def test_default_updates_multiple(self):
x = shared(0)
y = shared(1)
x.default_update = x - 1
y.default_update = y + 1
f1 = pfunc([], [x, y])
f1()
assert x.get_value() == -1
assert y.get_value() == 2
f2 = pfunc([], [x, y], updates=[(x, (x - 2))], no_default_updates=[y])
f2()
assert x.get_value() == -3
assert y.get_value() == 2
f3 = pfunc([], [x, y], updates=[(x, (x - 2))], no_default_updates=True)
f3()
assert x.get_value() == -5
assert y.get_value() == 2
f4 = pfunc([], [x, y], updates=[(y, (y - 2))])
f4()
assert x.get_value() == -6
assert y.get_value() == 0
def test_doc(self):
"""Ensure the code given in pfunc.txt works as expected"""
# Example #1.
a = lscalar()
b = shared(1)
f1 = pfunc([a], (a + b))
f2 = pfunc([In(a, value=44)], a + b, updates={b: b + 1})
self.assertTrue(b.get_value() == 1)
self.assertTrue(f1(3) == 4)
self.assertTrue(f2(3) == 4)
self.assertTrue(b.get_value() == 2)
self.assertTrue(f1(3) == 5)
b.set_value(0)
self.assertTrue(f1(3) == 3)
# Example #2.
a = tensor.lscalar()
b = shared(7)
f1 = pfunc([a], a + b)
f2 = pfunc([a], a * b)
self.assertTrue(f1(5) == 12)
b.set_value(8)
self.assertTrue(f1(5) == 13)
self.assertTrue(f2(4) == 32)
def test_no_default_updates(self):
x = shared(0)
y = shared(1)
x.default_update = x + 2
# Test that the default update is taken into account in the right cases
f1 = pfunc([], [x], no_default_updates=True)
f1()
assert x.get_value() == 0
f2 = pfunc([], [x], no_default_updates=[x])
f2()
assert x.get_value() == 0
f3 = pfunc([], [x], no_default_updates=[x, y])
f3()
assert x.get_value() == 0
f4 = pfunc([], [x], no_default_updates=[y])
f4()
assert x.get_value() == 2
f5 = pfunc([], [x], no_default_updates=[])
f5()
assert x.get_value() == 4
f5 = pfunc([], [x], no_default_updates=False)
f5()
assert x.get_value() == 6
self.assertRaises(TypeError, pfunc, [], [x], no_default_updates=(x))
self.assertRaises(TypeError, pfunc, [], [x], no_default_updates=x)
self.assertRaises(TypeError,
pfunc, [], [x],
no_default_updates='canard')
# Mix explicit updates and no_default_updates
g1 = pfunc([], [x], updates=[(x, (x - 1))], no_default_updates=True)
g1()
assert x.get_value() == 5
g2 = pfunc([], [x], updates=[(x, (x - 1))], no_default_updates=[x])
g2()
assert x.get_value() == 4
g3 = pfunc([], [x], updates=[(x, (x - 1))], no_default_updates=[x, y])
g3()
assert x.get_value() == 3
g4 = pfunc([], [x], updates=[(x, (x - 1))], no_default_updates=[y])
g4()
assert x.get_value() == 2
g5 = pfunc([], [x], updates=[(x, (x - 1))], no_default_updates=[])
g5()
assert x.get_value() == 1
g5 = pfunc([], [x], updates=[(x, (x - 1))], no_default_updates=False)
g5()
assert x.get_value() == 0
def test_doc(self):
# Ensure the code given in pfunc.txt works as expected
# Example #1.
a = lscalar()
b = shared(1)
f1 = pfunc([a], (a + b))
f2 = pfunc([In(a, value=44)], a + b, updates={b: b + 1})
assert b.get_value() == 1
assert f1(3) == 4
assert f2(3) == 4
assert b.get_value() == 2
assert f1(3) == 5
b.set_value(0)
assert f1(3) == 3
# Example #2.
a = tensor.lscalar()
b = shared(7)
f1 = pfunc([a], a + b)
f2 = pfunc([a], a * b)
assert f1(5) == 12
b.set_value(8)
assert f1(5) == 13
assert f2(4) == 32
def test_default_updates_multiple(self):
x = shared(0)
y = shared(1)
x.default_update = x - 1
y.default_update = y + 1
f1 = pfunc([], [x, y])
f1()
assert x.get_value() == -1
assert y.get_value() == 2
f2 = pfunc([], [x, y], updates=[(x, (x - 2))], no_default_updates=[y])
f2()
assert x.get_value() == -3
assert y.get_value() == 2
f3 = pfunc([], [x, y], updates=[(x, (x - 2))], no_default_updates=True)
f3()
assert x.get_value() == -5
assert y.get_value() == 2
f4 = pfunc([], [x, y], updates=[(y, (y - 2))])
f4()
assert x.get_value() == -6
assert y.get_value() == 0
def test_doc(self):
# Ensure the code given in pfunc.txt works as expected
# Example #1.
a = lscalar()
b = shared(1)
f1 = pfunc([a], (a + b))
f2 = pfunc([In(a, value=44)], a + b, updates={b: b + 1})
self.assertTrue(b.get_value() == 1)
self.assertTrue(f1(3) == 4)
self.assertTrue(f2(3) == 4)
self.assertTrue(b.get_value() == 2)
self.assertTrue(f1(3) == 5)
b.set_value(0)
self.assertTrue(f1(3) == 3)
