Skip to content

Commit

Permalink
Fix DepthWiseConv2dNative (#267)
Browse files Browse the repository at this point in the history 
* fix shape inference of generic_op_define/DepthwiseConv2dNative

* onnx frontend support for DepthwiseConv2dNative op

* fix bug in TensorFlow frontend for DepthwiseConv2dNative op

* fix bug in onnx frontend for DepthwiseConv2dNative

Co-authored-by: Lingxiao Ma <lingm@microsoft.com>
  • Loading branch information
@xysmlx
xysmlx and Lingxiao Ma authored May 20, 2021
1 parent 12130ab commit bd3d9a9
Show file tree
Hide file tree
Showing 3 changed files with 104 additions and 46 deletions.
23 changes: 14 additions & 9 deletions src/nnfusion/core/operators/generic_op/generic_op_define/DepthwiseConv2dNative.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -10,13 +10,6 @@ REGISTER_OP(DepthwiseConv2dNative)
.attr<nnfusion::op::OpConfig::any>("dilations")
.attr<nnfusion::op::OpConfig::any>("padding_before")
.attr<nnfusion::op::OpConfig::any>("padding_after")
.constrait([](const nnfusion::op::OpConfig::any& config) -> bool {
if (config["padding_type"] != "SAME")
{
return false;
}
return true;
})
.infershape([](std::shared_ptr<graph::GNode> gnode) -> void {
NNFUSION_CHECK(gnode->get_input_size() == 2);
auto op = std::dynamic_pointer_cast<nnfusion::op::GenericOp>(gnode->get_op_ptr());
Expand All @@ -39,11 +32,23 @@ REGISTER_OP(DepthwiseConv2dNative)
const int64_t filter_rows = filter_shape[0];
const int64_t filter_cols = filter_shape[1];
const int64_t batch = input_shape[0];
auto padding_before = op->localOpConfig.getRoot()["padding_before"];
auto padding_after = op->localOpConfig.getRoot()["padding_after"];
const int64_t padding_h = padding_before[0];
const int64_t padding_w = padding_before[1];
const int64_t dilation_h = op->localOpConfig.getRoot()["dilations"][0];
const int64_t dilation_w = op->localOpConfig.getRoot()["dilations"][1];

std::vector<int64_t> strides = op->localOpConfig.getRoot()["strides"];
NNFUSION_CHECK(strides.size() == 2);
const int64_t out_rows = (input_rows + strides[0] - 1) / strides[0];
const int64_t out_cols = (input_cols + strides[1] - 1) / strides[1];
const int64_t out_rows =
(int64_t)((float)(input_rows + 2 * padding_h - dilation_h * (filter_rows - 1) - 1) /
(float)strides[0] +
1);
const int64_t out_cols =
(int64_t)((float)(input_cols + 2 * padding_w - dilation_w * (filter_cols - 1) - 1) /
(float)strides[1] +
1);

Shape output_shape(
is_nhwc
Expand Down
125 changes: 89 additions & 36 deletions src/nnfusion/frontend/onnx_import/op/conv.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -22,6 +22,7 @@
#include "conv.hpp"

#include <unordered_map>
#include "nnfusion/core/operators/generic_op/generic_op.hpp"
#include "nnfusion/frontend/onnx_import/util/broadcasting.hpp"

namespace nnfusion
Expand Down Expand Up @@ -143,44 +144,96 @@ namespace nnfusion
// split data and filters for group conv
std::size_t n_data_channels{data_shape.at(1)};
std::size_t n_filters_channels{filters_shape.at(0)};
NNFUSION_CHECK(n_data_channels % groups == 0 &&
n_filters_channels & groups == 0);
std::size_t data_group_size{n_data_channels / groups};
std::size_t filters_group_size{n_filters_channels / groups};

std::vector<std::size_t> data_lower_bounds(data_shape.size(), 0);
std::vector<std::size_t> data_upper_bounds{data_shape};
std::vector<std::size_t> filters_lower_bounds(filters_shape.size(), 0);
std::vector<std::size_t> filters_upper_bounds{filters_shape};

std::vector<std::shared_ptr<nnfusion::graph::GNode>> convolution_nodes;
for (std::size_t group = 0; group < groups; ++group)
if (n_data_channels == groups)
{
// slice data
data_lower_bounds[1] = group * data_group_size;
data_upper_bounds[1] = (group + 1) * data_group_size;
auto sliced_data_op =
std::make_shared<op::Slice>(data_lower_bounds, data_upper_bounds);
auto sliced_data = m_graph->add_node_and_edge(sliced_data_op, {data});
// slice filters
filters_lower_bounds[0] = group * filters_group_size;
filters_upper_bounds[0] = (group + 1) * filters_group_size;
auto sliced_filters_op = std::make_shared<op::Slice>(
filters_lower_bounds, filters_upper_bounds);
auto sliced_filters =
m_graph->add_node_and_edge(sliced_filters_op, {filters});

