diff --git a/.gitignore b/.gitignore
index d6ad30bcdedca78251643619cd173047e59c9e00..2f0db776988e536a42d7c307433d642b5d77bf3e 100644
--- a/.gitignore
+++ b/.gitignore
@@ -4,6 +4,7 @@ __pycache__/
 *$py.class
 .vscode
 sftp.json
+data
 
 # C extensions
 *.so
diff --git a/README.md b/README.md
index fe9146dc5b245cc0405c745c1bc69f941d1d47c4..70a87dea35a2d8524f1675be3e3d997ab4f591ec 100644
--- a/README.md
+++ b/README.md
@@ -1,4 +1,4 @@
-# Natural Language Processing with PyTorch 中文版
+# PyTorch 自然语言处理（Natural Language Processing with PyTorch 中文版）
 
 ![](cover.jpg)
 
@@ -20,8 +20,31 @@
 * [Chapter 8.用于自然语言处理的高级 Sequence](docs/8.md)
 * [Chapter 9.经典, 前沿和后续步骤](docs/9.md)
 
----
+代码地址: <https://github.com/joosthub/PyTorchNLPBook>
 
 * [在线阅读](https://nlp-pt.apachecn.org)
 * [ApacheCN 机器学习交流群 629470233](http://shang.qq.com/wpa/qunwpa?idkey=30e5f1123a79867570f665aa3a483ca404b1c3f77737bc01ec520ed5f078ddef)
 * [ApacheCN 学习资源](http://www.apachecn.org/)
+
+## 精品推荐
+
+> 深度学习必学
+
+1. [反向传递](/docs/dl/反向传递.md): <https://www.cnblogs.com/charlotte77/p/5629865.html>
+2. [CNN原理](/docs/dl/CNN原理.md): <http://www.cnblogs.com/charlotte77/p/7759802.html>
+3. [RNN原理](/docs/dl/RNN原理.md): <https://blog.csdn.net/qq_39422642/article/details/78676567>
+4. [LSTM原理](/docs/dl/LSTM原理.md): <https://blog.csdn.net/weixin_42111770/article/details/80900575>
+
+> 自然语言处理
+
+* Python 自然语言处理 第二版: <https://usyiyi.github.io/nlp-py-2e-zh>
+* 推荐一个[liuhuanyong大佬](https://github.com/liuhuanyong)整理的nlp全面知识体系: <https://liuhuanyong.github.io>
+* 开源 - 词向量库集合: 
+  * <https://github.com/Embedding/Chinese-Word-Vectors>
+  * <https://github.com/brightmart/nlp_chinese_corpus>
+  * <https://github.com/codemayq/chinese_chatbot_corpus>
+  * <https://github.com/candlewill/Dialog_Corpus>
+
+---
+
+内容由 [ApacheCN](https://github.com/apachecn) 团队提供支持
diff --git a/docs/chatbot.md b/docs/chatbot.md
index 70f7deee133f53ed681c567ed7a9feb0da93d621..15711295a5ef705fc3d31a2b35c486939d20e1a6 100644
--- a/docs/chatbot.md
+++ b/docs/chatbot.md
@@ -102,3 +102,58 @@ u0 +++$+++ u2 +++$+++ m0 +++$+++ ['L207', 'L208']
 4. 打印输出 pairs
 5. voc.trim, 过滤掉词频数据（有利于让训练更快收敛的策略是去除词汇表中很少使用的单词。减少特征空间也会降低模型学习目标函数的难度）
 
+```
+['there .', 'where ?']
+['you have my word . as a gentleman', 'you re sweet .']
+['hi .', 'looks like things worked out tonight huh ?']
+['you know chastity ?', 'i believe we share an art instructor']
+['have fun tonight ?', 'tons']
+['well no . . .', 'then that s all you had to say .']
+['then that s all you had to say .', 'but']
+['but', 'you always been this selfish ?']
+['do you listen to this crap ?', 'what crap ?']
+['what good stuff ?', 'the real you .']
+```
+
+### 为模型格式化数据
+
+1. 加速训练，利用GPU并行计算能力，则需要使用小批量 `mini-batches`
+2. 为了保证数据长短一致，设置 `(max_length，batch_size)`, 短于 max_length 的句子在 EOS_token 之后进行零填充 `(zero padded)`
+3. 矩阵转置（以便跨第一维的索引返回批处理中所有句子的时间步长）
+
+![](https://pytorch.apachecn.org/docs/1.0/img/b2f1969c698070d055c23fc81ab07b1b.jpg)
+
+## 定义模型
+
+Seq2seq模型的目标是将可变长度序列作为输入，并使用固定大小的模型将可变长度序列作为输出返回。
+
+* Seq2Seq模型: 
+    1. 编码器，其将可变长度输入序列编码为固定长度上下文向量。 
+    2. 解码器，它接收输入文字和上下文矢量，并返回序列中下一句文字的概率和在下一次迭代中使用的隐藏状态。
+
+![](https://pytorch.apachecn.org/docs/1.0/img/32a87cf8d0353ceb0037776f833b92a7.jpg)
+
+
+* 编码器:
+
+如果将填充的一批序列传递给RNN模块，我们必须分别使用torch.nn.utils.rnn.pack_padded_sequence和torch.nn.utils.rnn.pad_packed_sequence在RNN传递时分别进行填充和反填充。
+
+```py
+def forward(self, input_seq, input_lengths, hidden=None):
+    # Convert word indexes to embeddings
+    embedded = self.embedding(input_seq)
+    # Pack padded batch of sequences for RNN module
+    packed = torch.nn.utils.rnn.pack_padded_sequence(embedded, input_lengths)
+    # Forward pass through GRU
+    outputs, hidden = self.gru(packed, hidden)
+    # Unpack padding
+    outputs, _ = torch.nn.utils.rnn.pad_packed_sequence(outputs)
+    # Sum bidirectional GRU outputs
+    outputs = outputs[:, :, :self.hidden_size] + outputs[:, : ,self.hidden_size:]
+    # Return output and final hidden state
+    return outputs, hidden
+```
+
+![](https://pytorch.apachecn.org/docs/1.0/img/c653271eb5fb762482bceb5e2464e680.jpg)
+
+* 解码器:
\ No newline at end of file
diff --git a/requirements.txt b/requirements.txt
new file mode 100644
index 0000000000000000000000000000000000000000..a8924c42403c87fb038c41f19aedf2f59af2b369
--- /dev/null
+++ b/requirements.txt
@@ -0,0 +1,2 @@
+fire
+torch
diff --git a/src/Chinese_ChatBot/QQ_ETL.py b/src/Chinese_ChatBot/QQ_ETL.py
new file mode 100644
index 0000000000000000000000000000000000000000..a22674e16841e01e77d0705dd492ecef06ef3a8a
--- /dev/null
+++ b/src/Chinese_ChatBot/QQ_ETL.py
@@ -0,0 +1,159 @@
+#!/usr/bin/python
+# coding: utf-8
+import re
+import codecs
+import pandas as pd
+ 
+ 
+script_name = "QQ聊天记录整理"
+ 
+# 1、通过正则语句，提取出所有的记录头和记录内容两个数组。一条记录头对应一条记录内容，所以两个数组长度应该相等。
+# 2、处理记录内容
+#     2.1、windows的换行为'\r\n'，单'\n'体现不出换行效果。手机端导出的记录有的换行是\n，需要替换一下。
+#     2.2、记录头放在了每条记录末行后面，为了记录头整齐美观，需要计算一下记录头前补多少空格。windows记事本显示中文字符占两格，英文占1格，而python中文字符长度是却是1，如果想要显示整齐，还需要计算一下，然后补齐空格数。补齐后记录头距离行首位置为100的整数倍。
+# 3、读和写文件的时候注意编码转换
+ 
+ 
+ 
+def length_w(text):
+    '''计算字符串在windows记事本中的实际显示长度'''
+    # 取文本长度，中文按2格计算。
+    length      = len(text)                          # 取其长度(中文字符长度为1，英文1)
+    utf8_length = len(text.encode('utf-8'))          # 取其长度(中文长3，英文1)
+    length      = int((utf8_length-length)/2)+length # 按(中文2英文1)计算长度
+    
+    # 这个写法实际上还是有问题的，有些特殊字符会导致计算长度和实际显示长度不一致。所以下面计算换行问题的代码中换了另一种写法，避免因特殊字符导致每行实际显示长度超出限定值，虽然还是不精确，但是不会超出限定值。
+    # 比如：
+    # '°'在记事本中显示占2格，b'\xc2\xb0'utf-8编码长度为2。
+    # '�'在记事本中显示占1格，b'\xef\xbf\xbd'utf-8编码长度为3。
+    # ''在记事本中显示占2格，b'\x01'utf-8编码长度为1。（特殊字符无法显示）
+    # 至于特殊'\t'制表符最好最开始就用四个空格替换掉，避免其自动缩进带来的影响
+    
+    return length
+ 
+def chinese_linefeed(text,limit):
+    '''中英文混合排版，限制单行长度，超出长度换行'''
+    text_format= '' # 结果变量，初始化
+    text = text.replace('\t','    ')
+    text = text.replace('\r\n','\n')
+    text_arr = text.split('\n') # 按行分割文本
+    for line in text_arr:
+        # 逐行处理
+        text_format+='\r\n'
+        num = 0 # 长度计数变量，初始化
+        for i in line:
+            # 从该行第一个字符起计算长度
+            # 中文长度为2
+            # asc2码(英文及其字符等)长度为1
+            # 其他长度为2（一些特殊）
+            if i >= u'\u4e00' and i <= u'\u9fa5':
+                char_len=2
+            elif i >= u'\u001c' and i <= u'\u00ff':
+                char_len=1
+            else:
+                char_len=2
+            # 累计长度小于limit，直接保存至结果变量，计数变量累加
+            # 累计长度大于limit，换行后再保存，计数变量重置
+            if num+char_len<=limit:
+                text_format+=i
+                num+=char_len
+            else:
+                text_format+='\r\n'+i
+                num=char_len
+    return text_format.strip()
+ 
+
+def format_chat_data(infile, outfile):
+    """
+    # QQ聊天记录手机端导出文本
+    """
+
+    # 读取文件
+    fp = codecs.open(infile,'r','utf-8')
+    txt = fp.read()
+    fp.close()
+    
+    re_pat = r'20[\d-]{8}\s[\d:]{7,8}\s+[^\n]+(?:\d{5,11}|@\w+\.[comnet]{2,3})\)'  # 正则语句，匹配记录头
+    log_title_arr = re.findall(re_pat, txt) # 记录头数组['2016-06-24 15:42:52  张某(40**21)',…]
+    log_content_arr = re.split(re_pat, txt) # 记录内容数组['\n', '\n选修的\n\n', '\n就怕这次…]
+    log_content_arr.pop(0)                  # 剔除掉第一个（分割造成的冗余部分）
+    
+    # 数组长度
+    l1 = len(log_title_arr)
+    l2 = len(log_content_arr)
+    print('记录头数: %d\n记录内容: %d'%(l1,l2))
+    
+    if l1==l2:
+        # 整理后的记录
+        log_format = ''
+        
+        # 开始整理
+        for i in range(0,l1):
+            title       = log_title_arr[i]                   # 记录头
+            content     = log_content_arr[i].strip()         # 删记录内容首尾空白字符
+            content     = content.replace('\r\n','\n')       # 记录中的'\n'，替换为'\r\n'
+            content     = content.replace('\n','\r\n')
+            content     = chinese_linefeed(content,100)      # 每行过长自动换行
+            lastline    = content.split('\r\n')[-1]          # 取记录内容最后一行
+            length      = length_w(lastline)                 # 取其长度
+            # space = (100-(length%100))*' ' if length%100!=0 else ''# 该行记录头前补空格，变整齐为100整数倍；余数为0则不用补空格
+            space = ' | ' # 该行记录头前补空格，变整齐为100整数倍；余数为0则不用补空格
+            log_format += content + space + '['+title+']\r\n'# 拼接合成记录
+    
+        # 写到文件
+        fp = codecs.open(outfile, 'w', 'utf-8')
+        fp.write(log_format)
+        fp.close()
+    
+        print("整理完毕~^_^~")
+    else:
+        print('记录头和记录内容条数不匹配，请修正代码')
+    
+
+def split_line(line):
