def test_verify_grad_with_zeros(self):
# including zeros, as the case with zeros is important
# (and special cases: 1 zero in the row, more than 1 zero in the row)
x_val = numpy.asarray([[1., 2., 3.], [0., 5., 6.], [0., 0., 9.]],
dtype='float32')
x = theano.tensor.dmatrix()
# sanity check
x2 = theano.tensor.dmatrix()
p = Prod(axis=1)(x)
p2 = Prod(axis=1)(x2)
fn = theano.function([x, x2], [p - p2], mode=self.mode)
#print "hand computed diff for each row"
x2_val = numpy.asarray([[1., 2., 3.003], [0.003, 5., 6],
[0., 0., 9.01]])
#print fn(x_val, x2_val)
fn2 = theano.function([x], [theano.tensor.grad(p.sum(), x)],
mode=self.mode)
#print "real grad"
#print fn2(x_val)
fn3 = theano.function([x], [p], mode=self.mode)
assert numpy.allclose(fn3(x_val), [6., 0., 0.])
# now with verify_grad
unittest_tools.verify_grad(Prod(axis=1), [x_val], mode=self.mode)
def test_other_grad_tests(self):
x = theano.tensor.dmatrix()
x_val1 = numpy.array([[1, 2, 3], [0, 5, 6], [0, 0, 9]],
dtype='float32')
x_val2 = numpy.array([[1, 2, 0], [0, 5, 6], [7, 8, 9], [9, 10, 0]],
dtype='float32')
rng = rng = numpy.random.RandomState(43)
p = Prod(axis=1)
grad_p = theano.tensor.grad(p(x).sum(), x)
grad_fn = theano.function([x], grad_p, mode=self.mode)
assert numpy.allclose(grad_fn(x_val1),
[[6., 3., 2.], [30., 0., 0.], [0., 0., 0.]])
assert numpy.allclose(
grad_fn(x_val2),
[[0., 0., 2.], [30., 0., 0.], [72., 63., 56.], [0., 0., 90.]])
p_axis0 = Prod(axis=0)
grad_p_axis0 = theano.tensor.grad(p_axis0(x).sum(), x)
grad_fn_axis0 = theano.function([x], grad_p_axis0, mode=self.mode)
assert numpy.allclose(
grad_fn_axis0(x_val2),
[[0., 400., 0.], [63., 160., 0.], [0., 100., 0.], [0., 80., 0.]])
tensor.verify_grad(p, [x_val1], rng=rng, mode=self.mode)
def test_verify_grad_with_zeros(self):
# including zeros, as the case with zeros is important
# (and special cases: 1 zero in the row, more than 1 zero in the row)
x_val = numpy.asarray([[1., 2., 3.], [0., 5., 6.], [0., 0., 9.]],
dtype='float32')
x = theano.tensor.dmatrix()
# sanity check
x2 = theano.tensor.dmatrix()
p = Prod(axis=1)(x)
p2 = Prod(axis=1)(x2)
fn = theano.function([x, x2], [p - p2], mode=self.mode)
#print "hand computed diff for each row"
x2_val = numpy.asarray([[1., 2., 3.003], [0.003, 5., 6], [
0., 0., 9.01]])
#print fn(x_val, x2_val)
fn2 = theano.function([x], [theano.tensor.grad(p.sum(), x)],
mode=self.mode)
#print "real grad"
#print fn2(x_val)
fn3 = theano.function([x], [p], mode=self.mode)
assert numpy.allclose(fn3(x_val), [6., 0., 0.])
# now with verify_grad
unittest_tools.verify_grad(Prod(axis=1), [x_val], mode=self.mode)
def test_prod_no_zeros_in_input(self):
x = theano.tensor.dmatrix()
x_val = numpy.array([[1, 2, 3], [4, 5, 6], [7, 8, 9]], dtype='float32')
pwz = Prod(axis=1, no_zeros_in_input=True)(x)
fn = theano.function([x], pwz, mode=self.mode)
assert numpy.allclose(fn(x_val), [6, 120, 504])
pwz = Prod(no_zeros_in_input=True)(x)
g = theano.grad(pwz, x)
gg = theano.grad(g.sum(), x)
fn = theano.function([x], g, mode=self.mode)
assert numpy.allclose(
fn(x_val), [[362880., 181440., 120960.], [90720., 72576., 60480.],
[51840., 45360., 40320.]])
fn = theano.function([x], gg, mode=self.mode)
assert numpy.allclose(
fn(x_val),
[[663696., 422568., 301872.], [233964., 190800., 161016.],
[139248., 122652., 109584.]])
unittest_tools.verify_grad(Prod(axis=1, no_zeros_in_input=True),
[x_val],
mode=self.mode)
unittest_tools.verify_grad(Prod(no_zeros_in_input=True), [x_val],
mode=self.mode)
def second_deriv(x):
return theano.grad(Prod(no_zeros_in_input=True)(x), x)
unittest_tools.verify_grad(second_deriv, [x_val], mode=self.mode)
def test_verify_grad(self):
# including zeros, as the case with zeros is important
# (and special cases: 1 zero in the row, more than 1 zero in the row)
x_val = numpy.asarray([[1, 2, 3], [4, 5, 6], [7, 8, 9]],
dtype='float32')
# now with verify_grad
unittest_tools.verify_grad(Prod(axis=1), [x_val], mode=self.mode)
# second time, with some added complexity
# verify_grad takes the sum of the matrices anyway
def fn(x2):
return theano.tensor.sqr(Prod(axis=1)(x2))
unittest_tools.verify_grad(fn, [x_val], mode=self.mode)
def test_pickle_bug(self):
# Regression test for bug fixed in 24d4fd291054.
o = Prod()
s = cPickle.dumps(o, protocol=-1)
o = cPickle.loads(s)
cPickle.dumps(o)
def fn(x2):
return theano.tensor.sqr(Prod(axis=1)(x2))
def second_deriv(x):
return theano.grad(Prod(no_zeros_in_input=True)(x), x)
