From: Павел Ильютченко/AI Tools Lab /SRR/Engineer/삼성전자
Date: Mon, 5 Aug 2019 13:49:31 +0000 (+0300)
Subject: [mir2loco] Support Conv2D operation (#5899)
X-Git-Tag: submit/tizen/20190809.050447~156
X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=ca288c68a34535f8ced0399c90330985657b850a;p=platform%2Fcore%2Fml%2Fnnfw.git
[mir2loco] Support Conv2D operation (#5899)
* Implemented `mir::Conv2DOp` transformation to `loco::Conv2D`
* Added test for this
Signed-off-by: Pavel Iliutchenko
---
diff --git a/compiler/mir2loco/src/mir2loco.cpp b/compiler/mir2loco/src/mir2loco.cpp
index c005b30..6f70dd7 100644
--- a/compiler/mir2loco/src/mir2loco.cpp
+++ b/compiler/mir2loco/src/mir2loco.cpp
@@ -19,6 +19,7 @@
#include "mir/ops/BiasAddOp.h"
#include "mir/ops/ConcatOp.h"
#include "mir/ops/ConstantOp.h"
+#include "mir/ops/Conv2DOp.h"
#include "mir/ops/PoolOp.h"
#include "mir/ops/ReluOp.h"
#include "mir/ops/ReshapeOp.h"
@@ -200,7 +201,52 @@ void Transformer::visit(mir::ops::ConstantOp &op)
_mir2loco_map.emplace(&op, const_node);
}
-void Transformer::visit(mir::ops::Conv2DOp &op) { throw std::runtime_error("NYI"); }
+void Transformer::visit(mir::ops::Conv2DOp &op)
+{
+ auto input = op.getInput(0)->getProducer()->getNode();
+ auto kernel = op.getInput(1)->getProducer()->getNode();
+ // Get ConstantOp
+ auto const_node = _mir2loco_map.at(kernel);
+
+ auto filter_enc = _loco_graph->nodes()->create();
+ {
+ auto enc = stdex::make_unique>();
+
+ // mir using filter convention as TF
+ // In TensorFlow, conv2d filter is a 4-D tensor of following shape:
+ // [filter_height, filter_width, in_channels, out_channels] -> HWIO (HWCN)
+ enc->perm()->axis(loco::FilterAxis::Height) = 0;
+ enc->perm()->axis(loco::FilterAxis::Width) = 1;
+ enc->perm()->axis(loco::FilterAxis::Depth) = 2;
+ enc->perm()->axis(loco::FilterAxis::Count) = 3;
+
+ filter_enc->encoder(std::move(enc));
+ }
+ // Set filter input
+ filter_enc->input(const_node);
+ // Setting up conv2d
+
+ // FeatureEncode
+ auto encode_node = createNHWCFeatureEncode(_loco_graph.get());
+ // Set Input
+ auto loco_it = _mir2loco_map.find(input);
+ assert(loco_it != _mir2loco_map.end()); // can't find the input
+ encode_node->input(loco_it->second);
+ // Conv2D
+ auto conv2d_node = _loco_graph->nodes()->create();
+ setupStride(op.getStrides(), conv2d_node->stride());
+ setupPad(op.getPaddingBefore(), op.getPaddingAfter(), conv2d_node->pad());
+ // Set Input
+ conv2d_node->ifm(encode_node);
+ conv2d_node->ker(filter_enc);
+ // FeatureDecode
+ auto decode_node = createNHWCFeatureDecode(_loco_graph.get());
+ // Set Input
+ decode_node->input(conv2d_node);
+ // Not set Shape
+ // Add to map
+ _mir2loco_map.emplace(&op, decode_node);
+}
void Transformer::visit(mir::ops::DeConv2DOp &op) { throw std::runtime_error("NYI"); }
diff --git a/compiler/mir2loco/src/mir2loco.test.cpp b/compiler/mir2loco/src/mir2loco.test.cpp
index b8d6770..077ee80 100644
--- a/compiler/mir2loco/src/mir2loco.test.cpp
+++ b/compiler/mir2loco/src/mir2loco.test.cpp
@@ -19,6 +19,7 @@
#include "mir/ops/BiasAddOp.h"
#include "mir/ops/ConcatOp.h"
#include "mir/ops/ConstantOp.h"
+#include "mir/ops/Conv2DOp.h"
#include "mir/ops/PoolOp.h"
#include "mir/ops/ReluOp.h"
#include "mir/ops/ReshapeOp.h"
@@ -323,3 +324,53 @@ TEST_F(TestTransformer_mir2loco, Bias_Add_Test)
ASSERT_EQ(bias_add_node->axis(), 3);
}
+
+TEST_F(TestTransformer_mir2loco, Conv2D_Test)
+{
+ mir::Graph mir_graph;
+
+ mir::Shape input_shape = mir::Shape({7, 7, 9, 1});
+ auto *input = mir_graph.create("input", input_shape);
+ mir::Shape shape = mir::Shape({2, 3, 1, 1});
+ const float data[] = {5.9, 6.7, 5.32, 54.11231, 43.2444, 3.409};
+ auto mir_tensor = mir::TensorVariant(mir::DTYPE::FLOAT32, shape, (const void *)data);
+ auto *constant = mir_graph.create("constant", mir_tensor);
+ auto *conv = mir_graph.create(
+ "conv", input->getOutput(0), constant->getOutput(0), mir::Shape{2, 3},
+ std::vector{5, 9}, std::vector{7, 4});
+ auto *output = mir_graph.create("output", conv->getOutput(0));
+
+ mir2loco::Transformer transformer;
+ auto loco_graph = transformer.transform(&mir_graph);
+
+ loco::Pull *pull_node = dynamic_cast(loco_graph->nodes()->at(0));
+ loco::ConstGen *const_node = dynamic_cast(loco_graph->nodes()->at(1));
+ loco::FilterEncode *filter_node = dynamic_cast(loco_graph->nodes()->at(2));
+ loco::FeatureEncode *encode_node =
+ dynamic_cast(loco_graph->nodes()->at(3));
+ loco::Conv2D *conv_node = dynamic_cast(loco_graph->nodes()->at(4));
+ loco::FeatureDecode *decode_node =
+ dynamic_cast(loco_graph->nodes()->at(5));
+ loco::Push *push_node = dynamic_cast(loco_graph->nodes()->at(6));
+
+ ASSERT_NE(pull_node, nullptr);
+ ASSERT_NE(const_node, nullptr);
+ ASSERT_NE(filter_node, nullptr);
+ ASSERT_NE(encode_node, nullptr);
+ ASSERT_NE(conv_node, nullptr);
+ ASSERT_NE(decode_node, nullptr);
+ ASSERT_NE(push_node, nullptr);
+ ASSERT_EQ(encode_node->input(), pull_node);
+ ASSERT_EQ(filter_node->input(), const_node);
+ ASSERT_EQ(conv_node->ifm(), encode_node);
+ ASSERT_EQ(conv_node->ker(), filter_node);
+ ASSERT_EQ(decode_node->input(), conv_node);
+ ASSERT_EQ(push_node->from(), decode_node);
+ // Check params
+ ASSERT_EQ(conv_node->pad()->left(), 5);
+ ASSERT_EQ(conv_node->pad()->top(), 9);
+ ASSERT_EQ(conv_node->pad()->right(), 7);
+ ASSERT_EQ(conv_node->pad()->bottom(), 4);
+ ASSERT_EQ(conv_node->stride()->horizontal(), 2);
+ ASSERT_EQ(conv_node->stride()->vertical(), 3);
+}