OQ/eUUZzY8QA4UQ1mBhJDeczpVJQvU8MFm8eimU+UODHO+G2gmKqzKNIhyxUnbUNj37m3jFC5KEScqWe+rVyPhoGPP3IOh1gkXuk00kInyRzJgpGHxOl+1wHYGQSZlagkGiF9/p8xxyJv0iOrMgskbFDCmTRnGL55bciWbh4pnmDC78BdseMHSW/S5GMHpAvj3daKw92XCMfvLRZigq3S/bqhSa/Gq869RzbBcD4CXHYEs4HHAY1kbff9tbv86ldRr6E4LGteIeABNIckhEFSySM8gYapsUhjz+dIgzYAjy9sbJONDYU+4iNwRZjh9Ume7d8R4oGkawgl/9v705j7avO+oEbY+uYGBP1jb7znUYTjRrTGuOLauvQ2kajLxzjkBRJLVilpQ1SqlBqkQ6p1L8ah5bpB6VQWoS//TdtlaGAhdIqpUAQKFAGWxTBFGg5/3w2fn9dbPa555x79xnvc5J91757XOtZz/Bdz/OstQ8EFIVhhmZR8igyAvMQJNeIxSP6MpB/vIpmLK4aKFLIRktZw1G+TRTDWArH8yiZRx/19ZXPTe5/8M7JQ5+3iO3TIR/vGfOdnsXDIOfTjO5VJ1h7H5CKd5aVowi8mCWML8Ojs0pAsT8CBLR4FJc52ouCmOVRdB3DQ2bn9Sq6TpuWsaAsegGulsNYltyP/dx4EY477rgu75g+WaVncV4DDkBKNfh/H7picvXHPjz58Ed8ku+SDhy2z3AdYPnxG6+a3P/gZyYPff62ya233zC54v8+/a1obWPILWUEMIrMGPACdnIVx/Ay0ifZxtKJm/6ctBf9HnvssY6eDzzwQAcQ5emZwQogHnQZpravN3FfhM+AnH7ZttzEeBPhilm2YdHzdKN893nCzvQ5h8kyfwcCipY0GAKKjMvLXvayhbwQhMPncKD0sZW75yEkzwjv1yoFBlAk7PIveU/GntAShUjxPL3/laRy/3/l+LgL1loXLTOe10FPfHL88ccvLa9Vfxnh8nTPK+SUnZX02xFggOJDDz20TDnunj0PULTUBoNvFNrWsz9Szf+u0QbyOS8dFrkOD5notW1GAriVTwUwMnabCBbJJf0Tr6P9IVl1/GlP5aUdaORpbK+1Lypi403F52TPcd8ON2jkBUu0JCAouqnKp/Vw6IJOQDZZtG+wARhKxTC72UA4oea2H4b6bhuPpU1mOVvA3vyEZQ34l4ElDBbV15J3lvlaRN/Ncy1vIrtDB++lo52jx3/mZ35m6bblQEAR+BqaRcmwSIIX6p2HMK7huaGElgEU4wWA/I2yKfVVCxhlbZRgJL7tipNy8/k+BmMdHloGywyvk046aSk5LVEEUit4vOblYXkqL3zhC58h3GTBYErIbtm/eYCiOvAgLeJVpJB8IgotlhGCpmxPP/30rQk7xfgAt9YQxB8GoPiSEYwhXLWOWeb70iYTKni6Lrjggk7+GHlg2RJgdALwAzgCQm2IOiAp5bbrwKH6p21t6Tr6ktfQ4NrPRB3AnH4BDn3p5+///u87cIi++jH0XmafrvrZkQ2OEyBRu8nONg0S2QZOhHe84x0dSFwE48xjR+hXMpXI1F5AkW2Bv3xOdNm/AwFFFTSjs98YDVj001+AotlDy0hm1blALUOKKTHqKoWEgABUgNVtt912VIEOKZtNP0YJArvSBEz2WHVoBC0ZKl4oeVPLHFiYqLNIkrKBiMkfrTwIHTi2it+8QFFdfD5vXrCoPWRa/iUeHhssUrY+VcXjTk7Ja8DYppfqK0XBwJjuYgDjXdxVY0938noBxjxfF110Uae7Jd/zEBvA+bLK7bff3kVQ8JuBEvAoDxFwavVcC6yy357fpP3Ury1TP+3SPqDQV7gAZu02iQwv8BYabMk98/EH4BoNbUnf2UWewS9pF+/7kSNHuqgGHZS0k22Q+YBEpagMfp8H/M17Db3KmSQay260diQRnpTOyTWX27mK34GAotHP0HIbjApEvAghCQ8FA60vg2kocc/nuVj1MgOEBKDSLuDGyJKiiYLZtlL4HCiKclsl6PYuitXkIBOUlgEk9BMFBhQR3HkEnZDzDrThAsK8ikTjKIpFgKJ75MAsGoIGiJYBFtGYguQp3kYPA8DIE6APRA8AqADGVcvHKt5Hp8X4ayePqsEw2ZS3DBSRHQ4Aq01wAgCPN9544+S+++7rPk1HjwBUNqHrfNsY4JK7F73YgrJ593NvW857b3td7lcf9eIlNXkQ6G3rr+5yDE1a4y2UliU1QQiRrpTGYoY/OhlMsEHRny0tV9F3q3xHeIQ8sIG+3CONxedRyQz9vQx7P7ZdUEf1lR4HJM5rF+axHa4JSDQxZl6QaHCNB1fxOxBQtDbckFeCgRSqYlDmJRRhEsLQGctgHJ4nHgu5NetYIodwEhbGWedGAW1TSYEaLUtuR0ftWaXSybsoHJ6+ZXgTo2CAlUUmshB0RgHv2wIYATGKcRW/RYGi/DKz5RYBi5QYBbUssEgJmySF/vp3GbogfTxWmToaXMi5wjfWiAQYeL93GTBGJpUBBQAQIAQ40hFyHvUn8Ehv8CDbeOx51tAPiASgLJVz7733dkAR+AIiRTCy8dDRn4AlsOYawM3Gm5cNoOtvOafMPe6n0wL8PD/vUnq/80CidWNvvfXWyTXXXNO1iVPDwMYycTH0ZFCbOCakOQHOAdCHARiGH8ILAYjoRC6E27dJtqMjwr+A3NggEUbyzEVAouiOb4Cv6ncgoHj33Xd36JcHMS7RGMgXvOAF3UiKAM0DFnkfKY+MMtJBY5cAKYUURg5jL7v0PkpDmEobKSTAa5uAorqq97HHHtvlKEXxLZt27fMZ3YsvvrjzVi+bVxZd9sBSOmShBYrCAwZAq/gtChTVSVjMBJxFwaIBD7A+r3zPowNyDTryGAOKy4owjK1XPA/gUV+byQkMIw8KDyOwQF7ogVXrnlZ+VrXftpPMGtwBj4BzwIPlfSwDw7sk0oNWZOWYY47pbAYewNNClOjJKymKIJ9P2pP7PevKK6+cXH/99ZNPfvKTk0996lPdBtBZaSCblJ+cU1533XWTq6++uqsPQA+o8vipi6gPDyBQwyuoHvhc6X8rWNDj8tR8Vg8g1Lfqon30vPZqN3q3tFgV/dfxnrRTuzljeE8NDgygpQqR522btKLOQKLBgFxqGCV6aowSX9GjVolpnQwtnsq+89HTsNcqfwcCikZlQsy8DGlMgCKDuUjyuw4Qfl4WI1HiFAzlvQ4hyjspEl4cI9g2vLLpgDHeRF5kAkIRpE2rLCkfxoLg4pV4c8Y0/JSDiVUUwyJAiDeslQWCDSieccYZK5Hp/QBFFePBAfzUtQW6rUy3+9qlnUa1wu1jKMz2GWhuQAcwMtrkNkBszH5exrPCj8Ai/rSEGI8ygCEnDagBmFoQsUr5Wde7AiJSqgcaAFgBkYCWDdBCI8cD4kyeAeT0mfAfmTLrFH0BTBMjgBEeSwNZX8Lqb+SDF9N1ruf1BQh5AD3Pc/UfYKCv5F7yhqoHvaNebf0cU//0pTalfcp10XqV7017vRM9RJtMlgOm6U8l72rseuRjGbI31jNTRxECtgbQtbU6aox9ek66inSevfQufes8/WyGcxamX4lR+d+XHAgomuFGmQflxpjEkABlixhaz0onjdXpeY7nYlaKwjFMvWph9j4KkEISokK/bfIqCsXoI8qTclw1/SgjipqSt/YWQJf+HavEJxSb/llEORjoROBbOZDDK3l7Fb/9AsXUjfIziw4IJMNpx1AZGZdiQs6FThaR9XkULZoaace7GA/ysnTEWDzkOakjXgI8AF7g12QPKTDkx6CRHgrQWIc8rRJUDL1Lm/tb/zr0sQVUBrjRpbYAOLoBTWdtrmsBn2foB8+1BfylX9r69Ot6mPsMXdArfMwbyztmYATEk9dtBIiJCnBGyJteRqg5jjHRnL30bfQsvSzdR8rFOn4HAooqLFEZ0u0bFo0XupvHIOQaIEQS8DIAAOXtucIXRpTrylMkXN7NaPism/yXTQeL6udndGUETzGsS0FS6kZhYxr19lkMPAWHRxbxllEm077KYpS9it9BgaI6CqmRXZtR7BBIzDEy7xqbvEX9Elkeq4x3kS7hOdpW7wTdkxmeltQBGHm/GFeyRKbwNnASMNIClMO6H1qsszystG/bHfo7BlTjV4CarpQagJ95cC3Ijs/JaQZMGTS1enaT9lO/RAF4mA1+OQrGHvx6HsBHn9GjdGcfO7X6lR6Gr+R4rvN3YKA47essCMDDsgga1zlyGsJky2AmChtIw+itIKxqn8ARMDkx2iuxWqL0poaegURhcqBWKEF+56po1X8PIyo/ESChoCLgY/KJZ/IAyZNaZDKWkfTQGopGgvKjVvEbAyiqpz6WHpGJatMUWavQyLs0FAZDvvHYCtYI3LNtwoXb5q3Ao+FXXhabcKeQNH3EyIo0CLVmEgg9QQZaI92Xifr/8rXpo12mfctz9C6vbfJs8bJBDr6lJ4EYNptMsq/h8zH18jKelXoCiHS+PFnOAR5R+maswa7n0IfxIsahME2vZgAOIL7oRS/qsMoq7Mde7zgwUERgxrDf6BiORYwtxuPJoUTTiWMzCGYGNORRDIUWViH8hJDgWYeQd8EPUNw0z6L6ALJyKeWwoZdN/VdBp/47jGSNWK0dtWweoTDmVRaUgOtN4GrlwGhQaHZVSxiMBRTxo36XX0S2pZbsNfIFGKPclAaIZBn9xgaMBp4Z7Zvg0AeMy9IbY+qh1JFRxccMlUgK3SevG90MmOXJCZEC7v0QdWvI+3JS/xd4XIQHwkvR63Q8fhP5snFqSPUBCnnZpG/5P979OHbC12PKytjPUsfUk9zxfpI1mEBEc+xBbvSf54u67OVFpDtt9C29CxvABZvwOzBQlJwdz0M8DDEcFhpGoHmROSPAwEheT2eOzSiYgwE0egDWIhyLCNYY13qvEZq6YFS/TQKLQKLlJXgTeXHM7hMaWye9GEw8AmAzsmPzhud5rhC7UeW8fEsZAC8R9PC/ESHFuqrfmEAxdTZ5CfAj48lF1s5W1tt954BKoNnkHvKcEfW89JznOnyALyl3s1MZK4PW8MWy9MeYPNcaLfWmmwBf/AcEM1r4kIHG82b5MtwJU8fjSB+Ry2xj6Kd6xm6CzfBIq8fxEWBoUIK/zNLHm7yG+I8Nl/aBBzMwi5yRh22Rteh3gzPtEGKmQ7SRrM2jdxa9RqRJJE60ieOgtRHRmzlGv9Kz9O1VV10VFbwR5YGBosVTGQVESMOVabyZzEHVs4jsOsnrLROOqZjzLF8VwRiA4joVImGVv8bACzv58d6t27MYTyKQyCCjFyXSKpdV080oV1iOUBsFLkM5eSZjTYEsMpEFX+Pb1uuG/wFFawKu6rcMoJi6k0kz7iiyaQov8h/Zj0dVPk4GjPPqglm6oj2vr4BS4VueuRizyPsyeCXPHqtUx7aeAb7ALwPN20iX8jhaQsbkmHPPPbeTSbJpY+yHwOOqZbXetzkgM8AwfRJQyPbhGf+bbBceM7AjT3hMagRZwoP0Yuxyn1fHkoFlPCcylUGYJY146oA3g/tlRD3oJgDRILadrEIvRke2GImepFddK2K2ib8DA0Xr6jGS0HCfCAiAWK1Sn7Xv+mV9yi+MiPkBRaGdVrFGmFZd8ixy5/Ms+moLT966wKL3ZoFZRgmA5cFYJ0j0bkrNTGTAi9BHAaRPxyg9E2/IGwuwmcWvzlM2PlkJKEYGKAUznnnHV/VbJlDUBnx55plndgNDIHgWYIwyRBe5NsAcui5LOTNy9AclzSOnPxm5bQqNtXwcHmegeUHiMdUHPDxAN3oqLfXCE3nhhRd2g1/yQm4BAjrOQIteC3BYpzyvWr8ehvf1+1V/63dRIHxg878lf/CTuQU81ZFHK0kYfOA/fIbfthEYqn/khtzT59phFja9Y9Ca9I55dPu812QAjJ5AqHxteq91HrS2gX0IQORoE8rflDDzkL06MFD0UESCiPuIGZG4UXXQvARHaKMZHZwOb5XnGPsY6JRTTuk+6bdur2KUGLAIBAEp8sNWPRs6wBRQ9XUE4J/njsHZBKOCPryJQOIYPDDtGZ6/SH4ivt5rIsuqZjyTw2UDxSgQX6igeMl7CxijCPul62z0gZxNvIXGvNVRsPPqh1nXeZ5BoJASj4FVGSTb61f83Bq/aTywicfpwuhDbdCeeHoARLPV5e6GH9HW/wyQ9Qd544EFcqQEIApAbo7nL3ZgVjkNEPImp3+Fj1139tlndwM7s5LxBduKL9hXURPeq4ApwDADKvwfXttEWRiqU+pLNmAHG48oB0x0wSLzJWbpmZynb+Abg1Qg1IB4GkDMwBlAjN7k/BCV3fTfKEBRuGcaUOROXQ0xiPwAACAASURBVKSDEB3iX1Z4MUyGobjYAaFZwrmq8wRc/YT4fDPUL6HoADmjjjE3z7VZPN1PLiLPj7US15mT2NKcQeP9xUfL5As8QXkCpFEEs0qKQkgwSiAgiTIwUnzwwQc7uq7iz6qAYtqCD60JmNwaym9oBB2aBDCGNmZWo130w9igUd95NsDI04g+jAceaj2NZK4FYtERm1z268vIJ/cKiLTsjrYafALLdCp9R78CDs7xdgOQBssGNEAkfQh0WB+PVyqeyMhjC1Syn3NVLg48Q8N+2dIy3kH9oq/0k3ztFhAKHfMqRx/pa+uPyt8VoZM+hC/wSEAhfdfyfsDWJvP9UH21gzyTa3SQW5ncQ+B4lg6f53zr7EJjmwEp/WWtRfMx6LXouL7OoxfpRziJvjR49QnVbfmNAhQlXiIAYoRASkRDQN6DeY2A67huKbplMSyBwFgUqNE4ASSorXCuY18dgDNKmmETPmOMhfctoRNQNwZQDPDkvXzqqac6LyLhkgvFaFBOm0ITCtHXFnhHotyWwRsUqZAMOsyjPFwTfo2SCN9TCozzKn+rBoppG/7RH2b1RRmGHnRAqxNCH8fpC9cJ0xiNx5Pb1xWtkp63X4auS1iIfkEryjoeFX0f3uqDsGXw2tjPTJ1bY689aR8jystCr5IjoCIAkq7Bq2TspJNO6tJN6EefyTOxIQAFiLQBKwGT8UrOAygDhtahW1f1zrRxqOzXIQCQvkdPuj80DmB3Tz4vKL9OJCxLKhnQAoTkBfjHz+wZIJj+bvk6PBdeyf/bULZ1xtfahZ+VBquZeEOPZALdkA5Y5Fhf76CzY/QImckyN0O6LoAxAJFeFF3VDt8P37bfKEARcInSb41CjMEiHhodyVALbWGCVvGNzdCUJm8DYNQX4nX9T8FQIEaMFAPaEf4HHnigy1/0EXv03g9ozD28lFZ496UVYWZGw8hI/gpQpu3qsS4a5L3qwBAxWEAE4zU2D7TPo3gACKHkeRUKxUEBtIMkfG95A2GFVf7WBRTbNjJ4AAd6zFpap1WmlK1BJS8jmdQHUdQtcMyxRfqnvTbPirKn8HleDBB4k+OVSxiuNVAtr2z6furdlqlza2hb7ypvDL0LdOhDYUpGF614TwAUHiqAhG5y73nnndd96xigBHiASCBHGNTmfwAowNI108Bl9M4QyNqkY9FPKenrbNrWAr+WHgHZzmuPJdpMJOHZDYDnARYaxrNkgN7jFZdPaBAvZCyCpy/pq/QfWxmeda7t9+yn/7elbOudAY82847iU/qOrjDww6d0Q+Q7Mn9QfZHn6AOgnPdQ+gx9RcfRYUOYh9eQ/nMN/YIvtvk3ClBEAJ0FNfcJh1DW4KN0QvRZpU4x2gUMMMsyGNtzPV9dMCFFRngj/Osu1SUKVpgIkKMkrrnmmg7k+fyfWckmGAB+LXgMIIzXkDeS59BIhvfH9ZSU/A39gg4AImbeJBroA8ZGrpWJPq0iXAZPULYGKYsol6H8RDLAw37JJZesVDdsAlBMgy0yjm+BaJN66IahkXeUbEBjFLBcHx5K/UEfkFNGoG8IZumSvc7nWZ7vPQZlDI5Pd1k+C4+RDcZpm72NQ7ISI5zSNfFABiyn7eQOkPH9c3Sx4LmZo2aZM4J0CCCZfgIy0dAAlEHn8THY9Rz5kr7bTDbyKdB40gIu6SLhVaUtxwO0Aj7dl43+zgaITdtyTcrcr/TcFth5b+rQr0/qQmcCyfSpAa28QHoEvUw6ord4adEDXfBjaKVEP3Sk44BEuaboDAyhF95LP/RDxvos/acc6udtOpa2hBfjGcWXdIkJN3iMnGYwGb2wl5wvci46hg3QP+yunGopdHTUkA5zHM5xjp6j70xuxPccMbvwGw0oUh5DeYqISOlHiczbaZhhFUyOOTEdRbBucNh/f0BblJjRI4VjlGnUT9neeuutHfAzCaX/Awp5DXkjb7rppk6B8QoIM+kvio3Co1S9O+/r12Nd/xul+xILYaU0lskPlLJcHgOeAIhZvOo6dYvyCOjJSPOOO+7od8lS/98koJiG4kuyJTREPxhl90EjHRHaKaN4Q0f6g7JG6+iReftorz4cGhA4Rr7ohBgjnhyGCo+QQcbbfvgxBm4XjLU2te1p2xQgCQiRx3ZzzHkAibcRSOJ5NLgFGBl5PBCwBFwaZAHm8Zrp2zgUXAdA0Xfu5cmUigN4Aauez7NmszIDIApc2dRLHw1t6uca1/Mgt8/xXEDNe8wq917vN6BOvfGTequrequ//9t68wjiGc/gqfZM/ON93o1O02iIr5wPb/X7oz2+rfstf6UN2qzfeFeV+lWfo2tkkd0C3voyPSTH/Wtm/R994vneI09fSgx9RA9FF0U/0VG2gMPoNTxrsDFkj6MTt7EcDSgCHJD0kNJ3LLkUszos543w5WMQqFZZhbHGKD3XaA2TeA9AtmlgqQVwAB06q6dROQVkRMqA2jJC1R4bBYaOFBplJ7Rk5GukjJkBsfb56wKEQ+/VDwAGZc3YLJsPKCg0CxAJH84qKRTKIkAHrwNCQpqr/m0iUGxpwPuNb9GY4gUc9wKNaJkNjd1j4VrhH/pESDQGWt/N6qv9nCdT5McAwuadmU3MuMXbgz9bAx9jOIae2qRnpF1tOVQ/tAB6bGiTrQVxoZ1jLUACpoBN+gqQA7LkVQKJ0jkAOV46HjgfLLABFfoFsANIp23Ou871uQ+g8zwbm+M9Bo3e6/3qEZCnrdpm06bW42dfW2xpb2iQe4Zo1dLS/tA123ws7UsbQjt0wgNohM6cH+QtsgawRb73I7vtPUNgks5wnJ2kT0QweA7p8uib6J+2bMGhfYMFek2kbld/owFFOW/i8i3yDvp2jEFdRJljGExC+MJgyyqNLjEK8DQEWjblGPAUIAvkAYwJ0wCQwiDCOb7gIBziPFAo31EJaLbgMM/alPalHuqlLZK4eZYpk2X1fZ5LsfNatcpl1j4eTUJzCxQBIMZm1b9NB4otPcwGt2A0GpodnhF5X0GHrtElFLZrbHKF5DQyLPFKpc/6umbIUOTaeUrPs6kvw2AgkAk4vPQ8R0KGjGALFvwfHusbzBzfxTJt7Zfa2h7bq+0BVykDutoygEwZus9TtvfZb59pP+9MuVc92/bstb/XM3blXNqf9qBf+oYtR1seYCkJQLuBFzkiwy0wbOX3oLJLviO/nhUZZlukxhh8wih0SgBhq29yzqCWnnLOYJdzwVJhh+E3GlBELB2AmIjdKniEtuSL8/Mo5XQs5jGyxGxhvGWUFItRJuMOcG0qgAqQSqme2XIMEAwYdCznt6VNqbMJRoy/kSblQgEto+89lyITiud9nZc/XQcwRKGE3/E6oKj+q/5tE1BsaXPvvfd2C0XzAgkzJ6exHXj2dUoUeoAlsClUlLzG1tsYfdL27UGNT2t4vEt4DHjEQ/QcfcITHvDISMbj1OqzvmFdBo/v8jNDv4OWu0yjZbUtNM/zAwrD6/4HCtlVkSEyxyFDTsgL/RmPYQsMyemi8jl0fWTUO+I15AwwuM9yNrPAYdJlgEkDQpO37rnnnlZ9HYr9UYEizw/EPU2pM/x9hmiVd38fUwkLMORhxmWV3mGEwRvXAq0AsCoXXyNsPzQDaHl2KRi8gqcopGX1u+fKO8ObQ8qmz5Pt/wYyUTQBiv63z8O+6t+2AsWWTsLTPN/CgLyF9Mm8IWog3Ub/MAQMgkgGPcJQAHAxHq0e0u/p+5RtPw/tD13nmY4zTIwg0EinKHkg5C8Jmwpp4mkGtc15DICMAe6Xy5SBevb/WaqO2Tb69nkv/2sHPjV4D/8m4iOvkGfdgI+skbvw/yyP/5CMDR2L3Cmz77pWrgMMeSh5K+kR0Qd6odURwSlKG93N0RVwSH/AHyY+mRB6mH+jAkUzHSl2nRHDqUwn6LRWQQ8xQnsMI2CwMOKyhI0QAIqSry1DIFS7TR64/QCyTbwHzYF0gskzRBEtq889V79TevKPFuVNinAo7EzRUFTr+O0CUOzT7ZZbbun6SA4u2vI2TgOOUfgp6aEAeYbCTET97Fm8gPqQvpHnO00vtcao1U2L7uMJ7/RuxjNeFXUQguN9lAvHAAtpBUDSS44FREYeYrhT5niVBfjm5YHwjrJ/T8Ag/gsgxJd4kb6UH2/CDjmIlxBf2+9784dkZV65ynUp8yzySm79H2BItgzehZMDDMl/Bo/RCyn7wBB2UX+0uPnmm/uq6FD/PypQREmEFjLSGS1Y1FlCQ4tMFsAMPAIJ3/SZecz/CQvljMkJA69WgcXVeBEDWtEbSDfxJhNYxuzj/rP0uUGIpT4MSKKEZpX40mhVuLPlc/uUDa/0On67CBRbOloOSk6uAZ3QFfpn9A9E0jF9o5CBqr5xjm5iIHgcpcMYkDBw8XjEIE0Djngj18zik72uy/MDIHkdgVj6Tn14QM2e5YE0sUIKDv6VJsNw2+wXkCxg2Ndr+b8FgtnPOWULBsNXSvZP3j7PdzyE+JTM4U+bfTKTwVb4eUgmyMFesjB0T/+Y5+cd3undZCahZMCwlW/y3t8CDBOloD+0Q2hcPr81ius3TIHRgSKGpLx1UgsU02kMbDq8zwxD/wNurgfiPLtl9GXsExKzKoGXTVxXMKBq10ogUX4og2gwsWwvcniHsqQshnhvr2OUJcXT8rn/betSOLsOFPsqTCI5GaXoeRMo/hgBwFFftMAx+ii6yLn0mXvNeBSm0reMUHKoGMkYqmm6a1FDOHR93oHvAiANvAFHnpIAWueBSMvO8OzI42XY8XIMfsDkvEByFbo1MlflwcFtgF9btnQdAoEZXDhnZrlQsUGXNA9ebR46PB9+4/XGf5EDPNvyaKsfh/