tensorboard可视化(二)

2019-09-20 15:11:06 浏览数 (1)

tensorboard可视化(二)

1.导包

代码语言:javascript复制
import tensorflow as tf
import numpy as np

2.make up some data

代码语言:javascript复制
x_data = np.linspace(-1, 1, 300, dtype=np.float32)[:, np.newaxis]
noise = np.random.normal(0, 0.05, x_data.shape).astype(np.float32)
y_data = np.square(x_data) - 0.5   noise

3.将xs和ys包含进来,形成一个大的图层,图层名字叫做inputs

代码语言:javascript复制
with tf.name_scope('inputs'):
    # 为xs指定名称x_input
    xs = tf.placeholder(tf.float32, [None, 1],name='x_input')
    # 为ys指定名称y_input
    ys = tf.placeholder(tf.float32, [None, 1],name='y_input')

4.在 layer 中为 Weights, biases 设置变化图表

代码语言:javascript复制
# add_layer多加一个n_layer参数(表示第几层)
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):
         # 对weights进行绘制图标
         with tf.name_scope('weights'):
              Weights= tf.Variable(tf.random_normal([in_size, out_size]),name='W')
              tf.summary.histogram(layer_name   '/weights', Weights)
          # 对biases进行绘制图标
         with tf.name_scope('biases'):
              biases = tf.Variable(tf.zeros([1,out_size]) 0.1, name='b')
              tf.summary.histogram(layer_name   '/biases', biases)
         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)
         # 对outputs进行绘制图标
         tf.summary.histogram(layer_name   '/outputs', outputs)
    return outputs

5.修改隐藏层与输出层

代码语言:javascript复制
# 由于我们对addlayer 添加了一个参数, 所以修改之前调用addlayer()函数的地方. 对此处进行修改:
# 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)

6.设置loss的变化图

代码语言:javascript复制
#  loss是在tesnorBorad 的event下面的, 这是由于我们使用的是tf.scalar_summary() 方法.
with tf.name_scope('loss'):
    loss = tf.reduce_mean(tf.reduce_sum(
        tf.square(ys - prediction), reduction_indices=[1]))
    tf.summary.scalar('loss', loss)  # tensorflow >= 0.12

7.给所有训练图合并

代码语言:javascript复制
# 机器学习提升准确率
with tf.name_scope('train'):
    train_step=tf.train.GradientDescentOptimizer(0.1).minimize(loss) # 0.1表示学习效率

# 初始化
sess= tf.Session()
merged = tf.summary.merge_all()
writer = tf.summary.FileWriter("logs/", sess.graph) #
sess.run(tf.global_variables_initializer())

8.训练数据

代码语言:javascript复制
for i in range(1000):
   sess.run(train_step, feed_dict={xs:x_data, ys:y_data})
   if iP == 0:
      rs = sess.run(merged,feed_dict={xs:x_data,ys:y_data})
      writer.add_summary(rs, i)

0 人点赞