Merge branch 'master' of https://github.com/dragen1860/Deep-Learning-with-TensorFlow-book

.github/workflows/pythonapp.yml

+name: Python application
+on: [push, pull_request]
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v1
+    - name: Set up Python 3
+      uses: actions/setup-python@v1
+      with:
+        python-version: 3.x
+    - name: Install dependencies
+      run: |
+        python -m pip install --upgrade pip
+        pip install flake8 pytest
+        pip install -r requirements.txt || true
+    - name: Lint with flake8
+      run: |
+        # stop the build if there are Python syntax errors or undefined names
+        flake8 . --count --select=E9,F63,F7,F82 --show-source --statistics
+        # exit-zero treats all errors as warnings. The GitHub editor is 127 chars wide
+        flake8 . --count --exit-zero --max-complexity=10 --max-line-length=127 --statistics
+    - name: Test with pytest
+      run: |
+        pytest || true

.gitignore

 *.DS_Store
+*.bak
\ No newline at end of file

README.md

@@ -2,7 +2,15 @@
 
 基于TensorFlow 2.0正式版！！！
 
-包含电子书，配套源代码等，时间仓促，源代码还没有整理完全。
+包含电子书，配套源代码等，时间仓促，源代码还没有整理完全~~
+
+**Anyone interested in translating the Chinese version book into English is Welcome! pls contact me. We can co-author.**
+
+-  **本书被“机器之心”，“量子位”等权威媒体报导！**
+-  **本库在Github趋势日榜单连续多天全球排名第1！**
+
+
+>主页上方有个“Clone or Download”绿色按钮，下载整个仓库即可。之所以显示在线打不开是因为Github在国外，连接不稳定就会提示打不开。
 
 <p align="center">
   <img src="assets/1.jpg" align="center" width="600">
@@ -18,6 +26,12 @@ https://github.com/dragen1860/TensorFlow-2.x-Tutorials
 
 -	联系邮箱(一般问题建议Github issues交流)：liangqu.long AT gmail.com
 
+- 使用本书本的任何内容时(仅限个人的非商业用途)，请注明作者和Github链接
+
+# 介绍短视频
+
+https://www.bilibili.com/video/av75331861
+
 # 目录
 
 <p align="center">
@@ -29,11 +43,13 @@ https://github.com/dragen1860/TensorFlow-2.x-Tutorials
 
 # 初学者交流QQ群
 
-人工智能101学院：295208768
+- 人工智能101学院-免费2群：535205236
+
+- 人工智能101学院-免费1群：295208768(已满)
 
 #	配套视频课程
 
-收费，提供答疑等全方位服务，不喜勿喷。
+收费，适合零基础、希望快速入门AI的朋友，提供答疑、指导等全方位服务。
 
 - 深度学习与TensorFlow入门实战
 https://study.163.com/course/courseMain.htm?share=2&shareId=480000001847407&courseId=1209092816&_trace_c_p_k2_=9e74eb6f891d47cfaa6f00b5cb5f617c

ch01/tf1.py

-import tensorflow as tf
-assert tf.__version__.startswith('1.')
+import tensorflow.compat.v1 as tf
+tf.disable_v2_behavior()
+assert tf.__version__.startswith('2.')
 
