Documentation
Download

aclnnLerps &·aclnnInplaceLerps

📄 View source code

Supported Products

ProductSupported
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: Perform linear interpolation between the start and end tensors based on the given weight and return the interpolated tensor.

  • Formula:

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

Prototype

  • aclnnLerps and aclnnInplaceLerps implement the same function in different ways. Select a proper operator based on your requirements.

    • aclnnLerps: An output tensor object needs to be created to store the computation result.
    • aclnnInplaceLerps: 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, aclnnLerpsGetWorkspaceSize or aclnnInplaceLerpsGetWorkspaceSize is called to obtain the workspace size required for computation and the executor that contains the operator computation process. Then, aclnnLerps or aclnnInplaceLerps is called to perform computation.

    • aclnnStatus aclnnLerpsGetWorkspaceSize(const aclTensor* self, const aclTensor* end, const aclScalar* weight, aclTensor* out, uint64_t* workspaceSize, aclOpExecutor** executor)
    • aclnnStatus aclnnLerps(void* workspace, uint64_t workspaceSize, aclOpExecutor* executor, const aclrtStream stream)
    • aclnnStatus aclnnInplaceLerpsGetWorkspaceSize(aclTensor* selfRef, const aclTensor* end, const aclScalar* weight, uint64_t* workspaceSize, aclOpExecutor** executor)
    • aclnnStatus aclnnInplaceLerps(void* workspace, uint64_t workspaceSize, aclOpExecutor* executor, const aclrtStream stream)

aclnnLerpsGetWorkspaceSize

  • Parameters

    • self (aclTensor*, computation input): tensor that contains the start value, input start in the formula, and aclTensor on the device. The data type of self is the same as that of end, and the shapes of self and end must meet the broadcast relationship. Non-contiguous tensors are supported. The data format can be ND.
      • Atlas inference products and Atlas training products: The data type can be FLOAT16 or FLOAT.
      • Atlas A2 training products/Atlas A2 inference products and Atlas A3 training products/Atlas A3 inference products: The data type can be BFLOAT16, FLOAT16, or FLOAT.
    • end (aclTensor*, computation input): tensor that contains the end value, input end in the formula, aclTensor on the device. The data type of self is the same as that of end, and the shapes of self and end must meet the broadcast relationship. Non-contiguous tensors are supported. The data format can be ND.
      • Atlas inference products and Atlas training products: The data type can be FLOAT16 or FLOAT.
      • Atlas A2 training products/Atlas A2 inference products and Atlas A3 training products/Atlas A3 inference products: The data type can be BFLOAT16, FLOAT16, or FLOAT.
    • weight (aclScalar*, computation input): weight, input weight in the formula, aclScalar on the host. The data type must be a data type that can be converted to that of self (for details, see Conversion Relationship).
    • out (aclTensor*, computation output): tensor that contains the result, out in the formula, aclTensor on the device. The shape of out is the same as the tensor shape obtained by broadcasting self and end. 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, or DOUBLE.
      • Atlas A2 training products/Atlas A2 inference products and Atlas A3 training products/Atlas A3 inference products: The data type can be BFLOAT16, FLOAT16, FLOAT, or DOUBLE.
    • 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.

   The first-phase API implements input parameter verification. The following errors may be thrown:
 161001 (ACLNN_ERR_PARAM_NULLPTR): 1. The passed self, end, weight, or out is a null pointer.
 161002 (ACLNN_ERR_PARAM_INVALID): 1. The data type of self, end, or out is not supported.
                                       2. The data types of self and end are inconsistent.
                                       3. Broadcasting cannot be performed between self and end.
                                       4. The shapes of self and end after broadcasting are different from that of out.

aclnnLerps

  • 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 aclnnLerpsGetWorkspaceSize.
    • 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.

aclnnInplaceLerpsGetWorkspaceSize

  • Parameters

    • selfRef (aclTensor*, computation input/output): tensor containing the start value and tensor containing the result, start and out in the formula, aclTensor on the device. The data types of selfRef and end are the same. The shapes of selfRef and end meet the broadcast relationship, and the shape after broadcasting is the same as that of selfRef. Non-contiguous tensors are supported. The data format can be ND.
      • Atlas inference products and Atlas training products: The data type can be FLOAT16 or FLOAT.
      • Atlas A2 training products/Atlas A2 inference products and Atlas A3 training products/Atlas A3 inference products: The data type can be BFLOAT16, FLOAT16, or FLOAT.
    • end (aclTensor*, computation input): tensor that contains the end value, input end in the formula, and aclTensor on the device. The data type of selfRef is the same as that of end, and the shapes of selfRef and end must meet the broadcast relationship. Non-contiguous tensors are supported. The data format can be ND.
      • Atlas inference products and Atlas training products: The data type can be FLOAT16 or FLOAT.
      • Atlas A2 training products/Atlas A2 inference products and Atlas A3 training products/Atlas A3 inference products: The data type can be BFLOAT16, FLOAT16, or FLOAT.
    • weight (aclScalar*, computation input): weight, input weight in the formula, aclScalar on the host. The data type must be a data type that can be converted to that of selfRef (for details, see Conversion Relationship).
    • 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.

