Merge branch 'master' into 'master'

enable conv_winograd, fix conv_gemmlike bug, and update the unit tests of conv op

See merge request paddlepaddle/paddlelitearmbackend!1

paddle/fluid/lite/arm/math/CMakeLists.txt

@@ -26,5 +26,6 @@ cc_library(math_arm SRCS
     conv_depthwise.cc
     conv_gemmlike.cc
     conv_winograd_3x3.cc
+    conv_winograd.cc
     DEPS ${lite_kernel_deps} eigen3)
  

paddle/fluid/lite/kernels/arm/conv_compute.cc

@@ -13,10 +13,6 @@
 // limitations under the License.
 
 #include "paddle/fluid/lite/kernels/arm/conv_compute.h"
-#include "paddle/fluid/lite/arm/math/conv_direct.h"
-#include "paddle/fluid/lite/arm/math/conv_depthwise.h"
-#include "paddle/fluid/lite/arm/math/conv_gemmlike.h"
-#include "paddle/fluid/lite/arm/math/funcs.h"
 #include "paddle/fluid/lite/core/op_registry.h"
 #include "paddle/fluid/lite/core/type_system.h"
 
@@ -25,7 +21,7 @@ namespace lite {
 namespace kernels {
 namespace arm {
 
-void ConvCompute::Run() {
+void ConvCompute::PrepareForRun() {
   auto& param = this->Param<param_t>();
   auto x_dims = param.x->dims();
   auto w_dims = param.filter->dims();
@@ -61,44 +57,42 @@ void ConvCompute::Run() {
   bool flag_dw = flag_dw_3x3 || flag_dw_5x5;
 
   // select conv impl
-  // TODO(xxx): enable more
   if (param.groups == ic && ic == oc && kps_equal && no_dilation && flag_dw) {
     // dw conv impl
     impl_ = new lite::arm::math::DepthwiseConv<PRECISION(kFloat)>;
-    LOG(INFO) << "invoking dw conv";
+    // LOG(INFO) << "invoking dw conv";
   } else if (param.groups == 1 && kw == 3 && stride == 1 && kps_equal &&
              no_dilation) {
     if (ic >= 32 && oc >= 32 && oh > 16 && ow > 16) {
       // winograd conv impl
-      // impl_ = new lite::arm::math::WinogradConv<PRECISION(kFloat)>;
-      LOG(FATAL) << "TODO!!! winograd conv";
+      impl_ = new lite::arm::math::WinogradConv<PRECISION(kFloat)>;
+      // LOG(INFO) << "invoking winograd conv";
     } else {
       // direct conv impl
       impl_ = new lite::arm::math::DirectConv<PRECISION(kFloat)>;
-      LOG(INFO) << "invoking direct conv";
+      // LOG(INFO) << "invoking direct conv";
     }
   } else if (param.groups == 1 && kw == 3 && stride == 2 && kps_equal &&
              no_dilation) {
     // direct conv impl
     impl_ = new lite::arm::math::DirectConv<PRECISION(kFloat)>;
+    // LOG(INFO) << "invoking direct conv";
   } else {
     impl_ = new lite::arm::math::GemmLikeConv<PRECISION(kFloat)>;
-    LOG(INFO) << "invoking gemm like conv";
+    // LOG(INFO) << "invoking gemm like conv";
   }
-  this->impl_->create(param, &ctx);
+  CHECK(this->impl_->create(param, &ctx));
+}
 
+void ConvCompute::Run() {
+  auto& param = this->Param<param_t>();
   CHECK(impl_);
   impl_->run(param);
-
   // if (this->act_ != nullptr) {
   //   this->act_->run(outputs, outputs, param.activation_param);
   // }
 }
 
-TargetType ConvCompute::target() const { return TARGET(kARM); }
-
-PrecisionType ConvCompute::precision() const { return PRECISION(kFloat); }
-
 }  // namespace arm
 }  // namespace kernels
 }  // namespace lite

paddle/fluid/lite/kernels/arm/conv_compute.h

@@ -13,8 +13,7 @@
 // limitations under the License.
 
 #pragma once
-
-#include "paddle/fluid/lite/arm/math/conv_impl.h"
+#include "paddle/fluid/lite/arm/math/funcs.h"
 #include "paddle/fluid/lite/core/kernel.h"
 #include "paddle/fluid/lite/operators/conv_op.h"
 
@@ -27,10 +26,9 @@ class ConvCompute : public KernelLite<TARGET(kARM), PRECISION(kFloat)> {
  public:
   using param_t = operators::ConvParam;
 
-  void Run() override;
+  void PrepareForRun() override;
 
-  TargetType target() const override;
-  PrecisionType precision() const override;
+  void Run() override;
 
   virtual ~ConvCompute() = default;
 

paddle/fluid/lite/kernels/arm/conv_compute_test.cc

@@ -17,7 +17,6 @@
 #include <memory>
 #include <utility>
 #include <vector>
-#include "paddle/fluid/lite/arm/math/funcs.h"
 #include "paddle/fluid/lite/core/op_registry.h"
 
 namespace paddle {
@@ -76,7 +75,9 @@ void conv_compute_ref(const operators::ConvParam& param) {
             int out_idx = n * groups * out_c_group * hout * wout +
                           g * out_c_group * hout * wout + oc * hout * wout +
                           oh * wout + ow;
-            output_data[out_idx] = 0.0f;
+            output_data[out_idx] =
+                flag_bias ? static_cast<float>(bias_data[g * out_c_group + oc])
+                          : 0.f;
             for (int ic = 0; ic < in_c_group; ++ic) {
               for (int kh = 0; kh < kernel_h; ++kh) {
                 for (int kw = 0; kw < kernel_w; ++kw) {
@@ -97,9 +98,6 @@ void conv_compute_ref(const operators::ConvParam& param) {
                 }
               }
             }
-            output_data[out_idx] +=
-                flag_bias ? static_cast<float>(bias_data[g * out_c_group + oc])
-                          : 0.f;
             if (flag_relu) {
               output_data[out_idx] =
                   output_data[out_idx] > 0.f ? output_data[out_idx] : 0.f;
@@ -112,8 +110,8 @@ void conv_compute_ref(const operators::ConvParam& param) {
 }
 
 TEST(conv_arm, retrive_op) {
-  auto conv =
-      KernelRegistry::Global().Create<TARGET(kARM), PRECISION(kFloat)>("conv2d");
+  auto conv = KernelRegistry::Global().Create<TARGET(kARM), PRECISION(kFloat)>(
+      "conv2d");
   ASSERT_FALSE(conv.empty());
   ASSERT_TRUE(conv.front());
 }
@@ -125,73 +123,72 @@ TEST(conv_arm, init) {
 }
 
