Add alternative scale parametrization to Exponential

[ricardoV94]

Description

scale = 1 / lam
PyMC currently uses lam, and others (e.g., Numpy) use scale

This should be relatively simple, here is an existing example of a distribution with multiple parametrizations:

pymc/pymc/distributions/discrete.py

Lines 126 to 132 in 380e68c

	if p is not None and logit_p is not None:
	raise ValueError("Incompatible parametrization. Can't specify both p and logit_p.")
	elif p is None and logit_p is None:
	raise ValueError("Incompatible parametrization. Must specify either p or logit_p.")
	if logit_p is not None:
	p = pt.sigmoid(logit_p)

Test example:

pymc/tests/distributions/test_discrete.py

Lines 962 to 966 in 9a3f504

	class TestBernoulliLogitP(BaseTestDistributionRandom):
	pymc_dist = pm.Bernoulli
	pymc_dist_params = {"logit_p": 1.0}
	expected_rv_op_params = {"p": sp.expit(1.0)}
	checks_to_run = ["check_pymc_params_match_rv_op"]

See also docstrings examples of how multiple parametrizations are described.

[manulpatel]

Hi! @ricardoV94. I would like to work on the issue.

[ricardoV94]

Sure, go ahead