aclnnClampMax&aclnnInplaceClampMax

Supported Products

Product	Supported
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: Limits all input elements within the range of [-inf, max].
Formula:
${y}_{i} = min({{x}_{i}},max)$

Prototype

aclnnClampMax and aclnnInplaceClampMax implement the same function in different ways. Select a proper operator based on your requirements.
- aclnnClampMax: An output tensor object needs to be created to store the computation result.
- aclnnInplaceClampMax: No output tensor object needs to be created, and the computation result is stored in the memory of the input tensor.
Each operator has two-phase API calls. First, aclnnClampMaxGetWorkspaceSize or aclnnInplaceClampMaxGetWorkspaceSize is called to obtain the workspace size required for computation and the executor that contains the operator computation process. Then, aclnnClampMax or aclnnInplaceClampMax is called to perform computation.
- aclnnStatus aclnnClampMaxGetWorkspaceSize(const aclTensor *self, const aclScalar *clipValueMax, aclTensor *out, uint64_t *workspaceSize, aclOpExecutor **executor)
- aclnnStatus aclnnClampMax(void *workspace, uint64_t workspaceSize, aclOpExecutor *executor, aclrtStream stream)
- aclnnStatus aclnnInplaceClampMaxGetWorkspaceSize(const aclTensor *selfRef, const aclScalar *clipValueMax, uint64_t *workspaceSize, aclOpExecutor **executor)
- aclnnStatus aclnnInplaceClampMax(void *workspace, uint64_t workspaceSize, aclOpExecutor *executor, aclrtStream stream)

aclnnClampMaxGetWorkspaceSize

Parameters
- self (aclTensor*, computation input): input tensor, aclTensor on the device. The shape must be 1D to 8D. Non-contiguous tensors are supported. The data format can be ND.
  - Atlas inference products and Atlas training products: The data type can be FLOAT16, FLOAT, DOUBLE, INT8, UINT8, INT16, INT32, or INT64.
  - Atlas A2 training products/Atlas A2 inference products and Atlas A3 training products/Atlas A3 inference products: The data type can be FLOAT16, FLOAT, DOUBLE, INT8, UINT8, INT16, INT32, INT64, or BFLOAT16.
- clipValueMax (aclScalar*, computation input): upper bound.
  - Atlas inference products and Atlas training products: The data type can be FLOAT16, FLOAT, DOUBLE, INT8, UINT8, INT16, INT32, INT64, or BOOL.
  - Atlas A2 training products/Atlas A2 inference products and Atlas A3 training products/Atlas A3 inference products: The data type can be FLOAT16, FLOAT, DOUBLE, INT8, UINT8, INT16, INT32, INT64, BFLOAT16, or BOOL.
- out (aclTensor *, computation output): output tensor. The shape must be identical to that of self. The data format can be ND.
  - Atlas inference products and Atlas training products: The data type can be FLOAT16, FLOAT, DOUBLE, INT8, UINT8, INT16, INT32, or INT64. The data type must be the same as that of self.
  - Atlas A2 training products/Atlas A2 inference products and Atlas A3 training products/Atlas A3 inference products: The data type can be FLOAT16, FLOAT, DOUBLE, INT8, UINT8, INT16, INT32, INT64, or BFLOAT16. The data type must be the same as that of self.
- workspaceSize (uint64_t*, output): size of the workspace to be allocated on the device.
- executor (aclOpExecutor**, output): operator executor, containing the operator computation process.

Returns

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

(161001) ACLNN_ERR_PARAM_NULLPTR: 1. The passed self or out is a null pointer, or max is a null pointer.
(161002) ACLNN_ERR_PARAM_INVALID: 1. The data type of self or out is not supported.
                                  2. The shapes of self and out are inconsistent.

aclnnClampMax

Parameters
- workspace (void *, input): address of the workspace to be allocated on the device.
- workspaceSize (uint64_t, input): size of the workspace to be allocated on the device, which is obtained by calling the first-phase API aclnnClampMaxGetWorkspaceSize.
- executor (aclOpExecutor *, input): operator executor, containing the operator computation process.
- stream (aclrtStream, input): stream for executing the task.
Returns

