Computes Stochastic gradient descent in the batch processing mode.
More...
- Parameters
-
fptype | Data type to use in intermediate computations for the Stochastic gradient descent algorithm, double or float |
method | Stochastic gradient descent computation method |
- Enumerations
- Method Computation methods for Stochastic gradient descent
- iterative_solver.InputId Identifiers of input objects for Stochastic gradient descent
- iterative_solver.ResultId Result identifiers for the Stochastic gradient descent
- References
-
- Aliases
Batch_Float64DefaultDense
is an alias of Batch(fptype=float64, method=daal.algorithms.optimization_solver.sgd.defaultDense)
Batch_Float64MiniBatch
is an alias of Batch(fptype=float64, method=daal.algorithms.optimization_solver.sgd.miniBatch)
Batch_Float64Momentum
is an alias of Batch(fptype=float64, method=daal.algorithms.optimization_solver.sgd.momentum)
Batch_Float32DefaultDense
is an alias of Batch(fptype=float32, method=daal.algorithms.optimization_solver.sgd.defaultDense)
Batch_Float32MiniBatch
is an alias of Batch(fptype=float32, method=daal.algorithms.optimization_solver.sgd.miniBatch)
Batch_Float32Momentum
is an alias of Batch(fptype=float32, method=daal.algorithms.optimization_solver.sgd.momentum)
def __init__ |
( |
|
self, |
|
|
|
args |
|
) |
| |
- Variant 1
Constructs the SGD algorithm with the input objective function
- Parameters
-
objectiveFunction | Objective function that can be represented as a sum of functions |
- Variant 2
Constructs the SGD algorithm with the input objective function
- Parameters
-
objectiveFunction | Objective function that can be represented as a sum of functions |
- Variant 3
Constructs a Stochastic gradient descent algorithm by copying input objects of another Stochastic gradient descent algorithm
- Parameters
-
other | An algorithm to be used as the source to initialize the input objects and parameters of the algorithm |
Returns a pointer to the newly allocated Stochastic gradient descent algorithm with a copy of input objects of this Stochastic gradient descent algorithm
- Returns
- Pointer to the newly allocated algorithm
Creates the instance of the class
- Returns
- New instance of the class
Creates user-allocated memory to store results of the iterative solver algorithm
- Returns
- Status of computations
Get input objects for the iterative solver algorithm
- Returns
- Input objects for the iterative solver algorithm
Returns method of the algorithm
- Returns
- Method of the algorithm
Get parameters of the iterative solver algorithm
- Returns
- Parameters of the iterative solver algorithm
The documentation for this class was generated from the following file: