aclnnRemainderTensorScalar&aclnnInplaceRemainderTensorScalar

📄 View Source Code

Applicable Products

Function

• Description: Converts each element in the tensor self to the remainder obtained after division by the scalar other. The result has the same sign as the divisor other, and its absolute value is less than the absolute value of other.

• Formula: The actual computation of remainder(self, other) is equivalent to the following formula:

  $$out_i = self_i - floor(self_i / other) * other$$

• Example:

  self = tensor([[-1, -2],
                 [-3, -4]]).type(int32)
  other = 3.5   # float
  result = remainder(self, other)
  
  # Value of result.
  # tensor([[2.5000, 1.5000],
  #         [0.5000, 3.0000]])   float
  
  # For the element -1 in self, the calculation result is remainder(-1, 3.5) = -1 - floor(-1 / 3.5) * 3.5 = -1 - (-1) * 3.5 = -1 + 3.5 = 2.5.
  # As shown, the absolute value of the final result 2.5 is less than that of other 3.5.

Function Prototype

• aclnnRemainderTensorScalar and aclnnInplaceRemainderTensorScalar implement the same function. The differences in usage are as follows. Select the appropriate operator based on your actual scenario.

  • aclnnRemainderTensorScalar: Requires creating a new output tensor object to store the computation result.

  • aclnnInplaceRemainderTensorScalar: Does not require creating a new output tensor object. The computation result is stored directly in the memory of the input tensor.

• Each API is divided into a two-phase API. You must first call the "aclnnRemainderTensorScalarGetWorkspaceSize" or "aclnnInplaceRemainderTensorScalarGetWorkspaceSize" API to obtain the workspace size required for computation and the executor that contains the operator computation process, and then call the "aclnnRemainderTensorScalar" or "aclnnInplaceRemainderTensorScalar" API to execute the computation.

  aclnnStatus aclnnRemainderTensorScalarGetWorkspaceSize(
    const aclTensor* self,
    const aclScalar* other,
    aclTensor*       out,
    uint64_t*        workspaceSize,
    aclOpExecutor**  executor)

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

  aclnnStatus aclnnInplaceRemainderTensorScalarGetWorkspaceSize(
    aclTensor*       selfRef,
    const aclScalar* other,
    uint64_t*        workspaceSize,
    aclOpExecutor**  executor)

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

aclnnRemainderTensorScalarGetWorkspaceSize

• Parameters:

  Parameter	Input/Output	Description	Instruction	Data Type	Data Format	Dimension (Shape)	Non-contiguous Tensor
  self (aclTensor*)	Input	Input self in the formula.	Supports empty tensors. The data types of self and other must satisfy the data type deduction rules (see TensorScalar Deduction Relationship/Deduction Relationship).	INT32, INT64, FLOAT16, FLOAT, DOUBLE, BFLOAT16	ND	0-8	√
  other (aclScalar*)	Input	Represents the input "other" in the formula.	The data type of other must satisfy the data type deduction rules with the data type of self (see TensorScalar Deduction Relationship/Deduction Relationship).	INT32, INT64, FLOAT16, FLOAT, DOUBLE, BFLOAT16	-	-	-
  out (aclTensor*)	Output	Output out in the formula.	Supports empty tensors. The data type must be a data type that can be converted from the deduction of self and other. The shape must be the same as self.	UINT8, INT8, INT16, INT32, INT64, FLOAT16, FLOAT, DOUBLE, COMPLEX64, COMPLEX128, BFLOAT16	ND	0-8	√
  workspaceSize (uint64_t*)	Output	Returns the workspace size to be requested on the Device side.	-	-	-	-	-
  executor (aclOpExecutor**)	Output	Returns the op executor, which contains the operator computation flow.	-	-	-	-	-

  • Atlas training products: BFLOAT16 data type is not supported.

• Return value

  aclnnStatus: return 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 input self, other, or out is a null pointer.
  ACLNN_ERR_PARAM_INVALID	161002	The shapes of self and out are different.
  		The data type cannot be deduced from self and other.
  		The data type deduced from self and other is not in the supported data types.
  		The data type derived from self and other cannot be converted to the specified output out type.
  		The number of dimensions of self and out exceeds 8.
  		The data formats of self and out are inconsistent.

aclnnRemainderTensorScalar

• 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 by the first-phase API aclnnRemainderTensorScalarGetWorkspaceSize.
  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 Codes.

aclnnInplaceRemainderTensorScalarGetWorkspaceSize

• Parameters

  Parameter	Input/Output	Description	Instruction	Data Type	Data Format	Shape	Non-contiguous Tensor
  selfRef (aclTensor*)	Input/Output	Input/output tensor.	Supports empty tensors. The data type of selfRef and other must satisfy the data type deduction rules (see TensorScalar Deduction Relationship/Deduction Relationship), and must be a convertible data type after deduction.	INT32, INT64, FLOAT16, FLOAT, DOUBLE, BFLOAT16	ND	0-8	√
  other (aclScalar*)	Input	The input "other" in the formula.	The data type of other and the data type of selfRef must satisfy the data type deduction rules (see TensorScalar Deduction Relationship/Deduction Relationship).	INT32, INT64, FLOAT16, FLOAT, DOUBLE, BFLOAT16	-	-	-
  workspaceSize (uint64_t*)	Output	Returns the workspace size to be allocated on the Device side.	-	-	-	-	-
  executor (aclOpExecutor**)	Output	Returns the operator executor, which contains the operator computation flow.	-	-	-	-	-

  • Atlas training products: selfRef and other do not support BFLOAT16.

