aclnnModulate

Applicable Products

Function

• Description: Performs adaptive scaling and shifting on features.

• Formula:

  $$out = self \times (1 + scaleOptional) + shiftOptional$$

  Where $self \in R ^ {B \times L \times D}$, $scaleOptional \in R ^ {B \times 1 \times D}$, $shiftOptional \in R ^ {B \times 1 \times D}$ (or $scaleOptional \in R ^ {B \times D}$, $shiftOptional \in R ^ {B \times D}$).

Function Prototype

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

aclnnStatus aclnnModulateGetWorkspaceSize(
    const aclTensor* self,
    const aclTensor* scaleOptional,
    const aclTensor* shiftOptional,
    aclTensor*       out,
    uint64_t*        workspaceSize,
    aclOpExecutor**  executor)

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

aclnnModulateGetWorkspaceSize

• Parameters:

  Parameter	Input/Output	Description	Instruction	Data Type	Data Format	Dimension (Shape)	Non-contiguous Tensor
  self	Input	Represents the input feature tensor, which is self in the formula.	-	FLOAT, FLOAT16, BFLOAT16	ND	(batch_size, seq_len, hidden_dim)	√
  scaleOptional	Input	Optional parameter that specifies the scaling coefficient, i.e., scaleOptional in the formula.	The data type must be the same as self.	FLOAT, FLOAT16, BFLOAT16	ND	(batch_size, 1, hidden_dim) or (batch_size, hidden_dim)	√
  shiftOptional	Input	Optional parameter that represents the shift coefficient, i.e., shiftOptional in the formula.	The data type must be the same as self.	FLOAT, FLOAT16, BFLOAT16	ND	(batch_size, 1, hidden_dim) or (batch_size, hidden_dim)	√
  out	Output	Represents the output tensor after scaling, i.e., out in the formula.	Must have the same data type and shape as self.	FLOAT, FLOAT16, BFLOAT16	ND	-	√
  workspaceSize	Output	Returns the workspace size to be allocated on the Device side.	-	-	-	-	-
  executor	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	Indicates that self or out is a null pointer.
  ACLNN_ERR_PARAM_INVALID	161002	The data types of self, scaleOptional, and shiftOptional are not within the supported range.
  		The shapes of self, scaleOptional, and shiftOptional do not satisfy the constraints.
  		self is an empty tensor while scaleOptional or shiftOptional is not an empty tensor.

aclnnModulate

• Parameters:

  Parameter	Input/Output	Description
  workspace	Input	Address of the workspace memory applied for on the device side.
  workspaceSize	Input	Size of the workspace applied for on the device side, obtained by the first-phase API aclnnMultiScaleDeformableAttentionGradGetWorkspaceSize.
  executor	Input	Operator executor that contains the operator computation flow.
  stream	Input	Specifies the stream on which the task is executed.

• Return value:

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

Constraints

• Deterministic computation:

  • aclnnModulate 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_modulate.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: initialize resources.
  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-side 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 the host-side data to the 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];
  }

  // Create an aclTensor by calling the aclCreateTensor API.
  *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. See the acl API manual.
  // Set deviceId based on your 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. Custom construction is required based on the API interface.
  std::vector<int64_t> selfShape = {2, 1, 1};
  std::vector<int64_t> scaleOptionalShape = {2, 1, 1};
  std::vector<int64_t> shiftOptionalShape = {2, 1, 1};
  std::vector<int64_t> outShape = {2, 1, 1};
  void* selfDeviceAddr = nullptr;
  void* scaleOptionalDeviceAddr = nullptr;
  void* shiftOptionalDeviceAddr = nullptr;
  void* outDeviceAddr = nullptr;
  aclTensor* self = nullptr;
  aclTensor* scaleOptional = nullptr;
  aclTensor* shiftOptional = nullptr;
  aclTensor* out = nullptr;
  std::vector<float> selfHostData{10, 20};
  std::vector<float> scaleOptionalHostData{20, 30};
  std::vector<float> shiftOptionalHostData{30, 40};
  std::vector<float> outHostData{0, 0};
  // Create the self aclTensor.
  ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create the scaleOptional aclTensor.
  ret = CreateAclTensor(scaleOptionalHostData, scaleOptionalShape, &scaleOptionalDeviceAddr, aclDataType::ACL_FLOAT, &scaleOptional);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create the shiftOptional aclTensor.
  ret = CreateAclTensor(shiftOptionalHostData, shiftOptionalShape, &shiftOptionalDeviceAddr, aclDataType::ACL_FLOAT, &shiftOptional);
  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);

  // 3. Call the CANN operator library API. Replace with the specific API name.
  uint64_t workspaceSize = 0;
  aclOpExecutor* executor = nullptr;
  // Call the first-phase API of aclnnModulate.
  ret = aclnnModulateGetWorkspaceSize(self, scaleOptional, shiftOptional, out, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnModulateGetWorkspaceSize 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 aclnnModulate.
  ret = aclnnModulate(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnModulate 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 side memory to the host side. Modify based on the specific API definition.
  auto size = GetShapeSize(outShape);
  std::vector<float> resultData(size, 0);
  ret = aclrtMemcpy(resultData.data(), resultData.size() * sizeof(resultData[0]), outDeviceAddr,
                    size * sizeof(resultData[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("result[%ld] is: %f\n", i, resultData[i]);
  }

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

  // 7. Release device resources. Modify based on the specific API definition.
  aclrtFree(selfDeviceAddr);
  aclrtFree(scaleOptionalDeviceAddr);
  aclrtFree(shiftOptionalDeviceAddr);
  aclrtFree(outDeviceAddr);
  if (workspaceSize > 0) {
    aclrtFree(workspaceAddr);
  }
  aclrtDestroyStream(stream);
  aclrtResetDevice(deviceId);
  aclFinalize();
  return 0;
}