Skip to content

Saving

Module: saving.py#

The saving.py module provides functionality for saving processed data and metadata to disk. It includes methods to save image frames as TIFF files and metadata as JSON files.

Dependencies#

  • json: For handling JSON serialization and deserialization.
  • os: For interacting with the operating system, such as file and directory operations.
  • numpy: For handling numerical data, particularly image frames.
  • tifffile.TiffWriter: For writing image data to TIFF files.
  • OutputParameters: A model from the configio.configmodels module, representing output parameters.
  • MetaData: A model from the metadata.metadata module, representing metadata associated with the data.

Functions#

save_config_frames#

def save_config_frames(
    config: MetaData, frames: np.ndarray, outputparams: OutputParameters
) -> None:

Saves the provided image frames and associated metadata to disk.

Parameters#
  • config (MetaData):

  • Metadata associated with the image frames. This includes information such as axes definitions and other relevant metadata.

  • frames (np.ndarray):

  • A NumPy array containing the image frames to be saved.

  • outputparams (OutputParameters):

  • An instance of OutputParameters containing the output path and file name.
Returns#
  • None
Behavior#
  1. Directory Creation:

  2. If the directory specified in outputparams.output_path does not exist, it will be created.

  3. TIFF File Writing:

  4. The image frames are saved as a TIFF file using the TiffWriter class from the tifffile library. The file is saved with the name specified in outputparams.output_name and the .ome.tiff extension. The bigtiff=True parameter ensures that the file can handle large datasets.

  5. The metadata from the config object, excluding any notes, is embedded in the TIFF file.

  6. JSON Metadata Saving:

  7. The complete metadata from the config object is serialized to a JSON file named metadata.json in the same directory as the TIFF file.
Example#
from ..configio.configmodels import OutputParameters
from ..metadata.metadata import MetaData
import numpy as np

# Example usage
config = MetaData(axes="TYX", notes="Example metadata")
frames = np.random.rand(10, 256, 256)  # Example frames
outputparams = OutputParameters(output_path="./output", output_name="example")

save_config_frames(config, frames, outputparams)

This will save the frames as example.ome.tiff in the ./output directory and the metadata as metadata.json in the same directory.

Notes#

  • The function assumes that the MetaData and OutputParameters models have appropriate methods (model_dump) for serializing their data.
  • The TiffWriter class is used for writing TIFF files, and it is assumed that the tifffile library is installed and properly configured.
  • The function handles large datasets by using the bigtiff=True parameter, which is necessary for datasets that exceed the 4GB limit of standard TIFF files.