!5850 [MSLITE] strided slice support neg strides

Merge pull request !5850 from zhaozhenlong/lite/issue/strided_slice

mindspore/lite/nnacl/strided_slice.c

@@ -22,14 +22,22 @@ void PadStridedSliceParameterTo4D(StridedSliceParameter *param) {
   int32_t ends[DIMENSION_4D];
   int32_t strides[DIMENSION_4D];
   int32_t input_shape[DIMENSION_4D];
-  for (int32_t i = 0; i < param->num_axes_; ++i) {
+  int32_t i;
+  for (i = 0; i < param->num_axes_; ++i) {
     begins[i] = param->begins_[i];
-    ends[i] = param->ends_[i];
+    ends[i] = MSMIN(param->ends_[i], param->in_shape_[i]);
     strides[i] = param->strides_[i];
     input_shape[i] = param->in_shape_[i];
   }
-  int32_t real_index = param->num_axes_ - 1;
-  for (int32_t i = DIMENSION_4D - 1; i >= 0; --i) {
+  for (i = param->num_axes_; i < param->in_shape_length_; ++i) {
+    input_shape[i] = param->in_shape_[i];
+    begins[i] = 0;
+    ends[i] = param->in_shape_[i];
+    strides[i] = 1;
+  }
+
+  int32_t real_index = param->in_shape_length_ - 1;
+  for (i = DIMENSION_4D - 1; i >= 0; --i) {
     if (real_index >= 0) {
       param->begins_[i] = begins[real_index];
       param->ends_[i] = ends[real_index];
@@ -43,8 +51,23 @@ void PadStridedSliceParameterTo4D(StridedSliceParameter *param) {
     }
   }
   param->num_axes_ = DIMENSION_4D;
+  param->in_shape_length_ = DIMENSION_4D;
 }
 
+void ChangeNegToPositive(StridedSliceParameter *param) {
+  int i;
+  for (i = 0; i < DIMENSION_4D; ++i) {
+    if (param->begins_[i] < 0) {
+      param->begins_[i] += param->in_shape_[i];
+    }
+    if (param->ends_[i] < 0) {
+      param->ends_[i] += param->in_shape_[i];
+    }
+  }
+}
+
+inline bool LoopContinue(int stride, int i, int end) { return stride > 0 ? i < end : i > end; }
+
 int DoStridedSlice(const void *in_data, void *out_data, StridedSliceParameter *param) {
   if (in_data == NULL || out_data == NULL || param == NULL) {
     return NNACL_NULL_PTR;
@@ -61,16 +84,18 @@ int DoStridedSlice(const void *in_data, void *out_data, StridedSliceParameter *p
   if (param->num_axes_ < DIMENSION_4D) {
     PadStridedSliceParameterTo4D(param);
   }
+  ChangeNegToPositive(param);
 
   size_t dim_offset[DIMENSION_4D - 1];
   dim_offset[2] = in_shape[3];
   dim_offset[1] = dim_offset[2] * in_shape[2];
   dim_offset[0] = dim_offset[1] * in_shape[1];
   size_t out_offset = 0;
-  for (int32_t dim0 = begins[0]; dim0 < ends[0]; dim0 += strides[0]) {
-    for (int32_t dim1 = begins[1]; dim1 < ends[1]; dim1 += strides[1]) {
-      for (int32_t dim2 = begins[2]; dim2 < ends[2]; dim2 += strides[2]) {
-        for (int32_t dim3 = begins[3]; dim3 < ends[3]; dim3 += strides[3]) {
+  int32_t dim0, dim1, dim2, dim3;
+  for (dim0 = begins[0]; LoopContinue(strides[0], dim0, ends[0]); dim0 += strides[0]) {
+    for (dim1 = begins[1]; LoopContinue(strides[1], dim1, ends[1]); dim1 += strides[1]) {
+      for (dim2 = begins[2]; LoopContinue(strides[2], dim2, ends[2]); dim2 += strides[2]) {
+        for (dim3 = begins[3]; LoopContinue(strides[3], dim3, ends[3]); dim3 += strides[3]) {
           int32_t in_offset = dim0 * dim_offset[0] + dim1 * dim_offset[1] + dim2 * dim_offset[2] + dim3;
           if (param->data_type == kDataTypeFloat) {
             *((float *)out_data + out_offset) = *((float *)in_data + in_offset);

mindspore/lite/nnacl/strided_slice.h

@@ -25,6 +25,7 @@ typedef struct StridedSliceParameter {
   int strides_[8];
   int isScale;
   int num_axes_;
+  int in_shape_length_;
   int in_shape_[8];
   LiteDataType data_type;
 } StridedSliceParameter;

mindspore/lite/src/runtime/kernel/arm/base/strided_slice.cc

@@ -44,6 +44,11 @@ int StridedSliceCPUKernel::ReSize() {
   MS_ASSERT(input);
   MS_ASSERT(parameter);
   parameter->data_type = input->data_type() == kNumberTypeInt8 ? kDataTypeInt8 : kDataTypeFloat;
+  auto input_shape = input->shape();
+  for (size_t i = 0; i < input_shape.size(); ++i) {
+    parameter->in_shape_[i] = input_shape[i];
+  }
+  parameter->in_shape_length_ = static_cast<int>(input_shape.size());
   return RET_OK;
 }
 

mindspore/lite/test/ut/src/runtime/kernel/arm/fp32/strided_slice_fp32_tests.cc

@@ -0,0 +1,434 @@
+/**
+ * Copyright 2020 Huawei Technologies Co., Ltd
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include <iostream>
+#include <memory>
+#include "utils/log_adapter.h"
+#include "common/common_test.h"
+#include "mindspore/lite/src/common/utils.h"
+#include "nnacl/strided_slice.h"
+#include "mindspore/lite/src/kernel_registry.h"
+#include "mindspore/lite/src/lite_kernel.h"
+
+namespace mindspore {
+class TestStridedSliceFp32 : public mindspore::CommonTest {
+ public:
+  TestStridedSliceFp32() {}
+};
+
+void InitStridedSliceParam(StridedSliceParameter *param, const std::vector<int> &in_shape,
+                           const std::vector<int> &begins, const std::vector<int> &ends,
+                           const std::vector<int> &strides, const int begins_length, const int in_shape_length) {
+  for (auto i = 0; i < begins_length; i++) {
+    param->in_shape_[i] = in_shape[i];
+    param->begins_[i] = begins[i];
+    param->ends_[i] = ends[i];
+    param->strides_[i] = strides[i];
+  }
+  param->num_axes_ = begins_length;
+  param->in_shape_length_ = in_shape_length;
+}
+
+TEST_F(TestStridedSliceFp32, StridedSlice1) {
+  // prepare stage
+  auto strided_slice_param = new StridedSliceParameter();
+  std::vector<int> in_shape{1, 2, 4};
+  std::vector<int> begins{0, 0, 0};
+  std::vector<int> ends{1, 2, 4};
+  std::vector<int> strides{1, 2, 2};
+  int length = 3;
+  int in_shape_length = 3;
+  InitStridedSliceParam(strided_slice_param, in_shape, begins, ends, strides, length, in_shape_length);
+
+  float input_data[8] = {0.2390374, 0.92039955, 0.05051243, 0.49574447, 0.8355223, 0.02647042, 0.08811307, 0.4566604};
+
+  float correct[2] = {0.2390374, 0.05051243};
+
+  float output_data[2];
+
+  // runtime part
+  printf("Calculating runtime cost...\n");
+  uint64_t time_avg = 0;
+
+  // warm up loop
+  for (int i = 0; i < 3; i++) {
+    DoStridedSlice(input_data, output_data, strided_slice_param);
+  }
+
+  int loop_count = 100;
+  auto time_start = mindspore::lite::GetTimeUs();
+  for (int i = 0; i < loop_count; i++) {
