def test_exclude_metafeature_groups(self):
SUBSET_LENGTH = 3
test_failures = {}
test_name = inspect.stack()[0][3]
for dataset_filename, dataset in self.datasets.items():
groups = random.sample(
[group.value for group in consts.MetafeatureGroup],
SUBSET_LENGTH)
computed_mfs = Metafeatures().compute(
X=dataset["X"],
Y=dataset["Y"],
column_types=dataset["column_types"],
seed=CORRECTNESS_SEED,
exclude_groups=groups,
)
known_metafeatures = dataset["known_metafeatures"]
required_checks = [
(self._check_correctness,
[computed_mfs, known_metafeatures, dataset_filename])
]
test_failures.update(self._perform_checks(required_checks))
metafeature_ids = set(
mf_id for group in groups
for mf_id in Metafeatures.list_metafeatures(group))
if any(mf_id in computed_mfs.keys() for mf_id in metafeature_ids):
self.fail('Metafeatures computed an excluded metafeature')
self._report_test_failures(test_failures, test_name)
def test_request_metafeatures(self):
SUBSET_LENGTH = 20
test_failures = {}
test_name = inspect.stack()[0][3]
for dataset_filename, dataset in self.datasets.items():
metafeature_ids = random.sample(Metafeatures.IDS, SUBSET_LENGTH)
computed_mfs = Metafeatures().compute(
X=dataset["X"],
Y=dataset["Y"],
seed=CORRECTNESS_SEED,
metafeature_ids=metafeature_ids,
column_types=dataset["column_types"])
known_metafeatures = dataset["known_metafeatures"]
required_checks = [
(self._check_correctness,
[computed_mfs, known_metafeatures, dataset_filename])
]
test_failures.update(self._perform_checks(required_checks))
self.assertEqual(set(metafeature_ids), set(computed_mfs.keys()),
"Compute did not return requested metafeatures")
self._report_test_failures(test_failures, test_name)
def test_exclude_metafeatures(self):
SUBSET_LENGTH = 20
test_failures = {}
test_name = inspect.stack()[0][3]
for dataset_filename, dataset in self.datasets.items():
metafeature_ids = random.sample(Metafeatures.IDS, SUBSET_LENGTH)
computed_mfs = Metafeatures().compute(
X=dataset["X"],
Y=dataset["Y"],
seed=CORRECTNESS_SEED,
exclude=metafeature_ids,
column_types=dataset["column_types"])
known_metafeatures = dataset["known_metafeatures"]
required_checks = [
(self._check_correctness,
[computed_mfs, known_metafeatures, dataset_filename])
]
test_failures.update(self._perform_checks(required_checks))
if any(mf_id in computed_mfs.keys() for mf_id in metafeature_ids):
self.assertTrue(
False, "Metafeatures computed an excluded metafeature")
self._report_test_failures(test_failures, test_name)
def compute_dataset_metafeatures():
metadata = json.load(open(METADATA_PATH, "r"))
for dataset_metadata in metadata:
dataset_filename = dataset_metadata["filename"]
choice = None
while not choice in ["y", "v", "n"]:
choice = input(dataset_filename + " [(y)es, (v)erbose, (n)o]: ")
if choice == "n":
continue
X, Y, column_types = read_dataset(dataset_metadata)
start_time = time.time()
computed_mfs = Metafeatures().compute(X=X,
Y=Y,
column_types=column_types,
seed=CORRECTNESS_SEED)
run_time = time.time() - start_time
if choice == "v":
known_mf_path = get_dataset_metafeatures_path(dataset_filename)
with open(known_mf_path, 'r') as fp:
known_mfs = json.load(fp)
new_mfs = {}
deleted_mfs = {}
updated_mfs = {}
same_mfs = {}
all_mf_names = set(
list(computed_mfs.keys()) + list(known_mfs.keys()))
for mf in all_mf_names:
if mf not in known_mfs.keys():
new_mfs[mf] = computed_mfs[mf]
elif mf not in computed_mfs.keys():
deleted_mfs[mf] = known_mfs[mf]
elif is_close(computed_mfs[mf]['value'],
known_mfs[mf]['value']):
same_mfs[mf] = computed_mfs[mf]
else:
updated_mfs[mf] = {
'known': known_mfs[mf],
'computed': computed_mfs[mf]
}
print('UNCHANGED METAFEATURES')
print(json.dumps(same_mfs, sort_keys=True, indent=4))
print('DELETED METAFEATURES')
print(json.dumps(deleted_mfs, sort_keys=True, indent=4))
print('NEW METAFEATURES')
print(json.dumps(new_mfs, sort_keys=True, indent=4))
print('UPDATED METAFEATURES')
print(json.dumps(updated_mfs, sort_keys=True, indent=4))
print("Runtime: " + str(run_time))
choice = None
while not choice in ["y", "n"]:
choice = input(
f"Update {dataset_filename} metafeatures? [(y)es, (n)o]: ")
if choice == "y":
mf_file_path = get_dataset_metafeatures_path(dataset_filename)
with open(mf_file_path, 'w') as fp:
json.dump(computed_mfs, fp, sort_keys=True, indent=4)
