diff --git a/.gitignore b/.gitignore index 32b46f9..760e3e8 100644 --- a/.gitignore +++ b/.gitignore @@ -1,3 +1,4 @@ .ipynb_checkpoints .idea - +.tar.gz +.pth diff --git a/2_pytorch/PyTorch快速入门.ipynb b/2_pytorch/PyTorch快速入门.ipynb index 5b1a4f4..eb55881 100644 --- a/2_pytorch/PyTorch快速入门.ipynb +++ b/2_pytorch/PyTorch快速入门.ipynb @@ -4,9 +4,9 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## 2.2 PyTorch第一步\n", + "# PyTorch快速入门\n", "\n", - "PyTorch的简洁设计使得它入门很简单，在深入介绍PyTorch之前，本节将先介绍一些PyTorch的基础知识，使得读者能够对PyTorch有一个大致的了解，并能够用PyTorch搭建一个简单的神经网络。部分内容读者可能暂时不太理解，可先不予以深究，本书的第3章和第4章将会对此进行深入讲解。\n", + "PyTorch的简洁设计使得它入门很简单，在深入介绍PyTorch之前，本节将先介绍一些PyTorch的基础知识，使得读者能够对PyTorch有一个大致的了解，并能够用PyTorch搭建一个简单的神经网络。部分内容读者可能暂时不太理解，可先不予以深究，后续的课程将会对此进行深入讲解。\n", "\n", "本节内容参考了PyTorch官方教程[^1]并做了相应的增删修改，使得内容更贴合新版本的PyTorch接口，同时也更适合新手快速入门。另外本书需要读者先掌握基础的Numpy使用，其他相关知识推荐读者参考CS231n的教程[^2]。\n", "\n", @@ -25,21 +25,9 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 1, "metadata": {}, - "outputs": [ - { - "ename": "ImportError", - "evalue": "No module named 'torch'", - "traceback": [ - "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", - "\u001b[0;31mImportError\u001b[0m Traceback (most recent call last)", - "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m()\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0m__future__\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mprint_function\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0;32mimport\u001b[0m \u001b[0mtorch\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mt\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", - "\u001b[0;31mImportError\u001b[0m: No module named 'torch'" - ], - "output_type": "error" - } - ], + "outputs": [], "source": [ "from __future__ import print_function\n", "import torch as t" @@ -588,7 +576,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "注意：`grad`在反向传播过程中是累加的(accumulated)，这意味着每一次运行反向传播，梯度都会累加之前的梯度，所以反向传播之前需把梯度清零。" + "注意：`grad`在反向传播过程中是累加的(accumulated)，**这意味着每一次运行反向传播，梯度都会累加之前的梯度，所以反向传播之前需把梯度清零。**" ] }, { @@ -1090,10 +1078,11 @@ }, { "cell_type": "code", - "execution_count": 34, + "execution_count": 3, "metadata": {}, "outputs": [], "source": [ + "import torch as t\n", "import torchvision as tv\n", "import torchvision.transforms as transforms\n", "from torchvision.transforms import ToPILImage\n", @@ -1102,7 +1091,7 @@ }, { "cell_type": "code", - "execution_count": 35, + "execution_count": 4, "metadata": {}, "outputs": [ { @@ -1127,7 +1116,7 @@ "\n", "# 训练集\n", "trainset = tv.datasets.CIFAR10(\n", - " root='/home/cy/tmp/data/', \n", + " root='../data/', \n", " train=True, \n", " download=True,\n", " transform=transform)\n", @@ -1140,7 +1129,7 @@ "\n", "# 测试集\n", "testset = tv.datasets.CIFAR10(\n", - " '/home/cy/tmp/data/',\n", + " '../data/',\n", " train=False, \n", " download=True, \n", " transform=transform)\n", @@ -1164,7 +1153,7 @@ }, { "cell_type": "code", - "execution_count": 36, + "execution_count": 5, "metadata": {}, "outputs": [ { @@ -1176,12 +1165,12 @@ }, { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAGQAAABkCAIAAAD/gAIDAAALVElEQVR4nO1cW3MVxxGe2d1zk46E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\n", + "image/png": "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\n", "text/plain": [ - "" + "" ] }, - "execution_count": 36, + "execution_count": 5, "metadata": {}, "output_type": "execute_result" } @@ -1203,24 +1192,24 @@ }, { "cell_type": "code", - "execution_count": 37, + "execution_count": 6, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - " bird cat ship ship\n" + " cat deer horse plane\n" ] }, { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZAAAABkCAIAAAAnqfEgAAA3wElEQVR4nO19SY9kWZrVeYPNg5v5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\n+D9big2JqKJg6QAAAABJRU5ErkJggg==\n", + "image/png": "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\n", "text/plain": [ - "" + "" ] }, - "execution_count": 37, + "execution_count": 6, "metadata": {}, "output_type": "execute_result" } @@ -1243,7 +1232,7 @@ }, { "cell_type": "code", - "execution_count": 38, + "execution_count": 7, "metadata": {}, "outputs": [ { @@ -1251,11 +1240,11 @@ "output_type": "stream", "text": [ "Net(\n", - " (conv1): Conv2d (3, 6, kernel_size=(5, 5), stride=(1, 1))\n", - " (conv2): Conv2d (6, 16, kernel_size=(5, 5), stride=(1, 1))\n", - " (fc1): Linear(in_features=400, out_features=120)\n", - " (fc2): Linear(in_features=120, out_features=84)\n", - " (fc3): Linear(in_features=84, out_features=10)\n", + " (conv1): Conv2d(3, 6, kernel_size=(5, 5), stride=(1, 1))\n", + " (conv2): Conv2d(6, 16, kernel_size=(5, 5), stride=(1, 1))\n", + " (fc1): Linear(in_features=400, out_features=120, bias=True)\n", + " (fc2): Linear(in_features=120, out_features=84, bias=True)\n", + " (fc3): Linear(in_features=84, out_features=10, bias=True)\n", ")\n" ] } @@ -1296,7 +1285,7 @@ }, { "cell_type": "code", - "execution_count": 39, + "execution_count": 8, "metadata": {}, "outputs": [], "source": [ @@ -1320,30 +1309,39 @@ }, { "cell_type": "code", - "execution_count": 40, + "execution_count": 10, "metadata": {}, "outputs": [ { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/bushuhui/.virtualenv/dl/lib/python3.5/site-packages/ipykernel_launcher.py:25: UserWarning: invalid index of a 0-dim tensor. This will be an error in PyTorch 0.5. Use tensor.item() to convert a 0-dim tensor to a Python number\n" + ] + }, + { "name": "stdout", "output_type": "stream", "text": [ - "[1, 2000] loss: 2.238\n", - "[1, 4000] loss: 1.916\n", - "[1, 6000] loss: 1.703\n", - "[1, 8000] loss: 1.582\n", - "[1, 10000] loss: 1.544\n", - "[1, 12000] loss: 1.465\n", - "[2, 2000] loss: 1.425\n", - "[2, 4000] loss: 1.377\n", - "[2, 6000] loss: 1.364\n", - "[2, 8000] loss: 1.330\n", - "[2, 10000] loss: 1.331\n", - "[2, 12000] loss: 1.298\n", + "[1, 2000] loss: 2.210\n", + "[1, 4000] loss: 1.958\n", + "[1, 6000] loss: 1.723\n", + "[1, 8000] loss: 1.590\n", + "[1, 10000] loss: 1.532\n", + "[1, 12000] loss: 1.467\n", + "[2, 2000] loss: 1.408\n", + "[2, 4000] loss: 1.374\n", + "[2, 6000] loss: 1.345\n", + "[2, 8000] loss: 1.331\n", + "[2, 10000] loss: 1.338\n", + "[2, 12000] loss: 1.286\n", "Finished Training\n" ] } ], "source": [ + "from torch.autograd import Variable\n", + "\n", "t.set_num_threads(8)\n", "for epoch in range(2): \n", " \n", @@ -1383,7 +1381,7 @@ }, { "cell_type": "code", - "execution_count": 41, + "execution_count": 11, "metadata": {}, "outputs": [ { @@ -1395,12 +1393,12 @@ }, { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZAAAABkCAIAAAAnqfEgAAA0bklEQVR4nO19WZMc6XXdqcysvbq6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\nEyf/z8r/Ab+8NWulkQjIAAAAAElFTkSuQmCC\n", + "image/png": "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\n", "text/plain": [ - "" + "" ] }, - "execution_count": 41, + "execution_count": 11, "metadata": {}, "output_type": "execute_result" } @@ -1422,14 +1420,14 @@ }, { "cell_type": "code", - "execution_count": 42, + "execution_count": 12, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "预测结果: cat ship plane ship\n" + "预测结果: cat ship ship ship\n" ] } ], @@ -1452,7 +1450,7 @@ }, { "cell_type": "code", - "execution_count": 43, + "execution_count": 13, "metadata": {}, "outputs": [ { diff --git a/2_pytorch/PyTorch快速入门.py b/2_pytorch/PyTorch快速入门.py new file mode 100644 index 0000000..5b3d22b --- /dev/null +++ b/2_pytorch/PyTorch快速入门.py @@ -0,0 +1,533 @@ +# -*- coding: utf-8 -*- +# --- +# jupyter: +# jupytext_format_version: '1.2' +# kernelspec: +# display_name: Python 3 +# 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: 3.5.2 +# --- + +# # PyTorch快速入门 +# +# PyTorch的简洁设计使得它入门很简单，在深入介绍PyTorch之前，本节将先介绍一些PyTorch的基础知识，使得读者能够对PyTorch有一个大致的了解，并能够用PyTorch搭建一个简单的神经网络。部分内容读者可能暂时不太理解，可先不予以深究，后续的课程将会对此进行深入讲解。 +# +# 本节内容参考了PyTorch官方教程[^1]并做了相应的增删修改，使得内容更贴合新版本的PyTorch接口，同时也更适合新手快速入门。另外本书需要读者先掌握基础的Numpy使用，其他相关知识推荐读者参考CS231n的教程[^2]。 +# +# [^1]: http://pytorch.org/tutorials/beginner/deep_learning_60min_blitz.html +# [^2]: http://cs231n.github.io/python-numpy-tutorial/ + +# ### Tensor +# +# Tensor是PyTorch中重要的数据结构，可认为是一个高维数组。它可以是一个数（标量）、一维数组（向量）、二维数组（矩阵）以及更高维的数组。Tensor和Numpy的ndarrays类似，但Tensor可以使用GPU进行加速。Tensor的使用和Numpy及Matlab的接口十分相似，下面通过几个例子来看看Tensor的基本使用。 + +from __future__ import print_function +import torch as t + +# 构建 5x3 矩阵，只是分配了空间，未初始化 +x = t.Tensor(5, 3) +x + +# 使用[0,1]均匀分布随机初始化二维数组 +x = t.rand(5, 3) +x + +print(x.size()) # 查看x的形状 +x.size()[1], x.size(1) # 查看列的个数, 两种写法等价 + +# `torch.Size` 是tuple对象的子类，因此它支持tuple的所有操作，如x.size()[0] + +y = t.rand(5, 3) +# 加法的第一种写法 +x + y + +# 加法的第二种写法 +t.add(x, y) + +# 加法的第三种写法：指定加法结果的输出目标为result +result = t.Tensor(5, 3) # 预先分配空间 +t.add(x, y, out=result) # 输入到result +result + +# + +print('最初y') +print(y) + +print('第一种加法，y的结果') +y.add(x) # 普通加法，不改变y的内容 +print(y) + +print('第二种加法，y的结果') +y.add_(x) # inplace 加法，y变了 +print(y) +# - + +# 注意，函数名后面带下划线**`_`** 的函数会修改Tensor本身。例如，`x.add_(y)`和`x.t_()`会改变 `x`，但`x.add(y)`和`x.t()`返回一个新的Tensor， 而`x`不变。 + +# Tensor的选取操作与Numpy类似 +x[:, 1] + +# Tensor还支持很多操作，包括数学运算、线性代数、选择、切片等等，其接口设计与Numpy极为相似。更详细的使用方法，会在第三章系统讲解。 +# +# Tensor和Numpy的数组之间的互操作非常容易且快速。对于Tensor不支持的操作，可以先转为Numpy数组处理，之后再转回Tensor。 + +a = t.ones(5) # 新建一个全1的Tensor +a + +b = a.numpy() # Tensor -> Numpy +b + +import numpy as np +a = np.ones(5) +b = t.from_numpy(a) # Numpy->Tensor +print(a) +print(b) + +# Tensor和numpy对象共享内存，所以他们之间的转换很快，而且几乎不会消耗什么资源。但这也意味着，如果其中一个变了，另外一个也会随之改变。 + +b.add_(1) # 以`_`结尾的函数会修改自身 +print(a) +print(b) # Tensor和Numpy共享内存 + +# Tensor可通过`.cuda` 方法转为GPU的Tensor，从而享受GPU带来的加速运算。 + +# 在不支持CUDA的机器下，下一步不会运行 +if t.cuda.is_available(): + x = x.cuda() + y = y.cuda() + x + y + +# 此处可能发现GPU运算的速度并未提升太多，这是因为x和y太小且运算也较为简单，而且将数据从内存转移到显存还需要花费额外的开销。GPU的优势需在大规模数据和复杂运算下才能体现出来。 +# +# ### Autograd: 自动微分 +# +# 深度学习的算法本质上是通过反向传播求导数，而PyTorch的**`Autograd`**模块则实现了此功能。在Tensor上的所有操作，Autograd都能为它们自动提供微分，避免了手动计算导数的复杂过程。 +# +# `autograd.Variable`是Autograd中的核心类，它简单封装了Tensor，并支持几乎所有Tensor有的操作。Tensor在被封装为Variable之后，可以调用它的`.backward`实现反向传播，自动计算所有梯度。Variable的数据结构如图2-6所示。 +# +# +# ![图2-6:Variable的数据结构](imgs/autograd_Variable.svg) +# +# +# Variable主要包含三个属性。 +# - `data`：保存Variable所包含的Tensor +# - `grad`：保存`data`对应的梯度，`grad`也是个Variable，而不是Tensor，它和`data`的形状一样。 +# - `grad_fn`：指向一个`Function`对象，这个`Function`用来反向传播计算输入的梯度，具体细节会在下一章讲解。 + +from torch.autograd import Variable + +# + {"scrolled": true} +# 使用Tensor新建一个Variable +x = Variable(t.ones(2, 2), requires_grad = True) +x + +# + {"scrolled": true} +y = x.sum() +y +# - + +y.grad_fn + +y.backward() # 反向传播,计算梯度 + +# y = x.sum() = (x[0][0] + x[0][1] + x[1][0] + x[1][1]) +# 每个值的梯度都为1 +x.grad + +# 注意：`grad`在反向传播过程中是累加的(accumulated)，**这意味着每一次运行反向传播，梯度都会累加之前的梯度，所以反向传播之前需把梯度清零。** + +y.backward() +x.grad + +# + {"scrolled": true} +y.backward() +x.grad +# - + +# 以下划线结束的函数是inplace操作，就像add_ +x.grad.data.zero_() + +y.backward() +x.grad + +# Variable和Tensor具有近乎一致的接口，在实际使用中可以无缝切换。 + +x = Variable(t.ones(4,5)) +y = t.cos(x) +x_tensor_cos = t.cos(x.data) +print(y) +x_tensor_cos + +# ### 神经网络 +# +# Autograd实现了反向传播功能，但是直接用来写深度学习的代码在很多情况下还是稍显复杂，torch.nn是专门为神经网络设计的模块化接口。nn构建于 Autograd之上，可用来定义和运行神经网络。nn.Module是nn中最重要的类，可把它看成是一个网络的封装，包含网络各层定义以及forward方法，调用forward(input)方法，可返回前向传播的结果。下面就以最早的卷积神经网络：LeNet为例，来看看如何用`nn.Module`实现。LeNet的网络结构如图2-7所示。 +# +# ![图2-7:LeNet网络结构](imgs/nn_lenet.png) +# +# 这是一个基础的前向传播(feed-forward)网络: 接收输入，经过层层传递运算，得到输出。 +# +# #### 定义网络 +# +# 定义网络时，需要继承`nn.Module`，并实现它的forward方法，把网络中具有可学习参数的层放在构造函数`__init__`中。如果某一层(如ReLU)不具有可学习的参数，则既可以放在构造函数中，也可以不放，但建议不放在其中，而在forward中使用`nn.functional`代替。 + +# + +import torch.nn as nn +import torch.nn.functional as F + +class Net(nn.Module): + def __init__(self): + # nn.Module子类的函数必须在构造函数中执行父类的构造函数 + # 下式等价于nn.Module.__init__(self) + super(Net, self).__init__() + + # 卷积层 '1'表示输入图片为单通道, '6'表示输出通道数，'5'表示卷积核为5*5 + self.conv1 = nn.Conv2d(1, 6, 5) + # 卷积层 + self.conv2 = nn.Conv2d(6, 16, 5) + # 仿射层/全连接层，y = Wx + b + self.fc1 = nn.Linear(16*5*5, 120) + self.fc2 = nn.Linear(120, 84) + self.fc3 = nn.Linear(84, 10) + + def forward(self, x): + # 卷积 -> 激活 -> 池化 + x = F.max_pool2d(F.relu(self.conv1(x)), (2, 2)) + x = F.max_pool2d(F.relu(self.conv2(x)), 2) + # reshape，‘-1’表示自适应 + x = x.view(x.size()[0], -1) + x = F.relu(self.fc1(x)) + x = F.relu(self.fc2(x)) + x = self.fc3(x) + return x + +net = Net() +print(net) +# - + +# 只要在nn.Module的子类中定义了forward函数，backward函数就会自动被实现(利用`Autograd`)。在`forward` 函数中可使用任何Variable支持的函数，还可以使用if、for循环、print、log等Python语法，写法和标准的Python写法一致。 +# +# 网络的可学习参数通过`net.parameters()`返回，`net.named_parameters`可同时返回可学习的参数及名称。 + +params = list(net.parameters()) +print(len(params)) + +for name,parameters in net.named_parameters(): + print(name,':',parameters.size()) + +# forward函数的输入和输出都是Variable，只有Variable才具有自动求导功能，而Tensor是没有的，所以在输入时，需把Tensor封装成Variable。 + +# + {"scrolled": true} +input = Variable(t.randn(1, 1, 32, 32)) +out = net(input) +out.size() +# - + +net.zero_grad() # 所有参数的梯度清零 +out.backward(Variable(t.ones(1,10))) # 反向传播 + +# 需要注意的是，torch.nn只支持mini-batches，不支持一次只输入一个样本，即一次必须是一个batch。但如果只想输入一个样本，则用 `input.unsqueeze(0)`将batch_size设为１。例如 `nn.Conv2d` 输入必须是4维的，形如$nSamples \times nChannels \times Height \times Width$。可将nSample设为1，即$1 \times nChannels \times Height \times Width$。 + +# #### 损失函数 +# +# nn实现了神经网络中大多数的损失函数，例如nn.MSELoss用来计算均方误差，nn.CrossEntropyLoss用来计算交叉熵损失。 + +# + {"scrolled": true} +output = net(input) +target = Variable(t.arange(0,10)) +criterion = nn.MSELoss() +loss = criterion(output, target) +loss +# - + +# 如果对loss进行反向传播溯源(使用`gradfn`属性)，可看到它的计算图如下： +# +# ``` +# input -> conv2d -> relu -> maxpool2d -> conv2d -> relu -> maxpool2d +# -> view -> linear -> relu -> linear -> relu -> linear +# -> MSELoss +# -> loss +# ``` +# +# 当调用`loss.backward()`时，该图会动态生成并自动微分，也即会自动计算图中参数(Parameter)的导数。 + +# 运行.backward，观察调用之前和调用之后的grad +net.zero_grad() # 把net中所有可学习参数的梯度清零 +print('反向传播之前 conv1.bias的梯度') +print(net.conv1.bias.grad) +loss.backward() +print('反向传播之后 conv1.bias的梯度') +print(net.conv1.bias.grad) + +# #### 优化器 + +# 在反向传播计算完所有参数的梯度后，还需要使用优化方法来更新网络的权重和参数，例如随机梯度下降法(SGD)的更新策略如下： +# ``` +# weight = weight - learning_rate * gradient +# ``` +# +# 手动实现如下： +# +# ```python +# learning_rate = 0.01 +# for f in net.parameters(): +# f.data.sub_(f.grad.data * learning_rate)# inplace 减法 +# ``` +# +# `torch.optim`中实现了深度学习中绝大多数的优化方法，例如RMSProp、Adam、SGD等，更便于使用，因此大多数时候并不需要手动写上述代码。 + +# + +import torch.optim as optim +#新建一个优化器，指定要调整的参数和学习率 +optimizer = optim.SGD(net.parameters(), lr = 0.01) + +# 在训练过程中 +# 先梯度清零(与net.zero_grad()效果一样) +optimizer.zero_grad() + +# 计算损失 +output = net(input) +loss = criterion(output, target) + +#反向传播 +loss.backward() + +#更新参数 +optimizer.step() +# - + +# +# +# #### 数据加载与预处理 +# +# 在深度学习中数据加载及预处理是非常复杂繁琐的，但PyTorch提供了一些可极大简化和加快数据处理流程的工具。同时，对于常用的数据集，PyTorch也提供了封装好的接口供用户快速调用，这些数据集主要保存在torchvison中。 +# +# `torchvision`实现了常用的图像数据加载功能，例如Imagenet、CIFAR10、MNIST等，以及常用的数据转换操作，这极大地方便了数据加载，并且代码具有可重用性。 +# +# +# ### 小试牛刀：CIFAR-10分类 +# +# 下面我们来尝试实现对CIFAR-10数据集的分类，步骤如下: +# +# 1. 使用torchvision加载并预处理CIFAR-10数据集 +# 2. 定义网络 +# 3. 定义损失函数和优化器 +# 4. 训练网络并更新网络参数 +# 5. 测试网络 +# +# #### CIFAR-10数据加载及预处理 +# +# CIFAR-10[^3]是一个常用的彩色图片数据集，它有10个类别: 'airplane', 'automobile', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck'。每张图片都是$3\times32\times32$，也即3-通道彩色图片，分辨率为$32\times32$。 +# +# [^3]: http://www.cs.toronto.edu/~kriz/cifar.html + +import torch as t +import torchvision as tv +import torchvision.transforms as transforms +from torchvision.transforms import ToPILImage +show = ToPILImage() # 可以把Tensor转成Image，方便可视化 + +# + +# 第一次运行程序torchvision会自动下载CIFAR-10数据集， +# 大约100M，需花费一定的时间， +# 如果已经下载有CIFAR-10，可通过root参数指定 + +# 定义对数据的预处理 +transform = transforms.Compose([ + transforms.ToTensor(), # 转为Tensor + transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5)), # 归一化 + ]) + +# 训练集 +trainset = tv.datasets.CIFAR10( + root='../data/', + train=True, + download=True, + transform=transform) + +trainloader = t.utils.data.DataLoader( + trainset, + batch_size=4, + shuffle=True, + num_workers=2) + +# 测试集 +testset = tv.datasets.CIFAR10( + '../data/', + train=False, + download=True, + transform=transform) + +testloader = t.utils.data.DataLoader( + testset, + batch_size=4, + shuffle=False, + num_workers=2) + +classes = ('plane', 'car', 'bird', 'cat', + 'deer', 'dog', 'frog', 'horse', 'ship', 'truck') +# - + +# Dataset对象是一个数据集，可以按下标访问，返回形如(data, label)的数据。 + +# + +(data, label) = trainset[100] +print(classes[label]) + +# (data + 1) / 2是为了还原被归一化的数据 +show((data + 1) / 2).resize((100, 100)) +# - + +# Dataloader是一个可迭代的对象，它将dataset返回的每一条数据拼接成一个batch，并提供多线程加速优化和数据打乱等操作。当程序对dataset的所有数据遍历完一遍之后，相应的对Dataloader也完成了一次迭代。 + +dataiter = iter(trainloader) +images, labels = dataiter.next() # 返回4张图片及标签 +print(' '.join('%11s'%classes[labels[j]] for j in range(4))) +show(tv.utils.make_grid((images+1)/2)).resize((400,100)) + +# #### 定义网络 +# +# 拷贝上面的LeNet网络，修改self.conv1第一个参数为3通道，因CIFAR-10是3通道彩图。 + +# + +import torch.nn as nn +import torch.nn.functional as F + +class Net(nn.Module): + def __init__(self): + super(Net, self).__init__() + self.conv1 = nn.Conv2d(3, 6, 5) + self.conv2 = nn.Conv2d(6, 16, 5) + self.fc1 = nn.Linear(16*5*5, 120) + self.fc2 = nn.Linear(120, 84) + self.fc3 = nn.Linear(84, 10) + + def forward(self, x): + x = F.max_pool2d(F.relu(self.conv1(x)), (2, 2)) + x = F.max_pool2d(F.relu(self.conv2(x)), 2) + x = x.view(x.size()[0], -1) + x = F.relu(self.fc1(x)) + x = F.relu(self.fc2(x)) + x = self.fc3(x) + return x + + +net = Net() +print(net) +# - + +# #### 定义损失函数和优化器(loss和optimizer) + +from torch import optim +criterion = nn.CrossEntropyLoss() # 交叉熵损失函数 +optimizer = optim.SGD(net.parameters(), lr=0.001, momentum=0.9) + +# ### 训练网络 +# +# 所有网络的训练流程都是类似的，不断地执行如下流程： +# +# - 输入数据 +# - 前向传播+反向传播 +# - 更新参数 +# + +# + +from torch.autograd import Variable + +t.set_num_threads(8) +for epoch in range(2): + + running_loss = 0.0 + for i, data in enumerate(trainloader, 0): + + # 输入数据 + inputs, labels = data + inputs, labels = Variable(inputs), Variable(labels) + + # 梯度清零 + optimizer.zero_grad() + + # forward + backward + outputs = net(inputs) + loss = criterion(outputs, labels) + loss.backward() + + # 更新参数 + optimizer.step() + + # 打印log信息 + running_loss += loss.data[0] + if i % 2000 == 1999: # 每2000个batch打印一下训练状态 + print('[%d, %5d] loss: %.3f' \ + % (epoch+1, i+1, running_loss / 2000)) + running_loss = 0.0 +print('Finished Training') +# - + +# 此处仅训练了2个epoch（遍历完一遍数据集称为一个epoch），来看看网络有没有效果。将测试图片输入到网络中，计算它的label，然后与实际的label进行比较。 + +dataiter = iter(testloader) +images, labels = dataiter.next() # 一个batch返回4张图片 +print('实际的label: ', ' '.join(\ + '%08s'%classes[labels[j]] for j in range(4))) +show(tv.utils.make_grid(images / 2 - 0.5)).resize((400,100)) + + +# 接着计算网络预测的label： + +# + +# 计算图片在每个类别上的分数 +outputs = net(Variable(images)) +# 得分最高的那个类 +_, predicted = t.max(outputs.data, 1) + +print('预测结果: ', ' '.join('%5s'\ + % classes[predicted[j]] for j in range(4))) +# - + +# 已经可以看出效果，准确率50%，但这只是一部分的图片，再来看看在整个测试集上的效果。 + +# + +correct = 0 # 预测正确的图片数 +total = 0 # 总共的图片数 +for data in testloader: + images, labels = data + outputs = net(Variable(images)) + _, predicted = t.max(outputs.data, 1) + total += labels.size(0) + correct += (predicted == labels).sum() + +print('10000张测试集中的准确率为: %d %%' % (100 * correct / total)) +# - + +# 训练的准确率远比随机猜测(准确率10%)好，证明网络确实学到了东西。 + +# #### 在GPU训练 +# 就像之前把Tensor从CPU转到GPU一样，模型也可以类似地从CPU转到GPU。 + +if t.cuda.is_available(): + net.cuda() + images = images.cuda() + labels = labels.cuda() + output = net(Variable(images)) + loss= criterion(output,Variable(labels)) + +# 如果发现在GPU上并没有比CPU提速很多，实际上是因为网络比较小，GPU没有完全发挥自己的真正实力。 + +# 对PyTorch的基础介绍至此结束。总结一下，本节主要包含以下内容。 +# +# 1. Tensor: 类似Numpy数组的数据结构，与Numpy接口类似，可方便地互相转换。 +# 2. autograd/Variable: Variable封装了Tensor，并提供自动求导功能。 +# 3. nn: 专门为神经网络设计的接口，提供了很多有用的功能(神经网络层，损失函数，优化器等)。 +# 4. 神经网络训练: 以CIFAR-10分类为例演示了神经网络的训练流程，包括数据加载、网络搭建、训练及测试。 +# +# 通过本节的学习，相信读者可以体会出PyTorch具有接口简单、使用灵活等特点。从下一章开始，本书将深入系统地讲解PyTorch的各部分知识。 diff --git a/data/cifar-10-batches-py/batches.meta b/data/cifar-10-batches-py/batches.meta new file mode 100644 index 0000000..4467a6e Binary files /dev/null and b/data/cifar-10-batches-py/batches.meta differ diff --git a/data/cifar-10-batches-py/readme.html b/data/cifar-10-batches-py/readme.html new file mode 100644 index 0000000..e377ade --- /dev/null +++ b/data/cifar-10-batches-py/readme.html @@ -0,0 +1 @@ + diff --git a/data/cifar-10-batches-py/test_batch b/data/cifar-10-batches-py/test_batch new file mode 100644 index 0000000..3e03f1f Binary files /dev/null and b/data/cifar-10-batches-py/test_batch differ diff --git a/data/mnist/processed/test.pt b/data/mnist/processed/test.pt new file mode 100644 index 0000000..aa3d55f Binary files /dev/null and b/data/mnist/processed/test.pt differ diff --git a/data/mnist/raw/t10k-images-idx3-ubyte b/data/mnist/raw/t10k-images-idx3-ubyte new file mode 100644 index 0000000..1170b2c Binary files /dev/null and b/data/mnist/raw/t10k-images-idx3-ubyte differ diff --git a/data/mnist/raw/t10k-labels-idx1-ubyte b/data/mnist/raw/t10k-labels-idx1-ubyte new file mode 100644 index 0000000..d1c3a97 Binary files /dev/null and b/data/mnist/raw/t10k-labels-idx1-ubyte differ diff --git a/data/mnist/raw/train-images-idx3-ubyte b/data/mnist/raw/train-images-idx3-ubyte new file mode 100644 index 0000000..bbce276 Binary files /dev/null and b/data/mnist/raw/train-images-idx3-ubyte differ diff --git a/data/mnist/raw/train-labels-idx1-ubyte b/data/mnist/raw/train-labels-idx1-ubyte new file mode 100644 index 0000000..d6b4c5d Binary files /dev/null and b/data/mnist/raw/train-labels-idx1-ubyte differ diff --git a/demo_code/Linear_Regression.py b/demo_code/Linear_Regression.py new file mode 100644 index 0000000..e14ede5 --- /dev/null +++ b/demo_code/Linear_Regression.py @@ -0,0 +1,64 @@ + +import torch as t +from torch import nn, optim +from torch.autograd import Variable +import numpy as np +import matplotlib.pyplot as plt + +# create numpy data +x_train = np.linspace(0, 10, 100) +y_train = 10*x_train + 4.5 + +# convert to tensor (need to change nx1, float32 dtype) +x_train = t.from_numpy(x_train.reshape(-1, 1).astype("float32")) +y_train = t.from_numpy(y_train.reshape(-1, 1).astype("float32")) + + +# Linear Regression Model +class LinearRegression(nn.Module): + def __init__(self): + super(LinearRegression, self).__init__() + self.linear = nn.Linear(1, 1) # input and output is 1 dimension + + def forward(self, x): + out = self.linear(x) + return out + +# create the model +model = LinearRegression() + +# 定义loss和优化函数 +criterion = nn.MSELoss() +optimizer = optim.SGD(model.parameters(), lr=1e-4) + +# 开始训练 +num_epochs = 1000 +for epoch in range(num_epochs): + inputs = Variable(x_train) + target = Variable(y_train) + + # forward + out = model(inputs) + loss = criterion(out, target) + + # backward + optimizer.zero_grad() + loss.backward() + optimizer.step() + + if (epoch+1) % 20 == 0: + print('Epoch[{}/{}], loss: {:.6f}' + .format(epoch+1, num_epochs, loss.data[0])) + +model.eval() +predict = model(Variable(x_train)) +predict = predict.data.numpy() +plt.plot(x_train.numpy(), y_train.numpy(), 'ro', label='Original data') +plt.plot(x_train.numpy(), predict, label='Fitting Line') + +# 显示图例 +plt.legend() +plt.show() + +# 保存模型 +t.save(model.state_dict(), './model_LinearRegression.pth') diff --git a/demo_code/Logistic_Regression.py b/demo_code/Logistic_Regression.py new file mode 100644 index 0000000..21334a7 --- /dev/null +++ b/demo_code/Logistic_Regression.py @@ -0,0 +1,115 @@ + +import torch as t +from torch import nn, optim +import torch.nn.functional as F +from torch.autograd import Variable +from torch.utils.data import DataLoader +from torchvision import transforms +from torchvision import datasets +import time + +# 定义超参数 +batch_size = 32 +learning_rate = 1e-3 +num_epoches = 100 + +# 下载训练集 MNIST 手写数字训练集 +dataset_path = "../data/mnist" + +train_dataset = datasets.MNIST( + root=dataset_path, train=True, transform=transforms.ToTensor(), download=True) + +test_dataset = datasets.MNIST( + root=dataset_path, train=False, transform=transforms.ToTensor()) + +train_loader = DataLoader(train_dataset, batch_size=batch_size, shuffle=True) +test_loader = DataLoader(test_dataset, batch_size=batch_size, shuffle=False) + + +# 定义 Logistic Regression 模型 +class Logstic_Regression(nn.Module): + def __init__(self, in_dim, n_class): + super(Logstic_Regression, self).__init__() + self.logstic = nn.Linear(in_dim, n_class) + + def forward(self, x): + out = self.logstic(x) + return out + + +model = Logstic_Regression(28 * 28, 10) # 图片大小是28x28 +use_gpu = t.cuda.is_available() # 判断是否有GPU加速 +if use_gpu: + model = model.cuda() + +# 定义loss和optimizer +criterion = nn.CrossEntropyLoss() +optimizer = optim.SGD(model.parameters(), lr=learning_rate) + +# 开始训练 +for epoch in range(num_epoches): + print('*' * 10) + print('epoch {}'.format(epoch + 1)) + + since = time.time() + running_loss = 0.0 + running_acc = 0.0 + for i, data in enumerate(train_loader, 1): + img, label = data + img = img.view(img.size(0), -1) # 将图片展开成 28x28 + if use_gpu: + img = Variable(img).cuda() + label = Variable(label).cuda() + else: + img = Variable(img) + label = Variable(label) + + # 向前传播 + out = model(img) + loss = criterion(out, label) + running_loss += loss.data[0] * label.size(0) + _, pred = t.max(out, 1) + num_correct = (pred == label).sum() + running_acc += num_correct.data[0] + + # 向后传播 + optimizer.zero_grad() + loss.backward() + optimizer.step() + + if i % 300 == 0: + print('[{}/{}] Loss: {:.6f}, Acc: {:.6f}'.format( + epoch + 1, num_epoches, running_loss / (batch_size * i), + running_acc / (batch_size * i))) + + print('Finish {} epoch, Loss: {:.6f}, Acc: {:.6f}'.format( + epoch + 1, running_loss / (len(train_dataset)), running_acc / (len( + train_dataset)))) + + model.eval() + eval_loss = 0. + eval_acc = 0. + + for data in test_loader: + img, label = data + img = img.view(img.size(0), -1) + if use_gpu: + img = Variable(img, volatile=True).cuda() + label = Variable(label, volatile=True).cuda() + else: + img = Variable(img, volatile=True) + label = Variable(label, volatile=True) + out = model(img) + loss = criterion(out, label) + eval_loss += loss.data[0] * label.size(0) + _, pred = t.max(out, 1) + num_correct = (pred == label).sum() + eval_acc += num_correct.data[0] + + print('Test Loss: {:.6f}, Acc: {:.6f}'.format(eval_loss / (len( + test_dataset)), eval_acc / (len(test_dataset)))) + print('Time:{:.1f} s'.format(time.time() - since)) + print() + +# 保存模型 +t.save(model.state_dict(), './model_LogsticRegression.pth') diff --git a/demo_code/Neural_Network.py b/demo_code/Neural_Network.py new file mode 100644 index 0000000..fcf95b4 --- /dev/null +++ b/demo_code/Neural_Network.py @@ -0,0 +1,104 @@ +import torch +from torch import nn, optim + +from torch.autograd import Variable +from torch.utils.data import DataLoader +from torchvision import transforms +from torchvision import datasets + +batch_size = 32 +learning_rate = 1e-2 +num_epoches = 50 + +# 下载训练集 MNIST 手写数字训练集 +dataset_path = "../data/mnist" + +train_dataset = datasets.MNIST( + root=dataset_path, train=True, transform=transforms.ToTensor(), download=True) + +test_dataset = datasets.MNIST( + root=dataset_path, train=False, transform=transforms.ToTensor()) + +train_loader = DataLoader(train_dataset, batch_size=batch_size, shuffle=True) +test_loader = DataLoader(test_dataset, batch_size=batch_size, shuffle=False) + + +# 定义简单的前馈神经网络 +class Neuralnetwork(nn.Module): + def __init__(self, in_dim, n_hidden_1, n_hidden_2, out_dim): + super(Neuralnetwork, self).