aclnnUpsampleBilinear2dAA

Applicable Products

Product	Supported
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: Applies 2D bilinear anti-aliasing upsampling to an input signal consisting of multiple input channels.
Corresponding Formula: For a 2D interpolation point $(N, C, H, W)$ , the interpolated value $I(N, C, H, W)$ can be expressed as:
${I(N, C, H, W)} = \sum_{i=0}^{kW}\sum_{j=0}^{kH}{w(i) * w(j)} * {f(h_i, w_j)}/\sum_{i=0}^{kW}w(i)/\sum_{j=0}^{kH}w(j)$ $scaleH =\begin{cases} (input.dim(2)-1) / (outputSize[0]-1) & alignCorners=true \\ 1 / scalesH & alignCorners=false\&scalesH>0\\ input.dim(2) / outputSize[0] & otherwise \end{cases}$ $scaleW =\begin{cases} (input.dim(3)-1) / (outputSize[1]-1) & alignCorners=true \\ 1 / scalesW & alignCorners=false\&scalesW>0\\ input.dim(3) / outputSize[1] & otherwise \end{cases}$
Where:
- $kW$ and $kH$ represent the number of points that affect the interpolation point size in the W and H dimensions, respectively.
- If $scaleH >= 1$ , then $kH = floor(scaleH) * 2 + 1$ ; otherwise, $kH = 3$ .
- If $scaleW >= 1$ , then $kW = floor(scaleW) * 2 + 1$ ; otherwise, $kW = 3$ .
- $f(h_i, w_j)$ is the pixel value of the original image at $(h_i, w_j)$
- $w(i)$ and $w(j)$ are the weights in the W and H directions for bilinear anti-aliasing interpolation, calculated as follows: $w(i) = \begin{cases} 1 - |h_i - h| & |h_i -h| < 1 \\ 0 & otherwise \end{cases}$ $w(j) = \begin{cases} 1 - |w_j - w| & |w_j -w| < 1 \\ 0 & otherwise \end{cases}$

Function Prototype

Each operator is divided into a two-phase API. You must first call the "aclnnUpsampleBilinear2dAAGetWorkspaceSize" API to obtain the workspace size required for computation and the executor that contains the operator computation process, and then call the "aclnnUpsampleBilinear2dAA" API to execute the computation.

aclnnStatus aclnnUpsampleBilinear2dAAGetWorkspaceSize(
  const aclTensor   *input,
  const aclIntArray *outputSize,
  bool               alignCorners,
  double             scalesH,
  double             scalesW,
  aclTensor         *out,
  uint64_t          *workspaceSize,
  aclOpExecutor    **executor)

aclnnStatus aclnnUpsampleBilinear2dAA(
  void          *workspace,
  uint64_t       workspaceSize,
  aclOpExecutor *executor,
  aclrtStream    stream)

aclnnUpsampleBilinear2dAAGetWorkspaceSize

Parameters

Parameter Name	Input/Output	Description	Usage Notes	Data Type	Data Format	Dimensions (Shape)	Non-Contiguous Tensor
input (aclTensor*)	Input	Input tensor for sampling, corresponding to `input` in the formula.	Empty tensors are not supported. When the data format is ND, it is processed in NCHW format by default.	FLOAT32, FLOAT16, BFLOAT16	NCHW, ND	4	√
outputSize (aclIntArray*)	Input	Specifies the output spatial size, indicating the spatial size of `out` in the H and W dimensions. Corresponds to `outputSize` in the formula.	The size is 2, and each element is greater than zero.	INT64	-	-	-
alignCorners (bool)	Input	Determines whether to align corner pixels, corresponding to `alignCorners` in the formula.	If set to `true`, the input and output tensors are aligned by the center points of their corner pixels, preserving the values at the corner pixels. If set to `false`, the input and output tensors are aligned by the corner points of their corner pixels, and edge values are used to pad values outside the boundaries.	-	-	-	-
scalesH (double)	Input	The height dimension multiplier for the specified spatial size, corresponding to `scalesH` in the formula.	Negative values are not allowed.	-	-	-	-
scalesW (double)	Input	Specifies the multiplier for the width dimension of the spatial size, corresponding to `scalesW` in the formula.	Negative values are not allowed.	-	-	-	-
out (aclTensor*)	Output	Represents the output tensor after sampling, corresponding to `I` in the formula.	Empty tensors are not supported. The data type, data format, and shape must be consistent with the input parameter `input`.	FLOAT32, FLOAT16, BFLOAT16	NCHW, ND	4	√
workspaceSize (uint64_t*)	Output	Size of the workspace to be requested on the Device side.	-	-	-	-	-
executor (aclOpExecutor**)	Output	Returns the op executor, which contains the operator computation flow.	-	-	-	-	-