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

aclnnInplaceClampMaxGetWorkspaceSize

Parameters
- selfRef (aclTensor *, computation input | computation output): input tensor. Non-contiguous tensors are supported. The data format can be ND.
  - Atlas inference products and Atlas training products: The data type can be FLOAT16, FLOAT, DOUBLE, INT8, UINT8, INT16, INT32, or INT64.
  - Atlas A2 training products/Atlas A2 inference products and Atlas A3 training products/Atlas A3 inference products: The data type can be FLOAT16, FLOAT, DOUBLE, INT8, UINT8, INT16, INT32, INT64, or BFLOAT16.
- clipValueMax (aclScalar*, computation input): upper bound.
  - Atlas inference products and Atlas training products: The data type can be FLOAT16, FLOAT, DOUBLE, INT8, UINT8, INT16, INT32, or INT64.
  - Atlas A2 training products/Atlas A2 inference products and Atlas A3 training products/Atlas A3 inference products: The data type can be FLOAT16, FLOAT, DOUBLE, INT8, UINT8, INT16, INT32, INT64, or BFLOAT16.
- workspaceSize (uint64_t*, output): size of the workspace to be allocated on the device.
- executor (aclOpExecutor**, output): operator executor, containing the operator computation process.

Returns

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

(161001) ACLNN_ERR_PARAM_NULLPTR: 1. The passed selfRef or max is a null pointer.
(161002) ACLNN_ERR_PARAM_INVALID: 1. The data type of selfRef is not supported.

aclnnInplaceClampMax

Parameters
- workspace (void *, input): address of the workspace to be allocated on the device.
- workspaceSize (uint64_t, input): size of the workspace to be allocated on the device, which is obtained by calling the first-phase API aclnnInplaceClampMaxGetWorkspaceSize.
- executor (aclOpExecutor *, input): operator executor, containing the operator computation process.
- stream (aclrtStream, input): stream for executing the task.
Returns

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

Constraints

Deterministic computation:
- aclnnClampMax and aclnnInplaceClampMax default to a deterministic implementation.

Examples

The following examples are for reference only. For details, see Compilation and Running Sample.

aclNNClampMax Sample Code

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_clamp.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 allocate 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);
  // Call aclrtMemcpy to copy host data 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);

  // Compute 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 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 PrepareInputAndOutput(
    std::vector<int64_t>& selfShape, std::vector<int64_t>& outShape, void** selfDeviceAddr, aclTensor** self, aclScalar** max,
    void** outDeviceAddr, aclTensor** out)
{
  std::vector<int8_t> selfHostData = {0, 1, 2, 3, 4, 5, 6, 7};
  std::vector<int8_t> outHostData(8, 0);
  int8_t maxValue=4;
  // Create a self aclTensor.
  auto ret = CreateAclTensor(selfHostData, selfShape, selfDeviceAddr, aclDataType::ACL_INT8, self);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create a max aclScalar.
  *max = aclCreateScalar(&maxValue, aclDataType::ACL_INT8);
  CHECK_RET(*max != nullptr, return ret);
  // Create an out aclTensor.
  ret = CreateAclTensor(outHostData, outShape, outDeviceAddr, aclDataType::ACL_INT8, out);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  return ACL_SUCCESS;
}

void ReleaseTensorAndScalar(aclTensor* self, aclScalar* max, aclTensor* out)
{
  aclDestroyTensor(self);
  aclDestroyScalar(max);
  aclDestroyTensor(out);
}

void ReleaseDevice(
    void* selfDeviceAddr, void* outDeviceAddr, uint64_t workspaceSize, void* workspaceAddr, aclrtStream stream,
    int32_t deviceId)
{
  aclrtFree(selfDeviceAddr);
  aclrtFree(outDeviceAddr);
  if (workspaceSize > 0) {
    aclrtFree(workspaceAddr);
  }
  aclrtDestroyStream(stream);
  aclrtResetDevice(deviceId);
  aclFinalize();
}

int main() {
  // 1. (Boilerplate) Initialize the device and stream. For details, see the AscendCL API manual.
  // Set the device ID 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 inputs and outputs based on the API definition.
  std::vector<int64_t> selfShape = {4, 2};
  std::vector<int64_t> outShape = {4, 2};
  void* selfDeviceAddr = nullptr;
  void* outDeviceAddr = nullptr;
  aclTensor* self = nullptr;
  aclScalar* max = nullptr;
  aclTensor* out = nullptr;

