aclnnElu&aclnnInplaceElu

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 the exponential linear unit (ELU) activation function to each element x in the input tensor self, and stores the result in the output tensor out.
Formula:
$ELU(x) = \begin{cases} scale \ast x, \quad x > 0\\ \alpha \ast scale \ast (exp(x \ast inputScale)-1), \quad x \leq 0 \end{cases}$

Function Prototype

aclnnElu and aclnnInplaceElu implement the same function. The differences in usage are as follows. Select the appropriate operator based on your actual scenario.
- aclnnElu: Create a new output tensor object to store the computation result.
- aclnnInplaceElu: No need to create a new output tensor object. The computation result is stored directly in the memory of the input tensor.
Each operator is divided into a two-phase API. You must first call the "aclnnEluGetWorkspaceSize" or "aclnnInplaceEluGetWorkspaceSize" API to obtain the workspace size required for computation and the executor that contains the operator computation process, and then call the "aclnnElu" or "aclnnInplaceElu" API to perform the computation.

aclnnStatus aclnnEluGetWorkspaceSize(
  const aclTensor *self,
  const aclScalar *alpha,
  const aclScalar *scale,
  const aclScalar *inputScale,
  aclTensor       *out,
  uint64_t        *workspaceSize,
  aclOpExecutor  **executor)

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

aclnnStatus aclnnInplaceEluGetWorkspaceSize(
  aclTensor       *selfRef,
  const aclScalar *alpha,
  const aclScalar *scale,
  const aclScalar *inputScale,
  uint64_t        *workspaceSize,
  aclOpExecutor  **executor)

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

aclnnEluGetWorkspaceSize

Parameters:

Parameter	Input/Output	Description	Instruction	Data Type	Data Format	Dimension (Shape)	Non-contiguous Tensor
self (aclTensor*)	Input	Represents the input of the ELU activation function, i.e., x in the formula.	Supports empty tensors.	FLOAT, FLOAT16, BFLOAT16	ND	0-8	√
alpha (aclScalar*)	Input	Represents the activation coefficient of the ELU activation function, i.e., α in the formula.	The data type must be convertible to FLOAT.	-	-	-	-
scale (aclScalar*)	Input	Represents the scaling factor of the ELU activation function, i.e., scale in the formula.	The data type must be convertible to FLOAT.	-	-	-	-
inputScale (aclScalar*)	Input	Represents the scaling factor of the ELU activation function input, which is inputScale in the formula.	The data type must be convertible to FLOAT.	-	-	-	-
out (aclTensor*)	Output	Represents the output of the ELU activation function.	The shape must be consistent with self.	FLOAT, FLOAT16, BFLOAT16	ND	0-8	√
workspaceSize (uint64_t*)	Output	Returns the workspace size to be requested on the Device side.	-	-	-	-	-
executor (aclOpExecutor**)	Output	Returns the op executor, which contains the operator computation flow.	-	-	-	-	-

Atlas training products: The data type supports FLOAT and FLOAT16.

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 self, alpha, scale, inputScale, and out parameters are null pointers.
ACLNN_ERR_PARAM_INVALID	161002	The data type of the self parameter is not within the supported range.
		The data types of alpha, scale, and inputScale cannot be converted to FLOAT.
		The data type of out is not convertible from self.
		The shapes of self and out are inconsistent.
		The dimensions of self and out exceed 8.

aclnnElu

Parameters:

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

Return value

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

aclnnInplaceEluGetWorkspaceSize

Parameters:

Parameter	Input/Output	Description	Instruction	Data Type	Data Format	Dimension (Shape)	Non-contiguous Tensor
selfRef (aclTensor*)	Input	Represents the input of the ELU activation function, i.e., x in the formula.	Supports empty tensors.	FLOAT, FLOAT16, DOUBLE, BFLOAT16	ND	0-8	√
alpha (aclScalar*)	Input	Represents the activation coefficient of the ELU activation function, i.e., α in the formula.	Data type that is convertible to FLOAT.	-	-	-	-
scale (aclScalar*)	Input	Represents the activation coefficient of the ELU activation function, i.e., scale in the formula.	The data type must be convertible to FLOAT.	-	-	-	-
inputScale (aclScalar*)	Input	Represents the scaling factor for the input of the ELU activation function.	The data type must be convertible to FLOAT.	-	-	-	-
workspaceSize (uint64_t*)	Output	Returns the workspace size to be requested on the Device side.	-	-	-	-	-
executor (aclOpExecutor**)	Output	Returns the operator executor, which contains the operator computation process.	-	-	-	-	-

Atlas training products: The data type supports FLOAT, FLOAT16, and DOUBLE.

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 parameters selfRef, alpha, scale, and inputScale are null pointers.
ACLNN_ERR_PARAM_INVALID	161002	The data type of the parameter selfRef is not within the supported range.
		The data types of the parameters alpha, scale, and inputScale cannot be converted to FLOAT.
		The dimension of the parameter selfRef is greater than 8.

aclnnInplaceElu

Parameters:

Parameter	Input/Output	Description
workspace	Input	Address of the workspace memory applied on the Device side.
workspaceSize	Input	Size of the workspace applied on the Device side, obtained by the first-phase API aclnnInplaceEluGetWorkspaceSize.
executor	Input	Operator executor that contains the operator computation flow.
stream	Input	Specifies the stream to execute the task.

