aclnnChannelShuffle
Applicable Products
- Ascend 950PR&950DT products: Supported
- Atlas A3 products: Supported
- Atlas A2 products: Supported
- Atlas 200I/500 A2 inference products: Not supported
- Atlas inference products: Not supported
- Atlas training products: Supported
Function
Divides the channels of a (*, C, H, W) tensor into g groups, randomly shuffles the channels within each group, and then merges all channels for output, while keeping the shape of the final output tensor consistent with the input tensor.
Function Prototype
Each operator uses a two-phase API. You must first call the "aclnnChannelShuffleGetWorkspaceSize" API to obtain the workspace size required for computation and an executor containing the computation process of the operator, and then call the "aclnnChannelShuffle" API to perform the computation.
aclnnStatus aclnnChannelShuffleGetWorkspaceSize(
const aclTensor *self,
int64_t groups,
aclTensor *out,
uint64_t *workspaceSize,
aclOpExecutor **executor)
aclnnStatus aclnnChannelShuffle(
void *workspace,
uint64_t workspaceSize,
aclOpExecutor *executor,
aclrtStream stream)
aclnnChannelShuffleGetWorkspaceSize
-
Parameters:
Parameter Input/Output Description Instruction Data Type Data Format Dimension (shape) Non-contiguous Tensor self (aclTensor*) Input Input tensor. The data dimension is greater than 2, and more than 7 dimensions are not supported. FLOAT16, FLOAT, DOUBLE, UINT8, INT8, INT16, INT32, INT64, COMPLEX64, COMPLEX128, BOOL ND 3-7 √ groups (int64_t) Input Specifies the number of groups to divide the channels of input self into. The value must be greater than 0 and divisible by the channels of self. INT64 - - - out (aclTensor*) Output Output tensor. The dimensions must be greater than 2, and 7 or more dimensions are not supported. FLOAT16, FLOAT, DOUBLE, UINT8, INT8, INT16, INT32, INT64, COMPLEX64, COMPLEX128, BOOL, BFLOAT16. ND 3-7 √ workspaceSize (uint64_t*) Output Returns the workspace size that needs to be allocated on the Device side. - - - - - executor (aclOpExecutor**) Output Returns the op executor containing the operator computation flow. - - - - - -
Return value
aclnnStatus: return status code. For details, see aclnn Return Codes.
The first-phase API performs input parameter validation and returns an error in the following scenarios:
Return Value Error Code Description ACLNN_ERR_PARAM_NULLPTR 161001 Parameters self and out are null pointers. ACLNN_ERR_PARAM_INVALID 161002 The data type of out is inconsistent with that of self. The data type of out is inconsistent with that of self. There is a shape inconsistency between parameters self and out. The number of dimensions of parameters self and out is greater than 7 or less than 3. The value of parameter groups is less than or equal to 0 or cannot be divided evenly by the channels of self.
aclnnChannelShuffle
-
Parameters:
Parameter Input/Output Description workspace Input Workspace memory address allocated on the Device side. workspaceSize Input Workspace size allocated on the Device side, obtained by the first-phase API aclnnChannelShuffleGetWorkspaceSize. executor Input op executor that contains the operator computation process. stream Input Stream where the task is executed. -
Return value
aclnnStatus: return status code. For details, see aclnn Return Codes.
Constraints
- Deterministic computation:
- aclnnChannelShuffle defaults to a deterministic implementation.
Example
The sample code is as follows and is for reference only. For the compilation and execution process, see Compile and Run Samples.
#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_channel_shuffle.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** tensor) {
auto size = GetShapeSize(shape) * sizeof(T);
// Call aclrtMalloc to allocate memory on the device side.
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 the strides of the 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. For details, see the acl API documentation.
// 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 inputs and outputs, which need to be custom-built based on the API.
std::vector<int64_t> selfShape = {1, 4, 2, 2};
std::vector<int64_t> outShape = {1, 4, 2, 2};
void* selfDeviceAddr = nullptr;
void* outDeviceAddr = nullptr;
aclTensor* self = nullptr;
aclTensor* out = nullptr;
std::vector<float> selfHostData = {0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15};
std::vector<float> outHostData = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0};
// Create the self aclTensor.
ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
CHECK_RET(ret == ACL_SUCCESS, 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. Change it to the specific API name.
int64_t groups = 2;
uint64_t workspaceSize = 0;
aclOpExecutor* executor;
// Call the first-phase API of aclnnChannelShuffle.
ret = aclnnChannelShuffleGetWorkspaceSize(self, groups, out, &workspaceSize, &executor);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnChannelShuffleGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);
// Allocate device memory based on the workspaceSize computed 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 aclnnChannelShuffle.
ret = aclnnChannelShuffle(workspaceAddr, workspaceSize, executor, stream);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnChannelShuffle failed. ERROR: %d\n", ret); return ret);
// 4. (boilerplate) Synchronize and wait until the task is 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 in device memory to the host side. Modify the code as required by 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 the code as required by the specific API definition.
aclDestroyTensor(self);
aclDestroyTensor(out);
// 7. Release device resources. Modify the code as required by the specific API definition.
aclrtFree(selfDeviceAddr);
aclrtFree(outDeviceAddr);
if (workspaceSize > 0) {
aclrtFree(workspaceAddr);
}
aclrtDestroyStream(stream);
aclrtResetDevice(deviceId);
aclFinalize();
return 0;
}