1
votes

I have a problem with random.choice which I can't understand. I pass 3 arguments to the function which is allowed to have 4 (http://docs.scipy.org/doc/numpy-1.10.0/reference/generated/numpy.random.choice.html), but it writes I am allowed to give only 2 and 4 were given.

def load_data():
    dataset = load_boston()
    num_samples = size(dataset.data, 0)
    test_set_sz = int(1.0 * num_samples / 10)
    tst_sub_inds = random.choice(range(num_samples), test_set_sz, False)
    data_test, label_test = dataset.data[tst_sub_inds, :], dataset.target[tst_sub_inds]
    trn_sub_inds = list(set(range(num_samples)) - set(tst_sub_inds)) 
    data_train, label_train = dataset.data[trn_sub_inds, :], dataset.target[trn_sub_inds]
    return ((data_train, label_train), (data_test, label_test))

The Error:

tst_sub_inds = random.choice(range(num_samples), test_set_sz, False) TypeError: choice() takes 2 positional arguments but 4 were given Blockquote

What is the problem? Maybe it due to an old version of python?

Thanks, Eli

1
Show us how you do your imports. I suspect that you are calling Python's random library and not Numpy's.DeepSpace
I just made "import random"Eli Borodach

1 Answers

9
votes

As you clarified in the comments, you are using import random which imports Python's random library.

You should use from numpy import random, which will import Numpy's random.choice which is the one that you expect.