   The first-phase API implements input parameter verification. The following errors may be thrown:
 161001 (ACLNN_ERR_PARAM_NULLPTR): 1. The passed selfRef, end, or weight is a null pointer.
 161002 (ACLNN_ERR_PARAM_INVALID): 1. The data type of selfRef or end is not supported.
                                       2. The data types of selfRef and end are inconsistent.
                                       3. Broadcasting cannot be performed between selfRef and end.
                                       4. The shapes of selfRef and end after broadcasting are different from the shape of selfRef.

aclnnInplaceLerps

  • 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 aclnnInplaceLerpsGetWorkspaceSize.
    • 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:
    • aclnnLerps & aclnnInplaceLerps default to deterministic implementation.

Example

The following example is for reference only. For details, see Compilation and Running Sample.

aclnnLerps

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_lerp_scalar.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 memory on the device.
    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 the data from the host to the memory on the device.
    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 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 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 and stream. For details, 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 inputs and outputs based on the API definition.
    std::vector<int64_t> selfShape = {4, 2};
    std::vector<int64_t> endShape = {4, 2};
    std::vector<int64_t> outShape = {4, 2};
    void* selfDeviceAddr = nullptr;
    void* endDeviceAddr = nullptr;
    void* outDeviceAddr = nullptr;
    aclTensor* self = nullptr;
    aclTensor* end = nullptr;
    aclScalar* 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> outHostData = {0, 0, 0, 0, 0, 0, 0, 0};
    float weightValue = 2.0f;
    // Create a self aclTensor.
    ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
    CHECK_RET(ret == ACL_SUCCESS, return ret);
    // Create an end aclTensor.
    ret = CreateAclTensor(endHostData, endShape, &endDeviceAddr, aclDataType::ACL_FLOAT, &end);
    CHECK_RET(ret == ACL_SUCCESS, return ret);
    //Create a weight aclScalar.
    weight = aclCreateScalar(&weightValue, aclDataType::ACL_FLOAT);
    CHECK_RET(weight != nullptr, return ret);
    // Create an 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.
    uint64_t workspaceSize = 0;
    aclOpExecutor* executor;
    // Call the first-phase API of aclnnLerps.
    ret = aclnnLerpsGetWorkspaceSize(self, end, weight, out, &workspaceSize, &executor);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnLerpsGetWorkspaceSize 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 aclnnLerps.
    ret = aclnnLerps(workspaceAddr, workspaceSize, executor, stream);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnLerps failed. ERROR: %d\n", ret); return ret);

    // 4. (Boilerplate) Wait until the task execution is 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(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. Destroy aclTensor and aclScalar.
    aclDestroyTensor(self);
    aclDestroyTensor(end);
    aclDestroyScalar(weight);
    aclDestroyTensor(out);

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

    return 0;
}

aclnnInplaceLerps

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_lerp_scalar.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 memory on the device.
    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 the data from the host to the memory on the device.
    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 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 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 and stream. For details, 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 inputs and outputs based on the API definition.
    std::vector<int64_t> selfShape = {4, 2};
    std::vector<int64_t> endShape = {4, 2};
    void* selfDeviceAddr = nullptr;
    void* endDeviceAddr = nullptr;
    aclTensor* self = nullptr;
    aclTensor* end = nullptr;
    aclScalar* 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};
    float weightValue = 2.0f;
    // Create a self aclTensor.
    ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
    CHECK_RET(ret == ACL_SUCCESS, return ret);
    // Create an end aclTensor.
    ret = CreateAclTensor(endHostData, endShape, &endDeviceAddr, aclDataType::ACL_FLOAT, &end);
    CHECK_RET(ret == ACL_SUCCESS, return ret);
    //Create a weight aclScalar.
    weight = aclCreateScalar(&weightValue, aclDataType::ACL_FLOAT);
    CHECK_RET(weight != nullptr, return ret);

    // 3. Call the CANN operator library API.
    uint64_t workspaceSize = 0;
    aclOpExecutor* executor;
    // Call the first-phase API of aclnnInplaceLerps.
    ret = aclnnInplaceLerpsGetWorkspaceSize(self, end, weight, &workspaceSize, &executor);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnLerpsGetWorkspaceSize 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 aclnnInplaceLerps.
    ret = aclnnInplaceLerps(workspaceAddr, workspaceSize, executor, stream);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnLerps failed. ERROR: %d\n", ret); return ret);

    // 4. (Boilerplate) Wait until the task execution is 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. Destroy aclTensor and aclScalar.
    aclDestroyTensor(self);
    aclDestroyTensor(end);
    aclDestroyScalar(weight);

    // 7. Release device resources. Modify the configuration based on the API definition.
    aclrtFree(selfDeviceAddr);
    aclrtFree(endDeviceAddr);
    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