aclnnIsClose
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: supported
- Atlas training products: supported
Function
-
Description: Returns a new tensor with Boolean elements, determining whether the given self and other are close to each other. If the values are close, True is returned; otherwise, False is returned.
-
Formula: Returns a new tensor out with the data type BOOL, indicating whether each element of the input self is "close" to the corresponding element of the input other. Closeness is defined as:
∣selfi−otheri∣≤atol+rtol×∣otheri∣\left | self_{i}-other_{i}\right | \le atol + rtol\times \left | other_{i} \right |
The above formula holds when both self and other are finite values. When both self and other are non-finite, the result is True only if they are equal. When equal_nan is True, NaNs are considered close; when equal_nan is False, NaNs are considered not close.
Function Prototype
Each operator uses a two-phase API. First, call the "aclnnIsCloseGetWorkspaceSize" API to obtain the workspace size required for computation and the executor that contains the operator computation process. Then, call the "aclnnIsClose" API to perform the computation.
aclnnStatus aclnnIsCloseGetWorkspaceSize(
const aclTensor* self,
const aclTensor* other,
double rtol,
double atol,
bool equalNan,
aclTensor* out,
uint64_t* workspaceSize,
aclOpExecutor** executor)
aclnnStatus aclnnIsClose(
void* workspace,
uint64_t workspaceSize,
aclOpExecutor* executor,
aclrtStream stream)
aclnnIsCloseGetWorkspaceSize
-
Parameters:
Parameter Input/Output Description Instruction Data Type Data Format Dimensions (shape) Non-contiguous Tensor self (aclTensor*) Input Input tensor, self in the formula. - The data type needs to be with the data type of other.
- The shape needs to be with other to satisfy the broadcast relationship.
FLOAT, INT32, INT64, FLOAT16, INT16, INT8, UINT8, DOUBLE, BOOL, BFLOAT16 ND 0-8 √ other (aclTensor*) Input Input tensor, other in the formula. - The data type needs to be with the data type of self.
- The shape needs to be with self, satisfying the broadcast relationship.
FLOAT, INT32, INT64, FLOAT16, INT16, INT8, UINT8, DOUBLE, BOOL, BFLOAT16 ND 0-8 √ rtol (double) Input Relative tolerance, rtol in the formula. - - - - - atol (double) Input Absolute tolerance, atol in the formula. - - - - - equalNan (bool) Input Whether two NaNs are considered equal, corresponding to equalNan in the formula. - - - - - out (aclTensor*) Output Output tensor, which is out in the formula. The data type is BOOL. The shape is the broadcast shape of self and other. BOOL ND 0-8 √ workspaceSize (uint64_t*) Output Returns the required size of the Device-allocated workspace. - - - - - executor (aclOpExecutor**) Output Returns the op executor containing the operator computation flow. - - - - - - Atlas training products and Atlas inference products: The BFLOAT16 data type is not supported.
-
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:
Error Code Error Code Description ACLNN_ERR_PARAM_NULLPTR 161001 self, other, or out is a null pointer. ACLNN_ERR_PARAM_INVALID 161002 The data type of self, other, or out is not supported. The data types of self and other do not comply with the data type deduction rules. The shapes of self and other cannot be broadcast. The shape of out is not the broadcast shape of self and other.
aclnnIsClose
-
Parameters:
Parameter Input/Output Description workspace Input Memory address of the workspace side-applied on the Device. workspaceSize Input Size of the workspace side-applied on the Device, obtained by the first-phase API aclnnIsCloseGetWorkspaceSize. executor Input Executor of the op, containing the operator computation process. stream Input Stream for executing tasks. -
Return value
aclnnStatus: return status code. For details, see aclnn Return Codes.
Constraints
- Deterministic computation:
- aclnnIsClose defaults to a deterministic implementation.
Example
The following sample code is for reference only. For details about the compilation and execution process, see Compile and Run Samples.
#include <cinttypes>
#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_isclose.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 code, 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);
// 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 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 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 device/stream. For details, see the acl API manual.
// Fill in deviceId based on the actual device.
int32_t deviceId = 0;
aclrtStream stream;
auto ret = Init(deviceId, &stream);
// Handle check as needed.
CHECK_RET(ret == 0, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);
// 2. Construct inputs and outputs, which need to be customized based on the API.
std::vector<int64_t> selfShape = {4, 2};
std::vector<int64_t> otherShape = {4, 2};
std::vector<int64_t> outShape = {4, 2};
void* selfDeviceAddr = nullptr;
void* otherDeviceAddr = nullptr;
void* outDeviceAddr = nullptr;
aclTensor* self = nullptr;
aclTensor* other = nullptr;
aclTensor* out = nullptr;
std::vector<float> selfHostData = {0, 1, 2, 3, 4, 5, 6, 7};
std::vector<float> otherHostData = {1, 1, 1, 2, 1, 2, 3, 3};
std::vector<uint8_t> outHostData = {0, 0, 0, 0, 0, 0, 0, 0};
double rtol = 1.0;
double atol = 1.0;
bool equal_nan = false;
// Create gradOutput aclTensor.
ret = CreateAclTensor(otherHostData, otherShape, &otherDeviceAddr, aclDataType::ACL_FLOAT, &other);
CHECK_RET(ret == ACL_SUCCESS, return ret);
// Create 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_BOOL, &out);
CHECK_RET(ret == ACL_SUCCESS, return ret);
// 3. Call the CANN operator library API. Replace it with the specific API.
uint64_t workspaceSize = 0;
aclOpExecutor* executor;
// Call the first-phase API of aclnnIsClose.
ret = aclnnIsCloseGetWorkspaceSize(self, other, rtol, atol, equal_nan, out, &workspaceSize, &executor);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnIsCloseGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);
// Allocate 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 aclnnIsClose.
ret = aclnnIsClose(workspaceAddr, workspaceSize, executor, stream);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnIsClose failed. ERROR: %d\n", ret); return ret);
// 4. (Boilerplate) Synchronize and 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 device memory to host memory. Modify based on the specific API definition.
auto size = GetShapeSize(outShape);
std::vector<uint8_t> 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[%" PRId64 "] is: %d\n", i, resultData[i]);
}
// 6. Release aclTensor and aclScalar. Modify based on the specific API definition.
aclDestroyTensor(other);
aclDestroyTensor(self);
aclDestroyTensor(out);
// 7. Release device resources. Modify based on the specific API definition.
aclrtFree(otherDeviceAddr);
aclrtFree(selfDeviceAddr);
aclrtFree(outDeviceAddr);
if (workspaceSize > 0) {
aclrtFree(workspaceAddr);
}
aclrtDestroyStream(stream);
aclrtResetDevice(deviceId);
aclFinalize();
return 0;
}