Provides methods to access partial result obtained with the compute() method of the neural network training algorithm in the distributed processing mode.
More...
◆ allocate()
Registers user-allocated memory to store partial results of the neural network model based training
- Parameters
-
[in] | input | Pointer to an object containing input data |
[in] | parameter | Parameter of the neural network training |
[in] | method | Computation method for the algorithm |
- Returns
- Status of computations
◆ check()
Checks result of the neural network algorithm
- Parameters
-
[in] | input | Input object of algorithm |
[in] | par | Parameter of algorithm |
[in] | method | Computation method |
- Returns
- Status of computations
Reimplemented from PartialResult.
◆ get()
Returns partial result of the neural network model based training
- Parameters
-
[in] | id | Identifier of the result |
- Returns
- Result that corresponds to the given identifier
◆ set()
Sets partial result of neural network model based training
- Parameters
-
[in] | id | Identifier of the result |
[in] | value | Result |
The documentation for this class was generated from the following file: