提交 d684b49c 编写于 作者: F fengjiayi

Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_add_doc

......@@ -22,6 +22,7 @@
| jczaja | Jacek Czaja |
| JiayiFeng | Jia-Yi Feng |
| kbinias | Krzysztof Binias |
| kexinzhao | Ke-Xin Zhao |
| kuke | Yi-Bing Liu |
| lcy-seso | Ying Cao |
| lipeng-unisound | Peng Li |
......
......@@ -155,6 +155,15 @@ copy(inference_lib DEPS paddle_fluid_shared paddle_fluid
DSTS ${dst_dir}/${module} ${dst_dir}/${module}
)
if(WITH_CONTRIB)
set(contrib_dst_dir "${FLUID_INSTALL_DIR}/contrib/inference")
copy(contrib_inference_lib DEPS paddle_inference_api
SRCS ${PADDLE_SOURCE_DIR}/paddle/contrib/inference/paddle_inference_api.h
${PADDLE_BINARY_DIR}/paddle/contrib/inference/libpaddle_inference_api.*
DSTS ${contrib_dst_dir} ${contrib_dst_dir}
)
endif()
set(module "platform")
copy(platform_lib DEPS profiler_py_proto
SRCS ${src_dir}/${module}/*.h ${src_dir}/${module}/dynload/*.h ${src_dir}/${module}/details/*.h
......
......@@ -342,6 +342,12 @@ conv2d
.. autofunction:: paddle.fluid.layers.conv2d
:noindex:
conv3d
------
.. autofunction:: paddle.fluid.layers.conv3d
:noindex:
sequence_pool
-------------
......@@ -366,6 +372,12 @@ pool2d
.. autofunction:: paddle.fluid.layers.pool2d
:noindex:
pool3d
------
.. autofunction:: paddle.fluid.layers.pool3d
:noindex:
batch_norm
----------
......@@ -384,6 +396,13 @@ conv2d_transpose
.. autofunction:: paddle.fluid.layers.conv2d_transpose
:noindex:
conv3d_transpose
----------------
.. autofunction:: paddle.fluid.layers.conv2d_transpose
:noindex:
sequence_expand
---------------
......
......@@ -104,7 +104,7 @@ no changes added to commit (use "git add" and/or "git commit -a")
➜ docker run -it -v $(pwd):/paddle paddle:latest-dev bash -c "cd /paddle/build && ctest"
```
关于构建和测试的更多信息,请参见[这篇文档](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/getstarted/build_and_install/docker_install_cn.rst)
关于构建和测试的更多信息,请参见[使用Docker安装运行](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/v2/build_and_install/docker_install_cn.rst)
## 提交(commit)
......
......@@ -50,7 +50,7 @@ cc_test(test_paddle_inference_api
inference_api_test(test_paddle_inference_api_impl
ARGS test_word2vec test_image_classification)
if (WITH_ANAKIN)
if (WITH_ANAKIN AND WITH_TESTING) # only needed in CI
# Due to Anakin do not have official library releases and the versions of protobuf and cuda do not match Paddle's,
# so anakin library will not be merged to our official inference library. To use anakin prediction API, one need to
# compile the libinference_anakin_api.a and compile with anakin.so.
......
......@@ -17,7 +17,7 @@ if(APPLE)
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wno-error=pessimizing-move")
endif(APPLE)
cc_library(tape_variable SRCS variable.cc DEPS ${FLUID_CORE_MODULES})
cc_library(tape_variable SRCS variable.cc DEPS ${FLUID_CORE_MODULES} device_context framework_proto proto_desc operator)
cc_library(tape SRCS tape.cc DEPS ${FLUID_CORE_MODULES} ${GLOB_OP_LIB} tape_variable)
cc_test(test_tape
......
