Most efficient way of reshaping a list-array structure

[midarq]

Hi!

I'm dealing with huge data and I'm looking for an efficient way to reshape my list of arrays. Here is my solution which I find to be slow

1 2 3 4

#x is an input of shape [file,measurement, array(n,r)], i.e. #len(x) returns file #len(x[0]) returns measurement #x[0][0].shape returns (n,r)

the output I require:

1	#new_x is a list of arrays, its shape is [file*measurement*n, array(r)]

I'm using the following code:

1 2 3 4 5

new_x=[] for ii in range(len(x)):     for jj in range(len(x[0])):         for kk in range(len(x[0][0])):             new_x.append(x[ii][jj][kk])

Is there a more efficient way?
Thank you for your help.

[Gribouillis]

What about

1 2 3

for xi in x:     for xij in xi:         new_x.extend(xij)

[midarq]

Much faster, thank you!