def testMaxPoolGradEagerShapeErrors(self):
with context.eager_mode():
orig_in = array_ops.ones((1, 1, 1, 1, 1))
# Test invalid orig_out shape
orig_out = array_ops.ones((1, 1, 1, 1, 2))
grad = array_ops.ones((1, 1, 1, 1, 1))
with self.assertRaisesRegex(
errors_impl.InvalidArgumentError,
r"Expected orig_output shape to be \[1,1,1,1,1\], but got "
r"\[1,1,1,1,2\]"):
gen_nn_ops.max_pool3d_grad(orig_in,
orig_out,
grad,
ksize=[1, 1, 1, 1, 1],
strides=[1, 1, 1, 1, 1],
padding="VALID")
with self.assertRaisesRegex(
errors_impl.InvalidArgumentError,
r"Expected orig_output shape to be \[1,1,1,1,1\], but got "
r"\[1,1,1,1,2\]"):
gen_nn_ops.max_pool3d_grad_grad(orig_in,
orig_out,
grad,
ksize=[1, 1, 1, 1, 1],
strides=[1, 1, 1, 1, 1],
padding="VALID")
# Test invalid grad shape
orig_out = array_ops.ones((1, 1, 1, 1, 1))
grad = array_ops.ones((1, 1, 1, 1, 2))
with self.assertRaisesRegex(
errors_impl.InvalidArgumentError,
r"Expected grad shape to be \[1,1,1,1,1\], but got \[1,1,1,1,2\]"
):
gen_nn_ops.max_pool3d_grad(orig_in,
orig_out,
grad,
ksize=[1, 1, 1, 1, 1],
strides=[1, 1, 1, 1, 1],
padding="VALID")
with self.assertRaisesRegex(
errors_impl.InvalidArgumentError,
r"Expected grad shape to be \[1,1,1,1,1\], but got \[1,1,1,1,2\]"
):
gen_nn_ops.max_pool3d_grad_grad(orig_in,
orig_out,
grad,
ksize=[1, 1, 1, 1, 1],
strides=[1, 1, 1, 1, 1],
padding="VALID")
def _MaxPool3DGrad(op, grad):
return gen_nn_ops.max_pool3d_grad(op.inputs[0],
op.outputs[0],
grad,
ksize=op.get_attr("ksize"),
strides=op.get_attr("strides"),
padding=op.get_attr("padding"),
data_format=op.get_attr("data_format"))
def _MaxPool3DGrad(op, grad):
return gen_nn_ops.max_pool3d_grad(
op.inputs[0],
op.outputs[0],
grad,
ksize=op.get_attr("ksize"),
strides=op.get_attr("strides"),
padding=op.get_attr("padding"),
data_format=op.get_attr("data_format").decode())
def _MaxPool3DGradGradGrad(op, grad):
return (array_ops.zeros_like(op.inputs[0]),
array_ops.zeros_like(op.inputs[1]),
gen_nn_ops.max_pool3d_grad(
op.inputs[0],
op.inputs[1],
grad,
op.get_attr("ksize"),
op.get_attr("strides"),
padding=op.get_attr("padding"),
data_format=op.get_attr("data_format").decode()))
def _MaxPool3DGradGradGrad(op, grad):
return (array_ops.zeros(shape=array_ops.shape(op.inputs[0]),
dtype=op.inputs[0].dtype),
array_ops.zeros(shape=array_ops.shape(op.inputs[1]),
dtype=op.inputs[1].dtype),
gen_nn_ops.max_pool3d_grad(op.inputs[0],
op.inputs[1],
grad,
op.get_attr("ksize"),
op.get_attr("strides"),
padding=op.get_attr("padding"),
data_format=op.get_attr("data_format")))
def _MaxPool3DGradGradGrad(op, grad):
return (array_ops.zeros(
shape=array_ops.shape(op.inputs[0]), dtype=op.inputs[0].dtype),
array_ops.zeros(
shape=array_ops.shape(op.inputs[1]), dtype=op.inputs[1].dtype),
gen_nn_ops.max_pool3d_grad(
op.inputs[0],
op.inputs[1],
grad,
op.get_attr("ksize"),
op.get_attr("strides"),
padding=op.get_attr("padding"),
data_format=op.get_attr("data_format").decode()))
def _z_dtd(layer, R, parameter1, parameter2):
print('_maxpooling3d_z_dtd')
Z = layer.output + 1e-12
S = R / Z
C = gen_nn_ops.max_pool3d_grad(
layer.input,
Z,
S,
ksize=(1, ) + layer.pool_size + (1, ),
strides=(1, ) + layer.strides + (1, ),
padding=K.tensorflow_backend._preprocess_padding(layer.padding))
return layer.input * C