+    DoStridedSlice(input_data, output_data, strided_slice_param);
+  }
+  auto time_end = mindspore::lite::GetTimeUs();
+  auto cost = time_end - time_start;
+  time_avg = cost / loop_count;
+  printf("single thread running time : %f ms\n", time_avg / 1000.0f);
+
+  printf("==================output data=================\n");
+  std::cout << output_data[0] << " , " << output_data[1];
+  std::cout << std::endl;
+  printf("==================corret data=================\n");
+  std::cout << correct[0] << " , " << correct[1];
+  std::cout << std::endl;
+
+  CompareOutputData(output_data, correct, 2, 0.00001);
+
+  delete strided_slice_param;
+  MS_LOG(INFO) << "Teststrided_sliceFp32 passed";
+}
+
+TEST_F(TestStridedSliceFp32, StridedSlice2) {
+  // prepare stage
+  auto strided_slice_param = new StridedSliceParameter();
+  std::vector<int> in_shape{1, 3, 3};
+  std::vector<int> begins{0, 0, 0};
+  std::vector<int> ends{2, 4, 4};
+  std::vector<int> strides{1, 1, 1};
+  int length = 3;
+  int in_shape_length = 3;
+  InitStridedSliceParam(strided_slice_param, in_shape, begins, ends, strides, length, in_shape_length);
+
+  float input_data[9] = {0.0f, 1.0f, 2.0f, 3.0f, 4.0f, 5.0f, 6.0f, 7.0f, 8.0f};
+
+  float correct[9] = {0.0f, 1.0f, 2.0f, 3.0f, 4.0f, 5.0f, 6.0f, 7.0f, 8.0f};
+
+  float output_data[9];
+
+  // runtime part
+  DoStridedSlice(input_data, output_data, strided_slice_param);
+
+  CompareOutputData(output_data, correct, 9, 0.00001);
+
+  delete strided_slice_param;
+}
+
+TEST_F(TestStridedSliceFp32, StridedSlice3) {
+  // prepare stage
+  auto strided_slice_param = new StridedSliceParameter();
+  std::vector<int> in_shape{1, 2, 4};
+  std::vector<int> begins{0, 0, 0};
+  std::vector<int> ends{1, 2, 4};
+  std::vector<int> strides{1, 2, 2};
+  int length = 3;
+  int in_shape_length = 3;
+  InitStridedSliceParam(strided_slice_param, in_shape, begins, ends, strides, length, in_shape_length);
+
+  // int input_shape[3] = {1, 2, 4};
+  std::vector<int> input_shape = {1, 2, 4};
+  float input_data[8] = {0.2390374, 0.92039955, 0.05051243, 0.49574447, 0.8355223, 0.02647042, 0.08811307, 0.4566604};
+
+  float correct[2] = {0.2390374, 0.05051243};
+  float output_data[2];
+  // int out_shape[3] = {1, 1, 2};
+  std::vector<int> output_shape = {1, 1, 2};
+
+  lite::Tensor input_tensor;
+  input_tensor.SetData(input_data);
+  input_tensor.set_shape(input_shape);
+  std::vector<lite::Tensor *> inputs_tensor(1);
+  inputs_tensor[0] = &input_tensor;
+
+  std::vector<lite::Tensor *> outputs_tensor;
+  lite::Tensor output_tensor;
+  outputs_tensor.push_back(&output_tensor);
+  output_tensor.SetData(output_data);
+  output_tensor.set_data_type(input_tensor.data_type());
+  output_tensor.set_shape(output_shape);
+
+  lite::Context *ctx = new lite::Context();
+  ctx->thread_num_ = 2;
+  strided_slice_param->op_parameter_.type_ = schema::PrimitiveType_StridedSlice;
+  kernel::KernelKey desc = {kernel::KERNEL_ARCH::kCPU, kNumberTypeFloat32, schema::PrimitiveType_StridedSlice};
+  auto creator = lite::KernelRegistry::GetInstance()->GetCreator(desc);
+  ASSERT_NE(creator, nullptr);
+  kernel::LiteKernel *kernel =
+    creator(inputs_tensor, outputs_tensor, reinterpret_cast<OpParameter *>(strided_slice_param), ctx, desc, nullptr);
+  ASSERT_NE(kernel, nullptr);
+  kernel->Run();
+  delete ctx;
+
+  CompareOutputData(output_data, correct, 2, 0.000001);
+  input_tensor.SetData(nullptr);
+  output_tensor.SetData(nullptr);
+}
+
+TEST_F(TestStridedSliceFp32, StridedSlice4) {
+  // prepare stage
+  auto strided_slice_param = new StridedSliceParameter();
+  std::vector<int> in_shape{5, 5};
+  std::vector<int> begins{-5, -3};
+  std::vector<int> ends{-1, -1};
+  std::vector<int> strides{2, 1};
+  int length = 2;
+  int in_shape_length = 2;
+  InitStridedSliceParam(strided_slice_param, in_shape, begins, ends, strides, length, in_shape_length);
+
+  std::vector<int> input_shape = {5, 5};
+  float input_data[25] = {0.0,  1.0,  2.0,  3.0,  4.0,  5.0,  6.0,  7.0,  8.0,  9.0,  10.0, 11.0, 12.0,
+                          13.0, 14.0, 15.0, 16.0, 17.0, 18.0, 19.0, 20.0, 21.0, 22.0, 23.0, 24.0};
+
+  float correct[4] = {2.0, 3.0, 12.0, 13.0};
+  float output_data[4];
+  std::vector<int> output_shape = {2, 2};
+
+  lite::Tensor input_tensor;
+  input_tensor.SetData(input_data);
+  input_tensor.set_shape(input_shape);
+  std::vector<lite::Tensor *> inputs_tensor(1);
+  inputs_tensor[0] = &input_tensor;
+
+  std::vector<lite::Tensor *> outputs_tensor;
+  lite::Tensor output_tensor;
+  outputs_tensor.push_back(&output_tensor);
+  output_tensor.SetData(output_data);
+  output_tensor.set_data_type(input_tensor.data_type());
+  output_tensor.set_shape(output_shape);
+
+  lite::Context *ctx = new lite::Context();
+  ctx->thread_num_ = 2;
+  strided_slice_param->op_parameter_.type_ = schema::PrimitiveType_StridedSlice;
+  kernel::KernelKey desc = {kernel::KERNEL_ARCH::kCPU, kNumberTypeFloat32, schema::PrimitiveType_StridedSlice};
+  auto creator = lite::KernelRegistry::GetInstance()->GetCreator(desc);
+  ASSERT_NE(creator, nullptr);
+  kernel::LiteKernel *kernel =
+    creator(inputs_tensor, outputs_tensor, reinterpret_cast<OpParameter *>(strided_slice_param), ctx, desc, nullptr);
+  ASSERT_NE(kernel, nullptr);
+  kernel->Run();
+  delete ctx;
+
+  CompareOutputData(output_data, correct, 4, 0.000001);
+  input_tensor.SetData(nullptr);
+  output_tensor.SetData(nullptr);
+}
+
+TEST_F(TestStridedSliceFp32, StridedSlice5) {
+  // prepare stage
+  auto strided_slice_param = new StridedSliceParameter();
+  std::vector<int> in_shape{5, 5, 5};
+  std::vector<int> begins{1, -1, 1};
+  std::vector<int> ends{4, -3, 3};
+  std::vector<int> strides{1, -1, 1};
+  int length = 3;
+  int in_shape_length = 3;
+  InitStridedSliceParam(strided_slice_param, in_shape, begins, ends, strides, length, in_shape_length);
+
+  std::vector<int> input_shape = {5, 5, 5};
+  float input_data[125] = {
+    0.0,   1.0,   2.0,   3.0,   4.0,   5.0,   6.0,   7.0,   8.0,   9.0,   10.0,  11.0,  12.0,  13.0,  14.0,  15.0,
+    16.0,  17.0,  18.0,  19.0,  20.0,  21.0,  22.0,  23.0,  24.0,  25.0,  26.0,  27.0,  28.0,  29.0,  30.0,  31.0,