• Return value

  aclnnStatus: return 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 selfRef or other is a null pointer.
  ACLNN_ERR_PARAM_INVALID	161002	The data type cannot be deduced from selfRef and other.
  		The data type deduced from selfRef and other is not a supported data type.
  		The data type deduced from selfRef and other cannot be converted to the data type of selfRef.
  		The number of dimensions of selfRef is greater than 8.

aclnnInplaceRemainderTensorScalar

• 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 through the first-phase API aclnnInplaceRemainderTensorScalarGetWorkspaceSize.
  executor	Input	Operator executor, which contains the operator computation process.
  stream	Input	Specifies the stream for task execution.

• Return value

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

Constraints

• Deterministic computation:

  • aclnnRemainderTensorScalar&aclnnInplaceRemainderTensorScalar default to a deterministic implementation.

  • When the data type of self is INT32, functionality and precision within the range [-2^24, 2^24] are prioritized.

  • When other is 0 and the data type of self is an integer type, the result in out is self.

Example

The following provides example code for reference only. For details about compilation and running, see Compile and Run Samples.

aclnnRemainderTensorScalar sample code:

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_remainder.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: 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 host-side data 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);

  // 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];
  }

  // 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.
  // Enter the deviceId based on the 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.
  std::vector<int64_t> selfShape = {3, 3};
  std::vector<int64_t> outShape = {3, 3};
  void* selfDeviceAddr = nullptr;
  void* outDeviceAddr = nullptr;
  aclTensor* self = nullptr;
  aclScalar* other = nullptr;
  aclTensor* out = nullptr;
  std::vector<int64_t> selfHostData = {0, 1, 2, 3, 4, 5, 6, 7, 8};
  std::vector<int64_t> outHostData = {0, 0, 0, 0, 0, 0, 0, 0, 0};
  int64_t Other = 3;

  // Create the self aclTensor.
  ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_INT64, &self);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create the other aclScalar.
  other = aclCreateScalar(&Other, aclDataType::ACL_INT64);
  CHECK_RET(other != nullptr, return ret);
  // Create the out aclTensor.
  ret = CreateAclTensor(outHostData, outShape, &outDeviceAddr, aclDataType::ACL_INT64, &out);
  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 aclnnRemainderTensorScalar.
  ret = aclnnRemainderTensorScalarGetWorkspaceSize(self, other, out, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnRemainderTensorScalarGetWorkspaceSize 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 aclnnRemainderTensorScalar.
  ret = aclnnRemainderTensorScalar(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnRemainderTensorScalar 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 based on the specific API definition.
  auto size = GetShapeSize(outShape);
  std::vector<int64_t> 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: %ld\n", i, resultData[i]);
  }

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

  // 7. Release device resources.
  aclrtFree(selfDeviceAddr);
  aclrtFree(outDeviceAddr);
  if (workspaceSize > 0) {
    aclrtFree(workspaceAddr);
  }
  aclrtDestroyStream(stream);
  aclrtResetDevice(deviceId);
  aclFinalize();
  return 0;
}

aclnnInplaceRemainderTensorScalar sample code:

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_remainder.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-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 host-side data to 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);

  // 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];
  }

  // 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.
  // Fill in the 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 inputs and outputs. Customize the construction based on the API.
  std::vector<int64_t> selfRefShape = {3, 3};
  void* selfRefDeviceAddr = nullptr;
  aclTensor* selfRef = nullptr;
  aclScalar* other = nullptr;
  std::vector<int64_t> selfRefHostData = {0, 1, 2, 3, 4, 5, 6, 7, 8};
  int64_t Other = 3;

  // Create the self aclTensor.
  ret = CreateAclTensor(selfRefHostData, selfRefShape, &selfRefDeviceAddr, aclDataType::ACL_INT64, &selfRef);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create the other aclScalar.
  other = aclCreateScalar(&Other, aclDataType::ACL_INT64);
  CHECK_RET(other != nullptr, 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 aclnnInplaceRemainderTensorScalar.
  ret = aclnnInplaceRemainderTensorScalarGetWorkspaceSize(selfRef, other, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnInplaceRemainderTensorScalarGetWorkspaceSize 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 aclnnInplaceRemainderTensorScalar.
  ret = aclnnInplaceRemainderTensorScalar(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnInplaceRemainderTensorScalar 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(selfRefShape);
  std::vector<int64_t> resultData(size, 0);
  ret = aclrtMemcpy(resultData.data(), resultData.size() * sizeof(resultData[0]), selfRefDeviceAddr,
                    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: %ld\n", i, resultData[i]);
  }

  // 6. Release aclTensor and aclScalar. Modify based on the specific API definition.
  aclDestroyTensor(selfRef);
  aclDestroyScalar(other);

  // 7. Release device resources.
  aclrtFree(selfRefDeviceAddr);
  if (workspaceSize > 0) {
    aclrtFree(workspaceAddr);
  }
  aclrtDestroyStream(stream);
  aclrtResetDevice(deviceId);
  aclFinalize();

  return 0;
}