fixed disruptive nn bug
split nn assignment plots by plate and include samples with insufficient targets to be run through NN as blanks and order samples by well_id
deduplicated output files for nn and vae and renamed
made a start at aggregation script for all results -- currently only merges nn and vae results, still have to include prep, qc and plasm
we also need to discuss the default directory structure and which columns should be included in the results file (e.g. I now include the standard deviations inferred by the VAE, but I don't think that's relevant )
fixed disruptive nn bug split nn assignment plots by plate and include samples with insufficient targets to be run through NN as blanks and order samples by well_id deduplicated output files for nn and vae and renamed made a start at aggregation script for all results -- currently only merges nn and vae results, still have to include prep, qc and plasm we also need to discuss the default directory structure and which columns should be included in the results file (e.g. I now include the standard deviations inferred by the VAE, but I don't think that's relevant )