aclnnMaxUnpool2d

Applicable Products

Function

• Operator function: the inverse operation of aclnnMaxPool in 2D. The H and W axis sizes of out are determined by outputSize, and the element values of self are filled into out according to the indices, with all other positions set to 0.

• Formula:

  $$out[n][c][indices[n][c][i]] = self[n][c][i]$$

  Here, out, indices, and self are obtained by reshaping the last two axes into a single axis, where i ∈ [0, H*W).

Function Prototype

Each operator is divided into a two-phase API. The "aclnnMaxUnpool2dGetWorkspaceSize" API must be called first to obtain the workspace size required for computation and the executor that encapsulates the operator computation flow. Then, the "aclnnMaxUnpool2d" API is called to perform the computation.

aclnnStatus aclnnMaxUnpool2dGetWorkspaceSize(
  const aclTensor*     self,
  const aclTensor*     indices,
  const aclIntArray*   outputSize,
  aclTensor*           out,
  uint64_t*            workspaceSize,
  aclOpExecutor**      executor)

aclnnStatus aclnnMaxUnpool2d(
  void*          workspace,
  uint64_t       workspaceSize,
  aclOpExecutor* executor,
  aclrtStream    stream)

aclnnMaxUnpool2dGetWorkspaceSize

• Parameters:

  Parameter	Input/Output	Description	Instruction	Data Type	Data Format	Dimension (Shape)	Non-Contiguous Tensor
  self (aclTensor*)	Input	self in the formula.	The data type is the same as that of out. The shape is the same as that of indices. When the dimension is 3, the dimensions represent C, H, and W in sequence. When the dimension is 4, the dimensions represent N, C, H, and W in sequence, where C stands for channel, N for batch, H for height, and W for width.	FLOAT, FLOAT16, INT16, INT32, INT64, INT8, UINT8, DOUBLE	ND, NCHW	3-4	√
  indices (aclTensor*)	Input	Indices in the formula, representing the index positions of the input self elements in the output result.	The shape is consistent with self. When the dimension is 3, each dimension represents C, H, and W in sequence. When the dimension is 4, each dimension represents N, C, H, and W in sequence, where C stands for Channel, N stands for Batch, H stands for Height, and W stands for Width.	INT64, INT32	ND, NCHW	3-4	√
  outputSize (aclIntArray*)	Input	Specifies the spatial size of the output in the H and W dimensions.	The size is 2, and the product of the two elements must be greater than or equal to the product of the size of self in the H and W dimensions.	-	-	-	-
  out (aclTensor*)	Output	out in the formula.	The data type is the same as that of self. When the dimension is 3, the dimensions represent C, H, and W in order. When the dimension is 4, the dimensions represent N, C, H, and W in order, where C stands for channel, N stands for Batch, H stands for Height, and W stands for Width.	FLOAT, FLOAT16, INT16, INT32, INT64, INT8, UINT8, DOUBLE	ND, NCHW	3-4	√
  workspaceSize (uint64_t*)	Output	Returns the workspace size to be allocated on the Device side.	-	-	-	-	-
  executor (aclOpExecutor**)	Output	Returns the operator executor, which contains the operator computation flow.	-	-	-	-	-

• Return value

  aclnnStatus: return code. For details, see aclnn Return Code.

  The first-phase API performs input parameter validation and returns an error in the following scenarios:

  Return Value	Error Code	Description
  ACLNN_ERR_PARAM_NULLPTR	161001	The passed self, indices, outputSize, or out is a null pointer.
  ACLNN_ERR_PARAM_INVALID	161002	The data types of self and indices are not within the supported range.
  		The data types of self and out are inconsistent.
  		The dimension of self is not 3D or 4D.
  		The shapes of self and indices are inconsistent.
  		The size of outputSize is not equal to 2.
  		The product of the two elements in outputSize is less than the product of the sizes of self in the H and W dimensions.
  		The sizes of out in the N and C dimensions are not exactly the same as those of self.
  		The sizes of out in the H and W dimensions are not equal to the two element values in outputSize.
  		The product of outputSize is less than the product of the H-axis and W-axis of self.

aclnnMaxUnpool2d

• Parameters:

  Parameter	Input/Output	Description
  workspace	Input	Address of the workspace memory applied on the Device side.
  workspaceSize	Input	Size of the workspace applied on the Device side, obtained through the first-phase API aclnnLogAddExpGetWorkspaceSize.
  executor	Input	Operator executor that contains the operator computation flow.
  stream	Input	Specifies the stream for task execution.

• Return value

  aclnnStatus: return code. For details, see aclnn Return Code.

Constraints

• Deterministic computation:

  • aclnnMaxUnpool2d defaults to a deterministic implementation.

Example

The following provides example code for reference only. For details about compilation and running, see Compile and Run Samples.

