tensorflow学习（2）

最新推荐文章于 2024-08-20 19:00:16 发布

onejuliar

最新推荐文章于 2024-08-20 19:00:16 发布

阅读量345

点赞数

分类专栏： TensorFlow

本文链接：https://blog.csdn.net/juliarjuliar/article/details/89149084

版权

TensorFlow 专栏收录该内容

9 篇文章 0 订阅

订阅专栏

这篇博客介绍了如何使用TensorFlow构建简单的神经网络，包括激励函数、添加层、构建神经网络，并探讨了可视化工具TensorBoard的使用，以及在PyCharm中遇到的问题和解决方法。还提到了优化器和TensorFlow的7种优化器，以及通过TensorBoard监控训练过程。

摘要由CSDN通过智能技术生成

莫凡2017tensorflow（使用更简便版）https://github.com/MorvanZhou/Tensorflow-Tutorial

10.激励函数activation function：即让某一部分的神经元先激活起来，把激活效应的信息传递到后面一层的神经系统里面。（激励方程、存放位置）

11.添加层：如何定义并添加一个神经层

https://morvanzhou.github.io/tutorials/machine-learning/tensorflow/3-1-add-layer/

import tensorflow as tf

def add_layer(inputs, in_size, out_size, activation_function=None):
    Weights = tf.Variable(tf.random_normal([in_size, out_size]))
    biases = tf.Variable(tf.zeros([1, out_size]) + 0.1)
    Wx_plus_b = tf.matmul(inputs, Weights) + biases
    if activation_function is None:
        outputs = Wx_plus_b
    else:
        outputs = activation_function(Wx_plus_b)
    return outputs

其中，定义添加神经层的函数def add_layer(),它有四个参数：输入值、输入的大小、输出的大小和激励函数，我们设定默认的激励函数是None。在生成初始参数时，随机变量(normal distribution)会比全部为0要好很多，所以我们这里的weights为一个in_size行, out_size列的随机变量矩阵。在机器学习中，biases的推荐值不为0，所以我们这里是在0向量的基础上又加了0.1。

12.建造神经网络：在11的基础上，继续讲解如何构造神经层

xs = tf.placeholder(tf.float32, [None, 1])

tf.placeholder()就是代表占位符，这里的None代表无论输入有多少都可以，因为输入只有一个特征，所以这里是1。

通常神经层都包括输入层、隐藏层和输出层。这里的输入层只有一个属性，所以我们就只有一个输入；隐藏层我们可以自己假设，这里我们假设隐藏层有10个神经元；输出层和输入层的结构是一样的，所以我们的输出层也是只有一层。所以，我们构建的是——输入层1个、隐藏层10个、输出层1个的神经网络。

import tensorflow as tf
import numpy as np

def add_layer(inputs, in_size, out_size, activation_function=None):
    # add one more layer and return the output of this layer
    Weights = tf.Variable(tf.random_normal([in_size, out_size]))
    biases = tf.Variable(tf.zeros([1, out_size]) + 0.1)
    Wx_plus_b = tf.matmul(inputs, Weights) + biases
    if activation_function is None:
        outputs = Wx_plus_b
    else:
        outputs = activation_function(Wx_plus_b)
    return outputs

# Make up some real data
x_data = np.linspace(-1,1,300)[:, np.newaxis]  
noise = np.random.normal(0, 0.05, x_data.shape)  #噪声生成
y_data = np.square(x_data) - 0.5 + noise

# define placeholder for inputs to network
xs = tf.placeholder(tf.float32, [None, 1])
ys = tf.placeholder(tf.float32, [None, 1])
# add hidden layer
l1 = add_layer(xs, 1, 10, activation_function=tf.nn.relu)
# add output layer
prediction = add_layer(l1, 10, 1, activation_function=None)

# the error between prediciton and real data
loss = tf.reduce_mean(tf.reduce_sum(tf.square(ys - prediction),
                     reduction_indices=[1]))  #对二者差的平方求和再取平均
train_step = tf.train.GradientDescentOptimizer(0.1).minimize(loss)

# important step
init = tf.initialize_all_variables()
sess = tf.Session()
sess.run(init)

for i in range(1000):
    # training
    sess.run(train_step, feed_dict={xs: x_data, ys: y_data})
    if i % 50 == 0:
        # to see the step improvement
        print(sess.run(loss, feed_dict={xs: x_data, ys: y_data}))
'''
0.066639826
0.009021594
0.006373957
0.005147705
0.004446703
0.004053277
0.0037726625
0.003591398
0.0034521723
0.0033391833
0.003239072
0.0031602075
0.0031008963
0.003050547
0.0030073707
0.0029714515
0.002940909
0.0029167237
0.0028977306
0.0028768817
'''

13.可视化

注意：plt.ion()用于连续显示。（一般情况下，plt.show()之后的程序不会执行；plt.ion()这个函数，使matplotlib的显示模式转换为交互（interactive）模式）https://blog.csdn.net/zbrwhut/article/details/80625702
try的用法
抹除上次绘制的曲线

 

