def test_train_exact(self):
test_cases = [(self.m11, self.m21, self.ph1, 5 / (3.0)),
(self.m12, self.m22, self.ph2, 5 / (3.0)),
(self.m13, self.m23, self.ph3, 6),
(self.m14, self.m24, self.ph4, 6),
(self.m15, self.m25, self.ph5, 5 / (3.0)),
(self.m16, self.m26, self.ph6, 2),
(self.m17, self.m27, self.ph7, 2)]
for arg1, arg2, phrase, lambda_ in test_cases:
m = Dilation()
m._solve(arg1, arg2, phrase)
self.assertAlmostEqual(m._lambda, lambda_)
def test_train_exact(self):
test_cases = [(self.m11, self.m21, self.ph1, 5 / (3.0)),
(self.m12, self.m22, self.ph2, 5 / (3.0)),
(self.m13, self.m23, self.ph3, 6),
(self.m14, self.m24, self.ph4, 6),
(self.m15, self.m25, self.ph5, 5 / (3.0)),
(self.m16, self.m26, self.ph6, 2),
(self.m17, self.m27, self.ph7, 2)
]
for arg1, arg2, phrase, lambda_ in test_cases:
m = Dilation()
m._train(arg1, arg2, phrase)
self.assertAlmostEqual(m._lambda, lambda_)
def test_train_random(self):
test_cases = [1.0,2.0,3.0]
rows = 4
cols = 3
m1 = np.random.rand(rows,cols)
m2 = np.random.rand(rows,cols)
for lambda_ in test_cases:
m = Dilation(lambda_)
result_p = m._compose(DenseMatrix(m1), DenseMatrix(m2))
m = Dilation()
m._train(DenseMatrix(m1),DenseMatrix(m2),result_p)
self.assertAlmostEqual(lambda_, m._lambda)
def test_dilation(self):
self.m12 = DenseMatrix(np.mat([[3, 1], [9, 2]]))
self.m22 = DenseMatrix(np.mat([[4, 3], [2, 1]]))
self.ph2 = DenseMatrix(np.mat([[18, 11], [24, 7]]))
self.row = ["a", "b"]
self.ft = ["f1", "f2"]
self.space1 = Space(DenseMatrix(self.m12), self.row, self.ft)
self.space2 = Space(DenseMatrix(self.ph2), ["a_a", "a_b"], self.ft)
m = Dilation()
m.export(self.prefix + ".dil1")
m.train([("a", "b", "a_b")], self.space1, self.space2)
m.export(self.prefix + ".dil2")
def test_compose_exact(self):
test_cases = [(self.m11, self.m21, self.ph1, 5 / (3.0)),
(self.m13, self.m23, self.ph3, 6),
(self.m14, self.m24, self.ph4, 6)]
for arg1, arg2, phrase, lambda_ in test_cases:
m = Dilation()
m._solve(arg1, arg2, phrase)
res = m._compose(arg1, arg2)
np.testing.assert_array_almost_equal(res.mat, phrase.mat, 2)
m = Dilation(lambda_)
res = m._compose(arg1, arg2)
np.testing.assert_array_almost_equal(res.mat, phrase.mat, 2)
def test_dilation(self):
self.m12 = DenseMatrix(np.mat([[3,1],[9,2]]))
self.m22 = DenseMatrix(np.mat([[4,3],[2,1]]))
self.ph2 = DenseMatrix(np.mat([[18,11],[24,7]]))
self.row = ["a", "b"]
self.ft = ["f1","f2"]
self.space1 = Space(DenseMatrix(self.m12), self.row, self.ft)
self.space2 = Space(DenseMatrix(self.ph2), ["a_a","a_b"], self.ft)
m = Dilation()
m.export(self.prefix + ".dil1")
m.train([("a","b","a_b")], self.space1, self.space2)
m.export(self.prefix + ".dil2")
def test_train_random(self):
test_cases = [1.0, 2.0, 3.0]
rows = 4
cols = 3
m1 = np.random.rand(rows, cols)
m2 = np.random.rand(rows, cols)
for lambda_ in test_cases:
m = Dilation(lambda_)
result_p = m._compose(DenseMatrix(m1), DenseMatrix(m2))
m = Dilation()
m._solve(DenseMatrix(m1), DenseMatrix(m2), result_p)
self.assertAlmostEqual(lambda_, m._lambda)
def create_model(model, alpha, beta, lambda_):
#TODO: IMPORTANT here: if alpha, beta of lambda are none
model_dict = {
"weighted_add": WeightedAdditive,
"dilation": Dilation,
"mult": Multiplicative
}
if not model in model_dict:
raise ValueError("Invalid model:%s" % model)
if model == "weighted_add":
model_obj = WeightedAdditive(alpha, beta)
elif model == "dilation":
model_obj = Dilation(lambda_)
else:
model_obj = Multiplicative()
