{
"cells": [
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"$$\n",
"\\newcommand{\\mat}[1]{\\boldsymbol {#1}}\n",
"\\newcommand{\\mattr}[1]{\\boldsymbol {#1}^\\top}\n",
"\\newcommand{\\matinv}[1]{\\boldsymbol {#1}^{-1}}\n",
"\\newcommand{\\vec}[1]{\\boldsymbol {#1}}\n",
"\\newcommand{\\vectr}[1]{\\boldsymbol {#1}^\\top}\n",
"\\newcommand{\\rvar}[1]{\\mathrm {#1}}\n",
"\\newcommand{\\rvec}[1]{\\boldsymbol{\\mathrm{#1}}}\n",
"\\newcommand{\\diag}{\\mathop{\\mathrm {diag}}}\n",
"\\newcommand{\\set}[1]{\\mathbb {#1}}\n",
"\\newcommand{\\norm}[1]{\\left\\lVert#1\\right\\rVert}\n",
"\\newcommand{\\pderiv}[2]{\\frac{\\partial #1}{\\partial #2}}\n",
"\\newcommand{\\bb}[1]{\\boldsymbol{#1}}\n",
"$$\n",
"\n",
"\n",
"# CS236781: Deep Learning\n",
"# Tutorial 8: Sequence Models"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"## Introduction\n",
"\n",
"In this tutorial, we will cover:\n",
"\n",
"- What RNNs are and how they work\n",
"- Implementing basic RNNs models\n",
"- Application example: sentiment analysis of movie reviews\n",
"- Bonus: Stock prediction\n"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [],
"source": [
"# Setup\n",
"%matplotlib inline\n",
"import os\n",
"import sys\n",
"import time\n",
"import torch\n",
"import matplotlib.pyplot as plt\n",
"import warnings\n",
"warnings.simplefilter(\"ignore\")\n",
"\n",
"# pytorch\n",
"import torch\n",
"import torch.nn as nn\n",
"import torchtext\n",
"import torchtext.data as data\n",
"import torchtext.datasets as datasets\n",
"import torch.nn.functional as f"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [],
"source": [
"plt.rcParams['font.size'] = 20\n",
"data_dir = os.path.expanduser('~/.pytorch-datasets')\n",
"device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"## Theory Reminders"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"source": [
"Thus far, our models have been composed of fully connected (linear) layers or convolutional layers."
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"- Fully connected layers\n",
" - Each layer $l$ operates on the output of the previous layer ($\\vec{y}_{l-1}$) and calculates,\n",
" $$\n",
" \\vec{y}_l = \\varphi\\left( \\mat{W}_l \\vec{y}_{l-1} + \\vec{b}_l \\right),~\n",
" \\mat{W}_l\\in\\set{R}^{n_{l}\\times n_{l-1}},~ \\vec{b}_l\\in\\set{R}^{n_l}.\n",
" $$\n",
" - FC's have completely pre-fixed input and output dimensions.\n",
" \n",
"
"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"How can we model a dynamical system?\n",
"
\n",
"\n",
"\n",
"E.g., a linear system such as\n",
"\n",
"$$\\vec{y}_t = a_0 + a_1 \\vec{y}_{t-1}+\\dots+a_P \\vec{y}_{t-P} + b_0 \\vec{x}_t+\\dots+b_{t-Q}\\vec{x}_{t-Q}$$\n",
"\n",
"Many use cases and examples: signal processing, text translation, sentiment analysis, scene classification in video, etc."
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"## NLP Task summery\n",
"\n",
"\n",
"
"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"#### A quick review of the Naive Bayes (NB) method for text classification:\n",
"\n",
"input vector $\\mathbf{x} = \\langle x_1, x_2, \\ldots, x_m \\rangle$ of feature values, and a set of classes $\\{ c_1, c_2, \\ldots, c_k \\}$ the text should be classified as. \n",
"\n",
"We will predict the class $c_i$ that maximizes $Pr(c_i \\mid \\mathbf{x})$.\n",
"\n",
"By Bayes rule (this is the \"Bayes\" part in the name \"Naive Bayes\"), \n",
"\n",
"$$ argmax_i Pr(c_i | x) = argmax_i \\frac{Pr(\\mathbf{x} | c_i) \\cdot Pr(c_i)}{Pr(\\mathbf{x})} = argmax_i Pr(\\mathbf{x} | c_i) \\cdot Pr(c_i) $$\n",
"\n",
"What's naive?"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"source": [
"NB called like that because it used the bayes rule, and because it's navie...assuming I.I.D, i.e:\n",
"$$ Pr(\\mathbf{x} | c_i) = \\prod_{j=1}^m Pr(x_j | c_i) $$\n",
"\n",
"What is $Pr(x_j | c_i)$ ?"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"$$ Pr(x_j | c_i) \\approx \\frac{\\# (x_j, c_i)}{\\sum_k \\# (x_k, c_i)} $$\n",
"\n",
"This counting basic approch. is called **Bag Of Words (BOW)**.\n",
"\n",
"in this approch, we count all the words in the document and use each one as a feature.\n",
"in general, expressing features as raw words is very easy to use with naive base.\n",
"\n",
"
\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"you probably see the problem right away. \n",
"The strongest features are now very expresive.. 'the', ',' and 'very' doesn't say much about this review.\n",
"\n",
"That's why we need to express, not only the document frequncies, but also the general word frequncies. \n",
"something like that is called **TF-IDF** as **Term frequency–inverse document frequency** \n",
"\n",
"
\n",
"\n",
"\n",
"**TF** Like BOW, **IDF** define as $log(\\frac{N}{Df_x})$ where N is the num of docs, and Df is the number of documents containing the term x.\n",
"\n",
"$$\\# (x_j, c_i) log(\\frac{\\sum_i \\# (c_i)}{\\sum_i \\# (c_i,x_j)}) $$\n",
"\n",
"and we solved one problem, however, order of words still matter and naive base, ever with smarter representation, is still naive in the notion of corrolation and order."
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"#### Now text generation, in a nutshell:\n",
"\n",
"A model that predict the next token in a sentence, called **N-gram**.\n",
"\n",
"recall that we have $m$ possible tokens in our corpus, we want to choose the word that maximize:\n",
"$$\n",
"Pr(x_n | x_1, x_2, ..., x_{n-1}) \\approx \\frac{\\#{x_1, x_2, ..., x_n}}{\\#{x_1, x_2, ..., x_{n-1}}} = \\frac{\\#{x_1, x_2, ..., x_n}}{\\sum_{x'} \\#{x_1, x_2, ..., x_{n-1}, x'}}\n",
"$$\n",
"\n",
"were the lenght of the sequnce that we depand on is the **N**. \n",
"\n",
"
\n",
"\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"source": [
"Now Let's dive into the components that we have in deep learning"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"## Recurrent layers"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"An RNN layer is similar to a regular FC layer, but it has two inputs:\n",
"- Current sample, $\\vec{x}_t \\in\\set{R}^{d_{i}}$.\n",
"- Previous **state**, $\\vec{h}_{t-1}\\in\\set{R}^{d_{h}}$.\n",
"\n",
"and it produces two outputs which depend on both:\n",
"- Current layer output, $\\vec{y}_t\\in\\set{R}^{d_o}$.\n",
"- Current **state**, $\\vec{h}_{t}\\in\\set{R}^{d_{h}}$."
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"
\n",
"\n",
"Crucially,\n",
"- The layer itself is not time-dependent (but is parametrized).\n",
"- The same layer (function) is applied at successive time steps, propagating the hidden state."
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"A basic RNN can be defined as follows.\n",
"\n",
"$$\n",
"\\begin{align}\n",
"\\forall t \\geq 0:\\\\\n",
"\\vec{h}_t &= \\varphi_h\\left( \\mat{W}_{hh} \\vec{h}_{t-1} + \\mat{W}_{xh} \\vec{x}_t + \\vec{b}_h\\right) \\\\\n",
"\\vec{y}_t &= \\varphi_y\\left(\\mat{W}_{hy}\\vec{h}_t + \\vec{b}_y \\right)\n",
"\\end{align}\n",
"$$\n",
"\n",
"where,\n",
"- $\\vec{x}_t \\in\\set{R}^{d_{i}}$ is the input at time $t$.\n",
"- $\\vec{h}_{t-1}\\in\\set{R}^{d_{h}}$ is the **hidden state** of a fixed dimension.\n",
"- $\\vec{y}_t\\in\\set{R}^{d_o}$ is the output at time $t$.\n",
"- $\\mat{W}_{hh}\\in\\set{R}^{d_h\\times d_h}$, $\\mat{W}_{xh}\\in\\set{R}^{d_h\\times d_i}$, $\\mat{W}_{hy}\\in\\set{R}^{d_o\\times d_h}$, $\\vec{b}_h\\in\\set{R}^{d_h}$ and $\\vec{b}_y\\in\\set{R}^{d_o}$ are the model weights and biases.\n",
"- $\\varphi_h$ and $\\varphi_y$ are some non-linear functions. In many cases $\\varphi_y$ is not used."
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"### Modeling time-dependence\n",
"\n",
"If we imagine **unrolling** a single RNN layer through time,\n",
"
\n",
"\n",
"We can see how late outputs can now be influenced by early inputs, through the hidden state."
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"RNN models are very flexible in terms of input and output meaning.\n",
"\n",
"Common applications include image captioning, sentiment analysis, machine translation and signal processing. \n",
"\n",
"
\n",
"\n",
"1. Calculated loss from each output and accumulate\n",
"2. Calculate Gradient of loss w.r.t. each parameter at each timestep\n",
"3. For each parameter, accumulate gradients from all timesteps"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"This is known as **Backpropagation through time**, or BPTT.\n",
"\n",
"$$\n",
"\\pderiv{L_t}{\\mat{W}} = \\sum_{k=1}^{t}\n",
"\\pderiv{L_t}{\\hat y_t} \\cdot\n",
"\\pderiv{\\hat y_t}{\\vec{h}_t} \\cdot\n",
"\\pderiv{\\vec{h}_t}{\\vec{h}_k} \\cdot\n",
"\\pderiv{\\vec{h}_k}{\\mat{W}}\n",
"$$\n",
"\n",
"But how far back do we go? What's the limiting factor?"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"We're limited in depth by vanishing and exploding gradients controlled by the eigenvalues of $\\mat{W}$.\n",
"\n",
"One pragmatic solution is to limit the number of timesteps involved in the backpropagation.\n",
"\n",
"
\n",
"\n",
"This is known as **Truncated backpropagation through time**, or TBPTT."
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"### Gradient clipping\n",
"A popular method to avoid exploading gradients, is to use `gradients cliping`\n",
"\n",
"the idea is super simple, yet efficient only for exploading and not for vanishing gradients\n",
"\n",
"the idea is as such:\n",
"look at a layer $\\matrix{L}$ with gradient matrix $\\matrix{G}$\n",
"\n",
"we would simply look at the norm of the gradients and when they exceed some threshold, we would clip it down.\n",
"\n",
"i.e: if $\\norm{\\matrix{G}} \\ge c$\n",
"$$\n",
"\\matrix{G} = C \\frac{\\matrix{G}}{\\norm{\\matrix{G}}}\n",
"$$\n",
"\n",
"and the process would look sort of like this:\n",
"
"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"### Multi-layered (deep) RNN"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"RNNs layers can be stacked to build a deep RNN model.\n",
"\n",
"
\n",
"\n",
"- As with MLPs, adding depth allows us to model intricate hierarchical features.\n",
"- However, now we also have a time dimension which makes the representation time-dependent."
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"## RNN Implementation"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"source": [
"Based on the above equations, let's create a simple RNN layer with PyTorch."
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [],
"source": [
"import torch.nn as nn\n",
"\n",
"class RNNLayer(nn.Module):\n",
" def __init__(self, in_dim, h_dim, out_dim, phi_h=torch.tanh, phi_y=torch.sigmoid):\n",
" super().__init__()\n",
" self.phi_h, self.phi_y = phi_h, phi_y\n",
" \n",
" self.fc_xh = nn.Linear(in_dim, h_dim, bias=False)\n",
" self.fc_hh = nn.Linear(h_dim, h_dim, bias=True)\n",
" self.fc_hy = nn.Linear(h_dim, out_dim, bias=True)\n",
" \n",
" def forward(self, xt, h_prev=None):\n",
" if h_prev is None:\n",
" h_prev = torch.zeros(xt.shape[0], self.fc_hh.in_features)\n",
" \n",
" ht = self.phi_h(self.fc_xh(xt) + self.fc_hh(h_prev))\n",
" \n",
" yt = self.fc_hy(ht)\n",
" \n",
" if self.phi_y is not None:\n",
" yt = self.phi_y(yt)\n",
" \n",
" return yt, ht\n",
" "
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"We'll instantiate our model"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [
{
"data": {
"text/plain": [
"RNNLayer(\n",
" (fc_xh): Linear(in_features=1024, out_features=10, bias=False)\n",
" (fc_hh): Linear(in_features=10, out_features=10, bias=True)\n",
" (fc_hy): Linear(in_features=10, out_features=1, bias=True)\n",
")"
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"N = 3 # batch size\n",
"in_dim, h_dim, out_dim = 1024, 10, 1\n",
"\n",
"rnn = RNNLayer(in_dim, h_dim, out_dim)\n",
"rnn"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"And manually \"run\" a few time steps"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"y1 ((3, 1)):\n",
"tensor([[0.5200],\n",
" [0.4882],\n",
" [0.4751]], grad_fn=)\n",
"h1 ((3, 10)):\n",
"tensor([[ 0.1455, -0.5251, 0.0520, 0.1736, 0.2086, -0.3636, 0.3703, 0.5832,\n",
" -0.0415, -0.3400],\n",
" [ 0.5498, 0.7924, 0.2217, -0.7610, -0.0093, 0.5861, -0.2828, 0.5821,\n",
" -0.8824, -0.0300],\n",
" [ 0.3849, -0.5013, 0.8767, 0.2038, -0.0695, -0.7919, 0.2710, 0.6845,\n",
" -0.2400, -0.3216]], grad_fn=)\n",
"\n",
"y2 ((3, 1)):\n",
"tensor([[0.5586],\n",
" [0.5398],\n",
" [0.5812]], grad_fn=)\n",
"h2 ((3, 10)):\n",
"tensor([[ 0.8280, -0.6389, 0.5552, -0.1514, 0.7150, -0.9303, -0.6738, 0.6184,\n",
" -0.7540, -0.2250],\n",
" [ 0.4347, -0.2782, 0.9537, 0.2481, 0.3160, 0.1474, -0.7193, -0.4658,\n",
" -0.4856, 0.7382],\n",
" [-0.4521, -0.7508, 0.2275, 0.0179, -0.4076, 0.3369, 0.0820, -0.8876,\n",
" -0.0320, -0.6079]], grad_fn=)\n",
"\n"
]
}
],
"source": [
"# t=1\n",
"x1 = torch.randn(N, in_dim, requires_grad=True) # requiring grad just for torchviz\n",
"y1, h1 = rnn(x1)\n",
"print(f'y1 ({tuple(y1.shape)}):\\n{y1}')\n",
"print(f'h1 ({tuple(h1.shape)}):\\n{h1}\\n')\n",
"\n",
"# t=2\n",
"x2 = torch.randn(N, in_dim, requires_grad=True)\n",
"y2, h2 = rnn(x2, h1)\n",
"print(f'y2 ({tuple(y2.shape)}):\\n{y2}')\n",
"print(f'h2 ({tuple(h2.shape)}):\\n{h2}\\n')"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"As usual, let's visualize the computation graph and see what happened when we used the same RNN block twice, by looking at the graph from both $y_1$ and $y_2$."
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"data": {
"image/svg+xml": [
"\n",
"\n",
"\n",
"\n",
"\n"
],
"text/plain": [
""
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"import torchviz\n",
"\n",
"torchviz.make_dot(\n",
" y2, # Note: Change here to y2 to see the fullly unrolled graph!\n",
" params=dict(list(rnn.named_parameters()) + [('x1', x1), ('x2', x2)])\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"## LSTM and GRU\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"We talked about the vanishing gradients problem with RNN, however, we didn't mention how to deal with those.\n",
"Luckyly, Gradient based methods invented long time ago...\n",
"Today we will talk about **Long Short term Memory (LSTM)** and **Gated Recurrent Unit (GRU)**. \n",
"\n",
"both are a type of recurrent cell that tries to preserve long term information. The idea of LSTM was presented back in 1997, but flourished in the age of deep learning.\n",
"\n",
"The cell have `memory` or `context` that's base on the `state` of basic RNN and on 3 main gates: \n",
"\n",
"- Input gate: decides when to read data into the cell.\n",
"- Output gate: outputs the entries from the cell.\n",
"- Forget gate: a mechanism to reset the content of the cell.\n",
"These gates learn which information is relevant to forget or remember during the training process. The gates contain a non-linear activision function (sigmoid).\n",
"\n",
"\n",
"* we denote the input $X_t \\in \\mathbb{R}^{n\\times d}$ at timestemp $t$ and the hidden state of the previous time step is $H_{t−1}\\in \\mathbb{R}^{n\\times h}$.\n",
"\n",
"\n",
"\n",
"
\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"- **Forget gate**: $$ F_t = \\sigma(X_tW_{xf} +H_{t-1}W_{hf} +b_f) \\in \\mathbb{R}^{n\\times f}, $$\n",
"- **Input gate**: $$ I_t = \\sigma(X_tW_{xi} +H_{t-1}W_{hi} +b_i) \\in \\mathbb{R}^{n\\times h},$$\n",
"- **Output gate**: $$ O_t = \\sigma(X_tW_{xo} +H_{t-1}W_{ho} +b_o) \\in \\mathbb{R}^{n\\times o}, $$\n",
"\n",
"- **candidate memory**:$$ \\tilde{C}_t = \\text{tanh}(X_tW_{xc} +H_{t-1}W_{hc} +b_c) \\in \\mathbb{R}^{n\\times c}$$\n",
"\n",
"- **memory / context**: $$ C_t = F_t \\odot C_{t-1} + I_{t} \\odot \\tilde{C}_t$$\n",
" \n",
"- **hidden state**: $$ H_t = O_t \\odot \\text{tanh}(C_t)$$\n",
"\n",
"\n",
"as you can see, we save 2 diffrent states, `hidden` and `context` (we have the output as well).\n",
"\n",
"the idea really help with vanishing gradients, yet a big set of weights is added to the simple RNN."
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"input shape = torch.Size([5, 3, 10])\n",
"hidden shape = torch.Size([2, 3, 20])\n",
"output shape = torch.Size([5, 3, 20])\n"
]
}
],
"source": [
"lstm = nn.LSTM(10, 20, 2) #LSTM is not an encoder decoder, thus we can project the output size, but default we use the hidden size\n",
"input = torch.randn(5, 3, 10) #5 batch size, seq of 3, input dim 10\n",
"\n",
"h0 = torch.randn(2, 3, 20) #2 for 2 layers of the LSTM\n",
"c0 = torch.randn(2, 3, 20)\n",
"output, (hn, cn) = lstm(input, (h0, c0))\n",
"\n",
"print('input shape =',input.shape)\n",
"print('hidden shape =',h0.shape)\n",
"print('output shape =',output.shape)"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"We now have 8 sets of weights to learn (i.e., the U and W for each of the 4\n",
"gates within each unit), whereas with simple recurrent units we only had 2. Training\n",
"these additional parameters imposes a much significantly higher training cost. \n",
"\n",
"**Gated Recurrent Unit** or **GRU** ease this burden by dispensing with the\n",
"use of a separate context vector, and by reducing the number of gates to 2 — a reset gate, r and an update gate, z.\n",
"\n",
"
"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"- **reset gate**:\n",
"$$ R_t = \\sigma(X_tW_{xr} + H_{t-1}W_{hr} +b_r) \\in \\mathbb{R}^{n \\times h}$$\n",
"- **update gate**:\n",
"$$ Z_t = \\sigma(X_tW_{xz} + H_{t-1}W_{hz} +b_z) \\in \\mathbb{R}^{n \\times h} $$\n",
"\n",
"- **candidate hidden state**: \n",
" $$ \\tilde{H}_{t} = \\text{tanh}\\left(X_t W_{xh} + (R_t \\odot H_{t-1})W_{hh} \\right) + b_h$$\n",
"- **hidden state**:\n",
"$$ H_t = Z_t \\odot H_{t-1} +(1-Z_t) \\odot \\tilde{H}_t. $$"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"source": [
"\n",
"* Whenever the update gate $Z_t$ is close to 1, we simply retain the old state. In this case the information from $X_t$ is essentially ignored, effectively skipping time step $t$ in the dependency chain. \n",
"* In contrast, whenever $Z_t$ is close to 0, the new latent state $H_t$ approaches the candidate latent state $\\tilde{H}_t$.\n",
"* These designs can help us cope with the vanishing gradient problem in RNNs and better capture dependencies for sequences with large time step distances."
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"RNN params: 40\n",
"LSTM params: 160\n",
"GRU params: 120\n"
]
}
],
"source": [
"def num_params(layer):\n",
" return sum([p.numel() for p in layer.parameters()])\n",
"\n",
"rnn = nn.RNN(10, 2, 2)\n",
"lstm = nn.LSTM(10, 2, 2)\n",
"gru = nn.GRU(10,2,2)\n",
"print(f'RNN params: {num_params(rnn)}')\n",
"print(f'LSTM params: {num_params(lstm)}')\n",
"print(f'GRU params: {num_params(gru)}')"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"## Part 1: Sentiment analysis for movie reviews"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"The task: Given a review about a movie written by some user, decide whether it's **positive**, **negative** or **neutral**.\n",
"\n",
"
\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"Classically this is considered a challenging task if approached based on keywords alone.\n",
"\n",
"Consider:\n",
"\n",
" \"This movie was actually neither that funny, nor super witty.\"\n",
" \n",
"To comprehend such a sentence, it's intuitive to see that some \"state\" must be kept when \"reading\" it."
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"### Dataset\n",
"\n",
"We'll use the [`torchtext`](https://github.com/pytorch/text) package, which provides useful tools for working ith textual data, and also includes some built-in datasets and dataloaders (similar to `torchvision`).\n",
"\n",
"Out dataset will be the [Stanford Sentiment Treebank](https://nlp.stanford.edu/sentiment/treebank.html) (SST) dataset, which contains ~10,000 **labeled** movie reviews.\n",
"\n",
"The label of each review is either \"positive\", \"neutral\" or \"negative\".\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"#### Loading and tokenizing text samples\n",
"\n",
"The `torchtext.data.Field` class takes care of splitting text into unique \"tokens\"\n",
"(~words) and converting it a numerical representation as a sequence of numbers representing\n",
"the tokens in the text."
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"scrolled": true,
"slideshow": {
"slide_type": "skip"
}
},
"outputs": [],
"source": [
"#update enviroment locally\n",
"\n",
"#!pip install torchtext==0.6.0\n",
"#!pip install nltk\n",
"\n",
"##If you want to run on apple:\n",
"\n",
"#!pip install -U pip setuptools wheel\n",
"#!pip install -U 'spacy[apple]'\n",
"#!python -m spacy download en_core_web_sm\n",
"\n"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [],
"source": [
"import torchtext.data\n",
"\n",
"# torchtext Field objects parse text (e.g. a review) and create a tensor representation\n",
"\n",
"# This Field object will be used for tokenizing the movie reviews text\n",
"# For this application, tokens ~= words\n",
"\n",
"review_parser = torchtext.data.Field(\n",
" sequential=True, use_vocab=True, lower=True,\n",
" init_token='', eos_token='', dtype=torch.long,\n",
" tokenize='spacy', tokenizer_language='en_core_web_sm'\n",
")\n",
"\n",
"# This Field object converts the text labels into numeric values (0,1,2)\n",
"label_parser = torchtext.data.Field(\n",
" is_target=True, sequential=False, unk_token=None, use_vocab=True\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Number of training samples: 8544\n",
"Number of test samples: 2210\n"
]
}
],
"source": [
"import torchtext.datasets\n",
"\n",
"# Load SST, tokenize the samples and labels\n",
"# ds_X are Dataset objects which will use the parsers to return tensors\n",
"ds_train, ds_valid, ds_test = torchtext.datasets.SST.splits(\n",
" review_parser, label_parser, root=data_dir\n",
")\n",
"\n",
"n_train = len(ds_train)\n",
"print(f'Number of training samples: {n_train}')\n",
"print(f'Number of test samples: {len(ds_test)}')"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"Lets print some examples from our training data:"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"sample#0111 [positive]:\n",
" > the film aims to be funny , uplifting and moving , sometimes all at once .\n",
"\n",
"sample#4321 [neutral ]:\n",
" > the most anti - human big studio picture since 3000 miles to graceland .\n",
"\n",
"sample#7777 [negative]:\n",
" > an ugly , revolting movie .\n",
"\n",
"sample#0000 [positive]:\n",
" > the rock is destined to be the 21st century 's new ` ` conan '' and that he 's going to make a splash even greater than arnold schwarzenegger , jean - claud van damme or steven segal .\n",
"\n"
]
}
],
"source": [
"for i in ([111, 4321, 7777, 0]):\n",
" example = ds_train[i]\n",
" label = example.label\n",
" review = str.join(\" \", example.text)\n",
" print(f'sample#{i:04d} [{label:8s}]:\\n > {review}\\n')"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"#### Building a vocabulary\n",
"\n",
"The `Field` object can build a **vocabulary** for us,\n",
"which is simply a bi-directional mapping between a unique index and a token.\n",
"\n",
"We'll only include words from the training set in our vocabulary."
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Number of tokens in training samples: 15451\n",
"Number of tokens in training labels: 3\n"
]
}
],
"source": [
"review_parser.build_vocab(ds_train)\n",
"label_parser.build_vocab(ds_train)\n",
"\n",
"print(f\"Number of tokens in training samples: {len(review_parser.vocab)}\")\n",
"print(f\"Number of tokens in training labels: {len(label_parser.vocab)}\")"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"first 20 tokens:\n",
" ['', '', '', '', '.', 'the', ',', 'a', 'and', 'of', 'to', '-', 'is', \"'s\", 'it', 'that', 'in', 'as', 'but', 'film']\n",
"\n"
]
}
],
"source": [
"print(f'first 20 tokens:\\n', review_parser.vocab.itos[:20], end='\\n\\n')"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"source": [
"Note the **special tokens**, ``, ``, `` and `` at indexes `0-3`.\n",
"These were automatically created by the tokenizer."
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"word=film index=19\n",
"word=actor index=493\n",
"word=schwarzenegger index=3406\n",
"word=spielberg index=718\n"
]
}
],
"source": [
"# Show that some words exist in the vocab\n",
"for w in ['film', 'actor', 'schwarzenegger', 'spielberg']:\n",
" print(f'word={w:15s} index={review_parser.vocab.stoi[w]}')"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"labels vocab:\n",
" {'positive': 0, 'negative': 1, 'neutral': 2}\n"
]
}
],
"source": [
"print(f'labels vocab:\\n', dict(label_parser.vocab.stoi))"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"hist = [ds_train[i].label for i in range(len(ds_train))]\n",
"plt.hist(hist);"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"#### Data loaders (iterators)\n",
"\n",
"The `torchtext` package comes with `Iterator`s, similar to the `DataLoaders` we previously worked with.\n",
"\n",
"A key issue when working with text sequences is that each sample is of a different length.\n",
"\n",
"So, how can we work with **batches** of data?"
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [],
"source": [
"BATCH_SIZE = 4\n",
"\n",
"# BucketIterator creates batches with samples of similar length\n",
"# to minimize the number of tokens in the batch.\n",
"dl_train, dl_valid, dl_test = torchtext.data.BucketIterator.splits(\n",
" (ds_train, ds_valid, ds_test), batch_size=BATCH_SIZE,\n",
" shuffle=True, device=device)"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"source": [
"Lets look at a single batch."
]
},
{
"cell_type": "code",
"execution_count": 77,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"X = \n",
" tensor([[ 2, 2, 2, 2],\n",
" [ 2550, 5, 14, 19],\n",
" [ 6, 334, 37, 207],\n",
" [ 5, 225, 438, 28],\n",
" [ 412, 12, 1782, 196],\n",
" [ 11, 64, 15, 2757],\n",
" [ 8, 14129, 5, 25],\n",
" [ 11, 607, 1107, 3217],\n",
" [ 638, 4, 13, 2774],\n",
" [ 5835, 3, 105, 6],\n",
" [ 1181, 1, 246, 18],\n",
" [ 9, 1, 40, 500],\n",
" [ 79, 1, 144, 14208],\n",
" [ 591, 1, 516, 77],\n",
" [ 2853, 1, 4, 759],\n",
" [ 294, 1, 3, 4],\n",
" [ 785, 1, 1, 3],\n",
" [ 4229, 1, 1, 1],\n",
" [ 4, 1, 1, 1],\n",
" [ 3, 1, 1, 1]]) torch.Size([20, 4])\n",
"\n",
"y = \n",
" tensor([2, 1, 1, 2]) torch.Size([4])\n"
]
}
],
"source": [
"batch = next(iter(dl_train))\n",
"\n",
"X, y = batch.text, batch.label\n",
"print('X = \\n', X, X.shape, end='\\n\\n')\n",
"print('y = \\n', y, y.shape)"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"What are we looking at?"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"Our sample tensor `X` is of shape `(sentence_length, batch_size)`.\n",
"\n",
"Note that:\n",
"1. `sentence_length` changes every batch! You can re-run the previous block to see this.\n",
"2. Sequence dimension first (not batch). When we implement the model, you'll see why it's easier to work this way."
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"### Model\n",
"\n",
"We'll now create our sentiment analysis model based on the simple `RNNLayer` we've implemented above."
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"The model will:\n",
"- Take an input batch of tokenized sentences.\n",
"- Compute a dense word-embedding of each token.\n",
"- Process the sentence **sequentially** through the RNN layer.\n",
"- Produce a `(B, 3)` tensor, which we'll interpret as class probabilities for each sentence in the batch."
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"What is a **word embedding**? How do we get one?"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"source": [
"Embeddings encode tokens as tensors in a way that maintain some **semantic** meaning for our task.\n",
"\n",
"
"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"### How we embed:\n",
"\n",
"1. Embedding Layer: We train an embeding layer with our desired size, take a long time and require a lot of data, but learn the specific relations in our corpus\n",
"\n",
"\n",
"2. Word2Vec: statistical method for efficiently learning a standalone word embedding from a text corpus. it's based on CBOW (continues bag of words) and C-SKIP-GRAM (continues skip-gram).\n",
"\n",
"\n",
"3. GloVe (Global Vectors for Word Representation):extension to the word2vec method for efficiently learning word vectors. using matrix factorization techniques such as Latent Semantic Analysis (LSA). if Word2Vec use a window to determine the context, GloVe used the statistics of a word to accure in the global text.\n",
"\n",
"\n",
"\n",
"Here we'll train an Embedding together with our model:"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"tensor([0, 4, 1, 0, 4, 1])\n"
]
},
{
"data": {
"text/plain": [
"tensor([[-3.2321e-02, 1.3966e+00, 1.3337e-03, -1.0144e+00, -1.2862e+00,\n",
" -8.0038e-01, -3.7318e-01, 5.3016e-01],\n",
" [ 1.8137e-01, 1.8534e+00, -5.5121e-01, 1.0237e+00, 1.8915e+00,\n",
" -1.6838e+00, 1.0199e+00, 3.8242e-01],\n",
" [ 6.4400e-01, 3.7613e-01, -9.7896e-01, 3.5215e-01, -7.0531e-01,\n",
" -1.2142e+00, 9.5350e-01, -4.0627e-01],\n",
" [-3.2321e-02, 1.3966e+00, 1.3337e-03, -1.0144e+00, -1.2862e+00,\n",
" -8.0038e-01, -3.7318e-01, 5.3016e-01],\n",
" [ 1.8137e-01, 1.8534e+00, -5.5121e-01, 1.0237e+00, 1.8915e+00,\n",
" -1.6838e+00, 1.0199e+00, 3.8242e-01],\n",
" [ 6.4400e-01, 3.7613e-01, -9.7896e-01, 3.5215e-01, -7.0531e-01,\n",
" -1.2142e+00, 9.5350e-01, -4.0627e-01]], grad_fn=)"
]
},
"execution_count": 20,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"embedding_layer = nn.Embedding(num_embeddings=5, embedding_dim=8)\n",
"\n",
"token_idx = torch.randint(low=0, high=5, size=(6,))\n",
"print(token_idx)\n",
"embedding_layer(token_idx)"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"OK, model time:"
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [],
"source": [
"class SentimentRNN(nn.Module):\n",
" def __init__(self, vocab_dim, embedding_dim, h_dim, out_dim):\n",
" super().__init__()\n",
" \n",
" # nn.Embedding converts from token index to dense tensor\n",
" self.embedding = nn.Embedding(vocab_dim, embedding_dim)\n",
" \n",
" # Our own Vanilla RNN layer, without phi_y so it outputs a class score\n",
" self.rnn = RNNLayer(in_dim=embedding_dim, h_dim=h_dim, out_dim=out_dim, phi_y=None)\n",
" \n",
" # To convert class scores to log-probability we'll apply log-softmax\n",
" self.log_softmax = nn.LogSoftmax(dim=1)\n",
" \n",
" def forward(self, X):\n",
" # X shape: (S, B) Note batch dim is not first!\n",
" \n",
" embedded = self.embedding(X) # embedded shape: (S, B, E)\n",
" \n",
" # Loop over (batch of) tokens in the sentence(s)\n",
" ht = None\n",
" for xt in embedded: # xt is (B, E)\n",
" yt, ht = self.rnn(xt, ht) # yt is (B, D_out)\n",
" \n",
" # Class scores to log-probability\n",
" yt_log_proba = self.log_softmax(yt)\n",
" \n",
" return yt_log_proba"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"source": [
"Let's instantiate our model."
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"data": {
"text/plain": [
"SentimentRNN(\n",
" (embedding): Embedding(15451, 100)\n",
" (rnn): RNNLayer(\n",
" (fc_xh): Linear(in_features=100, out_features=128, bias=False)\n",
" (fc_hh): Linear(in_features=128, out_features=128, bias=True)\n",
" (fc_hy): Linear(in_features=128, out_features=3, bias=True)\n",
" )\n",
" (log_softmax): LogSoftmax(dim=1)\n",
")"
]
},
"execution_count": 22,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"INPUT_DIM = len(review_parser.vocab)\n",
"EMBEDDING_DIM = 100\n",
"HIDDEN_DIM = 128\n",
"OUTPUT_DIM = 3\n",
"\n",
"model = SentimentRNN(INPUT_DIM, EMBEDDING_DIM, HIDDEN_DIM, OUTPUT_DIM)\n",
"model"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"Test a manual forward pass:"
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {
"scrolled": true,
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"model(X) = \n",
" tensor([[-0.9014, -1.3483, -1.0957],\n",
" [-1.0641, -1.0480, -1.1897],\n",
" [-0.9029, -1.3476, -1.0944],\n",
" [-0.9016, -1.3481, -1.0956]], grad_fn=) torch.Size([4, 3])\n",
"labels = tensor([0, 0, 0, 2])\n"
]
}
],
"source": [
"print(f'model(X) = \\n', model(X), model(X).shape)\n",
"print(f'labels = ', y)"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"How big is our model?"
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"The RNN model has 1,574,799 trainable weights.\n"
]
}
],
"source": [
"def count_parameters(model):\n",
" return sum(p.numel() for p in model.parameters() if p.requires_grad)\n",
"\n",
"print(f'The RNN model has {count_parameters(model):,} trainable weights.')"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"source": [
"Why so many? We used only one RNN layer.\n",
"\n",
"Where are most of the weights?"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"### Training\n",
"\n",
"Let's complete the example by showing the regular pytorch-style train loop with this model.\n",
"\n",
"We'll run only a few epochs on a small subset just to test that it works."
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [],
"source": [
"def train(model, optimizer, loss_fn, dataloader, max_epochs=100, max_batches=200):\n",
" for epoch_idx in range(max_epochs):\n",
" total_loss, num_correct = 0, 0\n",
" start_time = time.time()\n",
"\n",
" for batch_idx, batch in enumerate(dataloader):\n",
" X, y = batch.text, batch.label\n",
"\n",
" # Forward pass\n",
" y_pred_log_proba = model(X)\n",
"\n",
" # Backward pass\n",
" optimizer.zero_grad()\n",
" loss = loss_fn(y_pred_log_proba, y)\n",
" loss.backward()\n",
"\n",
" # Weight updates\n",
" optimizer.step()\n",
"\n",
" # Calculate accuracy\n",
" total_loss += loss.item()\n",
" y_pred = torch.argmax(y_pred_log_proba, dim=1)\n",
" num_correct += torch.sum(y_pred == y).float().item()\n",
"\n",
" if batch_idx == max_batches-1:\n",
" break\n",
" \n",
" print(f\"Epoch #{epoch_idx}, loss={total_loss /(max_batches):.3f}, accuracy={num_correct /(max_batches*BATCH_SIZE):.3f}, elapsed={time.time()-start_time:.1f} sec\")"
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch #0, loss=1.085, accuracy=0.406, elapsed=0.8 sec\n",
"Epoch #1, loss=1.064, accuracy=0.398, elapsed=0.8 sec\n",
"Epoch #2, loss=1.096, accuracy=0.396, elapsed=0.8 sec\n",
"Epoch #3, loss=1.057, accuracy=0.427, elapsed=0.8 sec\n"
]
}
],
"source": [
"import torch.optim as optim\n",
"\n",
"rnn_model = SentimentRNN(INPUT_DIM, EMBEDDING_DIM, HIDDEN_DIM, OUTPUT_DIM).to(device)\n",
"\n",
"optimizer = optim.Adam(rnn_model.parameters(), lr=1e-3)\n",
"\n",
"# Recall: LogSoftmax + NLL is equiv to CrossEntropy on the class scores\n",
"loss_fn = nn.NLLLoss()\n",
"\n",
"train(rnn_model, optimizer, loss_fn, dl_train, max_epochs=4) # just a demo"
]
},
{
"cell_type": "code",
"execution_count": 27,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
" precision recall f1-score support\n",
"\n",
" 0 0.438 0.273 0.336 3610\n",
" 1 0.387 0.735 0.507 3310\n",
" 2 0.000 0.000 0.000 1624\n",
"\n",
" accuracy 0.400 8544\n",
" macro avg 0.275 0.336 0.281 8544\n",
"weighted avg 0.335 0.400 0.338 8544\n",
"\n"
]
}
],
"source": [
"from sklearn import metrics\n",
"\n",
"def print_stats(model, dataloader):\n",
" model.eval() # put in evaluation mode\n",
" trues = []\n",
" preds = []\n",
" with torch.no_grad():\n",
" for data in dataloader:\n",
" X, y = data.text.to(device), data.label.to(device)\n",
" trues+=list(y.cpu())\n",
" y_pred_log_proba = model(X)\n",
" predicted = torch.argmax(y_pred_log_proba, dim=1)\n",
" preds+= list(predicted.cpu()) \n",
" print(metrics.classification_report(trues, preds, digits=3))\n",
" \n",
"print_stats(rnn_model, dl_train)"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"#### Limitations\n",
"\n",
"As usual this is a very naïve model, just for demonstration.\n",
"It lacks many tricks of the NLP trade, such was pre-trained embeddings,\n",
"gated RNN units, deep or bi-directional models, dropout, etc.\n",
"\n",
"Don't expect SotA results :)"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "slide"
}
},
"source": [
"## Part 2: Time-Series Forecasting: Predicting Stock Prices Using An LSTM Model"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"Now we learned how to work with words and documents, how about making money by forecasting in NASDAQ?\n",
"\n",
"as a disclaimer, i will try to explain, why it's nearly impossible to beat the market, but it's still a great tool to be familiar with, and who knows, maybe you would think out of the box and beat the market, just don't forget to tip your favorit T.A after :)\n",
"\n",
"**this is the oposite of a recomandation to invest with predicting the price of stocks**"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"we will run this section regardless on the tutorial, so we're going to import all libraries again...\n",
"\n",
"note that `alpha_vantage` just help us to get the stocks data from https://www.alphavantage.co/.\n",
"\n",
"if you don't want it as part of your enviroment, just use `conda update` with the json file again...\n",
"\n",
"**this is also not a recommandation to use alpha_vantage API, it's you can use yahoo database from pandas or whatever other API, or even scrape the data yourself**"
]
},
{
"cell_type": "code",
"execution_count": 41,
"metadata": {},
"outputs": [],
"source": [
"! pip install alpha_vantage -q\n",
"\n",
"# pip install numpy\n",
"import numpy as np\n",
"\n",
"# pip install torch\n",
"import torch\n",
"import torch.nn as nn\n",
"import torch.nn.functional as F\n",
"import torch.optim as optim\n",
"from torch.utils.data import Dataset\n",
"from torch.utils.data import DataLoader\n",
"\n",
"# pip install matplotlib\n",
"import matplotlib.pyplot as plt\n",
"from matplotlib.pyplot import figure\n",
"\n",
"# pip install alpha_vantage\n",
"from alpha_vantage.timeseries import TimeSeries \n",
"\n",
"device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"now we need to determine some parameters, and we're going to use a dictionary config for that. \n",
"when usually we deal only with batch size and learning rate, here we devide the keys to some sections (so we actually have a dict of dicts) \n",
"- **alpha_vantage**: what params we want to use (here we use only the closing price and only IBM symbole)\n",
"- **data**: we determine window size, as the input seq len for the model and the split size for train and validation\n",
"- **plots**: just some parameters for the plots\n",
"- **model**: some model params\n",
"- **training**: some parameters for the training process"
]
},
{
"cell_type": "code",
"execution_count": 61,
"metadata": {
"slideshow": {
"slide_type": "fragment"
}
},
"outputs": [],
"source": [
"config = {\n",
" \"alpha_vantage\": {\n",
" \"key\": 'demo', # Claim your free API key here: https://www.alphavantage.co/support/#api-key\n",
" \"symbol\": \"IBM\", #can use APPL or anything you would like\n",
" \"outputsize\": \"full\",\n",
" \"key_adjusted_close\": \"5. adjusted close\",\n",
" },\n",
" \"data\": {\n",
" \"window_size\": 4,\n",
" \"train_split_size\": 0.90,\n",
" }, \n",
" \"plots\": {\n",
" \"show_plots\": True,\n",
" \"xticks_interval\": 90,\n",
" \"color_actual\": \"#001f3f\",\n",
" \"color_train\": \"#3D9970\",\n",
" \"color_val\": \"#0074D9\",\n",
" \"color_pred_train\": \"#3D9970\",\n",
" \"color_pred_val\": \"#0074D9\",\n",
" \"color_pred_test\": \"#FF4136\",\n",
" },\n",
" \"model\": {\n",
" \"input_size\": 1, # since we are only using 1 feature, close price\n",
" \"num_lstm_layers\": 2,\n",
" \"lstm_size\": 32,\n",
" \"dropout\": 0.2,\n",
" },\n",
" \"training\": {\n",
" \"device\": \"cpu\", # \"cuda\" or \"cpu\"\n",
" \"batch_size\": 64,\n",
" \"num_epoch\": 100,\n",
" \"learning_rate\": 0.01,\n",
" \"scheduler_step_size\": 40,\n",
" }\n",
"}"
]
},
{
"cell_type": "code",
"execution_count": 62,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Number data points: 6120 from 1999-11-01 to 2024-02-28\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"def download_data(config, plot=True):\n",
" # get the data from alpha vantage\n",
"\n",
" ts = TimeSeries(key=config[\"alpha_vantage\"][\"key\"])\n",
" data, meta_data = ts.get_daily_adjusted(config[\"alpha_vantage\"][\"symbol\"], outputsize=config[\"alpha_vantage\"][\"outputsize\"])\n",
"\n",
" data_date = [date for date in data.keys()]\n",
" data_date.reverse()\n",
"\n",
" data_close_price = [float(data[date][config[\"alpha_vantage\"][\"key_adjusted_close\"]]) for date in data.keys()]\n",
" data_close_price.reverse()\n",
" data_close_price = np.array(data_close_price)\n",
"\n",
" num_data_points = len(data_date)\n",
" display_date_range = \"from \" + data_date[0] + \" to \" + data_date[num_data_points-1]\n",
" print(\"Number data points:\", num_data_points, display_date_range)\n",
"\n",
" if plot:\n",
" fig = figure(figsize=(25, 5), dpi=80)\n",
" fig.patch.set_facecolor((1.0, 1.0, 1.0))\n",
" plt.plot(data_date, data_close_price, color=config[\"plots\"][\"color_actual\"])\n",
" xticks = [data_date[i] if ((i%config[\"plots\"][\"xticks_interval\"]==0 and (num_data_points-i) > config[\"plots\"][\"xticks_interval\"]) or i==num_data_points-1) else None for i in range(num_data_points)] # make x ticks nice\n",
" x = np.arange(0,len(xticks))\n",
" plt.xticks(x, xticks, rotation='vertical')\n",
" plt.title(\"Daily close price for \" + config[\"alpha_vantage\"][\"symbol\"] + \", \" + display_date_range)\n",
" plt.grid(visible=None, axis='y', linestyle='--')\n",
" plt.show()\n",
"\n",
" return data_date, data_close_price, num_data_points, display_date_range\n",
"\n",
"data_date, data_close_price, num_data_points, display_date_range = download_data(config)"
]
},
{
"cell_type": "code",
"execution_count": 63,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [],
"source": [
"class Normalizer():\n",
" def __init__(self):\n",
" self.mu = None\n",
" self.sd = None\n",
"\n",
" def fit_transform(self, x):\n",
" self.mu = np.mean(x, axis=(0), keepdims=True)\n",
" self.sd = np.std(x, axis=(0), keepdims=True)\n",
" normalized_x = (x - self.mu)/self.sd\n",
" return normalized_x\n",
"\n",
" def inverse_transform(self, x):\n",
" return (x*self.sd) + self.mu\n",
"\n",
"# normalize\n",
"scaler = Normalizer()\n",
"normalized_data_close_price = scaler.fit_transform(data_close_price)"
]
},
{
"cell_type": "code",
"execution_count": 64,
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"outputs": [],
"source": [
"def prepare_data_x(x, window_size):\n",
" # perform windowing\n",
" n_row = x.shape[0] - window_size + 1\n",
" output = np.lib.stride_tricks.as_strided(x, shape=(n_row,window_size), strides=(x.strides[0],x.strides[0]))\n",
" return output[:-1], output[-1]\n",
"\n",
"def prepare_data_y(x, window_size):\n",
" # # perform simple moving average\n",
" # output = np.convolve(x, np.ones(window_size), 'valid') / window_size\n",
"\n",
" #even easier, use the next day as label\n",
" output = x[window_size:]\n",
" return output\n",
"\n",
"def prepare_data(normalized_data_close_price, config, plot=False):\n",
" data_x, data_x_unseen = prepare_data_x(normalized_data_close_price, window_size=config[\"data\"][\"window_size\"])\n",
" data_y = prepare_data_y(normalized_data_close_price, window_size=config[\"data\"][\"window_size\"])\n",
"\n",
" # split dataset\n",
"\n",
" split_index = int(data_y.shape[0]*config[\"data\"][\"train_split_size\"])\n",
" data_x_train = data_x[:split_index]\n",
" data_x_val = data_x[split_index:]\n",
" data_y_train = data_y[:split_index]\n",
" data_y_val = data_y[split_index:]\n",
"\n",
" if plot:\n",
" # prepare data for plotting\n",
"\n",
" to_plot_data_y_train = np.zeros(num_data_points)\n",
" to_plot_data_y_val = np.zeros(num_data_points)\n",
"\n",
" to_plot_data_y_train[config[\"data\"][\"window_size\"]:split_index+config[\"data\"][\"window_size\"]] = scaler.inverse_transform(data_y_train)\n",
" to_plot_data_y_val[split_index+config[\"data\"][\"window_size\"]:] = scaler.inverse_transform(data_y_val)\n",
"\n",
" to_plot_data_y_train = np.where(to_plot_data_y_train == 0, None, to_plot_data_y_train)\n",
" to_plot_data_y_val = np.where(to_plot_data_y_val == 0, None, to_plot_data_y_val)\n",
"\n",
" ## plots\n",
"\n",
" fig = figure(figsize=(25, 5), dpi=80)\n",
" fig.patch.set_facecolor((1.0, 1.0, 1.0))\n",
" plt.plot(data_date, to_plot_data_y_train, label=\"Prices (train)\", color=config[\"plots\"][\"color_train\"])\n",
" plt.plot(data_date, to_plot_data_y_val, label=\"Prices (validation)\", color=config[\"plots\"][\"color_val\"])\n",
" xticks = [data_date[i] if ((i%config[\"plots\"][\"xticks_interval\"]==0 and (num_data_points-i) > config[\"plots\"][\"xticks_interval\"]) or i==num_data_points-1) else None for i in range(num_data_points)] # make x ticks nice\n",
" x = np.arange(0,len(xticks))\n",
" plt.xticks(x, xticks, rotation='vertical')\n",
" plt.title(\"Daily close prices for \" + config[\"alpha_vantage\"][\"symbol\"] + \" - showing training and validation data\")\n",
" plt.grid(b=None, which='major', axis='y', linestyle='--')\n",
" plt.legend()\n",
" plt.show()\n",
"\n",
" return split_index, data_x_train, data_y_train, data_x_val, data_y_val, data_x_unseen\n",
"\n",
"split_index, data_x_train, data_y_train, data_x_val, data_y_val, data_x_unseen = prepare_data(normalized_data_close_price, config)"
]
},
{
"cell_type": "code",
"execution_count": 65,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Train data shape (5504, 4, 1) (5504,)\n",
"Validation data shape (612, 4, 1) (612,)\n"
]
}
],
"source": [
"class TimeSeriesDataset(Dataset):\n",
" def __init__(self, x, y):\n",
" x = np.expand_dims(x, 2) # in our case, we have only 1 feature, so we need to convert `x` into [batch, sequence, features] for the model\n",
" self.x = x.astype(np.float32)\n",
" self.y = y.astype(np.float32)\n",
" \n",
" def __len__(self):\n",
" return len(self.x)\n",
"\n",
" def __getitem__(self, idx):\n",
" return (self.x[idx], self.y[idx])\n",
"\n",
"dataset_train = TimeSeriesDataset(data_x_train, data_y_train)\n",
"dataset_val = TimeSeriesDataset(data_x_val, data_y_val)\n",
"\n",
"print(\"Train data shape\", dataset_train.x.shape, dataset_train.y.shape)\n",
"print(\"Validation data shape\", dataset_val.x.shape, dataset_val.y.shape)"
]
},
{
"cell_type": "code",
"execution_count": 66,
"metadata": {},
"outputs": [],
"source": [
"class GRU_stocks_Model(nn.Module):\n",
" def __init__(self, input_size=1, hidden_layer_size=32, num_layers=2, output_size=1, dropout=0.2):\n",
" super().__init__()\n",
" self.hidden_layer_size = hidden_layer_size\n",
"\n",
" self.linear_1 = nn.Linear(input_size, hidden_layer_size)\n",
" self.relu = nn.ReLU()\n",
" self.gru = nn.GRU(hidden_layer_size, hidden_size=self.hidden_layer_size, num_layers=num_layers, batch_first=True)\n",
" self.dropout = nn.Dropout(dropout)\n",
" self.linear_2 = nn.Linear(num_layers*hidden_layer_size, output_size)\n",
" \n",
" self.init_weights()\n",
"\n",
" def init_weights(self):\n",
" for name, param in self.gru.named_parameters():\n",
" if 'bias' in name:\n",
" nn.init.constant_(param, 0.0)\n",
" elif 'weight_ih' in name:\n",
" nn.init.kaiming_normal_(param)\n",
" elif 'weight_hh' in name:\n",
" nn.init.orthogonal_(param)\n",
"\n",
" def forward(self, x):\n",
" batchsize = x.shape[0]\n",
"\n",
" # layer 1\n",
" x = self.linear_1(x)\n",
" x = self.relu(x)\n",
" \n",
" #gru\n",
" #if you choose to use lstm:\n",
" #lstm_out, (h_n, c_n) = self.lstm(x)\n",
" gru_out, h_n = self.gru(x)\n",
" \n",
" # reshape output from hidden cell into [batch, features] for `linear_2`\n",
" x = h_n.permute(1, 0, 2).reshape(batchsize, -1) \n",
" \n",
" # layer 2\n",
" x = self.dropout(x)\n",
" predictions = self.linear_2(x)\n",
" return predictions[:,-1]\n",
"\n",
"stocks_Model = GRU_stocks_Model(input_size=config[\"model\"][\"input_size\"], hidden_layer_size=config[\"model\"][\"lstm_size\"], num_layers=config[\"model\"][\"num_lstm_layers\"], output_size=1, dropout=config[\"model\"][\"dropout\"])\n",
"stocks_Model = stocks_Model.to(config[\"training\"][\"device\"])"
]
},
{
"cell_type": "code",
"execution_count": 67,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch[1/100] | loss train:0.032057, test:0.023425 | lr:0.010000\n",
"Epoch[11/100] | loss train:0.008802, test:0.022679 | lr:0.010000\n",
"Epoch[21/100] | loss train:0.008724, test:0.009327 | lr:0.010000\n",
"Epoch[31/100] | loss train:0.008208, test:0.005004 | lr:0.010000\n",
"Epoch[41/100] | loss train:0.007271, test:0.003234 | lr:0.001000\n",
"Epoch[51/100] | loss train:0.006513, test:0.003825 | lr:0.001000\n",
"Epoch[61/100] | loss train:0.006666, test:0.003743 | lr:0.001000\n",
"Epoch[71/100] | loss train:0.006652, test:0.004491 | lr:0.001000\n",
"Epoch[81/100] | loss train:0.006306, test:0.006072 | lr:0.000100\n",
"Epoch[91/100] | loss train:0.006361, test:0.005072 | lr:0.000100\n"
]
}
],
"source": [
"def run_epoch(dataloader, is_training=False):\n",
" epoch_loss = 0\n",
"\n",
" if is_training:\n",
" stocks_Model.train()\n",
" else:\n",
" stocks_Model.eval()\n",
"\n",
" for idx, (x, y) in enumerate(dataloader):\n",
" if is_training:\n",
" optimizer.zero_grad()\n",
"\n",
" batchsize = x.shape[0]\n",
"\n",
" x = x.to(config[\"training\"][\"device\"])\n",
" y = y.to(config[\"training\"][\"device\"])\n",
"\n",
" out = stocks_Model(x)\n",
" loss = criterion(out.contiguous(), y.contiguous())\n",
"\n",
" if is_training:\n",
" loss.backward()\n",
" optimizer.step()\n",
"\n",
" epoch_loss += (loss.detach().item() / batchsize)\n",
"\n",
" lr = scheduler.get_last_lr()[0]\n",
"\n",
" return epoch_loss, lr\n",
"\n",
"# create `DataLoader`\n",
"train_dataloader = DataLoader(dataset_train, batch_size=config[\"training\"][\"batch_size\"], shuffle=True)\n",
"val_dataloader = DataLoader(dataset_val, batch_size=config[\"training\"][\"batch_size\"], shuffle=True)\n",
"\n",
"# define optimizer, scheduler and loss function\n",
"criterion = nn.MSELoss()\n",
"optimizer = optim.Adam(stocks_Model.parameters(), lr=config[\"training\"][\"learning_rate\"], betas=(0.9, 0.98), eps=1e-9)\n",
"scheduler = optim.lr_scheduler.StepLR(optimizer, step_size=config[\"training\"][\"scheduler_step_size\"], gamma=0.1)\n",
"\n",
"# begin training\n",
"for epoch in range(config[\"training\"][\"num_epoch\"]):\n",
" loss_train, lr_train = run_epoch(train_dataloader, is_training=True)\n",
" loss_val, lr_val = run_epoch(val_dataloader)\n",
" scheduler.step()\n",
" if epoch%10 ==0:\n",
" print('Epoch[{}/{}] | loss train:{:.6f}, test:{:.6f} | lr:{:.6f}'\n",
" .format(epoch+1, config[\"training\"][\"num_epoch\"], loss_train, loss_val, lr_train))\n"
]
},
{
"cell_type": "code",
"execution_count": 70,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAABlAAAAHnCAYAAADHHBCXAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/OQEPoAAAACXBIWXMAAAxOAAAMTgF/d4wjAAEAAElEQVR4nOzdd3gUVdvH8e+m9w4BEpJQQ5cOQmgCUsQGFlSaYHkUXhTsWAAboCgiiqI8gFgQpQjyiGAhKB2ll1ADAUJNgATSk3n/iLtks5tKIIC/z3XlcjNzzpl72gbnnnOOyTAMAxEREREREREREREREbFwKO8ARERERERERERERERErjVKoIiIiIiIiIiIiIiIiOSjBIqIiIiIiIiIiIiIiEg+SqCIiIiIiIiIiIiIiIjkowSKiIiIiIiIiIiIiIhIPkqgiIiIiIiIiIiIiIiI5KMEioiIiIiIiIiIiIiISD5KoIiIiMh15dSpU7z33nt0796diIgIPD09cXd3JyQkhO7duzNu3DgOHz5c3mGKFCkiIgKTycSYMWNs1g0aNAiTyUTHjh2velxXm8lkwmQyMWvWrPIOpUizZs2yxCtyLYuOjrZcq4cOHSrvcIo0ZswYTCYTERER5R2KiIiIiBUlUEREROS6kJOTw9ixY6levTrPPvssy5Yt4/Dhw6SkpJCWlkZ8fDzLli1j1KhRVK9enYcffpikpKTyDlvkmqKHlFJS19uD+CupsKSniIiIiNyYlEARERGRa156ejp33XUXY8aM4eLFi4SEhPDWW2+xfv164uPjOXXqFNu2bWPq1Kl06NCBnJwcZs2axcGDB8s7dBERERERERG5TimBIiIiIte8ESNG8OOPPwLw4IMPsm/fPkaNGkXLli2pXLkyFSpUoGHDhjzxxBNER0fzyy+/EBoaWs5Ri5TerFmzMAyD6Ojo8g5F8hg0aBCGYWAYRnmHInJDGTNmDIZh/Ot7OYmIiMi1x6m8AxAREREpzJ9//sknn3wCQOfOnfnqq6+KnH+gS5cubNq06WqEJyIiIiIiIiI3KPVAERERkWvahAkTgNyJpj/55JNiT95coUIFKlSoYHfd2rVr6d+/PxEREbi5ueHn50fTpk0ZPXo0iYmJBbaZf2LvTZs28eCDDxIaGoq7uzu1atVi1KhRnD9/3lInPT2dDz/8kObNm+Pr64u3tzcdOnRg6dKlxd7O6tWr6dOnD1WqVMHNzY2IiAiefPJJjh49WmAbWVlZREdH88wzz9C8eXP8/f1xdnYmMDCQqKgo3nvvPS5evFhg/fxzZezevZvHHnuM6tWr4+bmZvc8HDhwgKeffpoGDRrg4+ODu7s7NWvW5LHHHmPPnj0Fbqso+edgSExM5MUXXyQyMhIPDw8CAwPp3r07P/30U4Ft5J/8++jRo4wYMYLIyEg8PT3tzu9w4sQJXn75ZZo1a4a/vz+urq6EhYXRr18/NmzYUGTcMTExDBgwgJCQEFxdXalatSoDBw5kx44dRdYt7iTya9asYciQIdSuXRtvb2+8vLyoU6cOffr0Yc6cOaSkpACXjuHYsWMBOHz4sOV4mH8Kmhflcs/r+vXr6dOnDxUrVsTNzY3q1aszdOhQ4uLiiqxblPznNS4ujqFDh1qu04oVK9KnTx/Wrl1bYBsludaLO4n84cOHef7552nWrBkBAQG4ublRrVo1OnfuzOTJkzlx4kSBdaOjo+nfvz/VqlXD3d0dHx8fGjduzGuvvcbZs2cLrGcYBt999x29evUiJCQEFxcXvL29qV69Oh06dGDMmDHs3Lmz0LjzM5lMdOrUyfJ7tWrVbK4bez0Gdu3axeOPP06tWrXw8PDA29ub+vXr88wzzxT6vVUcycnJzJ8/n0GDBlG3bl08PT1xcXGhcuXK9OrVi++++65YPYQyMjL47LPP6NmzJ1WqVMHV1ZWKFSvStGlTRowYwbp16yxlzffj4cOHARg7dqzNccg7L0px79+OHTtiMpkYNGiQzTrDMNiwYQOvvPIKbdq0ITAwEGdnZ/z9/WnRogWjR48mISGhWMesNPLfFzt27GDgwIFUrVoVV1dXQkJCGDBgALt37y6wjfzHYf369Tz00EOEhYXh4uJi9Z1T3PmZdu7cybBhw2jQoAG+vr54eHhQs2ZNbrvtNqZPn865c+cKrLto0SLuueceqlatipubG/7+/rRu3Zp3333X8l1pT1ZWFp9//jldunQhODgYZ2dn/Pz8qFmzJl27dmXChAmWa0NERERuQIaIiIjINerChQuGk5OTARidOnW67PZycnKMZ5991gAK/AkMDDRWr15tt/7AgQMNwOjQoYPx5ZdfGs7OznbbaNasmZGUlGQkJCQYbdu2tVvGZDIZX3zxRZHb+eyzzwxHR0e7bfj4+BQY6wcffFDofgJGZGSkcejQIbv1R48ebQBGeHi48cMPPxju7u429fOaPHlygccDMJycnIwZM2aU4GxdsmLFCks7v//+uxEWFlbgdp599lm7bcycOdNSZt26dUZAQIBN3djYWEv57777zvD09Cz0+L3++usFxvzDDz8Yrq6uduu5u7sbS5YsMcLDww3AGD16tE39vNeAPampqZYyhf0sXLjQ5hgW9BMeHm6zncs9rx999JHh4OBgt25AQICxceNGy+8zZ84ssJ2C5D2vGzdutHteAcPBwcH48MMP7bZRkms97/YKUtQxA4ynnnrKpl5aWprRv3//QusFBwcbGzdutKmbnZ1t3H333UWe48cff7xEx7eo9vLfN4aR+91T0HeW+fpfsGBBieLI66677ioypjvuuMNIT08vsI3t27cbNWrUKLQNX19fS/ni3Gt57+Oi7l+zDh06GIAxcOBAm3U//PBDkdusVKmSsXnzZrtt573n85+j4sh7XyxZssTufQEYLi4uxvz58+22kfc4TJ061ea6yPudk3d79mRnZxsvvviiYTKZCj0mkyZNsql77tw549Zbby20Xq1atYwDBw7Y1L1w4YLRpk2bIs/FuHHjSnyMRURE5PqgHigiIiJyzVq7di1ZWVkAtG/f/rLbe+edd5g4cSIArVq1YtmyZZw6dYrY2FgmTZqEj48PCQkJ9OjRg9jY2ALb2bdvH4888ghRUVH8/vvvnD59mgMHDvDSSy8B8Pfff/Puu+/yyCOPsHXrVt599132799PQkICv/76K3Xq1MEwDP7v//6v0B4v+/fvZ9iwYTRq1IilS5dy6tQpDh48yMSJE/H09CQpKYnbb7+dkydP2tR1dXW1vJG7bt06Dh06xOnTp9m6dSvvvfceoaGh7Nmzh759+xZ6zM6ePWvprfP9998THx/PiRMnWLhwoaXMp59+ylNPPUVmZiZdunRhyZIlxMfHc+bMGf744w969epFVlYWjzzyCL/++muh2yvKkCFDSExMZOLEiRw8eJBTp07x008/0bhxYwAmTpzI559/Xmgbffr0wcPDgxkzZhAXF8epU6f45Zdf8Pf3B+Cnn37i/vvv5+LFizRv3pzvvvuOuLg4EhMT2bhxIwMHDgTgtddeY+bMmTbtx8TEcP/995Oenk5gYCCfffYZR44c4cSJE3z33XdUqlSJ/v37W/VUKqn+/fvzxRdfANCpUycWL17M0aNHSUhIYPv27XzyySe0a9fO0lOiXbt2JCcnW67RsLAwkpOTrX527dpltY3LPa+///47//d//0dOTg7h4eF8++23HD9+nKNHjzJ9+nQcHR257777Sn0M8rvvvvtwcnJi+vTpluM9d+5cqlWrRk5ODsOHD+eXX34psH5xrvWiTJkyxXLMatSowfTp09m/fz+JiYkcOHCAuXPnct999+Hs7GxTd+DAgXz55Zc4ODjw+OOPs2bNGk6fPs3x48eZN28e9erV4+TJk9x22202PVhmzpxpifPee+8lOjqaI0eOcOrUKTZt2sT333/P/fffj4eHR7H3BXJ7e+Tt2bVz506b6yY8PNyy/rvvvuPpp58mOzubyMhI5s+fz4kTJzh69CgzZsygUqVKpKamct9997F+/foSxWIWEBDAwIEDmTdvHps2bSI+Pp74+HjWrVvHyJEjcXd3Z/HixYwePdpu/UOHDtGxY0cOHDiAq6srzz77LBs3buTMmTMcP36cFStW8NJLL1GlShVLnWnTppGcnExYWBgAL730ks1xGDVqVKn2pyCOjo506tSJDz/8kFWrVnHgwAHOnDnDzp07+eyzz6hTpw4nTpygd+/epKWllem28zp//jz9+/encuXKfPfdd5w4cYIjR47w+eefU6FCBTIyMnjggQcK7d0UExPD8OHDadmyJUuXLuXkyZMcPXqUjz/+uNhxPPfcc4wfPx7DMGjSpAlz5szh8OHDJCYmEhMTw6xZs+jZsycODtaPOLKysujVqxfLly/H1dWVF154gb///puEhASOHDnCrFmzCA0NZd++ffTq1cumJ8r48eNZs2YNAE888QTr1q0jPj6e48ePs3HjRmbPns1tt92Gi4tLCY6qiIiIXFfKO4MjIiIiUpDPPvvM8nbnt99+e1ltnTx50tIj4OabbzZSU1NtyqxevdrS46VPnz426/O+hdyjRw8jKyvLpsxDDz1kQO6b+U5OTsaqVatsyuzZs8fyFu20adMK3U6DBg2M5ORkmzK//PKLpY3//Oc/xT0MFvHx8Ya/v78Bub068jO/DQwYtWvXNs6dO2e3nePHjxtubm4GYDz55JMFbu/BBx80AKNhw4YljjXvm9Qmk8lYsWKFTZmkpCSjXr16BmD4+/sbKSkpVuvz9hwIDAw04uLi7G4rNTXVCA4ONgDj9ttvNzIzM+2We+mllwzAqFChgs21dPvttxuA4erqamzdutWm7rFjxyzbgJL3QJk7d65Vj4KcnBy7MRqGYRN/UW95m5XFeW3YsKHlGB09etRm/Y4dOyzbgMvvgeLh4WHs2rXLpkze412nTh2b9cW91vNvL78jR44YLi4uBmC0aNHCOH/+fIHt5D8vCxcutLQ7d+5cu3XOnz9vREZGGoAxbNgwq3W9e/c2AKNp06aFXg+lUdyeDOnp6ZbjXKNGDSMhIcGmzL59+wwfHx8DcnvqXQlLly41AMPLy8tISkqyWd+jRw8DMJydnY3o6OgC27F37xfWayyvsuiBUpTk5GRLLxp7vcDKqgcKYFSuXNk4fvy4TZm893D37t1t1uf9WxYVFVVor6DCvpvWr19vaadXr16FtpP/vE2aNMlyvv/880+7dY4cOWJUqFDBAIyJEydarWvatKkBGHfffXeB2xQREZEbm3qgiIiIyDUrb+8MPz+/y2rryy+/JD09HYAPPvgANzc3mzJt2rSxjEX/ww8/cPr06QLb++CDD3B0dLRZbu7RkZWVxf3330/btm1tytSuXZumTZsCFPkW9oQJE/Dy8rJZ3qVLF+6++24AvvrqKzIyMgptJ7/KlSvTpUsXgELfygd4/fXX8fX1tbvu008/JS0tjcDAQN57770C2xg3bhwA27dvZ9u2bSWKNa977rnH7rwC3t7ejB8/HsjtSVBYr4HnnnuOqlWr2l337bffcvLkSZycnJg2bRpOTk52y7366qt4enpy+vRpli9fbll+6tQp/ve//wHw2GOP0ahRI5u6VapU4eWXXy4wvqJ88MEHAFStWpUPP/yw0Pk4Coq/KJd7Xv/66y+2b98OwIsvvkhISIhN3fr16/Pkk0+WKj57hg4dSt26dW2W5z3eMTExlrfJ7SnsWi/K1KlTycjIwGQyMWvWLHx8fAosm/+8mM9pz549C+yV4+PjY+nl8M0331jN82HuqVepUqVizxNV1n788UdLb7i3336bgIAAmzI1a9bk+eefB3J76m3evLnM4+jevTsVKlTgwoULNnPf7N271zL/1IgRI+jQoUOB7ZT23rlavLy86N27N1D0d/jleuWVV6hUqZLN8rz38LJlywqd32bixIml7qVhvj88PDyYNWtWoe3kP2+TJ08G4NFHHyUqKspundDQUIYNGwbA119/bbXOfG9Vrly5VLGLiIjI9U8JFBEREflX+PPPP4HcoYtatmxZYLn7778fgOzs7AInnq5evTq1a9e2u65GjRqWz926dStwO+Zyx48fL7CMp6dnoW306dMHgAsXLrBlyxab9SkpKXz88cd069bNMgF93kmPv//+e4BCJwI3mUz06NGjwPXmB3cdOnQgKyuLCxcu2P0JCAggKCgIgI0bNxbYXlHMDwzt6dGjh2WIolWrVhVY7rbbbitwnXl/brrpJry9vQvcn+zsbOrUqQNY78+aNWvIyckpMlbzuSup5ORkywT2DzzwwBUbNuZyz2ve438ljoM9xd1OQddGUdd6UczDmLVo0YJ69eoVu15KSoolqdO1a9cCj/WFCxcs7SYmJnLw4EFLG02aNAHg559/ZtKkSVy4cKHU+1Fa5u9YZ2dn7rjjjgLLmb9j89YpqaNHj/Lqq69aTa6e97vNnPzO/92Wd6g5exO3X2uysrL44osvuOOOOwgLC8PDw8NqP999912g8O/wslCce8swjAKTk4GBgbRq1arU2//tt98A6NWrF4GBgcWut2/fPg4dOgRA586dC723GjRoAMDWrVutXkgw31szZ860ehFDRERE/j2u7ddqRERE5F8t7xvM586du6y2Dh8+DFDkg8369etbPpsfvOSXd2z8/Nzd3UtULjU1tcAytWrVstvLxSzvvhw6dMgqMbRv3z66detW6FwuZoXNxREUFFTom/TmB3cLFixgwYIFRW4LKLRnT1Hs9TAwc3JyolatWmzdurXAcwe5CbCCmPfn77//xtvbu1gx5d2fvNstLNYqVarg6+tb4nlQDh06RHZ2NoBl3pcr4XLPq/k4uLm5ERERUWCdkiQailLc413QtVHUtV6UAwcOACU/LwcPHiQzMxPI7RUxYsSIYtU7ffq0JRE7YsQIZs2axeHDhxk5ciQvvfQSrVu3pnXr1nTo0IFOnTrZ7XVXlszfsTVq1Ch0WzVr1sTV1ZX09PRC79OCLFq0iH79+hUrSZT//jKfI1dXV0sC9Fp1+vRpunfvzqZNm4oseznzKRXFz8/Pbu8Ts/x/h+wp