aclnnMultilabelMarginLoss

📄 View Source Code

Applicable Products

Function

• Description: Computes the negative log-likelihood loss value.

• Formula:

self is the input with shape (N, C) or (C), where N indicates the batch size and C indicates the number of classes. target indicates the true labels with shape (N, C) or (C), where each element ranges from [-1, C - 1]. To ensure the same shape as the input, padding with -1 is applied, meaning that the labels before the first -1 represent the true labels yTrue of the sample. For example, if y = [0, 3, -1, 1], the true labels yTrue are [0, 3]. The formula for each sample is as follows:

When reduction is none,

If reduction is not none, then

Function Prototype

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

aclnnStatus aclnnMultilabelMarginLossGetWorkspaceSize(
    const aclTensor* self,
    const aclTensor* target,
    int64_t          reduction,
    aclTensor*       out,
    aclTensor*       isTarget,
    uint64_t*        workspaceSize,
    aclOpExecutor**  executor)

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

aclnnMultilabelMarginLossGetWorkspaceSize

• Parameters

  Parameter	Input/Output	Description	Instruction	Data Type	Data Format	Dimension (shape)	Non-contiguous Tensor
  self (aclTensor*)	Input	Input tensor, which is the input x in the formula.	Shape is (N, C) or (C), where N indicates the batch size and C indicates the number of classes. When the number of elements in `self` exceeds 15000*20000, a 507034 Vector Core timeout may occur.	FLOAT, FLOAT16, BFLOAT16	ND	1, 2	√
  target (aclTensor*)	Input	Ground-truth labels, i.e., the input y in the formula.	Shape (N, C) or (C), where each element is in the range [-1, C - 1], padded with -1, meaning the labels before the first -1 represent the true labels of the sample.	INT32, INT64	ND	1, 2	√
  reduction (int64_t)	Input	Specifies the reduction to apply to the output, corresponding to the parameter reduction in the formula.	Supports 0 (none) | 1 (mean) | 2 (sum). 'none' means no reduction is applied. 'mean' means the sum of the output will be divided by the number of elements in the output. 'sum' means the output will be summed.	INT64	-	-	√
  out (aclTensor*)	Output	Output loss, ℓ(x,y) in the formula.	Shape is (N) or ().	Consistent with self and isTarget.	ND	0, 1
  isTarget (aclTensor*)	Output	Output `istarget` in the formula.	Shape is (N, C) or (C).	Consistent with self and out.	ND	1, 2	√
  workspaceSize (uint64_t*)	Output	Size of the workspace to be requested on the Device side.	-	-	-	-	-
  executor (aclOpExecutor**)	Output	Returns the op 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 input self, target, out, or isTarget is a null pointer.
  ACLNN_ERR_PARAM_INVALID	161002	The data type of self, out, or isTarget is not within the supported range.
  		The data types of self, out, and isTarget are inconsistent.
  		The data type of target is not within the supported range.
  		The shapes of self, target, and isTarget do not meet the requirements in Parameters.
  		The shape of out does not meet the requirements in Parameters.

aclnnMultilabelMarginLoss

• 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 through the first-phase API aclnnMultilabelMarginLossGetWorkspaceSize.
  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:

  • aclnnMultilabelMarginLoss defaults to a non-deterministic implementation. Deterministic computation can be enabled by calling aclrtCtxSetSysParamOpt.

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_multilabel_margin_loss.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 for 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 data from the host to the device 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);

  // Calculate 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 aclCreateTensor.
  *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. Customize the construction based on the API.
  std::vector<int64_t> selfShape = {2, 3};
  std::vector<int64_t> targetShape = {2, 3};
  std::vector<int64_t> outShape = {2};
  std::vector<int64_t> istargetShape = {2, 3};
  void* selfDeviceAddr = nullptr;
  void* targetDeviceAddr = nullptr;
  void* outDeviceAddr = nullptr;
  void* istargetDeviceAddr = nullptr;
  aclTensor* self = nullptr;
  aclTensor* target = nullptr;
  aclTensor* out = nullptr;
  aclTensor* istarget = nullptr;
  std::vector<float> selfHostData = {0, 1, 2, 3, 4, 5};
  std::vector<int32_t> targetHostData = {0, 1, 2, 3, 4, 5};
  std::vector<float> outHostData(2, 0);
  std::vector<float> istargetHostData(6, 0);
  int64_t reduction = 0;
  // Create the self aclTensor.
  ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create the target aclTensor.
  ret = CreateAclTensor(targetHostData, targetShape, &targetDeviceAddr, aclDataType::ACL_INT32, &target);
  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);
  // Create the istarget aclTensor.
  ret = CreateAclTensor(istargetHostData, istargetShape, &istargetDeviceAddr, aclDataType::ACL_FLOAT,
                        &istarget);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  // 3. Call the CANN operator library API. Replace with the actual API name.
  uint64_t workspaceSize = 0;
  aclOpExecutor* executor;
  // Call the first-phase API of aclnnMultilabelMarginLoss.
  ret = aclnnMultilabelMarginLossGetWorkspaceSize(self, target, reduction, out, istarget, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnMultilabelMarginLossGetWorkspaceSize 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 aclnnMultilabelMarginLoss.
  ret = aclnnMultilabelMarginLoss(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnMultilabelMarginLoss 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 memory. Modify based on the specific API definition.
  auto outSize = GetShapeSize(outShape);
  std::vector<float> outData(outSize, 0);
  ret = aclrtMemcpy(outData.data(), outData.size() * sizeof(outData[0]), outDeviceAddr,
                    outSize * 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 < outSize; i++) {
    LOG_PRINT("out[%ld] is: %f\n", i, outData[i]);
  }

  auto istargetSize = GetShapeSize(istargetShape);
  std::vector<float> istargetData(istargetSize, 0);
  ret = aclrtMemcpy(istargetData.data(), istargetData.size() * sizeof(istargetData[0]), istargetDeviceAddr,
                    istargetSize * sizeof(istargetData[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 < istargetSize; i++) {
    LOG_PRINT("istarget[%ld] is: %f\n", i, istargetData[i]);
  }

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

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