__init__() + self.layer1 = nn.Linear(in_dim, n_hidden_1) + self.layer2 = nn.Linear(n_hidden_1, n_hidden_2) + self.layer3 = nn.Linear(n_hidden_2, out_dim) + + def forward(self, x): + x = self.layer1(x) + x = self.layer2(x) + x = self.layer3(x) + return x + + +model = Neuralnetwork(28 * 28, 300, 100, 10) +if torch.cuda.is_available(): + model = model.cuda() + +criterion = nn.CrossEntropyLoss() +optimizer = optim.SGD(model.parameters(), lr=learning_rate) + +for epoch in range(num_epoches): + print('epoch {}'.format(epoch + 1)) + print('*' * 10) + running_loss = 0.0 + running_acc = 0.0 + for i, data in enumerate(train_loader, 1): + img, label = data + img = img.view(img.size(0), -1) + if torch.cuda.is_available(): + img = Variable(img).cuda() + label = Variable(label).cuda() + else: + img = Variable(img) + label = Variable(label) + # 向前传播 + out = model(img) + loss = criterion(out, label) + running_loss += loss.data[0] * label.size(0) + _, pred = torch.max(out, 1) + num_correct = (pred == label).sum() + running_acc += num_correct.data[0] + # 向后传播 + optimizer.zero_grad() + loss.backward() + optimizer.step() + + if i % 300 == 0: + print('[{}/{}] Loss: {:.6f}, Acc: {:.6f}'.format( + epoch + 1, num_epoches, running_loss / (batch_size * i), + running_acc / (batch_size * i))) + print('Finish {} epoch, Loss: {:.6f}, Acc: {:.6f}'.format( + epoch + 1, running_loss / (len(train_dataset)), running_acc / (len( + train_dataset)))) + model.eval() + eval_loss = 0. + eval_acc = 0. + for data in test_loader: + img, label = data + img = img.view(img.size(0), -1) + if torch.cuda.is_available(): + img = Variable(img, volatile=True).cuda() + label = Variable(label, volatile=True).cuda() + else: + img = Variable(img, volatile=True) + label = Variable(label, volatile=True) + out = model(img) + loss = criterion(out, label) + eval_loss += loss.data[0] * label.size(0) + _, pred = torch.max(out, 1) + num_correct = (pred == label).sum() + eval_acc += num_correct.data[0] + print('Test Loss: {:.6f}, Acc: {:.6f}'.format(eval_loss / (len( + test_dataset)), eval_acc / (len(test_dataset)))) + print() + +# 保存模型 +torch.save(model.state_dict(), './neural_network.pth') diff --git a/demo_code/model_LinearRegression.pth b/demo_code/model_LinearRegression.pth new file mode 100644 index 0000000..8fce4cc Binary files /dev/null and b/demo_code/model_LinearRegression.pth differ diff --git a/demo_code/model_LogsticRegression.pth b/demo_code/model_LogsticRegression.pth new file mode 100644 index 0000000..18749fb Binary files /dev/null and b/demo_code/model_LogsticRegression.pth differ diff --git a/references/SciPy.ipynb b/references/SciPy.ipynb index 876190a..2567d51 100644 --- a/references/SciPy.ipynb +++ b/references/SciPy.ipynb @@ -85,410 +85,7 @@ "outputs": [ { "data": { - "image/png": [ - "iVBORw0KGgoAAAANSUhEUgAAAX0AAAEACAYAAABfxaZOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\n", - "AAALEgAACxIB0t1+/AAAIABJREFUeJzsnXl8U1X6/z8nS9ukS5rue4GyyCKLrIJIAVnc+CGIIFtR\n", - "x8EZF2ZcvoMiWJFRZ9QZBxQHVxQGGVBUFhdcKKAg4ChFsOy0BVq6L+maND2/P24TkrZpcrfkpL3v\n", - "1yuQ5J57zunJvc957nOe8zyEUgoFBQUFha6BytcdUFBQUFDwHorQV1BQUOhCKEJfQUFBoQuhCH0F\n", - "BQWFLoQi9BUUFBS6EIrQV1BQUOhCiBb6hJB3CSFFhJBfXRxPJ4RUEUJ+aXk9LbZNBQUFBQVhaCSo\n", - "4z0AawB80EGZvZTSaRK0paCgoKAgAtGaPqV0P4AKN8WI2HYUFBQUFMTjDZs+BTCaEJJNCPmcENLP\n", - "C20qKCgoKLSDFOYdd/wMIJlSWkcIuRnApwB6e6FdBQUFBYVWyC70KaUmh/dfEELWEkIiKKXljuUI\n", - "IUoQIAUFBQUBUEo9NqHLbt4hhMQSQkjL+xEASGuBb4NSqrwoxTPPPOPzPrDyUsZCGQtlLDp+8UW0\n", - "pk8I+RDAOABRhJCLAJ4BoG0R4usA3AngD4SQJgB1AOaIbVNBQUFBQRiihT6l9G43x18H8LrYdhQU\n", - "FBQUxKPsyGWQ9PR0X3eBGZSxuIoyFldRxkI4RIhNSA4IIZSVvigoKCj4C4QQUB4Lud5w2VRQUGiH\n", - "Fv8GBQWPkUIxVoS+goIPUZ5uFTxFKiVBsekrKCgodCEUoa+goKDQhVCEvoKCgkIXQhH6CgoKsrF+\n", - "/XqMHTvW191wS2ZmJhYsWODrbngFRegrKCh0ebqSJ5Ui9BUUFBS6EIrQV1BQcEKlUuH8+fP2z4sW\n", - "LcLy5csBAFlZWUhKSsI//vEPxMbGIiEhAevXr7eXLSsrw7Rp02AwGDBy5EicO3fOqe4lS5YgJSUF\n", - "BoMBw4YNw/fff28/lpmZiVmzZmHBggUICwvDwIEDcebMGbzwwguIjY1Famoqvv76a3v59PR0PPnk\n", - "kxg5ciQMBgOmT5+Oioqr+Zx+/PFHjB49GkajEYMHD8bevXvtxy5cuIBx48YhLCwMkydPRmlpqWTj\n", - "xzqK0FdQYJCmpibk5uaKfjU1NYnuCyHEyfxRVFSE6upqFBQU4J133sGDDz6IqqoqAMCDDz4IvV6P\n", - "K1eu4N1338V7773ndO6IESOQnZ2NiooKzJ07F7NmzYLZbLYf37lzJxYuXIiKigoMGTIEkyZNAgAU\n", - "FBRg+fLlWLx4sVPfNmzYgPfeew+FhYXQaDR45JFHAACXL1/GbbfdhhUrVqCiogIvv/wyZs6cibKy\n", - "MgDA3LlzMXz4cJSVlWH58uV4//33u4yJRxH6CgoKbnHcRKbVarFixQqo1WrcfPPNCAkJwalTp2C1\n", - "WrFt2zasXLkSOp0O/fv3R0ZGhtO58+bNg9FohEqlwqOPPorGxkacOnXKfvzGG2/EpEmToFarceed\n", - "d6KsrAxLly6FWq3G7NmzkZubi+rqagDcZLRw4UL069cPer0ezz33HLZs2YLm5mZs3LgRt9xyC6ZO\n", - "nQoAuOmmmzBs2DDs2rUL+fn5+Omnn/Dcc89Bq9Vi7NixuP3227vMRjllR66CAoNoNBp069bN191o\n", - "l8jISKhUV/VFvV6PmpoalJSUoKmpCcnJyfZjKSkpTue+/PLLePfdd1FQUABCCKqrq51MKzExMfb3\n", - "Op0OUVFRdg1cp9MBAGpqahAWFgYAbdqyWCwoLS1FXl4etm7dih07dtiPNzU1YcKECSgoKIDRaLTX\n", - "BwCpqam4ePGiqHHxFxShr6Cg4IRer0ddXZ39c2FhoZNwdUV0dDQ0Gg3y8/PRp08fAEB+fr79+P79\n", - "+/HSSy/hu+++Q//+/QEAERERojRsx/rz8/Oh1WoRHR2NlJQULFiwAG+++Wabc/Ly8lBRUYG6ujro\n", - "9Xr7d2q1WnA//AnFvKOgoODE4MGD8Z///AdWqxVffvkl9u3b59F5arUaM2bMQGZmJurr6/Hbb785\n", - "2cpNJhM0Gg2ioqJgNpuxcuVKu6lGCJRSbNy4ETk5Oairq8OKFSswa9YsEEIwf/587NixA7t374bV\n", - "akVDQwOysrJw+fJlpKamYtiwYXjmmWdgsVjw/fffY+fOnYL74W8oQl9BQcGJf/3rX9ixYweMRiM2\n", - "bdqEO+64w+l4Rwuer732GmpqahAXF4d7770X9957r/3Y1KlTMXXqVPTu3RvdunWDTqdzMv+0XjBu\n", - "ry3Hz4QQLFiwAIsWLUJ8fDzMZjNWr14NAEhKSsJnn32G559/HjExMUhJScErr7yC5uZmAMCmTZtw\n", - "6NAhREREYOXKlcjIyOA5Sv6LEk9fQcFHtMRB93U3/Jbx48djwYIFThNLZ8bV9cI3nr6i6SsoKPgt\n", - "yqTJH0XoKygo+C1dxbdeShTzjoKCj1DMOwp8UMw7CgoKCgq8UYS+goKCQhdCEfoKCgoKXQhF6Cso\n", - "KCh0IRShr6CgoNCFUIS+goKCrMidMtFfUh065iXwJYrQV1BQ8Gv8xVe/vTATvkAR+goKCsxii5XT\n", - "WWBhX4Yi9BUUFNogZ8rEkydPYtKkSYiMjMQ111yDrVu3OrXzhz/8AbfccgtCQkKQlZXVpm/uUh3O\n", - "mjUL8fHxCA8Px7hx4/Dbb7851f/HP/4Rt9xyC0JDQzF27FhcuXIFS5YsgdFoRN++fXH06FF7+W7d\n", - "uuHFF19E//79ERERgXvvvReNjY324zt37sTgwYNhNBoxZswY/Prrr/Zjv/zyC6677jqEhYVhzpw5\n", - "aGho8HD05UUR+goKCm6RKmVibW0tJk2ahPnz56OkpASbN2/GH//4R+Tk5Njr/vDDD7F8+XLU1NRg\n", - "zJgxbfriLtXhrbfeirNnz6KkpATXXXcd5s2b53T+1q1b8de//hWlpaUICAjAqFGjMHz4cJSXl+PO\n", - "O+/Eo48+6lR+06ZN2L17N86dO4fTp09j1apVADihft999+Gtt95CeXk5Fi9ejGnTpsFiscBsNmP6\n", - "9OnIyMhARUUFZs2ahY8//pgJ8w4opUy8uK4oKHQd3F7zgDQvARBC6Llz5+yfFy1aRJ9++mlKKaV7\n", - "9uyhOp2OWq1W+/GYmBh66NAh2tTURLVaLT116pT92FNPPUVvuOEGSimlmzdvpmPHjnVq6/e//z19\n", - "9tlnKaWUZmRk0IyMDJf9ysvLoxqNhtbV1dm/mzt3Lp0/f3675SsqKighhFZXV9v/jt///vf242vW\n", - "rKH9+vWzfz527BgNDw+3f+7WrRtdt26d/fPnn39O09LSKKWUPvDAA3T58uVO7fXp04fu3buX7t27\n", - "lyYkJDgdGz16dJvyfHB1vbR877GsVTJnKSiwCgP2X1cITZmYl5eHQ4cOwWg02r9ramrCwoULAXBP\n", - "FElJSS7bdZfq0Gq1YtmyZfjoo49QUlJi72NpaSlCQ0MBOKdkDAoKapOisaamxqnN1n9LQUGB/W/5\n", - "4IMPsGbNGvtxi8WCwsJCUEqRmJjoVE9qaqpi01dQUGCT9lImemKacEyZaMPxfUpKCsaNG4eKigr7\n", - "y2Qy4fXXX/eoX/Hx8fZUhzby8vLsfdu0aRO2b9+Ob7/9FlVVVbhw4QIAcQuorf8WmzBPSUnBsmXL\n", - "nP6WmpoazJ49G/Hx8bh8+bJTPY799CWK0FdQUGiDXCkTb731Vpw+fRobN26ExWKBxWLBkSNHcPLk\n", - "SQDuhbO7VIc1NTUIDAxEREQEamtr8dRTTzmdz1f4U0qxdu1aXL58GeXl5fjrX/+K2bNnAwDuv/9+\n", - "/Pvf/8bhw4dBKUVtbS127dqFmpoajB49GhqNBqtXr4bFYsG2bdtw5MgRXm3LhSL0FRQU2iBXysTQ\n", - "0FDs3r0bmzdvRmJiIuLj4/Hkk0/CbDbb63WnDXeU6nDhwoVITU1FYmIiBgwYgOuvv75NisWOPrf+\n", - "2wghmDt3LiZPnoy0tDT06tULTz/9NABg6NCheOutt/DQQw8hIiICvXr1wgcffAAA0Gq12LZtG9av\n", - "X4/IyEhs2bIFM2fO7PDv8hai4+kTQt4FcCuAYkrptS7KrAZwM4A6AIsopb+0U4ayYO9SUPAWSjx9\n", - "9unevTveeecdTJgwwdddYSqe/nsApro6SAi5BUBPSmkvAL8H8IYEbSooKCgoCEC00KeU7gdQ0UGR\n", - "aQDebyl7CEA4ISRWbLsKCgoKCvzxhstmIoCLDp8vAUgCUOSFthUUFBQEY/P+6Ux4y0+/tb2pXUPm\n", - "suXLoFVrAQDp6elIT0+XuVtXOXToEC5fvow77rjDZ25VZrMZn3zyCfr3748BAwb4pA8AcODAAZSW\n", - "lmLatGk+60NDQwM+/fRTDBkyBH369PFZP3744QdUVlbi1ltv9VkfFBQcycrKQlZWFmpqauy7oPng\n", - "De+dywCSHT4ntXzXhuLkCmRmZiIzM9OrAh8ATCYTxo4di+zsbK+268jBgwdxxx132Dd/+AKb69l1\n", - "113ntDXe2/z444+YOXOmfdONL6CUor6+Hr1790Zubq7P+qGg4Eh6ejoyMzMxdepUvPXWW7zP94bQ\n", - "3w5gIQAQQkYBqKSUtmva2ZS3GdlXvC90CwoKkJiYiOjoaJSXl3u9fRtWqxUBAQHQ6XR2FzZvc+HC\n", - "BfTs2RNJSUkoLCz0SR8Abiy0Wi3UarXPIi0ePXoUQ4YMQa9evdoEDVNQ8DWOO6J5nSe2YULIhwAO\n", - "AOhDCLlICLmXELKYELIYACilnwM4Twg5C2AdgD+6qitiz2O4f8f9sDZbxXaLFydPnsQ111zj1TY7\n", - "YuDAgU7R+rzJhQsX0L17d5+0bYNSajexDRw4EMeOHfNJPyorKxEZGemTthUUOqK2thbBwcGCzpXC\n", - "e+duSmkCpTSAUppMKX2XUrqOUrrOocxDlNKelNJBlNKfXdVVefBhBJJgrDm8xlUR2bAJGbVajaam\n", - "Jq+3f+XKFcTHxwMADAaDIFudFLRez/CFH/mlS5fs8VoiIyNRUdGRc5h8OP7tQrUqBQU5yM7OxuDB\n", - "gwWdy9SVPFL9E+arXsGqfauQW5nrtXYdBV2PHj18smJ/5swZ9OrVq90+eRPHdpOTk3Hp0iWv9+Hs\n", - "2bNIS0vzersdERoaiurqal93g0n+85//YMqUKS6P79+/X5In6W7duuHbb7/1qKzcKRpdkZ+fj9DQ\n", - "UNmVJbPZjMDAQEHnMiX0B0acR+FeIx4f/Tge2PmAT7TMpKQknwg6q9UKjca3QU9b2867d+/uM5c1\n", - "XwemMpvNCAgIsH/u16+fTxe2WWbevHn46quv7J9bJ2AZO3asPbaOGFhJN9gRKSkpMJlMTPeTKaHf\n", - "v1sJjvxE8dj1j+FKzRVs+nWT7G2WlJQgKirK/pmlH8vbk15eXh66detm/6zRaGC1end9hRXOnz+P\n", - "Hj162D+3jjqpwOHq+lDCS7ALU0J/xLhg/HQ+AhqVFm/d/hYe2/0YSmpLZG3z9OnTTmYVX9H6JomP\n", - "j8eVK1e82of8/HykpqZ6tU1PCAsL87pppbCw0L7G0hW5ePEiZsyYgZiYGERFReHhhx8GwJlNxowZ\n", - "g0cffRRRUVHIzMx0MqXceOONAIBBgwYhNDQUW7duRVZWllNMeld1nzt3DhMmTEBUVBSio6Mxf/58\n", - "j9e23KVoXLJkCVJSUmAwGDBs2DB8//339mOZmZmYNWsWFixYgLCwMAwcOBBnzpzBCy+8gNjYWKSm\n", - "puLrr7+2l09PT8eTTz6JkSNHwmAwYPr06fZ1p9zcXKhUKvtTc3p6OlasWIEbbrgBYWFhmDJlCsrK\n", - "yux1ffDBB0hNTUVUVBRWrVrlsQlLzKTKlNDvNy4NtMmKy5eB4YnDMe/aefjzV3+WtU2z2YygoCBZ\n", - "2/CE1k8YPXr08LqbIKW0zYKltzW29trr0aOHk7nAW/1g6anPm1itVtx2223o3r078vLycPnyZdx9\n", - "993244cPH0ZaWhqKi4uxbNkyp3NtIZiPHTsGk8mEWbNmua17zpw59uPLli1DYWEhcnJycPHiRWRm\n", - "ZnrU545SNALAiBEjkJ2djYqKCsydOxezZs1ycoveuXMnFi5ciIqKCgwZMgSTJk0CwLlzL1++HIsX\n", - "L3Zqb8OGDXjvvfdQWFgIjUaDRx55xGXfPvzwQ6xfvx7FxcUwm814+eWXAQC//fYbHnzwQXz44Yco\n", - "LCxEVVUVCgoK3F53NTU1CAkJ8Whc2oMpoa/umYbB6mP4pSUG58rxK/HjpR/x6clPfdsxmWlqaoJa\n", - "rXb6LiAgABaLxUc9uoq3BV9lZaVTViUAMBqNqKys9Go/WIAQaV58OXz4MAoLC/HSSy9Bp9MhMDAQ\n", - "o0ePth9PSEjAgw8+CJVKxVthaq9uWx7ctLQ0TJw4EVqtFlFRUfjzn/+MvXv3uq3TarVi27ZtWLly\n", - "JXQ6Hfr374+MjAwnBWLevHkwGo1QqVR49NFH0djYiFOnTtmP33jjjZg0aRLUajXuvPNOlJWVYenS\n", - "pVCr1Zg9ezZyc3PtT5uEECxcuBD9+vWDXq/Hc889hy1btriMgHnPPfegZ8+eCAoKwl133WVPvP7R\n", - "Rx9h2rRpGD16NLRaLVauXOnR/Zafn++UjYwvTAl9dOuGgZafceJX7tEoOCAY709/H3/Y9QcU1xZ7\n", - "rRtardarAregoAAJCQlea48v3tT2L1686GQK8BUsaPlSJcnly8WLF5GamurSTVXM79NR3UVFRZgz\n", - "Zw6SkpJgMBiwYMECJ1OIK9ylaASAl19+Gf369UN4eDiMRiOqqqpQWlpqP946ZWJUVJT9GrClZnRM\n", - "o9i6LYvF4lSfI3FxcU512+opKChwSg2p0+k82hdSXFyM6Ohot+VcwZbQDwrCNfoLOP5Tvf2rMSlj\n", - "sGjQIvx+x++9Jny87cHDiqBrj5iYGBQXe2/CLS8vR0REhNfa40N4eLjP9gx4k+TkZOTn57tcpBUz\n", - "IXZU91NPPQW1Wo3jx4+jqqoKGzZs8Gg3trsUjfv378dLL72ErVu3orKyEhUVFTAYDJKmULQ9nfAh\n", - "ISHBSc7U19d7NMm1Z4blA1tCH0Cv2BIcP+b8Y2SmZ+JC5QV8kP2BV/qQmJjoVaHf2j2QJVJTU5GX\n", - "l+fVNlnQstsjKSmpTd7TzsjIkSMRHx+PpUuXoq6uDg0NDThw4IDH58fGxrpcjxoxYoTLumtqahAc\n", - "HIywsDBcvnwZL730kkftuUvRaDKZoNFoEBUVBbPZjJUrV4pyDKCUYuPGjcjJyUFdXR1WrFiBWbNm\n", - "ubxuXU0uM2fOxI4dO3Dw4EGYzWZkZmZ6RbFlTugnp9bhVH4QHDfFBmoCseGODXji6yeQVymdAHI1\n", - "wFqt1ie7cn1Je+sKABASEoLa2lqv9YMFV7/6+vp2bdWRkZEuH+E7EyqVCjt27MDZs2eRkpKC5ORk\n", - "bNmyBYDr9IKO32VmZiIjIwNGoxEfffSR03G1Wu2y7meeeQY///wzDAYDbr/9dsycOdNjBaCjFI1T\n", - "p07F1KlT0bt3b3Tr1g06nc7J/OMuZWLrz4QQLFiwAIsWLUJ8fDzMZjNWr17t8bm2z/3798eaNWsw\n", - "Z84cJCQkIDQ0FDExMYI3XXkMpZSJF9cVSi9kZNA0YynNyaFteHH/i3Tsu2OpxWppe1AApaWlNDs7\n", - "u91j3333nSRteIKrtrzZh9zcXHru3Dmf98NVW3v27PFaH86cOUPz8/PbPSblWNiueQX/Ij09nb7z\n", - "zjuS12symahGo6G5ubntHrddL62vwZbvPZa1zGn62h490Fd/HsePtz32xJgnEKgJxHN7n5OkLbGr\n", - "4HITEhICk8nklbYuXbrktKjkCAvmFpVK5bWNYgUFBV3aR1/BPVSiJ9IdO3agrq4OtbW1ePzxxzFw\n", - "4EDZ98owJ/TD+vbFNTjertBXERU23LEBb/38FvZc2CO6rcrKShgMBtH1yIU3N2hZLBaX6wpSXeCe\n", - "4GqC8eaCMgshMRTYRipFaPv27UhMTERiYiLOnTuHzZs3d1ieSrB/hDmhH9KnD/o3/tyu0AeAuJA4\n", - "rJ++Hgs+WSDJbl0WtFhXxMbGoqio62SVpFdNfW1ITEz0aXIZBQUbe/bscVozEMNbb72FiooKVFZW\n", - "4uuvv3YbHaCioqLNPha+MCf0SWIiBtUeQkfh5CenTcb8gfOR8WkGmqlvEmxIhdVqbXcBFfD+fgFX\n", - "eGtirK6uRlhYWLvHQkNDvWbq6giWlQSFzk9RURFiY2NF1cGc0EdEBK6xnEBeHkVDg+tiz41/DpUN\n", - "lfjb93+TpRveurlLS0t5+/d6G2+Zd4qLizu8oL1pZlJQYJHWASKFwJ7QJwSINqBHkgUOu6TboFVr\n", - "sWXWFqw5vAZfnf3KdUGBeGsRVYqZW240Go1XnjiKi4uddkayCCHEZ+kbFRSkWG9iT+gDaIyKQv+k\n", - "Kpw40XG5pLAkbL5zMxZ+uhDnK6QNyBUXF+eVHLHl5eWibXRyExMTg5ISeaOdAh0vJrNCV/HVV+i8\n", - "sCv0I6+4FfoAcGPqjVg2dhlm/HcG6izSxTuPjY31ircIFbml2hvEx8d7ZQJ0Z77xlsmto3a8NQEq\n", - "KMgFk9LGHBmJ/voLHgl9AHh4xMO4NvZa3L/jfsnsvqwsonoDd2PWlVIFWiwWlwvrAKfpexIfRUE6\n", - "WmfiYpHWcfRZhkmh3xgVhf7kN4+FPiEE625bh5ySHLx84GWPzmloaJB/u7Of4M4NrCt5rJSWlnYY\n", - "wVCtVvvFja2g4AomhX5zdDR61B/DpUtAfb378gCg1+rx2ZzP8K9D/8K2nG1uy7vzFGEFbywcFhUV\n", - "OYV/7cr4w8J6Z6GrxbeSAikUMCaFvq5bN5DCfKSlAXzyKScbkvHZnM+weOdiHLl8pMOyntzcLGi4\n", - "4eHhHqeME0pJSYlHcbzlxt14e8OLqLKyEuHh4bK24Q/k5OQgPT0dRqMRAwYMwI4dOwAAhw4dQnx8\n", - "vJNJ8JNPPsGgQYMAAM3NzXjxxRfRs2dPREVFYfbs2W1SCb777rtITU3FTTfd1G7bL730EhISEpCU\n", - "lIR3333X6diuXbswZMgQGAwGpKSk4Nlnn7Ufs9W/fv16pKSkIDIyEv/+979x5MgRDBw4EEaj0Z6a\n", - "Ebia+vHhhx9GeHg4+vbti++++85+vKqqCvfdd5+9L8uXL7crYM3NzXj88ccRHR2NtLQ07Nq1S8xw\n", - "exUmhX5Yr15ovnIF/fvDYxOPjaEJQ/H27W9j+n+ndxiRU2zKMW8RFRUlu7eIJ25gLEyA3hgLgI2/\n", - "1ZdYLBbcfvvtmDp1KkpKSrBmzRrMmzcPZ86cwciRIxEcHOyUx3XTpk2YN28eAGDNmjXYvn079u3b\n", - "h8LCQhiNRjz44INO9e/btw8nT57EV1+1dbX+8ssv8corr+Cbb77B6dOn8c033zgdDwkJwcaNG1FV\n", - "VYVdu3bhjTfewGeffeZU5vDhwzh79iw2b96MJUuW4Pnnn8d3332HEydOYMuWLfaUjrayPXv2RFlZ\n", - "GZ599lnMmDHDnqVt0aJFCAgIwLlz5/DLL79g9+7dePvttwEAb775Jnbt2oWjR4/ip59+skcTlRtJ\n", - "1iz5RGeT8wWHiIPWggLaGB5On32W0qVL3YSmc8E/D/6TDlg7gFbWV7Z73JNoid6I7OiuH/X19fTA\n", - "gQM+7QOl3hkLd22UlpbSo0ePytoHb44F3ETZRCYkefFl3759NC4uzum7u+++m2ZmZlJKKX366afp\n", - "vffeSymltLq6mgYHB9ujkvbt25d+++239vMKCgqoVqulVquVXrhwgRJC6IULF1y2fc8999Ann3zS\n", - "/vn06dOUEOIyAuySJUvon//8Z0optddfUFBgPx4ZGUm3bNli/zxz5kz66quvUkopfe+992hCQoJT\n", - "fSNGjKAbNmygV65coYGBgbS+vt5+bNOmTXT8+PGUUkrHjx9P161bZz+2e/duSgihVqvV5d8mFgD0\n", - "8OHD7X5PechaJqNKqaKjoTGZ0P8aK97f6NqToiOWjFyCc+XnMP2/0/HFvC8QpPF98nMhBAUFobGx\n", - "0dfdkB1X8fwdMRqN+LWj+BydDPqMb3YgFxQUtMnklpqaak8gc/fdd2PMmDF44403sG3bNgwdOtRe\n", - "Pjc3F3fccYeTG7JGo3GKIdVRlrjCwkIMHz7c/rl1FNxDhw5h6dKlOHHiBMxmMxobG3HXXXc5lXE0\n", - "2+p0ujafHfNDJCYmtvk7CwoKkJ+fD4vF4hRttbm52d6fwsLCDtMzyoUUZlgmzTvQaNAUGor+8eW8\n", - "zTs2CCH4183/QlxIHGZ/NBtNzewtGikeRFcpKytze0GrVColFIMXSEhIwMWLF53GOi8vzx56u1+/\n", - "fkhNTcUXX3yBTZs2Ye7cufZyKSkp+PLLL1FRUWF/1dXVOQnPjswg8fHxLtMeAsDcuXMxffp0XLp0\n", - "CZWVlXjggQdEOTq0zoSWl5eHxMREJCcnIzAwEGVlZfa/o6qqyq50uOunXHReoQ/AbDSiZ3AhCgqA\n", - "OoF7rlREhfenvw+z1Yzfbf+doOBscgqZsrIyj+JoyC3oWLBhs5wbt6sxatQo6PV6/P3vf4fFYkFW\n", - "VhZ27tyJOXPm2MvMnTsXr776Kvbv349Zs2bZv3/ggQfw1FNP2YVgSUkJtm/f7nHbd911F9avX29P\n", - "Rei4UAtwa3FGoxEBAQE4fPgwNm3axPv6dbyfiouLsXr1algsFmzduhUnT57ELbfcgri4OEyePBmP\n", - "PvooTCYTmpubce7cOft6wF133YXVq1fj8uXLqKiowIsvvsirD0JxFZCQD8wKfUt4ODRlRejVC8jJ\n", - "EV5PgDoAH836CKfLTuOJ3U/wEqAGg0HW+DueaLesILfnTFlZmV8J/c7sq6/VarFjxw588cUXiI6O\n", - "xkMPPYQNGzagd+/e9jJ333039u3bh4kTJzr9bkuWLMG0adMwefJkhIWF4frrr8fhw4ftx90J6KlT\n", - "p+JPf/oTJkyYgN69e2PixIlO56xduxYrVqxAWFgYnnvuOcyePdvpfE8mAMcyI0eOxJkzZxAdHY3l\n", - "y5fj448/tu9Z+eCDD2A2m9GvXz9ERERg1qxZ9vwW999/P6ZMmYJBgwZh2LBhvFI7ikGSNvgsAMj5\n", - "QqtFrSstzjdiAAAgAElEQVQTJlC6YQOdM4fS998XsuzhTFldGR2wdgB9Zs8zlFLPFuzy8/Pp2bNn\n", - "xTfugr1799Kmpia35eROV+hJ/Tk5ObSwsNCnfeBTTs5+HD16lJaWlopuq/U1r+Bd3nvvPXrDDTf4\n", - "uhse4+p6gb+nS7RhNhqBoiJBbpvtEaGLwDcLvsGWE1uwcu9Kj86R20WwubnZ7eIlK3jLXdIdLJii\n", - "oqKiUF5e7utuKCgIglmhb4mIAIqLJRP6ABAbEos9GXuw+fhmbMjb4La8TqdDQ0dB/b0EC4LOaDQy\n", - "IeiojOsbntbNylgoiIMQwsS95W2YFfqN4eGSavo2YkNi8V3Gd/i2+FuPEqzLKWRYoKPMXY7IHXOG\n", - "hZuvrq4OwcHBbsvpdDrUCfUuUGCGjIwMp41aXQVmhb6lxbyTlgYUFQE1NdLVbVAb8PqI17Hp+Cas\n", - "2LOCecEuZ/9YCTvAwm/gqQdRV9UQFToHzAp9xMaiuagIajXQu7c4D57WlJWVoXdCb+xdtBc7Tu/A\n", - "ki+X+CTXrqeCTs6ga/7kQQTI+0Tgb2OhoCAE0UKfEDKVEHKSEHKGEPKXdo6nE0KqCCG/tLye9qRe\n", - "fbduoC3uUVKbeGw3d0xwDLIysnD0ylFkfJoBi5XN+PkGg0G2ePZ8BB0L2q1arZYtOmNVVRUMBoNH\n", - "ZVl4MlFQEIIooU8IUQN4DcBUAP0A3E0I6dtO0b2U0iEtr1We1B3WsydIaSlAqeRC39GkYQgy4Mv5\n", - "X6K8vhwztsxAvcXDWM5exGg02iMVSk1DQwOCgjwLUcGCoDMajfaAWFLT3Nzs9SxmNlOR8lJe7l5S\n", - "IfYKHwHgLKU0l1JqAbAZwP9rpxzvHhvj49Gs1QLV1Rg0CMjOFtlTB1q7Suq1enw6+1OEBYZh4gcT\n", - "UVJ7NR2elIMtlIiIiE7vLeLpOHcmzxlXftR79uzxyN/6zJkz9nAJUr887QPfsnxf3333nUflsrOz\n", - "UVpa6tM+yDkWFotFskVnsUI/EcBFh8+XWr5zhAIYTQjJJoR8Tgjp50nFWq0WlvBwoLgY110H/Pwz\n", - "QGVUNLVqLTbcsQETuk/AqHdG4WQpF8ifytmoh4SGhsq2M5iFSY0PrEyAco1bc3MzExMgC9d9XV0d\n", - "dDqdR2XlfBpmAXfZ7fggVuh7cmX8DCCZUjoIwBoAn3pauTk8HCgpQXw8oNUCFy+6P0cMKqLCqgmr\n", - "8PTYpzFu/Thk5WbJ1panrpIAJ2BYuAnlorGxEQEBAR6V1el0qPc0nZofwmddITw8XDZTFwuUl5d7\n", - "vN7EijIgF1I6GYgNrXwZgGOc1GRw2r4dSqnJ4f0XhJC1hJAISmmbXygzM9P+Pj09HQNaNH0Adm3f\n", - "GxFM7xlyD1LDUzH7o9m4J+kepCNd8jYqKys9vrnlhIXJhG+wNRaeTuQaNz43t5x7J1gY47Kysg7D\n", - "MDui1+tl2zvBwliUl5ejR48eAICsrCxkZWUJrkus0P8JQC9CSDcABQBmA7jbsQAhJBZAMaWUEkJG\n", - "ACDtCXzAWegDQGErof/LL8D06SJ77CETuk9AVkYWJr83GTWf1+AfU/6BALVn2qgnVFZWMiHoWLmg\n", - "O0pG3pUoLy/HgAEDPC4v1+TDgjJQWVmJa6+91qOyLFzHcmI2m+1Pw+np6UhPT7cfax2J1B2izDuU\n", - "0iYADwH4CsBvAP5LKc0hhCwmhCxuKXYngF8JIUcBvApgTvu1tcVm3gGuavpS4OkF0je6LzaO24jz\n", - "Zecx8YOJKDQVStMB8N8UxcJNqNVqZYm06Y9hleUyufGxY3d2fOFN1R4s3Hssee+AUvoFpbQPpbQn\n", - "pfSFlu/WUUrXtbx/nVI6gFI6mFI6mlL6o6d1mw2GNuYdb5MSk4LXbngNk3tMxvC3huPAxQOS1Gsy\n", - "mZjI0cvngg4PD5dlsaypqcltjl5HWLgJQ0JCnDIwSQWllNcNzoKGq1KpYLVaJa+X798m13XBwhhL\n", - "ie+n0Q6wGI12TT8lBaivB1r2a3mNiIgIVFVWYfm45Xjz9jdxx3/vwCsHXhG9g9cfb+7O7iHBB7k8\n", - "Z1j4nQF+/ZBz74Sv4TuRBAUFMe9owLTQNzvY9AkBhg4F/vc/7/YhNDTUvhv2ll634NDvDuHjnI9x\n", - "66ZbUVRT5OZs1/ijFiOX0PdHQceKt4gc1wVfsworeyfkuI4aGhp4mdsiIiKYV4yYFvoWB5s+AIwa\n", - "BRw86N0+tM7L2i28G/Yu2ouh8UMxZN0Q7D6327sdkhC+AqMzh5rmm6/YYDCgqqpK8n6wQHV1Na+0\n", - "fP4g6ITCd+2NlQmwI5gW+o6aPgCMHi1e6PPZ/OIKrVqLVRNWYeOMjbj3s3vx6FePos7Cz12Mr8CV\n", - "Q9DV19czsWjIgo2ej388wE6SdjmuC76CrjOHmhYi9Fk3dTEt9JvCw0FLS4EWX+RRo4AjRwAx8bZM\n", - "JpMkyYUBzq3z6ANHUVhTiMH/HizZIm97yCFgWAmrzBdlLK7CwliwsnlQrrHgowxoNBrZAgJKBdNC\n", - "PyQiAggOBlpmTqMRSE4Gfv1VeJ1S39xR+ih8OPNDvHjTi5i5ZSYe++oxWYK2yXExVVZWSra125vI\n", - "IWSqqqp4XxcsCDo54Gve6cw0NjZ6HJBQTqS81pgW+kajEU0taRNtXH+9OBOPXBrdjL4z8OsffsVl\n", - "02UMXjfYbQgHvo/lcmy5r6ioYGJXMN+x0Ov1kntI+Kt/vBzukqz4x/srrCsDTP+yERERaDQYnBZz\n", - "R48GDoiwolRVVcmmxUTpo7D5zs34201/w8JPFmLBJwtEefg4IofnjNls5rV4CbBxQRsMBlnspqx4\n", - "EfHBYDDIFoyPDyyMXVBQEBOOBqzDtNAPCwtDQ2hom8XcH34QXiefQGc2+Aq66ddMx28P/ob4kHgM\n", - "eGMA1h5ZC2uzszbGt045XARZuFGFIMdTj5DJjIXxk2sC5AsrykBn9KiSwvnEEaaFvkql4jR9B6Hf\n", - "pw+3SSs313v9EDLgIQEh+Pukv2NPxh5sPr4ZI98eif15+wX3oTN7SPAVGJ315rbFTudDZ420abFY\n", - "eO3SBrixYOG6kFoZqKmpQWhoqGT1MS30gba++oQAEycC33zjw07xYEDMAOxdtBePXv8o5n8yHzP+\n", - "OwNnys7wvjCkzp4DsKGdWa1W3vbjoKAgNDY2ytQjz5F6/Orq6hAcHMzrnLCwMNlSafJB6mtTyMI6\n", - "K089UiP1OqR/CH0HTR8AJk0Cvv7ae30Qe3MTQjD32rk4+eBJjEgcgevfuR5rz69FWV2ZRD30X0wm\n", - "k6DFZKkFLgumGiE3t1qtliXuja8RMhZ6vV6WeEi+pusJfYf4OzYmTgS+/dbuvu8VpBAyOq0OS29Y\n", - "iqy7sqAOUKPPa33wbNazqGrwzSOpvwo6VpB6/Px5LKRGyFjI8TQsBDnciaV0PmFe6Jtb2fQBzlc/\n", - "Kgo4epR/fUIuCqldBLWNWrwy4RX8+Lsfcb7yPHqu6YlV+1ahutH3j+nukNpFUKigY+HmlhqhiXWk\n", - "Hgsh9Ukt6GpqanibuuToBwtYrVbe6xsdwb7Qb2XTtzFpkvfs+lIvltk2RfWM6In3p7+PH+79ASdL\n", - "T6Ln6p54fv/zqGzwjl1SyA0SFhYmqYtgVVWVpItUQhEyFhqNRtL8AhaLhbcLLcCOoOuMJjchsN5v\n", - "/xD6rTR9AJgyBdi1S8KGTp0CFiwAHnkEKC11OiS1V0B9fb3TLr/ekb2xccZG7F20F7+V/Ia01Wl4\n", - "fPfjuFTtlHmSiZtb6gmwubmZtwstwMZYdGYvIr7o9XomfORZELiEENnSWEoB80LfGh4OlJcDrUwK\n", - "N90EZGe3+xDAn8JCYMIEoH9/wGzmZhQHc47UXgGuYun3je6LjTM24pfFv6CZNmPgGwOx6NNFOF58\n", - "XLK2HRFyg0g9AbJwkwJsjIVQWBjDzug5I1SxkPppWGqYF/pUrQZsgt+BoCBg8mRg+3YJGlm2DJg3\n", - "D1i6FHjjDaBbN+D55x3a8q6LYIohBf+Y8g+ce+Qc+kT2waQNkzBpwyTsL92PpmbfBnMKCwtjQtCx\n", - "QGf0kRe6EagzjgXfWPo2WJ8AmRf6AIDo6HZNPHfcAXzyici6Cwq4SpYt4z4TArz6KvD660CZb10q\n", - "jTojnhz7JHKX5OKewfdg66Wt6P6v7nhu73Oi8/UKffxUq9VMPLpKqd0K2QgEcCkTWdbohCA0Ci0r\n", - "Tz1SItTJgPUJkHmhTykFYmLatePccguwfz/gaUiadgNJbdgA3Hkn4Og1kZzMzSivvSai567hK7AC\n", - "NYGYe+1crBmyBjvv3olL1ZfQb20/3LnlTuw6vUuQ9i/1Lj+hsGCb5xtL3wYrIYWlRKig64xxb4SO\n", - "BetPw8wLfQCgLjR9gwGYOhX47389q6dNyFhKgfXrgXvuaVv4sceAdevswftZubkHxQ3CutvXIXdJ\n", - "Lib1mIRV+1ch+Z/JeGL3E7xs//4aVtmGlL+HGP94FuzpUqLsFbiKUBdajUbD9IY55oV+SEgImtrZ\n", - "oGUjI4OT257Q5oI+cwYwmbh4za3p14/Lxi7D1l+hAsvRR94QZMDiYYtx8L6DyMrIglatxdSNUzHs\n", - "zWH458F/4mLVxQ7rUm7uqwjZ8t9ZEbMRiAXFSMo+iImlz7IywLzQNxgMqG8VadORyZOBvDzg5En3\n", - "dbURdF9+yT0quPqBFi3yfEbxAq7irPSJ6oPnJz6PvD/l4YWJL+BEyQkMWTcEo98Z7XICqKysZCJR\n", - "htCbIzAwULLFdZPJhJCQEEHnsiDoAOn6IdSFVsEZVq6L9mBe6IeHh6NWr3cp9DUa4N57gbVr3dfV\n", - "xrxjE/qumDWLKyNxwg6hgs7dYplapcaktEl4e9rbKHysECvGrcDx4uMYvG4wRr8zGi8feBknS0+C\n", - "Uir5Lj9v01ldR4UQHBzMRARWFsaws8YikhLmhb7BYEB1UFCHDvl//COwcaM9q6JLnBZyLRZuFXji\n", - "RNcnREYCw4YBu3czcUHzcQXTqrWY2nMq3vl/76DwsUIsv3E5zpWfw6QNk9BrTS+8fu51fHP+G5it\n", - "Zpl7LQ9SusWJ0cr87bqQExa0W1aijkqF1LH0AT8Q+oGBgagLDnap6QNAYiJw883AW2/xqDg7G+je\n", - "nUu82xF33AFs28ajYveI2fQh5IIOUAfg5l43443b3kD+n/Lx8V0fw6A1YPme5Yh5KQYz/jsDa4+s\n", - "xanSU16/cYW2x7pbHF+ExNK30dncJYW60AKd77qQw8uOeaEPtB9pszX/93/AP/4BeBxZ9ccfgVGj\n", - "3JebPh3YuRNEoqTkYoSqFI+uhBAMihuE+SnzcfC+gzjz8BnM6DsDRwqOYNKGSUj+ZzIyPs3AB9kf\n", - "4HL1ZVFtucPVzmRP0Ov1kpk0xGhSUk2S9fX10Ov1gs5lRdOXSiMVs7DOygQo1VjI4XDhF0bd9iJt\n", - "tmbQIGDcOGD1auDJJz2o9McfudAL7khKAnr2RFh2Nhf7QST19fWCogfKRXRwNOYPnI/5A+eDUoqz\n", - "5Wfx7YVvsf3Udvz5qz8jQheBMcljuFfKGFwTdY1kbYvRYlgwq0iJmJs7MDAQZrN/munaQ8xYhIaG\n", - "dirzTkVFBZKTkyWt0y+EviU0FKiu5nzmO3jsW7kSGDOGW9iNjW173EkrO3gQeOopzzpw++2IOnCA\n", - "890XiVDfXxtSCbv26iGEoFdkL/SK7IUHhj2AZtqME8Un8MPFH7Avfx9e+P4FVDVWobeuN6app2FU\n", - "0ihcF38dDEHC/p7O4DYqpUYXGRkpSV3+TmVlJa65RphywcqOcamoqqrCgAEDJK3TL4Q+1GogIoKL\n", - "fhkX57JY797cPqvHHuMWdl1SXMyFWPD0wpo2DVFr16LJYoFGq+XX91b4082tIipcG3stro29Fg8M\n", - "ewAAUGgqxJtfvoni2mI8vedpZF/JRkJoAoYmDMXQ+KEYljAM18Vfh7BA9+6glZWVSEhIENw/qUwr\n", - "LCxAVlVVoUePHr7uBhOmLqGx9KWGhbGQw4XWP4Q+cDX+TgdCHwCeeQYYMAD4/HMuTEO7HD4MjBgB\n", - "eJqb9dproSYENYcPI3zMGH79bgUrN7dQ4kPjMSV5CgYPHoygoCBYm604WXoSPxX8hP8V/g/bcrbh\n", - "WNExJIQmYFDcIAyIHoABMdwrLSINGtXVS06MRscKtjC6fPP8tkbMRqDOSGcz37GEXwh9QojL+Dut\n", - "CQ4G3n8fuOsu4MgRLoxOG7KzgSFD+HQAjZMno/nTTzn7kQg6w81tWziMi4uDWqVG/5j+6B/THxmD\n", - "MwAATc1NOFl6Er8W/YrjxcfxwbEPcLz4OApNhegT1QcDYgbg2phrYSmyILoqGj2MPRCgDuDdDxYE\n", - "g82jyt/NVFKg0WjQ1NTk1/s/pMIWl4mFa7Q1fvHrUEpdRtpsjxtvBP70J2DGDGDPHqDNZstjx4Bp\n", - "03j1gUyfjoDly4GXXuJ1HquIefy0ucXFuXjq0qg0du3ekRpzDXJKcnC8+Dh+Lf4VBwoPYP2H63Gx\n", - "6iISwxLRK6IXekX0Qu/I3tzaQkQvpIanOj0dyIGYG9PmLaII/atJZXxtvpTCtCK2DlsAOiGhmeXG\n", - "L4Q+AI81fRt/+Qtw9iwn+D/9FHDyhsvOBp5+mlfz+ilTYJ0/H7hyxa2JSTC5uVxkz/37uUXr664D\n", - "fvc7YORIexEW7M8GgwH5+fm8zwsJCMHwxOEYnjgcAJAVmIX09HRYrBZcqLyAM2VncKb8DHJKc7D9\n", - "9HacLjuNKzVXEB8Sj9TwVKQaUtEtvBtSDalIDU/FlboraGxqRKCGf4pBqQgPD0d+fj5SU1N91gep\n", - "ELsRyKYM+FroS4FYgW1TBhShLwYemj7AhdP597+5hd2bbnJItlJXxwXr6dOHV/OqoCCUDhuGmF27\n", - "gPvu43WuR7z/PrcCfd993IYDrRbYt48L+zx2LBffX6KomGIfO3U6naTb/rVqLXpH9kbvyN5tjpmt\n", - "Zlysuoi8qjzkVeYhryoP+/P3Y+OvG3Hqyinc9/N9iNRFIiksCQmhCfZXfEi80+dIfSRUpK3d3Wq1\n", - "irLHsx5Glw9iNwKFh4fj7NmzEvZIGFKYVMR6ljmaQFnDf4R+TAzw88+8TtFoOFm6bBkwbBjFn/5k\n", - "xPjQ3ziBH8Dfhlw6ejRiduwQJfTb1dRfew145RVOw+/b9+r3I0ZwMSaeegoYOhT47DPB7TpSU1Mj\n", - "OMAYIJ0t3ZOnlgB1ANIi0pAWkdbmWFZWFsbeOBYFpoI2r+8vfo8CUwEKTYUoMBWgurEacSFxiA+N\n", - "R0xwDKL10YjWRyNYFQzUAjVnahClj+K+D45GsDbYo7+zM8V6ESvoWEkqI8XTcGVlJSIiIgSfHx4e\n", - "jry8PNH9kOPJXrTQJ4RMBfAqADWAtymlf2unzGoANwOoA7CIUvoLzzZ4m3dsqFTACy8AgwfX4qGH\n", - "+uP4NefxbM9xSORdE1A2YgQnoOvrAake23bt4lIzHjjApWlsjV7PZfIaMQK46SYYli0Dxo8X1aTQ\n", - "pCEsolapkWxIRrKh4w0sDU0NuFJzBQWmApTUlqCkrgSldaU4W3gW5Q3lOHj4oP37ktoSUFBE66MR\n", - "pY+CUWdEeFA4jEHc/47vjTojzlefR1xJnP2YTsveI70nVFZWIikpSfD5KpWKCXu6FFRWVqJ79+6C\n", - "z2c5qYwooU8IUQN4DcBNAC4DOEII2U4pzXEocwuAnpTSXoSQkQDeAOBB/AOHTmo0sISHQ8vDvNOa\n", - "kSNL8c03BJvvy8e1n7+I2zOAxYu5SAyePt03hYVxXj/ffgvcdpugfjhpjwUFnP3ps8/aF/iOzJ0L\n", - "REVhwF13gfboASKwfYC7oGPb273mZ/ARDkGaIHQL74Zu4d2cvj969Ci6devWRsOtNdeitK4UpXWl\n", - "qGyoREVDBfd/Pfd/oanQ/l1eUR5ev/S6/TgAhAaGIjQgFCEBIQgNbPk/oNX/DmXyivPQcLYBoQGh\n", - "0Gv10Gl10Gl00Gl13GeNDmqVvCGPq6qq0L9/f1F1sCCwpXgSNZvNorzsWPTasSFW0x8B4CylNBcA\n", - "CCGbAfw/ADkOZaYBeB8AKKWHCCHhhJBYSmmRp42Eh4fDZDIhQoCmb6OqqgopKSl4wfh3PPYuwfuX\n", - "b8LvfselWrz5Zi6Ew9Ch3H6tDj3Opk3jFghECF0AXNauxYuBBx5oP4lLe0yejNOvvIL+997LxZuY\n", - "M0dQ01VVVejdu639HJcucUlj9u3jVsGLirgF5bAwLqpd797A8OHcU4cEN7fYG0MKF8E24bZbCA4I\n", - "RnBAMFLD3S/QZmVxC9I26i31qDHXwGQ2wdRosr+vMdfA1Ghyel9oKoTJbML5kvM4+ONBmBpNqLPU\n", - "ob6pHvWWeqf3GpXGaRKwTQx6rR71pnokXknkjmv0CNIEIVATiAB1AALUAQhUc+9t37X+HKAOQE5p\n", - "DlSXVR2W0ag09ld7ayRSwILAZGHykguxQj8RgGOGjksARnpQJgmAx0LfYDCgsqEBESI0fXtGoJMn\n", - "ETUyDY9159ZNz53jNnJ99RVnZbl4kUuYlZrKveLiuPVToxG4dCkKSL0LQX99AIG/b0agToWgICAw\n", - "kHup1dwCskp19eX4mRDAauXkJdm0CcjPBz7+mNffob3hBpRu3ozoBQuAqipu4uCJ2WxGgOOaxuHD\n", - "wN/+BmRlcave48dzCWTi4rgZsLoauHwZOHGCizj6l79gTG0tkJ4O3HAD9xoyhFt8FkNtLff0Y3uV\n", - "lwMNDdzLbL460EFBgF6PlLIy1BUWIiwxEQgNdX4FeubR0+HGKquVa7uxkXuZzVd/ULWae6lUUFdX\n", - "c2PU8r1OrYYuKALR+ijXCXpa0XriaA2lFGar2WkScJwYfsr+Ccndk9FEmlBnqUNDUwPMVjMamxph\n", - "tppRa6lFRUPF1e+arx5rtHL/F5UWYWv5VqfvWpexNlthabagqbkJKqJymgQ0Kg1oE4XuF53Td1qV\n", - "tk05Vy+tWovS4lK8V/keVEQFFVRQq9Tce6KCmnDv2/vO9n1ebh72791vn5jclW/v+5PFJ1H6WykI\n", - "CAghICBQEZX9veP/KqJq8x0BwdHyo0Au2j3maV3na88jqiiqw/J8ISJjic8EMJVSen/L5/kARlJK\n", - "H3YoswPAi5TSH1o+fwPg/yilP7eqi7rqi8lkwqmcHAwbM4YTDAIWYffs2YPxw4dzawMmE3fDtkNt\n", - "Lefck5vL/V9UxMXpr6gAzp4thUYThcbD2WhM6YlGdTAaG6/Khebmqy9KXb2naG6+Kgj4KzUOY0Sp\n", - "QwV86nSowxanhJCWE12f7FQvbb7aB/vvRrjT7XW5ap46/U9AnZ8cbOd6Uoerzx123naO/R8FnlBQ\n", - "gDQDqibXL7Wl1XdW+3vapnyrsqQZIJQ7hzQDxPZ/s8N3bb+n7X7f7LJ8x99buT6AOvzf3HKLUOdj\n", - "pLntd6CgHRyz/40e1tW2PLj/V58DpdRjSSJW078MwHEFLRmcJt9RmaSW79qQmZlpf5+enm7XfEJC\n", - "QlBTV8e5bZaUcKYGnhBCgNOngZ49XQp8gNvR268f92pNVtZxrk9LP+S02uee492PrKy9SN+/H/TX\n", - "46CbPczo7kBDQyOOHj2KUaNGAZcLODPTNddwSdzDwtxbXSwW/JaZiX67dnGC8IknuAxhbrT01vVm\n", - "Ze1z1kwrK4FDh4AffuCC2V26BBQWArU13HGi4uInRUVxv2P37jivUqH7+AlAWhqX2yA6mtcsWF1d\n", - "jbNnz+K6665re7CxkZvcTSagpoZ7mUzcHxIYyCkOAQE4kp2N4aNHw+mRzfbSaDzqz969ezFu3Lj2\n", - "B81qdf1qbrbX/8OBAxgzZkz7E17r79r5v6CgAPX19Ujr2bPjsh0c2//99xg7dqxnbQIA0QBwVsDc\n", - "PbG4o6mpCT/++CNuuOEGz05o54I/ffo0goODkShATthw+Zt60L6Nffv24cYbbxTcB5PJhNOnT2Po\n", - "0KHOfdu3F3v37bV1AKuwile9YoX+TwB6EUK6ASgAMBvA3a3KbAfwEIDNhJBRACpd2fMdhb4jti3N\n", - "dl99oT/myZOeB1nriNtv51wpBQj9gLIy4NVXQX76CUJMonp9EBob67nF5+RE4McD3PbjQddyJppZ\n", - "s9qf1EpLgbffBt54A0mRkVC/9CIwZYqQRw0AgFpNndc+osKBW6dwr/ZweirhnnguZ2WhtwhPJKMx\n", - "BA0NpvbnK20gEBIIxEd1WIe5sQbagX07LOMOjYa6mDMJuFus49usoaEBgclx0CbGCO5DZJAaOTk5\n", - "0EYKz3us0mmgDRG30U0bqII2ULit31Rbg4gog6g6ouMiUVpaCm2Q8IVvbZBa1PkAoA5UQasTLmLr\n", - "ymoRFRfZpo6bpkzETVOuZvxb9bwXhT6ltIkQ8hCAr8C5bL5DKc0hhCxuOb6OUvo5IeQWQshZALUA\n", - "7hHcoEC3TTsihb7d/DRqFKfF5ua697pxoKmpCd3Xr+diP4twB3NCp+O0/L17gaVLuWwyt90G9O/P\n", - "ae+XLnH+/z//DMycCWzbhmyTSZQ2BghYbGtVXoq8AiqVSnQYXRYW7KQI4xAcHIxajzMIsYsUY2Ew\n", - "GERvEmPhupDLy060nz6l9AsAX7T6bl2rzw+JbYcQwntXbqs+AKdO8Y650y5qNXDrrcCOHcDDD7sv\n", - "30LNwYNcXP4NG8T3oTXjxnFmlWPHgO++4xZdGxuBhARuxXr8eM52BXALtj5GbF4BqZDCU0SsgJBi\n", - "LOxPwyLrEIsUYyE2Cm1gYCAaGxtF1SEFYsfCpZedSPxnRy4gjab/f/8n+HSnyHnTpnEbtTwV+pQi\n", - "4MknYXrkEUTIGZxr4EDuxThSxWgRK6hY0OiqqqokyY7EwliI7YPJZJI8J6y/0sbLTiL8IkcuwD/S\n", - "Zhuam4EzZ3jH3HHE6RH65ps5rdrTx8gvvoDq0iVoHhL90MMEKpVKVPgBVjR9KRCrZdfV1TEZmMsX\n", - "SJGbAGDD15+FPrSH3wh9MaEYACCouJjzHhERc8Yp6XJQELeb9o033J9osQCPPYYzixcjRKKgab4m\n", - "LEZCJ0QAABjhSURBVCxMVJwVsTsepUKKGzMkJES0PZ0FASFFH2xJZRTEI9c14TdCHwAn9AVq+sEX\n", - "L4r23AkPD0dFRcXVL/7wBy6im7sbfu1aIDERZaNGdRotxhZFUOFqSGF/RwrzTmhoKGpqaiTojf+j\n", - "VqvR1NTk6260wb+Evgjzji4/X7TQbyPouncHJkzgQiK44sIFzrXztdcEu0eyiNNTjwBYsKUD0vSD\n", - "FaEv5m+R6vdgZSzEYLFYJMn+JfYekQv/EvoizDv6/HxR9nwACAgIgMVicf5y1SouLHJ7k1FTExeG\n", - "+YknpNkfwBCdIY68VOnsOsNTj1RZnjqD0K+urpZkvUnsWMilGPmX0Beo6VNKoZfAvNMuvXsD99/P\n", - "2fcdbZmUAo8/zrl3PvaY9O0KRCotRq1Wi7LdsmCiEps0xIZWq22rDPgZYmPp2+gMyoBUY8HqBOhf\n", - "Qj8sjAt6VV/P67T6+noEX7okidBvV1itXMlt8b/nHi5IT1kZl+Zw/37gv/91E7aTP2I0ACWf61Wk\n", - "urkBNiYxMUg1FhqNxu+Tykg1FmKTynT5hVyVSgVrc/PV+Ds8qL54EZraWuHhG9yh1XKhOrVaICmJ\n", - "C8+pUnGboERk33GFmItBSldJFmzIYuhMbqOAuOuiMyXWEYvYzHI2pNgwJwd+I/QNBgP32CjArl9/\n", - "9CisaWmeZ0sRQkgIF9umqooLPvbWW1yIXxkQ4xUgpXbLAmJCMSiC7ioNDQ1MuNCygFRrPaziN0Lf\n", - "bh8T4LbZdOIEVO2FzZQDjUZyc05rxNgKpdJixCLVTRUaGir4EbqpqQlasTkAWmBBo2PVRdBXsPCb\n", - "iKHLL+TaBZ2AxVzt+fNQi0wDxxJihL6UWgwL2hCri2W+gFUXQV+g0+mYzVHra/xG6IeGhqK6ulqQ\n", - "eUcKd02WMBqNiqBrQewE2Jnw9wlQyt/D38eisbERgR5mf+OL3wh9lUolOP5OsAQbs1giODiYiV2P\n", - "LCzksuIiKPSpx2q1Qt1BUh8++Lugq62tlcz06O9jIed6k98IfTuxscCVK56Xb2pCUEEB0KuXJM0L\n", - "FVZSajGsegV4ipRajJj9AiyYp1wlZheCWBdBXyOloPP3DXNyOlz4n9BPTuYSg3jKhQswR0YCer18\n", - "ffKAuro60UlDWEOo0FT2ClxF2StwFSnHIigoiImY+kKR8x7xT6Gfn+95+ZMnUZeSImkXhGjZUs/c\n", - "LNzgQp82WHEblfJpyd/HggUUF9qr1NbWQi+ToupXQp8QclXT9/SRXmKhLzSKYGfUboX6yHe2TVE2\n", - "hAh+qZOGsGD2E9oHuZKG+CtdfkeuHb2e2wjlqQePxEJfqOcMK4JOygvJYDBwHlU8kVOL4YOUY6HX\n", - "6wW5CEqVNIQllJj6VxF6jck5efvn1ZaSAly86FlZiYW+UK8AqbUYFjQ6oWNBCGHCPCUl/u4tIiVK\n", - "TP2raDQa5oLx+afQ99SuTymQkyOp0Pd3rwApaZNUxkewMIEoQv8qrIyFEMVI6icUoRvm5Lym/VPo\n", - "e6rpt5iAzBKaVViJIsiCoAsLCxNk3pH6KUVIfZRSSfvBijIg5LqQKty2DTFPgL5GqnDbNliZAB3x\n", - "T6HvqaZ/8iS3KYuBi4kFpBa29g1zfkh9fb2k6wpBQUF+u+1faq8ZoRMgC9eS1A4XitCXCk81/ZMn\n", - "0dy7t6RaDCsIuUHkCLYmpB8saHSseFN1xrFg5WlYCFI7XAjdPa8s5LZgjyLIQ9OvT01lwmuGBVgR\n", - "dCwgh3+8EAHOgnbLyl4BViZAKeUFC39Ta/xK6NtdBFNSPBP6v/0GU1ISExe01AQGBvI2J8jhNsqC\n", - "oBMSloIVF1oWMJlMTITbZmECtFgskoXbFoOykNuC3T6WkMClJXSXNjE7G6UJCZLf3CyYNIxGI2+v\n", - "gM664zE4OBi1tbW8zmElaYjU14XQmPosaqRSwMJEwhr+KfTVaqBHD+DsWdeFi4uBhgaUBQdLuhov\n", - "FKkvPiELRBaLRfIdjyxMgELGQg4hJ9SLSEqEbJhjRTBK/ZsIianfWSc/R/xK6DtFEezdGzh92nXh\n", - "Y8eAgQPRTGmn2/EICPORZ+XmlhoWPSR8BStjwYLwFDIWLNwjVqtVMe/YcHIRdCf0s7OBQYNk6QcL\n", - "F3RQUBDq3Zm3vAALNn17/mQf9gHgPxZy5GIV4i7JwvVsNpslt6WzMgHyRW6HC78S+k54IvQHDvRe\n", - "fzpALgHD92Zl4eZuamqSLGmIDa1W65e5YWtrayUPt23PMMcDFrTbiooKGI1GSev01/SRcoyFI51X\n", - "6B87Jpumzxc5bm4hsHBzd9bFZCFUVFQgIiJC0jqFbJhjQRkoLy+XfCz8dcOcHGPhiH8L/VOnuPg6\n", - "rWls5CaE/v1luaD5RhGU4+b2VyoqKhAZGSl5vSxMaHwpLy+XVaPzJ+TWbv2JmpoaWZVE/xX6MTGc\n", - "F8/ly22P/fwzF35Br5dFGPCNIijXBc2CRsc3imBnFnSs7BVgQXPnixJL3xllIbc9CAGGDgX+97+2\n", - "xw4cAEaPlq1pvgtErGwEkmMC5DsWrMTSl4PAwEBeKfpoJ/Uss+GPT19dAf++4oYOBX76qe33jAn9\n", - "5uZmyRcvATY0Or5jwUosfTn64K8Lh3KgxNS/Cmsx9QULfUJIBCHka0LIaULIbkJIuz5GhJBcQsgx\n", - "QsgvhJDDwrvK4aQ9tKfpUyq70A8LC/PLm1sOQSc0k5jUsDCRhIeHo7y83Nfd4KVhy6WN+6O7ZGNj\n", - "IwIDAyWvl68yIPe1LEbTXwrga0ppbwDftnxuDwognVI6hFI6QkR7bbFp+o4Xbk4OEBAApKZK2pQj\n", - "Go2GCRdBvje3HDe4Xq/nHQLB11gsFlmevCIiIpgQ+nyQOkevDaPRyMRY8Lnm5Vp7Y20CFCP0pwF4\n", - "v+X9+wCmd1BWnqkrKQkIDQV+/fXqd59/DtxyC0AIrFZrp7aZ8qGurk4WjwAhwc58TWVlpSw3d1BQ\n", - "EC+bPgvI5VkmZMOcHPDRmuUaC1YmQBtiJGIspbSo5X0RgFgX5SiAbwghPxFC7hfRHoBWPyIhwM03\n", - "c4LexrZtwO23A1BCCTvS2V3i+Ew8cvpBs2Bm4tMHubyp1Gq13yVHl2ssWHsa7jC7CCHkawBx7Rxa\n", - "5viBUkoJIa7uujGU0kJCSDSArwkhJyml+9srmJmZaX+fnp6O9PT0NmVsIYXtERJnzAAeeQT4y184\n", - "005uLjB5MgB5/eNZubk93cpfUVGB+Ph4L/SKfSoqKpAiYd5kf6aqqgphYWGy1M3CEyCfPsjlNiq1\n", - "rMjKykJWVpbg8zsU+pTSSa6OEUKKCCFxlNIrhJB4AMUu6ihs+b+EEPIJgBEA3Ap9V0RGRqKsrAyJ\n", - "iYncF+PHcxr/J58A777LTQAtmbLKy8vRt29ft3X6K7ZHaE+eZiorKzv1WOh0OtTX10On07kt29DQ\n", - "4FE5f8W2edBT06ZiAvUvWivEzz77LK/zxfza2wFktLzPAPBp6wKEED0hJLTlfTCAyQB+bV2ODzah\n", - "79AI8MYbwD33ALW1wKOP2g/JvbPN17QZiw5oamqSLW0kC089ERERHo9FZ4ePPV1ObZyF6yIgIABm\n", - "s9nX3WBiLGyIEfovAphECDkNYELLZxBCEgghu1rKxAHYTwg5CuAQgJ2U0t1iOtzuSvgNN3BJVfbs\n", - "4Tx3HGBpsKWGj9DvzOMAcGPh6WIZC2YHOeEzFnLi6TjL+XvwuUdYoLGxUfadyYJVP0ppOYCb2vm+\n", - "AMCtLe/PAxgsuHft4HKBiOFHVLkuar1ej7q6Op/2gU/dcoQStmEwGHD8+HFZ6uaDp2NhsVhke/KK\n", - "iIjAiRMnkJaW5rYsC8qAyWSSbV0hMjISRUVFfrOe5Q2HC3YlZSeBj22VLyzcsHyora2VLRerv3mL\n", - "yOlkwIoy4On1KWc8JqPR6HGyITnvJ0/H2RvBGTu10GfhR+wKoYT53NwsRBtlYbLs7G6jgOf3iJxj\n", - "odFoYLVaZalbDrwRkLBTC305tRhPM1d1BUHnaajpzhxhky+sjAULE4Sc5h1/Q4481q3p1EJfTjxd\n", - "ICotLUVUVJQXeuQ7PPUWkdMnnA9yKgOeeovIFeeFL3Kbdzypv7m5mYnJh4UFfm/0oVMLfTkvJE+F\n", - "fn19facNJWwjMjISpaWlbst19lDCgH/G35GLsLAw3qkbOyueKgPemPw69x0oI54GUZJ75mZBQ2LF\n", - "LY4FTY2VsfAEucfLn/ZOyLmPBfBcGVA0fYZhxVvEk4tE7gtJr9d7lIuUhQnKarXKEmHTBmvBtTpC\n", - "riB8NljZL+DJdVdWViZLGk8bLCkDnVboNzU1yXpz+xNyukraYGHy8QS5F9a1Wq1HYbdZmABLSkoQ\n", - "HR0tW/0hISEeJVKReyw8ue7kXntjyeznl0LfE28RuWduf0Lum9ufUMbiKnKPhacLuSwoA2VlZbIq\n", - "A6zk4AD8VOh7Yk8vLS1l4uaWW4vxxHW0K3gQAZ4JGUUZuEptbW2njk3FB6vVKqtN31OUhVwXeGIf\n", - "Y8UPWm6ioqLcjoU3kpGzYK7wRBnwxs3NwlhotVpmvEVY6IM7ZYCFcQCUhVyXeCL0Wbm55f4RPV0g\n", - "YuWilpOYmBgUF7cb4Zs55L4uYmJiUFJSImsbUiBXGk9HDAaDW9dRFkxMcnsQ2fBLoa/T6TzyFpEb\n", - "Fi4UVtLSuRsLb4xVdHS0W0HHwuQnZ7A1GzExMSgqKnJf0MfU1NTIkqPXkdjYWCbGwt215y3To18K\n", - "fX/BGx5EKpWKCddRd1RXV8u+G9dTzxm5cTfBeWONhZWNUe4EnTcW1j1RBljAW2tvfiv0WdDYgI5v\n", - "cG9EzGMFFm5ugI2nL3d4YyxYuT/cUVJSIrug02q1sFgssrYhBd6K0+W3Qp+FmzssLAwmk8nl8aKi\n", - "IsTGusoX37nQaDQd3lhdyVXSnRdRV1IG3CH3BjFPYWGSVGz6fkBcXByuXLni8nhpaWmXcQ90t4jq\n", - "lMy+k+POi6grxCCyERgYyMT6mzu8oUS621/krYmna1x5MuFO0HWlmzsuLs4vFsu8cXPHxsa6vS68\n", - "AQtPw/7gUWW1Wr1yn7KyK9dvJRILF7S7hUMWHhnlTFHoSGhoKBMLhx1dF956fI6KimJ+4dBkMske\n", - "mgNwPwGyQHFxsVfMsKx4Efmt0O+I+vr6LmNKcIfc28ttuJtYWJgAvbXGwooXUUcUFhYiISFB9nY8\n", - "jb/jSwoLC72SQ5eVoGt+K/SDgoJc2gq99SOygkqlcpkS7sqVK4iLi/Nyj9rCwpNZQUFBl7ouOqKo\n", - "qAgxMTG+7obX6Oj681ZyH1Yi8/qt0O/oUYkVQectOtIgutJiMsBt3HOVFNwb0UZZoiNloCvFmvEE\n", - "VvrhDfxa6LvynDGbzUykovMWHXkRsbKY7K2bKjk5GZcuXfJKW0Lx1liwYkN2hTef/lhfc/JmKHjf\n", - "SwOBBAcHu9ToWDAleBPWE3d4yzsCcD8B+hpvrjclJyfj4sWLXmlLCBUVFV4LipiSksLEWLia8L25\n", - "p8dvhT7r1NXVQafTeaUtlUrFhEBzhbe8IwB27Kau8NYCKsC+YuTNsegoFIM3TTuuxt2b65CK0BeJ\n", - "q3j2XW0xGXBtQy4oKPDazc0KrqJHFhYWdqn1JsC1oPPmelNHu6RZmAC9EXjOhiL0RZKSkoL8/Pw2\n", - "33dFQRcfH4/CwsI235tM/7+9s4uNMivj+O+fLgXaBRehWSlT0mlpC4ilEJBGY0zMXhCjq3ux6kbj\n", - "xhiv/FiNMYoXxku9MGpivFB3N2vULaaKWbOKru6CJiZGwkeB0tIvylCyQP2iNHT46OPFvDMOw7SF\n", - "aTnnHd7zS0jmfZl5z9Mz5zzznOfjnClnAzouzFWIc/PmTWpra53JEYcA5VzB9bjEm1wyX3DdmQxe\n", - "W18kcRjQc/mQXRUCxYm4+5BdEueA8uzsrNO5E+e+yGazTn+EGxsbyxpGLqlqpV9uWTY7O+vUeojz\n", - "1sbZbNZpFtOKFSvIZrPO2rsfXOVi54lzcN11SvNchlEc3CqZTIaNGzc6ay+VSnk3jKpa6ZfzpyfR\n", - "lw65AGbpsjGTydDU1ORJov8Th8k9NjZGS0uLs/bmsqTj0Bfj4+M0Nzc7a6/c2PRBue/EtRt25cqV\n", - "ZQ0jl+OiqpV+uWXj+fPnnf5yz4Xryd3Y2MjFixfvuJfEuALktrwuPU3MZXpgnjgo+GXLlt2lZHzs\n", - "eBqHvii3y+Xs7Kyz/Pi4UNVKv1zgMKkDupw/PQ4D2sUZqKW0tbUxNDTktM1ylFqWrn3pAK2trYyM\n", - "jDhtsxylf7eLIyNLSaVSdxmJPuZuaZvXrl1zeqZAVSv9uORklw5oH9kqy5cvj4U/vXRAuzoNqJhV\n", - "q1bNe7iNLyYmJpyvvOJalTs8PExra6vTNpubmzl37pzTNu+FgYEBNm/e7Ky9qlb6EA8re+3atUxO\n", - "ThauBwcH6ejo8CiRP+rq6pieni5cJ7kvampq7thtc3R01Lmii0OGW57iueoj9haX2EIpLs6PLqbq\n", - "lX4caG9v5+zZs4Xrqakpp19intIJ7uMHccuWLZw5c6Zwnc1mnVUmF1P8t/syDNLpNKOjo4Xr27dv\n", - "e3e3+ZIhlUoxMTFRuJbk/QdpZmbGyx5dpUaiaypW+pKelnRa0m1JO+d5315JA5KGJH210vbmeX5h\n", - "Uk9PT3tRMKXbPPtSMsXtTk5OejmHdfXq1XdsbBWHlZjrtLw85XzIPij+DnysNgBaWlruiC3EYY74\n", - "WoW2t7czODjovN08i7H0TwJPAX+Z6w2SaoAfAHuBrcAzkrYsos276OjoKFiWx48fp6uraykff88s\n", - "5SA+dOhQRZ9rampifHwcgFOnTrFt27Ylk+l+WEoru9K+aGhoKJzYNDIyQjqdXpQclbDUK69K+6I4\n", - "mymTyXjJ6Cp2rSxFcL/SvihO8/a17XhxPYuP4H7FSt/MBszs7AJveycwbGbnzOwm0AN8qNI2y7F+\n", - "/fpC4Uc2m/V2YlZ+xXH9+vVFrzYqHdDFmRpxyNy5cuUK69atW9QzKu2LrVu30t/fD7gv2Csmr9wy\n", - "mQypVGpRz6q0Lzo7O+nr6ytc+3arDA0N0dbWtqhnVNoX27dv58SJE4D/fgA/xtmDngkbgOI8wgvR\n", - "vSUln8HjM5Mnv+I4cuQIO3bs8CJDPg/ZzLwGrPJL+b6+Pjo7O73IkA+i+s7uSqfTDA8PMzg4SHt7\n", - "uxcZamtrmZmZ4datW14V3Zo1a5icnPRaS1NfX8/U1BTZbNbrNin19fVcvXqVy5cv09DQ4LTteZW+\n", - "pNcknSzz74P3+HwnjrtNmzbR09PjTcEAbNiwgbGxMW7cuOH1fN7m5mZ6enrYtWuXNxnS6TT9/f3U\n", - "1NR4VTKpVIr9+/ezZ88ebzK0tLRw+vRpamtrvfZFY2Mjvb29dHd3e5Ohq6uLw4cPU1dX500GyB1c\n", - "f+DAAa99sXv3bg4ePOjlJDct1rcm6Q3gy2Z2tMz/dQPfNLO90fU+YNbMvl3mvf4jfoFAIFCFmNk9\n", - "WxRLtb6Zq8EjQJukZuAi8FHgmXJvvB+hA4FAIFAZi0nZfEpSBugGXpX0++h+o6RXAczsFvA54A9A\n", - "P7DfzM7M9cxAIBAIPFgW7d4JBAKBQPXgvSL3QRdvVQuSmiS9ERW8nZL0Bd8y+UZSjaRjkn7rWxaf\n", - "SHpMUq+kM5L6o1hZIpG0L5ojJyX9QpL7klpPSHpB0iVJJ4vuvTVKuDkr6Y+SHlvoOV6VvovirSri\n", - "JvAlM3s7OZfZZxPcF3meI+cWTPpy9PvA78xsC9AJJNJFGsUGPwPsNLN3ADXAx3zK5JgXyenKYr4G\n", - "vGZm7cCfo+t58W3pP/DirWrBzN40s+PR62vkJnbyNsOPkJQC3g/8hLkTBR56JL0FeI+ZvQC5OJmZ\n", - "/XeBjz2sXCVnHNVJegSoAybm/8jDg5n9Ffh3ye0ngZei1y8BH17oOb6VvpPirWojsmh2AH/3K4lX\n", - "vgt8BfC/d7Zf0sAVSS9KOirpx5L8Jrp7wsz+BXwHOE8uG/A/ZvYnv1J553Ezy++dfQl4fKEP+Fb6\n", - "SV+234WkR4Fe4LnI4k8ckj4AXDazYyTYyo94BNgJ/NDMdgLT3MMS/mFEUivwRaCZ3Cr4UUkf9ypU\n", - "jLBcVs6COtW30p8Aig9xbSJn7ScSScuAXwE/M7Pf+JbHI+8CnpQ0BrwMvE/STz3L5IsLwAUz+0d0\n", - "3UvuRyCJ7AL+Zmb/jNLBf01urCSZS5LeBiBpPXB5oQ/4VvqF4i1JteSKt17xLJMXlKvRfx7oN7Pv\n", - "+ZbHJ2b2dTNrMrM0uUDd62b2Sd9y+cDM3gQykvIb9zwBnPYokk8GgG5JK6P58gS5QH+SeQV4Nnr9\n", - "LLCgsehvxyFyQSlJ+eKtGuD5BBdvvRv4BNAn6Vh0b5+ZHfQoU1xIuhvw88DPI8NoBPiUZ3m8YGYn\n", - "ohXfEXKxnqPAj/xK5Q5JLwPvBdZFhbHfAL4F/FLSp4FzwEcWfE4ozgoEAoHk4Nu9EwgEAgGHBKUf\n", - "CAQCCSIo/UAgEEgQQekHAoFAgghKPxAIBBJEUPqBQCCQIILSDwQCgQQRlH4gEAgkiP8BYYh8Ti/Q\n", - "duUAAAAASUVORK5CYII=\n" - ], + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAEACAYAAABfxaZOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nduUAAAAASUVORK5CYII=\n", "text/plain": [ "" ] @@ -548,140 +145,7 @@ "outputs": [ { "data": { - "image/png": [ - "iVBORw0KGgoAAAANSUhEUgAAAhkAAADICAYAAABF5/MoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\n", - "AAALEgAACxIB0t1+/AAAHORJREFUeJzt3X+QXGWd7/H3J5PJLwIEDCQRsiuuICC4icoPF10bhK0U\n", - "xSKuW4oWGr2sy+IuUtRdFdjadaK1V/FeUUtXdpUfFbn3cpcSNguCSkTa1bouAW4CkZAFXAKJMpPw\n", - "IyG/M8l87x/P6UzPpKf7dM/0dPfM51V1qk+fc7r7yUnS85nv85znKCIwMzMzG2tTWt0AMzMzm5gc\n", - "MszMzKwpHDLMzMysKRwyzMzMrCkcMszMzKwpHDLMzMysKXKFDEldklZLujd73iNpU7ZttaQlzW2m\n", - "mZmZdZqpOY+7GlgHHJ49D+DGiLixKa0yMzOzjlezkiHpeOBC4GZApc1l62ZmZmaHyNNd8jXgM8BA\n", - "2bYArpL0uKRbJM1pSuvMzMysY1XtLpF0EbA5IlZLKpTtugn4Qrb+ReCrwOUVXu85y83MzCaQiMjd\n", - "k1GrkvEHwMWSngPuAM6T9L2I2BwZUjfKmVUa42UMls9//vMtb8NEWnw+fT7befH59Lls16VeVUNG\n", - "RFwfEQsj4gTgUuCnEfExSQvKDns/sLbuTzYzM7MJLe/VJZAGepZizFck/X72/DngirFumJmZmXW2\n", - "3CEjIopAMVv/aJPaYyMoFAqtbsKE4vM5tnw+x5bP59jxuWwtNdLHkvvNpWjm+5uZmdn4kUSM4cBP\n", - "M7MhPvEJ+O1vR95/113wne+MX3vMrH25kmFmdZHge9+Dj47QaXrssbBlC/i/vtnE40qGmTXdwMDI\n", - "+6b4W8XMMv46MLO6VatSOGSYWYm/Dsysbq5kmFke/jows7o5ZJhZHv46MLO6OWSYWR7+OjCzulUL\n", - "GV1d49cOM2tvDhlmVjcP/DSzPHJ9HUjqkrRa0r3Z86MlrZT0tKQHJM1pbjPNrJ1Uq2Qo9xX0ZjbR\n", - "5f2d42pgHYM3SLsWWBkRJwEPZs/NbIIrVTAOHBj5GFcyzKyk5teBpOOBC4GbSXdiBbgYWJ6tLwcu\n", - "aUrrzKyt7N+fHvfuHfkYhwwzK8nzdfA14DNAeYF0XkT0Zet9wLyxbpiZtZ/+/vS4b9/IxzhkmFlJ\n", - "1Vu9S7oI2BwRqyUVKh0TESFpxGFgPT09B9cLhYJvu2vWwUohw5UMs8mhWCxSLBYbfn3VG6RJ+m/A\n", - "R4H9wAzgCOBu4AygEBG9khYAD0XEyRVe7xukmU0gL70ExxwDn/0s3HBD5WMWL4Y1a3yDNLOJaExv\n", - "kBYR10fEwog4AbgU+GlEfBS4B1iaHbYUWNFog82sc7i7xMzqUe/XQel3ky8DF0h6Gjgve25mE1ye\n", - "gZ++hNXMSqqOySgXET8DfpatvwKc36xGmVl7ciXDzOrhrwMzy60UMqrNk+FKhpmVOGSYWW4OGWZW\n", - "D4cMM8utFDJKYzMqKV1V4qtLzMwhw8xyK4WLapWM0n1Nqh1jZpODQ4aZ5ZanklHaV+0YM5scHDLM\n", - "LLc8YzJKx5QezWzycsgws9xcyTCzejhkmFlueSoZDhlmVuKQYWa57d8P06a5kmFm+dQMGZJmSHpY\n", - "0hpJ6yR9KdveI2mTpNXZsqT5zTWzVurvh5kz81UyPCbDzGpOKx4ReySdGxG7JE0FfiHpXaT7mNwY\n", - "ETc2vZVm1hb6+2HGjNqVjJkzXckws5zdJRGxK1udBnQBr2bPPbef2SRSChm1Khm1goiZTQ65Qoak\n", - "KZLWAH3AQxHxZLbrKkmPS7pF0pymtdLM2kLeSsaMGe4uMbP8lYyBiFgEHA/8oaQCcBNwArAIeBH4\n", - "arMaaWbtodQVUmueDHeXmBnUcat3gIjYJuk+4B0RUSxtl3QzcG+l1/T09BxcLxQKFAqFRtppZm2g\n", - "VMnYvXvkYzwmw2ziKBaLFIvFhl+vqHEXI0lzgf0RsVXSTODHwDLgyYjozY65BjgjIj4y7LVR6/3N\n", - "rHN84xuwYgW89BKsXVv5mK4uWLwYvvUtOPvs8W2fmTWXJCIi93jMPJWMBcBySVNI3Su3R8SDkr4n\n", - "aRHpKpPngCsaarGZdYxaAz8HBtIyfborGWaW7xLWtcDbKmz/WFNaZGZtq9Z4iwMHYOpU6O52yDAz\n", - "z/hpZnUoXTkyUiVj//4UMqZO9dUlZuaQYWZ1qFXJKA8ZrmSYmUOGmeVWa0xGf7+7S8xskEOGmeVW\n", - "azIud5eYWTmHDDPLrVYlY//+VMVwd4mZgUOGmdUh75gMd5eYGThkmFkd6rm6xCHDzBwyzCw3j8kw\n", - "s3o4ZJhZbnnGZLi7xMxKHDLMLLdaYzJKl7C6u8TMoEbIkDRD0sOS1khaJ+lL2fajJa2U9LSkByTN\n", - "GZ/mmlkrlSoZAwNQ6d6H7i4xs3JVQ0ZE7AHOjYhFwFuBcyW9C7gWWBkRJwEPZs/NbIIrXaI6ZUrl\n", - "LpPSfneXmBnk6C6JiF3Z6jSgC3gVuBhYnm1fDlzSlNaZWVvp7x+cB2OkkOHuEjMrqRkyJE2RtAbo\n", - "Ax6KiCeBeRHRlx3SB8xrYhvNrE2Uxlx0dVUOEe4uMbNyeW71PgAsknQk8GNJ5w7bH5Iq9M4mPT09\n", - "B9cLhQKFQqHhxppZa+WtZLi7xGxiKBaLFIvFhl9fM2SURMQ2SfcBbwf6JM2PiF5JC4DNI72uPGSY\n", - "WWcrhYw8lYxduw7db2adZXhxYNmyZXW9vtbVJXNLV45ImglcAKwG7gGWZoctBVbU9alm1pHK701S\n", - "a0yGu0vMrFYlYwGwXNIUUiC5PSIelLQauFPS5cAG4IPNbaaZtYNaYzJ8q3czK1c1ZETEWuBtFba/\n", - "ApzfrEaZWXvKMybDd2E1sxLP+GlmudUzJsPdJWbmkGFmuXmeDDOrh0OGmeVW6g6pVcnwmAwzA4cM\n", - "M6tD+Q3QXMkws1ocMswsN4/JMLN6OGSYWW61xmT4ElYzK+eQYWa51VPJcMgwM4cMM8stz4yfpf3u\n", - "LjEzhwwzyyUi/11Y3V1iZuCQYWY5HTgAU6akxVeXmFkeNUOGpIWSHpL0pKRfSfp0tr1H0iZJq7Nl\n", - "SfOba2atUhqPAb66xMzyyXOr937gmohYI2k28JiklUAAN0bEjU1toZm1hfKQUa2SMWuWu0vMLKkZ\n", - "MiKiF+jN1ndIego4LtutJrbNzNpIqUoBte/C6kqGmUGdYzIkvQFYDPx7tukqSY9LukXSnDFum5m1\n", - "kb17Yfr0tF5rTMa0abBv3/i2z8zaT57uEgCyrpLvA1dnFY2bgC9ku78IfBW4fPjrenp6Dq4XCgUK\n", - "hcIommtmrbJvXwoPUH1MRne3Q4bZRFEsFikWiw2/PlfIkNQN3AX8z4hYARARm8v23wzcW+m15SHD\n", - "zDpXPZWM6dPT8WbW2YYXB5YtW1bX6/NcXSLgFmBdRHy9bPuCssPeD6yt65PNrKOUh4xaV5c4ZJgZ\n", - "5KtknANcBjwhaXW27Xrgw5IWka4yeQ64ojlNNLN24EqGmdUrz9Ulv6ByxeOHY98cM2tXecdkeOCn\n", - "mZV4xk8zy8WVDDOrl0OGmeWSZ0xGaZ6MUiUjYnzbaGbtxSHDzHLJW8no7k73N+nudpeJ2WTnkGFm\n", - "udQzJgM8LsPMHDLMLKfhlYyRuku6utK6x2WYmUOGmeVSHjK6uyvfm6S/f7Da4ZBhZg4ZZpZLeciY\n", - "Ns0hw8xqc8gws1zKx2SMNKhz377B28F7TIaZOWSYWS6uZJhZvfLcu2ShpIckPSnpV5I+nW0/WtJK\n", - "SU9LesC3ejeb2IaPyahVyXDIMLM8lYx+4JqIeAtwNvCXkk4BrgVWRsRJwIPZczOboHbvhpkz03qe\n", - "SsasWbBr1/i1z8zaT82QERG9EbEmW98BPAUcB1wMLM8OWw5c0qxGmlnrlYeMPJWMww6DnTvHr31m\n", - "1n7qGpMh6Q3AYuBhYF5E9GW7+oB5Y9oyM2sru3en6gRUv4S1FDJcyTCzPLd6B0DSbOAu4OqI2C7p\n", - "4L6ICEkV71LQ09NzcL1QKFAoFBptq5m10K5dQ7tLRqpklLpLXMkw63zFYpFisdjw63OFDEndpIBx\n", - "e0SsyDb3SZofEb2SFgCbK722PGSYWeca3l1Sq5LhkGHW+YYXB5YtW1bX6/NcXSLgFmBdRHy9bNc9\n", - "wNJsfSmwYvhrzWzi2LVrsLskz8BPhwwzy1PJOAe4DHhC0ups23XAl4E7JV0ObAA+2JQWmllbqHfg\n", - "56xZDhlmk13NkBERv2Dkisf5Y9scM2tXtS5hjTi0u2TLlvFto5m1F8/4aWa5lHeXVKpkHDgA0uBd\n", - "WN1dYmYOGWaWS61KRvl4DHDIMDOHDDPLqdaYjPLxGOAxGWbmkGFmOe3cmaoTkL+S4cm4zCY3hwwz\n", - "qykCtm+H2bPT8zyVDHeXmJlDhpnVtHdvGtBZqlRUmvGz/MoScMgwM4cMM8th+3Y4/PDB55VCRvmt\n", - "4MFjMszMIcPMctixY2jImDED9uwZesyePWl7icdkmJlDhpnVVD4eA/KHDFcyzCY3hwwzq6lSJWPv\n", - "3qHH7N3rkGFmQ+W5QdqtkvokrS3b1iNpk6TV2bKkuc00s1YaXsmYPj1VLiIGtw2vZJSCyIED49dO\n", - "M2sveSoZtwHDQ0QAN0bE4mz50dg3zczaxfCBn11dMHXq0LkyhoeMKVM8+NNssqsZMiLi58CrFXZp\n", - "7JtjZu1o61aYM2fotuHjMoaHDICjjkqvNbPJaTRjMq6S9LikWyTNqX24mXWqrVtTYCiXJ2QcfTS8\n", - "/HLz22dm7anmrd5HcBPwhWz9i8BXgcsrHdjT03NwvVAoUCgUGvxIM2uVV19trJLxutfBK680v31m\n", - "1hzFYpFisdjw6xsKGRGxubQu6Wbg3pGOLQ8ZZtaZtm6F179+6LbS4M+SPXuGTsYFrmSYdbrhxYFl\n", - "y5bV9fqGukskLSh7+n5g7UjHmlnncyXDzBpRs5Ih6Q7gPcBcSRuBzwMFSYtIV5k8B1zR1FaaWUs1\n", - "OvDTlQyzya1myIiID1fYfGsT2mJmberll1NgKDd8Qq7hk3FBqmS8+GLz22dm7ckzfppZTVu2wDHH\n", - "DN02Ywbs3j34fPduj8kws6EcMsyspkohY/hEWzt3pqnEy3lMhtnk5pBhZlXt2ZNu637EEUO3z559\n", - "aMgon3ocUiXDIcNs8nLIMLOqXnoJ5s4FDZvjd/gN0EaqZLi7xGzycsgws6p6e+HYYw/dnidkHHMM\n", - "9PU1t31m1r4cMsysqk2bYOHCQ7fnCRlz58KuXWkxs8nHIcPMqtq0CY4//tDthx0GO3YMPt+x49CQ\n", - "IcFxx8FvftPcNppZe3LIMLOqNm5svJIBKaBs2tS89plZ+3LIMLOqqlUy8oSMhQsdMswmq5ohQ9Kt\n", - "kvokrS3bdrSklZKelvSAb/VuNnGNFDLyXMIK6bUbNzavfWbWvvJUMm4Dlgzbdi2wMiJOAh7MnpvZ\n", - "BDRSd8ns2bB9e1qPcHeJmR2qZsiIiJ8Drw7bfDGwPFtfDlwyxu0yszYwMAC//W0avDncnDmwbVta\n", - "37EjTTPe3X3oca5kmE1ejY7JmBcRpavf+4B5Y9QeM2sjL7yQLkOdOfPQfXPmpLuzQuW7tJa86U3w\n", - "zDPNa6OZta9RD/yMiCDd8t3MJpj16+GUUyrvO/LIwUrG1q3peSUnnggbNkB/f1OaaGZtrOat3kfQ\n", - "J2l+RPRKWgBsHunAnp6eg+uFQoFCodDgR5rZeFu/Hk4+ufK+8krGtm0jVzKmT09jOn7965Hfy8za\n", - "U7FYpFgsNvz6RkPGPcBS4IbsccVIB5aHDDPrLOvXw+mnV9532GGwd2+qUFTrLoFUDakWWMysPQ0v\n", - "Dixbtqyu1+e5hPUO4P8Cb5a0UdIngC8DF0h6Gjgve25mE8xTT40cDKTBaka17hJI7/HUU81po5m1\n", - "r5qVjIj48Ai7zh/jtphZGxkYgCeegNNOG/mY0l1WS3dqHcnpp8N99419G82svXnGTzOr6NlnU3Vi\n", - "XpVrx+bNS3dZ7eurftyZZ8KqVWPfRjNrbw4ZZlbRI4+kcFBN3pBx4onwyiuwZcvYttHM2ptDhplV\n", - "tGpVvpDR21s7ZEyZAmec4WqG2WTjkGFmFf3sZ3DOOdWPmT8fXnwxBY3586sfe8458G//NnbtM7P2\n", - "55BhZofo7YXnn0/Vh2p+7/fS2I1nnknr1VxwATzwwNi10czan0OGmR3iJz+Bc8+FqTWuP3vzm1PF\n", - "Y/p0OPro6seeeSY891zqWjGzycEhw8wO8a//ChdeWPu4E09MgzkXLap9bHd3qmbce+/o22dmncEh\n", - "w8yG2LEjdWv8yZ/UPvbww+Fv/xauuirfe3/oQ/DP/zy69plZ51C6v1mT3lyKZr6/mY2922+HO+6A\n", - "++8f+/fetSvdNv7JJ+H1rx/79zez5pJERCjv8a5kmNkQ//iP8Od/3pz3njULPvIRuOmm5ry/mbWX\n", - "UVUyJG0AXgMOAP0Rceaw/a5kmHWQRx9N3ST/+Z+1B3026j/+A979bnjhBZgxozmfYWbNMd6VjAAK\n", - "EbF4eMAws87zd38H117bvIAB6YqUM86AW25p3meYWXsYbSXjOeAdEfHyCPtdyTDrEL/4BVx2Wao0\n", - "TJ/e3M9aswaWLEmfVe3urWbWXlpRyfiJpEclfXKU72VmLbJvH3zqU/ClLzU/YEC65PXCC+ELX2j+\n", - "Z5lZ64y2KHpORLwo6RhgpaT1EfHz8gN6enoOrhcKBQqFwig/0szG2t//Pfzu78Kll47fZ95wQwob\n", - "F18M73nP+H2umeVXLBYpFosNv37MLmGV9HlgR0R8tWybu0vM2twPfwh/9mfprqvjfVnp/ffDX/wF\n", - "PPwwLFgwvp9tZvUbt+4SSbMkHZ6tHwb8EbC20fczs/H32GOwdGmaIKsV81ZceCFccQX88R/Da6+N\n", - "/+ebWXONZkzGPODnktYADwM/iAjf/sisQzz2WPoh/93vwrve1bp2XH89nH02nH8+vPJK69phZmPP\n", - "M36aTUJ33ZW6Kb77Xbjkkla3BiLgc5+DH/wA7r4bTj651S0ys0rq7S5p4tXwZtZuduyA666De+6B\n", - "H/0I3v72VrcokeArX0lzaLz73Wn94x9P282sc3lacbNJIAL+5V/gtNNS0Fizpn0CRrnLL4eVK+Ef\n", - "/gHOOy/d48TMOpdDhtkENjAA992XxjwsWwbf+Q7cdhscdVSrWzayRYvS1Sbvf38KGh/5iMOGWafy\n", - "mAyzCai3N91J9dvfTjNq/vVfwwc/CFM67NeK7dvhm99My8knw5VXpnk1fM8Ts9aod0yGQ4bZBLFh\n", - "QxpnceedsHp1uiz0yitTFaPTxzbs2wcrVsA//dPgVTF/+qfw3vd6WnKz8eSQYTYJRMBzz8GqVeme\n", - "Iw88ANu2wQUXwAc+kO4LMnNmq1vZHH19aXzJ3XfDL3+Zxpm8973pMtwzzoDXva7VLTSbuBwyzCaY\n", - "XbvSjcSeegrWrUuDNletgu5uOOsseOc7U7h461s7rztktPbsSUHjJz9Jj48+Cscck8LG4sVwyilw\n", - "6qlwwgnQ1dXq1pp1PocMsw5z4EAaQ/H882l54YX0uGEDrF8PL74IJ544+APz9NNTuDjuuFa3vP0M\n", - "DKRAtmoVPPHEYDDbvBne9CY46aR0j5bhy5w5nd+lZDYeHDLMWmxgIF0munUrvPoqbNmSfsj19Q19\n", - "LK339qarPSr98Hvzm+GNb4SpntFmVHbuTOHj2WcHw1wpyD3/fOp+mjcvLfPnD66Xlrlz09/RUUel\n", - "QDJrlkOJTU4OGWZ1GhiA3bvTD6KdO1P3RK31115LIaLS8tpr6YfQnDlpUOKxx6Zl3rzKjwsW+GqJ\n", - "VopIf2elwNfXd+jy8sspML76avo7PnAg/f2WgkcpfBx+OMyePXQ57LBDt5W2z5yZ/u67ux1arDOM\n", - "a8iQtAT4OtAF3BwRNwzb75AxRorFIoVCodXNqCoi/cDevx/6+9PjSEu1/dX27dsHe/cOPuZdhh//\n", - "2mtFoMDOnalff8aM9KU/a1Z6HL4+/PkRR6QfKuXLkUcOPk62ykMn/PscS3v2DFaqysPHjh0jLzt3\n", - "Dn2+fXt6nz170v+bGTMGl4giRx1VGLKt0tLdPbhMmzb0eaPbu7oGl6lThz4faWnngDTZ/m0227hN\n", - "Ky6pC/gWcD7wG+ARSfdExFONvmcr3HwzfPKTrW5FHkWg0OI2tLfp0ysv06alxyOOGNz2618XWby4\n", - "cPC3ybwDJvfvT1dxbNsGGzc298/TSX75yyLvfGeh1c1oO6UwMHdu9eP270/htxQ6Hn+8yHHHFQ4+\n", - "Lw8kpWX37vQ6q6VIu393nn9+mul2IhrN71tnAs9GxAYASf8HeB/QUSFj/nz4nd9J69JgIh++Xmt/\n", - "nvXRvMfGjUPbOWVK+g1iypT8Sz3H1/vepd94OuXqhn374B3vaHUrJo516wb/fdrobduWLkPuBKUK\n", - "5oEDg8v+/UOfV1tKrx/NY7V9GzbAwoX1vWf5n63SY95teY8/9dRR/RW0tdGEjOOA8t/lNgFnja45\n", - "4++ii9LS7np60mJjY9s2uOaaVrdi4vD5HFs+n2PH352t1fCYDEkfAJZExCez55cBZ0XEVWXHeECG\n", - "mZnZBDJet3r/DbCw7PlCUjWjoYaYmZnZxDKaHvRHgRMlvUHSNOBDwD1j0ywzMzPrdA1XMiJiv6S/\n", - "An5MuoT1lk67ssTMzMyap6mTcZmZmdnk1ZQLDiVdJekpSb+SdEPZ9uskPSNpvaQ/asZnT1SS/quk\n", - "AUlHl23z+ayTpP+e/dt8XNLdko4s2+fzWSdJS7Lz9Yykz7W6PZ1G0kJJD0l6Mvu+/HS2/WhJKyU9\n", - "LekBSXNa3dZOIalL0mpJ92bPfS4bJGmOpO9n35nrJJ1V7/kc85Ah6VzgYuCtEXEa8D+y7aeSxm2c\n", - "CiwBvi2pQ2ZVaC1JC4ELgOfLtvl8NuYB4C0R8fvA08B14PPZiLIJ+ZaQztuHJZ3S2lZ1nH7gmoh4\n", - "C3A28JfZObwWWBkRJwEPZs8tn6uBdUCpTO9z2bhvAPdHxCnAW4H11Hk+m/EleiXwpYjoB4iILdn2\n", - "9wF3RER/NoHXs6QJvay2G4HPDtvm89mAiFgZEaXpdh4Gjs/WfT7rd3BCvuz/e2lCPsspInojYk22\n", - "voM0meFxpF/UlmeHLQcuaU0LO4uk44ELgZuB0tWNPpcNyKq8746IWyGNw4yIbdR5PpsRMk4E/lDS\n", - "v0sqSirNq/h6hl7iuon0n8mqkPQ+YFNEPDFsl8/n6P0X4P5s3eezfpUm5PM5a5CkNwCLSeF3XkT0\n", - "Zbv6gHktalan+RrwGaBs3k6fywadAGyRdJuk/yfpu5IOo87z2dDVJZJWAvMr7Pqb7D2PioizJZ0B\n", - "3Am8cYS38qhTap7P64Dy8QHV5h7x+aTq+bw+Ikr9tH8D7IuI/13lrXw+q/P5GSOSZgN3AVdHxHaV\n", - "3XEsIsITG9Ym6SJgc0SsllSodIzPZV2mAm8D/ioiHpH0dYZ1jeQ5nw2FjIi4YKR9kq4E7s6OeyQb\n", - "rDiXQyfvOj7bNumNdD4lnUZKk49nXzrHA49JOgufzxFV+/cJIOnjpJLqe8s2+3zWr+aEfFabpG5S\n", - "wLg9IlZkm/skzY+IXkkLgM2ta2HH+APgYkkXAjOAIyTdjs9lozaRquiPZM+/T/qlt7ee89mM7pIV\n", - "wHkAkk4CpkXES6SJui6VNE3SCaRulVVN+PwJIyJ+FRHzIuKEiDiB9Jf+tqxU5fPZAElLSOXU90XE\n", - "nrJdPp/184R8o6T028MtwLqI+HrZrnuApdn6UtL3qlUREddHxMLsu/JS4KcR8VF8LhsSEb3Axuzn\n", - "OKQ7rj8J3Esd53M004qP5FbgVklrgX3Ax7IGr5N0J2nU737gU+FJOup18Hz5fDbsm8A0YGVWHfpl\n", - "RHzK57N+npBvTJwDXAY8IWl1tu064MvAnZIuBzYAH2xN8zpa6f+vz2XjrgL+V/ZLxK+BT5D+r+c+\n", - "n56My8zMzJrC8wCYmZlZUzhkmJmZWVM4ZJiZmVlTOGSYmZlZUzhkmJmZWVM4ZJiZmVlTOGSYmZlZ\n", - "U/x/8QPGQJ9lUrYAAAAASUVORK5CYII=\n" - ], + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhkAAADICAYAAABF5/MoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nU/x/8QPGQJ9lUrYAAAAASUVORK5CYII=\n", "text/plain": [ "" ] @@ -856,150 +320,7 @@ "outputs": [ { "data": { - "image/png": [ - "iVBORw0KGgoAAAANSUhEUgAAAXYAAAEACAYAAACnJV25AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\n", - "AAALEgAACxIB0t1+/AAAHxdJREFUeJzt3XmYVNWZx/HvyyIIKMgiIIsQBSKKKxIRI+2SiIlCYozG\n", - "zIxGjfGJjMk4WUazIJk8Gk2iMTGT5YkGTVSMS1zQqCjSEREhJKBCQwABpYk2CDayr+/8cartsu2G\n", - "qu7qOrdu/z7Pcx9quVX3paF/dercc841d0dERNKjVewCRESksBTsIiIpo2AXEUkZBbuISMoo2EVE\n", - "UkbBLiKSMjkFu5n1M7PpZrbQzBaY2dcyj3c1s2fNbImZTTWzLlmvuc7MlprZYjP7ZHP9BURE5IMs\n", - "l3HsZtYL6OXu882sE/B34DPApcA77v5jM/sf4CB3v9bMhgL3AScCfYDngMHuvqe5/iIiIhLk1GJ3\n", - "97fdfX7m9iZgESGwxwJ3Z3a7mxD2AOOAye6+091XAsuAEQWsW0REGpB3H7uZDQCOA2YDPd29KvNU\n", - "FdAzc/sQoDLrZZWEDwIREWlmeQV7phvmYeDr7r4x+zkPfTp769fR2gUiIkXQJtcdzawtIdT/6O6P\n", - "Zh6uMrNe7v62mfUG1mQeXw30y3p538xj2e+noBcRaQR3t709n+uoGAPuBCrc/baspx4HLsncvgR4\n", - "NOvxL5jZfmY2EBgEzKmnuMRv119/ffQaVKfqVJ2qsWbLRa4t9lHAvwOvmtm8zGPXATcBD5jZ5cBK\n", - "4IJMYFeY2QNABbALuMpzrUhERJokp2B39xdpuHV/ZgOvuRG4sZF1iYhII2nm6T6UlZXFLiEnqrOw\n", - "VGdhlUKdpVBjrnKaoNQsBzZT74yISJ7MDC/EyVMRESkdCnYRkRKxJ8dFWRTsIiIl4sUXc9tPwS4i\n", - "UiJmzcptPwW7iEiJyDXYNSpGRKQEuEOvXrBmjUbFiIikwooV0CbHtQIU7CIiJWDWLBg5Mrd9Fewi\n", - "IiVAwS4ikjL5BLtOnoqIJNzmzXDwwbBuHey/v06eioiUvLlzYdgwaN8+t/0V7CIiCZdPNwwo2EVE\n", - "Eu/ll/MLdvWxi4gkWM3EpLlzoV8/LdsrIlLyVqyAtm1DqOcqarCrwS4isnf59q9D5GBfsSLm0UVE\n", - "km/WLDjppPxeEzXYX3op5tFFRJJvxgw45ZT8XqNgFxFJqOpqWL4cjj8+v9cp2EVEEmrmTBgxIpw8\n", - "zUfUYF+6FN57L2YFIiLJNWMGfPzj+b8uarAffzzMmROzAhGR5CrJYD/5ZHXHiIjUZ+tWmD8//xEx\n", - "kIBgnzkzZgUiIsk0Zw4cdRR07Jj/a6MG+8iRYQ2E3btjViEikjyN7YaByMF+8MFhq6iIWYWISPKU\n", - "bLCD+tlFROratSv0ZuQ7MalG9GAfNUrBLiKS7ZVXoG9f6Natca+PHuxqsYuIfFBTumEgAcE+dCis\n", - "XQtr1sSuREQkGV54ocSDvVWrME5TrXYRkbCc+YsvlniwQzhBMGNG7CpEROKrqAhj1/v3b/x7JCLY\n", - "R48OXz1ERFq66dPh9NOb9h45BbuZ/d7MqszstazHJppZpZnNy2xnZz13nZktNbPFZvbJfb3/iSfC\n", - "okWwcWPj/hIiImnx/PNFCnZgEjCmzmMO3Orux2W2pwDMbChwITA085pfmdlej9O+PZxwgpYXEJGW\n", - "bc8e+Otf4bTTmvY+OQW7u88A3q3nqfqulD0OmOzuO919JbAMGLGvY6g7RkRauldegR494JBDmvY+\n", - "Te1jv9rMXjGzO82sS+axQ4DKrH0qgT77eqNTT1Wwi0jLNn1601vrAG2a8NpfA/+buf1D4Bbg8gb2\n", - "9foenDhx4vu3P/axMubPL2PrVth//yZUJSJSop5/Hi655IOPlZeXU15entf7mHu9mfvhHc0GAFPc\n", - "fdjenjOzawHc/abMc08D17v77Dqv8brHHjkSbryxMJ9YIiKlZNeusITAsmWhO6YhZoa719cN/r5G\n", - "d8WYWe+su58FakbMPA58wcz2M7OBwCAgp+skqTtGRFqqv/8dDj1076Geq5y6YsxsMjAa6G5mq4Dr\n", - "gTIzO5bQzbICuBLA3SvM7AGgAtgFXPWhpnkDTj0Vbr01/7+EiEipK1T/OuTRFVNo9XXFbNgQVjRb\n", - "tw722y9KWSIiUXzykzB+PIwbt/f9mrUrpjl07gyDBsHcubErEREpnh07wvrro0cX5v0SFeygfnYR\n", - "aXnmzIHBg6FLl33vm4vEBfvo0WHmlYhIS/Hcc01fRiBbIoN95szw1UREpCV45hk466zCvV/igr1r\n", - "19DPPienAZIiIqXt3Xdh4cLGX9+0PokLdoAzzoBp02JXISLS/KZNC6Herl3h3lPBLiIS0dSpYahj\n", - "ISVqHHuNzZuhZ0+oqgpXEhERSSN3GDAAnn4ajjgit9eU3Dj2Gh07hvXZdbk8EUmzJUvCGuwf/Whh\n", - "3zeRwQ7qjhGR9KsZDWN7bX/nT8EuIhJJc/SvQ0L72AF27oTu3WH58rCUpYhImmzfHlZyXLkyDPPO\n", - "Vcn2sQO0bRuGAE2fHrsSEZHCe+mlcMI0n1DPVWKDHdQdIyLp1VzdMKBgFxGJ4umnC7uMQLZEB/uw\n", - "YWG67Ztvxq5ERKRwKitDrp10UvO8f6KDvVWr8FXlmWdiVyIiUjhPPglnnw1tcrqGXf4SHewAY8aE\n", - "rywiImnxxBPw6U833/sndrhjjaoqGDIE1q4NI2VERErZli3Qq1f+wxxrlPRwxxo9e8Jhh8GsWbEr\n", - "ERFpuunT4bjjmmeYY43EBzuEvih1x4hIGjz5JJxzTvMeoySCXf3sIpIG7qF/XcFOGBK0YgW8/Xbs\n", - "SkREGu+118K5wkKv5lhXSQR7mzZhstLUqbErERFpvJrWeqFXc6yrJIId1B0jIqWvGN0wUALDHWus\n", - "WhXOJFdVQevWzViYiEgzWLsWDj8c1qxp2vVNUzHcsUa/fmHs59y5sSsREcnf44+HtWEKedHqhpRM\n", - "sEPojnnqqdhViIjk789/hvPOK86xSirYzz0XpkyJXYWISH42bAjXcP7Up4pzvJIK9lGjwjTcysrY\n", - "lYiI5O4vf4FTT4UDDyzO8Uoq2Nu0CZ94arWLSCkpZjcMlFiwA4wdG05CiIiUgi1bwhycsWOLd8yS\n", - "C/azzoIXX4SNG2NXIiKyb1OnwvDh0L178Y5ZcsF+4IFw8smahSoipaHY3TBQgsEO6o4RkdKwY0eY\n", - "bfqZzxT3uDkFu5n93syqzOy1rMe6mtmzZrbEzKaaWZes564zs6VmttjMCn4d7nPPDUtf7tpV6HcW\n", - "ESmc8vJwoaA+fYp73Fxb7JOAMXUeuxZ41t0HA9My9zGzocCFwNDMa35lZgX9ZtC/f5iJqotviEiS\n", - "PfRQ8bthIMdgd/cZwLt1Hh4L3J25fTdQ82VjHDDZ3Xe6+0pgGTCi6aXWObi6Y0QkwXbsCP3rF15Y\n", - "/GM3pSXd092rMrergJ6Z24cA2VOIKoGCfxEZOxYeeywsXC8ikjRTp4Z11/v3L/6xC9JFklmmcW8R\n", - "W/D4Pf542L4dFiwo9DuLiDTd5Mlw0UVxjt2mCa+tMrNe7v62mfUG1mQeXw30y9qvb+axD5k4ceL7\n", - "t8vKyigrK8v54GZw/vmhD2vYsDwrFxFpRlu2hAEeP/tZ09+rvLyc8vLyvF6T83rsZjYAmOLuwzL3\n", - "fwysc/ebzexaoIu7X5s5eXofoV+9D/AccHjdxdfzXY+9Pi+/DJddBhUVTXobEZGC+tOf4Pe/h2ee\n", - "Kfx7F2w9djObDLwEDDGzVWZ2KXAT8AkzWwKcnrmPu1cADwAVwFPAVU1O8AaMGBFmoCrYRSRJ7r8f\n", - "vvCFeMcvmSsoNeSaa+Cgg2DChAIUJSLSRNXVcOih8MYb0KXLvvfPV6quoNSQ88+HBx+MXYWISPDI\n", - "I3D66c0T6rkq+WAfORLWr4fFi2NXIiISdzRMjZIP9lat4HOfC6NjRERiWr0a/vY3OOecuHWUfLBD\n", - "7bBHEZGY7rkn5FGHDnHrSEWwjxoFVVWwdGnsSkSkpXKHu+6CL30pdiUpCfbWreHznw99WyIiMcye\n", - "Dbt3h+tFxJaKYAf4t3+De+/V2jEiEkdNa932OhCxOFIT7CNGwJ49MHdu7EpEpKXZujUMu7744tiV\n", - "BKkJdrPaVruISDE9+mi4rmnfvrErCVIT7BCC/f77dWUlESmuu+6CSy+NXUWtVAX7oEFhKu+0abEr\n", - "EZGWorIydAGPGxe7klqpCnYIrfZ77oldhYi0FHfeGa6StP/+sSupVfKLgNVVVRUuHrt6NXTsWPC3\n", - "FxF5365dMGAAPPVU8a4L0SIWAaurZ8+wfsxjj8WuRETSbsqUEOxJu9hP6oId4D/+A/7wh9hViEja\n", - "/frX8NWvxq7iw1LXFQNhTGnfvjBvXpwLyYpI+i1dGpYzWbUK2rUr3nFbZFcMhJMYF10EkybFrkRE\n", - "0uq3vw0zTYsZ6rlKZYsdYP78MPxo+fKwloyISKFs3Rp6A15+GQ47rLjHbrEtdoBjj4Xu3TWmXUQK\n", - "78EH4YQTih/quUptsAN8+ctwxx2xqxCRNHGHX/4SrroqdiUNS21XDISLyg4YEE5y9OjRrIcSkRbi\n", - "xRfD8gGLF8fp5m3RXTEQLiY7dqxmoopI4dxyC1xzTbLP3aW6xQ7wwgthnOmCBclYJ1lESteyZWEC\n", - "5MqV8Wa2t/gWO8DHPx7WaX/hhdiViEip+/nP4StfSf5yJalvsUM40fHXv4Yz2SIijbF+PRx+OCxc\n", - "CL17x6sjlxZ7iwj2jRvDcr6vvAL9+hXlkCKSMjfdFE6Y3nVX3DoU7Fm+9jU44AC44YaiHVJEUmL7\n", - "dvjIR8IqjkcfHbcW9bFnGT8+jGnfti12JSJSau6+OwR67FDPVYsJ9iFDwmzUBx6IXYmIlJKdO+FH\n", - "P4IJE2JXkrsWE+wAV18Nt98eZo6JiOTinnvC0gEjR8auJHctKtjPPhvWrYPZs2NXIiKlYNeucF7u\n", - "+9+PXUl+WlSwt24dTqLeckvsSkSkFNx/P/TpA6NHx64kPy1mVEyNTZtg4EB46SUYNKjohxeRErF7\n", - "Nxx5ZJgHc+aZsauppVEx9ejUKSwx8NOfxq5ERJLswQfhoIPgjDNiV5K/FtdiB1i7NoySqaiAXr2i\n", - "lCAiCbZzJwwdGq5pmqTWOqjF3qAePeCLXwzrPoiI1HXHHaHLNmmhnqsmt9jNbCXwHrAb2OnuI8ys\n", - "K/An4FBgJXCBu1fXeV20FjvAihUwfHi4dF7nztHKEJGE2bQJBg+GJ56A44+PXc2HFavF7kCZux/n\n", - "7iMyj10LPOvug4FpmfuJMnAgjBkTLkgrIlLjttvCKJgkhnquCtFiXwEMd/d1WY8tBka7e5WZ9QLK\n", - "3f2jdV4XtcUO8OqrcNZZ8Prr0KFD1FJEJAHWroUjjghzXZJ6PdNittifM7O5ZnZF5rGe7l6VuV0F\n", - "9CzAcQru6KNh1Cj41a9iVyIiSXDDDXDRRckN9VwVosXe293fMrMewLPA1cDj7n5Q1j7r3b1rnddF\n", - "b7FDWFv59NPDlVEOOCB2NSISy+LF4cI8CxZAz0Q2RYNcWuxtmnoQd38r8+daM3sEGAFUmVkvd3/b\n", - "zHoDa+p77cSJE9+/XVZWRllZWVPLyduRR4ZxqrffDt/5TtEPLyIJ4B7Wkvre95IX6uXl5ZSXl+f1\n", - "mia12M2sA9Da3TeaWUdgKvAD4ExgnbvfbGbXAl3c/do6r01Eix3gn/+EU04JrXaNkBFpef78Z7j+\n", - "epg3D9o0ubnbvJr9QhtmNhB4JHO3DXCvu/8oM9zxAaA/CR3uWNeXvgQDBkDWlwgRaQG2bAmTke66\n", - "CyJ0GuRNV1DKw/LlMGJEaL136xa7GhEplgkTYMmSsOBXKVCw52n8eGjbNoxjFZH0W7YMTjoJ5s+H\n", - "vn1jV5MbBXue1q4NX8lmzgwzz0QkvfbsCSPixo6F//7v2NXkTmvF5KlHD/j2t+Fb34pdiYg0t9/8\n", - "Jlyk+utfj11J4anFXse2baHVfscd4dNcRNJn5cqwVtSMGWGmaSlRi70R2reHm2+Gb3wjLLQvIuni\n", - "Dl/5SvgdL7VQz5WCvR7nnw8dO4bhTyKSLnfeGa59/M1vxq6k+agrpgH/+Ee4+PXChdC9e+xqRKQQ\n", - "aiYjPv88DBsWu5rG0aiYJrrmGqiuhkmTYlciIk21fTuMHAlXXBEuj1mqFOxNtHFjWEvmD38ojRlp\n", - "ItKw//ovePNNePhhsL3GYrIVZRGwNDvggLA42JVXhrXb27WLXZGINMYTT8Ajj4S1YEo51HOlk6f7\n", - "MG5cGP54002xKxGRxli5Er78ZbjnHujadZ+7p4K6YnKwalW4TNZzz8Exx8SuRkRytXkznHwyXHpp\n", - "6IpJA/WxF9Ddd8Mtt8CcOWGsu4gkmztccEEYujxpUnq6YDRBqYAuvhgGDYLvfz92JSKSixtuCN+2\n", - "f/Ob9IR6rtRiz8M774SumHvv1SgZkSR7+OGwBszf/ga9e8euprDUYi+w7t3hd78LF+XYsCF2NSJS\n", - "n+nTwzj1KVPSF+q5Uou9Ef7zP+Gtt+Chh1reVzyRJPvHP2DMGHjggfR+q1aLvZnccgtUVsJPfxq7\n", - "EhGpsXQpnHMO/Pa36Q31XKnF3khvvhkupXf//fpPJBLbsmVw5pnwve+FMetpphZ7M+rfP0x4+OIX\n", - "YfXq2NWItFyLFoXG1Xe+k/5Qz5WCvQnOPDP0t593XrjSuYgU16uvwhlnwI03hjXWJVBXTBO5wyWX\n", - "wHvvhSFWrVvHrkikZZg5Ez73Ofj5z+HCC2NXUzzqiikCs3AZvffeC1OWU/BZJZJ4994Ln/1suBhO\n", - "Swr1XKnFXiDV1WEB/8suK60rnouUEnf4wQ/CEh9TpsBRR8WuqPi0bG8RdekCf/kLjBoVbl92WeyK\n", - "RNKlujqcHK2shJdfhp49Y1eUXOqKKaD+/WHaNJgwQVddEimkOXPCCqu9e0N5uUJ9X9RiL7DBg0O4\n", - "n346tGoVTqyKSOPs3g0/+xn8+MdhMa/zzotdUWlQsDeDIUNCuJ9xBuzaBZdfHrsikdKzYEHoetlv\n", - "P5g9GwYOjF1R6VBXTDP56EfDYkQ33AATJ2q0jEiutm0L3ZmnnRbOVZWXK9TzpWBvRoMHw6xZ4XqL\n", - "l18OO3fGrkgkufbsgT/+MXzjXbgQ5s8Pk45aKaXypuGORbBpUxhru3MnTJ4M3brFrkgkOdzh6afD\n", - "kgDt28NPfhKGDkv9NEEpITp1gsceg6OPhhNOCP2FIi3drl1w331w3HHw7W+HYH/pJYV6IajFXmSP\n", - "PAJXXhn6EMeP13ru0vJUVoYZo3fcAYceGkL9U5/S70KudDHrhHr9dfj856FXr7B2dL9+sSsSaV7V\n", - "1eFc0733hm+sF14YzjsNHx67stKjYE+wHTvg5pvhF78II2euuEItlubmDuvXQ1VVuH5tzVZdHbYN\n", - "G8KaP5s3h23LFti+Pfxbbd8eug727Ambe/j3atUqbG3bhmF57dqFbf/9oUOHsB1wQO3WuXOYmdyl\n", - "Cxx0EHTtGrZu3cLr0sI9LKf7/PPw+ONhpmhZGVxwQRiL3qFD7ApLl4K9BCxYEIZ0tW8frsx04omx\n", - "KypN7rBmDaxcCW+8Ef6srKzd/vWv8HynTnDwwdCjR7iGbbduIWA7dw7bgQdCx45h69Ah/Lu0axdC\n", - "u23b2iCvOaZ7mESza1f4ANixA7Zurd02b4aNG2u3996r/SBZvz5s69aFP9u1CzV17x7qq6nz4IPD\n", - "TMvsP3v0CDUlxbvvwrx5YZszJwxR7NgRRo+GT386XK6uU6fYVaZD1GA3szHAbUBr4A53v7nO8wr2\n", - "jN27Q5/jhAmhVXPjjaHvUT7IPVxrdsmS2m3ZstC1tXx5aCUPHBh+doceGpZ46NMH+vaFQw4JoZjU\n", - "VrF7CP533oG1a2u3NWvCN4y1a8OfNbfXrg1B2aPHBz8EunWr3bp2rf1m0Llz+MbQqVN+Hwju4dvK\n", - "hg3hA+itt8K2enX4uS9bFi5JV10NxxwTpv0PHw6nngoDBjTbj6tFixbsZtYa+CdwJrAa+Btwkbsv\n", - "ytpHwV7Hpk3hOqq33x6+rn7zm2FMb0uzezesWAEVFWFbtChsixeHFvSQIWGOwKBBYTvsMPjIR0Jr\n", - "u6XYsyeE6Zo1H/wgWLcufDisWxda0TXfDqqrw/+vjRtDF1L79rXfRtq2re1WgjAst6YLqmb/zp3D\n", - "B0Xv3rXbYYeFn//hh4cPUV2LoDhiBvtI4Hp3H5O5fy2Au9+UtY+CvQHvvAP/939hO+UUuPrq8JU2\n", - "bRM13GHVqtAdlb0tXhxaoEOHwpFHwhFHhG3IkBAu0njuIbC3bQvhvW1bCPKabiX30KKv2Tp1Ch8A\n", - "khwxg/184Cx3vyJz/9+Bj7n71Vn7KNj3YfPmsErk734X+mYvuQQuvji0TkuJO7z9dphNuHBhbYBX\n", - "VITgOOqoEOBHHRW2oUPVHyvSkJjrseeU2BMnTnz/dllZGWVlZc1UTmnq2DFcU3X8+HBSatIkOOmk\n", - "0Jo999ywnXhi+CqdBHv2hBOVNV0nixaF8F64MHzbGDoUhg0L/bAXXxzCXC1wkb0rLy+nvLw8r9c0\n", - "V4v9JGBiVlfMdcCe7BOoarE3zu7dYdTBlCnw5JPhpOHw4eECHyeeGMJy4MDm6+/cujWE9xtv1J60\n", - "fP312hOZXbqEBdBquk9qulMOPljDOUUKIWZXTBvCydMzgH8Bc9DJ02ZRXR0WGps5M7TqKyrCyIlB\n", - "g8LEp759w8iQbt1qh/R17BiCv02b0JKuGae9fXs4wVYzprtmzHfNVlkZHu/bN5wsO+yw2hOXgweH\n", - "k2gHHBD7JyKSbrGHO55N7XDHO939R3WeV7A3k02bwhC07HHc69eHUN6wIfTd14y93rPngxNrOnUK\n", - "re7OncMwuZ49a7e+fUPLO20ncUVKiSYoiYikjFZ3FBFpgRTsIiIpo2AXEUkZBbuISMoo2EVEUkbB\n", - "LiKSMgp2EZGUUbCLiKSMgl1EJGUU7CIiKaNgFxFJGQW7iEjKKNhFRFJGwS4ikjIKdhGRlFGwi4ik\n", - "jIJdRCRlFOwiIimjYBcRSRkFu4hIyijYRURSRsEuIpIyCnYRkZRRsIuIpIyCXUQkZRTsIiIpo2AX\n", - "EUkZBbuISMoo2EVEUkbBLiKSMgp2EZGUUbCLiKSMgl1EJGUU7CIiKaNgFxFJmUYHu5lNNLNKM5uX\n", - "2c7Oeu46M1tqZovN7JOFKVVERHLRlBa7A7e6+3GZ7SkAMxsKXAgMBcYAvzKzkv1mUF5eHruEnKjO\n", - "wlKdhVUKdZZCjblqauBaPY+NAya7+053XwksA0Y08TjRlMo/tuosLNVZWKVQZynUmKumBvvVZvaK\n", - "md1pZl0yjx0CVGbtUwn0aeJxREQkR3sNdjN71sxeq2cbC/waGAgcC7wF3LKXt/LClSwiIntj7k3P\n", - "XDMbAExx92Fmdi2Au9+Uee5p4Hp3n13nNQp7EZFGcPf6usHf16axb2xmvd39rczdzwKvZW4/Dtxn\n", - "ZrcSumAGAXPyLUxERBqn0cEO3GxmxxK6WVYAVwK4e4WZPQBUALuAq7wQXwtERCQnBemKERGR5Ig6\n", - "vryeSU5jYtazL2b2DTPbY2ZdY9dSHzP7YWaU0nwzm2Zm/WLXVB8z+4mZLcrU+mcz6xy7pvqY2efN\n", - "bKGZ7Taz42PXk83MxmQmAC41s/+JXU99zOz3ZlZlZq/te+94zKyfmU3P/FsvMLOvxa6pPmbW3sxm\n", - "Z36/K8zsRw3tG3viUN1JTk9HrqdBmZD8BPBG7Fr24sfufoy7Hws8Clwfu6AGTAWOdPdjgCXAdZHr\n", - "achrhPNHL8QuJJuZtQZ+SZgAOBS4yMyOiFtVvSYRaky6ncA17n4kcBIwPok/T3ffBpyW+f0+GjjN\n", - "zE6pb9/YwQ71T3JKoluBb8cuYm/cfWPW3U7AO7Fq2Rt3f9bd92Tuzgb6xqynIe6+2N2XxK6jHiOA\n", - "Ze6+0t13AvcTJgYmirvPAN6NXce+uPvb7j4/c3sTsIgwHydx3H1L5uZ+QGtgfX37JSHY65vklChm\n", - "Ng6odPdXY9eyL2Z2g5m9CVwC3BS7nhxcBvwldhElpg+wKuu+JgEWSGbo9nGEBkfimFkrM5sPVAHT\n", - "3b2ivv2aMiom10KeBXrV89R3CZOc/jdz/4eESU6XN3dN9dlHndcB2YuZRfuWsZc6v+PuU9z9u8B3\n", - "M/MJfgZcWtQCM/ZVZ2af7wI73P2+ohaXJZc6E0gjHpqBmXUCHgK+nmm5J07mm+6xmfNSz5hZmbuX\n", - "192v2YPd3T+Ry35mdgcQ7RepoTrN7CjCDNtXzAxCt8HfzWyEu68pYolA7j9P4D4itoT3VaeZfQn4\n", - "FHBGUQpqQB4/zyRZDWSfGO/HB5fxkDyZWVvgYeAed380dj374u4bzOxJYDhQXvf52KNiemfdzZ7k\n", - "lBjuvsDde7r7QHcfSPgFOj5GqO+LmQ3KujsOmBerlr3JjH76FjAuc0KoFCTpXNBcYJCZDTCz/Qir\n", - "qT4euaaSZaHFdidQ4e63xa6nIWbWvaa72sz2JwzmqPd3POo4djP7A2GtmfcnObl7VbSCcmBmy4Hh\n", - "7l7vSYuYzOwhYAiwG3gd+GpCP4CWEk7+1PwMZ7n7VRFLqpeZfRb4BdAd2ADMc/ez9/6q4shc/+A2\n", - "wgm0O929waFvsZjZZGA00A1YA0xw90lxq/qwzMiSF4BXqe3mui5po/TMbBhwN6FB3gr4o7v/pN59\n", - "NUFJRCRdkjAqRkRECkjBLiKSMgp2EZGUUbCLiKSMgl1EJGUU7CIiKaNgFxFJGQW7iEjK/D+UMwo2\n", - "QcnZWQAAAABJRU5ErkJggg==\n" - ], + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXYAAAEACAYAAACnJV25AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQcnZWQAAAABJRU5ErkJggg==\n", "text/plain": [ "" ] @@ -1078,184 +399,7 @@ "outputs": [ { "data": { - "image/png": [ - "iVBORw0KGgoAAAANSUhEUgAAAXgAAAEACAYAAAC57G0KAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\n", - "AAALEgAACxIB0t1+/AAAIABJREFUeJzt3XmYFdW19/HvApwQBBWuGHEgiorGCQUxmNAKKJrEMTdK\n", - "xAGjMTjGmDjmSpt7jXGKRknUOBBnVERF4wRix+FVQEVQEecRHJhEQFGG9f6xq+HQdvc5fbrq1DnV\n", - "v8/z1MMZqk7tortX79619trm7oiISPa0SrsBIiKSDAV4EZGMUoAXEckoBXgRkYxSgBcRySgFeBGR\n", - "jMob4M1skJnNMLO3zOysRvbrZWbLzOzQph4rIiLxazTAm1lrYAQwCNgOGGxmPRrY72Lg0aYeKyIi\n", - "ycjXg+8NvO3u77v7UmAUcGA9+50CjAZmF3GsiIgkIF+A3wT4KOf5x9FrK5nZJoTAfU30Uu3U2LzH\n", - "iohIcvIF+ELqGFwJnO2h5oFFW6HHiohIQtrkeX8msGnO800JPfFcuwKjzAygE7CfmS0t8FjMTL8I\n", - "RESK4O6Wb4cGN8IvgHeALYA1gZeBHo3sPxI4pCnHhiZk1/Dhw9NuQqIq+fref9/9gAPcN9rI/cIL\n", - "3efN++4+ude3ZIn7jTe6b7ONe69e7s8/X7q2JqGSv3aFyPr1RbGz0Rje6BCNuy8DTgYeA6YDd7n7\n", - "62Z2gpmdUMyxjf62ESkBd7jpJthtN+jTB957D849F9Zfv/Hj1loLjj0Wpk+H3/4WDjgAzjsPvvmm\n", - "NO0WaapCJjp5zrYCwN2vc/frAMzsQDObamZTgB2BL3KOvQb4BlgIHBxnw0WKsWgRHHwwjBgBEybA\n", - "OefAOus07TNatYJf/hKmToVXX4Xdd4cPPkimvSLNEUce/Hh338nddwGOAf6Z854DVe6+i7v3jq/Z\n", - "laOqqirtJiSqkq7vk0+gXz/o1Amefx522CH/MY1dX5cucP/9cPTR0LcvvPRSfG0thUr62hUj69dX\n", - "CPNGFvwwsz2A4e4+KHp+NoC7/6WR/a9w9z7R8/eA3dx9biPn8MbaIBKH6dNh//3h+OPDcIw1fmuq\n", - "ycaMgRNOgJtvDucRSZqZ5b3J2uw8+OhEB5nZ68AjwKk5bzkw3sxeMLPjC2u2SLxmzIABA+BPfwpj\n", - "5nEHd4BDDoGxY2HoUHj44fg/X6QYceTB4+73u3sP4GfArTlv9Y2GbvYDTjKzHxXXTJHivPMODBwI\n", - "F10ERx2V7Ln22AMeeCAM2TzxRLLnEilEHHnwK7n702bWxsw2dPe57v5J9PpsM7uPUL7g6brHVVdX\n", - "r3xcVVWlsTOJxUcfhZ77eeeFoFsKffrA6NHw85+H8fm+fUtzXsm+mpoaampqmnRMvjH4NsAbQH9g\n", - "FjAJGJyb7mhmWwLvurubWU/gHnff0szaAq3dfaGZrQs8Dlzg7o/XOYfG4CV2CxeG4DpkCJx5ZunP\n", - "/+ijcMwx8MwzsNVWpT+/ZF8hY/CN9uDdfZmZ1eaytwZurM2Dj96/DjgUOCqavboIODw6vAswJprh\n", - "2ga4vW5wF0nC8uUweHDoTf/hD+m0YdAgqK6Gn/4Unnsuf469SBIa7cGXpAHqwUvMTj8dpk0Lveg1\n", - "1lBbJJviyKLJu2hH7kQnM3vRzPYu9FiRuI0cGbJYRo8uj4B62WVhItXpp6fdEmmJ8o3BtyaMwQ8g\n", - "3HCdzHfH4Nd198XR4x2A+9x9q0KOjY5RD15iMWUK7LMP/Oc/sN12abdmlQULQlmE6mo44oi0WyNZ\n", - "EUcPPu+iHbXBPdIOmFPosSJxmT8/ZK6MGFFewR2gQwe4995Qv+bVV9NujbQkSU500oIfUhIrVoQc\n", - "95/9DA47LO3W1G/HHeHyy+HQQ+HLL9NujbQU+fLgC57oBNwfTWS61cy2bUojlAcvzXHFFTBnTugl\n", - "l7Ojjgppk8OGwW23JTOjVrIriTz4PkB1Ti2ac4AV7n5xI8e8Qxie6V7IsRqDl+aYPBl+8hOYNAm2\n", - "2CLt1uT31VfQuzf8/vchT16kWHGMwb8AdDezLcxsTeAwYGydk2xpUbJ7NNGJqLhY3mNFmuPLL0O+\n", - "+9//XhnBHaBtWxg1KgT4GTPSbo1kXWITnRo6NrlLkZZm2DDo3x/++7/TbknT/OAHcOGFcPjhoWzx\n", - "2mun3SLJqmYv+EFIgVxOCOKtCAt81NKCH5KI228PaZFXXJF2S4rz61/DlluGOjkiSYkjD34PYLq7\n", - "LzCzQYRx99x68Lu6+7xGzqExeGmS99+HXr1g3DjYeee0W1O8uXNhp53C5KyBA9NujVSaUuXBP+fu\n", - "C6KnE4GuddvRhDaLNGrZslBA7KyzKju4A2y4IfzrX6GG/NwGl8QRKV4sefA5fgXkLnegBT8kVhdf\n", - "HBa//t3v0m5JPAYMCGPxxx8fFgMXiVMsefAAZrYXcCyQWwG7r7t/YmadgXFmNsPdVQ9eijJ5Mlx1\n", - "Fbz4Ylj4OisuvDAs3F3bmxepT2p58Ga2IzAGGOTubzfwWcOBRe5+eZ3XNQYveS1eDD17wv/9X+Vl\n", - "zRTi1Vdhr71CVs2WW6bdGqkEpcqD34wQ3IfkBncza2tm7aPH6wL7AK80/TJEQt54nz7ZDO4QUif/\n", - "+Ec48shwn0EkDo0GeHdfBtTmsk8H7qrNg6/NhQfOB9YHrolKBk+KXu8CPG1mLxNuvj6kBT+kGA89\n", - "FOqpX3112i1J1imnQLt28Oc/p90SyYo48uCfJNx8bQ18BRwX7fMucDawNrBW7bEiTfHpp+EG5C23\n", - "wHrrpd2aZLVqFcbh//GPMFQj0lyJ5cGrHrw014oVoc7MbrvB//5v2q0pnfvuC0NSU6Zk/5eaFC/t\n", - "PHjVg5dmGTEi1Hk///y0W1JaBx8cSjCcemr+fUUak2QevOrBS9FeeSX02m+/vTyW3iu1K64Ii3WP\n", - "GpV2S6SSJZkHX/CxyoOXXIsXh4U7Lrus5aYMrrsu3HknDBoUcuS7dUu7RZK2ssqDb8KxGoOX1Rx/\n", - "PCxZEm6stvRFMa68MgT6Z55pmX/JSMNSzYMv5FiRukaNCotm/+MfCu4Ap50GnTuHHHmRpoqjHnxu\n", - "HjzAUnfvrXrw0lTvvBNywR99FNq3T7s15cEsVJvcZZcw03XQoLRbJJUkjjz4y4A3gR7Abe7eO+dY\n", - "1YOXgnz9dViQevhw2HXXtFtTXjp3hjvuCEv8ffhh2q2RShJHHnxnYHPgIGB+bq0Z1YOXQv3qV2G9\n", - "0jvu0NBMQy69FEaPhqeeChU1pWUrVR78bHd/AVjaUDsKbbC0TDfdFFICr79ewb0xv/89fO974V+R\n", - "QsSdB1+X6sFLoyZPDot33HtvqMMiDTMLpQwefTRkGInkE1sefAMKqgcvLdMnn8Ahh4See48eabem\n", - "MnToAPffH264br11qLAp0pB8AX4msGnO800JvfiCuPsn0b+zzew+wpCPFvwQliwJU/JPOAEOOijt\n", - "1lSW7beHG28MN6UnTYJNND+8RUhiolMbwk3W/sAsYBL1FAyL9q0GFtbeZDWztkBrd18Y1YN/HLig\n", - "bslg3WRtedzh6KNDkL/rLo27F+uii2DMGKipCTNfpWUp5CZrowE++pD9gCtZlct+UW4evJl1IWTX\n", - "rEdIo1wIbAf8F2ECFIS/FG5394vq+XwF+Bbmj3+EceNgwgQFpuZwD6mT8+aFCpRt8v09LpkSRxYN\n", - "5M+D70gYtlkL+D9338zdF6kevNTnuutCr/3BBxXcm8ss3L/45hs46SQt2i3flVgevOrBS11jx4Yx\n", - "96efhq22Srs12bFwIfz4x+GG9f/8T9qtkVJJOw9e9eBlpcceg+OOC0FewT1e7dvDww+H1Mkrrki7\n", - "NVJO8o3a1ZcHv3uBn92cYyVDJkyAIUNCel+vXmm3Jps23hieeAL69YM11wxDNiJJ5sFr3EX4z39C\n", - "bfd77oG+ffPvL8XbbLPwy7RfP2jdGn7zm7RbJGlLMg++4GOVB59NY8eGGjN33QX6kpZGt24hyA8c\n", - "CF98EWYJKw01G8otD76gY3WTNZtuuQXOPDNky2hYpvRmzoR994X99oNLLlGQz6JU8+DdfVF9x9bz\n", - "+QrwGbJiBVx4IdxwQ7ixuu22abeo5Zo3D37yk9Crv+EGaNs27RZJnGIJ8ElTgM+OxYvDDNWZM8MM\n", - "y403TrtF8vXXIXvpjTfCTe6uXdNukcQllolOZjbIzGaY2VtmdlYD+1wVvT/VzHbJef19M5tmZlPM\n", - "bFLTL0EqxRtvhJuo7duHqfMK7uVhnXXgttvgF7+A3r3hySfTbpGUUqMBPpqsNAIYRCg/MNjMetTZ\n", - "Z39gK3fvDvyasIpTLQeq3H2XOis9SUa4hz//99wzTGK66SYtRlFuzML9kH/9K6Srnn02fPtt2q2S\n", - "Umj2RCfgAOBmAHefCHQ0s41y3tftnYyaOTPMnhwxIqRDDhumm3nlbJ994OWX4bXX4Ic/hKlT026R\n", - "JC2OBT8a20cLfmTQsmVw1VWw887wgx/AxImw3XZpt0oK0blzSF8dNiwE/DPOCKUOJJvimujUUL9t\n", - "T3eflW/BD+XBVwZ3+Pe/4bzzYMMNQ00ZZclUHrMwP+GAA8LQTY8eoYbNscfCGmuk3TppSBJ58H2A\n", - "ancfFD0/B1jh7hfn7HMtUOPuo6LnM4B+7v5Znc8aDizKXZQ7el1ZNGXOPZT3veACWLAgpEEecICG\n", - "Y7Ji8mQ491x47z04/3w4/PBQ7kDKWxxZNC8A3c1sCzNbEzgMGFtnn7HAUdEJ+wBfuPtnZtbWzNpH\n", - "r68L7AO8UsR1SEq++iqsHLTDDuFP+WHDwrjtgQcquGdJr17hF/g//wk33xzy5v/8Z5gzJ+2WSXM1\n", - "OkTj7svM7GTgMVZNVno9d6KTuz9sZvub2dvAYmBodHgXYIyFSFC74Mfj3z2LlJPly8PQy623hlz2\n", - "vn3hyiuhf38F9azbe++wTZ0avuZbbRVKTBx1FOy/P6y9dtotlKaKY8EPotc9Z3+04EfQ1DGzNMyZ\n", - "A6NHh9WBunSB3/423DSdPh0eeggGDGg4uFfC9TVHlq+voWvbaScYORI+/DD8tXb11bDRRmHt3Btv\n", - "hPffL2kzi5blr12hEsuDL+TYlqDcvsmWLIEXXwwrAf3mN2EB5y23DPnru+0GL7wQUunOOKOwyUrl\n", - "dn1xy/L15bu29daDoUPD5Kh334Wf/zwM5ey+O2y+ORx5JPztb+Evvi+/LE2bmyLLX7tC5cuiWZkH\n", - "D2BmtXnwuQXDVsuDN7OOUX2abgUcKzH7+mv49NOwzZoFH3wQelzvvgszZsDHH0P37rDrrtCzZ5jG\n", - "vvPOWs9TGrfhhmGS1JAh4ab7m2+GwP7SS3DnnfDKK7D++iGraqutwi+AzTcPpRG6dAmdhXbtNMxX\n", - "anEs+NFQHvz3Cji2pGbNClPqi1Ffok/ua7WP3VdtAG+/DY88suq9FStWbbXPly8PueXLl696vHRp\n", - "2L79NmxLloTt669DzZfFi2HRotBzWrAA5s+HuXPD8V26rPqh2mKLcNOsf/+QDtetm1LhpHnMYJtt\n", - "wlZrxQr46KPw8/XWW6FjMWVK+Jmr7XAsXQobbBC2Dh3CXwjt24e1edddNxRDW3vtMBN6rbVCJs8a\n", - "a4StTZuwtW69amvVavXNbPXt7bdDwbva57Vtz/0l09AvnGJ+EW28cRmmDbt7gxtwKHB9zvMhwNV1\n", - "9nkQ6JvzfDywayHHRq+7Nm3atGlr+tZY/Hb3WBb8qLtP12ifNQo4Nm8ep4iIFCexPPgCjxURkYQk\n", - "lgff0LFJXoyIiKyS+oIfIiKSjEImOomISAVSgBcRySgFeBGRjFKAFxHJqKIDvJndZGafmVmDJYAb\n", - "WoxbRESS15we/EhCIbF65VmMW0REElZ0gI+W3pvfyC75FuMWEZEEJTkGX18Rsq4Jnk9ERHIkXSS2\n", - "bp2Z78yqMjPNtBIRKUK+Wl5JBvj6ipDNrG/HLM+mra6uprq6Ou1mJEbXVznmzw9rArz5ZijpO3Zs\n", - "NeutV81HH4Vyvu3ahZWbOneGTp1CWd/114eOHVeV9m3XLmxt28I666xe3re2xG+bNqtK/LZunV4N\n", - "+Cx97epjBfzHJhngxwInA6PqFCETkYR9/jlMnBi2KVNg2rSwbsC228LWW4dFX7bcMizPuNlmoZa5\n", - "1lzNnqIDvJndCfQDOpnZR8BwQongfItxi0jM5s4Ny+k9+WTYZs+G3r3D8nrHHx/WWd1887AwRq3q\n", - "aujXL7UmSwkUHeDdfXAB+5xc7OdnRVVVVdpNSJSuLz3vvgv33BMWRp86FaqqYO+94cQTYYcdVg/m\n", - "9Snna4tD1q+vEKlXkzQzT7sNIpVi7ly47Ta44w547z049FA48MAQ3DXE0rKYWd6brArwImXOHZ55\n", - "Bq69Fv79b/jpT+HII8M6u1osveVSgBepYMuWwb33wuWXwxdfwEknhcC+wQZpt0zKQSEBXr//RcrM\n", - "smVhCOZPfwrZLX/8Y+i15xtTF6mrWd8yZjbIzGZEBcXOquf9Tmb2qJm9bGavmtkxzTmfSJa5w+jR\n", - "sP32cMMNYXv6aTjgAAV3KU7RQzRm1hp4AxhAmMA0GRicu+6qmVUDa7n7OWbWKdp/I3dflrOPhmik\n", - "xZs0CX73O1i0CC69FAYMSG+CkFSGQoZomtMv6A287e7vu/tSYBRwYJ19PgHWix6vB8zNDe4iLd28\n", - "eSFP/aCD4Nhj4cUXYeBABXeJR3MCfH3FxDaps8/1wPZmNguYCpzWjPOJZIZ7SHfcbruQ3vj66yHA\n", - "t26ddsskS5pzk7WQcZVzgZfdvcrMtgTGmdlO7r6wGecVqWiffAK//jV8+CGMHRtmnIokoTkBvm4x\n", - "sU0JvfhcPwQuBHD3d8zsPWAb4IXcnXILAlVVVWkGmmTW3XfDKaeEAH/vvaE4l0ghampqqKmpadIx\n", - "zbnJ2oZw07Q/MAuYxHdvsv4VWODuF0SLfbwI7Oju83L20U1WybzFi+HUU8OEpdtug1690m6RVLpE\n", - "b7JGN0tPBh4DpgN3ufvrZnaCmZ0Q7fZnYDczmwqMB87MDe4iLcG0abDbbrB8ebiJquAupaKZrCIJ\n", - "uuUWOOMMuOIKGDIk7dZIlmgmq0hKvvkm5LWPGwc1NWHykkipKcCLxOzzz+Hgg8PKSJMnQ4cOabdI\n", - "WipNgBaJ0dSpIe1xwAAYM0bBXdKlHrxITMaOheOOgxEj4Be/SLs1Is3owecrNBbtU2VmU6JCYzVF\n", - "t1KkzF11FQwbFuq1K7hLuSgqi6bAQmMdgWeBfd39YzPr5O5z6vksZdFIxVq+PNxMHT8+BPcttki7\n", - "RdJSJJlFs7LQWHSi2kJjr+fs80vgXnf/GKC+4C5SyZYsgSOOgPnz4dlnoWPHtFsksrpih2gKKTTW\n", - "HdjAzJ40sxfM7MgizyVSdubPh333hTXWgEceUXCX8lRsgC9kTGUNoCewP7Av8D9m1r3I84mUjZkz\n", - "4cc/hp49w8pLa62VdotE6lfsEE0hhcY+Aua4+9fA12b2FLAT8FbdD1OxMakUb70F++wDJ5wAZ52l\n", - "uu1SOiUrNlZgobFtgRGE3vtawETgMHefXuezdJNVKsKUKfCTn4S1Uo87Lu3WSEuX2E1Wd19mZrWF\n", - "xloDN9YWGovev87dZ5jZo8A0YAVwfd3gLlIpnnoKfv5zuPZaOOSQtFsjUhgVGxPJ4+GH4Zhjwnj7\n", - "gAFpt0YkSHpNVpHMGzUKhg4Ns1QV3KXSKMCLNOC660Kp3/HjoU+ftFsj0nSqRSNSj0sugWuugf/8\n", - "B7baKu3WiBRHAV4khzucdx7cdx88/TR07Zp2i0SKl2ixsWi/Xma2zMyUeyBlbcUKOPFEePxxBXfJ\n", - "hqJ68FGxsRHkFBszs7G5efA5+10MPApoSoiUraVL4eijYdYsmDAB1lsv7RaJNF+xPfiVxcbcfSlQ\n", - "W2ysrlOA0cDsIs8jkrivvoIDD4SFC0NdGQV3yYrEio2Z2SaEoH9N9JKS3aXszJ8PAweG5fXGjIF1\n", - "1km7RSLxSbLY2JXA2dEsJkNDNFJmaouG9ekDI0eGypAiWZJksbFdgVEWqjF1AvYzs6XuPrbuh6nY\n", - "mJTa66/DfvuFm6p/+IOKhkn5K6tiY3X2Hwk86O5j6nlPpQqkpJ59NtSTuewyOFKrFEiFSrXYWDGf\n", - "K5K0e+6Bk06CW28NC3aIZJmKjUmL4A6XXgpXXw0PPQQ77ZR2i0SaJ8k1WUUqxrffhl77pEnw3HOa\n", - "wCQthwK8ZNrs2aGOe4cO8Mwz0L592i0SKR1Vk5TMmjYNdt8d+vaF++9XcJeWRz14yaTbboPTT4e/\n", - "/Q1++cu0WyOSjmb14PMVHDOzI8xsqplNM7NnzWzH5pxPJJ9vv4VTToELLoAnnlBwl5at6B58gQXH\n", - "3gV+7O4LzGwQ8E9ASydIIt5+GwYPhk02gcmToWPHtFskkq7m9ODzFhxz9+fcfUH0dCKg/AVJxO23\n", - "wx57hIqQ992n4C4CzRuDr6/g2O6N7P8r4OFmnE/kO+bMCSmQ06bBuHGw885pt0ikfDSnB1/w7CQz\n", - "2ws4FmhwYRCRpho7FnbcMeS1v/SSgrtIXc3pwRdScIzoxur1wCB3n1/fB6nYmDTFxx/DaaeFXvtd\n", - "d8GPfpR2i0SSV7JiY1BYwTEz2wyYAAxx9+cb+ByVKpCCfPst/P3vcOGFYVjmnHNg7bXTbpVIOhIt\n", - "VVBgwbHzgfWBa6KywUvdvXex55SWyR0eeCCU9e3ePVSD3GabtFslUv5UbEzK2pNPwvnnh5WXLr9c\n", - "FSBFaqnYmFQkdxg/PgzFzJwJw4eH/PbWrdNumUhlUYCXsrFkSbhp+te/wooV8PvfwxFHQBt9l4oU\n", - "RT86krrXXoMbbgj1Y3r2hEsugX320TJ6Is2lAC+pePdduPtuGDUqlPQdOjTUa+/WLe2WiWSHbrJK\n", - "SXzzDTz/PDzySFhR6fPPw7qogwfDnntqfF2kqQq5yZpoNclon6ui96ea2S7NOV8laurEhErT0PXN\n", - "nRuCeXU19O8PG24YxtRbtw7DMZ9+CtdeC/36lXdwz/LXL8vXBtm/vkIUHeBzqkkOArYDBptZjzr7\n", - "7A9s5e7dgV8D1zSjrRUpy99k7vDIIzVMnBgWsT7vPPjZz2DzzcNQy6WXhp77b38bZp9OnhwyY/r0\n", - "gVYVstRMlr9+Wb42yP71FaI5Y/Arq0kCmFltNcnccsEHADcDuPtEM+toZhu5+2fNOK8kaNkyWLgQ\n", - "vvgibPPmhTHyOXPgs89g1iz45BP48EP44IMwu3TCBNh667ANHRrqw3z/+5UTxEWyKulqkvXt0xVI\n", - "JcBPmRKGDZKWe0vhqadCr7Wh9xt7XPu89nHttmLFdx8vXx4eL18etmXLVt+WLg3BeOnS0KtesiRs\n", - "X38NX30FixfDokVhn3btYP31w9axI3TuHLb/+q+wBN7GG8Nmm4We+pVXhmEYESk/zalFcyihgNjx\n", - "0fMhwO7ufkrOPg8Cf3H3Z6Pn44Ez3f2lnH10h1VEpAhJzmQtpJpk3X26Rq8V3EARESlOc0ZJXwC6\n", - "m9kWZrYmcBgwts4+Y4GjAMysD/CFxt9FREoj0WqS7v6wme1vZm8Di4GhsbRaRETySn2ik4iIJEOJ\n", - "bCIiGaUALyKSUQrwIiIZpQAvIpJRsQR4M9vGzKbkbAvM7FQz28DMxpnZm2b2uJl1jON8IiKSX+xZ\n", - "NGbWijCZqTdwCjDH3S+Jqk2u7+5nx3pCERGpVxJDNAMIRcg+IqfYWPTvQQmcT0RE6pFEgD8cuDN6\n", - "nFs58jNgowTOJyIi9Yh1iCYqWTAT2M7dZ5vZfHdfP+f9ee6+QZ1jNNNKRKQIia7oVI/9gBfdfXb0\n", - "/DMz6wJgZhsDn9d3kLtndhs+fHjqbdD16fpa2rW1hOsrRNwBfjCrhmcgFBs7Onp8NHB/zOcTEZEG\n", - "xBbgzWxdwg3WMTkv/wUYaGZvAntHz0VSY2bf2USyqjn14Ffj7ouBTnVem0cI+i1WVVVV2k1IVGVe\n", - "X+6ft40H+Mq8vsJk+dog+9dXiNhuskaTmG4Atif8BA0F3gLuAjYH3gd+4e5f1DnO42qDSD6hx756\n", - "gM/9/quvR6/vTylHZoaX8Cbr34CH3b0HsCMwAzgbGOfuWwNPRM9FypznbCKVK5YevJl1AKa4+/fr\n", - "vD4D6Ofutdk0Ne6+bZ191IOXxNQ/xp6vB9/w+yLlopQ9+G7AbDMbaWYvmdn10U1XTXSSMqAeubRM\n", - "cQX4NkBP4B/u3pOwPN9qwzFRN10/YSIiJRJXFs3HwMfuPjl6Pho4B/jUzLq4+6eNTXSqrq5e+biq\n", - "qkp3v6Vs6SaspKWmpoaampomHRNnFs1TwHHu/qaZVQNto7fmuvvFZnY20NHrVJPUGLwkqb4x9eaM\n", - "wWuMXspFIWPwcQb4nQhpkmsC7xDSJFsDdwOboTRJSYECvGRVSQN8sRTgJUkK8JJVpc6DFxGRMhJb\n", - "qQIzex/4ElgOLHX33ma2AXlmsoqISDLi7ME7UOXuu7h77+g1zWQVEUlJ3EM0dceDtGSfiEhK4u7B\n", - "jzezF8zs+Og1zWSVklIpYJFVYhuDB/q6+ydm1hkYF9WhWcndvaHl+TTRSYpVSK2ZZM4hUlqpTnRa\n", - "7UPNhgOLgOMJ4/K1M1mfVLExiVMhaZDNTZNsyvH1t291+n6XOJQsTdLM2ppZ++jxusA+wCtoyT4R\n", - "VOxM0hLXEM1GwH1Rb6UNcLu7P25mLwB3m9mviNIkYzqfSGw0BCNZpZmsUtHiGKKJc4insPatTt//\n", - "UoySz2Q1s9ZmNsXMHoyeb2Bm48zsTTN7PFrWT6Ro2ciS0ZCNlEbcefCnAdNZ9Z2riU6SAAVIkULE\n", - "FuDNrCuwP6GiZG3XShOdRERSEmcP/grgD8CKnNc00UlEJCWxZNGY2U+Bz919iplV1bePJjqJiBQv\n", - "tYlOZvZn4EhgGbA2sB4wBuiFJjpJjMohayb/RKm6is/CEWlIybJo3P1cd9/U3bsBhwMT3P1INNFJ\n", - "WiTdBJbykNSCH7Xf2X8BBprZm8De0XMRESkBTXSSilIZQzTxTZQSaUgpa9GsbWYTzexlM5tuZhdF\n", - "r2uik0ge2Zi8JeUorjH4JcBe7r4zsCOwl5ntiSY6SQvQ/OCsMXtJRmxj8O7+VfRwTaA1MB9NdJIW\n", - "Id4ArR7+AnP0AAAGsElEQVS9xCXOmaytzOxlwoSmJ939NTTRSaQI6tFLPGJb0cndVwA7m1kH4DEz\n", - "26vO+w1OdBJpiHqwIsWLc8k+ANx9gZn9G9gV+MzMuuRMdPq8vmM0k1UaF+8SfCKVKM2ZrJ2AZe7+\n", - "hZmtAzwGXADsC8x194vN7Gygo7ufXedYpUlKg8oxLbKc6s1Ly1VImmRcPfiNgZvNrBVhXP9Wd3/C\n", - "zKagFZ2kCTQkIxIfTXSSsqIeu3rwUpiSr+gkIiLlI66ZrJua2ZNm9pqZvWpmp0avayarSIzq5shr\n", - "SEsaE1cPfilwurtvD/QBTjKzHmgmq0gClCcvhYmrVMGn7v5y9HgR8DqwCZrJKiKSmtjH4M1sC2AX\n", - "YCKaySoikppYJzqZWTvgXuA0d1+YOz6oJftapvrGiJUVUjr6/8+O1CY6AZjZGsBDwCPufmX02gy0\n", - "ZF+LVl/aY2Nfb6VJNr3efFP/P/Xzlg2lrAdvwI3A9NrgHtGSfdIoZYSIJCeuUgV7Ak8B01jVXTgH\n", - "mATcDWxGNJPV3b+oc6x68BmWrwepHrt68FKcQnrwmskqiVKAV4CXZJRyiOYmM/vMzF7JeU2TnERi\n", - "EPcQlobFWo640iRHAoPqvKZJTiKxaHxiU9MDtiZKtRRxTXR6mrBEXy5NchIpCQVsqV/sC37k0CQn\n", - "kRRo2EVqJRngV8q3XJ8mOrUsCkBJq3vTVrIg7YlOWwAPuvsO0fO8k5yi/ZRFkyH1B+/yykrRc2XV\n", - "ZEHa9eA1yanF0phwJVFWTXbFNdHpTqAf0Ikw3n4+8AB5JjlFx6oHnyHKa8/C89Xp57M8aaKTJE5D\n", - "Mtl/rp/P8pT2EE1tIwaZ2Qwze8vMzkr6fJIGDcmIlKNEA7yZtQZGECZBbQcMjlZ6ajGaetc7bvmW\n", - "eCvmfY3XZkVN2g1IVNo/e+Ug6R58b+Btd3/f3ZcCo4ADEz5nWSmPb7J8PezV3/9uAK97vHrs2VBT\n", - "0F7N6SCkqTx+9tKVdB78JsBHOc8/BnZP+Jxla+TIm3nggfErn7dqBSeeeDQDBgwoaTvy/xAqj1py\n", - "5ft+0PdLuUo6wKuLl+P//b/JPPDAbau9NmDAHrEG+MJrkaw8IrZzi0D+78HvVsds+H1pnkSzaMys\n", - "D1Dt7oOi5+cAK9z94px99NUUESlCqmmSZtYGeAPoD8wiLAAy2N1fT+ykIiICJDxE4+7LzOxk4DGg\n", - "NXCjgruISGmkPtFJRESSkfhEp0KZ2Slm9rqZvWpmF+c/ovKY2RlmtsLMNki7LXExs0ujr9tUMxtj\n", - "Zh3SblMcsjxBz8w2NbMnzey16Oft1LTblAQza21mU8zswbTbEicz62hmo6Ofu+nRvc56lUWAN7O9\n", - "CAuE7OjuPwAuS7lJsTOzTYGBwAdptyVmjwPbu/tOwJuExdYrWguYoLcUON3dtwf6ACdl7PpqnQZM\n", - "J3vZfH8DHnb3HsCOQIPD3mUR4IFhwEXRZCjcfXbK7UnCX4Ez025E3Nx9nLuviJ5OBLqm2Z6YZHqC\n", - "nrt/6u4vR48XEQLE99JtVbzMrCuwP3ADGcoFjv5C/pG73wThPqe7L2ho/3IJ8N2BH5vZ82ZWY2a7\n", - "pd2gOJnZgcDH7j4t7bYk7Fjg4bQbEYP6JuhtklJbEhWt47AL4ZdzllwB/AFYkW/HCtMNmG1mI83s\n", - "JTO73szaNrRzSVZ0AjCzcUCXet46L2rH+u7ex8x6EcoMf79UbYtDnus7B9gnd/eSNComjVzbue7+\n", - "YLTPecC37n5HSRuXjKz9SV8vM2sHjAZOi3rymWBmPwU+d/cpZlaVdnti1gboCZzs7pPN7ErgbEKJ\n", - "9np3Lgl3H9jQe2Y2DBgT7Tc5uhG5obvPLVX7mquh6zOzHxB+606NZu11BV40s97u/nkJm1i0xr52\n", - "AGZ2DOHP4f4laVDyZgKb5jzflNCLzwwzWwO4F7jN3bO2GM8PgQPMbH9gbWA9M7vF3Y9KuV1x+Jgw\n", - "GjA5ej6aEODrVS5DNPcDewOY2dbAmpUU3Bvj7q+6+0bu3s3duxG+QD0rJbjnY2aDCH8KH+juS9Ju\n", - "T0xeALqb2RZmtiZwGGGFskyw0NO4EZju7lem3Z64ufu57r5p9PN2ODAhI8Edd/8U+CiKkwADgNca\n", - "2r9kPfg8bgJuMrNXgG+BTHwxGpC1P/+vBtYExkV/oTzn7iem26TmaQET9PoCQ4BpZjYleu0cd380\n", - "xTYlKWs/c6cAt0edj3eAoQ3tqIlOIiIZVS5DNCIiEjMFeBGRjFKAFxHJKAV4EZGMUoAXEckoBXgR\n", - "kYxSgBcRySgFeBGRjPr/SYBJHv8M/N8AAAAASUVORK5CYII=\n" - ], + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXgAAAEACAYAAAC57G0KAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nkYxSgBcRySgFeBGRjPr/SYBJHv8M/N8AAAAASUVORK5CYII=\n", "text/plain": [ "" ] diff --git a/tips/dataset_CIFAR-10.py b/tips/dataset_CIFAR-10.py new file mode 100644 index 0000000..5f4a25a --- /dev/null +++ b/tips/dataset_CIFAR-10.py @@ -0,0 +1,54 @@ + +import os + +import torch as t +import torchvision as tv +import torchvision.transforms as transforms +from torchvision.transforms import ToPILImage +show = ToPILImage() # 可以把Tensor转成Image，方便可视化 + +# 第一次运行程序torchvision会自动下载CIFAR-10数据集， +# 大约100M，需花费一定的时间， +# 如果已经下载有CIFAR-10，可通过root参数指定 + +# 定义对数据的预处理 +transform = transforms.Compose([ + transforms.ToTensor(), # 转为Tensor + transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5)), # 归一化 + ]) + +# set data storage dir & check whether do download +data_path = "../data/" +p = os.path.join(data_path, "cifar-10-batches-py") +do_download = True +if os.path.isdir(p): + do_download = False + +# 训练集 +trainset = tv.datasets.CIFAR10( + root=data_path, + train=True, + download=do_download, + transform=transform) + +trainloader = t.utils.data.DataLoader( + trainset, + batch_size=4, + shuffle=True, + num_workers=2) + +# 测试集 +testset = tv.datasets.CIFAR10( + root=data_path, + train=False, + download=do_download, + transform=transform) + +testloader = t.utils.data.DataLoader( + testset, + batch_size=4, + shuffle=False, + num_workers=2) + +classes = ('plane', 'car', 'bird', 'cat', 'deer', + 'dog', 'frog', 'horse', 'ship', 'truck') diff --git a/tips/datasets.ipynb b/tips/datasets.ipynb index 3bd818b..64e57ec 100644 --- a/tips/datasets.ipynb +++ b/tips/datasets.ipynb @@ -314,6 +314,75 @@ "data = np.concatenate((x, yy), axis=1)\n", "np.savetxt(\"dataset_circles.csv\", data, delimiter=\",\")" ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## CIFAR-10数据\n", + "\n", + "CIFAR-10[^3]是一个常用的彩色图片数据集，它有10个类别: 'airplane', 'automobile', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck'。每张图片都是$3\\times32\\times32$，也即3-通道彩色图片，分辨率为$32\\times32$。\n", + "\n", + "[^3]: http://www.cs.toronto.edu/~kriz/cifar.html" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import torchvision as tv\n", + "import torchvision.transforms as transforms\n", + "from torchvision.transforms import ToPILImage\n", + "show = ToPILImage() # 可以把Tensor转成Image，方便可视化" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# 第一次运行程序torchvision会自动下载CIFAR-10数据集，\n", + "# 大约100M，需花费一定的时间，\n", + "# 如果已经下载有CIFAR-10，可通过root参数指定\n", + "\n", + "# 定义对数据的预处理\n", + "transform = transforms.Compose([\n", + " transforms.ToTensor(), # 转为Tensor\n", + " transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5)), # 归一化\n", + " ])\n", + "\n", + "# 训练集\n", + "trainset = tv.datasets.CIFAR10(\n", + " root='../data/', \n", + " train=True, \n", + " download=True,\n", + " transform=transform)\n", + "\n", + "trainloader = t.utils.data.DataLoader(\n", + " trainset, \n", + " batch_size=4,\n", + " shuffle=True, \n", + " num_workers=2)\n", + "\n", + "# 测试集\n", + "testset = tv.datasets.CIFAR10(\n", + " '../data/',\n", + " train=False, \n", + " download=True, \n", + " transform=transform)\n", + "\n", + "testloader = t.utils.data.DataLoader(\n", + " testset,\n", + " batch_size=4, \n", + " shuffle=False,\n", + " num_workers=2)\n", + "\n", + "classes = ('plane', 'car', 'bird', 'cat', 'deer', \n", + " 'dog', 'frog', 'horse', 'ship', 'truck')" + ] } ], "metadata": { diff --git a/tips/datasets.py b/tips/datasets.py index 571f46c..86c8443 100644 --- a/tips/datasets.py +++ b/tips/datasets.py @@ -174,3 +174,55 @@ plt.show() yy = y.reshape(-1, 1) data = np.concatenate((x, yy), axis=1) np.savetxt("dataset_circles.csv", data, delimiter=",") +# - + +# ## CIFAR-10数据 +# +# CIFAR-10[^3]是一个常用的彩色图片数据集，它有10个类别: 'airplane', 'automobile', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck'。每张图片都是$3\times32\times32$，也即3-通道彩色图片，分辨率为$32\times32$。 +# +# [^3]: http://www.cs.toronto.edu/~kriz/cifar.html + +import torchvision as tv +import torchvision.transforms as transforms +from torchvision.transforms import ToPILImage +show = ToPILImage() # 可以把Tensor转成Image，方便可视化 + +# + +# 第一次运行程序torchvision会自动下载CIFAR-10数据集， +# 大约100M，需花费一定的时间， +# 如果已经下载有CIFAR-10，可通过root参数指定 + +# 定义对数据的预处理 +transform = transforms.Compose([ + transforms.ToTensor(), # 转为Tensor + transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5)), # 归一化 + ]) + +# 训练集 +trainset = tv.datasets.CIFAR10( + root='../data/', + train=True, + download=True, + transform=transform) + +trainloader = t.utils.data.DataLoader( + trainset, + batch_size=4, + shuffle=True, + num_workers=2) + +# 测试集 +testset = tv.datasets.CIFAR10( + '../data/', + train=False, + download=True, + transform=transform) + +testloader = t.utils.data.DataLoader( + testset, + batch_size=4, + shuffle=False, + num_workers=2) + +classes = ('plane', 'car', 'bird', 'cat', 'deer', + 'dog', 'frog', 'horse', 'ship', 'truck')