i5PT/vfuRL6ZlAIQygHvEWcjgZ6EV+WxnvyzY9QCeIRLjHc8ycN9nzsExE6evN/fw/OlC0rp+Os0Uxp6SUjeAXYSoMY9QQRdgKw9j7hI5iffOb39yNWDZxbcVdA4jaAyQy+NaFpJwosrH7tv88fc2oMraUx7yKzHWALFDR8jgFRSnJQ1vX77ABxT6d77///m6wIb+VjuFlYCRs+qY/OSYgvzUu+pSecoxhYZSEslrwqP8DHvHDWMZTnYd0Y4yndznv/YynQTdAy7Db8HEG1rzylhux3AuPpLB29BvAYCBmC3hYBFAGnPRlqv7fPwAMTYfKlq50o01/ZVDQ9qXjniE8bEkhoMhSQPIGAUGeaoMqg2M8o/Q/MIivWv5r+a7Vj+017fFp+9Oub+XGNX1Q6EMdvP7kGIbIoI6MRmZTOka+W2DoPnJqICXv2XrC9dsfBUYHiqrBLazDdGJAotL/Zg9h1mlMNXSc8pM7uAqvIiGjPM1alMgKLFYIevkhaCFnRm/Zs5xbpUupeuc0RTbEi5Sbe6Kgwt/+N2r9wAc+sD9JHOGuww4U+yQUqr7hhhsm7373u7v0AjmK8S4oRT4Sik5/9ksGqAWPPI9mZsfzyNACbHQaoxSeaY1gjqVs+a3dz/l+Oc81uUcdYnDpYeCxDyYTMvT1KwCCnrNciQFy1gskG/QgnZucSYAEOLE5L/oS4DJtcDcEfA56rJXhZewftH7t/dPq19INHdEztJ1Gd8/VP0LC+ss6kuyUxcDxCLCnvwMC9bv/8ScbmgFOeEU5DQw6Nw/ftde0+/13tPKAR8lLwsdSLgzGDL7JFxmM3Cn7Mgkwklvya0vespxxnlE8+/GPf7z7cllfJ9T/+6PAUoAiRqeIdXAMaYCiTjaS2YtBWyYLM2OmLJVDYKYJ4FjHMRtGJtCVr7g8oAiEX3nllZ2AG/ninVX0L8Us/4sy7fPbXv9ThpRaFFjL1wZB6wo7E/8CirOV4H/8x39Mrrrqqk6X8LAwLnRSjI7SINex9HHfUMWIucbmPO8F7wcvCF0VAMlAA24xoq3B3IvPnMs9065zftY17b3Rue5hrFvPJEDBmy+8ly2hx3gpecsBE3KaL474TBlvZXIogZ4W8MRrmZJ85zwZbDf3tlsLpob2x9L1njP0/PZYWy/7bb3tp03aZ2u9fAHYrnOv94mQoRv6+SIMz6/ohnxpy7D0vX/pE2X6hSewBYDhh/Rz2/fT9nPPtPOOz3ON6/q8je/7gJB88NKTF3ITGYpMtbLmmPPkkVy2gzty651oyMnw0EMPzRb+umLfFFgKUKSMGU2d3AJF+zpfJ2OivZizf85oiaIidKsAElEeQk4XXnjhxn0LehdC0ECiyQlCY4wWz8Wq+pYyZ8z7fLbX/xShwYOQJIUW3rZPkZmAs85fAcX9Ud86joyNz7sx1AwxQ5YwFgNlP6FrOqw1aNmPYaP37OMPz2EYeSCl3cTI02fxQjLE+K5vaPfixXmMd56b58xzT78O7omXUn3Vm4cKsDTg156AGADZfsLg8WR5hufyfAmJA5vtZ/mATh4yuaZAk5Ap76boQj7RB4jaAKx8gk+pz/pbdHcL9LLfvzb/t8/Nu3jvvN+mPjb1k9envsCdwSaPrDA/3pE7qp2AHrrTa+iQcC96hU6hndKA1XXhC/f1+6vfN+nXfum+9lj///ac/f57+uf9n3enzH0ZcKi3Nup7qWL43cAJDiAHAYSRjciMkqxEdvoyR37Qh6dQf/twBHmt32opsBSgqAmUAeOJEWJQlfmfosF0Q0w57RglRIgJPWWwzA1gMfqjLCh4IUV5DhWGHse7iI6EHn0pGB7cZfZnnq1fjewpeQpoUR4cmsQSJed72uv8FVAcj/q+HHP99dd3C+wCBww/bwj9BTjGu8GwxQjiA+f7W2sIXeu8wQYjypjyUDOu+JFeNBjpe4qGjPVevDuP8R/Ss4veF+CQsn2mZwVkAj427QIqtDEASpsN2toN4AgQRZd233X+VwJYtvZez5u2tde19+Z57Xu8N+/O+fb+vEM7tEe70m/ayhmi/ejQ0iW0Stme22t/0b7Js2bdl3q0pXvb/otnUJvRhB5MDiE+zuA5ujA838pCjpEBchMvIVlyjhyQMxO0zjnnnMl1111XeYXjqbQDPWlpQPHqq6/uGAED9IFimIIQhZnnKQmjnJpVeRUDKri3KXJAsWZCHxwoAolmOANsFLAyQG4VJVDqvX0FvhcPUqKMAKVI+YWn7VN6eHPdvwKKy+2Bxx9/fHLfffd1oet3vetd3UQBOkzuYzwhyZfyf/IfYyBbo2k/x5Wtp0WeFk+KnC3RFwaZYcazDDVeCyAJGMG7raHP/l487VxARF8Wph3vP2/e6/r3tf+nrkNle90i+229sr/I/e21Q/XKsfa6RfZTJ+Ve97XXtde2x/e637nUtS0dx7s2oDZAMF5BXk+rf+BtfIgf2/zB8Gt4uOXt8DP+DyAkFzbHElEUgSFHvmLlayfkq36bSYGlAUXNFXLBGJgohjVeRUxnJIZ5ZzF6zrsWA8eruIowZcCicIiRlHzFAov7B4tAIhoeOXKkA98JCa2qL4WcJYEzuovwHh50TzxC4We8bWR89tlnr13CCyiurwsefvjhyY033jh53/ve14VMhR8NLvEJj0kAZLwofSDZGtrsxwjH8OZ/z6Q/TaqhYxlzRp1uDJjshzEBggCNFjDw4OT/6Nl5y4CV9voca8v2/NB+e21/f+j6VR/r16n9f1Zd2muz37/H8f6xWf+nz4b6z/NaAGhQgR94UPGHyJwomfAw/ul7BMNvfTAY/st5th0/h6fD5/EQ4n8TpoTsyQX5+MIXvrA+Ia0375sCSwWK3MeYpg8UKTpMuOhSOYTHyMcIaFWhSh4uIAbAkENjRA8oVhh6cbAYT+L555/fGTe0BRRXBRK9S04Sg7ooSMRzQ0viUJaU7Re/+MV9C+FYNxZQHIuS4z7HshyM5MUXX9x9PEBajjClUBtPDV0YYxtAGSCJv2KgAyBT5rj7s7XX07NCgi2g5J2kd8mAga9QKi9SQqf0qy2eyj64CUBpy1mgZpXnU9+9ylXWZ553tbRs9ZI2uF+pP2z6JqF7/Qb86Uf9qV/1b/ID9Tv+wgfhFfwRXlHmeL90nS0eQXbchk/d57n4F9/IH5THedFFF3WzjWt9wnH1xyY8balA8cknnzyqCPtg0f+UGKZvhWOWYLmWUEgmNtNsFSCjDxYp1/e///3dau7Azy5MLFlFG0xcQUsjTaUc0FX1n/dJWbDMEp6bxWfteTxHGVOmLR/bpzgl3W/Cr4DiJvTC/HX48pe/PDHx7zOf+Uw3qevcc8/teMlEDx4h4WY5jPiMwU74Ds/ZHLPF8LfG3j3Z2uO5tgUJrguoBAACLIUIeSsDLtUnk1XoQINmYCXeS56rDOTbHL14MwPe+rLlfzLW3/Y6Pu1c/xnz/p86DV0/7V05nntbYGdfu20AHnrY0Aed6CB0Qz90bL196NyCPv1gMGqgGu9fGwbWv22/Zr/t93bf+T4IfN7zntfxF37CD95HT2uHT/WZnc3xQ4dbK9ksY/xbv8NBgaUCRSSU30epRRlRSDb/Y9j9eBUxL7AIKAIAq9qAGu8kNITICEquXYHF2d5F61GiG6UIIK7Kk4g39Js8SLOrGTrGIMp9npJy73sT8TD+xdub8sH4Aoq7pbQNtB955JEuf4tHUiRDigPvjSVqhB0BNkACPzL+DH6Mfh9QOj8EIujiFkj0993TP+ae6HSgxZYJOoAmeQE2AN0ATp6ugE76EyAK+CSXBmNCo9qUiSb0RQtIgatsvFn9DQADxKZtzvfv8X+eGeAbABcQpz7qpX7qqb5skC3Aji2TAqCdNu1uQR66AHrAeGgWW9jvg6F+6vdB/s+1AX/Anr6PFzD8EBCoTtpBD77mNa/pBifWGhUls+7oZz/72Y7v8F/9igIosHSg6GsJhAAzR7EQjgiIkqAuarwJtqUIhKBX6ZUK6LCcAqVHwPIFlwKMzwSMoYd16yyBQXkL4a8SJGYQAeBT+kb384DD9ppp3kSK12h7U34FFDelJ1Zbj6eeeqoz7HfeeWdn6H231pJedJTZ/ZakMdgBaIAUAAOQaEFlQEXCiwGZSgAkW0CJMkBlP+U8z2EvhrbYj5QBXfsp84yUQ+9zbFYb055Z1/XP577Ql06x39I/fZI+4llO37nOM/Ur8E3H6m8TReTyX3DBBV3kCwD893//945P8Ev9igKLUGDpQFFlhIkxNIGLQCr9T1CMwuK2bw30XvuAJcVndL3qECaw6J02I0vtAxavuOKK8i5e/jRYBBI/+MEPdhNX9K2k5uSVrhrYA6fWOeMRWGRA4lqKt+9NDN9S6kbfm/IroLgpPbHZ9TC7VMj7jjvumFjSyazTSy+9tJuBKgIEZMij5LE0iAdAeMh4CAGSlv/JQEBNgExbAjX533W5XmkDilLaZw9SBkT1yz7YGuP//jvyv7qkPtlPmTa07Upb0+6UOR5a5V7vQU90RV90Ru84T6zNqD/0i3VmpTzpr0984hOT22+/vevHJ554YrMZrmq39RRYCVCU3CosEaHog0X/c/UvYsSBSN4hhlzuWTxHqyqBHZ4x4IeBlosiz89oPp60VeT9bdo70nZrJPK2UnqA9LpAYiawCAstyl94TGipz7cUO0Vv9vQm/QooblJv7F5dAMz//u//nlhj8u67757ccsst3SQdi5ULi5vZSualeVgf1UoRQI4FqQFPnq7o7YR6eevJmOiMyRHAkoFZwrTsho23MM6FAMOAuZQBcHuVuTZlnkWmY5fyTmHihNHVSf3UU33Vm80C6NghusUXVUzs0F7fWJbqgg7ogS4mM6ETekklQD90NNlJikEtD7N7MrMrLVoJUESsaV5FwklYjaQWDQsSTl4i4b9Vra04BESBIIuE8nCqx2ENRQOJ8ly0X38L9Qp/rXLSUb9/9AejJJF8Lw91/xzeks/EaLRGxD4jg2+F+jbpV0Bxk3qj6jIPBb70pS9NbL7LbbPMkJxfy6gApCZNAFLCpnfddVcncyb/AFnZ/H/zzTdP/uVf/qWbdes7v0Ob867r3//pT3+6e67ne490Ke/1fvVQH/VKHeXuVf7ePL1b1+wKBVYGFIU6GF2jvYzcUjK+wGISbPtGe6//GXReK6NXI7dVhjUDSrwz7wZKhAsSdgWe4mXbNO/fWPVJ+wBEE3wALN4DAHEd+Yj6RZ8A8PJYAVZ8shcf9c/xEEuJCChseVX4SM7lpv0KKG5aj1R9igJFgaLA9lNgZUARqXjdGNnWQxMDDChy7wslL2rU5cAJLQJr6wQm3s2DBUSoj7DDlVde2YUbgLIAqrEA2rqfk/YIt5v9bQadnM23vvWtR9MB1gXcgVTrXgoTLcpPrhdWwqctr9o30BEGEzLatF8BxU3rkapPUaAoUBTYfgqsFCgiFyPLS9MaYGDR/47zDi46sYU3SGiRJyu5cPH2rbIMKAJYARWAQw7Le97znm79KWHZXQCMLUCUiyi8rN/k7Em4XhdYT197vzoBiX1P4az/9ZkZopb1MHjJQCY8aqADCG/ir4DiJvZK1akoUBQoCmw3BVYOFC3bYAZY3wjHECuFb/fjBbLWlaTpdeYrJuypBFrNyuadEoo977zzthowBiD6BJ9NG+MBtr/OXETvz2aWM4AO8M0Chv3zBilyTfuDmXgTrUG2qQvNFlDcbmVctS8KFAWKAptIgZUDRUSQMyaE1/cqAokAJGPMyC8KFhl9oc/TTz99I0ALD6MldAAXdQKqLDlh0Vz5fPIYt8HDCCBa+kd42WapBt5bS94IrwPmQFo8qgFsqy69n0cTKDdoWJR/XG+Q0p/AEr40wLnkkks2UY67OhVQ3NiuqYoVBYoCRYGtpcBagOK1117b5Sr2JwrEq+g4wNf39szzP4+QkKPFRjPBZNWApX1fwFMAo8+9AcFA41lnndV55nwWybIJAY3x3K0jB9G72/cDs5ZzUBczmU04AsR8M3ldi2e39M0+OgOsJ598cucR3A9IlL7gk2UGK+0gJt5ES2Js8q+A4ib3TtWtKFAUKApsJwXWAhSR7YNaBAAAGtJJREFU6qSTTpo6sYVhlsu4n7UVgUkzVoUPgQgALWAtoGIdZeogf05IGpC17hZQe/zxx3deRuFcHjvgjAdvVcCxBYcAayan2JfrZ9kiwF0OYvIv0VCb0q510DTvVAeg1dplciUXBYl4JgOMoZCzYzzgFrjd5F8BxU3unapbUaAoUBTYTgqsDSg+9thj3QKmFi5uvTe8ign1mVAgxLkfw887xPsFKK57ckUATQuu1Am44QWztA9wK+z5ute9rgNfvpQwBBxb8Nh6/mZ5HwMGU7oeGA0wNCkFQDXxRmjZsjLWqOQ9FF4W0gUS2za07VrXPpAoNxINDQ7m8Tr3r8FfQ7Ocw4cmsKDJpv8KKG56D1X9igJFgaLA9lFgbUARqYAVRnjaxBbHLTMDRPWN+zz/C++a1ABIAGbrAjPT3htvHDCbPD9LCAFn6s7LBTi63/qEyRMUquZ55PkzkxrYsznf3xx3DRAIeAKE7nfvZZddNjn//PO7Wby+KOCdvIcWMD/jjDM6urXh5dR3WntWfVx9gFezkHlm0WsevmivARK12xch+nxoAGMgY8CxqRNYWpVTQLGlRu0XBYoCRYGiwBgUWCtQ1AATIqZ5FRlqG0PdGvd594EAXjpAE+ABZDYN7KROqRfQCNja5DBaC5B3T5szk/fYY4/twsAAkvt9HurIkSPdJ7R8RsvGI+mTUYCgULecQl63E044ofPQ8tQCSGgUYBgvZwusU69Vg8BZ71MvtNIuIHG/gwnA2GCkH3LmTUzI+ZprrhlD1pb+jAKKSydxvaAoUBQoChw6CqwdKPos0rS1FRlrQNEs1P0smQNcAUJmWR9zzDFd+DTAbBYQWdd5AKgFZ8AQrxmgGwD39re/vVu8/MQTT+zyGwFHbQX+0AmgVGbm+Cte8Ypu2SA5hsBiJvp4ps3zW49rvw7rosW09wYkqjeQuOinHzPQ4IE0QaU/eaUNOb/xjW/cGqVQQHFruqoqWhQoChQFtoYCaweKKCUUmhA0YMhQtxtDLjS433xFYNFyKQAC7xpQ1IKxaYBkE46nninVSf0DIIFHYE/o2qZ92XIsINM9LSD0LM/NtgntnacOAKJJNkDeot9wDkgMTxik9HnO/yavmMCzTb8CitvUW1XXokBRoCiwHRTYCKCIVDw3AYstSLTPcAsDvvSlL+28R4x8DP68ZTyLvGzAFdDUgq95AMomXhOQN0+5ifVfpE7aqO8sYg7EHQQkGnS88IUvHMxLxGu2u+66azuk+H9rWUBxq7qrKlsUKAoUBbaCAhsDFE0WMMt22kLcwCLPopmt+81HAxYTlgVQdgUsLgK2tvVaIJFnVOj8IDmJeMBgwaLu00LOFtb2FZ1t+xVQ3LYeq/oWBYoCRYHNp8DGAEWkuuOOOzrv4V5gEWDc7+QW3kdAwSQOoNRkEWFMIGRbAdSu1zt9I5xuwW8TT/Yzuzl9zwvJM81jiJda7zXgCCS++tWv3nzJHahhAcUBotShokBRoChQFDgQBTYKKGqJZVsY6yFDnjA0A5+cw3lDz/3rTICwOLOZwzxVuw64trF9LUg0EQdIBPT7fTnv/+nzId7CUwYoJj5t66+A4rb2XNW7KFAUKApsLgU2DigilTX8puUrAos8P2ZCWyR5XpAwdB3P1Mte9rLJG97whqPrGAacbCOw2qU66wepAUC8dSX3+0nH9Lu+FrIeCjcHJL7oRS+a3HvvvZsrrTNqVkBxBoHqdFGgKFAUKAosTIGNBIpasdf6igGLJiMcFCwCEgCEJWR2aZLLNoNGIBFAtIyPnFJevoN4EoFEQDMeaWU2IBF4fP7znz/56Ec/urAAbdINBRQ3qTeqLkWBokBRYDcosLFA8fHHH++WP9krX5GB5wXa7xqL8TYBIQCnzefqgBRgpbyL/2elIfnQ2+cCfbfZxKWsBZm+WrQEEuW0AoS2AMSARmFoIFGfb/uvgOK292DVvyhQFCgKbB4FNhYoItWDDz448b3nVYBFACRf6TjllFO2YnHubfYa9usOJAo1W/uRd5eXd1FQ2L/e7Hi5rNNAooGGFIdtWlR7LxVSQHEv6tS5okBRoChQFNgPBTYaKGrQLbfc0n25ZV6w2AcL+/kfSBH6NiPaBtTE29UHOPX/wbyOoSuAyJsrzHzQULM+50m0IPcQSORNDEh81atetR+52ch7CihuZLdUpYoCRYGiwFZTYOOBIurKHeP5GZqtmlAiwz9WzmJC0ZbQsSQLELNNX3PZFvAKJAaM+940gG75moPkIwKJPIlyEvcCiQYePMhPPfXUVgtwW/kCii01ar8oUBQoChQFxqDAVgBFDb388svnAos/+ZM/2XmkeJT2401s7wE4eKWUPhkHMJZ38WAexNAP8LY24mmnnXbUi9jSfj/7AKa+AjingUTHgUQh6SeffHIMGdqYZxRQ3JiuqIoUBYoCRYGdocDWAEUUf+9737vnGosJKfp+LyAwBlgEWEyoMLGC1wtYrHD0/sBiwswmjgCM8fwp9wMM23uARN5I/cS7DBDG25wyINGSSE888cTOCHEaUkAxlKiyKFAUKAoUBcaiwFYBRY0+cuRI5xHaKwwNEAALPEtAyEFDmQCJZ8id42E0+QFgNPkiHjJlbcM0CEC0/BCQ+JrXvKb7OspBZzS3QNHXdl7ykpdMTU/AEz/2Yz/WAcnHHntsLPnZqOcUUNyo7qjKFAWKAkWBnaDA1gFFVJ8XLAKTvuYxRt5bQElCm57pu8OAT3kY9waIaCTUfNJJJ3We3jEmq6Q/lNZa9O3maYMHgwYgUc7pww8/vBOCO9SIAopDVKljRYGiQFGgKHAQCmwlUNTgSy+9dGbOYjyLL37xizswMYZnETDxHBMheBd5sk499dQOMPKYlYfxKzPEAURe1xNPPLEDaYDaWOkA+sGzpBiYxDQt3Oy4iVDW2txVT2IUQAHFUKLKokBRoChQFBiLAlsLFBHgQx/6UOcpmrZ0jtw0YJGnycLcPFljAhVgBWC0oLPFuq2/aIJGCxgTdt3lsLQ2pp08hwnLy+kEpseme4B6vrYyBBIzSOBJ1DeH4VdA8TD0crWxKFAUKAqslgJbDRSR6tprr52Y6TwLLAITgCPgAtyN5V0EFuPdAhh5zYRYAUOgMXmMu+hpbMEhcKy91kI87rjjulxAXrzQZ6xSv/36r/96l+M4BBDbwQFP4rHHHrtaiVrj2woorpH49eqiQFGgKLCjFNh6oKhf7rzzzi4XkfcogDAzXVPyMMW7+HM/93NdKHos8JLnADEJh5pIY9/yL7xsCcMGMAZkbZunsa233EztOuussyave93rOpqadQzIhSZjlS1tf+qnfmpqPmL62MDhTW96046K7XCzCigO06WOFgWKAkWBosD+KbATQFHzH3300S7EyIs0zdPUepvktfH+jTUrug+IABsTXuJl/L3f+71u8gtw1S6xE+C4qYARMAw4BHh5DtXfvu8xA8NCwMLLy6Il2qLlz//8z3de4b0GAwCiVIN3vetd+5eKLb2zgOKWdlxVuyhQFCgKbDAFdgYohsYnnHBCt9biLLCY89bUW4YHLMAxnjDvSC7jK1/5yu6LL+985zu7cG1y+gIaA8xakLZsIJl3pfQ+YDDAlufw7W9/exdWP+aYY7pJJIA2AKet2pk2j1WGdnIMffM7fRYvccrkIwKIrvnIRz4SdjhUZQHFQ9Xd1diiQFGgKLASCuwcUEQ1oIZnMXmLgERARVsmTCnHEehZVu5iC5x44KwfaLYuTxwQBDiefvrpXT6jPL94HOU3Amt9kBgw15b9a9r/2+va/fYa7/G+5Bqqh5Dy61//+glgiD7yO+UdAofa0W9X+/9B94FEdGq9iEP9GJAo7cBSSPfee+9KBGcTX1JAcRN7pepUFCgKFAW2mwI7CRR1yQ033NB5oZK3OAQyAhoDGC3YDAgBQcvwkPXBUzxmltgBwoAiawKagPHa1762W9j7bW97WwcUATgAWGnj6bMFTAJ60zbX2HJPnuV5jvFsvuUtb5m84Q1v6L4+8/KXv7wDhvINAdp4Dfv1X8b/aCKE7b0//dM/vSfYT78ZFFjE+7D/Cigedg6o9hcFigJFgfEpsLNAEaksrgz0ARLTFmPug8Wf+Imf6GbsAm/LAELtM1uvHIAU4MiTBrDy4vE6Cln73/lXvepV3cSRk08+uZso8+Y3v3ly5plndrONfY+6vwGaf/Znf9Z5LP/kT/5k8kd/9EcdCOXFBMjkFgKp3gOcAaoBht7X1neZ+2m7dppsFE9h+qctc47H2PGzzz57fMnYwicWUNzCTqsqFwWKAkWBDafATgPF0J4HTe5aQtEt6Ojvx0tl3UUALaBpmSDJs1vQmHcBTwFrzguNy3UUrgbwADsbQGlT3/6Wc0rXus/9wKjnDb0rx4bqlHNjlQGIgLnQMaC+F6gPSAT+tfXWW29NNx/6soDioWeBIkBRoChQFBidAocCKKLav/3bv01MXJnXuwhY2kyiALAAxoC2sUDSrOcAavOAtVn1mnVePeZ916w6z3s+dQIQeTPliaI3INgH7/nfOWD/R3/0R7uw/OjSsOUPLKC45R1Y1S8KFAWKAhtIgUMDFEN76xrKW5w3dxF44eFaJ2AcAl8BdimHrsmxXJMyx9dRtgDRWpOWKYoHcRpIdNw1QD6wf91116U7q2woUECxIUbtFgWKAkWBosAoFDh0QBHVbrrppi7nb17vIqDSehjl9AndAloBPusAXdv2TkBV6NwkmXkBIronZeBP//RPR2H6XX1IAcVd7dlqV1GgKFAUWB8FDiVQDLnN+BXW5F2MVythzqGyBYxyGHnEhE4BoAKMT3/KsAWvoYlcSHmRchDbzy1O8yCifetFBMo/9alPpduqnEKBAopTCFOHiwJFgaJAUWDfFDjUQBHVHnzwwW5JGHlvwCIP1hBIbI+1gBHweelLX9pNEsnElwCkgCZAMvu7Vg61TfsdB6JNOPnZn/3ZjqZoOwscBiAC7z7V9573vGffzH3YbiygeNh6vNpbFCgKFAWWT4FDDxRD4htvvLEDNQCjkPQsUBOvV0Cj0rp/JmZYYsbSM0BhHzTuAlCcBg61Ld5Da0JmgkpoOQ0kOm4D1JM/agHyL3/5y+meKuegQAHFOYhUlxQFigJFgaLAQhQooNgj12WXXdbl0AGLcuMCclqPYn8/QMe1wI7zvGjzgMYh0LWJYFI9+3UNCOZJtewOcCgkHzqELn165f+cdz2A6P8TTzxxct999/V6pf6dhwIFFOehUl1TFCgKFAWKAotQoIDiFGpdeOGFXQ7iIoCxBUABSy94wQs60GgCBzAl3w7gArKytcBwCJC151ex3weE3tnWlbfUpBRrM5qFLEQM5CV0bz+0GCoDEF0f+vqyyu233z6lN+rwPBQooDgPleqaokBRoChQFFiEAgUU96DWk08+OTly5EgHGJ///OfP7WEMOAogAhpt/udx8+UR3kZfIeGNa8PULSDrg8IhANe/Zt7/9wKkqYPSdeon3xDQNYHnxS9+cRdW1p55wSGahB7u4UHksQUQrXFZv4NToIDiwWlYTygKFAWKAkWBZ1KggOIz6TH1vw9/+MPdzF2AMSAnwEcZcDitzLUBjUrXyuMDvPJdZeCR1xE4A9Ja0GZ/XiA467qh53of4AoUqoeJKGYqC6P7Yoo2pP5pz7T2tsddG3DIg2hfDuI999wzld51YnEKFFBcnGZ1R1GgKFAUKArsTYECinvT51lnLdPyB3/wBxMhZaAHaIy3ECBqAdJe+wFaSvcDTzb/A2UW+DabGlCzbqOvw5gkI+QLSAZMApQBlcBhPIXK9n/XAIHuAwTzKUDhY15C73nJS17STcjRtrZObfvmaWPbNm0yQQid5DBakuiJJ554Fl3rwMEpUEDx4DSsJxQFigJFgaLAMylQQPGZ9Jj7v//6r/+a/N3f/V3nZcxn5drJL/MAqgDJXNsCrBaoAVv+dz0QZ7FqIWyzrHn7eCRtQtrAZTbAz3HXAJ5yCd3Li+k53jf0ntQj9Zu3zH3qCxgCiN4nvHzttdfOTdu6cH8UKKC4P7rVXUWBokBRoCgwnQIFFKfTZu4zH//4xye+GsIrF+9ZwF3A07xga9p1eU5bAnn9zXuz9c/5v70/+9PeOc/xPMOzgUNeVsde/vKXT84555zJQw89NDcd68KDUaCA4sHoV3cXBYoCRYGiwLMpUEDx2TTZ9xHr/vGcnXrqqZ0Xr81nBN4CquYBYJt2jbqrU9oAGPKgxnOoFNY+++yzJ5/73Of2TcO6cf8UKKC4f9rVnUWBokBRoCgwTIECisN0GeXov/7rv07OOuusLr8QUORty4Le/TB1gNimAMQAwpQBhqm/UghbvuY//MM/TD7/+c+PQrN6yP4pUEBx/7SrO4sCRYGiQFFgmAIFFIfpMvrR//zP/5x88IMfnJx22mndRBI5hvHGAV2AIzA5LTy8DCAZENiW3h9QGGCrfo7JfTzuuOO6CSmf/OQnR6dRPfBgFCigeDD61d1FgaJAUaAo8GwKFFB8Nk1WcuThhx+efOxjH5v81V/91eS1r31tNynGZBOAUcgaOLMBkwGRAZIBcy3Am2c/9yk9y3NtAYR5r2dZSPuYY46ZvPGNb5y8973vndx2222TL33pSyuhTb1kfxQooLg/utVdRYGiQFGgKDCdAgUUp9Nm5WceeeSRyc033zz5wAc+0Hnt/viP/3jyu7/7u93yOGYPA3CA4/Oe97xuA+wC7gIs22M5l2Pucx1waNazGdGWyRE+futb3zo577zzJv/0T//UrW/4+OOPr7z99cKDUaCA4sHoV3cXBYoCRYHDRIF57XwBxS3hCt48eYB33333RO7jVVddNbn88ssn73//+yfnn39+t1TPX/7lX07aDfC74IILuhzCD33oQ5Prr7++8wz6lvKjjz66JS3fjWr+z//8z+T7v//7J1/1VV/VbT/4gz84uoe2gOJu8Eq1oihQFCgKLJMCJp1+0zd9U2eLnvvc53YTcPd6XwHFvahT5w4tBayTecYZZ0weeOCBUWgQgNiWz3nOc0Z5dh5SQDGUqLIoUBQoCuwGBUT5LrnkktEcC5xHrR3KvhS4ab8CitMoU8cPNQV4XBOi97nDiy++eGJC0n5+ACdh/Pqv//oJz6KfUZxj1t8c61dAcSxK1nOKAkWBosBmUOCyyy7rUs18NOOUU06ZXHfddQeq2Ld8y7d0tseX2/yszBKwOC0UXUDxQCSvm3eVAo899lj3JRvfuzbb20QjX7g58cQTJ777vchnCL/3e7+3E8Tf+q3fOkoun2QknD/wAz9w9NhBdwooHpSCdX9RoChQFNgsCkgxAxJ9atfX18xX8Ondt73tbZPPfOYzC1X2s5/97FFQeM899xy9N0DR/Iih31Sg+Ja3vGXo+jpWFDg0FAAOCeSrX/3q7rvZEVSfTvTdapONbrrpppn0+Nqv/dpOOAl2fu94xzu6Y9/8zd+cQwcuAxRvuOGGAz+rHlAUKAoUBYoC66fARz7yke6TvSa2/v7v/343AdWqJCajskW/8Ru/Mfnbv/3byRe+8IWZlb300kuPAsX24u/4ju/ojr/+9a9vDx/dfxZQtFyL2bVezkCaEVtb0eAw8cAf/uEfTmy/8Au/MPnlX/7lTjB/5Vd+pftm9fHHHz/51V/91W50R1ithwlA/sVf/MXgF2l8rSejte/7vu+b/Nqv/Vp3/w/90A91x7/ma75mtIlFvozj+97HHnvs5IQTTii5Ld1VPFA8UDywxTxAj1umjmOCXhfhkgrFPvlM7i/90i91dgpoZIsc44F88sknj4K8dufP//zPj9qjX/zFX+w+BsK2fcM3fEN3vI16tfc9Cyjefvvt3TIp1trz4tqKBoeRB4zUCCWgaNAEDPqWNw/jySef3AkvIXWNcAB5kdMIMLY/IeoAxaHyq7/6qycmzozx8/lIQFRo4jD2WbW5dFXxQPHArvGA1CegDlBkfzgwbBwDb3rTmzrHAzvERglRWwLPPUMfxRApHrJDOcbeDf2eBRRdJKHxr//6ryfnnnvu5JxzzqmtaHCoeMCyQu9+97u7RccJjhGcUR1BfcUrXtGBQ95EYJIHD3C8+uqrh+Rr8tRTTx0VTM+SmCwP5Ld/+7e741/3dV83+eIXvzh4734Oer66l+yW3irdXTxQPLDdPGCG8qmnnto5I6QW/c7v/E5ni0R72RDOC44K4Jhn8J3vfOfkrrvummo6ODICCvGGj2kIR3/bt31bd9w7hn6DQHHowjpWFDhsFJCjaKRGeHjqjNiAQ9+4fuUrX9kBvnnWozTbmXC2M5wJv2Pf/u3fftjIWu0tChQFigJFgTkpYA1ktug3f/M3O5BoImQmtSh9Pe0Tn/jEXE/jpAhQbG/41m/91u74mWee2R4+ul9A8Sgpaqco8BUKZNYzlz+vobxd+0ZhFj5f5Cc3kXAa/eVnBrVjvpBTv6JAUaAoUBQoCgxRQM4hB4Vws1Ka03HHHTf553/+5y5iNXTPtGM+thGgeMsttxy9LMekLw39CigOUaWOHXoK8BT65CFwZ5R155137psm0jgiiB/96Ecntvxvfcb+74orrhicGNO/rv4vChQFigJFgd2mwPve977uU71CzRdddNFCS7MNUeY7v/M7O/sjl/Hhhx/u8h7Zo2/8xm981uVmUv/jP/7jpIDis0hTB4oCkwmPovUSx/p9z/d8z1FwGJD4Iz/yI896/N/8zd9011nuoH5FgaJAUaAocLgp8OlPf3pikvFYP8unxQa1pa+/9H/f9V3f9fS1/RP1f1GgKLAcCpx22mmTH/7hH55YGscK+/2fpQ0iuEeOHOmfrv+LAkWBokBRoChwYArcdttt3Yoe3/3d392tx3jllVc+45nWB5Y/zx51X3J5xtn6pyhQFFgLBXgvCWXC1AUU19IN9dKiQFGgKHDoKeBrYj/+4z/epV51kzEPPUWKAEWBDaBAvgF9//33d4CxgOIGdEpVoShQFCgKHEIKPPLII12rrdv4nOc8p3IUDyEPVJM3mAJyUXgWCyhucCdV1YoCRYGiwCGggJU+nvvc5xZQPAR9XU3cIgoUUNyizqqqFgWKAkWBHaZAAcUd7txq2vZSoIDi9vZd1bwoUBQoCuwSBQoo7lJvVlt2hgIFFHemK6shRYGiQFFgqykQoPj/Acn/JhKSOW9XAAAAAElFTkSuQmCC'>\n",
        "</p>\n",
        "Figure 4-7. Gradient Descent with (left) and without (right) feature scaling"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "q83qBIkEhtC8"
      },
      "source": [
        "Gradient Descent goes down the steep slope quite fast, but then it takes a very long time to go down the valley. In contrast, momentum optimization will roll down the valley faster and faster until it reaches the bottom (the optimum)\n",
        "\n",
        "**In deep neural networks that don’t use Batch Normalization, the upper layers will often end up having inputs with very different scales, so using momentum optimization helps a lot. It can also help roll past local optima**\n",
        "\n",
        "Due to the momentum, the optimizer may overshoot a bit, then come back, overshoot again, and oscillate like this many times before stabilizing at the minimum. This is one of the reasons it’s good to have a bit of friction in the system: it gets rid of these oscillations and thus speeds up convergence\n",
        "\n",
        "Implementing momentum optimization in Keras is a no-brainer: just use the SGD optimizer and set its momentum hyperparameter"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lk5hYcghh_hF"
      },
      "source": [
        "optimizer = keras.optimizers.SGD(learning_rate=0.001, momentum=0.9)"
      ],
      "execution_count": 87,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "w7FS3hFQibfu"
      },
      "source": [
        "The one drawback of momentum optimization is that it adds yet another hyperparameter to tune. However, the momentum value of 0.9 usually works well in practice and almost always goes faster than regular Gradient Descent"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8YvpQ_RNig9a"
      },
      "source": [
        "## Nesterov Accelerated Gradient"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3Ux7RM6LijIl"
      },
      "source": [
        "One small variant to momentum optimization, proposed by [Yurii Nesterov in 1983](https://scholar.google.com/scholar?q=A+method+for+solving+the+convex+programming+problem+with+convergence+rate+author%3Anesterov), is almost always faster than vanilla momentum optimization\n",
        "\n",
        "The Nesterov Accelerated Gradient (NAG) method, also known as Nesterov momentum optimization, **measures the gradient of the cost function not at the local position $\\theta$ but slightly ahead in the direction of the momentum, at $\\theta + \\beta m$** (see Equation 11-5)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qMkZWU0xjL1q"
      },
      "source": [
        "Equation 11-5. Nesterov Accelerated Gradient algorithm\n",
        "\n",
        "$$m \\gets \\beta m - \\eta \\nabla_\\theta J(\\theta + \\beta m) \\\\ \\theta \\gets \\theta + m $$"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DPJ5NdGKjhn5"
      },
      "source": [
        "This small tweak works because in general **the momentum vector will be pointing in the right direction (i.e., toward the optimum)**, so it will be slightly more accurate to use the gradient measured a bit farther in that direction rather than the gradient at the original position, as you can see in Figure 11-6\n",
        "\n",
        "$\\nabla_1$ represents the gradient of the cost function measured at the starting point $\\theta$, and $\\nabla_2$ represents the gradient at the point located at $\\theta + \\beta m$)\n",
        "\n",
        "As you can see, the Nesterov update ends up slightly closer to the optimum. After a while, these small improvements add up and NAG ends up being significantly faster than regular momentum optimization\n",
        "\n",
        "Moreover, note that when the momentum pushes the weights across a valley, $\\nabla_1$ continues to push farther across the valley, while $\\nabla_2$ pushes back toward the bottom of the valley. This helps reduce oscillations and thus NAG converges faster"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hYjPkN7SkUcF"
      },
      "source": [
        "NAG is generally faster than regular momentum optimization. To use it, simply set nesterov=True when creating the SGD optimizer:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XRWTH_1QkY_y"
      },
      "source": [
        "optimizer = keras.optimizers.SGD(learning_rate=0.001, momentum=0.9, nesterov=True)"
      ],
      "execution_count": 88,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "uaVwf_4Wkej4"
      },
      "source": [
        "<p align='center'>\n",
        "<img src='data:image/png;base64,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'>\n",
        "</p>\n",
        "Figure 11-6. Regular versus Nesterov momentum optimization: the former applies the gradients computed before the momentum step, while the latter applies the gradients computed after"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RVWTYgpPk0z4"
      },
      "source": [
        "## AdaGrad"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NHfImCYDk2QV"
      },
      "source": [
        "Consider the elongated bowl problem again: Gradient Descent starts by quickly going down the steepest slope, which does not point straight toward the global optimum, then it very slowly goes down to the bottom of the valley\n",
        "\n",
        "It would be nice if the algorithm **could correct its direction earlier to point a bit more toward the global optimum**. The [AdaGrad algorithm](https://jmlr.org/papers/v12/duchi11a) **achieves this correction by scaling down the gradient vector along the steepest dimensions** (see Equation 11-6)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CW_8l2AIetng"
      },
      "source": [
        "Equation 11-6. AdaGrad algorithm\n",
        "\n",
        "$$ s \\gets s + \\nabla_\\theta J(\\theta) \\otimes \\nabla_\\theta J(\\theta) \\\\ \\theta \\gets \\theta - \\eta \\nabla_\\theta J(\\theta) ⊘ \\sqrt{s + \\epsilon}  $$"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xiOvpOtTfwlh"
      },
      "source": [
        "The first step accumulates the square of the gradients into the vector s (recall that the $\\otimes$ symbol represents the element-wise multiplication)\n",
        "\n",
        "This vectorized form is equivalent to computing\n",
        "\n",
        "$$s_i \\gets s_i + (\\partial J(\\theta) / \\partial \\theta_i)^2$$\n",
        "\n",
        "for each element $s_i$ of the vector s; in other words, each $s_i$ accumulates the squares of the partial derivative of the cost function with regard to parameter $\\theta_i$. If the cost function is steep along the $i^{th}$ dimension, then $s_i$ will get larger and larger at each iteration\n",
        "\n",
        "The second step is almost identical to Gradient Descent, but with one big difference: the gradient vector is scaled down by a factor of $\\sqrt{s + \\epsilon}$ (the ⊘ symbol represents the element-wise division, and $\\epsilon$ is a smoothing term to avoid division by zero, typically set to $10^{-10}$\n",
        "\n",
        "This vectorized form is equivalent to simultaneously computing\n",
        "\n",