......@@ -18,6 +18,7 @@ limitations under the License. */
#include "paddle/fluid/framework/init.h"
#include "paddle/fluid/framework/operator.h"
#include "paddle/fluid/operators/math/blas.h"
#include "paddle/fluid/platform/device_context.h"
#include "paddle/fluid/platform/place.h"
#include "paddle/fluid/string/piece.h"
......@@ -113,6 +114,9 @@ void InitDevices(bool init_p2p, const std::vector<int> devices) {
}
places.emplace_back(platform::CPUPlace());
platform::DeviceContextPool::Init(places);
#ifndef PADDLE_WITH_MKLDNN
operators::math::SetNumThreads(1);
#endif
}
void InitGLOG(const std::string &prog_name) {
......
......@@ -107,7 +107,13 @@ REGISTER_OPERATOR(concat, ops::ConcatOp, ops::ConcatOpMaker,
false> /* set false to disable empty grad */);
REGISTER_OPERATOR(concat_grad, ops::ConcatOpGrad);
REGISTER_OP_CPU_KERNEL(
concat, ops::ConcatKernel<paddle::platform::CPUDeviceContext, float>);
concat, ops::ConcatKernel<paddle::platform::CPUDeviceContext, double>,
ops::ConcatKernel<paddle::platform::CPUDeviceContext, float>,
ops::ConcatKernel<paddle::platform::CPUDeviceContext, int64_t>,
ops::ConcatKernel<paddle::platform::CPUDeviceContext, int>);
REGISTER_OP_CPU_KERNEL(
concat_grad,
ops::ConcatGradKernel<paddle::platform::CPUDeviceContext, float>);
ops::ConcatGradKernel<paddle::platform::CPUDeviceContext, double>,
ops::ConcatGradKernel<paddle::platform::CPUDeviceContext, float>,
ops::ConcatGradKernel<paddle::platform::CPUDeviceContext, int64_t>,
ops::ConcatGradKernel<paddle::platform::CPUDeviceContext, int>);
......@@ -15,7 +15,13 @@ limitations under the License. */
#include "paddle/fluid/operators/concat_op.h"
namespace ops = paddle::operators;
REGISTER_OP_CUDA_KERNEL(
concat, ops::ConcatKernel<paddle::platform::CUDADeviceContext, float>);
concat, ops::ConcatKernel<paddle::platform::CUDADeviceContext, double>,
ops::ConcatKernel<paddle::platform::CUDADeviceContext, float>,
ops::ConcatKernel<paddle::platform::CUDADeviceContext, int64_t>,
ops::ConcatKernel<paddle::platform::CUDADeviceContext, int>);
REGISTER_OP_CUDA_KERNEL(
concat_grad,
ops::ConcatGradKernel<paddle::platform::CUDADeviceContext, float>);
ops::ConcatGradKernel<paddle::platform::CUDADeviceContext, double>,
ops::ConcatGradKernel<paddle::platform::CUDADeviceContext, float>,
ops::ConcatGradKernel<paddle::platform::CUDADeviceContext, int64_t>,
ops::ConcatGradKernel<paddle::platform::CUDADeviceContext, int>);
......@@ -85,7 +85,7 @@ class GetPlacesOpProtoMaker : public framework::OpProtoAndCheckerMaker {
.InEnum({"CUDA", "CPU", "AUTO"})
.SetDefault("AUTO");
AddComment(R"DOC(
Returns a list of places based on flags. The list will be used for parallel
Returns a list of places based on arguments. The list will be used for parallel
execution.
)DOC");
}
......