# Example #2.
a = tensor.lscalar()
b = shared(7)
f1 = pfunc([a], a + b)
f2 = pfunc([a], a * b)
self.assertTrue(f1(5) == 12)
b.set_value(8)
self.assertTrue(f1(5) == 13)
self.assertTrue(f2(4) == 32)
def test_no_default_updates(self):
x = shared(0)
y = shared(1)
x.default_update = x + 2
# Test that the default update is taken into account in the right cases
f1 = pfunc([], [x], no_default_updates=True)
f1()
assert x.get_value() == 0
f2 = pfunc([], [x], no_default_updates=[x])
f2()
assert x.get_value() == 0
f3 = pfunc([], [x], no_default_updates=[x, y])
f3()
assert x.get_value() == 0
f4 = pfunc([], [x], no_default_updates=[y])
f4()
assert x.get_value() == 2
f5 = pfunc([], [x], no_default_updates=[])
f5()
assert x.get_value() == 4
f5 = pfunc([], [x], no_default_updates=False)
f5()
assert x.get_value() == 6
self.assertRaises(TypeError, pfunc, [], [x], no_default_updates=(x))
self.assertRaises(TypeError, pfunc, [], [x], no_default_updates=x)
self.assertRaises(TypeError, pfunc, [], [x],
no_default_updates='canard')
# Mix explicit updates and no_default_updates
g1 = pfunc([], [x], updates=[(x, (x - 1))], no_default_updates=True)
g1()
assert x.get_value() == 5
g2 = pfunc([], [x], updates=[(x, (x - 1))], no_default_updates=[x])
g2()
assert x.get_value() == 4
g3 = pfunc([], [x], updates=[(x, (x - 1))], no_default_updates=[x, y])
g3()
assert x.get_value() == 3
g4 = pfunc([], [x], updates=[(x, (x - 1))], no_default_updates=[y])
g4()
assert x.get_value() == 2
g5 = pfunc([], [x], updates=[(x, (x - 1))], no_default_updates=[])
g5()
assert x.get_value() == 1
g5 = pfunc([], [x], updates=[(x, (x - 1))], no_default_updates=False)
g5()
assert x.get_value() == 0
def new(cls, input, n_in, n_out, dtype=None, name=None):
if dtype is None:
dtype = input.dtype
if name is None:
name = cls.__name__
w = shared(np.zeros((n_in, n_out), dtype=dtype), name='%s.w'%name)
b = shared(np.zeros((n_out,), dtype=dtype), name='%s.b'%name)
return cls(input, w, b, params=[w,b])
def test_clone0(self):
x = shared(np.asarray([4, 4, 4]))
y = shared(np.asarray([4, 4, 4]))
z = shared(np.asarray([2, 2, 2]))
up = pfunc(
[], [], updates={x: (x * 5), y: ((x * 5) + y), z: (((x * 5) + y) ** z)}
)
up()
assert np.all(x.get_value() == 20)
assert np.all(y.get_value() == 24)
assert np.all(z.get_value() == (24 ** 2))
def test_clone0(self):
x = shared(numpy.asarray([4, 4, 4]))
y = shared(numpy.asarray([4, 4, 4]))
z = shared(numpy.asarray([2, 2, 2]))
up = pfunc([], [], updates={
x: (x * 5),
y: ((x * 5) + y),
z: (((x * 5) + y) ** z)})
up()
assert numpy.all(x.get_value() == 20)
assert numpy.all(y.get_value() == 24)
assert numpy.all(z.get_value() == (24 ** 2))
def test_shared_mutable(self):
bval = np.arange(5)
b = shared(bval)
b_out = b * 2
# shared vars copy args.
assert b.get_value(borrow=True) is not bval
# so we do this to get at the underlying data
bval = data_of(b)
# by default, shared are not mutable unless doing an explicit update
f = pfunc([], [b_out], mode="FAST_RUN")
assert (f() == np.arange(5) * 2).all()
assert np.all(b.get_value(borrow=True) == np.arange(5))
# using updates, b is now a mutable parameter
f = pfunc([], [b_out], updates=[(b, b_out)], mode="FAST_RUN")
assert (f() == (np.arange(5) * 2)).all()
# because of the update
assert (b.get_value(borrow=True) == (np.arange(5) * 2)).all()
assert (bval == (np.arange(5) * 2)).all() # because of mutable=True
# do not depend on updates being in-place though!
bval = np.arange(5)
b.set_value(bval, borrow=True)
bval = data_of(b)
f = pfunc([], [b_out], updates=[(b, (b_out + 3))], mode="FAST_RUN")
assert (f() == (np.arange(5) * 2)).all()
# because of the update
assert (b.get_value(borrow=True) == ((np.arange(5) * 2) + 3)).all()
# bval got modified to something...
assert not (bval == np.arange(5)).all()
# ... but not to b.value !
assert not (bval == b.get_value(borrow=True)).all()
def test_shared(self):
# CHECK: two functions (f1 and f2) can share w
w = shared(numpy.random.rand(2, 2), 'w')
wval = w.get_value(borrow=False)
x = dmatrix()
out1 = w + x
out2 = w * x
f1 = pfunc([x], [out1])
f2 = pfunc([x], [out2])
xval = numpy.random.rand(2, 2)
assert numpy.all(f1(xval) == xval + wval)
assert numpy.all(f2(xval) == xval * wval)
# CHECK: updating a shared value
f3 = pfunc([x], out1, updates=[(w, (w - 1))])
# f3 changes the value of w
assert numpy.all(f3(xval) == xval + wval)
# this same value is read by f1
assert numpy.all(f1(xval) == xval + (wval - 1))
w.set_value(w.get_value(borrow=True) * 10, borrow=True)
# this same value is read by f1
assert numpy.all(f1(xval) == xval + w.get_value(borrow=True))
def test_shared_mutable(self):
bval = numpy.arange(5)
b = shared(bval)
b_out = b * 2
# shared vars copy args.
assert b.get_value(borrow=True) is not bval
# so we do this to get at the underlying data
bval = data_of(b)
# by default, shared are not mutable unless doing an explicit update
f = pfunc([], [b_out], mode='FAST_RUN')
assert (f() == numpy.arange(5) * 2).all()
assert numpy.all(b.get_value(borrow=True) == numpy.arange(5))
# using updates, b is now a mutable parameter
f = pfunc([], [b_out], updates=[(b, b_out)], mode='FAST_RUN')
assert (f() == (numpy.arange(5) * 2)).all()
# because of the update
assert (b.get_value(borrow=True) == (numpy.arange(5) * 2)).all()
assert (bval == (numpy.arange(5) * 2)).all() # because of mutable=True
# do not depend on updates being in-place though!
bval = numpy.arange(5)
b.set_value(bval, borrow=True)
bval = data_of(b)
f = pfunc([], [b_out], updates=[(b, (b_out + 3))], mode='FAST_RUN')
assert (f() == (numpy.arange(5) * 2)).all()
# because of the update
assert (b.get_value(borrow=True) == ((numpy.arange(5) * 2) + 3)).all()
# bval got modified to something...
assert not (bval == numpy.arange(5)).all()
# ... but not to b.value !
assert not (bval == b.get_value(borrow=True)).all()
def test_shared(self):
# CHECK: two functions (f1 and f2) can share w
w = shared(np.random.rand(2, 2), "w")
wval = w.get_value(borrow=False)
x = dmatrix()
out1 = w + x
out2 = w * x
f1 = pfunc([x], [out1])
f2 = pfunc([x], [out2])
xval = np.random.rand(2, 2)
assert np.all(f1(xval) == xval + wval)
assert np.all(f2(xval) == xval * wval)
# CHECK: updating a shared value
f3 = pfunc([x], out1, updates=[(w, (w - 1))])
# f3 changes the value of w
assert np.all(f3(xval) == xval + wval)
# this same value is read by f1
assert np.all(f1(xval) == xval + (wval - 1))
w.set_value(w.get_value(borrow=True) * 10, borrow=True)
# this same value is read by f1
assert np.all(f1(xval) == xval + w.get_value(borrow=True))
def test_givens_replaces_shared_variable2(self):
a = shared(1., 'a')
a.default_update = a + 3
c = a + 10
f = pfunc([], c, givens={a: (a + 10)})
assert f() == 21
assert f() == 34
def test_default_scalar_container(self):
# Similar in spirit to test_default_container, but updating a scalar
# variable. This is a sanity check for non mutable types.
x = shared(0.0, "x")
f = pfunc([], x)
assert f() == 0
x.set_value(x.get_value(borrow=True) + 1, borrow=True)
assert f() == 1
def test_default_scalar_container(self):
# Similar in spirit to test_default_container, but updating a scalar
# variable. This is a sanity check for non mutable types.
x = shared(0.0, 'x')
f = pfunc([], x)
assert f() == 0
x.set_value(x.get_value(borrow=True) + 1, borrow=True)
assert f() == 1
def test_update_equiv(self):
# Like test_update_same, but the update expression is simplified until
# it is found to be equal to the original variable
a = shared(1.0, "a")
b = shared(np.ones((2, 3)), "b")
# See comment in test_update_same about why we try both
# shared variables.
f = theano.function([], [], updates=[(a, a), (b, (2 * b - b))])
g = theano.function([], [], updates=[(a, (a * 2 - a)), (b, b)])
f()
assert a.get_value(borrow=True).shape == (), a.get_value()
assert b.get_value(borrow=True).shape == (2, 3), b.get_value()
g()
assert a.get_value(borrow=True).shape == (), a.get_value()
assert b.get_value(borrow=True).shape == (2, 3), b.get_value()
def test_update_equiv(self):
# Like test_update_same, but the update expression is simplified until
# it is found to be equal to the original variable
a = shared(1., 'a')
b = shared(numpy.ones((2, 3)), 'b')
# See comment in test_update_same about why we try both
# shared variables.
f = theano.function([], [], updates=[(a, a), (b, (2 * b - b))])
g = theano.function([], [], updates=[(a, (a * 2 - a)), (b, b)])
f()
assert a.get_value(borrow=True).shape == (), a.get_value()
assert b.get_value(borrow=True).shape == (2, 3), b.get_value()
g()
assert a.get_value(borrow=True).shape == (), a.get_value()
assert b.get_value(borrow=True).shape == (2, 3), b.get_value()
def test_givens_replaces_shared_variable2(self):
a = shared(1.0, "a")
a.default_update = a + 3
c = a + 10
f = pfunc([], c, givens={a: (a + 10)})
assert f() == 21
assert f() == 34
def __init__(self,
window_size=HYPERPARAMETERS["WINDOW_SIZE"],
vocab_size=vocabulary.wordmap.len,
embedding_size=HYPERPARAMETERS["EMBEDDING_SIZE"],
hidden_size=HYPERPARAMETERS["HIDDEN_SIZE"],
seed=miscglobals.RANDOMSEED):
"""
Initialize L{Model} parameters.
"""
self.vocab_size = vocab_size
self.window_size = window_size
self.embedding_size = embedding_size
if LBL:
self.hidden_size = hidden_size
self.output_size = self.embedding_size
else:
self.hidden_size = hidden_size
self.output_size = 1
import numpy
import hyperparameters
numpy.random.seed(seed)
self.embeddings = numpy.asarray(
(numpy.random.rand(self.vocab_size,
HYPERPARAMETERS["EMBEDDING_SIZE"]) - 0.5) * 2 *
HYPERPARAMETERS["INITIAL_EMBEDDING_RANGE"],
dtype=floatX)
if HYPERPARAMETERS["NORMALIZE_EMBEDDINGS"]:
self.normalize(range(self.vocab_size))
if LBL:
self.output_weights = shared(
numpy.asarray(random_weights(
self.input_size,
self.output_size,
scale_by=HYPERPARAMETERS["SCALE_INITIAL_WEIGHTS_BY"]),
dtype=floatX))
self.output_biases = shared(
numpy.asarray(numpy.zeros((1, self.output_size)),
dtype=floatX))
self.score_biases = shared(
numpy.asarray(numpy.zeros(self.vocab_size), dtype=floatX))
else:
self.hidden_weights = shared(
numpy.asarray(random_weights(
self.input_size,
self.hidden_size,
scale_by=HYPERPARAMETERS["SCALE_INITIAL_WEIGHTS_BY"]),
dtype=floatX))
self.output_weights = shared(
numpy.asarray(random_weights(
self.hidden_size,
self.output_size,
scale_by=HYPERPARAMETERS["SCALE_INITIAL_WEIGHTS_BY"]),
dtype=floatX))
self.hidden_biases = shared(
numpy.asarray(numpy.zeros((self.hidden_size, )), dtype=floatX))
self.output_biases = shared(
numpy.asarray(numpy.zeros((self.output_size, )), dtype=floatX))
def test_givens_replaces_shared_variable(self):
a = shared(1., 'a')
a.default_update = a + 3.
b = tensor.dscalar('b')
c = a + 10
f = pfunc([b], c, givens={a: b})
assert len(f.maker.fgraph.inputs) == 1
assert len(f.maker.fgraph.outputs) == 1
def test_default_updates_partial_graph(self):
a = shared(0)
a.default_update = a + 1 # Increment a each time it is used
b = 2 * a
# Use only the tip of the graph, a is not used
f = pfunc([b], b)
assert a.get_value() == 0
f(21)
assert a.get_value() == 0
def test_update_err_broadcast(self):
# Test that broadcastable dimensions raise error
data = numpy.random.rand(10, 10).astype('float32')
output_var = shared(name="output", value=data)
# the update_var has type matrix, and the update expression
# is a broadcasted scalar, and that should be allowed.
self.assertRaises(TypeError, theano.function, inputs=[], outputs=[],
updates={output_var: output_var.sum().dimshuffle('x', 'x')})
def test_default_updates_partial_graph(self):
a = shared(0)
a.default_update = a + 1 # Increment a each time it is used
b = 2 * a
# Use only the tip of the graph, a is not used
f = pfunc([b], b)
assert a.get_value() == 0
f(21)
assert a.get_value() == 0
def test_no_shared_as_input(self):
# Test that shared variables cannot be used as function inputs.
w_init = np.random.rand(2, 2)
w = shared(w_init.copy(), "w")
with pytest.raises(
TypeError,
match=r"^Cannot use a shared variable \(w\) as explicit input"
):
pfunc([w], theano.tensor.sum(w * w))
def test_givens_replaces_shared_variable(self):
a = shared(1.0, "a")
a.default_update = a + 3.0
b = tensor.dscalar("b")
c = a + 10
f = pfunc([b], c, givens={a: b})
assert len(f.maker.fgraph.inputs) == 1
assert len(f.maker.fgraph.outputs) == 1
def test_default_updates_expressions(self):
x = shared(0)
y = shared(1)
a = lscalar("a")
z = a * x
x.default_update = x + y
f1 = pfunc([a], z)
f1(12)
assert x.get_value() == 1
f2 = pfunc([a], z, no_default_updates=True)
assert f2(7) == 7
assert x.get_value() == 1
f3 = pfunc([a], z, no_default_updates=[x])
assert f3(9) == 9
assert x.get_value() == 1
def test_default_updates_expressions(self):
x = shared(0)
y = shared(1)
a = lscalar('a')
z = a * x
x.default_update = x + y
f1 = pfunc([a], z)
f1(12)
assert x.get_value() == 1
f2 = pfunc([a], z, no_default_updates=True)
assert f2(7) == 7
assert x.get_value() == 1
f3 = pfunc([a], z, no_default_updates=[x])
assert f3(9) == 9
assert x.get_value() == 1
def test_no_shared_as_input(self):
# Test that shared variables cannot be used as function inputs.
w_init = np.random.rand(2, 2)
w = shared(w_init.copy(), "w")
try:
pfunc([w], theano.tensor.sum(w * w))
assert False
except TypeError as e:
msg = "Cannot use a shared variable (w) as explicit input"
if str(e).find(msg) < 0:
raise
def test_no_shared_as_input(self):
"""Test that shared variables cannot be used as function inputs."""
w_init = numpy.random.rand(2, 2)
w = shared(w_init.copy(), 'w')
try:
pfunc([w], theano.tensor.sum(w * w))
assert False
except TypeError as e:
msg = 'Cannot use a shared variable (w) as explicit input'
if str(e).find(msg) < 0:
raise
def test_update(self):
"""Test update mechanism in different settings."""
# Simple value assignment.
x = shared(0)
assign = pfunc([], [], updates={x: 3})
assign()
self.assertTrue(x.get_value() == 3)
# Basic increment function.
x.set_value(0)
inc = pfunc([], [], updates={x: x + 1})
inc()
self.assertTrue(x.get_value() == 1)
# Increment by a constant value.
x.set_value(-1)
y = shared(2)
inc_by_y = pfunc([], [], updates={x: x + y})
inc_by_y()
self.assertTrue(x.get_value() == 1)
def test_update(self):
# Test update mechanism in different settings.
# Simple value assignment.
x = shared(0)
assign = pfunc([], [], updates={x: 3})
assign()
assert x.get_value() == 3
# Basic increment function.
x.set_value(0)
inc = pfunc([], [], updates={x: x + 1})
inc()
assert x.get_value() == 1
# Increment by a constant value.
x.set_value(-1)
y = shared(2)
inc_by_y = pfunc([], [], updates={x: x + y})
inc_by_y()
assert x.get_value() == 1
def test_default_updates_input(self):
x = shared(0)
y = shared(1)
if theano.configdefaults.python_int_bitwidth() == 32:
a = iscalar("a")
else:
a = lscalar("a")
x.default_update = y
y.default_update = y + a
f1 = pfunc([], x, no_default_updates=True)
f1()
assert x.get_value() == 0
assert y.get_value() == 1
f2 = pfunc([], x, no_default_updates=[x])
f2()
assert x.get_value() == 0
assert y.get_value() == 1
f3 = pfunc([], x, no_default_updates=[y])
f3()
assert x.get_value() == 1
assert y.get_value() == 1
f4 = pfunc([a], x)
f4(2)
assert x.get_value() == 1
assert y.get_value() == 3
f5 = pfunc([], x, updates={y: (y - 1)})
f5()
assert x.get_value() == 3
assert y.get_value() == 2
# a is needed as input if y.default_update is used
with pytest.raises(theano.gof.MissingInputError):
pfunc([], x)
def test_update_err_broadcast(self):
# Test that broadcastable dimensions raise error
data = np.random.rand(10, 10).astype("float32")
output_var = shared(name="output", value=data)
# the update_var has type matrix, and the update expression
# is a broadcasted scalar, and that should be allowed.
with pytest.raises(TypeError):
theano.function(
inputs=[],
outputs=[],
updates={output_var: output_var.sum().dimshuffle("x", "x")},
)
def test_default_updates_chained(self):
x = shared(2)
y = shared(1)
z = shared(-1)
x.default_update = x - y
y.default_update = z
z.default_update = z - 1
f1 = pfunc([], [x])
f1()
assert x.get_value() == 1
assert y.get_value() == -1
assert z.get_value() == -2
f2 = pfunc([], [x, y])
f2()
assert x.get_value() == 2
assert y.get_value() == -2
assert z.get_value() == -3
f3 = pfunc([], [y])
f3()
assert x.get_value() == 2
assert y.get_value() == -3
assert z.get_value() == -4
f4 = pfunc([], [x, y], no_default_updates=[x])
f4()
assert x.get_value() == 2
assert y.get_value() == -4
assert z.get_value() == -5
f5 = pfunc([], [x, y, z], no_default_updates=[z])
f5()
assert x.get_value() == 6
assert y.get_value() == -5
assert z.get_value() == -5
def test_default_updates_input(self):
x = shared(0)
y = shared(1)
if theano.configdefaults.python_int_bitwidth() == 32:
a = iscalar('a')
else:
a = lscalar('a')
x.default_update = y
y.default_update = y + a
f1 = pfunc([], x, no_default_updates=True)
f1()
assert x.get_value() == 0
assert y.get_value() == 1
f2 = pfunc([], x, no_default_updates=[x])
f2()
assert x.get_value() == 0
assert y.get_value() == 1
f3 = pfunc([], x, no_default_updates=[y])
f3()
assert x.get_value() == 1
assert y.get_value() == 1
f4 = pfunc([a], x)
f4(2)
assert x.get_value() == 1
assert y.get_value() == 3
f5 = pfunc([], x, updates={y: (y - 1)})
f5()
assert x.get_value() == 3
assert y.get_value() == 2
# a is needed as input if y.default_update is used
self.assertRaises(theano.gof.MissingInputError, pfunc, [], x)
def test_default_updates_chained(self):
x = shared(2)
y = shared(1)
z = shared(-1)
x.default_update = x - y
y.default_update = z
z.default_update = z - 1
f1 = pfunc([], [x])
f1()
assert x.get_value() == 1
assert y.get_value() == -1
assert z.get_value() == -2
f2 = pfunc([], [x, y])
f2()
assert x.get_value() == 2
assert y.get_value() == -2
assert z.get_value() == -3
f3 = pfunc([], [y])
f3()
assert x.get_value() == 2
assert y.get_value() == -3
assert z.get_value() == -4
f4 = pfunc([], [x, y], no_default_updates=[x])
f4()
assert x.get_value() == 2
assert y.get_value() == -4
assert z.get_value() == -5
f5 = pfunc([], [x, y, z], no_default_updates=[z])
f5()
assert x.get_value() == 6
assert y.get_value() == -5
assert z.get_value() == -5
def test_default_container(self):
# Ensure it is possible to (implicitly) use a shared variable in a
# function, as a 'state' that can be updated at will.
rng = numpy.random.RandomState(1827)
w_init = rng.rand(5)
w = shared(w_init.copy(), 'w')
reg = theano.tensor.sum(w * w)
f = pfunc([], reg)
assert f() == numpy.sum(w_init * w_init)
# Change the value of w and ensure the output changes accordingly.
w.set_value(w.get_value(borrow=True) + 1.0, borrow=True)
assert f() == numpy.sum((w_init + 1) ** 2)
def test_default_container(self):
# Ensure it is possible to (implicitly) use a shared variable in a
# function, as a 'state' that can be updated at will.
rng = np.random.RandomState(1827)
w_init = rng.rand(5)
w = shared(w_init.copy(), "w")
reg = theano.tensor.sum(w * w)
f = pfunc([], reg)
assert f() == np.sum(w_init * w_init)
# Change the value of w and ensure the output changes accordingly.
w.set_value(w.get_value(borrow=True) + 1.0, borrow=True)
assert f() == np.sum((w_init + 1) ** 2)
def test_default_updates(self):
x = shared(0)
x.default_update = x + 1
f = pfunc([], [x])
f()
assert x.get_value() == 1
del x.default_update
f()
assert x.get_value() == 2
g = pfunc([], [x])
g()
assert x.get_value() == 2
def test_default_updates(self):
x = shared(0)
x.default_update = x + 1
f = pfunc([], [x])
f()
assert x.get_value() == 1
del x.default_update
f()
assert x.get_value() == 2
g = pfunc([], [x])
g()
assert x.get_value() == 2
def test_update_same(self):
# There was a bug in CVM, triggered when a shared variable
# was its own update expression.
a = shared(1., 'a')
b = shared(numpy.ones((2, 3)), 'b')
# The order of the variables is not determined, so we try
# both shared variables.
# TODO: explain the above comment. By "not determined" does
# this mean "not deterministic"?
# This test originally wrote the updates using dictionaries,
# and iterating over the dictionary was not deterministic.
# Is that all the comment above meant, or is the CVM intended
# to add extra non-determinism? Or is the CVM meant to
# deterministically but arbitrarily pick an order for the updates?
f = theano.function([], [], updates=[(a, a), (b, (2 * b))])
g = theano.function([], [], updates=[(a, (a * 2)), (b, b)])
f()
assert a.get_value(borrow=True).shape == (), a.get_value()
assert b.get_value(borrow=True).shape == (2, 3), b.get_value()
g()
assert a.get_value(borrow=True).shape == (), a.get_value()
assert b.get_value(borrow=True).shape == (2, 3), b.get_value()
def test_update_same(self):
# There was a bug in CVM, triggered when a shared variable
# was its own update expression.
a = shared(1.0, "a")
b = shared(np.ones((2, 3)), "b")
# The order of the variables is not determined, so we try
# both shared variables.
# TODO: explain the above comment. By "not determined" does
# this mean "not deterministic"?
# This test originally wrote the updates using dictionaries,
# and iterating over the dictionary was not deterministic.
# Is that all the comment above meant, or is the CVM intended
# to add extra non-determinism? Or is the CVM meant to
# deterministically but arbitrarily pick an order for the updates?
f = theano.function([], [], updates=[(a, a), (b, (2 * b))])
g = theano.function([], [], updates=[(a, (a * 2)), (b, b)])
f()
assert a.get_value(borrow=True).shape == (), a.get_value()
assert b.get_value(borrow=True).shape == (2, 3), b.get_value()
g()
assert a.get_value(borrow=True).shape == (), a.get_value()
assert b.get_value(borrow=True).shape == (2, 3), b.get_value()
def test_givens(self):
x = shared(0)
assign = pfunc([], x, givens={x: 3})
assert assign() == 3
assert x.get_value(borrow=True) == 0
y = tensor.ivector()
f = pfunc([y], (y * x), givens={x: 6})
assert np.all(f([1, 1, 1]) == [6, 6, 6])
assert x.get_value() == 0
z = tensor.ivector()
c = z * y
f = pfunc([y], (c + 7), givens={z: theano._asarray([4, 4, 4], dtype="int32")})
assert np.all(f([1, 1, 1]) == [11, 11, 11])
assert x.get_value() == 0
def __init__(self, window_size, vocab_size, embedding_size, hidden_size, seed, initial_embeddings, two_hidden_layers):
"""
Initialize L{Model} parameters.
"""
self.vocab_size = vocab_size
self.window_size = window_size
self.embedding_size = embedding_size
self.two_hidden_layers = two_hidden_layers
if LBL:
self.hidden_size = hidden_size
self.output_size = self.embedding_size
else:
self.hidden_size = hidden_size
self.output_size = 1
import numpy
import hyperparameters
numpy.random.seed(seed)
if initial_embeddings is None:
self.embeddings = numpy.asarray((numpy.random.rand(self.vocab_size, HYPERPARAMETERS["EMBEDDING_SIZE"]) - 0.5)*2 * HYPERPARAMETERS["INITIAL_EMBEDDING_RANGE"], dtype=floatX)
else:
assert initial_embeddings.shape == (self.vocab_size, HYPERPARAMETERS["EMBEDDING_SIZE"])
self.embeddings = copy.copy(initial_embeddings)
if HYPERPARAMETERS["NORMALIZE_EMBEDDINGS"]: self.normalize(range(self.vocab_size))
if LBL:
self.output_weights = shared(numpy.asarray(random_weights(self.input_size, self.output_size, scale_by=HYPERPARAMETERS["SCALE_INITIAL_WEIGHTS_BY"]), dtype=floatX))
self.output_biases = shared(numpy.asarray(numpy.zeros((1, self.output_size)), dtype=floatX))
self.score_biases = shared(numpy.asarray(numpy.zeros(self.vocab_size), dtype=floatX))
assert not self.two_hidden_layers
else:
self.hidden_weights = shared(numpy.asarray(random_weights(self.input_size, self.hidden_size, scale_by=HYPERPARAMETERS["SCALE_INITIAL_WEIGHTS_BY"]), dtype=floatX))
self.hidden_biases = shared(numpy.asarray(numpy.zeros((self.hidden_size,)), dtype=floatX))
if self.two_hidden_layers:
self.hidden2_weights = shared(numpy.asarray(random_weights(self.hidden_size, self.hidden_size, scale_by=HYPERPARAMETERS["SCALE_INITIAL_WEIGHTS_BY"]), dtype=floatX))
self.hidden2_biases = shared(numpy.asarray(numpy.zeros((self.hidden_size,)), dtype=floatX))
self.output_weights = shared(numpy.asarray(random_weights(self.hidden_size, self.output_size, scale_by=HYPERPARAMETERS["SCALE_INITIAL_WEIGHTS_BY"]), dtype=floatX))
self.output_biases = shared(numpy.asarray(numpy.zeros((self.output_size,)), dtype=floatX))
def test_givens(self):
x = shared(0)
assign = pfunc([], x, givens={x: 3})
assert assign() == 3
assert x.get_value(borrow=True) == 0
y = tensor.ivector()
f = pfunc([y], (y * x), givens={x: 6})
assert numpy.all(f([1, 1, 1]) == [6, 6, 6])
assert x.get_value() == 0
z = tensor.ivector()
c = z * y
f = pfunc([y], (c + 7),
givens={z: theano._asarray([4, 4, 4], dtype='int32')})
assert numpy.all(f([1, 1, 1]) == [11, 11, 11])
assert x.get_value() == 0
def test_param_strict(self):
a = tensor.dvector()
b = shared(7)
out = a + b
f = pfunc([In(a, strict=False)], [out])
# works, rand generates float64 by default
f(numpy.random.rand(8))
# works, casting is allowed
f(numpy.array([1, 2, 3, 4], dtype='int32'))
f = pfunc([In(a, strict=True)], [out])
try:
# fails, f expects float64
f(numpy.array([1, 2, 3, 4], dtype='int32'))
except TypeError:
pass
def test_param_strict(self):
a = tensor.dvector()
b = shared(7)
out = a + b
f = pfunc([In(a, strict=False)], [out])
# works, rand generates float64 by default
f(np.random.rand(8))
# works, casting is allowed
f(np.array([1, 2, 3, 4], dtype="int32"))
f = pfunc([In(a, strict=True)], [out])
try:
# fails, f expects float64
f(np.array([1, 2, 3, 4], dtype="int32"))
except TypeError:
pass
def _shared_uniform(rng, low, high, size, dtype, name=None):
return shared(theano._asarray(rng.uniform(low=low, high=high, size=size), dtype=dtype), name)
def _shared_uniform(rng, low, high, size, dtype, name=None):
return shared(
theano._asarray(rng.uniform(low=low, high=high, size=size),
dtype=dtype), name)
def __init__(self, window_size, vocab_size, embedding_size, hidden_size,
seed, initial_embeddings, two_hidden_layers):
"""
Initialize L{Model} parameters.
"""
self.vocab_size = vocab_size
self.window_size = window_size
self.embedding_size = embedding_size
self.two_hidden_layers = two_hidden_layers
if LBL:
self.hidden_size = hidden_size
self.output_size = self.embedding_size
else:
self.hidden_size = hidden_size
self.output_size = 1
import numpy
import hyperparameters
numpy.random.seed(seed)
if initial_embeddings is None:
self.embeddings = numpy.asarray(
(numpy.random.rand(self.vocab_size,
HYPERPARAMETERS["EMBEDDING_SIZE"]) - 0.5) *
2 * HYPERPARAMETERS["INITIAL_EMBEDDING_RANGE"],
dtype=floatX)
else:
assert initial_embeddings.shape == (
self.vocab_size, HYPERPARAMETERS["EMBEDDING_SIZE"])
self.embeddings = copy.copy(initial_embeddings)
if HYPERPARAMETERS["NORMALIZE_EMBEDDINGS"]:
self.normalize(range(self.vocab_size))
if LBL:
self.output_weights = shared(
numpy.asarray(random_weights(
self.input_size,
self.output_size,
scale_by=HYPERPARAMETERS["SCALE_INITIAL_WEIGHTS_BY"]),
dtype=floatX))
self.output_biases = shared(
numpy.asarray(numpy.zeros((1, self.output_size)),
dtype=floatX))
self.score_biases = shared(
numpy.asarray(numpy.zeros(self.vocab_size), dtype=floatX))
assert not self.two_hidden_layers
else:
self.hidden_weights = shared(
numpy.asarray(random_weights(
self.input_size,
self.hidden_size,
scale_by=HYPERPARAMETERS["SCALE_INITIAL_WEIGHTS_BY"]),
dtype=floatX))
self.hidden_biases = shared(
numpy.asarray(numpy.zeros((self.hidden_size, )), dtype=floatX))
if self.two_hidden_layers:
self.hidden2_weights = shared(
numpy.asarray(random_weights(
self.hidden_size,
self.hidden_size,
scale_by=HYPERPARAMETERS["SCALE_INITIAL_WEIGHTS_BY"]),
dtype=floatX))
self.hidden2_biases = shared(
numpy.asarray(numpy.zeros((self.hidden_size, )),
dtype=floatX))
self.output_weights = shared(
numpy.asarray(random_weights(
self.hidden_size,
self.output_size,
scale_by=HYPERPARAMETERS["SCALE_INITIAL_WEIGHTS_BY"]),
dtype=floatX))
self.output_biases = shared(
numpy.asarray(numpy.zeros((self.output_size, )), dtype=floatX))
def test_duplicate_updates(self):
x, y = dmatrices('x', 'y')
z = shared(numpy.ones((2, 3)))
self.assertRaises(ValueError, theano.function, [x, y], [z],
updates=[(z, (z + x + y)), (z, (z - x))])
def test_duplicate_updates(self):
x, y = dmatrices("x", "y")
z = shared(np.ones((2, 3)))
with pytest.raises(ValueError):
theano.function([x, y], [z], updates=[(z, (z + x + y)), (z, (z - x))])
def __init__(
self,
window_size=HYPERPARAMETERS["WINDOW_SIZE"],
vocab_size=vocabulary.wordmap.len,
embedding_size=HYPERPARAMETERS["EMBEDDING_SIZE"],
hidden_size=HYPERPARAMETERS["HIDDEN_SIZE"],
seed=miscglobals.RANDOMSEED,
):
"""
Initialize L{Model} parameters.
"""
self.vocab_size = vocab_size
self.window_size = window_size
self.embedding_size = embedding_size
if LBL:
self.hidden_size = hidden_size
self.output_size = self.embedding_size
else:
self.hidden_size = hidden_size
self.output_size = 1
import numpy
import hyperparameters
numpy.random.seed(seed)
self.embeddings = numpy.asarray(
(numpy.random.rand(self.vocab_size, HYPERPARAMETERS["EMBEDDING_SIZE"]) - 0.5)
* 2
* HYPERPARAMETERS["INITIAL_EMBEDDING_RANGE"],
dtype=floatX,
)
if HYPERPARAMETERS["NORMALIZE_EMBEDDINGS"]:
self.normalize(range(self.vocab_size))
if LBL:
self.output_weights = shared(
numpy.asarray(
random_weights(
self.input_size, self.output_size, scale_by=HYPERPARAMETERS["SCALE_INITIAL_WEIGHTS_BY"]
),
dtype=floatX,
)
)
self.output_biases = shared(numpy.asarray(numpy.zeros((1, self.output_size)), dtype=floatX))
self.score_biases = shared(numpy.asarray(numpy.zeros(self.vocab_size), dtype=floatX))
else:
self.hidden_weights = shared(
numpy.asarray(
random_weights(
self.input_size, self.hidden_size, scale_by=HYPERPARAMETERS["SCALE_INITIAL_WEIGHTS_BY"]
),
dtype=floatX,
)
)
self.output_weights = shared(
numpy.asarray(
random_weights(
self.hidden_size, self.output_size, scale_by=HYPERPARAMETERS["SCALE_INITIAL_WEIGHTS_BY"]
),
dtype=floatX,
)
)
self.hidden_biases = shared(numpy.asarray(numpy.zeros((self.hidden_size,)), dtype=floatX))
self.output_biases = shared(numpy.asarray(numpy.zeros((self.output_size,)), dtype=floatX))