7Du3KMnJyZw6dQoo+b2VN7FU3GRtTk4OiYmJln0eM2YMixYt4ty5cwwYMIAnnniCtm3b0rp1azp27Ei7du2u+d5KIiIicnnUA0VERESuWdWqVbN83r1792W1lZycDGB3OKy88j40N9fJr7CkRknL5R2GJ7+iYs27Pm+s2dnZ9O7dm9jYWDw9PXn55ZdZuXKlZSJ086THDzzwAHBpiBJ7ipp0ujQP7i5nwuPiHpOCzh0Uvk+Xuz95H+qW5PwVV1JSkuVzcRM8pVFWx+FKHIPStlXUtVHSCdbzM5+bkp6X0j78znu8fX192bhxI0899RSBgYGkp6ezcuVKJkyYQM+ePalUqRKvvPLKFX17vrjfsXnLFHaf2nPo0CH69u3LhQsXiIiI4MMPP+Svv/7i+PHjnD9/3vLdFhoaCth+t5nPkZeXV7kNdVZcAwYMYNOmTTg5OTF8+HB++eUXYmNjSUhIsOzniy++CBT+HX65Svt3KK/Lubcu5zuvLO6tiIgINm/ezMMPP4y3tzcXL15k+fLlvP7669xyyy2EhoYyadIkS89DERERufHoVQkRERG5Zt188804OTmRlZXFH3/8cVltmR+8FPXWct71V/IBdXGUNtaVK1eyY8cOAObNm0f37t3t1r948eJlx+jl5cW5c+d46qmnLOPUX0nFPSalPXfmh4F33nknP/zwQ6nrm2Px9PQssGxphlnK20OipA+fS+Jyz6v5OJTkGr5cFy5cKHT+ksu9Nori4+NjSVCWRN5rZuHChdx1112l2n6FChX44IMPeP/999m+fTvr1q0jOjqapUuXcv78ed566y02b95smaOnrBX3OxYuffeU9FzMmDGDtLQ0fHx8WLduHcHBwXbL5X3onpf5/rlw4QKGYVyxJEpx2y0o8XHw4EF+/vlnAKZMmcJ//vMfu+VSUlJKF2AJlPffzMv5zst7b23evLnUvfYiIiKYMWMG06ZNY9OmTZZ7a9myZZw8eZKRI0eyf/9+Pv7441K1LyIiItc29UARERGRa1beOUCio6PZt29fqdsyDyO0c+fOQsuZEw9565SXffv2WYZrsmfXrl2Wz3ljNc+H4u/vX2DyBLBM8n05zEMIFWeYmbJQWE+krKwsyzVS2nNn3p/STm6dd7uFxRofH1+qt6MjIiIsPZvszXtTVi73vJqPQ1paWqHDNOW9hi9XcY/3lbqva9asCZT8vERERODgkPu/ZWVxHzk4OHDTTTfx+OOPM2fOHI4dO2aZw+Knn35i/fr1l70Ne8zH9cCBA4X2Mtu/f79lfUnPhfnY3nLLLQUmT+Li4gpMoJjPUXp6OjExMSXadkkUZ4hGgGPHjtldnvcaMvcUtKcsvsOLcu7cOU6cOFHg+oL+DpUVb29vKlasCJT83so7J1lZ3FvOzs60atWKp556ioULFxIXF0ebNm0A+OSTTwqd00xERESuX0qgiIiIyDXt+eefB3KHunryyScLHfIqr9OnT1vNydCuXTsAjhw5YpmE2x7zxOqOjo6WByPl5eLFiyxbtqzA9fPnzwdy37LN+2ateZiewpIvq1evLtb8KEW59dZbgdzJ06/0RMZAofNxLF261PJGdlRUVKnaN+9PXFyc1YTTxdWmTRvLw/DCYjWfu5Ly9vamdevWAHz77bdWkx0Xh7OzM1D4tQGXf17zHv8rcRzsKe52SnttFKVr164AbNy4sURDDvr6+lom2P7qq68s86GUFU9PT1555RXL7yUdDtF8zUDh1435OzYzM5PFixcXWM78HZu3TnEV57vtyy+/LHBdly5dLJ+/+OKLEm0bin//VK5cGchNghc0tFNMTEyBycW8Q60VtK24uLjL7plZXMW5t0wm0xX7m2m+t/73v/+RmJhY7Hr169e3zEU2c+bMMo8rKCiIZ599Fsj9N8rV+BsoIiIiV58SKCIiInJNa9++PY899hgAv/76K/369Svyrd5ff/2VZs2aWb3d269fP1xdXQF4+umn7c4FsH79embMmAHA3XffTVBQUFntRqm9+OKLdofa+u2331i4cCGQu28uLi6WdeYJe5OSklixYoVN3aSkJJ588skyiW/o0KG4ubmRnZ1N//79C3zz2+xy3/qeP38+0dHRNssvXLhgmQ/A39+fu+++u1Tt9+vXz/Jm+2OPPUZ8fHyh5Q8dOmR1LVWsWJHbbrsNgM8++4xt27bZ1Dl+/DhvvfVWqeKD3OsXch+gFjXheP4hgszX9OnTpwudN+Fyz2vz5s1p2LAhAOPHj7f7pv3OnTv55JNPCm23JD7++GO7yYG8x7tOnTpX7CHvE088gYuLC4Zh8PDDDxc63FD+Y//MM88AEBsby/Dhw4ucTyH/g9qikiL79++3fA4MDCy0bH55vwcLux969epluXdGjRrF2bNnbcocPHiQCRMmANCsWTOaNGlSoljM322rV68mISHBZv2OHTsYP358gfVr1apluT8nTZrEqlWrCixr7/4wH4uivhfMCbGzZ8/aHQowMzOT4cOHF1g/76TrixYtslv/0UcfLTKRU1befPNNu71Q8t7D3bp1s8w9U9aeeuopIPelgsGDBxeaZMx73kwmEyNHjgRg1apVjBs3rtDtZGdnW90rcGXvLREREbk+KIEiIiIi17zJkyfTs2dPAL755htq167NuHHj2LhxIydOnODMmTNs376dTz/9lI4dO9K1a1eOHDli1UbFihUZO3YsAGvXrqVjx4788ssvnDlzhsOHD/Phhx/SrVs3srKy8PHx4d13373q+5lfSEgIMTExtG/fnp9//tkS6/vvv89dd92FYRgEBAQwZswYq3rdu3e3zAXxwAMPMHv2bOLi4jhx4gTz5s2jdevW7Nixg8jIyDKJccqUKUDum/c33XQTH3/8MTExMZw7d46TJ0+yceNGPvroIzp16kSLFi0ua3thYWHcfvvtvP/++xw+fJgzZ87w888/0759e8tQMhMmTLAMoVNSHh4ezJo1C0dHR2JjY2ncuDETJkxg27ZtnD17llOnTrFlyxamT5/O7bffTs2aNW0elL/zzju4urqSnp5O586d+fzzzzl27BgnT55k3rx5REVFkZ6ejp+fX6livOeee7jnnnsAmDp1Kl26dGHJkiXEx8dz9uxZdu3axfTp0+nYsaPNfBfNmzcHct9wf/PNNzl16hRZWVlkZWVZPYwti/P6wQcfYDKZOH36NFFRUXz33XecPHmS+Ph4ZsyYQadOnahUqVKpjoE9FStWpGPHjsyYMYP4+HhOnjzJ999/T1RUFCdPngTgww8/LLPt5RcSEsJ7770H5CZjmzVrxsyZM4mNjeXcuXPExsayYMECHnroIUaNGmVVt0+fPjz00EMAfPrpp7Rv357vv/+ew4cPc+7cOY4ePcqKFSsYO3YsdevWtSRczHr27EmLFi145513WL16NcePHycxMZGYmBgmT57Mo48+CkBwcDCdO3cu0X7VrFnTcq2+++67HDx4kIyMDMt1Y+bi4sLkyZOB3GG82rRpw8KFCy3n/IsvviAqKorz58/j5ORUqvki7r//fgASExPp1q0bv/zyC6dOnSI2NpbJkyfTvn17vLy8CAgIKLCNjz/+mICAADIyMujatSsvvPACf//9N4mJiZw6dYpVq1bx6quv2p0vw3z//PDDD0RHR3Px4kXLccib9OrcuTPVqlUDYMiQIcyaNYsTJ05w6tQpli5dSqdOnfjzzz8JCQmxG2Pz5s0tSZSnnnqKDz/8kAMHDnD69Gl+/vlnOnTowPLly6lXr16Jj2FJ+fn5kZqaSlRUFPPmzePkyZMcO3aM//73v3Tq1InU1FRcXFyYOHHiFYuhRYsWlmt+0aJFtGnThu+++44jR45w7tw59u3bx9dff80dd9zB1KlTreoOHz6cjh07ArmJvdtvv50lS5Zw7Ngxzp07x+HDh1m2bBkvvPAC1atXt5nzqX79+nTq1IkpU6awceNGTp48SUJCAtu3b+f111+39O5q2LChJWksIiIiNxhDRERE5DqQlZVlvPrqq4aHh4cBFPrj6OhoPPbYY0ZSUpJVGzk5Ocazzz5baN3AwEBj9erVdmMYOHCgARgdOnQoMM7Y2FhLWytWrCiwXGFt5V336aefGg4ODnZj9fHxKTDWOXPmGI6OjnbrOTg4GB9++GGhMYwePdoAjPDw8AL3Ia///ve/hru7e5HnJiAgoFjt5bVixQpL/d9//92oWrVqge0/88wzdtuYOXOmpUxxLFmyxAgICCjWtZaYmGhTf+HChYarq6vdOm5ubsaPP/5ohIeHG4AxevRom/pFXWupqanGQw89VGR8CxcutKnbsWNHu2XtnevLPa8fffRRgdevv7+/sWHDBsvvM2fOLOSM2Jf3vG7YsMHw9/cv9Jq3pyTXenGuo/fff99wcnIq9Hg99dRTNvUyMjKMYcOGGSaTqcjj3bt3b6u65mupqO+2VatWFbmP9piPkb2f2NhYq7KTJk0q8LsHMNzd3Y0FCxaUKg7DMIwnn3yywLb9/f2NP//8s9B7yzAMY9u2bUZEREShx8vX19em3q5duww3Nze75fNva+XKlQXeO25ubsb3339vdOjQwQCMgQMH2myrsPqA8eyzzxZ67eb93sx/joojb9s//vhjgfvt4uJizJ8/324bxfmbaW979mRnZxvPPvtskffHpEmTbOomJSUZ99xzT5H3CGCMHDnSqm5x6oSHhxu7d+8uch9FRETk+qQeKCIiInJdcHR05PXXX+fgwYO8++67dO3alapVq+Lu7o6rqytVqlTh1ltvZdy4cRw6dIhp06bh7e1t1YbJZOLdd99lzZo1PPTQQ4SFheHq6oqPjw9NmjTh1VdfZe/eveU+90lejz/+OCtWrOCuu+6iUqVKuLi4EB4ezn/+8x927txZYKx9+/Zl5cqV9OrVC39/f1xcXAgNDeW+++7jjz/+4P/+7//KNM7BgwcTGxvLmDFjuPnmmwkMDMTR0RFPT08iIyN58MEHmTt3LocPH76s7VSrVo1NmzbxzDPPUKtWLdzc3PD39+fWW2/lf//7X5m9BX3bbbdZrrVOnTpRoUIFnJyc8PDwoEaNGtx9993MnDmTkydP4u/vb1P/rrvuYsuWLfTv35/KlSvj4uJCSEgIDz30EOvXr6dXr16XFZ+bmxtfffUV0dHR9OvXj4iICNzc3PD29iYyMpJ77rmHb7/9lm7dutnUXbx4MaNGjaJhw4Z4enpiMpkK3M7lntehQ4eyZs0a7r77bipUqICrqyvh4eE8/vjjbNq06bJ7JOXVokULNm3axOOPP054eDiurq5UqFCBu+++m1WrVpX5NV+QESNGEBMTw9NPP039+vXx9vbGzc2NatWq0aVLF6ZMmWLTAwVy59eYMmUKW7ZsYejQoTRo0AAfHx8cHR3x8/OjSZMmDBs2jN9++425c+da1V22bBlTpkyhd+/e1KtXj4CAAJycnPD39+fmm2/mjTfeYM+ePbRt27ZU+zR69GimTZtG27Zt8fPzs8zzY8/TTz/Ntm3bePTRR6lRowbu7u54enpSr149RowYwd69e0s9xB7k9iCZNWsWrVu3xtPTE3d3d2rWrMn//d//sXnz5mLNcdOwYUN2797NlClTuOWWWwgKCsLZ2Zng4GCaNm3KyJEj+eWXX2zq1a1bl9WrV3PfffcRGhpqNT9Mfu3bt2fDhg307duX4OBgnJ2dCQkJoV+/fmzcuNHSi6w49StWrIizszOVKlWiV69eLFmy5Kr2kuzVqxfr16/noYceIiQkBBcXFypXrky/fv3YsmULvXv3vuIxODg48O6777Jp0yYeffRRatWqhYeHB56entSsWZPbbruNGTNm8PDDD9vU9fb25vvvv+fPP/9k8ODB1K5dGy8vL5ycnAgMDKRVq1Y8//zzrF271uZvyKZNm3j33Xfp2bMnkZGR+Pr64uTkRIUKFejYsSOTJk1i586d1KlT54ofAxERESkfJsMo5kysIiIiInJVDBo0iC+++IIOHTrYne/j3yY6OppOnToBuXNERERElG9Acs2YNWuW5YGp/rdGpOyMGTOGsWPHEh4eXuBk9yIiIiL/BuqBIiIiIiIiIiIiIiIiko8SKCIiIiIiIiIiIiIiIvkogSIiIiIiIiIiIiIiIpKPEigiIiIiIiIiIiIiIiL5KIEiIiIiIiIiIiIiIiKSj8kwDKO8g7gaXF1dqVChQnmHISIiIiIiIiIiIiIi14DTp0+Tnp5e4HqnqxhLuapQoQJHjx4t7zBEREREREREREREROQaEBoaWuh6DeElIiIiIiIiIiIiIiKSjxIoIiIiIiIiIiIiIiIi+SiBIiIiIiIiIiIiIiIiko8SKCIiIiIiIiIiIiIiIvkogSIiIiIiIiIiIiIiIpKPEigiIiIiIiIiIiIiIiL5KIEiIiIiIiIiIiIiIiKSjxIoIiIiIiIiIiIiIiIi+TiVdwDXA8MwLD8iIvLvYTKZLD8iIiIiIiIiIvLvogRKAQzD4Pz58yQlJXHx4sXyDkdERMqRp6cnPj4++Pr6KpkiIiIiIiIiIvIvoQSKHYZhcPz4cVJSUggICCA4OBgnJx0qEZF/o6ysLC5evMiZM2dISUmhcuXKSqKIiIiIiIiIiPwLKCtgx/nz50lJSSEiIkKJExGRfzlHR0dcXV3x8fHh0KFDnD9/Hj8/v/IOS0RERERERETEyrbdBxjxxsfM+2Qs/r7e5R3ODUGTyNuRlJREQECAkiciImLh5OSEv78/SUlJ5R2KiIiIiIiIiIiNe4eO4fc1m/h49sLyDuWGoQRKPoZhcPHiRTw9Pcs7FBERucZ4eXlx8eJFDMMo71BERERERERERKy4u7oAkJqWAcBzb39Cv6ffLM+QrntKoORjfiim3iciIpKf+W+DEigiIiIiIiIicq1xd3MFICU1DYCJn83l6x9+Lc+QrntKoOSjh2IiIlIU/a0QERERERERkWuNOYGybvMuPvvmx3KO5sagbhYiIiIiIiIiIiIiItc5D/dLCZR1m3dZli9buYFuHVqWV1jXNfVAERERERERERERERG5zoVWqmB3efeBz/PutG+vcjQ3BiVQRERERERERERERESuc6GV7SdQAJydHK9iJDcOJVBESmjWrFmYTCYGDRpU3qFYDBo0CJPJxKxZs8o7FBERERERERERESkHOTkFz9natEHtqxjJjUMJFCkTDRs2xGQy4e7uTlJSUpm2PWvWLMaMGcOhQ4fKtF0RERERERERERGRG0VOTk6B625uWv8qRnLjUAJFLtuWLVvYsWMHAGlpacybN69M2581axZjx45VAqUQlStXJjIyEl9f3/IORURERERERERERMpBjmG/B0qXqGY4Oztd5WhuDEqgyGX78ssvAfDz87P6Xa6ecePGERMTw913313eoYiIiIiIiIiIiEg5yN8DxcXFGQCjgMSKFE0JFLks2dnZzJkzB4CPPvoIR0dHVq5cSVxcXDlHJiIiIiIiIiIiIvLvkX8OlEA/HwCUPyk9JVDksvz6668cP36cSpUq0bdvX2655RYMw+Drr78utF5KSgoTJ06kdevW+Pn54eHhQa1atejfvz8rV64EIDo6GpPJZPm9U6dOmEwmy495wnRzuY4dO9rd1qFDhzCZTERERNisW7duHc8//zzNmzenYsWKuLq6UrVqVfr378/OnTtLfVwKi8EwDKZMmULDhg3x8PCgYsWK9O/fv8Ckk3l/AebPn0/79u3x8/PDZDJZhjUrahL5PXv28Nhjj1GzZk3c3d0JDAykWbNmjB49muPHj9uUT0xM5OWXX6ZBgwZ4enri7e1N69at+fzzz+2OpZiVlcXkyZNp2bIl3t7euLq6UqVKFdq0acPo0aM5d+5cqY6biIiIiIiIiIiIFE/e53ZVq1SkY+vGgHqgXA4NfCaXZfbs2QDcf//9ODo68tBDD/HLL7/w5Zdf8tJLL9mtExcXR/fu3dm9ezcAtWrVwtvbm0OHDvHVV19x5MgRoqOj8fX1pW3btmzfvp2kpCQaNGhgNcdHcHDwZcffr18/Dhw4QGBgIJUrV6ZKlSqWOObPn89PP/1UYGKmtIYOHconn3xCWFgY9erVY+fOnXz11VcsW7aMP//8k8jISLv1JkyYwIsvvkhwcDC1a9cu9pwwX3/9NYMHDyYjIwN3d3fq1atHSkoKO3fuZNOmTVSrVo1BgwZZyu/cuZNu3bpx7NgxXFxcqFmzJunp6WzYsIH169ezfPlyvvvuO0tSB6Bv377Mnz8fgBo1ahAQEMCJEyfYsGEDa9eu5e6776Zx48alPWQiIiIiIiIiIiJSBPMcKLF/ziE8tBJDX/0AUALlcqgHipTahQsX+OGHHwB46KGHAOjduzfu7u7s3r2bv//+26ZOdnY2vXv3Zvfu3TRv3pxdu3axd+9e/v77bxISEti8eTP3338/AE2aNGHVqlU0adIEgClTprBq1SrLT48ePS57H1577TUOHDjAmTNn2L59O1u2bOHMmTNMnz6dzMxMhgwZYrfHRWkdO3aM6dOnM2fOHA4fPsxff/3F0aNH6dKlC6dPn2bAgAEFfqG99tprfPbZZxw/fpwNGzYQHx9PaGhoodv766+/ePjhh8nIyOD555/n9OnT/P333+zevZvk5GTmzJlDzZo1LeUvXrzInXfeybFjxxg+fDinT59m586d7N+/nx07dlC/fn3mzZvH1KlTLXX+/vtv5s+fT9WqVdm1axf79+9nw4YNxMXFkZiYyOeff05gYGDZHEARERERERERERGxy/wc08HBwWpUGyVQSk89UErhjkdGceBwfHmHUWI1wquwePrbZdbe/PnzSUlJoWbNmrRo0QIAb29vevXqxffff8+XX35Js2bNrOosWLCAv//+m4oVK/Lzzz/bPFhv3LjxVe2pMGDAAJtlTk5ODBkyhOjoaL766ivWrVtHmzZtymR7WVlZDB8+nL59+1qWBQYG8vXXXxMeHs6GDRuIjo6mU6dONnUff/xxHn30Uas4izJ69GgyMzMZPHgwEyZMsFrn7OxsFQfAjBkzOHDgAHfffTeTJ0+2WlevXj2++eYbGjduzPvvv8/QoUMB2LdvHwD33HMPdevWtarj4+PDI488UmScIiIiIiIiIiIicnnMc6A4OOQmTgL9c+dAqRDoV14hXffUA0VK7csvvwTgwQcftFpu7o0yZ84csrKyrNYtWrQIgMGDB18zvRJiYmIYPXo0vXv3pmPHjkRFRREVFWWZe2Xr1q1luj1z4iGvihUrcs899wCwbNkyu/XsJXsKk5qayi+//ALA888/X6w6CxYsACgw6dGoUSMiIiI4ePAgR48eBaBq1aoA/PbbbyQmJpYoRhERERERERERESkb2dmXeqAAjHryIV568iE+G/dseYZ1XVMPlFIoy14c16tjx46xYsUKwDaB0qNHD/z9/Tl16hTLly+nZ8+elnXmeU9at2599YItxLhx43jllVcKHaarLJMCzs7OVkNm5WXuvbF3795C1xfX/v37yczMxM/Pr8B5VfLbvn07kDtc2Ntv27/Oz5w5A+ReA6Ghodx88820atWK9evXU7VqVbp27Ur79u3p0KEDTZs2tZorRURERERERERERK6MHCP3GafjPwkUNzdX3n7+0cKqSBHUA0VK5euvvyYnJ4emTZvaPJx3cXHh3nvvBS71UjFLSkoCwM/P76rEWZg//viDUaNGYTKZGDduHDt37uTChQvk5ORgGAYvv/wyAJmZmWW2zcDAQEsGOL/g4GAAkpOT7a739PQs0bZKc6zPnz8P5M5rsnr1ars/5vhSU1OB3Iz20qVLeeqpp3B3d2fRokU888wzNG/enGrVqjFr1qwSxS0iIiIiIiIiIiIld2kILz32LyvqgSKlYk6MbNq0qdAeBosWLSIpKQkfn9zx9ry9vQE4d+5cmcVS1GRIFy9etLv866+/BuC5557jxRdftFl/5MiRMorwkoSEBHJycux+iZ06dQq4dIwuV2mOtZeXF+fOnWPfvn0F9pSxx9/fnw8++IBJkyaxdetW/vjjD3744QdWrFjBww8/jJeXl2WIMhERERERERERESl7lyaR14gwZUWpKCmxzZs3s2PHDkwmE8HBwQX+uLi4kJqayvz58y1169evD8C6deuKvb2ihoAy98w4ffq03fX79++3u/zQoUMABU4QX9Zzn0Bub5YDBw7YXWce3qx27dplsq1atWrh4uLCuXPn2LNnT7Hq1KtXD4AdO3aUapsmk4nGjRszfPhwfv/9d0ti6vPPPy9VeyIiIiIiIiIiIlI86oFS9nQkpcTMvU/at2/PiRMnCvx55plnrMoD3HXXXQDMmDGj2HOLuLu7A5eGjMqvevXqABw8eJCEhASb9dOnTy+03ZMnT9qsW758+RVJoABMnTrVZtnp06f5/vvvAbj11lvLZDvu7u6WtiZOnFisOr179wbgww8/LLBHT0mY57qJj4+/7LZERERERERERESkYOY5UBw0J3GZUQJFSiQ7O5s5c+YA0L9//0LL9uvXD4Do6GjLcFh33XUXzZs359SpU/Ts2dOmZ8TWrVv55JNPrJaZEyQrV660u52AgABatmxJeno6I0eOtMxZkp2dzfjx41m2bJndelFRUQCMHz+e2NhYy/KNGzcyePBg3NzcCt2/0nBycmLq1KmWZAnkTlLfr18/0tLSaN68OZ06dSqz7Y0ePRpnZ2emT5/OqFGjSElJsazLzMxk7ty5rFq1yrLs8ccfp3r16qxYsYKHHnqI48ePW7V34cIFvvvuO0aOHGlZ9vXXX/PGG29YevSYJSQk8OGHHwLQtGnTMtsnERERERERERERsaUeKGVPR1JK5JdffuHEiRO4ubkVOadFvXr1aNKkCYZhWOYbcXR0ZP78+URGRrJ+/Xrq1KlDZGQkzZs3JygoiMaNGzN37lyrdu6//34AJkyYQGRkJB06dKBjx478/PPPljITJkzAycmJ2bNnU7FiRVq0aEFwcDCvvvoq77//vt34HnvsMapXr86BAweoU6cOjRo1ok6dOrRs2RJfX1+efPLJyzlUdoWEhDBkyBDuu+8+IiIiaNGiBaGhoSxfvpzAwEBmz55d5JBlJdG8eXNmzJiBs7Mz48aNo0KFCjRr1ox69erh4+ND3759rYY48/Ly4n//+x/VqlVjzpw5hIaGUq9ePVq3bk1kZCR+fn7cf//9rFmzxlLn9OnTvPbaa1SrVo3Q0FBatmxJw4YNqVKlCr///jshISG88cYbZbZPIiIiIiIiIiIiYktzoJQ9JVCkRMzDcd1+++34+voWWd7cCyXvMF5hYWH8/fffjBs3jqZNmxIfH8/u3bsJCAhg4MCBNg/b27VrxzfffEPLli05duwYf/zxBytXruTEiROWMh07dmTZsmVERUWRkZHB3r17adq0KdHR0fTq1ctubD4+PqxatYoBAwbg4+PDnj17yMjIYOTIkaxdu7bMJnPP7+OPP2by5Ml4e3uzY8cOPD09eeihh/j777+pW7dumW+vX79+bNmyhYcffpigoCB27NjB6dOnqV+/PmPGjKF79+5W5evUqcPWrVsZP348LVq04NixY2zZsoWMjAw6dOjAxIkT+fbbby3l+/Tpw4QJE+jatSuOjo5s376d48eP06BBA95880127NhBWFhYme+XiIiIiIiIiIiIXKIeKGXPZJTFRAfXgdDQUI4ePVpkuezsbPbu3Uvt2rVxdHS8CpHJv8GhQ4eoVq0a4eHhNkNdicj1Q38jRERERERERORa1X/EW3y18BfS9izH1dWlvMO5LhSVN1AqSkRERERERERERETkOqceKGVPR1JERERERERERERE5Dp39nwyoDlQypISKCIiIiIiIiIiIiIi17ml0esB9UApS07lHYCIiIiIiIiIiIiIiJRednY2JpOJCoF+mEzqgVJWlEARuQoiIiIwDKO8wxAREREREREREZEb0MkzZzEMg3t7dijvUG4o6ssjIiIiIiIiIiIiInIdO3biDAChlSqUcyQ3FiVQRERERERERERERESuY0ePnwYgRAmUMqUEioiIiIiIiIiIiIjIdex88gUA/H29yjmSG4sSKCIiIiIiIiIiIiIi17H0jEwAXF1cyjmSG4sSKCIiIiIiIiIiIiIi1zFzAsXNVQmUsqQEioiIiIiIiIiIiIjIdSwtPQMAVxfnco7kxqIEioiIiIiIiIiIiIjIdezSEF5KoJQlJVBERERERERERERERK5j6Rm5PVA0hFfZUgJFREREREREREREROQ6piG8rgwlUERERERERERERERErlMZGZm89/l3ALiqB0qZUgJFbniDBg3CZDIxa9Ysq+VjxozBZDIxZsyYcomrLBS0b+XJZDJhMpnKO4wyNXXqVEwmE99++215h2IlIiICk8nEoUOHyqS9W265BT8/PxISEsqkPRERERERERERufKOHD9l+ezv401qZk45RnNjUQJFSsX84Dbvj7u7OzVq1GDw4MHs3LmzvEO8qsaMGXNdJ2KkYBcuXOD111+nTp063HfffVbroqOjGTNmDNHR0eUTXBl79dVXOX/+PG+++WZ5hyIiIiIiIiIiIsV09Phpy+dxK5PwGLWf2MRMxi5PoOb4WBJTsssxuuubEihyWWrVqkXbtm1p27YtNWrU4OjRo8ycOZNmzZrx448/lnd4hQoKCiIyMpKgoKDLbmvs2LGMHTu2DKK6/kVGRhIZGVneYZSZSZMmcfLkSV588UUcHKy/MqOjoxk7dmy5JVBq1KhBZGQkzs5lM7Zlp06daN26NVOnTiUuLq5M2hQRERERERERkbKVkppGTs6lXibnky8CMH3Cc7z+ayIA1cfFMuaXBA4kZBI4+gCT/zxbLrFe75RAkcsyatQoVq1axapVq9ixYwdxcXF06dKF9PR0Hn74YS5cuFDeIRZo2LBhxMTEMGzYsPIO5YYSExNDTExMeYdRJrKzs/n000/x8PDgnnvuKe9wbPz222/ExMQQEhJSZm0OHDiQjIwMPv/88zJrU0REREREREREykbC2fN41u3OU2OnWJalpqUD4OvtWWA9hxtrxP2rRgkUKVPBwcF8+eWXuLq6kpCQwC+//FLeIYmU2pIlS4iPj+eOO+7A07PgP0A3knvuuQcnJydmzZpFdra6d4qIiIiIiIiIXEsOxh0H4KMvFnL46Am+XfwbKam5CRR3N1cq+zjarde/mc9Vi/FGogSKlLlKlSpRq1YtAPbt2wfAoUOHMJlMREREAPD555/TokULvL29bSYcP3r0KMOHD6d27dq4u7vj5+dHp06dmDdvXoHbvHjxIi+99BLVqlXDzc2NiIgInnnmmUJ7wBQ1ifyxY8cYOXIk9erVw9PTE19fXxo2bMizzz5r2S9zG2b554XJP7n31dq3wuTd7xMnTjBkyBCqVKmCm5sbdevWZeLEiWRlZdnUmzVrFiaTiUGDBnHx4kVGjRpF7dq1cXNzo2PHjjbHwB7DMPj+++/p2bMnFStWxNXVlbCwMHr06MGsWbPs1tmwYQN9+/YlJCQEFxcXgoODuffee9m8ebPd8ocPH+bxxx+nevXquLq64u3tTfXq1bn77rtLPAn83LlzAbjtttts1plMJsuwbWPHjrU674MGDbKUyzvR+4oVK+jRowdBQUGYTCbL0F/nzp3jv//9L3feeSc1a9bE3d0dX19fWrVqxYcffmj3fORvO6+OHTta2o+JieHee+8lKCgId3d3mjVrxnfffVfgPgcFBdGqVSuOHj3K6tWrS3C0RERERERERETkSks8l2T5fNvgF3lg+Bss+2MDAK6urpy5mE3NIGdahbkxf0Bl7qjnydcPVsLP3X5iRQrnVN4ByI3JMIwC1z3xxBN8+umnVK1alTp16rB//37LupUrV3LnnXdy/vx53N3dqVWrFufOnSM6Opro6GieeeYZJk6caNXexYsXueWWW9iwYQMmk4n69euTk5PDpEmTiI6Opnbt2iWO/7fffqN3794kJSXh7OxM3bp1ycnJ4eDBg7z33nt4eXkxZswYwsLCaNu2reVBc9u2ba3acXNzu+b2zSwhIYGWLVsSHx9Pw4YN8fb2JiYmhueee47Vq1czf/58mzk/AFJTU2nfvj2bN2+mTp061KtXD1dX1yK3l5GRQd++fVm4cCEAlStX5qabbiI+Pp5ly5bx888/WyUeIHf+kWeeeQbDMAgICKBBgwbExcUxb948Fi1axLfffkvv3r0t5Q8dOkSLFi04c+YMHh4eREZG4ujoSFxcHD/88AOxsbH07du32Mfozz//BKBly5Y269q2bUtcXBxHjhyhatWqhIWFWdbZOy9z5szhlVdewdfX15IkMVuyZAmPPPIILi4uVK5cmYYNG5KQkMBff/3Fhg0bWL58OYsXL7Z7Pgrz999/WxJmtWvXJi4ujk2bNnH//feTkZFBv3797NZr2bIlq1ev5o8//qB9+/Yl2qaIiIiIiIiIiFw5pxLOWT7v3HsIgO17YgFIx5XMbOhS04NP+gQD0Luh99UO8cZi/EuEhIQUq1xWVpaxa9cuIysr6wpHdH0LDw83AGPmzJk2644fP264uroagDF//nzDMAwjNjbWAAxHR0fD09PTWLRokaV8SkqKYRiGcezYMSMgIMAwmUzG22+/baSlpVnKrF692ggJCTEA48cff7Ta3ogRIwzACA8PN3bs2GFZvmXLFiMkJMRwdna2G+vo0aMNwBg9erTV8sOHDxu+vr4GYAwYMMBISEiwrMvOzjaWLFliLF682KoOYBR2O13tfSuMeb+dnJyMhg0bGrGxsZZ1K1eutOz7Rx99ZFVv5syZlnNYu3ZtY9euXZZ1qampRR6Lp59+2gCMoKAgY+nSpTbHJ/95WLp0qWEymYygoCDLdWQ2ffp0w8nJyfD29jbi4+Mty4cNG2YAxsCBA43k5GSrOrt37zamTZtW+MHJ4/DhwwZgeHt7Gzk5OXbLFHQN5WW+VxwdHY2xY8camZmZhmEYRk5OjuU62Lp1q7FkyRKr68IwDOPAgQNG+/btDcCYNWtWgW3nPYeGYRgdOnQwAMPZ2dkYNmyY5fzk5OQYL7zwggEYVapUKfB77ptvvjEAo1u3bgXu17+V/kaIiIiIiIiISHl6d9ocg/AOVj/e9XsYhHcw5qzca/DsHuONX86Ud5jXjaLyBuqBUgov/jyd+OQz5R1GiVXxDmJ890eu6DZOnTpF//79SU9Px9/fn65du1qtz87O5vXXX+eOO+6wLDO/if/ee++RmJjIiBEjeOmll6zqtWnThk8//ZTbb7+dSZMm0atXLwCSk5OZNm0aAFOnTqV+/fqWOjfddBNTpkyx6qFQHBMmTOD8+fN07tzZMmyVmYODg93hnIpyrexbXllZWcyaNcsyrBpA+/bteeONNxg+fDgTJ07kySeftBmOKzs7mzlz5lC3bl3Lsrw9beyJj4/n448/BmDBggW0a9fOan2VKlVshlJ7+eWXMQyD//73v1bXC8CQIUPYvXs37733HtOnT+fVV18FLg0ZN3LkSLy8vKzq1KlThzp16hQaZ16HDx8GcoekK2hIspLo2bMnr732muV3k8lk6bnTqFEjGjVqZFOnevXqzJgxg5o1a/L1118zcODAEm2zXr16TJ482dJzxWQy8cYbb/DFF18QHx/Ptm3baNKkiU29ypUrA5eOgYiIiIiIiIiIlD/DMPhm0W82y5MvpADg4OoOpODjppk7yoqOpFyWt99+m6ioKKKiomjQoAFVq1bl119/xdnZmc8//xxvb9suYgMGDLDb1oIFCwB45BH7SZ7u3bvj4uLCmjVrLHNC/Pnnn6SkpBAeHk6PHj1s6tx5552EhISUaJ8WLVoEwHPPPVcmD87h2tm3vG6++WaaNm1qs3zw4MG4ublx6NAh9uzZY7O+fv36dusV5qeffiIzM5PWrVvbJE/sOXz4MJs2baJixYo2yRMz8/KVK1dallWtWhWAefPmFTqMXHGcOZObJA0ICLisdswKuu7N0tPT+eabb3j00Ufp1q0b7dq1IyoqypI02bp1a4m3OXjwYJthv5ydnbnpppsAOHjwoN165n0+ffp0ibcpIiIiIiIiIiJXxmff/MjmnfsKXO/lnTtRvLtz2TzTFM2BUipXuhfH9WTfvn2Wt/5dXFyoVKkS7du355lnnqFx48Y25YOCgggKCrJZfuHCBctE2I899lih20xLSyMhIYHg4GD27t0L5PYusJfscHBwoHbt2hw7dqxY+5OcnGwp27p162LVKcq1sm/55e1BkpenpydVq1Zl37597N2716bXRkH1CrN7926g+Md0+/btQO7xiIqKslsmLS0NwGr/hw4dyhdffMEbb7zB7Nmz6d69O+3ataNTp05UqVKlRDGb2y/O/C7FUdhxi4uL49Zbb7WbsDJLTEws8TZr1Khhd3nFihWB3GvTHnOvsNTU1BJvU0RERERERERErozZC5YXuj49O/cZooez+k2UFSVQ5LLMnDnTZuLvwnh6etpdfv78ectn84TshTE/2DU/AK5QoUKBZYODg4sdX1JSkuWzr69vsesV5lrZt/zMD9ELanffvn0kJyfbrCvoHBbGfFz9/PyKVd58zJKSkoo8Znkf8jdu3Jg//viD0aNH8/vvvzNt2jSmTZuGyWSia9eufPDBB8VOAJl7YZw7d65Y5YtS2HEbNGgQe/bsoVWrVowdO5bGjRsTEBCAs7MzWVlZlv+W1TbNvVIK6qVjTtbYS3aKiIiIiIiIiEj5yPzn+VBk9arsOXjEZn1qVu6zHvVAKTtKoMg1Ie98FRkZGTg7O5eoXmFDDZ06darYceQdcuz8+fNlkkS5VvYtv+K0a28IttIwt1PcZIR539u2bcuqVatKtK3WrVuzbNkyLly4wOrVq1mxYgXffPMNy5cvp2vXruzYsaNYiRxzgqk0PT9KIj4+nhUrVuDh4cFPP/1kM2TYkSO2fwyvNPM+F5a8ExERERERERGRqys1LR0XF2eWfzmR2CPHOXL8FP1HvA1Anx7tSc3MTaCoB0rZ0ZGUa4Kvr69liKWdO3cWu17t2rUB2LNnj9236XNycgodFik/Hx8fQkNDAVi3bl2x6xXmWtm3/MzDauWXkpJCXFycVQyXq379+kDxj2m9evWA3BhzcnJKtU0vLy+6devG+PHjiYmJoUaNGhw7doylS5cWq37dunVxcXHh2LFjVj2T8iqLOXLME7XXqVPH7nwrpZn75HLt2rULoMRz3YiIiIiIiIiIyJVz9nwyTerVJCwkmA6tG9Pv7luZPPr/AHjhPw9yIT33OZp6oJQdJVDkmtG7d28APvjgg2LXiYqKwsPDg0OHDrFs2TKb9YsXLy7xHCF33XUXAO+9916x6xQ1Z8S1sm95rVmzhi1bttgsnzFjBmlpaYSHhxMZGVnq9vPq2bMnzs7OrFu3rljDmNWqVYsGDRqQmJjI7NmzL3v7Hh4eNGzYEMjt8VEcbm5uNG/eHMMw2LRpk90yZTFXiLmNU6dO2U2UvfPOO6Vuu7Q2bNgAQLt27a76tkVERERERERExL5zSRfw8/GyWvZ/g3pzfvv/aHFTHbafSAegVpBLeYR3Q1ICRa4ZL7zwAgEBAXzxxReMHDnSZrinxMREZsyYwZtvvmlZ5uPjw6OPPgrAk08+adWrYtu2bQwfPrzYQ2aZPffcc/j6+vLLL78wZMgQzp49a1mXk5PDTz/9xJIlS6zqVK9eHYCVK1de0/uWl5OTE4MGDbL0gABYtWoVr732GgDPPvtsmfSwAKhcuTLDhg0DcpNJy5dbT3gVHx/P66+/brVswoQJmEwmhg4dyvTp023mADl48CBvvfUWCxYssCx74oknmDt3LikpKVZl//jjD3777TegZL0qbr31VoAChxEzn/c1a9aUao4SyO2d4+/vz9GjR3nrrbcsSZS0tDSeeuopNm/eXKp2S8swDNasWYOjoyOdO3e+qtsWERERERERERH7MjOzuJiShr+v9ZD7JpMJH+/ceXA3xKUR6utEFV/N3FFWlECRa0ZoaCiLFy8mKCiISZMmUbFiRRo1akTr1q2pUaMGQUFBDBkyhB07dljVe/PNN2nWrBmxsbHUr1+fRo0a0bBhQxo3bkyFChXo06dPieIICwtj3rx5eHt7M2PGDIKDg2ncuDGNGjXCx8eH2267jb/++suqzv333w9Ar169aNq0KR07dqRjx46cOHHimtq3vB5//HESExOpWbMmTZo0oU6dOrRr146zZ89y++238+STT5a6bXvGjRvHnXfeyalTp+jWrRshISG0bNmSqlWrEhoayujRo63K9+zZkylTppCens6jjz5KQEAAzZs3p0WLFlSqVIkaNWrwyiuvWM0Ds3btWvr27Yuvry/16tWjVatWRERE0KFDB5KTk+nXrx+dOnUqdswPP/wwDg4OzJ071+76W2+9FX9/f1atWkVYWBhRUVF07NiR8ePHF3sbzs7OvPHGGwC8+uqrVKlShRYtWhAcHMyUKVOYMmVKsdsqC2vWrOHo0aN069bNMvSciIiIiIiIiIiUr3NJFwBseqCYGYbB3jOZNKik3idlSQkUuaa0bduWXbt28fLLL1OvXj1iY2PZtm0bDg4OdO/enalTpzJ58mSrOl5eXkRHR/PCCy8QFhbGnj17SE5OZsSIEaxcuRJXV9cSx9GlSxd27NjBsGHDCA8PJyYmhiNHjlCjRg2ee+45+vfvb1X+xRdfZPTo0dSsWZNdu3axcuVKVq5cSVpa2jW3b2ZBQUFs2LCBAQMGcPLkSWJjY4mMjGTChAksWLAAB4ey/XpwdXVl4cKFfP3113Tu3Jm0tDS2bt2Kg4MDPXv2tDtU19ChQ9myZQuPPPIIFSpUYOfOnezbt4+goCAeeOABvv/+ewYMGGApP2nSJJ566ikaNWrEmTNnLEOUdevWjcWLF5d4OLCwsDC6devGjh072LZtm816Hx8fli9fTo8ePUhPT2ft2rWsXLmSmJiYEm1n6NChfPXVVzRu3JjExET2799P8+bN+emnn3jkkUdK1NblmjNnDoCl95OIiIiIiIiIiJS/ohIopy9mk55lEO5f+hFrxJbJsDfo/g0oNDSUo0ePFlkuOzubvXv3Urt2bRwdHa9CZCJX15gxYxg7diyjR49mzJgx5R3ONW/16tVERUUxZMgQpk+fXt7hXFHnzp0jIiKCsLAwtmzZUuZJtBuB/kaIiIiIiIiISHn4acU6bnv4RSaP/j+GP2w7Ks1fR9Jo8WEcb3UPZFTnwHKI8PpUVN5AT8dERArRtm1b7rrrLmbPnm01X8yNaPLkyZw/f57x48creSIiIiIiIiIicg0Z/8k3ALRuUs/u+phTGQBU9VMPlLJU7CdksbGxfP755zz66KPcdNNNODk5YTKZrCa9zm/MmDGYTKZCfwob6mb37t089NBDVK5cGTc3N2rUqMGzzz5rMwG3iMiVNHHiREaNGsWRI0fKO5Qryt/fn/fff5+ePXuWdygiIiIiIiIiIvKPw0dP8OeGbTSsU52WjevaLfPh6rMAtKjqdjVDu+E5Fbfg5MmTbeZnKK6qVasSFhZmd52Hh4fd5StWrOC2224jNTWVChUqUL9+fWJiYnjvvfdYuHAha9asITg4uFTxiIiURI0aNf4Vw50NHz68vEMQEREREREREZF8HnnxXQA6t2kKwMYjabg6mmhUxZWUjBw8XBw4kJBJrSBn6lTUJPJlqdgJlKCgIHr16kXLli1p0aIF06dPZ/78+cWqO3jw4BI9fExOTub+++8nNTWV4cOHM3HiRJydnUlISODOO+9k9erVDBkyhCVLlhS7TRERERERERERERGR682vq/4G4Ka6NTAMg5YfxgHw/u0VeGbJad7sFkhiSg631fEszzBvSMVOoLzyyitWv3/77bdlHozZp59+yunTp6lbty7vv/++ZaLewMBAvvnmG2rUqMH//vc/Nm3aRNOmTa9YHCI3ojFjxvwrelOIiIiIiIiIiIhc77Kzsy2fB97TnSPnsiy/j/zxNACfrjsPQLi/5j8pa9fkLMELFiwAYNCgQZbkiVlYWBhdunQBYN68eVc9NhERERERERERERGRq2HLrv1Wv/f47zGbMuakSo1AJVDKWrF7oFyOFStWsHPnThISEggICKBly5YMGDCASpUq2ZTNysri779zuyS1bdvWbntt27bl559/Zv369Vc0bhERERERERERERGR8nLsxBkAZr77ApuPpbPrZAYAdSq6EHMqw6rsLTXtzzcupXdVEih//PGH1e/z589nzJgxTJ06lUGDBlmtO3ToEJmZmQBUr17dbnvm5fv27Sv7YEVERERERERERERErgHnky8C4OfjxdI9uZ9bh7nxzUOVWbA9GRMmnlmSO5RXmIbwKnNXNIFSuXJlRo0axd1330316tVxd3dn8+bNvPnmmyxdupTBgwcTGBjI7bffbqlz9uxZy2d/f3+77ZqX5y2b3/vvv8/7779v+T0pKYlly5YB4OvrS+vWrVm3bh3nz+eOD1ejRg3CwsJYsWIFFSpUICkpCW9vb3JyckhNTQXAZDLh4+NDSkqKJcnj7OyMh4cHSUlJGIYBgLu7Ow4ODly8eNGyfW9vbzIyMkhPTwfA0dERLy8vLly4YBnHzs3NDWdnZ5KTky31vLy8yM7OtsTg4OCAt7e3VQwuLi64u7tbxeDh4YHJZLKKwcfHh/T0dEsMTk5OeHp6FiuGrKws0tLSrGK4ePEiWVlZVjGYj6c5BoCUlBSrGNLS0sjIyLCKITk5mZycHEsMTk5OXLhwocQxuLm5kZSUVGgMvr6+pKamljgGb29vMjMzLTGYz2HeGFxdXXF1dbWKwdPTE8MwLDGYr6O8MZivo7wxuLu74+joWKwY8p5DV1dXXFxcrM6hp6fnVb2Wr4UYdD/pfroS91NOTg45OTmcOHGCHTt2WPa5c+fObN++nfj4eAAqVarETTfdxO+//265tho2bIiHh4dV78lOnToRFxfHgQMHLMezsL9PZq1atSIlJYXt27db9vmWW25h69atnDhxAoAqVarQsGFDfvvtN8sxbdSoEW5ubmzYsMHS1i233EJsbCyxsbEA+Pn50apVK9asWWPZ71q1alGlShVWrlxpqXfzzTeTlJTEzp07LceqY8eObNmyhZMnTwIQGhpK/fr1+fXXXy3HtHHjxjg7O7Nx40ZLW126dGH//v0cOnQIgICAAFq0aMHq1ast56x27dpUqlTJ6qWMNm3acPbsWXbv3g3kXmsdOnRg06ZNnD6d+4/HqlWrUq9ePZYvX265p5s0aYKDg4OlxyvArbfeSkxMDHFxuRPxBQUF0axZM/7880/L9VanTh2CgoJYtWqVpV5UVBRnzpwhJiYGyL1P2rVrx99//82ZM7lvCYWFhVGnTh2WL19uqdesWTNycnLYvHkzkHst33rrrezatYsjR44AUKFCBZo2bcrKlSst12ndunXx9/dnzZo1lrbat2/PiRMn2Lt3L5B7j7dt25aNGzeSmJgIQEREBDVr1uTXX3+11GvRogWZmZls2bIFyL0HunTpws6dOzl69CgAwcHBNG7cmOjoaMv3Xf369fHx8WHt2rWWtjp06EB8fLzl5RZvb2/atGnD+vXrOXfuHADVqlWjWrVq/P7775Z6LVu2JC0tjW3btgG6n3Q/6X4C3U9mup9y6X7S/aT7SfcT6H4y0/2US/fTv+9+WrfhLwAcTAbbjuRe689W28PBzft4pnNntm3bzrA6yURVvMjWram6nyjZ/VQUk2G+G0po0KBBfPHFF7zxxhs2E8wXxTAM+vTpw8KFC6lRowb79u3DZDIB8Oeff9K+fXsgd4IcBwfbaVp+//13OnfujKOjo+UkFCU0NNRyMRcmOzubvXv3Urt2bZv5V0RE5N9NfyNERERERERE5Gp666MveWXif9m05HNG/OnG1uPpnH29ZnmHdcMoKm9QLpPIm0wmxo8fD8CBAwcsWWjIzX6amd8mzs+cwXN3d7+CUYqIiIiIiIiIiIiIlB/zEF6+Pp7EJmZSLUDDdF1N5ZJAgdyuZAEBAQDs37/fsjzvsF0FDdFlXl7QEF8iIiIiIiIiIiIiIte7d6d9C8D+JBfizmXRINilnCP6dym3BArkjpEGWA3DFRERYVl+8OBBu/XMy2vVqnWFIxQRERERERERERERKXtpaems/Xtngesvzb5hotsXiTg7wlPt1Kngaiq3BMqZM2c4deoUkDvOmJmTkxNNmzYFYPXq1Xbrmpe3atXqCkcpIiIiIiIiIiIiIlL2Hn1pIm36DCV67Wa769PS/5niIqQJAE2quNEs1M1uWbkyyi2B8v7772MYBr6+vrRo0cJqXe/evQGYNWsW2dnZVuvi4uL49ddfAejTp8/VCVZEREREREREREREpAz9tGI9ALv2H7a7fuAz4wAIuqkzAMOj/K5KXHLJFUug7Ny5kyeffJKdO627IKWlpfH2228zYcIEAF544QVcXKzHbfvPf/5DUFAQu3fvZuTIkWRmZgKQkJDAgw8+SFZWFj169KBZs2ZXKny5gQwaNAiTycSsWbOslo8ZMwaTycSYMWPKJa6yUNC+lSeTyYTJZCrvMMrU1KlTMZlMfPvtt+UWQ8eOHTGZTERHR1stL+11HB0djclkomPHjmUWY1EK2oerKSkpCX9/f6KiosotBhERERERERH5d/tpxTrq3NKfxHNJAGRlZduUMQyD7/8XDZg47xlBBU9H7m3kfXUDleInUFavXk1QUJDlx/wgcdy4cVbLjxw5AkBmZiaffPIJDRo0oGLFijRv3pzmzZsTGBjIyy+/TE5ODkOGDOHFF1+02ZaPjw/ffvstbm5ufPjhh4SEhNC8eXPCwsJYvXo1ERERzJgxo4wOgZRGRESE5UG5+cfd3Z0aNWowePBgm8TZjW7MmDHXdSJGCnbhwgVef/116tSpw3333Vfe4VyzoqOjGTNmTLkmR4rDx8eH4cOHs3r1ahYtWlTe4YiIiIiIiIjIv0zM/sPc9vCL7Dl4xLIsK9s2gXIu6ULuB1dvMnHmvpu8cHG6sV5avh4UO4GSmZlJQkKC5Sc9PR2AlJQUq+XmIbciIiJ444036NGjB15eXuzZs4ft27cTEBDAPffcw88//8z06dMLfFO9c+fO/PXXX/Tt2xeTycT27dsJDg5m5MiRbNq0iUqVKpXB7svlqlWrFm3btqVt27bUqFGDo0ePMnPmTJo1a8aPP/5Y3uEVKigoiMjISIKCgi67rbFjxzJ27NgyiOr6FxkZSWRkZHmHUWYmTZrEyZMnefHFF3FwKLdRDwtUltfx5YiOjmbs2LGFJlDCwsKIjIzEw8Pj6gVmx9NPP42HhwcvvfRSnsnYRERERERERESuvNV/77BZlpKaZrPsSHzu/OENGjcGoJK30xWNS+wr9lHv2LFjiR40+fn58corr5QqKLP69eszZ86cy2pDrqxRo0YxaNAgy+8nT56kX79+/Prrrzz88MMcOnQILy+v8guwEMOGDWPYsGHlHcYNJyYmprxDKDPZ2dl8+umneHh4cM8995R3OHZdT9fx7NmzyzsEAPz9/bn99tuZO3cuv//+O507dy7vkERERERERETkX+K9z7+zWbb6L9ukytETpwG45ZaO7NivBEp5ufZep5brWnBwMF9++SWurq4kJCTwyy+/lHdIIqW2ZMkS4uPjueOOO/D09CzvcKQM9e3bF4Dp06eXcyQiIiIiIiIi8m+RlpbOgbh4XF2crZZv2bXfpuyxE2cAmBVXBYBK3o5XPkCxoQSKlLlKlSpRq1YtAPbt2wfAoUOHMJlMREREAPD555/TokULvL29bYZxO3r0KMOHD6d27dq4u7vj5+dHp06dmDdvXoHbvHjxIi+99BLVqlXDzc2NiIgInnnmGS5cuFBgnaIm3z527BgjR46kXr16eHp64uvrS8OGDXn22Wct+2Vuwyz/vDCHDh0ql30rTN79PnHiBEOGDKFKlSq4ublRt25dJk6cSFZWlk29WbNmYTKZGDRoEBcvXmTUqFHUrl0bNzc3q4nIC5tE3jAMvv/+e3r27EnFihVxdXUlLCyMHj16MGvWLLt1NmzYQN++fQkJCcHFxYXg4GDuvfdeNm/ebLf84cOHefzxx6levTqurq54e3tTvXp17r777hJPAj937lwAbrvtNpt1zZs3x2QyFXrupkyZgslkonfv3pZlqampzJkzh759+xIZGYmXlxdeXl40btyYN998k4sXL5YoxqKu44ULF9KmTRs8PT0JDAykV69e/PXXX4W2+csvvzBs2DBuuukmAgICcHNzo0aNGjzxxBPExcXZlDeZTJYh7MaOHWt1D+TtoVbYJPKGYfDVV1/RoUMH/Pz8cHd3p06dOrzwwgskJibajTPvtbZ06VLat2+Pt7c3vr6+9OjRo8BrBKBbt244OTnxww8/WIakFBEREREREZHrm2EYZGbaPte6VmzYGkNGRiajnxrEuoVTwcEJmjxAsuFuU/bo8dweKEkZuY/wAzyUQCkPSqDIFVHYcG9PPPEEjz32GCdPnqROnTr4+flZ1q1cuZIGDRowZcoUjh49Sq1atfDx8SE6Opp7772XZ5991qa9ixcvcssttzB+/HgOHz5MrVq18PT0ZNKkSXTo0KFUD0d/++036tWrx6RJk9i/fz81a9YkLCyMgwcP8t577/H1118DuXM6tG3b1lLPPB+M+cfNze2a2zezhIQEWrZsyRdffEFwcDDh4eHExMTw3HPPce+995KTk2O3XmpqKu3bt2f8+PE4OTlRr149XF1di9xeRkYGffr04b777mPp0qU4OTlx0003kZOTw7Jly3j44Ydt6kyaNInWrVszd+5c0tLSaNCgAdnZ2cybN49WrVqxYMECq/KHDh2iefPmfPbZZ5w8eZLIyEhq1qzJ+fPn+eGHHxg/fnyJjtGff/4JQMuWLW3WPfjggwCFDjNoXvfAAw9Ylv399988+OCDzJ8/n5SUFOrWrUuVKlXYuXMnr776Ku3btyc1NbVEcRbknXfeoXfv3qxduxZfX1+qVavGypUriYqKYtWqVQXW69GjB1OnTuXEiROEh4dTq1YtTp48yaeffkrTpk3ZtWuXVfm2bdtStWpVAKpWrWp1D9SuXbvIOA3DoF+/fvTv358//viDwMBA6tWrR2xsLO+88w5Nmzbl4MGDBdb/9NNPue2229i/fz+1a9cmOzubn3/+mfbt2xc4pJy7uzsNGzYkLS2NjRs3FhmjiIiIiIiIiFz7Hhz+Bi61uljm6b4WZGZmkXD2PAA79x4CoHmjSAKr1oAGd0HNTlxsPMjmeeq+Q0ehdlfL7zeHuyHlwPiXCAkJKVa5rKwsY9euXUZWVtYVjuj6Fh4ebgDGzJkzbdYdP37ccHV1NQBj/vz5hmEYRmxsrAEYjo6Ohqenp7Fo0SJL+ZSUFMMwDOPYsWNGQECAYTKZjLfffttIS0uzlFm9erUREhJiAMaPP/5otb0RI0YYgBEeHm7s2LHDsnzLli1GSEiI4ezsbDfW0aNHG4AxevRoq+WHDx82fH19DcAYMGCAkZCQYFmXnZ1tLFmyxFi8eLFVHcAo7Ha62vtWGPN+Ozk5GQ0bNjRiY2Mt61auXGnZ948++siq3syZMy3nsHbt2sauXbss61JTU4s8Fk8//bQBGEFBQcbSpUttjk/+87B06VLDZDIZQUFBluvIbPr06YaTk5Ph7e1txMfHW5YPGzbMAIyBAwcaycnJVnV2795tTJs2rfCDk8fhw4cNwPD29jZycnJs1h87dsxwcHAw3NzcjPPnz9usj42NNUwmk+Ht7W25xg3DMA4dOmR89913NvEdP37cuOeeewzAGDNmjE17HTp0MABjxYoVVssLuo43bdpkODo6GiaTyfjoo48s+5CcnGzcf//9lmunQ4cONtuaNm2acezYMatlKSkpxltvvWUARseOHW3qFBRHcfZhypQplmO9fPlyq2PStm1bAzBatWpl0575WvPw8LC6B5KSkozOnTsbgHH//fcXGM9jjz1mAMa4ceMKLJOf/kaIiIiIiIiIXLsI72AQ3sHYve9QeYdi8dzbnxiEdzBi4+KNNz78wiC8g9Fx8h6DZ/P8DF1vpKWlW9VrfvtjhunprQbP7jG8Xt5bTtHf+IrKG2jmmVK4Y+YxDiRklncYJVYj0JnFD4dc0W2cOnWK/v37k56ejr+/P127drVan52dzeuvv84dd9xhWebunttF7b333iMxMZERI0bw0ksvWdVr06YNn376KbfffjuTJk2iV69eACQnJzNt2jQApk6dSv369S11brrpJqZMmWI1fFJxTJgwgfPnz9O5c2fLsFVmDg4OdodzKsq1sm95ZWVlMWvWLMuwagDt27fnjTfeYPjw4UycOJEnn3zSZjiu7Oxs5syZQ926dS3L8va0sSc+Pp6PP/4YgAULFtCuXTur9VWqVLEZgurll1/GMAz++9//Wl0vAEOGDGH37t289957TJ8+nVdffRW4NGTcyJEj8fLysqpTp04d6tSpU2iceR0+fBjIHZLO3pBkVapUoUOHDqxYsYKFCxcycOBAq/XffvsthmFw1113Wa5xgPDwcMLDw23aq1SpErNnz2bx4sV8/fXXjB49utix2vP++++TnZ3Nvffey9ChQy3Lvby8mDVrFitWrODUqVN26z722GM2y9zd3Rk1ahRLly4lOjqaY8eOERJy+d8nhmHwzjvvAPD6669bfWdUqlSJuXPnUr16ddavX8/vv//OLbfcYtPGkCFDrIYK8/b2ZtKkSTRq1Iiff/65wG1XrlwZuHSuRUREREREROT6M+3rxWyLOciZxPOWZYnnk8sxImvvTssdUn7D1hjiTyaAmx/RR/IVcnIlJS0dV1cXAH5b/Td/bduDqUZur5SV/6l6NUOWPDSEl1yWt99+m6ioKKKiomjQoAFVq1bl119/xdnZmc8//xxvb2+bOgMGDLDblnk4pkceecTu+u7du+Pi4sKaNWssc3T8+eefpKSkEB4eTo8ePWzq3HnnnSV+yLto0SIAnnvuuQLn8iipa2Xf8rr55ptp2rSpzfLBgwfj5ubGoUOH2LNnj836+vXr261XmJ9++onMzExat25tkzyx5/Dhw2zatImKFSvaJE/MzMtXrlxpWWYeRmrevHmFDiNXHGfO5E7UFRAQUGCZwobxMi8zl8krJyeHRYsWMXToUHr06EG7du2Iioqia9eumEwm9u3bR0pKymXFv3z5ciB3yLz83NzcGDx4cKH1//rrL1588UXuuOMOOnToYLnP9+7dC8C2bdsuKz6z3bt3c+TIEdzc3Hj00Udt1oeEhNCnTx/g0j7lZ+++atiwIW5ubpw/f56EhAS79czn9vTp06UNX0RERERERETK2X9efp+pX/7Ad/9bYVl2Lc6Dcv+wsaz+aweO9W2f8+HszqaYwwQ3v5uX353OA8PfACdXDCd3+jX1pmmohu8qL+qBUgpXuhfH9WTfvn2Wt/5dXFyoVKkS7du355lnnqFx48Y25YOCgggKCrJZfuHCBcuE6/befs8rLS2NhIQEgoODLQ9z69SpYzfZ4eDgQO3atTl27Fix9ic5OdlStnXr1sWqU5RrZd/yy9uDJC9PT0+qVq3Kvn372Lt3r02vjYLqFWb37t1A8Y/p9u3bgdzjERUVZbdMWloagNX+Dx06lC+++II33niD2bNn0717d9q1a0enTp2oUqVKiWI2t1/Y/C733HMPQ4cO5bfffuP06dNUqFABgF27drFt2zYqVKhAly5drOqcO3eOnj17snbt2kK3f/bsWTw8PEoUc95tmHuXFHS+ClpuGAbDhg1j6tSphW6joIndS8p8nYeFheHp6Wm3jLn3lblsfjVq1LC7vEKFChw5coQLFy4QGBhos97cM6is5pwRERERERERkauroBdos66ROVBGvfO51e/bYg7gdudjZAMvdPJnwoqzlnXf/76VU2fO8vbHX1G/dgSns3JHV6kR6Hw1Q5Z8lECRyzJz5kyroXOKUtAD0vPnL3WxW716dZHtmB94XrhwAcDy4Nqe4ODgYseXlJRk+ezr61vseoW5VvYtv4oVKxba7r59+0hOtu3uWNA5LIz5uPr5+RWrvPmYJSUlFXnM8j78bty4MX/88QejR4/m999/Z9q0aUybNg2TyUTXrl354IMPip0AMvdOOHfuXIFl/Pz86NGjB4sWLeL777/nySefBC71Prn33ntxcrL+mh05ciRr164lMjKSt99+m9atWxMUFISLS24XzdDQUI4dO0ZmZumHCTRfO1Dw9VPQtfPll18ydepUPD09effdd+natSshISGWZEO/fv34+uuvLys+e7EWdT0Cdq9HKPiadHDI7WRZ0D+mzEkge0ldEREREREREbn2vT/9O+sFXsEQeStJqddGD5RxU7/O/WByBCMbPAJIc/Hn0Va+jO9ZwSqBMm3uMsvn5g0j2ZmSO6JGzUCXqxqzWNMQXnJNyDtfRUZGBoZhFPpjnrfDXK+wIXgKmufBnrxDjuVNfFyOa2Xf8itOu/aGYCsNczuFJSPyMu9727Ztizxe5t49Zq1bt2bZsmWcPXuWn3/+mRdeeIHQ0FCWL19O165dix2D+YF+UT0tHnjgAcB6GK9vv/3Wap1ZVlYW332X+4d90aJF9O7dmypVqliSJ1lZWZw4caJY8RUm7zVX0Hku6Nr5+uvcP+zvvfceTzzxBDVr1rSaw+XIkfyDdF4ec6yFXcsnT54Eyu56NDOf28KSlCIiIiIiIiJy7fr2x9+tFzR9EKq3Y/6B8k86nDz9zzOlCpFwzydQ+SbwyH1ht15wbnzLHgmhmr9jbjmn3GG6qlWtTICfDwREANCiqobvKk9KoMg1wdfX1zLE0s6dO4tdr3bt2gDs2bPH7lvmOTk5dufxKIiPjw+hoaEArFu3rtj1CnOt7Ft+5mG18ktJSSEuLs4qhstlHoKpuMe0Xr16QG6MOTk5pdqml5cX3bp1Y/z48cTExFCjRg2OHTvG0qVLi1W/bt26uLi4cOzYMaueSfndcccdeHl5sXr1auLi4tiwYQP79+8nLCyMtm3bWpU9ffo0Fy9eJCAggMjISJu2duzYQXYZdDH18/OzJIBiYmLslino/JsTUm3atLFZl5mZWWC90s4XZL7G4uLirHrO5GW+b8rqejTbtWsXQInn9BERERERERGRa0PjejWhRgfo9Q7c8hIE5448cjHz0rO0vQePsP/Q0ase2+79hwGo16l37oKooRBUE4DK3rkjltwa6ckrXf4Zdty7InhVJNOzEplZWeAfDkC4vwaRKk9KoMg1o3fv3C+TDz74oNh1oqKi8PDw4NChQyxbtsxm/eLFi0s8R8hdd90F5L6BX1xFzaVwrexbXmvWrGHLli02y2fMmEFaWhrh4eF2H/KXRs+ePXF2dmbdunXFGsasVq1aNGjQgMTERGbPnn3Z2/fw8KBhw4YAxMfHF6uOm5sbzZs3xzAMNm3aVGA5d3d37rrrLgzD4Ntvv7X0ROnbt69NUsF8nSQlJdm9Vt55551ixVYcXbt2BeDTTz+1WZeens6MGTPs1jPHaO71kdfMmTML7NFS2vlE6tatS1hYGGlpaUyfPt1mfXx8PPPnzwegW7duJWq7KBs3bgSgXbt2ZdquiIiIiIiIiFwd6RmZ0PQhcPeDwGqW5W6m3BdUMzIyibylP7U69rvqsSWeTwYnN3YZl+Jyr5U712+TkEtz7jr9MwQ5dXpAjzc52vAJPvpxC1TMfS7n5qxH+OVJR1+uGS+88AIBAQF88cUXjBw50maopcTERGbMmMGbb75pWebj48Ojjz4KwJNPPmn1dvy2bdsYPnw4zs4lm2jpueeew9fXl19++YUhQ4Zw9uylsQhzcnL46aefWLJkiVWd6tWrA7By5cpret/ycnJyYtCgQRw+fNiybNWqVbz22msAPPvss6XuVZBf5cqVGTZsGJCbTFq+fLnV+vj4eF5//XWrZRMmTMBkMjF06FCmT59OVpb12JUHDx7krbfeYsGCBZZlTzzxBHPnziUlJcWq7B9//MFvv/0GlKy3wa233grkHpfCPPjgg0Du8FfmIbrMy/Ly8/Ojfv36ZGVlMWLECDIyMgDIzs5mwoQJzJ071zKc1+UaMWIEDg4OfPfdd3z66aeWXkwXL15k8ODBBQ5NFhWV+4f8lVdesUqW/Pzzzzz33HO4udnvNmq+B9asWWNzrgpjMpl47rnnABg9erTlPEFuEqdv375kZGTQunVrOnXqVOx2i7J//35OnjxJnTp1qFq1apm1KyIiIiIiIiJXT/xp+6OGZGbnPgf5acWl0VDS0tKvSkxmCWeT4JYXrJaluuWOGFK7wqXnP30a/TMUu2fgpYLela54fFI8SqDINSM0NJTFixcTFBTEpEmTqFixIo0aNaJ169bUqFGDoKAghgwZwo4dO6zqvfnmmzRr1ozY2Fjq169Po0aNaNiwIY0bN6ZChQr06dOnRHGEhYUxb948vL29mTFjBsHBwTRu3JhGjRrh4+PDbbfdxl9//WVV5/777wegV69eNG3alI4dO9KxY0fLfBbXyr7l9fjjj5OYmEjNmjVp0qQJderUoV27dpw9e5bbb7/dMiF6WRk3bhx33nknp06dolu3boSEhNCyZUuqVq1KaGgoo0ePtirfs2dPpkyZQnp6Oo8++igBAQE0b96cFi1aUKlSJWrUqMErr7xiNXfG2rVr6du3L76+vtSrV49WrVoRERFBhw4dSE5Opl+/fiV6CP/www/j4ODA3LlzCy3XtWtXKlSowLZt24iPj6du3brcdNNNBR4Hk8nEtGnTqFy5smV/XnzxRV5++WUqV65c7PgK06xZM958800Mw+CJJ54gNDSUFi1aULlyZebPn29JlOX3/PPPExAQwPr16wkPD6dJkyZUq1aNHj160KxZswKvuVtvvRV/f39WrVpFWFgYUVFRdOzYkfHjxxcZ69ChQ3nwwQdJSkqiS5cu1KpVi2bNmhEWFsaff/5JWFiYZW6WsmI+p4MHDy7TdkVERERERETk6rhwMYWN8Rl212Vk5w4J//vazZZlF1PTrkpcZsdOnAbfkCLLebo4UCPQGUx5HtVn5SZ7+jXxvFLhSTEpgSLXlLZt27Jr1y5efvll6tWrR2xsLNu2bcPBwYHu3bszdepUJk+ebFXHy8uL6OhoXnjhBcLCwtizZw/JycmMGDGClStX4urqWsDWCtalSxd27NjBsGHDCA8PJyYmhiNHjlCjRg2ee+45+vfvb1X+xRdfZPTo0dSsWZNdu3axcuVKVq5cSVrapS/ma2XfzIKCgtiwYQMDBgzg5MmTxMbGEhkZyYQJE1iwYAEODmX79eDq6srChQv5+uuv6dy5M2lpaWzduhUHBwd69uxpd6iuoUOHsmXLFh555BEqVKjAzp072bdvH0FBQTzwwAN8//33DBgwwFJ+0qRJPPXUUzRq1IgzZ85Yhijr1q0bixcvLvFwYGFhYXTr1o0dO3awbdu2Ass5OTlx7733Wn631/vE7Pbbb2fp0qW0adOG1NRU9uzZQ82aNfnqq69seuFcrpdeeol58+bRqlUrzp49y4EDB2jXrh2rVq2y9DTJLywsjLVr19K7d29cXFyIiYnBzc2NsWPH8vPPP+PkZH/cTR8fH5YvX06PHj1IT09n7dq1rFy5ssA5WPIymUx89dVXzJ49m3bt2nHq1Cl27txJeHg4zz33HJs2bbL0cCkrc+bMwdnZmYEDB5ZpuyIiIiIiIiJydfy2ehNJadbzBg+IzH0Wl/nPFLNTZi2AwOpQpzupV7kHyqbdsZbPKW/XtHy+x9zjJI8DCZnWCyrUAqBHHSVQypvJsDc79Q0oNDSUo0eLniwoOzubvXv3Urt2bRwdHa9CZCJX15gxYxg7diyjR49mzJgx5R3ONW/16tVERUUxZMgQu3N0yPVnxYoV3HLLLTz55JN8/PHHJaqrvxEiIiIiIiIi14ZPv1rEE7O2QpO+hPk58dItAfhejOXBZZ50DkrgpxHNca19K9w7DYDf+ntwS6PQqxZfpQ5DONnyBZ5u58ekOyoyYvEp1selseI/obg6Wb+4bHpur902FgyozN0Nva9GuP9aReUN1ANFRKQQbdu25a677mL27NlW88XI9ev111/Hy8urwGHMREREREREROTad+J0IvjkzhXy++Oh/OdmP9ycc192PHz8NN0HPg8V61jK7z1tf7ivKyH+5BlOJucOIxbslTuax6Q7KrJmWJhN8gQg/tXqzLo/mJc7B1gtd3Uqm/mJpfSUQBERKcLEiRMZNWoUR44cKe9Q5DIlJSXRsWNHZs+eTXBwcHmHIyIiIiIiIiKldOTUeYhoS1VfB8L9nQHwcsudnH1/3AlWrN0MTfpays/+4fcrEsc7n86h9V1PkHegp7+27QH/cAAaVnYpqKpFZR8nBjb35XxajtVye8kWubrsD2YvIiIWNWrU0HBnNwgfHx9Gjx5d3mGIiIiIiIiIyGWKTcwEH2cGNPPFyTG3p0bFQB/gApj+GXbbp7Kl/Note9h78Ai1q1ctsxgMw+CF8blDhB0/lUCV4CDgn94xgbnzubYOcy92ezdVtp7vWD1Qyp9SWCIiIiIiIiIiIiJyXTmZmptciAi41MOjUqBP7ofq7eDez6wreFZgw9bdZRrDzr2x4OQGEW3Zfyjesjz5YioE1SLM2yDQs/hzqA5p6cPT7fwsv19Izym4sFwVSqCI/MuMGTMGwzDUo0JEREREAEhKvsj9Q8ey96CGKxUREZHrR3JObs+OML9Lgyz5e3sUXKHhXfx1tGznQTl6/DR0GAktBjJvZ6pleVySAZ6B3BxassfvJpOJ5ztemgelcRXXQkrL1aAEioiIiIiIyL/Y9z9F893/VhB5S3+27NzHsROnyzskERERuY4ZhsHYD2axct0WAFZt3Mb6zbvKfDspptwESmieBIqzY+FDXm05712mMRw+dhICIgCITbiUnDmbkgVA2D9zs5REZR8n+jT04q3ugVTy0Qwc5U0JFBERERERkX8xDzdXavUMo1KTQJr0epTQ1veyY8/B8g5LRERErlPxJ88w5oNZdOz7NIZh0O7e4bS++8ky306GkTs0lo/rpUfcJpNtAuWpKD/L5zMZLhw4fAxTREcW/vznZcew6/BJy+fk1EzL531HEwAI9Cp6Anl75g2owqjOgZcXnJQJJVBERERERET+xRIvJlOpcSC1eoRRq0cYoa0r8p/R7zN/6cryDk1ERESuQ6fOnINGfaDJAySeSwLvyhDWmuzs7DLbRuK5JEvCwtOl4EfcVf2cuLfRpV4nmTnw/f+iwacKvYdPuOw4/oo9Z/l8Oi03eXPydCJb9ufOhxIS4HXZ25DypQRKPvaylCIiInnpb4WIiNxIzlw4b/lcqXEg1W4J4WzVFO77vzHsP3S0HCMTERGR69HpxHMQ2Q1qduLImYvQfSy0GoxTvbvsvqBhGAZ/btiGYRjF3saCn/+AgOoAeLjY/3/0Ac18WD20KjeHu/FE89yhtLJywMnZBbqNyf25DOnpGayJuTT06dFzGZgiOlKpRW/SKtwEgL+HHr9f73QG8zGZTDg4OJCenl7eoYiIyDUmPT0dBwcHJVBEROSGcjwx0WZZYC1f2j7fmO7/eb4cIhIREZHr2cGjl4a12nPs7KUVtbpwz8gPbMp/PmcJ7e8bzvvTvytW+9t2H2DXvjjwCwXAJd+8J38/Fcb/tfVjxn3BVPVzxsHBxEvtPCEjhfMX05j146rcgm4+nEvJzN98sZ08cxbq3275PSndBFVbQodnoGozALrUKmRSe7kuKIGSj8lkwtfXl8TExBJlPUVE5MZmGAaJiYn4+voqgSIiIjeMhLPn+W75igLXV+kdbLPsYkoqre96gv4j3rqSoYmIiMh16q/9l3pl7D12qacrdXtArwmM+nCuVfn12w9Az3F8vOY8xdGmz1Amfb3M8nv+/0dvGurGh3dVxNHh0nIXZyfIySLh/EV2nnOzLP9i2dZibdOeUwlnwcUTAIcL/0wm3/oRqBhpKePqpMfv1zun8g7gWhQYGEhcXBxxcXH4+fnh7u6Oo6NjeYclIiLlIDs7m9TUVM6dO0dWVhbBwbYPkkRERK4UwzCIP3kGZycnJn4+l+cf70tQgF+ZtR+9bgt1+1Qvdnm32l1Jz8ikUpNAkhzSOHXmLBWD/MssHhEREbm+JV9I4b/LtkGbtgAcOJUMBFmVGbcmh7eHX/r9TI4XeAYS69mxWNu4GBIFje8rUVwuLs5gZEOF2rk//5gTY+KpErV0yWdzloBDV+r7pHDg/EXSStmOXNuUQLHD2dmZiIgIEhISSExMJC1Nl7+IyL+Zm5sbnp6eBAUF4eCgt0dEROTqca7ZGQe33L89Fer6cfidk8wdP7rM2l+zZSeOgbntf3XfSwxbPIXzaRcoqC9+xeYBRHSsYvl9ye51DG7Xo8ziERERkevXtt0H2LRjL9z8uGVZXIKd56pVm3E2JRt/j9wX1s+kmsC1BBuK7Gb5+H53z2JVcXZyhBzbSezPpJX+//HXx6VBFXcaVfPkwIEDNuun3FWh1G3LtUMJlAI4OjpSsWJFIPetL/OPiIj8e5hMJsuPiIhIeajeLYRKjS+9tXmcc2Xa/u6ThyEQbq/TmjC/iiwe8AY/7l7Lu39eGoM8OycHRwcHDMOwSp4ALPjjTyVQRERESuh0wjkC/X2u6xf0jsSfolbHh5j70WjuvDWK0wnnuKnP8+DgCD3etJTbeOA0hNa1qX8yKQN/D/fczxcpWQIl4QCENgWgVrBXsaq4ODuDR4DNcm+H0r84f9w5DID+zf1Y8L8LNuufvNmv1G3LtUMJlGLQwzMREREREUlJTSM1LZ1Af9+rsr3s7Gyr5MmVcCYjCTfciIpoaFl2e92bub3uzXR8cwQ5FeF82kUCPLxJTUsnJzsHB8dLD3uSKqRy6ORxIoIrX9E4RUREbhR/bthGz+Ev8FS/Prw57JHyDqfUvv8pmnSvUO566n2M3VGcOXsebhtnU+6CXx279ePPpVGnUm4CJTG9hIkkh0uPtKv4FS/z4uzsBKZL27m7tgML9+aQamcOecMw2LprP1WrVCz0332p2bk9aFpUdSM7I92yvHcDL6b2roiDg54n3wiu3zSniIiIiIjIVdTmwWE0feQxki5evKLbycnJYcqsBfy8cgPJ8Vd2WynOuf+zXzOwis06x38eZpxPu8B3S1ZQ+dY+pCam25Tr9uaLjHz/4ysap4iIyI3i0bETaTK4Dn+47CzvUC5Ldo4BnV+CXuM5fSGL7Gzr4bG6VEjI/eBV0W79sym55Q3DICmn+N1Pzp5PBqdL5X3civd4O+/L8VPvrsikHl6QmUqa7ahezJ6/jCajlhPU571C28z4p2+Cr5sDWTk5ANTwSmf+wCoEe6vfwo1CCRQREREREZEiLPz5T2jkQHi3KkTv2QJAWlo6z749lV9X/VWm2/p28e9M2bCQEfM+wcXb2WooYc9st1K1ee58Mhu27Gb95l2WZSmpaTj5OuGQZSLIw/btSod//ncxIyuLoR9P5qb+tfGs4I6b4UL3Ks0t5YIbBvCX1/5SxSUi1nJyckg4e574k2f4Yt7PGkpc5AbkU694Q05d61KzLz1W/nnTYeb9tNJqfUSQh9XvFTwdrX5Py8wC4HzSBbJdfCzL48+mFrrdH39fDxUjLb/7FjOBktejrXzx9/WG7AxSMnJs1n+6bDfU7grN+hX4Pbx1134yHDxwyMnE2dGEeQI5J/U6ueEogSIiIiIiIlKInJwcnv7qY3xCcicp/WbpbwR1vZMvFi4jmm289Pt/y3R7i7espVLjICo1DsTV2wWP9EtvWTrnOBZS0759sUcJ6XIvj3z/Hk8smcyZ5PMAHD1+Gjd/FzwNN7tDFjv8s2zFus3UvbuaZXkNn8p0DWta4jhEpGgfzJpHy+eepPO4Z/lv4jJe/2Z2eYckImUsO+fSA/uLGaWff6O8Xci49G+HAW99z9jd1azWhwdbv5wR/2p1MsfX4maXvQCkZ+Z2/Th09AR4BFrKPf7Z+kK3e+aidcKjuD1QAF7tEsA9jbxwcjTh4+2JM1mcyXRlx/4jVuXc8vTMfX/xNrtt3fzwW1AxkhwHZwD63NoagKoVvYsdj1wflEAREREREREpwKEjx3nzoy8Jb39pjo8Djsepd291Ptv/Ex5B7rhXcuP0hXNlts3EjCSr32tWqILTP2955uTYviVZlI1bY2j2aF38wr3xCvZg9rrlucu3x+Di6Yyvq6fdeuYEyurNO6yWt6/XiIa1q9uUzzFKHpuIWPt+40pCWlakQj1/AH67uMVmWBwRuX5lZ2eTbsqw/L7/1LFyiSMrK4unXp/K4t83lrqN5Kw8L1806g2+1sOBhgZY90BxcjTh5Ggi8J/hQ1Mzcr/b9h85CR7+8M+/I5acsR1WNK/j53OTTiFeBk/c7IurU/Efb7/eLYjv+19q/8GGrhiuPjQcv4cH315oWb73eLLl80urbYcXS76QQmrlllbLZgxuyKDmPnzRN6TY8cj1QQkUERERERGRAjzz8Sf84rLZapl7QO4wWp7B7pZlu48eLrNtJqWnAJCTkvsgIcMhi09ufxrDMMguRZIiJcf67db56/4kJyeH52ZPA8Dfw/5QIg7/TLS6x+PSw50Iv2B614/Czc2Vx5r3tCp/McN2fpRr3aEjx7l71KscOnGivEMRAXLv9/xmrFhaDpGIyJWwN/YobgEult/3xR8tlzj+2LCdD0+34L7vLr208b8V6zF1H4+p79fsiEsoso30zMKHGKzk707zINt/G7g4OvxTPzeBsuVQIpgcCHa6UKzYdxw6DcCgm1yY2ju4WHUKMuWhepCVAf5hzDlbH4DVf20nPqitpUyIS7JNvQdf+hhqdQagRWju+fRxc2Tm/ZWo4qu5T240SqCIiIiIiIgU4GDqCUz/jGUd7hFMVprtw02APceO2F1eUhcupnDRJR1yYPJdQwlw8uaZW+6lko8/mSlZ5JRiPoSU9Ayr3x2DHWn1zlBqdqsKQKopw141SwLFzfefBwPOHsy+70VcnXJ/79e0K24Olx4CJaVf2Qnvr4SWA54gIeICg+e+y+GjJzgSf6rEbTw65l0en/J+oWUuXEwpVe8h+XdJT88gzc32ftwQt6ccohGRsrDol1Ws3XJpsvjV23bgEeSO8U/yYcr8hbR/cjhzfv3tqsZ16FQSuPmS7l3Vsuyh95ZDw95QtQUjfij6xYLMnCISKN7OrH22AfGvVuf0mBqW5c7/JFDSMnP/Lu4+nvvvh4HNcl/o8Cal0HZ3JuQOZxpVp0KRMRbF29MNUhKtln2y+C9w+GfI1JwsMjNtewEuibmUGFo4SD1ObnRKoIiIiIiIiNiRlpZOjtelhwPv3fUfko7aTxLsPVn6ITiSL6Rw4lTum57HTyXiEeiKZ44bzSJq88Pg16lTIQxXF2eMnNL1QEnPyn0g2zawvmWZuRcNQP2QCLv1HPLNi3JLRGObMgsHjMU5PvchQ1Ja4Q88rjWnE85R54Hc8dqz3HOIeu4p7p71Glk5xR8uacXaTeypEs9u1yPEJh63W+ZgXDzNnv0PD334dpnELTeumANxeAa745HtysiWfWjil/vAce/58nlDXUQu32t/fMELG6aTkpY7MfqRhNxEvVdG7t9ho4YJGjvy8b7FpWo/KyuLnJwcHhr9BY9N+bXY9Y4mWv/N3rRjL27+lSy/uzsV/cJGRlbBZUxGNg0ru+LkaKKyjxNBeSaQd3Ey/VM/9+/tgYRMAFrVDIKkeBwM+y+rmF00cl/eaFTZrdByxeZyaaix1MwcjqXlaTclkWM5AWRm5fu3QeCloUz93Us+P51cX5RAERERERERseNUwjlcfJxxzHHg/Z7/oaKXn22hf54/bD66r9TbaTd8OHfMfI1TSeeIP5WAs6cz3k7uVmVcXJwxciCHkidQ0jJzH0y4ObuQlWqbHBja5k679fInULw9PWzKeLq44ZGd+yDjXGru0BvZ2dm88PE0dh6OLXGsV8Ovq/7iydc/YPd+62HXqnWqgkeQO4kptkN1FOTB19+61O6Wvy2f240Yzi1jniEnJ4f73nyd4CaBHPNIKFFyRv49Js74loETxvHG9Nm4+bkS6hXEXY2jeK7dfQAYnjB0xuRyjlLEVmZmFq+9P4OOz4zgr/3qKZVfdnY2vlVze1U8v/BzAL5an5vkqOjhZ1XWwdH+I9pDR48TE2t/mNCsrCycO79C82fm8U3KzXweF4ZRzJ6qJ85d6kFx4WIazXo9xsnzl5a5GPZ7p+ZVWA8UZ7JwdDDZXefimLs8PTOHwa99zlb33MnXa1fyguxMMg0Tf23fxwsz/rC7P+k5uQkLX/eyeazdLDDV8vmbdcdJ/ad30OIBFcHNF4Co8ZfmiklKvgjufpbfPVz0eP1GpzMsIiIiIiJix5mz53H2cMIPL5qHRgLQILSaVZkAF29ysnPI8Mtm9MJZJd7GufPJ+LT2xs3fhfkbV/L/7N11nBT1Gwfwz2zXdRfc0d3dJY0oKIqiIgjmT6VMQARBEBGxUBAUAxUVA1EJ6e7uvDuue297d+b3x9zN3LC7F1zB8bxfL19OfGfmu3tbfJ/5Ps/p61chkzMIMwRI2slkMoDlSj0wUpTVyQ+IqBVK/DZ2Fia0ltYu0ShUng4TUngV6luvjcd2WgVfXDU9LwcAsOqXDdirPIdnN3yEBz94+5b6XJnGvDcPp8KvY/rxrz3uP3rpItJzcko8D8uyiOwmpg85lHABAD+jiGkshzOSRa8vp0DXTgyGTV6ztFx9JzUPx3H4MWUbrgakIa0RH7yrF8qng4kIDRLanXReg8Vx59UZIjXb7I9X4W/bQbANgcmbP6/u7tx28k3iwPwJ41X0ePtlhLcOBgC0jW5Q4vFWqw2DF7+BiZs+9Pj+v3A1EegwDkcVrcRr2kr3nZtudAjLn228AMR2ARr2F7blmIufBQIADlcxARTG+w0faiX/++LUlSR8ZekpbA/WKwD/GJhlPmj/zj68dzYca47muB1vhxIM54JO6TlAU1a73uyK9gZ+ZtCVNBOsBY/LT6/CzPZ8HbkTGXIMn/cPcs127Dx8FgiMrZBrkzsDBVAIIYQQQggh5CaHT1/AoyvmQaGWw1clDoAvf3oK2hnqC+v+Wj1U1/lioVvTj7vNMCgueLD05z9w708zhfXzqYnYduY4AKBxZC33AziARdmDETYnP0iiVigR5heIx9rfg5AEnxKPk8vEfy76OnWoExjhsV2gjj/XjcwMAEBKjphLPNWQgy1nj5a5z5UpuHFAsfvnHliNkWvehquEmiWTF38GfYj42six8undHnxzltdjjuVdLn1HyV3BZLZAFyydcdayNp+6Sy6Xo5EzWti+7vieKu0bISU5n5oAXVBBuiMaYXRz7nK8dEMEA3lB8KBHbHO39kV/M3z9179oOXUCghv6AwDis91rdKWmZ7tty7GWbqZjZpEAyat7NED7sZL9yXkOlMRZzNekvJjYhkrBPwcnbkjTiAXoZEDhzRu1OgIALmdYJW04joODUUHBOcAwFRNA0Shl6BfBf4evP3gFx9XtAQA6pQxvP9IeyE2CVemHP7PrYuDHp5GQkQ/I+N9+o5q7z84lNQ99vBFCCCGEEELITca8NxchBQPtQ1t2Erb7qHXoFNJYWA/Q+mDl81OF9V1XTgrLcQ+PRs/lk/Htjk0er7Fy+7+S9SP5l5Bs52uh9G/ezv0AFuCYWwigFKTw0qrUwrYfZs4AALQMq+PxGACQF5mBopErvbaLDubvpv3hwla88s0XSMiRDvK8vetb7L540u24pNQM9J05BZtOHCrFo6g4LnvpBpeSjJnF7j94Q0xX47Q6ke+0wGazw9zUfdCpvV680zjP4rmODrk7pWflum0b2LC9sDxj5GPC8idH/sTYJQvw7baNVdK3inQ9MQUWK82gqWkM/tqSG93FJn3yqcftWqcKgf6+wrojlw9mFK1ztjJpA8JaiLPQLqW411rLNbun2Vqw/lKJ/eI4Dluzwrw3yEtGgsVzfZGVf+3DPXO3wOHiip2BwnhJ3wUAGiUffMhxqiXb1QoZ7tGekbaVSWfC5OTlA0qt2/byignWAwCOu8TfRToV/xiK3E+CfRl6pOTyQZ1H6lvx0+NikJvUXBRAIYQQQgghhJCbWCAOSvRu1Fqyr0vrZsKyQa9BbEwEcq/wqXe+3PmPsC+mD1+Mdfm5v93Ob7XaYPfn//GvcIjFR5lYfjlQ6+t2DOcEOBkwfOZ0PLfyQ4/9zjeZcejCeXzz5wa0e/MZnL8RD7uzsAaKGARRKZX498n5WDzsOY/nAQrShhXQKNRe28WFiTNT9lnP4XRAglubtcd3uW0bN+c9OKJZzNn3vddzVwaXXXrLrMNLmpKcEmqh5LNiahab0QELa8fGPWIw6KMBzwMANC4lHmwipigZ+u10vPDFR2XuN6mZsnLy4LCIr0GNQym5qzomLBSPB/YT1q9oU7Ds/N9wuCp28LAynb8cj6GfTseoT2ZXd1dIBZMpxO8JxlkxswFqkgxZnsftYzsOgJ+PXlh3ZRcEUFgX8owmjJz3ltsxWSb376Rci3sAZdflkut4XUtMAfy9D/wzdiNsnPuNEyazBeO3B2JzTjR2XDLCUexETe+vB38DH3gzgv//kiF+ODctFgDQOkgaaHU6pJ91aRnZgFILfSmK3JdFt6buz4e2YLbQpPbS4fOLqfyNEME+nlOgkpqHAiiEEEIIIYQQUsTAqa+gVtdwYT1AK013FREo3hH6YJteAIBBtToAAOIdaTifyAcQzJn8HYpqp/sgxGe//AG/Wgao7Ur899xCt/16lYeAhQuAAsiONuGU03NB2bHvL8DkbZ/jo9O/Q1dbizf+XYl0E3+He6ifNHWVTqmGQib3dBoA0iLyeqX3AEpkQLDXfbH2UACAEgq3fcZGVrdtVcFlE2eg9I9sKxm8Liop2/sMlMTkNPg14V8XUzs/CH2IFk61C59u+QMA0Du0JVrVroftEz7AxmffQ6M6MZLjTzBXy/swSA1htkgHC/t5qDV0f99uknWGYZBh9jwwezvacvAoAuJ8kOtrvu1qIpHycbjEz1OWYeFiSzfDr6bLzjXinY+/QXh76ffjpLYj8esjb2F02z7w8zVgWEQnPNG0P5iC4VkH68Lv23YiPdj9/Z1lNOLtb1chIUuc5Zlrdp/xWExdd8HhU5cA1oVAucVtX11/DkrOAQfj/tvlq7/2C8sJmSY4C2agzLnHHw9EpUvacsUEUMIC+eARp+d/I3Sp64uGoXwwYsZT90ravrrFgbNJ+cJ6rtEEKLUwKCv2s6RZg9qALV+yzVAwA2XeY53RWXMNjI0PTm0oSN3aJNq/QvtAbl8UQCGEEEIIIYSQArl5+cgOl6ZYKloLpHB9x8TF2DZhEVpE8aketBDvQpzw9wfIt5qFmQ0uhwuPvPcOVvzHz0S5lpCM3/L5WgYKmQwMw8B5U85ylYeUWYwLYBTigMS2s8ck+1f9/S/SIvhBl8Kc9OmuXGSE8//gb127PspCzojBldpB3lN9dGnTFDnXPd/x2j2az/FuskmDJU6nE3KV9+DNrUjJzESWseRB5aJ/z+lDxwj56G+WnJXlcfs36zdg6OfThfU+DcUZSuYY/m7gAF8+uFI4k8DfzwePxvQpsW/k7mMyW6BQ8+8Fzs7h0Y593doEBfhh+4QPsP6JuQhN52en/bR3a5X2szysLvEu+W6fvIT4rNRq7A2pSEVnQjFyBmO/eq8ae3P7eOjlt7E2Z7fb9vvbdkOIwV9YnzbsIYzvOgiygmDDo8vn4esd0hR9veUtAQC/XtuJ/yzH8Ogv78Lq5N9TeRY+gGJwiakAXcWX7wIAbD+TAsjkGNpQjcmtpEGUV/sEQ8W4wDFy2G4qcnI5VQwwvPpXmlADpU20Bi93D5S0tbLeh5ybRRqAIunK6gWLv6EMei1w45ik/fM/i7VkbHYHoNBAU7E/IcAwDJQX/5FsC9DyF1GplNgzpz/qOy8AADK1cWA4Fk91LSYNGqlRKIBCCCGEEEIIIQWyc42SrBMvdbzfa1tZkRohT9w7ULJvyNw34RdjAAA4tSwS/TPx9aWNmLF4Bbq/8rKQG9yk4O8+f6/7BMnxngqjGvOlxVZn7lyFnzZuQff5L+FEwhV8cvJP7w+MBcINxRdPv5m8SP7yAS081GQpoFarMO+e8W7b5/Yfh0ADH0iwOKR32f/5n7QY9oZTB8vUN8l1vvoWz33+IQZ/+gbu+8E97cnNAgP4Aeh+tfg7/UcVzCKCC3ikVh/Ys/kBqYtpiR6Pf/fPH+AbJaZe0SnVqGsMl7S5r2VXt+P6NZHOLCipSD25O+SZzWBkDBpoorBp4gJE+Yd4bMcwDHzUOkwZOgoOixO/nt+J0/HXqrazt8jmFO+Sl6vlWLHnn2JakzuJg5XO4Lvuci90fjeKl6XBEFb64uKyguHZbGU+nPWlMysC9Hq39jsunAAAfL/1HABgSCMt/n0QgMtRuhko8fyNIj0bBWPRoy2BlNMAAL3Mgac6BcJPw3///3NKOhMz1Sh+l6c5tEIARaOUoUubRpK2jQO8z0aKCPEXi8UD8NdKoyE+MmlQx2ITX2dWuwOQyaGu4AAKAAQapLNtFXLpbzGdrOCzTKkBx8gkv5NIzUYBFEIIIYQQQggpkGs0Qa6SI9Dlgx0TF2Nkyx6lOi42OhxLe70orHMx7m0YhsGPF7YhrneksE3h4kcAurVuDruJ/4d5XZ8I94MBaIPcC7q+++tqMIEyvPDPx9AGeE+zpYbSY1CmOEUDRKE+xQdfQgP8JOvPth2G7rHNEWjggxU3B1B2XpAWlZ+7Z3WZ+lbUP5ZDOCW7DkM4P1hlc7rnhN+09yCGLXgDfx/YB1Vd/k7XyX0eAAC83Gsk1o6ZhR3PLsYzA4ehl74F38eMU8gx5+O/g4fx3fZNwrkUOnHUZvnwyZAxMsT4ioPea0bPQGyQNKACAHVrR4E9Jg4CHYm/cMuPmdQcuRZ+IFOnVEPjKXXfTTq1aIK8BBMYNYNn/11S2d2rEGa79P1vt9859VtI8RwFKbtGNixIM+eeUequZLmpNsniQc9i+X2TvLa32dy/twLsBiy652nUCnGf5fDOru9hslhwLnQQAMDXoEO4QQZwLNhSpMm7nMP/v31sQYpSMx8o8VfYwTAMhjXmgzZ/HE4RH5PVhh8SpN9tx5nGAAC1Qg6GYXD+lVikv1UHnw2QYetLTbxeP8hfrPN2fz33QMve2f0xLEIM3uRb7Thy8gJW/3sAJiv/+aF0zwxabo/2aiwsB8vz3fa7LDkVf1FyR6AACiGEEEIIIaRGO38lATO//gosV/Id/4mp6VCo5dAqyl4YtEn9WGRfLj6FlMZfHCBVWGVYMlQs4j6u3gDobWosGv6Mx2M7aRtK1uWsDEpZ6UYQnms/rFTtirLZxZGwMJ/AYloC7Zo3QrQiGM0D4zCu5UA83KY3AKBWRChYF4cELgPv/Pmd0D7Pyg8aa0xiqrJbKYp95NwFSRFjAEgz5bq1e/HbT5EbYMH8Yz8J2/RKPiAll8kQrBMDQIM7dxSW7/1uBt4++h2Wnf8bxxMvAwBkaj4Q9dXwqWgYxkfKHrunv3DMzTVzivp93hxh+cwNz3VsyN2lMICiV2lL1Z5hGOTdENMMFp3dcbtKzs6QrFs9DBaTO5OzoAZKt7rNYbxqAqfgqM4NAFWR72aNTIVWUfXQMLSW1/ZGk3SGqZKT448X5qB9XCNE+Ivfv44E8f3+zPKPhOXpg2Igl8sAjoOrhKff6XQiw2UAw7FoGML/1tGDv77RxX8nt47hb0hIyRPfqx0mfgIYCm4WsPOfQS4lH2hRF6QXbRCiQrBBgWf71UOAzvsUEY1G/C205qlGbvub1onAH5M6CeunjL5o+84BPPqfP24Y+d9yannFz/54/7nB2PSEL/Lm1EPKvNZu+3s1jfRwFLkbUACFEEIIIYQQUmM5nU5M2PwBttlPYPGmX7y2e2bhB+jw3nNYcHENACDaz3thdG8YhsHYlv2LbaNUigMKI5t1R/OYOsL6xGHD8M//5iNQ5+vpUDwzRFpYFQ7AqJSmueCcHIYGdxDWfx8zGzsmLsb9bbuX9mEIbFZxoEYlLz5QI5PJsHrcm/j0gRcxtuMAYbZLbHQ4uIKE7BtTDgvtC2uivHzPCJiu84/BZC97Ufknv1vots3qkA7O5uQaoY1wv7Pf24ycHm1bwppjc9t+6Mp5fmBQwYBxMKgbFiXsq1dbXFYr3OvXFAr290cHRwMAwE8nt3ltR+4eR69dAgDEBHlO3eXJ0mdeEpZvnt11u9lx6DjO+ydJtp3LjvfSmtxpnBwfQFErVZCDr+llc93+Qb3K1qpRPWF541ML3Gqp3UzlI/3eUBSpQdapTVO0Yurgf62GY8Or84Xt5+ViIDXSVwG5XA5wLEqKX12JTwbrG4kguRmqgsBHkIq/gSGvIIDiq+UDKxaHeOPJqVwxyPv3w9KAr4+67MPLKwa4sGSA2i1NViGGYfBy4yIpxGrxv20S8vgHqFZU/JA2wzDo1ywcPhrP6bkWv3Q/3m6Xhxebm7BzYmiFX5/cviiAQgghhBBCCKmxEpLTheVzWQke26z4bT3O+CVIZoe8NmT0LV0v2FecyaCEHG+1HCPZLy/yD/7IgKAynTs6Qhxgtac54FKzCG0mnRnSM7wFXhoyUlj317rnTi+tkgZ8SnUOuRxNHbWFdXNBkMTs5Ad9g3z8EKjkZ2xkGHPR++VJmPbV56U6N8dx8KtlcNt+cwqvoW+84ZaLvpau+IGPMfXdC3lvPnkERy5ehEwpEwr+FmIYBh/1eQ6fDHyhxH43CI8GAOTLyh4wIjUDx3FoNn4cei2ajHRbDgCge6PmpT5+aM8usFznXz9O1nudgdvBso3r3bYV1n4id77C159SJoeC4wf9b/egXlVwFMzMmdr1wVK1f2kA/71tSuFvKHAVmTHLMAw+mvA/PNihF3wNBsTv5tNq2Zz893s9HxsUcqbgO7vkGShHzicA+mDU9RcbNgkvDIjw320+Wj6QYnWKbXwKZsL8+UQYerWOA3LE31QGddlng4zr1xgv9qtdbJvF4zoDaeck27LMfLBHo6j6+iNyuRwzH2qHJY+3Rrf6/lV+fVJ9KIBCCCGEEEIIqbHOXxXvdC6c9VDU8ws+xKr0zZJtj9brgyC9n1vb0ggpEkCZ3utR9O3YFq0hzjIpOu4eF+q51ok3gf6+GODbFt2DmqFXaAuPbWqFhEKtVqGFPA6dghtL6piUlUbpfSZFWSybNAXyeP6BJ+Xxd5OaCwbY/HR6mIz8gNFjn78LVxNgv+M8LmeJd6xnZuei4yvPYewXC6TnXbPO4/VMVv58uXn56PD8s5C1dJ8907iYVCoA8OL9I2E9Ln293FBmYtK2pZArZVBw7s9rq3r10aJW3WLPCwDj7xsMp5UfXLuccqPE9qRmMOab8fTiRdh0/BD2HD6FwI5+YH04WH35u/VLm8KrkJrl35+3ewDlXKbn2Sb5dunsuczsXBw9ewE30sWgt8PhRMzDo3D/ghmV2seaYPXGzTiXWPUzewpTyKnkCiFtlZkCKEJKyloBpZul8FS/wfjz8Tl4uF4vAED/2LZe2373vzcAAPaCAMqUnvzNGHwKL7bEIvIXkvlUo7GB4nf8tzMfQT32CpYN5T+HfHX8DSU2JzB27o+o/7+1MMr9oGRtGNbMD1qNGjKbmC7zVmaglFYzX5NkPSOff27V1RBAIXcvCqAQQgghhBBCaqyT168KyxlmaX0SjuOwLeOEsO7L6LBj4mI83afs9UIKBfmIARR/PT87YsnE/6EH0wwAoPTnByzknAzNIuLKfP43Hx6DuSPH47HeYqowHaNGe9QHYweGt+oKAPhk/It4b8TEW34cADBiYI9yHV9UhIG/c3Xc2vfx5poVyHPy+daDdX5ILJglpAwXB3NSjdnC8vqd+6Cup8YVJkWYwQIAh655LsKeb+XbjH13ATQtNcL2wAxxNs7UPqNK7PPbY8aCsQDPNBwq2a4L1kDOeM/tXhK5XA6Fhj9+2vplt3wecmdZvvYvnNUnYs7+77Hr7Clxhx8/CKhXabwc6Zm8IDjquI0DKKnpWXAaWMicRQY6c/jR3WyzUdK24+Tn8dLOpRj92zwhKJSelYO4PlHIDHAv5kxEZy9dx+fX1mPi34ur/No2lp/xp1dpoZLxn+FmOwVQHCw/yK8qJqXjzfw1Bkx75GGsHf0Wpvb3/h3VrV1zIIUTAigNI/m0n3IZH0ApqQLNt5uOAwAi/MRZt8EBfri4aCAm9OTrehm0aoB1wupksSqjCS5pmgFBdeEjF2d4Kp1iYEOvqrzhZcNNwZl1x1IBAA4PNzIQUlno1UYIIYQQQgi5YyUkpeHTH3/3uv9U4jVh2emUDjSmZ+ZAHyre9f3NI6+Wuz8RoWJargCdWEy8MF2XQs0PnH/z0KtQyG59ED46TKzR0jGkERZNfA7bX1iMMJ+AWz7nzbo15lMK9a3rXki1rB5oLwZjduacgqYuP3AToPXBp8+/5NbeUHA3/vXEFCw5+ZuwPSFXvDs9z8EHYaZ2ehAGszj4nF8wA+WiSZzdEa0JRis/MSd9cXVKCg3q3BHbX1qMER27ue0zKytmgNDM0UDj3eJk/BVheb3lgNt+ndK9Tk9xCgMorioOoOw8cALHTl8sVdvktEzogtTwk+kxtfuDqOUXCo2Jf+8V1jwyW6xwOJyI7CymKMy38e/hfJPF/aREguM49J8+TVh/87sVVXp9R0ENFINKA42cr5uRbzUXd8hdoTAIqJaXfSZnsI9/ibNHu4c3g93Fp6aMCeCfd74GCgcXV/zMjEuh/A0YrMJ70FarUQEuB/KtTqBIOz+VmFps4UONAQDD6zg91gupKCEGlWTdGcnPzjmQVjGzZAkpDQqgEEIIIYQQQu5YfV6egp/ytuPzLWI6p5xcI779fQNYlsXhBH6WAuvi4JRJBxqvJaagcIzi08H/Q6Dec/H2sqgVFYYIcwBCZH6o7R8mbA/185e0K8tdqZ74+oizKeJCw8t1Lm/8NQb8PmY2Xut5a/VgiurXoS3YLNZtu1wmw7CuXWBMlg64sSzfdt53q6EPEYNcSVkZwnJyQTqwumGR+PPFd2A5zA+2rj2wCwCgUfODLt0im+HbMa/jqRGDAQD1DWLB99LQaNTwzdGCKfLy0TAq7weUgjmTHzz2V7nXcCE106H84oMO2rIGUApqFFVlCi+O4zBy1lt48JPZ4EqqVA0gL98MuUoOrVyFext3wXcPvQ6tnH+cueZ8ZOca0Wj8E2jz3NOS40wO/v2x6+DJin8QNcyXP65HbK9IYX2n+VQxrSuek3EBHP/61Sr5z8Ws/LwSjqqZir4nnAWBJZXcPYVkRWgWFQe7kw+gRPjy1yhM4VXiWzOPr6HyfE/vaUT1Oi3gcsBkcwEWcUZoiE4MlPxveDtwCxvg92eb3OKjKJ0u9fw9bl/9WPGpOAmpSBRAIYQQQgghhNyxZGH8P2kOpZ4Xtt03dQaWp/2Lt9Z8DXmQHDIXA9sNG6AU85IDwLUbyVDplYhTh6N5dB23c9+qn16eiV+fmgWGEQcaQv2lM0O0ivINwDMMA1sun0qja4Nm5TpXcQJ1PqWarVESXx89dr76IVKOZ3rc39a3vmQ9PjkNm3cdQoolS7I9sUgAJZfj04fUCYyAQiZH04Z8SrSrSAHHcbDLnAALvDPkSchlMkRHhOK/8QuxfPTkMvf/r1fmYfuzi9FV0wQyJ4OVo6aW+RxFDQpuDwCoqy9bHZyKxLIscvMoNVJVMOabEVCHD9DW84t02//z6JllnpFWmEausFh1VcjMzkWj+2IR0ToYFqe9xPZZ+UYwMgZahRgc0iv5u9kzjXlITE5HbK9IBLSXBq8LZ6C89fuqCux9yY6fuQRj/p01e+J08lUwlXj3f3EcDiegYMCwDBiGga4gDV226e77XNm27xhqj3kYr3+8HCazRQygVMD3pyeNatdCVpIMcJiF9Fl8Ci+uxBooctYKGedEgxDvv0OiI0KgcBiRrYoAtOLvl1CfygkIFWfyqG4A534DRsdaZUt7SEh5VP0rnxBCCCGEEEIqSJEYBQ6fPI9/du8H257feDDrAtS+KugZDZRqOWxw4rUVy7Bo4nMAgEwzP7jnr678WQARwUGS9bLebe7JwqFPI9WShfrB0eU+V1VgGAad6jfGNaQBAL4eIaad6VCnIS6kiim3XvzoI0S2C4XNbIMeOsQiFNeQhuRsPqCSZzRB4acA4xCfyzdGPIKnNn4ArVOF42cuQR2ghIpVSFKhKMt5N/C7j08Ax3GS4NitCDT4AEbA5WFQqLKxLAuZTIYZH6/EP7kHobDIsOGNBQjw8T4DKyk1AyGB/lAqaQihrMa8NRdJxkygEb++cNjTCNT6IDEnHa+uWY5HuvS5pdR7ioIZKPaCIt5V4cLVRGE512oqMe1Ydj5f50RXpL5L0UF2hd1z0OiHnf9h1oixiOvtHmyqLInJaXh42TsI8ffDjllLquy65WVyFdSFyuKAwKoNpOQa86FQy6AoqEVhUPOzBXNMpuIOq5E+XrsWsT0jsRtnMOi7N2D3c0EBZaXNQKkXGwUnl4SAIim1ZEINFO/BWLvdAZdcAy2KD4AyDIP6qgychXTGZstaFZcmtLRUKiWibvyDG9FDJNs1SpoTQKoO/fohhBBCCCGE3JHyjCbhztvE5HQ8eXIh/GPFuiMOsxNKrRxamQoalRLpyMNBXETTqeMwrvNAFI4flHc2SGnUrS0dCCzvQD4A9GjcotznqGpfvjAVq49sQf9G7RDpKwaVIgKCgFSxXa1u/MwMRX0+RUl9QzSu5afhr8R9mGgagu//2gR9qFZSLDc6LATmDAsUPnIcvXgJ2kAN6mgrPr1ZeYMnAKBU8H9/ZxXOHgCA7Fwjmo57Ej4ROvhE6mAI55/f+X//iAUPTfR4zOGT5/HY1/PRPrYhVv3vtars7h3PZLYgPkqcNdXatx6CdHygKiYgFKuffvOWz52bZwITKUdqZjZQRXGG64nimzTbYkSET2Cx7bNNfADFUCSAoi547e86ehLbL51AcEN/t+O2ZBzHLAB2kwMqfdXUOdh59CRCmvCDw1anHZoq+F6oCLaCWZUTew3BshN/e2yTlJoBg04LXx89vlz7F06mXsWSZ/9Xrus+/dYibD91HIZ2eigZ/m/qUxBAyTXffTNQgoP9kA6jsK4L4l/zqluogVIaPgYdfn2+FbgiiYXkchkADmwxNVB+3bQPMIRBKyv5u8dfJs7GahdkwZz76qB3PX0xR1QerqRpNYRUMgrXEUIIIYQQQu5I42a+Jwx4ZZhzJcETADDJrJCr5FDKFFAz4iBGUAM//JG5FxYbX8Bbraz8ATqZjP7pBfDpTMZ2GCAJngBAmH/xd7U2LDLLZvnu9biRww9K+8h1wnadVgOXhYVT5sKFGwn8eQ3FD/BWl8IAmout2hkoW/cfRb0BMQhrEQRdsFhbZm/uWeR4Sbvzzd8bEdzQH1fVqR73E+/Gz14oWS8cYK4IV2/wdQxW7dxYYecsibXgMxMA0nNz3Pb/tGkLpn32OQbNfRX5VjNyzfxMBB+N+D5VFwQmDiVdcAueNPYvqGnAATdS0sXgSSXHGfNNZsz6SkwXlmW6c2p42J18AMVfb0BevAlwSvdzHIdesyej/9JXAQCfnlyHo/IrOJlytVzX3ceeQ+iAIOiCNFAVBFB8tfzfOe8uLCLvZDy/SCtrBgoAjGjhh5EtxN898oIZKN6+VcwWKx75VwOoDbBzJacM1MnEF5NMLsfARj5QK6rntwxbippLhFQm+hVPCCGEEEIIqXYcx/H51MvgaNolYVmmcv+njcrAD76pZAqoWPdBjCQrX4+jqu40tmTbSm50l9KqpKmAHBbxtWCQadAgOkZY//PKXvx8fAcA4KEmPSXHyZ0yQAmsu7wfABDm719JPS6fwgBKVRYAB4C0vGyv+xZs/NHj9tR878fUZOmZOej42nNYvP6XWz7HOVWiZL0iAyh972kLAEhQpOPoleIL1FcUGyumC0vNlb4u7HYHPr26DvsV52EKsWPl/g3CQHrhwDoAaAoKjct93Qdw3+0/HuZMK1QOBd78cEWRPZU7eNr6uacRdo8Y1L2WfucECx0s/1mpVanBsAAY6XNlMlsQ3jIIKn8lbmSlwyeC/1skZWfcfKoy8Y0SZyIUzkDx1/Pb8q2Wcp37TpRvs3rcXtbaRuVRtIg8y7KY+eUGXErKEfZv3nsC0PgBAPKh83IW0b1dGgrLVlf11NkpFBcVKll/o0vVPa+EABRAIYQQQgghhNwGNA37o93bz2LcioUlNy7QNC5WWFZovd/lqVIoofKQvTjJXBBAUVZNAGV2ryfQylAHG56cXyXXu5M0qlsLMqs4QKMs8vdUMHLUrSXmKGJkDCLaBAMAYkPCJOdxGvmARGAjPk3SgOYdKq3P5aGU84M/rioOoKTn5Xrdtzv9NKweCoNn2YweWtd8Mz5ZAXUdNX67sbvMx166loiG4x6Hb7Q03Y2PtuRBy9J6c8AjwvJLGz+rsPMWx+QQB4lvfi0lp2VK1u12hzCQ7q8X60yxDv7+eJWfdOafmlPAR6+Dy+aCXeXElTpiEKOYjETlxrIsIroGS7ZdSr7hpfXtpzCAoldrwHAMIGfAFblbPyNL/DuN/mWesPzDvi2YuWIlLiQl3Np1iwS5C4ME/jr+72yyew4m1GRmh+fHXBEpH0tLLpfzReTB4JPv/sKc83Fo9YE402jL6RRhuX1Eyf168YFuaOQ4AwDoV796UncV+uvtkWipFT8Tgg1Vk9qPkEIUQCGEEEIIIYRUu+AWfvCrZcAlV1Kpj1EoxTsQFWrvdyOq5UoMaNvebXuWL5+y6ObZD5VlUMeO+OiR/1VIAfmaRqlUYNuLH6ChMcpt37Reo+Bj0GGEbxe3faF+0tRfbaPqS9bjAiu+BkpFUCmrJ4XXvlNnJOv/azkcOafEAMkLqz6CySoOBCYkpcJar2wzw+50B86exZUbSTjud03Ytu/yGXAch6c/+gCfb/qzxHMs/OonhHXiZzQUBgwAwFdbcYOQDWJiMK35KH5FBjy88p0KO7c3Dpf4WsjMl6a5Onv1umQ9PjMN+66fBQAE6MQ0Q3lG9wLjelaDbx5+DUqlwnMwvBLHoK/fcJ9tkpCRVnkXrGCOgjpKOrUGsoInqujMtswcz+nIrjhSsM11Ek/99cEtXddlE1/XyoIASoCB/zubHXffbEtLQfD5mdZDhW319FVUnKiAXCgiD+w+y7+uTYz4mROfKf5ddvyvbqnOeWrRMPw3Lgjv3V+7QvtaVoG+Oux9Q7whwldHv6NI1aIACiGEEEIIIaTa1e0vpmjKMHu/S76o/FLe5XrVnIKR9/QE6+IHfGIgvdtYq7ozigXfDRRyaSBsx8TF6F63BQDg+ZH3w35FOkPCX2OQrH8xbTLMmQWvC7Zq06eUhaqgkLaLq9oAirERP4A2ullvbJuwCA927IXjS5Yj6wz/nrvguoFB37yOHssm4f1/1+Czn/8QjpVVcwqXqrD7yElM3bkMT/ywAEqdOJD/yn/LcebiNZzVJGD11a0lnifHIQYJdBZxoK9dbH1PzW/ZsM6dkX6AT6WV5MwsoXX5OYsE/HLM0plJa/Zul6wfy7uMgOb8TLCiM8VcDulrflKrEfj76XmI8OMDTtoA94FRhmHAVsJ7Jd9kxrBPprttj8+u2gBKZnYuer09Ccu3ri/TcfkmM0wx/HtaIZfzM1AgDXSlZGRVXEeLcNnFII1Sxr9X9Bq+cLqTvbuCrgBgY/nvph4NWgjbZg5+rEr7wKfw4sBxDEyc+++apDw+Bd9vY4KhUZZuOFgul6NP4yDIZdX/+a/VqLH2Pg7Do7IxpmNIdXeH3GUogEIIIYQQQgipdtZ08c7IV35fVqpj0myeazNM7DAEPZXNhXUjx6eRiVbx/+B+uGsf2PLEgXgL6562iFQPZZGCuzO6Pyrdp1Rg1ztLkHdELHYe7iMtEq9QKGCK5//eKq7yiveWl1JR9TNQitYYGt26D2QMPxzAMAyWjHrerf2f8Xuh9REHs2Wumj98MPEL/m58xpd/rPY08bNh4vrFpT6PQiMG7p7pMxQdmIYY12AAGoVX/F3cp5ethMvBwksNaxw4dxaZ+blYv3cv5qz9tlyBiKID84fyLuLd9auF9XNZfCoodbb7+65RWC1h2WKVzk4Y1q5LqdIc3UoR6fVb9+DajWS37S6XC99v3YwrN5IRWJcP8sitMkxqMhIOixOJlvLVBymrtuOeBhsBfHtxc5mO+3PLHmE5UOsDecEQn71IAOOrdf8AAOrKI2BKq5jaJBzHSQJhchl/XY1KDY7lJIG2sjp/JQEJyXdODRoAsNnsyFPwz61BpYUplV8O1ftXaT9kMhnkShXyZT6SmSfWgr9VOl+SCHEh1ZuOqzzu79oQv7/csdqK2ZO7F73iCCGEEEIIIdXOkSsO+FzKT4LdVfIdrJwKYO3SgZo6snA80rIP+jRuLWybe884AMAno/+HKd0ewNAmndBX20rYf2+zzuXsPakoeq1GWG4f28htv0KhgB7Fp+6wO/m7bH1VFVdvoqIJM1CqsAaKyWyBKc0CpV0Of6105k6D2BiPx2QYxdlgXCUX8q5uu46dQFBbf8m20R374MYBfjaCXC8GRQprTPyw9T+cT4p3O5dcKQYEBjbrgPcnPIOxvQZWQq/5QVNTkhk3xxesVhsavvA4pu5YhvtXz8KCk2uwKeMIVh/ecsvXKkwXVeifGwfhZF14bME8KOooAAew4ZUFSNgt1lr48t7JksDo2Pukz8PNs8Q41vPrrKzvlfTMHMw9+SMe/e1dt32Pv/Muvri4Hi//LdaOWTNuBoZ06AhLpg35qJoi6L9s2oYeyyah1qCIMh979so1fLt3EwCgpV8dhPsECkHRTUcPCe1OOvjUalaVA4enf44dExejdmaRu/dv4SMoMztXMkPLyfDfw2qVEqyThZO79c+1kV/MwqPrbp86YT9s3Iwp3y2V1JW5Wc/JL8Mvhv9M9VHrsPGlBfjqvqnQqTRej6kMDMPAZeDTVm5zijNhlm7nU6Nm2/nXR4Tv7Tkzk5DbGQVQCCGEEEIIIdWOYzhwHAe7iR/8vpGb7r0tx2HK0s/gF2sAw0rvXP76qVchY2QI9vcTttUN4gengnV+GN6kKxiGQfdm/OBCkMsHkf7SlF6k+jSuI96h76fxfJfsA116AgDqyD3XN2kYzgcDOkU3ruDeVRyVgi+A67qFu+pvldlqg0wpg4JxHzyLDAvyeEy6/e4JoDz9/YfCssPiBOfg8ETX/jiw0L1A+/nkBLzx8XIsvfgXJvzlPjPFUTDY3zG8UZUMospYBpycQ0pmJpwFQY7zVxIQ1sL977rsyHr0WDYJp5Kuuu0rSWFtDdbIwmHmg9xp+dk4kc+fa0TdbpAxMlxc8S3s+fxnef2waMk5+rRvg7wbfIozJdxnq9z8lrDf4GcBlfW9YrHaoNDIIVfJkZSagdV/bIbT6USn6c8jIYJPd+bU8QP/jdQxCDH4Q6VSgrEDnIIrdsC8ory14RvJuo9LW+pjn1y6EKZY/rkZ0aYbAMBp5/8mH58UU++F+fB1op7o2F/YZjc7xBMx/OO02x2YtOJTbDwpBl+8+WXrDuiCxNe1A/x1VSolWBcnqcFSVoUzgspzjor06aV1OGi+gAsZiV7bqFryQf3aujDIZTIE+fqhbqh7Pa/qcuAKP1s3n1UBHItgHQVQCCmr23dOMyGEEEIIIeSuwckAhmUQlO0Do96KEzeuIi7Q8125Zy5ew0H5RQCAAjK4CgZ2JzQfLLQJ9BeLFnsq2j6oZ0cEBvugVcOKrUlAymdou0749Oif0LBKr2l9Jj3+INoebYD2TRp63L/m9bew/fRxDGjbvjK7Wi5qZUEApQoHCb//YxNUBqXHAIpcLkf6sWyEtArAM02H4qeNW5EdZcINP7F+Qk0OoHAcB20AXzOgiboWPp84SdypA7qjKbZnnIRd4YDGX41n/1wC13Un0IC/J/VSciLqRYiBAlvBLKhhTatmdpuMk4FhGIz69R2oLHLEGSLQJ7plscc899dH2DGx9GnJAPDBGQXwYMue+Hf7fph0djz841xo/FSQ22V4eeBIAPxMsc8GvwhfP53H97HdyD8/Df2j3fZZc2zQBWkQqwzDvOHj8cD7swAUn+7u0z9/R73akRjQUiwybbWJ6df6TpqCkF6B+Pz9v6Cq5V4bYmirTsKyjJMBDIOUzExEBFducJ25+ZbmUma+4jgObIj4fmwYxgeN7TYn3MJ1BedsHCqmUTPlF6kfJmPAcRw+/vk3HHZdwuG9l9C/eTuv1zZbrPj9+G4gmv9bafzVyLPzuaFUSgU4lrvlNHGuIjOcrE47DKrSB5Qqi6wgVZRK7nn4dNvBo8Jyw6DqD5rM7ALM3iPd9vdFB95etg4OvzhoOStkt0E9E0LuNDQDhRBCCCGEEFJtElPSkJiWBjAAOODkpSsAgEV7f/Z6zHvf/Cgsq2RKYTlALwZNIkKDhFQwOg8BFADo2LQJ1Aqlx32kegT6+uLXR9/CX8+9U2y7Hq1bQqv2fGe/QafFkPadbtsC8gCgVfOvyYq6y9rlcmHn8RPFDjIv/vdXyJUymGU2j/vXTJqJeV3G4ZGufRHhK60t47S6wNXgMbesnDw4rS7InTJ8/sQkt/1zJz6FXW8sEWZVuGQsXErxuR63bhEyTOJsnXO5fD0QX23VpJGL0IozTexaF867ErH0Ol+QPJwNwOSmI/GwT49yX6fw9aqQyyC3i8NJal8V9Iz0/dimUQNJUKkouZo/1l9ncNv3SvtRaK6IxddjX0F0YIj4XvGS1nHbwaP4KWU75u7/QajRkpGVg+EfzxDahPQqeD0Hii9iRwbf1mDTYHDzjmLfCobJhn4xXVLzpSJtPXAUrWZMREAdX3AsxwfjgFKlvsrKycX4RQuh8Re/1wprQUXoAt3aOzj+3Gq5+F3XvG6cpI2LY7Hr0klh3el0Ii3Lc42xZz9aDHM0H5yKdPDXi9PzswH5FF4cnLcYQDGZxcCO1Vn+2mQulwvjX3kP2/cdK/OxV+KTcObKtWLbsCyLR+bPFdZf6T+6zNepaE+2E38HvdhBBb/0Q8iDAbMuNgRUevgoHMUcTQjxhgIohBBCCCGEkGrTf+GreOT3d8EoAIZj0KduKwCAmvUe2LBpxAEAlUyBkHRfOE0utIgWB4VkMhmmt3oULzQdDrXC/Y5jcvuKDg+Bj/7OLXJbGjqtBqyTdaspcasemfEO3tz/FV79ZRkAvmB8wxcex9AlbwDg62HUG8jfpe6SeR7cbF6/Dro1aw4A8FFL7/y25zn4IGcNdS0xBdpANQyy4tNtsS7xrn95nPSO9KQ8Pi2UMd+M4Eb+AACDumruoO/ToJXXfYObd8B9XbvhgXt6ITfeCDaZf83VloeW+TqFARSVXIGpD4yCwyIGGAzK0qcq07n4wf9mEbFu+0b1641Px70k1PPw9+E/C9bt3SPMbDh46hxGfPAWErLS8N32/4Rj/7twBADw7+6DCGkc4PX6I2p3w57X+Rk4f//vXUmwVVkwQ0sfosWlzKRSP6ayeGr5+/CN0YORMVA45fhz6juwZFlhUdnRffpL6LFsEracPuLx2IFvvIZLvsnCutzGCM9Vz/rirKPCFGSugiInmiLfg3OeGYfxsQNhiecDFjaHHVesYt2aZz9YjAd+mY1fD+1AUloGNh84jPvem4Fnli/GcdMVod3XL7+GiXGD8d4DEwEAKqUSnItDlsWIfUdOl/l5MZrMwnLhLK7yOHbmEtad2osRM2aW6TizxYrB77+OZzYvEbaxHlK6HTx+DvUG8J+r0zo/6HWWSlWqHR2O4MT/wLBOvDU4Cp2jpTcSBKhr7kxCQioTBVAIIYQQQggh1YJlWQQ39AcA+EYbwACYMHwoOI5DoMzH63FFC8z7qw349c23sfvlJagVGCZpN6BTe4zq2qsSek5I+Wg1arjsrHB3OAAkpWZg75nT4DgOL3/yCb7bvqlU50rPzEFybA4A4ED2eQBAakYWwloEIU9rAcdxeP2T5UL7ZkGxJZ6zea06wnILTRw4KwsWHFb8tR77zp8pVb/uJOcT4qHUKRGudb+Dv6hOhkZe981Z9y2crAuXr98QttUNjKywPhZnSMdOXvc1qRULAAgNDsCRt7/AvEHjAUColVIWLmEGigK92rfG8kHibJ17mrUt9XnWvfYOJjYdjNFt+pTY1k/Fz1L58vK/ePrbxcg3mTHuu4XIMOThqTWLcCL9stD26z0bAADH4y97PFehUZ16ek0RqGTEQXCVomIGxK8lpmDIvNfxz7H9MJktiGovBq9cKhYatQraQD4AxdTih+ne27XG47myWaOwHC4PwJ/j5wjrU58cBVseP3Oj8HuyMMWlqshsS4VCgSf6DwDj5J+D9OxcKH3Egfbz/nzg6LsTm9HrrUmYfew7ZPnn4wwXD99o/u/RL7g1/PUGjLnnHmiUfHBGrVJCG6iGxk+FYW+8WebnKd9kEZYrYgaKXCZDo/ti0fTBumU6LiMrF6HNpJ8FLg+zaorWyRnQpIPb/urAMAzSf3gW7KImCNQrERckDWyG+dy+MzMJuZ1RAIUQQgghhBBSLTKyciXrrJKDVqPmU4CwnlOnTP1oKTIixAGkEB++WLy3wTBCbkdajQouBwsnXNh+5Dh+2bINw5ZPx6u7vsTfR/bjiOoylp3/G7ZSDCKeulikGHjB2yA9U3xvjXr/bRzQXhTWp9/zaInnfGxYfzxVeyC+GPIyPnn8RXBqQKZgsCppM17ZvrzE4+8U76z4Fn/t24tzyXzKrdoBxc/K+OK1KW7b2jn4OkqpyMG/Zw5i99FTAIBoJrjKPpfqx0XDlC4OPv/5+BwM1rdHjCsEbaLEOk9yuRw6jRqsi4XDy2esN0t++gWHbZcAiPUgmjUQA21NomqX+lyhgYEY0/WeUj0/RQurn7cmouv8lxBQhy80bpHZgSAGrIUf3M625AMANl8VZ2/IHe7DXpF+3mubFJ1FUBFpAK1WG/rNngpjsBXvHvgRn//1p2S/v0wPtcp9xmXR4Gohu90hBFoAYETzbvDRibP1FAoFLDf4FH2Hr16A0+kEZ+AH+Yum8CokK/jAGP7pDOiCtXDZpUG1TKcR4a09P1czRzzutq3o42g8Is5tf0nyzeJrOMuYV+bjb6ZUli4Alpqehawc8Xq5RhNYpzRg4qmuS56ZnzETKvO/LWafeNK6tr9k/aOHqO4bIbei1AGUq1evYvny5ZgwYQJatmwJhUIBhmHwzjvF56YFgL1792L48OEICQmBVqtFkyZNMGfOHFit1mKPO3v2LB599FFERERAo9Ggbt26mDp1KnJyckrbbUIIIYQQQsht6sCJs2Bd0kEJjVrFpzZiXcgzmtyO2WE7JVn3M9TsVE+kZtJpNWAdLIysBTMOfY2PLv0BfSg/ULz7rPgaf371R3C5XLh0PdHruY5fld5tn5aTjcT0NGE91S8XcqX4T399KQozMwyDxwcMQOOCQXFDWNXU8qhKKWmZ2Og6gvdOrMGFFP75bRIdW+wxDMMg55oYwA1kDZgwYIiw/saXX+In8w4AgF5V+pRW5cUwDD4c/Bz6BrTCf+MXwl9jwGuPPoLvn33DLQigVqkK6lSUbQbKN5c2QxfMPyalQjznIH07hDr80Da6YfkfiAcus7SffjHSuikafzWitMEwxptglttgtljBFflaaB4sDuR31DbEe70nFHs9jVxMdcUWU1OotCbMfx+R7UKE9T9y9gEAdCYVwAIrHpoChYeZLnbOid1HT2LWd18LqdOuJiRDF6yBzqHGzw/MwEPte7tfsOCt+trWL/H4nPnCe1cucx/+kxUEsAqfU3+416QR+pMuptRqrov12EatLl+6zBxjvrB8LS21TMeyLOv293I6S/cabzDiMfScPwnbTx/n+5GXj5szdnlK4ZVaUCfGX3n7/g4Z1a8tEH8AABChMqFlLd9q7hEhd6ZSh0iXLFmCJUuWlNzwJt9//z2eeOIJuFwuREVFISYmBqdOncLMmTOxbt06bNu2DTqd+4+xrVu3YsiQIbBYLAgJCUHTpk1x7tw5LFq0CL/99hv27NmDsLAwD1ckhBBCCCGE3AnO3LgOmVwc1JnV4zGoVXwO9fjMNDSd8CRWTpqGFf/8i+ZN4zCxz1Bo/KQDNN3qNq/qbhNSbhq1Ci67SxiQLmqXWawdcMF6A3WffBS1ukUAAJYPnYSGkbUk7c8mXQeKZLx7YM1s9JR7f19obqEmkC3FDnX4nV9LaMycubikTwbDMdDeUAIFGbkSczMgC5GjZe2SU/0ce2cZDlw6h/b1GgmD0splMjjqsfBvIv4hbq4jU9m6t26B7q1blNhOo1aBc7FweqmFc7N/9u3DJ+v/gD5GfDxFgzKvP1ryjKbyePvpsZj45WIkyTKh9uVfg1HyIKhS5LgawgcKm4bHIjU+C6wcOHzlPDR+KqgsCnQLb4rXRz6KzjNegCFOh9dHPopAnff0kADgtIkzPzylbSoLm82Ok9x1+EI6wK50yvHXS/Mkz+M/Y9/FoQvnkGc1Y+GRn8HIGbx+cCUAIHCrLxqoovH24W+hMihRSxuCsEDP6eYYmTirJzEqs9j+FdZOKWRUWTCl/kgsuvirZHvn0MYIVfjhDycf/Pl0zEuer13OGVfZ+WJwMjE7vUzHhna8DwGhvri4/jthm91RullWLR9rAACYsftrPJU2CDkJRknQGRDT1xX18kefoOmoush3Wtz23S78fA3gfhqD/45dR+v6sdXdHULuWKWegRIcHIyhQ4di9uzZ+OeffzBy5MgSj7l27RrGjx8Pl8uF9957DwkJCThy5AguXryIhg0b4uDBg3jllVfcjjMajXjooYdgsVjw4osv4saNGzh8+DDi4+PRtWtXXLlyBePHjy/bIyWEEEIIIYRUmW+3bMIv+7Zj6vLPkWvO99gmzZgDAOge1AzfjXgNfRq1gUatgsqghG+0HnF9ojDn+GrEB6VjQ/Zh7D0vLUq7sN9E9G3YprIfCiEVTiaTQakr3f2MhcETAJj579du+0+mXQMA5CaI77PtrpNu7YxnTGCsuKVUMz+Ml9YzqIgCz1XtemIK4sMyoDIoofRRwNlIvKPcrOLTHkX4BpV4HplMhk4Nmkju6G8X18CtXdd6zSqg1xVPrVKCY4E8pdljWiKAL6I9ZslcDPvgTbx74icYY6TZQyL8S36eKkpIYAB+e2U2gjLF2RGto+ojPU1MU9e9UXNE6fk+/XhwG1QGJUJ1fpj10FioFUrsnPEhVj/wRonBEwBITMkQll3lnIFy8MQ5+EbxwZNHGvZGbjwfIIhQB7rNDNKrNOjZrBWGteuCrMvS9FXpxlyMX7JQ+MyoGxzl9ZpMfOn7p1OqJevRymAM790Nc7s8iRa2WGH7awNGY8KDQ6HLUOHpZkNQWT77/Q9h2epwT1945Mx5/Ll7NwDg5y3b8MYPX8JRUOsl7oFoRAwPgdlhE9rbHeLnlLfPLO6mmSVfXvwHv1h3AQCi1MHwvcEHDh1O92BMWAv+NWdmbG77bjd9W9VGoP7OD4ITUl1KHUCZPn061q1bhxkzZmDgwIEwGLxP7Su0cOFC2Gw29O/fH9OmTROi0bVr18bKlXwkfdmyZUhNlU7N+/zzz5Geno7GjRvjgw8+gFLJ51EMCgrC6tWroVAosH79ehw5cgSEEEIIIYSQ20tSagaWX/obH534HQe48xj3/fse22Wa+EGiZtFxqBXMzy5XKd3ztCvU/EDT3svS4tXtYisnZQwhVYGL5wdnTWkW2PMdeKR2byQdSofd5MBzTYZ5PuimO7ydTifMais4jsPGlxZ4vVYw54ujHy7D9hcX39Jd4nWiIvFM7BDkXOcHgO9Z+QruWfYKEnPKdpd4dfphyxav+wy1+KwYNw8ol1ZmRpEBbzPwcrP7MbJdj1s6V2ULDfIXCo1vPn/YY5vlf65HvDYDuQazZPuKIVPwdofH0SmuSaX382YuizjQPbh5B3Rt3FRY71KnKWIN4QCAoxl8nZYwgzhDQ6/TIjpQTKNVHBsnDrSXdwZKeq4Y5JnYYygcFn4WQ0m1OcY3HiBZdzic8PETM7cMaua9YPmn/3vRbduqB9xvXAaAcFmAsCxPY/DpqP8BALo3a4HJwx+E5aIVIRZfBGh94Oujx79vLMCjXfoV23entWy1dYqS6cUhSqfLfcbHmC/fxfunf0FCThpm/P4VdhlPY/qfXwEAtAH8e9dkF2eD2OxiEGbXJfegss1mx48bvH8uuBgWsoJAadH+bNl/BN3efhE+ev5vMnfwuFI9PkLInavSishzHIfffvsNADzOFunSpQsaNWoEh8OBP/74Q7Jv7dq1AICxY8dCLpdG5WvVqoV+/fgP7F9++aUyuk4IIYQQQggph7U7dkrW07lczCgY5Cjqei5/I1WzmFhhW1CA9/zcZ9Okt9Z6yulOyJ3i79nv4t0O47Bn2kf4bsRreGbAvfhj2jv4+aEZeLhbH6Tvy3I7JtmRhb1XxJlYc776Fr7RBoAFwoIDceGv65L2gYkGhKf64/OHXi53fx/p3w8tZGI9CRscmPnX1+U+b0VITEnDyI9n4eC185j703d4fvVHbneW77t6FgDwcO1eSDrsHvjpFnDrM0aeHzFcWO4e2wwjutyewRMA0GjU6GxoDACYv/0n2JzSO/0vJSbi9+u7JdtsRjsmNByE+lHR6N2qdblTNd2Kfs1aAwDi2DA0i4rDaxMewdOhg/Hbg7OgVijx6MB+sBntUPjywYmYwNBbus6A+u2EZU+zDspi91W+nlF7nwaQMTI4TPz5dKriA3WtG0lnNJkdVviG8jNZfnl4JlrE1PF6rEHnnjrO28yqh7r3RvrZbPTStcDW6R8gQC9+/9apFYmDC5fi15feLravN0v9T0wb5vSQ9qqQy+Vye4/K1eL4n6djgxvxAZ9zqQmQq/jv/5PJV/Di8o+FNia7OFvKUiSA8vaOb5GQI9aGAoAFX/+ApfF/ee2jmlFAXpDmzOFy4eC5s3j66w/wyqplkEXIoY7hZ3RE+gZ7PQchpGaotH9xxMfHIzk5GQDQtWtXj20Kt+/fv1/Y5nQ6cfjw4TIfRwghhBBCCLk9fLb3T7dt25NPID5bnHlusdqQozIBdqBxeG1hu0wmgzHJvXg8AGQ7PKcCI+ROFBIYgK6tmkOtVqFebDQAoGn9WMRE8rOxmtURgxUhVnFg8/VNKwAAn/z8G7ZyJwAAzX1jAQB/vTNPco21b7+NNTPeQqh/ACrCVzNfhTVHTFdjtJiLaV113vxyJdLVuZiy8XNsyD2Mk/lXYSlI5bPtyDFkGXOR7+DvTO/bpg1iFWFug7dzH7j1u8hbN20A1wEnjEkmTOpXcrrz6vZQN774uJNx4Z6Vr+KrPf8CAO5/fwbG/b0ILn/+ubGl2jEyoht2v7QEj/XsX239BYCp4x/G9CaPYtmTkwEAcrkcj953D4IC/AAAkaFBUPuIKYra1K53S9dZ9OpzcJ3hAx02x62nqvtty04c4C4AAGJD+NkxL/UYgdwLRjzWpvhZHL4GaZ3gxIwMIJABHECIj3+xx3qa3eKt7tGQPp2x9fVFmD3myWLPWRbPP3SfsJxj8fydnZqehc6L/ofhn82QbLeyYsDDWUz6tHNJ8ZAp+WBLvsKKY9wVYd/xq5fx4vcfI99ugdUuDQ7+cmiHZP144uViH8vU7qMgKwgWOpxOPPvbRzhrTwDCpQFEX7V7XWdCSM1SaQGUixcvAgDUajUiIyM9tqlTp46kLcDXTXEUfEkV7i/NcYQQQgghhJDq53A4ofH3MFjDAGN+no/DifyAUkp6FlQ+ShgYjVsu+JvGNcXtBRlZ7PkO9FKXXCyZkDtZ0UHPBhExwjLLcNhz6RSWHV0vbBvVoRcAoG3zhuBY8Q10c5Ho8lIqFZjT5Un4HtfAaXXB6Lo9iid7SiW27cxx7Dx2AjMPrcJ9P8xCvg9/Z7pBrcWauW9hmKYjci/zA7wyO1PuWRWbPl6IzS8tRLDBv1znqQo9O7VC1p4cYf2rUxsw8dsPkOkrHfD+6YUZeGnYSCgUZa+bU9EYhkH/bu2gVnsOBvj6iIXa5TYZejZseUvXUSoVCA30BwDYy1Hr56N1a4XlBuF8gPSFR+7HsYXL0alBU2+HAQD0Wo2w7LS5kKHJg8ZPBShLLtQeGx0O40kT8m54vhHhZuGhFVvPZtLoB2HP4J+31Pxsj22OnbkEjb8aOUppH21skfRpN6XwOnjinLD888UdCIjzXM9m/s4fccx0BZ/t+BPWmwNgN8VkVGoxZejD4T0RdNEAnGfxWHhfbB3/PlrG1hVmun61Z4Pw20YXrpGcR3kLdaUIIXeWSgugZGfzH5T+/v5eP+ADAgIkbW9eLtxfmuMIIYQQQggh1e/bvzbCEC7ejanJViL5qFiU9/B1PoCSnJYBpVYBH4V7uhGvEZQCL3cbgdlPVNwds4TcjooGUNrUqS/Z99qWFfCPFQcQowP4+g4Mw+DiX2WoIn0LBnRrj69mv4q8GyZYYHebyVHIxbLINOd53FfRGLX7mMP8Az/izR1i6kBVAD9YqldpEOjvi1eeGA2Xgx+kVUDudnxZ6bQaYTbEnWDeY08h7ZSYJu6cJcGtTZihYmYuVQWZTCakZqsTFFGugJi8IKh/qym8Plq3Fq7G4nrfJm2E5dL0q+gskryEss281Gk1OPrxMqiTyv+avhUMwyDM4Q8AyPMyQ83icC+63u+FKVDVET/zDp2/IPls2X7iuPt5UtzPo9Tyz53ZaoXZxgdNHQUBHftNf09rQfq6aR0exHP33offFs7BjkVLMOHeoUI5AauVb3PVngKZnNKGEnK3qrQwqdXKf1CpVJ7vDgD42SkAYLGId60UHlfcsZ6Ou9kHH3yADz74QFjPy8vDhg0bAAB+fn7o1KkT9u3bh9yCol5169ZFrVq1sHXrVuGYjh07wmw24+RJvtiUUqlEnz59cPz4caSkpAAAIiMj0bx5c/z3339wFnwYt2jRAhqNBgcOHBDO1adPH1y9ehVXr14FwAeWOnbsiD179sBo5Avx1a9fH5GRkdi+fbtwXOfOnZGXl4fTp08Lj71Xr144duwYUlP5FAjR0dFo2rQpNm/eLETpW7VqBaVSiYMHDwrn6tevHy5duoRr164BAAIDA9G+fXvs3r0b+fn8l3KDBg0QHh6OHTvEqY1dunRBdnY2zp7lc8ZqNBr07NkTR44cQXo6/wMlJiYGTZo0wcaNG4UvudatW0Mmkwkp2QCgf//+OHfuHOLj+R/1wcHBaNu2LXbu3Amzmf9ybdSoEYKDg7Fr1y7huG7duiEjIwPnzvF3Heh0OnTv3h2HDx9GRgb/D/JatWqhUaNG2Lhxo3Bc27ZtwbIsjh49CqDgrpX+/XHmzBkkJPA/EENCQtCmTRts375deP01btwYAQEB2LNnj3CuHj16ICUlBRcu8P/oNxgM6Nq1Kw4ePIisLP6HZ2xsLOrVq4fNmzcLx7Vv3x4OhwPHjh0DwE817tevH06fPo3ExEQAQFhYGFq1aoVt27bBZuN/BDRt2hS+vr7Yu3evcK6ePXsiKSlJmH3l4+ODLl26YP/+/cjJyQEAxMXFIS4uDluKFEns0KEDrFYrTpzgp/grFAr07dsXJ0+eRFJSEgAgPDwcLVu2xJYtW4RZYM2bN4dOp5Oky+vduzfi4+Nx+TI/3ZXeT9cA0PuJ3k/0fqL3E72f6P1E76dCb//6DWr3iACb5YQiicOsRx6Cs48W049+DQA4cfoMrkc3x8pt/4LxYeAy2bFhwwbJ+ynzQi5f18GLlOvx2JDN/7an9xOP3k+8mvR+yk7PAAoyd127eAnedJLXR4TWX/j3bmAA/95hOAjbKvr7qWXLlnDkO8DJOPz573q0btpC8n5iWQ5zj/wNWZAMk2v1hVauqrT3U2ZuLuy1vdRakN4oDo7jYFBphfeTw86/BhWMXHiugLvj/XRv367wkTvx+OefIq5PlNA2IEWFEXXbgOM4OCw2ZKam3zHvp7aqGJw8n4JH2rZFWlraLf/eKwziHzpyBPZraWX6vffz2t/x5eG/4R/rA7vJgXc6jcF/m/jXd2m/nxo2bIiYawGIDfPHXw4xcDCuYR/hdVrS+ynOEIArOTkYVb87UlNTq/T7SQY+SHTg4AHknU9wez/tOXYaKIjNFT4eewvp9BB7qBPTvv8M04c+iv3792PnicPATVnZnmvSA59e3A5d0E1vdPCBj70nDgPRgLwgbpJrzBWup1QqYS2YYXTj2lVsSDV6fD+ZTMWnKXSZXdi/f/9d9/0E0L+fAPq9V5N+75WE4bzdLlKCsWPHYtWqVZgzZw6mT5/utv/nn3/GqFGjEBYWJjx5N1u6dCmee+45NGvWTHjSDx48iA4dOgDgAyQajfsH4T///IPBgwfDYDAIT2ZJoqOjhRczIYQQQgghpHRcLhfW7NqGezt3hV7F/zZ3Op3gOPdc62euXsMzm5YAAN7tNx5d64hFmZ99+wOcDLmOcEUA3ugyGi/v+RwA0NgnBl+Mniw5z4qf1mNV7mZ48/nwl9AkLLYiHh4ht603v1iBnQxfhHrhgIl46vNFHtPW/PX4O/DViOmLZn36NdYbD+Cp1oPw9IBhlda/Zi+MQ2ALP/zyyFsIvSltVVpGNh5YOxsAsGTIc2gdVd/DGcrvwtV4PPn3IsiVpb8zfMfExcJyg+ceR3irIATBB79NnF0ZXbzt2e0O/LZhJz5NXgcAeKfXk+jR4M5MkchxHOx2h9c0X6X16Px3kBCYiVc6jcLQFp3LdOwHX67BT7k7oJOrsOKBqagT7TmlfWn1emUS2ILAwfcPvYEYv5BSHWe3O3AjNQNxMRHluv6tGPXO20gJzcHMHmPQr1Fbt/1PLHgXVwP4gu6F78eWMybAL8aAWqZgxOvFWauF+wfPfA350TbkJZrgG60X9r278ntkuYzYf/AM0E6cddNcH4ukCxnIjMpHQJ4e2b4m9A5tibfvGyu06fPaFDjrsPj8vpfRJFSsxVZU38mT4WgkDptq45Ww1BJTg214fD60GnVZnyJCyG2mpLhBpc0/K0yzlZOT43VKb2EKrqKpuooue0vR5ek4QgghhBBCSMWb8flKLD3/FyZ8vwgZ+fxv+w5Tnkffr6YhxypNLfLqymXCcr3QKMm+hVOfQW6CCWnIFYInABDrG+52zab1Y922XfjrurBcdLCYkJqqdnCosKxXa/DB0KcRcM69WLFOJb3pcOazj+OXMTMrNXgCANqCFGNmh9VtX0q6mBrqpfWfIc1LLYTy+mnzNiF4Ys6wIvN4DlJPZrm1c9k9z1CRq/hj9Ur3GzfvFiqVEr06tRLWu9Qrvj7H7YxhmHIHTwBAzvAD8TenfCqNWZ98DbWPEtE+IeUOngBAq7C6wrJGoSympZRKpayW4AkAFCYpu3bD883UGUVS+7286lPkWkyQO/j3Yt+6bTwek2njj+kY0FCy/fVxj2LhhGcw5p57JNtPmq4hM4r/jRKh5wuoOVzSv6e9oOaKWu79eXXc9NkxemAfKM4wSNyXioaWKAqeEHKXqLQASv36/B0mNptNmEJ0sytXrkjaAvwUKaVSKdlfmuMIIYQQQgghFW/vuTMAgERHBkasfhubzx2GoSk/iHstix8c+XHrFtyzcBpyY8QUu343BTkMeh3y46WpMOrawzFt4ENu17TZ3Qv3Pjt8uLDsr/We3ouQmqJuuBiE1CrVuKdre/zxwVw4zNJBQIVMWutAJpOhdrR7YLKi6QqCDmYP9QxOXbkqWd985kil9GHbJTG90cIHJuL0p1+5PT8AH1wBgNaBdSXbGTk/1KtV3d2DoKHBAWhtrIMnYvq5vZ7uRpqCdPJmq/truyQDB/EZVbo1bl4hfRnSsaPYL8Wd8To9cIxPzfTmoi8l209fuIov/loHk1MMuh6xXcLTPy4GK+MAl3swI9/O/66wyRwAC3SNawp7vgMRdukN1c3qxHntT5MAfnbJzQEUJ8cHR7RK78/rzXVwaoeGYcuHH+DKytVY/tJUr8cRQmqWSgug1KpVC+Hh/I+23bt3e2xTuL1jkS8EhUKBNm3alPk4QgghhBBCSMViORZcjHTb0u3riuzncDE+EZ9dXAebnzjIMDy4s6QAdqFZox6XrC8a9xwUcvfBugA/9zRFmiKDKgaVh8LzhNQw0eHBwrK6yJ3nlixx8DFcVX1ZGXw0fCD12d8/RJ7VhL8P7sNzq5cgz2rCjJ++krS9kFw56bQtMnGA21fPB21lCvci3RHpAchPMeOJ1v09nkddTO3WuwHDMFgy5X8YP2hIdXfltmBQ898xueayFXAHAJeKz8BSOzCsQvoSEuAvLGuVd8br1OXk65ncnIvm3oXT8X3SFjiCpLM6Ei3p0EVowLCAUiFNDZqWn8PXT1AxkLkYjLm3P74cMhmrn31T0q5L22Y488sVOG03zTbL46AquEnb4RL3LfztJ6jq88+nQeV9BpqmyGeDy8aifWwj7w+cEFJjVVoAhWEY3H///QCAFStWuO3fs2cPzp07B6VSiXvvvVeyb8SIEQCAr7/+WijyUig+Pl4o4DNy5MjK6DohhBBCCCEEwKxVq8D6S7dlyMTUG3+d2ouJSxcJ63mJ+WjBxGHKiFEez/fE8IEY7dMTAGBJsyLAy0ySFo3r4vGYfsg+JV7LZeUHZJjMW3kkhNx5osKDkXWZfw+EGsRAySOxfZB+lk+JNaZZ32rpGwDoC2ZtcAAe/2EB5h/9Cafyr+Grvf9CpuKDGG3kfPEGu9N9VllFsDH8efNTzWhaUBdJoVG4tft10Wz8+uQstKnbQLLdaeXHG/zU7qnRyN3LT8u/Hn66sh0WDzOsiuME/5oqblC+LPQ6LVgXB9bJ3jGzg+QMP9SoC5E+B8EN/fn9hptmzSlkUKjl4JQAbioBMPWPz/H70d1QauVQcgrIZDI0a1gHcg83X5z9YRW+HjINaacK0vi5gD+emi0EoB0s/7cZvfgdrEvfJxx3cxrEosbcL6YG2/2/JXQDByF3qUoLoADAtGnToFKpsHHjRixcuFCohXL9+nWMGzcOAPDUU08JM1UKPfPMMwgODsbZs2cxefJkOBz8j6LMzEw88sgjcDqdGDRoENq2dS9GRQghhBBCCCk/h8uJPTdOAwA6+zRGzoZctzabrx+FqyBrxhudRuO3CbPx8VP/K/a8z46+D5NjR2Ld03PBMO53ihd6atAQyEwF+zngf2NGoFlKLawe/6bXYwipSQx6HcbWuwdvNHkYKrkYFHhl3Gj899r7eCnmPtzbrmu19U9bJJ1QlsMoLB9MvICgYD8AQKQuCABgd5W9lkRJWJYFqwI4C4e1Y2cJs97Ygrvf/fP4QXAul4VWo/ZYD2JS95GwXrJicq8HKrx/5M7lrxNnQa4/u79Mx7oKBukLZz2UV73YKPSwNMWCLk9VyPmqQmEApVZX6Vif2rfkGTR2u/SzIsORh4+O/g5toAYapvjnNCTIHw3qxEBlKfi8lAMB/r5CAMVZ8Dl0Qy+9E6O4wFT35i34x+If6rUNIaTmK3UAZffu3QgODhb++/HHHwEA7777rmR7QkKCcExcXByWL18OmUyGV155BTExMWjTpg3q16+P8+fPo23btli4cKHbtXx9ffHjjz9Co9Hgo48+QlRUFNq1a4datWph9+7diI2NxcqVKyvg4RNCCCGEEEI8mbJqKewh/EDQ8wOH49s5b7g3KnJjbr9mbREXE1FsUKTQff27ITwksMR2LMsPhMrAwMegw2czJyEqNLiEowipOaZOeBgDu7mnro4MC8bIQT2roUcib3UD4i1pUETxA5g+BXdr31x7oCJcTU6BT6QOWpkKsUWCI0+1HgjTSQsWj3oO+mNKzO07zus5nn14OA68txRBvn4V3j9y54rwE7+fZGzJ32lFuTj+e0tVhoLvxWEYBvMmT0DXVhVTU6UqyDz8DsjMlt6E4bA4kXIsAy4HK91u9/5ZEaTxLdX11fn850+ck0+jplYVBFAKflNwrDjLxXbVXuy5GgRH462+j2PBwAmlujYhpGYqdQDF4XAgMzNT+M9m4/+1ZDabJdtvTrn1+OOPY+fOnRg6dCgsFgvOnDmDOnXqYNasWdi1axf0er2ny6Fv3744dOgQHn74YTAMg5MnTyIsLAyTJ0/GkSNH3GatEEIIIYQQQirOv4cOCMuhPgGICg/GxX/ipY0Kxk+jHUGVklrEJecHOzS4M/K+E3I30ReT9kam5ocaCmtJHDVeRrIxq0Kv//H6tQCAetpIyfZXnnoEhz/+HHVrReGfz95DjzYtK/S6pOarEykG5EwWazEt3RUGUO6UdFuVQa5wf+zxN9Ik62q1Ehtffw+mNItk+80zUIqK8i3dDRSrZr8Ow2EV3ntoIgAIs9McrPu59837pMTz9a3butTXJoTUTO7JQb3o1auXkIKrrLp06YJ169aV3PAmTZs2xQ8//HBL1ySEEEIIIYTcOoYR77XSKFTQBKrwzeTX8c3aDbh+IxUpwdkIb8UPKPRq3KpS+mC5akNOqBGPd+5XKecnhNw6jbzkwKZWJc5SeXfzanx0/wsVdv0LSYlAJDBx4NAKOychANCkfizOTLuCJiPrwGg1l+lYMYBSqRnzb2sKlXsAJSM7R7IuYxjJzDEA8Luhhb2O93pJjaJjSnX9JvVj8fcXC4R1fx8DOJbDNWcqzFYrGJk4Q6Y0s2YJIeTu/UQnhBBCCCGEeKXS8fdaRV0XU5nc070dvl38Jnas+Qh+erEAfJdGTSulD38tnYdBPu3x5OCBlXJ+QsitUzLSQdKgfB+c+O6iZJumSB2IpOyMCru2w+FETqAJABAVQHeGk4ql02rw9IhhAOAxgOJwOeFiWbftAMCCv/FYKSv1/co1jqcZKClZ0hloyoJ6JkL8Ip7FyhdfwYAeHXB4+VkEHte5naNuRKTbttII9PcVgiZ/ndoHp5XPnBNlLDmVKCGEABRAIYQQQgghhNzE5XKBUTJgnMAPc2d4bCMv8k+JphGxldKPpg3isOD1pyGX372pUAi5XRUtIg8A3z/7Ovau/BR2E38H+Zg6fSUzUJwuFkabOBj9+46d6PnJJBy4fq7M177vjelQaPgB6kBt6eoiEFIWgTr+dZWcly3Znm8yo++KaXjqp/c9Hlc4A0V+F6fwKjrDo1COxSRZL5yho/HnZ7J1bdEcIUH+qB8XDdPBfzFhpPvMslCfgFvqT6Cfj7B8LTkFAIfsq3lY+fwrt3Q+QsjdhwIohBBCCCGEEAljvhl+tQyQc97/uWA184VXGZahFBiE3IV8deId4q81GgWdVovG9WqjTno4uAQWj3TrA41aTPOVxRkxZNWbuJadAgCY/PXn4FTA1A1fwGjj6yBkG42wOmwlXvt02nVhWX4Xp0oiladz4yZwWJy4kp0k2b509Z8AgMvGZI/HceADKMq7OPAfrXSfFWa0SGfyKBk+AFo4G2Rkx+6S/XVrReLiP/GQnRS3+Wk811AuSVCAL9LP8oGwlev+gUKjQFRQCLQadQlHEkII7+6dU0gIIYQQQgjxaNbyVWB8GThlnlOUAHyQxRcGaGVU4J2Qu1FIkD+Ml81ALofBEzsL21fPmy4sKxXuQw7HEy4jNiAcLodL2DZlzVI0CIjG6iP/wS/EgC3PvF9sYLYwaDKr22MV8VAIcVM/LhqWdTbkh1mwdtcOrL24C5+NfgnxmalAQeYnJ+uSFIt3uVxg4vj1u7mI/FfzXkGvBVMQWFucHWa0WSS3cBsUGgDA6TVX4ButR9sJDSTnaFSvNrYu/ACx0eEI6X0//GsbEDLR/5b6o1AoMLx+F+zBWYS24/94dpf3WiuEEHIzCqAQQgghhBBCJC7nJAG+QB1NuNc2roK7bP3Ut3ZHKCHkznZf/244ePwcnnx+kNc2sdHhuDQ7AfUGisWfD147j0PHzqN2d7GA9DlLAs5ZEmAI18EFFlkWI4J0nlNzWaw2+NbWw2V3oXP9yqm/REhwoB9cFhccMhc+PPMbAGDf9bP4Zf8O1B9UCwBwIOEcutQWX4OJyenCMoO7d2ZmcKA/n6KryFNgslkArbj+dv8nAAAJ/60Bx3EeA6aN6tUGAFz687tyz3T11eqAIjGT5/sOL9f5CCF3F5rrSgghhBBCCJFwMnxwZPI9D3ptk3EuB8ZkMya0HFxV3SKE3Ebkcjnmv/Y0Gtat5bVNaHAA9n/ymWTbqeRr+OPcHgBA9lWjx+OSjVketwPA1YRkaIM08GV10CopBQ+pHAzDQA6ZpJ7H76d3C8ETAJAz4pCay+XCJ5t/F9YDdWLdjbsTIwmgmAtS8+Ul5sPvghZ1I6IAAEEBfggO9C/2TGEhgQgNvrX6J4V8tdKbPQL8DeU6HyHk7kIBFEIIIYQQQoiEjeVv09SrNF7brJg+DXWSw9C7easq6hUh5E6k12nAOvmgrDnTCpPDArWfCqY0C6a2fQAcy7kdczU9yW2bsC8pGQq1HKH68g2oElISGSed9XAq45pk/bUNX8JSEBiY/91q7LefBwC0cMbe1Sm8AAAcAEZ8b19J5mvGPNy8N36cPaPKuxPi6ydZ71SbZq8RQkqPAiiEEEIIIYQQCQfHB1B0xQRQHhrWB79+Phvyu7hQLiGkZHqtBvs+PInsP3LgMDthVTigDVAjNjAMg3p1hNPmcjtm/en9Hs/FcRzm7fsBAOB/iwWlCSktGVf8kJmLY7Hp/CEAwIajh4Tti556plL7dSdgAIBhwHJ88DTPagIADOneGXqd1vuBlSQ6NERYDk73gY5mrxFCyoACKIQQQgghhBCJwuLxhmICKIQQUhpqtQpJe37F/l+WIutirpASKTYwHKHBAci9ni+0ZdKA3IR8nMtO8Hiue16cCnkAH7QNNvh5bENIRVEw4g0C+SlmYdllF4N+ey+dBQA41Pw2+zk71CoanEfB5BOb3YFlq9ch18Y/f37VFPiMCA0Slnu0bVktfSCE3LkogEIIIYQQQggRWO02KGspATMHH7WuurtDCKkBQoMDoNNqwKaywrY6wXwR+csbExC/OwXn113HkpHPwWV0gZVxyLdbJOfIN5lhby4eH+rrXyV9J3cvZZE0XBGuQGE55Vgmzq69CgDIzMsDAFhkNnAsh42z36vaTt6mmIICKK/O/xxPv7EInJyPqPioq372CQD4GsTfM6G+FHwlhJQNBVAIIYQQQgghAIAjFy+g/9evAQD8GSqwSgipWO++NEFY7lKXr0Fw+NdleP+hp3Hh+2/QqnF9qDgFAGDw12/AxYoBk92HTknO5aOnAC+pXEpGKSx3aSHWzLinRzu0DasPALA4bTCZLZDr5VA65fA1UGo5QKwfv3HXIbR4rD4i2/IptNRypfeDKpFOK86oDQug+kmEkLKhAAohhBBCCCEEAPDy1qXC8ujWvauxJ4SQmqhudKSw3Cq2HgCgWcM6eGBwL/j58kHbXLNJaPPj0a3CckpOluRc0UEhIKQyqYrMQAkPFGegNImsjRn/exwuuwtGpxkXryZC7aeCQVY9sytuSxwfQpGHKOAXI96QoZQrqqU7Wo2YVi06KLRa+kAIuXNRAIUQQgghhJC7wIzfVqLHsknYdvmY274cSz4e+nqOuH7NiAe69qzC3hFC7gYxEaFIPZGJjHPZYBjGY5v0MznC8pX0JGHZZLMCAFKPZ+Ha30noXqd5pfaVELVCJSxHhwRj/0cncfGfeAxv3BV6rQZgGGQhH9vOHodCLUekb1AxZ7u7CO/um0qpKYoEpaoSwzBgXXwasRh/Cr4SQsqmekK/hBBCCCGEkCqz+cAhbE8/CQB4f9vPmPnfKkRoAhETEIoDSeeKjHQA2MHiy6lTqu0uUUJIzRUTGYo+fq0wpE8nr22+m/UGnvtyCer0jQLn4ITtJjsfQLm/dzfMf2CCt8MJqTAahZhuKjwgCP8sew/5JgtqR4XBarUhLzEfAXG+uJKXAgAI0PpUV1dvO4U1UJxOl3S7l8BpVTi87Ay0ARroJmpKbkwIIUXQv4oIIYQQQgip4faePSMs57nMAIBkaxaSk7MkwZPI1AD8+N3Mqu4eIeQuwTAMls6dXGybe+/pig37DuAk4pGenytst9rtAACtSu3tUEIqVNEZKL5aHVp2qSus160dieTDGQiI88Ux52UAgEpJQ2yFCn9aOB2uYttVpS6NmyE4gArIE0LKjj7dCSGEEEIIqeGSczPd0mjcLG+XERuWza+aDhFCSDEGd+2EExeuI9tiBABsPn0YyfYsQAZolaoSjiakYliNdiCYX9YopYE7hUKBkff0wFkkCtvsnLMqu3dbs1js0EADF9jq7orgv9UfVHcXCCF3KKqBQgghhBBCSA2VZTYiJSMT+43nwXEcEvamSvabM61I+j4Vr9R9EAe+XCopskoIIdUlPCQQDosTRrsF12+kYPbu73BKdh0AzUAhVSdI7oO009lIOZ6JCJ9At/3+OoNkXSavvvRUtxuTmU+5J1PSsCMh5M5HM1AIIYQQQgipgTYeOIh3jq1GsMsX+hAt8i6bEGyS5mdvEBaN1duXVlMPCSHEswA/H7AOFna1A/8eOCDZp6MACqkirz49GtxSDv8bO8Lj/kC9D2AX19uHNayint3+OJavX8TJuRJaEkLI7Y9CwYQQQgghhNRAk3/hAyMZ8jwAQO+2rfD5jMnI+jsHCauTcfHveAyL8F7ImRBCqotWowbrZOHkWCRmZkj2+en01dQrcreJjYnA5/OmoGmDOI/7gw1iPY0r/93AoBYdqqprt72wFvyMnZjOYcI2Y7K5urpDCCHlQjNQCCGEEEIIqYG0gdK7tJuHxKFt84Y49ftXMFus+GfbfowY2KOaekcIId5p1Cq4HBxYsEg1Z0tqOEUFBldfxwgpIsTPH8jil/9ePB9qNdXnKaTSK922japLvzkIIXcmmoFCCCGEEEJIDfPTjq3QBoojjvZ8B57o0V9Y12k1GDmoJxiG8rUTQm4/GrUKrJOFA06c0yRK9oX6BVRTrwiR8vcVa6AE6n2KaXn3yblulKwfXXYOEwYPqabeEEJI+VAAhRBCCCGEkBpm6bY/JeszuzwGuVxeTb0hhJCyUauUYB0soHQP8gZoaKCa3B6KBlB81Lpq7Mnt58qmG5L11L2/IzjQv3o6Qwgh5UQpvAghhBBCCKlBTl68AkegC850Fy7+lwCOZTF4Ysfq7hYhhJSaTCYD6/JcfNpXQwPV5Pbg5yPW49GrNMW0vPu4HKxkXaug9GaEkDsXBVAIIYQQQgipQX7Y+h/kKjm6xDXFyFHdMLRv5+ruEiGElFnivlQE1eeLdOenWmAI0wIAFDKaTUduD0UDKPS6lGKd0gAKpQwlhNzJKIBCCCGEEEJIDWFz2LELZwAA7eo1wEMdeldzjwgh5NbkJZiE5bZ+9XAeN4ppTUjV8/XR49DnZ6DQyIGJ1d2b28uEUUNxAtequxuEEFIhKIBCCCGEEEJIDTHmw3kAf8M2utRpUr2dIYSQcsg+vg6xfUdDFarEkPs74DxHARRye5HJZBjetSs6tmpc3V257WjVlLKLEFJzUACFEEIIIYSQGsDlcuF8diL8/XyQfSUPMcFh1d0lQgi5Zf5+Prj23w+wO5z4Zu0GAIDD6KzmXhEi9cPHM6u7C7clOSuDJcsGbaC6urtCCCHlJqvuDhBCCCGEEEJKz+q0e9x+7mo8/GN9YMuz48dx06u4V4QQUvH8/XwQGhyAOrUiceCTU4i6FljdXSKElALHAVc2J1Z3NwghpEJQAIUQQgghhJBS6jVpEu5ZOA0sx5bcuBjf/74JT732HjiOK9Nx837+Hv1Xvoqfj21z2/fAolkAALWvCk0bxJWrf