"""
Please note, this code is only for python 3+. If you are using python 2+, please modify the code accordingly.
"""
import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt

def add_layer(inputs, in_size, out_size, activation_function=None):
    # add one more layer and return the output of this layer
    Weights = tf.Variable(tf.random_normal([in_size, out_size]))
    biases = tf.Variable(tf.zeros([1, out_size]) + 0.1)
    Wx_plus_b = tf.matmul(inputs, Weights) + biases
    if activation_function is None:
        outputs = Wx_plus_b
    else:
        outputs = activation_function(Wx_plus_b)
    return outputs

# Make up some real data
x_data = np.linspace(-1,1,300)[:, np.newaxis]
noise = np.random.normal(0, 0.05, x_data.shape)
y_data = np.square(x_data) - 0.5 + noise

# define placeholder for inputs to network
xs = tf.placeholder(tf.float32, [None, 1])
ys = tf.placeholder(tf.float32, [None, 1])
# add hidden layer
l1 = add_layer(xs, 1, 10, activation_function=tf.nn.relu)
# add output layer
prediction = add_layer(l1, 10, 1, activation_function=None)

# the error between prediciton and real data
loss = tf.reduce_mean(tf.reduce_sum(tf.square(ys - prediction),
                     reduction_indices=[1]))
train_step = tf.train.GradientDescentOptimizer(0.1).minimize(loss)

# important step
init = tf.initialize_all_variables()
sess = tf.Session()
sess.run(init)

# plot the real data
fig = plt.figure()
ax = fig.add_subplot(1,1,1)
ax.scatter(x_data, y_data)
plt.ion()
plt.show()

for i in range(1000):
    # training
    sess.run(train_step, feed_dict={xs: x_data, ys: y_data})
    if i % 50 == 0:
        # to visualize the result and improvement
        try:
            ax.lines.remove(lines[0])
        except Exception:
            pass
        prediction_value = sess.run(prediction, feed_dict={xs: x_data})
        # plot the prediction
        lines = ax.plot(x_data, prediction_value, 'r-', lw=5)
        plt.pause(0.1)

其中一个运行结果截图如下：

pycharm运行时，开始会有不出现红色这条动态线的情况，参考https://morvanzhou.github.io/tutorials/machine-learning/tensorflow/3-3-visualize-result/，将Setting->Tools->Python Scientific->Show Toolwindow ,然后取消勾选就可以里，sciview就可以弹出来，然后就好用了。

14.优化器optimizer

tensorflow的7种优化器

莫凡的更多关于optimizer的解释，参考https://morvanzhou.github.io/tutorials/machine-learning/ML-intro/3-06-speed-up-learning/

15.可视化好助手tensorboard

https://morvanzhou.github.io/tutorials/machine-learning/tensorflow/4-1-tensorboard1/

学会用 Tensorflow 自带的 tensorboard 去可视化我们所建造出来的神经网络是一个很好的学习理解方式. 用最直观的流程图告诉你你的神经网络是长怎样,有助于你发现编程中间的问题和疑问.与 tensorboard 兼容的浏览器是 “Google Chrome”. 使用其他的浏览器不保证所有内容都能正常显示.

 

"""
Please note, this code is only for python 3+. If you are using python 2+, please modify the code accordingly.
"""
import tensorflow as tf
import os

os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'

def add_layer(inputs, in_size, out_size, activation_function=None):
    # add one more layer and return the output of this layer
    with tf.name_scope('layer'):
        with tf.name_scope('weights'):
            Weights = tf.Variable(tf.random_normal([in_size, out_size]), name='W')
        with tf.name_scope('biases'):
            biases = tf.Variable(tf.zeros([1, out_size]) + 0.1, name='b')
        with tf.name_scope('Wx_plus_b'):
            Wx_plus_b = tf.add(tf.matmul(inputs, Weights), biases)
        if activation_function is None:
            outputs = Wx_plus_b
        else:
            outputs = activation_function(Wx_plus_b, )
        return outputs


# define placeholder for inputs to network
with tf.name_scope('inputs'):
    xs = tf.placeholder(tf.float32, [None, 1], name='x_input')
    ys = tf.placeholder(tf.float32, [None, 1], name='y_input')

# add hidden layer
l1 = add_layer(xs, 1, 10, activation_function=tf.nn.relu)
# add output layer
prediction = add_layer(l1, 10, 1, activation_function=None)

# the error between prediciton and real data
with tf.name_scope('loss'):
    loss = tf.reduce_mean(tf.reduce_sum(tf.square(ys - prediction),
                                        reduction_indices=[1]))

with tf.name_scope('train'):
    train_step = tf.train.GradientDescentOptimizer(0.1).minimize(loss)

sess = tf.Session()
#writer = tf.train.SummaryWriter("logs/", sess.graph)
writer = tf.summary.FileWriter("logs/", sess.graph)  #https://blog.csdn.net/zeuseign/article/details/72771598
# important step
sess.run(tf.initialize_all_variables())

其中

os.environ[“TF_CPP_MIN_LOG_LEVEL”]=‘1’ # 这是默认的显示等级，显示所有信息
os.environ[“TF_CPP_MIN_LOG_LEVEL”]=‘2’ # 只显示 warning 和 Error
os.environ[“TF_CPP_MIN_LOG_LEVEL”]=‘3’ # 只显示 Error

