Commit 426ca09a authored by Robert P. Goldman's avatar Robert P. Goldman Committed by Robert P. Goldman

Fixed incorrect reshaping.

I believe that this reshaping -- which flattened the samples -- was a hold-over from the pointwise variant.  It was causing test failures.
parent 0ae72c57
......@@ -424,8 +424,8 @@ class _PosteriorPredictiveSampler(AbstractContextManager):
evaluated[i] = drawn[(var, samples)]
# We filter out Deterministics by checking for `model` attribute
elif name is not None and hasattr(var, 'model') and name in trace.varnames:
# is in point
drawn[(var, samples)] = evaluated[i] = trace[cast(str, name)].reshape(-1)
# is in the trace. Record it as drawn and evaluated
drawn[(var, samples)] = evaluated[i] = trace[cast(str, name)]
# param still needs to be drawn
symbolic_params.append((i, var))
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment