重塑茶叶

Question

我无法清楚地理解theano的reshape。我有一个形状的图像矩阵：

    [batch_size, stack1_size, stack2_size, height, width]

，其中有 stack2_size 叠图像，每叠图像有 stack1_size 个通道。我现在想将它们转换成以下形状：

    [batch_size, stack1_size*stack2_size, 1 , height, width]

这样所有的堆栈将被组合在一起成为所有通道的堆栈。我不确定重塑是否会为我做这件事。我看到如果像素在中间的尺寸中混合，则 reshape 似乎不会按字典顺序排列像素。我一直试图通过 dimshuffle、reshape 和 concatenate 的组合来实现这一点，但无济于事。我将不胜感激。

谢谢。

Answer 1

Theano reshape works just like numpy reshape 及其默认值 order，即 'C':

‘C’ means to read / write the elements using C-like index order, with the last axis index changing fastest, back to the first axis index changing slowest.

这是一个示例，显示图像像素在通过 numpy 或 Theano 重塑后保持相同的顺序。

import numpy
import theano
import theano.tensor


def main():
    batch_size = 2
    stack1_size = 3
    stack2_size = 4
    height = 5
    width = 6
    data = numpy.arange(batch_size * stack1_size * stack2_size * height * width).reshape(
        (batch_size, stack1_size, stack2_size, height, width))
    reshaped_data = data.reshape([batch_size, stack1_size * stack2_size, 1, height, width])
    print data[0, 0, 0]
    print reshaped_data[0, 0, 0]

    x = theano.tensor.TensorType('int64', (False,) * 5)()
    reshaped_x = x.reshape((x.shape[0], x.shape[1] * x.shape[2], 1, x.shape[3], x.shape[4]))
    f = theano.function(inputs=[x], outputs=reshaped_x)
    print f(data)[0, 0, 0]


main()

重塑茶叶

Theano Reshaping

reshape

theano