pyfli.io.data_saving

Save fitted maps, metadata, figures, and session artifacts to disk.

This module belongs to pyfli.io and is part of PyFLI detector importers, file readers, saving helpers, and processed-data loaders. Public API includes classes DataSaver; functions filter_vars().

Functions

filter_vars(local_vars, keys)

Run the filter vars routine.

Classes

DataSaver(path[, folder_name, new_session])

Persist PyFLI fitting outputs and analysis artifacts.

filter_vars(local_vars, keys)

Run the filter vars routine.

Parameters:

• local_vars (np.ndarray) – Local variable dictionary filtered before saving.

• keys (np.ndarray) – Dataset keys to include in the saved output.

Returns:

Object produced by filter vars.

Return type:

Any

class DataSaver(path, folder_name='_pyfli_Analysis', new_session=False)

Bases: object

Persist PyFLI fitting outputs and analysis artifacts. The class creates a session directory and writes arrays, metadata, figures, and tabular summaries in a consistent layout.

Parameters:

• path (str) – Filesystem path loaded or saved by the routine.

• folder_name (str) – Output folder where session files are written.

• new_session (bool) – If True, create a new timestamped session folder.

classmethod load(save_dir)

Convenience constructor for reconnecting to an existing session folder for data loading. Equivalent to:

DataSaver(path=save_dir, folder_name=None, new_session=False)

Parameters:

save_dir (str)

Return type:

DataSaver

classmethod resolve_path(path, folder_name)

Public helper: returns the save_dir that DataSaver(path, folder_name) would resolve to, without creating it. Use this to derive the load path for an auto-saved session before deciding whether to load from or write into it.

Parameters:

• path (str)

• folder_name (str)

Return type:

str

log(message)

Run the log routine.

Parameters:

message (Any) – Message text displayed to the user.

Returns:

No object is returned; the function perform log.

Return type:

None

log_to_file(message)

Append to the log file only, without echoing through logging/console.

Use for bulk/verbose content (e.g. a printed comparison table) that belongs in the session log but would be noisy to also emit via the logger on every line. For normal status messages, use log().

Parameters:

message (Any) – Message text appended to the log file.

Returns:

No object is returned; the message is written to self.log_file.

Return type:

None

save_plot(name, fig=None, dpi=300, close=True)

Save plot.

Parameters:

• name (str) – Dataset, experiment, figure, or output name.

• fig (Any | None) – Matplotlib figure object to update or save.

• dpi (int) – Resolution used when saving a figure.

• close (bool) – Whether to close the figure after saving.

Returns:

No object is returned; the function save plot.

Return type:

None

save_json(name, data_dict)

Saves settings/dictionaries as JSON.

Parameters:

• name (str)

• data_dict (ndarray)

Return type:

None

load_json(name)

Loads a dictionary previously saved with save_json.

Parameters:

name (str)

Return type:

Any

save_npy(name, array)

Save npy.

Parameters:

• name (str) – Dataset, experiment, figure, or output name.

• array (np.ndarray) – Array processed by the routine.

Returns:

No object is returned; the function save npy.

Return type:

None

load_npy(name, allow_pickle=True)

Loads an array/dict previously saved with save_npy.

Parameters:

• name (str)

• allow_pickle (bool)

Return type:

Any

save_params(**kwargs)

Save params.

Parameters:

**kwargs (Any) – Additional keyword options forwarded to the underlying implementation.

Returns:

No object is returned; the function save params.

Return type:

None

save_config(config_dict, name='fitting_config')

Save config.

Parameters:

• config_dict (np.ndarray) – Configuration dictionary written to disk.

• name (str) – Dataset, experiment, figure, or output name.

Returns:

No object is returned; the function save config.

Return type:

None

load_config(name='fitting_config')

Loads a config dict previously saved with save_config.

Parameters:

name (str)

Return type:

Any