@@ -74,11 +74,11 @@ class Simulator(NoDistribution):
74
74
75
75
Parameters
76
76
----------
77
- fn: callable
77
+ fn : callable
78
78
Python random simulator function. Should expect the following signature
79
79
``(rng, arg1, arg2, ... argn, size)``, where rng is a ``numpy.random.RandomStream()``
80
80
and ``size`` defines the size of the desired sample.
81
- params: list
81
+ params : list
82
82
Parameters used by the Simulator random function. Parameters can also
83
83
be passed by order, in which case the keyword argument ``params`` is
84
84
ignored. Alternatively, each parameter can be passed by order after fn,
@@ -98,17 +98,17 @@ class Simulator(NoDistribution):
98
98
``distance="gaussian"`` + ``sum_stat="sort"`` is equivalent to the 1D 2-wasserstein distance
99
99
100
100
``distance="laplace"`` + ``sum_stat="sort"`` is equivalent to the the 1D 1-wasserstein distance
101
- sum_stat: Aesara Op, callable or str
101
+ sum_stat : Aesara Op, callable or str
102
102
Summary statistic function. Available options are ``"indentity"`` (default),
103
103
``"sort"``, ``"mean"``, ``"median"``. If a callable (or Aesara Op) is defined,
104
104
it should return a 1d numpy array (or Aesara vector).
105
- epsilon: float or array
105
+ epsilon : float or array
106
106
Scaling parameter for the distance functions. It should be a float or
107
107
an array of the same size of the output of ``sum_stat``. Defaults to ``1.0``
108
108
ndim_supp : int
109
109
Number of dimensions of the SimulatorRV (0 for scalar, 1 for vector, etc.)
110
110
Defaults to ``0``.
111
- ndims_params: list[int]
111
+ ndims_params : list[int]
112
112
Number of minimum dimensions of each parameter of the RV. For example,
113
113
if the Simulator accepts two scalar inputs, it should be ``[0, 0]``.
114
114
Defaults to ``0`` for each parameter.
0 commit comments