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_max_unpool2d.h"

#define CHECK_RET(cond, return_expr) \
  do {                               \
    if (!(cond)) {                   \
      return_expr;                   \
    }                                \
  } while (0)

#define LOG_PRINT(message, ...)     \
  do {                              \
    printf(message, ##__VA_ARGS__); \
  } while (0)

int64_t GetShapeSize(const std::vector<int64_t>& shape) {
  int64_t shapeSize = 1;
  for (auto i : shape) {
    shapeSize *= i;
  }
  return shapeSize;
}

int Init(int32_t deviceId, aclrtStream* stream) {
  // Boilerplate: resource initialization.
  auto ret = aclInit(nullptr);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclInit failed. ERROR: %d\n", ret); return ret);
  ret = aclrtSetDevice(deviceId);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSetDevice failed. ERROR: %d\n", ret); return ret);
  ret = aclrtCreateStream(stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtCreateStream failed. ERROR: %d\n", ret); return ret);
  return 0;
}

template <typename T>
int CreateAclTensor(const std::vector<T>& hostData, const std::vector<int64_t>& shape, void** deviceAddr,
                    aclDataType dataType, aclTensor** tensor) {
  auto size = GetShapeSize(shape) * sizeof(T);
  // Call aclrtMalloc to apply for device memory.
  auto ret = aclrtMalloc(deviceAddr, size, ACL_MEM_MALLOC_HUGE_FIRST);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtMalloc failed. ERROR: %d\n", ret); return ret);
  // Copy host-side data to device-side memory by calling aclrtMemcpy.
  ret = aclrtMemcpy(*deviceAddr, size, hostData.data(), size, ACL_MEMCPY_HOST_TO_DEVICE);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtMemcpy failed. ERROR: %d\n", ret); return ret);

  // Compute the strides of a contiguous tensor.
  std::vector<int64_t> strides(shape.size(), 1);
  for (int64_t i = shape.size() - 2; i >= 0; i--) {
    strides[i] = shape[i + 1] * strides[i + 1];
  }

  // Call the aclCreateTensor API to create an aclTensor.
  *tensor = aclCreateTensor(shape.data(), shape.size(), dataType, strides.data(), 0, aclFormat::ACL_FORMAT_ND,
                            shape.data(), shape.size(), *deviceAddr);
  return 0;
}

int main() {
  // 1. (Boilerplate) Initialize the device/stream. Refer to the ACL API manual.
  // Set deviceId based on the actual device.
  int32_t deviceId = 0;
  aclrtStream stream;
  auto ret = Init(deviceId, &stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);

  // 2. Construct the input and output. Customize the construction based on the API interface.
  std::vector<int64_t> selfShape = {1, 1, 2, 2};
  std::vector<int64_t> outShape = {1, 1, 4, 4};
  void* selfDeviceAddr = nullptr;
  void* indicesDeviceAddr = nullptr;
  void* outDeviceAddr = nullptr;
  aclTensor* self = nullptr;
  aclTensor* out = nullptr;
  aclTensor* indices = nullptr;
  std::vector<float> selfHostData = {1, 2, 3, 4};
  std::vector<float> outHostData = {0, 0, 0, 0.0, 0, 0, 0, 0,
                                    0, 0, 0, 0.0, 0, 0, 0, 0};
  std::vector<int64_t> indicesHostData = {3, 8, 11, 13};
  // Create the self aclTensor.
  ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  // Create the indices aclTensor.
  ret = CreateAclTensor(indicesHostData, selfShape, &indicesDeviceAddr, aclDataType::ACL_INT64, &indices);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  // Create the out aclTensor.
  ret = CreateAclTensor(outHostData, outShape, &outDeviceAddr, aclDataType::ACL_FLOAT, &out);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  std::vector<int64_t> arraySize = {4, 4};
  const aclIntArray *outputSize = aclCreateIntArray(arraySize.data(), arraySize.size());
  CHECK_RET(outputSize != nullptr, return ACL_ERROR_INTERNAL_ERROR);

  // 3. Call the CANN operator library API. Replace with the specific API name.
  uint64_t workspaceSize = 0;
  aclOpExecutor* executor;
  // Call the first-phase API of aclnnMaxUnpool2d.
  ret = aclnnMaxUnpool2dGetWorkspaceSize(self, indices, outputSize, out, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnMaxUnpool2dGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);
  // Allocate device memory based on the workspaceSize calculated by the first-phase API.
  void* workspaceAddr = nullptr;
  if (workspaceSize > 0) {
    ret = aclrtMalloc(&workspaceAddr, workspaceSize, ACL_MEM_MALLOC_HUGE_FIRST);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("allocate workspace failed. ERROR: %d\n", ret); return ret);
  }
  // Call the second-phase API of aclnnMaxUnpool2d.
  ret = aclnnMaxUnpool2d(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnMaxUnpool2d failed. ERROR: %d\n", ret); return ret);

  // 4. (Boilerplate) Synchronize and wait for task execution to complete.
  ret = aclrtSynchronizeStream(stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSynchronizeStream failed. ERROR: %d\n", ret); return ret);

  // 5. Obtain the output value and copy the result from the device memory to the host side. Modify this based on the specific API definition.
  auto size = GetShapeSize(outShape);
  std::vector<float> outData(size, 0);
  ret = aclrtMemcpy(outData.data(), outData.size() * sizeof(outData[0]), outDeviceAddr,
                    size * sizeof(outData[0]), ACL_MEMCPY_DEVICE_TO_HOST);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("copy result from device to host failed. ERROR: %d\n", ret); return ret);

  for (int64_t i = 0; i < size; i++) {
    LOG_PRINT("out[%ld] is: %f\n", i, outData[i]);
  }

  // 6. Release aclTensor and aclScalar. Modify based on the specific API definition.
  aclDestroyTensor(self);
  aclDestroyTensor(out);
  aclDestroyTensor(indices);
  aclDestroyIntArray(outputSize);

  // 7. Release device resources.
  aclrtFree(selfDeviceAddr);
  aclrtFree(indicesDeviceAddr);
  aclrtFree(outDeviceAddr);
  if (workspaceSize > 0) {
    aclrtFree(workspaceAddr);
  }
  aclrtDestroyStream(stream);
  aclrtResetDevice(deviceId);
  aclFinalize();
  return 0;
}