+    l = re.sub(r"[\[\]]+", "", str(line).strip()).split(" | ")
+    if len(l) == 2:
+        content = l[0]
+        names = l[1].split(" ")
+        if names == 3:
+            c_time = names[0] + names[1]
+            c_id = names[2]
+            print([content, c_time, c_id])
+            # return "%s | %s | %s" % (content, c_time, c_id)
+            return content
+        return content
+    return ""
+
+
+# Extracts pairs of sentences from conversations
+def extractSentencePairs(conversation):
+    qa_pairs = []
+    for i in range(len(conversation) - 1):  # We ignore the last line (no answer for it)
+        inputLine = conversation[i].strip()
+        targetLine = conversation[i+1].strip()
+        # Filter wrong samples (if one of the lists is empty)
+        if inputLine and targetLine:
+            qa_pairs.append("%s | %s" % (inputLine, targetLine))
+    return qa_pairs
+
+
+def format_2(infile, outfile):
+    df = pd.read_csv(infile, sep='\00001', header=None, names=["txt"])
+    # print(df["txt"].head(5))
+    df["content"] = df["txt"].apply(lambda line: split_line(line))
+    # df.query("content!=''")["content"].to_csv(outfile, sep="\t", header=False, index=False)
+
+    lines = df.query("content!=''")["content"].tolist()
+    # print(lines)
+    chats = extractSentencePairs(lines)
+    df_chats = pd.DataFrame(chats, columns=['lines'])
+    df_chats.to_csv(outfile, sep="\t", header=False, index=False)
+    print(">>> 数据合并成功: %s" % outfile)
+
+
+if __name__ == "__main__":
+    infile = r'data/QQChat/ML_ApacheCN.csv'
+    outfile_1 = r'data/QQChat/format_1.csv'
+    outfile_2 = r'data/QQChat/format_2.csv'
+    # format_chat_data(infile, outfile_1)
+    format_2(outfile_1, outfile_2)
\ No newline at end of file
diff --git a/src/Chinese_ChatBot/run_demo.py b/src/Chinese_ChatBot/run_demo.py
new file mode 100644
index 0000000000000000000000000000000000000000..b1cfb508b6e8fa798839a01b35af56e693ca4c65
--- /dev/null
+++ b/src/Chinese_ChatBot/run_demo.py
@@ -0,0 +1,101 @@
+#!/usr/bin/python
+# coding: utf-8
+import os
+import re
+import unicodedata
+import torch
+import torch.nn as nn
+from u_tools import normalizeString, load_model
+from u_class import Voc, GreedySearchDecoder
+
+
+# Default word tokens
+PAD_token = 0  # Used for padding short sentences
+SOS_token = 1  # Start-of-sentence token
+EOS_token = 2  # End-of-sentence token
+MAX_LENGTH = 50  # Maximum sentence length to consider
+
+
+def indexesFromSentence(voc, sentence):
+    return [voc.word2index[word] for word in sentence.split(' ')] + [EOS_token]
+
+
+def evaluate(encoder, decoder, searcher, voc, sentence, max_length=MAX_LENGTH):
+    ### Format input sentence as a batch
+    # words -> indexes
+    indexes_batch = [indexesFromSentence(voc, sentence)]
+    # Create lengths tensor
+    lengths = torch.tensor([len(indexes) for indexes in indexes_batch])
+    # Transpose dimensions of batch to match models' expectations
+    input_batch = torch.LongTensor(indexes_batch).transpose(0, 1)
+    # Use appropriate device
+    input_batch = input_batch.to(device)
+    lengths = lengths.to(device)
+    # Decode sentence with searcher
+    tokens, scores = searcher(input_batch, lengths, max_length)
+    # indexes -> words
+    decoded_words = [voc.index2word[token.item()] for token in tokens]
+    return decoded_words
+
+
+def evaluateInput(encoder, decoder, searcher, voc):
+    input_sentence = ''
+    while(1):
+        try:
+            # Get input sentence
+            input_sentence = input('> ')
+            # Check if it is quit case
+            if input_sentence == 'q' or input_sentence == 'quit': break
+            # Normalize sentence
+            input_sentence = normalizeString(input_sentence)
+            # Evaluate sentence
+            output_words = evaluate(encoder, decoder, searcher, voc, input_sentence)
+            # Format and print response sentence
+            output_words[:] = [x for x in output_words if not (x == 'EOS' or x == 'PAD')]
+            # print('Bot:', ' '.join(output_words))
+            print('Bot:', ''.join(output_words))
+
+        except KeyError:
+            print("Error: Encountered unknown word.")