return model_obj
def test_compose_exact(self):
test_cases = [(self.m11, self.m21, self.ph1, 5 / (3.0)),
(self.m13, self.m23, self.ph3, 6),
(self.m14, self.m24, self.ph4, 6)
]
for arg1, arg2, phrase, lambda_ in test_cases:
m = Dilation()
m._train(arg1, arg2, phrase)
res = m._compose(arg1, arg2)
np.testing.assert_array_almost_equal(res.mat, phrase.mat, 2)
m = Dilation(lambda_)
res = m._compose(arg1, arg2)
np.testing.assert_array_almost_equal(res.mat, phrase.mat, 2)
comp_model = FullAdditive(learner=RidgeRegressionLearner(param=2)) comp_model.train(train_data, space, per_space) composed_space = comp_model.compose(test_phrases, space) pred = composed_space.get_sims(test_pairs, CosSimilarity()) print scoring_utils.score(gold, pred, "spearman") print "Training Weighted Additive composition model..." comp_model = WeightedAdditive() comp_model.train(train_data, space, per_space) print "alpha, beta:", comp_model.alpha, comp_model.beta composed_space = comp_model.compose(test_phrases, space) pred = composed_space.get_sims(test_pairs, CosSimilarity()) print scoring_utils.score(gold, pred, "spearman") print "Training Dilation composition model..." comp_model = Dilation() comp_model.train(train_data, space, per_space) print "lambda:", comp_model._lambda composed_space = comp_model.compose(test_phrases, space) pred = composed_space.get_sims(test_pairs, CosSimilarity()) print scoring_utils.score(gold, pred, "spearman") print "Multiplicative composition model..." comp_model = Multiplicative() composed_space = comp_model.compose(test_phrases, space) pred = composed_space.get_sims(test_pairs, CosSimilarity()) print scoring_utils.score(gold, pred, "spearman") print "Simple additive composition model..." comp_model = WeightedAdditive(1, 1) composed_space = comp_model.compose(test_phrases, space)
#-------
from composes.utils import io_utils
from composes.composition.dilation import Dilation
#training data
train_data = [("good", "car", "good_car"),
("good", "book", "good_book")
]
#load an argument space
arg_space = io_utils.load("./data/out/ex10.pkl")
#load a phrase space
phrase_space = io_utils.load("data/out/PHRASE_SS.ex10.pkl")
print "Training phrase space"
print phrase_space.id2row
print phrase_space.cooccurrence_matrix
#train a Dilation model on the data
my_comp = Dilation()
my_comp.train(train_data, arg_space, phrase_space)
#print its parameters
print "\nlambda:", my_comp._lambda
#use the model to compose the train data
composed_space = my_comp.compose([("good", "bike", "good_bike")],
arg_space)
print "\nComposed space:"
print composed_space.id2row
print composed_space.cooccurrence_matrix
#ex14.py
#-------
from composes.utils import io_utils
from composes.composition.dilation import Dilation
#training data
train_data = [("good", "car", "good_car"), ("good", "book", "good_book")]
#load an argument space
arg_space = io_utils.load("./data/out/ex10.pkl")
#load a phrase space
phrase_space = io_utils.load("data/out/PHRASE_SS.ex10.pkl")
print "Training phrase space"
print phrase_space.id2row
print phrase_space.cooccurrence_matrix
#train a Dilation model on the data
my_comp = Dilation()
my_comp.train(train_data, arg_space, phrase_space)
#print its parameters
print "\nlambda:", my_comp._lambda
#use the model to compose the train data
composed_space = my_comp.compose([("good", "bike", "good_bike")], arg_space)
print "\nComposed space:"
print composed_space.id2row
print composed_space.cooccurrence_matrix
