C++ API Reference for Intel® Data Analytics Acceleration Library 2020 Update 1
Contains classes that implement algorithms for data analysis(data mining), and data modeling(training and prediction). These algorithms include matrix decompositions, clustering algorithms, classification and regression algorithms, as well as association rules discovery. More...
Namespaces | |
daal::algorithms::interface1 | |
Contains version 1.0 of Intel(R) Data Analytics Acceleration Library (Intel(R) DAAL) interface. | |
Classes | |
class | AlgorithmIface |
Abstract class which defines interface for the library component related to data processing involving execution of the algorithms for analysis, modeling, and prediction. More... | |
class | AlgorithmIfaceImpl |
Implements the abstract interface AlgorithmIface. AlgorithmIfaceImpl is, in turn, the base class for the classes interfacing the major compute modes: batch, online and distributed. More... | |
class | Algorithm< mode > |
Implements the abstract interface AlgorithmIface. Algorithm is, in turn, the base class for the classes interfacing the major stages of data processing: Analysis, Training and Prediction. More... | |
class | Algorithm< batch > |
Implements the abstract interface AlgorithmIface. Algorithm<batch> is, in turn, the base class for the classes interfacing the major stages of data processing in batch mode: Analysis<batch>, Training<batch> and Prediction. More... | |
class | AlgorithmContainer< batch > |
Abstract interface class that provides virtual methods to access and run implementations of the algorithms in batch mode. It is associated with the Algorithm<batch> class and supports the methods for computation of the algorithm results. The methods of the container are defined in derivative containers defined for each algorithm. More... | |
class | AlgorithmContainerImpl< batch > |
Abstract interface class that provides virtual methods to access and run implementations of the algorithms in batch mode. It is associated with the Algorithm<batch> class and supports the methods for computation of the algorithm results. The methods of the container are defined in derivative containers defined for each algorithm. More... | |
class | AlgorithmDispatchContainer< batch, sse2Container > |
Implements a container to dispatch batch processing algorithms to CPU-specific implementations. More... | |
class | AlgorithmContainerIface |
Implements the abstract interface AlgorithmContainerIface. It is associated with the Algorithm class and supports the methods for computation and finalization of the algorithm results in the batch, distributed, and online modes. More... | |
class | AlgorithmContainerIfaceImpl |
Implements the abstract interface AlgorithmContainerIfaceImpl. It is associated with the Algorithm class and supports the methods for computation and finalization of the algorithm results in the batch, distributed, and online modes. More... | |
class | AlgorithmContainer< mode > |
Abstract interface class that provides virtual methods to access and run implementations of the algorithms. It is associated with the Algorithm class and supports the methods for computation and finalization of the algorithm results in the batch, distributed, and online modes. The methods of the container are defined in derivative containers defined for each algorithm. More... | |
class | AlgorithmContainerImpl< mode > |
Abstract interface class that provides virtual methods to access and run implementations of the algorithms. It is associated with the Algorithm class and supports the methods for computation and finalization of the algorithm results in the batch, distributed, and online modes. The methods of the container are defined in derivative containers defined for each algorithm. More... | |
class | AlgorithmDispatchContainer< mode, sse2Container > |
Implements a container to dispatch algorithms to cpu-specific implementations. More... | |
class | Kernel |
Base class to represent algorithm implementation More... | |
class | Batch |
Provides methods to compute quality metrics of an algorithm in the batch processing mode. Quality metric is a numerical characteristic or a set of connected numerical characteristics that represents the qualitative aspect of a computed statistical estimate, model, or decision-making result. More... | |
class | InputAlgorithmsCollection |
Class that implements functionality of the collection of quality metrics algorithms. More... | |
class | InputDataCollection |
Class that implements functionality of the collection of input objects of the quality metrics algorithm. More... | |
class | ResultCollection |
Class that implements functionality of the collection of result objects of the quality metrics algorithm. More... | |
struct | Parameter |
Base class to represent computation parameters. Algorithm-specific parameters are represented as derivative classes of the Parameter class. More... | |
struct | ValidationMetricIface |
class | Model |
The base class for the classes that represent the models, such as linear_regression::Model or svm::Model. More... | |
class | PredictionContainerIface |
Abstract interface class that provides virtual methods to access and run implementations of the algorithms for model based prediction. Is associated with the Prediction class and supports the methods for computing the prediction results based on the trained model. The methods of the container are defined in derivative containers defined for each prediction algorithm. More... | |
class | DistributedPredictionContainerIface |
class | Prediction |
Provides prediction methods depending on the model such as linear_regression::Model. The methods of the class support different computation modes: batch, distributed, and online(see ComputeMode). Classes that implement specific algorithms of the model based data prediction are derived classes of the Prediction class. The class additionally provides virtual methods for validation of input and output parameters of the algorithms. More... | |
class | DistributedPrediction |
class | TrainingContainerIface< mode > |
Abstract interface class that provides virtual methods to access and run implementations of the model training algorithms. The class is associated with the Training class and supports the methods for computation and finalization of the training output in the batch, distributed, and online modes. The methods of the container are defined in derivative containers defined for each training algorithm. More... | |
class | Training< mode > |
Provides methods to train models that depend on the data provided. For example, these methods enable training the linear regression model. The methods of the class support different computation modes: batch, distributed, and online(see ComputeMode). Classes that implement specific algorithms of model training are derived classes of the Training class. The class additionally provides methods for validation of input and output parameters of the algorithms. More... | |
For more complete information about compiler optimizations, see our Optimization Notice.