        "$$\\theta_i \\gets \\theta_i - \\eta \\partial J(\\theta) / \\partial \\theta_i / \\sqrt{s_i + \\epsilon}$$\n",
        "\n",
        "for all parameters $\\theta_i$\n",
        "\n",
        "**In short, this algorithm decays the learning rate, but it does so faster for steep dimensions than for dimensions with gentler slopes. This is called an adaptive learning rate. It helps point the resulting updates more directly toward the global optimum** (see Figure 11-7)\n",
        "\n",
        "One additional benefit is that it requires much less tuning of the learning rate hyperparameter $\\eta$"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "K1EjEcOCikNS"
      },
      "source": [
        "<p align='center'>\n",
        "<img src='data:image/png;base64,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'>\n",
        "</p>\n",
        "Figure 11-7. AdaGrad versus Gradient Descent: the former can correct its direction earlier to point to the optimum"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8MiepZ4YjJxJ"
      },
      "source": [
        "**AdaGrad frequently performs well for simple quadratic problems, but it often stops too early when training neural networks**. The learning rate gets scaled down so much that the algorithm ends up stopping entirely before reaching the global optimum\n",
        "\n",
        "So even though Keras has an Adagrad optimizer, you should not use it to train deep neural networks (it may be efficient for simpler tasks such as Linear Regression)\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "g9lRvPmqjan7"
      },
      "source": [
        "## RMSProp"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "epH6jTAtjcc2"
      },
      "source": [
        "As we’ve seen, AdaGrad runs the risk of slowing down a bit too fast and never converging to the global optimum\n",
        "\n",
        "The RMSProp algorithm fixes this by **accumulating only the gradients from the most recent iterations** (as opposed to all the gradients since the beginning of training). It does so by **using exponential decay in the first step** (see Equation 11-7)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "njlUnnmioO4B"
      },
      "source": [
        "Equation 11-7. RMSProp algorithm\n",
        "\n",
        "$$ s $\\gets  \\beta s + (1 - \\beta) \\nabla_\\theta J(\\theta) \\otimes \\nabla_\\theta J(\\theta) \\\\ \\theta \\gets \\theta - \\eta \\nabla_\\theta J(\\theta) ⊘ \\sqrt{s + \\epsilon} $$"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fJL-EA70ox9n"
      },
      "source": [
        "The decay rate $\\beta$ is typically set to 0.9, this default value often works well, so you may not need to tune it at all\n",
        "\n",
        "Keras has an optimizer RMSprop"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2mZKGzczo-pY"
      },
      "source": [
        "optimizer = keras.optimizers.RMSprop(learning_rate=0.001, rho=0.9)"
      ],
      "execution_count": 89,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nBSoqHWkpHgH"
      },
      "source": [
        "Except on very simple problems, **this optimizer almost always performs much better than AdaGrad**. In fact, **it was the preferred optimization algorithm of many researchers until Adam optimization came around**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Wo7-hrzUpOhe"
      },
      "source": [
        "## Adam and Nadam Optimization"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "eirjBymJpZrQ"
      },
      "source": [
        "[Adam](https://arxiv.org/abs/1412.6980), which stands for **adaptive moment estimation, combines the ideas of momentum optimization and RMSProp**\n",
        "\n",
        "Just like momentum optimization, it **keeps track of an exponentially decaying average of past gradients**; and just like RMSProp, it **keeps track of an exponentially decaying average of past squared gradients** (see Equation 11-8)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "g4iDlpdYqBGL"
      },
      "source": [
        "Equation 11-8. Adam algorithm\n",
        "\n",
        "$$m \\gets \\beta_1 m - (1 - \\beta_1) \\nabla_\\theta J(\\theta) \\\\\n",
        "s \\gets \\beta_2 s + (1 - \\beta_2) \\nabla_\\theta J(\\theta) \\otimes \\nabla_\\theta J(\\theta) \\\\\n",
        "\\hat m \\gets \\frac {m} {1 - \\beta_1^T}\\\\\n",
        "\\hat s \\gets \\frac {s} {1 - \\beta_2^T} \\\\ \n",
        "\\theta \\gets \\theta + \\eta \\hat m ⊘ \\sqrt{\\hat s + \\epsilon}$$"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QhqSP-sArMnN"
      },
      "source": [
        "* t represents the iteration number (starting at 1)\n",
        "\n",
        "If you just look at steps 1, 2, and 5, you will notice Adam’s close similarity to both momentum optimization and RMSProp\n",
        "\n",
        "The only difference is that step 1 computes an exponentially decaying average rather than an exponentially decaying sum, but these are actually equivalent except for a constant factor (the decaying average is just 1 - $\\beta_1$ times the decaying sum)\n",
        "\n",
        "Steps 3 and 4 are somewhat of a technical detail: since m and s are initialized at 0, they will be biased toward 0 at the beginning of training, so these two steps will help boost m and s at the beginning of training\n",
        "\n",
        "The momentum decay hyperparameter $\\beta_1$ is typically initialized to 0.9, while the scaling decay hyperparameter $\\beta_2$ is often initialized to 0.999. As earlier, the smoothing term ε is usually initialized to a tiny number such as $10^{-7}$\n",
        "\n",
        "These are the default values for the Adam class ((to be precise, epsilon defaults to None, which tells Keras to use keras.backend.epsilon(), which defaults to $10^{-7}$; you can change it using keras.backend.set_epsilon())\n",
        "\n",
        "Here is how to create an Adam optimizer using Keras:\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NA3w4fsFuKJD"
      },
      "source": [
        "optimizer = keras.optimizers.Adam(learning_rate=0.001, beta_1=0.9, beta_2=0.999)"
      ],
      "execution_count": 90,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "d-EAei_TudVL"
      },
      "source": [
        "Since Adam is an adaptive learning rate algorithm (like AdaGrad and RMSProp), it requires less tuning of the learning rate hyperparameter $\\eta$\n",
        "\n",
        "You can often use the default value $\\eta = 0.001$, making Adam even easier to use than Gradient Descent"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HZksxEeKus98"
      },
      "source": [
        "Finally, two variants of Adam are worth mentioning\n",
        "\n",
        "**AdaMax**\n",
        "\n",
        "* Notice that in step 2 of Equation 11-8, Adam accumulates the squares of the gradients in s (with a greater weight for more recent weights)\n",
        "\n",
        "* In step 5, if we ignore ε and steps 3 and 4 (which are technical details anyway), Adam scales down the parameter updates by the square root of s\n",
        "\n",
        "* In short, Adam scales down the parameter updates by the $l_2$ norm of the time-decayed gradients (recall that the $l_2$ norm is the square root of the sum of squares)\n",
        "\n",
        "* AdaMax, introduced in the same paper as Adam, replaces the $l_2$ norm with the $l_\\infty$ norm (a fancy way of saying the max). Specifically, it replaces step 2 in Equation 11-8 with $s \\gets \\max (\\beta_2 s, \\nabla_\\theta J(\\theta))$ it drops step 4, and in step 5 it scales down the gradient updates by a factor of s, which is just the max of the time decayed gradients\n",
        "\n",
        "* In practice, this can make AdaMax more stable than Adam, but it really depends on the dataset, and in general Adam performs better. So, this is just one more optimizer you can try if you experience problems with Adam on some task\n",
        "\n",
        "**Nadam**\n",
        "\n",
        "* **Nadam optimization is Adam optimization plus the Nesterov trick**, so it will often converge slightly faster than Adam\n",
        "\n",
        "* In [his report introducing this technique](http://cs229.stanford.edu/proj2015/054_report.pdf), the researcher Timothy Dozat compares many different optimizers on various tasks and finds that Nadam generally outperforms Adam but is sometimes outperformed by RMSProp\n",
        "\n",
        "All the optimization techniques discussed so far only rely on the first order partial derivatives (Jacobians) The optimization literature also contains amazing algorithms based on the second-order partial derivatives (the Hessians, which are the partial derivatives of the Jacobians)\n",
        "\n",
        "Unfortunately, these algorithms are very hard to apply to deep neural networks because there are $n^2$ Hessians per output (where n is the number of parameters), as opposed to just n Jacobians per output. Since DNNs typically have tens of thousands of parameters, the second-order optimization algorithms often don’t even fit in memory, and even when they do, computing the Hessians is just too slow\n",
        "\n",
        "Table 11-2 compares all the optimizers we’ve discussed so far (* is bad, ** is average, and *** is good)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0mHX8mCU6WeL"
      },
      "source": [
        "Table 11-2. Optimizer comparison\n",
        "<p align='center'>\n",
        "<img src='data:image/png;base64,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'>\n",
        "</p>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nFFlxaPg6u4O"
      },
      "source": [
        "## Learning Rate Scheduling"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "whYdijQid5uP"
      },
      "source": [
        "Finding a good learning rate is very important. If you set it much too high, training may diverge (as we discussed in “Gradient Descent”). If you set it too low, training will eventually converge to the optimum, but it will take a very long time. If you set it slightly too high, it will make progress very quickly at first, but it will end up dancing around the optimum, never really settling down\n",
        "\n",
        "If you have a limited computing budget, you may have to interrupt training before it has converged properly, yielding a suboptimal solution (see Figure 11-8)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "r1ngskYQecO3"
      },
      "source": [
        "<p align='center'>\n",
        "<img src='data:image/png;base64,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'>\n",
        "</p>\n",
        "Figure 11-8. Learning curves for various learning rates η"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5w3KtDCXey0N"
      },
      "source": [
        "As we discussed in Chapter 10, you can **find a good learning rate by training the model for a few hundred iterations, exponentially increasing the learning rate from a very small value to a very large value, and then looking at the learning curve and picking a learning rate slightly lower than the one at which the learning curve starts shooting back up. You can then reinitialize your model and train it with that learning rate**\n",
        "\n",
        "But **you can do better than a constant learning rate**: if you start with a large learning rate and then reduce it once training stops making fast progress, you can reach a good solution faster than with the optimal constant learning rate\n",
        "\n",
        "**There are many different strategies to reduce the learning rate during training. It can also be beneficial to start with a low learning rate, increase it, then drop it again. These strategies are called learning schedules**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wLSOZy_ffwns"
      },
      "source": [
        "**Power scheduling**\n",
        "\n",
        "Set the learning rate to a function of the iteration number:\n",
        "$$t: \\eta (t) = \\eta_0 / (1 + t/s)^c$$\n",
        "\n",
        "The initial learning rate $\\eta_0$, the power c (typically set to 1), and the steps s are hyperparameters. The learning rate drops at each step \n",
        "\n",
        "After s steps, it is down to $\\eta_0 / 2$. After $s$ more steps, it is down to $\\eta_0 /3$, then it goes down to $\\eta_0 /4$ then $\\eta_0 / 5$ and so on\n",
        "\n",
        "As you can see, the schedule **first drops quickly, then more and more slowly**. Of course, power scheduling requires tuning $\\eta_0$ and $s$ (and possibly c)\n",
        " "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aCWEuxZMvBnB"
      },
      "source": [
        "**Exponential scheduling**\n",
        "\n",
        "Set the learning rate to\n",
        "$$\\eta (t) = \\eta_0 0.1^{t/s}$$\n",
        "\n",
        "The learning rate will gradually drop by a factor of 10 every s steps. While power scheduling reduces the learning rate more and more slowly, **exponential scheduling keeps slashing it by a factor of 10 every s steps**\n",
        "\n",
        "**Piecewise constant scheduling**\n",
        "\n",
        "Use a constant learning rate for a number of epochs (e.g., $\\eta_0 = 0.1$ for 5 epochs), then a smaller learning rate for another number of epochs (e.g., $\\eta_1 = 0.001$ for 50 epochs), and so on\n",
        "\n",
        "Although this solution can work very well, it requires fiddling around to figure out the right sequence of learning rates and how long to use each of them\n",
        "\n",
        "**Performance scheduling**\n",
        "\n",
        "Measure the validation error every N steps (just like for early stopping), and reduce the learning rate by a factor of $\\lambda$ when the error stops dropping\n",
        "\n",
        "**1cycle scheduling**\n",
        "\n",
        "1cycle (introduced in a [2018 paper](https://arxiv.org/abs/1803.09820) by Leslie Smith) starts by increasing the initial learning rate $\\eta_0$, growing linearly up to $\\eta_1$ halfway through training\n",
        "\n",
        "Then it decreases the learning rate linearly down to $\\eta_0$ again during the second half of training, finishing the last few epochs by dropping the rate down by several orders of magnitude (still linearly)\n",
        "\n",
        "The maximum learning rate $\\eta_1$ is chosen using the same approach we used to find the optimal learning rate, and the initial learning rate $\\eta_0$ is chosen to be roughly 10 times lower\n",
        "\n",
        "When using a momentum, we start with a high momentum first (e.g., 0.95), then drop it down to a lower momentum during the first half of training (e.g., down to 0.85, linearly), and then bring it back up to the maximum value (e.g., 0.95) during the second half of training, finishing the last few epochs with that maximum value\n",
        "\n",
        "Smith did many experiments showing that this approach was often able to speed up training considerably and reach better performance. For example, on the popular CIFAR10 image dataset, this approach reached 91.9% validation accuracy in just 100 epochs, instead of 90.3% accuracy in 800 epochs through a standard approach (with the same neural network architecture)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "I7xmwCoxrVL4"
      },
      "source": [
        "A [2013 paper](https://static.googleusercontent.com/media/research.google.com/en//pubs/archive/40808.pdf) by Andrew Senior et al. compared the performance of some of the most popular learning schedules when using momentum optimization to train deep neural networks for speech recognition. The authors concluded that, in this setting, both performance scheduling and exponential scheduling performed well\n",
        "\n",
        "They favored exponential scheduling because it was easy to tune and it converged slightly faster to the optimal solution (they also mentioned that it was easier to implement than performance scheduling, but in Keras both options are easy). That said, the 1cycle approach seems to perform even better\n",
        "\n",
        "**Implementing power scheduling in Keras is the easiest option: just set the decay hyperparameter when creating an optimizer**:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eATAbuHFrsKA"
      },
      "source": [
        "optimizer = keras.optimizers.SGD(learning_rate=0.01, decay=1e-4)"
      ],
      "execution_count": 91,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RXijMlJxrwXT"
      },
      "source": [
        "The decay is the inverse of s (the number of steps it takes to divide the decay learning rate by one more unit), and Keras assumes that c is equal to 1\n",
        "\n",
        "Exponential scheduling and piecewise scheduling are quite simple too. You first need to define a function that takes the current epoch and returns the learning rate. For example, let’s implement exponential scheduling:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Wzd-Xl3JsJ-1"
      },
      "source": [
        "def exponential_decay_fn(epoch):\n",
        "  return 0.01 * 0.1**(epoch/ 20)"
      ],
      "execution_count": 92,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6iQQyueRsQ6B"
      },
      "source": [
        "If you do not want to hardcode $\\eta_0$ and $s$, you can create a function that returns a configured function:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VAuM8l3fsfG7"
      },
      "source": [
        "def exponential_decay(lr0, s):\n",
        "  def exponential_decay_fn(epoch):\n",
        "    return lr0 * 0.1**(epoch / s)\n",
        "  return exponential_decay_fn"
      ],
      "execution_count": 93,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IQpnsRkmsXj8"
      },
      "source": [
        "Next, create a LearningRateScheduler callback, giving it the schedule function, and pass this callback to the fit() method:\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rXZDnsWotC_w"
      },
      "source": [
        "\n",
        "\n",
        "```\n",
        "lr_scheduler = keras.callbacks.LearningRateScheduler(exponential_decay_fn)\n",
        "history = model.fit(X_train_scaled, y_train, epochs=n_epochs,\n",
        "                    validation_data=(X_valid_scaled, y_valid),\n",
        "                    callbacks=[lr_scheduler])\n",
        "\n",
        "```\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oE4TUKCHtRDS"
      },
      "source": [
        "The LearningRateScheduler will update the optimizer’s learning_rate attribute at the beginning of each epoch. Updating the learning rate once per epoch is usually enough, but if you want it to be updated more often, for example at every step, you can always write your own callback \n",
        "\n",
        "Updating the learning rate at every step makes sense if there are many steps per epoch. Alternatively, you can use the keras.optimizers.schedules approach\n",
        "\n",
        "The schedule function can optionally take the current learning rate as a second argument. For example, the following schedule function multiplies the previous learning rate by $0.1^{1/20}$, which results in the same exponential decay (except the decay now starts at the beginning of epoch 0 instead of 1):"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mbTJATXuuetB"
      },
      "source": [
        "def exponential_decay_fn(epoch, lr):\n",
        "  return lr * 0.1**(1 / 20)"
      ],
      "execution_count": 94,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vPfXKze7uq1R"
      },
      "source": [
        "This implementation relies on the optimizer’s initial learning rate (contrary to the previous implementation), so make sure to set it appropriately\n",
        "\n",
        "**When you save a model, the optimizer and its learning rate get saved along with it**. This means that with this new schedule function, you could just load a trained model and continue training where it left off, no problem\n",
        "\n",
        "Things are not so simple if your schedule function uses the epoch argument, however: **the epoch does not get saved**, and it gets reset to 0 every time you call the fit() method. If you were to continue training a model where it left off, this could lead to a very large learning rate, which would likely damage your model’s weights\n",
        "\n",
        "One solution is to manually set the fit() method's initial_epoch argument so the epoch starts at the right value\n",
        "\n",
        "For piecewise constant scheduling, you can use a schedule function like the following one, then create a LearningRateScheduler callback with this function and pass it to the fit() method"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AtrgdT-3UWVA"
      },
      "source": [
        "def piecewise_constant_fn(epoch):\n",
        "  if epoch < 5:\n",
        "    return 0.01\n",
        "  elif epoch < 15:\n",
        "    return 0.005\n",
        "  else:\n",
        "    return 0.001"
      ],
      "execution_count": 95,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PS07QkdlUjmB"
      },
      "source": [
        "For performance scheduling, use the ReduceLROnPlateau callback. For example, if you pass the following callback to the fit() method, it will multiply the learning rate by 0.5 whenever the best validation loss does not improve for five consecutive epochs"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Hix9LPpKU7sw"
      },
      "source": [
        "lr_scheduler = keras.callbacks.ReduceLROnPlateau(factor=0.5, patience=5)"
      ],
      "execution_count": 96,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "J4pXKNayVCnk"
      },
      "source": [
        "Lastly, tf.keras offers an alternative way to implement learning rate scheduling: define the learning rate using one of the schedules available in keras.optimizers.schedules, then pass this learning rate to any optimizer. This approach updates the learning rate at each step rather than at each epoch\n",
        "\n",
        "For example, here is how to implement the same exponential schedule as the exponential_decay_fn() function we defined earlier:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fmrHer9VVqi6"
      },
      "source": [
        "s = 20 * len(X_train) // 32 # number of steps in 20 epochs (batch size = 32)\n",
        "learning_rate = keras.optimizers.schedules.ExponentialDecay(0.01, s, 0.1)\n",
