Niek de Jonge

Yes I agree, generating a sparse matrix (or graph) will solve the scalability issue, without repeatedly calculating the same scores. Before switching to new ways of pair generating I hope...

A new filter was created in #622, to only repair the adduct, without repairing the parent mass on the smiles. I planned to split the two functions, but now I...

My first idea was actually to keep the EmbeddingEvaluator and the LinearModel as separate classes, but to save them to the same file. However, I like the idea of having...

@florian-huber I am implementing this now. We have a few settings that are now stored in SettingsMS2Deepscore (e.g. settings.evaluator_num_filters), but others have to be provided separately and are therefore not...

And if we want to add them, shall I add them all to SettingsMS2Deepscore or shall I make a new Class EmbeddingsEvaluationModelSettings, or just add them as separate parameters in...

Yes, I noticed this issue as well. You would actually like to have more of a tree-like structure of filters. One of the cases this was the case was running...

Please do not review/merge yet. First #43590 should be merged.

@hechth @florian-huber I agree with both points, it makes sense to remove this and make this a function. But it will indeed also break functionality of Spec2Vec and MS2Query. For...

Yes I agree that we can close this issue. Pickle still has some issues: - Security issues (mainly an issue if you want to share spectra, so not that important)...

@hechth No if the Spectrum class does not change this does not give any issues and the Spectrum object can be loaded in fine. However if any method name or...