Documentation
Download

aclnnLerp&aclnnInplaceLerp

📄 View Source Code

Applicable Products

ProductSupported
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: Performs linear interpolation between the start and end tensors based on a given weight, and returns the interpolated tensor.

  • Formula:

 out i= start i+ weight i×( end i− start i)\text { out }_i=\text { start }_i+\text { weight }_i \times\left(\text { end }_i-\text { start }_i\right)

Function Prototype

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

    • aclnnLerp: Create a new output tensor object to store the computation result.

    • aclnnInplaceLerp: No need to create a new output tensor object. The computation result is stored directly in the memory of the input tensor.

  • Each operator uses a two-phase API. You must first call the "aclnnLerpGetWorkspaceSize" or "aclnnInplaceLerpGetWorkspaceSize" API to obtain the workspace size required for computation and the executor that contains the operator computation process, and then call the "aclnnLerp" or "aclnnInplaceLerp" API to perform the computation.

aclnnStatus aclnnLerpGetWorkspaceSize(
  const aclTensor*          self,
  const aclTensor*          end,
  const aclTensor*          weight,
  aclTensor*                out,
  uint64_t*                 workspaceSize,
  aclOpExecutor**           executor)
aclnnStatus aclnnLerp(
  void*                     workspace,
  uint64_t                  workspaceSize,
  aclOpExecutor*            executor,
  const aclrtStream         stream)
aclnnStatus aclnnInplaceLerpGetWorkspaceSize(
  aclTensor*                selfRef,
  const aclTensor*          end,
  const aclTensor*          weight,
  uint64_t*                 workspaceSize,
  aclOpExecutor**           executor)
aclnnStatus aclnnInplaceLerp(
  void*                     workspace,
  uint64_t                  workspaceSize,
  aclOpExecutor*            executor,
  const aclrtStream         stream)

aclnnLerpGetWorkspaceSize

  • Parameters:

    Parameter Input/Output Description Instruction Data Type Data Format Dimension Non-contiguous Tensor
    self (aclTensor*) Input The input start in the formula. The data type must be the same as that of end, weight, and out. The shape must satisfy the broadcast relationship with end and weight. FLOAT, FLOAT16, BFLOAT16 ND 0-8 √
    end (aclTensor*) Input The input end in the formula. Data type consistent with self, weight, and out. The shape must meet the broadcast relationship with self and weight. FLOAT, FLOAT16, BFLOAT16 ND 0-8 √
    weight (aclTensor*) Input The input weight in the formula. The data type is the same as that of self, end, and out. The shape must satisfy the broadcast relationship with self and end. FLOAT, FLOAT16, BFLOAT16 ND 0-8 √
    out (aclTensor*) Output The out in the formula. The data type is the same as self, end, and weight. The shape must be the same as the shape after self, end, and weight are broadcast. FLOAT, FLOAT16, BFLOAT16 ND 0-8 √
    workspaceSize (uint64_t*) Output Returns the workspace size to be requested on the Device side. - - - - -
    Executor (aclOpExecutor**) Output Returns the operator executor, which contains the operator computation process. - - - - -
    • Atlas inference products, Atlas training products: The 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, end, weight, and out are null pointers.
    ACLNN_ERR_PARAM_INVALID 161002 The data types of self, end, weight, and out are not within the supported range.
    The data types of self, end, weight, and out are inconsistent.
    self, end, and weight cannot be broadcast.
    The shape after broadcasting self, end, and weight is inconsistent with the shape of out.

aclnnLerp

  • 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 from the first-phase API aclnnLerpGetWorkspaceSize.
    Executor Input Operator executor, which contains the operator computation flow.
    stream Input Specifies the stream for task execution.

  • Return value

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

aclnnInplaceLerpGetWorkspaceSize

  • Parameters:

    Parameter Input/Output Description Instruction Data Type Data Format Dimension Non-contiguous Tensor
    selfRef (aclTensor*) Input/Output The input start and output out in the formula. The data type is the same as that of end and weight. The shape must satisfy the broadcast relationship with end and weight, and the shape after broadcast must be the same as that of selfRef. FLOAT, FLOAT16, BFLOAT16 ND 0-8 √
    end (aclTensor*) Input The input end in the formula. The data type is the same as that of selfRef and weight. The shape must satisfy the broadcast relationship with selfRef and weight. FLOAT, FLOAT16, BFLOAT16 ND 0-8 √
    weight (aclTensor*) Input The input weight in the formula. Its data type is the same as that of selfRef and end. Its shape must satisfy the broadcast relationship with selfRef and end. FLOAT, FLOAT16, BFLOAT16 ND 0-8 √
    workspaceSize (uint64_t*) Output Returns the workspace size to be applied for on the Device side. - - - - -
    executor (aclOpExecutor**) Output Returns the op executor, which contains the operator computation flow. - - - - -
    • Atlas inference products, Atlas training products: The 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 passed selfRef, end, and weight are null pointers.
    ACLNN_ERR_PARAM_INVALID 161002 The data types of selfRef, end, and weight are not within the supported range.
    The data types of selfRef, end, and weight are inconsistent.
    selfRef, end, and weight cannot be broadcast.
    The shape after broadcasting selfRef, end, and weight is inconsistent with the shape of selfRef.

aclnnInplaceLerp

  • 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 from the first-phase API aclnnInplaceLerpGetWorkspaceSize.
    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.