convolution_nodes.push_back(m_graph->add_node_and_edge(
std::make_shared<op::Convolution>(strides,
dilations,
padding_below,
padding_above,
conv_data_format),
{sliced_data, sliced_filters}));
// depthwise convolution
NNFUSION_CHECK(n_filters_channels == groups)
<< "Currently only support depth_multiplier = 1 in DepthwiseConv2d";

auto filter_gnode = GetInputNode(all_ng_nodes, node_proto, 1);
auto reshape_filter_op = std::make_shared<nnfusion::op::Reshape>(
nnfusion::AxisVector{2, 3, 0, 1},
nnfusion::Shape({filters_shape[2],
filters_shape[3],
filters_shape[0],
filters_shape[1]}));
reshape_filter_op->set_name(filter_gnode->get_name() +
"_filters_reshape");
auto reshape_filter_gnode =
m_graph->add_node_and_edge(reshape_filter_op, {filter_gnode});

size_t depth_multiplier = 1;
nnfusion::op::OpConfig::any myConfig;
myConfig["data_format"] = "NCHW";
myConfig["strides"] = strides;
myConfig["dilations"] = dilations;
myConfig["padding_before"] = padding_below;
myConfig["padding_after"] = padding_above;

if ((2 * padding_below[0] - dilations[0] * (filters_shape[2] - 1) ==
0) &&
(2 * padding_below[1] - dilations[1] * (filters_shape[3] - 1) == 0))
{
myConfig["padding_type"] = "SAME";
}
else if (padding_below[0] == 0 && padding_below[1] == 0)
{
myConfig["padding_type"] = "VALID";
}
else
{
NNFUSION_CHECK_FAIL() << "Currently only support SAME and VALID "
"padding in DepthwiseConv2dNative";
}

auto conv_op = std::make_shared<nnfusion::op::GenericOp>(
node_proto.name(), "DepthwiseConv2dNative", myConfig);
conv_node = m_graph->add_node_and_edge(
conv_op, {data, GNodeIndex{reshape_filter_gnode, 0}});
}
else
{
NNFUSION_CHECK(n_data_channels % groups == 0 &&
n_filters_channels & groups == 0);
std::size_t data_group_size{n_data_channels / groups};
std::size_t filters_group_size{n_filters_channels / groups};

std::vector<std::size_t> data_lower_bounds(data_shape.size(), 0);
std::vector<std::size_t> data_upper_bounds{data_shape};
std::vector<std::size_t> filters_lower_bounds(filters_shape.size(), 0);
std::vector<std::size_t> filters_upper_bounds{filters_shape};

std::vector<std::shared_ptr<nnfusion::graph::GNode>> convolution_nodes;
for (std::size_t group = 0; group < groups; ++group)
{
// slice data
data_lower_bounds[1] = group * data_group_size;
data_upper_bounds[1] = (group + 1) * data_group_size;
auto sliced_data_op = std::make_shared<op::Slice>(
data_lower_bounds, data_upper_bounds);
auto sliced_data =
m_graph->add_node_and_edge(sliced_data_op, {data});
// slice filters
filters_lower_bounds[0] = group * filters_group_size;
filters_upper_bounds[0] = (group + 1) * filters_group_size;
auto sliced_filters_op = std::make_shared<op::Slice>(
filters_lower_bounds, filters_upper_bounds);
auto sliced_filters =
m_graph->add_node_and_edge(sliced_filters_op, {filters});

convolution_nodes.push_back(m_graph->add_node_and_edge(
std::make_shared<op::Convolution>(strides,
dilations,
padding_below,
padding_above,
conv_data_format),
{sliced_data, sliced_filters}));
}
std::size_t concatenation_axis = 1;
conv_node = m_graph->add_node_and_edge(
std::make_shared<op::Concat>(concatenation_axis),
convolution_nodes);
}
std::size_t concatenation_axis = 1;
conv_node = m_graph->add_node_and_edge(
std::make_shared<op::Concat>(concatenation_axis), convolution_nodes);
}

// add bias
Expand Down
2 changes: 1 addition & 1 deletion src/nnfusion/frontend/tensorflow_import/util/graph_convert.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -869,7 +869,7 @@ namespace nnfusion
ng_padding_below,
ng_padding_above);

NNFUSION_CHECK(ng_padding_above == ng_padding_above)
NNFUSION_CHECK(ng_padding_below == ng_padding_above)
<< "Asymetric padding is not supported by now.";
nnfusion::op::OpConfig::any op_config;
op_config["data_format"] = tf_data_format;
Expand Down

0 comments on commit bd3d9a9

Please sign in to comment.