All Classes

Class	Description
AbstractActivationFunction	Implements the boiler plate code for applying functions on container classes like vectors and matrices by applying the function on every element.
AbstractClassifier	Abstract base class for classifiers.
AbstractKNearestNeighbours	K nearest neighbour classification algorithm that is seeded with a "database" of known examples and predicts based on the k-nearest neighbours majority vote for a class.
AbstractMiniBatchCostFunction	Mini Batch cost function.
AbstractMinimizer	Abstract minimizer class that adds functionality that can be shared between many minimizers.
AbstractPredictor
AbstractTreeNode
ActivationFunction	Squashing function interface to provide multiple activation functions, e.G.
ActivationFunctionSelector	Singleton helper to get the activation functions as singleton.
AgglomerativeClustering	"Bottom Up" clustering (agglomerative) using average single linkage clustering.
AgglomerativeClustering.ClusterNode	Tree structure for containing information about linkages and distances.
ArrayIterator<E>	Generic ArrayIterator.
ArrayJoiner	A Joiner utility that works for primitive arrays which Guava's Joiner can't deal with.
ArrayUtils	Array utils for stuff that isn't included in Arrays.
ArticleContentExtrator	Extractor for news articles.
ArticleContentExtrator.ContentFetchResult	Article content fetch result.
AsyncBufferedOutputStream	BufferedOutputStream that asynchronously flushes to disk, so callers don't have to wait until the flush happens.
BasicFeatureExtractor	Basic feature extraction for sequence learning, takes the current word into account and the previous label - as well as the joint version of both.
BigramTokenizer	Advanced tokenizer that lowercases, adds start and end tags, deduplicates tokens and builds bigrams.
BitSetWritable
BlockPartitioner	This partitioner partitions connected ranges from 0 to numberOfRows into sizeOfCluster buckets.
Boundaries
Boundaries.Range
ByteBufferInputStream
CanopyClustering	Sequential canopy clusterer.
Classifier	Classifier interface for predicting categorial variables.
ClassifierFactory<A extends Classifier>	Factory interface for building new classifiers, majorly used in crossvalidation to generate new classifiers when needed.
Cluster
CollectionInputProvider<T>	Provider for generic collections to read.
ConditionalLikelihoodCostFunction	Conditional likelihood cost function, used in a maximum entropy markov model to optimize the weights.
ConsoleResultWriter<T extends FetchResult>	Simple class that outputs to console.
CosineDistance
CostFunction	Cost function interface to be implemented when using with a optimizer like conjugate gradient for example.
CostGradientTuple	More readable variant of the before used Tuple<> in CostFunction.
CountMinSketch<T>
Crawler<T extends FetchResult>	Basic Crawler Interface, all implements should implicit give a constructor with the same arguments like setup and redirect the call to it.
CrossEntropyLoss	Cross entropy error function, for example to be used with the SoftMaxActivationFunction.
CsvDatasetReader	Binary dataset reader from CSVs.
Dataset	Simplistic dataset to carry information about them.
DBSCAN	Sequential version of DBSCAN to evaluate if this algorithm is suitable for arbitrary parallelization paradigms that can crunch graphs.
DBSCANClustering	Plain sequential DBSCAN clustering.
DecisionTree	A decision tree that can be used for classification with numerical or categorical features.
DenseMatrixFolder
DiskList<E extends org.apache.hadoop.io.Writable>	A file backed disk for adding elements and reading from them in a sequential fashion.
DiskList.IORuntimeException
DistanceMeasurer
DistanceResult<TYPE>	Immutable generic distance result that contains a document type object and its distance (to some artificial queried document).
DocumentSimilarity	Simply distance measure wrapper for debug string similarity measuring.
ElliotActivationFunction	Implementation of the elliot activation function.
EuclidianDistance
EvaluationListener<A extends Classifier>	The evaluation listener is majorly used to track the overfitting of a classifier while training.
EvaluationSplit	Split data class that contains the division of train/test vectors.
Evaluator	Binary-/Multi-class classification evaluator utility that takes care of test/train splitting and its evaluation with various metrics.
Evaluator.EvaluationResult
Extractor<T extends FetchResult>	Simple extraction logic interface for a site and a result.
FeatureOutcomePair
FeatureType	Denotes a feature type, either numerical or nominal.
FetchResult	Fetch Result class, contains the origin url and its outlinks for further crawling.
FetchResultPersister<T extends FetchResult>	Asynchronous persister thread, taking a resultwriter and handles the logic behind asynchronous writing to disk or an arbitrary sink implemented by the ResultWriter.
FetchThread<T extends FetchResult>	Callable fetcher that extracts, for a given list of URLs and with a given Extractor, the content from the list of urls.