Return Value

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

The first API performs input parameter validation and reports an error in the following scenarios:

Return Code	Error Code	Description
ACLNN_ERR_PARAM_NULLPTR	161001	The input, outputSize, or out passed in is a null pointer.
ACLNN_ERR_PARAM_INVALID	161002	The data type of input or out is not within the supported range.
		The data types of input and out are inconsistent.
		The data formats of input and out are inconsistent.
		The shape of input is not 4-dimensional.
		The value of outputSize is less than or equal to 0.
		The value of scalesH or scalesW is less than 0.

aclnnUpsampleBilinear2dAA

Parameters

Parameter Name	Input/Output	Description
workspace	Input	Address of the workspace memory applied for on the Device side.
workspaceSize	Input	Size of the workspace applied on the Device side, obtained through the first-phase API aclnnUpsampleBilinear2dAAGetWorkspaceSize.
executor	Input	Operator executor, which contains the operator computation flow.
stream	Input	Specifies the stream for task execution.

Return Value

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

Constraints

Atlas A3 training products/Atlas A3 inference products, Atlas A2 training products/Atlas A2 inference products:

In the input data scaling scenario, the downscaling factor must be less than or equal to 50, i.e.:
$H\_input / outputSize\_H \leq 50$ $W\_input / outputSize\_W \leq 50$
The parameters input, outputSize, scalesH, and scalesW must satisfy the following constraints:
$outputSize\_H = floor(input\_H * scalesH)$ $outputSize\_W = floor(input\_W * scalesW)$
Deterministic computation:
- aclnnUpsampleBilinear2dAA uses deterministic implementation by default.

Example

The following sample code is for reference only. For details about compilation and execution, see Compile and Run Samples.

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_upsample_bilinear2d_aa.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)
{
    // Standard 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 CreateAclNchwTensor(const std::vector<T> &hostData, const std::vector<int64_t> &shape, void **deviceAddr,
    aclDataType dataType, aclTensor **tensor)
{
    auto size = GetShapeSize(shape) * sizeof(T);
    // Allocate device-side memory using aclrtMalloc.
    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-side memory using 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 the aclCreateTensor API to create an aclTensor.
    *tensor = aclCreateTensor(shape.data(),
        shape.size(),
        dataType,
        strides.data(),
        0,
        aclFormat::ACL_FORMAT_NCHW,
        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 input and output. Customize the construction based on the API.
    std::vector<int64_t> inputShape = {1, 1, 4, 2};
    std::vector<int64_t> outShape = {1, 1, 8, 4};
    void *inputDeviceAddr = nullptr;
    void *outDeviceAddr = nullptr;
    aclTensor *input = nullptr;
    aclTensor *out = nullptr;
    std::vector<float> inputHostData = {0, 1, 2, 3, 4, 5, 6, 7};
    std::vector<float> outHostData(32, 0);
    std::vector<int64_t> outputSize = {8, 4};
    bool alignCorners = true;
    double scalesH = 2.0;
    double scalesW = 2.0;
    // Create input aclTensor.
    ret = CreateAclNchwTensor(inputHostData, inputShape, &inputDeviceAddr, aclDataType::ACL_FLOAT, &input);
    CHECK_RET(ret == ACL_SUCCESS, return ret);
    // Create input aclIntArray.
    auto outputSizeArray = aclCreateIntArray(outputSize.data(), 2);
    // Create out aclTensor.
    ret = CreateAclNchwTensor(outHostData, outShape, &outDeviceAddr, aclDataType::ACL_FLOAT, &out);
    CHECK_RET(ret == ACL_SUCCESS, 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 aclnnUpsampleBilinear2dAA.
    ret = aclnnUpsampleBilinear2dAAGetWorkspaceSize(
        input, outputSizeArray, alignCorners, scalesH, scalesW, out, &workspaceSize, &executor);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnUpsampleBilinear2dAAGetWorkspaceSize 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 aclnnUpsampleBilinear2dAA.
    ret = aclnnUpsampleBilinear2dAA(workspaceAddr, workspaceSize, executor, stream);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnUpsampleBilinear2dAA 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 device memory to host memory.
    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. Modify based on the specific API definition.
    aclDestroyTensor(input);
    aclDestroyTensor(out);

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

    return 0;
}