4QQcju5f0B3rH5vBlYteK26u0IIKYUWjevA5SjfbyVCCLldUACFEEIIIYSQYuw8dQIPr5iDKxlJYBsDNj8nsizGMp1jw8EDGP75DKTlZwMAXvrsE5yrnYTfjuyStMs15yPFmCXZ5mRdAIB8kxn/Zh8CAPxybKe0jdMJTRifJmOgoW2Z+kYIIbc7hmEwYmAP6HXa6u4KIaQURt/bF03r0s0chJCagQIohBBCCCGEeJFnNOHNPV8hyZWFsWsXCtuzzKUPoOQZTZh79Adky/Lxwu+f4EJGIgzhOsjkDD48vBYulr9Dk2VZ9PtoGkb9MAeLt/8CAHji4/nos2wqjiRcQJ9XpgjnzLTl4VpmsrD+wa+/wCdSD41ZiddHP1reh00IIYQQcstkMhk6tmhc3d0ghJAKQQEUQgghhBBCvDhz+brH7ZnmvFKf47dt4myRFHMWnlq7CFH1g4VtFzL4HOH/7NoPbaCGP+b8bmTm5eKqOhWQAS//sxS5jEk4xs448fiv78FstyIpNQN/5e4HAPSObQWGYUr/AAkhhBBCKoFGrqzuLhBCSIVQVHcHCCGEEEIIud2kZ2XjsVULYLZbgQD3gERqXpbbNhfLQi6TweZ0YOb3X2Gv7azX88vDxJ/hp25cRYRPID7fsQ4IFds8+dNCyTFhLYL4a5/IFJYvZNzA7q0nhDZTh40q3QMkhBBCCKlEOpmmurtACCEVggIohBBCCCHkrudwOPDO79/j/s7d0DyyDp755EOYw22Ang+esEeckLURfzpfuJEINOOXM3Nysf3kcXx49jc80fQerDq9yeM1WBcHmdw9GHPkwgV8fPB3IXhiybZBG6BGDmdya6vIlaEpUwtJphyo9Eqs3rIZ+8znAABd/JtAKaef94QQQgipfnq1BqA68oSQGoBSeBFCCCGEkLvK6s3/4e21q2B3OYVtry1fjq3Zx/Hi35+i95dTkB4uTdG19ePFeETXC72dLQAAyTn8DBSr1Yae8ybhw7O/AYBb8MScaRWWl/V7CcO4jkg7nS1p88+Zg8KywinD+T89pw2z37Bjxeip+OmDWejJ8dGbXWmnhf3NomJL9fgJIYQQQirbPd3b4sCnpxF3NbTkxoQQchujAAohhBBCSA3kYlk4igQIiGjJ8d/wX8Yx7LnKBx8Onj6Hg4qLHtsak0zooWsOpVKBZ8YMR/M6cQCAeFMaOI7D6YvXEVjPz+2469uS8Wnv/2HLC++jSXI0Pu75PJrUj8O0px+Gwi79Ce4bpReW9dDgz/ffgSndwm9w8n0AgAEdOyAuOgJKpQKDu3QCx3FQGOTCsSPad7/1J4UQQgghpAI1a1gHp/5Yic/fmlzdXSGEkHKhOf6EEEIIIXcQlmUxfvFCBEX4YeHopz0WDG848jGEDQiGyqZAJ79GuGZPxUePvIBAnW819Pj2sn7nPqh9+KKmWXm5AIC5q78Daru3jcoLxICm/fH4yAHCtmA/PzhtLqQjF6O/egetNHUBAPZ8B/RnVHB04AAAR75chqCC5/vzt6ZIT2zjvPZPLVehcb3acKxzAiGAnJUh61AufO7V48HmYoCkZeO64I5wYBT83//d3uOgU1GucUIIIYTcPurFRld3FwghpNxoBgohhBBCyB1i2fp1+PC3X3HZJwUH8s9j8tqlHtuFDQgGANjVTuywnkI8m44vd/xdlV29bW09flRYNtr49Fq5nBkAMCK6K9SH5Eg6nI7V97+O1VOmY+yDgyCTiT+ZgwL8cOmfBABAkjMLf+fz6bee7XgvNi17H9HXAtHeVV8Innii5Lzfw+Sr1CEkyB8OMz97SMHIsX/lZ5jX+kk0i6kjtPP380HmhVxhvWPdJqV+DgghhBBCCCGElA7NQCGEEEIIuQOkpGXiuxtbJNsOZ15EpjlPMlhvtzvgtLmgUMslbf+K349xpkEI1runm7qbpBqzgSB+2WTl02SZWSvUUOOxXv0xodcQmMxWhAYHeDy+eaM6GNysI64gVbJ9WKtOkMlkWD1vRol9UCuUwvK+D0+izYRGSDqYDp1cgwWvT4BMJoMj2wEAsCkcCA8NQnhokNt5bCdssNdyIJIJgkImd9tPCCGEEEIIIaR8aAYKIYQQQsgdYMvhox633//tW/hi6zphPSsnDw6Tw2PbV/9YXil9u5MY7WZhee3lXQAAu8wJjuXgrzFAr9N6DZ4UmjJulGSdtbII1Jc+PVqUJhgOixP1zZHIO7Ie48IG4Mo33+Pct6vQuQk/k4TJ4VNzhXDeA16H1n6BFUOnYO2kt0t9bUIIIYQQQgghpUcBFEIIIYSQO8D2cyck69YcG7/AAN9f3AIn6wIA7Dt5Bmo/NfJumJCXaEL6mWw4rfy+i/k3sP74Ppjs1irte3W7lHQDVzOSwXEcMkKMwnY760SPZZPgE6uH2qGEXFa6n8a1IkNhTBYDMZPbjSxTf5bPmYrOeY3w0cQXoNGoMfGRYVAopBPDf3xnBkw7TZg78Emv5wn090XTBnFlujYhhBBCCCGEkNJjOI7zXsWyBomOjkZiYmJ1d4NUMJfLha92/Ys29RugTWT96u4OIYQQUikSklPx6Lr5km3xu1JQq1u4ZNtfj7+DD75fgy2uE+gR2AzD6nZGh1aN0X7UM9DdoxPahSsCsGbczCrpe3Vbt30P5h1dDaVOib5RrfDfjWMe20UhCD9MnF7q86765V/I5TIwDINH77ungnpLCCGEEEIIIaQqlRQ3oBoo5I7FcRyaPT0eIR0C8M35zVg5cirqBUVVd7cIIYSQCvHPzv2Yc/Q7KDQKBBoNgI90/xfPTcI7x1aDkTHCtvu+eAtOLT/bpE5YBDq25tNB7fvhU7Sb8xx8o/QAgBRndtU8iGrAcRxyrSb4anSQMTJsPH4ISh1fc6QweKIzqnDxbCKiOoQKx7WOqlem6zzxwMAK6zMhhBBCCCGEkNsTBVDIHevwyfMI6SDmKE/Lz6EACiGEkBrjh91boAjkf6pl+eQDADrqG+Lnr7dhWO8uGNCpAx6b/S6CGvohpDH/fVgYPAGAuNAIYVmhUCDjVI4QQKlpzsfHY/qGr5HKSQND8/qNQ1p+DqABHBYnlFr++ezftB2GhnTC99lboNQqoE1RYuqEUR7OTAghhBBCCCHkbkYBFHLHOn39mmTdYr278rkTQgip2dLMOUCgdNtLg0di4aPPCOv7v/gMxnwzejzzElo8Kk1l2SSytmT9i+cn4dPvf4exgw1yV80qgzd5xVIYo9x/B3y883ckIRNgGbiOOaDszP/0HdCiPZpGxqLB7mjUjg5Dw9q1qrrLhBBCCCGEEELuADXrX8/krpKUnSlZT8/LraaeEEIIIRXP7JQGBDgrh+iAEMm2uJgItGhcF4teeBYcKy1rF+Yjjb4M6dMZf69YgNzLRrByDmaHDQ6XE9O/W4ExX70Lh8tZOQ+kkm09f8xj8MRmtCPZlgUmWAZtngo/zJqB+N0pyLlmROMIPmDSv2t7Cp4QQgghhBBCCPGKZqCQO1Z6Xg6gAizXrdDW1iAxJ6O6u0QIIYRUGCtnhwYanF1xBdq6GuxatMRr22b14sBki7VQVg6e4rWt3CUHGGDgitfAyBhw4AMvRxIvoGPtJhX3AKrAfweP4O2j3wrrNw6kIbRZILgkFjeupaNOPz6154Cm7VG3dhSmD3oUIYH+kDF0DxEhhBBCCCGEkJLRvx7JHSvLZAQA1A4MAwD8eXUvUvNrblFcQggh1edS4g0s2fgrWI6ttGtM/WYpBix7FSa7FVuPHoUmTgO4OFz5bzU2zH4PtSLCvR7btEEsko/yNxL0NrREvehor23VjoL7Z2QQgicAkGHMq5gHUoWm/PW5sDzKrwd2zl2ChomR+OGFN9Euqj6cNhfSz2bj+X7DAQBPPTwUw/t3q67uEkIIIYQQQgi5w1AAhVSaZ+cvxvyfVlfa+XNsJgBAlG+wsG3P5dOVdr2a4tdtO/DSN5/AxVbeICAhhNyp2CKfjTm5RvSaPxkvrv4Y4/5+H79e24Utl45WynXnrfoOB6wXYIEdx5Mu4Y0dK/kdcgY+Bj0a16td7PEGvQ6180OQ8nM6Zo1+oti2IQp/j9vXnd57K12vNvkmM1QGJQCgiaIWXnjofkSGBePr919HXK1IfDljGiaGD8LJhV9Co1ZVc28JIYQQQgghhNyJKIBCKgXHcTgdGI+/cw/C5XJVyjXybRYAQEyRfPBfHf6XAgMleHfPahy1XsbQr96s7q4QQki1+OLPdfjnyH637XNWfYPun07CtF+/gM3pwAdrfgYbyOFY/hWhTa7JXGH9sNnsyDOacOjMefxrOyxsX7VnE5TaglkinJeDPdj87Qe4sOk7MAxTbLuOTRt73H4mN770F7sNJKVmwpZnBwAsGfO82/7gQH+MfXAQFArKWEsIIYQQQggh5NbQvyhJpbBYbcJysynjMW3IQxh3z6AKvYbZZYUCSoRo/WBKs0AfqkWO04RN5w9hYOMOFXqtO92Xf6/H0oProI/QQhesBQCYXO4FdwkhpKYz5pvxfcoWIAUY2LqDJNjw+4Hd0LfUYX/mOdyz8hWPx+ebKy6A0m/yVDAt5bDnO4SZFABwNl8MZESwARV2vUJhvgFADch4mWs0QamVw8BpoFbRDBNCCCGEEEIIIRWPZqCQSmEyi4PzIU0C8PXVjRV27rX/7cADH8yC058Fx3K4t1tXZG/IFfbP2/lDhV2rJti8+zC+SdwMfYTWbV++3VINPSKEkOrz2sIvhOXE3HTJPk7v/TjOxU8FiU9PQ7OXxuHFVR+Xuy/2OH6GZtHgSVH5V834fMykcl/nZrXDwir8nNUhL98EhUYBjZyCJ4QQQgghhBBCKgcFUEilMJndB+azzMYKOfdrv36JNEMuNIFqqB1KBPj54OzmVUjcnya04bgy5DypwdIysjHus/e87k811oBbkAkhpAz2qy8Iy3svnxGWs3LyIAuQ/iySWRn0cjVHziUj2jsbAAA2pR5BYFM/HLNdKXdBeWuOXbLez68Vrm5NAgDUc0XgyLtfIMDHp1zX8KRv1zboaGqIwAtixMiYbIbCfmf9LMzNM0GhlUOv0FR3VwghhBBCCCGE1FB31r+USbW7lpCMnjMn4a21XxfbzmRxTw91NvW61/YpaZn44Ms1pauXUiS1u0HJz6pQq1XoGNJI2G512m8+6q40d8V3iO0ZKawbk/nUM+ZM/u/zx8Hd1dIvQgipDizLQhcsDrZ/vWMDLqXc4Jf/+Be6IA3CnP4IveQLl4PFqz0ewtvPPIlj85dhbL8Bbuf798zBW+4Lx3FwWJySbTMfegLbFyzGt4NfxYpnpt3yuUsil8uxcNIzaBfXQNjGOliwzJ1180GmMRcyuQw+avcZloQQQgghhBBCSEWgAAopkx82bQEXDWzNOF5su6IpvApdT0/12v7pWR9gdfpWRA5/EHVGPyKpoVIUx3HQ+4qDXw+17iksL58xBWmn+RkVt1tqqkXf/YS+C6biSnpSlV7X108nLGtsKux//RMs6jQRbRR1AQC/x+/B4YQL3g4nhJAaxWyxwpQmfj/k660Y9+f7eG3tcvyYth0AoPfRYs38t/DTiDcxqFVHMAwDmUyGyLAgt/OtO7j3lvtisdqg1CrAOvlZLHW5CABAnVqRqB0dXmIh+IpQKzwM8btT4HtRA87FgZXdvgGUJxcuwEOfzZbM+snMzwMA+KqLyb1GCCGEEEIIIYSUAwVQSKn1fm4S/nEeEtZvpKSj9fiJWLt7p1vbPJN7kd3ErHS3bYVcOha6IA0a3lsb0b3D8OW+vz22W7n2b/jE6SEzMZjU8H6M7thX2GfQ6xCm8QNQcenCblVSagaenL8AGXm5OH7mEv4w74MjwIXX16+o0n44wQ80Bdl9sOGF+VCplGjfojHseeJdzyeSrlRpnwghpLoYTRbIFDK4cqWzHfdknIHGj6+joVWoIJPJEB0RKmkT6O+LS/8mAADqOMPhMDtxJT/Z43VmLFuBTgtfwJr927z2Jc/Ip5/Sciq803IsVkyYWo5HdmseG9EfMwaPwQ9vzQDDMoDs9kuBuXj1zxg85zVc9ktBsiIbF9IThX1ZJv67PkBX8WnOCCGEEEIIIYQQgAIoxIPM7Fz0mTYZv+/fJWwzmS1wtZK2a/v0M/DpqMeHp9e6nSPH5B7AyDHne71mTKR0oCo+V6xnwrIszAUpwTadPQIAaBlVB/f37OF2nkhDMABg/Jr30e+9qcizugdyysrhcCI1PQssy5Z6YOnx2fNxOTAFI36chVd/Xi5sT7ZnoceySRj6+ZuwVVKasRMXLmP9gb1wsS7kWU0AgIHNO0juZp78yIPCssxV+Xc5E0LI7SA9MwdKnRwqxnPRdgCYe+84j9tlMhnWvT0X3w99HSufmQZTmgUWmfRzvHD25I+HtkHlp8Qnx//wep28fDOUWgVUjBI9OraETFb1P8kYhsH4h4bAx6CDjGMABnCwpUilWYW+ubgZ+WHirNQlm8TfHDkF33FBBt8q7xchhBBCCCGEkLsDBVCIm7lffAdnfQ4fHP9V2HbhSqJbuwZDagnLz61eItmXa+YHNWTJQOvsOgAAo9V7Wi2b0yFZNxYJfLw0/xN0mvMCjl25hGwbH5h5qvtgj+dpGB7DX1cpg93fxLjIUQAArDRJREFUhTVHt3q9Zmk8v+hDNH5+LB74dTZ6fTkFjyyfW6rjzqaJ9V7MEe6BkjyZGX8ev/XUL8V5dOU8LDi2Br2/nIqTOr4fQQbp3bktGteF7RAflDJabq90Z4QQUlnW7tkJhUaBSJ8gJOxJRcLeVGRfzRP2d/JvhEC998H4ts0bIiYyFDKZDIyLAeTijI1fNm7DgG9eQ8+lkyDXyoVj7C5xxt+ZS9fw7i+rkZKRhX0nzkCpU8BHeXvU75AX/CS0Ojyn0KwqdrsDLCum6VIZFMKyLc+B06brGP3RO4jPTBNukgjx9avyfhJCCCGEEEIIuTsoSm5C7jZnUq4DgeJ6emYO7pk8FU0eqOP1mFP51yTreRZ+UKNjk8bo4NcIR69dgcnuXhel0M0BlHybOKj/z7UDiGwXgtn/fQtLDN8u1CfA43lqBYcCRTKFeSpmX1p2uwMnfa4jsl2IsO0GMks87ti5i6h7T3SJ7T4+/DuOHLuA4FA/TBk26pb7WRTHcQio4z74FxcW4bZtYIcO2MqeQK7F+8wgQgipCTiOw/AZ05FTi/9ueqrXYLzU8X40rBODpLRMPL3lQwBAtwbNSn3OwhkbdpcDaoUKX236F6gPcHIgtKn4HZVizEItf36W5f0L30JY20CsX7MfMoUMjIxBmMHz91lVk3MFARSnA9U1nyPfZEb3eS8jPDwQQyM64qtzG6EL5euetVHXQ54zH5eQghuaTIz59V34OfjgU6ivfzX1mBBCCCGEEEJITUczUIibDDZPsv7K4i8kwZOsS7kejyt6l21aXg4AwEejQ6DBAI7lYHZ6v6vVzkoDKPHWdGw5fxQAwDj5FFMZNrFf3gacaoeGS9Zz8289hdfxc5c8bi9awPZml6/fwIs7PvO4z5zJB3NYk3j8btcZ/JG8F9svHL/lfhb16tIvPG4PMfi7bfPXGQAAuZbypzkjhJDqtnHvQTz+8buweJhBkZKehbRA8burQ1wjdGjVGH6+BjSuVxuzWo1BT9/mGNKic6mvVxhwsDjseHPZlzDW9/wdl56fAwD4a+9eBDXlQxMyhfjzq3ZIWKmvWZkUDD9rprLSS5bGifNX4FNbD5Pahp+ydgjBEwD48Inn0adZa0n7PDV/s0WAnmqgEEIIIYQQQgipHBRAIW5kqoKXRUG5Dxcj5kP3lelwf91ukva5V/gZDEabOBC///pZAEDvJq3g52uA0+aCtbgAitPptm3W9m+QYc6FiimYKMXX90X/qLZezxMZFiRZP3j1HPLMtxYg2HbCc1CjuJk0Z65c87ovP4Xvh1aphjlDeo5Pd3vPk18WG+MPAwAijQHIOJ8jbC+8+7mowpzxRf9uhBByp3p62WJcU6dh4PLXMGX1UthdTiRlZGDmT19hy9GjUBmUgBl4v+dE6FUaybF9OrTFnIfHQV6GOiQK8AEHs8OG9YkH3Pbb8/gbA1Jzs5GZnYv3Tq6BQuM+8Xd8V88pKauasuC7Nq8avxMmfLao2P1Du3eG6bjYP65ggqivWleZ3SKEEEIIIYQQchejAApxUxivQEFt8aLFxwfVaY+4IrM8Qlhf+HB8Co3CguX5JjOyVSbACXSMawwfvQ4umwv5ditGL5yDL7evd0vZlR7Ezy65sllaayXbnA+DVpofPtJfGiQpKjTIX7KeqzXj0VXzin/AXlzPS/O4PbfgcXqSbxUDIx0MDXFtWxIAIFYZCoeZDxI1CIzG03UGw5Yn3uWbV85ZICzL4tiVi4CagczO4McpM1Hbzo8s3Rvp+Y7qwIK6KOZiAkKEEHInmPv9d4jpzM/k4BTAwfwLePXXZej11mRsyz2BL66vBwAMqNsWHRo2rpBrFs7YePjHd6DQyN32Bzr5z9iN5w/h2FnPMxoBwE+rr5D+lJee44NKl9JvVNk1XS4XsvL57/+Vm/5BUAd/j+06+jUEAPj7+eDgx0vR0hIn2R96m6RBI4QQQgghhBBS81AAhbhxMWKKKY7joFSLd8w+3+c+RAaIAYwnOg2Ar5y/8zM+iw84XLqeBH2oBoHwgUImh69BB5VBCbvGiRt+Wfjm/Ga89efXwjkOnj8HhZoffDq07AuknswS9uVZTfDxld5ZGurnfaAkKMAPGedyJNty5d4DHt58/NdanFBclWzLT+WDHMUHUMTaLXMefBLTh49Bxt5sLLh3Ip7vei/yT5swfeCjmPDwMHwx5GV80fdl5N0wwcY5vJ6zNO6dNR0vbv4MvlF6qApKG615dxYm1x2JqUM911cJ9POFy8EWm1qNEELuBL9e3um27UjKRYS3lAbcOzdoWmHXtJnFz21dMB98GFd/AFJP8LWyojn+2leTU5BpNLodb75hweSmIyusP+VVPzwSAHAhKbGElhXn4flzcN/qt9Dji0n4+upGj21MaRa8++AEYV0mkyHGVzqrUiFzD2ARQgghhBBCCCEVgQIoxI2L4YTlActfQ6IfPxjU2Z+/a7dT6ybCfo1WhbggfkbKjC1fAwAupiRCppAhWMOniPIx6CT53gFgT8YZpBqzwbIspmzn63bIXAzC/QMRaxSLtl9MuwGuSH8AICrQ+wwUAMg5nIec6+6DVWXx0frfJOvOeCeiHcEAgMTsdE+HAABMBTNQOgc0hlapxsSHhuHMV18jIigIk8c+hCNLliGioP+tGtdH47q1wdpZsIy0roqL9V5nxZNklRh0KkxB42PQ4b6+3bwdggA/H8iVMqQiB+fT4j22cbEsOI7zuI8QQm4H+0+cgS5E47adu2mTPkuN7vWaV9h1Waf75/TY3gPx6egXMS5yAMb2Hwh7vgNGpwULj60p6BRgzuAD7SM79sB9Xb1/Rle1mGA+KJGal11l17wqS+UXGOl2524HTOkW+KVo8fu42W4Bkvt6doXTxqcX1TpUVdFVQgghhBBCCCF3Kfdk3OSux8nEAXMr7ELtkdgQPlDi7+eDk6svoXaPCHSNa4ZNOCgey3HYfoWvHRLjxw/GaDVqj9d58IfZknUZww/8/++xEZhzbDXUPkrsOn8SDtYFgIE11466iEDryPrF9n/f6s+w+o/N+BuHAAAqrnwv8y5ZjfC/F0bg2U8+BADM3bEaAxq199i2sD6KRln6AR3Gxaec6bFsEkY26IYG6ijMO/4jxrcaiLGdBpbqHBwr/s2GNSxdEeTwkEC4HCzkShkm/L4YL7a5D18f3ohpfUehZ72W2HXsJN44sBItg+pg9uCxCNBSkV5CyO1l3b49WHjiZ8jkMvQNbQ3fPC22XzqBeGUaDGH87MUBhjaYOHAoQgIrNs2Ty+E50N2vazsAQFJqBhwWJ5whLmi1/PegTq5G4t5U2Jo5MH5U6T7fq0pUUDBwA8g2l+8GhPLyt+rx56p3kJKWibCQQEka0UIN6sTAcdCODMaEY58vq4ZeEkIIIYQQQgi5W9AMFCLYeuYo3vxrBWQq98EKAMh3iumpTv/0FVY/9gb0Kg1a1a0nbO+5fDJOsNcAAB0bNAIAj4MfnsgK2g3t2wX1E/lgTY7ZBBf4u0w/G/4ifpg2vcTz1IoKw+MjBwjraihLdf2iOCcfkIh0BmH+a08jKjwEWzYdEfY/unQePt3uXvjdYufTYelUnoNGnsg48W3464VdWPTrz2BkDFae2ACO43Dk8gXM/vtbr6nDLFYbGDn/3OVdNeHp3sNKdV1/XwOOfXVeWP/oyO/IY8xYsPUnAMD8NasBAMczr2D4tzNhdzlL/ZgIIaSycRyHBUd+EtZHd+qNSY8/iN9nz0F+ivh99erDj1R48AQAnHb+M5F1srDm2tFf11qy39/XAIdJ+rkZqw7D3pWfYM34mQj1D6zwPpVHeGAgWCeLfFvl18VasWE9eiybBF2QdJqQ5owC03rxaSfDQ4OK/f1wYNVSnP/6G+huqpNGCCGEEEIIIYRUJAqgEMH0LV9hZ9IpGMJ1sGeLud1Tjmci55oRgxp0ELZFhgWjSf1YAMAjw/vh+s5kt/M1jYn1eJ2001ketxetvfLiEyPhtLlgcVrhLAigNKwdU+rHolaJQRPZzblBSiEsmB9s++jxF4RtL9x/n7CcIE/HT+e3uR1nFgIo7ulkvJG5pP2zNxCfh57LJ+Pl/5Zic+IRfHtwk8fjk1IzoPZVQWVUYP1Lc4UUXiVhGAafvPKS23YFJ8PLKz5BXqx0EG3v9dOlOi8hhJQGx3FYf26/EBx2uJxwsq5SH//eLz8K6SEfCuqJBuHid8TTrQfj/J/XsaT7s5VWHyPvigmX/k3Ak6H9sX/qJ5g+5nHJfq1GDVee+Hhyrhnx7n3jERYSiMb1aldKn8rD10cPl52FzVW+mlxFpeXneExJOf/3HyXr5/64hgcCu2HjhwvRvUWLUp2bYZhS36BBCCGEEEIIIYTcKgqgEEHRUhc+Gi1Orr6I6BtB2PvOx/j+8TfQrHacx+MYhkGj6Fpu22P8xSKvhQXYAeC14aORdTnPrX24U7xD2N/XAI7lkM7kQR3Hz+bQKks/q0MSQOHK/jJnwT8ZRQfe/jd2hHs7TjowZHWUfQaKli1d27NJ1z1u/+/kUSjUctTyDUFUeIjHNt50aNXIfaML+OfoAbfN3x/Z4vZ4CSHkVlgddvReNgULdvyIp39ZjBxTPvqumIY+X07FykP/YvvVEyWe49etOwAAAVYDnh95n2TflKceRupfa9G6cYPK6D4A4JdP38ageh0wZtg9HgfyGYYBisShN7z+HgJ8fCutP+Wl12rgcrCwsxUz2/DPw7vxwOq38eOxLW775Iz4vRzjCkHa+t/w4gMjK+S6hBBCCCGEEEJIRaIAChHY88W7TrUKNTL++wPfz3oTQQF+aN6oTrHHBumkg0IPR/WSrJ/64bKw3KNlC+iCxRkag5xtoTusxKIHnhG2+fsaoNRKa5eU5S5itUqsQVJcEXSjzeIxKMCC36aUi9f00evc2lmd0jt1rQ47AECvKf0MFD+l+3lteXa3bddyUz0ef/gSn4ara4Nmpb5mIU+Pyc46JYOBURf5wNa5rHhsunC4zNcghJCbzfv2O7AM/9mcZM7EsGVvCvu+PrIBMzZ9hWNJl70dDgAIDfUHAMweObayulmsDq0aY8V7r0Ct9l7zypovfpZrFbd3sXODXguX3QUnVzEBlDe/XQkA+OLQerfvYR8/8bunebTnmzMIIYQQQgghhJDbAQVQiMCWJwYDdCoN5HJ5qdNjhPtK88s/O/heyfqHr7+Aq1tuwJxpRbR/KDR+/ECSn02H158bg3+/eA+RocFCez8f/a0+DACAQiHHhfXxAAD2poGb9//5CT2WTcKGkwcxZNUb+Oawe2qswmPkRYI2Br0W8btTJO0sBTNOChUGVAwa98CEN6oiNVqubUuCYgcws80YJKxLRvwuMTWa2WlzO5ZlWZzVJQIA6kVElfqahXRaNTIv5iLvhglpv2fCmGSCnXFAa+BnxdRLicD44YOF9rsunCrzNQghpCiO47DNeVKyjdG7/xxZc2ib13PYnHbkF0zvCDLcvrM62gbXR8b5HCTuT4Ofpnzfa5XNoNPCZWeFtJnl8c2+jZD5iH/Ti5k3JPsLb1IAgKigoHJfjxBCCCGEEEIIqSyKkpuU39ixY7Fq1api21gsFmg83LW/d+9ezJ8/H3v27EF+fj7i4uIwevRoTJs2zWN7cms4joNMIQZL2obVK6a1u5jAEByyXAIAaHNVboGX5x67T7gDVSVX4MR3FxHUwA8/zl3i8Xw339FrPJxfpv4wDIPU45mo3T0cnE4Ls9WKkd+9DZZhYXHxdwTPXvst5BFyfH/kP4xtN0ByPFdwZ7SySABFJpOBuSnz2M0BFFtBAKUsM1Ac+fzdvsoMOY4v/RIBfj4AgCF9OqPOiEeEdi4Fi/PpCWgYIub5f/L99wB/frl+WNkDKIH+vmhpi8O93bvgwZm90GLaU3BGspBpZIADWDFjGpxOFw4/Px1tn2+MvclnynwNUnWup6fg55M7sPfKGaSzuQCArx6YhrqBkdXcs/LLsebDoNJWWj0LUnUe/2w+CuPGDrMTSh3/U6SFMw72BDvOxF+DtZkLB9nzbsdmGHOwYPNP2J9+Dih4Wd/OgYllc6dg8+7D6Nau+W1fr0Ov04B1sHAx3mdtlsba/3bgy8v/wC/GIGw7m3wdDYKjAfC/NxgtH1zJTchH/1HtynU9QgghhBBCCCGkMlVJAKVQ/fr1ERoa6nGfzEPh6++//x5PPPEEXC4XoqKiEBMTg1OnTmHmzJlYt24dtm3bBp2u9Hf6E+/MFivkKjkcef9v77zDoyi7Nn7P7Kb3BoSEnoQOgdB7s6Ei2FGxIagIvtg/RcGKr1LsBUFRwYKo2EVBihogdBQp0ntPgNRNO98f+2bDZndDEuZJ9mTO77r2upIp99xn5nnOnNlnZ6YAKfmJeODOayu1fuO6scBe+99Trhrtdpn7bh3m+HvBf58DgVC/brTbZQH7Y6z8Qn1hO5SP36e5H2gpj4UfvYyn0j5EMRVj5BtTkR3m/FL03DwbghEIguuXRe7egQIAzerXR/45v5zNKTOAkl9kHwwJDah4u3xwxHW47O7H8Ov7UxyDJyW8cs8Y3PPqK2hyaRys/haMWjAdv49+xTF/99nD0MLtHmNDK/8rXl3X8enrTzn+9y+2D1wFxPjBUqRD0zT4+Fix+YfZuO3nl5GPQvxzbC9a121c6W0J6rn5vRehxzjn0ge/eQff3vlcDTmqOrvSD6NucAT8rb648cPncbzwNC5r0hmPX3TT+VcWvJKi4mKMn/cW9vkcBwDkbc7DWS0HdVpHovBoIaY8erf9xetFRUh+fDRskQUopmLomo6dJw8htzAfD3z8FvLDne+QCPL13h9ThIeF4NrB/WraRoWwWq2gQgJZCeO+egOvXX0fbIUFmP7bfNzV63LUDY44vwiAxWnrgDKn9j3HjmJbvf1oFlkfE+fNRmA9+zF7ostw1A2PNDoUQRAEQRAEQRAEQTCMah1AeeKJJ3D77bdXaNm9e/di5MiRKCoqwssvv4yHH34YmqZh3759uOSSS7BmzRo8+uijePPNN9WaNglns3Jg8dUR6h+ENx/+T6XXjwwt/eK/XVKz8y7fv0eH8y6z9ZM9QACw+4dPERJc+YGyVomNQSsJxSCs+msLGvSu6zQ/uIlnTdIAFJPLL4aDfQKQjmzH/7kFNqzY9Q/S887iitbdUVBsvwMl0K/iX+gN6NERtr9dHyMGAFdd3AtXXdwLTe64CQ2613WZn6Pnw7/AD5M732HIr5sj/UJhQyF0qw7fotL00CiuLgrzimD1t+DomXQZQPFCiMhp8CQ/qwC+wT7IKMjCqZyziAwI8fpfwJfw1o/fYN6h5S7Tf96zBo8U3yB3oTBk+8kDeObzj3HQ96R9QhZh9evvYMQjL+Dnr1bj+M8LHD+ksFgs8Cm2ABpw1paDk6fP4NbPXoI1yOK44+5cdE2eRmoUOtlzxKZTuzHjzx/w/h8/ITA6APt+PI73bniwQhrk5/qjhK93/4mvd/+Ji5um4LdDGx2P8bzhygHGmRcEQRAEQRAEQRAEBXjttw5TpkyBzWbDxRdfjEceecTxxV+jRo3wwQf2F5O+9957OHbM/Yu1hcpxNjMbVj8dPnrVxtSaN22AzZ/vRKPDMYZ5OrHqG5xY+g2iI8OrtL7FooOKAQIhIjz4/Cs4Qf8bRXHmoZuux/bv9uLUhtMAgDX7t+H/fpuFl1O/wLhP30BO/P8e4WXwL6KfvvxWFBcRtHN+eP3mjwsQFB8Av2If9O2SbMh26oWU/sLY95y2oOs6zqzKBACcOHvGkG0JxvL+Tz85/i60FWHW5Q/CshWABgybOwl9Zz6I1Qe21ZzB87D76GFMXzIfeYX5+HTPEpf5J7ZkAADmbVha3daEKjIvbSkmLvwQH/z2M0Z9Pd0xeHJo9XE8O+B2AMCcKRNwYuE3Lneh+sH+BfuQj5/Cnd9NtQ+e/I/9qUeRm2G/+6++Te5eMJKi/NI7LD/buhSB0QEAgMK8ir0XZcVf/+CfkAOO/3NOOd/5uWzXJsfgyYyLKv9jDUEQBEEQBEEQBEGobrxyAIWIsGDBAgDAyJEjXeb36NEDLVq0QEFBAb799tvqtlcryczOgcXXAn+Lz/kXdkO9OlHYveBTzH7yUcM8+fn5urwLpTJYLRZQsf0OlPyAQo/LFZ3zSK4SSCO4ebIXendph2M/LIBfkX0/fbxpsWPepqzdjr8jAio7YFM+t117KTJ3ZoMsQKYtB7l5Nnxx6HcAgO5j3F0FZ8+U3l0T4e/8OLGLuqYAAI6fyTBse4JxLPtnIwAgfdMZTOt/N9o0b4rkes53g72y6MsacFYxhn/8Ar7ZuQIXf/AYLL72L8tz0/Nw9mA2bqjXF3e1vwx5p22Yse5H3PPVKyAi2AoLsP3kASzbthELt66u4QiEErLz83Dr7P/irU3fYdn+Tfhw16+OeQPDk7Fr1ifol5zsmObuzqhoH9cXwzc6FIMzv2fi73fex4Jbn0av/FaYcfsDSmIwK7acAqf/87P+9z9V7L0o//fODMff2Sdy8WQP50fu5Wv2c7E134KWTRpX3aggCIIgCIIgCIIgVBPV+givL7/8Et988w3Onj2LOnXqoGfPnrj11lsRFhbmtNz+/ftx5MgRAEDPnj3davXs2RPbtm1DWloaRo92/84NoXyICBlnMrH76BFM+/NL6FYd/paqD1iEl3l/R01jtVqQn12A/OhC+IX5Or2s+O/PdqLt8AQA9gGUI5npiA0555fMGlDee3SPZWQgGJ5fXBzsG2BIDOcSotk1h3z8FC4NTnFM7xXZ2riNnPNj4clXOQ9ehgYGAgRk5uYatz3BMPZlHYcWrmPxf6ei3v/a8vibr0WX0fciLjAavj38cCjgFAqLi2DVLcgtsGHh1jXom9AeBcWFFX6/gQq27z0Av3Dn3FNcUIzVj72NnNw8BAUGgIgw84Yf4T/ID1tO7cfv2//CU79/6LROYp14RAaGICLAu3KRWdhx/CCe+vFDHC445Xb+kLBuePDa6yqk1b1payyktY7/m52uh9nPPOb4PyQwEJPHjroww4IL+WedB1B8g+0/FsjOz3O3uIN3fv4Ovx3egOjIcJyE/W7Fj697DK2TmmDigo8Q3sq5T16d6L62EwRBEARBEARBEARvo1rvQPnxxx/x7bffYunSpZg3bx7uv/9+NGnSBAsXLnRabseOHQAAPz8/1K9f361W06ZNnZYVKs9vqesQlTwEI2a8iENW+xdeQT7e+zLeymLRdZzZlwXoQFBMAJBfOu+bF51fqr3p8E7H30VFRdCsGjQ3j/AqIfu450GEwvRCJe+aaBRQx+6PirHm5L8AgK7W5ph03W2GbeOVR+4DrSjEm/3GIj7a+XFs/j72L7hthfnuVhWqQGFxEQ6cOeH4P7fAdt4vKgHg9e++Rp/3HsCNc5/HS4s/x8bDO5Hrmw/kw2kgpFF8PRz7aQHWfjEDp/49A1iAAbMexr8nD2LcO2/glVVfYejcibju02dxKueskhjPBxFhzOevAgAOrTmO7d/sxf7Uo5g28B5omoagQPvAoaZp+OHlycg6lgMAeGrJhy5ad3w1BVfNmYj3035GboGtukIQAPyzew9GfjPNafCk6HgRzmzMdPw/ZthVLo/q8sSgzh0df39343OY/ehj5SwtGEWI1f3gf25R+Xn/0/1LcLzoNPT69uMbWxSB1klNAAB1LGEuy/dp0+4CnQqCIAiCIAiCIAhC9VAtd6A0a9YMkydPxuWXX44mTZpA0zSsXLkSTz31FNLS0jB06FD8+eef6NSpEwAgI8P+iKDw8HCPX0RHREQ4LVuW6dOnY/r06Y7/z549i19++QUAEBYWhm7dumHVqlU4c+aMw2PDhg2xdGnp8/W7du2KnJwc/P333wAAHx8fDBgwAJs2bcLRo0cBAPXr10fbtm3x22+/obDQ/miKdu3awd/fH6tXlz5SZsCAAdizZw/27NnjiK1r165YsWIFMjPtXzAlJiaifv36WL689OXJ3bt3x9mzZ/HPP/8AsA8q9evXDxs3bnS8/yU+Ph6tW7fG4sWLUVRkf055cnIyfHx8sGbNGofWoEGDsHPnTuzduxcAkG2zP7rK6lf6bHm9kJCbm4vff//dMa1Hjx7IyMjA1q1bAQD+/v7o27cv1q9fjxMn7F/+NmjQAK1atcKvv/4K+t+jPjp06ABd17Fu3TqH1sUXX4xt27Zh//79AIDo6GikpKTgjz/+QE6O/YvRFi1aIDo6Gn/++adjvV69euHkyZPYts3+DofAwED07t0b69atw8mT9ufqN2zYEC1atMCvv9ofF5NfUIjc9HO+RC0sfVTXsaNHcC4HTpzA8qPLMXXNT7AFFyMwJgDWIoujzQBAnz59cPToUfz7778YlpyCXXD9wvnsjiz89Mhkp/U6d+6MgoICbNy4EYD9BcmDBg3CP//8g4MHDwIA6tati+TkZCxbtgw2m91z69atERoaipUrVwIAfPJKfxl83M/ebptE2ftIWloaTp8+bZ/WpAmaNGmCJUtK3yPRpUsX5OXl4a+//gIAWK1WDBw4EH///TcOHz4MAKhXrx7at2+PZ24dhiN7duLInp1o27YtAgMDkZaWhv379gGxQG5+Pnbu3Ildu3YBkP5U0p8iIyPRuXNnpKamIisrCwCQlJSEevXqeexPb25YgjOR+Zh19UNYm7YB7x5aAhQDL3Ydjp4durjtT2lpq/Hl0T8AAIdzTuHw7lP4cXcaAuv4w5qtQdM0t/0psrD0i9G7vp4GlBkr/ffEQfiezPXYnwAgJSUFxcXF2LBhAwD7oMbFF1+MLVu24MAB+3sPYmJi0LFjRyxfvhx5efbBoJYtWyIiIgIrVqxwaJX0p4VL/4QttBBFp4uQNvUt7Pl3B9LT03F6335s13QkJCRg8eLSR+X1s7bBWux2nL1O7TgD3yArQuqX3hH20aZf8dGmX/FM4hCsPbYbjUOi0L19J6f+BAB9+/bF4cOHHQPxISEh6NGjh6H9acmSJSgosPfdc/tTCf3798f+/fvZ96cZfywFGtqnFxcRGu8PwdVdO+HfyHS8segndE5oij+W2tetSH8KDQtDwF9WJDWqg7SVqQDO358APucnQE1/+vdf++B6cHAwevbsiTVr1iA9PR0A0LhxY5f+VPb85OdhnMRWXOD2/PTHij9RWFDsqNWOn80AonQ087f769GjB8L9/HEcOU56uzZtwZHN9h8uSH+y423nJ0D604X2p6rWe4Ccn0qQ/mRH+pP0J+lP0p8A6U8lSH+yI/1J+pP0J2P70/nQiCr4YGsF5Ofno3fv3li9ejUGDBiA3377DQAwZ84c3HrrrWjQoIGjI5blgw8+wMiRI9GsWTPs3LnT7TLnEh8f72jMgp38/AKEtBmMrg+3cUxLDI7D+zc9XIOujKX/3Q+gKMV1+kspd+GxdbMc//eMbo2H+1+HYfOfdlru99GvuNUlIjS6+UY07ud8h9R1Tftg3KBhF+zbHS/N+Aw/as7veXh/2ENIjIlXsr2yvP3FN/j89HKkhCbilRvHVMs2azt93rO/v+GelCvw3g8/oDjWPv3eDldieOcBAOx3pWw7sR8d6idix7GDGLlgmsd7B0MLA/DDmMlu51376CQcT3Ad9Du46hjiu9UFAIxOuRy3pLg/ceQU2JCRfRY5hTYkRhvX5p6fMwe/5q5Hz7BWePGG8z+SqaCgEE1H3IQmA+LQwdoMY/sPRVCgP+6aPBVbdu5FWMcQhDVwfgdRgO6LX+56yTDPRkJESu5Yq06Ki4uR8vQ9CIkPQrusxnjpnlEICgysaVtCFVj71zZc/MCjCI0PQsOe9ZzmfTviWafH4z364Qysyt+GkU0vxfu7ne8kHpN8JW7sYs9htz37IvbUO+40f9moadA1r3wNnyAIgiAIgiAIgmAyzjduUKNXr76+vnjuOfujlJYtW+a4m8Tf3/7T6Px8z4+MKBnFCwgw/l0TZsHX1weJjeOcpt3Qsm8NuVHDVf1Kn7Ou5QE7ftqPE0vT0SKhIda9txWHVtu/1Ek9+Q9ybWXaWzlPANI0DXH1ol2mB/ura4+jb7wCuxY5d+amUe4fcaeCAF8/AEB+UcF5lhQqQnZO6WPgPlz1i2PwBACOZJxCMdnvmLri7Qn4zw9v46u1v2Pkt6WDJx20puickYCDaaVfTOZbCj1ub1ByCmxnS9t4xqazGBE3EHPHPOGY9t66H10e5VVcXIyLX3oUl87+Pwz/YjJGfj2tQo8ZOx9Z+bn4YsVS/Jq7HgBwVUrF3ong42PFylfexIwB4/HanWOR2CQe9etG46fX/ou9P36OD65/GPu/db7DLLc4H7vSD1+wZyN57qs56PPeA+g780E889NHNW3ngkg/fRY+wVYgk/Dmg/+RwRPGdGrXAum/fYeOTRJd5m0+usfx96w/fsKqfPsvxn7evdpl2ZLzBQCcynAduJXBE0EQBEEQBEEQBIELNX4F2717dwD2L+l2794NoPTxXKdPn4anG2RKBltKlhWqRmDc/57jQ8BnQ57AxR0616whg2nXpKnj71eHjcFrd43FktemIyIsBNlrF+LpwaXvDzl+NgP5WfbBgeNp6Xh5YPm/hj+V5/qlUFCAunfIRISFoHujVk7TLBV8n4ARlA6geP6SvjazaedO9HnvAVz+wRPnX/g8HDx1HNd/UPoenjwfe7vzO2x/nN7y3X+h38yH0Oe9B1AQYL/N8LX1CxzLF+QWYvrI+zDtsfuwbvq72PbtXgBAqOb5i+u7h1+Jl3rehZc7j8LN0f3x91vvY9TlV6BTu+b454tdjuUe+u5dx/tDFm1fh8nff4q8COdBs5+3uH5hWhkKCwsx+M3H8ebm7wAAvtlWdG3WssLrx8fWQcuERm7npbRtjrnPTED2iVwUFxYjd519oOqz1UvcLl8TLEj9A4tOrXf8/9vBjSgsLqpBRxfGoWMn4Bfm63R3gsCbIN/Sc1n6Tvut2xMWzba/t+nwcXy8dZFj/vGc0y7r65bSu6o6NEjE0Y0nEbtX6jVBEARBEARBEASBHzU+gOLj4+P4u+SZaImJ9l8+2mw2x/PbylIy2FKyrFA1fEv2vwbE1Yspf2GGJDUqfdRQZEgorr+iP2LrRDmmXXNZH9gy7b/KfzPtW2gWDWGFQdj2/kfo1qqVi965ZB4ufaZ7yUBf92blr3OhtGtcOiDkV1AtrzByEOhnH0DZazuGbSf24/DZk9W6/aqy9fh+ZNpyz7/gebjrlakAgMzCXBzLykB6TuZ51vDM9TOeQ6av3VNBTumA1MjOl8F2Nh8ZepbHdd/qOw4r//OGY/AsIiwE7z/0CIq3FGH60Hs9rqdpGvp174BuHVrh7quHOKb7+vrg1OLv0KfQ/ii/3WeP4P++nYlvU1Px3PK5+PWY/RmncemR2L3YfgfUjFU/VOkulPz8Agx580k88uV7QHDpF6zzRz5l6GOsenVqi0c6XI+p3Ufj+ZvuRG6GDcv3/eVxQL66eeH3TwEA+gHg9D57O1qzf3tNWrogZi39GZqmoWFQnZq2IhhE2uotjr+7NCgd3Nx56hCOn3J+91xhoP1uuX1/lN75VUCleW3iuFsx/Zp78eZD92PP0sOgHXwHCwVBEARBEARBEATzUeMDKCUvewHszxsD7C8LqlfP/uzt1NRUt+uVTO/atatih7Wb7gn2L/zr50fWsBM1RIaHOv4O8nG9OyQwwB9nD2YDAHbmHoZPgBXWCj5apE98W8ff7fIbo0deSzQIV/sF4hP33IxW++KRvvo0nh94p9JtlSXA1xcAYKMCjF7wCsbOf6Nat++J/IJ8vLP8OxzJTHeZ997PP+Dub17B6C+mX9A2Pv7zVwS0L30823WfPouhcyfiryO73C5f3hf1B48ehzW6dPDrIp8OsJ3Nh3YIGDqwF2IPRCA/uwBFBcUoyrd/0bj96704eygbD7S5Bm2bN3XRvHJAD/z56utoWKduVUPE1b17I2OP/a6qDem7MO2fL53mP3XrCPz70RycWZ8Jm16AeeuWwlZYgKOZ6bjvo9dw7xevutxFcSYrC1e88QRu++Ql2Arz8X9zZuK0bzbWnd3hWKZBUQwiQkNhNDcOHoBuHVqjf/cOyD6QC5tW4LaNVDfrtmxHYJQ9Fy149FlcHGN/SdN3G1aUt5pXs+mY/QcND11+bQ07EYxiWLdejr/bNmjs+Ptk1hn8tmeD23XeGTve8XcRSnOgj48VVwzsgZiocCyZPA0/Tfqv4X4FQRAEQRAEQRAEQRU1PoAybdo0AECLFi0QF2d/H4emaRg2zP4i7vfff99lnRUrVmDbtm3w8fHBkCFDXOYLFWfyA6NwfUBvvD+y9rw4/lwC/Eufwx7o6/7xWoV5zl/6WjVLhbSnPH4vMnafRW66DS+NGo3/3j+66kYriNVqxbsvPITNs2aja4uKP/LICPz+N4BSwsmisziRfbpaPbjj/2bNwmfbl+KGz55zmfdW6jcAgEO5Vb9bJis7B7O2/Ox23rfrnb/03nnsICb98CEGvvsIPli50O06H/70i+Pvm+r1x7P33Yn3Bj+A7x9+Hv7+fpg/9WncXGcAZl36AD4d+gQ+vuxRHP35a/zxyKsY1qOXW00j6Ng2CR/d/Bi2f7fPaXr9gxHoQs3Rqn5jBAb446FLrwcVE37eshpXvPMErv/sOfxt24t/Tu/Dc4vm4nRuFhZuX41DZ0/i0Q/fw1m/XOzJPophsych7dQ2J+2cI3l4e8T9ymIC7HfYtIxqAAD4flPVBimKioux5uB25FThrhsichpYemXRVwCAvpHtEBESgktSOqG4qBh7Ttl/vb//6DEMfOdhPPrDe/hyw/Lz3jXz195d+M+XbyHTloPftq939Mn8okJk5Fb9LqnyKCouxqYju0BEmP/HMiBOA9kIjaLqnW9VgQmP33uz4+/E2NI7OSf8/AF+OJrmdp3+HTvg7K+ZOLElA5e3cP/jlsQm8QgJlnfkCIIgCIIgCIIgCHxQ/gygRYsWYcmSJRg9ejSaNGnimH7mzBk89dRT+OyzzwAAEydOdFrvkUcewfvvv49ff/0VU6ZMwcMPPwxN07Bv3z7ceaf9l/d33XWX404VoWr4+FgxdsTVNW1DGZqmYd8fRxDdIhwBVl+3y8SdjULxOf9b9YoNoDSoXwcDA5MxrH9vBPvX/i+EGtavg72vHUbjfqUvrn/0m/cw++ZHseX4Pqw9sB3Xt+8Hfw/7WRVrLaV3M+QU2BDoYx80W7ppA8Ialr6TwVZYAD+rj8v6JRRTMbLz8xDiV3osp/3wBb49vNLjOulZmSAirDv0LxpH1sMdX0yF5qcBFuDLv5fjzu6Xuqzz98k9QDjQ3b8l7vnfAHD7VgmO+T4+Voy72bVPVseXjh3bJGHD++8h5Z67kXBJAzzT61YMaNPRaZl+ndtj2rvzoTXQAD/n9Zfu24ilczaWTvhfU8hNzwMiAUu4BcXHilAQWAS/EF9c0qoTwoKD1QYFYFjHXnj38E/4feffuLv3lZVef+rX8/Bjuv29L36aD9rVa4LHB9yE6KCwctfLys/FQ5+8i60F+/H8RXcgwhKM/QEnAABDOvYAADSKqwd9q47Dxel4+Nt3sW3HfhSEFmHV4a1YdXgrWtRvhDZ1G7vV/37dCkxZNx8AcPlHEwAAgZofesS1xh97/4bNWoCmfrHYl38MTw+4DX2btat07GXZdewQ7vjW/ji7BiExOJBpj+fy+l0MfQybULOEhZb2y+jw0nZOZarG9F1nEdksFJRVjAAfP6yc+xayc3I9/mBBEARBEARBEARBELih/A6U7Oxs/Pe//0XTpk0RHx+PLl26oEOHDqhTpw7eeOMNaJqGSZMmYfjw4U7rNWnSBDNnzoSu63j00UfRoEEDdOzYEYmJidi+fTtSUlIwZcoU1faFWsBrt9yHKf1He/xy76dZzo8TqeiL2TVNw6sTx6Fv1+QLtciCxg1iERft/J6cA2dP4Hh6Bu755lXMWvczJv/2abW+Z2Lbfue7JQ6dKb3T5J5XX3Gal5mfg/IYP+ctXP7RBOw7fQyA/c6BcwdPTu/MxLjGQ3Bg5THHtHVndqDvzAfx4E/v4uq5T9sHT/5HIRWjqLh0aC4rPxcZuZk4esb+/oBxVw2raJjVSv260fjx2RcxttEQl8ETAGgYVxeZh7KdpmVvykHOKfd3Z0Rlh6BDfjPH//f0vhIhu/yRsfksRva+zFjzHujXJRnZx3JwOPdUldb/YVfpL+5tVIA1R/7F1Z88jWs/fAapeze7XScrOweDP3wCWwv2AwCeXDQb9y20P/auML0QnZs2B2AfiD32l93X6mPbcTbU+X09O48dxOGzJ1FMxU7Ti4uLMfnPz1y2m0M2LD64HjZrAQBgt+0IiqgYT/02G/lFhS7LV5S/Du/Gs7/NwS0fl+bLksGT4rxi/N81N1VZW/A+NE2D7az9/WDR4aE49e8Z14WOE2bd8CAaH4rB5yOeBGC/6zM6MrwanQqCIAiCIAiCIAiCWpTfgZKSkoIJEyZg5cqV2LlzJzZv3gwiQlxcHHr37o0xY8Z4fI/JrbfeioSEBLz44otYsWIFtmzZgqZNm2L48OF47LHH4O8vv3AUzs/QS3qXOz8wwB/FhcXQrf8bONHlV9Se2Lf3KBLa2B+HVJBTCAQC4754wzEUu2zfJqw+sA1dG1bP48Xemv8NUHqTCbYd3YfEaPujAHPybE7LZufnITrQ/R0DRISNefb3OLz6x1d45coxmDJ/ntMySydMQ1RYGC7q2gnvf/ETvqVV5XrL0/LRf9ZDmH3tI2gaEYsr33oSRUHFQCP7/DD/oEpEWr10bJuEjm2TPM7vGdsGG/fsRt5pG965YzwGjuqIpGtHIPASe06+KbY/Tu07g8Wr1uOTmU/gxKnTSLnzbvTs2Ba3jL4It/S7CERUbXcsxMfWQXEeoUgrPv/CZfhw0UJoEXafdbaF4N+AwwhvZG90x/NP4/Ff38d/L7kLv2xdg3b1m+Kadn0AAN8v9/y4sEmDbnX8HeDvhxsS+2He38tRt23pu6j2px5Fw571MH3VV0CZpjakeXdkHMqET4AVtrP52PLVHnS4ozlsmfnwC3G+AyznVB4svjr8Qnwx6dePcMaWhQkDbkZcaHSF98Hx0xkY+4N98McnzLlsyD6ei9evua/CWgIf6u+NwBk9Bwmj43FyfQaikpzzZ8dWiejWsTW6dWxdQw4FQRAEQRAEQRAEQT0aVefPxWuQ+Ph4HDx4sKZtCF5Kj9fvh9Xf/uiu/7Qchmt696lhR97J5LfmYurXX+CRa2/A2z9/iyZXxLksE+Ufiq9HPF3hL8c/SvsV72/6Ge9f/bBj8KM88gpsGPv1G7it6yX4+qffsc66E9hVDGqiISk0DlOG3oPIwBD0fvI/0BrqyN2Rh4BEf7w7dDxa1WnkVvPjP37FrK2l7zm5sV1/fLDgJwQ2C0D2iVxMv/Qe9E1Jdsy32fLRZvxIxHZw/RI66VB9pJ75BzGtIgAAHeo0w4h2F+HBxe86Lbds1DToWo2/hqpKnDmbhaMn0tG8WUPHtOMnM9Dh3tGITAjFpskzvS625MdGIbRZcKX3+1UvPImMmGwUZhRixWNvICs7B/UvuhbJtzd3u/z9PYbBV7Niaqr90VqZ+7LxWLcb8faR7wEAl9brjCeGuN6tsXXnPuTl5ePlOZ8jsk4ILmnfCZP//hy+QZ4fOwcAL/S6A52aNsfJ9DN4c+4CrAy3v2fG8hcw7vJh6Nc9Ga9/8hUW65ugW0r75A+3Po/QCgziPf/tHPx6bL3TtLZ6YxzbcgpX9O+Oto2bIqWt+30h1B7uePy/2NXkmNO0MW2G4MYe/WvIkSAIgiAIgiAIgiAYw/nGDWQARRAAdHtlLHyDfDCqw2CM6HxRTdvxagoLC2G1WrF15z5c/dYkxLS0DxTsXnwITQfZB0DeGnI/2tZrUp6Mgz7vPeD4+/fRr5SzpJ0X5s7FLznrAAAxOWE4EXgG19bvjU+2/+b49f3MYQ/i1rf/C986Pgg56I