Python contained subsampling

Question

I'm trying to subsample a dataset in a contained manner, as in not getting the entirety of the samples while bagging.

Example:

dataset

dataset = array([[ 1.1,  1.2,  1.3,  1.4,  1.5,  1.6,  1.7,  1.8,  1.9],
   [ 2.1,  2.2,  2.3,  2.4,  2.5,  2.6,  2.7,  2.8,  2.9],
   [ 3.1,  3.2,  3.3,  3.4,  3.5,  3.6,  3.7,  3.8,  3.9],
   [ 4.1,  4.2,  4.3,  4.4,  4.5,  4.6,  4.7,  4.8,  4.9],
   [ 5.1,  5.2,  5.3,  5.4,  5.5,  5.6,  5.7,  5.8,  5.9],
   [ 6.1,  6.2,  6.3,  6.4,  6.5,  6.6,  6.7,  6.8,  6.9],
   [ 7.1,  7.2,  7.3,  7.4,  7.5,  7.6,  7.7,  7.8,  7.9],
   [ 8.1,  8.2,  8.3,  8.4,  8.5,  8.6,  8.7,  8.8,  8.9],
   [ 9.1,  9.2,  9.3,  9.4,  9.5,  9.6,  9.7,  9.8,  9.9],
   [10.1, 10.2, 10.3, 10.4, 10.5, 10.6, 10.7, 10.8, 10.9],
   [11.1, 11.2, 11.3, 11.4, 11.5, 11.6, 11.7, 11.8, 11.9],
   [12.1, 12.2, 12.3, 12.4, 12.5, 12.6, 12.7, 12.8, 12.9],
   [13.1, 13.2, 13.3, 13.4, 13.5, 13.6, 13.7, 13.8, 13.9],
   [14.1, 14.2, 14.3, 14.4, 14.5, 14.6, 14.7, 14.8, 14.9],
   [15.1, 15.2, 15.3, 15.4, 15.5, 15.6, 15.7, 15.8, 15.9],
   [16.1, 16.2, 16.3, 16.4, 16.5, 16.6, 16.7, 16.8, 16.9],
   [17.1, 17.2, 17.3, 17.4, 17.5, 17.6, 17.7, 17.8, 17.9]])

Desired subsample:

array([[5.5, 5.6, 5.7],
     [6.5, 6.6, 6.7]])

This can be done by double slicing:

dataset[4:, 4:][:2, :3]

Now, this way to subsample seems everything but optimal (it is fairly slow). I was wandering if there was any better way to do this, maybe using a list comprehension with np.sample or np.take.

EDIT: I'm looking to take multiple subsamples from the dataset, each of them being random.

EDIT 2: Regarding the number of features per subsample, > 2 and <= number of features. Regarding the number of samples it should contain about 60% of the given dataset.

EDIT 3: The shapes of all subsamples should be the same. shape = (X, 0.6*len(dataset)) where X is in range [2, number_of_columns]

Answer 1

If you always sample contiguous rectangles from your data, then indexing using

dataset[4:6, 4:7]

should be "better" (faster) than

dataset[4:, 4:][:2, :3]

since the former avoids creating the intermediate view, and iterates over the 2d-ndarray directly.

This can be confirmed using the ipython %timeit magic:

In [11]: %timeit dataset[4:6, 4:7]                                              
216 ns ± 0.896 ns per loop (mean ± std. dev. of 7 runs, 1000000 loops each)

In [12]: %timeit dataset[4:, 4:][:2, :3]                                        
419 ns ± 11.9 ns per loop (mean ± std. dev. of 7 runs, 1000000 loops each)