转置卷积(反卷积)算术

时间:2017-10-13 17:21:13

标签: python machine-learning tensorflow conv-neural-network deconvolution

我使用tensorflow来构建卷积神经网络。给定一个形状的张量(无,16,16,4,192)我想执行一个转置卷积,导致形状(无,32,32,7,192)。

过滤器大小[2,2,4,192,192]和步幅[2,2,1,1,1]会产生我想要的输出形状吗?

1 个答案:

答案 0 :(得分:1)

是的,你几乎是对的。

一个小的修正是tf.nn.conv3d_transpose期望NCDHWNDHWC输入格式(您的显示为NHWDC),滤镜形状预计为{{1} }。这会影响[depth, height, width, output_channels, in_channels]filter

中的维度顺序
stride

哪个输出:

# Original format: NHWDC.
original = tf.placeholder(dtype=tf.float32, shape=[None, 16, 16, 4, 192])
print original.shape

# Convert to NDHWC format.
input = tf.reshape(original, shape=[-1, 4, 16, 16, 192])
print input.shape

# input shape:  [batch, depth, height, width, in_channels].
# filter shape: [depth, height, width, output_channels, in_channels].
# output shape: [batch, depth, height, width, output_channels].
filter = tf.get_variable('filter', shape=[4, 2, 2, 192, 192], dtype=tf.float32)
conv = tf.nn.conv3d_transpose(input,
                              filter=filter,
                              output_shape=[-1, 7, 32, 32, 192],
                              strides=[1, 1, 2, 2, 1],
                              padding='SAME')
print conv.shape

final = tf.reshape(conv, shape=[-1, 32, 32, 7, 192])
print final.shape