tensorflow下载安装

最新推荐文章于 2025-05-14 12:38:28 发布

青年有志

最新推荐文章于 2025-05-14 12:38:28 发布

阅读量5.1k

点赞数

CC 4.0 BY-SA版权

文章标签： tensorflow keras 深度学习

本文链接：https://blog.csdn.net/qq_46450354/article/details/125143741

本文档详细介绍了TensorFlow的下载安装过程，以及如何使用Session执行计算。通过示例解释了Variable的概念，展示了如何初始化和更新Variable。此外，还讲解了Placeholder的基本用法，允许在运行时传递参数。最后，提到了在构建多层神经网络时需要注意的梯度问题。文章以一个简单的神经网络搭建为例，演示了TensorFlow的训练过程。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

tensorflow 官网

一、tensorflow 下载安装

在这里插入图片描述 tensorflow详细安装过程
 大佬介绍

下载完 tensorflow 后需要下载相应的 keras 版本
在这里插入图片描述

tensorflow 与 keras 对应的版本

二、Session

执行命令的东西

import tensorflow as tf

matrix1 = tf.constant([[3, 3]])
matrix2 = tf.constant([[2],
                       [2]])
# print(matrix1)
# print(matrix2)

# matrix multiply np.dot(m1, m2) Go ahead and multiply the column
product = tf.matmul(matrix1, matrix2)

# method 1 . tf.Session() have removed
sess = tf.compat.v1.Session()

# Each run executes a result
result = sess.run(product)

# method 2
with tf.Session() as sess:
    result2 = sess.run(product)
    print(result2)

三、Variable

import tensorflow as tf
__all__ = [tf]

state = tf.Variable(0, name='counter')

print(state.name)

one = tf.constant(1)
new_value = tf.add(state, one)

# new_value load in state, so state is new_value
updata = tf.assign(state, new_value)

# must have if define variable
init = tf.initialize_all_variables()

with tf.Session() as sess:
    # At first must have the run(init) to initial
    sess.run(init)
    for _ in range(3):
        sess.run(updata)

        # you should print though this if you want to see result
        print(sess.run(state))

四、placeholder

import tensorflow as tf

# placeholder show you can get a parameter when sess is running
input1 = tf.placeholder(tf.float32)
input2 = tf.placeholder(tf.float32)

output = tf.multiply(input1, input2)

with tf.Session() as sess:
    # Through type dict to express the value
    print(sess.run(output, feed_dict={input1:[7.], input2:[2.]}))

在这里插入图片描述
当使用多层神经网络是，慎重考虑激励函数，因为可能导致梯度消失和梯度爆炸

五、 tensorflow 搭建神经网络

import tensorflow as tf
import numpy as np

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

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

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

l1 = add_layer(xs, 1, 10, activation_function=tf.nn.relu)
predition = add_layer(l1, 10, 1, activation_function=None)

# MSE,
loss = tf.reduce_mean(tf.reduce_sum(tf.square(ys - predition), reduction_indices=[1]))

# by 0.1 step upgrade
train_step = tf.train.GradientDescentOptimizer(0.1).minimize(loss)

# must initial the variable
init = tf.initialize_all_variables()
sess = tf.Session()
sess.run(init)

for i in range(1000):
    sess.run(train_step, feed_dict={xs:x_data, ys:y_data})
    if i % 100 == 0:
        print(sess.run(loss, feed_dict={xs:x_data, ys:y_data}))