aclnnUpsampleNearestExact1dBackward

📄 View Source Code

Applicable Products

Function

• Description: backward propagation of aclnnUpsampleNearestExact1d. Maps points from the output gradient tensor to positions in the input gradient tensor, and accumulates the output gradient values onto the input gradient tensor.
• Formula:$$gradInput(N, C, floor ( scales * ( L + 0.5 ))) += gradOutput( N, C, L)$$

Function Prototype

Each operator is divided into a two-phase API. You must first call the aclnnUpsampleNearestExact1dBackwardGetWorkspaceSize API to obtain the workspace size required for computation and the executor that includes the operator computation process, and then call the aclnnUpsampleNearestExact1dBackward API to perform the computation.

aclnnStatus aclnnUpsampleNearestExact1dBackwardGetWorkspaceSize(
  const aclTensor   *gradOutput,
  const aclIntArray *outputSize,
  const aclIntArray *inputSize,
  double             scales,
  aclTensor         *out,
  uint64_t          *workspaceSize,
  aclOpExecutor    **executor)

aclnnStatus aclnnUpsampleNearestExact1dBackward(
  void             *workspace,
  uint64_t          workspaceSize,
  aclOpExecutor    *executor,
  aclrtStream       stream)

aclnnUpsampleNearestExact1dBackwardGetWorkspaceSize

• Parameters

  Parameter Name	Input/Output	Description	Usage Notes	Data Type	Data Format	Dimension (shape)	Non-contiguous Tensor
  gradOutput (aclTensor*)	Input	The input `gradOutput` in the formula represents the gradient tensor for backward computation.	Empty tensors are not supported. When the data format is ND, it is processed in NCL format by default.	FLOAT32, FLOAT16, BFLOAT16	NCL, ND	3	√
  outputSize (aclIntArray*)	Input	Indicates the spatial size of the input gradOutput in the L dimension.	The size is 1, and the value must be greater than zero.	INT64	-	-	-
  inputSize (aclIntArray*)	Input	Indicates the spatial sizes of the output out in the N, C, and L dimensions, respectively.	The size is 3, and each element is greater than zero.	INT64	-	-	-
  scales (double)	Input	Indicates the scaling factor for the output out.	Negative values are not allowed.	-	-	-	-
  out (aclTensor*)	Output	The output `gradInput` in the formula, representing the output tensor of the backward computation.	Empty tensors are not supported. The data type and data format must be consistent with the input `gradOutput`. The N and C axes of the shape must be consistent with the input `gradOutput`.	FLOAT32, FLOAT16, BFLOAT16	NCL, ND	3	√
  workspaceSize (uint64_t*)	Output	Returns the workspace size that the user needs to apply for on the Device side.	-	-	-	-	-
  executor (aclOpExecutor**)	output	Returns the operator executor, which includes the computation process of the operator.	-	-	-	-	-

• Return Value

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

  The first-phase 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 parameter is a required input, output, or required attribute, and is a null pointer.
  ACLNN_ERR_PARAM_INVALID	161002	The data type of gradOutput or out is not within the supported range.
  		The data type of gradOutput is inconsistent with that of out.
  		The data format of gradOutput is inconsistent with that of out.
  		The shape of gradOutput is not 3-dimensional.
  		The value of the L axis of inputSize is less than 1.
  		The value of scales is less than 0.

aclnnUpsampleNearestExact1dBackward

• Parameters

  Parameter Name	Input/Output	Description
  workspace	Input	Workspace memory address applied on the Device side.
  workspaceSize	Input	Size of the workspace applied on the Device side, obtained through the first-phase API aclnnUpsampleNearestExact1dBackwardGetWorkspaceSize.
  executor	Input	Operator executor, which includes the operator computation flow.
  stream	Input	Specifies the stream for task execution.

• Return Value

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

Constraints

• In the input data scaling scenario, the scaling factor must be less than or equal to 50, that is:

  $$outputSize[0] / height L of the output shape <= 50$$

• The parameters inputSize, outputSize, and scales must satisfy the following constraint:

  $$outputSize = floor(inputSize\_L * scales)$$

• Deterministic computation:

  • aclnnUpsampleNearestExact1dBackward 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_nearest_exact1d_backward.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 CreateAclNchTensor(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-side 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-side data to the device-side 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 for 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_NCL,
        shape.data(),
        shape.size(),
        *deviceAddr);
    return 0;
}

int main()
{
    // 1. (Boilerplate) Initialize 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 input and output. Customize the construction based on the API.
    std::vector<int64_t> inputShape = {1, 1, 2};
    std::vector<int64_t> outShape = {1, 1, 4};
    void *inputDeviceAddr = nullptr;
    void *outDeviceAddr = nullptr;
    aclTensor *input = nullptr;
    aclTensor *out = nullptr;
    std::vector<float> inputHostData = {1, 2};
    std::vector<float> outHostData(4, 0);
    std::vector<int64_t> outputSize = {2};
    std::vector<int64_t> inputSize = {1, 1, 4};
    double scales = 0.5;

    // Create the input aclTensor.
    ret = CreateAclNchTensor(inputHostData, inputShape, &inputDeviceAddr, aclDataType::ACL_FLOAT, &input);
    CHECK_RET(ret == ACL_SUCCESS, return ret);
    // Create the input aclIntArray.
    auto outputSizeArray = aclCreateIntArray(outputSize.data(), 1);
    auto inputSizeArray = aclCreateIntArray(inputSize.data(), 3);
    // Create the output aclTensor.
    ret = CreateAclNchTensor(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 aclnnUpsampleNearestExact1dBackward.
    ret = aclnnUpsampleNearestExact1dBackwardGetWorkspaceSize(
        input, outputSizeArray, inputSizeArray, scales, out, &workspaceSize, &executor);
    CHECK_RET(
        ret == ACL_SUCCESS, LOG_PRINT("aclnnUpsampleNearestExact1dBackwardGetWorkspaceSize 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 aclnnUpsampleNearestExact1dBackward.
    ret = aclnnUpsampleNearestExact1dBackward(workspaceAddr, workspaceSize, executor, stream);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnUpsampleNearestExact1dBackward 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 memory. Modify based on the specific 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 aclTensor and aclIntArray. 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;
}