 TEST(conv_arm, compute) {
-  ConvCompute conv;
-  operators::ConvParam param;
-
   lite::Tensor input;
   lite::Tensor filter;
   lite::Tensor bias;
   lite::Tensor output;
   lite::Tensor output_ref;
-
   DeviceInfo::Init();
-  std::unique_ptr<KernelContext> ctx(new KernelContext);
-  ctx->As<ARMContext>();
-  conv.SetContext(std::move(ctx));
   for (auto n : {1, 2}) {
-    for (auto chin : {3, 8, /*32, 128*/}) {
-      for (auto chout : {3, 8, /*32, 128*/}) {
-        for (auto hin : {7, 14, 28, /*56 , 112, 224, 512*/}) {
-          for (auto win : {7, 14, 28, /*56, 112, 224, 512*/}) {
-            for (auto flag_bias : {false , true}) {
-              for (auto flag_relu : {false , true}) {
+    for (auto ic : {6, 32 /*, 128*/}) {
+      for (auto oc : {6, 32 /*, 128*/}) {
+        for (auto ih : {9, 18 /*, 56 , 112, 224, 512*/}) {
+          for (auto iw : {9, 18 /*, 56, 112, 224, 512*/}) {
+            for (auto flag_bias : {false, true}) {
+              for (auto flag_relu : {false, true}) {
                 for (auto depthwise : {false, true}) {
-                  for (auto dilation : {1 /*, 2*/}) {
+                  for (auto dilation : {1, 2}) {
                     for (auto stride : {1, 2}) {
-                      for (auto padding : {0, 1}) {
-                        for (auto ks : {/*1, */3/*, 5*/}) {
+                      for (auto padding : {0, 1, 2}) {
+                        for (auto ks : {1, 3, 5}) {
                           int group = 1;
-                          if (depthwise) {  // depthwise conv ?
-                            group = chin;
-                            chout = chin;
-                            // remove the follow code if
-                            // all kernels are implemented.
-                            if (ks == 5) {
-                              stride = 2;
-                              padding = 2;
-                            }
+                          if (depthwise) {  // depthwise convolution ?
+                            group = oc = ic;
                           }
                           // get input, filter and output shape
-                          std::vector<int64_t> input_shape = {n, chin, hin,
-                                                              win};
-                          std::vector<int64_t> filter_shape = {
-                              chout, chin / group, ks, ks};
-                          std::vector<int64_t> output_shape({n, chout});
+                          std::vector<int64_t> input_shape = {n, ic, ih, iw};
+                          std::vector<int64_t> filter_shape = {oc, ic / group,
+                                                               ks, ks};
+                          std::vector<int64_t> output_shape({n, oc});
                           const int dkernel = dilation * (ks - 1) + 1;
                           output_shape.push_back(
-                              (hin + 2 * padding - dkernel) / stride + 1);
+                              (ih + 2 * padding - dkernel) / stride + 1);
                           output_shape.push_back(
-                              (win + 2 * padding - dkernel) / stride + 1);
+                              (iw + 2 * padding - dkernel) / stride + 1);
                           // resize input, filter and output
-                          input.Resize(DDim(input_shape));
-                          filter.Resize(DDim(filter_shape));
-                          output.Resize(DDim(output_shape));
-                          output_ref.Resize(DDim(output_shape));
+                          input.Resize(input_shape);
+                          filter.Resize(filter_shape);
+                          output.Resize(output_shape);
+                          output_ref.Resize(output_shape);
                           auto* input_data = input.mutable_data<float>();
                           auto* filter_data = filter.mutable_data<float>();
                           auto* output_data = output.mutable_data<float>();
-                          auto* output_ref_data =
-                              output_ref.mutable_data<float>();
                           for (int i = 0; i < input.dims().production(); i++) {
                             input_data[i] = static_cast<float>(i % 128);
                           }
                           for (int i = 0; i < filter.dims().production(); i++) {
-                            filter_data[i] = i / 1000.0f;
+                            filter_data[i] =
+                                i * 0.001f /
+                                static_cast<float>(filter.dims().production());
                           }
+                          // prepare kernel params and run
+                          ConvCompute conv;
+                          std::unique_ptr<KernelContext> ctx(new KernelContext);
+                          ctx->As<ARMContext>();
+                          conv.SetContext(std::move(ctx));
+                          operators::ConvParam param;
                           param.x = &input;
                           param.filter = &filter;
                           param.output = &output;
                           param.bias = nullptr;
+                          if (flag_bias) {
+                            bias.Resize({oc});
+                            auto* bias_data = bias.mutable_data<float>();
+                            for (int i = 0; i < bias.dims().production(); i++) {
+                              bias_data[i] = static_cast<float>(i);
+                            }
+                            param.bias = &bias;
+                          }
                           // TODO(hong19860320) param.relu = flag_relu;
                           param.paddings = std::vector<int>({padding, padding});
                           param.strides = std::vector<int>({stride, stride});
@@ -199,9 +196,12 @@ TEST(conv_arm, compute) {
                               std::vector<int>({dilation, dilation});
                           param.groups = group;
                           conv.SetParam(param);
-                          conv.Run();
+                          conv.Launch();
+                          // invoking ref implementation and compare results
                           param.output = &output_ref;
                           conv_compute_ref<float>(param);
+                          auto* output_ref_data =
+                              output_ref.mutable_data<float>();
                           for (int i = 0; i < output.dims().production(); i++) {
                             EXPECT_NEAR(output_data[i], output_ref_data[i],
                                         1e-3);
@@ -218,59 +218,6 @@ TEST(conv_arm, compute) {
       }
     }
   }
-#if 0
-// for testing gemm like conv
-  int n = 1;
-  int chin = 8;
-  int chout = 8;
-  int hin = 14;
-  int win = 14;
-  int flag_bias = false;
-  int flag_relu = false;
-  int dilation = 1;
-  int stride = 1;
-  int padding = 1;
-  int ks = 5;
-  int group = 1;
-  // get input, filter and output shape
-  std::vector<int64_t> input_shape = {n, chin, hin, win};
-  std::vector<int64_t> filter_shape = {chout, chin / group, ks, ks};
-  std::vector<int64_t> output_shape({n, chout});
-  const int dkernel = dilation * (ks - 1) + 1;
-  output_shape.push_back((hin + 2 * padding - dkernel) / stride + 1);
-  output_shape.push_back((win + 2 * padding - dkernel) / stride + 1);
-  // resize input, filter and output
-  input.Resize(DDim(input_shape));
-  filter.Resize(DDim(filter_shape));
-  output.Resize(DDim(output_shape));
-  output_ref.Resize(DDim(output_shape));
-  auto* input_data = input.mutable_data<float>();
-  auto* filter_data = filter.mutable_data<float>();
-  auto* output_data = output.mutable_data<float>();
-  auto* output_ref_data = output_ref.mutable_data<float>();
-  for (int i = 0; i < input.dims().production(); i++) {
-    input_data[i] = static_cast<float>(i % 128);
-  }
-  for (int i = 0; i < filter.dims().production(); i++) {
-    filter_data[i] = i / 1000.0f;
-  }
-  param.x = &input;
-  param.filter = &filter;
-  param.output = &output;
-  param.bias = nullptr;
-  // TODO(hong19860320) param.relu = flag_relu;
-  param.paddings = std::vector<int>({padding, padding});
-  param.strides = std::vector<int>({stride, stride});
-  param.dilations = std::vector<int>({dilation, dilation});
-  param.groups = group;
-  conv.SetParam(param);
-  conv.Run();
-  param.output = &output_ref;
-  conv_compute_ref<float>(param);
-  for (int i = 0; i < output.dims().production(); i++) {
-    EXPECT_NEAR(output_data[i], output_ref_data[i], 1e-3);
-  }
-#endif
 }
 
 }  // namespace arm