aclnnScale

Applicable Products

Function

• Description: Computes the result of scale and bias on the input.

• Formula:

  If bias is not provided, then

  $$y=x*scale$$

  If bias is input, then

  $$y=x*scale + bias$$

NOTE scale/bias supports broadcasting with X. The shape rules for scale/bias are as follows:

• When scaleFromBlob is True (axis is converted to a positive number, and numAxes being -1 indicates up to the last axis):

  scaleShape is xShape[axis:axis + numAxes]

  biasShape is xShape[axis:axis + numAxes]

  • When scaleFromBlob is False (axis is converted to positive, and numAxes being -1 indicates up to the foremost axis):

    scaleShape is xShape[axis:axis + rank(scaleShape)]

    biasShape is xShape[axis:axis + rank(scaleShape)]

  • Example:

    • scaleFromBlob = True:

      xShape = [a, b, c, d, e, f] axis = 3 numAxes = 2 --> scaleShape = [d, e]

      xShape = [a, b, c, d, e, f] axis = 3 numAxes = 3 --> scaleShape = [d, e, f]

      xShape = [a, b, c, d, e, f] axis = 3 numAxes = -1 --> scaleShape = [d, e, f]

    • scaleFromBlob = False:

      xShape = [a, b, c, d, e, f] axis = 3 rank(scaleShape) = 2 --> scaleShape = [d, e]

      xShape = [a, b, c, d, e, f] axis = 3 rank(scaleShape) = 3 --> scaleShape = [d, e, f]

      xShape = [a, b, c, d, e, f] axis = 3 rank(scaleShape) = 1 --> scaleShape = [d]

Function Prototype

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

aclnnStatus aclnnScaleGetWorkspaceSize(
  const aclTensor* x,
  const aclTensor* scale,
  const aclTensor* bias,
  int64_t          axis,
  int64_t          numAxes,
  bool             scaleFromBlob,
  aclTensor*       y,
  uint64_t*        workspaceSize,
  aclOpExecutor**  executor)

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

aclnnScaleGetWorkspaceSize

• Parameters:

  Name	Input/Output	Description	Usage	Data Type	Data Format	Dimension (Shape)	Non-contiguous Tensor
  x (aclTensor*)	Input	Input tensor,	Empty tensors are supported.	FLOAT, FLOAT16, BFLOAT16	ND	0-8	√
  scale (aclTensor*)	Input	Input tensor.	Empty tensors are supported. The data type must be the same as that of x. The shape meets the broadcast requirements. For details, see funtion.	FLOAT, FLOAT16, BFLOAT16	ND	0-8	√
  bias (aclTensor*)	Optional input	Input tensor,	Empty tensors are supported. If the value is not empty, the data type must be the same as that of the scale, and the shape must be the same as that of the scale.	FLOAT, FLOAT16, BFLOAT16	ND	0-8	√
  axis (int64_t)	Input	Start axis for scaling.	The value range is [–x_rank, x_rank) (x_rank indicates the shape dimension of x).	-	-	-	-
  numAxes (int64_t)	Input	Axis length for scaling.	The value is greater than or equal to –1. numAxes = –1 indicates that scaling is performed from the axis to the last axis.	-	-	-	-
  scaleFromBlob (bool)	Input	Type of the scaleFromBlob to be specified.	True: scale from blob, and numAxes + axis is used for scaling. False: scale from input scale, and scale the input scale length from `axis` and ignore the value of `numAxes`.	-	-	-	-
  y (aclTensor*)	Output	Output tensor.	Empty tensors are supported. The data type must be the same as that of x, and the shape dimension must be the same as that of x.	FLOAT, FLOAT16, BFLOAT16	ND	0-8	√
  workspaceSize (uint64_t*)	Output	Size of the workspace to be allocated on the device.	-	-	-	-	-
  executor (aclOpExecutor**)	Output	Operator executor, containing the operator computation process.	-	-	-	-	-

• Return Value:

  aclnnStatus: status code. For details, see aclnn Return Codes.

