Parameter class for LBFGS algorithm
More...
def __init__ |
( |
|
self, |
|
|
|
args |
|
) |
| |
Constructs the parameters of LBFGS algorithm
- Parameters
-
function | Objective function that can be represented as sum |
nIterations | Maximal number of iterations of the algorithm |
accuracyThreshold | Accuracy of the LBFGS algorithm |
batchSize | Number of observations to compute the stochastic gradient |
correctionPairBatchSize_ | The number of observations to compute the sub-sampled Hessian for correction pairs computation |
m | Memory parameter of LBFGS |
L | The number of iterations between the curvature estimates calculations |
seed | Seed for random choosing terms from objective function |
- Deprecated:
- This item will be removed in a future release. Use engine instead.
Checks the correctness of the parameter
- Returns
- Status of computations
correctionPairBatchIndices = ... |
|
static |
Numeric table of size (nIterations / L) x correctionPairBatchSize that represent indices that will be used
correctionPairBatchSize = ... |
|
static |
Number of observations to compute the sub-sampled Hessian for correction pairs computation
Engine for random choosing terms from objective function.
The number of iterations between the curvature estimates calculations
Memory parameter of LBFGS.
Seed for random choosing terms from objective function.
- Deprecated:
- This item will be removed in a future release. Use engine instead.
The documentation for this class was generated from the following file: