diff --git a/README.md b/README.md
index 9a34e7c..f416dc9 100644
--- a/README.md
+++ b/README.md
@@ -35,7 +35,7 @@
 4. 安装PyTorch
 访问[PyTorch官网](https://pytorch.org/get-started/locally/)，选择合适的版本安装PyTorch，有条件的小伙伴可以下载GPU版本
 ```shell
-pip3 install torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cu117
+pip3 install torch==1.12.1+cu116 torchvision==0.13.1+cu116 torchaudio==0.12.1+cu116 -f https://download.pytorch.org/whl/torch_stable.html
 ```
 
 5. docsify框架运行
@@ -162,9 +162,10 @@ requirements.txt-----------------------------------运行环境依赖包
 - [王昊文](https://github.com/whw199833) （帝国理工学院-算法工程师）
 
 **其他**
-1. 特别感谢 [@Sm1les](https://github.com/Sm1les)、[@LSGOMYP](https://github.com/LSGOMYP) 对本项目的帮助与支持
-2. 感谢[@GYHHAHA](https://github.com/GYHHAHA)，指出了第7章习题7.4的解答问题，并完善了该题的解答
-3. 感谢范佳慧、汪健麟、张宇明、范致远、兰坤、李拙等同学对项目提供的完善建议
+1. 特别感谢 [@Sm1les](https://github.com/Sm1les)、[@LSGOMYP](https://github.com/LSGOMYP) 对本项目的帮助与支持；
+2. 感谢[@GYHHAHA](https://github.com/GYHHAHA)，指出了第7章习题7.4的解答问题，并完善了该题的解答；
+3. 感谢范佳慧、汪健麟、张宇明、兰坤、李拙等同学对项目提供的完善建议；
+4. 感觉张帆同学对习题27.1解答的帮助，解决了ELMo预训练模型的代码问题。
 
 ## 参考文献
 1. [李航《统计学习方法笔记》中的代码、notebook、参考文献、Errata](https://github.com/SmirkCao/Lihang)  
diff --git a/codes/ch27/bi-lstm-text-classification.py b/codes/ch27/bi-lstm-text-classification.py
index 5f35dee..458780c 100644
--- a/codes/ch27/bi-lstm-text-classification.py
+++ b/codes/ch27/bi-lstm-text-classification.py
@@ -5,158 +5,143 @@
 @file: bi-lstm-text-classification.py
 @time: 2023/3/15 14:30
 @project: statistical-learning-method-solutions-manual
-@desc: 习题27.1 基于双向LSTM的预训练语言模型，假设下游任务是文本分类
+@desc: 习题27.1 基于双向LSTM的ELMo预训练语言模型，假设下游任务是文本分类
 """
+import os
 import time
 
 import torch
 import torch.nn as nn
+import wget
+from allennlp.modules.elmo import Elmo
+from allennlp.modules.elmo import batch_to_ids
 from torch.utils.data import DataLoader
 from torch.utils.data.dataset import random_split
 from torchtext.data.functional import to_map_style_dataset
-from torchtext.data.utils import get_tokenizer
 from torchtext.datasets import AG_NEWS
-from torchtext.vocab import build_vocab_from_iterator
+
+
+def get_elmo_model():
+    elmo_options_file = './data/elmo_2x1024_128_2048cnn_1xhighway_options.json'
+    elmo_weight_file = './data/elmo_2x1024_128_2048cnn_1xhighway_weights.hdf5'
+    url = "https://s3-us-west-2.amazonaws.com/allennlp/models/elmo/2x1024_128_2048cnn_1xhighway/elmo_2x1024_128_2048cnn_1xhighway_options.json"
+    if (not os.path.exists(elmo_options_file)):
+        wget.download(url, elmo_options_file)
+    url = "https://s3-us-west-2.amazonaws.com/allennlp/models/elmo/2x1024_128_2048cnn_1xhighway/elmo_2x1024_128_2048cnn_1xhighway_weights.hdf5"
+    if (not os.path.exists(elmo_weight_file)):
+        wget.download(url, elmo_weight_file)
+
+    elmo = Elmo(elmo_options_file, elmo_weight_file, 1)
+    return elmo
+
+
+# 加载ELMo模型
+elmo = get_elmo_model()
 
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 
-# 加载AG_NEWS数据集
-train_iter, test_iter = AG_NEWS(root='./data')
-# 定义tokenizer
-tokenizer = get_tokenizer('basic_english')
-
-
-# 定义数据处理函数
-def yield_tokens(data_iter):
-    for _, text in data_iter:
-        yield tokenizer(text)
-
-
-# 构建词汇表
-vocab = build_vocab_from_iterator(yield_tokens(train_iter), specials=["<unk>"])
-vocab.set_default_index(vocab["<unk>"])
-
-# 将数据集映射到MapStyleDataset格式
-train_dataset = to_map_style_dataset(train_iter)
-test_dataset = to_map_style_dataset(test_iter)
-# 划分验证集
-num_train = int(len(train_dataset) * 0.95)
-split_train_, split_valid_ = random_split(train_dataset, [num_train, len(train_dataset) - num_train])
-
-# 设置文本和标签的处理函数
-text_pipeline = lambda x: vocab(tokenizer(x))
 label_pipeline = lambda x: int(x) - 1
 
 
 def collate_batch(batch):
-    """
-    对数据集进行数据处理
-    """
-    label_list, text_list, offsets = [], [], [0]
+    label_list, text_list = [], []
     for (_label, _text) in batch:
         label_list.append(label_pipeline(_label))
-        processed_text = torch.tensor(text_pipeline(_text), dtype=torch.int64)
-        text_list.append(processed_text)
-        offsets.append(processed_text.size(0))
+        text_list.append(_text.split())
     label_list = torch.tensor(label_list, dtype=torch.int64)
-    offsets = torch.tensor(offsets[:-1]).cumsum(dim=0)
-    text_list = torch.cat(text_list)
-    return label_list.to(device), text_list.to(device), offsets.to(device)
+    return label_list.to(device), text_list
 
 
-# 构建数据集的数据加载器
-BATCH_SIZE = 256
+# 加载AG_NEWS数据集
+train_iter, test_iter = AG_NEWS(root='./data')
+train_dataset = to_map_style_dataset(train_iter)
+test_dataset = to_map_style_dataset(test_iter)
+num_train = int(len(train_dataset) * 0.95)
+split_train_, split_valid_ = \
+    random_split(train_dataset, [num_train, len(train_dataset) - num_train])
+
+BATCH_SIZE = 128
 train_dataloader = DataLoader(split_train_, batch_size=BATCH_SIZE,
                               shuffle=True, collate_fn=collate_batch)
 valid_dataloader = DataLoader(split_valid_, batch_size=BATCH_SIZE,
-                              shuffle=True, collate_fn=collate_batch)
+                              shuffle=False, collate_fn=collate_batch)
 test_dataloader = DataLoader(test_dataset, batch_size=BATCH_SIZE,
-                             shuffle=True, collate_fn=collate_batch)
+                             shuffle=False, collate_fn=collate_batch)
 
 
 class TextClassifier(nn.Module):
-    """
-    基于双向LSTM的文本分类模型
-    """
-
-    def __init__(self, vocab_size, embedding_dim, hidden_dim, num_classes):
+    def __init__(self, embedding_dim, hidden_dim, num_classes):
         super().__init__()
-        self.embedding = nn.EmbeddingBag(vocab_size, embedding_dim, sparse=False)
+        # 使用预训练的ELMO
+        self.elmo = elmo
+
+        # 使用双向LSTM
         self.lstm = nn.LSTM(embedding_dim, hidden_dim, bidirectional=True, batch_first=True)
+
+        # 使用线性函数进行文本分类任务
         self.fc = nn.Linear(hidden_dim * 2, num_classes)
+
+        self.dropout = nn.Dropout(0.5)
         self.init_weights()
 
     def init_weights(self):
-        initrange = 0.5
-        self.embedding.weight.data.uniform_(-initrange, initrange)
+        initrange = 0.1
         self.fc.weight.data.uniform_(-initrange, initrange)
-        self.fc.bias.data.zero_()
+        self.fc.bias.data.uniform_(-initrange, initrange)
 
-    def load_elmo_weights(self, elmo):
-        self.embedding.weight.data.copy_(elmo.embedding.weight.data)
-        self.embedding.weight.requires_grad = False
-        self.lstm.weight_ih_l0.data.copy_(elmo.lstm.weight_ih_l0.data)
-        self.lstm.weight_hh_l0.data.copy_(elmo.lstm.weight_hh_l0.data)
-        self.lstm.bias_ih_l0.data.copy_(elmo.lstm.bias_ih_l0.data)
-        self.lstm.bias_hh_l0.data.copy_(elmo.lstm.bias_hh_l0.data)
-        self.lstm.weight_ih_l0_reverse.data.copy_(elmo.lstm.weight_ih_l0_reverse.data)
-        self.lstm.weight_hh_l0_reverse.data.copy_(elmo.lstm.weight_hh_l0_reverse.data)
-        self.lstm.bias_ih_l0_reverse.data.copy_(elmo.lstm.bias_ih_l0_reverse.data)
-        self.lstm.bias_hh_l0_reverse.data.copy_(elmo.lstm.bias_hh_l0_reverse.data)
-        self.fc.weight.data.copy_(elmo.fc.weight.data)
-        self.fc.bias.data.copy_(elmo.fc.bias.data)
+    def forward(self, sentence_lists):
+        character_ids = batch_to_ids(sentence_lists)
+        character_ids = character_ids.to(device)
+
+        embeddings = self.elmo(character_ids)
+        embedded = embeddings['elmo_representations'][0]
 
-    def forward(self, text, offsets):
-        embedded = self.embedding(text, offsets)
         x, _ = self.lstm(embedded)
+        x = x.mean(1)
+        x = self.dropout(x)
         x = self.fc(x)
         return x
 
 
-# 设置超参数
-EMBED_DIM = 64
+EMBED_DIM = 256
 HIDDEN_DIM = 64
 NUM_CLASSES = 4
 LEARNING_RATE = 1e-2
-NUM_EPOCHS = 10
+NUM_EPOCHS = 1
 
-# 创建模型、优化器和损失函数
-model = TextClassifier(len(vocab), EMBED_DIM, HIDDEN_DIM, NUM_CLASSES).to(device)
-criterion = nn.CrossEntropyLoss().to(device)
-optimizer = torch.optim.Adam(model.parameters(), lr=LEARNING_RATE)
+model = TextClassifier(EMBED_DIM, HIDDEN_DIM, NUM_CLASSES).to(device)
 
 
 def train(dataloader):
-    """
-    模型训练
-    """
     model.train()
 
-    for idx, (label, text, offsets) in enumerate(dataloader):
+    for idx, (label, text) in enumerate(dataloader):
         optimizer.zero_grad()
-        predicted_label = model(text, offsets)
+        predicted_label = model(text)
         loss = criterion(predicted_label, label)
         loss.backward()
-        torch.nn.utils.clip_grad_norm_(model.parameters(), 0.1)
+        torch.nn.utils.clip_grad_norm_(model.parameters(), 1.0)
         optimizer.step()
 
 
 def evaluate(dataloader):
-    """
-    模型验证
-    """
     model.eval()
     total_acc, total_count = 0, 0
 
     with torch.no_grad():
-        for idx, (label, text, offsets) in enumerate(dataloader):
-            predicted_label = model(text, offsets)
-            criterion(predicted_label, label)
+        for idx, (label, text) in enumerate(dataloader):
+            predicted_label = model(text)
             total_acc += (predicted_label.argmax(1) == label).sum().item()
             total_count += label.size(0)
+
     return total_acc / total_count
 
 
+# 使用交叉熵损失函数
+criterion = nn.CrossEntropyLoss().to(device)
+# 设置优化器
+optimizer = torch.optim.Adam(model.parameters(), lr=LEARNING_RATE)
+
 for epoch in range(1, NUM_EPOCHS + 1):
     epoch_start_time = time.time()
     train(train_dataloader)
@@ -168,26 +153,24 @@ for epoch in range(1, NUM_EPOCHS + 1):
                                             accu_val * 100))
     print('-' * 59)
 
-# 新闻的分类标签
-ag_news_label = {1: "World",
-                 2: "Sports",
-                 3: "Business",
-                 4: "Sci/Tec"}
+ag_news_label = {1: "World", 2: "Sports", 3: "Business", 4: "Sci/Tec"}
 
 
-def predict(text, text_pipeline):
+def predict(text):
     with torch.no_grad():
-        text = torch.tensor(text_pipeline(text))
-        output = model(text, torch.tensor([0]))
+        output = model([text])
         return output.argmax(1).item() + 1
 
 
-# 预测一个文本的类别
 ex_text_str = """
-Our younger Fox Cubs (Y2-Y4) also had a great second experience of swimming competition in February when they travelled 
-over to NIS at the end of February to compete in the SSL Development Series R2 event. For students aged 9 and under 
-these SSL Development Series events are a great introduction to competitive swimming, focussed on fun and participation 
-whilst also building basic skills and confidence as students build up to joining the full SSL team in Year 5 and beyond.
+Our younger Fox Cubs (Y2-Y4) also had a great second experience 
+of swimming competition in February when they travelled over to 
+NIS at the end of February to compete in the SSL Development 
+Series R2 event. For students aged 9 and under these SSL 
+Development Series events are a great introduction to 
+competitive swimming, focussed on fun and participation whilst 
+also building basic skills and confidence as students build up 
+to joining the full SSL team in Year 5 and beyond.
 """
-model = model.to("cpu")
-print("This is a %s news" % ag_news_label[predict(ex_text_str, text_pipeline)])
+
+print("This is a %s news" % ag_news_label[predict(ex_text_str)])
diff --git a/docs/README.md b/docs/README.md
index d40a79d..2e6cdfb 100644
--- a/docs/README.md
+++ b/docs/README.md
@@ -118,9 +118,10 @@ requirements.txt-----------------------------------运行环境依赖包
 - [王昊文](https://github.com/whw199833) （帝国理工学院-算法工程师）
 
 **其他**
-1. 特别感谢 [@Sm1les](https://github.com/Sm1les)、[@LSGOMYP](https://github.com/LSGOMYP) 对本项目的帮助与支持
-2. 感谢[@GYHHAHA](https://github.com/GYHHAHA)，指出了第7章习题7.4的解答问题，并完善了该题的解答
-3. 感谢范佳慧、汪健麟、张宇明、范致远、兰坤、李拙等同学对项目提供的完善建议
+1. 特别感谢 [@Sm1les](https://github.com/Sm1les)、[@LSGOMYP](https://github.com/LSGOMYP) 对本项目的帮助与支持；
+2. 感谢[@GYHHAHA](https://github.com/GYHHAHA)，指出了第7章习题7.4的解答问题，并完善了该题的解答；
+3. 感谢范佳慧、汪健麟、张宇明、兰坤、李拙等同学对项目提供的完善建议；
+4. 感觉张帆同学对习题27.1解答的帮助，解决了ELMo预训练模型的代码问题。
 
