aimz.ImpactModel.log_likelihood#

ImpactModel.log_likelihood(X, y=None, *, shard_axis='obs', batch_size=None, output_dir=None, progress=True, **kwargs)[source]#

Compute the log-likelihood of the data under the given model.

Results are written to disk in the Zarr format, with computing and file writing decoupled and executed concurrently.

Parameters:

  • X (ArrayLike | ArrayLoader) – Input data. If array-like, the leading axis is

  • Alternatively (the observation axis.)

  • array-like (a data loader that holds all)

  • internally. (objects and handles batching)

  • y (ArrayLike | None) – Output data. The leading axis is the observation axis. Must be None if X is a data loader.

  • shard_axis (Literal['obs', 'draw']) – Multi-device sharding strategy; no effect on a single device. "obs" (default) shards the input across devices and replicates the posterior. "draw" shards the posterior across devices and replicates the input, which must be an array, not a data loader. If the model has no posterior samples, the data path is used regardless of shard_axis.

  • batch_size (int | None) – Size of each batch, taken from the input under shard_axis="obs" and from the draws under shard_axis="draw". Also used as the chunk size when storing results. If None, it is determined automatically from the input size and number of samples. Ignored if X is a data loader, in which case the data loader is expected to handle batching internally.

  • output_dir (str | Path | None) – The directory where the outputs will be saved. If the specified directory does not exist, it will be created automatically. If None, a model-owned temporary directory is used. A subdirectory is generated within this directory to store the outputs; its path is recorded in the returned tree’s artifact_path attribute (on the root and the group node). The temporary directory is removed by cleanup(), cleanup_models(), or when the model is garbage-collected. Pass an explicit output_dir to keep results beyond the model’s lifetime.

  • progress (bool) – Whether to display a progress bar.

  • **kwargs (object) – Additional arguments passed to the model.

Returns:

Log-likelihood values. Posterior samples are included if available.

Raises:

  • TypeError – If shard_axis="draw" is used with a data loader X.

  • ValueError – If shard_axis is not "obs" or "draw".

  • NotImplementedError – If a return site’s axis-1 size does not match the input batch size (shard_axis="obs" only).

Return type:

xr.DataTree

See also

cleanup() to remove the temporary directory if created.