def test_select_proportional_to_weight(self):
"""
Tests that multinomial_wo_replacement selects elements, on average,
proportional to the their probabilities
"""
th_rng = RandomStreams(12345)
p = tensor.fmatrix()
n = tensor.iscalar()
m = th_rng.multinomial_wo_replacement(pvals=p, n=n)
f = function([p, n], m, allow_input_downcast=True)
n_elements = 100
n_selected = 10
mean_rtol = 0.0005
numpy.random.seed(12345)
pvals = numpy.random.randint(1, 100, (1, n_elements)).astype(config.floatX)
pvals /= pvals.sum(1)
avg_pvals = numpy.zeros((n_elements,), dtype=config.floatX)
for rep in range(10000):
res = f(pvals, n_selected)
res = numpy.squeeze(res)
avg_pvals[res] += 1
avg_pvals /= avg_pvals.sum()
avg_diff = numpy.mean(abs(avg_pvals - pvals))
assert avg_diff < mean_rtol
def test_select_proportional_to_weight(self):
"""
Tests that multinomial_wo_replacement selects elements, on average,
proportional to the their probabilities
"""
th_rng = RandomStreams(12345)
p = tensor.fmatrix()
n = tensor.iscalar()
m = th_rng.multinomial_wo_replacement(pvals=p, n=n)
f = function([p, n], m, allow_input_downcast=True)
n_elements = 100
n_selected = 10
mean_rtol = 0.0005
numpy.random.seed(12345)
pvals = numpy.random.randint(1, 100,
(1, n_elements)).astype(config.floatX)
pvals /= pvals.sum(1)
avg_pvals = numpy.zeros((n_elements, ), dtype=config.floatX)
for rep in range(10000):
res = f(pvals, n_selected)
res = numpy.squeeze(res)
avg_pvals[res] += 1
avg_pvals /= avg_pvals.sum()
avg_diff = numpy.mean(abs(avg_pvals - pvals))
assert avg_diff < mean_rtol
class RVal(Elem, TensorWrapped, Masked): # random value
def __init__(self, seed=None, **kw):
super(RVal, self).__init__(**kw)
if seed is None:
seed = np.random.randint(0, 1e6)
self.rng = RandomStreams(seed=seed)
self.value = None
def binomial(self, shape, n=1, p=0.5, ndim=None, dtype="int32"):
if isinstance(shape, Elem):
shape = shape.d
self.value = self.rng.binomial(shape, n, p, ndim, dtype)
return self
def normal(self, shape, avg=0.0, std=1.0, ndim=None, dtype=None):
if isinstance(shape, Elem):
shape = shape.d
self.value = self.rng.normal(shape, avg, std, ndim, dtype)
return self
def multinomial(self, shape, n=1, pvals=None, without_replacement=False, ndim=None, dtype="int32"):
if isinstance(shape, Elem):
shape = shape.d
if without_replacement:
self.value = self.rng.multinomial_wo_replacement(shape, n, pvals, ndim, dtype)
else:
self.value = self.rng.multinomial(shape, n, pvals, ndim, dtype)
return self
def gumbel(self, shape, eps=1e-10):
if isinstance(shape, Elem):
shape = shape.d
x = self.rng.uniform(shape, 0.0, 1.0)
self.value = -theano.tensor.log(-theano.tensor.log(x + eps) + eps)
return self
@property
def d(self):
return self.value
@property
def v(self):
return self.value.eval()
@property
def allparams(self):
return set()
@property
def allupdates(self):
return {}
@property
def all_extra_outs(self):
return {}
def test_target_parameter():
srng = MRG_RandomStreams()
pvals = np.array([[.98, .01, .01], [.01, .49, .50]])
def basic_target_parameter_test(x):
f = theano.function([], x)
assert isinstance(f(), np.ndarray)
basic_target_parameter_test(srng.uniform((3, 2), target='cpu'))
basic_target_parameter_test(srng.binomial((3, 2), target='cpu'))
basic_target_parameter_test(srng.multinomial(pvals=pvals.astype('float32'), target='cpu'))
basic_target_parameter_test(srng.choice(p=pvals.astype('float32'), replace=False, target='cpu'))
basic_target_parameter_test(srng.multinomial_wo_replacement(pvals=pvals.astype('float32'), target='cpu'))
def test_target_parameter():
srng = MRG_RandomStreams()
pvals = np.array([[.98, .01, .01], [.01, .49, .50]])
