Documentation
Download
文档中心
aclnnMseLossOut

aclnnAscendAntiQuant

📄 View Source Code

Applicable Products

ProductSupported
Ascend 950PR/Ascend 950DTNo
Atlas A3 training products/Atlas A3 inference productsYes
Atlas A2 training products/Atlas A2 inference productsYes
Atlas 200I/500 A2 inference productsNo
Atlas inference productsYes
Atlas training productsNo

Function

  • Description: Dequantizes the input x based on the input scale and offset.

  • Formula:

  • When sqrtMode is false, the formula is:

    y=float16((x+offset)∗scale)y = float16((x+offset) * scale)
    • When sqrtMode is true, the formula is:y=float16((x+offset)∗scale∗scale)y = float16((x+offset) * scale * scale)

Function Prototype

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

aclnnStatus aclnnAscendAntiQuantGetWorkspaceSize(
  const aclTensor* x,
  const aclTensor* scale,
  const aclTensor* offset,
  int64_t          dstType,
  bool             sqrtMode,
  const aclTensor* y,
  uint64_t*        workspaceSize,
  aclOpExecutor**  executor)
aclnnStatus aclnnAscendAntiQuant(
  void          *workspace,
  uint64_t       workspaceSize,
  aclOpExecutor *executor,
  aclrtStream    stream)

aclnnAscendAntiQuantGetWorkspaceSize

  • Parameters:

    Parameter Input/Output Description Instruction Data Type Data Format Dimension (shape) Non-contiguous Tensor
    x (aclTensor*) Input Indicates the input to be dequantized. Corresponds to `x` in the formula.
    • Supports empty tensors.
    • When the data type is INT32, each piece of data is used as eight INT4 data.
    • When the data type is INT4, the last axis of the shape must be an even number.
    		 INT4, INT8, INT32 ND 0-8 √
    scale (aclTensor*) Input The scale value in anti-quantization. Corresponds to `scale` in the formula.
    • Supports empty tensors.
    • The size of shape can be 1. If it is not 1, when the input `x` is of INT4/INT8 type, it equals the size of the last dimension of the input `x`; when the input `x` is of INT32 type, it equals 8 times the size of the last dimension of the input `x`.
    		 FLOAT32, BFLOAT16 ND 1 √
    offset (aclTensor*) Input The offset value in dequantization, corresponding to `offset` in the formula.
    • Supports empty tensors.
    • The data type and shape must be consistent with `scale`.
    		 FLOAT32, BFLOAT16 ND 1 √
    dstType (int64_t) Input Specifies the data type of the output. Supports values 1 and 27, representing FLOAT16 and BFLOAT16 respectively. - - - -
    sqrtMode (bool) Input Specifies the logic for scale participation in computation. Corresponds to `sqrtMode` in the formula. - - - - -
    y（aclTensor*） Output The computed output of anti-quantization. Corresponds to `y` in the formula.
    • Supports empty tensors.
    • When the input `x` is of INT4 or INT8 type, the shape is the same as that of the input `x`.
    • When the input `x` is of INT32 type, the size of the last axis of the shape is eight times the size of the last axis of the input `x`.
    		 FLOAT16, BFLOAT16 ND 0-8 √
    workspaceSize (uint64_t*) Output Returns the workspace size to be applied for on the Device side. - - - - -
    executor (aclOpExecutor**) Output Returns the op executor, which contains the operator computation flow. - - - - -

    • Atlas inference products:

      • Data type:

        • The input x supports only INT8.

        • The inputs scale and offset, and the output y do not support BFLOAT16.

      • The input dstType supports only the value 1, which indicates 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 x, scale, or y is a null pointer.
    ACLNN_ERR_PARAM_INVALID 161002 The data type or data format of x, scale, offset, or y is not within the supported range.
    dstType is not within the valid value range.
    When the data type of x is INT4, the size of the last axis of the shape of x is not an even number.
    When the data type of x is INT32, the last axis of the shape of y is not 8 times the size of the last axis of the shape of x, or the sizes of the non-last axes of the shapes of x and y are inconsistent.

aclnnAscendAntiQuant

  • Parameters:

    Parameter Input/Output Description
    workspace Input Memory address of the workspace applied for on the device.
    workspaceSize Input Size of the workspace applied for on the device, obtained through the first-phase API aclnnAscendAntiQuantGetWorkspaceSize.
    executor Input Operator executor, which contains the operator computation process.
    stream Input Specifies the stream for task execution.

  • Return value

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

Constraints

  • Deterministic computation:

    • aclnnAscendAntiQuant 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_ascend_anti_quant.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 shape_size = 1;
    for (auto i : shape) {
        shape_size *= i;
    }
    return shape_size;
}

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** tensor)
{
    auto size = GetShapeSize(shape) * sizeof(T);
    // Call aclrtMalloc to apply for device 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 data from the host side to the device memory by calling 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);

    // Compute 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];
    }

    // Create an aclTensor by calling aclCreateTensor.
    *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. Refer to the ACL API manual.
    // Fill in the deviceId based on your actual device.
    int32_t deviceId = 0;
    aclrtStream stream;
    auto ret = Init(deviceId, &stream);
    // Handle based on your own needs.
    CHECK_RET(ret == 0, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);
    // 2. Construct inputs and outputs based on the API's interface.
    std::vector<int64_t> selfShape = {4, 2};
    std::vector<int64_t> scaleShape = {1};
    std::vector<int64_t> offsetShape = {1};
    std::vector<int64_t> outShape = {4, 2};
    void* selfDeviceAddr = nullptr;
    void* scaleDeviceAddr = nullptr;
    void* offsetDeviceAddr = nullptr;
    void* outDeviceAddr = nullptr;
    aclTensor* self = nullptr;
    aclTensor* scale = nullptr;
    aclTensor* offset = nullptr;
    aclTensor* out = nullptr;
    std::vector<int8_t> selfHostData = {0, 1, 2, 3, 4, 5, 6, 7};
    std::vector<float> scaleHostData = {1};
    std::vector<float> offsetHostData = {1};
    std::vector<short> outHostData = {0, 0, 0, 0, 0, 0, 0, 0};
    // Create the self aclTensor.
    ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_INT8, &self);
    CHECK_RET(ret == ACL_SUCCESS, return ret);
    // Create the scale aclTensor.
    ret = CreateAclTensor(scaleHostData, scaleShape, &scaleDeviceAddr, aclDataType::ACL_FLOAT, &scale);
    CHECK_RET(ret == ACL_SUCCESS, return ret);
    // Create the offset aclTensor.
    ret = CreateAclTensor(offsetHostData, offsetShape, &offsetDeviceAddr, aclDataType::ACL_FLOAT, &offset);
    CHECK_RET(ret == ACL_SUCCESS, return ret);
    // Create the out aclTensor.
    ret = CreateAclTensor(outHostData, outShape, &outDeviceAddr, aclDataType::ACL_FLOAT16, &out);
    CHECK_RET(ret == ACL_SUCCESS, return ret);

    int64_t dstType = 1;
    bool sqrtMode = false;

    // 3. Call the CANN operator library API. Replace with the specific API.
    uint64_t workspaceSize = 0;
    aclOpExecutor* executor;
    // Call the first-phase API of aclnnAscendAntiQuant.
    ret = aclnnAscendAntiQuantGetWorkspaceSize(self, scale, offset, dstType, sqrtMode, out, &workspaceSize, &executor);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnAscendAntiQuantGetWorkspaceSize 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 aclnnAscendAntiQuant.
    ret = aclnnAscendAntiQuant(workspaceAddr, workspaceSize, executor, stream);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnAscendAntiQuant failed. ERROR: %d\n", ret); return ret);
    // 4. (Boilerplate) Synchronously wait for task execution to finish.
    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 and check the data in resultData.
    auto size = GetShapeSize(outShape);
    std::vector<short> resultData(size, 0);
    ret = aclrtMemcpy(
        resultData.data(), resultData.size() * sizeof(resultData[0]), outDeviceAddr, size * sizeof(short),
        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: %hd\n", i, resultData[i]);
    }

    // 6. Release aclTensor. Modify based on the specific API definition.
    aclDestroyTensor(self);
    aclDestroyTensor(scale);
    aclDestroyTensor(offset);
    aclDestroyTensor(out);

    // 7. Release device resources. Modify based on the specific API definition.
    aclrtFree(selfDeviceAddr);
    aclrtFree(scaleDeviceAddr);
    aclrtFree(offsetDeviceAddr);
    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