Fmincg	Minimize a continuous differentialble multivariate function.
GradientDescent	Gradient descent implementation with some neat features like momentum, divergence detection, delta breaks and bold driver and scheduled annealing adaptive learning rates.
GradientDescent.GradientDescentBuilder
HaversineDistance	Haversine distance implementation that picks up lat/lng in degrees at array/vector index 0 and 1 and returns the distance in meters between those two vectors.
HingeLoss	Hinge-loss for linear SVMs.
HMM	Hidden Markov Model implementation for multiple observations for all three types of problems HMM aims to solve (Decoding, likelihood estimation, unsupervised/supervised learning).
HtmlExtrator	Extractor for raw html.
HtmlExtrator.HtmlFetchResult	Article content fetch result.
HuffmanTree<VALUE>	Huffman tree coder that takes input from a Multiset and returns huffman codes HuffmanTree.getHuffmanCodes().
ImageReader	BufferedImage reader that exports the raw bytes as a feature vectors with different encodings.
InputProvider<T>	A provider that provides generic input from a generic source as an iterator that can be read over and over again.
IntegerFunnel
InvertedIndex<DOCUMENT_TYPE,KEY_TYPE>	Inverted Index, mainly developed for sparse vectors to speedup dimension lookups for fast distance measurement and search space reduction.
InvertedIndex.DocumentDistanceMeasurer<DOCUMENT_TYPE,KEY_TYPE>	Measurer that measures distance of two documents.
InvertedIndex.DocumentMapper<DOCUMENT_TYPE,KEY_TYPE>	Mapper that maps a document to its keys.
IrisReader	Dataset vectorizer for the iris dataset.
Iterables	Some fancy utilities for iterables, e.G.
IterationCompletionListener	Callback that should be triggered when a iteration was finished.
IterativeSimilarityAggregation	Iterative similarity aggregation for named entity recognition and set expansion based on the paper "SEISA: Set Expansion by Iterative Similarity Aggregation".
JaccardDistance
KMeansClustering	Sequential version of k-means clustering.
KNearestNeighbours	K nearest neighbour classification algorithm that is seeded with a "database" of known examples and predicts based on the k-nearest neighbours majority vote for a class.
LeafNode
LinearActivationFunction	Linear activation function.
ListUtils	List util class for some fancy operations on generic lists.
LogActivationFunction	Log activation function, guarded against NaN and infinity edge cases.
LogisticRegression
LogisticRegressionCostFunction
LogLoss	Logistic error function implementation.
LongFunnel
LossFunction	Calculates the error, for example in the last layer of a neural net.
LRUCache<K,V>	Normal LRU cache based on LinkedHashMap.
LUVColorSpace	Represents the LUV colorspace.
ManhattanDistance
MarkovChain	Markov chain, that can "learn" the state transition probabilities by a given input and returns the probability for a given sequence of states.
MathUtils	Math utils that features normalizations and other fancy stuff.
MathUtils.PredictionOutcomePair
MatrixWritable	Writable class for dense and sparse matrices.
MaxEntMarkovModel	Maximum entropy markov model for named entity recognition (classifying labels in sequence learning).
MeanAbsoluteLoss	MAE for regression problems.
MeanShiftClustering	Sequential Mean Shift Clustering using a gaussian kernel and euclidian distance measurement.
Merger<M extends org.apache.hadoop.io.WritableComparable>	Sorted segment merger on disk.
MinHash	Linear MinHash algorithm to find near duplicates faster or to speedup nearest neighbour searches.
MinHash.HashType
Minimizer	Minimizer interface for various function minimizers.
MLPWeightMapper
MNISTReader	MNIST CSV reader from kaggle: www.kaggle.com/c/digit-recognizer/
MultilayerPerceptron	Multilayer perceptron implementation that works on GPU via JCuda and CPU.
MultilayerPerceptron.MultilayerPerceptronBuilder	Configuration for training a neural net through the Classifier
MultilayerPerceptronCostFunction	Neural network costfunction for a multilayer perceptron.
MultilayerPerceptronCostFunction.NetworkConfiguration
MultinomialNaiveBayes	Multinomial naive bayes classifier.
MultithreadedCrawler<T extends FetchResult>	Fast multithreaded crawler, will start a fixed threadpool of 32 threads each will be fed by 10 urls at once.
MushroomReader	Dataset vectorizer for the mushroom dataset.
NegatedCostFunction	Negated cost function to implement maximization problems.
NominalNode
NumericalNode
OnePassExclusiveClustering	A one pass exclusive clustering algorithm.
OutlinkExtractor	Outlink extractor, parses a page just for its outlinks.
OWLQN	Java translation of C++ code of "Orthant-Wise Limited-memory Quasi-Newton Optimizer for L1-regularized Objectives" (@see http://research.microsoft.com/).
