Closed chaozg closed 1 month ago
Description
The _acc attribute of NUTS has some extra None values. Maybe @amal-ghamdi you know the reason behind it?
_acc
None
Example to reproduce
import cuqi import matplotlib.pyplot as plt import numpy as np gaussian = cuqi.distribution.Gaussian(np.array([0.0]), np.array([1.0])) sampler = cuqi.experimental.mcmc.NUTS(gaussian, initial_point=np.array([0.1])) sampler.sample(10) sampler._acc
sampler._acc has some None values as seen here:
sampler._acc
[None, 0, None, 0, None, 0, None, 0, None, 1, None, 1, None, 0, None, 1, None, 0, None, 0, None]
Note: we're drawing 10 samples, but there are 21 elements in _acc, including 10 extra None.
Description
The
_acc
attribute of NUTS has some extraNone
values. Maybe @amal-ghamdi you know the reason behind it?Example to reproduce
sampler._acc
has someNone
values as seen here:Note: we're drawing 10 samples, but there are 21 elements in
_acc
, including 10 extraNone
.