aclnnNLLLoss2dBackward
Applicable Products
| Product | Supported |
|---|---|
| √ | |
| √ | |
| √ | |
| × | |
| × | |
| √ |
Function
Performs backward propagation of the Negative Log-Likelihood Loss.
Function Prototype
Each operator is divided into a two-phase API. You must first call the "aclnnNLLLoss2dBackwardGetWorkspaceSize" API to obtain the workspace size required for computation and the executor that contains the operator computation process, and then call the "aclnnNLLLoss2dBackward" API to perform the computation.
aclnnStatus aclnnNLLLoss2dBackwardGetWorkspaceSize(
const aclTensor *gradOutput,
const aclTensor *self,
const aclTensor *target,
const aclTensor *weight,
int64_t reduction,
int64_t ignoreIndex,
aclTensor *totalWeight,
aclTensor *out,
uint64_t *workspaceSize,
aclOpExecutor **executor)aclnnStatus aclnnNLLLoss2dBackward(
void *workspace,
uint64_t workspaceSize,
aclOpExecutor *executor,
aclrtStream stream)aclnnNLLLoss2dBackwardGetWorkspaceSize
Parameters
Parameter Input/Output Description Instruction Data Type Data Format Dimension (Shape) Non-Contiguous Tensor gradOutput (aclTensor*) Input Input aclTensor. The shape is three-dimensional (the first dimension is N) or one-dimensional (with only one element). FLOAT, FLOAT16, BFLOAT16 ND - √ self (aclTensor*) Input Input aclTensor. - The shape is four-dimensional, where the first dimension N represents the batch size and the second dimension C represents the classes.
- Requires that the shapes of the 0th, 2nd, and 3rd dimensions of self be consistent with the shapes of the 0th, 1st, and 2nd dimensions of target, respectively. Otherwise, returns false.
FLOAT, FLOAT16, BFLOAT16 ND - √ target (aclTensor*) Input Represents the ground-truth labels. - y in the formula, with a 3-dimensional shape.
- The shapes of the 0th, 1st, and 2nd dimensions of target are consistent with those of the 0th, 2nd, and 3rd dimensions of self, respectively.
- The value range of each element is [0, C - 1].
INT64, UINT8, INT32 ND - √ weight（aclTensor*） Input Scaling weight for each class. w in the formula, with a shape of (C,). The data type is consistent with self. ND (C,) √ reduction (int64_t) Input Specifies the reduction to apply to the output. - Supports 0 (none), 1 (mean), and 2 (sum).
- 'none' indicates that no reduction is applied.
- 'mean' indicates that the sum of the output is divided by the number of elements in the output.
- 'sum' indicates that the output is summed.
- When reduction is 0, the shape of target must be consistent with the shape of gradOutput; otherwise, false is returned.
INT64 - - - ignoreIndex（int64_t） Input Specifies a target value that is ignored and does not affect the input gradient. - INT64 - - - totalWeight（aclTensor*） Input - Only when reduction is mean, totalWeight is obtained by taking the weight at the corresponding position through target, then removing the weight corresponding to ignoreIndex, and summing the remaining weights. When reduction is set to other values, this parameter is not processed by default. The data type is the same as weight. ND (1,) - out (aclTensor*) Output - Has the same shape as self. Has the same data type as self. ND - √ workspaceSize (uint64_t*) Output Size of the workspace to be applied for on the Device side. - - - - - executor (aclOpExecutor**) Output Returns the operator executor, which contains the operator computation process. - - - - - 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 Value Error Code Description ACLNN_ERR_PARAM_NULLPTR 161001 The input gradOutput, self, target, weight, totalWeight, or out is a null pointer. ACLNN_ERR_PARAM_INVALID 161002 The data type of gradOutput, self, target, weight, totalWeight, or out is not within the supported range. The data types of gradOutput, self, weight, totalWeight, and out are inconsistent. target is not a 3D tensor, or self is not a 4D tensor. The number of elements in weight is not C. The number of elements in dimensions 0, 2, and 3 of self is not equal to the number of elements in dimensions 0, 1, and 2 of target. The number of elements in totalWeight is not 1. When reduction is none, the number of dimensions of gradOutput is not 3, or the number of elements in dimensions 0, 1, and 2 of gradOutput is not equal to the number of elements in dimensions 0, 1, and 2 of target. When reduction is not none, the number of dimensions of gradOutput is greater than 1, or the number of elements is not 1. The reduction value is not within the range of 0 to 2.