 ## 参考文献
 1. [李航《统计学习方法笔记》中的代码、notebook、参考文献、Errata](https://github.com/SmirkCao/Lihang)  
diff --git a/notebook/part03/notes/ch27.ipynb b/notebook/part03/notes/ch27.ipynb
index 3c9aa69..968071f 100644
--- a/notebook/part03/notes/ch27.ipynb
+++ b/notebook/part03/notes/ch27.ipynb
@@ -3,7 +3,8 @@
   {
    "cell_type": "markdown",
    "metadata": {
-    "collapsed": true
+    "collapsed": true,
+    "id": "zpNlbEG5HWn0"
    },
    "source": [
     "# 第27章预训练语言模型"
@@ -12,7 +13,8 @@
   {
    "cell_type": "markdown",
    "metadata": {
-    "collapsed": true
+    "collapsed": true,
+    "id": "waoEGAtfHWn6"
    },
    "source": [
     "## 习题27.1\n",
@@ -21,56 +23,240 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "tT2dkcZrHWn6"
+   },
    "source": [
     "**解答：**  "
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "**解答思路：**\n",
-    "\n",
-    "1. 结合习题25.4可给出双向LSTM模型\n",
-    "2. 给出下游任务是文本分类的模型微调的描述\n",
-    "3. 自编程实现双向LSTM的预训练语言模型"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "A_Qohfg3HWn7"
+   },
    "source": [
     "**解答步骤：**  "
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "yQuo79vYHWn7"
+   },
    "source": [
-    "**第1步：结合习题25.4可给出双向LSTM模型**"
+    "**解答思路：**\n",
+    "\n",
+    "1. 设计基于双向LSTM的预训练模型ELMO，并加载模型权重\n",
+    "2. 设计基于ELMO的文本分类模型\n",
+    "3. 自编程实现基于ELMO的文本分类"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "UsRdKonCIOTr"
+   },
    "source": [
-    "&emsp;&emsp;前向的LSTM的隐层（状态）是\n",
+    "**第1步：设计基于双向LSTM的预训练模型ELMO，并加载模型权重**"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "id": "UsRdKonCIOTr"
+   },
+   "source": [
+    "&emsp;&emsp;ELMO是一个双向LSTM模型，它首先使用大规模文本语料库进行预训练，当用于下游任务时，其输出表示通常作为上下文相关的词向量。（论文参考链接：https://arxiv.org/abs/1802.05365）\n",
+    "\n",
+    "&emsp;&emsp;ELMO有多个不同模型大小的变体，这里我们以small版本为例，详见：https://allenai.org/allennlp/software/elmo."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import wget\n",
+    "import os\n",
+    "from allennlp.modules.elmo import Elmo"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "elmo_options_file = './data/elmo_2x1024_128_2048cnn_1xhighway_options.json'\n",
+    "elmo_weight_file = './data/elmo_2x1024_128_2048cnn_1xhighway_weights.hdf5'\n",
+    "url = \"https://s3-us-west-2.amazonaws.com/allennlp/models/elmo/2x1024_128_2048cnn_1xhighway/elmo_2x1024_128_2048cnn_1xhighway_options.json\"\n",
+    "if(not os.path.exists(elmo_options_file)):\n",
+    "    wget.download(url, elmo_options_file)\n",
+    "url = \"https://s3-us-west-2.amazonaws.com/allennlp/models/elmo/2x1024_128_2048cnn_1xhighway/elmo_2x1024_128_2048cnn_1xhighway_weights.hdf5\"\n",
+    "if(not os.path.exists(elmo_weight_file)):\n",
+    "    wget.download(url, elmo_weight_file)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# 加载模型\n",
+    "elmo = Elmo(elmo_options_file, elmo_weight_file, 1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Elmo(\n",
+       "  (_elmo_lstm): _ElmoBiLm(\n",
+       "    (_token_embedder): _ElmoCharacterEncoder(\n",
+       "      (char_conv_0): Conv1d(16, 32, kernel_size=(1,), stride=(1,))\n",
+       "      (char_conv_1): Conv1d(16, 32, kernel_size=(2,), stride=(1,))\n",
+       "      (char_conv_2): Conv1d(16, 64, kernel_size=(3,), stride=(1,))\n",
+       "      (char_conv_3): Conv1d(16, 128, kernel_size=(4,), stride=(1,))\n",
+       "      (char_conv_4): Conv1d(16, 256, kernel_size=(5,), stride=(1,))\n",
+       "      (char_conv_5): Conv1d(16, 512, kernel_size=(6,), stride=(1,))\n",
+       "      (char_conv_6): Conv1d(16, 1024, kernel_size=(7,), stride=(1,))\n",
+       "      (_highways): Highway(\n",
+       "        (_layers): ModuleList(\n",
+       "          (0): Linear(in_features=2048, out_features=4096, bias=True)\n",
+       "        )\n",
+       "      )\n",
+       "      (_projection): Linear(in_features=2048, out_features=128, bias=True)\n",
+       "    )\n",
+       "    (_elmo_lstm): ElmoLstm(\n",
+       "      (forward_layer_0): LstmCellWithProjection(\n",
+       "        (input_linearity): Linear(in_features=128, out_features=4096, bias=False)\n",
+       "        (state_linearity): Linear(in_features=128, out_features=4096, bias=True)\n",
+       "        (state_projection): Linear(in_features=1024, out_features=128, bias=False)\n",
+       "      )\n",
+       "      (backward_layer_0): LstmCellWithProjection(\n",
+       "        (input_linearity): Linear(in_features=128, out_features=4096, bias=False)\n",
+       "        (state_linearity): Linear(in_features=128, out_features=4096, bias=True)\n",
+       "        (state_projection): Linear(in_features=1024, out_features=128, bias=False)\n",
+       "      )\n",
+       "      (forward_layer_1): LstmCellWithProjection(\n",
+       "        (input_linearity): Linear(in_features=128, out_features=4096, bias=False)\n",
+       "        (state_linearity): Linear(in_features=128, out_features=4096, bias=True)\n",
+       "        (state_projection): Linear(in_features=1024, out_features=128, bias=False)\n",
+       "      )\n",
+       "      (backward_layer_1): LstmCellWithProjection(\n",
+       "        (input_linearity): Linear(in_features=128, out_features=4096, bias=False)\n",
+       "        (state_linearity): Linear(in_features=128, out_features=4096, bias=True)\n",
+       "        (state_projection): Linear(in_features=1024, out_features=128, bias=False)\n",
+       "      )\n",
+       "    )\n",
+       "  )\n",
+       "  (_dropout): Dropout(p=0.5, inplace=False)\n",
+       "  (scalar_mix_0): ScalarMix(\n",
+       "    (scalar_parameters): ParameterList(\n",
+       "        (0): Parameter containing: [torch.FloatTensor of size 1]\n",
+       "        (1): Parameter containing: [torch.FloatTensor of size 1]\n",
+       "        (2): Parameter containing: [torch.FloatTensor of size 1]\n",
+       "    )\n",
+       "  )\n",
+       ")"
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "elmo"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "id": "UEs7rScQcVXm"
+   },
+   "source": [
+    "&emsp;&emsp;从模型结构可以看到，ELMO主要由一个token编码器、一个双向双层LSTM和输出层构成。其中token编码器为输入的token计算输入向量表示，然后双向双层LSTM计算隐层（状态），最后，输出层将输入向量表示以及两层LSTM的隐层（状态）进行混合，得到最后的输出表示。"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "id": "ymFQJbs_HWn8"
+   },
+   "source": [
+    "**第2步：设计基于ELMO的文本分类模型**"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "id": "ymFQJbs_HWn8"
+   },
+   "source": [
+    "&emsp;&emsp;文本分类模型主要由预训练模型ELMO、基于双向LSTM的编码器、全连接层分类器三部分组成。"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "id": "ymFQJbs_HWn8"
+   },
+   "source": [
+    "&emsp;&emsp;假设输入单词序列是$\\text{x} = \\{\\text{x}_1, \\text{x}_2, \\cdots, \\text{x}_l$\\}，输出是类别 $\\text{y}$，文本分类任务旨在计算条件概率 $P(\\text{y}|\\text{x}) = P(\\text{y}|\\text{x}_1, \\text{x}_2, \\cdots, \\text{x}_l)$。"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "id": "ymFQJbs_HWn8"
+   },
+   "source": [
+    "&emsp;&emsp;首先，将单词序列 $\\text{x}$ 输入到预训练模型ELMO，获得每个token的输入表示$x = \\{x_1, x_2, \\cdots, x_l$ \\}。"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "id": "ymFQJbs_HWn8"
+   },
+   "source": [
+    "&emsp;&emsp;结合习题25.4，基于双向LSTM的编码器如下所示："
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "id": "7KQP3wPEHWn8"
+   },
+   "source": [
+    "&emsp;&emsp;前向的LSTM的隐层（状态）:\n",
     "$$\n",
     "h_t^f = \\text{LSTM}_f (x_t, h_{t-1}^f)\n",
     "$$\n",
-    "&emsp;&emsp;后向的LSTM的隐层（状态）是\n",
+    "\n",
+    "&emsp;&emsp;后向的LSTM的隐层（状态）:\n",
     "$$\n",
     "h_t^b = \\text{LSTM}_b (x_t, h_{t+1}^b)\n",
     "$$\n",
+    "\n",
     "其中，$\\text{LSTM}_f$ 和 $\\text{LSTM}_b$ 分别表示前向LSTM和后向LSTM，$h_{t-1}^f$ 和 $h_{t+1}^b$ 分别表示前向LSTM和后向LSTM的上一个位置的隐藏状态。"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "1G13vlt3HWn8"
+   },
    "source": [
-    "两者的拼接是\n",
+    "这里，将两者的拼接表示为:\n",
     "$$\n",
     "h_t = [h_i^f; h_i^b]\n",
     "$$\n",
@@ -79,48 +265,54 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "04rIMfCrY6Wr"
+   },
    "source": [
+    "&emsp;&emsp;然后，将所有token的表示进行Mean-Pooling，得到输入单词序列$\\text{x}$的表示$h$:\n",
+    "\n",
     "$$\n",
-    "p_t = \\text{softmax}(V \\cdot h_t + c) \n",
+    "h = \\text{Mean-Pool}(h_1, h_2, \\cdots, h_l)\n",
     "$$"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "04rIMfCrY6Wr"
+   },
    "source": [
-    "**第2步：给出下游任务是文本分类的模型微调的描述**"
+    "&emsp;&emsp;最后,将这一表示喂入分类器，获得每个类别的概率:\n",
+    "\n",
+    "$$\n",
+    "p = \\text{softmax}(\\text{fc}(h)) \n",
+    "$$"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "dVold10AiaGR"
+   },
    "source": [
-    "&emsp;&emsp;根据书中第498页下游任务是文本分类的模型微调的描述：\n",
-    "> &emsp;&emsp;假设下游任务是文本分类，输入单词序列是$x' = x_0, x_1, \\cdots, x_l$，输出是类别 $y$，计算条件概率 $P(y|x_0,x_1,\\cdots, x_l)$：\n",
-    "> $$\n",
-    "P_{\\theta,\\phi} = \\text{softmax} ({W_y}^T h_{cls}^{(L)}) = \\frac{\\exp {w_y^T \\cdot h_{cls}^{(L)}}}{\\sum_{y'} \\exp {w_{y'}^T \\cdot h_{cls}^{(L)}}} \\tag{27.30}\n",
-    "$$\n",
-    ">其中，$h_{cls}^{(L)}$ 是第 $L$ 层的类别特殊字符\\<cls\\>的表示向量，$W_y$ 是类别的权重矩阵，$\\phi$ 表示分类的参数。这时单词序列 $x_0, x_1, \\cdots ,x_l$ 是一个句子或一段文章，以特殊字符\\<cls\\>开始，以特殊字符\\<sep\\>结束。\n",
-    "> 微调损失函数为：\n",
-    "> $$\n",
-    "L_{\\text{FT}} = -\\log P_{\\theta,\\phi}(y|x')\n",
-    "$$\n",
-    "> 微调中，预训练模型的参数 $\\theta$ 作为初始值，在这个过程中进一步得到学习，以帮助更好地分类；同时分类的参数 $\\phi$ 也得到学习。"
+    "&emsp;&emsp;在训练时，通常将预训练模型ELMO的参数进行固定，然后使用交叉熵作为损失函数，优化基于双向LSTM的编码器和全连接层分类器。"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "D7-dA3DCHWn9"
+   },
    "source": [
     "**第3步：自编程实现双向LSTM的预训练语言模型**"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
-   "metadata": {},
+   "execution_count": 5,
+   "metadata": {
+    "id": "JjTkIP7nHWn-"
+   },
    "outputs": [],
    "source": [
     "import time\n",
@@ -130,64 +322,46 @@
     "from torch.utils.data import DataLoader\n",
     "from torch.utils.data.dataset import random_split\n",
     "from torchtext.data.functional import to_map_style_dataset\n",
-    "from torchtext.data.utils import get_tokenizer\n",
     "from torchtext.datasets import AG_NEWS\n",
-    "from torchtext.vocab import build_vocab_from_iterator"
+    "from allennlp.modules.elmo import batch_to_ids"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
-   "metadata": {},
+   "execution_count": 6,
+   "metadata": {
+    "id": "f-yEKyAGHWoB"
+   },
    "outputs": [],
    "source": [
-    "# 得到词向量\n",
-    "train_iter = AG_NEWS(split='train', root='./data')\n",
-    "tokenizer = get_tokenizer('basic_english')\n",
-    "def yield_tokens(data_iter):\n",
-    "    for _, text in data_iter:\n",
-    "        yield tokenizer(text)\n",
-    "\n",
-    "\n",
-    "vocab = build_vocab_from_iterator(yield_tokens(train_iter), specials=[\"<unk>\"])\n",
-    "vocab.set_default_index(vocab[\"<unk>\"])"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 3,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "text_pipeline = lambda x : vocab(tokenizer(x))\n",
     "label_pipeline = lambda x : int(x) - 1"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
-   "metadata": {},
+   "execution_count": 7,
+   "metadata": {
+    "id": "Kg4pQc7eHWoB"
+   },
    "outputs": [],
    "source": [
     "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
     "\n",
     "def collate_batch(batch):\n",
-    "    label_list, text_list, offsets = [], [], [0]\n",
+    "    label_list, text_list = [], []\n",
     "    for (_label, _text) in batch:\n",
     "        label_list.append(label_pipeline(_label))\n",
-    "        processed_text = torch.tensor(text_pipeline(_text), dtype=torch.int64)\n",
-    "        text_list.append(processed_text)\n",
-    "        offsets.append(processed_text.size(0))\n",
+    "        text_list.append(_text.split())\n",
     "    label_list = torch.tensor(label_list, dtype=torch.int64)\n",
-    "    offsets = torch.tensor(offsets[:-1]).cumsum(dim=0)\n",
-    "    text_list = torch.cat(text_list)\n",
-    "    return label_list.to(device), text_list.to(device), offsets.to(device)"
+    "    return label_list.to(device), text_list"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
-   "metadata": {},
+   "execution_count": 8,
+   "metadata": {
+    "id": "RhGwcP-gHWoB"
+   },
    "outputs": [],
    "source": [
     "# 加载AG_NEWS数据集\n",
@@ -198,111 +372,112 @@
     "split_train_, split_valid_ = \\\n",
     "    random_split(train_dataset, [num_train, len(train_dataset) - num_train])\n",
     "\n",
-    "BATCH_SIZE = 256\n",
+    "\n",
+    "BATCH_SIZE = 128\n",
     "train_dataloader = DataLoader(split_train_, batch_size=BATCH_SIZE,\n",
     "                              shuffle=True, collate_fn=collate_batch)\n",
     "valid_dataloader = DataLoader(split_valid_, batch_size=BATCH_SIZE,\n",
-    "                              shuffle=True, collate_fn=collate_batch)\n",
+    "                              shuffle=False, collate_fn=collate_batch)\n",
     "test_dataloader = DataLoader(test_dataset, batch_size=BATCH_SIZE,\n",
-    "                             shuffle=True, collate_fn=collate_batch)"
+    "                             shuffle=False, collate_fn=collate_batch)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
-   "metadata": {},
+   "execution_count": 9,
+   "metadata": {
+    "id": "FMLBRD10HWoC"
+   },
    "outputs": [],
    "source": [
     "class TextClassifier(nn.Module):\n",
-    "    def __init__(self, vocab_size, embedding_dim, hidden_dim, num_classes):\n",
+    "    def __init__(self, embedding_dim, hidden_dim, num_classes):\n",
     "        super().__init__()\n",
-    "        self.embedding = nn.EmbeddingBag(vocab_size, embedding_dim, sparse=False)\n",
+    "        # 使用预训练的ELMO\n",
+    "        self.elmo = elmo\n",
+    "\n",
     "        # 使用双向LSTM\n",
     "        self.lstm = nn.LSTM(embedding_dim, hidden_dim, bidirectional=True, batch_first=True)\n",
+    "\n",
     "        # 使用线性函数进行文本分类任务\n",
     "        self.fc = nn.Linear(hidden_dim * 2, num_classes)\n",
+    "\n",
+    "        self.dropout = nn.Dropout(0.5)\n",
     "        self.init_weights()\n",
-    "    \n",
+    "\n",
     "    def init_weights(self):\n",
-    "        initrange = 0.5\n",
-    "        self.embedding.weight.data.uniform_(-initrange, initrange)\n",
+    "        initrange = 0.1\n",
     "        self.fc.weight.data.uniform_(-initrange, initrange)\n",
-    "        self.fc.bias.data.zero_()\n",
+    "        self.fc.bias.data.uniform_(-initrange, initrange)\n",
     "    \n",
-    "    def load_elmo_weights(self, elmo):\n",
-    "        self.embedding.weight.data.copy_(elmo.embedding.weight.data)\n",
-    "        self.embedding.weight.requires_grad = False\n",
-    "        self.lstm.weight_ih_l0.data.copy_(elmo.lstm.weight_ih_l0.data)\n",
-    "        self.lstm.weight_hh_l0.data.copy_(elmo.lstm.weight_hh_l0.data)\n",
-    "        self.lstm.bias_ih_l0.data.copy_(elmo.lstm.bias_ih_l0.data)\n",
-    "        self.lstm.bias_hh_l0.data.copy_(elmo.lstm.bias_hh_l0.data)\n",
-    "        self.lstm.weight_ih_l0_reverse.data.copy_(elmo.lstm.weight_ih_l0_reverse.data)\n",
-    "        self.lstm.weight_hh_l0_reverse.data.copy_(elmo.lstm.weight_hh_l0_reverse.data)\n",
-    "        self.lstm.bias_ih_l0_reverse.data.copy_(elmo.lstm.bias_ih_l0_reverse.data)\n",
-    "        self.lstm.bias_hh_l0_reverse.data.copy_(elmo.lstm.bias_hh_l0_reverse.data)\n",
-    "        self.fc.weight.data.copy_(elmo.fc.weight.data)\n",
-    "        self.fc.bias.data.copy_(elmo.fc.bias.data)\n",
-    "    \n",
-    "    def forward(self, text, offsets):\n",
-    "        embedded = self.embedding(text, offsets)\n",
+    "    def forward(self, sentence_lists):\n",
+    "        character_ids = batch_to_ids(sentence_lists)\n",
+    "        character_ids = character_ids.to(device)\n",
+    "\n",
+    "        embeddings = self.elmo(character_ids)\n",
+    "        embedded = embeddings['elmo_representations'][0]\n",
+    "\n",
     "        x, _ = self.lstm(embedded)\n",
+    "        x = x.mean(1)\n",
+    "        x = self.dropout(x)\n",
     "        x = self.fc(x)\n",
     "        return x"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
-   "metadata": {},
+   "execution_count": 10,
+   "metadata": {
+    "id": "3zh34EgtHWoC"
+   },
    "outputs": [],
    "source": [
-    "EMBED_DIM = 64\n",
+    "EMBED_DIM = 256\n",
     "HIDDEN_DIM = 64\n",
     "NUM_CLASSES = 4\n",
     "LEARNING_RATE = 1e-2\n",
-    "NUM_EPOCHS = 10"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 8,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 9,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "model = TextClassifier(len(vocab), EMBED_DIM, HIDDEN_DIM, NUM_CLASSES).to(device)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 10,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "def train(dataloader):\n",
-    "    model.train()\n",
-    "\n",
-    "    for idx, (label, text, offsets) in enumerate(dataloader):\n",
-    "        optimizer.zero_grad()\n",
-    "        predicted_label = model(text, offsets)\n",
-    "        loss = criterion(predicted_label, label)\n",
-    "        loss.backward()\n",
-    "        torch.nn.utils.clip_grad_norm_(model.parameters(), 0.1)\n",
-    "        optimizer.step()"
+    "# 由于单个epochs运行时间达到5分钟，\n",
+    "# 有条件的小伙伴可以设置更多的epoch，达到更好的训练效果\n",
+    "NUM_EPOCHS = 1"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 11,
-   "metadata": {},
+   "metadata": {
+    "id": "vC-XbFOAHWoD"
+   },
+   "outputs": [],
+   "source": [
+    "model = TextClassifier(EMBED_DIM, HIDDEN_DIM, NUM_CLASSES).to(device)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {
+    "id": "92Z_GvDxHWoD"
+   },
+   "outputs": [],
+   "source": [
+    "def train(dataloader):\n",
+    "    model.train()\n",
+    "\n",
+    "    for idx, (label, text) in enumerate(dataloader):\n",
+    "        optimizer.zero_grad()\n",
+    "        predicted_label = model(text)\n",
+    "        loss = criterion(predicted_label, label)\n",
+    "        loss.backward()\n",
+    "        torch.nn.utils.clip_grad_norm_(model.parameters(), 1.0)\n",
+    "        optimizer.step()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {
+    "id": "RyDGi3l0HWoE"
+   },
    "outputs": [],
    "source": [
     "def evaluate(dataloader):\n",
@@ -310,18 +485,20 @@
     "    total_acc, total_count = 0, 0\n",
     "\n",
     "    with torch.no_grad():\n",
-    "        for idx, (label, text, offsets) in enumerate(dataloader):\n",
-    "            predicted_label = model(text, offsets)\n",
-    "            loss = criterion(predicted_label, label)\n",
+    "        for idx, (label, text) in enumerate(dataloader):\n",
+    "            predicted_label = model(text)\n",
     "            total_acc += (predicted_label.argmax(1) == label).sum().item()\n",
     "            total_count += label.size(0)\n",
+    "\n",
     "    return total_acc/total_count"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
-   "metadata": {},
+   "execution_count": 14,
+   "metadata": {
+    "id": "4O0AtMO6HWoE"
+   },
    "outputs": [],
    "source": [
     "# 使用交叉熵损失函数\n",
@@ -332,8 +509,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 15,
    "metadata": {
+    "colab": {
+     "base_uri": "https://localhost:8080/",
+     "height": 677
+    },
+    "id": "9VdHJhC9HWoF",
+    "outputId": "ff49937d-fda2-48ab-e8cb-53b537b84a8d",
     "scrolled": false
    },
    "outputs": [
@@ -342,34 +525,7 @@
      "output_type": "stream",
      "text": [
       "-----------------------------------------------------------\n",
-      "| end of epoch   1 | time:  5.74s | valid accuracy     92.1% \n",
-      "-----------------------------------------------------------\n",
-      "-----------------------------------------------------------\n",
-      "| end of epoch   2 | time:  5.49s | valid accuracy     91.2% \n",
-      "-----------------------------------------------------------\n",
-      "-----------------------------------------------------------\n",
-      "| end of epoch   3 | time:  5.36s | valid accuracy     90.9% \n",
-      "-----------------------------------------------------------\n",
-      "-----------------------------------------------------------\n",
-      "| end of epoch   4 | time:  5.25s | valid accuracy     90.5% \n",
-      "-----------------------------------------------------------\n",
-      "-----------------------------------------------------------\n",
-      "| end of epoch   5 | time:  5.20s | valid accuracy     90.5% \n",
-      "-----------------------------------------------------------\n",
-      "-----------------------------------------------------------\n",
-      "| end of epoch   6 | time:  5.21s | valid accuracy     90.5% \n",
-      "-----------------------------------------------------------\n",
-      "-----------------------------------------------------------\n",
-      "| end of epoch   7 | time:  5.22s | valid accuracy     90.6% \n",
-      "-----------------------------------------------------------\n",
-      "-----------------------------------------------------------\n",
-      "| end of epoch   8 | time:  5.20s | valid accuracy     90.4% \n",
-      "-----------------------------------------------------------\n",
-      "-----------------------------------------------------------\n",
-      "| end of epoch   9 | time:  5.22s | valid accuracy     90.0% \n",
-      "-----------------------------------------------------------\n",
-      "-----------------------------------------------------------\n",
-      "| end of epoch  10 | time:  5.21s | valid accuracy     90.0% \n",
+      "| end of epoch   1 | time: 321.49s | valid accuracy     90.4% \n",
       "-----------------------------------------------------------\n"
      ]
     }
@@ -389,39 +545,41 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
-   "metadata": {},
+   "execution_count": 16,
+   "metadata": {
+    "id": "_oPGcPsbHWoF"
+   },
    "outputs": [],
    "source": [
-    "ag_news_label = {1: \"World\",\n",
-    "                 2: \"Sports\",\n",
-    "                 3: \"Business\",\n",
-    "                 4: \"Sci/Tec\"}"
+    "ag_news_label = {1: \"World\", 2: \"Sports\", 3: \"Business\", 4: \"Sci/Tec\"}"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
-   "metadata": {},
+   "execution_count": 17,
+   "metadata": {
+    "id": "JcVF6XTZHWoG"
+   },
    "outputs": [],
    "source": [
-    "def predict(text, text_pipeline):\n",
+    "def predict(text):\n",
     "    with torch.no_grad():\n",
-    "        text = torch.tensor(text_pipeline(text))\n",
-    "        output = model(text, torch.tensor([0]))\n",
+    "        output = model([text])\n",
     "        return output.argmax(1).item() + 1"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
-   "metadata": {},
+   "execution_count": 18,
+   "metadata": {
+    "id": "X4rRKBDuHWoG"
+   },
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "This is a Sci/Tec news\n"
+      "This is a Business news\n"
      ]
     }
    ],
@@ -436,13 +594,15 @@
     "also building basic skills and confidence as students build up \n",
     "to joining the full SSL team in Year 5 and beyond.\n",
     "\"\"\"\n",
-    "model = model.to(\"cpu\")\n",
-    "print(\"This is a %s news\" %ag_news_label[predict(ex_text_str, text_pipeline)])"
+    "\n",
+    "print(\"This is a %s news\" %ag_news_label[predict(ex_text_str)])"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "W9N3Ec3oHWoH"
+   },
    "source": [
     "## 习题27.2\n",
     "&emsp;&emsp;假设GPT微调的下游任务是两个文本的匹配，写出学习的目标函数。"
@@ -450,14 +610,18 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "A8ICO0vbHWoH"
+   },
    "source": [
     "**解答：**"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "hve-xrFPHWoH"
+   },
    "source": [
     "**解答思路：**  \n",
     "\n",
@@ -468,21 +632,27 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "5eU_8FtAHWoH"
+   },
    "source": [
     "**解答步骤：**   "
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "0LTkOHWFHWoH"
+   },
    "source": [
     "**第1步：给出GPT模型的描述**"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "eejZM8uIHWoI"
+   },
    "source": [
     "&emsp;&emsp;根据书中第501页GPT模型的描述：\n",
     "> &emsp;&emsp;GPT模型的输入是单词序列，可以是一个句子或一段文章。首先经过输入层，产生初始的单词表示向量的序列。之后经过 $L$ 个Transformer解码层，得到单词表示向量的序列，GPT模型的输出是在单词序列各个位置上的条件概率。  \n",
@@ -495,14 +665,18 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "8MDBXdAUHWoI"
+   },
    "source": [
     "**第2步：给出文本匹配任务的描述**"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "LdIT08ZNHWoI"
+   },
    "source": [
     "&emsp;&emsp;根据书中第499页文本匹配任务的描述：\n",
     "> &emsp;&emsp;如果下游任务是文本匹配，如判断两句话是否形成一问一答。输入单词序列是 $x_0, x_1, \\cdots, x_l$，输出是类别 $y$，仍然计算条件概率 $P(y | x_0, x_1, \\cdots, x_l)$。类别有两类，表示匹配或不匹配。这时单词序列 $x_0, x_1, \\cdots, x_l$ 是两个单词序列合成的序列，如一个问句和一个答句合并而成。以特殊字符\\<cls\\>开始，中间以特殊字符\\<sep\\>间隔，最后以特殊字符\\<sep\\>结束。"
@@ -510,21 +684,27 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "OnY3RRIbHWoI"
+   },
    "source": [
     "**第3步：写出GPT微调的下游任务是两个文本匹配时的学习的目标函数**"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "Ov00EG4PHWoI"
+   },
    "source": [
     "&emsp;&emsp;假设分别有两个文本，矩阵$X_1 = (x_1^{(1)}, x_2^{(1)}, \\cdots, x_{m - 1}^{(1)})$和矩阵$X_2 = (x_{m + 1}^{(2)}, x_{m + 2}^{(2)}, \\cdots, x_{m + n - 1}^{(2)})$分别表示两个文本的词嵌入的序列，矩阵$E_1=(e_1^{(1)}, e_2^{(1)}, \\cdots, e_{m - 1}^{(1)})$和矩阵$E_2=(e_{m + 1}^{(2)}, e_{m + 2}^{(2)}, \\cdots, e_{m + n - 1}^{(2)})$分别表示两个文本的位置嵌入序列。"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "eHZmIDxNHWoI"
+   },
    "source": [
     "&emsp;&emsp;经过输入层，产生初始的单词表示向量的序列，分别表示为\n",
     "$$\n",
@@ -535,7 +715,9 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "PBpurXL2HWoI"
+   },
    "source": [
     "&emsp;&emsp;将两个句子分别作为前句或后句，构造两个完整的文本：  \n",
     "- 文本1：\\<Start\\>, $x_1^{(1)}, x_2^{(1)}, \\cdots, x_{m - 1}^{(1)},$ \\<Delim\\>, $x_{m + 1}^{(2)}, x_{m + 2}^{(2)}, \\cdots, x_{m + n - 1}^{(2)}$, \\<Extract\\>\n",
@@ -544,7 +726,9 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "rV8qV1mzHWoJ"
+   },
    "source": [
     "分别进行GPT预训练模型：\n",
     "$$\n",
@@ -555,7 +739,9 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "E-kfJoeEHWoJ"
+   },
    "source": [
     "将得到单词表示向量的序列进行融合，使用线性层得到GPT模型的输出：\n",
     "$$\n",
@@ -566,7 +752,9 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "e3k9STWuHWoJ"
+   },
    "source": [
     "在预训练阶段，目标函数是\n",
     "$$\n",
@@ -576,14 +764,18 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "zsq2ImsoHWoJ"
+   },
    "source": [
     "&emsp;&emsp;微调时，下游任务是文本匹配，输入的单词序列是两个文本合的序列$x_1, x_2, \\cdots,x_{m + n}$，输出是类别$y$，以特殊字符\\<Start\\>开始，中间以特殊字符\\<Delim\\>间隔，最后以特殊字符\\<Extract\\>结束。"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "JWWiMvDyHWoJ"
+   },
    "source": [
     "可得到微调的目标函数为\n",
     "$$\n",
@@ -593,7 +785,9 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "OSuANbEcHWoJ"
+   },
    "source": [
     "## 习题27.3\n",
     "&emsp;&emsp;设计一个2层卷积神经网络编码器和2层卷积神经网络解码器组成的自动编码器（使用第28章介绍的转置卷积）。"
@@ -601,14 +795,18 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "zb2t6Q5uHWoJ"
+   },
    "source": [
     "**解答：**"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "n_tmLh39HWoK"
+   },
    "source": [
     "**解答思路：**  \n",
     "\n",
@@ -619,21 +817,27 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "7u8yGJkSHWoK"
+   },
    "source": [
     "**解答步骤：**   "
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "LevojTnFHWoK"
+   },
    "source": [
     "**第1步：自动编码器的原理**"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "5KfOi2QpHWoK"
+   },
    "source": [
     "&emsp;&emsp;根据书中第493页关于自动编码器的描述：\n",
     "> &emsp;&emsp;自动编码器（auto encoder）是用于数据表示的无监督学习的一种神经网络。自动编码器由编码器网络和解码器网络组成。\n",
@@ -651,14 +855,18 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "punKi_CQHWoK"
+   },
    "source": [
     "**第2步： 转置卷积的定义**"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "UcYQQKUnHWoK"
+   },
    "source": [
     "&emsp;&emsp;根据书中第509页转置卷积的定义：\n",
     "> &emsp;&emsp;转置卷积（transposed convolution）也称为微步卷积（fractionally strided convolution）或反卷积（deconvolution），在图像生成网络、图像自动编码器等模型中广泛使用。卷积可以用于图像数据尺寸的缩小，而转置卷积可以用于图像数据尺寸的放大，又分别称为下采样或上采样。"
@@ -666,7 +874,9 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "99QRJpf5HWoK"
+   },
    "source": [
     "&emsp;&emsp;根据书中第514页转置卷积的运算：\n",
     "> 对于任意一个卷积运算，存在对应的线性变换的矩阵 $C$。针对转置矩阵 $C^T$，引入新的卷积运算，称为转置卷积。原始卷积和转置卷积是相互对应、互为反向的运算。原始卷积的卷积核是 $W$ 时，转置卷积的卷积核是 $\\text{rot180}(W)$。卷积核和转置卷积核之间有 $\\text{rot180}(\\text{rot180}(W)) = W$ 成立。"
@@ -674,15 +884,19 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "qMOSuDlKHWoL"
+   },
    "source": [
     "**第3步：自编程实现2层卷积自动编码器**"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
-   "metadata": {},
+   "execution_count": 19,
+   "metadata": {
+    "id": "ZEE8dTTuHWoL"
+   },
    "outputs": [],
    "source": [
     "import torch\n",
@@ -698,8 +912,10 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
-   "metadata": {},
+   "execution_count": 20,
+   "metadata": {
+    "id": "3ysY5kL7HWoL"
+   },
    "outputs": [],
    "source": [
     "class AutoEncoder(nn.Module):\n",
@@ -730,8 +946,10 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
-   "metadata": {},
+   "execution_count": 21,
+   "metadata": {
+    "id": "MX-y4Fh-HWoL"
+   },
    "outputs": [],
    "source": [
     "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
@@ -745,8 +963,10 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
-   "metadata": {},
+   "execution_count": 22,
+   "metadata": {
+    "id": "mFt8PfOFHWoL"
+   },
    "outputs": [],
    "source": [
     "model = AutoEncoder().to(device)\n",
@@ -759,8 +979,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": 23,
    "metadata": {
+    "id": "uKl4cEWbHWoL",
     "scrolled": false
    },
    "outputs": [
@@ -783,7 +1004,7 @@
        ")"
       ]
      },
-     "execution_count": 21,
+     "execution_count": 23,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -795,8 +1016,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": 24,
    "metadata": {
+    "id": "Q3aOx4M9HWoM",
     "scrolled": true
    },
    "outputs": [
@@ -804,16 +1026,16 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1875/1875 [00:10<00:00, 175.31it/s]\n",
-      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1875/1875 [00:09<00:00, 188.10it/s]\n",
-      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1875/1875 [00:09<00:00, 188.67it/s]\n",
-      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1875/1875 [00:10<00:00, 187.49it/s]\n",
-      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1875/1875 [00:09<00:00, 188.35it/s]\n",
-      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1875/1875 [00:10<00:00, 186.48it/s]\n",
-      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1875/1875 [00:10<00:00, 187.44it/s]\n",
-      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1875/1875 [00:09<00:00, 188.00it/s]\n",
-      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1875/1875 [00:09<00:00, 188.76it/s]\n",
-      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1875/1875 [00:10<00:00, 187.21it/s]\n"
+      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1875/1875 [00:10<00:00, 180.98it/s]\n",
+      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1875/1875 [00:09<00:00, 188.63it/s]\n",
+      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1875/1875 [00:09<00:00, 189.27it/s]\n",
+      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1875/1875 [00:10<00:00, 178.80it/s]\n",
+      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1875/1875 [00:10<00:00, 185.14it/s]\n",
+      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1875/1875 [00:09<00:00, 187.51it/s]\n",
+      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1875/1875 [00:10<00:00, 179.53it/s]\n",
+      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1875/1875 [00:10<00:00, 186.49it/s]\n",
+      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1875/1875 [00:09<00:00, 189.39it/s]\n",
+      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1875/1875 [00:09<00:00, 189.29it/s]\n"
      ]
     }
    ],
@@ -834,14 +1056,15 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": 25,
    "metadata": {
+    "id": "z5Z7x7emHWoM",
     "scrolled": false
    },
    "outputs": [
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 800x200 with 16 Axes>"
       ]
@@ -851,7 +1074,7 @@
     },
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAoAAAACpCAYAAAC78Yz0AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy88F64QAAAACXBIWXMAAA9hAAAPYQGoP6dpAAC99UlEQVR4nOz9eZRk133fCX7uvW+LF2tGZuSete/YQYALuIkUKVKUZMuLZMtWu0dttd3d42XGbR8f95yemXYfu49Pe3p6uuecGY/bbi+yLEs2aYkSxX0nFoLEVqgCUHtV7mvsES/ecu/88SKzCiBAAmAVkJn1PufUASozIuq9G/fd+7u/5fsTxhhDRkZGRkZGRkbGXYN8ty8gIyMjIyMjIyPjnSUzADMyMjIyMjIy7jIyAzAjIyMjIyMj4y4jMwAzMjIyMjIyMu4yMgMwIyMjIyMjI+MuIzMAMzIyMjIyMjLuMjIDMCMjIyMjIyPjLiMzADMyMjIyMjIy7jKsN/vCT8pfuZPXsWf4iv69t/yebOxSsrF7+2Rj9/Z5O2MH2fhtk829t082dm+fbOzePm927DIPYEZGRkZGRkbGXUZmAGZkZGRkZGRk3GVkBmBGRkZGRkZGxl1GZgBmZGRkZGRkZNxlZAZgRkZGRkZGRsZdxpuuAs7IyMh4FUIgCwXE7CTGsTCWBCEQUYLoh4gwIllexQwG7/aVZmRkZGS8hrvHABQCjHm3ryIjY18gLAthWST3HeHSX7GYmqxTy3UpOX1e2pxkY2GU3KLFod+xSC5fB6Oz5y8jIyNjF7E/DUAhXvP37Uj3cBMSAqEUAEa/ZlPKNqq7k9fOmVt/NZwrKIUQAmMMaJPOlSHm1r/fBfNHWBYil2Mw6vL+4xf55doznHJWqcmYz1dO8NvOe7kuaxjfRUiBSd7tK87IyPixbO+LQiKUBClBp2uaSTQmCt/lC9yFCJGOlxQ7doZQQ3tD/pgMO63TfQTSvQQwyXCRfAdtkL1vAA6/AOm56Qbt5xCe+6rfG9cBJRGtLqbVhokx2vfW0LbA24iwOiEyiNKwVX9Asrr+o5P9xxgIGXuU7bmT85C1UYylEJ0eJgjAdhCei/E9+odHiPKS9qwiGDO4dUFhUaMig4wMQhucrRCr0UN0+iRr65go3p+HieEmoR88wfqDeVpH4ddGLnCPs0JZJthCUI/zrDaLqJZChDFam/03Dhn7hyw6BEKgThyldd8oQVnSPAFJXmN1JFZXULmgKX/hHLrbA52d5rbXQTUxjinlicYLdKZdorygcwDiHGhPY+xXzyuRCNBgddNxVQG4Wwa7Zyhf6qI225hmi2Sr/o7Myb1tAG5b30ohch7CtjEjJZKCl/7eGIwUxAUHYwncNQuZJITTZdbeI0lcKF53ces2TkfjNiKs5gCxVX+1AZgZf/sTIRG2hfB94skK2lbYGxaibWFyLkkhRzjqsXGfzaBqqNy/zp+ePcc3Vk+w8MIkMpSoQCAjyC9ZFJZtnE0H0WiCNphY/+Rr2GsICUrROuLT/FDAoclNHvMvc9hSaAQJhnbiEbRcvI6EOHmVpzQj413htWv49ua6/fO72QgcjsFgtszaI5J4esB//77f52F3nq/3TnKuO8OXv/0gI9/yMcEAsx8Ptm8HpdBjZQY1n+YRh8YZA6MBv3bfD7jfv8EpZ5VplaBIxzfB0NSGrrF4vHeUF7uzXGzVuLgwDg0HYfIUHIWtNWzV35Fb2FsG4C0eG+G5iHKJaLJMnLPoTDvEOYgKgjg3fLkBI0C7BiPBaVVwmmX6NYE82cZ3IhojBUTXwuorrK6Fv+wxvlyAXu/mvzsMG2fsE4anN3nkIL2jVcKyonlYoh2wujmsALQNiQOxD+HhAL8w4JHxeU7nFmmNerROuAwiiyi0CBNJf9qh2bKx2w7+vfdi9zTFKx3UVgfTaJE0Gnt+0RS2g5qdQpd82gckJ2dWOVTYYiku09ADPt94iJdak5y/Nk3hJQd/zSA6vT1/3xl7gFsiQWkUyCOZHCHxLIJxl0HxZjhOJga3oVGhRvUT5CBGtQPMjSVMGGHi6K6cs1FBEdUixqodaqpFRWoOOetERvHHlZhkagxl2+j1DXQQvNuXe0eR+TxiZhLj2CRFF+0q4pwizkuMACMF2oLulCQsG8JaQnWmwUShwwlvmRmrjo0mMoYlrbgWVVmPS3yrcYK1fpGr66OEdQ/VVuTXBVYH8st9rK0uptt/x+5z7xiA2/kJSiFro+hKgeaJEmsPC+JqzM/c/yIPFOeZs7eYtBpExiI0ae6WIxKk0MxHoyxFFWpWm4e9G3gioactQiSNxKehff6nS5/EfL8Kq2uv/vfvwgVhXyIEwnEQlsXm+8apf6bHZLXF3zn4PebsTbraJTIWCQJtJJ6MmLO28EVMUWryQvJhb5HfHP0uGkHXWGhzc3NZjEd4sT/LufYU5/7oJJXLRUqv+Ih2O83x2KvzSAhkzqN9/wTtWYvkPW3+4aHP0TIu3+2c5HKvxhN/dD/jP4w4vhFg3VjADEKSdvvdvvKMu4DtZ1qOVUnGKwQ1j9VHbQZjCffff42/MPXkzmsXwlF+98bDbDXzxJsedt2jeL3ARH+AabbRne7dle82dHAEI4qHjl/lvvISx+06VeXysLPBcXuT/zj3II0zk/irHl4wgMFg765lbwI5UWPtg+OEZUF3xpCUYooTHe4fX8aVMXlrgCtj7vPnmbHreCIiLyJsoakpjY2goTVbWvF47yifW36I+XoF6/ES+WXNgfkAZ3EV4gTT70OiMf0+OtHv6D6xew3A7RNd3keWimApTD6HcSz6YznCskV7TpJMBYyMdHlf6QoPejeYtvpUpUXCgGgYelIIpBBMqhus2JuUxICDlsHGRisNJLRVnYZuMplvM5Cj7+69Z9wxhFKokQqm4NOvCQ6Ob3GytMaj3nUmVHpiS4CBgba2sYWmKBMkoIHGMCm6KDWa9IeJEFRlTFEqJtQSFdXFVwN+MHUMK1B4Gz6O40AU723vghTEOUlUhKIfMKEiwkRyPahyuTWGt2HILXaQjQ7JZj1dyLJ8oYw7hLAsZLEIjg2VEtp3CUY8+jWbYEQQTMV4o33eO3KNx7zFnfct2Ru8NDrFNbfKol2m73n0AodoagTLdRDxHn9O3ybahslcm3G7hSdAIvGlAp1QcgNaRYHds8hZ6t2+1DuOybn0JwSDEYOZCqgU+xwfXed95avYIgZACsOk1aQi+wTGYiUpAbCeRCQIroU1VqMyP2gc4PJKDb3pMLGsyS+H2CtN9Oo6Jkkw4fCw8S7Mt91pAAqB9H2E5xI8dJiV97uEFU355BYThQaTzjJFO6DmdDidSzfcU/YGZSnwhIVEYguBN4zayqHedU0O8EQHG1DDW9/+nSckealxVEygxN2dE7JfkQpZLLL5icO0Dkus99T5Px38CjXVZloZXGGjhU7z2GLDYlIm0A5t7RFom8vBOKuDIlWnx4xbRxvJWlgkMoqPl1/iQ94qRSm4z2kxZ72A/pjkwnsneCJ/P4evjUOvnxpGe3VzUYpBWRCMaQ7lO3hCci0a48sv3oOzbDN7IUTcWEGHYXqPGW+f7ZST15snP+53t75GyH1diKRmpth6bIZgVNJ4z4BjB9Yo22sccvoU7YAzfro33OcsU5XOzts9EfJ/qH2HxqhPezZHV7v88cl7eXbsGN5angOft+DcK+/iDb47RL7g/cVLnHBW8cT2vmhhS8Xp0grnjx0iziuKL/rv8pXeeXoHShz69FUeq17hoLtBRXWpyB411WclyfO19j3UI59vDk7SiV2u1UdoLxd3ijxEIihekRQXE1Rfc7gdoQY91FoD0wswvR46GLzrz+fuNAAhTc53HPrjNsHxgInxJv+343/Ivc4meSGxhdwx3gBskdv5f43e+Z0cJmAqIXGFRUK0k5R56+vS16ZWfXoBkn2hXfE6kjhCvn4+444kzj7eNITr0JmRBEcHfGxynve5m/jCRgkbSCvytTEMjGI9LtHVLsthhX5i81JrktVOgUouoF3w0AiWeyUSIznprxC4K3hCMCI9ylLza+VnWS84fGPqHnQxhzQGoSQm2btzK3EFiZ9QsAdIIWgkeaw1G39Z4Gz2SZqt/Tl/3kmkuik99Lq/F6+SIfoRCaJXyVzt3bn2hgwLkUw+R2dW0p/Q/Py95/jrta+TlxpfiJ2oj0JgC/tV+4EtFPc6AyLTxxYShSAvB1w9VmXLL5MUXYQY+vzvonlsLJix69RUH1tYO/ujFFC1uiSlhKgtwd61ZsNtI85L/sTE8/xS/gJFaWGL9HlKjE1gAppxjrVBkSuNUdo9j8GyT+mKQsYgYoOMYOyZJpx9JX0+dYIB4nf3tn6E3fdNbm/UIxV0OU/7gOSjJy9yKr/CEXuLolTYqJ0vZBuNJjEGJQQSufN3jaahYwYGzoaTnO/PAGDLGFskVFQPT0SERhEYhwubNSaiJK102mtewNcYezKXQ+TzCMcehs9tgtki/VFFVBD0xwQIEAkIDYUFTX45wm6HqLUGDEKSeiN1Ue+lcXgdZC6HrI0ST1ToHov4xOmX+ZnKy/jCRqPZSmK2tMU/XPwMzy3OMGh62JsWIgY1EIgErB6owLBlCdbd2fRzo7TQ6B/fO8nnj97PI9Ub/I3RJylLh6KQSBVy+OQy1/7ULPn5ChNf1bDVwPT7mHi3LQc/AcsiyoOqhFSdHgOj2YiKeBuC/EqC6O7vxPA7xtBbp0ariHyO/olxNs84JB6EJYOx0jmGuPkMilhgdwQqhLHnQ/zn53f0KYWl0OMj6JyNtVwnWVze2YT2NNvjdOQAwZFRWgdtxAfrPDi6zs+WzzOqDDapcwDY2QcW4gGb2mVUDphSDkqInT1ke7845azwiwfO8W3vGFF5HFep1G7eb8bz6zEcV6MgL0KKMjWSLRQxCYkxPLF1hLEnLfIrEWw23u0rvuPIyPBCZ44Ze4sz9gY1BVs6ZiVx+YPmo/zR1x7F3RTYbcgPDKNNTW69D0kqCya0Qa1sEie7e/7sLgNw24BRCl3OE0z49GYS/ubEV5m2YsrSwUKhxKsFFhOT5m5pNAqFRKCBgYmJMCwlLm3t8c3WKR5fOQyArRKUMIz7bYr2gEgrtBHUNwtMRq13+MZvE0Pv3rYnT/g5GKugXZvBuE/sSzbus+gfChmbaPHrB57HkxFbcZ5W7PH5Fx6gcN4lt+EwogSqPUAGAUkY7j1j+DWInEcyXqE343Py6AJ/Z/LLVCX4Mk9HB2xpi5fDCZ5+6gSTTxhy6xHutWGS7iBMN88o3knQ3RbtFEohHJvCz53myvoBFk5V+IuVpygIQ0G6+Gj+09nH+cNPPMDT548w9kwJGQxSo3ovGYBCIqQkLhiq5S5Vu0toDJtRHn/N4C8Fw4rfTPLlLTOUI2KsQjiaZ/U9Dsc/fZlDhU1+qfIs06qNJzR2Gl0CYCux+Xr3NFf6Nb7uvIeDyyOIWEOi0Z5N+1iJsCgZAeT6Zjp397hndlukOJwdYf1Bh86RmP/nPZ/nMW8VXyhc4b3q9RqNNrCYFHh5MM1RZ5WybOGhcIWNROzsJSfshL888hQjVpffK34KT6Xh87tiOm/vGxKKMsQfGn9KSAY6JiLh4sYYh76xhF5dJwn2dwEIgBoYLrZqTLqHqBXbVGTESuJyfjDD15ZPcOiPApxXljBBkFaOJ0mq/XoL8R543t55A1AIhGWnId6ZSXQhhxxEiCCEQYje3MIYg4gS1EAj4rQiMzSGwMQoEqSRKCFo6pCVRNHWDucGh2gmOTpJmq/Vjj1W+kWCxKYe5BhEFvWNItZ6GuozEhCGhdw4xtLDIzb4NyxEZ1iGvcu/vB2GYyoLeXBsTK1KUnTpjLt0phTaEYQlSDxDfLDP3HiD4+V1jrhreDJizPLoaZcXZme4ntToTyn6YwWsfoHc5ghWX2N3EqxuhGoFmGsLaf7CXvIoCIG2JdoCT8X4wqCQJEazmsT8QesRXmxP469I/LU+Vr2fluPrBBNGqSJ+kux4WXZU25MEtCa3FlK85tLIFfmdo49yT26Bx7xFaspl0mpyurjCC5VpohEPt1tE9Pp7pkfudtcPU/CJKpoTI+tUrS4bic3qoITd1ajW4F1NZt6rCNdFVUcw+RyNB0Zpz0p6x0PuLS9xwN1kUnUoSo0jBLceexMZcdDZQArNFyYS2sfLaEsQe4LEhc4cJL4Bk2esNYts90hW1vZm/ul2QaDvg+vSnUqNv8pUi5pq4Qq5ExHajvxs6ZDHgxlWowpfWj/D1c0qh0e3+FMTzzJpN3jA2aAqHeyhwwDABmyR3L2SXwYCowhMhCte0+nIiPQwnNwdup5OY8CFV6ZZbJYZO9VmspDmhFZUD8+KMTKtE0g7pAxF//fSfjjkHTcAhWUjSwVEqcjKxydoHwS3LsitG3KbCfknI2i3Eb0Auy5RfZeGzqWSLUajMNgifVifCcf44/oD3OiNcPblOayGhbslcOsGt2Uo3Ogjw4SRfgRRzFS0Dq+x0lFyuMAM/wxCkrWNH10kd+uisK1pl89hZiZISi7Lj+XpHkyoHGzwl44+RUX1qFktPBFRkX18EeMJTX6YC7i9sXz0xEW2jnkExmYzKdDTLhf6kzRinyeWDtFYLFG4UuDAv+vB+gY6ZO9MeqVIchZxTlKwBzt5pAMT83Qwxz/7/odwlm3mvh9gPXMJkgQ9NGhelRu5zXB+GJ228LF/eJGJl3IU5w/yr60P4Uz0+AcP/j6/mN/kQbfBUfsprs6N8tKheyhaI+TaXfQekUgRrousjRJOl5k4ssH/eerLrCdFnhvMcakxRnmxj7ixhO4HP51x8UbP2F4zWN4M20ZNpUxwZobeuE3rz7b566e/yRFnjfuc+rCQTQEWkUmLk2wh8YSFJxLe5y1xylnl6/ef4qx9ALsccO/MEjWvwwdLF5m0mvy1mV9D26MUlsv47TZJK9l7YU0hkY4NkzV02WfjAcHf/sgfc8hZ54wd4At3x4gbmISGjnkqmOa/ffZPEq341H4gmHuxxepDh/iHH59kcqzJ3z36RR511/CHKUUajScknkiLl4wxP9omdL9iNEZLZAzrSZGi3MITUVoUxy1r3vAQfDegXpnn5D+foTeT57f/q/dy5vgieRFy0l5jym+x5I3jug6i20PvYaP4HTcAZc6DiTGiEZ/ehCCaDEHYWD2B3Ut7EGpjIIoRYYzdETzVPcaY3UYNJ6MnI2wR81z3IM9uzrDZyuOu2Lh1yK1rvK0Etz5AXV6GwQDdD26eXN7IsLs1fLoXjBqpEFIgcjlk3scU8wQzBcKSoj9h8Ka6PDS+yCfzL1GUyU5idGA0EZAYaGiwMdSUhStsPBExaQIS+mhaBMZwxFmjkfj0E5vvx4pBq4QuFxDBANFqYQZ7YKyGGLHj6EUKQc8kdHXMlcE49oaNvyqwGwG607nlTa8zX177M2PS93R7eOvjeCslAumzHhcJzGrqoZCGmtPhhYIgLCpyjn1nb/Y2IhyHZCTPoGJT87tMq5C2jmgnOfqhTWWQoAeDm17Rt/r5lrXTZ5nt4oftPppRhA6HFcV74bl8k6SpAw4i79OdtOmPS46PbfBR/yJFqSlLh4GJuRJDT9sExiY0Ck9G5EVIgqKnfbrGIdYSbE0hH/BI5QbTTp33565TlVAuBMRugcSRt/RE30MIkUaLPJd4tMBg1CUeibnHXWRCdXBFuoXFpHNjKUk4H07yZOco0YpPbkVSWBwgry1RmPCpL3ssJ4InJ44BcNTeZFpty3oIPBkxKAkqUxOYZpuk/s50ZNgNyAhuRKP4ckBVdl+VZ29ZCabgI3tFdLu99/KX3yKm30ctb+ELwbVmgSvhOHP2JhWriaviYcXo8Hnao8YfvJMG4NBTFd97mIu/7uLV+rxn5jzH82v8m/OPolfz7OjpaoOuNxCdDrNf9fjs6sdBpsUKmLRaSSuw+uA1EyYHBm+9hQhiZC+AQYgZDNDtThq2+3HCits/N0nq5n4jdosXYlghKKsVRM6je88Em6dtBmOG8r2bTBTafGLkBofddY44a0wrQ4RgaRgq/1bnNC93J2iFORqDHGO5Dn975kvc60SsJjHrSQ5fRkyqBE8IDlkdItWhN/Ii426br9gnWf1wFX+tTPmHy8TXbrzbI/Lm0AYVpf17g8SipxO+0jvEFzfv4+nrB5n5Tkxuvo1YXPvx3/WPnUcaeW2ZA180dA4V+PJ9Z3h/7gpVFVGVFsdzq3z2TMygYlG4UoLrd+RObz+TYyz9TJnepOHTI1cpS4eucbjYH6fbc5Fh/+2dgreT+ifG0SMljK3Qvo2xBLGnMFKQW+qgljYgClOB3r0spg07B045WoWxERr3jDD41QaPTCzwK2NPM6mga+B6HPOt3nH+0fc+g71uoUKBiEE7hmTY6lzGpH1Fe4JSH+qnykyfqHPcWcHG0DPQ7HjU1jXeVphu2ntss9ou3kqqJa7/go882eEXDl3goNXaiWAMTLp2NbTDf3Plz7H4lQO4dcORVwbYjQC5ukXS6eK/uMSR1hhR0eaLT36Q3698iJGPrvAPT3yWUdnnoCV40F3A/jNrnH/vBJPfmKb82WfTEN8+Onz8CMaASSgsJvzj5z7JwfEt/u+Hf58zdoAtJDaKe2srnP/UafIrY1S+d4N4cendvuo7ig4jqDdQUqKePcDf7/8ijx2/wt+b/mMAEldici44NkKpYSX+3psj76ABmJbuBzWXDzxwkZ+pvsIJZ4Wq6vHlkVO0ZX7HO4PRaTipp5HnrjBxo5DmXvWDND/QTVXfTRRhur3UXR/FGJ3wUy1vu31j2a6Qti1E3kcXc7RnLDqn06KO//7U73Pc3qQoBe5Q3sAWNm0dspX4rCVFXmjPcH59gqDvELUdlsslVibLnDRrNLTDSlKmRoua7OEJiTs85Zxxl/FkxHKtxJNzVRJHUXo59xMueBdhdKrPpCHWigi4PJjg7NoUZjGHf3ENfXUe/dPo1xlDsrmFaDQp9Y+w0K6wqX182UIiqVlt/FqXflJA+za7NKngR9AFj84BjTPV5Yi7hissImOxFeZJBgq0Hi6Ab/L5GRpB2519TD5HPJoj9hRRQaW5bLm01ZIKfbyGhwkE9AME7AMjUCJyHuFIju6k5C8fe4I/WTxHUUh8adNOBqxrnxe7s1SesylfjVCBRoYJiaeIiumyrcJ0Tic5SeIIurMyTfdQfZSAyEAcWtidBNWL92b4bjg/ohEPc7THf3Ly+zzsX6OqFAqR5vxhhmtXiYvzE5z4WhtZ72CWVtH9fhpRAuLFJeTKKp7rkrs6SVLxuXB4lJUjZRwrASImlOY3D32P8+PTfOnK+6lYVprzu58NwCFuI8bM+1zVkvUDJSK7j8RgC8GcX+f7BzXalpR9b88XBf5EdILu9xGdLoVFg3Y8LtXGSKYFEoNWYCy5U5gEyZ4ckztvAA49VvqR02yd9mmcgj83colD9jpfbt3LjX6VlZfGOXA+wtkK0N3eTV0rY9IKm14/NQCjaJiEr9OFIY7TP7fqYO1jrOkp2o/MMigp2ocEgxGNPdvhZw5c40Bui2mriStgPrZpa4/nggM82TjCUqfM/NUaqiNxNyVOC4qRQQ2gP27z7+ceIRh9nn86/2GuXJ2gOtnkrx77LgfsTc44m5SlIi9iJlWTx8qXufaeUZY2KvRfKeFd9VPjeze3ThICM1lj6UM5+lMJvzhyjaKQrAxKdBeL5NcFYhDenjm0XSEcRqyvVfn9iYf5RPkcs7kGk1aD+yaXOScniQo+zk/+tF1BVHQoHmzyyOQ8h+wNNPBE5xjfe+kYuasOohe8ubETAnXsMM0Ha0S+pF9LCxbCEU1SSkAZhB0hlcGyY5TSbG75WFuzuHXByCsJTjPGe2WZeHl17+kNSoUaKSM8j/r7pll9Pzizbe7z5vGE4JXIZVPn+fcbH+dbF46jVlzmXhngLXfS3GWtsW0L17UxQoASGEuydqRA84Rm5MgWc9YWiRH8VvNhLvbGca565BbryFaPJIr31ngBMu/TO1CmO2kxPbrJvbl5pq1mqvOHJCKhrRN+p/4YT28cxL3qojY3Me3uj1Rl7jybcYxotrHiBG+hyL9aeowHKwvMVp/AE4J73EVGrQ6fO/UQrV+4j9xaiP3Di2maxx4bvzeNEPQmbMbuW+PUyBqHrM00dQVFYgyejNDe0Pss92AqwdvF6KGsS3po9URC1enSqynsboGcECjHSefF9gHLstK51umgu/1d3RHpzhqA2x4rx2b94Tz2Z9b5yOgyv1R4icjAf7d6jKWrY4w/K/C+9zImGKQemFseMhOFJK/xypgw3L8q92+EEMQzoyx+TKImevzFM0/zqeJZqiqgJlPRU4DICK5FY7wSTPG5Gw/QfmYUb11w+uubsLqRSppEt4znqSM8+WAqjbP0nVmOf6XL2iNj/GHpfu4vL1Ir/5CyTIZexT4T+Vf46MmLPH7gCP+fb/4pcvk89Hq71wAcjkswU6Dy0RU+XbvGZ4ovUJAui70K+RuK/JLGBIPb51kyBhGE2MsO36gcZ9xp84lcg2nV4zOjL1C2+5wr3o+zR06MYdnilw69yJ8oP8MhK0Tj8NT6IUZ+YJNfSTCd7k++j2G4t3d8lOVfiBkZbfPrh57ltLfIAavOhIpISGVOJOANc1YbWtPUNr/beJTf+c4HyK24zLWqiI3NNL1jD+UiCaUQ5RLJSJ61RwT/wy/8NgesLU47ISB5YTDHc50DfOuHZzj82Rh7q4W8upCGvrelnWQ6jkJJRLmU5hDO5vnFD/6Qe/OLHLMTlmL43I0H2JivMPmyRlxd3LPdWUzBp3XAojcpeGx0gUfdNTwhsYWd6rwaQ1MrvnTjFIMXK4y+rNEra2/cZcEYzGBAsraOdF1KV6c4d36OzcM+v1p5mjmlud9JuJ8tfvb0y3x1cIbCZY8DF4updude9z6/HsM1sjsp+R+O/xFHrC3mLIkrnJ2qal+GGC8h8STcBa3ggPR71gahh1q5wlAUhimnSTAOMraAPK5nIZKhY0oItKPAgL2sEIlGRBF6sDttlTtjAG5XuDk24sAMuuzTmzK8r7rKUX99WIDgsLJRxr9hkduMMGH4kx+u7d8JcVd4/LbH0RofQ4+N0Diax5rqcWhsi2PeKhOqP2x3J9hKEp4PJ1mJynxh/T6u1UfoXC8zsgjeVoJoddONJB5q2YnUfS2jhKRtc6M9gtUF1YtwWoZL62PEWvKRwsscs1vYpKLRCoMSmqLqoxUIS91M3N+FCMtGODZJTjKa6zHjNrCFJjIJ/dhGBanm020Pj8UxTkvQ2fRZmq2kGpUCSiqgoAapjMBuZ1sgVoIvQzyR0NWG0AzY7PqUNjRuPfrRyvrtt1tWKk1kO+gD4wzGcmzcZzM7tcyBYp1j7ioHrDplGeEJSWA0bS1RwqS9SIUYitJGHPXWcKd69JVPVHJxLAsDe0dLUaYH4XBuhN6ESzIWMmPV8WXERpLQ1jbf3DrJc8szeEsKu9FBtns7GmPbmAQQaUcZkfPQxRxxXnMkt86MvYVCMDCSZjuHs6lwOsnwmd+D66UQGM8lGBUMRjVjdgdbiJ0uUAEh84nkSjRGe63AyCLkNqI3Lvi7FWMwiSa3FePP26wWyiwdL5MXm0wojS0U014DrxoQrln73+s1fM6LMqAoNfI1pkFiJBiGlXS7z5C57Wx303Fd+qOS/lTC6UITWwim7TrBwQFR3mFQtbE79jC9yIAAbaVre37Fw9us4Wz1UQurEEboXm9XHSLuiAG4XeEma6MsfXKCzpzh6Htv8HenvoTE0DYWZwezFJ7OMfPldUS9Rfzjuk28zilu33OLXmL7/QdZfVQRHwz4+w/+EaecZeasiIJw6JiILa35Zu8Y/+i5TxFveIz9UDJxqc9Mp4XcbEEU3ezoAamHSgmEkhDF5BYt5k2N8VWDbPUo3sgRPFHi4kSRF2tzvM87hy0kLhItNIFJ8ESEdgR4LuxioWhZKsBImd6o4mOVeR7xryAxbOiQjU6e3KbGbaZhodt5/SYIqFxMsDsOz0zPEkwmSGBSNRl32mnB0y4cr1sRSiEsC20LylYPTyRcicupRND1Ege+v4jpdEk63de8cXgALOTRxw8wGPW49suCX3jP85z0V/ho/hXyIk6NOwSBgYbWrGuXs8EcSmjucxeoypC8FFSlw8f9S1Qe6PF4+xjf/sH7GD9fwPT6qUj5m6nUfjeRad6uHKmw9FiO4J4+nzh+gYNWj7aWfLN3jJf7U/zwm6eoPaPxl7rIa8uYMBp6sm65l+HhVzgO0UyVoObgTXf5TOEcvjCAxab2Mdd9Rs8bcktpKHQ3bTpviuEGHNV8Bvf1OD6xwUP+NTyRdvCQCDaShD9oPcJzzVlGn7aY+OJ8Oh/fZOciE0fkf3gD/2qZ5fYYX7v3DN3CZR7zlhhTNu/NX6ZzxOWznYcw7t6p2n+7GAUVGVCUCldYKCGJDCQkaAQkYliIuYfm0VtlOz/ZshGeC9UyzQdDfu3h7/PB4gXK0uMT/gK1j/wL2jrtIx0ZRWgsIqNQaHw5IDIWn197gIsbY8QvlTnwFRerESAXVlNHzC4pLLozHkClEH4Ok88R1EDPBJwqrzJrwVaiOT+Y5Gx3ltymhuW1H13kMlIPnZNOwv6IIpwOOThe5z53kYOWwcZGCcFmLLgY1Xi2c5BkOUduXVK6HuBcWsYMQpJeL01ifj0B2KHnTkYgg7QVUFLOk7gSEadu74RtrcBUbHVgND1j6GkXsdudCkIgPI+k7BPnBWNWh5IYECFpaMkgtCj1Daqf3H4PiTbYXY3dkfTDdPNQgBQaKTQ/ruB8tyAcB+HnSByBJyIUhkaSZzWqoHoCU2+ie72bC9n24uk4yJyHqJTpT+TojynGZjf5S2Pf3WnHBQ5NHdI1htUkx6bOsxhVeaEzhxSa0FiMWy0OWZvMWhGegDPOCg3f5+s5Aa5zM5XhVoNvF+p1CttClkqYcoGgpjkwscUxfw1PCDaM4uX+FC+1JvHWBIUbvbSAodtL5+TrRTpE2o81LNsEFUU536cmBSAITEIjyWN3BG4jRvZC9F6MlohUtiZxJOVihwOFOkUZvKp3e2AUV3tjLLQreHWdCl2/FUPXGHSrjRiEuPUqi/0K826VwF1CIqjIHpNuE8u9S8ShBdhoFG/g7dwLi9ZtQCiVrl+jI0RjBQrVHg/mrzOjmmgknlCctJtEpkmEIDKSZPhfTyTUVPq8BcamYB/le40c3UkX11Pk4hqqk8e0OuhON9VgfBcPZ7ffABQCNV6je98UnSmL6vtX+DOzz3GvN09bJ/yz+gf4vT/6EP6yYPKFrZtJkhmvQjo2HJ0jrPps3Wf4iw99nxPeMpMqwcamZyIibfg7136FK187jNOAgy8NsDsR1sImutVOxzVJXi1oessmLQp5dN4jzoEuxmz+bELj04qC3+JoZYvJXJuP51/CG2ptaTRXYvjdxgf4wdYB/DWNabTSjha70YAXkvDYBGsP5midjjnjLVCWEV/onuZifwJ9pUDx3Dqi00vbG91mjEg7zkhhUg1GUsO5l+xy43k79Hv6MPVTRbbuhTl7Ewn8sHuI5xppUcZOAdbwPTKXQzg2+tgcG/cU6dcE5rEmh6pb/NrU9zlihWgEGzpkKc7xT9Z+nqutUa5fHid/3UL1wW0aMPDd3CNoW9B+X5+/fP/jzDqbfDB3jUm7wWAEotnRNMem2Xr1+rELDUF5cJblT07QnzA89OhFfn3ySeasLQC+0zvB73/5/eQXBBPP9pA3VlO1g53ittd6N7fbxlVZ+rCFONLlP515EV/aXI9DHu8f5uv1U5SuavLnVzHN9t7z/pGufyKXIywp7hlb4YPli8yoDkqkGjgDE3MunOTr50/hLNscWOnzdrpU6GCASBLyyxFPnD3Oi5OTHL93hWmrDai0MwikIeO3Uum+lxhKWAkNPWMRmXinE4j+6XQ19hwyl0sr9O8/xJU/ZWON9/mvTjzOo+4ibWPxROCylhT5YfcwW2Ge59ZnaLR8hDBIpRktdfmrh77NUWeNUdXhfeWrTD3U5NyRKXqRw0IvRxCMkP/OHBOPN1HNbtqr+13yCN5eA3C46JqiT3vWojsj+NPT5/jLlXP0dELPwDP1OWa/FuJdWUevb+7e4oF3G9smHMvTH7exZ7v8xsgTFKWgLNPa0Ugb2kZw/to0J77QRDZ7mOU1zGBA/HpV0a8KI6WSPMJxMK5COwbpJfzMsYv8ubGnKMmAmurjCqhIC8lNRfj1JM/Tmwe5ujbKTCtJPUC71IAXUtCvOXQOa0qTbSZVB0/AlX6N57dm8NYE5voCyZ0yYLc1xoVJi3RMeiocaAuxy/cRIQXBhE/zmMTM9BlXHZSAG/0qN+ojOB3zIxtu6rH26E+m1f7xdMD/eO/n+WhumaJ0cEWepu6znsC1aIynFg4SrOSZeFJQfWoF0QvSVpCJTqWOHJsF5x6enDvMvSWXD+euUZE94oJhMOKgmt6PuQN2jfBxUs3TuDemNNXm1ya+zy/4TQZGExi42B+n9oyh9EoDsbhGspkahm80H4UUaVi+6CGPdvjTx5/nQ4VXcIVNT2vOdme52Kjhr0bENxb3ZqGcEGDbCNch9iRH/Q1OOcsU5bDy1yREJCxFI7gLNoV5sBpB6ul8q/eqE8wgwa4H+NfLdOISiyerJF6LZOgJE9sP6170pL5ZTHrwiowiIiIyyauEoAF2/aL10yLSZ0t4Hp0Zh8984Fk+Vn6J93lLTCmfc1HI5XCcG+Eo3984yFbXp3uxQm5NYBRoC5bHfc5NzFJSAUXVp6o6fNi/wOHxdO60dcJGYvOnmn+N8hUfTwnEqpUagO9C6srtNQCNSatgcjbBmCAcTShbPQAuxgWe7x/k2maVA+0Q0+3vqQq+dxytkZFGRgatf9SbEQGRkfilgOaJIk4nj1srIYyhN51jUBom7QIyMagQMIawKIlzgjgHURHigqZ2Zo2j5U0eLV1l0mrjiWS42LITcmnrkKY2fLfzAJdemCW3InFXm0ONrN27MAQVSe5Ai5Nja0gMW1rxlfmT9C9UGJ+/A6HfIcYYRGKQMQy0JDKayEBoFJHZvUUztxL7kkFVU8wHSGEIjOBSY4zeUoFic9vTJkGA9FzCB4/SnnNpnITqA+scKW9yyN4A4KVQs5K4fK31EH987QzdRo7CeYfRTUPpah9aHcxgsLMmbC+IVs+w0CxTsgN6RtHVLjIUqCBBRDFvOPOGXo3dQOJZuKN9jlY3GFVpl5meSdhMBBuDAnY3Qbb7ac9peP1NYHi4VrPT9E5N0DpgcWLiKvf784zKPgNj8WT/CJ87+xDWssORzdYbj80eQOZ9zGiFsCioWl2KMmRbOXM5CbkYjfJ4/SiFG1C6ESGbnR29v7eDMAaZgIjFzvOZFyGTVhPXizCei/T9VJ92F+Ru3XZu8ZgnxqCFRiNIhmOaGHkzB3AfI/wcplIkygtGnQ55OeDxYIZA2/ybxfdz8fwMqi9x6gI1gLHVYQ75sPgjvCH5XO8D/F7hfRiVFoW41T73Ty8x4vR4qHADXw6YPrjJjV+oYbWK5BcfwGkbqs/VEcubmF4vday8A8bg7Q8BG0PiO/SnE7zJLjWrhTaGZ/qH+P2lBxgsFFCbKySt1o/qNGWkbHcLiBLUwKCTmwZDYgya1JgIjGK20uDqvSVU38JtWGgFnUf63HNgGW0EsZYMEov1dgGtBfdPLfFQaZ5ZZ4v73EVsNI7QSCA/TMqXQ/0nNbwOjWY9EVyPq3xt5SRzX03IXW8gFldJdnMfRJFqzf3Jw2c56q3hCM3FaJTouREOfGeAu9AcXv8deNB0arzLyBDHisCYYb6IddMA3G3FCtsMr2lQlFgTXQ6O1LHRtLXN6nKF0iVFbi0Nme/IkuTzrL7XY/Bwh/cfvMbfm/oiFakpSovICL7XP8b36sd44uxxDv6+YWYzwJofSpyEYZq4f8u/vV3s4LQMqxtFrjoRm5M5esZFBWB1Y0QY/fjDxy4Z2zinODWxyEdGLzKhOoBLWxuux1VW+0WceojZqv/4XsrDVpWDQ2Ms/oxFPDXgr088w0dz89hCEJiYr2+dYvwrNoWlAXJhjWQvev8gTT8oFQimCgxGYMJuUJUJtrDQaF6O0h7wz96Y48iLXdTFhR8tRHqraJ3mQoepAajRVGTMnL1JOReQFItYhTwijvdU+8u3SoIgARJe3dkiMgoRC2QC7MWK8p/EdtpLwSes5RlUBDNOnaLs83sbj/JKY5zm1yc5/a+vpIolYZTKvmxrSw4pCMGo44AUYDsIJQnOzHLufScJRwwXHhrnnsoK//mh73LfyQWuRWN8v3uEZ7bm2JSzjD4rkGvDtYA7//zekSKQxJWIcsh4qUNJBkghUoX6XIdrxZhwuoIDmGYL0w92v5Dwu0GSIHshdsci6VhcjEapqTZHrBgpxFArLeFgYYsLM5OEA0lYURgFM+MN7i0tkSCJtaSvHcpOQGwk9xaXOO0tMmk1mVMaKQTRMLnXHnYP2SYwMRtJQtdYfLVzL4/XjzB/Y4zjWyGy3UWH4e4PMQnwVYgSqQHTSHysHtjNwZsXMH47yFRWwUiQ0nCrz0/tEs/UT8JIsO0ER8YoYYYyEOzkL6Yq+MNwnecSlg1zYw2O+huMKgMIXokkm0mRL63fw7n5KfwbFt56C9nspcbfT9BWkwkQSoLIoq09GomPCkH131h+ZtchwbMiijLAeU3ypzYCZYDXM2S3pSiUQo2NYkp52nMuyUzAxFiLSauJJyQRhp5OaIUebktjtQa3var9HWM7jcixiQsK7Ro8EWEPf54Yw8XBJE+tHiRZ95CdBiYY/PRpKEOtNzToYU9SW0BeROTtkKhoY5UKiF4/zXnep6g38BuvDkp46wpvwyDC1yko3Ots50EOIlQvwtt0+e359zLun+a5+Vmihsv4skE3Wzcl617nMwzAcH4Iy0YoibMxQn7JweoLrpXHWR4psTGTp13JERlFUQXUvA6LcwIVlim5Fqrbw8Qxut+/o2N9RwzAwYjFzxw/x4crFzhpb+IJlw/nrnBoaoN/Ywd8+1P3k1v3GX1xFHexiWy2STY2f3Qj2HZLv+GpWNwUhP5xr9trGIMOI9SNZdw1j/zVo/zzlQ9xorDGr1W+vyPMXMHwfxz/Bj8/cnYYWrRQaObsTaoySF34CLQRBCb9qqsqoCLBRuDL7erUhAQzbB2niExCz0Rsafj3zfdwsTfOd753D9Pf1hzfCLFevkHS66Ue3N085iattpUYAm1zOarxYn+W3JpBXl+9o252IQTakcSewLHiVL/MGGyxR4wWwFjguxFlO8AThkgkCMugbdCuRPhpK0CR90nGSpijXf7OoS8xbTXxheKVSPIP5n+By1tjyK+OcPyJJrLVxKysY6Io7bf5RuG0YdK9Cg2qK2m1fV7sz7EWFsmtG8TC6tBjtssPIICRgrLdZ9TqYA+XtPQAF+HIJC2ufB1dSKEUslhE+Dm2PnqA+kmJPtnhHzz0eebsTU7afVxh00hC1rXLarvIxHIPtbiRVv/vYeKKR3vaIhxNGFUdfJHmIUck/Pv5h0n+cJSZVY1Y3STp/5QHOSEQWiPDNGUDwBaKsjR4IuZQcZMfHpjDyFH8/iDtCAK7ft69HaTQOELsRIACExMZzQ9XZ5n7UhtrtYFeXX+3L/POYAzJ2jqq2WJiuUR0bpS6qnCkGyHDALHVJH4jgfHXfA6kMkMmBnl5ntp6Axwb/dU8Jmdz6YGTPHPsJPFkyM+efpkpr4n1mRfpRC4vffU4h8JZZKuHWVxO5dvu0Fy7I1XA2hYc9jc44qztJO6OKYUrWtxTWOSrU6cxto2/ZiPDAjYgg6EncDsPRg5Pv1rfzNN67UM+VMTf7gX8TrhM3zGMRne6iDDCaRgubY2hjeBivsakalJTfXxhqEnNpLcKsFO8vz1KidEkpPIj24bHTR2tmwnykvT0s/2zBEPPGLaSHC93J7hQHye/ICk+v4Tp9tHt9t7I3xzmpymhSZC0tUcncVGhwXRfp1XUbcZIATJNIk89q+nclOI1h5xdOmeNBEsluCoe5oMakGbo2RRpIZFMwyZx0aVSbHPG2cQGAiNYT8pc2hqjvVLkwLUYnn+F5JY2j2/uItI/xgjaiUcrzqXfX6+/+w8gPwYl0mfStWI6voVdLCItC8LopjFoO4hiHl3w6E5KBrMh90yu81hunjHpYIu0CCYwkkbiM4gsZC/E9Pq7tjDrzSBk2k0h9sG4CZ6IUUKgDWhjqHdz1OZj3M0BJrjNh4BbPsZGgYCCGhD7gtgfdsG41elwl9AfOFhLWyQrq/tatcMMBmk6SqeLXF5Jf5YkO7mQb+3D0vfobhfdHaYoDPV9K/l70ZZLG4flQyXsfMJj5ctUVI+/N3WYwVgORwnkloeGO1YlfPurgIVEW4IRq0tFBkPVdoEnLJQUfDz/Mr33uqyEJZ55cI6lbo7+4iiFa+M4bUP5SogMNYNRmygnsXsapxkjI43qDBCxTmPv2qALLmHZQ4YJzrV1TKeTtpPbdtHv0c0BuJlbEIaMP9Wi1RjhWr7K/2XiBHHeYI52ma62uHdkmY+UX0YbSWBsQmOxERXpJC4DbTHQNhW7x8eK56mpLpMqYUS8WtRUiVSvZDvn71Kk+A+N93G+NcUr3zhK8Zph4kIXs9W42UlktyMVQkm0Y5iy6zgioatdQm0hEm6K495BRGKQEQwSlXpYRer1cWV8U1JrF8/ROCc4Vdpi1q2jAI1AWZrEHardJwl6osryz1TpTxg+NXGWopA8Exb5Wusevrt2FOtLFQ7eiMm/tE6c3JTU+IkMw59hUWImAsrFHgNt0Y5dZGTSMMwuLj66FTXQvLg1RUENmK7WmVCaslR4dp+/MPkU/9ff/CWixjQiFIg4rShEGYxlkPkIZSccm7jG6dIKD+RvUBTpsyoRaAzPDOb4av0MwXIe0W2SDAZ74xn9MYQVi/6UpjCaCpCDRWQ0gdH0Wh7+1Sai2Unlm36aZ2gn5GwRFQVxIRXyvRUp0kOPVtzsQrWLn9u3jQAHjY0Yzq80JWingeB+lsK5FZPmPxrzEyKQb/OzTRzhXlplolGmfNVn6fohLtcOwy/Bnx99ko8/dJ7vjBxBLxaZfPwU3maEe3GFZG3jtvcVvs0GYJqobCTk5YCiiLGHLe8tFJZQ3GPD/WOvEJmE+sQ3aWvDPzvyGP9x+n7aGz5GOlh96E4LwrLBaVrk1hRqYPC2bFS43ZzZMKjY9GoKKzCMNIrI7ZZH217Eve4RNCb1tD13ntJZhcjlkNUKupxn6WNVFg74bB3NMep0SIykEfkMtMV8t0I7dAlji0GsqPp9Drib4KxSkc3X/ae2jT+AxbjC4+tHuLFS5eATIblnrmN6vZthpd0+pkKkxQlSoi0Yt9qERtHSOSKtkMmPSpjcCWRiEEma57X9LzkieZW22K5keJBLXJjJNZiw0zmTIJBKox2TGinGkBRcmqdjKtMtHilcxZc289Eo31o9xtK1MU59r445fyk1/t7G/cY5KJd7jOZ7REYRxPYt398uHb/XoAYJ640CF3PjNMo+ieniC4eyVHzKX6T6yL+loX224gJt7aXtuFRAUfY5am/ii2Rn/vgCXJGuqUpItEm4FEzwwvoUzpZKe1rfwZDRO0XkC5JqxFihiz3Mm0wwqTHSV7Cwkqah3KZIhLFVqoea0zefzyHbBuCeaN/4U2AE2EIPc8xTx83O7wzDooe7yPN5p54hY4gXFmFhEc/38c+NMDg+wY2Pj1Cr9fjPx7/Fr4w9zb+ae4znN07jr0pq60XE5laa1nQbv4LbLAOjMVpidzVf3LyXRsXnU/nzHLTYad8DaWgSwBUSpOZ+f57V2RILIxUuehOYUOKNBJRzAzp9l42WC5FE9hUiUQidaqolOQ2lEDOQtA5XsbpV3IbBbWlyqyH22SvofrAvFsRtb6Dp9pBA5XIJp6Xor4/wr69/GGFIu3dogQpARmLYnxAWqhW+nj9FvZSnWDjP1C0VCRpNT0cERvOF7jGebB3lh6uzdF+oUtwQeCt1TBDcXGj3wDgKx0EOe1BH1YSaatM1DltJASneIJ3gduPYtGdsOgcEB8tNXCEJTEJLe7Rib3dLag2/Y6FhoG9WLSsMQpib3ktt0k3R0eTdEE9GSCS+HFB0BmCnzdHfsqdOKlS5hPBz9McEH5xYQApNN3ZpDHKImD0xD7exN3o4z45ybsnn37nvQ489zUGrzhE79cSPq7RYriJ7RChskuFBISYykqaRfD84zMX+BAfcLe7z5qnIPgetHlIInm/OUL86QnmVvdMb+Y3Ylh1xBF4hZNTrYg/jsgoxTFdJZZZuqwdYa2QMIhE7+n+3IswtMnh7aO69FYRJpcW0SXY0TFMvoN5Nuur7CpMkmH4fe7PH+ecP8pvBr/PQ6CIfKb/MTK7B9+/r0Z9xkVGV0kgOZ6lBfG3+tnkBb78OoEnIbUQ89dIRLk6MMX2yTkUuUpQWBZnmrGwbgAXhUhDwi/llPuEvEBhD+6hEI/BFgi0Yyp3cfCC3Cxu22674IiEwkuvxCJtJgS9t3cO5jUnWzo5ybHUMub6F1mZvVxlvJ5Vu5ydsNfDXN/FVGubEsm4uStuLotFDwWdJfHKOp0aPsHkwz0FngwectZ0TXmQMDa3Z0g7/r1c+hn5yhMK8Zup7i5h2Z0emYy8tejLn0blnjM6Uoja3wRErZkuHzIvRtJp1mzt4TyKXo3kScqfqfKx2AV84bJk+K3GF9bCAjHf5eBqNSKCfOAQmNVQkBikNsbrpDTGWxPYjxnJdKrKHRFCSAeO5NpdzNYx862LMQimoVYkrPv2DEX994mtcDMf5t6vvZbPrMxbtLS+EuXydg7/dJZmo8HX/DIP7LT428jKT6gpSCA7bGghRw2DbdqgzMOy0yfu3C49y7co4hfEuP3vgAjNunc8UXqQoE15YmKH2tKSwOMD8OCmZPUSUF8xWGxz0t24WzgiBs22Z3C5v1FC7dluzU4Q3dQA1GjUsokur3/dx+NMY0BAYi8BEuLcIQdtCpmLY+/Xe30VMGJLUY1QUc+RzPv1ajT/6uVFOPrbCPf4iH3z0IovRCP/Y/xTtAz7jz9l480uY25SGcEeqgK1OhLNcYCuu8AejD7FeKXGPu8hJu5VWQw7lRlxhI4f/tYXCN4aKTB9qW7hI5FDzLklPIsNjSGQSEpLhZ+SITIInNmlbTVaLZWKteKJaIR4rYGmNCMO9bQDeytDI1v3+q3/2WoRAFYuIUoE4p1BeTNEO8GS6yaSdPSQdE/FUMMeVwTjtlSK1ZYO/HmO2Guheb0+F2nYY5qFqW2CrJO3CMSS5E3kdtyIV0rExOZeolHCi0mDMaqNJjeynW4d5eWsCP9gDRsxrTv0aQRRa2D2BCtMQugpi4o0cl9wx5sdHGXg38ETElNeikA8YTBTx56Yx9Waq1fYmFi7h2IQzZXoTDm6lS17EBMbmaqNKZyPPZLC3vFwmjjHNFsq28JZLPD1ygE6UNpH35YCSCrBFvOP520wKrMRletphPqjSiHIsrI+gOoqwYuHKGF+GREbS1YakZ+HVE+x2mBpG+wSJeVXBlDaGxBiEuQOFUyZN1xBDGZjEmDTkbDSRUUPvIHtvLXyzbMvsvOahT/dfnTYjyNyAt5+h/IwJQ+yNHiLJ4ax4fGn9DAfydT5deYFR1aFU7dLuKQZXFd5tTEW4IwagvHiDo78zSVJ0uXDyNGdHztB9b4//7N4nGLPbnHKXKIkBs1ZAWXo7rX1g283/aq/B9t+3/2sLkLc8iEoIqkpRNAm/WDzLY/5FQq145bET5FdyjH5Po6+9M8ra7xg/7l6kQihFcuoga48W6c4Z/szpJ/i50lmOWE3AJTGGgJBnBlX+3nf/DN51h7nnY4rPLWB6wd41/mDHeyUSiBJFZDShkXS1Sz+x70wfXqkQUiCLRahV6R8e4eCxNf7a7NeZsxr0tOEPW4/wnS8+QH7J4F/dYLen6RsFJauPL8OhJIRCr3qUL2tyS310P0BdWuTYb88STOT55//FY7z/1FVqqsuvVZ5izG7zTz75SfKn5xj/wRj2i9cwYfjj5VuEQNZGufLLDofvWeI3Jl4iQfDt5gnir44xvZTgXN9kL5mAJo7RvR5iJebw7znElRwbU4f4J1NHiXMwGDUkjkHnNFgGZ82icANUAF4jQcaG8pTFoCroTyp+rnSWSdUhMpKlpIi9ZlM4v4zp9kh28p/3KMP8U5FAJ3LoJu5NVYPtHMDbvSQZg4gSVB9UcNMD2NMJDS1ZCUrYbYPTjG/JL99HbMupAQnyVetSQ8c0tSKOFUiJUGrv7gu7lR3ZtyXsFZvD/SlaP5zjW6cPI/+04XR+iV8/9n16h13+7dZHqVjD1nHmp99B7owQdKMJrQ4y5zEaHCQa8ehP5njh4AyTXgtHxFRUD1esAUHaJcFsC29KbAGKV7fL2vb+pZVvN9nuUesJCxuDpzQTKuRUcZXnx44jY4nxnOEE3+NFIW8SYVsIy2Iw4tKdMcSTIQ/51zlpNynKm+GNyGjW4iLuok3lkiZ/tUV8YyH9kP0wTrfcgh6GcrR56yHJN8N2j1bhuSRln7CkOFxocMbZRAKBgYVghMINQ3EhQrR+ys4F7wBGgCU1jkhlYBIEaiBwOgkyiDBJgm62sV66RmFzjAtbJRbjEuOqw6wVc8ZbRE736ZIjuO7g5H1QEjEYYPQtz+P2pq8UwrbQJR9/rs2fm/kBM/YWPW2x3C9TXEwoXO9hftquD+8CJo7TPNqXLyOloDw7jT9TJSpYdGZsElcQ5S20A4UFw8j5NrI7gLXNNNfyQ8eJihZCGOasFhUJ12KLRpLH6glMvZkKFO+TJH1hDImWRPoN2ia+A96onoEt7dGJXKyBQQ0SjNa7Wrrpp0UbSbSdTXRLxykAbCvtHx8MbovxkXELOiFpt0FIVBxTXCkQ5eZYCYoczDmc8RYZlV3+Vekj8DbSat6IO2IAAmlByGCAXN7AbXjMfHOCC9dP8rIt+GruvRgFg6om8XUqYCcMudE+f+X0dznlLnPG2WRCucCrK1Rjkh1xSoXAHmazKZGKGGsE0hiOeauoIx3auVRHa78sjD8JYTuIU0cIJvOsvtfhxGNXOVZc54y7TH4Yek+MoaFjFuIcZ3tzFOahfKGDXG+k5vReX9zEsAuHBWqYUiAx2CK5KcFyOxdxIZCVMqJYoH90jNVHXfqTmj9fvkpZKp4Pc5wN5nhq+QC1qyHufB3dat+ef/tOIBXCcdAOzDp1alYLW6Q5t1EloTupcBo51FAPzQQDRL1F9cvj/LXrv4lzssXfPvMVEiP5pRNnWT9U4DvlE+ROH6R0TTP6+DL0g1TLzxjEZI1ktEB3OsfmacWgpvmzh5/kfneeP2w9yPfWj3D9lUlOXOuilrfQvf5PvofditGYBMxWAzvRWK6Nu+FjlEQ7Ks2pbATI1S1MnECSIHI51h+0mPjIIr9Uu0JNChpa83v193KuOUVu3aR9lPdKZ5Q3gVECz4rJqXAnHpSu9wYjTboJCnnbDJFwokDj4ZCxyRZnvEUA/l3zPfze1YdoX6xw7HIXtdZAtzt7f318LcMwpN01fKtzitO5RT7sLVJVLkUpsEXImZkVFj59mPzqFKUnrxOvrO6/cdgNDNdTtMFtJLy8PoEUhiPj61TtzbS/8G2sxr6DBmAqYZKsrgHg3FhkXN50NQvHRhyYJinn0LYicSX1EyW+OnaaXtWlptpMvObwp9EkxgyTpA2OAIzGFmCT5hNKFFIYDtkbPDizyAtymjjv8AbnyP2FEAjbonegSOOIRXymy3938PepqZCqtJC3fN1tLbkWjXG1O0phKUZemh8q6u+Ph9qoVE9NDUWYlTB4MsKSyY/ktv3UCLnTDaN10IFHmtw7tsmjuSsUhMt8NMp36sdpLxeZu7ZGcmNh9+q03dJ+TNuGmtWiInsoBJ5IUMWI/phFoWhhKYmJNTqMEI0mta9cZ+zxHAu/NMH3Zo5zIr/Cb45+l6LU/Iv8Gt85fIzLz85SvlhCNW2EEKA14UyF9pxL4yQ8+vHznCks85niC8xaMS+3J5h/YYryVYmcXyNe39zbh7ntKtdWC9qvPgTI4dqoTRrXEEohfR9ch8HJPv/y5L/BF4KSzLGUDHhi7TCLyyNMb+mb7an2y/M7bJ+XU9HO2i2FSAXVJQg1DEfepmT4oGrznpOXed/IVY7YW4DNt9aPE/ywSvWGQV5dIt5q3BEx3l2BMVh9ONuaRiN40F2ibAy+sClKwftHrvJP3zNHf9GmeKEMq+vcLRG1d4xt4ejBABHH2O2I7obPZXuUoGZTkaQG4G0c8ztnAL6WoUQM2xkGsUC2e6hEo2wL5VjkNm2ubIziqYiH/WtIejtGX89EvBLl2EwKfKt1ipdak9xXWeLPVJ6mIkNqUuPLbX+gwBYxeRViWUlasbjP1duF7SCrFSgVqJ+waJ8OeWB6hYoM8bdFPW9p7fNUcJh/fv2DzC+McrwRDZXG98nDbDRWT2N3BP3IShubD1HbMia3KyVgW7B4bpT6yRyt4/Dw5DInC6vYIqGuI77ZOMVTLx8hf92CwR4QMDZ6KIsh2EwKVFSPhAE2htlaneuHLXrLFjnbHlYPxjsefyEEhQXNV5+5h68XTvCFyXvJ2yEXlseJtzwKyxKkQBc9otkyiSfZPG3TPZjgTXW5r7jIiNXlm72TDLTNM1cPULkkKCzHaY/N/STCu30fOy0vX2d9UgpjKaQ02KResIGJWU/yLK2M4F138bYGu39OvRWMxl/TvHJhhq05n1+tfB9PREghcIWkMNGh8zMn8dZDrLNX0uKit2uY3dK8YNTtUrPaqfSODllolCldM/irUZr7t4/3D0Sq32vJVAdRYV4VeVN3JHE641Vsp7n5PsL36Y06VCZanB5dQ6FZii1EJG9r+sM7aACaV7nrzSAhXlpJBXtVWrRQkofZfKXED9suH6pe4pO5K2ggImEhtvg3m49xsVVj4TtzjL2Q8Nn3HqT8mT7HvBUedJeoEuILG1865MWAqtOl6A0wVv4du813C1nIEx+doj/hYX10k//tnn/PpOowpRzsYfu3xGgGJqSh4XeWHqXzuUlm1zT29eV90T1gGxPFuFsDjHJZ6Xo7OaMKneqIDYWi31b06LWhYyERtsXGgz7Rx5s8PLHMfzPzBaoypq0lV2KHr710ioOfFbibHXSjubuNmO0ONHGMHMC1YAxbJDzgbFKUgv/i4Ld4ZWKK317/KGOuk7YcSxKMNphOF9MPGPlWTOW5Itp3CabHabqSmZ5GBREiCTFSMBjPsfJ+i8F4wocfeJG/OfnVHVmn63GJ//r8n2VrscL444qxL1/BBAFJax+G3+CN70lIcGxwbJSV4EuFNoaOiXh5cJjCCy615wZ4VzeId/OceisM51/lhS3sToW1h2ucPTpLUV5lQgl84fDrx57mX/zF9xNfLnB8vYa4PsBEvD0jcJh7mjiCY/4aR5w1AqO4HjsMrhY59LXrmF5/WMG+D8b39RgawUZCXoUUVJAWWr5WD9Fs/3mL7Rwz3hRCpW0G5ViVpFamddDiLx59msf8i6zEFZ4bzGH1xW1Nf3jnDMDXQyfpoSpJQClkL8TqCKK2RSfx0JidHnxSGCyRoITGyG0vTnoycYYnlltjewmCfuIQxhZechdMVMcmKjkMSpLxQofjdhNfCJRIv+LEaGISLkY5Xh5Mc32zyui6xlsPU6HnfeVBMMh+jNW1SCI1LDJSSKGxZIKRpIcOePOdBIY9HIWSCM9F5PNgKYxjY1yH3oTh5OgmJwurTKoEieCpcJIrYQ25YeOud1HNPnqvCPVqg92B5+qzREbx4dwlKlIzaTXBg3A0QR+eRnYCxEYjFSDe1gZMNKLZRkUxVsVNHa0S4rxC24LEkQzKgsFETGG8y6n8KnNWREPDK9Eo54MZttZKeCsWuY0I3Wim39N+Db+9AcK2oFomGs3jOgEKQc8krCY2C2EVp2WwmwEEg5/8YXsM0e3jbuZw6xbPdA7iywGPuov4Fsw6WxytbXCu7mF8F+m6qdbrW50fQqBG0tzdsCwoqz55EdLQOdo6h+oLdKs91EHdvx4w6boIxyHxBBW7R1EG2D/5bfsa4bppXrcQ6f4YpUVcZnCHnjUxLCJ0HHSlQDCRY1CBCatJXkRcGkxwoTuB1Ra3Verp3TUAb8Fog2j3qFxKcNqK8++ZYlB9DonEExYHrZj/bPS7rFUK/P8+9lFeODHNAzNLfKJwjqoMqSkLd9jjNjIJi3GF5zZn2FgrMbLHdMPeFsU89RM2vSnDh8vLVOW24WeIiAhMwpaGv3HuN+g+M0bpiqH8zBKm00M32/tqc9XBAHVjmdyGj1id5Xw4ii1iSjJgxOoRewJZyGP6wU/Om7qlQlWNj2HKBdonR1h/SBLnDboW4uYiPnX4B/zZ6tOMyj4FYfNiJPjb3/5VChccZs9GqCtLmEE4LN/f5cb20Asz+USLRn2GL9wzy6E/scHDuWtMqA5H7BZ/6QPf47MTD9BZHWH8e2N49QSrlyDDBDVIEP2IpOyxddIlrAi6RyKqU00quYDjhS0qdo/78gtUVQclNM8MqvzexqN84/v34G4oDv4gIrdYR67Vifd7+O31EAI5PsbCp2r0pg0/P/sirrB5KsrzL9c+yA8WDzB5I0JeX0HvN++UMeiNLaxeQE3N8OXaI3x+4kH++ge/xm+UX+RBd4HR2Q7/i/lZOgdnKQwmkCsbJI3ozY/D8EDX+fAx1h6yMKc6vNe7SlFG/EHrIV5qT+JtijS3ci88s28TYVmIowcJx/O0D2s+U3meOatFUd40DdJqYAVa7G8txG2EQB6aY/kT4yQuVK7EuBsh9koTfW3+p8u13Un3ePX7hWUjpybQJZ+VD47QeDhkYnqNcavNtXiU/+93PkblRYvp88FtrfbfHQbgthjiIMTbjNG2oBHmCEyCJ8DFoSAUp52Ew6bFUu0FRt0uDxVuMGdF+MIiJxyUkKmmoEloJD71bg7RVYho/8gjvC5CgGMzqEA0kjDutHGFvZM/mWDoGcNWkmNrscLM2QR/MSBZWtlzXT7eFDohqTeR/QCrN8dmUqAo++TlAF8NSJy0UwfaIKL4J88NpRCWhSn4xJUcnSkFp9pMlLo8Nn6VGbfOY/5FHnBg+5HaSgr4VxxqzwzwrtdJthp7K2yiE+T8GtVeSOyNcqk3wbRdZ87tMKFc/kTpWaZONvhi9V4uXj9K4iqcjsQKVJp/qQRhyWZQFQyqmiNHVvnT089y1FnjYXcLV0jsYXr/D0KH88EM57Ymqbwk8dcS8meXiRcW90dV+ltlOyTne3RnDPahDif9FWyhWI9LXKiP09/wcep9dLO9b1I3bkX3etAPsFcKlC/l6HZtrjxcIyobaspQU1ucKK3xZOEAuuAhnaHP6s1W9w9TN7oTiuR4j/umlhlTERpY6I9wrVnF6pk0OrWf9w6liCs5+jWHpBJzxGpSU9ZOF5BtmbXESNCkIeD9FC16Lds5oaUc7UOpNqfVVxjhYnVyaehVmLcffhUyTT/SwzD6LUWxuuQTjeToTxqOHV7lSHEDiWYzLpC/YVF7tou90njbfdVfj91hAELqdej1cOfrqKDIy/OT/KvafZzxFvmA18BGpVIvKM64i/iVAQesdCORSGISYpPwRODyXHCU37n+CPa3ypTXNNZqIxWO3YcbibBSzb+o6hOd6nHfzAoP5G7s/F6jWYoNv9N8Hy+1JylctiheaiAbHZJ9VDX4IxiNSRJkKFgIq0zZdQ7ZWzycu8b//nNtLhw7gN0SuHVerRfopALIMgQrSJvARwWBtmFQNcSlhNJknT976EXG7DbH3RUqsseoHNDTkrORz2frj/CD9QOUrmm8+SY02nvL+BtiOl2ENlTP5/j6f3wPXxh/iD/xoR/wn1QfJ0FwnzePHDf81s/4NPoevVihddotRAcO0o05MLlGLdfhwyOXeNC7TkUOkMBSbPgPrfu53KvxncvHUFc9/BXB2NkeVivAbFfI7rEx+6kRAun7yFKR/nSR/PEGn5h7hXvcRQYm4qVgmrUro/hLCtkJhs/wPjRQhk4B2l1K1wPUwOXbi0f4rdw6J70lHnA2GLF6dOYkQucZaZXgzVSIC4F0XeTkOLqcp3UUPnX8JY7k1tlIbC5HNb7+3BkKly0mLg5TY/bxHJSuy+oDPo37Yx44No83zP3bzv9LHSqagbYQRuzuHuY/LUJgTU6gx0fYPFNg+v4VxnIdnpWHCcYshC6Qv2hDyFurPh92hxJ5H314miR/M8AeFm26E4rYF7SPaHQ5plrbZCLXZr47wt+8/ucJGh6zryTYKw3MbZYP2z0GIOmpT1xbwGpVsK8f5fNT99Ecz3Gv8xRFCYr0ZHKvnXDSXkl1ocTNUOfAxDzevZcvLp9h82yN43+8Auub+zuBVymE5xKOOHzs6Dl+ZfT7nHHqSHw0qXr+UlLkD2/cw9ZKmbmLMbx8hWRbmHY/ow0ygsVBhbLqUZMxc6rNP334X7Fyf4UnO0f5weYBEi3RRiCFYSzXoewELHXLLLeLuFbCmZENqk6P+/PzHHdXqKkuRyxe1Z6wpyVdo3mqd5Q/ePF+1LLL0Ysd9LX5oRdh780/3e1Ct4tqtzl8vUoyPcoXp07zYP4Gx50VztgBJ+2LfPLUBQDyUmAj6BpNW0uUMBSFwR7Kd0gh0CbN1V1NCvzulYdoLxWZ+rZk5DvXMYMBut1JBab36Jj9VGynG/g+ulahN2nz6YMv8Veq38UXqSDv5W6NwlVFflkjOv09ebB40xiDabZwLtuoToWV62X+uHgPvZrDKXuDstWjN51ghKJ0OfeTC7u2x9dziaZHCGou9pE2f2P86wRGsZ7k+WH3EKM/VIx/bwM2Gzt96/clQoDr0jyd8Jc+8D3uz83jy1cLpkVG0x0agDsewP2KkOhahfaxEs3j8HcPf41D9gb/bfLLXKyM01nPUbCttOgteZMqEkNpNpHLwUiZxskCQTUtuAEIxg3qZJvxUoe/deC7POgtcHYww7neDC9tjuN9o8joWkLx7BrJwvJtl3raVQbgTnVRGJFfhOsXJvmGlnwgf4maajNnhbhDMWOFIMEQ6JAtDY/3D7MQVvndKw/RvVKmfA1EL0CH0f4qcNhmezE7fpjWqTJbpxS/mF9hXHXwhkZJU4csxRbP9w9SvzZCflHhbvQwid6fY3Irw7nkbRm+MX+chdEKZ7xFarKHJ2JmrDqncssMRiy0kamAOIZRp0NBBYw5HUa9CjkVcdRfp6gCDjkb1FSXooiROAD0dESE4ZlBlbPBLH+0dB/uZY/cmkG2+sPk9L091iaOMb0+stnDXBjn/2F/grlKg4dH5vFlSNnqkZcD7nMXmFYhC3GOFwZzO0a1uqV3z3JU4XKvxrV2leDlCsU1QW5tkIpCh+FORfG+NWreBMJziUoeUT4tTPAF2K+RfjC3TwliV2MSjQkCZDsgP1/mkjNFzor4xeLzFGVAbqZDz/VpnCwwkpzE2mwT31h8dU7zrW0ax0bQlTwb9/sENcGx2ga+MKwnDk90j/NcYxanZRDd/p1L+N8FCNtBloswVoFSxFF3lUmrAaRRo55JGBjN5ztH+WHnEF+/epzCVUVu3SB6+0cv9kcQIq0lFeDJkKKMOF1aIdGSSydn8T59GquvcRsRIjYYKTAqfY8ZFsEJbTBCkOQksZdKDCWuIPKhfRjiYrzzb6hSyKGRBiNej6uDcZpJnj9eu4dXFieQSx6TywneVojobh/2bu+BZHcZgJCe/jtdJr++Ru25PAsfm+K3ih/gaH6dny2eY1J1qUgoCJuGDllKXJ7uH+EfP/kpnCWb2rOauWeW09L9za19JY66w3ZFqm2x/PEq9//ai/xSYZk/V3qeCeUi8QCYj22+1L6Pzy/ey4EvaPLnl9BbDXQUvss38M5gkoTqiz22KHP2aJnPfrLOg/kbvM+7xpytOWZf5lP5SwA7JopN2nkmMYbAgBI3fyYZ9qoe5sdEJCwlgk3t8z/f+ASvXJiheNHi8B+uQaOVVq/ug7E2cUzSbCH7fY7+S4Mu+zSOHuBzRw4RFQ3xzIBcYcCvHnuWny89z+ea7+Hzl+8ljhRxqDD6prXiLDiMvmhwWgnHLq8hWh10t0