Predict( model, 'koning filip ', k = 10, type = 'knn ') 'de migranten komen naar europa de asielcentra.

Text <- c( 'de nmbs heeft het treinaanbod uitgebreid via onteigening. Similarity = 'cosine ', negSearchLimit = 50, Lr = 0.01, epoch = 40, loss = 'softmax ', adagrad = TRUE, Model <- embed_tagspace( x = dekamer $ x, y = dekamer $ y, Get words/labels part of the model dictionaryĬosine/dot product similarity between embeddings - top-n most similar textīuild a Starspace model which calculates word/ngram embeddingsīuild a Starspace model which calculates sentence embeddingsīuild a Starspace model for embedding articles - sentence-article similaritiesīuild a Starspace model for multi-label classificationīuild a Starspace model for content-based recommendationīuild a Starspace model for interest-based recommendationīuild a Starspace model for entity relationship completionĮxample Short example showing word embeddingsĭata( 'dekamer ', package = 'ruimtehol ')ĭekamer $ x <- strsplit( dekamer $ question, ' \W ')ĭekamer $ x <- sapply( dekamer $ x, FUN = function( x) paste(setdiff( x, ' '), collapse = ' '))ĭekamer $ y <- strsplit( dekamer $ question_theme, split = ', ')ĭekamer $ y <- lapply( dekamer $ y, FUN = function( x) gsub( ' ', '- ', x)) Get embeddings of documents/words/ngrams/labelsįind k-nearest neighbouring information for new text Load a pre-trained model or a tab-separated file Low-level interface to build a Starspace model The following functions are made available. classification / ranking) / Get nearest neighbours similarity This R package allows to Build Starspace models on your own text / Get embeddings of words/ngrams/sentences/documents/labels / Get predictions from a model (e.g. Vignette('ground-control-to-ruimtehol', package = 'ruimtehol')

Tagspace alternative