Skip to content

bugfix for fit_laplace absent dims #609

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
Dekermanjian wants to merge 3 commits into
base: main
Choose a base branch
from
+45 −4
Open

bugfix for fit_laplace absent dims #609

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
52d8a92
updated dim_shape assignment logic in fit_laplace to handle absent di…
Dekermanjian Nov 7, 2025
1e41758
reverted to numpy arange
Dekermanjian Nov 8, 2025
5abe4ff
updated dim handling in model_to_laplace_approx to not force dims on …
Dekermanjian Nov 8, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
11 changes: 7 additions & 4 deletions pymc_extras/inference/laplace_approx/laplace.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -224,12 +224,15 @@ def model_to_laplace_approx(
elif name in model.named_vars_to_dims:
dims = (*batch_dims, *model.named_vars_to_dims[name])
else:
dims = (*batch_dims, *[f"{name}_dim_{i}" for i in range(batched_rv.ndim - 2)])
initval = initial_point.get(name, None)
dim_shapes = initval.shape if initval is not None else batched_rv.type.shape[2:]
laplace_model.add_coords(
{name: np.arange(shape) for name, shape in zip(dims[2:], dim_shapes)}
)
if dim_shapes[0] is not None:
Copy link
Member

@ricardoV94 ricardoV94 Dec 10, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I don't follow what's hapening here. If dim_shapes is coming from rv.type.shape, you could have dim_shapes[0] not be None, and a follow up be None. Example x = pm.Data("x", np.zeros(5, 3), shape=(5, None) is valid.

The whole approach seems a bit backwards. Instead of trying to plug the sample dims in the model that may not have properly defined dims to begin with, why not sample first and then attach those in the final inference data object, which always has dims, explicit or automatic?

If you just want to patch for now be more exhaustive and check not any(d is None for d in dim_shapes)

dims = (*batch_dims, *[f"{name}_dim_{i}" for i in range(batched_rv.ndim - 2)])
laplace_model.add_coords(
{name: np.arange(shape) for name, shape in zip(dims[2:], dim_shapes)}
)
else:
dims = None

pm.Deterministic(name, batched_rv, dims=dims)

Expand Down
38 changes: 38 additions & 0 deletions tests/inference/laplace_approx/test_laplace.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -193,6 +193,44 @@ def test_fit_laplace_ragged_coords(rng):
assert (idata["posterior"].beta.sel(feature=1).to_numpy() > 0).all()


def test_fit_laplace_no_data_or_deterministic_dims(rng):
coords = {"city": ["A", "B", "C"], "feature": [0, 1], "obs_idx": np.arange(100)}
with pm.Model(coords=coords) as ragged_dim_model:
X = pm.Data("X", np.ones((100, 2)))
beta = pm.Normal(
"beta", mu=[[-100.0, 100.0], [-100.0, 100.0], [-100.0, 100.0]], dims=["city", "feature"]
)
mu = pm.Deterministic("mu", (X[:, None, :] * beta[None]).sum(axis=-1))
sigma = pm.Normal("sigma", mu=1.5, sigma=0.5, dims=["city"])

obs = pm.Normal(
"obs",
mu=mu,
sigma=sigma,
observed=rng.normal(loc=3, scale=1.5, size=(100, 3)),
dims=["obs_idx", "city"],
)

idata = fit_laplace(
optimize_method="Newton-CG",
progressbar=False,
use_grad=True,
use_hessp=True,
)

# These should have been dropped when the laplace idata was created
assert "laplace_approximation" not in list(idata.posterior.data_vars.keys())
assert "unpacked_var_names" not in list(idata.posterior.coords.keys())

assert idata["posterior"].beta.shape[-2:] == (3, 2)
assert idata["posterior"].sigma.shape[-1:] == (3,)

# Check that everything got unraveled correctly -- feature 0 should be strictly negative, feature 1
# strictly positive
assert (idata["posterior"].beta.sel(feature=0).to_numpy() < 0).all()
assert (idata["posterior"].beta.sel(feature=1).to_numpy() > 0).all()


def test_model_with_nonstandard_dimensionality(rng):
y_obs = np.concatenate(
[rng.normal(-1, 2, size=150), rng.normal(3, 1, size=350), rng.normal(5, 4, size=50)]
Expand Down