fS7e1vb9ZbwJTydA549McF006dsnR2crFsmaCtVOic29gUfrdi4rQKXPT6zH0hzVc796uHWJwpc8hZ5+/f93nW4yL/U+kTNE6WGD1bpLy2gQ4GO0agdOy0uvLINJsPlOhNCg7/3FU+PvYKj/kXqUqHr4fj/NuL76G/VODYjT7Jytqe9dz/WIaOA1kqEB+fpT/pcXhmmc/kr6MQaAMDNA0NK0mef/iDn6f0RI7xxZjiMzcwvR5Js/Vu38Xt53UOWJ6ImFCSvzr2HdpVm3Mz05z96BxXu6M8e30OHVioXIKyYiwrwbVjjBFEiUIIw33jyzxQWkgPyaqLIxJ8OUAJjU2qaNLVLm2dYzUq89mFB1lrFPAfL3Dim1uIXgM2G+n62+/fkYjdrjMAd3TI6i2sQUhuI8/5jQkaYY4xq8OWs8mctUVNhVyLCzzfP8gz7QPYqzb+kiC/2CeZX0y9CPuosvW1yJwHrsugAh8feZkjzhpVaWGxnbxrWElKnO9MsV4vcmwjQK+uo/djM/Mfg9UK8NdcBmWLs/VptBEcd1aYG2pGFoeb6LbckBqKZutbvFbbv+uYiJ7RBFoMQ74u5wazbERFrqyN4i1b5NYMrG2mi+R+mn86QQ80cmUdWXcoOLPEnk9UFHSUS69o8VT1EL4a8MzWHMFqHhGmvYN38oYM5JegeL2HagWY5TWSTmf4u3220f4UGFsRe4LEM3giQglBNOyA1I0d1ABUaCDZx+HJbbb3gzBEbDaw+wPcjTzf7x5lym5w3F0hLwdMVlvMd216aw6VmUmsXoAJgrTQq1zE+B79qTy9SUF/MuE9IzdS408FdAwsRyP0tnzcukT1I/SbKQ7by0iF9tJ5NmKna2FTh6xri7Z2uBhOshBWkcsexcUYf6lPsra+P4sGb0HEGjUwWH3By4NpivLKUGIOJq0mTj5hxOqxVivSDW0quYC8HeKpiLwVkhhBK8whhea9las8nLuGQ4InYkIkW0mB0CgQad/lts5xdVBjPhhhZbOM3nDJr2jEtSVMGKb79R08GO8+AxDAaHSrhej1GP+uR3d5hPWRKv/rPYeIygnV2QanR9d46tohcj/wcRuGgxf62PU+rG2lidH7GOm66JMHCcY9gtmIM+4iNTnAFmnv5L4JiYzmX69+krN/dIqRFYNamSeJ9/mi9lqMgcUVip0e/lKF+toM36zNcuWXxvgvZ7/BjGoya0U7+WnATiN0AD2soA6MoWcEX+qc4YXOLK80xpmfH0W2LUqXJG5DM7MR4240kc1eKsuxn4y/bYzB9PuYYIB1cYmx9SLYFknJQ1uSzugsn8sdwO5ojtUHYAwiTufbthEoOwGsb0EYYfr9fb2ZvC2MJsnZBFVBVEo9BgALMczHVZ5fnOHg0z2stRZ6q353jN/QCNSNJqLTZe7LJf544SO0Dwoe/fSLPFhc4JdnnocZ+Pqxk5x7cAbZtChdlFiBoX4a5FyX8coGnxm/yrjT4qP5l6nJAc8MpnklmOJ3rjzM2BMW/lqS9kTf595oYSmivEXk30wX+nfte/nta4+ysVmk+EOP3Kbm8JUe9vxm6vnbz8bfdqOKpTUKnR5Wb5z/xftForLGrvUp5QMK7oBRr4tvhXxk4hK+DJmwm1RUj8goQqNo6xyv9Cbpxi7nOjNc6E2y0i+y0K7Q6nroKwXs1lAvVYDdA2/DoELDwY0I1e9jL9VJer13xIm1Sw3AYXPzwQBeuULuikWhNoaK5uiPWtSjEZ6LbKwLPjNfbyCbXfTaBnq7m8V+naTbKMVg1KMzaeFWOtTkgIoc9hFlu1ey5uLWGKPnYrz1AabdvjvG5jUkzRa0OsitBmObo4TTFS6+t8aV2gTKNVTVJhKDQxrivdU8TjUUoa0VXWPxYnea59Zn2FioUD5v4W1qRh9fJFlaGSYGJzvewv3KdhgiWV+HjY2d+SSB3I973/C/+9Asvn2YtEuNsSSxD8ZLRe4TY9jUORajEcKGi31tkWRtAxPfRd78W/YE+cJFxq7k8R89wsX315jyWvxs8Rz3OXXOeIt8t3KC861JnpNHsbqCuQeX+I2573HI2eABp49CDNdIWInLPNeapb1SZPzKAGetg2l33u27vXNsV1dLiXZSdQNtBG0jONeZYeNKldySYubL65jri5gwJN7vxYK3kDQaiHYbD6hVphmUJO3DeTZHcmyWI1ojLhOFDh+tXGDGrjOpWpRlRM8oGtplMymwqEYYaIvVfolm6LHaLBKs5rEbksmnE7zVwU7PCqsZYG6k3r5t7+o7Odq70wC8FaNTFe5Oh9KlDv6Kg7flEIyUqS7GyI1m6pUIo7vDwBECkfPYvMemfTrkw3PXKUqBPdQT6ugBn+0c4fnuHPUrVWoLbVS9m+bE3I1sa0yGYVpRCBS+PsX/+9JnSBzQOZ0m8L5G30BsZ9kbIBEIDe6GxGnBeMNQWBxgdcK0LH9o/O37ufda7rb7vdMMvdCDUZvBkQHTk3WqqsPAxHy59R6+tnwCb9G+KQR7l46/iWPoB/g3Wmx9ZYL/WB3nd8cfxS4NqFU6nBlZxbciZk+tEmlJLdfhme4hzvbm+KJI2IzyPL1ygHYnh1n28NYl44sGZ3ED0e6i97Ona4hpdyhcaeNtulz/wmE+feBv4S0rJq6kBQ5sNdN9d59H014PkySYVofi5Ta+b5Oru0Q5SZxziX2PeW+MfzR6EG0bjAVGmR2RbBkLrK5AxKAikBF4fUOpbbB7mvzVNrLVu5lz2A/Q22Lj7wJ7wABMXbNJownPdLCkoKwUFSEwiSbePgXv8wf2VkQuR+e+AX/9kW/wcO4aRZlWpCbG0Daaz60+xLmr04y8JJAX51N38t1ooGwz9B4kgwFs1an9i2XG1VDu4Me11Rl6VbdfY7bHb7uq1+j9raWY8a7Qryref/wiZ4rLjMoBgYFvrBxn87lxRq6atIPNXeSVeS1mKGElLlxldmkN4TokszXCisvK+ydZ/kSfA/k6f/vIl6ioHr9ff5jn6rPUezmaLR9Td5h4XDC6NMBeXMUsrabe++28v/3+PBtD0m4jXryAEpLZH6QFhSaKMcOcs7t2XRvec1KvQ7OFAHwpdgSbAZACIYY9eV9n/zCvN27DPUMnSZpecKsg9LuYLrT7DcBbuaV3sBHy7nhY3wChNL4cYIthSG7YJH4rsbm6WcW94ZLbTNLF8i6X1XgVxqTewFv+/pa5tWIsG9eM28UwBIyBILEItE1oJLbQBJGFCgQquotyeH8CJknS6E8cIxtdnMSQX3Q498osLxcmOV+bxLMirqyOEdVdZCCx2xK7Df7aAGuzD+1hdGQ7N/pueZ63i2tIhcR3Imj7MXf57TIci9dLmzfw5rvObL8Wbnm9ftvNRG4ne8sA3GY7YfMuJ9VYSyfUwMRcix3OBnPwgzKHv1BHbrWJb5FDyBjy0y7yd8smkfGuoELDaq9I2QloGxtlQrp9F7cFdlffYije5fNwWJnOYIAIQ4RlUVvKU/teEZTEODkgx7GwB3EbkejUYxMnmE4HE0ZpuPduXR+H88fEcTaf3g5vZbxe+9pdMtZ70wDMQMeSepyn6zgkpk9gNEvxCAthFbduEEsb6CC4u6p+MzL2ASqCVt+jnvPpaRdPJMSxIheBSHbHxrFr2DZitosGez1YW3/V715F5r1/fbKxuCvJDMC9hjGYYIB/weV/cz/IpaPjzE1+mbODaf7HSz/H2nqJuYUE0+2miaXZg52RsXcwBn91wOYLFZ6fKPDl0r0c8jZImg5Oy2D1kh+ft3q385PWu2w9zMjYITMA9yJRiL9qiIo5zlcn2Kp5XBpMsHq9irds4W6+Jq8lIyNjz2A1Ago3PERicalbS1M9uhKrr5GDYbhSyCwNJiMj46ciMwD3ICYYUL4ywO45NDqT/IWVvwJNm9GzAq+RYK230ibm2Wk3I2PPIbfaVK7k8Jo2z9sneMY/zugLgsJ8D7XZQSe3vydoRkbG3UdmAO5BdBBgP3meslKMlIqYUgERxZh6M20fMxhkxl9Gxh4lWV7B3tzCsSzKTxRASkyvh+kH6CRJBaCz5zsjI+OnJDMA9yg6jBAyxvQthFKYKNqRRLgbxTszMvYL271vTZIgpQClMMFgmNObefYzMjJuD8K8rmphRkZGRkZGRkbGfkX+5JdkZGRkZGRkZGTsJzIDMCMjIyMjIyPjLiMzADMyMjIyMjIy7jIyAzAjIyMjIyMj4y4jMwAzMjIyMjIyMu4yMgMwIyMjIyMjI+MuIzMAMzIyMjIyMjLuMjIDMCMjIyMjIyPjLiMzADMyMjIyMjIy7jIyAzAjIyMjIyMj4y4jMwAzMjIyMjIyMu4yMgMwIyMjIyMjI+MuIzMAMzIyMjIyMjLuMjIDMCMjIyMjIyPjLiMzADMyMjIyMjIy7jIyAzAjIyMjIyMj4y4jMwAzMjIyMjIyMu4yrDf7wk/KX7mT17Fn+Ir+vbf8nmzsUrKxe/tkY/f2eTtjB9n4bZPNvbdPNnZvn2zs3j5vduwyD2BGRkZGRkZGxl1GZgBmZGRkZGRkZNxlZAZgRkZGRkZGRsZdRmYAZmRkZGRkZGTcZWQGYEZGRkZGRkbGXcabrgLOyMjIeBVCIH0fMT2BcR2MrcCSiDBG9EPEICRZWcMMBu/2lWZkZGRkvIbMAMzIyHjrSIWwLfQ9R7jwGznyE12OjG5ywK/z7OYMSzdG8RZsjvyWIr5y7d2+2oyMjIyM17A/DUAhXv/nxuz8XiiV/kgPf2b0j74u4+5he87c+t0Pf7Y9VxASpABtXj1fGM6j7Z/dBfNH2BbSdelXPQ4dW+XTk+f5+eJZTts2/6E0xj+zP8QlMYnJue/2pWZkZLwZtvdFIRFKgpSgNcYYSBJMHL/bV7g7ESIdMyl29gixvZ/IN8iy2x5XSPcTwCRJuoe8g/vH/jAAhUDmcqAUspDH5HM/+hJtoNdHd3uIiTHa942jbYHbiFH9BNUJke0eIghJVlazyX4XIX0fMTsFloLVDZKtOqpYRIyUMfkcvYNl4rykO6EYVMBpQX4lQSbDDzDgbYVY9T6y1SNZWNq/80cqhFKYB06w9kCB1lH4y5Pn+Wj+ZWoyBmzmoypXV0exNywIo3f7ijMyMt4E6sRRmvePMqhI6mcMOp9gb1nYLcHIhYTiHz6PDoJ3+zJ3D0JgTU5ginmi8SLdWY8wL+jOCBLfEOc1xtXwWn+UAdVSWF2JCsDugN01jJ7toNYamGaLpNF8R25hnxiAEpHzELaNHq0QV7zXfZm95SCMIZqqsPawJPHAX3FwWgav7pBbdbDaA8RWff9u4Bk/gvBzDGYrJK4k3wsQzRYi75OMVxiMemw8YDOoGLyTDd43Nc9zqzOsvTyC3J4iBgrzHoUlG2/VRiyvwj6dP0Klod/m0TxbHwyZnqzz88WznLQV4KLRrIUl9IZLri4Q0f4ch4yM/cZgtszaIxIz1+d/f/+/4BEn5Lfah/h2/QTf//ZpSl/LQWYA7iCUwlTLDCYKNI861M8YTHXAL95zljP+Eo/5lzlpK+Rram0jk/DVfoUnO8e41hvlpY0JNut5rH6eMqASDZkB+AYIgRqtIooFdD5HPJIjySmaNYvYEwxGBFHxdd5mwGn6OI1R+hMCTnTIuxHt8RwmUMiuwu54+EseUytFdK/3zt9bxjuCsCwQEnl4jv6RKsGIonFcol1D7tgsbnOGsCAIyxAXDMmhPvl8wMOTCzxcvIElND8AYp0+2FoLtsbytA8r7FaR/JmHsXuG4pUOstGFepNkc+vdvenbgLAd5JEDxNU8rcOS43OrHCjUuRKNsRIP+HbnFFe6Yzx56TDllxX+eoLpZxtGxp1HWFYaAcp5iEIBHJtktEjiKvoTLoPyzU1YRuDVE1SoUYMEEWmsZgDzy5gwRA8Gd0Uax2uJc4p4NKJW7pIXIbZQHHVW6ZUdvjd2DDMzgeU6JJt1TBS+25d7R5HFImJqHOPYaN/G2Io4bxHl1Y5HTytoH1AMRgzhWMLYbINavsO9+UUO2etUZIzEZiHu88xgmkbi83J/iq0wz9Mrc7RXishAYjcl+Q7klwNko4vp99+x+9xzBqBQiuTINJ2DPp1pRetkjCxGPHToGofzmzyYv8EpZ/l13/vD4BAvdmeYdFp8tPASeRFhC43EECGJjOS/vfbL6CeqsLL6Dt9ZxjuCEAjXRXgu6x8ap/cLLWYrTf7LmSeZtJpsJQVaOkdeDsjLAb4YMGc1yUuNJwSukESl8/Smbm4QiYEt7dDQObraZSWu8EJnjq9+4T1ULhaonPNhq77nNxWZ89h6ZIzWYUnpsTX+12P/jvUkx2frj3C5U+Olbx+h+pLh6PwA+/zLmGBAkhmAGXeaYTU6rouZGKV3oMigoqifEoQVzWOPvMJfm/waCYIEybVwjH9y7SMsbZWI2zaqoyhcyzP9VY1stDFb9buycj0YURw7vMw9lWWqMsQWBT7kBbzPfZkvHrmH5r1z+Cs+ztloXxxofxyyNsrqh8eJSoJ+zRDnDdZEjzNTq0iR5np7KubnRs/xoDuPLTS+SFACylKhENgizX/+Tv8Q//OFn6XRyOOf9/A2DRMXAuYu3MBoDUkCicb0++g4TnMB3yF2rwE4zDUSnovM+2BZGN/DODatw3k6M5L+hCE/2WWs0OXRynWOuauccVY4Zr9+4rktrlCSfaqqw0GrR15IfGm/ykU74zeZt0beqbvMeKcRElkdwZTy9McFJ2trHC+u8x5vnprStHWd0NycD0oYisJgC4EGejp9OO1bPtIWMGdFHCQiMIa2vUFJ9vmjyfvp9Gz8tRxuLoeJ4r19cpaCOCeICoZqrseEkrR1zHJQZqlTwtsQ5JcG2GttknoT9Du3kGXcfQjLSj01roMeGyEpugTjLp0pRVQSDKYi3HLAByqXeb+XFnJFJmFOXePJ0WNctkPW8gU6XY9+P0c4WcR2LGS3R3IXGoDahim/yYxbxxt6uVxh4wqbittnoyiwOxaOtXvNhtuF9j16k4KwYkjGQhw/4nBtk0dGrgMQaBtbJExaTXwZExjFunbRRjI/PGg0Ep+edvnC5v3Ul8pYDUV+yZDbjLGXmsTLK+/yXe5WA1AIVG0Ukffpnqqxea/NYMRQuGeLyWKL+/LXmXabVK0uR9xV8iJk2mrjC0NRqjf82DmlqebmcYSgIByUED8Sn8/Y36hCnvWfnaN5FEbes8Z/PfslqjJg2hJ4wqEgDBrN9dhwLpwk0DYbcYlA23QSl05y83AhhUGhkcLw6fILfMgL8I2hIjV5Oc/f+tCXeeXhSb5SfJgja4dQzS7J/OLezS8VktgXxKWYitNHIViMK/zgxgGSlRxzFyOcF6+nYd/M+Pv/t/fnUZZk930f+Lk39nj7y5d7ZmVW1r71jm6gwUYTjY0AIYEgQWlsWqYlaiiNLR157JE5tuVFFm2Z59DSjGYsLyJNmiYFEyQFkACIfW0AvS/V1bWvmVm5v32PFxH3+o+XtfWGBrqqqzLrfc7JPyor38sXkRFxf/e3fL8DbhWb06rG1ASln5mgMySp39NjYrLMqL/M+1IbpIwuO+wiaaPLEXuVWPtXX543DH698D2qeY+Wtmkph7+YvZ8fpffhrrnMfkHB8fptPMDbQ5AV/DvDT7HHqpA3bkyizPlFXty1j8gzSZ7w3+Qdtg+dmRT3feIkH8idIWu0cWWILwLSsktV+bzYnqUS+fzLpQ+y0U5Qa3l0ai5EAtE1kD1Bal6QWIkx24q9pQ6yFyOrTQh6qHrjdh8icMcGgBLhOqiUR3vEpLkrJDXa5J8c/HMesIv40sAX9mte9MaDH9eTlA4+N8p3qNf+W7+JhMx2YXNk/eo/Zf94XyeHs8XLlW+K49AaF6hdbd47eomHnBjz+mtHxBgYdLVmOczRjF0utgt0Yotqz6MTWTe8nRQaS8Yc8RfBXcESBhKTccPiM6njVBMn+fL4EcKci6UUGMbWHRCRAmUCtsIx+sfQVg5hzcErS5xii7hYur2fcTtw3T165f58I264Z9/ofhViW97HwjAQpolK+jSmJd1RxWMHz/CrIz9kzGiy0zQwrpMCk3go+udBoXCFyQEbINj8amDwIqfnRig6GeLU3SldFDtwxK4wbiZf939JIyBKKcKkgTbfPMmyXQgTBr80/AJ/NVG5+r227tFQMTJWKASt2OFCaYhW0ceomSSKAhmC3dAYAQy9UCI+fvrq69Xm153EHRkACinQCY8w59HYIfjZe0+xx19nj1XClwYWb/8CDHXMfBRRUw4LUZ7lMIfSb571e3ZlB9OdkO2Qv5C+j0yn+uVz1wbLpDuZJshb9JKC7pC4NqKuIHVZkVgJMJo9jPVavyG6XN3aZctNpOsih/LEYznacyEf3nWWR1NnkUgCHbEW9ygrm99a+jjHVibo1hzMooWMwGwJZARGADK8cUHVQqBN+K8fGOPre8/zQHqBv5l5laR0SEkTQ8TsnV1l4fEdJBYdRssNVLmC6gZbL0smBMoCYStMoejqmOUwh7dkklzQGLXOtrhvbgtCYIwMI3yP7lyByl6byO/3HylH3ygloUEGAn9VYDU1ubNd7DMr6DBEN1tgWYiJUVTSwVivEa9vbCsdNzk7TWfXELWdFu6jRR4srPLJoaPMmjV8AYa4tqzFWnM0VFzojTBrFTlsx1ji9evHfnuNvzb7At/y99NLj2O97ifubp6v7mDkKYG/GkC5ers/zi3Hasb8ycZDVOMT/Kx/jh2mx1qsONEb4yuVI3znW/fhFgVeRZNuacxujNXoB4cyiBFhDMXKj/9Ft5k7MgBESFTCoZc16U6F/JOJrzBsOMi3keV7LSExp8MRFntDvNjYwZnq8Ftm+VqLKUR3ewyACN9DjeTQtkmYdog9SfGIRWtHhDfS5rEd5/GMkEgZBMrkm0cPkj7u4ZUcMobEaPUQ7c62CACF6xCP5OiMJ5iZWefvjnyHUaOHxCfQEYtxklPBBC/9aC+jzyicSoSzWoIwQjRa/cW10+0Hbte/rxQI28b99D08Vd3PmT3DfOrwKyQlJKWLpxWfHDvGH7/XZiU7wvDzaUTQQ8QxOth64ZI2wLRiHBkRas1amCZxWZO50EXUm7f7421ZhGHAUJbecIL1Bx3ST6yyN7vB/3PsG+w25Q0ZrVhrzkWK/2bxk5wpDbP+nRwTtTyi1UVojbBtWrtydHMGGUMiqzUIw02h2a2fEQwns6w/YNPeGfI7B/+M9zmdzfPz+sydQvFyd44fVHfzSOYi+6zTbxgA7rZM/oPsSSasKv9T6jODAPA1nC0Ns+Nbl4hW14i3wTX047AaES/M76Dec5meLjFltlmOUrzUnuF7l3az+w/LxCfPvc4Q4Apa6y2xGX73A0AhEKaFsK3+mHXSQ7QDRLuLDnrXpou0RsSAEgQaujoi1P2CrYVACkE5jlmMk3SVRTlO0lIOxShFLfJoxg7lXoJ2ZHGpmqfdtQmqLkbNRLzF9ZtaErAVJxelgZFMgG3ByBBx2qU17NKYMohtQZgEZUNvJmB4uM6OdIXDiWUcGaK0oKstTuwYZVkP0WoYtMZSmF3w17JYLYXZjjHaIUaji764uPWkEoREWxJlC1JWj6zs4W4uqhtK89XaPbxan8BbE3hrQV8PstGGKEK3O+g4Rvd6r8vaaS0gjEishnTPOxTNPH88/SAHvSUecVYpGB7Tdol7hpZZHc7QK/g43QwyCLZOo7kQSMdBJBN0hxUHxtcZsptciHwutoZw6gqz0kF3t8jx3CmI/uZBZjMI36NyJE9zStLcHfLY0DI7vb6UhCU8JAJDSGLd7znNyh5H0svYMubZqSy1fWlknMIICkSuoLLPIExrlJlkqD2JbLbRW1mg/IrvtG3TGHdoz4YUJmoMyTaGeH1FpxwHPNmd5HJviM8tPMDaYo5TU6MYOxUTVoX3OOuv63MzhMC444p0dw5avbsuFbcTsxEgz2c42ZrkpcIsB+2XAIeC2cSxI5RnIV1na1ZyruNdDwCFaSGzGcimWHtilOY0eGuC1FKMW+phvtjt6+DEGhFrRChZjX26OqCqPGIkaRHgy4gn27v5/Nr9VLoeK+tZdNvEXTVxi2DXNcnLPYwgZqQbIsIQ0WkhOm+9SOmgR1y581O3r0UmfPSOCaKsy+qjPq2ZmOx0lc/MvkzebDJrF0nJDinZIyEiLAGp1zw4/8qBV2nss2hpi6ryqcYJnqrvZrWb4uT6KJ3VJMkLKab/uI1e20BH4dZ5IJgmyutrRRbcJlOmc3UA6FgwzmeffwRn2WL62Q7mi2fQcUwcxz/e4k1rdNjDfvoUE6/4ZB+a5V8Zj5MYbfGbR/6cj3plHnfXeWT0mwA8te8B0n6eZLsL9a3RaC4dB1kYIhzPMXlwjf/fzj/haG+ML9fu45XlCXaeq6PPXkQNRJ/fPpsqB0YuS/fgFJ1hi+Kn2/y7B59lr7vCe5wlErI/rAb0e9i0utrLVpA2v557lkZW8F8/Ck8Pz2HYMalkh4zX5T+ceo699ip/Z+bfQdk5ksspvFIF3bgzms9/UoRhIMZHUNkExfsE/6/Hvsoue50pMwJe2w8OJ8IM/+jFTxEv+4w9DQdeXKP60Ci/9aFPkB1p8NuH/4SfMbbgRn/Au4K4cJm5P1F0J1L82cR93LtvAUtE3OvNM5M7QH14msToMBTLqC16T8G7HQAKgUx46NEhopzXH7Me7yFDG7csMF0DSwg0IMIYo6swWybPdebwZY9imCLUBnmzRcZo8XR9F+fWC/TaNsaqjdkRJJY1/nqMUw2xLxUh6KGDHjqKUN1gW5Qz4ZqYsXAdhGMj0inaUymCjEF7TJGYaHDvyDJPJE+QkcHmlKtJWylCBLHWNLTCAvKGg4mBL2O6OsKghySiqUu4ImTRz6O04JQWdOspdCaJ7HRQ9eaWOp9aCrQESyhMDDq6R0NFXOjtwSqaeOtg1jo/lQi4arWg1cJdH8FdydASCZbCHIG7jiNM0tJkwqkSpiHISJLWndl98UYI20ZlU/RyNmOJOlOmx+kwpBk7RKGJ6Hb6GeGf9v03RXyFEH3vzOvKnTqK+v1rSm/pnfZruSJxpVMJWuMWnWHJzuEyH0ieYli2yRsGoVbMRxHBazYeMQIDjRQmSgtMGWM6EclEl4OFNcbdGu/1LjBnQi7VpusmiR15zZ90K7E58Ss8D5VNEBQ8wmzMEXeRYdnG3Sznxro/vV9WEcuRxzOt3cSrPv6KJLHURs0vkZjMYq+6VHWKZ3buJiVfZcwIKMh+AGkIgSVigowkMzmBqje29OL+kyJDmI88DNEiJ90bSuWWEUMmhdHpEjdb2+pefCNUECDXyjiGwVLd53xvhEmrwqxZImkGVG2Btq2+Z/IW5l1bhYRpImyb8PBOzv91B3O4wwfnXuL+5AK/7X0ENe+hDQFSoOMYeXkFt+iyQ03y+0ufQGhwqgoZQS8lCH2BU9OMr0bInsJsNhGhQra6iE7QH2BoNPt9L5tZnHdTYPFWIkwTY3IcnfRp7MtS3mfQy2lyh4qMJRt8MneRfe4Kk2aF3VYXBVyOBC0N32sd4VRznErPY62dYthr8Z/t+BKHrYhXegbHg53M2hs84rTwhcGYWQXgg0Nn2J9a42vOflYeL5BYz5J5bplofvG2nou3jRQoQ6DM/oIJ8M1Ogc+tv4dn52eY/l6EN1+FlfV39msurTDzRWjtTPLFvfdwxF1k1mwyaZjsdtZo7w/oZWyyp1Nw4SYc17vB5ChLH8vTHlf8fPbSzX1vITDGx4hHs8SOSS9t3TDw4JQDjGID0QmI19a3bgnzNRiFPPHEEOWDacxfWeODwwv8XOYYe8wmbQ0XQ4unO7v57Zc/gio5iEggrlQnNWgTVDJCmBpRtrDrkupOg4d2XWLOXiclIrpa0Ow6JGoaqxlvyeef9P2rmb/LH07Tmo14z+Hz7DGb/YFAYRBrzXwUsaF8/vv5T3Dh+7M4Zdj5Sher3EEsF4mjEPv0MrPBKL2sw2fPfIjfH3qCg49e4L+d+QIpGVOQNgftVby/vsqJ908w/g2T1B8/s3WqHO+Q9HzMr73077KnUOSfznyeA/Y1uZcHRy/z9F+9l8TKMEPfWyS6vHQbP+mtR/d6qGoNw7Iwj8/w/+UJPrz7NP/F6DcZd2tcSErijIdRfn32eSvxLgeAFt0Rh3vuucQHhs7yseRxdlsmf5iv0zW8aw9+rTfNkGvYYcjE2hAiCImX+lY95ugIOpNCNNv9RWGzuVnDlmi8fMcYBiqVIMp51GcMgsMdJgpV/rNdf8keq0TBMEgKh/4J9airLiXlsRGlOVqf5vjGGM2WS1xxWMtmWJ7IcdAqshANc6w9BcBDThMLg4QIyRotEjJgxt5gZTjDD6eHULZB+uTW0YMSQqBNiTL64s4AZ4NRjq5OIBY8/FNLRJcW3vHviUtlKJVJNedYrGVYDnMMyQ4YkDeaFIbrbEQZYt/cMgqUccqluTPGG28yYxevfl9p+dOvjZvZKGEYqEyS7ohH5Eu6WYm+knjQoCyBH2uklFtbQue1eC69IY/WhOAfz32TTyauyOc4dKOA1TjNifYEzqs+iSWN0dPI6NrJVpagM2QTOyCj/ld3zGTaKrPLKmEJ6GpNr2eQ6WqMYGsOgAjLRGV8goJHazZi755lHsudI29ca+EICSkrl+Uwx6nFMea+2w/8uHAZ1WxePe5obR1RquAlE0xUpwmGHF6dGac07QEdChKGDc3f2/ltzoyP89lzT5A2jG2z6fhxuOWI3vk0xwOL8vSNA5e7/XW+uSsidkzyz3u36RO+i2iNDgJ0s4m/pqmlfRbGc6SkSdrsEjsC5ZiYW1wS59YHgJu9LvGD+ykd9qjt1Xx66Cy7nVU+X7+f1SDDyqkRdl4IsCpddC+84eW600WUa/0gbzPQU602Qmt0p7vZo7X1Hmw/DebkBM0HpuhmDWq7Bb18THK6wiemzjHtlpmzyvgCXu05bMRpVsMMl3t5zrWGeebELoyagVOS2HXIBBqrrWmPenxu9j10Cy/z/zn/ITZOFfj6jv24R77CpFkhJbsMGy2qyqUX+xxMLnP54SyLpSzts2m8hRS617vjrZNUIcfaQxadiYhDif7udSnI0V5PkKgIiG7u1kGEEc31BF8avRdZUOy3KgwZLfbn1wgjg8hLvUHn0p1JnLBITtW5b3SJSbPfH/tMaxdfO3sA86KL6BZ/zDtcw5ybpX7vKKEv6AxLIg+CIUWcjRB2iJcIrq8AU665iHoCt5hi6HgBuxrhnrhMtBWtGoXAyOcQnkf5vWOsPapJTlWYtYqEWvFqz2IhyvOl0r388NwuxLrDxPEIb62LiFTfc3AT7Rg4NYfQl1T2GXRmQqZ3FJm2+oHkH9Xu51x7BHE2Qep8HVlrb8keTeF5dEd9WqMm+ckSnxw7xgPeRSQSiSAipq1jvlB9kKfWd+JccLGX1/vJgd5r2lO07g9zdTqYq1Vk28e4mONfzjzB/elF/mb2ZRwh2W+vMWQ0+b1D76P2yw/hr4XYz5zqt3lsY5oTNrMPXuae7BLTZht4vR7g3YwUGokkZXTpZQTdYRujU8B03uBJrjW63kQ1W5tVyDszNXXLA0BhGAjbYuM+H/eTa3xoaJnPpF8h1PBPz32ctQsFCi8J7GPz6KCHeo0Rsmq34TU9WarRgLuoN+MK0eQQix8V2KNN/vq+F/lg8iQTZoMpw8IQAhOPju7xXGeOZ6s7uVjPs7aewVxx2PcndeSllX7QHEZXBxuy+3fzzIOzKAS1p0fY9c0Wa+9J82cjD3BvZomPp48ybcSEukeZJI/45/mbe1/hxSDLb3z//07iaBrdbN7xE6290QSJ9xZ5YnSe93rnAZPL7Szekom3rtHBTe5lDCPcFYtnUrPM+iU+nSgzbPR4PHsGQ2jO+Qe3TAAYJk0+MHmBj+eOMmO2AY8XqjtwjvoklxS69fZ7Jju7C1z+mMYrtPjF3Uc57F1mwqps9nMpUlJgXFcDLitFSTl8ofognx1/L+6qw0ylsCW9uoVhQCFPmE+w/hD8s4/9ERNmhd1WTKg1T7b38lx1ludf3s30VzVOuYN54hJxrc5rt7jCtkmMjaBSCdbfk+MXH3iBg/4yc2aPtVjyp/P3U1zMMnFMw9HTxFtVAsZzaY2ZtMYFf2XqDL+ePYdEbpZ+FaGOaSjN1xf20zmRpXBSoecvE3ffZMBDxahujJpfRNg22dM5nhnazepcms+kXyJjmhyyDJTV4zOHXuLP5H3YJ312nkhu+wCwNSn417s/yw7TwxKD4O8KWvS/rpA0unTzGqMjQSdwkq9/kotYYy8ZiChChCGqe7cFgNJA2hZiahyV8WmPax7IrbHbXyfWUFU268U03mUDrxRBL+zLbAy4xqYjgDGUh3yG6s4E7niTuUKJve4qo0YTX2hCYspxzOkww2o0zL+5fD+XV/JQN3FKBu4GGJUWcbPVD/6u241IpRACbBkhNIhIYTc0J1bHaIQujyTOs8es4IuIvNEkK7v4wiIhA95CT/uOQVg2wjKJfIOs12HI7mvVdXSPRs/FaoHVuW7S9yahowizCd2azUbv2sPUEAqJZksYzmxef8oUODLEIibQUFFdVhpp/FWNV4zftCwrTBPp+2BbqOkxekMuxSM2+cki0+kaB7xlZq0iKdnDFoqulpQjEwNN3ghxhcARMCwDdrnr+KMt2iJBmHOxfb+fed5C5TlhmvQm0rTGbXShx4RZwRUxF0NJSfl8Z2MfZ5ZH8ZYMnFILs9bpV0TeJHDTjo3yLaJkzJy3waRVwRKSUEtqDQ+7ZGA1oy2t/6cdiyArCDOavNm6mvmD/j08H2nOhyPU15PkFsErhm/vmtAa4hivGOHPWyz4eS7tymCJGqObQ3EFq0Em1aaR8PrtB9scLcAXvKFO4l3HFbk6zyPIC9RIlym/iiEEY2aVcEdA3bfpDhuY7defLxFDcszGX81hNHuYxTvTWOGWBIDCNBGehxzKsfqhMZrTMPu+RX5j/GsANLTJy90dJF/wmPxGCVGuETVbb25pdJcibBthmrQfmmXjfov2rh6/fc9fcMRZIS/BFxY11WM+Ejzd2cu/OPlBWkWfkSdN9r5aRwYhotWBoEdcqfYD7NecXy0ljttjyq3yjKPRhiSxEqG+meTyWIpjf22KD3lt5qRiWneRSByxdSZYZSYF+SztgsHeZJldzhoxgrU4YqmWIbOk8NZ7Nz8DGASkFxQyNjm9axQ1ufX0xa7odSpL4MgIQygWozQxgtLFHAe+OY9utYhfKwC9ObkpkwnUril6OZeLv2jw0YeO8Ul/jccTp0jJkGEpcIRJUcWUY4vjvQm+Vj6EITSPZ08zaxWZNuvsMD2e8M9hHFI8Nb2LF168j5GFUUS1vqWs50QyweojLp3DHX5u30n2mCHzkcHvl9/PmfoI81+fZeqVEHe1hjizgA7DN52ulo5DbyJDZ9giPVHnU8mTuKLvZ11VDnreZ+iExltubWlluzjj0dgbkR5rsNvpZ32vSOIsxzGfrb6XV2qTFJ4yGfnKBXS7Q/w2NwU6ivB/dIbZV5Ks/dwO/vzgAzySusDj3jyjhsdBd4nHJnL8+UYattDU/oB3zhW5OjWSo