def test_pooling_with_tensor_vars(self):
x = ftensor4()
window_size = ivector()
stride = ivector()
padding = ivector()
data = np.random.normal(0, 1, (1, 1, 5, 5)).astype("float32")
# checking variable params vs fixed params
for ignore_border in [True, False]:
for mode in ["max", "sum", "average_inc_pad", "average_exc_pad"]:
y = pool_2d(x, window_size, ignore_border, stride, padding,
mode)
dx = aesara.gradient.grad(y.sum(), x)
var_fct = aesara.function([x, window_size, stride, padding],
[y, dx])
for ws in (4, 2, 5):
for st in (2, 3):
for pad in (0, 1):
if (pad > st or st > ws
or (pad != 0 and not ignore_border) or
(mode == "average_exc_pad" and pad != 0)):
continue
y = pool_2d(x, (ws, ws), ignore_border, (st, st),
(pad, pad), mode)
dx = aesara.gradient.grad(y.sum(), x)
fix_fct = aesara.function([x], [y, dx])
var_y, var_dx = var_fct(data, (ws, ws), (st, st),
(pad, pad))
fix_y, fix_dx = fix_fct(data)
utt.assert_allclose(var_y, fix_y)
utt.assert_allclose(var_dx, fix_dx)
def test_max_pool_2d_2D(self):
rng = np.random.RandomState(utt.fetch_seed())
maxpoolshps = ((1, 1), (3, 2))
imval = rng.rand(4, 5)
images = dmatrix()
for maxpoolshp, ignore_border, mode in product(
maxpoolshps,
[True, False],
["max", "sum", "average_inc_pad", "average_exc_pad"],
):
# print 'maxpoolshp =', maxpoolshp
# print 'ignore_border =', ignore_border
numpy_output_val = self.numpy_max_pool_2d(imval,
maxpoolshp,
ignore_border,
mode=mode)
output = pool_2d(images, maxpoolshp, ignore_border, mode=mode)
output_val = function([images], output)(imval)
utt.assert_allclose(output_val, numpy_output_val)
def mp(input):
return pool_2d(input, maxpoolshp, ignore_border, mode=mode)
utt.verify_grad(mp, [imval], rng=rng)
def test_DownsampleFactorMax_hessian(self):
# Example provided by Frans Cronje, see
# https://groups.google.com/d/msg/theano-users/qpqUy_3glhw/JMwIvlN5wX4J
x_vec = vector("x")
z = aet.dot(x_vec.dimshuffle(0, "x"), x_vec.dimshuffle("x", 0))
y = pool_2d(input=z, ws=(2, 2), ignore_border=True)
C = aet.exp(aet_sum(y))
grad_hess = aesara.gradient.hessian(cost=C, wrt=x_vec)
fn_hess = function(inputs=[x_vec], outputs=grad_hess)
# The value has been manually computed from the theoretical gradient,
# and confirmed by the implementation.
assert np.allclose(fn_hess([1, 2]), [[0.0, 0.0], [0.0, 982.7667]])
def test_max_pool_2d_3D(self):
rng = np.random.RandomState(utt.fetch_seed())
maxpoolshps = [(1, 2)]
imval = rng.rand(2, 3, 4)
images = dtensor3()
for maxpoolshp, ignore_border, mode in product(
maxpoolshps,
[True, False],
["max", "sum", "average_inc_pad", "average_exc_pad"],
):
# print 'maxpoolshp =', maxpoolshp
# print 'ignore_border =', ignore_border
numpy_output_val = self.numpy_max_pool_2d(imval, maxpoolshp,
ignore_border, mode)
output = pool_2d(images, maxpoolshp, ignore_border, mode=mode)
output_val = function([images], output)(imval)
utt.assert_allclose(output_val, numpy_output_val)
def test_DownsampleFactorMax(self):
rng = np.random.RandomState(utt.fetch_seed())
# maxpool, input size
examples = (
((2, ), (16, )),
(
(2, ),
(
4,
16,
),
),
(
(2, ),
(
4,
2,
16,
),
),
((1, 1), (4, 2, 16, 16)),
((2, 2), (4, 2, 16, 16)),
((3, 3), (4, 2, 16, 16)),
((3, 2), (4, 2, 16, 16)),
((3, 2, 2), (3, 2, 16, 16, 16)),
((2, 2, 3, 2), (3, 2, 6, 6, 6, 5)),
)
for example, ignore_border, mode in product(
examples,
[True, False],
["max", "sum", "average_inc_pad", "average_exc_pad"],
):
(maxpoolshp, inputsize) = example
imval = rng.rand(*inputsize)
images = aesara.shared(imval)
# Pure Numpy computation
numpy_output_val = self.numpy_max_pool_nd(imval,
maxpoolshp,
ignore_border,
mode=mode)
# The pool_2d or pool_3d helper methods
if len(maxpoolshp) == 2:
output = pool_2d(images, maxpoolshp, ignore_border, mode=mode)
f = function(
[],
[
output,
],
)
output_val = f()
utt.assert_allclose(output_val, numpy_output_val)
elif len(maxpoolshp) == 3:
output = pool_3d(images, maxpoolshp, ignore_border, mode=mode)
f = function(
[],
[
output,
],
)
output_val = f()
utt.assert_allclose(output_val, numpy_output_val)
# Pool op
maxpool_op = Pool(ndim=len(maxpoolshp),
ignore_border=ignore_border,
mode=mode)(images, maxpoolshp)
output_shape = Pool.out_shape(
imval.shape,
maxpoolshp,
ndim=len(maxpoolshp),
ignore_border=ignore_border,
)
utt.assert_allclose(np.asarray(output_shape),
numpy_output_val.shape)
f = function([], maxpool_op)
output_val = f()
utt.assert_allclose(output_val, numpy_output_val)
def mp(input):
return pool_2d(input, maxpoolshp, ignore_border, mode=mode)