posted on 2024-02-20, 18:26authored byDylan D. Ray, Lex Flagel, Daniel R. Schrider
The confusion matrix shows IntroUNET’s classification performance when applying the same trained network used for the simple bidirectional introgression scenario whose parameters are laid out in Table 1 to data simulated under the the D. melanogaster BGS model specified in the Methods. Note that the shading represents the number of examples in each entry of the confusion matrix rather than the fraction of examples.