......@@ -115,4 +115,7 @@ USE_CPU_ONLY_OP(concat);
REGISTER_OPERATOR(split, ops::SplitOp, ops::SplitOpMaker, ops::SplitGradMaker);
REGISTER_OP_CPU_KERNEL(split,
ops::SplitOpKernel<paddle::platform::CPUPlace, float>);
ops::SplitOpKernel<paddle::platform::CPUPlace, double>,
ops::SplitOpKernel<paddle::platform::CPUPlace, float>,
ops::SplitOpKernel<paddle::platform::CPUPlace, int64_t>,
ops::SplitOpKernel<paddle::platform::CPUPlace, int>);
......@@ -15,4 +15,7 @@ limitations under the License. */
#include "paddle/fluid/operators/split_op.h"
namespace ops = paddle::operators;
REGISTER_OP_CUDA_KERNEL(
split, ops::SplitOpKernel<paddle::platform::CUDADeviceContext, float>);
split, ops::SplitOpKernel<paddle::platform::CUDADeviceContext, double>,
ops::SplitOpKernel<paddle::platform::CUDADeviceContext, float>,
ops::SplitOpKernel<paddle::platform::CUDADeviceContext, int64_t>,
ops::SplitOpKernel<paddle::platform::CUDADeviceContext, int>);
......@@ -1224,6 +1224,34 @@ class IfElseBlockGuard(object):
class IfElse(object):
"""
if-else control flow.
Args:
cond (Variable): condition used to compare.
name (str, default None): The name of this layer.
Examples:
.. code-block:: python
limit = fluid.layers.fill_constant_batch_size_like(
input=label, dtype='int64', shape=[1], value=5.0)
cond = fluid.layers.less_than(x=label, y=limit)
ie = fluid.layers.IfElse(cond)
with ie.true_block():
true_image = ie.input(image)
hidden = fluid.layers.fc(input=true_image, size=100, act='tanh')
prob = fluid.layers.fc(input=hidden, size=10, act='softmax')
ie.output(prob)
with ie.false_block():
false_image = ie.input(image)
hidden = fluid.layers.fc(
input=false_image, size=200, act='tanh')
prob = fluid.layers.fc(input=hidden, size=10, act='softmax')
ie.output(prob)
prob = ie()
"""
OUT_IF_ELSE_BLOCKS = 0
IN_IF_ELSE_TRUE_BLOCKS = 1
IN_IF_ELSE_FALSE_BLOCKS = 2
......
......@@ -648,6 +648,26 @@ def read_file(reader):
class Preprocessor(object):
"""
A block for data pre-processing in reader.
Args:
reader (Variable): A reader variable.
name (str, default None): The name of the reader.
Examples:
.. code-block:: python
preprocessor = fluid.layers.io.Preprocessor(reader=reader)
with preprocessor.block():
img, lbl = preprocessor.inputs()
img_out = img / 2
lbl_out = lbl + 1
preprocessor.outputs(img_out, lbl_out)
data_file = fluid.layers.io.double_buffer(preprocessor())
"""
BEFORE_SUB_BLOCK = 0
IN_SUB_BLOCK = 1
AFTER_SUB_BLOCK = 2
......
......@@ -25,20 +25,72 @@ import utils
import random
__all__ = [
'fc', 'embedding', 'dynamic_lstm', 'dynamic_lstmp', 'dynamic_gru',
'gru_unit', 'linear_chain_crf', 'crf_decoding', 'cos_sim', 'cross_entropy',
'square_error_cost', 'chunk_eval', 'sequence_conv', 'conv2d',
'sequence_pool', 'sequence_softmax', 'softmax', 'pool2d', 'batch_norm',
'beam_search_decode', 'conv2d_transpose', 'sequence_expand', 'lstm_unit',
'reduce_sum', 'reduce_mean', 'reduce_max', 'reduce_min', 'reduce_prod',
'sequence_first_step', 'sequence_last_step', 'dropout', 'split',
'ctc_greedy_decoder', 'edit_distance', 'l2_normalize', 'matmul', 'topk',
'warpctc', 'sequence_reshape', 'transpose', 'im2sequence', 'nce',
'beam_search', 'row_conv', 'multiplex', 'layer_norm',
'softmax_with_cross_entropy', 'smooth_l1', 'one_hot',
'autoincreased_step_counter', 'reshape', 'lod_reset', 'lrn', 'pad',
'label_smooth', 'roi_pool', 'dice_loss', 'image_resize',
'image_resize_short', 'resize_bilinear', 'gather', 'random_crop', 'mean_iou'
'fc',
'embedding',
'dynamic_lstm',
'dynamic_lstmp',
'dynamic_gru',
'gru_unit',
'linear_chain_crf',
'crf_decoding',
'cos_sim',
'cross_entropy',
'square_error_cost',
'chunk_eval',
'sequence_conv',
'conv2d',
'conv3d',
'sequence_pool',
'sequence_softmax',
'softmax',
'pool2d',
'pool3d',
'batch_norm',
'beam_search_decode',
'conv2d_transpose',
'conv3d_transpose',
'sequence_expand',
'lstm_unit',
'reduce_sum',
'reduce_mean',
'reduce_max',
'reduce_min',
'reduce_prod',
'sequence_first_step',
'sequence_last_step',
'dropout',
'split',
'ctc_greedy_decoder',
'edit_distance',
'l2_normalize',
'matmul',
'topk',
'warpctc',
'sequence_reshape',
'transpose',
'im2sequence',
'nce',
'beam_search',
'row_conv',
'multiplex',
'layer_norm',
'softmax_with_cross_entropy',
'smooth_l1',
'one_hot',
'autoincreased_step_counter',
'reshape',
'lod_reset',
'lrn',
'pad',
'label_smooth',
'roi_pool',
'dice_loss',
'image_resize',
'image_resize_short',
'resize_bilinear',
'gather',
'random_crop',
'mean_iou',
]
......@@ -1275,8 +1327,6 @@ def conv2d(input,
conv2d = fluid.layers.conv2d(
input=data, num_filters=2, filter_size=3, act="relu")
"""
if stride is None:
stride = [1, 1]
num_channels = input.shape[1]
......@@ -1339,6 +1389,171 @@ def conv2d(input,
return helper.append_activation(pre_act)
def conv3d(input,
num_filters,
filter_size,
stride=1,
padding=0,
dilation=1,
groups=None,
param_attr=None,
bias_attr=None,
use_cudnn=True,
use_mkldnn=False,
act=None,
name=None):
"""
**Convlution3D Layer**
The convolution3D layer calculates the output based on the input, filter
and strides, paddings, dilations, groups parameters. Input(Input) and
Output(Output) are in NCDHW format. Where N is batch size C is the number of
channels, D is the depth of the feature, H is the height of the feature,
and W is the width of the feature. Convlution3D is similar with Convlution2D
but adds one dimension(depth). If bias attribution and activation type are
provided, bias is added to the output of the convolution, and the
corresponding activation function is applied to the final result.
For each input :math:`X`, the equation is:
.. math::
Out = \sigma (W \\ast X + b)
In the above equation:
* :math:`X`: Input value, a tensor with NCDHW format.
* :math:`W`: Filter value, a tensor with MCDHW format.
* :math:`\\ast`: Convolution operation.
* :math:`b`: Bias value, a 2-D tensor with shape [M, 1].
* :math:`\\sigma`: Activation function.
* :math:`Out`: Output value, the shape of :math:`Out` and :math:`X` may be
different.
Example:
- Input:
Input shape: :math:`(N, C_{in}, D_{in}, H_{in}, W_{in})`
Filter shape: :math:`(C_{out}, C_{in}, D_f, H_f, W_f)`
- Output:
Output shape: :math:`(N, C_{out}, D_{out}, H_{out}, W_{out})`
Where
.. math::
D_{out}&= \\frac{(D_{in} + 2 * paddings[0] - (dilations[0] * (D_f - 1) + 1))}{strides[0]} + 1 \\\\
H_{out}&= \\frac{(H_{in} + 2 * paddings[1] - (dilations[1] * (H_f - 1) + 1))}{strides[1]} + 1 \\\\
W_{out}&= \\frac{(W_{in} + 2 * paddings[2] - (dilations[2] * (W_f - 1) + 1))}{strides[2]} + 1
Args:
input (Variable): The input image with [N, C, D, H, W] format.
num_filters(int): The number of filter. It is as same as the output
image channel.
filter_size (int|tuple|None): The filter size. If filter_size is a tuple,
it must contain three integers, (filter_size_D, filter_size_H, filter_size_W).
