diff --git a/README.md b/README.md
index 042b5d2..500f30e 100644
--- a/README.md
+++ b/README.md
@@ -101,7 +101,8 @@ Jeff回信原文：
 这样的一个高技术领域的文档，我们在翻译的过程中，难免会有不完善的地方，希望请大家一起帮助我们持续改进文档的翻译质量，帮助更多的人，方法：
 
 - 在GitHub上提Issue或Pull Request，地址为: [https://github.com/jikexueyuanwiki/tensorflow-zh](https://github.com/jikexueyuanwiki/tensorflow-zh)
-- 加入我们的QQ群提建议--协同翻译群：248320884，技术交流群：551830261
+- 加入TensorFlow技术交流群，与TensorFlower们一起研究交流技术干货--TensorFlow技术交流群：782484288
+- 对翻译感兴趣？加入协同翻译群：248320884，与翻译大神一道研究TensorFlow的本地化
 - 给我们写邮件： wiki@jikexueyuan.com
 
 ## 感谢支持
diff --git a/SOURCE/how_tos/variable_scope/index.md b/SOURCE/how_tos/variable_scope/index.md
index aadc0c0..abfbe15 100755
--- a/SOURCE/how_tos/variable_scope/index.md
+++ b/SOURCE/how_tos/variable_scope/index.md
@@ -55,7 +55,7 @@ result1 = my_image_filter(image1, variables_dict)
 result2 = my_image_filter(image2, variables_dict)
 ```
 虽然使用上面的方式创建变量是很方便的，但是在这个模块代码之外却破坏了其封装性：
-*  在构建试图的代码中标明变量的名字，类型，形状来创建.
+*  在构建视图的代码中标明变量的名字，类型，形状来创建.
 *  当代码改变了，调用的地方也许就会产生或多或少或不同类型的变量.
 
 解决此类问题的方法之一就是使用类来创建模块，在需要的地方使用类来小心地管理他们需要的变量.
diff --git a/SOURCE/tutorials/mnist_beginners.md b/SOURCE/tutorials/mnist_beginners.md
index f518f7e..a277e35 100755
--- a/SOURCE/tutorials/mnist_beginners.md
+++ b/SOURCE/tutorials/mnist_beginners.md
@@ -20,7 +20,7 @@ MNIST是一个入门级的计算机视觉数据集，它包含各种手写数字
 MNIST数据集的官网是[Yann LeCun's website](http://yann.lecun.com/exdb/mnist/)。在这里，我们提供了一份python源代码用于自动下载和安装这个数据集。你可以下载[这份代码](https://raw.githubusercontent.com/tensorflow/tensorflow/master/tensorflow/examples/tutorials/mnist/input_data.py)，然后用下面的代码导入到你的项目里面，也可以直接复制粘贴到你的代码文件里面。
 