        "optimizer = keras.optimizers.SGD(learning_rate)"
      ],
      "execution_count": 97,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jr6KzK-JWHth"
      },
      "source": [
        "This is nice and simple, plus when you save the model, the learning rate and its schedule (including its state) get saved as well. This approach, however, is not part of the Keras API; it is specific to tf.keras\n",
        "\n",
        "As for the 1cycle approach, the implementation poses no particular difficulty: just create a custom callback that modifies the learning rate at each iteration (you can update the optimizer’s learning rate by changing self.model.optimizer.lr)\n",
        "\n",
        "**To sum up, exponential decay, performance scheduling, and 1cycle can considerably speed up convergence**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XnchBmTIWdcO"
      },
      "source": [
        "# Avoiding Overfitting Through Regularization"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "241BbhGvWomM"
      },
      "source": [
        "Deep neural networks typically have tens of thousands of parameters, sometimes even millions. This gives them an incredible amount of freedom and means they can fit a huge variety of complex datasets. But this great flexibility also makes the network prone to overfitting the training set. We need regularization"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HU-KFElraYML"
      },
      "source": [
        "## $l_1$ and $l_2$ Regularization"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HLSlGuqFaeQL"
      },
      "source": [
        "You can use $l_2$ regularization to constrain a neural network's connection weights, and/or $l_1$ regularization if you want a sparse model (with many weights equal to 0)\n",
        "\n",
        "Here is how to apply $l_2$ regularization to a Keras layer’s connection weights, using a regularization factor of 0.01:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZuoBUM9NbQh6"
      },
      "source": [
        "layer = keras.layers.Dense(100, activation='elu',\n",
        "                           kernel_initializer='he_normal',\n",
        "kernel_regularizer=keras.regularizers.l2(0.01))"
      ],
      "execution_count": 98,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YrRyFElmb8qw"
      },
      "source": [
        "The l2() function returns a regularizer that will be called at each step during training to compute the regularization loss. This is then added to the final loss\n",
        "\n",
        "As you might expect, you can just use keras.regularizers.l1() if you want $l_1$ regularization; if you want both $l_1$ and $l_2$ regularization, use keras.regularizers.l1_l2() (specifying both regularization factors)\n",
        "\n",
        "Since you will typically want to apply the same regularizer to all layers in your network, as well as using the same activation function and the same initialization strategy in all hidden layers, you may find yourself repeating the same arguments\n",
        "\n",
        "This makes the code ugly and error-prone. To avoid this, **you can try refactoring your code to use loops**. Another option is to use **Python’s functools.partial() function**, which lets you create a thin wrapper for any callable, with some default argument values:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AIOX1HIccuf-"
      },
      "source": [
        "from functools import partial\n",
        "\n",
        "RegularizedDense = partial(keras.layers.Dense,\n",
        "                           activation='elu',\n",
        "                           kernel_initializer='he_normal',\n",
        "kernel_regularizer=keras.regularizers.l2(0.01))\n",
        "\n",
        "model = keras.models.Sequential([\n",
        "    keras.layers.Flatten(input_shape=[28, 28]),\n",
        "    RegularizedDense(300),\n",
        "    RegularizedDense(100),\n",
        "    RegularizedDense(10, activation='softmax',\n",
        "                     kernel_initializer='glorot_uniform')\n",
        "])"
      ],
      "execution_count": 99,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Ule4npL6dWck"
      },
      "source": [
        "## Dropout"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "X06Szxe2dYIy"
      },
      "source": [
        "Dropout is one of the most popular regularization techniques for deep neural networks. It was [proposed in a paper](https://arxiv.org/abs/1207.0580) by Geoffrey Hinton in 2012 and further detailed in a [2014 paper](https://jmlr.org/papers/v15/srivastava14a.html) by Nitish Srivastava et al., and it has proven to be highly successful: **even the state-of-the-art neural networks get a 1–2% accuracy boost simply by adding dropout**\n",
        "\n",
        "This may not sound like a lot, but when a model already has 95% accuracy, getting a 2% accuracy boost means dropping the error rate by almost 40% (going from 5% error to roughly 3%)\n",
        "\n",
        "It is a fairly simple algorithm: **at every training step, every neuron (including the input neurons, but always excluding the output neurons) has a probability p of being temporarily “dropped out,” meaning it will be entirely ignored during this training step, but it may be active during the next step** (see Figure 11-9)\n",
        "\n",
        "The hyperparameter p is called the dropout rate, and it is typically set between 10% and 50%: closer to 20–30% in recurrent neural nets (see Chapter 15), and closer to 40–50% in convolutional neural networks (see Chapter 14). After training, neurons don’t get dropped anymore"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Jy0s5vhSeoxA"
      },
      "source": [
        "<p align='center'>\n",
        "<img src='data:image/png;base64,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'>\n",
        "</p>\n",
        "Figure 11-9. With dropout regularization, at each training iteration a random subset of all neurons in one or more layers—except the output layer—are “dropped out”; these neurons output 0 at this iteration (represented by the dashed arrows)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pmPiymjwfcAt"
      },
      "source": [
        "Neurons trained with dropout cannot co-adapt with their neighboring neurons; they have to be as useful as possible on their own. They also cannot rely excessively on just a few input neurons; they must pay attention to each of their input neurons. **They end up being less sensitive to slight changes in the inputs**. In the end, you get a more robust network that generalizes better\n",
        "\n",
        "Another way to understand the power of dropout is to realize that **a unique neural network is generated at each training step. Since each neuron can be either present or absent, there are a total of $2^N$ possible networks** (where N is the total number of droppable neurons)\n",
        "\n",
        "This is such a huge number that it is virtually impossible for the same neural network to be sampled twice. Once you have run 10,000 training steps, you have essentially trained 10,000 different neural networks (each with just one training instance). These neural networks are obviously not independent because they share many of their weights, but they are nevertheless all different. **The resulting neural network can be seen as an averaging ensemble of all these smaller neural networks**\n",
        "\n",
        "In practice, you can usually apply dropout only to the neurons in the top one to three layers (excluding the output layer)\n",
        "\n",
        "There is one small but important technical detail. Suppose p = 50%, in which case during testing a neuron would be connected to twice as many input neurons as it would be (on average) during training\n",
        "\n",
        "To compensate for this fact, we need to multiply each neuron’s input connection weights by 0.5 after training. If we don’t, each neuron will get a total input signal roughly twice as large as what the network was trained on and will be unlikely to perform well\n",
        "\n",
        "More generally, we need to multiply each input connection weight by the keep probability (1 – p) after training. Alternatively, we can divide each neuron’s output by the keep probability during training (these alternatives are not perfectly equivalent, but they work equally well)\n",
        "\n",
        "To implement dropout using Keras, you can use the keras.layers.Dropout layer. **During training, it randomly drops some inputs (setting them to 0) and divides the remaining inputs by the keep probability. After training, it does nothing at all; it just passes the inputs to the next layer**\n",
        "\n",
        "The following code applies dropout regularization before every Dense layer, using a dropout rate of 0.2:\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GPLnrr-6HmgB",
        "outputId": "6970fc01-4cf0-4d98-c13d-f0de2dfd405e"
      },
      "source": [
        "model = keras.models.Sequential([\n",
        "    keras.layers.Flatten(input_shape=[28, 28]),\n",
        "    keras.layers.Dropout(rate=0.2),\n",
        "    keras.layers.Dense(300, activation=\"elu\", kernel_initializer=\"he_normal\"),\n",
        "    keras.layers.Dropout(rate=0.2),\n",
        "    keras.layers.Dense(100, activation=\"elu\", kernel_initializer=\"he_normal\"),\n",
        "    keras.layers.Dropout(rate=0.2),\n",
        "    keras.layers.Dense(10, activation=\"softmax\")\n",
        "])\n",
        "model.compile(loss=\"sparse_categorical_crossentropy\", optimizer=\"nadam\", metrics=[\"accuracy\"])\n",
        "n_epochs = 2\n",
        "history = model.fit(X_train_scaled, y_train, epochs=n_epochs,\n",
        "                    validation_data=(X_valid_scaled, y_valid))"
      ],
      "execution_count": 114,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/2\n",
            "1719/1719 [==============================] - 18s 10ms/step - loss: 0.5705 - accuracy: 0.8037 - val_loss: 0.3699 - val_accuracy: 0.8624\n",
            "Epoch 2/2\n",
            "1719/1719 [==============================] - 16s 9ms/step - loss: 0.4216 - accuracy: 0.8455 - val_loss: 0.3507 - val_accuracy: 0.8696\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "064_cUK5IO3u"
      },
      "source": [
        "Since dropout is only active during training, comparing the training loss and the validation loss can be misleading. In particular, a model may be overfitting the training set and yet have similar training and validation losses. So make sure to evaluate the training loss without dropout (e.g., after training)\n",
        "\n",
        "If you observe that the model is** overfitting, you can increase the dropout rate. Conversely, you should try decreasing the dropout rate if the model underfits the training set**. It can also help to increase the dropout rate for large layers, and reduce it for small ones. Moreover, many state-of-the-art architectures only use dropout after the last hidden layer, so you may want to try this if full dropout is too strong\n",
        "\n",
