acres.word2vec package

Package grouping modules related to the word2vec expansion strategy.

Submodules

acres.word2vec.test module

Module to apply/test a given word2vec model.

acres.word2vec.test.find_candidates(acronym, left_context='', right_context='', min_distance=0.0, max_rank=500)

Similar to robust_find_embeddings, this finds possible expansions of a given acronym.

Parameters

• acronym (str) –

• left_context (str) –

• right_context (str) –

• min_distance (float) –

• max_rank (int) –

Return type

Iterator[str]

Returns

acres.word2vec.train module

Trainer for word2vec embeddings based on an idea originally proposed by Johannes Hellrich (https://github.com/JULIELab/hellrich_dh2016).

acres.word2vec.train.train(ngram_size=6, min_count=1, net_size=100, alpha=0.025, sg=1, hs=0, negative=5)

Lazy load a word2vec model.

Parameters

• ngram_size (int) –

• min_count (int) –

• net_size (int) –

• alpha (float) –

• sg (int) –

• hs (int) –

• negative (int) –

Return type

Word2Vec

Returns