def basic_target_parameter_test(x):
f = theano.function([], x)
assert isinstance(f(), np.ndarray)
basic_target_parameter_test(srng.uniform((3, 2), target='cpu'))
basic_target_parameter_test(srng.binomial((3, 2), target='cpu'))
basic_target_parameter_test(
srng.multinomial(pvals=pvals.astype('float32'), target='cpu'))
basic_target_parameter_test(
srng.choice(p=pvals.astype('float32'), replace=False, target='cpu'))
basic_target_parameter_test(
srng.multinomial_wo_replacement(pvals=pvals.astype('float32'),
target='cpu'))
def test_fail_select_alot(self):
# Tests that multinomial_wo_replacement fails when asked to sample more
# elements than the actual number of elements
th_rng = RandomStreams(12345)
p = tensor.fmatrix()
n = tensor.iscalar()
m = th_rng.multinomial_wo_replacement(pvals=p, n=n)
f = function([p, n], m, allow_input_downcast=True)
n_elements = 100
n_selected = 200
np.random.seed(12345)
pvals = np.random.randint(1, 100, (1, n_elements)).astype(config.floatX)
pvals /= pvals.sum(1)
self.assertRaises(ValueError, f, pvals, n_selected)
def test_fail_select_alot(self):
# Tests that multinomial_wo_replacement fails when asked to sample more
# elements than the actual number of elements
th_rng = RandomStreams(12345)
p = tensor.fmatrix()
n = tensor.iscalar()
m = th_rng.multinomial_wo_replacement(pvals=p, n=n)
f = function([p, n], m, allow_input_downcast=True)
n_elements = 100
n_selected = 200
np.random.seed(12345)
pvals = np.random.randint(1, 100, (1, n_elements)).astype(config.floatX)
pvals /= pvals.sum(1)
self.assertRaises(ValueError, f, pvals, n_selected)
def test_target_parameter():
srng = MRG_RandomStreams()
pvals = np.array([[0.98, 0.01, 0.01], [0.01, 0.49, 0.50]])
def basic_target_parameter_test(x):
f = theano.function([], x)
assert isinstance(f(), np.ndarray)
basic_target_parameter_test(srng.uniform((3, 2), target="cpu"))
basic_target_parameter_test(srng.normal((3, 2), target="cpu"))
basic_target_parameter_test(srng.truncated_normal((3, 2), target="cpu"))
basic_target_parameter_test(srng.binomial((3, 2), target="cpu"))
basic_target_parameter_test(
srng.multinomial(pvals=pvals.astype("float32"), target="cpu"))
basic_target_parameter_test(
srng.choice(p=pvals.astype("float32"), replace=False, target="cpu"))
basic_target_parameter_test(
srng.multinomial_wo_replacement(pvals=pvals.astype("float32"),
target="cpu"))
def test_select_distinct(self):
# Tests that multinomial_wo_replacement always selects distinct elements
th_rng = RandomStreams(12345)
p = tensor.fmatrix()
n = tensor.iscalar()
m = th_rng.multinomial_wo_replacement(pvals=p, n=n)
f = function([p, n], m, allow_input_downcast=True)
n_elements = 1000
all_indices = range(n_elements)
np.random.seed(12345)
for i in [5, 10, 50, 100, 500, n_elements]:
pvals = np.random.randint(1, 100, (1, n_elements)).astype(config.floatX)
pvals /= pvals.sum(1)
res = f(pvals, i)
res = np.squeeze(res)
assert len(res) == i
assert np.all(np.in1d(np.unique(res), all_indices)), res
def test_select_distinct(self):
# Tests that multinomial_wo_replacement always selects distinct elements
th_rng = RandomStreams(12345)
p = tensor.fmatrix()
n = tensor.iscalar()
m = th_rng.multinomial_wo_replacement(pvals=p, n=n)
f = function([p, n], m, allow_input_downcast=True)
n_elements = 1000
all_indices = range(n_elements)
np.random.seed(12345)
for i in [5, 10, 50, 100, 500, n_elements]:
pvals = np.random.randint(1, 100, (1, n_elements)).astype(config.floatX)
pvals /= pvals.sum(1)
res = f(pvals, i)
res = np.squeeze(res)
assert len(res) == i
assert np.all(np.in1d(np.unique(res), all_indices)), res
