tidy3d.web.api.container.BatchData

class BatchData

Bases: Tidy3dBaseModel, Mapping

Holds a collection of SimulationData returned by Batch.

Parameters:

• task_paths (dict[str, str]) – Mapping of task_name to path to corresponding data for each task in batch.

• task_ids (dict[str, str]) – Mapping of task_name to task_id for each task in batch.

• verbose (bool = True) – Whether to print info messages and progressbars.

• cached_tasks (Optional[dict[str, bool]] = None) – Whether the data of a task came from the cache.

• lazy (bool = False) – Whether to load the actual data (lazy=False) or return a proxy that loads the data when accessed (lazy=True).

• is_downloaded (Optional[bool] = False) – Whether the simulation data was downloaded before.

• task_tree (Optional[BatchTaskTree] = None) – Optional nested mapping from container positions to flat batch task names.

Notes

When the batch is completed, the output is not a SimulationData but rather a BatchData. For flat batches created from dict[str, simulation], this behaves like the historical flat task-name mapping. For nested batches, indexing and iteration follow the original top-level container shape while flat task access remains available through load_sim_data() and task_items(). Converting nested results with dict(batch_results.items()) eagerly touches all top-level entries and can load all results.

See also

Batch:

Interface for submitting several Simulation objects to sever.

SimulationData:

Stores data from a collection of Monitor objects in a Simulation.

Notebooks

• Running simulations through the cloud

• Performing parallel / batch processing of simulations

Attributes

task_paths
task_ids
verbose
cached_tasks
lazy
is_downloaded
task_tree

Methods

load([path_dir, replace_existing])	Load Batch from file, download results, and load them.
load_sim_data(task_name)	Load a simulation data object from file by task name.
task_items()	Iterate over flat (task_name, sim_data) pairs.
task_names()	Return an iterator over flat batch task names.
task_values()	Iterate over flat simulation data values.

task_paths

task_ids

verbose

cached_tasks

lazy

is_downloaded

task_tree

load_sim_data(task_name)

Load a simulation data object from file by task name.

When config.batch_data_cache.enabled is True and the total size of all task files stays under the configured threshold, the loaded object is cached in memory for subsequent accesses.

task_names()

Return an iterator over flat batch task names.

task_items()

Iterate over flat (task_name, sim_data) pairs.

task_values()

Iterate over flat simulation data values.

classmethod load(path_dir='.', replace_existing=False)

Load Batch from file, download results, and load them.

Parameters:

• path_dir (PathLike = './') – Base directory where data will be downloaded, by default current working directory. A batch.hdf5 file must be present in the directory.

• replace_existing (bool = False) – Downloads the data even if path exists (overwriting the existing).

Returns:

Returns a BatchData mapping. When initialized from a nested container, __getitem__ also supports the original top-level dict keys and sequence indices.

Return type:

BatchData