Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Exposing use_cudnn #7536

Merged
merged 7 commits into from
Jan 18, 2018
Merged

Exposing use_cudnn #7536

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
78dc934
expose use_cudnn
chengduoZH Jan 15, 2018
79aa512
fix conv, pool, conv_trans to decide use cudnn or not
chengduoZH Jan 15, 2018
251c603
set use_cudnn as default
chengduoZH Jan 15, 2018
24f528a
follow comments
chengduoZH Jan 17, 2018
ed7e74a
follow comments and refine python doc
chengduoZH Jan 17, 2018
c9641a0
refine code
chengduoZH Jan 17, 2018
edd2132
remove conflict
chengduoZH Jan 18, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
15 changes: 15 additions & 0 deletions paddle/operators/conv_op.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -70,6 +70,13 @@ void ConvOp::InferShape(framework::InferShapeContext* ctx) const {
framework::OpKernelType ConvOp::GetExpectedKernelType(
const framework::ExecutionContext& ctx) const {
bool use_cudnn = ctx.Attr<bool>("use_cudnn");
use_cudnn &= platform::is_gpu_place(ctx.GetPlace());
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think use_cudnn means that user wants to force use cudnn operator, if env does not support cudnn, the framework should throw an error but not implicit change to use other operator.

Copy link
Contributor

@dzhwinter dzhwinter Jan 17, 2018 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Agree with the comment.
But we need a default policy right now, by default, we need to set use_cudnn True.
As we discussed offline, we reach the agreement that place is also a user configuration. we need to overwrite our default policy, not user specified.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

ok, I get the point, seems we still need a priority mechanism to resolve such kind of problem in the future.

#ifdef PADDLE_WITH_CUDA
if (platform::is_gpu_place(ctx.GetPlace())) {
auto& dev_ctx = ctx.template device_context<platform::CUDADeviceContext>();
use_cudnn &= dev_ctx.cudnn_handle() != nullptr;
}
#endif
framework::LibraryType library_;
if (use_cudnn) {
library_ = framework::LibraryType::kCUDNN;
Expand Down Expand Up @@ -283,6 +290,14 @@ void ConvOpGrad::InferShape(framework::InferShapeContext* ctx) const {
framework::OpKernelType ConvOpGrad::GetExpectedKernelType(
const framework::ExecutionContext& ctx) const {
bool use_cudnn = ctx.Attr<bool>("use_cudnn");
use_cudnn &= platform::is_gpu_place(ctx.GetPlace());
#ifdef PADDLE_WITH_CUDA
if (platform::is_gpu_place(ctx.GetPlace())) {
auto& dev_ctx = ctx.template device_context<platform::CUDADeviceContext>();
use_cudnn &= dev_ctx.cudnn_handle() != nullptr;
}
#endif

framework::LibraryType library_;
if (use_cudnn) {
library_ = framework::LibraryType::kCUDNN;
Expand Down
14 changes: 14 additions & 0 deletions paddle/operators/conv_transpose_op.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -61,6 +61,13 @@ void ConvTransposeOp::InferShape(framework::InferShapeContext* ctx) const {
framework::OpKernelType ConvTransposeOp::GetExpectedKernelType(
const framework::ExecutionContext& ctx) const {
bool use_cudnn = ctx.Attr<bool>("use_cudnn");
use_cudnn &= platform::is_gpu_place(ctx.GetPlace());
#ifdef PADDLE_WITH_CUDA
if (platform::is_gpu_place(ctx.GetPlace())) {
auto& dev_ctx = ctx.template device_context<platform::CUDADeviceContext>();
use_cudnn &= dev_ctx.cudnn_handle() != nullptr;
}
#endif
framework::LibraryType library_;
if (use_cudnn) {
library_ = framework::LibraryType::kCUDNN;
Expand Down Expand Up @@ -263,6 +270,13 @@ void ConvTransposeOpGrad::InferShape(framework::InferShapeContext* ctx) const {
framework::OpKernelType ConvTransposeOpGrad::GetExpectedKernelType(
const framework::ExecutionContext& ctx) const {
bool use_cudnn = ctx.Attr<bool>("use_cudnn");
use_cudnn &= platform::is_gpu_place(ctx.GetPlace());
#ifdef PADDLE_WITH_CUDA
if (platform::is_gpu_place(ctx.GetPlace())) {
auto& dev_ctx = ctx.template device_context<platform::CUDADeviceContext>();
use_cudnn &= dev_ctx.cudnn_handle() != nullptr;
}
#endif
framework::LibraryType library_;
if (use_cudnn) {
library_ = framework::LibraryType::kCUDNN;
Expand Down
14 changes: 14 additions & 0 deletions paddle/operators/pool_op.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -64,6 +64,13 @@ void PoolOp::InferShape(framework::InferShapeContext *ctx) const {
framework::OpKernelType PoolOp::GetExpectedKernelType(
const framework::ExecutionContext &ctx) const {
bool use_cudnn = ctx.Attr<bool>("use_cudnn");
use_cudnn &= platform::is_gpu_place(ctx.GetPlace());
#ifdef PADDLE_WITH_CUDA
if (platform::is_gpu_place(ctx.GetPlace())) {
auto &dev_ctx = ctx.template device_context<platform::CUDADeviceContext>();
use_cudnn &= dev_ctx.cudnn_handle() != nullptr;
}
#endif
framework::LibraryType library_;
if (use_cudnn) {
library_ = framework::LibraryType::kCUDNN;
Expand All @@ -88,6 +95,13 @@ void PoolOpGrad::InferShape(framework::InferShapeContext *ctx) const {
framework::OpKernelType PoolOpGrad::GetExpectedKernelType(
const framework::ExecutionContext &ctx) const {
bool use_cudnn = ctx.Attr<bool>("use_cudnn");
use_cudnn &= platform::is_gpu_place(ctx.GetPlace());
#ifdef PADDLE_WITH_CUDA
if (platform::is_gpu_place(ctx.GetPlace())) {
auto &dev_ctx = ctx.template device_context<platform::CUDADeviceContext>();
use_cudnn &= dev_ctx.cudnn_handle() != nullptr;
}
#endif
framework::LibraryType library_;
if (use_cudnn) {
library_ = framework::LibraryType::kCUDNN;
Expand Down
27 changes: 23 additions & 4 deletions python/paddle/v2/fluid/layers/nn.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -676,6 +676,7 @@ def conv2d(input,
groups=None,
param_attr=None,
bias_attr=None,
use_cudnn=True,
act=None):
"""
**Convlution2D Layer**
Expand Down Expand Up @@ -739,6 +740,8 @@ def conv2d(input,
connected to the second half of the input channels. Default: groups=1
param_attr(ParamAttr): The parameters to the Conv2d Layer. Default: None
bias_attr(ParamAttr): Bias parameter for the Conv2d 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