  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 x, scale, or y is a null pointer.
  ACLNN_ERR_PARAM_INVALID	161002	The data type of x is not within the supported range.
  		When bias is not empty, the data type of bias does not match that of scale.
  		The data type of scale does not match that of x.
  		The data type of y does not match that of x.
  		The shapes of x and y do not match.
  		When bias is not empty, the shape of bias does not match that of scale.
  		The shape dimension of x and scale is greater than 8.
  		The value of axis is not within the range [-x_rank, x_rank).
  		The value of numAxes is less than -1.
  		scaleFromBlob is True, numAxes equals 0, and the shape of scale is not [1].
  		The sum of axis converted to positive and numAxes exceeds x_rank.
  		The shape of scale does not match the expected shape (see Function for expected shape derivation).

aclnnScale

• Parameters:

  Parameter	Input/Output	Description
  workspace	Input	Workspace memory address applied for on the Device side.
  workspaceSize	Input	Workspace size applied for on the Device side, obtained from the first-phase API aclnnScaleGetWorkspaceSize.
  executor	Input	Operator executor that includes the operator computation flow.
  stream	Input	Specifies the stream for task execution.

• Return value

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

Constraints

• Deterministic computation:

  • aclnnScale 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_scale.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);
  // Call aclrtMemcpy to copy data from the host side to the device-side memory.
  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 aclCreateTensor 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. See the ACL API manual.
  // Enter the actual device ID for deviceId.
  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.
  std::vector<int64_t> selfShape = {4, 2};
  std::vector<int64_t> tensor1Shape = {4};
  std::vector<int64_t> tensor2Shape = {4};
  std::vector<int64_t> outShape = {4, 2};
  void* selfDeviceAddr = nullptr;
  void* tensor1DeviceAddr = nullptr;
  void* tensor2DeviceAddr = nullptr;
  void* outDeviceAddr = nullptr;
  aclTensor* self = nullptr;
  aclTensor* tensor1 = nullptr;
  aclTensor* tensor2 = nullptr;
  aclTensor* out = nullptr;

  std::vector<float> selfHostData = {0, 1, 2, 3, 4, 5, 6, 7};
  std::vector<float> tensor1HostData = {2, 2, 2, 2, 2, 2, 2, 2};
  std::vector<float> tensor2HostData = {2, 2, 2, 2, 2, 2, 2, 2};
  std::vector<float> outHostData = {0, 0, 0, 0, 0, 0, 0, 0};
  int64_t axis = 0;
  int64_t numAxes = 1;
  bool fromBlob = true;

  // Create self aclTensor.
  ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create tensor1 aclTensor.
  ret = CreateAclTensor(tensor1HostData, tensor1Shape, &tensor1DeviceAddr, aclDataType::ACL_FLOAT, &tensor1);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create tensor2 aclTensor.
  ret = CreateAclTensor(tensor2HostData, tensor2Shape, &tensor2DeviceAddr, aclDataType::ACL_FLOAT, &tensor2);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create 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;
  // Call the first-phase API of aclnnScale.
  ret = aclnnScaleGetWorkspaceSize(self, tensor1, tensor2, axis, numAxes, fromBlob,  out, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnScaleGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);
  // Apply for 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 aclnnScale.
  ret = aclnnScale(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnScale failed. ERROR: %d\n", ret); return ret);

  // 4. (Boilerplate) Synchronously wait for the 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 memory. Modify this 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 resultData from device to host failed. ERROR: %d\n", ret);
            return ret);
  for (int64_t i = 0; i < size; i++) {
    LOG_PRINT("resultData[%ld] is: %f\n", i, resultData[i]);
  }

  // 6. Release aclTensor and aclScalar. Modify based on the specific API definition.
  aclDestroyTensor(self);
  aclDestroyTensor(tensor1);
  aclDestroyTensor(tensor2);
  aclDestroyTensor(out);

  // 7. Release device resources. Modify based on the specific API definition.
  aclrtFree(selfDeviceAddr);
  aclrtFree(tensor1DeviceAddr);
  aclrtFree(tensor2DeviceAddr);
  aclrtFree(outDeviceAddr);
  if (workspaceSize > 0) {
    aclrtFree(workspaceAddr);
  }
  aclrtDestroyStream(stream);
  aclrtResetDevice(deviceId);
  aclFinalize();

  return 0;
}