def sample_from_scores(scores, proportion, exclude_list):
scores = normalized_scores(scores, exclude_list)
np.random.seed() #necessary for mutlithreaded code
n_nonzero = len([score for score in scores if score != 0])
size = min(int(len(scores) * proportion), n_nonzero)
sample = np.random.choice(len(scores), size, p=scores, replace=False) if len(scores) > 0 else []
for s in sample:
assert scores[s] != 0
return sample
def test_normalized_scores(self):
"""Test normalizing scores"""
scores = np.array([0.5, None, 2, 2, -1, -0.5, None])
exclude_list = np.array([2, 1])
norm_scores = normalized_scores(scores, exclude_list)
expected = np.array([0.3, 0.0, 0.0, 0.6, 0.0, 0.1, 0.0])
self.assertEquals(len(norm_scores), len(expected))
for i in range(len(norm_scores)):
self.assertEqual(norm_scores[i], expected[i])