Otherwise, the filter will be a square.
stride (int|tuple): The stride size. If stride is a tuple, it must
contain three integers, (stride_D, stride_H, stride_W). Otherwise, the
stride_D = stride_H = stride_W = stride. Default: stride = 1.
padding (int|tuple): The padding size. If padding is a tuple, it must
contain three integers, (padding_D, padding_H, padding_W). Otherwise, the
padding_D = padding_H = padding_W = padding. Default: padding = 0.
dilation (int|tuple): The dilation size. If dilation is a tuple, it must
contain three integers, (dilation_D, dilation_H, dilation_W). Otherwise, the
dilation_D = dilation_H = dilation_W = dilation. Default: dilation = 1.
groups (int): The groups number of the Conv3d Layer. According to grouped
convolution in Alex Krizhevsky's Deep CNN paper: when group=2,
the first half of the filters is only connected to the first half
of the input channels, while the second half of the filters is only
connected to the second half of the input channels. Default: groups=1
param_attr (ParamAttr): The parameters to the Conv3d Layer. Default: None
bias_attr (ParamAttr): Bias parameter for the Conv3d layer. Default: None
use_cudnn (bool): Use cudnn kernel or not, it is valid only when the cudnn
library is installed. Default: True
use_mkldnn (bool): Use mkldnn kernels or not.
act (str): Activation type. Default: None
name (str|None): A name for this layer(optional). If set None, the layer
will be named automatically.
Returns:
Variable: The tensor variable storing the convolution and \
non-linearity activation result.
Raises:
ValueError: If the shapes of input, filter_size, stride, padding and
groups mismatch.
Examples:
.. code-block:: python
data = fluid.layers.data(
name='data', shape=[3, 12, 32, 32], dtype='float32')
conv2d = fluid.layers.conv3d(
input=data, num_filters=2, filter_size=3, act="relu")
"""
l_type = 'conv3d'
helper = LayerHelper(l_type, **locals())
dtype = helper.input_dtype()
num_channels = input.shape[1]
if groups is None:
num_filter_channels = num_channels
else:
if num_channels % groups != 0:
raise ValueError("num_channels must be divisible by groups.")
num_filter_channels = num_channels / groups
filter_size = utils.convert_to_list(filter_size, 3, 'filter_size')
stride = utils.convert_to_list(stride, 3, 'stride')
padding = utils.convert_to_list(padding, 3, 'padding')
dilation = utils.convert_to_list(dilation, 3, 'dilation')
if not isinstance(use_cudnn, bool):
raise ValueError("use_cudnn should be True or False")
input_shape = input.shape
filter_shape = [num_filters, num_filter_channels] + filter_size
def _get_default_param_initializer():
std = (2.0 / (filter_size[0]**3 * num_channels))**0.5
return Normal(0.0, std, 0)
filter_param = helper.create_parameter(
attr=helper.param_attr,
shape=filter_shape,
dtype=dtype,
default_initializer=_get_default_param_initializer())
pre_bias = helper.create_tmp_variable(dtype)
helper.append_op(
type=l_type,
inputs={
'Input': input,
'Filter': filter_param,
},
outputs={"Output": pre_bias},
attrs={
'strides': stride,
'paddings': padding,
'dilations': dilation,
'groups': groups,
'use_cudnn': use_cudnn,
'use_mkldnn': use_mkldnn
})
pre_act = helper.append_bias_op(pre_bias, dim_start=1, dim_end=2)
return helper.append_activation(pre_act)
def sequence_pool(input, pool_type):
"""
This function add the operator for sequence pooling.
