Python* API Reference for Intel® Data Analytics Acceleration Library 2020 Update 1

Public Member Functions | Static Public Attributes | List of all members
Batch Class Reference

Algorithm for the multi-class classifier model training. More...

Public Member Functions

def __init__
 
def getInput
 
def getMethod
 
def getResult
 
def resetResult
 
def clone
 
def compute
 
- Public Member Functions inherited from Batch
def getInput
 
def setResult
 
def getResult
 
def resetResult
 
def clone
 
def compute
 
- Public Member Functions inherited from AlgorithmImpl
def computeNoThrow
 
def compute
 
def checkComputeParams
 
def checkResult
 
def setupCompute
 
def resetCompute
 
def enableResetOnCompute
 
def hostApp
 
def setHostApp
 
- Public Member Functions inherited from Algorithm
def checkComputeParams
 
def getBaseParameter
 
- Public Member Functions inherited from AlgorithmIfaceImpl
def enableChecks
 
def isChecksEnabled
 
- Public Member Functions inherited from AlgorithmIface
def checkComputeParams
 
def checkResult
 
def getMethod
 

Static Public Attributes

 parameter = ...
 
 input = ...
 

Detailed Description

Deprecated:
This item will be removed in a future release.
Parameters
fptypeData type to use in intermediate computations for the multi-class classifier training algorithm, double or float
methodComputation method for the algorithm, Method
Enumerations
  • Method Multi-class classifier training methods
  • classifier.training.InputId Identifiers of input objects for the multi-class classifier algprithm
  • classifier.training.ResultId Identifiers of multi-class classifier training results
References
Aliases
  • Batch_Float64OneAgainstOne is an alias of Batch(fptype=float64, method=daal.algorithms.multi_class_classifier.training.oneAgainstOne)
  • Batch_Float32OneAgainstOne is an alias of Batch(fptype=float32, method=daal.algorithms.multi_class_classifier.training.oneAgainstOne)

Constructor & Destructor Documentation

def __init__ (   self,
  args 
)

Variant 1

Default constructor

Deprecated:
This item will be removed in a future release.

Variant 2

Default constructor

Parameters
nClassesNumber of classes

Variant 3

Constructs multi-class classifier training algorithm by copying input objects and parameters of another multi-class classifier training algorithm

Parameters
otherAn algorithm to be used as the source to initialize the input objects and parameters of the algorithm

Member Function Documentation

def clone (   self)

Returns a pointer to the newly allocated multi-class classifier training algorithm with a copy of input objects and parameters of this multi-class classifier training algorithm

Returns
Pointer to the newly allocated algorithm
def compute (   self)

Invokes computations

def getInput (   self)

Get input objects for the multi-class classifier training algorithm

Returns
Input objects for the multi-class classifier training algorithm
def getMethod (   self)

Returns method of the algorithm

Returns
Method of the algorithm
def getResult (   self)

Returns the structure that contains the training results of the multi-class classifier algorithm

Returns
Structure that contains the training results of the multi-class classifier algorithm
def resetResult (   self)

Resets the training results of the classification algorithm

Member Data Documentation

input = ...
static

Input objects of the algorithm

parameter = ...
static

Parameters of the algorithm


The documentation for this class was generated from the following file:

For more complete information about compiler optimizations, see our Optimization Notice.