TensorFlow:tf.space_to_depth函数通俗解释
函数原型:
tf.space_to_depth( input, block_size, data_format='NHWC', name=None )
tf.space_to_depth将Input Tensor的height、weight维度的值移到Output Tensor的depth维度。
通俗的讲,就是实现如下的功能:
Input Tensor: [batch, height, width, channels]
Output Tensor: [batch, height / block_size, width / block_size, channels * block_size * block_size]
应用示例
x = [[[[1], [2]], [[3], [4]]]] y = tf.space_to_depth(x, 2) # [[[[1, 2, 3, 4]]]]
Input Tensor的shape为[1, 2, 2, 1],Output Tensor的shape为[1, 1, 1, 4]。
x = [[[[1], [2], [5], [6]], [[3], [4], [7], [8]], [[9], [10], [13], [14]], [[11], [12], [15], [16]]]] y = tf.space_to_depth(x, 2) [[ [[1, 2, 3, 4], [5, 6, 7, 8]], [[9, 10, 11, 12], [13, 14, 15, 16]] ]]
Input Tensor的shape为[1, 4, 4, 1],Output Tensor的shape为[1, 2, 2, 4]。
参考材料
https://www.tensorflow.org/api_docs/python/tf/nn/space_to_depth
除非注明,否则均为[半杯茶的小酒杯]原创文章,转载必须以链接形式标明本文链接