def test_write_to_location_1(tmpdir):
"""
Test that predicates are written to files in a target location with add_pos syntax.
"""
_db = Database()
_db.add_pos("a(b).")
_db.add_neg("a(c).")
_db.add_fact("d(b,c).")
_db.write(filename="train", location=pathlib.Path(tmpdir))
assert tmpdir.join("train_pos.txt").read() == "a(b).\n"
assert tmpdir.join("train_neg.txt").read() == "a(c).\n"
assert tmpdir.join("train_facts.txt").read() == "d(b,c).\n"
def test_initialize_database_1():
"""
Test initializing a Database with defualt settings.
"""
_db = Database()
assert _db.pos == []
assert _db.neg == []
assert _db.facts == []
def test_initialize_database_2():
"""
Test initializing a Database with a specific target.
"""
_db = Database()
assert _db.pos == []
assert _db.neg == []
assert _db.facts == []
def test_serialize_BoostedRDNRegressor(tmpdir):
"""Test serializing and inference with a regressor object."""
output_json = tmpdir.join("BostonHousingRDN.json")
train = Database.from_files(
pos="datasets/Boston/train/pos.pl",
neg="datasets/Boston/train/neg.pl",
facts="datasets/Boston/train/facts.pl",
)
bkg = Background(modes=[
"crim(+id,#varsrim).",
"zn(+id,#varzn).",
"indus(+id,#varindus).",
"chas(+id,#varchas).",
"nox(+id,#varnox).",
"rm(+id,#varrm).",
"age(+id,#varage).",
"dis(+id,#vardis).",
"rad(+id,#varrad).",
"tax(+id,#vartax).",
"ptratio(+id,#varptrat).",
"b(+id,#varb).",
"lstat(+id,#varlstat).",
"medv(+id).",
])
rdn = BoostedRDNRegressor(background=bkg, target="medv", n_estimators=5)
rdn.fit(train)
rdn.to_json(output_json)
# New BoostedRDN instance, loading from file, and running.
rdn2 = BoostedRDNRegressor()
rdn2.from_json(output_json)
test = Database.from_files(
pos="datasets/Boston/test/pos.pl",
neg="datasets/Boston/test/neg.pl",
facts="datasets/Boston/test/facts.pl",
)
_predictions = rdn2.predict(test)
assert len(_predictions) == 13
def test_initialize_from_files_lazy_paths():
"""Test initializing from pathlib.Path filenames with lazy loading."""
_db = Database.from_files(
pos=pathlib.Path("datasets/ToyFather/train/pos.pl"),
neg=pathlib.Path("datasets/ToyFather/train/neg.pl"),
facts=pathlib.Path("datasets/ToyFather/train/facts.pl"),
lazy_load=True,
)
assert _db.pos == pathlib.Path("datasets/ToyFather/train/pos.pl")
assert _db.neg == pathlib.Path("datasets/ToyFather/train/neg.pl")
assert _db.facts == pathlib.Path("datasets/ToyFather/train/facts.pl")
def test_initialize_from_files_lazy_strings():
"""Test initializing from string filename with lazy loading."""
_db = Database.from_files(
pos="datasets/ToyFather/train/pos.pl",
neg="datasets/ToyFather/train/neg.pl",
facts="datasets/ToyFather/train/facts.pl",
lazy_load=True,
)
assert _db.pos == "datasets/ToyFather/train/pos.pl"
assert _db.neg == "datasets/ToyFather/train/neg.pl"
assert _db.facts == "datasets/ToyFather/train/facts.pl"
def test_initialize_mix():
"""Test initializing from a mix of lazy and lists."""
_pos = "datasets/ToyFather/train/pos.pl"
_neg = "datasets/ToyFather/train/neg.pl"
_facts = pathlib.Path("datasets/ToyFather/train/facts.pl")
_db = Database.from_files(pos=_pos, neg=_neg, facts=_facts, lazy_load=True)
_db.neg = ["father(harrypotter,ronweasley)."]
assert isinstance(_db.pos, str)
assert isinstance(_db.neg, list)
assert isinstance(_db.facts, pathlib.Path)
def test_boston_predict_after_load(test_input):
"""Load a 0.5.2 BostonHousing json file and predict."""
clf = BoostedRDNRegressor()
clf.from_json("srlearn/tests/regression_tests/json/boston_{0}.json".format(
test_input))
test = Database.from_files(
pos="datasets/Boston/test/pos.pl",
neg="datasets/Boston/test/neg.pl",
facts="datasets/Boston/test/facts.pl",
)
_predictions = clf.predict(test)
assert len(_predictions) == 13
def test_lazy_write(tmpdir):
"""Test writing to a location after a lazy load."""
_pos = "datasets/ToyFather/train/pos.pl"
_neg = "datasets/ToyFather/train/neg.pl"
_facts = "datasets/ToyFather/train/facts.pl"
_db = Database.from_files(pos=_pos, neg=_neg, facts=_facts, lazy_load=True)
assert isinstance(_db.pos, str)
assert isinstance(_db.neg, str)
assert isinstance(_db.facts, str)
_db.write(filename="train", location=pathlib.Path(tmpdir))
assert tmpdir.join("train_pos.txt").read() == open(_pos).read()
assert tmpdir.join("train_neg.txt").read() == open(_neg).read()
assert tmpdir.join("train_facts.txt").read() == open(_facts).read()
def test_initialize_from_files_not_lazy():
"""Test initializing from files without lazy loading."""
_pos = "datasets/ToyFather/train/pos.pl"
_neg = "datasets/ToyFather/train/neg.pl"
_facts = "datasets/ToyFather/train/facts.pl"
_db = Database.from_files(pos=_pos,
neg=_neg,
facts=_facts,
lazy_load=False)
assert isinstance(_db.pos, list)
assert isinstance(_db.neg, list)
assert isinstance(_db.facts, list)
assert _db.pos == open(_pos).read().splitlines()
assert _db.neg == open(_neg).read().splitlines()
assert _db.facts == open(_facts).read().splitlines()
def test_mixed_write(tmpdir):
"""Test writing to a location with a mix of lazy, non-lazy, and lists."""
_pos = "datasets/ToyFather/train/pos.pl"
_neg = "datasets/ToyFather/train/neg.pl"
_facts = pathlib.Path("datasets/ToyFather/train/facts.pl")
_db = Database.from_files(pos=_pos, neg=_neg, facts=_facts, lazy_load=True)
_db.neg = ["father(harrypotter,ronweasley)."]
assert isinstance(_db.pos, str)
assert isinstance(_db.neg, list)
assert isinstance(_db.facts, pathlib.Path)
_db.write(filename="test", location=pathlib.Path(tmpdir))
assert tmpdir.join("test_pos.txt").read() == open(_pos).read()
assert tmpdir.join(
"test_neg.txt").read() == "father(harrypotter,ronweasley).\n"
assert tmpdir.join("test_facts.txt").read() == open(_facts).read()