tensorflow中的tf.nn.conv2d学习

tf.nn.conv2d是TensorFlow里面实现卷积的函数

conv2d学习网址

import tensorflow as tf


input = tf.Variable(tf.random_normal([1,3,3,1]))#1表示一次读几张图片,3*3的图像大小,1表示通道的个数
filter = tf.Variable(tf.random_normal([3,3,1,2]))#3*3表示卷积核的大小,1表示通道的个数,2表示卷积核个数
op = tf.nn.conv2d(input, filter, strides=[1, 1, 1, 1], padding='VALID')
sess = tf.InteractiveSession()
sess.run(tf.global_variables_initializer())
print("input",sess.run(input))
print("filter",sess.run(filter))
print("op",sess.run(op))

实验结果如下:

"C:\Program Files\Anaconda3\python.exe" D:/pycharmprogram/tensorflow_learn/cnn_learn/conv2d_learn.py
2018-05-15 10:15:13.014600: W C:\tf_jenkins\home\workspace\rel-win\M\windows\PY\35\tensorflow\core\platform\cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations.
2018-05-15 10:15:13.014600: W C:\tf_jenkins\home\workspace\rel-win\M\windows\PY\35\tensorflow\core\platform\cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX2 instructions, but these are available on your machine and could speed up CPU computations.
input [[[[ 0.3438858 ]
   [-0.4425889 ]
   [-0.28974482]]

  [[-0.82606524]
   [ 0.3106512 ]
   [ 0.24117461]]

  [[ 0.12498179]
   [-0.12209705]
   [ 0.17391145]]]]
filter [[[[-1.2464828  -0.6758195 ]]

  [[ 0.6434889   1.2502667 ]]

  [[ 0.44064078  1.5831428 ]]]


 [[[ 1.2931986   1.9008949 ]]

  [[ 0.04882597 -0.2704289 ]]

  [[-0.203322    0.8987799 ]]]


 [[[-0.36723176  0.97302365]]

  [[ 0.7012152   1.1584816 ]]

  [[-0.8887519   0.6498683 ]]]]
op [[[[-2.2293346 -2.5887933]]]]

Process finished with exit code 0

阅读更多
想对作者说点什么?

博主推荐

换一批

没有更多推荐了,返回首页