  ret = PrepareInputAndOutput(selfShape, outShape, &selfDeviceAddr, &self, &max, &outDeviceAddr, &out);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  // 3. Call the CANN operator library API. Replace it with the actual API name.
  uint64_t workspaceSize = 0;
  aclOpExecutor* executor;
  // Call the first-phase API of aclnnClampMax.
  ret = aclnnClampMaxGetWorkspaceSize(self, max, out, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnClampMaxGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);
  // Allocate device memory based on workspaceSize computed 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 aclnnClampMax.
  ret = aclnnClampMax(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnClampMax failed. ERROR: %d\n", ret); return ret);

  // 4. (Boilerplate) Synchronize the stream and wait for the task 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. Modify the code based on the 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 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 aclTensors and aclScalars. Modify the code based on the API definition. 
  ReleaseTensorAndScalar(self, max, out);
  
  // 7. Release device resources. Modify the code based on the API definition.
  ReleaseDevice(selfDeviceAddr, outDeviceAddr, workspaceSize, workspaceAddr, stream, deviceId);

  return 0;
}

aclnnInplaceClampMax Sample Code

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_clamp.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 allocate 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);
  // Call aclrtMemcpy to copy host data 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);

  // Compute 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 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 PrepareInputAndOutput(
    std::vector<int64_t>& selfShape, void** selfDeviceAddr, aclTensor** self, aclScalar** max)
{
  std::vector<int8_t> selfHostData = {0, 1, 2, 3, 4, 5, 6, 7};
  int8_t maxValue=4;
  // Create a self aclTensor.
  auto ret = CreateAclTensor(selfHostData, selfShape, selfDeviceAddr, aclDataType::ACL_INT8, self);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create a max aclScalar.
  *max = aclCreateScalar(&maxValue, aclDataType::ACL_INT8);
  CHECK_RET(*max != nullptr, return ret);

  return ACL_SUCCESS;
}

void ReleaseTensorAndScalar(aclTensor* self, aclScalar* max)
{
  aclDestroyTensor(self);
  aclDestroyScalar(max);
}

void ReleaseDevice(
    void* selfDeviceAddr, uint64_t workspaceSize, void* workspaceAddr, aclrtStream stream,
    int32_t deviceId)
{
  aclrtFree(selfDeviceAddr);
  if (workspaceSize > 0) {
    aclrtFree(workspaceAddr);
  }
  aclrtDestroyStream(stream);
  aclrtResetDevice(deviceId);
  aclFinalize();
}

int main() {
  // 1. (Boilerplate) Initialize the device and stream. For details, see the AscendCL API manual.
  // Set the device ID 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 inputs and outputs based on the API definition.
  std::vector<int64_t> selfShape = {4, 2};
  std::vector<int64_t> outShape = {4, 2};
  void* selfDeviceAddr = nullptr;
  aclTensor* self = nullptr;
  aclScalar* max = nullptr;
  
  ret = PrepareInputAndOutput(selfShape, &selfDeviceAddr, &self, &max);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  // 3. Call the CANN operator library API. Replace it with the actual API name.
  uint64_t workspaceSize = 0;
  aclOpExecutor* executor;
  // Call the first-phase API of aclnnInplaceClampMax.
  ret = aclnnInplaceClampMaxGetWorkspaceSize(self, max, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnInplaceClampMaxGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);
  // Allocate device memory based on workspaceSize computed 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 aclnnInplaceClampMax.
  ret = aclnnInplaceClampMax(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnInplaceClampMax failed. ERROR: %d\n", ret); return ret);

  // 4. (Boilerplate) Synchronize the stream and wait for the task 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. Modify the code based on the API definition.
  auto size = GetShapeSize(selfShape);
  std::vector<float> resultData(size, 0);
  ret = aclrtMemcpy(resultData.data(), resultData.size() * sizeof(resultData[0]), selfDeviceAddr,
                    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: %f\n", i, resultData[i]);
  }

  // 6. Release aclTensors and aclScalars. Modify the code based on the API definition. 
  ReleaseTensorAndScalar(self, max);

  // 7. Release device resources. Modify the code based on the API definition.
  ReleaseDevice(selfDeviceAddr, workspaceSize, workspaceAddr, stream, deviceId);

  return 0;
}