aclnnNLLLoss2dBackward
Parameters
Parameter Input/Output Description workspace Input Workspace memory address applied for on the Device side. workspaceSize Input Workspace size applied for on the Device side, obtained through the first-phase API aclnnNLLLoss2dBackwardGetWorkspaceSize. 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:
- aclnnNLLLoss2dBackward 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_nll_loss2d_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 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);
// Copy data from the host 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. For details, 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> gradShape = {3, 1, 1};
std::vector<int64_t> selfShape = {3, 5, 1, 1};
std::vector<int64_t> targetShape = {3, 1, 1};
std::vector<int64_t> weightShape = {5};
std::vector<int64_t> totalWeightShape = {1};
std::vector<int64_t> outShape = {3, 5, 1, 1};
void* gradDeviceAddr = nullptr;
void* selfDeviceAddr = nullptr;
void* targetDeviceAddr = nullptr;
void* weightDeviceAddr = nullptr;
void* totalWeightDeviceAddr = nullptr;
void* outDeviceAddr = nullptr;
aclTensor* grad = nullptr;
aclTensor* self = nullptr;
aclTensor* target = nullptr;
aclTensor* weight = nullptr;
aclTensor* totalWeight = nullptr;
aclTensor* out = nullptr;
std::vector<float> gradHostData = {2.7, 2.6, 2.5};
std::vector<float> selfHostData = {4.1, 4.2, 4.3, 4.4, 4.5, 4.6, 4.7, 4.8, 4.9, 5.0, 5.1, 5.2, 5.3, 5.4, 5.5};
std::vector<int64_t> targetHostData = {2, 3, 1};
std::vector<float> weightHostData = {1.0, 1.0, 1.0, 1.0, 1.0};
std::vector<float> totalWeightHostData = {1.0};
std::vector<float> outHostData = {0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0};
int64_t reduction = 0;
int64_t ignoreIndex = -100;
// Create the grad aclTensor.
ret = CreateAclTensor(gradHostData, gradShape, &gradDeviceAddr, aclDataType::ACL_FLOAT, &grad);
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 target aclTensor.
ret = CreateAclTensor(targetHostData, targetShape, &targetDeviceAddr, aclDataType::ACL_INT64, &target);
CHECK_RET(ret == ACL_SUCCESS, return ret);
// Create the weight aclTensor.
ret = CreateAclTensor(weightHostData, weightShape, &weightDeviceAddr, aclDataType::ACL_FLOAT, &weight);
CHECK_RET(ret == ACL_SUCCESS, return ret);
// Create the totalWeight aclTensor.
ret = CreateAclTensor(totalWeightHostData, totalWeightShape, &totalWeightDeviceAddr,
aclDataType::ACL_FLOAT, &totalWeight);
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. Replace with the specific API name.
uint64_t workspaceSize = 0;
aclOpExecutor* executor;
// Call the first-phase API of aclnnNLLLoss2dBackward.
ret = aclnnNLLLoss2dBackwardGetWorkspaceSize(grad, self, target, weight, reduction, ignoreIndex, totalWeight, out,
&workspaceSize, &executor);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnNLLLoss2dBackwardGetWorkspaceSize 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 aclnnNLLLoss2dBackward.
ret = aclnnNLLLoss2dBackward(workspaceAddr, workspaceSize, executor, stream);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnNLLLoss2dBackward failed. ERROR: %d\n", ret); return ret);
// 4. (Boilerplate) Synchronously 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. Modify based on the specific API definition.
aclDestroyTensor(grad);
aclDestroyTensor(self);
aclDestroyTensor(target);
aclDestroyTensor(weight);
aclDestroyTensor(totalWeight);
aclDestroyTensor(out);
// 7. Release device resources.
aclrtFree(gradDeviceAddr);
aclrtFree(selfDeviceAddr);
aclrtFree(targetDeviceAddr);
aclrtFree(weightDeviceAddr);
aclrtFree(totalWeightDeviceAddr);
aclrtFree(outDeviceAddr);
if (workspaceSize > 0) {
aclrtFree(workspaceAddr);
}
aclrtDestroyStream(stream);
aclrtResetDevice(deviceId);
aclFinalize();
return 0;
}