Closed rosemckeon closed 5 years ago
@rozeykex I can only seem to replicate the appearance of Inf
when genome values are very large (ca 10000000
). Is it possible that values are greatly exceeding 100, then causing Inf
(due to limitations of double
) and running into an issue with L65 where the maximum of 100 is hard coded?
@bradduthie hmmm. Weird that you're not seeing this happen, or maybe weirder that I am! I can't see how allele values could increase. They have only 2 opportunities to change; when they are set originally, and when they mutate. In both cases, random uniform numbers between 0 and 100 are chosen https://github.com/rozeykex/ploidy/blob/b11a87e82ace493410e404d7cab109a55eada85d/R/functions.R#L1002 and https://github.com/rozeykex/ploidy/blob/b11a87e82ace493410e404d7cab109a55eada85d/R/functions.R#L567
I'll try and plot the allele values to make sure.
@bradduthie yeah, I'm not getting the issue either anymore! Though I'm getting extinction appearing earlier so maybe need to tweak settings that allow the population to persist for longer. I've fixed the unnesting issues with genomes though to examine the values, and they're definitely constrained.
Let's just keep an eye on it for now.
@bradduthie I thought I'd managed to code what you explained for
get_growth_rate
to constrain it below a max value, but as generations continue values ofInf
begin to appear.Please can you take a look at traits.R and see what you think?