Returns:
Expand Down Expand Up @@ -774,6 +777,8 @@ def conv2d(input,
stride = [stride, stride]
if isinstance(padding, int):
padding = [padding, padding]
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
Expand All @@ -797,9 +802,12 @@ def _get_default_param_initializer():
'Filter': filter_param,
},
outputs={"Output": pre_bias},
attrs={'strides': stride,
'paddings': padding,
'groups': groups})
attrs={
'strides': stride,
'paddings': padding,
'groups': groups,
'use_cudnn': use_cudnn
})

pre_act = helper.append_bias_op(pre_bias, dim_start=1, dim_end=2)

Expand Down Expand Up @@ -948,6 +956,7 @@ def pool2d(input,
pool_stride=None,
pool_padding=None,
global_pooling=False,
use_cudnn=True,
name=None):
"""
This function adds the operator for pooling in 2 dimensions, using the
Expand All @@ -967,6 +976,8 @@ def pool2d(input,
pool_stride = [pool_stride, pool_stride]
if isinstance(pool_padding, int):
pool_padding = [pool_padding, pool_padding]
if not isinstance(use_cudnn, bool):
raise ValueError("use_cudnn should be True or False")

helper = LayerHelper('pool2d', **locals())
dtype = helper.input_dtype()
Expand All @@ -981,7 +992,8 @@ def pool2d(input,
"ksize": pool_size,
"global_pooling": global_pooling,
"strides": pool_stride,
"paddings": pool_padding
"paddings": pool_padding,
"use_cudnn": use_cudnn
})

return pool_out
Expand Down Expand Up @@ -1096,6 +1108,7 @@ def conv2d_transpose(input,
stride=None,
dilation=None,
param_attr=None,
use_cudnn=True,
name=None):
"""
The transpose of conv2d layer.
Expand Down Expand Up @@ -1123,6 +1136,8 @@ def conv2d_transpose(input,
contain two integers, (dilation_H, dilation_W). Otherwise, the
dilation_H = dilation_W = dilation.
param_attr: Parameter Attribute.
use_cudnn(bool): Use cudnn kernel or not, it is valid only when the cudnn
library is installed. Default: True
name(str|None): A name for this layer(optional). If set None, the layer
will be named automatically.

Expand Down Expand Up @@ -1151,6 +1166,10 @@ def conv2d_transpose(input,
elif dilation is not None:
op_attr['dilations'] = dilation

if not isinstance(use_cudnn, bool):
raise ValueError("use_cudnn should be True or False")
op_attr['use_cudnn'] = use_cudnn

if filter_size is None:
if output_size is None:
raise ValueError("output_size must be set when filter_size is None")
Expand Down
18 changes: 12 additions & 6 deletions python/paddle/v2/fluid/nets.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,19 +28,22 @@ def simple_img_conv_pool(input,
pool_stride,
act,
param_attr=None,
pool_type='max'):
pool_type='max',
use_cudnn=True):
conv_out = layers.conv2d(
input=input,
num_filters=num_filters,
filter_size=filter_size,
param_attr=param_attr,
act=act)
act=act,
use_cudnn=use_cudnn)

pool_out = layers.pool2d(
input=conv_out,
pool_size=pool_size,
pool_type=pool_type,
pool_stride=pool_stride)
pool_stride=pool_stride,
use_cudnn=use_cudnn)
return pool_out


Expand All @@ -54,7 +57,8 @@ def img_conv_group(input,
conv_with_batchnorm=False,
conv_batchnorm_drop_rate=None,
pool_stride=1,
pool_type=None):
pool_type=None,
use_cudnn=True):
"""
Image Convolution Group, Used for vgg net.
"""
Expand Down Expand Up @@ -85,7 +89,8 @@ def __extend_list__(obj):
filter_size=conv_filter_size[i],
padding=conv_padding[i],
param_attr=param_attr[i],
act=local_conv_act)
act=local_conv_act,
use_cudnn=use_cudnn)

if conv_with_batchnorm[i]:
tmp = layers.batch_norm(input=tmp, act=conv_act)
Expand All @@ -97,7 +102,8 @@ def __extend_list__(obj):
input=tmp,
pool_size=pool_size,
pool_type=pool_type,
pool_stride=pool_stride)
pool_stride=pool_stride,
use_cudnn=use_cudnn)
return pool_out


Expand Down