Capture-Recapture Models in Python with PyMC3

capture_recapture.ipynb

Hey,

Thanks for this awesome tutorial! I'm planning to make use of it for calculating birth rates for whales, so thanks so much for writing this up!

I have one question. In the Jolly-Seber section, you mentioned the following:

I read the above expression as "everything other than detecting an individual who was alive in the study period" which involves scenarios where individual is alive up to the end of the study but never captured, like you said, but also involves scenarios where individual dies before the end of study. So the description "the probability that an individual is alive at the end of the study period but never captured" is not entirely correct, is it?

I think the probability that an individual is alive at the end of the study period but never captured" is correct - the definition of $\Psi$ includes the survival probability $\phi$.

Hi,

I am new to Python and am trying to run the Jolly-Seber example as the building block of an analysis I want to do. I am having trouble when I try to run:

with js_model:
js_trace = pm.sample(**SAMPLE_KWARGS)

It gives me this traceback. I imagine it has to do with syntax changing since the example was produced but I am not sure how to diagnose given how new I am to python. This might be a silly problem, sorry!

Traceback (most recent call last):
File "", line 1, in
File "C:\Users\16307\anaconda3\lib\multiprocessing\spawn.py", line 116, in spawn_main
exitcode = _main(fd, parent_sentinel)
File "C:\Users\16307\anaconda3\lib\multiprocessing\spawn.py", line 126, in _main
self = reduction.pickle.load(from_parent)
AttributeError: Can't get attribute 'IncompleteMultinomial' on <module 'main' (built-in)>

EDIT: It seems that "njobs" has been deprecated in favor of "cores".