def test_singletask_index_k_fold_split(self):
"""
Test singletask IndexSplitter class.
"""
solubility_dataset = self.load_solubility_data()
index_splitter = IndexSplitter()
ids_set = set(solubility_dataset.ids)
K = 5
fold_dirs = [tempfile.mkdtemp() for i in range(K)]
fold_datasets = index_splitter.k_fold_split(solubility_dataset,
fold_dirs)
for fold in range(K):
fold_dataset = fold_datasets[fold]
# Verify lengths is 10/k == 2
assert len(fold_dataset) == 2
# Verify that compounds in this fold are subset of original compounds
fold_ids_set = set(fold_dataset.ids)
assert fold_ids_set.issubset(ids_set)
# Verify that no two folds have overlapping compounds.
for other_fold in range(K):
if fold == other_fold:
continue
other_fold_dataset = fold_datasets[other_fold]
other_fold_ids_set = set(other_fold_dataset.ids)
assert fold_ids_set.isdisjoint(other_fold_ids_set)
merge_dir = tempfile.mkdtemp()
merged_dataset = DiskDataset.merge(merge_dir, fold_datasets)
assert len(merged_dataset) == len(solubility_dataset)
assert sorted(merged_dataset.ids) == (sorted(solubility_dataset.ids))
def test_k_fold_splitter(self):
"""
Test that we can 5 fold index wise over 5 points
"""
ds = NumpyDataset(np.array(range(5)), np.array(range(5)))
index_splitter = IndexSplitter()
K = 5
fold_datasets = index_splitter.k_fold_split(ds, K)
for fold in range(K):
self.assertTrue(fold_datasets[fold][1].X[0] == fold)
train_data = set(list(fold_datasets[fold][0].X))
self.assertFalse(fold in train_data)
self.assertEqual(K - 1, len(train_data))
def test_k_fold_splitter(self):
"""
Test that we can 5 fold index wise over 5 points
"""
ds = NumpyDataset(np.array(range(5)), np.array(range(5)))
index_splitter = IndexSplitter()
K = 5
fold_datasets = index_splitter.k_fold_split(ds, K)
for fold in range(K):
train, cv = fold_datasets[fold][0], fold_datasets[fold][1]
self.assertTrue(cv.X[0] == fold)
train_data = set(list(train.X))
self.assertFalse(fold in train_data)
self.assertEqual(K - 1, len(train))
self.assertEqual(1, len(cv))