+sRjb0adwWz198p1vNi596BHkN3D/e6PvhzyEsxPU9HQl3j0D9FPsASq8zrfD+8p/Qq31bfPXCsxg54SXsaHTUZZ3MIzkIiQ2E31ErFk2sfY8gXLnuHwQF+qNdy2bnX7ia6fDoaIQkBOHXO1+q8Ht6iAh9Zz4IAPh06BOIr2N/hF1hYSHGPvs6Pv1tMZJvcx08KC4qhm6xD9K8MvBedGiSgCGjnsC+7OP4/b1XEREaet5tFxUVodv1Y3AIp9Cge134hdo95522wT/c/vIZ2182pL35tmOd9NNncdekKbjvpqEY2DXFMf3Q0RPoNuY+NLm0dICyWUQs+jRth8HNuyImKAwFRUUu7wjKyc3DpXMed/y//ft9iOsSg+/GPIf4qDrnjUGoPaSfPouBrz2CkNhAZB/Pxdav9yBr7c/QK/jIS0EQBEEQBEEQBEHwVs43biBXvoIA4MAK+y9ruzWunkdPccZqtT/Cp2VCI/tjvP7H0/fcjow9ZwEA9333OmyF+RXSy88ucPx97jtDPLFx9y7H3ycC7c/l79yqhdOji0YtmA6/WF9Ycy2I1O2PW/p979948Nt3AACnc7Mw5uvXsPnYXgCArdjuYe/yw2XMwe3gCQAU/M93eF4QJj8yCsd+WICvXngWADD9sfuwYuomHFxV+ovtUzvO4INrH8I9dS/HZ2MnnDdOjnRPae2VgycAYCH76S6vgu0SAD5ftsTxd8ngCWDvA+8++yCOLv4af760ESe2ZMBnp470Xfb2r1t02DLzYf1XQ0qzJOi6jh/e/y/+/vyDCg2eAIDFYsGar2bg0Jfz4fe3FYdSj+PzoRPw7uDxiFwbiEsKOmLplOlO60SGh+Lr155zGjwBgLh6MVg8ZSqu9+uNf76w959dGUcwe90vuO7TZ9Fv5kO46INHkVNgf+zdobMnse7Qv1i/dQcAoDCvEOH7A7F0+nQ83+sOGTwxIZHhoaBi++9t2kQ3lsETQRAEQRAEQRAEwTQofweKIHDg2ja9seDjP5E4Or6mrbCCikpvYGvesAEa5sYgE/YvYdcf2oHujcp/Nj4RwXYm3/GYoqNZ6ed9N0OO7vqy8nrhETi9NxPhjUOcpkf6B6NbYivsz/sdALD22L8gIgx+6wlYQy0Y8+1rWDZqGvL1QvjAB1d27o6/Ufpier2c9+Fc3qwLfl65Fq/+39Mu88JCg1H47xJ888sfeGj+uwiMCcD6F96Bn68fOrT2/G4RQR2W//1eILcgH8G+RQAAq27BpiO7YCsswIp9/6BP03ZoHt0Am4/tQYCPH5Zs2AAEA3Wy3L87JzDAH4U7fsNPS1ehb9dkDHzgIeB/40eD4jviyftuuWDfmqbh1xkvo7CoCMFBgahfJxrfvPdCpXWaN2uI5s0aIiQ4CO+fWuh2ma83/4EhLXtg+Od2/a6B9rtrOgQn4JV7xsDX1wetEhtXORaBNyVPZawXEymDJ4IgCIIgCIIgCIJpkAEUQQDw2qT78dqk+2vaBjvOfQBg50bN0b9DMr6zpQEAVvz7z3kHULKyc+ETWJqGdp08XO4AChHhlJ4JXzg/aig8IBi9/drgw5kLkTKq9C6iUL8g3Hb1JXj1oa9QP8V+B0HfmQ/CGmpxLNNv5kPwaeADyiHEBkY6DaDEw7OX15+6H8dPnUb9up6XGXpJb4QEByK2ThT8fP08Lieop+QOlFFfTMPZohwAwKuD7sX4xe84lvl6y5+IKgrBKUsmACAyLxgIBh666jqPurqu44qBPQAAdw0cjPf2/ASfACvuGXYlAvyNOeb+BukAwKCeHfHEuFmo2y4SuxcfQrubEx3z3lvzI95b86Pj/7Sc7QCAG7sPgK9v+e9iEWo/2v/eoRMcEFDDTgRBEARBEARBEASh+pCfEAqCUHX+N4ISqYfA38cXdYIjHLN+2bMO3/2zotzVT6Sfhk+gFYU2+x0BS7dvLHf5/YeOwSfKisK80keH+e+xIiIgBG9NGo/MtJ/QYGcUigvtjwIL8vdHSHAgnrrkZpzYklGudpg1CDdcNACFqwrQJ6c1Tn6djunX3+txeavVWu7gSQkDe6bIr/a9AP9s+wBAyeAJALy0aJ7LciWDJwCQHp0FAEiKbVChbYy64QrMuGI8PrjiYTSI8c7HXMXVi8HeTz/DL2NeQvrS73BTYD+0P9wY6TvPOC1X0ieLbEXo3rxVTVgVvIyifHtejQwOOc+SgiAIgiAIgiAIglB7kAEUQRCqTHGhfQDF/38vn77pqoFI2FkPuRk25CEfU1PnIyff9ZFbJRw+cRIWXwuKT9m/rP3twAZsPb4f2W7WsRXmY9qyL6FpGkIyA3Bg2VFcE9MLv75Y+jJ2XdfxyctP4uS20wCAuAj7AMdtV1+GXx55CftTS1/s7nPWgszD2Y7//X18kNgkHis+eBPPj78LWxZ+hDrRpQNCAm8u7dDFZdoRLR0AnN7l447IgIp9YaxpGlLaNkfLhEaVN1iN+PhYERMVDl3Xcc8tV+GNp/+DmKLSx5T10lvj5W53oX9OW3x29QRYdUs5aoJZ2P/rERzZcBI3dRxY01YEQRAEQRAEQRAEodqQARRBEKqMX7oVZ/Zn4aJY+0urrVYr3n/pURTmFTmW2Xv6mKfVseuo/aXtDcJKf61/9zev4NEf3nNZ9r5PXsfGHPsLsEf0vRg7Pp6D/wy7xq3u8Ob9cXrDWdzXd4hjWsO4uvjjxdcQ808IMnafxcP9r8Pcmx7HnqV2D1F+FXu5t8CTXiltHX83z4tzuuPioxseQ6/MVjiw8hjSd51FTlrpXSrt0QSa5vldOLWF/onJyDmVh5iTIZh8113o1bkdnhl/JxrG1a1pa4KXsPnb2fjorscQIXegCIIgCIIgCIIgCCZC3oEiCEKV+fCZ/8PPy9Nwx0WXOqZpmoaM3WcREhsIANh/6hjqBIUj05aDJpGxTuvvPnEEAJDcLAGzv/gZ9fvY31Py98k9Ltv6+8Qe+IX6AgC6t2gJq9Vz+pp03+2YhNtdpjeoXwdfTJ+EXfsOI6mp/bFMs8c+gic+eh93/OdSl+WF2kNkWOmXvq+PHot2994FJACFOYVIio3H5IdGYTJGgYiwfdd+XPHoBHRqnoRXn7+vBl1XH4/dMRwpqUm4uHfnmrYieClx9WIQVy+mpm0IgiAIgiAIgiAIQrUiAyiCIFSZ7imt0T3F9UXx02+6Fw9++g4a9qyH3ceO4MXfPwdphGWjpkHXSm98O5xxCggCGkfXRXCw84uJC4oK4WMpTVH52YWOAZR6oVFV9myxWByDJwAwqGcnDOrZqcp6Ag/CQ4Mdfwf4++PRS27Afxd+jil33e20nKZpaJHQCDu/nlvdFmsUPz9fXD6ge03bEARBEARBEARBEARB8CpkAEUQBMO5+pI+2H74AH4uWIvP/10G/O8JSC/8+Ak2ndqNno1aIzggANt8D0KDhsT6cdi37Qjadkx0aLy6/Cs8MuAGAMDZzGz4Bvs45gX5+ldnOEItoG5MJBL21sOlfe3vQhl145W464YrTPF4LkEQBEEQBEEQBEEQBKFqyDtQBEFQQrO69V2mLTq8Hsdtp7Hg31TM2bQYmo+GotwitI9PwC+vTsHWBXuw/0/7i97/3LHZsd6xkxmw+GgoyClEwqlYF11BOB+apuGDyY/h+kv6O00TBEEQBEEQBEEQBEEQBE/IAIogCEpoFFuxl0/flDQAflYfdEluiRM/f4OFE/+LrKM5yCrKdSyTZ8uHbtUR6xeJDx5/VJVlQRAEQRAEQRAEQRAEQRAEB/IIL0EQlFC/bnSFlhuQ3MHp/6SmDVCQU4QCrcgxbf3BHdCtOnIp31CPgiAIgiAIgiAIgiAIgiAInpA7UARBUEK9mEjs+Hm/07SCnEKX5RpF13P6X9d16EUaYAEKi+2DKNOWzgcAZGt5itwKgiAIgiAIgiAIgiAIgiA4IwMogiAowcfHisUvTcWqV//G/tSjeDb5VtxR52KX5QJ9/FymWcmemrLy7Y/x0nR5V4UgCIIgCIIgCIIgCIIgCNWLPMJLEARltE5qguMrF+DoiXS0SGiEzm1aYHyvN5FwaQOENw5BbrrN7XpWWAAAWbZchPsHQ971LQiCIAiCIAiCIAiCIAhCdSMDKIIgKCU8LAThYSEAgKDAABxd/jWWp23CvJ+X4parBrldx6/QBwCw/cQBxIfFIDQ+GAAQnx1VPaYFQRAEQRAEQRAEQRAEQTA9MoAiCEK1EhoShCsH9cCVg3p4XCYpNA7/4gieWTIH3eNbOaZPv21MdVgUBEEQBEEQBEEQBEEQBEGQd6AIguB9DOvX2/H3T1vTUJBbiOKiYtSLjKxBV4IgCIIgCIIgCIIgCIIgmAkZQBEEweuoXyfa8ffra76BxVdHWFFQDToSBEEQBEEQBEEQBEEQBMFsyCO8BEHwOiLCgp3+1y06fEnSlSAIgiAIgiAIgiAIgiAI1YfcgSIIgtcRERaCldP/cprmq/vUkBtBEARBEARBEARBEARBEMyIDKAIguB1hIUG4+E7bsDaGVsc04otxTXoSBAEQRAEQRAEQRAEQRAEsyEDKIIgeCXPPngnQrRA5JzMAwD4WuQOFEEQBEEQBEEQBEEQBEEQqg8ZQBEEwSvx8bFi0Zyp2PjhdhzZcBKX1u1U05YEQRAEQRAEQRAEQRAEQTARMoAiCILX0rZFUxTlF2PnzwcQGRBS03YEQRAEQRAEQRAEQRAEQTARMoAiCILXomkaVn79FoYPGYhBPVNq2o4gCIIgCIIgCIIgCIIgCCbCWtMGBEEQyqNbx9bo1rF1TdsQBEEQBEEQBEEQBEEQBMFkyB0ogiAIgiAIgiAIgiAIgiAIgiAIZZABFEEQBEEQBEEQBEEQBEEQBEEQhDLIAIogCIIgCIIgCIIgCIIgCIIgCEIZZABFEARBEARBEARBEARBEARBEAShDDKAIgiCIAiCIAiCIAiCIAiCIAiCUAYZQBEEQRAEQRAEQRAEQRAEQRAEQSiDDKAIgiAIgiAIgiAIgiAIgiAIgiCUQQZQBEEQBEEQBEEQBEEQBEEQBEEQyiADKIIgCIIgCIIgCIIgCIIgCIIgCGWQARRBEARBEARBEARBEARBEARBEIQyyACKIAiCIAiCIAiCIAiCIAiCIAhCGWQARRAEQRAEQRAEQRAEQRAEQRAEoQwaEVFNm6gO/Pz8EBMTU9M2vJIjR46AiKBpGgBU6u+qrqfi75rePgdfNb19Dr5qevscfNX09r3Vi7f6quntc/BV09vn4Kumt++tXrzVV01vn4Ovmt4+B181vX0Ovmp6+97qxVt91fT2Ofiq6e1z8FXT2/dWL97qq6a3z8FXTW+fg6+a3r6RXkJDQxEcHAzBmRMnTsBms3mcb5oBFMEzJR1JEARBEARBEARBEARBEARBqH3ExcXh4MGDNW2DHfIIL0EQBEEQBEEQBEEQBEEQBEEQhDLIAIogCIIgCIIgCIIgCIIgCIIgCEIZrDVtQKh56tati+zsbAQFBQFApf6u6noq/q7p7XPwVdPb5+CrprfPwVdNb99bvXirr5rePgdfNb19Dr5qevve6sVbfdX09jn4quntc/BV09vn4Kumt++tXrzVV01vn4Ovmt4+B181vX1v9eKtvmp6+xx81fT2Ofiq6e0b6eXBBx+EUHnkHSiCIAiCIAiCIAiCIAiCIAiCIAhlkEd4CYIgCIIgCIIgCIIgCIIgCIIglEEGUARBEARBEARBEARBEARBEARBEMogAyiCIAiCIAiCIAiCIAiCIAiCIAhlkAEUQRAEQRAEQRAEQRAEQRAEQRCEMsgAiiAIgiAIgiAIgiAIgiAIgiAIQhlkAEUQBEEQBEEQBEEQBEEQBEEQBKEMMoAiCIIgCIIgCIIgCIIgCIIgCIJQBmtNGxB4cuLECXz77bdIS0vDjh07kJGRgdzcXAQEBCAiIgKJiYno2rUrhgwZgjp16lRYd8OGDVi1ahW2b9/u0AwODka9evWQkpKCQYMGISwszLA4YmNjceLECRQWFl6w1t9//42dO3eisLAQ8fHx6Ny5M6zWC+9iFfU4Z84cdOvWDYmJiRe8zbJkZWUhNTUVhYWFSE5ORlxcnGNeZmYm3n77baxduxZ5eXlISkrCLbfcgg4dOlR6Ozk5OZg9ezYWLlyIvXv3oqioCHFxcRg4cCBGjRqFqKioC4rDyON9IZpG9x8V/VFVH3dHZfuOyrZeFtVt0h0rV65EQUEB+vTpY7i2kZTnU1XOMLpdGu2zunJlCRdy3snPz4fNZkNISIjLvBUrVmD9+vWOOPr27QtN02pEszJUpe8Y1ceNql9UtaHaeLwriora0gjN6jiXXajP6jzfnouKmrqqmirauYoay1t91obrk7JUpfbnkC+9/XrC269Ha2NbNxoV1+JlOV8tqGJfVsd1c3Vck1b2+HDIa1WBy7V4edSWnOF1kCB4YMWKFbR8+XKnabm5uTRu3Djy8/MjXddJ0zSPH13Xyc/Pj8aOHUs5OTnlbmvu3LmUkJBAuq67fEq0dF2nwMBAGj16NJ06dcqQGOvVq0e6rp93ue3bt9OBAwfczvvhhx8oMTHRxXdERAS99NJL1eaxZD9169aN3n77bcP20eeff07h4eGOuHx9fWnSpElERLR3716Kj493agsly7344otu9T744APq3LkzLVq0yGn6pk2bqHHjxm7bla7rFB0dTYsXL76gWCq6L1VpGt1/VPRHozVV9B2j23pNtkl3REdHk8Vi8Tj/rbfeouTkZAoMDKSoqCi66qqr6M8//yxXs1evXuVqGunT6JxBpKatG+3TaD1V550DBw7QFVdcQT4+PqTrOiUmJtL333/v2M+DBw920U1OTqbdu3dXq2ZV8NQmVfZxI+sXFX2H0/E2OrepqC2N1FRVtxnpU4VHFblNhaaKdq7iXObtPrlcn1SGytT+Ko6P0bmSw/UEh+tRTm29pq4lVFyLl6W86yij96XR7bymr0kreny41IFVhcu1uCdq8vxY29GIiGp6EEfwTmJiYpCRkeEYgbbZbOjbty/WrFkDIkKLFi3Qs2dPNG3aFBEREfDz84PNZkNGRgZ2796N1NRUbNu2DZqmoVOnTvjjjz/g6+vrsp377rsP7777LogIuq4jOjoaJ0+eRHFxMTRNw5VXXomQkBCsXbsW27dvh6ZpqFevHhYvXoyWLVu66I0ePbrCMX7yySfIy8vDyJEjHdM0TcOMGTOcltN1Hb1798by5cudpn/00UcYOXIkiouLAQCRkZHw8fHB8ePHQUTQNA133XWXi54qj+fO9/HxwWWXXYZbbrkFV155pdt9fz42bNiALl26oKioCNHR0WjUqBG2bduGnJwczJ07F7NmzcLSpUvRqlUr9OvXD0SE5cuXY8uWLdA0DUuXLnUZub/sssuwePFiHD161PFLiaysLLRo0QKHDx9GYGAgbrzxRrRs2RJ+fn7YvXs35s2bhyNHjiAoKAjr1q1DUlKSQ0/FvlShaXT/UdEfVWga3XdKNM/d1xfa1o1ukxdKTEwM0tPTUVRU5DLvzjvvxEcffYSyp25N03Dfffdh2rRp8PHxcVmvd+/eWLFihVtNI32qyBkq2qXRPlXEraLvZGVloX379ti7d69TG/Lx8cGyZcvw0Ucf4b333kPdunXRr18/FBYWYunSpUhPT0diYiL++usv+Pn5KdesKp76jqo+bmT9oqINcTreRuc2o2tLFZoq6jajfarwqKouMFJTRTtXcS7j4JPL9YmK2l/F8TE6V3K4nuBwPQrwaetGtyEVfedC8FQLGr0vVfQdFe3S6OPDpQ68ELhci7tDRc4QzkHFqIxQO4iOjnYagX7uuedI0zRq0aIFrVixokIaqamp1Lx5c9J1nZ577jmX+Z999hlpmkZxcXE0b948ys/PJyKi/Px8mjdvHsXFxVFkZCTt3buXiIg2b95MV155JWmaRg0bNqTMzEwXzZJR1PJ+AXC+Xwe40+zdu7fTtKNHj1JgYCBpmkZXX301/fvvv455J0+epMcff5wsFgvpuu7yKwJVHtu0aUNPPvkkNW7c2GnZiIgIuvvuu887Ml6Wm2++mTRNo3vuuYcKCwuJiOjUqVPUo0cPatq0Kem6TqNGjXLMIyIqLCykkSNHkqZpdP3117toxsXFUaNGjZymvfrqq6RpGnXq1ImOHj3qsk5eXh7ddtttpGka3XHHHdWyL43WNLr/qOiPKjSN7jslmka2daPb5IVSNveWsGDBAkec9957Ly1YsIA+++wzGjp0qGP6wIEDKSsry2XdXr16Gf6LL3c+VeQMFe3SaJ8q4lbRd0r2ZXJyMq1Zs4YyMzPpl19+ofr161P//v0pKCiIBg4cSNnZ2Y51zpw5Q926dSNd1+mtt96qFs2q4qnvqOjjRtcvKvuOtx9vo3ObitpSVb1qdN1mtE8VHlXVBUZqquw7Kuo2b/bJ5fpEZe1v1PFRUQdyuJ7gcD1KxKOtq2hDKvrOheCpFjR6X6roO6rapZHHh0MdeKFwuRZ3h4p8KZQid6AIHik78tq6dWvs2rULO3bsQIMGDSqss2/fPiQlJaFZs2bYsmWL07y+ffsiNTUVGzZsQNu2bV3WTUtLQ/fu3TFixAh89NFHjum333475syZg2eeeQZPPvmk0zq6rkPTNFx++eVISUkp19vUqVORk5ODiRMnOk2fNGmSi2avXr3w+++/O6ZNmTIFjz32GAYPHowffvjBrf7kyZPx5JNP4vrrr8fnn39erR7/+OMPzJkzB19++SVOnz7teOZk48aNMWLECNx8883nfUZro0aNcPr0aRw9ehQBAQGO6ampqejduzcCAwNx/PhxBAYGOq2XlZWFunXrIjIyEgcOHHCa5+/vj+TkZKxatcox7dprr8WCBQuwevVqj/sjLy8PsbGxCA4OdtJUtS+N1jS6/6jojyo0je477jQvtK0b3SaByv3apyxz5sxBfn6+yy9UBg8ejF9++QVvvPEGxowZ4zTv119/xS233IJTp06hU6dO+PnnnxEZGemYb8SvxiriU0XOUNEujfapIm4VfadTp07YuHEjNm/ejBYtWjimf/LJJxgxYgR0Xcc///yD5s2bO623atUq9OjRAwMHDsSiRYuUaqroOyr6uNH1i4o2xOF4A8bnNhW1pap61ei6zWifKjxWR11woZoq2rmKcxkHn1yuT1TU/kYfHxV1IIfrCQ7XowCPtq6iDanoOypqQaP3pYq+o6pdGnl8ONSBAJ9rcaNRkS+Fc6jpERxBLaNGjaryx9/f32mUNCAggFJSUqrko2PHjhQQEOAyPTw8nNq0aVPuuk2aNKG6des6TTt16hT5+PhQcnKyy/IzZ86kqKgoslqtNHbsWDpz5oxH7cq8X6TsL9uuvfZa0nWdVq5c6XG9vLw8CgoKooYNG9aIRyIim81GX375JV111VXk6+vrGB3XK/CMVl9fX+rQoYPL9OzsbNI0jdq3b+/RT/v27cnPz89let26dSkhIcFp2sCBA0nXdSooKCg3xi5durhoqtiXKjSN7j8q+qMKTaP7jidNoqq3daPbZIlHo3+NVadOHYqKivLoZd++fdSqVSvSNI3atm1LR44cccyrrl+NqcgZKtql0T5VxK2i74SEhFCTJk1cph89epQ0TaMGDRp41I2JiXE5F6vQVNF3VPRxo+sXFW2Iw/EmMj63qagtVWiqqNuM9qnCY3XVBReiqaKdqziXcfDJ5fpERe1v9PFRUQdyuJ7gcD1KxKOtq2hDqr7XMLoWNHpfqug7Ktql0ceHQx1IxOda3GhU5EuhFGtND+AIapk1axY0TXN5Rl9FKfmlBAAEBwfj+PHjVdI5fvw4goKCXKbn5+e7fUbgufj6+uLIkSNO0yIjI9GyZUvs3r3bZfm77roLQ4cOxfjx4/HWW2/hq6++wpQpU3DzzTdXybsn0tPTAQAdOnTwuIyfnx9atmyJzZs314hHwL7/rrnmGlxzzTVIT0/H559/jrlz52LVqlVIS0vD6tWr8cADD+Cyyy7DggULXNbNyspy0SyZdvbsWY/bPXv2LKxW1xSTnJyMRYsWYefOnUhISABgHykHgL179zqmlYWIsH//fsfzQEtQsS9VaBrdf1T0RxWa7riQvlMeVW3rRrfJEi8FBQUYN26c069PKsLLL7+MvLw8l+kZGRlITk72uF7Dhg3x559/4rLLLsPq1avRp08fLF68GA0bNvS4jtE+VeQMFe3SaJ8q4nbHhfad/Px8xMTEuEyvW7cuAJT7C7oGDRpUi6aKvqOijxtdv6hoQxyON2B8blNRW6rQLE+nqnVbdfm8EI/uUFEXXIiminau6jqKg093eNv1iYra3+jjo6IO5HA9weF6tDy8qa2raEMq+o6KWtDofami76hol0YfHw51IMDnWtxoquua1LTU5OiNoB4/Pz/SdZ3uv/9+evrppyv1CQwMdBolHTJkCOm6TtOmTauUhylTppCmaTRkyBCXeW3atCEfHx/av3+/23X//fdfslgsLiPxRPYR0pCQkHK3vWjRIkpISCBd16lfv360ZcsWp/kXcnfHddddR7quU0ZGRrnrdurUiUJDQ2vEY3ns2rWLJk2aRImJiR5H2Tt06EBWq9XpudFERLNnzyZN08hisdDmzZtd1vvrr79I13W3v4KcN28eaZpG/fv3p7y8PCIiWrp0KWmaRkOHDnV6HuO5vPDCC6RpGt10000eYzJqX6rQNLr/qOiPKjRV9B2j27qKNtmlSxfSdZ2+//77CvsswdNzV6Ojo6lZs2bnXT8rK4sGDBjg+BXQtm3bPP7qxWifKnKGinZptE8VcavoO3FxcdS4cWO365yvX6WkpFBERIRyTRV9R0UfN7p+UdGGOBxvIuNzm4raUoWmirrNaJ8qPFZXXXAhmirauYpzGQefXK5PzsWo2t/o46OiDuRwPcHlepRDW1fRhs7FqL6johY0el+q6DsqvychMub4cKgDifhcixuN6vOj2ZEBlFqOkYljxYoVZLVaSdd1uuyyy2j+/Pl0+PBht+sePnyY5s+fT5deeinpuk5Wq9XtLfRPPfUUaZpG7dq1o9WrVzvNW7NmDbVu3Zp0XacHHnjAZd3Q0FBq3rz5eePIy8ujCRMmkJ+fH/n6+tJjjz3meKlVZQYnGjVqRM8884zj07dvX9J1ndatW1fuunXq1DlvojXKY2WKtnNZsWIFjRkzxmX6M888Q5qmUatWrejHH3+kLVu20KxZsygiIoL69OlDLVq0oFatWtGaNWsc66xevZpatmxJuq7Tww8/7HZ7Q4YMIU3TqHXr1vTZZ5/RmTNn6PHHHydN0yghIYGmT59OP/30E/322280c+ZM6t27N+m6Tv7+/m4T/rkYsS9VaBrdf1T0RxWaKvqOirZudJscO3Ys6bpOEydOrLRHT0VbyX5z9wLBsthsNkdMderUoYYNG7rVNNqnipyhol0a7VNF3Cr6Tvfu3cnX19dxUXYu33zzDf3xxx8eNaOioqhFixbKNVX0HSLj+7jR9YuKNsTheBMZn9tU1JYqNFWcy4z2qcKjqrrASE0V7VzFuYyDT07XJ+diRO1v9PFRUQdyuJ7gcj3Koa2raENlMaLvqKgFjd6XKvoOkdrvSYgu/PhwqAOJ+FyLG011nB/NjAyg1HKMThxz584lf39/x68kdF2ngIAAio2NpUaNGlFsbCwFBAQ45mmaRn5+fvTxxx+73caZM2ecfnXRsGFD6t69uyPBaJpG9evXp2PHjjmt99tvv5GmaXT77bdXOJ4tW7ZQ7969HRdZX331VaUGJ8o+E7Hk7yeffNLjev/88w9pmkaXX355tXisatHmiczMTEpKSnI63rquk5+fH/3++++OkWxd1yk0NJRCQ0Md+yYqKsrpuY/nYrPZaPjw4Y51LRYLNWjQgHx9fZ22c247CgkJqdRA4IXsS1WaRvcfo/VUaKroOyrautFt8uOPPyZN0+iSSy6ptJeoqCi3bWrixImk6zpNnTq1QjqFhYV08803Ox1L1T5V5Qyj26XRPlXEraLvjBs3jnRdp19++cXj+uVp3njjjco1VfQdIuP7uNH1i4o2xOF4Exmf21TUlio0VZzLjPapwqOqusBITRXtnMj4cxkHn5yuT9xxIbW/0cdHRR1I5P3XE1yuRzm0dVVtyB0X0ndU1IIq2pGKvlMd35MQVf34cKgDifhcixtNdZ4fzYgMoNRyVCSOvXv30pgxY6h+/fouFyznfmJjY2nMmDG0Z8+ecrdz6NAh6t+/v1uNTp060fbt213W+fPPP+nVV1+ljRs3VjqukhdplSSKiiSy8h51NmPGDI/rjRkzhjRNoxdeeEG5RxVFGxHR8ePHaeTIkRQbG0vh4eHUp08fWrJkiWP+5MmTKSAgwOm4tW7dmtavX39e7Z9//pkuueQS8vHx8diOGjZsSA899FCFRvrdUZV9qVLT6P5jtJ7Rmir6jqq2TmRcmzx69Ci9+uqr9MEHH1Taw4EDB2jv3r0u0zds2ECaplFcXBzl5uZWWG/cuHEe26kKn6pyhtFt3WifRuup6DtLly6lW265hb7++uty901Zxo8fT5qm0cyZM5VrqmiT52Lkecfo+sXoNsTheBOpyW0qakujNVWdy4z0qcKjitxmtKaKdl6CkecyDj45Xp+4oyq1v9HHR0WuLMHbryc4XI9yaOsq25AnqtJ3VNWCKtqRir5DVD3fkxBV/vhwqQO5XIuroLrPj2ZCI6ri28UFFhw7dgyff/45QkNDcccdd1Rq3YMHD6KoqMjx4ip37N+/Hzt27EBGRgby8vLg7++PiIgIJCYmVvplSX/99RdSU1ORkZGBsLAwdO7cGV26dKmURkU5efIkHnnkEWzcuBEAsGHDBiXb2bJlCwoKCtC4cWOEhYVVat3q8mgEJ06cwNq1a5GZmYnExMRyX9jpjry8PGzatAlHjhxBdna2ox01b94ccXFxF+xPxb40QtPI/qNCT5VmRbiQvmMEqttkVdm9ezeICA0aNICvr2+F11u1ahVsNhv69u2r0F3FudCcUV3t8kJ9qtZzh4q+c+TIEeTn56NevXrw8/PzWs3KYGQfr876pTraUE0cb1W5TcWxqc7jfSFw8VlRVOQ2ozUr23dqqsbi4vNCMTpfVte1WXnHpzrqQK7XE95+PaqSysReE9cSZvleQ0U7r452WR3Hp6bqQKPh4vN8VMf1RG1EBlAEQRAEQRAEQRAEQRAEQRAEQRDKoNe0AUEQBEEQBEEQBEEQBEEQBEEQBG/DWtMGBHNw6NAhFBUVGXqbtwrNilBUVISlS5ciLS3NcYtmbm4uAgICHLdodu3aFf369YPVanwXM2vcQM3Ebta4y+PAgQPYtGkTCgsL0b59ezRp0qRC65l1X5o1bm+gtpx7zNqGVMQt+9JccavArPtS4jZX3OejKrWgmfdlTccubcg4pA3xbkNAze/LmqKm4zZr3wG8s/8IVaCmXr4iVD+FhYW0aNEiev755+m2226jIUOG0EUXXURDhgyh2267jZ5//nlatGgRFRQUGL7t6Ohoslgs1ar5448/0tChQ6lVq1aUkpJC//nPf2j37t3lal533XXUtGlTj/OnTZtGMTExpOu640VbZT8l82JiYmjq1KlUXFxc5RjdYda4icqPXeI2Ju4zZ87Q2LFjKT4+niIjI+miiy6itWvXEhFRUVERjRkzhqxWqyNuXdfpyiuvpFOnTpW7TTPuSyIecRMZH7uKfVkVqvvcI23IuNhVxM1hX0ob8u48ZNZ9KXGbK24VtaBZ9yWRd8Qubcg4pA25h0MbIjJ+X6o43tKGalfcRGquScvDW67FaxsygGISajpxREdHk67rhumdT3PixIkuseq6Tv7+/jRt2jSPmr169XKrWVxcTEOHDnXohYWF0aBBg2j06NH02GOP0cSJE+mxxx6j0aNH06BBgygsLMyxzSFDhlTbvqzNcRN5jl3iNibu/Px86tSpk0uOCA0NpX/++YcmTJhAmqaRv78/devWjTp16kQ+Pj6k6zp169aNioqKXDTNui+5xE1kfOxG610I1XnukTZkTOwq4uayL6UNeW8eMuu+lLjNFTeR8bWgmfelN8Uubcg4pA2pj5vLNamKax5pQ7UvbiI116Se8KZr8dqGDKDUcrwlcVTnl1hLly51JIrBgwfTK6+8Qi+++CJ17NjREduIESOosLDQZV1PSeONN94gTdOoXr169Omnn573Lp2CggL65JNPqF69eqTrOr3xxhtVD7QMZo2byH3sErdxcb/22mukaRo1aNCA5s+fT3///TfNmDGDQkND6eqrr6aIiAjq0KEDHT582LHOnj17qEWLFqTrOn388ccummbdlxziJjI+dhX78kKornOPtCHjYlcRN4d9KW3Iu/OQWfelxG2uuImMrwXNvC+9KXZpQ8YhbcgZDm2IyPh9qSJuaUO1M26i6htA8bZr8dqGRkRU048RE9Tx5ptv4v7770fdunUxffp0XHfddeU+16+wsBBffPEFHnroIRw/fhyvvfYaxo4dCwCYPHlylX1MnjwZubm5KCoqcplutOa1116LBQsW4PHHH8fzzz/vNO+9997D+PHjYbPZcMUVV+CLL76An5+fY37v3r2xYsUKF82UlBT89ddfWL9+Pdq2bVthj5s2bUJKSgratWuH9evXO3mvKpzjLvFfVdzFLnEbF3fPnj2xatUqrFy5El26dHFMf/vttzF27FhomoZVq1ahc+fOTustXLgQgwcPxuDBg/HDDz84zTPrvuQQN2B87Cr2JYdzj7Qh42JXETeHfSltyLvzkFn3pcRtrrgB42tBM+9Ls14/ShuSNlSCt7QhwPh9qSJuaUPeG3eJ/6riKXajUbEvhXOo6REcQS0dO3Ykq9VKf/31V6XW27hxI1ksFurQoYNjWsmIZVU+JeuWRYVm/fr1KTQ01OMo87p16yg2NpZ0XacBAwZQVlaWY56nUdegoCBKTk6u1D4soX379hQcHCxxK4pd4jYu7vDwcGrQoIHL9IMHD5KmaRQbG+sx7oiICIqLi3OZbtZ9ySFuFbGr2Jcczj3ShoyLXUXcHPaltCHvzkNm3ZcSt7niJjK+FjTzvjTr9aO0IWlDFxo3h2tSFXFLG/LeuFXFbjQq9qVQiudbEYRawfbt29GmTZtKjboCQPv27dGmTRvs2LHDMc1isaC4uBhXX301goODK6X3+eefIz8/32W6Cs2TJ0+ibdu2Hu+06dixI1JTU3HRRRdh2bJlGDRoEBYuXIiwsDCP2/Lz80NWVlal/JWQlZUFHx8fp2lmjRswPnaJ27i4c3JykJSU5DI9NjYWANCoUSOP6zZs2BDbt293mW7WfckhbsD42FXsSw7nHmlDxsWuIm4O+1LakHfnIbPuS4m78nCOGzC+FjTzvjTr9aO0IWlD5+INbQgwfl+qiFvakPfGDaiJ3WhU7EvhHGp6BEdQS2RkJCUkJFRp3WbNmlFERITj//bt25Ou6/TLL79UWsvTM/9UaIaHh1OLFi3Ou/6RI0eoTZs2pGkatWvXjo4ePepx1HXgwIGk6zrNmzevUh4/++wz0jSNBg0a5DTdrHETGR+7xF0+lYk7NjaWmjVr5lZH0zTq3bu3x+107tyZQkNDXaabdV9yiJvI+NhV7EsO5x5pQ8bFriJuDvtS2pB35yGz7kuJ21xxExlfC5p5X5r1+lHaUPlIG6od16Qq4pY2VD61sW4zGhX7UihFBlBqOUYmjlGjRpGu6/TCCy9U2oenhKFCs1u3bmS1Wun06dPn1UhPT6euXbuSpmmUmJhISUlJbjV//PFH0jSNrFYr3XvvvbRmzRoqKipyq1lUVERr1qyhe+65h6xWK+m6Tj/++KPTMmaNm8j42CVu4+JOSUmhgIAAty8Ve/XVV2n+/PketxEbG0uJiYku0826LznETWR87Cr2JYdzj7Qh42JXETeHfSltyLvzkFn3pcRtrriJjK8FzbwvzXr9KG1I2pAnatM1qYq4pQ15b9xEamI3GhX7UihFBlBqOUYmjpkzZ5KmaTRkyJBK+4iKinLbGVVoPvLII6TrOs2YMaNCOllZWTRgwACnZxq648UXX3TM13WdAgMDqUWLFtS9e3fq27cvde/enVq0aEGBgYFOy02ePNlFy6xxExkfu8RtXNx333036bpOy5cvr5S/3bt3k6ZpNHToULfzzbgvibw/biLjY1exLzmce6QNGRu70XGr0JQ25L1tSPald/dHidu741ZRC5p1XxKZ8/pR2pC0ofKoLdekKuKWNuS9cROpid1oVO1LwY4MoJgAoxLHnj17aPz48fT8889X2kNqaiotW7bMZboKzT///JM0TaOkpCSPg0VlsdlsNHToUEfi8MSKFSto8ODB5OfnR5qmefz4+vrS4MGDKTU11a2OWeMmMj52idu4uL/55hvq1asXffTRR5XyN2nSJNI0jd544w2Py5htX5bgzXETGR+7in3J4dwjbcj42I2MW4WmtCHvbUOyL727P0rc3h23qlrQjPuyBLNdP0obkjZ0PmrDNamKuKUNeW/cRGpiNxqV+1Ig0oiIavo9LIJ6Vq5cieeffx6//fZbuS8v8vHxwaBBgzBhwgT06NGjGh0ay5IlS0BE6Nq1a4Vf8FRcXIwvvvgCNpsNt912W7nLZmZmYsOGDdixYwcyMjKQl5cHf39/REREIDExER06dEBISIgRoVQKiVviPh+VibuirF+/HpmZmWjfvj3Cw8PLXdas+9Jb4waMj91b2mV1I21ITewq4vbWfSltiE8eMtO+PBeJ21xxV4aK1oJm3pfeGru0IeOQNsSjDQHG7EsVcUsbMlfcKvCW/lgbkQEUk2GmxCEIgiAIgiAIgiAIgiAIgiAIVUUGUARBEARBEARBEARBEARBEARBEMqg17QBofZx/fXXo1mzZqLphXpm1uTgUYUmB49cNDl45KLJwaMKTQ4euWhy8KhCk4NHLpocPHLR5OBRhSYHj1w0OXjkosnBowpNDh65aHLwyEWTg0cVmhw8ctI0Gg4evREZQBEAGNuBjhw5gr179xqiZXZNDh65aHLwqEKTg0cumhw8ctHk4FGFJgePXDQ5eFShycEjF00OHrlocvCoQpODRy6aHDxy0eTgUYUmB49cNDl45KLJwaMKTQ4eOWkaDQeP3ogMoAgApAMJgiAIgiAIgiAIgiAIgiAIwrnIAIogCIIgCIIgCIIgCIIgCIIgCEIZZABFEARBEARBEARBEARBEARBEAShDDKAIhgOEYGIRNML9cysycGjCk0OHrlocvDIRZODRxWaHDxy0eTgUYUmB49cNDl45KLJwaMKTQ4euWhy8MhFk4NHFZocPHLR5OCRiyYHjyo0OXjkpGk0HDx6IxrJXhMA9OrVCytWrEBxcXFNWxEEQRAEQRAEQRAEQRAEQRCEGkcGUARBEARBEARBEARBEARBEARBEMpgrWkDQu0kMzMTRUVFCA8Pr2kr1c4DDzyAs2fP4v33379grUOHDmHjxo3Iy8tDUlIS2rZta4BD7yMnJwf//vsvMjIykJubi+DgYNSrVw9JSUk1bc0tp06dwrJly7B3714UFRUhLi4O/fr1Q1xcXIU1srOzsXHjRuzYscMRd0BAACIiIpCYmIjk5GQEBQUpjKLyGBF3dWPWXGRkHgLMkYskD/HIQxyRPCR5qKJIHpI8pArJQ5KHKorkIclDqpA8JHmoonDLQwDP70qESkKCUEm+/PJLuuWWW+jqq6+mF154gc6ePeuYN2/ePEpMTCRd10nXdWrUqBFNmzbNo5au69SrVy+aMWMGZWRkVIP78unYsSM1bdr0gjTq1atHuq6fd7mjR4/SHXfcQfXq1aPo6GgaPHgwbdiwgYiIiouLady4ceTj4+PYl7quU4cOHWjz5s3l6h47doxeffVVGjNmDD300EP09ddfU1FRUbnrTJ48me64444Kx0hEdPbsWXr++eepY8eOFBoaSoGBgZSUlET33Xcf7d69+7zrnzlzhp5//nlq166dU4znfkJDQ2n48OH0559/VspbRfF0vP/++2966623aMuWLU7T8/Pz6aGHHqKAgAAXr1arlW677TY6c+ZMudtcu3YtDR061K3GuR9/f38aOnQorVmzplw9I4+3yrirwvn6o1G5SGUeqq7+WJaK5iEiNblI8lDFqQ15iMi4Y86tLpA85BnJQ3zyEJH7ti55yFx5iEhd7GbNQ1WlstekkodqXx4i8q5cVBvyEFHN5CLJQzWXh2rzdyXlUVPn3NqODKAIHnGXMEeNGkW6rpOmaaRpGum6Ts2bN6f09HT67LPPnOadu8zo0aPdbqNkfklhcs0119CCBQsoPz+/OkJ0ITo62u3JLTU1tcKfqKgo0nWdVqxY4TT9XM6ePUsJCQku+ys8PJy2bdtG//d//+eYFhISQsHBwY7/69SpQ0ePHnXr/4cffqDQ0FCXxN2yZUtKS0vzGHevXr3cxt29e3e69957Xabv2rXLrf+S4xkcHEw//fSTx+0tW7aMYmJiPK4fExNDFovFpQ0VFhZ61KwKno73nXfeSbqu019//eU0/eqrr3Z4DgoKojZt2lBKSgpFRkY6fHbp0oXy8vLcbm/atGlOcQUEBFBSUhJ17dqV+vTpQ127dqWkpCTy9/d3LGOxWOjll192q2f08VYVd1XxdHyIjM1FqvKQ0cfH6DxEpCYXSR6qHNzzEJGxx5xLXUAkeUjyUO3JQ0Tu27rkIXPlISI1sZs1D10I7o655CFz5SEi78pF3PMQkbHHR/KQM96Yh2r7dyWeMDpuoRQZQBE8UjZhfvPNN6RpGvn4+NB9991Hb7zxBg0fPpw0TaOHHnqImjRpQlFRUfTVV19RXl4eZWVl0Zw5cxydd+nSpS7b0DSNYmJiqGHDhk7JMCoqisaMGUMrVqyoxog9FwbnnsSr8rFYLE56kyZNIk3TKCkpib799lvavHkzvfvuuxQWFkbDhg2jkJAQat68Oa1cudKxTmpqKjVv3px0XadHH33UxePOnTspKCiINE2junXr0lVXXUWXXHKJo+jz8fGhmTNnuo27vKKtd+/eTtOKioqobdu2pGkaRURE0BNPPEHff/89LVy4kKZNm0aJiYmkaRqFhobSgQMHXDS3b9/u8HnZZZfRrFmz6Oeff6ZZs2bR4MGDSdM0uvzyyykrK4tWrVpFDz/8MAUHB5Ou63TDDTe4P3BVxNPxbt68OUVFRTlN++mnn0jTNAoMDKTXX3+dbDab0/wFCxZQ48aNSdd1eumll1w0f/75Z0fhPXr0aEpLS/N4wi8sLKS0tDQaNWoUWSwW0nWdfv75Z6dlVBxvFXFfCJ6Oj9G5SEUeUtUfjcxDRMbnIslDlYdzHiIy/phzqQskD0keqk15iMh9W5c8ZK48RGR87GbNQxeKu2MueagUM+QhIu/KRZzzEJGa84TkITvemIfM8F2JO1TELZQiAyiCR8omzMsvv5x0Xac5c+Y4Lffggw+Sr68v6bpOn3/+uYvOjBkzSNM0GjFihMu8cxPwkiVL6I477qCwsDCnE1JCQgI9++yztGvXrgr5rsyvAcp+wsPDyy3UGzRoQI0bNy73U1JYlZ1+LsnJyWS1Wmnr1q1O02fNmuUo5sqObBMRbdiwgXRdp7Zt27rMu/fee0nTNLriiisoMzPTMf3YsWN00003OWJwl7gr84XBvHnzSNM0io+Pp71797qsk5OTQ/369SNd1+mxxx5zmX/rrbeSrus0ffp0l3lERM8//zzpuu706569e/c6bvtdsGCB0/IqjndwcDB16NDBadpdd91Fuq7TBx984NY3EdHWrVvJYrFQ+/btXeYNGjSIdF2nzz77zOP67vjkk09I0zS66KKLnKarON4q4lZxfIzORSrykKr+aGQeIjI+F0keMlceIjL+mHOpCyQPSR7ytjxEZHxblzxkrjykInaz5iEi44+55KFSzJCHiIzvj2bNQ0RqzhOSh+x4Yx7i8l2J0aiIWyhFBlBqOUYmzLp167qMuBLZR4xLRpvdYbPZKDAwkBISElzmuUvAeXl5NG/ePLr88svJx8fH6URZkWdhXsivAUrWLUvv3r1J0zRKSUkp97Y3ooo94zIoKIgSExNdph87dow0TaMmTZp4XLdJkyYUFBTkMj0hIYH8/Pzo+PHjbtd7++23Hc/LfPzxx53mVeYLg5EjR7otks5l165dpOu625NEXFwcxcbGelyXiCg8PJxatmzpNG3FihWOXx+U9Wj08Q4NDXU5OV500UWk6zplZWWV6719+/YUGBjoMj0yMpKaNWtW7rqeaNasmUv/UnG8VcSt4vgYnYtU5CEVx8foPERkfC6SPGSuPERk/DHnUhdIHpI85G15qMSnkW1d8lApZshDKmI3ax4iMv6YSx4qxQx5iMj4/mjWPERk/PGRPFSKN+YhLt+VGI2KuIVSZACllmNkwvTx8aGUlBSXbdhsNsfJwxMtWrTwmNDLJuBzOXHiBL3++uvUpUsX0rTS2zdLnoX5zTffeIy5Ir8G8PTrAHfMnDmToqKiyGKx0D333EOnT592u1xFTpC+vr4uyZeIKDs7mzRNK3d0un379uTn5+cy3d/fn9q0aVPudr///nsKDAwkXddpzJgxjumVKdpKThInT54sd1stWrSg0NBQl+l+fn7UtWvXctdNTk4mf39/l+mNGjWiunXrung0+ni3b9+eAgICKCcnxzFt2LBhpOu6x+NeQps2bSgsLMxlelBQECUnJ5e7rifat29PwcHBTtNUHG8Vcas4PkbnIhV5SMXxITI2DxEZn4skD5krDxEZf8y51AWShyQPeVseKvFpZFuXPOSe2pqHVMRu1jxEZPwxlzxkrjxEZHx/NGseIlJzfCQP2fHGPMTluxKjUXXOFezIAEotx8iEGRUVRa1atfK4nfJOcl27dqWAgIBKr3cu//77Lz355JPUtGlTx4nS3bMjmzRpQrquOz0bsqKU93I0IvvJ+uabbyZNsz+Xc/bs2S7LVOQE2ahRIwoNDaXc3Fyn6SWj6EFBQZSdne2yXlZWFgUGBlJ8fLzLvKCgoArdFrh8+XIKCwsjXddpxIgRVFRUVKmi7bLLLiNd112e71iWbt26uT3J1a9fn6Kjoz0+ZzY/P5+Cg4MpJibGZV7nzp1dNFUc74kTJ5Ku6/Tcc885pr355pukaRq98cYbHvXWrVtHuq5Tz549XeaV3JL7999/V8rjpk2byGKxuBRTKo63irhVHB+jc5GKPKTi+JRgVB4iMj4XSR4yVx4iMv6Yc6kLJA9JHvK2PERkfFuXPHR+alMeqsh651KR2M2ah4iMP+aSh8yVh4iM749mzUNE6nKR5CHvzENcvisxGpXnXEEGUGo9RibMVq1aeRw1veqqq2jChAketZo1a0aNGjVymV6Zk+O5/PnnnzR69GiKjIx0mXfjjTeSruv0+uuvV1r3fAMoJSxatIgSEhJI13Xq3bu3U/FVkRPkLbfcQrqu09ixY6moqIiIiE6fPk29e/emuLg4CgkJoXvuuYeKi4sd6xQXF9Pdd99NmqbRtdde66LZpk0bCgwMpPz8/PP6X7t2rSPWoUOHUufOnT0WbV26dKF9+/Y5PiW3aO7YsaPcbSQmJlJcXJzL9JLj8/DDDzvFV8KYMWNI0zQaMmSIy7z69eu7aKo43qdPn6Z69eqRxWKhZ599lnJycigvL4/atWtHvr6+9PTTT1N6erpjeZvNRrNmzXIc+48++shF87XXXiNN0yg2NpbmzZvnsUAoobCwkD7//HOqX7++2/hUHG8Vcas4PkbnIhV5SMXxKcuF5iEi43OR5CFz5SEi4485l7pA8pAdyUPek4fO1TSqrUseqhzc8xCR8bGbNQ8RGX/MJQ+ZKw8RGd8fzZqHiNTnIslD3pWHuHxXYjTVcc41MzKAUssxMmHedNNNpOt6pX+pkZGRQbquU//+/V3mVfXkWIK7xDB9+nTSNI1uvvnmSutFRUVVOGnk5eXRhAkTyM/Pj3x8fOjBBx+kzMzMCp0gN27c6Hj2YL169ahLly4UGhpKuq7Tiy++6Hj5U9u2bWncuHE0btw4atu2Lem6/fFqS5YscdEseYHV999/XyH/W7Zsobi4OIemp6Lt3Pnnfso7AZw+fZosFgt169bNZd6GDRscL5Rr27YtTZw4kWbMmEETJ06kdu3aka7rZLFYaNmyZU7r7d27lzRNo0suucRpuqrjvX79esf8iIgIuuaaa+iee+5xeNd1nerUqUMNGjQgq9VKum5/7J27F+ER2Yubyy+/3LFPw8PD6eKLL6Z77rmHHn/8cZo0aRI9/vjjdM8999DFF1/seAeRpmk0ePBgl2JCxfFWEbeK42N0LlKRh1Qdn7JcSB4iMj4XSR4yVx4iMv6Yc6kLJA+VInnIO/IQkZq2Lnmo8nDNQ0TGx27WPERk/DGXPGSuPERkfH80ax4iqp5cJHnIe/IQl+9KjKa6zrlmRQZQajlGJsxXX32VNE2jadOmVUpn5syZpGkaPfvssy7zLvTk6I41a9ZQcnIyXXPNNZVe9+WXX6ann366Uuts2bLF8RKxuLg4x/MEz8f8+fMpIiLCcauprus0cuRIIiI6e/YstWvXzukEVbLcM88841bv66+/Jk3TaODAgRX2vmfPHkpISHBspyyNGjXy+Ii3W265xaPuO++8Q5qm0X/+8x+38+fOnUv+/v4uJ+CSW25fffVVl3Vmz55NycnJ9P777ztNV3m8Dx8+TDfccANZrVaH15LjUPZTt25devPNN8vdXlFREb344osUGRnpdNzLfkrmRUZG0uTJk93+KkrF8VYRt4rjY3QuUpGHVB4fd1Q1DxEZm4skD5krDxEZf8y51AWSh1yRPORKdeYhInW5SPLQhcMhDxEZH7tZ8xCR8cdc8pC58hCR8f3RrHmIqHpzkeQhV6o7D3H5rsRoqvucazY0IiIItZa1a9di1KhRaNasGb788stKrTtlyhTk5ORg0qRJjmlFRUXQdR2aplVY5+OPP8aePXtw8803IyEhwWneRx99hLp16+LSSy+tlDdvZNasWfi///s/pKenQ9M0FBUVnXed7OxspKamIjc3F23btkXTpk0d8/Lz8/Huu+8iNTUVmZmZSExMxM0334wuXbq41bLZbHjhhRdARHjwwQcRERFRId/Hjx/HU089hfz8fMyePbtiwZ6H6dOn48CBAxg+fLhHvzt37sTbb7+N1NRUZGRkICwsDJ07d8bdd9+N9u3bG+LDKA4fPoyffvoJ69atw5EjR5CdnQ1/f39ERESgefPm6N69O/r27QuLxVIhvYKCAixduhSrVq3Cjh07kJGRgby8PIdmYmIiunbtiv79+8PX19etRnUcb6PjNhIjc5GKPFRT/bEqeQgwLhdJHlKHN+YhwPhjzqkukDzkHslDpUgeKh/JQxeOt1+bmTUPeROSh8rHW/MQwCcXeXseAmomF0keKqW685BZvyvxphq4NiIDKIJgIKdPn8a+ffsAwOuKTUEQzIHkIUEQahrJQ4Ig1DSShwRBqGkkDwlC7UEGUARBEARBEARBEARBEARBEARBEMpgrWkDAm8OHTrkuNU1NzcXAQEBjltd4+LivEZTBUb7NGvcKjQ5eFSBWeMGeLQhFZj1mHOIm4NHVXDYlyqQuM3V1rnELe1S4va2NqQCLnFzaJccNDm0SYDHvlSBxO3dcXPoj2bWNBoOHtlQze9cEWoBBw8epPHjx1N8fLzbF62VfOLj42n8+PF04MCBGtGsCldddRUNGDCg2nyaNW4Vmhw8VpXyjo9Z41bh06xxq9KsLLUhbg4eq4r0x+rxada4VWlWltoSt7RLidvb2lBVqC1xc2iXHDS9oU0S1Z52WVnMGjdR7bgW59AfzaxpNBw8ckQGUASPuDtRzJs3j4KCgkjXddI0jTRNIz8/P6pbty41bNiQ6tatS35+fo55uq5TYGAgffbZZx63o0KzqkRHR5Ou69Xi06xxq9Dk4PFC8HR8zBq3Cp9mjdubYuceNwePF4L0R4lbpU9vib02xC3tUuL2tjZk5rg5tEsOmt7SJolqR7uUuI2JnUvcHPqjmTWNhoNHrsgAiuCRsieKVatWkdVqJU3T6OKLL6Z58+bRwYMH3a578OBBmjdvHl100UWkaRr5+PhQWlqay3IqNC8ETydHo32aNW4Vmhw8Xijujo9Z41bh06xxq9KsKpzj5uDxQpH+6IzEXTvbOve4pV1K3N7Whi4E7nFzaJccNL2pTRLxb5dVxaxxE/G+FufQH82saTQcPHJGBlAEj5Q9UVx11VWk6zpNmTKlUjovv/wyaZpGQ4cOdZmnQvPQoUNV/kRGRrotDIz2ada4VWhy8Ehk/PExa9wqfJo1bhWaZo2bg0ciHsdH4jZX3Co0zRq3Ck2ztkuzxq1C06xxq9Dk4FGFpvRH6Y9GaBodO5e4OfRHM2saDQePnNGIiGr6PSyCOg4fPlzlddu2bYvTp0+jqKgIAFCnTh34+/tj//79ldZq0KABbDYbjh8/7jRdhaau69A0rdJ6AEBE0DTNEbMqn2aNW4UmB4+A8cfHrHGr8GnWuFVomjVuDh4BHsdH4jZX3Co0zRq3Ck2ztkuzxq1C06xxq9Dk4FGFpvRH6Y9GaJr1WpxDfzSzptFw8MgZvaYNCGqJj49HgwYNqvTJyMhw0srKykKdOnWq5KNOnTrIyspyma5CE7Cf5Kry8YTRPs0atwpNDh5LMPL4mDVuFT7NGrcqTTPGzcFjCd5+fCRuY/RK8Pa4VWmaNW5pl8b4NGvcqjTNGjeHdslBU/qj9EcjNAFzXotz6I9m1jQaDh5ZQ0KtRtO0C/qce6tiq1atyN/fnw4cOFApD/v27SM/Pz9q2bKlyzwVmnFxcaTrOq1fv75SmkSen+1ptE+zxq1Ck4NHIuOPj1njVuHTrHGr0DRr3Bw8EvE4PhK3ueJWoWnWuFVomrVdmjVuFZpmjVuFJgePKjSlP0p/NELTrNfiHPqjmTWNhoNHzsgdKLWc+vXrQ9M0rFu3DsXFxZX6REVFOWndcMMNsNlsuPTSS7F69eoKbT8tLQ2XXXYZCgoKMHz4cJf5KjS7dOkCAFizZk2F9CqC0T7NGrcKTQ4eAeOPj1njVuHTrHGr0DRr3Bw8AjyOj8RtrrhVaJo1bhWaZm2XZo1bhaZZ41ahycGjCk3pj9IfjdA067U4h/5oZk2j4eCRM/IOlFrO1VdfjW+//RbvvPMORo8eXal1Y2JikJ6e7njWo81mQ8+ePbF+/XpomoaWLVuiZ8+eaNq0KSIiIuDr64v8/HxkZGRg9+7dSE1NxdatW0FE6NixI1JTU+Hn5+e0DRWa//3vf/HEE0/gjjvuwPvvv39BMavyada4uexLDsfHrHGr8GnWuFVomjVuDh65HB+J21xxq9A0a9wqNM3aLs0atwpNs8atQpODRy5xm7VdmjVuFbFziZtDfzSzptFw8MgaQ+9nEbyOF198kTRNozvvvLPS67q7VTE7O5vuvvtu8vHxcXrMl7uPpmnk4+NDo0ePpszMTI/bMVpz2bJlFBYWRr169ap0zGPHjqXbb7+9WnyaNW4O+1KFporjY9a4Vfg0a9xGa5o1bi4euRwfidtccRutada4VWmatV2aNW6jNc0atypNDh45xG3WdmnWuFXFziFuFZocPHLSNBoOHrkid6DUcpYvX46rrroKbdu2xR9//FGpdceNG4esrCzMnj3bZd7Ro0fx3XffYdWqVdixYwcyMjKQl5cHf39/REREIDExEV27dsWQIUMQGxtboe2p0FSB0T7NGrcKTQ4eVWDWuFX4NGvcqjSNhkPcHDyqgsO+VIHEba62ziVuaZcSt7e1IRVwiZtDu+SgyaFNqvLJIXaJ27vj5tAfzaxpNBw8ckMGUARBEARBEARBEARBEARBEARBEMogL5EXBEEQBEEQBEEQBEEQBEEQBEEogwygCIbz4Ycf4tlnnxVNL9QzsyYHjyo0OXjkosnBIxdNDh5VaHLwyEWTg0cVmhw8ctHk4JGLJgePKjQ5eOSiycEjF00OHlVocvDIRZODRy6aHDyq0OTgkZOm0XDw6JXU7CtYBG9h9uzZ9Mwzzxii1atXL9J1/fwLima165lZk4NHFZocPHLR5OCRiyYHjyo0OXjkosnBowpNDh65aHLwyEWTg0cVmhw8ctHk4JGLJgePKjQ5eOSiycEjF00OHlVocvDISdNoOHj0RuQOFAEA8P777+OZZ56paRuCIAiCIAiCIAiCIAiCIAiC4BXIAIogCIIgCIIgCIIgCIIgCIIgCEIZZABFEARBEARBEARBEARBEARBEAShDDKAIhgOEYmml+qZWZODRxWaHDxy0eTgkYsmB48qNDl45KLJwaMKTQ4euWhy8MhFk4NHFZocPHLR5OCRiyYHjyo0OXjkosnBIxdNDh5VaHLwyEnTaDh49EasNW1A8A6M7EBPPvkkjh07ZpiemTU5eOSiycGjCk0OHrlocvDIRZODRxWaHDxy0eTgUYUmB49cNDl45KLJwaMKTQ4euWhy8MhFk4NHFZocPHLR5OCRiyYHjyo0OXjkpGk0HDx6IxrJ0JMAYOHChTh27Bhuu+22mrYiCIIgCIIgCIIgCIIgCIIgCDWODKAIF0xRURHOnj2L3NxcBAQEIDQ0FBaLxRSagsCBVatWIT8/H3369DGdphk5deoUioqKEB0dDV035kmdRmuq8CgI3oDRtYbUQ4JgLEbXGlIPeS8c6iFBqK1wqV+kJhLMitQalUcGUIQq8e2332L+/PlIS0vD3r17UVxc7Jin6zoaN26Mrl274tprr8XQoUNrlWZlmTp1KnJycjBx4kSv1KuoZkZGBnbu3ImAgAC0atXqvBcNmzZtwpkzZ8pNyEZq2mw2/Pnnn8jJyUH79u3RsGFDx7yzZ8/izTffxPr161FYWIjk5GSMHDkSDRo0KHd7XDTPR2xsLE6cOIHCwsIL0uGiuW/fPnz33XfYuXMnCgsLER8fj0GDBqFz585eoVdVzeLiYuTl5cHPz8+lsN+wYQMmT56MRYsWITMzEwDg5+eHPn364LHHHkP//v2rRVOFx5EjR6JHjx649tprERYW5nH/VAYVmgDw+++/4+eff0ZhYSE6d+6M66+/3jFv1apVeOaZZ7B27Vrk5eUhKSkJt956K8aOHVvuhRoXTQDYv38/vvrqK6xatQrbt29HRkYGcnNzERwcjHr16iElJQVXXHEFLrnkkorsTsP1VGkaXWtIPVRzmrWhHgKMrzVqSz0EGF+/mKkeUqFZW+shwPhag1M9BBhfa5i5HlKhyaEe4qRZGWrL9zlmrIdUadYEKuqXWg8JpmHfvn00ffp0uv7666l9+/bUsGFDiomJoSZNmlD37t1p7NixtHDhwnI1tm/fTsnJyaTrOmmadt6PruvUvn172rZtG3vNqhIdHU26rnut3vk009PT6brrriOr1Uq6rpOu6xQVFUUvvPACFRQUeNTs1asXWSyWatH89ddfqU6dOg4tHx8f+r//+z8isrf7+vXrO7UFXdcpKCiIfvrpJ4/b4qJZEerVq2d4m6lJzRkzZtB3333nMr24uJgeeeQR8vHxcezjcz+XXXYZnTp1SrmeKs0JEyaQruv0ww8/OE2fO3cu+fr6esx3uq7T5MmTq0VThceS+QEBAXT99dfTd999R4WFhW6XrSgqNMeOHes4jiX6/fv3J5vNRosXLyY/Pz+3cV9xxRXsNbOzs+muu+5ytOtz13On1bJlS/r999+rTU+VptG1htRDNa/JvR4iMr7WqE31EJHx9UttqodUaJq1HiIyvtbgUg8RGV9rmLUeUqHJoR7ipFkVuH+fY9Z6SJVmTaGifqntyACKCTDqJHno0CGKjo4mTdMoOjqa7r33Xpo7dy6tWLGCtm7dSrt376atW7fSihUraO7cuXTvvfc6lo+JiaFDhw6x1bwQOA+g2Gw26tChg8e2kpKSQvv373er2atXr2rR3LVrFwUGBpKmaRQZGUkpKSkUHBxMuq7Txx9/TJdccglpmkZXXnklvfvuu/Tmm2/SwIEDSdM0CgsLo8OHD7tsh4NmYmJihT8lhci505KSklw8ctHUNI169+7tMr3kQkjTNGrQoAFdc801dOONN1Lbtm0d7atbt24uF4ZG66nS7N69OwUEBFBeXp5j2p49exwXeJ06daI5c+bQ2rVr6e+//6Zvv/2WrrrqKofu4sWLlWuq8OguT8TExNC4ceMoLS3NZfmKYLTm119/7dAZNmwYPfzww9SjRw/SdZ1eeOEFat68OVksFrr77rtp/vz59MUXX9CoUaPIYrGQruv04YcfstW02WzUpUsX0nWd/P39qVevXnTDDTdQr169KCAggHRdpxEjRtCsWbPo7rvvptjYWNI0jXx8fGjevHnK9VRpGl1rSD3kHZqc6yEi42sNDvUQkfG1hlnrIRWaZq2HSmI3stbgUA8RGV9rmLUeUqHJoR7ipFlVOH+fY9Z6SJWm0aioNYRSZACllmPkSfLuu+8mTdPo2muvpczMzAptPzMzk6655hrSNI3uuecel/lcNC8EzgMor7zyCmmaRnXr1qV58+ZRRkYGHTlyhF599VVHQREfH09btmxxWdfTycxozTFjxpCmaXTTTTdRfn4+ERGdOnWKunXrRgkJCWSxWGjcuHEuWrfeeivpuk5PPfWUyzwOmiUFRdkio6Ifd8eGk2bZi/GNGzeSpmlksVjo9ddfp6KiIqf5v/zyC0VGRpKu6/TBBx8o1VOlGRMT41LUPPnkk6RpGo0YMcJl+RLeeecd0jSNLr/8cuWaKjxqmkY9e/akxYsX06233kohISGOtqHrOjVv3pxeeOEF2rt3r0d91ZqXXnop6bpOs2bNcpp+++23Owrrd99912W9t99+mzRNo4EDB7LVnDx5MmmaRv3796d9+/Y5zdu3bx/17NmTLBYLLVu2jIiICgsLaerUqWS1WikwMJB27dqlVE+VptG1htRD3qHJuR4iMr7W4FAPERlfa5i1HlKhadZ6iMj4WoNDPURkfK1h1npIhSaHeoiTZlXh/H2OWeshVZpGo6LWEEqRAZRajpEnyYYNG1JERATl5ORUykN2djaFh4dTgwYNXOZx0azMSG7ZT8kvVVTqqdLs3r076bpOS5YscZl3+PBh6tWrF2ma/VcZa9eudZrv6WRmtGZSUhL5+/u73Oa/ZMkS0jSN/Pz8KD093WVb+/btI4vFQl26dHGZx0Gz5I6yoUOH0m+//UbLli1z+1m6dKnjIrTsvLJw0XR3Mf7444+Tpmn04IMPuixfQsmvzS666CKleqo0/fz8qGvXrk7TrrjiCtJ1/bwXtg0aNKCoqCjlmio8lt2XOTk59Omnn9Kll15KVqvV6UK/b9++NGvWLDp9+nS52zJaMyYmhurWresyfdu2baRpmtu4iOyPMImMjKTo6Gi2mm3btqWAgAA6efKk23UPHz5MPj4+dPHFFztNL7lYuu+++5TqqdI0utYwaz2kQtOs9RCR8bUGh3qIyPhaw6z1kApNs9ZDRMbXGhzqISLjaw2z1kMqNDnUQ1w0OdRDRMbXGmath1RpGo2KWkMoRQZQajlGniT9/PyoU6dOVfLRqVMn8vf3d5nORdPokVwVI8MqNMPCwtwWICXk5+fT8OHDSdPstyWe++g3TyczozUDAgKoRYsWLjoZGRmkaVq5tyHGx8dTRESEy3QOmlu2bKHevXuTpmmUnJxMK1eu9Lh+RZ9vyUXT3cX4lVdeSbqu0/bt28tdt06dOlSnTh2leqo0GzZsSLGxsU7TLrnkEtJ13fErGE906dLFbW4zWlOFR3f7soRjx47RK6+8QikpKU657HzP8jZa08fHhzp37uyilZeXR5qmUUpKise4U1JSyMfHx2U6F82goCC3mufSpk0bCgsLc5pms9koNDSUmjVrplRPlabRtYZZ6yEVmmath4iMrzU41ENExtcaZq2HVGiatR4iMr7W4FAPERlfa5i1HlKhyaEe4qLJoR4iMr7WMGs9pErTaFTUGkIp1pp+ib2glt27d6NNmzaIiopyOz82NhbNmzdHWlqa0/QxY8Zg0qRJWLhwoWNanTp1sGvXLuTl5cHf37/CHnJzc7Fz507ExMS4zOOiGRoaiszMTMyePRuNGzeusCYRYejQocjMzFSqp0ozNzcXzZs397iuj48PPv30U0RGRuLtt9/GpZdeii+//BKXXXaZx3WM1tR1HUFBQS7Tw8PDAQB169b1uK369etj48aNLDVbtmyJ33//HbNmzcL//d//oVevXrjjjjvw3//+12N/Px9cNN2Rk5MDAGjatGm5yzVp0gQbNmyodj0jNPv27YtPPvkES5YswYABAwAAbdq0waJFi7B69Wr07NnTrV5WVha2bt2K+Ph45ZoqPJZHnTp1MH78eIwfPx7btm3Dxx9/jE8//RT79+/H/Pnz8eWXXyI6OhrHjh1TqhkSEoKjR4+6aJVMO3TokMftHTx40G1u4KJptVpx5swZj+sBQEZGBgoLC52m+fr6IikpCf/8849SPVWaRtcaZq2HVGiatR4CjK81ONRDgPG1htRDajXNUA+dD6PrF2+phwDjaw2z1kMqNDnUQ1w0OdRDJf6NrDXMWg+p0jSa6qo1TEvNjd0I1UFYWNh5XwQUFxdHQUFBLtM7depEAQEBjv9HjRpFmqbR9ddfT1lZWRXaflZWFl177bWk6zqNHj3aZT4XzYEDB5Ku6/TJJ59USO9c3D2P0mg9VZrx8fEuv8byxBNPPEGappG/vz/NmzfP468BjNZs0qSJR73yflVFRNSuXTu3v2zjolnCiRMn6OabbyZNs99yPnPmTKf5Vfl1gTdruttft99+O+m6TkeOHCl33eTkZJfb8o3WU6W5Zs0aslgs1KhRI8evNnfs2EGBgYHUunVr2rlzp8s6GRkZNGzYMNJ1nR566CHlmio8nq9/uGPZsmU0cuRICgsL8/gLLyM1+/btS7qu08KFC52mP/HEE+Tr60u6rtM333zjorlgwQLSNI169OjhMo+LZs+ePR23f7vjxx9/9PhrzubNm1NkZKRSPVWaRtcaZq2HVGiatR4iMr7W4FYPERlfv5ipHlKhadZ6yFPs56O8WoNDPURkfK1h1npIhSaHeoiLJod6iMj4WsOs9ZAqTZWoqF/Mjgyg1HKMPEkeOnTI8Zy8mJgYuu++++jTTz+lVatW0fbt22nPnj20fft2WrVqFX366ad03333UUxMDOm6TlFRUXTw4EGXbXDRfPzxx0nXdRo/fnxldj8RuT+ZGa2nSrPkFvvNmzdXSOfll18mTdPIarV6LKqN1rzooovIYrG4fUzd6dOnPRZIRUVFFBgYSMnJyS7zuGiWZdGiRZSQkEC6rlP37t1pw4YNRHRhJ0dv1NQ0jcLDw6l///6OT7NmzUjXdVq6dKnH9YqLiyk4OJhatWqlVE+VJhHR888/T5qmUWBgIN1999303Xff0dSpU8nX15d8fHxoyJAh9Oijj9JTTz1FN998s6PPxMfHu30mqwpNo/WqcnFfQl5eHn3xxRfKNWfPnu2I+ZFHHqG3336bbrnlFtJ1nW688Ubq168fhYSE0JQpU2jdunW0bt06evnllykkJIR0Xafp06e7bIeL5syZMx231E+ZMoV2795NNpuNdu/eTVOnTqXw8HDSdZ2mTZvmtF5ubi75+flRx44dleqp0jS61jBrPaRC06z1EJHxtQbXeojI+PrFDPWQCk0z10NG1xoc6iEi42sNs9ZDKjQ51ENcNDnUQ0TG1xpmrYdUaVYHKuoXsyIDKLUco0+S//zzD7Vq1crxjMXzfTRNo5YtW5abDDlolvdrk/PRoUMHatKkiVI9VZol78J54IEHKqw1Y8YMx0vM3CVkozUnTJhAmqbRZ599VmE9IqKlS5eSpml07733uszjoumOvLw8mjBhAvn5+ZHVaqVx48ZRVFTUBZ0cvU2zvGe/jhw50uN6JS94Gz58uFI9VZolzJgxg0JDQ13ym7v/NU2jzp070549ezzqqdA0Uu9CLu49oUJz2LBhTvFpmkaxsbG0f/9+Sk1NJX9/f7fnno4dO5LNZmOrWVxcTFdeeaXH862madSrVy+XZ8jPnTuXNM31JcJG66nSJDK+1jBjPaRC06z1EJHxtQbneojI+PqlttdDKjTNXA8ZXWtwqYeIjK81zFgPqdDkUg9x0ORQDxEZX2uYtR5SpVldqKhfzIgMoNRyVJwki4qKaP78+TR8+HBq0qQJWSwWp2LYYrFQkyZN6MYbb6QvvviCioqKzuvT2zVzc3Np48aN9M8//5x3uxXBaD1VmgcPHqT4+HhKSkqijIyMCq/3xRdfOG5/Vq25c+dOmjVrFq1Zs6bCWkRE999/PzVu3Ji+//57l3lcNMvj3BeIlfT/C8VbNJctW+bxk5aW5nG9G2+8kcLDw2nGjBlK9VRpnsvx48fpxRdfpJ49e1JgYKBLbktISKARI0ZUqt0YrWmUXt++fen++++vcBwVQYVmcXExzZkzh4YPH05Dhw6liRMn0okTJxzzly5dSh06dHCce8PDw+nee++l06dPs9csLCykZ555hqKjo52Oc2hoKD300EOUm5vrss7Bgwdp48aNdOrUKeV6qjSJjK9fzFYPqdA0az1EZHytURvqISLj65faWg+p0DRzPWR0rcGlHiIyvtYwaz2kQpNLPeTtmhzqISLjaw2z1kOqNKsbFfWLmdCIiGr6PSyCWoqKivDCCy/gjTfewKlTpxzTQ0JCMGrUKDz//PMuL9I6dOgQTp48iQYNGiAyMrJc/YKCApw5c8bxQq6wsDD4+PhckGcumoLg7Xz00UeOF5a98sorptI0C0SEM2fOIDs725HbrFarV2mq8MiVnJwcZGVlISYmBpqm1SpNIsK2bduQkZGBsLAwtGjRAhaLpcoejNZTpXkuRtcaUg8JgnEYXWtIPeRdcKiHBGeMrl9qaz2kQpNbPcRJUxC8Hak1qoYMoJgI1SdJQRAEQRAEQRAEQRAEQRAEQagtyACKIAiCIAiCIAiCIAiCIAiCIAhCGeT+U6FKbNy4EV9++SXS0tKwY8cOZGRkIDc3FwEBAYiIiEBiYiK6du2Ka665Bh06dBBN5h65aHLwKHF7tyYHj1w0OXiUuL1bk4PHqvDFF18gLy8Pt956q1fqmVmTg0cumhw8qtDk4JGLJgePXDQ5eFShycEjF82K6p04cQIBAQEIDg4+r+bx48eRl5eHhg0bstfk4FGFJgePnDSNhoNHVlT3S1eEmiMnJ4e+/vprevTRR+mqq66iPn36UOfOnal///40fPhwmjp1Km3btq1cjePHj9OVV17p9BJ6T5+SZa644go6duyYaDL0yEWTg0eJ27s1OXjkosnBo8Tt3ZocPF4I0dHRZLFYvFbPzJocPHLR5OBRhSYHj1w0OXjkosnBowpNDh65aJanZ7PZ6JFHHqGoqChHDdW+fXv65JNPytXs1asXa00OHlVocvDISdNoOHjkijzCywQUFxfjxRdfxJQpU5CZmekyv6QJaP97KdqgQYPw1ltvISEhwWm5M2fOoFOnTti9ezcsFgsGDRqEnj17omnTpoiIiICfnx9sNhsyMjKwe/dupKamYvHixSgsLETTpk2xbt06hIWFiSYTj1w0OXiUuL1bk4NHLpocPErc3q3JweOFEhMTg/T0dBQVFXmlnpk1OXjkosnBowpNDh65aHLwyEWTg0cVmhw8ctH0pEdEuOyyy7Bo0SKU/WpR0zRcfvnlmDNnjts6qnfv3lixYgVLTQ4eJW7v1zQaDh5ZU21DNUKNUFxcTEOGDHH82jI+Pp66du1K8fHxjl9Z3njjjfTkk0/SpZdeSv7+/qRpGoWEhNDy5cudtB5++GHSNI169+5N+/fvr9D29+3bRz179iRd1+mRRx5xmW9WTQ4euWhy8KhCk4NHLpocPHLR5OBRhSYHj1w0OXi8UKKjo0nXda/VM7MmB49cNDl4VKHJwSMXTQ4euWhy8KhCk4NHLpqe9GbPnk2aplFgYCC99NJLtGHDBlq5ciWNHz+e/Pz8SNd1ateuHR09etRl3V69erHV5OBR4vZ+TaPh4JEzcgdKLeedd97Bfffdh9atW2PWrFno2rWrY97q1atx++23Y8+ePUhLS0O7du2QkZGBiRMn4q233kJkZCS2bNmCOnXqAAASEhJw7Ngx7N+/HxERERX2cOrUKTRq1Aj16tXDzp07neaZVZODRy6aHDyq0OTgkYsmB49cNDl4VKHJwSMXTQ4eAeDiiy+usE5Zli9fjsLCQqdfeBmtZ2ZNDh65aHLwqEKTg0cumhw8ctHk4FGFJgePXDRVeBwwYACWL1+OefPm4dprr3Wat3nzZlxzzTXYsWMHmjZtisWLF6Nx48aO+Z5+8c5Bk4NHidv7NY2Gg0fW1PQIjqCWzp07k4+PD+3bt8/t/O3bt5PFYqFhw4Y5TX/sscdI0zR69NFHHdP8/f2pU6dOVfLRqVMn8vf3d5luVk0OHrlocvCoQpODRy6aHDxy0eTgUYUmB49cNDl4JCLHXbzlvUvlfO9ZUalnZk0OHrlocvAocXu3JgePXDQ5eJS4vVtThcfIyEiqV6+ey/QSTp8+TX379iVNsz8NZevWrY55nn7xzkGTg0cVmhw8ctI0Gg4eOWOt6QEcQS3btm1DmzZt0LBhQ7fzk5KSkJCQgOXLlztNf+qpp/D666/j+++/x0svvQQAiIiIwIEDB1BUVASLxVJhD4WFhdi/fz/Cw8Nd5plVk4NHLpocPKrQ5OCRiyYHj1w0OXhUocnBIxdNDh4BICAgAHl5eZg8eTJiY2MrrAkA48aNQ3Z2tlI9M2ty8MhFk4NHFZocPHLR5OCRiyYHjyo0OXjkoqnCY2ZmJjp06OBxvbCwMPz666+4/vrr8d1336Fv375YuHBhuetw0OTgUYUmB4+cNI2Gg0fW1PQIjqCW4OBgatOmTbnLNGnShAIDA12md+jQgYKCghz/33TTTaTrOj3wwANUXFxcoe0XFxfTf/7zH9J1nW666SaX+WbV5OCRiyYHjyo0OXjkosnBIxdNDh5VaHLwyEWTg0ei0l9pffnllxXSOxd3zxE3Ws/Mmhw8ctHk4FGFJgePXDQ5eOSiycGjCk0OHrloqvBYr149atiw4XnXLywspFtuuYU0TaPw8HD6448/PP7inYMmB48qNDl45KRpNBw8ckYGUGo5KSkpZLFY6O+//3Y7Py0tjTRNo9atW7vMa926NYWFhTn+37p1KwUGBpKu69S6dWuaMmUKrVixgo4ePUo2m42IiGw2Gx09epRWrFhBU6ZModatW5Ou6xQYGEhbtmxx2YZZNTl45KLJwaPE7d2aHDxy0eTgUeL2bk0OHomIHnroIdJ1nR577DGXeefD3ZcQRuuZWZODRy6aHDyq0OTgkYsmB49cNDl4VKHJwSMXTRUeL774YtJ1nfbs2VMhnTFjxpCmaRQUFER169Zlq8nBowpNDh45aRoNB4+ckQGUWs7UqVMdz7ebP38+FRQUEBFRQUEBffnllxQfH0+6rtPEiROd1issLKTAwECXu1eWLFlCMTExjmdgnu+jaRrFxMTQb7/95tGjWTU5eOSiycGjxO3dmhw8ctHk4FHi9m5NDh7nzZtHmqZRv379PG7TE1FRUaTrzhcoRuuZWZODRy6aHDyq0OTgkYsmB49cNDl4VKHJwSMXTRUeJ0+eTLqu06RJkyqs9cQTTzjVYxw1OXhUocnBIydNo+HgkTMygFLLsdls1LVrV0eHsFqtFBsbSz4+Po4vCFq2bElnz551Wu+bb74hTdPo3nvvddHMzMykl19+mbp3706+vr6kaa4vGPP19aVu3brRyy+/7KLtDrNqcvDIRZODR4nbuzU5eOSiycGjxO3dmt7u8fTp0/TNN9/QL7/8ct7tVgSj9cysycEjF00OHlVocvDIRZODRy6aHDyq0OTgkYumCo///vsvaZpGkZGRdPr06QqvN23aNI9f2HLQ5OBRhSYHj5w0jYaDR85oREQ1/R4WQS1ZWVm4//77MWfOHBQVFTmma5qGYcOG4Z133kFMTIzTOps3b8auXbvQvn17NG7c2KN2UVER9u7di4yMDOTl5cHf3x8RERFo3LhxpV7UKpo8PHLR5OBRhSYHj1w0OXjkosnBowpNDh65aHLwKAiCIAiCYEaKi4tBRJWun44cOYL8/Hw0atSIpSYHjyo0OXjkpGk0HDxyRQZQTER6ejrS0tKQkZGBsLAwdOrUCXXr1q1pW4IgCIIgCIIgCIIgCIIgCILgdcgAiiAIgiAIgiAIgiAIgiAIgiAIQhmsNW1A4MmJEyfw7bffIi0tDTt27EBGRgZyc3MREBCAiIgIJCYmomvXrhgyZAjq1Kkjmsw9ctHk4FHi9m5NDh65aHLwKHF7tyYHjxK3d2ty8MhFk4NHidu7NTl45KLJwaPE7d2aHDxy0eTgUeL2fk2j4eCRHdX+1hWhRlm/fj29/fbb9J///IduvfVWuu666+iOO+6gxx9/nL788svzvmgoNzeXxo0bR35+fo6X0Hv66LpOfn5+NHbsWMrJyRFNhh65aHLwKHF7tyYHj1w0OXiUuL1bk4NHidu7NTl45KLJwaPE7d2aHDxy0eTgUeL2bk0OHrlocvAocXu/ptFw8MgVeYSXSfjkk0/w9NNPY/fu3S7ziAiapgEA/P39ccstt+DFF19EZGSk03I2mw19+/bFmjVrQERo0aIFevbsiaZNmyIiIgJ+fn6w2WzIyMjA7t27kZqaim3btkHTNHTq1Al//PEHfH19RZOJRy6aHDxK3N6tycEjF00OHiVu79bk4FHi9m5NDh65aHLwKHF7tyYHj1w0OXiUuL1bk4NHLpocPErc3q9pNBw8suZCRl8EHowZM8Yx8mixWKhu3bpksVgcI45XXXUV3XLLLdSiRQvHtPr169OWLVucdJ577jnSNI1atGhBK1asqNC2U1NTqXnz5qTrOj333HMu882qycEjF00OHlVocvDIRZODRy6aHDyq0OTgkYsmB48qNDl45KLJwSMXTQ4eVWhy8MhFk4NHLpocPKrQ5OCRiyYHj1w0OXhUocnBIydNo+HgkTMygFLL+eyzz0jTNIqLi6N58+ZRfn4+ERHl5+fTvHnzKC4ujiIjI2nv3r1ERLR582a68sorSdM0atiwIWVmZjq0WrVqRX5+frR///5Kedi7dy/5+vpSy5YtXeaZVZODRy6aHDyq0OTgkYsmB49cNDl4VKHJwSMXTQ4eVWhy8MhFk4NHLpocPKrQ5OCRiyYHj1w0OXhUocnBIxdNDh65aHLwqEKTg0dOmkbDwSNnZAClltOnTx+yWCz0119/uZ2/atUq0jSNbr31Vqfpt912m8sIZEBAAKWkpFTJR8eOHSkgIMBlulk1OXjkosnBowpNDh65aHLwyEWTg0cVmhw8ctHk4FGFJgePXDQ5eOSiycGjCk0OHrlocvDIRZODRxWaHDxy0eTgkYsmB48qNDl45KRpNBw8ckav6UeICWr566+/0LJlS7Rt29bt/K5du6Jx48b45ZdfnKZPnz4dFosFX331lWNacHAwjh8/XiUfx48fR1BQkMt0s2py8MhFk4NHFZocPHLR5OCRiyYHjyo0OXjkosnBowpNDh65aHLwyEWTg0cVmhw8ctHk4JGLJgePKjQ5eOSiycEjF00OHlVocvDISdNoOHjkjAyg1HLy8/Ph4+NT7jK+vr44c+aM07TIyEi0bNnS6aXz3bt3x6FDhzB9+vRKeZg6dSoOHTqEHj16uMwzqyYHj1w0OXhUocnBIxdNDh65aHLwqEKTg0cumhw8qtDk4JGLJgePXDQ5eFShycEjF00OHrlocvCoQpODRy6aHDxy0eTgUYUmB4+cNI2Gg0fW1PQtMIJa2rRpQz4+Ph6fgffvv/+SxWKhhIQEl3nt27enkJAQx/8rVqwgq9VKuq7TZZddRvPnz6fDhw+71T18+DDNnz+fLr30UtJ1naxWK61cudJlObNqcvDIRZODR4nbuzU5eOSiycGjxO3dmhw8StzercnBIxdNDh4lbu/W5OCRiyYHjxK3d2ty8MhFk4NHidv7NY2Gg0fOyABKLeepp54iTdOoXbt2tHr1aqd5a9asodatW5Ou6/TAAw+4rBsaGkrNmzd3mjZ37lzy9/cnTdNI13XSdZ0CAgIoNjaWGjVqRLGxsRQQEOCYp2ka+fn50ccff+zRo1k1OXjkosnBo8Tt3ZocPHLR5OBR4vZuTQ4eJW7v1uTgkYsmB48St3drcvDIRZODR4nbuzU5eOSiycGjxO39mkbDwSNXZACllnPmzBlKTEx0dJ6GDRtS9+7dqWHDho7OUr9+fTp27JjTer/99htpmka33367i+bevXtpzJgxVL9+fdI0zeMnNjaWxowZQ3v27DmvT7NqcvDIRZODR4nbuzU5eOSiycGjxO3dmhw8StzercnBIxdNDh4lbu/W5OCRiyYHjxK3d2ty8MhFk4NHidv7NY2Gg0eOaERENf0YMUEthw8fxi233IJly5a5zEtJScEnn3yCpKQkp+mpqalYu3Yt+vXrh/bt23vU3r9/P3bs2IGMjAzk5eXB398fERERSExMRMOGDavk16yaHDxy0eTgUYUmB49cNDl45KLJwaMKTQ4euWhy8KhCk4NHLpocPHLR5OBRhSYHj1w0OXjkosnBowpNDh65aHLwyEWTg0cVmhw8ctI0Gg4euSADKCbir7/+QmpqKjIyMhAWFobOnTujS5cuNW1LEARBEARBEARBEARBEARBELwOGUARBEEQBEEQBEEQBEEQBEEQBEEog17TBgTv5dChQ9i/f7/L9J9++gnDhg1D69at0alTJ4wfPx579uwpV+v6669Hs2bNPM43qyYHj1w0OXhUocnBIxdNDh65aHLwqEKTg0cumhw8qtDk4JGLJgePXDQ5eFShycEjF00OHrlocvCoQpODRy6aHDxy0eTgUYUmB4+cNI2Gg0eW1OQLWATvJjo6miwWi9O0iRMnkq7rjhfQa5r95fT+/v40bdo0j1q9evUiXdfdzjOrJgePXDQ5eJS4vVuTg0cumhw8StzercnBo8Tt3ZocPHLR5OBR4vZuTQ4euWhy8Chxe7cmB49cNDl4lLi9X9NoOHjkigygCB6Jjo526kBLly51dMDBgwfTK6+8Qi+++CJ17NjR0SlHjBhBhYWFLlqeOqNZNTl45KLJwaPE7d2aHDxy0eTgUeL2bk0OHiVu79bk4JGLJgePErd3a3LwyEWTg0eJ27s1OXjkosnBo8Tt/ZpGw8EjZ2QARfBI2QGUa665hnRdpwkTJrgsO2PGDAoICCBd12nIkCGUl5fnNN9TZzSrJgePXDQ5eJS4vVuTg0cumhw8StzercnBo8Tt3ZocPHLR5OBR4vZuTQ4euWhy8Chxe7cmB49cNDl4lLi9X9NoOHjkjAyg1HJeeOGFKn+CgoKcOlD9+vUpNDSUCgoK3G5r3bp1FBsbS7qu04ABAygrK8sxz1NnNKsmB49cNDl4lLi9W5ODRy6aHDxK3N6tycGjxO3dmhw8ctHk4FHi9m5NDh65aHLwKHF7tyYHj1w0OXiUuL1f02g4eOSMDKDUckpu06rKp2TdEnx9fSklJaXc7e3evZuaNWtGuq5Tt27d6PTp00TkuTOaVZODRy6aHDyq0OTgkYsmB49cNDl4VKHJwSMXTQ4eVWhy8MhFk4NHLpocPKrQ5OCRiyYHj1w0OXhUocnBIxdNDh65aHLwqEKTg0dOmkbDwSNnZACllmO1WknXdbr22mvp9ttvr9TH39/fqQOFh4dTixYtzrvNI0eOUJs2bUjTNGrXrh0dPXrUY2c0qyYHj1w0OXhUocnBIxdNDh65aHLwqEKTg0cumhw8qtDk4JGLJgePXDQ5eFShycEjF00OHrlocvCoQpODRy6aHDxy0eTgUYUmB4+cNI2Gg0fOyABKLad9+/ak6zr98ssvlV637DtQunXrRlar1TFCWR7p6enUtWtX0jSNEhMTKSkpyW1nNKsmB49cNDl4VKHJwSMXTQ4euWhy8KhCk4NHLpocPKrQ5OCRiyYHj1w0OXhUocnBIxdNDh65aHLwqEKTg0cumhw8ctHk4FGFJgePnDSNhoNHzsgASi1n1KhRpOs6vfDCC5Vet+wAyiOPPEK6rtOMGTMqtH5WVhYNGDCANK30MWJlMasmB49cNDl4VKHJwSMXTQ4euWhy8KhCk4NHLpocPKrQ5OCRiyYHj1w0OXhUocnBIxdNDh65aHLwqEKTg0cumhw8ctHk4FGFJgePnDSNhoNHzsgASi1n5syZpGkaDRkypNLrRkVFOXWgP//8kzRNo6SkJCoqKqqQhs1mo6FDhzo6ZFnMqsnBIxdNDh5VaHLwyEWTg0cumhw8qtDk4JGLJgePKjQ5eOSiycEjF00OHlVocvDIRZODRy6aHDyq0OTgkYsmB49cNDl4VKHJwSMnTaPh4JEzGhERhFrL3r178dprryE6OhoTJkyo1LorVqxAQUEB+vbt65i2ZMkSEBG6du2K4ODgCukUFxfjiy++gM1mw2233eYy36yaHDxy0eTgUYUmB49cNDl45KLJwaMKTQ4euWhy8KhCk4NHLpocPHLR5OBRhSYHj1w0OXjkosnBowpNDh65aHLwyEWTg0cVmhw8ctI0Gg4euSIDKIIgCIIgCIIgCIIgCIIgCIIgCGXQa9qAIAiCIAiCIAiCIAiCIAiCIAiCtyEDKIIgCIIgCIIgCIIgCIIgCIIgCGWQARRBEARBEARBEARBEARBEARBEIQyyACKIAiCIAiCIAiCIAiCIAiCIAhCGWQARRAEQRAEQRAEQRAEQRAEQRAEoQz/DzZwZEC8qoyIAAAAAElFTkSuQmCC",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# here we re-initialize dataloader so the data doesn't shuffled, so we can plot the values by date\n",
"\n",
"train_dataloader = DataLoader(dataset_train, batch_size=config[\"training\"][\"batch_size\"], shuffle=False)\n",
"val_dataloader = DataLoader(dataset_val, batch_size=config[\"training\"][\"batch_size\"], shuffle=False)\n",
"\n",
"stocks_Model.eval()\n",
"\n",
"# predict on the training data, to see how well the model managed to learn and memorize\n",
"\n",
"predicted_train = np.array([])\n",
"\n",
"for idx, (x, y) in enumerate(train_dataloader):\n",
" x = x.to(config[\"training\"][\"device\"])\n",
" out = stocks_Model(x)\n",
" out = out.cpu().detach().numpy()\n",
" predicted_train = np.concatenate((predicted_train, out))\n",
"\n",
"# predict on the validation data, to see how the model does\n",
"\n",
"predicted_val = np.array([])\n",
"\n",
"for idx, (x, y) in enumerate(val_dataloader):\n",
" x = x.to(config[\"training\"][\"device\"])\n",
" out = stocks_Model(x)\n",
" out = out.cpu().detach().numpy()\n",
" predicted_val = np.concatenate((predicted_val, out))\n",
"\n",
"if config[\"plots\"][\"show_plots\"]:\n",
"\n",
" # prepare data for plotting, show predicted prices\n",
"\n",
" to_plot_data_y_train_pred = np.zeros(num_data_points)\n",
" to_plot_data_y_val_pred = np.zeros(num_data_points)\n",
"\n",
" to_plot_data_y_train_pred[config[\"data\"][\"window_size\"]:split_index+config[\"data\"][\"window_size\"]] = scaler.inverse_transform(predicted_train)\n",
" to_plot_data_y_val_pred[split_index+config[\"data\"][\"window_size\"]:] = scaler.inverse_transform(predicted_val)\n",
"\n",
" to_plot_data_y_train_pred = np.where(to_plot_data_y_train_pred == 0, None, to_plot_data_y_train_pred)\n",
" to_plot_data_y_val_pred = np.where(to_plot_data_y_val_pred == 0, None, to_plot_data_y_val_pred)\n",
"\n",
" # plots\n",
"\n",
" fig = figure(figsize=(25, 5), dpi=80)\n",
" fig.patch.set_facecolor((1.0, 1.0, 1.0))\n",
" plt.plot(data_date, data_close_price, label=\"Actual prices\", color=config[\"plots\"][\"color_actual\"])\n",
" plt.plot(data_date, to_plot_data_y_train_pred, label=\"Predicted prices (train)\", color=config[\"plots\"][\"color_pred_train\"])\n",
" plt.plot(data_date, to_plot_data_y_val_pred, label=\"Predicted prices (validation)\", color=config[\"plots\"][\"color_pred_val\"])\n",
" plt.title(\"Compare predicted prices to actual prices\")\n",
" xticks = [data_date[i] if ((i%config[\"plots\"][\"xticks_interval\"]==0 and (num_data_points-i) > config[\"plots\"][\"xticks_interval\"]) or i==num_data_points-1) else None for i in range(num_data_points)] # make x ticks nice\n",
" x = np.arange(0,len(xticks))\n",
" plt.xticks(x, xticks, rotation='vertical')\n",
" plt.grid(visible=None, which='major', axis='y', linestyle='--')\n",
" plt.legend()\n",
" plt.show()\n",
"\n",
" # prepare data for plotting, zoom in validation\n",
"\n",
" to_plot_data_y_val_subset = scaler.inverse_transform(data_y_val)\n",
" to_plot_predicted_val = scaler.inverse_transform(predicted_val)\n",
" to_plot_data_date = data_date[split_index+config[\"data\"][\"window_size\"]:]\n",
"\n",
" # plots\n",
"\n",
" fig = figure(figsize=(25, 5), dpi=80)\n",
" fig.patch.set_facecolor((1.0, 1.0, 1.0))\n",
" plt.plot(to_plot_data_date, to_plot_data_y_val_subset, label=\"Actual prices\", color=config[\"plots\"][\"color_actual\"])\n",
" plt.plot(to_plot_data_date, to_plot_predicted_val, label=\"Predicted prices (validation)\", color=config[\"plots\"][\"color_pred_val\"])\n",
" plt.title(\"Zoom in to examine predicted price on validation data portion\")\n",
" xticks = [to_plot_data_date[i] if ((i%int(config[\"plots\"][\"xticks_interval\"]/5)==0 and (len(to_plot_data_date)-i) > config[\"plots\"][\"xticks_interval\"]/6) or i==len(to_plot_data_date)-1) else None for i in range(len(to_plot_data_date))] # make x ticks nice\n",
" xs = np.arange(0,len(xticks))\n",
" plt.xticks(xs, xticks, rotation='vertical')\n",
" plt.grid(visible=None, which='major', axis='y', linestyle='--')\n",
" plt.legend()\n",
" plt.show()"
]
},
{
"cell_type": "code",
"execution_count": 72,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Predicted close price of the next trading day: 163.78\n"
]
}
],
"source": [
"# predict on the unseen data, tomorrow's price \n",
"\n",
"stocks_Model.eval()\n",
"\n",
"x = torch.tensor(data_x_unseen).float().to(config[\"training\"][\"device\"]).unsqueeze(0).unsqueeze(2) # this is the data type and shape required, [batch, sequence, feature]\n",
"prediction = stocks_Model(x)\n",
"prediction = prediction.cpu().detach().numpy()\n",
"prediction = scaler.inverse_transform(prediction)[0]\n",
"\n",
"if config[\"plots\"][\"show_plots\"]:\n",
" \n",
" # prepare plots\n",
"\n",
" plot_range = 10\n",
" to_plot_data_y_val = np.zeros(plot_range)\n",
" to_plot_data_y_val_pred = np.zeros(plot_range)\n",
" to_plot_data_y_test_pred = np.zeros(plot_range)\n",
"\n",
" to_plot_data_y_val[:plot_range-1] = scaler.inverse_transform(data_y_val)[-plot_range+1:]\n",
" to_plot_data_y_val_pred[:plot_range-1] = scaler.inverse_transform(predicted_val)[-plot_range+1:]\n",
"\n",
" to_plot_data_y_test_pred[plot_range-1] = prediction\n",
"\n",
" to_plot_data_y_val = np.where(to_plot_data_y_val == 0, None, to_plot_data_y_val)\n",
" to_plot_data_y_val_pred = np.where(to_plot_data_y_val_pred == 0, None, to_plot_data_y_val_pred)\n",
" to_plot_data_y_test_pred = np.where(to_plot_data_y_test_pred == 0, None, to_plot_data_y_test_pred)\n",
"\n",
" # plot\n",
"\n",
" plot_date_test = data_date[-plot_range+1:]\n",
" plot_date_test.append(\"next trading day\")\n",
"\n",
" fig = figure(figsize=(25, 5), dpi=80)\n",
" fig.patch.set_facecolor((1.0, 1.0, 1.0))\n",
" plt.plot(plot_date_test, to_plot_data_y_val, label=\"Actual prices\", marker=\".\", markersize=10, color=config[\"plots\"][\"color_actual\"])\n",
" plt.plot(plot_date_test, to_plot_data_y_val_pred, label=\"Past predicted prices\", marker=\".\", markersize=10, color=config[\"plots\"][\"color_pred_val\"])\n",
" plt.plot(plot_date_test, to_plot_data_y_test_pred, label=\"Predicted price for next day\", marker=\".\", markersize=20, color=config[\"plots\"][\"color_pred_test\"])\n",
" plt.title(\"Predicted close price of the next trading day\")\n",
" plt.grid(visible=None, which='major', axis='y', linestyle='--')\n",
" plt.legend()\n",
" plt.show()\n",
"\n",
"print(\"Predicted close price of the next trading day:\", round(prediction, 2))"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "subslide"
}
},
"source": [
"#### Thanks!"
]
},
{
"cell_type": "markdown",
"metadata": {
"slideshow": {
"slide_type": "skip"
}
},
"source": [
"**Credits**\n",
"\n",
"This tutorial was written by [Moshe Kimhi](https://mkimhi.github.io/) and [Aviv A. Rosenberg](https://avivr.net). \n",
"To re-use, please provide attribution and link to the original.\n",
"\n",
"Some images in this tutorial were taken and/or adapted from the following sources:\n",
"\n",
"- Fundamentals of Deep Learning, Nikhil Buduma, Oreilly 2017\n",
"- Sebastian Ruder, \"On word embeddings - Part 1\", 2016, https://ruder.io\n",
"- Andrej Karpathy, http://karpathy.github.io\n",
"- MIT 6.S191\n",
"- Stanford cs231n\n",
"- S. Bai et al. 2018, http://arxiv.org/abs/1803.01271"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"celltoolbar": "Slideshow",
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.8.12"
}
},
"nbformat": 4,
"nbformat_minor": 4
}
![](\"img/mlp.png\")
![](\"https://raw.githubusercontent.com/mkimhi/cs3600-tutorials/bff05457bd30d5c798a6ec3bbc59d2ff23453eb2/T4/resources/rnn.gif\")
![](\"img/tasks.png\")
![](\"https://raw.githubusercontent.com/mkimhi/cs3600-tutorials/main/T4/resources/bow.png\")
![](\"https://raw.githubusercontent.com/mkimhi/cs3600-tutorials/main/T4/resources/tfidf.png\")
![](\"https://raw.githubusercontent.com/mkimhi/cs3600-tutorials/main/T4/resources/grams.png\")
![](\"img/rnn_cell.png\")
![](\"img/rnn_unrolled.png\")
![](\"img/rnn_use_cases.jpeg\")
![](\"img/bptt.png\")
![](\"img/tbptt.png\")
![](\"https://raw.githubusercontent.com/mkimhi/cs3600-tutorials/main/T4/resources/clipping.PNG\")
![](\"img/rnn_layered.png\")
![](\"https://raw.githubusercontent.com/mkimhi/cs3600-tutorials/main/T4/resources/LSTM.PNG\")
![](\"https://raw.githubusercontent.com/mkimhi/cs3600-tutorials/main/T4/resources/gru_1.svg\")
![](\"img/sentiment_analysis.png\")
![](\"img/word_embeddings.png\")