tensorflow 卷积/反卷积-池化/反池化操作详解

Plese see this answer for a detailed example of how tf.nn.conv2d_backprop_input and tf.nn.conv2d_backprop_filter in an example.

In tf.nn, there are 4 closely related 2d conv functions:

tf.nn.conv2d
tf.nn.conv2d_backprop_filter
tf.nn.conv2d_backprop_input
tf.nn.conv2d_transpose

def conv2d(input, filter, strides, padding, use_cudnn_on_gpu=True, data_format="NHWC", name=None):
  r"""Computes a 2-D convolution given 4-D `input` and `filter` tensors.

  Given an input tensor of shape `[batch, in_height, in_width, in_channels]`
  and a filter / kernel tensor of shape
  `[filter_height, filter_width, in_channels, out_channels]`, this op
  performs the following:

  1. Flattens the filter to a 2-D matrix with shape
     `[filter_height * filter_width * in_channels, output_channels]`.
  2. Extracts image patches from the input tensor to form a *virtual*
     tensor of shape `[batch, out_height, out_width,
     filter_height * filter_width * in_channels]`.
  3. For each patch, right-multiplies the filter matrix and the image patch
     vector.

  In detail, with the default NHWC format,

      output[b, i, j, k] =
          sum_{di, dj, q} input[b, strides[1] * i + di, strides[2] * j + dj, q] *
                          filter[di, dj, q, k]

  Must have `strides[0] = strides[3] = 1`.  For the most common case of the same
  horizontal and vertices strides, `strides = [1, stride, stride, 1]`.

Given out = conv2d(x, w) and the output gradient d_out:

Use tf.nn.conv2d_backprop_filter to compute the filter gradient d_w
Use tf.nn.conv2d_backprop_input to compute the filter gradient d_x
tf.nn.conv2d_backprop_input can be implemented by tf.nn.conv2d_transpose
All 4 functions above can be implemented by tf.nn.conv2d
Actually, use TF's autodiff is the fastest way to compute gradients

Long Answer

Now, let's give an actual working code example of how to use the 4 functions above to compute d_x and d_w given d_out. This shows how conv2d, conv2d_backprop_filter, conv2d_backprop_input, and conv2d_transpose are related to each other. Please find the full scripts here.

Computing d_x in 4 different ways:

# Method 1: TF's autodiff
d_x = tf.gradients(f, x)[0]

# Method 2: manually using conv2d
d_x_manual = tf.nn.conv2d(input=tf_pad_to_full_conv2d(d_out, w_size),
                          filter=tf_rot180(w),
                          strides=strides,
                          padding='VALID')

# Method 3: conv2d_backprop_input
d_x_backprop_input = tf.nn.conv2d_backprop_input(input_sizes=x_shape,
                                                 filter=w,
                                                 out_backprop=d_out,
                                                 strides=strides,
                                                 padding='VALID')

# Method 4: conv2d_transpose
d_x_transpose = tf.nn.conv2d_transpose(value=d_out,
                                       filter=w,
                                       output_shape=x_shape,
                                       strides=strides,
                                       padding='VALID')

Computing d_w in 3 different ways:

# Method 1: TF's autodiff
d_w = tf.gradients(f, w)[0]

# Method 2: manually using conv2d
d_w_manual = tf_NHWC_to_HWIO(tf.nn.conv2d(input=x,
                                          filter=tf_NHWC_to_HWIO(d_out),
                                          strides=strides,
                                          padding='VALID'))

# Method 3: conv2d_backprop_filter
d_w_backprop_filter = tf.nn.conv2d_backprop_filter(input=x,
                                                   filter_sizes=w_shape,
                                                   out_backprop=d_out,
                                                   strides=strides,
                                                   padding='VALID')

Please see the full scripts for the implementation of tf_rot180, tf_pad_to_full_conv2d, tf_NHWC_to_HWIO. In the scripts, we check that the final output values of different methods are the same; a numpy implementation is also available.