如何使用numpy.savez将带有子数组的数组保存到单独的.npy文件中

Question

I have just recently started using numpy and was wondering some things. I have a numpy array that looks like this after splitting it:

[array([1,2,3]),
array([4,5,6])]

I want to use numpy.savez to save the main array into the .npz archive with each subarray in its own .npy file.

I thought using this:

numpy.savez('dataFile', mainArray)

would work but it only creates the archive with a single .npy file called arr_0.npy.

Is there a way to do something like this? and if so is there a way so that I can use any array with any number of subarrays with that method. To get these arrays I am reading from a .bin file that could contain any number of elements that would split into any number of arrays. This is why I'm having a hard time.

Is there a way to add files to an already created .npz file?

Answer 1

After doing more research I came upon the answer to my main question. I found out that you can use the *arg to loop through the list of arrays to add them.

I changed the code to

numpy.savez('test', *[mainArray[x] for x in rang(len(mainArray))])

This gave me the solution i was looking for. Thank you for your help.

Answer 2

If you want to save the subarrays in your main array, then you probably need to use save manually, ie

mainArray = [np.array([1,2,3]), np.array([4,5,6])]
for i in range(len(mainArray)):
    np.save('dataFile_%i'%i, mainArray[i] )

Or you can use savez to save subarrays separately and load them later.

mainArray = [np.array([1,2,3]), np.array([4,5,6])]
np.savez('dataFile', mainArray[0], mainArray[1])

npzfile = np.load('dataFile.npz')
npzfile['arr_0']
npzfile['arr_1']