Pair<S,T>	Pair implementation, unlike Tuple this implements hashcode and equals on both parts of this pair.
ParticleSwarmOptimization	Particle Swarm Optimization algorithm to minimize costfunctions.
Partitioner	Used to partition a list/matrix-like structure to a number of cores / buckets.
Permutations<T extends java.lang.Comparable<? super T>>
Predictor
RandomForest	A decision tree forest, using bagging.
RBM	Class for training and stacking Restricted Boltzmann Machines (RBMs).
RBM.RBMBuilder
RBMCostFunction	Restricted Boltzmann machine implementation using Contrastive Divergence 1 (CD1).
ReferencedContext<REF_TYPE,CONTEXT_TYPE>	Reference and its context.
ReluActivationFunction	Rectified linear units implementation.
ResultWriter<T extends FetchResult>	Result writing interface.
ResultWriterAdapter<T extends FetchResult>	Empty Adapter class for a ResultWriter.
SequenceFeatureExtractor<K>	Interface for feature extraction in sequence learning.
SequenceFileResultWriter<T extends FetchResult>	Writes the result into a sequencefile "files/crawl/result.seq".
SequentialCrawler<T extends FetchResult>	Sequential crawler, mainly for debugging or development.
SigmoidActivationFunction	Implementation of the sigmoid function.
SimilarityMeasurer	Similarity measurer wrapper.
SingleLinkedList<T>	Single Linked list with less overhead in memory than the double linked list of Java utils.
SoftMaxActivationFunction	Softmax activation that only works on vectors, because it needs to sum and divide the probabilities.
SoftplusReluActivationFunction	Smoothed approximation to a ReluActivationFunction.
SortedFile<M extends org.apache.hadoop.io.WritableComparable>	A file that serializes WritableComparables to a buffer, once it hits a threshold this buffer will be sorted in memory.
SparseFeatureExtractorHelper<K>	Convenient helper for creating vectors out of text features for sequence learning.
SparseKNearestNeighbours	K nearest neighbour classification algorithm that is seeded with a "database" of known examples and predicts based on the k-nearest neighbours majority vote for a class.
SparseVectorDocumentMapper	Mapper that maps sparse vectors into a set of their indices so they can be used in the InvertedIndex for fast lookup.
Split	From Mahout, split class with better naming.
SquaredLoss	Squared mean error function for regression problems and LinearActivationFunction.
StackMap<K,V>	A stack that also provides random access lookup of values.
StandardTokenizer	Just a basic tokenizer by certain attributes with normalization.
Statistics	Small statistics utility to describe data based on its min/max/mean/median/deviation.
StepActivationFunction	Classic perceptron-like step function.
StepLoss	Calculates a step error function that can be used for StepActivationFunction.
StringPool	Simple map based StringPool that is considered faster than using String.intern(), but uses a bit more memory.
TanhActivationFunction	Implementation of the Tanh activation based on FastMath.
TestSetIterationCallback<T extends Classifier>	This callback is used to evaluate the performance on a held-out test set.
TextLineInputProvider	Line reader for plain text that contains data in lines.
Tokenizer	Standard tokenizer interface.
TokenizerUtils	Nifty text utility for majorly tokenizing tasks.
TrainingSplit
TrainingType	Train normally on the CPU or on the GPU via CUDA?
TreeCompiler	Compilation unit for the object tree structure of the DecisionTree.
TwentyNewsgroupReader	Reads the "20news-bydate" dataset into a vector space model as well as predictions based on the category.
UnrollableDoubleVector	Unrollable proxy double vector class, that wraps multiple vectors into one that can be later unrolled.
UntrainableClassifier
VectorDocumentDistanceMeasurer<T>	Document distance measurer on vectors (basically a proxy to the real DistanceMeasurer).
VectorFunnel	A funnel that funnels a DoubleVector into bytes by taking the non-zero items from a vector for sparse instances.
VectorizerUtils	Vectorizing utility for basic tf-idf and wordcount vectorizing of tokens/strings.
VectorWritable	New and updated VectorWritable class that has all the other fancy combinations of vectors that are possible in my math library. This class is not compatible to the one in the clustering package that has a totally different byte alignment in binary files.
ViterbiUtils	Viterbi Utilities for forward backward passes and his famous decoding algorithm for hidden markov models.
Voter<A extends Classifier>	Implementation of vote ensembling.
Voter.CombiningType
Voter.SelectionType
WeightMapper<A extends Classifier>	This interface helps to map minimizable weights of a CostFunction to a Classifier implementation.
WeightMatrix	Weight matrix wrapper to encapsulate the random initialization.
ZeroDistance