aclnnHardtanhBackward

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 the activation function aclnnHardtanh.
Formula:
$grad\_self_{i} = \begin{cases} 0,\ \ \ \ \ \ \ if \ \ self_{i}>max \\ 0,\ \ \ \ \ \ \ if\ \ self_{i}<min \\ 1,\ \ \ \ \ \ \ \ \ \ \ \ otherwise \\ \end{cases}$ $res_{i} = grad\_output_{i} \times grad\_self_{i}$

Function Prototype

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

aclnnStatus aclnnHardtanhBackwardGetWorkspaceSize(
  const aclTensor* gradOutput,
  const aclTensor* self,
  const aclScalar* min,
  const aclScalar* max,
  aclTensor*       out,
  uint64_t*        workspaceSize,
  aclOpExecutor**  executor)

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

aclnnHardtanhBackwardGetWorkspaceSize

Parameters:

Parameter	Input/Output	Description	Instruction	Data Type	Data Format	Dimension (shape)	Non-Contiguous Tensor
gradOutput (aclTensor*)	Input	The gradient output from the previous step in backward propagation, i.e., grad_output in the formula.	Null tensor is not supported. gradOutput, self, and out have the same shape. gradOutput, self, and out have the same data type.	BFLOAT16, FLOAT16, FLOAT	ND	0-8	√
self (aclTensor*)	Input	Forward input data, self in the formula.	Null Not Supported. Consistent shapes among gradOutput, self, and out. Consistent data types among gradOutput, self, and out.	BFLOAT16, FLOAT16, FLOAT	ND	0-8	√
min (aclScalar*)	Input	The lower bound of the linear range, min in the formula.	-	BFLOAT16, FLOAT16, FLOAT	-	-	-
max (aclScalar*)	Input	Upper limit of the linear range, max in the formula.	-	BFLOAT16, FLOAT16, FLOAT	-	-	-
out (aclTensor*)	Output	Computed gradient result.	Null Tensor is not supported. The shapes of gradOutput, self, and out must be consistent. The data types of gradOutput, self, and out must be consistent.	BFLOAT16, FLOAT16, FLOAT	ND	0-8	√
workspaceSize (uint64_t*)	Output	Size of the workspace to be applied on the Device side.	-	-	-	-	-
executor (aclOpExecutor**)	Output	Returns the op executor, which includes the calculation process.	-	-	-	-	-

Atlas inference products, Atlas training products: The supported data types are FLOAT16 and FLOAT.

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 input gradOutput, self, out, min, or max contains a null pointer.
ACLNN_ERR_PARAM_INVALID	161002	The data type of gradOutput, self, or out is not within the supported range.
		gradOutput and self have different data types.
		gradOutput and self have different shapes.
		The dimension of gradOutput or self is greater than 8.

aclnnHardtanhBackward

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 from the first-phase API aclnnHardtanhBackwardGetWorkspaceSize.
executor	Input	Operator executor, which includes 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:
- aclnnHardtanhBackward 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_hardtanh_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: 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 data from the host side to the device 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);

  // Calculate the strides of the 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.
  // Specify the deviceId based on your actual device.
  int32_t deviceId = 0;
  aclrtStream stream;
  auto ret = Init(deviceId, &stream);
  // Handle the check result based on your own needs.
  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.
  std::vector<int64_t> gradOutputShape = {4, 2};
  std::vector<int64_t> selfShape = {4, 2};
  std::vector<int64_t> outShape = {4, 2};
  void* gradOutputDeviceAddr = nullptr;
  void* selfDeviceAddr = nullptr;
  void* outDeviceAddr = nullptr;
  aclTensor* gradOutput = nullptr;
  aclTensor* self = nullptr;
  aclTensor* out = nullptr;
  aclScalar* clipValueMin = nullptr;
  aclScalar* clipValueMax = nullptr;
  std::vector<float> gradOutputHostData = {0, 1, 2, 3, 4, 5, 6, 7};
  std::vector<float> selfHostData = {1, 1, 1, 2, 1, 2, 3, 3};
  std::vector<float> outHostData = {0, 0, 0, 0, 0, 0, 0, 0};
  float clipValueMinValue = 1.2f;
  float clipValueMaxValue = 2.4f;
  // Create the gradOutput aclTensor.
  ret = CreateAclTensor(gradOutputHostData, gradOutputShape, &gradOutputDeviceAddr, aclDataType::ACL_FLOAT, &gradOutput);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create the self aclTensor.
  ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create the clipValueMin aclScalar.
  clipValueMin = aclCreateScalar(&clipValueMinValue, aclDataType::ACL_FLOAT);
  CHECK_RET(clipValueMin != nullptr, return ret);
  // Create clipValueMax aclScalar.
  clipValueMax = aclCreateScalar(&clipValueMaxValue, aclDataType::ACL_FLOAT);
  CHECK_RET(clipValueMax != nullptr, 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.
  uint64_t workspaceSize = 0;
  aclOpExecutor* executor;
  // Call the first-phase API of aclnnHardtanhBackward.
  ret = aclnnHardtanhBackwardGetWorkspaceSize(gradOutput, self, clipValueMin, clipValueMax, out, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnHardtanhBackwardGetWorkspaceSize 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 aclnnHardtanhBackward.
  ret = aclnnHardtanhBackward(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnHardtanhBackward failed. ERROR: %d\n", ret); return ret);
  // 4. (Boilerplate) Wait for the task execution to complete synchronously.
  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(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 and aclScalar. Modify based on the specific API definition.
  aclDestroyTensor(gradOutput);
  aclDestroyTensor(self);
  aclDestroyScalar(clipValueMin);
  aclDestroyScalar(clipValueMax);
  aclDestroyTensor(out);

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