Skip to content

Commit

Permalink
BUG Initialize adapt_full with a diagonal covariance matrix rather th…
Browse files Browse the repository at this point in the history
…an ones. (#3891)
  • Loading branch information
twiecki authored Apr 27, 2020
1 parent c1c7700 commit 60cca23
Showing 1 changed file with 1 addition and 1 deletion.
2 changes: 1 addition & 1 deletion pymc3/sampling.py
Original file line number Diff line number Diff line change
Expand Up @@ -1999,7 +1999,7 @@ def init_nuts(
elif init == "adapt_full":
start = [model.test_point] * chains
mean = np.mean([model.dict_to_array(vals) for vals in start], axis=0)
cov = np.ones((model.ndim, model.ndim))
cov = np.eye(model.ndim)
potential = quadpotential.QuadPotentialFullAdapt(model.ndim, mean, cov, 10)
else:
raise ValueError("Unknown initializer: {}.".format(init))
Expand Down

0 comments on commit 60cca23

Please sign in to comment.