+
+
+if __name__ == "__main__":
+    global device, corpus_name
+    USE_CUDA = torch.cuda.is_available()
+    device = torch.device("cuda" if USE_CUDA else "cpu")
+    corpus_name = "Chinese_ChatBot"
+
+
+    # Configure models
+    attn_model = 'dot'
+    #attn_model = 'general'
+    #attn_model = 'concat'
+    hidden_size = 500
+    encoder_n_layers = 2
+    decoder_n_layers = 2
+    dropout = 0.1
+    cp_start_iteration = 0
+    learning_rate = 0.0001
+    decoder_learning_ratio = 5.0
+    n_iteration = 8000
+
+    voc = Voc(corpus_name)
+    loadFilename = "data/save/cb_model/%s/2-2_500/%s_checkpoint.tar" % (corpus_name, n_iteration)
+    if os.path.exists(loadFilename):
+        checkpoint = torch.load(loadFilename)
+        voc.__dict__ = checkpoint['voc_dict']
+
+    cp_start_iteration, encoder, decoder, encoder_optimizer, decoder_optimizer, embedding = load_model(loadFilename, voc, cp_start_iteration, attn_model, hidden_size, encoder_n_layers, decoder_n_layers, dropout, learning_rate, decoder_learning_ratio)
+
+    # Use appropriate device
+    encoder = encoder.to(device)
+    decoder = decoder.to(device)
+    # Set dropout layers to eval mode
+    encoder.eval()
+    decoder.eval()
+
+    # Initialize search module
+    searcher = GreedySearchDecoder(encoder, decoder, device)
+
+    # Begin chatting (uncomment and run the following line to begin)
+    evaluateInput(encoder, decoder, searcher, voc)
diff --git a/src/Chinese_ChatBot/run_train.py b/src/Chinese_ChatBot/run_train.py
new file mode 100644
index 0000000000000000000000000000000000000000..e3b106a3f5e857ea3dbf6979738ff02a46e8f009
--- /dev/null
+++ b/src/Chinese_ChatBot/run_train.py
@@ -0,0 +1,350 @@
+#!/usr/bin/python
+# coding: utf-8
+import os
+import re
+import fire
+import random
+import unicodedata
+import itertools
+import csv
+import math
+import codecs
+import torch
+import torch.nn as nn
+from u_tools import normalizeString, load_model
+from u_class import Voc
+
+
+# Default word tokens
+PAD_token = 0  # Used for padding short sentences
+SOS_token = 1  # Start-of-sentence token
+EOS_token = 2  # End-of-sentence token
+MAX_LENGTH = 50  # Maximum sentence length to consider
+MIN_COUNT = 3    # Minimum word count threshold for trimming
+
+
+def printLines(file, n=10):
+    with open(file, 'rb') as datafile:
+        lines = datafile.readlines()
+    for line in lines[:n]:
+        print(line)
+
+
+def indexesFromSentence(voc, sentence):
+    return [voc.word2index[word] for word in sentence.split(' ')] + [EOS_token]
+
+def zeroPadding(l, fillvalue=PAD_token):
+    return list(itertools.zip_longest(*l, fillvalue=fillvalue))
+
+def binaryMatrix(l, value=PAD_token):
+    m = []
+    for i, seq in enumerate(l):
+        m.append([])
+        for token in seq:
+            if token == PAD_token:
+                m[i].append(0)
+            else:
+                m[i].append(1)
+    return m
+
+# Returns padded input sequence tensor and lengths
+def inputVar(l, voc):
+    indexes_batch = [indexesFromSentence(voc, sentence) for sentence in l]
+    lengths = torch.tensor([len(indexes) for indexes in indexes_batch])
+    padList = zeroPadding(indexes_batch)
+    padVar = torch.LongTensor(padList)
+    return padVar, lengths
+
+# Returns padded target sequence tensor, padding mask, and max target length
+def outputVar(l, voc):
+    indexes_batch = [indexesFromSentence(voc, sentence) for sentence in l]
+    max_target_len = max([len(indexes) for indexes in indexes_batch])
+    padList = zeroPadding(indexes_batch)
+    mask = binaryMatrix(padList)
+    mask = torch.ByteTensor(mask)
+    padVar = torch.LongTensor(padList)
+    return padVar, mask, max_target_len
+
+
+# 初始化Voc对象 和 格式化pairs对话存放到list中
+def readVocs(datafile):
+    print("Reading lines...")
+    # Read the file and split into lines
+    lines = open(datafile, encoding='utf-8').read().strip().split('\n')
+    # Split every line into pairs and normalize
+    pairs = [[normalizeString(s) for s in l.split(' | ')] for l in lines]
+    return pairs
+
+
+# 如果对 'p' 中的两个句子都低于 MAX_LENGTH 阈值，则返回True
+def filterPair(p):
+    # Input sequences need to preserve the last word for EOS token
+    return len(p[0].split(' ')) < MAX_LENGTH and len(p[1].split(' ')) < MAX_LENGTH
+
+
+# 过滤满足条件的 pairs 对话
+def filterPairs(pairs):
+    return [pair for pair in pairs if filterPair(pair)]
+
+# 使用上面定义的函数，返回一个填充的voc对象和对列表
+def loadPrepareData(corpus, corpus_name, datafile, voc, save_dir):
+    print("Start preparing training data ...")
+    pairs = readVocs(datafile)
+    print("Read {!s} sentence pairs".format(len(pairs)))
+    pairs = filterPairs(pairs)
+    print("Trimmed to {!s} sentence pairs".format(len(pairs)))
+    print("Counting words...")
