aclnnSoftplusBackward

Applicable Products

Product	Supported
Ascend 950PR/Ascend 950DT	×
Atlas A3 training products/Atlas A3 inference products	√
Atlas A2 training products/Atlas A2 inference products	√
Atlas 200I/500 A2 inference products	×
Atlas inference products	×
Atlas training products	√

Function

Description: backward propagation of aclnnSoftplus.
Formula:
$gradInput = gradOutput \cdot \begin{cases} \frac{1}{1+e^{(-\beta * self)}}, ~~~~~~~\beta * self \le threshold \\ 1, ~~~~~~~~~~~~~~~~~~~~~\beta * self > threshold \end{cases}$

Function Prototype

Each operator is divided into a two-phase API. First call the "aclnnSoftplusBackwardGetWorkspaceSize" API to obtain the workspace size required for computation and the executor that includes computation of the operator computation process, and then call the "aclnnSoftplusBackward" API to perform the computation.

aclnnStatus aclnnSoftplusBackwardGetWorkspaceSize(
  const aclTensor* gradOutput,
  const aclTensor* self,
  const aclScalar* beta,
  const aclScalar* threshold,
  aclTensor*       gradInput,
  uint64_t*        workspaceSize,
  aclOpExecutor**  executor)

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

aclnnSoftplusBackwardGetWorkspaceSize

Parameters:

Parameter	Input/Output	Description	Instruction	Data Type	Data Format	Dimension (Shape)	Non-contiguous Tensor
gradOutput (aclTensor*)	Input	Input gradient. gradOutput in the formula.	Supports null tensor. The shape needs to be consistent with self and gradInput. The shape needs to satisfy the broadcast relationship with self, and the broadcast shape must be equal to the shape of self.	FLOAT, FLOAT16, DOUBLE, BFLOAT16	ND	0-8	√
self (aclTensor*)	Input	Input data. self in the formula.	Supports null tensor. Forward input value of Softplus. The shape needs to be consistent with gradOutput and gradInput.	INT8, INT16, INT32, INT64, UINT8, BOOL, FLOAT16, FLOAT, DOUBLE, BFLOAT16	ND	0-8	√
beta (aclScalar*)	Input	beta in the formula, which indicates the degree of approximation to ReLU.	The data type must satisfy the data type deduction rules (see deduction relationship). Supports types convertible to FLOAT.	-	-	-	-
threshold (aclScalar*)	Input	threshold in the formula represents a threshold value; when exceeded, the function reverts to a linear function.	The data type must satisfy the data type deduction rules (see deduction relationship). Types convertible to FLOAT are supported.	-	-	-	-
gradInput (aclTensor*)	Output	gradInput in the formula.	Supports null tensor. The shape needs to be consistent with gradOutput and self. The shape and data type of gradInput are the same as those of self.	FLOAT, FLOAT16, DOUBLE, BFLOAT16	ND	0-8	√
workspaceSize (uint64_t*)	Output	Returns the workspace size to be applied on the device side.	-	-	-	-	-
executor (aclOpExecutor**)	Output	Returns the op executor, which includes the computation flow of the operator.	-	-	-	-	-

Atlas training products: The data type supports FLOAT, FLOAT16, and DOUBLE.

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 Code	Error Code	Description
ACLNN_ERR_PARAM_NULLPTR	161001	When the input gradOutput, self, beta, threshold, or gradInput is a null pointer.
ACLNN_ERR_PARAM_INVALID	161002	The data types of gradOutput, self, and gradInput are not within the supported range.
		The shapes of gradOutput, self, and gradInput are different.

aclnnSoftplusBackward

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 aclnnSoftplusBackwardGetWorkspaceSize.
executor	Input	Operator executor, includes the operator computation process.
stream	Input	Specifies the stream for task execution.

Return value

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

Constraints

Deterministic computation:
- aclnnSoftplusBackward 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_softplus_backward.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 shape_size = 1;
  for (auto i : shape) {
    shape_size *= i;
  }
  return shape_size;
}

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-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 to the device.
  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. See the ACL API manual.
  // Fill in the deviceId based on your actual device.
  int32_t deviceId = 0;
  aclrtStream stream;
  auto ret = Init(deviceId, &stream);
  // Handle the check as needed.
  CHECK_RET(ret == 0, 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> gradOutputShape = {4, 2};
  std::vector<int64_t> selfShape = {4, 2};
  std::vector<int64_t> gradInputShape = {4, 2};
  void* gradOutputDeviceAddr = nullptr;
  void* selfDeviceAddr = nullptr;
  void* gradInputDeviceAddr = nullptr;
  aclTensor* gradOutput = nullptr;
  aclTensor* self = nullptr;
  aclTensor* gradInput = nullptr;
  aclScalar* beta = nullptr;
  aclScalar* threshold = nullptr;

  std::vector<float> gradOutputHostData = {0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8};
  std::vector<float> selfHostData = {0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8};
  std::vector<float> gradInputHostData = {0, 0, 0, 0, 0, 0, 0, 0};
  float betaValue = 1.1f;
  float thresholdValue = 1.2f;

  // Create gradOutput, self, and gradInput aclTensor.
  ret = CreateAclTensor(gradOutputHostData, gradOutputShape, &gradOutputDeviceAddr, aclDataType::ACL_FLOAT, &gradOutput);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  ret = CreateAclTensor(gradInputHostData, gradInputShape, &gradInputDeviceAddr, aclDataType::ACL_FLOAT, &gradInput);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  beta = aclCreateScalar(&betaValue, aclDataType::ACL_FLOAT);
  CHECK_RET(beta != nullptr, return ret);
  threshold = aclCreateScalar(&thresholdValue, aclDataType::ACL_FLOAT);
  CHECK_RET(threshold != nullptr, return ret);

  // 3. Call the CANN operator library API. Replace with the specific API.
  uint64_t workspaceSize = 0;
  aclOpExecutor* executor;
  // Call the first-phase API of aclnnSoftplusBackward.
  ret = aclnnSoftplusBackwardGetWorkspaceSize(gradOutput, self, beta, threshold, gradInput, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnSoftplusBackwardGetWorkspaceSize 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 aclnnSoftplusBackward.
  ret = aclnnSoftplusBackward(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnSoftplusBackward failed. ERROR: %d\n", ret); return ret);
  // 4. (Boilerplate) Synchronously 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 memory. Modify based on the specific API definition.
  auto size = GetShapeSize(gradInputShape);
  std::vector<float> resultData(size, 0);
  ret = aclrtMemcpy(resultData.data(), resultData.size() * sizeof(resultData[0]), gradInputDeviceAddr,
                    size * sizeof(float), 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. Modify based on the specific API definition.
  aclDestroyTensor(gradOutput);
  aclDestroyTensor(self);
  aclDestroyTensor(gradInput);

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