search

iomega / spec2vec

Word2Vec based similarity measure of mass spectrometry data.
Apache License 2.0
62 stars 16 forks source link

reduce the size score after calling calculate_stores #86

Closed r00bit closed 4 months ago

r00bit commented 1 year ago

Hi, After training spec2vec model and creating spec2vec_similarity function, I used calculate_scores to find the similarity of spectra pairs scores = calculate_scores(reference_documents, query_spectrums, spec2vec_similarity) However the size scores is too big to process it, how can I reduce the size of scores by saving only the non-zero scores with their related reference and query scan numbers? Thanks!

florian-huber commented 1 year ago

We are currently working on such a sparse implementation. I hope this will be ready very soon (few weeks... again: I hope).

florian-huber commented 4 months ago

Done and outdated.