......@@ -1548,12 +1763,84 @@ def pool2d(input,
if not isinstance(use_cudnn, bool):
raise ValueError("use_cudnn should be True or False")
helper = LayerHelper('pool2d', **locals())
l_type = 'pool2d'
helper = LayerHelper(l_type, **locals())
dtype = helper.input_dtype()
pool_out = helper.create_tmp_variable(dtype)
helper.append_op(
type="pool2d",
type=l_type,
inputs={"X": input},
outputs={"Out": pool_out},
attrs={
"pooling_type": pool_type,
"ksize": pool_size,
"global_pooling": global_pooling,
"strides": pool_stride,
"paddings": pool_padding,
"use_cudnn": use_cudnn,
"ceil_mode": ceil_mode,
"use_mkldnn": use_mkldnn
})
return pool_out
def pool3d(input,
pool_size=-1,
pool_type="max",
pool_stride=1,
pool_padding=0,
global_pooling=False,
use_cudnn=True,
ceil_mode=False,
use_mkldnn=False,
name=None):
"""
This function adds the operator for pooling in 3-dimensions, using the
pooling configurations mentioned in input parameters.
Args:
input (Variable): ${input_comment}
pool_size (int): ${ksize_comment}
pool_type (str): ${pooling_type_comment}
pool_stride (int): stride of the pooling layer.
pool_padding (int): padding size.
global_pooling (bool): ${global_pooling_comment}
use_cudnn (bool): ${use_cudnn_comment}
ceil_mode (bool): ${ceil_mode_comment}
use_mkldnn (bool): ${use_mkldnn_comment}
name (str): A name for this layer(optional). If set None, the layer
will be named automatically.
Returns:
Variable: output of pool3d layer.
"""
if pool_type not in ["max", "avg"]:
raise ValueError(
"Unknown pool_type: '%s'. It can only be 'max' or 'avg'.",
str(pool_type))
if global_pooling is False and pool_size == -1:
raise ValueError(
"When the global_pooling is False, pool_size must be passed "
"and be a valid value. Received pool_size: " + str(pool_size))
pool_size = utils.convert_to_list(pool_size, 3, 'pool_size')
pool_padding = utils.convert_to_list(pool_padding, 3, 'pool_padding')
pool_stride = utils.convert_to_list(pool_stride, 3, 'pool_stride')
if not isinstance(use_cudnn, bool):
raise ValueError("use_cudnn should be True or False")
l_type = "pool3d"
helper = LayerHelper(l_type, **locals())
dtype = helper.input_dtype()
pool_out = helper.create_tmp_variable(dtype)
helper.append_op(
type=l_type,
inputs={"X": input},
outputs={"Out": pool_out},
attrs={
......@@ -1974,6 +2261,173 @@ def conv2d_transpose(input,
return out
def conv3d_transpose(input,
num_filters,
output_size=None,
filter_size=None,
padding=0,
stride=1,
dilation=1,
groups=None,
param_attr=None,
bias_attr=None,
use_cudnn=True,
act=None,
name=None):
"""
**Convlution3D transpose layer**
The convolution3D transpose layer calculates the output based on the input,
filter, and dilations, strides, paddings. Input(Input) and output(Output)
are in NCDHW format. Where N is batch size, C is the number of channels,
D is the depth of the feature, H is the height of the feature, and W
is the width of the feature. Parameters(dilations, strides, paddings) are
two elements. These two elements represent height and width, respectively.
The details of convolution transpose layer, please refer to the following
explanation and references `therein <http://www.matthewzeiler.com/wp-content/uploads/2017/07/cvpr2010.pdf>`_.
For each input :math:`X`, the equation is:
.. math::
Out = W \\ast X
In the above equation:
* :math:`X`: Input value, a tensor with NCDHW format.
* :math:`W`: Filter value, a tensor with MCDHW format.
* :math:`\\ast` : Convolution transpose operation.
* :math:`Out`: Output value, the shape of :math:`Out` and :math:`X` may be
different.
Example:
- Input:
Input shape: $(N, C_{in}, D_{in}, H_{in}, W_{in})$
Filter shape: $(C_{in}, C_{out}, D_f, H_f, W_f)$
- Output:
Output shape: $(N, C_{out}, D_{out}, H_{out}, W_{out})$
Where
.. math::
D_{out} &= (D_{in} - 1) * strides[0] - 2 * paddings[0] + dilations[0] * (D_f - 1) + 1 \\\\
H_{out} &= (H_{in} - 1) * strides[1] - 2 * paddings[1] + dilations[1] * (H_f - 1) + 1 \\\\
W_{out} &= (W_{in} - 1) * strides[2] - 2 * paddings[2] + dilations[2] * (W_f - 1) + 1
Args:
input(Variable): The input image with [N, C, D, H, W] format.
num_filters(int): The number of the filter. It is as same as the output
image channel.
output_size(int|tuple|None): The output image size. If output size is a
tuple, it must contain three integers, (image_D, image_H, image_W). This
parameter only works when filter_size is None.
filter_size(int|tuple|None): The filter size. If filter_size is a tuple,
it must contain three integers, (filter_size_D, filter_size_H, filter_size_W).
Otherwise, the filter will be a square. None if use output size to
calculate filter_size.
padding(int|tuple): The padding size. If padding is a tuple, it must
contain three integers, (padding_D, padding_H, padding_W). Otherwise, the
padding_D = padding_H = padding_W = padding. Default: padding = 0.
stride(int|tuple): The stride size. If stride is a tuple, it must
contain three integers, (stride_D, stride_H, stride_W). Otherwise, the
stride_D = stride_H = stride_W = stride. Default: stride = 1.
dilation(int|tuple): The dilation size. If dilation is a tuple, it must
contain three integers, (dilation_D, dilation_H, dilation_W). Otherwise, the
dilation_D = dilation_H = dilation_W = dilation. Default: dilation = 1.
groups(int): The groups number of the Conv3d transpose layer. Inspired by
grouped convolution in Alex Krizhevsky's Deep CNN paper, in which
when group=2, the first half of the filters is only connected to the
first half of the input channels, while the second half of the
filters is only connected to the second half of the input channels.
Default: groups=1
param_attr(ParamAttr): The parameters to the Conv3d_transpose Layer.
Default: None
bias_attr(ParamAttr): Bias parameter for the Conv3d layer. Default: None
use_cudnn(bool): Use cudnn kernel or not, it is valid only when the cudnn
library is installed. Default: True
act(str): Activation type. Default: None
name(str|None): A name for this layer(optional). If set None, the layer
will be named automatically.
Returns:
Variable: The tensor variable storing the convolution transpose result.
Raises:
ValueError: If the shapes of input, filter_size, stride, padding and
groups mismatch.
