aclnnSwishBackward

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: aclnnSwishBackward is the backward propagation of the aclnnSwish activation function, used to compute the gradient of the Swish activation function.
Formula:
- Swish function formula
$s(x) = x*\sigma(\beta x)$
- Derivative implementation of the Swish function formula
$s^\prime(x)= \beta s(x)+\sigma(\beta x)(1-\beta s(x))= \sigma(\beta x)*(1+\beta x(1-\sigma(\beta x)))$ $gradInput = gradOutput * s^\prime(x)$ $\sigma(x) = {\frac{1} {1+{e}^{-x}}}$
Where $\sigma(x)$ is the Sigmoid function, $s(x)$ is the Swish function, and $s^\prime(x)$ is the derivative of the Swish function.

Function Prototype

Each operator is divided into a two-phase API. The "aclnnSwishBackwardGetWorkspaceSize" API must be called first to obtain the workspace size required for computation and the executor that encapsulates the operator computation flow. Then, the "aclnnSwishBackward" API is called to perform the computation.

aclnnStatus aclnnSwishBackwardGetWorkspaceSize(
  const aclTensor* gradOutput,
  const aclTensor* self,
  const aclScalar* betaOptional,
  aclTensor*       gradInput,
  uint64_t*        workspaceSize,
  aclOpExecutor**  executor)

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

aclnnSwishBackwardGetWorkspaceSize

Parameters:

Parameter	Input/Output	Description	Instruction	Data Type	Data Format	Dimension (Shape)	Non-contiguous Tensor
gradOutput (aclTensor*)	Input	Represents the gradient of the forward output of the Swish activation function, i.e., gradOutput in the formula.	Supports empty tensors. gradOutput, self, and gradInput have the same shape. gradOutput, self, and gradInput have the same data type.	BFLOAT16, FLOAT16, FLOAT	ND	0-8	√
self (aclTensor*)	Input	Represents the tensor used to compute the activation function, i.e., x in the formula.	Supports empty tensors. The shapes of gradOutput, self, and gradInput are consistent. The data types of gradOutput, self, and gradInput are consistent.	BFLOAT16, FLOAT16, FLOAT	ND	0-8	√
betaOptional (aclScalar*)	Input	A scalar that represents an adjustable parameter for controlling the shape and slope of the Swish function, i.e., β in the formula.	The data type must be convertible to FLOAT (see Deduction Relationship). When betaOptional is a null pointer, the API performs computation with 1.0.	-	-	-	-
gradInput (aclTensor*)	Output	Output of the backward computation, representing the gradient of the Swish forward input, i.e., the result of taking the derivative with respect to the input. Corresponds to gradInput in the formula.	Supports empty tensors. gradOutput, self, and gradInput have the same shape. gradOutput, self, and gradInput have the same data type.	BFLOAT16, FLOAT16, FLOAT	ND	0-8	√
workspaceSize (uint64_t*)	Output	Returns the workspace size to be allocated on the Device side.	-	-	-	-	-
executor (aclOpExecutor**)	Output	Returns the operator executor, which contains the operator computation flow.	-	-	-	-	-

Atlas inference products and Atlas training products: The BFLOAT16 data type is not supported.

Return value

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

The first-phase API performs input parameter validation and returns an error in the following scenarios:

Return Code	Error Code	Description
ACLNN_ERR_PARAM_NULLPTR	161001	The passed gradOutput, self, or gradInput is a null pointer.
ACLNN_ERR_PARAM_INVALID	161002	The data type of gradOutput, self, betaOptional, or gradInput is not within the supported range.
		The data types of gradOutput, self, and gradInput differ.
		The shapes of gradOutput, self, and gradInput differ.

aclnnSwishBackward

Parameters:

Parameter	Input/Output	Description
workspace	Input	Address of the workspace memory applied for on the device.
workspaceSize	Input	Size of the workspace applied for on the device, obtained through the first-phase API aclnnSwishBackwardGetWorkspaceSize.
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 Code.

Constraints

Deterministic computation:
- aclnnSwishBackward defaults to a deterministic implementation.

Example

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

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_swish_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 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 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 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);
    // Copy host data to device 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_ND, shape.data(), shape.size(),
        *deviceAddr);
    return 0;
}

int main()
{
    // 1. (Boilerplate) Initialize the device/stream. See the ACL API manual.
    // Set deviceId based on the 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> gradOutputShape = {2, 3};
    std::vector<int64_t> selfShape = {2, 3};
    std::vector<int64_t> gradInputShape = {2, 3};
    void* gradOutDeviceAddr = nullptr;
    void* selfDeviceAddr = nullptr;
    void* gradInputDeviceAddr = nullptr;
    aclTensor* gradOut = nullptr;
    aclTensor* self = nullptr;
    aclScalar* betaOptional = nullptr;
    aclTensor* gradInput = nullptr;
    std::vector<float> gradOutHostData = {1, 1, 1, 1, 1, 1};
    std::vector<float> selfHostData = {1, 2, 3, 4, 5, 6};
    std::vector<float> gradInputHostData = {0, 0, 0, 0, 0, 0};
    float betaValue = 1.1f;

    // Create the gradOut aclTensor.
    ret = CreateAclTensor(gradOutHostData, gradOutputShape, &gradOutDeviceAddr, aclDataType::ACL_FLOAT, &gradOut);
    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // Create the self aclTensor.
    ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // Create the betaOptional aclScalar.
    betaOptional = aclCreateScalar(&betaValue, aclDataType::ACL_FLOAT);
    CHECK_RET(betaOptional != nullptr, return ret);

    // Create the gradInput aclTensor.
    ret = CreateAclTensor(gradInputHostData, gradInputShape, &gradInputDeviceAddr, aclDataType::ACL_FLOAT, &gradInput);
    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 3. Call the CANN operator library API. Replace with the actual API name.
    uint64_t workspaceSize = 0;
    aclOpExecutor* executor;
    // Call the first-phase API of aclnnSwishBackward.
    ret = aclnnSwishBackwardGetWorkspaceSize(gradOut, self, betaOptional, gradInput, &workspaceSize, &executor);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnSwishBackwardGetWorkspaceSize 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 aclnnSwishBackward.
    ret = aclnnSwishBackward(workspaceAddr, workspaceSize, executor, stream);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnSwishBackward 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-side memory to the host side. Modify based on the specific API definition.
    auto size = GetShapeSize(gradInputShape);
    std::vector<float> outData(size, 0);
    ret = aclrtMemcpy(
        outData.data(), outData.size() * sizeof(outData[0]), gradInputDeviceAddr, size * sizeof(outData[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("out result[%ld] is: %f\n", i, outData[i]);
    }

    // 6. Release aclTensor and aclScalar. Modify based on the specific API definition.
    aclDestroyTensor(gradOut);
    aclDestroyTensor(self);
    aclDestroyScalar(betaOptional);
    aclDestroyTensor(gradInput);

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