Fill missing values with NaN in observation_data_to_array (facebook#2430

)

Summary:

This would previously filter out the observations with missing metrics with a warning, which could lead to issues in downstream usage. Returning the observations with NaN lets the downstream users deal with NaNs while keeping the expected return shapes.

Differential Revision: D56949439

ax/modelbridge/modelbridge_utils.py

@@ -9,10 +9,8 @@
 from __future__ import annotations
 
 import warnings
-
 from copy import deepcopy
 from functools import partial
-
 from logging import Logger
 from typing import (
     Any,
@@ -55,7 +53,6 @@
     SearchSpaceDigest,
 )
 from ax.core.types import TBounds, TCandidateMetadata
-
 from ax.core.utils import (  # noqa F402: Temporary import for backward compatibility.
     get_pending_observation_features,  # noqa F401
     get_pending_observation_features_based_on_trial_status,  # noqa F401
@@ -1198,29 +1195,34 @@ def observation_data_to_array(
 ) -> Tuple[np.ndarray, np.ndarray]:
     """Convert a list of Observation data to arrays.
 
+    Any missing mean or covariance values will be returned as NaNs.
+
     Args:
-        observation_data: A list of n ObservationData
+        outcomes: A list of `m` outcomes to extract observation data for.
+        observation_data: A list of `n` ``ObservationData`` objects.
 
     Returns:
-        An array of n ObservationData, each containing
-            - f: An (n x m) array
-            - cov: An (n x m x m) array
+        - means: An (n x m) array of mean observations.
+        - cov: An (n x m x m) array of covariance observations.
     """
     means = []
     cov = []
-    for obsd in observation_data:
-        try:
-            metric_idxs = np.array([obsd.metric_names.index(m) for m in outcomes])
-        except ValueError:
-            missing = set(outcomes).difference(set(obsd.metric_names))
-            logger.warning(
-                f"{obsd} is missing the metrics {missing}. Ignoring the data "
-                "for the remaining metrics."
-            )
-            continue
-        means.append(obsd.means[metric_idxs])
-        cov.append(obsd.covariance[metric_idxs][:, metric_idxs])
-    return np.array(means), np.array(cov)
+    # Initialize arrays with all NaN values.
+    means = np.full((len(observation_data), len(outcomes)), np.nan)
+    cov = np.full((len(observation_data), len(outcomes), len(outcomes)), np.nan)
+    # Iterate over observations and extract the relevant data.
+    for i, obsd in enumerate(observation_data):
+        # Indices of outcomes that are observed.
+        outcome_idx = [j for j, o in enumerate(outcomes) if o in obsd.metric_names]
+        # Corresponding indices in the observation data.
+        observation_idx = [obsd.metric_names.index(outcomes[j]) for j in outcome_idx]
+        means[i, outcome_idx] = obsd.means[observation_idx]
+        # We can't use advanced indexing over two dimensions jointly for assignment,
+        # so this has to be done in two steps.
+        cov_pick = np.full((len(outcome_idx), len(outcomes)), np.nan)
+        cov_pick[:, outcome_idx] = obsd.covariance[observation_idx][:, observation_idx]
+        cov[i, outcome_idx] = cov_pick
+    return means, cov
 
 
 def observation_features_to_array(

ax/modelbridge/tests/test_utils.py

@@ -6,7 +6,6 @@
 
 # pyre-strict
 
-from unittest import mock
 from unittest.mock import patch
 
 import numpy as np
@@ -219,15 +218,25 @@ def test_observation_data_to_array(self) -> None:
             means=np.array([1, 2]),
             covariance=np.array([[1, 2], [4, 5]]),
         )
-        with mock.patch("ax.modelbridge.modelbridge_utils.logger.warning") as mock_warn:
-            Y, Ycov = observation_data_to_array(
-                outcomes=outcomes, observation_data=[obsd, obsd2]
-            )
-        self.assertTrue(np.array_equal(Y, np.array([[2, 3, 1]])))
+        Y, Ycov = observation_data_to_array(
+            outcomes=outcomes, observation_data=[obsd, obsd2]
+        )
+        nan = float("nan")
         self.assertTrue(
-            np.array_equal(Ycov, np.array([[[5, 6, 4], [8, 9, 7], [2, 3, 1]]]))
+            np.array_equal(Y, np.array([[2, 3, 1], [2, nan, 1]]), equal_nan=True)
+        )
+        self.assertTrue(
+            np.array_equal(
+                Ycov,
+                np.array(
+                    [
+                        [[5, 6, 4], [8, 9, 7], [2, 3, 1]],
+                        [[5, nan, 4], [nan, nan, nan], [2, nan, 1]],
+                    ]
+                ),
+                equal_nan=True,
+            )
         )
-        mock_warn.assert_called_once()
 
     def test_feasible_hypervolume(self) -> None:
         ma = Metric(name="a", lower_is_better=False)