我们一般用第二个就可以了。

使用with tf.name_scope('inputs')可以将xs和ys包含进来，形成一个大的图层，图层的名字就是with tf.name_scope()方法里的参数。

莫凡编译文件目录如下，

想直接运行，

1.打开终端

2.在终端中打开至logs上层目录

3.输入 tensorboard --logdir logs

出现以下问题：（http://PC201710130840，在谷歌浏览器中打不开）

且会报错如下：

参考https://morvanzhou.github.io/tutorials/machine-learning/tensorflow/4-1-tensorboard1/下评论的windows上的解决方法，https://blog.csdn.net/u011228094/article/details/85634261，直接在相应的F下执行

tensorboard --logdir=F:\BaiduNetdiskDownload\04-深度学习课程\TensorFlow教程\Tensorflow视频教程\tensorflowTUT源码\tf14_tensorboard\logs，出现了http://127.0.0.1:6006，但是在谷歌浏览器中仍然打不开https://blog.csdn.net/Sunny_Guang/article/details/79914068

考虑主要是因为Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 提示，原因是下载的tensorflow版本不支持cpu的AVX2编译。参考https://blog.csdn.net/yimixgg/article/details/80676619未找到解决办法。可能需要重新安装有AVX2编译的tensorflow版本https://github.com/fo40225/tensorflow-windows-wheel和https://github.com/lakshayg/tensorflow-build。

16.可视化好助手2

上一篇讲到了如何可视化TesorBorad整个神经网络结构的过程。其实tensorboard还可以可视化训练过程( biase变化过程) , 这节重点讲一下可视化训练过程的图标是如何做的。

 

"""
Please note, this code is only for python 3+. If you are using python 2+, please modify the code accordingly.
"""
import tensorflow as tf
import numpy as np
import os

os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'

def add_layer(inputs, in_size, out_size, n_layer, activation_function=None):
    # add one more layer and return the output of this layer
    layer_name = 'layer%s' % n_layer
    with tf.name_scope(layer_name):
        with tf.name_scope('weights'):
            Weights = tf.Variable(tf.random_normal([in_size, out_size]), name='W')
            #tf.histogram_summary(layer_name + '/weights', Weights)
            tf.summary.histogram(layer_name + '/weights', Weights)  # tensorflow >= 0.12
        with tf.name_scope('biases'):
            biases = tf.Variable(tf.zeros([1, out_size]) + 0.1, name='b')
            #tf.histogram_summary(layer_name + '/biases', biases)
            tf.summary.histogram(layer_name + '/biases', biases)  # Tensorflow >= 0.12
        with tf.name_scope('Wx_plus_b'):
            Wx_plus_b = tf.add(tf.matmul(inputs, Weights), biases)
        if activation_function is None:
            outputs = Wx_plus_b
        else:
            outputs = activation_function(Wx_plus_b, )
        #tf.histogram_summary(layer_name + '/outputs', outputs)
        tf.summary.histogram(layer_name + '/outputs', outputs)  # Tensorflow >= 0.12
        return outputs


# Make up some real data
x_data = np.linspace(-1, 1, 300)[:, np.newaxis]
noise = np.random.normal(0, 0.05, x_data.shape)
y_data = np.square(x_data) - 0.5 + noise

# define placeholder for inputs to network
with tf.name_scope('inputs'):
    xs = tf.placeholder(tf.float32, [None, 1], name='x_input')
    ys = tf.placeholder(tf.float32, [None, 1], name='y_input')

# add hidden layer
l1 = add_layer(xs, 1, 10, n_layer=1, activation_function=tf.nn.relu)
# add output layer
prediction = add_layer(l1, 10, 1, n_layer=2, activation_function=None)

# the error between prediciton and real data
with tf.name_scope('loss'):
    loss = tf.reduce_mean(tf.reduce_sum(tf.square(ys - prediction),
                                        reduction_indices=[1]))
    #tf.scalar_summary('loss', loss)
    tf.summary.scalar('loss', loss)  # tensorflow >= 0.12

with tf.name_scope('train'):
    train_step = tf.train.GradientDescentOptimizer(0.1).minimize(loss)

sess = tf.Session()
#merged = tf.merge_all_summaries()  # tensorflow < 0.12
merged = tf.summary.merge_all() # tensorflow >= 0.12
#writer = tf.train.SummaryWriter("logs/", sess.graph)
writer = tf.summary.FileWriter("logs/", sess.graph) # tensorflow >=0.12
# important step
#sess.run(tf.initialize_all_variables()) # tf.initialize_all_variables() # tf 马上就要废弃这种写法
sess.run(tf.global_variables_initializer())  # 替换成这样就好
for i in range(1000):
    sess.run(train_step, feed_dict={xs: x_data, ys: y_data})
    if i % 50 == 0:
        result = sess.run(merged,
                          feed_dict={xs: x_data, ys: y_data})
        writer.add_summary(result, i)

在.py文件平行文件中成功生成logs及其下级文件。