Is there an idiom or API for synchronized shuffling of Python arrays? -
is there api in numpy (or perhaps tensorflow) performing synchronized shuffling of several arrays (with same first dimension)?
for example, if 2 arrays dimensions (n, a) , (n, b), , want randomize ordering of n elements of each, while maintaining association between elements of first array , second.
is there api or python idiom accomplishing this?
note combining these single array of n tuples shuffled random.shuffle
might option i'd accept answer, can't work: getting original arrays (as near i've managed) messy since combined_array[:,0]
have dimension (n,) objects elements, rather dimension (n, a), unless manually rebuilt like [x x in combined_array[:,0]
permutation = numpy.random.permutation(n) arr1_shuffled = arr1[permutation] arr2_shuffled = arr2[permutation]
pick 1 permutation , use both arrays.
Comments
Post a Comment