Return value

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

Constraints

Deterministic computation:
- aclnnElu&aclnnInplaceElu default to a deterministic implementation.

Example

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

aclnnElu API Call Example:

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_elu.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 CreateAclTensor(
    const std::vector<T>& hostData, const std::vector<int64_t>& shape, void** deviceAddr, aclDataType dataType,
    aclTensor** selfOrResult)
{
    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 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);

    // Calculate the strides of the contiguous selfOrResult.
    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.
    *selfOrResult = 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.
    // 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> selfShape = {2, 2};
    std::vector<int64_t> outShape = {2, 2};
    void* selfDeviceAddr = nullptr;
    void* outDeviceAddr = nullptr;
    aclTensor* self = nullptr;
    aclScalar* alpha = nullptr;
    aclScalar* scale = nullptr;
    aclScalar* inputScale = nullptr;
    aclTensor* out = nullptr;
    std::vector<float> selfHostData = {-2, -1, 0, 1};
    std::vector<float> outHostData = {0, 0, 0, 0};
    float alphaValue = 1.0f;
    float scaleValue = 1.0f;
    float inputScaleValue = 1.0f;
    // Create the self aclTensor.
    ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
    CHECK_RET(ret == ACL_SUCCESS, return ret);
    // Create the alpha aclScalar.
    alpha = aclCreateScalar(&alphaValue, aclDataType::ACL_FLOAT);
    CHECK_RET(alpha != nullptr, return ret);
    // Create the scale aclScalar.
    scale = aclCreateScalar(&scaleValue, aclDataType::ACL_FLOAT);
    CHECK_RET(scale != nullptr, return ret);
    // Create the inputScale aclScalar.
    inputScale = aclCreateScalar(&inputScaleValue, aclDataType::ACL_FLOAT);
    CHECK_RET(inputScale != nullptr, return ret);
    // Create the 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. Replace with the specific API name.
    uint64_t workspaceSize = 0;
    aclOpExecutor* executor;
    // Call the first-phase API of aclnnElu.
    ret = aclnnEluGetWorkspaceSize(self, alpha, scale, inputScale, out, &workspaceSize, &executor);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnEluGetWorkspaceSize 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 aclnnElu.
    ret = aclnnElu(workspaceAddr, workspaceSize, executor, stream);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnElu failed. ERROR: %d\n", ret); return ret);

    // 4. (Boilerplate) Synchronously wait for the task 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 aclScalar. Modify based on the specific API definition.
    aclDestroyTensor(self);
    aclDestroyScalar(alpha);
    aclDestroyScalar(scale);
    aclDestroyScalar(inputScale);
    aclDestroyTensor(out);

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

aclnnInplaceElu API call sample code:

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_elu.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** selfOrResult)
{
    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);
    // 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);

    // Calculate the strides of the contiguous selfOrResult.
    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.
    *selfOrResult = 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.
    // Fill in the 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> selfRefShape = {2, 2};
    void* selfRefDeviceAddr = nullptr;
    aclTensor* selfRef = nullptr;
    aclScalar* alpha = nullptr;
    aclScalar* scale = nullptr;
    aclScalar* inputScale = nullptr;
    std::vector<float> selfRefHostData = {-2, -1, 0, 1};
    float alphaValue = 1.0f;
    float scaleValue = 1.0f;
    float inputScaleValue = 1.0f;
    // Create the selfRef aclTensor.
    ret = CreateAclTensor(selfRefHostData, selfRefShape, &selfRefDeviceAddr, aclDataType::ACL_FLOAT, &selfRef);
    CHECK_RET(ret == ACL_SUCCESS, return ret);
    // Create the alpha aclScalar.
    alpha = aclCreateScalar(&alphaValue, aclDataType::ACL_FLOAT);
    CHECK_RET(alpha != nullptr, return ret);
    // Create the scale aclScalar.
    scale = aclCreateScalar(&scaleValue, aclDataType::ACL_FLOAT);
    CHECK_RET(scale != nullptr, return ret);
    // Create the inputScale aclScalar.
    inputScale = aclCreateScalar(&inputScaleValue, aclDataType::ACL_FLOAT);
    CHECK_RET(inputScale != nullptr, 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 aclnnInplaceElu.
    ret = aclnnInplaceEluGetWorkspaceSize(selfRef, alpha, scale, inputScale, &workspaceSize, &executor);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnInplaceEluGetWorkspaceSize 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 aclnnInplaceElu.
    ret = aclnnInplaceElu(workspaceAddr, workspaceSize, executor, stream);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnInplaceElu 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(selfRefShape);
    std::vector<float> resultData(size, 0);
    ret = aclrtMemcpy(
        resultData.data(), resultData.size() * sizeof(resultData[0]), selfRefDeviceAddr, 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 based on the specific API definition.
    aclDestroyTensor(selfRef);
    aclDestroyScalar(alpha);
    aclDestroyScalar(scale);
    aclDestroyScalar(inputScale);

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