+    32.0,  33.0,  34.0,  35.0,  36.0,  37.0,  38.0,  39.0,  40.0,  41.0,  42.0,  43.0,  44.0,  45.0,  46.0,  47.0,
+    48.0,  49.0,  50.0,  51.0,  52.0,  53.0,  54.0,  55.0,  56.0,  57.0,  58.0,  59.0,  60.0,  61.0,  62.0,  63.0,
+    64.0,  65.0,  66.0,  67.0,  68.0,  69.0,  70.0,  71.0,  72.0,  73.0,  74.0,  75.0,  76.0,  77.0,  78.0,  79.0,
+    80.0,  81.0,  82.0,  83.0,  84.0,  85.0,  86.0,  87.0,  88.0,  89.0,  90.0,  91.0,  92.0,  93.0,  94.0,  95.0,
+    96.0,  97.0,  98.0,  99.0,  100.0, 101.0, 102.0, 103.0, 104.0, 105.0, 106.0, 107.0, 108.0, 109.0, 110.0, 111.0,
+    112.0, 113.0, 114.0, 115.0, 116.0, 117.0, 118.0, 119.0, 120.0, 121.0, 122.0, 123.0, 124.0};
+
+  float correct[12] = {46.0, 47.0, 41.0, 42.0, 71.0, 72.0, 66.0, 67.0, 96.0, 97.0, 91.0, 92.0};
+  float output_data[12];
+  std::vector<int> output_shape = {3, 2, 2};
+
+  lite::Tensor input_tensor;
+  input_tensor.SetData(input_data);
+  input_tensor.set_shape(input_shape);
+  std::vector<lite::Tensor *> inputs_tensor(1);
+  inputs_tensor[0] = &input_tensor;
+
+  std::vector<lite::Tensor *> outputs_tensor;
+  lite::Tensor output_tensor;
+  outputs_tensor.push_back(&output_tensor);
+  output_tensor.SetData(output_data);
+  output_tensor.set_data_type(input_tensor.data_type());
+  output_tensor.set_shape(output_shape);
+
+  lite::Context *ctx = new lite::Context();
+  ctx->thread_num_ = 2;
+  strided_slice_param->op_parameter_.type_ = schema::PrimitiveType_StridedSlice;
+  kernel::KernelKey desc = {kernel::KERNEL_ARCH::kCPU, kNumberTypeFloat32, schema::PrimitiveType_StridedSlice};
+  auto creator = lite::KernelRegistry::GetInstance()->GetCreator(desc);
+  ASSERT_NE(creator, nullptr);
+  kernel::LiteKernel *kernel =
+    creator(inputs_tensor, outputs_tensor, reinterpret_cast<OpParameter *>(strided_slice_param), ctx, desc, nullptr);
+  ASSERT_NE(kernel, nullptr);
+  kernel->Run();
+  delete ctx;
+
+  CompareOutputData(output_data, correct, 12, 0.000001);
+  input_tensor.SetData(nullptr);
+  output_tensor.SetData(nullptr);
+}
+
+TEST_F(TestStridedSliceFp32, StridedSlice6) {
+  // prepare stage
+  auto strided_slice_param = new StridedSliceParameter();
+  std::vector<int> in_shape{5, 5, 5};
+  std::vector<int> begins{-5, -3, 1};
+  std::vector<int> ends{-1, -1, 3};
+  std::vector<int> strides{2, 1, 1};
+  int length = 3;
+  int in_shape_length = 3;
+  InitStridedSliceParam(strided_slice_param, in_shape, begins, ends, strides, length, in_shape_length);
+
+  std::vector<int> input_shape = {5, 5, 5};
+  float input_data[125] = {
+    0.0,   1.0,   2.0,   3.0,   4.0,   5.0,   6.0,   7.0,   8.0,   9.0,   10.0,  11.0,  12.0,  13.0,  14.0,  15.0,
+    16.0,  17.0,  18.0,  19.0,  20.0,  21.0,  22.0,  23.0,  24.0,  25.0,  26.0,  27.0,  28.0,  29.0,  30.0,  31.0,
+    32.0,  33.0,  34.0,  35.0,  36.0,  37.0,  38.0,  39.0,  40.0,  41.0,  42.0,  43.0,  44.0,  45.0,  46.0,  47.0,
+    48.0,  49.0,  50.0,  51.0,  52.0,  53.0,  54.0,  55.0,  56.0,  57.0,  58.0,  59.0,  60.0,  61.0,  62.0,  63.0,
+    64.0,  65.0,  66.0,  67.0,  68.0,  69.0,  70.0,  71.0,  72.0,  73.0,  74.0,  75.0,  76.0,  77.0,  78.0,  79.0,
+    80.0,  81.0,  82.0,  83.0,  84.0,  85.0,  86.0,  87.0,  88.0,  89.0,  90.0,  91.0,  92.0,  93.0,  94.0,  95.0,
+    96.0,  97.0,  98.0,  99.0,  100.0, 101.0, 102.0, 103.0, 104.0, 105.0, 106.0, 107.0, 108.0, 109.0, 110.0, 111.0,
+    112.0, 113.0, 114.0, 115.0, 116.0, 117.0, 118.0, 119.0, 120.0, 121.0, 122.0, 123.0, 124.0};
+
+  float correct[8] = {11.0, 12.0, 16.0, 17.0, 61.0, 62.0, 66.0, 67.0};
+  float output_data[8];
+  std::vector<int> output_shape = {2, 2, 2};
+
+  lite::Tensor input_tensor;
+  input_tensor.SetData(input_data);
+  input_tensor.set_shape(input_shape);
+  std::vector<lite::Tensor *> inputs_tensor(1);
+  inputs_tensor[0] = &input_tensor;
+
+  std::vector<lite::Tensor *> outputs_tensor;
+  lite::Tensor output_tensor;
+  outputs_tensor.push_back(&output_tensor);
+  output_tensor.SetData(output_data);
+  output_tensor.set_data_type(input_tensor.data_type());
+  output_tensor.set_shape(output_shape);
+
+  lite::Context *ctx = new lite::Context();
+  ctx->thread_num_ = 2;
+  strided_slice_param->op_parameter_.type_ = schema::PrimitiveType_StridedSlice;
+  kernel::KernelKey desc = {kernel::KERNEL_ARCH::kCPU, kNumberTypeFloat32, schema::PrimitiveType_StridedSlice};
+  auto creator = lite::KernelRegistry::GetInstance()->GetCreator(desc);
+  ASSERT_NE(creator, nullptr);
+  kernel::LiteKernel *kernel =
+    creator(inputs_tensor, outputs_tensor, reinterpret_cast<OpParameter *>(strided_slice_param), ctx, desc, nullptr);
+  ASSERT_NE(kernel, nullptr);
+  kernel->Run();
+  delete ctx;
+
+  CompareOutputData(output_data, correct, 8, 0.000001);
+  input_tensor.SetData(nullptr);
+  output_tensor.SetData(nullptr);
+}
+
+TEST_F(TestStridedSliceFp32, StridedSlice7) {
+  // prepare stage
+  auto strided_slice_param = new StridedSliceParameter();
+  std::vector<int> in_shape{2, 3};
+  std::vector<int> begins{-1, 1};
+  std::vector<int> ends{2, 2};
+  std::vector<int> strides{1, 1};
+  int length = 2;
+  int in_shape_length = 2;
+  InitStridedSliceParam(strided_slice_param, in_shape, begins, ends, strides, length, in_shape_length);
+
+  std::vector<int> input_shape = {2, 3};
+  float input_data[6] = {0.0, 1.0, 2.0, 3.0, 4.0, 5.0};
+
+  float correct[1] = {4.0};
+  float output_data[1];
+  std::vector<int> output_shape = {1, 1};
+
+  lite::Tensor input_tensor;
+  input_tensor.SetData(input_data);
+  input_tensor.set_shape(input_shape);
+  std::vector<lite::Tensor *> inputs_tensor(1);
+  inputs_tensor[0] = &input_tensor;
+
+  std::vector<lite::Tensor *> outputs_tensor;
+  lite::Tensor output_tensor;
+  outputs_tensor.push_back(&output_tensor);
+  output_tensor.SetData(output_data);
+  output_tensor.set_data_type(input_tensor.data_type());
+  output_tensor.set_shape(output_shape);
+
+  lite::Context *ctx = new lite::Context();
+  ctx->thread_num_ = 2;
+  strided_slice_param->op_parameter_.type_ = schema::PrimitiveType_StridedSlice;
+  kernel::KernelKey desc = {kernel::KERNEL_ARCH::kCPU, kNumberTypeFloat32, schema::PrimitiveType_StridedSlice};
+  auto creator = lite::KernelRegistry::GetInstance()->GetCreator(desc);
+  ASSERT_NE(creator, nullptr);