Constraints

  • Deterministic computation:

    • aclnnLerp&aclnnInplaceLerp default to a deterministic implementation.

Example

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

aclnnLerp

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_lerp_tensor.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 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.
    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.
    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 = {4, 2};
    std::vector<int64_t> endShape = {4, 2};
    std::vector<int64_t> weightShape = {1};
    std::vector<int64_t> outShape = {4, 2};
    void* selfDeviceAddr = nullptr;
    void* endDeviceAddr = nullptr;
    void* weightDeviceAddr = nullptr;
    void* outDeviceAddr = nullptr;
    aclTensor* self = nullptr;
    aclTensor* end = nullptr;
    aclTensor* weight = nullptr;
    aclTensor* out = nullptr;
    std::vector<float> selfHostData = {1, 2, 3, 4, 5, 6, 7, 8};
    std::vector<float> endHostData = {4, 5, 6, 7, 8, 9, 10, 11};
    std::vector<float> weightHostData = {2};
    std::vector<float> outHostData = {0, 0, 0, 0, 0, 0, 0, 0};
    // Create the self aclTensor.
    ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
    CHECK_RET(ret == ACL_SUCCESS, return ret);
    // Create the end aclTensor.
    ret = CreateAclTensor(endHostData, endShape, &endDeviceAddr, aclDataType::ACL_FLOAT, &end);
    CHECK_RET(ret == ACL_SUCCESS, return ret);
    // Create the weight aclTensor.
    ret = CreateAclTensor(weightHostData, weightShape, &weightDeviceAddr, aclDataType::ACL_FLOAT, &weight);
    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);

    uint64_t workspaceSize = 0;
    aclOpExecutor* executor;

    // 3. Call the CANN operator library API.
    // Call the first-phase API of aclnnLerp.
    ret = aclnnLerpGetWorkspaceSize(self, end, weight, out, &workspaceSize, &executor);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnLerpGetWorkspaceSize 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 aclnnLerp.
    ret = aclnnLerp(workspaceAddr, workspaceSize, executor, stream);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnLerp 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 side.
    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 aclTensor and aclScalar.
    aclDestroyTensor(self);
    aclDestroyTensor(end);
    aclDestroyTensor(weight);
    aclDestroyTensor(out);

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

aclnnInplaceLerp

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_lerp_tensor.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 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 host-side data to 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];
    }

    // 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.
    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 inputs and outputs. Custom construction is required based on the API.
    std::vector<int64_t> selfShape = {4, 2};
    std::vector<int64_t> endShape = {4, 2};
    std::vector<int64_t> weightShape = {1};
    void* selfDeviceAddr = nullptr;
    void* endDeviceAddr = nullptr;
    void* weightDeviceAddr = nullptr;
    aclTensor* self = nullptr;
    aclTensor* end = nullptr;
    aclTensor* weight = nullptr;
    std::vector<float> selfHostData = {1, 2, 3, 4, 5, 6, 7, 8};
    std::vector<float> endHostData = {4, 5, 6, 7, 8, 9, 10, 11};
    std::vector<float> weightHostData = {2};
    // Create the self aclTensor.
    ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
    CHECK_RET(ret == ACL_SUCCESS, return ret);
    // Create the end aclTensor.
    ret = CreateAclTensor(endHostData, endShape, &endDeviceAddr, aclDataType::ACL_FLOAT, &end);
    CHECK_RET(ret == ACL_SUCCESS, return ret);
    // Create the weight aclTensor.
    ret = CreateAclTensor(weightHostData, weightShape, &weightDeviceAddr, aclDataType::ACL_FLOAT, &weight);
    CHECK_RET(ret == ACL_SUCCESS, return ret);

    uint64_t workspaceSize = 0;
    aclOpExecutor* executor;

    // 3. Call the CANN operator library API.
    // Call the first-phase API of aclnnInplaceLerp.
    ret = aclnnInplaceLerpGetWorkspaceSize(self, end, weight, &workspaceSize, &executor);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnInplaceLerpGetWorkspaceSize 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 aclnnInplaceLerp.
    ret = aclnnInplaceLerp(workspaceAddr, workspaceSize, executor, stream);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnInplaceLerp 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.
    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 aclTensor and aclScalar.
    aclDestroyTensor(self);
    aclDestroyTensor(end);
    aclDestroyTensor(weight);

    // 7. Release device resources. Modify based on the specific API definition.
    aclrtFree(selfDeviceAddr);
    aclrtFree(endDeviceAddr);
    aclrtFree(weightDeviceAddr);
    if (workspaceSize > 0) {
    aclrtFree(workspaceAddr);
    }
    aclrtDestroyStream(stream);
    aclrtResetDevice(deviceId);
    aclFinalize();
    return 0;
}
This site uses cookies to improve your browsing experience. By continuing to browse this site, you agree to our use of cookies.View Details