 # 1.创建计算图阶段
 # 创建2个输入端子，指定类型和名字

ch03/demo.py

+import  tensorflow as tf
+from    tensorflow.keras import datasets, layers, optimizers, Sequential, metrics
+
+
+# 设置GPU使用方式
+# 获取GPU列表
+gpus = tf.config.experimental.list_physical_devices('GPU')
+if gpus:
+  try:
+    # 设置GPU为增长式占用
+    for gpu in gpus:
+      tf.config.experimental.set_memory_growth(gpu, True) 
+  except RuntimeError as e:
+    # 打印异常
+    print(e)
+
+(xs, ys),_ = datasets.mnist.load_data()
+print('datasets:', xs.shape, ys.shape, xs.min(), xs.max())
+
+batch_size = 32
+
+xs = tf.convert_to_tensor(xs, dtype=tf.float32) / 255.
+db = tf.data.Dataset.from_tensor_slices((xs,ys))
+db = db.batch(batch_size).repeat(30)
+
+
+model = Sequential([layers.Dense(256, activation='relu'), 
+                     layers.Dense(128, activation='relu'),
+                     layers.Dense(10)])
+model.build(input_shape=(4, 28*28))
+model.summary()
+
+optimizer = optimizers.SGD(lr=0.01)
+acc_meter = metrics.Accuracy()
+
+for step, (x,y) in enumerate(db):
+
+    with tf.GradientTape() as tape:
+        # 打平操作，[b, 28, 28] => [b, 784]
+        x = tf.reshape(x, (-1, 28*28))
+        # Step1. 得到模型输出output [b, 784] => [b, 10]
+        out = model(x)
+        # [b] => [b, 10]
+        y_onehot = tf.one_hot(y, depth=10)
+        # 计算差的平方和，[b, 10]
+        loss = tf.square(out-y_onehot)
+        # 计算每个样本的平均误差，[b]
+        loss = tf.reduce_sum(loss) / x.shape[0]
+
+
+    acc_meter.update_state(tf.argmax(out, axis=1), y)
+
+    grads = tape.gradient(loss, model.trainable_variables)
+    optimizer.apply_gradients(zip(grads, model.trainable_variables))
+
+
+    if step % 200==0:
+
+        print(step, 'loss:', float(loss), 'acc:', acc_meter.result().numpy())
+        acc_meter.reset_states()

ch03/readMNIST.py

+#!/usr/bin/python
+# -*- coding: utf-8 -*-
+
+"""
+从MNIST中读取原始图片并保存、读取标签数据并保存。
+MNIST文件结构分析可以参考：https://blog.csdn.net/justidle/article/details/103149253
+"""
+"""
+使用方法：
+1、将MNIST的文件下载到本地。
+2、在py文件所在目录下，建立mnist_data目录。然后将MNIST的四个文件拷贝到mnist_data目录，并解压
+3、在py文件所在目录下，建立test目录，改目录用于存放解压出的图片文件和标签文件
+"""
+
+import struct
+import numpy as np
+import PIL.Image
+    
+def read_image(filename):
+    #打开文件
+    f = open(filename, 'rb')
+    
+    #读取文件内容
+    index = 0
+    buf = f.read()
+    
+    #关闭文件
+    f.close()
+    
+    #解析文件内容
+    #>IIII 表示使用大端规则，读取四个整型
+    magic, numImages, rows, columns = struct.unpack_from('>IIII', buf, index)
+    index += struct.calcsize('>IIII')
+    
+    for i in range(0, numImages):
+        # L代表灰度图片
+        image = PIL.Image.new('L', (columns, rows))
+        
+        for x in range(rows):
+            for y in range(columns):
+                # ‘>B' 读取一个字节
+                image.putpixel((y,x), int(struct.unpack_from('>B', buf, index)[0]))
+                index += struct.calcsize('>B')
+                
+        print('save ' + str(i) + 'image')
+        image.save('mnist_data/test/'+str(i)+'.png')
+        
+def read_label(filename, saveFilename):
+    f = open(filename, 'rb')
+    index = 0
+    buf = f.read()
+    f.close()
+    
+    magic, labels = struct.unpack_from('>II' , buf , index)
+    index += struct.calcsize('>II')
+    
+    labelArr = [0] * labels
+    
+    for x in range(labels):
+        labelArr[x] = int(struct.unpack_from('>B', buf, index)[0])
+        index += struct.calcsize('>B')
+    
+    save = open(saveFilename, 'w')
+    save.write(','.join(map(lambda x: str(x), labelArr)))
+    save.write('\n')
+    save.close()
+    print('save labels success')
+
+if __name__ == '__main__':
+    #注意t10k-images-idx3-ubyte里面一共有10,000张图片
+    read_image('mnist_data/t10k-images-idx3-ubyte')
+    read_label('mnist_data/t10k-labels-idx1-ubyte', 'mnist_data/test/label.txt')
+    
\ No newline at end of file

ch11/nb.py

@@ -16,7 +16,7 @@ out
 #%%
 net.embeddings
 net.embeddings.trainable
-net.trainable = Flase
+net.trainable = False
 #%%
 # 从预训练模型中加载词向量表
 embed_glove = load_embed('glove.6B.50d.txt')