+    for pair in pairs:
+        voc.addSentence(pair[0])
+        voc.addSentence(pair[1])
+    print("Counted words:", voc.num_words)
+    return voc, pairs
+
+
+# Returns all items for a given batch of pairs
+def batch2TrainData(voc, pair_batch):
+    pair_batch.sort(key=lambda x: len(x[0].split(" ")), reverse=True)
+    input_batch, output_batch = [], []
+    for pair in pair_batch:
+        input_batch.append(pair[0])
+        output_batch.append(pair[1])
+    inp, lengths = inputVar(input_batch, voc)
+    output, mask, max_target_len = outputVar(output_batch, voc)
+    return inp, lengths, output, mask, max_target_len
+
+
+def trimRareWords(voc, pairs, MIN_COUNT):
+    # Trim words used under the MIN_COUNT from the voc
+    voc.trim(MIN_COUNT)
+    # Filter out pairs with trimmed words
+    keep_pairs = []
+    for pair in pairs:
+        input_sentence = pair[0]
+        output_sentence = pair[1]
+        keep_input = True
+        keep_output = True
+        # Check input sentence
+        for word in input_sentence.split(' '):
+            if word not in voc.word2index:
+                keep_input = False
+                break
+        # Check output sentence
+        for word in output_sentence.split(' '):
+            if word not in voc.word2index:
+                keep_output = False
+                break
+
+        # Only keep pairs that do not contain trimmed word(s) in their input or output sentence
+        if keep_input and keep_output:
+            keep_pairs.append(pair)
+
+    print("Trimmed from {} pairs to {}, {:.4f} of total".format(len(pairs), len(keep_pairs), len(keep_pairs) / len(pairs)))
+    return keep_pairs
+
+
+def maskNLLLoss(inp, target, mask):
+    nTotal = mask.sum()
+    crossEntropy = -torch.log(torch.gather(inp, 1, target.view(-1, 1)).squeeze(1))
+    loss = crossEntropy.masked_select(mask).mean()
+    loss = loss.to(device)
+    return loss, nTotal.item()
+
+
+def train(input_variable, lengths, target_variable, mask, max_target_len, encoder, decoder, embedding, encoder_optimizer, decoder_optimizer, batch_size, clip, max_length=MAX_LENGTH):
+
+    # Zero gradients
+    encoder_optimizer.zero_grad()
+    decoder_optimizer.zero_grad()
+
+    # Set device options
+    input_variable = input_variable.to(device)
+    lengths = lengths.to(device)
+    target_variable = target_variable.to(device)
+    mask = mask.to(device)
+
+    # Initialize variables
+    loss = 0
+    print_losses = []
+    n_totals = 0
+
+    # Forward pass through encoder
+    encoder_outputs, encoder_hidden = encoder(input_variable, lengths)
+
+    # Create initial decoder input (start with SOS tokens for each sentence)
+    decoder_input = torch.LongTensor([[SOS_token for _ in range(batch_size)]])
+    decoder_input = decoder_input.to(device)
+
+    # Set initial decoder hidden state to the encoder's final hidden state
+    decoder_hidden = encoder_hidden[:decoder.n_layers]
+
+    # Determine if we are using teacher forcing this iteration
+    use_teacher_forcing = True if random.random() < teacher_forcing_ratio else False
+
+    # Forward batch of sequences through decoder one time step at a time
+    if use_teacher_forcing:
+        for t in range(max_target_len):
+            decoder_output, decoder_hidden = decoder(
+                decoder_input, decoder_hidden, encoder_outputs
+            )
+            # Teacher forcing: next input is current target
+            decoder_input = target_variable[t].view(1, -1)
+            # Calculate and accumulate loss
+            mask_loss, nTotal = maskNLLLoss(decoder_output, target_variable[t], mask[t])
+            loss += mask_loss
+            print_losses.append(mask_loss.item() * nTotal)
+            n_totals += nTotal
+    else:
+        for t in range(max_target_len):
+            decoder_output, decoder_hidden = decoder(
+                decoder_input, decoder_hidden, encoder_outputs
+            )
+            # No teacher forcing: next input is decoder's own current output
+            _, topi = decoder_output.topk(1)
+            decoder_input = torch.LongTensor([[topi[i][0] for i in range(batch_size)]])
+            decoder_input = decoder_input.to(device)
+            # Calculate and accumulate loss
+            mask_loss, nTotal = maskNLLLoss(decoder_output, target_variable[t], mask[t])
+            loss += mask_loss
+            print_losses.append(mask_loss.item() * nTotal)
+            n_totals += nTotal
+
+    # Perform backpropatation
+    loss.backward()
+
+    # Clip gradients: gradients are modified in place
+    _ = torch.nn.utils.clip_grad_norm_(encoder.parameters(), clip)
+    _ = torch.nn.utils.clip_grad_norm_(decoder.parameters(), clip)
+
+    # Adjust model weights
+    encoder_optimizer.step()
+    decoder_optimizer.step()
+
+    return sum(print_losses) / n_totals
+
+
+def trainIters(model_name, cp_start_iteration, voc, pairs, encoder, decoder, encoder_optimizer, decoder_optimizer, embedding, encoder_n_layers, decoder_n_layers, save_dir, n_iteration, batch_size, print_every, save_every, clip, corpus_name):
+
+    # Load batches for each iteration
+    training_batches = [batch2TrainData(voc, [random.choice(pairs) for _ in range(batch_size)]) for _ in range(n_iteration)]
+
+    # Initializations
+    print('Initializing ...')
+    # start_iteration = 1
+    print_loss = 0
+    start_iteration = cp_start_iteration + 1
+
+    # Training loop
+    print("Training...")
