Module org.nasdanika.ai

Package org.nasdanika.ai

package org.nasdanika.ai

Class	Description
AbstractDoubleFitter	Collects all samples to double[][] features and double[][] labels, then calls fit(double[][] features, double[][] labels) to fit Function<double[][],double[][]> which makes predictions for multiple features at once.
AbstractMapReduceDoubleFitter	Creates a Function<double[][],double[]> predictor for each label element, predicts label elements using provided predictors and then combines
AbstractRecursiveDoubleFitter	Creates a Function<double[][],double[]> predictor for each label element with features for later elements including labels for earlier.
BagOfWordsGenerator	Splits input by whitespace, lowercases and then computes frequency of each word
BufferedImageSimilarityComputer<S>
ByteArrayImageMetadataNarrator
CachingEmbeddingGenerator<S,E,K>	Caches image embeddings in a map which can be loaded and saved between runs Uses image digest as caching key
CachingImageEmbeddingGenerator<E>	Caches image embeddings in a map which can be loaded and saved between runs Uses image digest as caching key
CachingImageNarrator
CachingTextEmbeddingGenerator<E>	Caches text embeddings in a map which can be loaded and saved between runs Uses text digest as caching key
Chat
Chat.Message
Chat.Requirement	Chat requirement.
Chat.ResponseMessage
Chat.Role
ChatImageNarrator
CompositeFloatSimilarityComputer<T>
Coordinates	Model coordinates (identifier)
EmbeddingGenerator<S,E>	Generates an embedding from source.
EmbeddingGenerator.Requirement	EmbeddingGenerator requirement.
EmbeddingModel<S,E>
FileImageMetadataNarrator
FittedPredictor<F,L,E>	A predictor which is fitted (trained)
FittedPredictor.ErrorComputer<F,L,E>
FittedPredictor.Fitter<F,L,E>
FloatVectorEmbeddingGenerator<S>
FloatVectorSimilarityComputer
ImageEmbeddingGenerator<E>
ImageFloatVectorEmbeddingModel
ImageMetadataNarrator<S>
ImageNarrator	Converts image to text (String).
MapCachingEmbeddingGenerator<S,E,K>	Caches image embeddings in a map which can be loaded and saved between runs Uses image digest as caching key
Model	Base interface for interfaces to work with (large language) models.
Narrator<S>	Converts source to text (String).
Predictor<F,L>	Predicts output from input
Predictor.Sample<F,L>
SearchResult<D extends Comparable<D>>
SimilarityComputer<T,S>	Computes pair-wise similarity
SimilaritySearch<T,D extends Comparable<D>>
SimilaritySearch.EmbeddingsItem	Vector index item
SimilaritySearch.IndexId	Index id - item URI and embedding vector index for URIs with multiple vectors/chunks.
TextEmbeddingGenerator<E>
TextFloatVectorCharChunkingEmbeddings	A simple implementation which treats a character as a token.
TextFloatVectorChunkingEmbeddingModel<T>
TextFloatVectorEmbeddingModel	TextFloatVectorEmbeddingModel "business" interface focusing on ease of use and leaving token usage reporting to implementations.
TextFloatVectorEmbeddingResource	A collection of strings pre-computed embeddings, e.g. web site contents.
TextFloatVectorEmbeddingResourceContents	A pre-computed embeddings
TextFloatVectorEncodingChunkingEmbeddingModel
TextSimilarityComputer<S>
UriImageMetadataNarrator	Can handle data: URIs
UrlImageMetadataNarrator
VectorEmbeddingGenerator<S,E>
VectorSimilarityComputer<E,S>