def _GetVal(use_xla):
with self.cached_session():
t0 = array_ops.placeholder(np.float32, shape=input_sizes)
t1 = constant_op.constant(filter_sizes, shape=[len(filter_sizes)])
t2 = array_ops.placeholder(np.float32, shape=output_sizes)
native_t0 = t0
native_t2 = t2
strides = [1, stride, stride, 1]
if use_xla:
if data_format == "NCHW":
# Transpose from NWHC input to NCHW
# Ex. [4, 5, 5, 48] to [4, 48, 5, 5]
native_t0 = array_ops.transpose(t0, [0, 3, 1, 2])
native_t2 = array_ops.transpose(t2, [0, 3, 1, 2])
strides = [1, 1, stride, stride]
with self.test_scope():
backprop = nn_ops.depthwise_conv2d_native_backprop_filter(
native_t0,
t1,
native_t2,
strides=strides,
padding=padding,
data_format=data_format)
else:
# For CPU, the format NCHW is not supported. Therefore we always use
# NHWC here.
backprop = nn_ops.depthwise_conv2d_native_backprop_filter(
native_t0, t1, native_t2, strides=strides, padding=padding)
ret = backprop.eval({t0: x0, t2: x2})
self.assertShapeEqual(ret, backprop)
return ret
def _GetVal(use_xla):
with self.test_session():
t0 = array_ops.placeholder(np.float32, shape=input_sizes)
t1 = constant_op.constant(filter_sizes, shape=[len(filter_sizes)])
t2 = array_ops.placeholder(np.float32, shape=output_sizes)
if use_xla:
with self.test_scope():
backprop = nn_ops.depthwise_conv2d_native_backprop_filter(
t0, t1, t2, strides=[1, stride, stride, 1], padding=padding)
else:
backprop = nn_ops.depthwise_conv2d_native_backprop_filter(
t0, t1, t2, strides=[1, stride, stride, 1], padding=padding)
ret = backprop.eval({t0: x0, t2: x2})
self.assertShapeEqual(ret, backprop)
return ret
def _DepthwiseConv2dNativeBackpropInputGrad(op, grad):
"""The derivatives for deconvolution.
Args:
op: the Deconvolution op.
grad: the tensor representing the gradient w.r.t. the output
Returns:
the gradients w.r.t. the input and the filter
"""
return [
None,
nn_ops.depthwise_conv2d_native_backprop_filter(
grad,
array_ops.shape(op.inputs[1]),
op.inputs[2],
dilations=op.get_attr("dilations"),
strides=op.get_attr("strides"),
padding=op.get_attr("padding"),
data_format=op.get_attr("data_format")),
nn_ops.depthwise_conv2d_native(
grad,
op.inputs[1],
dilations=op.get_attr("dilations"),
strides=op.get_attr("strides"),
padding=op.get_attr("padding"),
data_format=op.get_attr("data_format"))
]
def _DepthwiseConv2dNativeGrad(op, grad):
return [
nn_ops.depthwise_conv2d_native_backprop_input(
array_ops.shape(op.inputs[0]), op.inputs[1], grad,
op.get_attr("strides"), op.get_attr("padding")),
nn_ops.depthwise_conv2d_native_backprop_filter(
op.inputs[0], array_ops.shape(op.inputs[1]), grad,
op.get_attr("strides"), op.get_attr("padding"))
]
def _GetVal(use_gpu):
with self.cached_session(use_gpu=use_gpu):
t0 = constant_op.constant(x0, shape=input_sizes)
t1 = constant_op.constant(filter_sizes, shape=[len(filter_sizes)])
t2 = constant_op.constant(x2, shape=output_sizes)
backprop = nn_ops.depthwise_conv2d_native_backprop_filter(
t0, t1, t2, strides=[1, stride, stride, 1], padding=padding)
ret = self.evaluate(backprop)
self.assertShapeEqual(ret, backprop)
return ret
def _DepthwiseConv2dNativeGrad(op, grad):
return [
nn_ops.depthwise_conv2d_native_backprop_input(
array_ops.shape(op.inputs[0]),
op.inputs[1],
grad,
op.get_attr("strides"),
op.get_attr("padding"),
data_format=op.get_attr("data_format")),
nn_ops.depthwise_conv2d_native_backprop_filter(
op.inputs[0],
array_ops.shape(op.inputs[1]),
grad,
op.get_attr("strides"),
op.get_attr("padding"),
data_format=op.get_attr("data_format"))
]
def _DepthwiseConv2dNativeBackpropInputGrad(op, grad):
"""The derivatives for deconvolution.
Args:
op: the Deconvolution op.
grad: the tensor representing the gradient w.r.t. the output
Returns:
the gradients w.r.t. the input and the filter
"""
return [
None,
nn_ops.depthwise_conv2d_native_backprop_filter(
grad, array_ops.shape(op.inputs[1]), op.inputs[2],
op.get_attr("strides"), op.get_attr("padding"),
op.get_attr("data_format")),
nn_ops.depthwise_conv2d_native(grad, op.inputs[1],
op.get_attr("strides"),
op.get_attr("padding"),
op.get_attr("data_format"))
]