Examples:
.. code-block:: python
data = fluid.layers.data(
name='data', shape=[3, 12, 32, 32], dtype='float32')
conv2d_transpose = fluid.layers.conv3d_transpose(
input=data, num_filters=2, filter_size=3)
"""
l_type = "conv3d_transpose"
helper = LayerHelper(l_type, **locals())
if not isinstance(input, Variable):
raise TypeError("Input of conv3d_transpose must be Variable")
input_channel = input.shape[1]
padding = utils.convert_to_list(padding, 3, 'padding')
stride = utils.convert_to_list(stride, 3, 'stride')
dilation = utils.convert_to_list(dilation, 3, 'dilation')
if not isinstance(use_cudnn, bool):
raise ValueError("use_cudnn should be True or False")
if filter_size is None:
if output_size is None:
raise ValueError("output_size must be set when filter_size is None")
if isinstance(output_size, int):
output_size = [output_size, output_size]
d_in = input.shape[2]
h_in = input.shape[3]
w_in = input.shape[4]
filter_size_d = (output_size[0] - (d_in - 1) * stride[0] + 2 *
padding[0] - 1) / dilation[0] + 1
filter_size_h = (output_size[1] - (h_in - 1) * stride[1] + 2 *
padding[1] - 1) / dilation[1] + 1
filter_size_w = (output_size[2] - (w_in - 1) * stride[2] + 2 *
padding[2] - 1) / dilation[2] + 1
filter_size = [filter_size_d, filter_size_h, filter_size_w]
else:
filter_size = utils.convert_to_list(filter_size, 3,
'conv3d_transpose.filter_size')
groups = 1 if groups is None else groups
filter_shape = [input_channel, num_filters / groups] + filter_size
img_filter = helper.create_parameter(
dtype=input.dtype, shape=filter_shape, attr=helper.param_attr)
pre_bias = helper.create_tmp_variable(dtype=input.dtype)
helper.append_op(
type=l_type,
inputs={'Input': [input],
'Filter': [img_filter]},
outputs={'Output': pre_bias},
attrs={
'strides': stride,
'paddings': padding,
'dilations': dilation,
'groups': groups,
'use_cudnn': use_cudnn
})
pre_act = helper.append_bias_op(pre_bias, dim_start=1, dim_end=2)
out = helper.append_activation(pre_act)
return out
def sequence_expand(x, y, ref_level=-1, name=None):
"""Sequence Expand Layer. This layer will expand the input variable **x**
according to specified level lod of **y**. Please note that lod level of
......
......@@ -76,8 +76,7 @@ def db_lstm(word, predicate, ctx_n2, ctx_n1, ctx_0, ctx_p1, ctx_p2, mark,
emb_layers.append(mark_embedding)
hidden_0_layers = [
fluid.layers.fc(input=emb, size=hidden_dim, act='tanh')
for emb in emb_layers
fluid.layers.fc(input=emb, size=hidden_dim) for emb in emb_layers
]
hidden_0 = fluid.layers.sums(input=hidden_0_layers)
......@@ -94,8 +93,8 @@ def db_lstm(word, predicate, ctx_n2, ctx_n1, ctx_0, ctx_p1, ctx_p2, mark,
for i in range(1, depth):
mix_hidden = fluid.layers.sums(input=[
fluid.layers.fc(input=input_tmp[0], size=hidden_dim, act='tanh'),
fluid.layers.fc(input=input_tmp[1], size=hidden_dim, act='tanh')
fluid.layers.fc(input=input_tmp[0], size=hidden_dim),
fluid.layers.fc(input=input_tmp[1], size=hidden_dim)
])
lstm = fluid.layers.dynamic_lstm(
......
......@@ -41,8 +41,8 @@ function(py_test_modules TARGET_NAME)
endfunction()
list(REMOVE_ITEM TEST_OPS test_warpctc_op)
list(REMOVE_ITEM TEST_OPS test_dist_train)
#list(REMOVE_ITEM TEST_OPS test_parallel_executor_crf)
#list(REMOVE_ITEM TEST_OPS test_parallel_executor_fetch_feed)
list(REMOVE_ITEM TEST_OPS test_parallel_executor_crf)
list(REMOVE_ITEM TEST_OPS test_parallel_executor_fetch_feed)
# TODO(wuyi): this test hungs on CI, will add it back later
list(REMOVE_ITEM TEST_OPS test_listen_and_serv_op)
foreach(TEST_OP ${TEST_OPS})
......@@ -50,3 +50,5 @@ foreach(TEST_OP ${TEST_OPS})
endforeach(TEST_OP)
py_test_modules(test_warpctc_op MODULES test_warpctc_op ENVS FLAGS_warpctc_dir=${WARPCTC_LIB_DIR} SERIAL)
py_test_modules(test_dist_train MODULES test_dist_train SERIAL)
py_test_modules(test_parallel_executor_crf MODULES test_parallel_executor_crf SERIAL)
py_test_modules(test_parallel_executor_fetch_feed MODULES test_parallel_executor_fetch_feed SERIAL)
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册