 ```python
-import tensorflow.examples.tutorials.mnist.input_data
+import tensorflow.examples.tutorials.mnist.input_data as input_data
 mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)
 ```
 
diff --git a/SOURCE/tutorials/word2vec/word2vec_basic.py b/SOURCE/tutorials/word2vec/word2vec_basic.py
index 7176aa8..3785600 100755
--- a/SOURCE/tutorials/word2vec/word2vec_basic.py
+++ b/SOURCE/tutorials/word2vec/word2vec_basic.py
@@ -138,7 +138,7 @@ def generate_batch(batch_size, num_skips, skip_window):
   # tf.nce_loss automatically draws a new sample of the negative labels each
   # time we evaluate the loss.
   loss = tf.reduce_mean(
-      tf.nn.nce_loss(nce_weights, nce_biases, embed, train_labels,
+      tf.nn.nce_loss(nce_weights, nce_biases, train_labels,embed,
                      num_sampled, vocabulary_size))
 
   # Construct the SGD optimizer using a learning rate of 1.0.
@@ -216,7 +216,7 @@ def plot_with_labels(low_dim_embs, labels, filename='tsne.png'):
   tsne = TSNE(perplexity=30, n_components=2, init='pca', n_iter=5000)
   plot_only = 500
   low_dim_embs = tsne.fit_transform(final_embeddings[:plot_only,:])
-  labels = dictionary.keys()[:plot_only]
+  labels = list(dictionary.keys())[:plot_only]
   plot_with_labels(low_dim_embs, labels)
 
 except ImportError:
diff --git a/tex_pdf/get_started/c1s01_introduction.tex b/tex_pdf/get_started/c1s01_introduction.tex
index f1c8512..51a5b03 100644
--- a/tex_pdf/get_started/c1s01_introduction.tex
+++ b/tex_pdf/get_started/c1s01_introduction.tex
@@ -41,7 +41,7 @@ \section{Introduction  ||  简介}
 train = optimizer.minimize(loss)
 
 # Before starting, initialize the variables. We will 'run' this first.
-init = tf.initialize_all_variables()
+init = tf.global_variables_initializer()
 
 # Launch the graph.
 sess = tf.Session()
diff --git a/tex_pdf/get_started/c1s03_basic_usage.tex b/tex_pdf/get_started/c1s03_basic_usage.tex
index c89b59a..9a84a61 100644
--- a/tex_pdf/get_started/c1s03_basic_usage.tex
+++ b/tex_pdf/get_started/c1s03_basic_usage.tex
@@ -99,7 +99,7 @@ \subsection{Overview  |  总览}
 
 Ⓔ \textcolor{etc}{The default graph now has three nodes: two constant() ops and one matmul() op. To actually multiply the matrices, and get the result of the multiplication, you must launch the graph in a session.}
 
-Ⓒ 默认图现在拥有三个节点，两个\lstinline{constant()} op，一个\lstinline{matmul()} op. 为了真正进行矩阵乘法运算，得到乘法结果, 你必须在一个会话(session)中载入动这个图。
+Ⓒ 默认图现在拥有三个节点，两个\lstinline{constant()} op，一个\lstinline{matmul()} op. 为了真正进行矩阵乘法运算，得到乘法结果, 你必须在一个会话(session)中载入启动这个图。
 
 
 %%%
@@ -201,7 +201,7 @@ \subsection{Interactive Usage  |  交互式使用}
 x.initializer.run()
 
 # Add an op to subtract 'a' from 'x'.  Run it and print the result
-sub = tf.sub(x, a)
+sub = tf.subtract(x, a)
 print(sub.eval())
 # ==> [-2. -1.]
 
@@ -239,7 +239,7 @@ \subsection{Variables  |  变量}
 
 # Variables must be initialized by running an `init` Op after having
 # launched the graph.  We first have to add the `init` Op to the graph.
-init_op = tf.initialize_all_variables()
+init_op = tf.global_variables_initializer()
 
 # Launch the graph and run the ops.
 with tf.Session() as sess:
@@ -282,7 +282,7 @@ \subsection{Fetches  |  取回}
 input2 = tf.constant(2.0)
 input3 = tf.constant(5.0)
 intermed = tf.add(input2, input3)
-mul = tf.mul(input1, intermed)
+mul = tf.multiply(input1, intermed)
 
 with tf.Session() as sess:
   result = sess.run([mul, intermed])
@@ -311,7 +311,7 @@ \subsection{Feeds  |  供给}
 \begin{lstlisting}
 input1 = tf.placeholder(tf.float32)
 input2 = tf.placeholder(tf.float32)
-output = tf.mul(input1, input2)
+output = tf.multiply(input1, input2)
 
 with tf.Session() as sess:
   print(sess.run([output], feed_dict={input1:[7.], input2:[2.]}))
@@ -324,4 +324,4 @@ \subsection{Feeds  |  供给}
 
 Ⓒ 如果没有正确供给, \lstinline{placeholder()} 操作将会产生一个错误提示.关于feed的规模更大的案例，参见\hyperref[minist_tf]{MNIST 全连通 feed 教程}以及其\href{https://tensorflow.googlesource.com/tensorflow/+/master/tensorflow/g3doc/tutorials/mnist/fully_connected_feed.py}{源代码}。
 
-\href{http://tensorflow.org/get_started/basic_usage.md}{原文：Basic Usage}
\ No newline at end of file
+\href{http://tensorflow.org/get_started/basic_usage.md}{原文：Basic Usage}