        "**Dropout does tend to significantly slow down convergence, but it usually results in a much better model when tuned properly**. So, it is generally well worth the extra time and effort"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GK1mrluTIrL5"
      },
      "source": [
        "## Monte Carlo (MC) Dropout"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8HApHsyFIy5f"
      },
      "source": [
        "In 2016, a [paper](https://arxiv.org/abs/1506.02142) by Yarin Gal and Zoubin Ghahramani added a few more good reasons to use dropout: \n",
        "\n",
        "* First, the paper established a profound connection between dropout networks (i.e., neural networks containing a Dropout layer before every weight layer) and approximate Bayesian inference, giving dropout a solid mathematical justification\n",
        "\n",
        "* Second, the authors introduced a powerful technique called MC Dropout, which can boost the performance of any trained dropout model without having to retrain it or even modify it at all, provides a much better measure of the model’s uncertainty, and is also amazingly simple to implement\n",
        "\n",
        "Take a look at the following code. It is the full implementation of MC Dropout, boosting the dropout model we trained earlier without retraining it"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "POwoqg50Ja12"
      },
      "source": [
        "y_probas = np.stack([model(X_test_scaled, training=True)\n",
        "                    for sample in range(100)])\n",
        "y_proba = y_probas.mean(axis=0)\n",
        "y_std = y_probas.std(axis=0)"
      ],
      "execution_count": 115,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RokDR4SsK2Ir"
      },
      "source": [
        "We just make 100 predictions over the test set, setting training=True to ensure that the Dropout layer is active, and stack the predictions. Since droupout is active, all the predictions will be different\n",
        "\n",
        "Recall that predict() returns a matrix with one row per instance and one column per class. Because there are 10,000 instances in the test set and 10 classes, this is a matrix of shape [10000, 10]\n",
        "\n",
        "We stack 100 such matrices, so y_probas is an array of shape [100, 10000, 10]. Once we average over the first dimension (axis=0), we get y_proba, an array of shape [10000, 10], like we would get with a single prediction\n",
        "\n",
        "That’s all! **Averaging over multiple predictions with dropout on gives us a Monte Carlo estimate that is generally more reliable than the result of a single prediction with dropout off**\n",
        "\n",
        "For example, let’s look at the model’s prediction for the first instance in the test set, with dropout off:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YjKtqXQbL77X",
        "outputId": "3783b433-fa08-471d-e742-74a478e8f7e7"
      },
      "source": [
        "np.round(model.predict(X_test_scaled[:1]), 2)"
      ],
      "execution_count": 116,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.  , 0.  , 0.  , 0.  , 0.  , 0.02, 0.  , 0.09, 0.  , 0.89]],\n",
              "      dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 116
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0EmpsMx1MRXm"
      },
      "source": [
        "The model seems almost certain that this image belongs to class 9 (ankle boot). Should you trust it? Is there really so little room for doubt? Compare this with the predictions made when dropout is activated:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "efaQkLMhMNPo",
        "outputId": "5d2d161f-3524-4f10-9c8a-0d7b740eb5d7"
      },
      "source": [
        "np.round(y_probas[:, :1], 2)"
      ],
      "execution_count": 117,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[[0.  , 0.  , 0.  , 0.  , 0.  , 0.06, 0.  , 0.09, 0.  , 0.85]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.06, 0.  , 0.08, 0.  , 0.86]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.01, 0.  , 0.25, 0.  , 0.73]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.19, 0.  , 0.39, 0.  , 0.42]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.07, 0.  , 0.28, 0.  , 0.65]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.35, 0.  , 0.09, 0.  , 0.56]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.19, 0.  , 0.44, 0.  , 0.37]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.06, 0.  , 0.35, 0.  , 0.6 ]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.01, 0.  , 0.03, 0.  , 0.96]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.02, 0.  , 0.11, 0.  , 0.87]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.14, 0.  , 0.43, 0.  , 0.43]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.19, 0.  , 0.4 , 0.  , 0.41]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.02, 0.  , 0.38, 0.  , 0.6 ]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.08, 0.  , 0.67, 0.  , 0.25]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.04, 0.  , 0.03, 0.  , 0.93]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.01, 0.  , 0.17, 0.  , 0.82]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.01, 0.  , 0.01, 0.  , 0.98]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.03, 0.  , 0.02, 0.  , 0.95]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.05, 0.  , 0.54, 0.  , 0.41]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.06, 0.  , 0.09, 0.  , 0.86]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.03, 0.  , 0.02, 0.  , 0.95]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.03, 0.  , 0.02, 0.  , 0.96]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.11, 0.  , 0.42, 0.  , 0.47]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.02, 0.  , 0.1 , 0.  , 0.88]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.01, 0.  , 0.1 , 0.  , 0.89]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.18, 0.  , 0.26, 0.  , 0.56]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.12, 0.  , 0.07, 0.  , 0.81]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.02, 0.  , 0.04, 0.  , 0.94]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.09, 0.  , 0.07, 0.  , 0.84]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.02, 0.  , 0.06, 0.  , 0.92]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.08, 0.  , 0.29, 0.  , 0.63]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.01, 0.  , 0.02, 0.  , 0.97]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.05, 0.  , 0.15, 0.  , 0.8 ]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.26, 0.  , 0.08, 0.  , 0.67]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.12, 0.  , 0.33, 0.  , 0.56]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.03, 0.  , 0.06, 0.  , 0.91]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.06, 0.  , 0.07, 0.  , 0.87]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.15, 0.  , 0.12, 0.  , 0.73]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.08, 0.  , 0.56, 0.  , 0.35]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.14, 0.  , 0.07, 0.  , 0.79]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.01, 0.  , 0.05, 0.  , 0.94]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.22, 0.  , 0.44, 0.  , 0.34]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.  , 0.  , 0.03, 0.  , 0.97]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.06, 0.  , 0.51, 0.  , 0.44]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.01, 0.  , 0.04, 0.  , 0.95]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.03, 0.  , 0.07, 0.  , 0.9 ]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.01, 0.  , 0.26, 0.  , 0.73]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.02, 0.  , 0.  , 0.  , 0.98]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.03, 0.  , 0.34, 0.  , 0.63]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.07, 0.  , 0.15, 0.  , 0.78]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.01, 0.  , 0.18, 0.  , 0.81]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.08, 0.  , 0.27, 0.  , 0.65]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.01, 0.  , 0.03, 0.  , 0.96]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.05, 0.  , 0.09, 0.  , 0.85]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.03, 0.  , 0.01, 0.  , 0.97]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.07, 0.  , 0.18, 0.  , 0.75]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.01, 0.  , 0.01, 0.  , 0.98]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.  , 0.  , 0.02, 0.  , 0.98]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.23, 0.  , 0.26, 0.  , 0.51]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.2 , 0.  , 0.53, 0.  , 0.27]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.02, 0.  , 0.07, 0.  , 0.92]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.01, 0.  , 0.06, 0.  , 0.93]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.12, 0.  , 0.01, 0.  , 0.87]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.12, 0.  , 0.21, 0.  , 0.67]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.08, 0.  , 0.16, 0.  , 0.75]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.02, 0.  , 0.02, 0.  , 0.96]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.01, 0.  , 0.1 , 0.  , 0.89]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.02, 0.  , 0.13, 0.  , 0.85]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.38, 0.  , 0.2 , 0.  , 0.42]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.01, 0.  , 0.13, 0.  , 0.86]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.  , 0.  , 0.13, 0.  , 0.87]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.59, 0.  , 0.03, 0.  , 0.38]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.04, 0.  , 0.7 , 0.  , 0.26]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.08, 0.  , 0.15, 0.  , 0.77]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.01, 0.  , 0.01, 0.  , 0.99]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.01, 0.  , 0.  , 0.  , 0.98]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.02, 0.  , 0.11, 0.  , 0.87]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.03, 0.  , 0.04, 0.  , 0.93]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.01, 0.  , 0.02, 0.  , 0.97]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.01, 0.  , 0.21, 0.  , 0.78]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.07, 0.  , 0.22, 0.  , 0.72]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.  , 0.  , 0.31, 0.  , 0.68]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.01, 0.  , 0.07, 0.  , 0.93]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.01, 0.  , 0.02, 0.  , 0.97]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.83, 0.  , 0.01, 0.  , 0.16]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.05, 0.  , 0.27, 0.  , 0.69]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.  , 0.  , 0.01, 0.  , 0.99]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.15, 0.  , 0.16, 0.  , 0.69]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.  , 0.  , 0.04, 0.  , 0.96]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.01, 0.  , 0.02, 0.  , 0.96]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.3 , 0.  , 0.06, 0.  , 0.63]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.02, 0.  , 0.07, 0.  , 0.92]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.05, 0.  , 0.43, 0.  , 0.53]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.18, 0.  , 0.09, 0.  , 0.73]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.  , 0.  , 0.1 , 0.  , 0.89]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.03, 0.  , 0.22, 0.  , 0.75]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.07, 0.  , 0.2 , 0.  , 0.73]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.35, 0.  , 0.46, 0.  , 0.19]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.03, 0.  , 0.06, 0.  , 0.91]],\n",