+  kernel::LiteKernel *kernel =
+    creator(inputs_tensor, outputs_tensor, reinterpret_cast<OpParameter *>(strided_slice_param), ctx, desc, nullptr);
+  ASSERT_NE(kernel, nullptr);
+  kernel->Run();
+  delete ctx;
+
+  CompareOutputData(output_data, correct, 1, 0.000001);
+  input_tensor.SetData(nullptr);
+  output_tensor.SetData(nullptr);
+}
+
+TEST_F(TestStridedSliceFp32, StridedSlice8) {
+  // prepare stage
+  auto strided_slice_param = new StridedSliceParameter();
+  std::vector<int> in_shape{5, 5, 5};
+  std::vector<int> begins{-2, -3};
+  std::vector<int> ends{-1, -2};
+  std::vector<int> strides{1, 1};
+  int length = 2;
+  int in_shape_length = 3;
+  InitStridedSliceParam(strided_slice_param, in_shape, begins, ends, strides, length, in_shape_length);
+
+  std::vector<int> input_shape = {5, 5, 5};
+  float input_data[125] = {
+    0.0,   1.0,   2.0,   3.0,   4.0,   5.0,   6.0,   7.0,   8.0,   9.0,   10.0,  11.0,  12.0,  13.0,  14.0,  15.0,
+    16.0,  17.0,  18.0,  19.0,  20.0,  21.0,  22.0,  23.0,  24.0,  25.0,  26.0,  27.0,  28.0,  29.0,  30.0,  31.0,
+    32.0,  33.0,  34.0,  35.0,  36.0,  37.0,  38.0,  39.0,  40.0,  41.0,  42.0,  43.0,  44.0,  45.0,  46.0,  47.0,
+    48.0,  49.0,  50.0,  51.0,  52.0,  53.0,  54.0,  55.0,  56.0,  57.0,  58.0,  59.0,  60.0,  61.0,  62.0,  63.0,
+    64.0,  65.0,  66.0,  67.0,  68.0,  69.0,  70.0,  71.0,  72.0,  73.0,  74.0,  75.0,  76.0,  77.0,  78.0,  79.0,
+    80.0,  81.0,  82.0,  83.0,  84.0,  85.0,  86.0,  87.0,  88.0,  89.0,  90.0,  91.0,  92.0,  93.0,  94.0,  95.0,
+    96.0,  97.0,  98.0,  99.0,  100.0, 101.0, 102.0, 103.0, 104.0, 105.0, 106.0, 107.0, 108.0, 109.0, 110.0, 111.0,
+    112.0, 113.0, 114.0, 115.0, 116.0, 117.0, 118.0, 119.0, 120.0, 121.0, 122.0, 123.0, 124.0};
+
+  float correct[5] = {85.0, 86.0, 87.0, 88.0, 89.0};
+  float output_data[5];
+  std::vector<int> output_shape = {1, 1, 5};
+
+  lite::Tensor input_tensor;
+  input_tensor.SetData(input_data);
+  input_tensor.set_shape(input_shape);
+  std::vector<lite::Tensor *> inputs_tensor(1);
+  inputs_tensor[0] = &input_tensor;
+
+  std::vector<lite::Tensor *> outputs_tensor;
+  lite::Tensor output_tensor;
+  outputs_tensor.push_back(&output_tensor);
+  output_tensor.SetData(output_data);
+  output_tensor.set_data_type(input_tensor.data_type());
+  output_tensor.set_shape(output_shape);
+
+  lite::Context *ctx = new lite::Context();
+  ctx->thread_num_ = 2;
+  strided_slice_param->op_parameter_.type_ = schema::PrimitiveType_StridedSlice;
+  kernel::KernelKey desc = {kernel::KERNEL_ARCH::kCPU, kNumberTypeFloat32, schema::PrimitiveType_StridedSlice};
+  auto creator = lite::KernelRegistry::GetInstance()->GetCreator(desc);
+  ASSERT_NE(creator, nullptr);
+  kernel::LiteKernel *kernel =
+    creator(inputs_tensor, outputs_tensor, reinterpret_cast<OpParameter *>(strided_slice_param), ctx, desc, nullptr);
+  ASSERT_NE(kernel, nullptr);
+  kernel->Run();
+  delete ctx;
+
+  CompareOutputData(output_data, correct, 5, 0.000001);
+  input_tensor.SetData(nullptr);
+  output_tensor.SetData(nullptr);
+}
+
+}  // namespace mindspore