error in multiprocessing when using current NumPy version

[thaynecurrie]

Following some of the recent Issues (#60, #62), I edited the orvara source code to remove instances of np.float and np.int and modify .join on the pyplot shareaxis calls.

[I also got Taichi Uyama's error of a mismatch of on the PA and rel-separation output plots between the top and bottom panels but fixed it with a hardcoded sharex command as here: https://stackoverflow.com/questions/77418896/attributeerror-grouperview-object-has-no-attribute-join.]

This gets orvara to run with the current version of NumPy (1.26.4) using one thread.

However, when I attempt to use multiple threads for parallelization, I get a
"AttributeError: 'NoneType' object has no attribute 'nInst'"
on line 206 of main.py .

Is there a quick workaround to this? I am running with Python 3.12 on an Macbook M3 Max and NumPy 1.26.4. On my older laptop running Numpy 1.22.6 and orvara 1.1.2 I get no errors.

---

Running MCMC ...
[ ] 0%multiprocessing.pool.RemoteTraceback:
"""
Traceback (most recent call last):
File "/Users/thaynecurrie/anaconda3/lib/python3.12/multiprocessing/pool.py", line 125, in worker
result = (True, func(*args, **kwds))
^^^^^^^^^^^^^^^^^^^
File "/Users/thaynecurrie/anaconda3/lib/python3.12/multiprocessing/pool.py", line 48, in mapstar
return list(map(*args))
^^^^^^^^^^^^^^^^
File "/Users/thaynecurrie/anaconda3/lib/python3.12/site-packages/ptemcee/sampler.py", line 137, in call
ll = self.logl(x, *self.loglargs, **self.loglkwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/thaynecurrie/Research/Github/repositories/orvara_py311/orvara/orvara/main.py", line 253, in avoid_pickle_lnprob
return lnprob(theta, **_loglkwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/thaynecurrie/Research/Github/repositories/orvara_py311/orvara/orvara/main.py", line 206, in lnprob
params = orbit.Params(theta, i, nplanets, data.nInst, njitters)
^^^^^^^^^^
AttributeError: 'NoneType' object has no attribute 'nInst'
"""