              "\n",
              "       [[0.  , 0.  , 0.  , 0.  , 0.  , 0.04, 0.  , 0.18, 0.  , 0.78]]],\n",
              "      dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 117
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rgk5TwcnMpkT"
      },
      "source": [
        "This tells a very different story: apparently, when we activate dropout, the model is not sure anymore. It still seems to prefer class 9, but sometimes it hesitates with classes 5 (sandal) and 7 (sneaker), which makes sense given they’re all footwear. Once we average over the first dimension, we get the following MC Dropout predictions:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nGR_edw6MuY6",
        "outputId": "aa326d79-7407-43bc-8920-f0bc5232e6b1"
      },
      "source": [
        "np.round(y_proba[:1], 2)"
      ],
      "execution_count": 118,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.  , 0.  , 0.  , 0.  , 0.  , 0.08, 0.  , 0.17, 0.  , 0.75]],\n",
              "      dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 118
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Z00vaKSxRPdu"
      },
      "source": [
        "The model still thinks this image belongs to class 9, but only with a (62%) confidence, which seems much more reasonable than (99%)\n",
        "\n",
        "Plus it’s useful to know exactly which other classes it thinks are likely. And you can also take a look at the [standard deviation of the probability estimates](https://xkcd.com/2110/):"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "S4CwHoumReMD",
        "outputId": "854bb99c-742e-4c6f-8ed9-e698805e15c7"
      },
      "source": [
        "s_std = y_probas.std(axis=0)\n",
        "np.round(y_std[:1], 2)"
      ],
      "execution_count": 119,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.  , 0.  , 0.  , 0.  , 0.  , 0.12, 0.  , 0.16, 0.  , 0.22]],\n",
              "      dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 119
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Du6RDyXDRlgW"
      },
      "source": [
        "Apparently there’s quite a lot of variance in the probability estimates: if you were building a risk-sensitive system (e.g., a medical or financial system), you should probably treat such an uncertain prediction with extreme caution\n",
        "\n",
        "You definitely would not treat it like a (99%) confident prediction. Moreover, the model’s accuracy got a small boost from 0.8455(86.8) to 0.8557(86.9):"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qrcCNxdzR3ei",
        "outputId": "d11b2797-5b5c-4d69-9154-227213571523"
      },
      "source": [
        "y_pred = np.argmax(y_proba, axis=1)\n",
        "accuracy = np.sum(y_pred == y_test) / len(y_test)\n",
        "accuracy"
      ],
      "execution_count": 121,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.8557"
            ]
          },
          "metadata": {},
          "execution_count": 121
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ogEF07DMSTAU"
      },
      "source": [
        "The number of Monte Carlo samples you use (100 in this example) is a hyperparameter you can tweak. The higher it is, the more accurate the predictions and their uncertainty estimates will be. However, if you double it, inference time will also be doubled. Moreover, above a certain number of samples, you will notice little improvement. So your job is to find the right trade-off between latency and accuracy, depending on your application\n",
        "\n",
        "**If your model contains other layers that behave in a special way during training (such as BatchNormalization layers), then you should not force training mode like we just did. Instead you should replace the Dropout layers with the following MCDropout class**:\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZpW-LoPyS5XK"
      },
      "source": [
        "class MCDropout(keras.layers.Dropout):\n",
        "  def call(self, inputs):\n",
        "    return super().call(inputs, training=True)"
      ],
      "execution_count": 122,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SnVV7MGKTA7O"
      },
      "source": [
        "Here, we just subclass the Dropout layer and override the call() method to force its training argument to True. Similarly, you could define an MCAlphaDropout class by subclassing AlphaDropout instead\n",
        "\n",
        "If you are creating a model from scratch, it’s just a matter of using MCDropout rather than Dropout. But if you have a model that was already trained using Dropout, you need to create a new model that's \n",
        "the existing model except that it replaces the Dropout layers with MCDropout, then copy the existing model's weights to your new model\n",
        "\n",
        "In short, MC Dropout is a fantastic technique that boosts dropout models and provides better uncertainty estimates. And of course, since it is just regular dropout during training, it also acts like a regularizer"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "M_7Nj0FeTpyR"
      },
      "source": [
        "## Max-Norm Regularization"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RlaAC3GITxV6"
      },
      "source": [
        "Another regularization technique that is popular for neural networks is called **max-norm regularization: for each neuron, it constrains the weights $w$ of the incoming connections such that $||w||_2 \\leq r$, where $r$ is the max-norm hyperparameter and $||.||_2$ is the $l_2$ norm**\n",
        "\n",
        "Max-norm regularization does not add a regularization loss term to the overall loss function. Instead, it is typically implemented by computing $||w||_2$ after each training step and rescaling $w$ if neexed ($w \\gets w \\frac{\\tau}{||w||_2}$)\n",
        "\n",
        "Reducing r increases the amount of regularization and helps reduce overfitting. Max-norm regularization can also help alleviate the unstable gradients problems (if you are not using Batch Normalization)\n",
        "\n",
        "To implement max-norm regularization in Keras, set the kernel_constraint argument of each hidden layer to a max_norm() constraint with the appropriate max value, like this:\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HRX3myZnVF7Z",
        "outputId": "0ce229c4-ae48-44b1-ed87-2a84b7e9a1d3"
      },
      "source": [
        "keras.layers.Dense(100, activation='elu', kernel_initializer='he_normal',\n",
        "                   kernel_constraint=keras.constraints.max_norm(1.))"
      ],
      "execution_count": 123,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.layers.core.Dense at 0x7efd36255c50>"
            ]
          },
          "metadata": {},
          "execution_count": 123
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3DEPLcKQVRLo"
      },
      "source": [
        "After each training iteration, the model’s fit() method will call the object returned by max_norm(), passing it the layer's weights and getting rescaled weights in return, which then replace the layer's weights\n",
        "\n",
        "As you’ll see in Chapter 12, you can define your own custom constraint function if necessary and use it as the kernel_constraint. You can also constrain the bias term by setting the bias_constraint argument\n",
        "\n",
        "The max_norm() function has an axis argument that defaults to 0. A Dense layer usually has weights of shape [number of inputs, number of neurons], so using axis=0 means that max-norm constraint will apply independently to each neuron's weight vector\n",
        "\n",
        "If you want to use max norm with convolutional layers (see Chapter 14), make sure to set the max_norm() constraint's axis argument appropriately (usualy axis=[0, 1, 2])"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "i5qVLfmvWEXQ"
      },
      "source": [
        "# Summary and Practical Guidelines"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tMfuzMcMWVmb"
      },
      "source": [
        "You may be wondering which ones you should use. This depends on the task, and there is no clear consensus yet, but I have found the configuration in Table 11-3 to work fine in most cases, without requiring much hyperparameter tuning"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tr7imgcMWW4l"
      },
      "source": [
        "Table 11-3. Default DNN configuration\n",
        "<p align='center'>\n",
        "<img src='data:image/png;base64,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'>\n",
        "</p>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "c251jlAtWlMl"
      },
      "source": [
        "the network is a simple stack of dense layers, then it can self-normalize, and you should use the configuration in Table 11-4 instead"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rxLdFar0Wpb-"
      },
      "source": [
        "Table 11-4. DNN configuration for a self-normalizing net\n",
        "<p align='center'>\n",
        "<img src='data:image/png;base64,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'>\n",
        "</p>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "l4KweKaodI07"
      },
      "source": [
        "Don’t forget to normalize the input features! You should also try to reuse parts of a pretrained neural network if you can find one that solves a similar problem, or use unsupervised pretraining if you have a lot of unlabeled data, or use pretraining on an auxiliary task if you have a lot of labeled data for a similar task\n",
        "\n",
        "While the previous guidelines should cover most cases, here are some exceptions:\n",
        "\n",
        "* **If you need a sparse model, you can use $l_1$ regularization** (and optionally zero out the tiny weights after training). If you need an even sparser model, you can use the TensorFlow Model Optimization Toolkit. This will break self-normalization, so you should use the default configuration in this case\n",
        "\n",
        "* **If you need a low-latency model (one that performs lightning-fast predictions), you may need to use fewer layers, fold the Batch Normalization layers into the previous layers, and possibly use a faster activation function such as leaky ReLU or just ReLU**. Having a sparse model will also help. Finally, you may want to reduce the float precision from 32 bits to 16 or even 8 bits\n",
        "\n",
        "* **If you are building a risk-sensitive application, or inference latency is not very important in your application, you can use MC Dropout to boost performance and get more reliable probability estimates, along with uncertainty estimates**"
      ]
    }
  ]
}