TensorFlow-定义简单神经网络

TensorFlow-定义简单隐藏层

硬件：NVIDIA-GTX1080

软件：Windows7、python3.6.5、tensorflow-gpu-1.4.0

一、基础知识

矩阵运算基础知识：[300, 1] * [1, 10] -> [300, 10]

训练过程loss下降

二、代码展示

import tensorflow as tf
import numpy as np

def add_layer(inputs, in_size, out_size, activate_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 activate_function is None:
        outputs = Wx_plus_b
    else:
        outputs = activate_function(Wx_plus_b)

    return outputs

#[300, 1]
x_data = np.linspace(-1, 1, 300)[:, np.newaxis].astype(np.float32)
noise = np.random.normal(0, 0.05, x_data.shape).astype(np.float32)
y_data = np.square(x_data) - 0.5 + noise

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

#matrix calculate difference with convolution
#[300, 1] * [1, 10] -> [300, 10]
hide_layer = add_layer(xs, 1, 10, activate_function = tf.nn.relu)
#[300, 10] * [10, 1] -> [300, 1]
prediction = add_layer(hide_layer, 10, 1, activate_function = None)

#tf.reduce_sum -> add same rows: [300 , 1]->[300]
#tf.reduce_mean-> mean same cols: [300]->[1]
loss = tf.reduce_mean(tf.reduce_sum(tf.square(ys - prediction),reduction_indices = [1]))
optimizer = tf.train.GradientDescentOptimizer(0.1)
train_step = optimizer.minimize(loss)

init = tf.global_variables_initializer()

with tf.Session() as sess:
    sess.run(init)
    for i in range(1000):
        sess.run(train_step, feed_dict = {xs: x_data, ys: y_data})
        if i%50 == 0:
            print(sess.run(loss, feed_dict = {xs: x_data, ys: y_data}))

三、结果展示

0.12531051
0.011452716
0.0075810845
0.0061033727
0.005264972
0.0047253426
0.0043737977
0.0041175643
0.0039112023
0.0037404771
0.003616743
0.0035241707
0.0034555118
0.003394439
0.0033438925
0.0032840546
0.0032283582
0.0031608832
0.003105799
0.0030622235

 

任何问题请加唯一QQ2258205918（名称samylee）！

或唯一VX：samylee_csdn