Tensorflow：深度卷积究竟做了什么？

Question

我想使用Tensorflow的。据我所知，它为每个通道执行常规2D卷积，每个通道都有depth_multiplier个特征。

然后我应该期望，如果depth_multiplier = 1，输入通道的数量与输出通道的数量相同。但为什么我有256个输入通道和512个输出通道？额外频道来自哪里？

Answer 1

过滤器的大小为[filter_height, filter_width, in_channels, channel_multiplier]。如果是channel_multiplier = 1，那么您将获得与输出相同数量的输入通道。如果是N，则N*input_channels为output channels，每个input channe l与N过滤器进行卷积。

例如，

inputs = tf.Variable(tf.random_normal((20, 64,64,100)))
filters = tf.Variable(tf.random_normal((5,5,100,10)))
out = tf.nn.depthwise_conv2d(
      inputs,
      filters,
      strides=[1,1,1,1],
      padding='SAME')

你得到out形状：shape=(20, 64, 64, 1000)

Answer 2

我修改了@vijay m的代码以进一步说明问题。他的回答绝对正确。但是，我仍然没有得到它。

快速回答是＆＃34;频道乘数＆＃34;这个论点是一个令人困惑的名字。它可以被称为＆＃34;您希望每个频道应用的过滤器数量＆＃34;。因此，请注意此代码段的大小：

# batch of 2 inputs of 13x13 pixels with 3 channels each. # Four 5x5 filters applied to each channel, so 12 total channels output inputs_np = np.ones((2, 13, 13, 3)) inputs = tf.constant(inputs_np) # Build the filters so that their behavior is easier to understand. For these filters # which are 5x5, I set the middle pixel (location 2,2) to some value and leave # the rest of the pixels at zero filters_np = np.zeros((5,5,3,4)) # 5x5 filters for 3 inputs and applying 4 such filters to each one. filters_np[2, 2, 0, 0] = 2.0 filters_np[2, 2, 0, 1] = 2.1 filters_np[2, 2, 0, 2] = 2.2 filters_np[2, 2, 0, 3] = 2.3 filters_np[2, 2, 1, 0] = 3.0 filters_np[2, 2, 1, 1] = 3.1 filters_np[2, 2, 1, 2] = 3.2 filters_np[2, 2, 1, 3] = 3.3 filters_np[2, 2, 2, 0] = 4.0 filters_np[2, 2, 2, 1] = 4.1 filters_np[2, 2, 2, 2] = 4.2 filters_np[2, 2, 2, 3] = 4.3 filters = tf.constant(filters_np) out = tf.nn.depthwise_conv2d( inputs, filters, strides=[1,1,1,1], padding='SAME') with tf.Session() as sess: sess.run(tf.global_variables_initializer()) out_val = out.eval() print("output cases 0 and 1 identical? {}".format(np.all(out_val[0]==out_val[1]))) print("One of the pixels for each of the 12 output {} ".format(out_val[0, 6, 6])) # Output: # output cases 0 and 1 identical? True # One of the pixels for each of the 12 output [ 2. 2.1 2.2 2.3 3. 3.1 3.2 3.3 4. 4.1 4.2 4.3]

它的大小允许您将10个不同的过滤器应用于每个输入通道。我已经创建了以前答案代码的一个版本，可能很有启发性：

final Logger rootLogger = Logger.getRootLogger();
for (Handler h: rootLogger.getHandlers())
    rootLogger.removeHandler(h);