To follow my wanderings through other decontamination/metagenomic software, see dataset_explorations.ipynb.
The three "results" directories contain some results from test datasets that I ran each of the programs with.
For a look at what I did to prep/clean the data I use to train the ML model, see data_for_ml.ipynb.
The parser that features prominently in this notebook is "parse_kraken_output.py"
The percent_plots directory has density plots of the percentage of classified kmers in kraken2 and krakenuniq results files.