tensorflow 卷积神经网络用到的一些函数

• tf.truncated_normal()

  产生截断正态分布随机数，取值范围为:[ mean - 2 * stddev, mean + 2 * stddev ]

  truncated_normal(
  shape, # 张量的维度
  mean=0.0, # 均值
  stddev=1.0, # 标准差
  dtype=tf.float32, # 数据类型
  seed=None, # 随机种子
  name=None # 名称
  )
  

• tf.nn.conv2d()

  https://blog.csdn.net/zuolixiangfisher/article/details/80528989

• tf.nn.bias_add()

  https://blog.csdn.net/mieleizhi0522/article/details/80416668

• tf.nn.relu()

  https://www.cnblogs.com/xh_chiang/p/9132524.html

• tf.nn.max_pool()

  https://www.jianshu.com/p/80e46fa0040f