neuralib.widefield.preproc.PreprocessOptions

class neuralib.widefield.preproc.PreprocessOptions

Bases: AbstractParser

__init__()

Initialize internal state variables.

Methods

__init__()	Initialize internal state variables.
calculate_dff()	Calculate ΔF/F with time-varying baseline using rolling window
compute_motion_transforms()	Compute motion correction transforms using the pre-computed reference frame.
load()	load TIF files and validate dataset consistency
main([args, parse_only, system_exit])	parsing the commandline input args and call run().
new_parser(**kwargs)	create an ArgumentParser.
run()	Main entry point - orchestrates entire pipeline.
save_metadata()

Attributes

DESCRIPTION	parser description.
EPILOG	parser epilog.
GROUP_ACCEL
GROUP_IO
GROUP_PROC
USAGE	parser usage.
chunk_size	number of frames per chunk for processing (default: 3000)
directory	directory for input files (default: None)
file	single input file (default: None)
force_compute	force recomputation even if output files exist (overwrite mode) (default: False)
max_shift	maximum allowed shift in pixels for motion correction (default: 20)
motion_correction	do the motion correction (default: False)
n_jobs	number of parallel jobs for processing (-1 = use all CPUs) (default: -1)
output_dir	output directory
percentile	percentile for baseline calculation (default: 10)
rotate	rotate the all sequences in degree (default: None)
save_f0	save F0 baseline to disk (can be disabled to save storage space) (default: False)
suffix_pattern	suffix for directory (default: '.tif')
use_gpu	use GPU acceleration with CuPy (requires NVIDIA GPU and cupy package) (default: False)
window_size	window size for rolling baseline (frames) (default: 100)

DESCRIPTION: str | None = 'Preprocessing pipeline for widefield calcium imaging dataset'

parser description. Could be override as a method if its content is dynamic-generated.

GROUP_IO = 'Data I/O Options'

GROUP_PROC = 'Processing Options'

GROUP_ACCEL = 'Acceleration Options'

file: Path | None

single input file (default: None)

directory: Path | None

directory for input files (default: None)

suffix_pattern: str

suffix for directory (default: ‘.tif’)

motion_correction: bool

do the motion correction (default: False)

max_shift: int

maximum allowed shift in pixels for motion correction (default: 20)

rotate: float | None

rotate the all sequences in degree (default: None)

chunk_size: int

number of frames per chunk for processing (default: 3000)

window_size: int

window size for rolling baseline (frames) (default: 100)

percentile: int

percentile for baseline calculation (default: 10)

n_jobs: int

number of parallel jobs for processing (-1 = use all CPUs) (default: -1)

force_compute: bool

force recomputation even if output files exist (overwrite mode) (default: False)

save_f0: bool

save F0 baseline to disk (can be disabled to save storage space) (default: False)

use_gpu: bool

use GPU acceleration with CuPy (requires NVIDIA GPU and cupy package) (default: False)

__init__()

Initialize internal state variables.

run()

Main entry point - orchestrates entire pipeline.

property output_dir: Path

output directory

load()

load TIF files and validate dataset consistency

compute_motion_transforms()

Compute motion correction transforms using the pre-computed reference frame. The reference frame should already be computed and stored in self._reference_frame.

calculate_dff()

Calculate ΔF/F with time-varying baseline using rolling window

save_metadata()