Coding for population size initial, upper and lower parameters about babette HOT 11 CLOSED

Great, I'll send you and email then.

Oh yes - I always forget about the private setting! haha

from babette.

Hi Gary, thanks for this awesome bug report!

The coalescent model is only a modestly tested model. And I'd be happy to fix this -probably- bug. Could you post the BEAST2 XML file? And, will you volunteer to test if it indeed works?

If yes, I will probably fix it before Sunday 12:00 👍

To answer you questions:

Am I right in saying that id and pop_size_distr are the only arguments to create_ccp_tree_prior()?

Yes, and I think that is still correct. I feel I need to add some parameters to create_log_normal_distr. Note that one never needs to initialize id 👍

Actually, in general is there a way of parsing the paper's XML using Babette so that I can find what functions to use - in the same way that I've managed to load to XML into Beauti so I can see explicitly what parameters they've used?

No, there is no support for that yes, although tiebeaur (the reverse of 'beautier') was an initial start of this idea. But if you want to help add this, sure, we'll have an online chat 👍

from babette.

Hi Richèl, many thanks for getting back to me so quickly.

Yes, I'd certainly be happy to test this!

The XML file can be found here: https://github.com/GaryNapier/transmission/blob/main/beast_xml/BEAST2_GTR_Model_London_TB_Outbreak.xml

Ok, thanks for letting me know about the 'reverse' XML upload.

It'd be good to have an online chat, thanks - I'm making my way through the "London study" XML and there are a couple of other settings/parameters that I doubt would be coded in Babette, so it might be quicker to let you know about these online.

Many thanks!

Gary

from babette.

Hi Gary, fun, lets meet online! You don't have an email on your GitHub profile, could you send a short email to me? We'll take care of things from there. Fun!

from babette.

P.S. The repo is private, so I cannot see it. You could make me a Collaborator 👼

from babette.

Added at ropensci/beautier#121, time for @GaryNapier to check my work 👍

from babette.

Hi Richel,

Many thanks again for your help, and so quick.

I'm just trying to run create_log_normal_distr() again after running remotes::install_github("ropensci/beautier").

My code is:

beautier::create_log_normal_distr(
      m = 1,
      s = 1.25,
      initial_value=100.0,
      lower = 0.0,
      upper = 200.0
    )

But the initial_value, lower and upper are described as unused arguments.

When I run ?create_log_normal_distr these arguments show up in the help file, but when I look at the function itself they don't seem to be there:

> create_log_normal_distr
function (id = NA, m = 0, s = 0) 

Perhaps the function hasn't been pushed to Github or something?

Thanks!

from babette.

Silly me, will fix tonight and send an email when done. Thanks for checking me 😊

from babette.

haha cheers!

from babette.

Added the trivial test below, which works:

test_that("log_normal_distr, complete use", {
  m <- 1.0
  s <- 1.25
  value <- 100.0
  lower <- 0.01
  upper <- 200.0
  log_normal_distr <- create_log_normal_distr(
    m = m,
    s = s,
    value = value,
    lower = lower,
    upper = upper
  )
  expect_equal(m, log_normal_distr$m$value)
  expect_equal(s, log_normal_distr$s$value)
  expect_equal(value, log_normal_distr$value)
  expect_equal(lower, log_normal_distr$lower)
  expect_equal(upper, log_normal_distr$upper)
})

Whatever it was. it is fixed now 👍

from babette.

It works, except there is another Issue at #93.

from babette.