revise readme

1-3,文本数据建模流程范例.md

@@ -180,7 +180,7 @@ _________________________________________________________________
 @tf.function
 def printbar():
     today_ts = tf.timestamp()%(24*60*60)
-
+    
     hour = tf.cast(today_ts//3600+8,tf.int32)%tf.constant(24)
     minite = tf.cast((today_ts%3600)//60,tf.int32)
     second = tf.cast(tf.floor(today_ts%60),tf.int32)

3-1,低阶API示范.md

@@ -26,7 +26,6 @@ def printbar():
                 timeformat(second)],separator = ":")
     tf.print("=========="*8+timestring)
 
-
 ```
 
 ```python
@@ -86,7 +85,7 @@ def data_iter(features, labels, batch_size=8):
     np.random.shuffle(indices)  #样本的读取顺序是随机的
     for i in range(0, num_examples, batch_size):
         indexs = indices[i: min(i + batch_size, num_examples)]
-        yield tf.gather(X,indexs), tf.gather(Y,indexs)
+        yield tf.gather(features,indexs), tf.gather(labels,indexs)
 
 # 测试数据管道效果   
 batch_size = 8
@@ -356,7 +355,7 @@ def data_iter(features, labels, batch_size=8):
     np.random.shuffle(indices)  #样本的读取顺序是随机的
     for i in range(0, num_examples, batch_size):
         indexs = indices[i: min(i + batch_size, num_examples)]
-        yield tf.gather(X,indexs), tf.gather(Y,indexs)
+        yield tf.gather(features,indexs), tf.gather(labels,indexs)
 
 # 测试数据管道效果   
 batch_size = 10

5-5,损失函数losses.md

@@ -33,7 +33,7 @@ model.add(layers.Dense(64, input_dim=64,
 model.add(layers.Dense(10,
         kernel_regularizer=regularizers.l1_l2(0.01,0.01),activation = "sigmoid"))
 model.compile(optimizer = "rmsprop",
-        loss = "sparse_categorical_crossentropy",metrics = ["AUC"])
+        loss = "binary_crossentropy",metrics = ["AUC"])
 model.summary()
 
 ```

6-2,训练模型的3种方法.md

@@ -14,8 +14,7 @@ from tensorflow.keras import *
 #打印时间分割线
 @tf.function
 def printbar():
-    ts = tf.timestamp()
-    today_ts = ts%(24*60*60)
+    today_ts = tf.timestamp()%(24*60*60)
 
     hour = tf.cast(today_ts//3600+8,tf.int32)%tf.constant(24)
     minite = tf.cast((today_ts%3600)//60,tf.int32)
@@ -29,8 +28,7 @@ def printbar():
 
     timestring = tf.strings.join([timeformat(hour),timeformat(minite),
                 timeformat(second)],separator = ":")
-    tf.print("=========="*8,end = "")
-    tf.print(timestring)
+    tf.print("=========="*8+timestring)
 
 ```
 

6-3,使用单GPU训练模型.md

@@ -38,8 +38,7 @@ from tensorflow.keras import *
 #打印时间分割线
 @tf.function
 def printbar():
-    ts = tf.timestamp()
-    today_ts = ts%(24*60*60)
+    today_ts = tf.timestamp()%(24*60*60)
 
     hour = tf.cast(today_ts//3600+8,tf.int32)%tf.constant(24)
     minite = tf.cast((today_ts%3600)//60,tf.int32)
@@ -53,8 +52,7 @@ def printbar():
 
     timestring = tf.strings.join([timeformat(hour),timeformat(minite),
                 timeformat(second)],separator = ":")
-    tf.print("=========="*8,end = "")
-    tf.print(timestring)
+    tf.print("=========="*8+timestring)
 
 ```
 

README.md

@@ -198,15 +198,14 @@ Please leave comments in the WeChat official account "Python与算法之美" (El
 ```python
 
 ```
-
-
-
 # 30天吃掉那只 TensorFlow2
 
 📚 gitbook电子书地址： https://lyhue1991.github.io/eat_tensorflow2_in_30_days
 
 🚀 github项目地址：https://github.com/lyhue1991/eat_tensorflow2_in_30_days
 
+🐳 kesci专栏地址：https://www.kesci.com/home/column/5d8ef3c3037db3002d3aa3a0
+
 
 ### 一，TensorFlow2 🍎 or Pytorch🔥
 
@@ -360,6 +359,10 @@ tf.keras绝大部分功能和兼容多种后端的Keras库用法完全一样，
 
 为了直接能够在jupyter中打开markdown文件，建议安装jupytext，将markdown转换成ipynb文件。
 
+**此外，本项目也与和鲸社区达成了合作，可以在和鲸专栏fork本项目，并直接在云笔记本上运行代码，避免环境配置痛苦。** 
+
+🐳和鲸专栏地址：https://www.kesci.com/home/column/5d8ef3c3037db3002d3aa3a0
+
 ```python
 #克隆本书源码到本地,使用码云镜像仓库国内下载速度更快
 #!git clone https://gitee.com/Python_Ai_Road/eat_tensorflow2_in_30_days
@@ -388,6 +391,9 @@ tensorflow version: 2.1.0
 hello tensorflow2
 ```
 
+
+
+
 ```python
 
 ```

push-to-github.py

@@ -27,7 +27,7 @@
 
 # !git push   origin master 
 
-# !git push gitee master 
+# !git push -f gitee master 
 
 # ## 创建pages分支