Documentation
Download

aclnnRound&aclnnInplaceRound

📄 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: Rounds the input value to the nearest integer. If the value is equidistant from two integers, round it to the even integer.

Prototype

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

    • aclnnRound: An output tensor object needs to be created to store the computation result.
    • aclnnInplaceRound: 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, aclnnRoundGetWorkspaceSize or aclnnInplaceRoundGetWorkspaceSize is called to obtain input parameters and compute the required workspace size based on the computation process. Then, aclnnRound or aclnnInplaceRound is called to perform computation.

    • aclnnStatus aclnnRoundGetWorkspaceSize(const aclTensor *self, aclTensor *out, uint64_t *workspaceSize, aclOpExecutor **executor)

    • aclnnStatus aclnnRound(void* workspace, uint64_t workspaceSize, aclOpExecutor* executor, const aclrtStream stream)

    • aclnnStatus aclnnInplaceRoundGetWorkspaceSize(const aclTensor *selfRef, uint64_t *workspaceSize, aclOpExecutor **executor)

    • aclnnStatus aclnnInplaceRound(void* workspace, uint64_t workspaceSize, aclOpExecutor* executor, const aclrtStream stream)

aclnnRoundGetWorkspaceSize

  • Parameters:

    • self (aclTensor*, compute input): aclTensor on the device. The data format supports ND, and the shape does not support 9D or higher.
      • Atlas training products: The data type can be FLOAT16, FLOAT32, DOUBLE, 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 BFLOAT16, FLOAT16, FLOAT32, DOUBLE, INT32, or INT64.
    • out (aclTensor *, compute output): aclTensor on the device. The shape and dtype are the same as those of self. The data format supports ND.
      • Atlas training products: The data type can be FLOAT16, FLOAT32, DOUBLE, 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 BFLOAT16, FLOAT16, FLOAT32, DOUBLE, INT32, or INT64.
    • 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 tensor is a null pointer.
161002 (ACLNN_ERR_PARAM_INVALID): 1. The data type or format of self or out is not supported.
                                 2. The shapes of self and out are inconsistent.
                                 3. The dtypes of self and out are inconsistent.

aclnnRound

  • 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 the first-phase API aclnnRoundGetWorkspaceSize.
    • executor (aclOpExecutor*, input): operator executor, including the operator computation process.
    • stream (aclrtStream, input): stream for executing the task.

  • Returns:

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

aclnnInplaceRoundGetWorkspaceSize

  • Parameters:

    • selfRef (aclTensor*, compute input): aclTensor on the device. The data format supports ND, and the shape does not support 9D or higher.
      • Atlas training products: The data type can be FLOAT16, FLOAT32, DOUBLE, 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 BFLOAT16, FLOAT16, FLOAT32, DOUBLE, INT32, or INT64.
    • 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): The input tensor is a null pointer.
561101 (ACLNN_ERR_INNER_CREATE_EXECUTOR): 1. Inner error. Failed to create the executor.

aclnnInplaceRound

  • 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 the first-phase API aclnnInplaceRoundGetWorkspaceSize.
    • 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 compute:
    • aclnnRound&aclnnInplaceRound defaults to a deterministic implementation.
  • If the input value is in the range of [–0.5, –0], the output value is 0.

Example

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

#include "acl/acl.h"
#include "aclnnop/aclnn_round.h"
#include <iostream>
#include <vector>

#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 shape_size = 1;
    for (auto i : shape) {
        shape_size *= i;
    }
    return shape_size;
}

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 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.
    // Set the device ID in use.
    int32_t deviceId = 0;
    aclrtStream stream;
    auto ret = Init(deviceId, &stream);
    // Handle the check as required.
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);
    // 2. Construct inputs and outputs based on the API definition.
    std::vector<int64_t> selfShape = {1, 3};
    std::vector<int64_t> outShape = {1, 3};

    void* selfDeviceAddr = nullptr;
    void* outDeviceAddr = nullptr;
    aclTensor* self = nullptr;
    aclTensor* out = nullptr;

    std::vector<double> selfHostData = {1.1, 2.8, 3.5};
    std::vector<double> outHostData = {0, 0, 0};

    // Create a self aclTensor.
    ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_DOUBLE, &self);
    CHECK_RET(ret == ACL_SUCCESS, return ret);
    // Create an out aclTensor.
    ret = CreateAclTensor(outHostData, outShape, &outDeviceAddr, aclDataType::ACL_DOUBLE, &out);
    CHECK_RET(ret == ACL_SUCCESS, return ret);

    LOG_PRINT("Example of aclnnRound API call \n");
    uint64_t workspaceSize = 0;
    aclOpExecutor* executor;
    // aclnnRound API call example
    // 3. Call the first-phase API of aclnnRound.
    ret = aclnnRoundGetWorkspaceSize(self, out, &workspaceSize, &executor);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnRoundGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);
    // Allocate device memory based on 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 aclnnRound.
    ret = aclnnRound(workspaceAddr, workspaceSize, executor, stream);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnRound 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 to the host. Modify the code based on the API definition.
    auto size = GetShapeSize(outShape);
    std::vector<double> 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]);
    }

    LOG_PRINT("Example of aclnnInplaceRound API call \n");
    // aclnnInplaceRound API call example
    // 3. Call the first-phase API of aclnnInplaceRound.
    ret = aclnnInplaceRoundGetWorkspaceSize(self, &workspaceSize, &executor);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnInplaceRoundGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);
    // Allocate device memory based on workspaceSize calculated by the first-phase API.
    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 aclnnInplaceRound.
    ret = aclnnInplaceRound(workspaceAddr, workspaceSize, executor, stream);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnInplaceRound 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. Modify the code based on the API definition.
    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. Modify the code based on the API definition.
    aclDestroyTensor(self);
    aclDestroyTensor(out);

    // 7. Release device resources.
    aclrtFree(selfDeviceAddr);
    aclrtFree(outDeviceAddr);
    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