+    for iteration in range(start_iteration, n_iteration+1):
+        training_batch = training_batches[iteration-1]
+        # Extract fields from batch
+        input_variable, lengths, target_variable, mask, max_target_len = training_batch
+
+        # Run a training iteration with batch
+        loss = train(input_variable, lengths, target_variable, mask, max_target_len, encoder, decoder, embedding, encoder_optimizer, decoder_optimizer, batch_size, clip)
+        print_loss += loss
+
+        # Print progress
+        if iteration % print_every == 0:
+            print_loss_avg = print_loss / print_every
+            print("Iteration: {}; Percent complete: {:.1f}%; Average loss: {:.4f}".format(iteration, iteration / n_iteration * 100, print_loss_avg))
+            print_loss = 0
+
+        # Save checkpoint
+        if (iteration % save_every == 0):
+            directory = os.path.join(save_dir, model_name, corpus_name, '{}-{}_{}'.format(encoder_n_layers, decoder_n_layers, hidden_size))
+            if not os.path.exists(directory):
+                os.makedirs(directory)
+            torch.save({
+                'iteration': iteration,
+                'state_dict_en': encoder.state_dict(),
+                'state_dict_de': decoder.state_dict(),
+                'state_dict_en_opt': encoder_optimizer.state_dict(),
+                'state_dict_de_opt': decoder_optimizer.state_dict(),
+                'loss': loss,
+                'voc_dict': voc.__dict__,
+                'embedding': embedding.state_dict()
+            }, os.path.join(directory, '{}_{}.tar'.format(iteration, 'checkpoint')))
+
+
+def TrainModel():
+    global device, corpus_name
+    USE_CUDA = torch.cuda.is_available()
+    device = torch.device("cuda" if USE_CUDA else "cpu")
+
+    corpus_name = "Chinese_ChatBot"
+    corpus = os.path.join("data", corpus_name)
+    datafile = os.path.join(corpus, "format_data.csv")
+    save_dir = os.path.join("data", "save")
+    
+    global teacher_forcing_ratio, hidden_size
+    # Configure models
+    model_name = 'cb_model'
+    attn_model = 'dot'
+    #attn_model = 'general'
+    #attn_model = 'concat'
+    hidden_size = 500
+    encoder_n_layers = 2
+    decoder_n_layers = 2
+    dropout = 0.1
+    cp_start_iteration = 0
+    learning_rate = 0.0001
+    decoder_learning_ratio = 5.0
+    teacher_forcing_ratio = 1.0
+    clip = 50.0
+    print_every = 1
+    batch_size = 64
+    save_every = 1000
+    n_iteration = 8000
+
+    voc = Voc(corpus_name)
+    loadFilename = "data/save/cb_model/%s/2-2_500/5000_checkpoint.tar" % (corpus_name)
+    if os.path.exists(loadFilename):
+        checkpoint = torch.load(loadFilename)
+        voc.__dict__ = checkpoint['voc_dict']
+
+    # Load/Assemble voc and pairs
+    voc, pairs = loadPrepareData(corpus, corpus_name, datafile, voc, save_dir)
+    # Print some pairs to validate
+    print("\npairs:")
+    for pair in pairs[:10]:
+        print(pair)
+
+    # Trim voc and pairs
+    pairs = trimRareWords(voc, pairs, MIN_COUNT)
+
+    # # Example for validation
+    # small_batch_size = 5
+    # batches = batch2TrainData(voc, [random.choice(pairs) for _ in range(small_batch_size)])
+    # input_variable, lengths, target_variable, mask, max_target_len = batches
+    # print("input_variable:", input_variable)
+    # print("lengths:", lengths)
+    # print("target_variable:", target_variable)
+    # print("mask:", mask)
+    # print("max_target_len:", max_target_len)
+
+    cp_start_iteration, encoder, decoder, encoder_optimizer, decoder_optimizer, embedding = load_model(loadFilename, voc, cp_start_iteration, attn_model, hidden_size, encoder_n_layers, decoder_n_layers, dropout, learning_rate, decoder_learning_ratio)
+    
+    # Use appropriate device
+    encoder = encoder.to(device)
+    decoder = decoder.to(device)
+    for state in encoder_optimizer.state.values():
+        for k, v in state.items():
+            if isinstance(v, torch.Tensor):
+                state[k] = v.cuda()
+
+    for state in decoder_optimizer.state.values():
+        for k, v in state.items():
+            if isinstance(v, torch.Tensor):
+                state[k] = v.cuda()
+
+    # Ensure dropout layers are in train mode
+    encoder.train()
+    decoder.train()
+
+    print("Starting Training!")
+    trainIters(model_name, cp_start_iteration, voc, pairs, encoder, decoder, encoder_optimizer, decoder_optimizer, embedding, encoder_n_layers, decoder_n_layers, save_dir, n_iteration, batch_size, print_every, save_every, clip, corpus_name)
+
+
+if __name__ == "__main__":
+    fire.Fire()
diff --git a/src/Chinese_ChatBot/u_class.py b/src/Chinese_ChatBot/u_class.py
new file mode 100644
index 0000000000000000000000000000000000000000..7050d76be939fa75e3716936e4f4aeee34b98222
--- /dev/null
+++ b/src/Chinese_ChatBot/u_class.py
@@ -0,0 +1,204 @@
+#!/usr/bin/python
+# coding: utf-8
+import re
+import unicodedata
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+
+# Default word tokens
+PAD_token = 0  # Used for padding short sentences
+SOS_token = 1  # Start-of-sentence token
+EOS_token = 2  # End-of-sentence token
+
+
+class Voc:
+    def __init__(self, name):
+        self.name = name
+        self.trimmed = False
+        self.word2index = {}
+        self.word2count = {}
+        self.index2word = {PAD_token: "PAD", SOS_token: "SOS", EOS_token: "EOS"}
+        self.num_words = 3  # Count SOS, EOS, PAD
+
+    def addSentence(self, sentence):
+        for word in sentence.split(' '):
+            self.addWord(word)
+
+    def addWord(self, word):
+        if word not in self.word2index:
+            self.word2index[word] = self.num_words
+            self.word2count[word] = 1
+            self.index2word[self.num_words] = word
+            self.num_words += 1
+        else:
+            self.word2count[word] += 1
+
+    # Remove words below a certain count threshold
+    def trim(self, min_count):
+        if self.trimmed:
+            return
+        self.trimmed = True
+
+        keep_words = []
+
+        for k, v in self.word2count.items():
+            if v >= min_count:
+                keep_words.append(k)
+
+        print('keep_words {} / {} = {:.4f}'.format(
+            len(keep_words), len(self.word2index), len(keep_words) / len(self.word2index)
+        ))
+
+        # Reinitialize dictionaries
+        self.word2index = {}
+        self.word2count = {}
+        self.index2word = {PAD_token: "PAD", SOS_token: "SOS", EOS_token: "EOS"}
+        self.num_words = 3 # Count default tokens
+
+        for word in keep_words:
+            self.addWord(word)
+
+
+class EncoderRNN(nn.Module):
+    def __init__(self, hidden_size, embedding, n_layers=1, dropout=0):
+        super(EncoderRNN, self).__init__()
+        self.n_layers = n_layers
+        self.hidden_size = hidden_size
+        self.embedding = embedding
+
+        # Initialize GRU; the input_size and hidden_size params are both set to 'hidden_size'
+        #   because our input size is a word embedding with number of features == hidden_size
+        self.gru = nn.GRU(hidden_size, hidden_size, n_layers,
+                          dropout=(0 if n_layers == 1 else dropout), bidirectional=True)
+
+    def forward(self, input_seq, input_lengths, hidden=None):
+        # Convert word indexes to embeddings
+        embedded = self.embedding(input_seq)
+        # Pack padded batch of sequences for RNN module
+        packed = torch.nn.utils.rnn.pack_padded_sequence(embedded, input_lengths)
+        # Forward pass through GRU
+        outputs, hidden = self.gru(packed, hidden)
+        # Unpack padding
+        outputs, _ = torch.nn.utils.rnn.pad_packed_sequence(outputs)
+        # Sum bidirectional GRU outputs
+        outputs = outputs[:, :, :self.hidden_size] + outputs[:, : ,self.hidden_size:]
+        # Return output and final hidden state
+        return outputs, hidden
+
+
+# Luong attention layer
+class Attn(nn.Module):
+    def __init__(self, method, hidden_size):
+        super(Attn, self).__init__()
+        self.method = method
+        if self.method not in ['dot', 'general', 'concat']:
+            raise ValueError(self.method, "is not an appropriate attention method.")
+        self.hidden_size = hidden_size
+        if self.method == 'general':
+            self.attn = torch.nn.Linear(self.hidden_size, hidden_size)
+        elif self.method == 'concat':
+            self.attn = torch.nn.Linear(self.hidden_size * 2, hidden_size)
+            self.v = torch.nn.Parameter(torch.FloatTensor(hidden_size))
+
+    def dot_score(self, hidden, encoder_output):
+        return torch.sum(hidden * encoder_output, dim=2)
+
+    def general_score(self, hidden, encoder_output):
+        energy = self.attn(encoder_output)
+        return torch.sum(hidden * energy, dim=2)
+
+    def concat_score(self, hidden, encoder_output):
+        energy = self.attn(torch.cat((hidden.expand(encoder_output.size(0), -1, -1), encoder_output), 2)).tanh()
+        return torch.sum(self.v * energy, dim=2)
+
+    def forward(self, hidden, encoder_outputs):
+        # Calculate the attention weights (energies) based on the given method
+        if self.method == 'general':
+            attn_energies = self.general_score(hidden, encoder_outputs)
+        elif self.method == 'concat':
+            attn_energies = self.concat_score(hidden, encoder_outputs)
+        elif self.method == 'dot':
+            attn_energies = self.dot_score(hidden, encoder_outputs)
+
+        # Transpose max_length and batch_size dimensions
+        attn_energies = attn_energies.t()
+
+        # Return the softmax normalized probability scores (with added dimension)
+        return F.softmax(attn_energies, dim=1).unsqueeze(1)
+
+
+class LuongAttnDecoderRNN(nn.Module):
+    def __init__(self, attn_model, embedding, hidden_size, output_size, n_layers=1, dropout=0.1):
+        super(LuongAttnDecoderRNN, self).__init__()
+
+        # Keep for reference
+        self.attn_model = attn_model
+        self.hidden_size = hidden_size
+        self.output_size = output_size
+        self.n_layers = n_layers
+        self.dropout = dropout
+
+        # Define layers
+        self.embedding = embedding
+        self.embedding_dropout = nn.Dropout(dropout)
+        self.gru = nn.GRU(hidden_size, hidden_size, n_layers, dropout=(0 if n_layers == 1 else dropout))
+        self.concat = nn.Linear(hidden_size * 2, hidden_size)
+        self.out = nn.Linear(hidden_size, output_size)
+
+        self.attn = Attn(attn_model, hidden_size)
+
+    def forward(self, input_step, last_hidden, encoder_outputs):
+        # Note: we run this one step (word) at a time
+        # Get embedding of current input word
+        embedded = self.embedding(input_step)
+        embedded = self.embedding_dropout(embedded)
+        # Forward through unidirectional GRU
+        rnn_output, hidden = self.gru(embedded, last_hidden)
+        # Calculate attention weights from the current GRU output
+        attn_weights = self.attn(rnn_output, encoder_outputs)
+        # Multiply attention weights to encoder outputs to get new "weighted sum" context vector
+        context = attn_weights.bmm(encoder_outputs.transpose(0, 1))
+        # Concatenate weighted context vector and GRU output using Luong eq. 5
+        rnn_output = rnn_output.squeeze(0)
+        context = context.squeeze(1)
+        concat_input = torch.cat((rnn_output, context), 1)
+        concat_output = torch.tanh(self.concat(concat_input))
+        # Predict next word using Luong eq. 6
+        output = self.out(concat_output)
+        output = F.softmax(output, dim=1)
+        # Return output and final hidden state
+        return output, hidden
+
+
+class GreedySearchDecoder(nn.Module):
+    def __init__(self, encoder, decoder, device):
+        super(GreedySearchDecoder, self).__init__()
+        self.encoder = encoder
+        self.decoder = decoder
+        self.device = device
+
+    def forward(self, input_seq, input_length, max_length):
+        # Forward input through encoder model
+        encoder_outputs, encoder_hidden = self.encoder(input_seq, input_length)
+        # Prepare encoder's final hidden layer to be first hidden input to the decoder
+        decoder_hidden = encoder_hidden[:self.decoder.n_layers]
+        # Initialize decoder input with SOS_token
+        decoder_input = torch.ones(1, 1, device=self.device, dtype=torch.long) * SOS_token
+        # Initialize tensors to append decoded words to
+        all_tokens = torch.zeros([0], device=self.device, dtype=torch.long)
+        all_scores = torch.zeros([0], device=self.device)
+        # Iteratively decode one word token at a time
+        for _ in range(max_length):
+            # Forward pass through decoder
+            decoder_output, decoder_hidden = self.decoder(decoder_input, decoder_hidden, encoder_outputs)
+            # Obtain most likely word token and its softmax score
+            decoder_scores, decoder_input = torch.max(decoder_output, dim=1)
+            # Record token and score
+            all_tokens = torch.cat((all_tokens, decoder_input), dim=0)
+            all_scores = torch.cat((all_scores, decoder_scores), dim=0)
+            # Prepare current token to be next decoder input (add a dimension)
+            decoder_input = torch.unsqueeze(decoder_input, 0)
+        # Return collections of word tokens and scores
+        return all_tokens, all_scores
diff --git a/src/Chinese_ChatBot/u_tools.py b/src/Chinese_ChatBot/u_tools.py
new file mode 100644
index 0000000000000000000000000000000000000000..0e6b57180bb513f3f321c01a2377b39ec2b91adb
--- /dev/null
+++ b/src/Chinese_ChatBot/u_tools.py
@@ -0,0 +1,68 @@
+#!/usr/bin/python
+# coding: utf-8
+import os
+import re
+import unicodedata
+import torch
+import torch.nn as nn
+from torch import optim
+from u_class import Voc, EncoderRNN, LuongAttnDecoderRNN
+
+
+# Turn a Unicode string to plain ASCII, thanks to
+# https://stackoverflow.com/a/518232/2809427
+def unicodeToAscii(s):
+    return ''.join(
+        c for c in unicodedata.normalize('NFD', s)
+        if unicodedata.category(c) != 'Mn'
+    )
+
+
+# Lowercase, trim, and remove non-letter characters
+def normalizeString(s):
+    s = unicodeToAscii(s.lower().strip())
+    s = re.sub(r"([.!? ])", r" \1", s)
+    # s = re.sub(r"[^a-zA-Z.!?]+", r" ", s)
+    s = re.sub(r"[^a-zA-Z.!? \u4E00-\u9FA5]+", r" ", s)
+    s = re.sub(r"\s+", r" ", s).strip()
+    # '咋死 ? ? ?红烧还是爆炒dddd' > '咋 死   ?   ?   ? 红 烧 还 是 爆 炒 d d d d'
+    s = " ".join(list(s))
+    return s
+
+
+def load_model(loadFilename, voc, cp_start_iteration, attn_model, hidden_size, encoder_n_layers, decoder_n_layers, 
+        dropout, learning_rate, decoder_learning_ratio):
+
+    # Load model if a loadFilename is provided
+    if os.path.exists(loadFilename):
+        # If loading on same machine the model was trained on
+        checkpoint = torch.load(loadFilename)
+        cp_start_iteration = checkpoint['iteration']
+        encoder_sd = checkpoint['state_dict_en']
+        decoder_sd = checkpoint['state_dict_de']
+        encoder_optimizer_sd = checkpoint['state_dict_en_opt']
+        decoder_optimizer_sd = checkpoint['state_dict_de_opt']
+        # loss = checkpoint['loss']
+        # voc.__dict__ = checkpoint['voc_dict']
+        embedding_sd = checkpoint['embedding']
+
+    print('Building encoder and decoder ...')
+    # Initialize word embeddings
+    embedding = nn.Embedding(voc.num_words, hidden_size)
+    if os.path.exists(loadFilename):
+        embedding.load_state_dict(embedding_sd)
+
+    print('Initialize encoder & decoder models')
+    encoder = EncoderRNN(hidden_size, embedding, encoder_n_layers, dropout)
+    decoder = LuongAttnDecoderRNN(attn_model, embedding, hidden_size, voc.num_words, decoder_n_layers, dropout)
+    print('Building optimizers ...')
+    encoder_optimizer = optim.Adam(encoder.parameters(), lr=learning_rate)
+    decoder_optimizer = optim.Adam(decoder.parameters(), lr=learning_rate * decoder_learning_ratio)
+    if os.path.exists(loadFilename):
+        encoder.load_state_dict(encoder_sd)
+        decoder.load_state_dict(decoder_sd)
+        encoder_optimizer.load_state_dict(encoder_optimizer_sd)
+        decoder_optimizer.load_state_dict(decoder_optimizer_sd)
+
+    print('Models built and ready to go!')
+    return cp_start_iteration, encoder, decoder, encoder_optimizer, decoder_optimizer, embedding
diff --git "a/src/1.\350\257\255\350\250\200\345\244\204\347\220\206\344\270\216Python/test.py" "b/src/Python\350\207\252\347\204\266\350\257\255\350\250\200\345\244\204\347\220\206/test.py"
similarity index 100%
rename from "src/1.\350\257\255\350\250\200\345\244\204\347\220\206\344\270\216Python/test.py"
rename to "src/Python\350\207\252\347\204\266\350\257\255\350\250\200\345\244\204\347\220\206/test.py"