3Oow9++90c8kuhnnx9zi/wfj/0OLeXQ1Rah7l8b8ebGxEDTUjb/x9L7OL80jLmUJv9qCrcS4z+niTc2bueh3cCtuaoNo+9CkfTpjPSlXvakN5gwDDbiiGO9EY61p3DLGlY3UAPz+NezKRIrHJtu3qAzEZMfqXPEWWGn2W/Q7cseGJwIxnmlOU17JYm7bpBe6CDPL6J7vTcvhQBIAyQoJWkrG2VCmLGJXYkM+7IASvd33JawcIRFqGOaOqCqcn2hbq37U9Z3KMJ1iTM+kS/IWm3SRpeutigp6HYtCs2+wDXqJi+TSmMECrMt6EZbc/EQtoVIJoiczQAQTVX5tJSD2ZLExdKNvZ+bzXvScRCeh8ik6I74dAom2ckKv1J4imHZZtqUSCwCHdHUIYuRz6VwmJdbOzhZGkMITcLoseGlwLtIwWjhCthjr1L2kzzjCXD6m6OtwJUNMdk03SHN9EiFnd4GjjDpaoMz9REulfL46xpvsYEsN4iuG154QwyDMGUSZCU5v8Ow4RBrTVuFVFUOqylwKhGi8/pN35ZBCJRlIJMhI6kmCRmgUEgkCk1bmVxqD3G5liFZiX+qvtC4WoNqDa88xUIrz6RTpe32h8ESokfBaiKt+AZpogHbHCGQnguFLL2CTzbb4F5/nmGjRVtrpBDsszooOvS0RnGtPdfY1NcN0VwsjBApybwq0Co5RK7AmyhgGhLd7suN3W55q1vyBDWGC7TumaA5bpJ7/yp/e+plZuwiJ0KDPyg+zje+9gDemmD85XK/SXILNiffaqTjwK5pgrxH8QH49PueY5+/yvCmOfxK3KGqTP7R/Kc5+dRO3LJg5pUQu9rBnF8nvnJxvZYri7TvI7MZItciPp/ki8ERzJ0tNvbGOFZIwW+zz2vy4eRxDHGtx+FcGPD7lUd5vrSDxFqMbjTfkQbcLUUIejtH2LjPo74/5rHUaXZZJf6yeZgzrTGM8x6JUyvoRuvOPYbbhRDovbNUD6QoH4J7/QWGjRb/pvIgxyp9x5QbAn8hMFIp8FyiXeNU9vc3f8Z7K+zKF/mFkZeYM5sEGi5EkuUoxb/eeISFRp75cyMkL5lYDU1iTYGG5xNDPO0I/uWjIb/y4DPscEo85p1jt7NKLw1h3sfuBP3r+Q4PcOTOHax+eJTOqOCeR87yK2PPMG2VaOqQJ9sHuPDkDInLUHiliVha72+I3+yYNrOrDGVZfdggnuvyV0bPIJHMx12+297Dt8v7yZxXJE6somtbQ2z8tUjXRXge7YLN7vElHi1cYMKswXU+8JeiIX54ajf2skVh/SfX7rwed6PH0eMznB8f4vCRRXaa27vfb8AbI10XkfDp3TPLhV+wMUfb/P25ZzhiF1mOPL7WzlOOkpxoT9CIXI6Xxqg1+1PRWkMm2eUzMy8xZZfJmG0+MHKO9tAC1QM+jcjh0i/laXQK2N/JMPbdMkarg1ovoqPoDSt0t5pbEgDqhEdzwqQ1Jfi3p17m7+fOcibscTYc5tn1HUx9u4d7qYRa27jjp0dvG5ZFUPBoj1rYM03+/cL3SUlBUrr9zF9ssRqnOb4wzvT3I5xiF3n8AqrV4seF01dFaNMJtGXgrgu62mXqPUv8g5lvkjXaTBpNHAF5eWOD61qc5IdrcyyvZZmtRqhO947WF+sWbJo7NP54kz1WkWFDM98pcKw0jrchiBeX76iejDuJYMSjNidRkx1mzRIpGXOhWWBhNU++9gZ9k54LqQStSZfKAWCyzf9y5HP8jHslC+2wFgdsRAlOBeM8PT9LtO4x+qwg/+waotUhWlm7tiOWBrH7CD/cMUcz7/CJxBnGzBqxq4l8A8vZGo6t8VCSyr0x6bEGvzL2DH81UaGpQhpac7EzTO6kJnO2iby0cs0K860QEu07RLNdfnbXWR5K9AUly7HLy80dnCkNk1/tbR2NzjfCshCuQ+gL9qQ3OOxdJisjrg8A16M09rJFchHMWvcdSYAZzR7+okdLpVjdnwVvEADejQjbRngezQmbxx85zuPZ0zzmXWDc8LgQWrzS3sFakObljUmaHYfoYhKndC07XM0leCo1x+GMy5Rd5h5vkVmryGFbYG5eu3XV5YHyPyB/MoFVNZD1BnTZHM58d9fSWxIAKt+hMywICjEZo78zW4rSPN/aSbmWIF/rohutLdXAfVuQAr2Z8Yu5sQQRI4gR2G5Ea9QlyCRwhg+Cgighia0bf/6KiGzoC2IHwpQgyGtiV2OMtUj4Ae8rXGTWKpMQESkpsBBXp+06ukdbxzzTvo+VEyMkViVOsYqK45s+QHHTEJJOXuLM1dk/vIZCsBxLnlyao3Uuw+jKHfzZ7wDCpEEwpEimulhCEWoodXx03cbs3rhTFbZN78AUjR0Olf0wde8Ku9JFJowGsTY5E2oWohzfrB3i6xf2E1Rd0ict3JImfaENtUZ/M/iav4fR1Ww0kqz6acI7O9H3pijHwM512ZkrkzX6gUWIpqEMyj0fpxpjFOv9zN+bsTmQY05P0N4/SmPaYs/kAo9mzjFtVgGLV4JpvnbqAHLJZbRS29K9fzKTJh7L0R2SzHkbzFpF/M3qxeWow4Uow/cre0kuQmohQtRvfsDmipCc2cJyIlTKQ6ZS6E5nsG5tY0TCRxUyBFnJHn+daavEy8EEP+zY/M78Y6wcHcPoglUXOD3IFhV2/VrQ1ktJzrR2cTwxR5xQaEeRKLR5ZGKevN3iwcQlUkaH2dl1Lv7iGGYjjbeWwWpohp+vIVaK6GbrXZs4vyUBYJy06UzF+ONNxqwqCsXZ3hhPFXei11yM9SJRsXjHl25uN1pwVRw71JLedbuDUBt0lU0m0aE2nQIp6GVAOQqn0CaXulYSiZUkCPt/6l35IruSRfZ4azzmncMQmuUoRVdb7LFK7DA9wLo6aXeFqopYjh2+V9zDxA80iYU6YnH1ju7dFFLQGRX8yp7nmbJLxAjO9kbonMwy/qwieaHeD2AHvB4h6aUkxniLmVx/WrWnJZWGj12SWK34hhYDYdsU73Gp3R/wnj2X+Gc7/pyUNHCFRVdHPN3Zzfcre3nq2B52fAmcUoB5YR5VrcGm7/IbPQ+stqZU8rnsZwnZmn1YkWeya2SN9w+dY8xoojDpak1JeZS6CbyVVj9b91bPQyERhkF31wiXn7BQE13+06kn+Zi/flUQ+bn6TrI/cEkuxci18tYNAIVA5VM0ZxO0xzQPeJc4YHG1FeVEWODLlft4dn6GmeMd7POrqFtQ6k7JHmNmjZQfEGUS2NkMKo4HAeA2RmdTdCYSdIfhkcQ5Dlo1fqt6Py8Wp6l8Y5w9v3vyWqlWKfRreuA9KcgaBhgGMpNG+y6tfUP86MF7CHKKFw9Nsy+zzt+b+Q6P7lvmQuTzneZBnq3Msqx3UnhZYKyaWzwAdCQy22M8Uycl+5OjQ0aTUa/BhXREOD2EZUh0s9X36d0CQsLvOnGM1YywfYNu0+F8OMSYWSMr+0LZrlBkjRazmTLPTOXA0LjpAMuK2JGtMupe00kMtaQbW0ihOZBcZaezzqxdZNq8ouPSoKsNUlLcEPhFxCxHAQ1t8mT7AM/XZzm1MMZcOUTW2q8T7b4T0RLczXO2ESdYjTJYdYFb7iFaAzP4t0KZ4DgRSTO46lqitUCq62zJhEA6DjKZIMhCvtBgd2KDrDSJ0ZzsKUoqyVc3DvPq5Qm8yyZusYlRaaMbzR973wsNKOjFBqtxgnKcxOgIzFaECO786w/6GzlTKiwRIzcV/a40jSst+gNIbxT8XRkEMwzkUB6d8mlM2+ipLpOFKsNmHUsY/cEsFVHpedh1jV0L0eHWODdvhnYseklJ7GksEWFcN4RxIRjl+fVpog0Xs9lAd4Nb0kfuirjvgmT3iBIZrKSHqG2NwaObzUovg7Nu4hU14mYrJtxBiG4PqxXhFi3+x6UnmPDqfOvSXrolj9EVhWo0fuwG4Po7WYQhTjGJv2xgtiQXvFEuZ7IEMybkoaVskkaXUbfBmRmQUZqsY2K2Wuhe2NdFvoWJsltyNXeHLD65/3keS51hzmwCHo96i8yOF8nbbb7+yQdwiwnyp0PctQ7GWrXvLTjICF5FdQPMM0uklzzK+6f4/bn3czC9Sir7DEOGZsqEGRr8V5NfYnksBfRLFgaalOxhCUWsBeq6gE6iSUi1qa8mcTZ31FOGRKFwxDX7H4WmHAf8buV9nKiP8cozuxl/SrGr1MN6dR7daqO2QAB4hXKc5HJviFcak2QuKqyjF9DdYHDNvQWxKxhLNxh3a5vWYq//Gel5iIlRwuE00f42/3DvN5iz17GEwYUw5jcv/zxnS8OIb+XY/VQdWa/D6kZfE+sn2PS1ujZfrd3DWpAmtaCxTi/1ReK38d9P2DbGyDA66bHxcIH6nEDvb/Jb93+BaavEHjMELNbiiNXY50JliMJ8B2uxhGpu7R62bsGlMSMQI10SIoTrjBO/sHIv4VeGmViNEUsbxLX6LWnlGDVMMrLOwdwqz06Og87gVxtQ35qDNe+Ep9dmmf1iHWO1Qrzx9p1/thrx8ipWpcrEhRTNFyY4bUwy2+whwhqiVCX6CbK/qtVGdDqYnS5jC0mwTHTCQ7kmLz18hO/sP4w50uHn9pxg1Knz8z//DB1l881v3s+snsWsthHn5m9pj/qt6QG0YI+3xn57jZTsNz7mpYlvdbknuchfThym7Vg4VRO0hxfGyLIPcYyON29keV25R/Ubzt9oqlVs/tzVQYTtsiBohW40IAhwqpNcrAwBcCGZp63rDBs9UkIwbUrmrH4mK948doUgRqK0JkZh0NdbM4S4QUj1Cq/V9YuIaauQDWVyqjHK2dIwyUVB+uU1aHeIKrU7uvR7AwKkUITaoBL5VAMPq6U2vaYHvBXKAM8M8YyQN5OGFaaJTnqEGZtcpsoRZ5mUjOlqg404ydnSMPXVFDsuRXDs7E9XQhN9qaJSmGAjSGK1Fape71tDbgGEhkhJQm1c3ZAZAgwUloyJE05/gvq1r/M9VD5FlHJojwmC6R73jq3xqLtM3nCA/hBMS5tsxGnaXRujGaDb7WvP0S2KsgWRp7Hs6KoFF0CoY4rNBOnlGHc9QHdvjYSYROBsSl+lzC6xK4hdCVtEeuhm0wps8ovrRGvrt/uj3FJ0EPRdrao1xOJlBPz0rRQqRiuI6/UbNw3SIDt0P5Fn01Iey1MZpK95LHWGYbPOV8YO0R1xcAWYnovS6paJud+aANAUTFslJgyNK/oPKUsYSCTv987zdx78Puu9FCcfGKPS9Vg4XyBz6l7MtsZfj5GRJkxIYkdgdjR2PUIGMVapdUPZR6U8wpyHDBXWcgXd7vRlSdrvTBLgjkBrVC9ExIqRp+s0ihkuJ3L8xugeooRGHmywe7jIXLLIfYkFetqkEiXoKou1Xppqz0MKjSVjUmaXw4klhowm9znLzJr+W/7qoz34g+IHOV4dY+27k6QWNJlzLXSx3F+8t8DghLBshOugzH5mdMBPjjYgbXXJmG0MITCERogbXUzEUI7V92VojwmeGFlk1FA8HwzxrfpBfrS+E/MrWWYWIhKnN4jCn/zaUSYIJ8ayIiJlECmJjPUNdoZ3OkYQM1/O8ZK9g0f88+wmJCUkE0abXxh7md/62x/HqBx63euUBWRCDDvmyNRZHsoucNBbIiWvPbZjrXm6M8c3SwcIlxPI2lLf8Sfa2te8NgTa0phmjHFdUc0QgmbDZfJkGVFvEb/GOvRmYAiFISRo1RedHjDgZqIV3uk1JipZugWXC2f2cKIgWP25NL9QeInH9p/lxV+dpr2YYmp4P+5GgHl2+ZYISN8aGRgJQ0aTnOETb95EEokUsNcy+I/ypwBQY4pYa/678Qf5bPYhdM2me9HE6EGQhcjX2HWJW5KYbUhZEqNzLXsQ5ByaE/2fz4YxRsVA9MLXeQdvWVSMVjG8dJzkS/2gppBJQS7D4qfGOLbL59J4jta4Q6AM1tppOpHFei1Jr20jDIU0NK7XozrmMeVWmTQrzP6Yv/qF3gjfWdhNZyXJzqcCnGMLfceHrXJeheiLGDs22tBYIiZGIsXgYf6TcCUAdEXEa13/rtgAqpRHfZdGTra5L7lARroshTm+vbSHyqUc+79fJD559sdKE70ZyhBIS2FK1c9sK9kXIN8qGWhA9hSdhstCIkdj1AVCHGEybCieSJwhfp+kEiVe9zpf9phz1snKNnNWnXHD2/wfC7WZl1AoznZGObE+hlOS6FZ7W/RTa9kP/k2prvZNXkG1LPTFxbcWuX8HyOtyPhKBsXXHaQbciWjdH/qaX8RzXRK5LOHOUY4/MMZjuSSfGnqZvzP6Xf6X8Z/lpcXDJJKSodUEbJUA0G4o/rf1D3A2e4Yn/AvXPbhuRCJBKO7xFzi3c5hSN8FCIUccS3w/IGOHNDoOlZaDDgxKVQcZOVdfH3salelBKKntTmG1Urgbo3jlGG+ti3jp9LZ4GF5BxzG600WaJrmzEXbDIJjP8d18FhQYgUDEYLXBDfoPUW1AL+txxumhspJqygdu3DUHOiLQEV9rT/Kjxm6eXNoFz2fIlTTOWhOC4I7W+nstwrYRM5NEOZ/ecMQue51SnGQpyL17H8K2aA8bNKfgQKJxtYQVa4lCsFVi0fhNTIuvDoFIiXI1vtfDFf1eFVeGpN2Asq1A/hSG0UL0ewsdh86o4KGZBUyh6MQWtcDFjbfIydvELDZJvTzC+ugof5R6H4w+xbRZZc6ElNDssVdpW87rXmeJiDGjXzr6w+qDvNKYZNKt8kByniGjyQNOGUdIjlUnCM6myazoN/Vl3lIISehJRL7LSLKJLd7dAExdt91RaOLXbX8GDLg56FihOx3MSpvgVIH/MX6cByYu88mho+TtNq17urSnbGQ8RnoohbFeJVq4fNPKwbckAPRXe3zn+UM8PT5D/p7PM+433vRnJZJP+Gs8seOLKKC7R29+vy/5GcNVq5XXykAYaCTQ1ZLT4QilKMmXN45wYnUMeTTF7Llkv56/XVAxqtVCtdskvlYjYRh9V4ArRuVXSmJK98fTN9H7Zjg9nEdpwWo+w/UBoEJTUz3KyuB/OPsRms8WSM1rJr69gG40UK1Ov5y0hXorpedSO5SjOWkwObPKQ06b+ajFK53pd+0zCMehMQPGgQb3Z/uCvLHuy/eEymCrVZb696DoTwFf99m1KdGJiKFEm4TsoehPp4/7dRaTebT5ky+ewjCQ6RQ6laA5F/Hb03/Bq70h/v+Xn6DS8Jnsba2MjLq0yORn6+jRPE+n9tI+YvOJkWPsTl0ibzg86jaAN35GSiRF1eP3Xn0f6e95vDgLT96zi9lMmeHJrzBjhpxZHGXqqRhvtfvWWoJbiF5GMDdR5J7cEs51j/34XXgOxVperVwNGHAr0WGPuBZiKM3Ut9K0X03yww/sYcfDZXY4Zf779/4Za2GWf5b8CLWdKYZfdnEWl2+aYPStsYJr9nBXPDpxii9O3U9Xv8oua4M589ruVAqBK0wk8mqz7U9LqGMMsUbVrHA5k6enTE6MJNATw5imSVypbqtMIFqj3kb5Q/o+MpMmSNoIPyLvtUnJG1/X1j1e7BW4EIxSXEsztKLx1yNUtYZqtbdUqe0qQqJMgbLAMaPN/tPwdaWkW4I0kK6DTvqEac1EpsGo1R84qSmLV1rTnKsWcIOtE8S8VoT8ekQvwiglWfIyLE/k6Oo1LGLG3Rp+oksw7OONj/UtHzudt9XMLEyTeGqYoOBhZQJS0qCnDRarWXoVF2OLyL9cQYcRulZH2jbecp5XcxMApGUHXwakZRd5XZarqyza2qGnDTaiNMUwRVS30aKf1U/YPVJmQFeb1FQP3TGx6jGy3Xtj+8ctiJbgGBGW6D9/1LtchlVoFP0WpUhJRKyRkd5SG+EBWwSt0b0edqULUuCsWXxvbTc702XmhtcZtaqkCy0agUEwb+BKcdPan29JACjPLzL7b3rEKZeXX76Hp4fuRfxMhf/m0BdxRYgl+jf2HrNJ3nCulsd+WixhMGEohmVINvscH08f5V/YH+HF4kH81Ryj31ohunDp5hzcFkLv28nKe9M0p+Fv3Pd9Ppo6xpzZBq4NgZwOTf7DZ/9vsOAx+bwi89xldLtLvFWDv9uMkUmjZ8ZpziSZ2LfO/2PmuxyyV5E4PNneyze+fT/+iiA/X3pH1lV3CmJxjbnPu3RGPf515iEe23+GIaPFX88/Q85q879/+IMk989RONrBOrmADnqoxptXBABkYYizv5TCPVDlr82+SlvFPNnYh/pRjtElhb1U+al7Cm8LKkYFCr2+wcyfOqhv+KzP7eSfzO0i8jW90QhhX/dEb5pYFYnVFGTPx5gdReKwSfVwzMhcif9051+Sll1KKsHZ3hj2hoGzVEY020RbqFXjTqetQhpasdFLYrXAbMVbXl9xwJ2JCgKMs4u48xazG2N0ni3w0pExRj9T50jiMv/Jga/DAfjHjb9G5s+Mm9bqcWuEoOt1aLaQCZ+heAe9IY9Ls2nO7h7FFRGWiPBlQFZ2MURvs9wrkEJgYVyVK/lJ6GcRIbn5socyl/jh5F60MFG++9Yv3m5sGsaHOZfWJEQTPR5OnOd+O8IS1/ox+1p/SfSyS+YCJC+1trZ/6O1GCITr0Mu6BBmDPekyR+wVhgyNQrPSy+AvC1KXY0Tzzh6oEdfJMBlvkTlVzRbmmSVS1TxnKikWohxjZo1pI+CAu0w8EdDEIblsY19OgmhDs3ljJmXT5kxIAYaBTvnE010+PnOSe/wFGlqw0s3gr2iSSz10684+d2+I1n2JiTPnAcjWdmG1hggykkbbIr7uEeVUwV9TONUI/5nz6E4Xa+d9yHzAXKbEA04VgKVOjuVeDrMjEO1uXxJli0xG33Fs9ktrQ2NsZmO7WlFVJs3QwehpZC++wfVhuyE0hECsN6egr/8/oRGW1Zd92g59pncaWvfjJkAGPZLrKXqpHax0M+zx1ng0cZEpw+IfZWOEEDetlnXrRI20QncDjKUiXtll+quj/OH5j121N1MWtKdiRKaHkP0LLJdp8Tdmn2XOWeOIXXzT4ZG3w5hZIznapBWl0N7WMI2/GQjTRBzaQ3cswdrDNrvef4kD6VX2WCUs4V7VACypDmdDj2/VD5I5Kyi80sJYKW+tzModhDE8DLk07bk86w9YdMZi/q3MJYYNzYXI5rthnm+v7CV3JsRfqKNvgXXVTUEIjFQK4XuEKc2cV2TCqmC9SRlYRyG61UKsQ/5ru/mHF38Vc3+df3joG8Ra8ulDL7OxJ8kPxnez/sAE2dMw9BXV97vcbMswZ6aJxrK0xj3K+w2CguLj+17ig6mTfL+xj99dfIwzZybYc66NtVxGNZrv5hm5NZQqJE5rfMcmddlDGdcWXKMbYdQDRDfoB7uWRWNG8+8dfprD3mVcYbARRzzd3MXx2jhORaNr9WsWVQN+YnpDHuHBNntHi8zZ6ygUn63fw+cWHmDjTIE955sYq5Ufm73eylgN+OP6vRxxL/M+t0pGXlt/Dw6vce7n9pFcmcB/fp54m+sB3k50N0AphVuJObo6QaQl+50VZsyb38Z2CwNAjQ57RKtrALgX5xmD/m6fvoNA9OA+2mM2Wgq0hMYOn6/4h3ggl2YsU2f8zdRn3wZ5o8lMrsLpjk3sGHfNHJcwTZpzaaq7DfQ9Df672c9TMEKGDecGAehSLHi5O8MrlUky53vIV84R97avxc8tRQjIpelNZqjNmoRHWswUqjzgXSInPVajJN+tHWBtJcuBs0XUpcU7dqpaGAYilUQnPCJfM2WXGTbrSCF4w8kVrfu6m50Oha8KCj9McPlT4zy7Y44D/gr/YPj7ZKTBH2fP8dS+XTzpHaLwgyT0wv4XEI1lqe32qewT3P/EKQ4kV/lU+iWmzIj/bXWYc0enyF6QmOcuEt0CKYTbQVwqQ6kM9AfeXvt8uj6PZ3geTHT5jaHjm9+xaGvN6cYoF4tDZKvXsgcDfjp6aYP37TzDo5nzTBtNYm3z7eI+qs8Pk78E8uIy0ebfa7tiNTVPlvZQy/gctp8ic91FeX96kWce3E37ssXsuTQMAsBbhg576LCHXQ1oryU4ZYxQGkkCWykAfDM2d6g6jLBKLRJK9wNAQxDbDufXCygteDhxnnvs/m5LoaipHi8GedajFMc7Uyy2c0z7FR5OXCBrtLjXrpORd1mp9zqEZSMzKcj2g7/mwYD3jC+TkSEJIa+W1CNiYq15pjvL7194H8XLWfbVOv1G9e1S3tAKs6sx29Dq2bRVSHCdC8PNRhgGvYkM5f0OjVk4NLHKruQGBoqVuM3Xq+/j62f3487biO5mo/4dmqnRSqO7XYQQGB3BfFDAEhFHrDauUOTTLdYKHkFa4tsWqse1XtEgQAhB+lLMV5+5l69lDvCdmb3k7A7Pr05TLyVILUm0IRH5LBycJfItivc4NHbFuONNDqeWyZktvt06QFdZPH9uluwZQWopgu00yPV2EALp+4hkAmlcG4bo6ojVOMmp5VH0go9b2X55e39dcfz0FMXpBL+QfQFX9PA3BwX94Rbdxw/jFDuIkxdvivC/loK0GZA12nS1pKx6zFdypC5BYiXaEr7nN5NBI8HtQ7ouwvNo5x28kTa780VCbXAhDBHBzU1l3TZfGx32UGcvIaVAbGYFh8qzNHfkOT3hcnJ4ko/7p1AoQh1zIXT5H+Y/ymIpi/Fyiuw5xdHdkq8/sp+duRL/+fSXuce+XUdz+5EJj3hugu6oh/kzZX7n8J8wYTSYMB3M64y82iqkpRWfW3kI/YUhdqzFyIU14lvoN/huo8MIpxoSO4Jy02NDaeraIda3IA8sDYRtUz7g0PrZFvdOLPOPp79IXsYsxg6v9ob4yxOHGP+yjbfeRVWqd/ZwjYqJyxWE3cItz/BibZpQG/yst0xCCt4zvMALQtO4OEY+4SNl96pheVyrQ71J6qtNDvwwiR7KsvTgbi4mBam1mOFSiNFugGEQjmaY/zmP3kTIR4+8xH8w8h0sFIbQXAjz/Marv0hjKc3oDwS5r57q98/dAteHOxlh28jhIeLhDK7bD0C6OqIcxxzrTuO8lGDo1RDvXHFbDBRdRcVkj5awG1nW7x/hudk5bO8cU2ZIStp8au4Y/+e//RDOhQxz6/mbEgAqEyacKsNmnZpy2FAGnXMZdnx9Ad1q94fiBgx4F5BDeeKRHLVZi1/c/TQPJ85TjX2+196D1RQ3SLy9U26rseEVk+Mrh2M2u1h1CBMGtejG/j8p+r6ZhqH6fYS6b0llGgrbiF/XqK6QhLGBUgJxh2ZbbiqmSZh16GYNRpJN9lg1UkJeDf6uyBqcDh3OhyNcLA4xvBHjFIPtl1nRGqMTYTVNwsCkrUy6ysIQfUcJbQDS6DfM/wTXhjBNMAyk4yASfn9gwbXBsemMCKaGquxNrjNhaKQwONUe50x3HFm08TZ6WJXO1mig1n2rNaupOVMaxpSKjYzEFzEjdoPpVJWX8qOoHePIVhejVIEwuibxojWq3kBaFmYni9rUAowdSey66CGXbt6gNx4yPFbjQGKFaUNRVopTYYET3Unq60m8FQOv1COuVu/YjOmtRNg28VCKbsEl4fTLj20Vsxz7LAU57JrGKQeIzja7fwHR7OBsuDhli1eaUyRkQEpeICVh0qkwOlJjrWahk33BcB1GP9XGykinEakkQUaQMdu4IqSqfKqxj9kBdaW38k7etL1DhOP0pZdcyNptMmabu6dr/o0RpolMpUCK/uDP5kzD25Ff+6mRfV1fnUnSHfPpZWHcrpI12jzb2sWl9hBWQ9zUQaQ7ytlaN1rkzobYDZMLjxWAvhCqJWDO7PGfz3yJ1cksvzf0fk4dGmX32AZ/d/p7jBk1ZsyY6w+nrlxWGylUw0KE2z9zINIpSgdsOuOaD+SWKEgbYzOzqtA0VUBZKf79V3+NzgtDpC9oUi8uotttVLN1mz/9zUV1A4zTiyQv+8if2cErwSQKScbokHdaFJOSZD6L7nRRrbd57EJgjI2isimau9IUD5tECU04HGEkQj60+yV+pfAUw7KNLy1O9hT/1VOfwj/jMHkswn51EYIAtVX6LLVi+Nkq9Wqaowdz/NEvFLkvMc/PJM7w4eRxfvfDAd/esRc2Uow8P4xTi7Gafc/uK3SzDrU5g15GEw4prHRINt1mf36NvN3mPcmLDBlNYgRPdgt8vvQg33v+IHbJYObpEH+hBMXKuyL+eycih3IsfChDa1LxS6MXADjWy/H7az/Di0tTTJ7rYZxe3DoWjT8BaqOI0WwxbM3w3a/fx9dHD/Mb7/9Lfi2zwBP+aYZ31fld+zEa+ydJWgZyab3fV/kTIEyT5hP7Wb/fhAMN3uudJyVD/qL2AMdr47gbAt0Ntryv8lshTBM5t4OwkKS2R/O3Rn7AjFknb7zemeZuwpgcp/z+SSJXYHY0RqhJnWvA0ZO3ZjMqDYzhIYTvsf7eIYoPx+Qni+x3lmkrh3/9w0fJHZWMn2jf1P7xOyoAJOzhlAOULagGHgrV9xBGkpEuDzuawF5nY+QV0tZeHs5e5HF3naR0gGsXrELRVg7dwEJ2JSLe/pru2rUJ8hAORYzbNRxx7U/b7xtSlJRD5XKG6aMR/lKHeHX9ahZ2W6Fi4koF0WhgNWcoRmkcGeLIkKTZI7ZFP4OnNSII3lbvo9iUJwkLPs1xg+7eLol0l/ePL7LDK/Oh1HHe7yjARaEpKRd33qHwakTifOWWGHnfUrRGLK2TaQdE3iinGqMUrCZHkivsthyqQ8+TOBjwzPoMzdVRQl/g1CRm59q5DDKSXlYTZmMmZko8WFjkgeQ8n07O425en7HWPBV4HAumOV4aI33awF9XJF5dIVq8fLuO/o5A+y7tCUVypsYut990vxGnOVctEJQ9rHKLuFK5zZ/y1qC6Xeh2sVYypC94tDoWCw8NobjEhCnIyiWeyS7z/ew0bsbD3vgp+n8Mg+aYgd7f5N6JZUaNHqGGxXaOxWoWq63flnD5lkZI4oxHZ8QmzkUcthsUjOTt/lS3FyHQSZ/GtCTywWoKjC64Jbcvj/NOXDjEG/ShC4m0LUTCR6V92qOCidki+7PrpGSXxXAIf8Fg+MU6xlqV6CZmo++oAFB3A8y1Gp7SnFkY41+M7Oded4HHvfbVIQZLGDzgXSJrtJm1ilcDnSuq7c/3bI51Z/nD+Yfxn0zir6v+SbudB3YLEaaJME16I0nseys8PnGJR/xzN/zMRhzwB9UHOdEcJ3XeJHmmiKi37irR2P71ssGYWeNrTxygMTON2QGzxdvy5dUSWhOaqBCSHa7w6akzDFkt9rkrZI0WWdllJY6Yj3y+1zzIs5VZkvMaf6GBqGzNCU3d7iC0Jnsmyemv7OHo8C4ufWCIXx36Ia4IeSJ9glm3yBc/eg/Vjku1ZxFF1/pNHSfknqENcnaHQ8kl5ux1Js0qFgblOODPm/u4GAzz+dP3IM4l8FcFw0c7mPXunSuT8y4gEwlkLkt7R4bMXIWPTJ/miNvX57wYDLOxkMO7bCLbwfbq/Xsj2h2SSyEysvjS/CGSRsAD/iUecSqMOTVaEwIZuQytJ2Hl7b2lcByMyXHiXIL6Lvjo3Bl2e+vUlMH5cIinTuwmcd6icDHY9rqKwrYoHk5QOaTZt3MFV7wD6Y1tgDk9RTSeY+NQEudnioz7Hc5dGMMsmSSXLZI/hQuHME2E5yETPtHsKJF/rcDey5o0xwxiDzojmjipSE5W2JvdoBQk+Duv/A1qNZ+p0xHGShl9kyWw7qgAUPVCWC9ihhHOpRk+V3iQymSCR91ncTaFKSWS+2243y5uvqp/wcZaE+iIp1qH+OrqQTZeHWHv11Zho0S8zUqcN2AYCM+jW7D4W3u+za+mT+BLC64b/FiLbb68fIiV1RxT5yLiE2du3+e9TaRkh8N2wGF7jf/ioS9z8tAEK900a+3025oOtmTMh4dP8ah/lrzRZca0N3PToi/yHEesxTY/au/hjy88QH0tya6LAeLi0pb1o1btNrTbyE6X2bVhwokc35jez72JRQ45SzzsdcGr8uuZMzdYdb2ZiHu/JaF/rjeUyedX7ufCWoHMtzxGv3m532xfrqK2cb/V20GmksTjeZrjJh+ZPs2v5X/IsBSAw2I3R2LexF/RiPb28P19K1SrjbdQw2wnKZ3L8KfyPsIZgw+4LzNtlemOxwhlkDv59hUgpOMQzORpj9g4c3V+rfB9YgRV5fBqZ5rMMYuR51pYl0tE2zn7R7/PtLYH3vfIKT6YO/WOLFm3PEIQjecoH0pSvkfzB4f+TyaMNn/f+GVOemMEZ1x+mtyosG1kOoXKpSkdShAMXVtvOqOKicMrjCXqfKJwjDl7neUwx2qU4c8b9xJ/P8/ImiL56hrR0vLNO9ZN7qgAEK3QUYTudkksacrpPN8x9vCh9HGGZJtRI8QVEkeYWMIg0CENFdHQgmPBOKtRhs9dup/q2TzpiwLRbKN64faRN7keaSCkQOzbSfVAlvIhwazdz4heWYCbOqAcx7zc3c3ypQLeoolb3H79Qm+GVhqnAn+xfA/7sqOMmd8nJULyRpMD3jIFq8Go8/aEXS0Rs9tZZdjokJLXnGo6ukeoFc91J3ihNcsP1nfRPpUlWRJYlWZ/6GOLZ1p1HKPbHcyKgzozxD83PsTuoSLvHzqHL3tkjX7z/EF7lQlTMB/FnOqNEiNQr5m8Xg5znOuMcKmZ5+zxSZyiQXIlRLfa6KC37TMubwvXoZdzCFOCvNnCFxprMzMTKQMRg4zvXCmhm0oYIpodTMsgseRRM3I8k5plLfssUiiSU3UadpLKgRT56CCyVCe6vHTDuZGui7BtcBxE0iceSlE87NIZ1RwoFEnJkOUoxTPtXTxd2YlT0ZjVNrq7NTdubwdh2chsBgpZonzEkdQS01bpBq3YUMd8uZ3h2eYcX7x0mOQFE2/9Lth4bO5TLRGTkoJDmRXC2ODc4SmcX3gQs6OwqyEivvasUqZEOf17VEb97/fSFpEniW2IPEHoCxpzCp261lPqpbuMJepkrC7nuqMs93I8XdnJ+eIQneUkY5cVbilEtG7NHMMdFgD27ZJUHDPynVUKLydZ+uAo/9z7KHPJIh/LHmPMqDNthuSEQVlFnOgNcSoY53dOv5/Whs/okwb7v3cZ3ekQV2vbs4dDCKTrIGyLyx/J87P/1nMc9Jd51F3Duc7q7XIET7b38SfLD7LjS5B89TK6XN3+ZaMraMXQ8S4b1gTfnBnDeSzqnyf/PI97JUL9k1nM+8LAEtd6TQMdshzHVJXN/7r4Ac6emiR1zmDPl9eg1kBVa9vCnUEHAfFGCVmrs+t/j1EZn5X9c/yrA3PEnkYP9XC8kL914Ed8JnWUP609zGdPP0i8WQ6+/ujlgkfhJY1dj9l/oYSoNVBXZDZ+wqns7YrKJKjvsGiPa2bsIsPXNeSHWiIjkBF3xblSQYBeWUWUHSaAKOdxWuzguclJEjLgnx7+PBtRmn+a+TkqBzIUjqZJf2EDfSXrLgSyMITKpgiHfRrTDq1xwe5PnOcTI8d40L3EhGHwo06B3zv5XnorCfacbaHOzd+xYu03A5lOEh6coj3qsG/XIr+ePYojTAxxrZeyprr8J8//LVLf9ckvxSRfuoRud4i3sRvKFbToW2DmpMd/XPghjbzm1EyBox+e4URznB+e3oVoO+jN/iGZDhnK1YmVoNnuZ6N/ducpPp47hkRhixiJImu0sTdXHSk0q1GKE8Ek6700X7t8gEotgf+Cx+TTLYx6DbGyju4GRJ1bE3TfWQHgJjqOoVLDCEK89STn1gvUA5cxu07JLhHqNabMDmfDDE+1dnO2OUJrLYG7ZpJYDu6K5nHhOAjfI8hqPpY9xqRRI/ma9P1GnODV1hQr1TTTxS5qdb1fZr+LMGsB/ppN5EleKU/SiW0OukvM0cWV9g0aiT+Opg5oxAEh0FKShrY4EexgLcxwcSOPu2rgrSv0yvr2s4xSMSpQGKsbyLJFKjlDL+MS+YJu4BAkLJ4anWPYbPBUaSe9VR8Ridf1V/rLgtR8C6PWQV9eIX67U9h3EdqUxK5AWQpXhkgkgQ4Jddj3pe1ojEBv+czy20LrflVIacxyAyuMcUoeP2rsYYdT5gHvEpZZYSJf41LLpr1qk5sch3an740sJPFwlmDUp1MwaE0IOiOKh3LzPOxeJCVDGgqWezmCsodblshmD7Udh+OuRwhixyByBWm7S87waasel6MmgYaSclgMx4lXPVKLEd5yi3htfWvIWL0DRHjFREBwPJjAFwu4QuMImDRqGP4lXBlybrRAo+MgpUYIzVCize50kUAZV9uKHkzN85CzejXJEGpoKIseEgNNrAVV5XOhM8xyJ0OpmEKULRIrCmt+A93tEldqt1SC6I4MAPs6Yk1Ep0vhhzbJyzm6hVF+774xomxMaqzBVKbGyUvjZJ93sOuaXYsBZqOFsVTctgMfVxC2jdo1SXfEozcZctAqkjcMLHHjJNxnS4/w3a/fh78sMFcWiXrhttazeh1aIy8tk6+myZxLU10Y49nhCRb/apa/O/099lgb7LfeXgAY6Ig/qO3nW8X9bHQSrJXTxHWb7DETr6SY3gixyjWMWmv7ChZfsX0LDJyTS4yvZdCmRLkW2jIofmsn/3NiF05dsbvUfUPnOLPWgfUS9ELULdrVbnWUYxImIfYVlogIdczpUHIpHOHowhR7nqtjrNeIK9Xb/VHfPVSMKpYQ9QZT3/L5wep7aE0IZh6fZ3eqyPuHL/CR0VP8YOcuTj4wjqhaZE8IzA5sPKyY2LPBmNdiT2qdEbvOx5LHGTNivt2e5rnmTr58/hDDTxn4GyGyWLkrnTD+tDnB/3TxcdZLaZLPe7glza5zbayFDXS7g9ruG44r60Utg7c+xH8b/zJhVpHdUWUqU6MXG4TKIGUFPDF+hozZoWA2SBkdqnGC9TBNO7ZRWtKNTf5y/Qhf4h46kUUjcGh0HIKLKczWtbYYqwn+qsYMNDuLIUang7VaRZXK6Fjd8vX6zgwAueaHx9kLWGfBGxtFmTvpFExarQynhj0SpxzGv72BqDaIi+W+9/Dt/uDvAsIwCPIurVETL91g1LDxZT/4i6/roTpXHyZ/QuOt99D15t0V/G0SVypQqSDXfIbWhwgn8px7eJjTw+OkZIfdVvdNhxauJ9ARLzemeXVpHFVx8JYMnLJm/KuXiS4tAP14Z7tffzqKIIr6Ht+bPt/Qb5vxNr/eirvvCvzJ0VIQ24ClMTYF3FfjLOeCUXTFRl6a3/a+tG/EFWkY88QlRleyNA+Ncn7fMAC/PP4Cj3oXuNef59XCNC/UdvBSby9mU7D/8CL/5cwXyRtdpgwLQwhiLQnRzPcKvFDaQW8lQeZCF6vYRN1Nrh8alBbEWnG8M8Xa+QL+ZYOpL1x7rm33Z9r1XFkvvKDHaGKKbtag3M1xfMy/+jNDuSa/OPoCu+x1Zsw2w4bDq70NfqT2AJAwA5QWXKzmqdYSqMBAtAyshmT0ZYW3cW3ja9YC5MJKX2S60wUVv6vn+44NAF+LarXJnGmRWLFJLpn0Ug7JpR6Ua+hOd1v3a7wW4bmUDtrUD4Q8NrlwVfAZ+oHK51vjvNzawcVT4+y+2MEst9DbNSv1NtG9fhBsCUH225P83rkP86/8/tj920KBv2iS3dBYbY1TDbEaEbq2zUq9A2473WGH3t4OMyMVho0GoVZ8t36Aby/twVs2ILybluTXo7sB1Bv4Cx7JH+RZzM7wmzsmsbJdhjItDubWsGXM1P3L9GIDKTT/YuXDeEZIyurSihyOlcb7GZlLKfxlyeiywlqpQqPV79vd5uhOF3e+ilX1OPWlvezeOYe7bDJ6VuFWwpsuN7LV0I0miYt1PN/GansE6WtT5j3f4zcnfpnY0Whbo02N7EjMlkDEAiMAofrZvVxHIyMwehqzq0hcaiKb19Zi0ekHfjqMbssA3NYJABsNeO5VTCFJyX7Ao5UmvguzWsJ1qR/u8e+950e8L3H2agYr1oq2Dvmj5Uc4eXaSoZclxtGz/ZLkXdAw/lboKOrv7qpVCr+/TMHol37FGwlzvtl7xOrqTXplsvxuvP4G3Fo6Q5KP7TvJocRSX5wYwY/Wd9I4NkR+UW37Pqwfx5VMoKg3GV8rI2ybcHqIXs5h/f4k5z8Qsytd5J/v/hx5I+TvXfwMTx3fffX1smlQeFkwvBHhXyqhF5b7z4dtMLD1dlGtFpw5D8DUSxbCkH2P2Ti+a9fV64nrdTjW39ynnjdIXf+fQvYt4ri2ftzgz3u96sh1QZ1WGq3VjVWQ23y9bZkAENj0GY0HShFCICxFwWqQkAGSfgq/o3uUFcyXc7hLFl453txZ3B0PtbfFZlM5m4vo4MwMuNMQMVR7HjXHJ9Rgi36ZTqh+ZmFAnyvyREQRZsVDRIrEksnCmVEW0nk2giRJM+DYmWn8i9cG5MwO+OshTqmLaLT7k8bqLpHVuZ7N4+23W93mz3IncuX8vMWGa6tfMVsrABzQRwhMJ2LaKpGXXQzRn+BajBQneuOoVzLs+FoTc61GtI19LAcM2I5Ybc3xjTEiLflk+iiJQefkG6NiVLPZt9JqtZGGwfBFn+Enk2AYhF6eshAcaJYQnes0/eIY3emgeyFxr3fXZ1QH3L0MAsCtiNao2KAa+wRmv5QZ6IjlOMV8r4BdBXO1etf3cQwYsBUxQk277VDqJmgom5Ro04sMZCQQ21HU/p2wWRVS3c0gudWC1/huD8LnAQPemEEAuAXR3S7OcY/flD/PX9l7jENjz3EidPl/n/wlimtpdpyPUBul/s72bitrDBiwxXFLPYyzPhfqNv8m9xDTbpnyUpahRY23Ed5VA28DBgy4dQwCwK1ILySxpKm7Pq+MTKLGnmExHKF0Pk/yssRbqfWbfAcMGLDlMJo9vDUPMDleG6ceeVhlA78UY9V7NzaZDxgwYMBPySAA3ILoXo/M+Q52y2G5M83B5V9HV2yGXxB4pRCjWL+rtJsGDNhOGMU6+dMuwbrJYjDDJXcHheMaf76FUWkQDTKAAwYMuAkMAsAtiOp2MZ5+laSQpCwTYZporfv6VUoPBj8GDNjCRAuXsZZWsDbvb4To26GFEdHAL3nAgAE3iUEAuEW5Mrmmt7tn5YABdxtXpIoY3N8DBgy4dQitB9vJAQMGDBgwYMCAu4kfb4I6YMCAAQMGDBgwYFsxCAAHDBgwYMCAAQPuMgYB4IABAwYMGDBgwF3GIAAcMGDAgAEDBgy4yxgEgAMGDBgwYMCAAXcZgwBwwIABAwYMGDDgLmMQAA4YMGDAgAEDBtxlDALAAQMGDBgwYMCAu4xBADhgwIABAwYMGHCX8X8BmVRyQsmJA6gAAAAASUVORK5CYII=",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 800x200 with 16 Axes>"
       ]
@@ -861,7 +1084,7 @@
     },
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 800x200 with 16 Axes>"
       ]
@@ -892,8 +1115,10 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 24,
-   "metadata": {},
+   "execution_count": 26,
+   "metadata": {
+    "id": "qR0-qnJBHWoN"
+   },
    "outputs": [],
    "source": [
     "def save_model_structure(model, device):\n",
@@ -907,8 +1132,10 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
-   "metadata": {},
+   "execution_count": 27,
+   "metadata": {
+    "id": "NedNeULcHWoN"
+   },
    "outputs": [],
    "source": [
     "# 保存模型架构图\n",
@@ -917,7 +1144,9 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "yBwLI-w_HWoN"
+   },
    "source": [
     "## 习题27.4\n",
     "\n",
@@ -926,14 +1155,18 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "p43z2xw9HWoN"
+   },
    "source": [
     "**解答：**"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "LTI5nBoLHWoN"
+   },
    "source": [
     "**解答思路：**  \n",
     "\n",
@@ -944,21 +1177,27 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "ZsPrnSnsHWoN"
+   },
    "source": [
     "**解答步骤：**   "
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "4jMYaKXKHWoN"
+   },
    "source": [
     "**第1步：主成分分析的定义**"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "oScHNCc8HWoO"
+   },
    "source": [
     "&emsp;&emsp;根据书中第250页主成分分析的基本思想\n",
     "> &emsp;&emsp;主成分分析中，首先对给定数据进行规范化，使得数据每一变量的平均值为0，方差为1，之后对数据进行正交变换，原来由线性相关变量表示的数据通过正交变换变成由若干个线性无关的新变量表示的数据。新变量是可能的正交变换中变量的方差的和（信息保存）最大的，方差表示在新变量上信息的大小。将新变量依次称为第一主成分、第二主成分等。这就是主成分分析的基本思想。"
@@ -966,7 +1205,9 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "tBTj79IkHWoO"
+   },
    "source": [
     "&emsp;&emsp;根据书中第266页算法16.1主成分分析算法\n",
     "> **算法16.1（主成分分析算法）**  \n",
@@ -991,14 +1232,18 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "2HllJmwtHWoO"
+   },
    "source": [
     "**第2步：自动编码器学习过程**"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "NHJxdvRwHWoO"
+   },
    "source": [
     "&emsp;&emsp;根据书中第501页关于自动编码器的描述：\n",
     "> &emsp;&emsp;自动编码器是用于数据表示的无监督学习的一种神经网络。自动编码器由编码器网络和解码器网络组成。学习时编码器将输入向量转换为中间表示向量，解码器再将中间表示向量转换为输出向量。编码器和解码器可以是\n",
@@ -1015,14 +1260,18 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "maLXPlOEHWoO"
+   },
    "source": [
     "**第3步：证明当编码器和解码器都是线性函数时，主成分分析可以作为自动编码器学习的方法**"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "R6W-talNHWoO"
+   },
    "source": [
     "&emsp;&emsp;假设给定样本矩阵$X = (x_1, x_2, \\cdots, x_N)$，根据主成分分析算法，可以得到主成分为$K$个对应的单位特征向量$V$，样本主成分矩阵$Y$满足\n",
     "$$\n",
@@ -1032,14 +1281,18 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "nslH-6t5HWoP"
+   },
    "source": [
     "这样，样本主成分矩阵$Y$可以作为样本矩阵$X$的低维表示，$V$维度是$N \\times K$，其中$V = (v_1, v_2, \\cdots, v_N)$"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "RZLRyoEwHWoP"
+   },
    "source": [
     "&emsp;&emsp;根据自动编码器中的编码阶段，编码器对数据进行压缩，当编码器是线性函数时，可用主成分矩阵$Y$表示编码器的输出，即\n",
     "$$\n",
@@ -1050,7 +1303,9 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "lzQJIH3EHWoP"
+   },
    "source": [
     "&emsp;&emsp;在解码阶段，解码器通过解压可以得到原始数据的近似，当解码器是线性函数时，可用单位特征向量$V$表示$W_D$，即\n",
     "$$\n",
@@ -1061,7 +1316,9 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "mr0WQUmPHWoP"
+   },
    "source": [
     "&emsp;&emsp;综上所述，当损失函数是平方损失函数时，自动编码器学习的目标函数是\n",
     "$$\n",
@@ -1076,14 +1333,18 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "bk_sJGmUHWoP"
+   },
    "source": [
     "&emsp;&emsp;由于单位特征向量$V$可以使得样本矩阵$X$的所有线性变换中方差最大，故单位向量特征$V$能够使目标函数最小，因此，当编码器和解码器都是线性函数，且$b_E = b_D = 0$时，主成分分析可以作为自动编码器的学习方法。"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "EjLkO-3YHWoQ"
+   },
    "source": [
     "## 习题27.5\n",
     "\n",
@@ -1092,14 +1353,18 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "tHRdTsGwHWoQ"
+   },
    "source": [
     "**解答：**"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "A6XwVS60HWoQ"
+   },
    "source": [
     "**解答思路：**  \n",
     "\n",
@@ -1110,21 +1375,27 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "F7dIntKpHWoQ"
+   },
    "source": [
     "**解答步骤：**   "
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "yneyya2VHWoQ"
+   },
    "source": [
     "**第1步：BERT预训练中的掩码语言模型化的描述**"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "r_HQ3LTLHWoQ"
+   },
    "source": [
     "&emsp;&emsp;根据书中第497页关于BERT预训练中的掩码语言模型化的描述：\n",
     "> &emsp;&emsp;BERT模型的预训练由两部分组成，掩码语言模型化和下句预测。掩码语言模型化的目标是复原输入单词序列中被掩码的单词。可以看作是去噪自动编码器学习，对被掩码的单词独立地进行复原。掩码单词序列表示为$\\tilde{x}$。  \n",
@@ -1137,14 +1408,18 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "SYtzyGv4HWoQ"
+   },
    "source": [
     "**第2步：去噪自动编码器的原理**"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "euetI50zHWoR"
+   },
    "source": [
     "&emsp;&emsp;根据书中第501页去噪自动编码器的描述：\n",
     "> &emsp;&emsp;去噪自动编码器是自动编码器的一种扩展，去噪自动编码器不仅可以用于数据表示学习，而且可以用户数据去噪。学习时首先根据对输入向量进行的随机变化，得到有噪声的输入向量。编码器将有噪声的输入向量转换为中间表示向量，解码器再将中间表示向量转换为输出向量。编码器、解码器、目标函数分别是\n",
@@ -1158,21 +1433,27 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "HUWvSl7tHWoR"
+   },
    "source": [
     "**第3步：解释BERT预训练中的掩码语言模型化是基于去噪自动编码器原理的原因**"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "OfYMepJqHWoR"
+   },
    "source": [
     "&emsp;&emsp;BERT在预训练阶段，输入序列中的一些单词被随机遮掩，掩盖单词的过程相当于向输入序列中添加了一些噪声，得到有噪声的输入向量，模型需要预测并复原这些单词。这个过程类似于去噪自动编码器的学习过程，其中模型需要从缺失的或噪声的数据中重构原始数据。因此，掩码语言模型化提供了一些“噪音”，让模型学会从不完整的输入中推断出缺失的信息，从而提高模型的泛化能力。这也就说明了BERT预训练中的掩码语言模型化是基于去噪自动编码器原理的。"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "B071n_ZwHWoR"
+   },
    "source": [
     "## 习题27.6\n",
     "\n",
@@ -1181,14 +1462,18 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "WPzxpNZuHWoR"
+   },
    "source": [
     "**解答：**"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "qjU05u9GHWoR"
+   },
    "source": [
     "**解答思路：**  \n",
     "\n",
@@ -1199,21 +1484,27 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "8PWM-glnHWoR"
+   },
    "source": [
     "**解答步骤：**   "
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "hKOB64FuHWoS"
+   },
    "source": [
     "**第1步：BERT模型的描述**"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "tkrlcH-fHWoS"
+   },
    "source": [
     "&emsp;&emsp;根据书中第495页关于BERT的描述：\n",
     "> &emsp;&emsp;BERT是双向Transformer编码器表示（Bidirectional Encoder Representations from Transformers）的缩写。BERT的模型基于Transformer的编码器，是双向语言模型。BERT的预训练主要是掩码语言模型化，使用大规模语料基于自去噪自动编码器原理进行模型的参数估计，学习的目标是复原给定的掩码单词序列中被掩码的每一个单词。学习和预测都是非自回归过程（Non-autoregressive process）。\n",
@@ -1226,7 +1517,9 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "uGR8dxWUHWoS"
+   },
    "source": [
     "&emsp;&emsp;根据书中第502页BERT模型的描述：\n",
     "> &emsp;&emsp;BERT模型的输入是两个合并的单词序列。首先经过输入层，产生初始的单词表示向量的序列。之后经过$L$个Transformer编码层，得到单词的表示向量的序列。BERT模型的输出是在单词序列的各个位置上的条件概率。  \n",
@@ -1239,14 +1532,18 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "YBGTGGaJHWoS"
+   },
    "source": [
     "**第2步：Transformer编码器的模型描述**"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "6EpQqmJUHWoS"
+   },
    "source": [
     "&emsp;&emsp;根据书中第476页关于Transformer编码器的描述：  \n",
     "> &emsp;&emsp;Transformer（转换器）由编码器和解码器组成。编码器有1个输入层、6个编码层（一般是$L$层）。解码器有1个输入层、6个解码层、1个输出层。编码器的输入层与第1个编码层连接，第1个编码层再与第2个编码层连接，依次连接，直到第6个编码层。解码器的输入层与第1个解码层连接，第1个解码层再与第2个解码层连接，依次连接，直到第6个解码层。第6个解码层再与输出层连接。第6个编码层与各个解码层之间也有连接。  \n",
@@ -1257,7 +1554,9 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "lvzayYDZHWoS"
+   },
    "source": [
     "&emsp;&emsp;根据书中第481页关于Transformer编码器计算公式：\n",
     "> &emsp;&emsp;在编码器的输入层通过线性变换获得单词的词嵌入。\n",
@@ -1296,14 +1595,18 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "9aIlF38yHWoS"
+   },
    "source": [
     "**第3步：比较BERT与Transformer编码器在模型上的异同**"
    ]
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "XkVXUMVmHWoT"
+   },
    "source": [
     "- 相同点：\n",
     "    1. 两者都是基于多头注意力机制的架构设计，使用了自注意力机制来处理输入序列。\n",
@@ -1313,7 +1616,9 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "id": "G0B504HEHWoT"
+   },
    "source": [
     "- 不同点：\n",
     "    1. BERT模型在Transformer模型的基础上添加了掩码语言模型和下句预测任务。\n",
@@ -1325,6 +1630,11 @@
   }
  ],
  "metadata": {
+  "accelerator": "GPU",
+  "colab": {
+   "provenance": []
+  },
+  "gpuClass": "standard",
   "interpreter": {
    "hash": "c50c7698ff25eb1d7dce4c52e3d5cc14b3b23d6a97d4fb70aad8c815eba6f63e"
   },
diff --git a/requirements.txt b/requirements.txt
index 5c286ac..08bbe85 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,3 +1,4 @@
+allennlp==2.10.1
 anyio==3.6.2
 argon2-cffi==21.3.0
 argon2-cffi-bindings==21.2.0
@@ -6,24 +7,51 @@ asttokens==2.2.1
 attrs==22.2.0
 autopep8==2.0.1
 backcall==0.2.0
+base58==2.1.1
 beautifulsoup4==4.11.2
 bleach==6.0.0
+blis==0.7.9
+boto3==1.26.114
+botocore==1.29.114
+cached-path==1.1.6
+cachetools==5.3.0
+catalogue==2.0.8
 certifi==2022.12.7
 cffi==1.15.1
 charset-normalizer==3.0.1
+click==8.1.3
 colorama==0.4.6
 comm==0.1.2
+commonmark==0.9.1
 contourpy==1.0.7
 cycler==0.11.0
+cymem==2.0.7
 debugpy==1.6.6
 decorator==5.1.1
 defusedxml==0.7.1
+dill==0.3.6
+docker-pycreds==0.4.0
+exceptiongroup==1.1.1
 executing==1.2.0
+fairscale==0.4.6
 fastjsonschema==2.16.3
+filelock==3.7.1
 fonttools==4.38.0
 fqdn==1.5.1
+gitdb==4.0.10
+GitPython==3.1.31
+google-api-core==2.11.0
+google-auth==2.17.3
+google-cloud-core==2.3.2
+google-cloud-storage==2.8.0
+google-crc32c==1.5.0
+google-resumable-media==2.4.1
+googleapis-common-protos==1.59.0
 graphviz==0.20.1
+h5py==3.8.0
+huggingface-hub==0.10.1
 idna==3.4
+iniconfig==2.0.0
 ipykernel==6.21.2
 ipython==8.11.0
 ipython-genutils==0.2.0
@@ -31,6 +59,7 @@ ipywidgets==8.0.4
 isoduration==20.11.0
 jedi==0.18.2
 Jinja2==3.1.2
+jmespath==1.0.1
 joblib==1.2.0
 jsonpointer==2.3
 jsonschema==4.17.3
@@ -44,15 +73,20 @@ jupyter_server_terminals==0.4.4
 jupyterlab-pygments==0.2.2
 jupyterlab-widgets==3.0.5
 kiwisolver==1.4.4
+langcodes==3.3.0
+lmdb==1.4.0
 MarkupSafe==2.1.2
 matplotlib==3.7.0
 matplotlib-inline==0.1.6
 mistune==2.0.5
+more-itertools==9.1.0
+murmurhash==1.0.9
 nbclassic==0.5.2
 nbclient==0.7.2
 nbconvert==7.2.9
 nbformat==5.7.3
 nest-asyncio==1.5.6
+nltk==3.8.1
 notebook==6.5.3
 notebook_shim==0.2.2
 numpy==1.24.2
@@ -60,19 +94,29 @@ packaging==23.0
 pandas==1.5.3
 pandocfilters==1.5.0
 parso==0.8.3
+pathtools==0.1.2
+pathy==0.10.1
 pickleshare==0.7.5
 Pillow==9.4.0
 platformdirs==3.0.0
+pluggy==1.0.0
 portalocker==2.7.0
+preshed==3.0.8
 prometheus-client==0.16.0
+promise==2.3
 prompt-toolkit==3.0.38
+protobuf==3.20.3
 psutil==5.9.4
 pure-eval==0.2.2
+pyasn1==0.4.8
+pyasn1-modules==0.2.8
 pycodestyle==2.10.0
 pycparser==2.21
+pydantic==1.8.2
 Pygments==2.14.0
 pyparsing==3.0.9
 pyrsistent==0.19.3
+pytest==7.2.2
 python-dateutil==2.8.2
 python-json-logger==2.0.7
 pytz==2022.7.1
@@ -80,34 +124,58 @@ pywin32==305
 pywinpty==2.0.10
 PyYAML==6.0
 pyzmq==25.0.0
+regex==2023.3.23
 requests==2.28.2
 rfc3339-validator==0.1.4
 rfc3986-validator==0.1.1
+rich==12.6.0
+rsa==4.9
+s3transfer==0.6.0
+sacremoses==0.0.53
 scikit-learn==1.2.1
 scipy==1.10.1
 Send2Trash==1.8.0
+sentencepiece==0.1.97
+sentry-sdk==1.19.1
+setproctitle==1.3.2
+shortuuid==1.0.11
 six==1.16.0
+smart-open==6.3.0
+smmap==5.0.0
 sniffio==1.3.0
 soupsieve==2.4
+spacy==3.3.2
+spacy-legacy==3.0.12
+spacy-loggers==1.0.4
+srsly==2.4.6
 stack-data==0.6.2
+tensorboardX==2.6
+termcolor==1.1.0
 terminado==0.17.1
+thinc==8.0.17
 threadpoolctl==3.1.0
 tinycss2==1.2.1
+tokenizers==0.12.1
 tomli==2.0.1
-torch==1.13.1+cu117
-torchaudio==0.13.1+cu117
-torchdata==0.5.1
-torchtext==0.14.1
-torchvision==0.14.1+cu117
+torch==1.12.1+cu116
+torchaudio==0.12.1+cu116
+torchdata==0.4.1
+torchtext==0.13.1
+torchvision==0.13.1+cu116
 torchviz==0.0.2
 tornado==6.2
 tqdm==4.65.0
 traitlets==5.9.0
+transformers==4.20.1
+typer==0.4.2
 typing_extensions==4.5.0
 uri-template==1.2.0
 urllib3==1.26.14
+wandb==0.12.21
+wasabi==0.10.1
 wcwidth==0.2.6
 webcolors==1.12
 webencodings==0.5.1
 websocket-client==1.5.1
+wget==3.2
 widgetsnbextension==4.0.5