Configuration#

EEGPrep keeps global options intentionally small. Use explicit function arguments for preprocessing choices, and use EEG_OPTIONS only for session, storage, compatibility, and GUI behavior that really is global.

EEG_OPTIONS is a dictionary, not a configuration object:

from eegprep import EEG_OPTIONS

print(dict(EEG_OPTIONS))
EEG_OPTIONS["option_allmenus"] = 1
EEG_OPTIONS["option_savetwofiles"] = 1

Save the previous values when a script changes global options:

from eegprep import EEG_OPTIONS

old_options = dict(EEG_OPTIONS)
try:
    EEG_OPTIONS["option_memmapdata"] = 1
    # Load or save datasets here.
finally:
    EEG_OPTIONS.clear()
    EEG_OPTIONS.update(old_options)

Common Global Options#

Key	Effect
`option_allmenus`	Show advanced and legacy menu entries in the main GUI.
`option_storedisk`	Keep saved non-current `ALLEEG` datasets on disk to reduce memory use.
`option_savetwofiles`	Make `pop_saveset` write a `.set` header plus `.fdt` sidecar by default.
`option_memmapdata`	Memory-map two-file `.fdt` data when loading with `pop_loadset`.
`option_boundary99`	Treat numeric `-99` events as boundary events for ERPLAB-style data.
`option_computeica`	Precompute ICA activations where supported.
`option_scaleicarms`	Scale ICA component activations to RMS microvolt during checkset paths.
`option_cachesize`	STUDY cache size in MB for implemented STUDY measure workflows.

For storage details, see Large-Dataset Storage.

Set Options From GUI or Console#

Use the Options menu in the GUI, or call pop_editoptions from eegprep-console:

from eegprep import EEG_OPTIONS, pop_editoptions

com = pop_editoptions(option_allmenus=1, option_savetwofiles=1)
print(com)
print(EEG_OPTIONS["option_allmenus"])

In a shared GUI/console session, the menu refresh uses the same EEG_OPTIONS dictionary.

Preprocessing Parameters#

Most analysis choices are ordinary function arguments. This keeps scripts readable and prevents hidden global state from changing scientific results.

from pathlib import Path

from eegprep import pop_clean_rawdata, pop_eegfiltnew, pop_loadset, pop_resample

EEG = pop_loadset(Path("sample_data") / "eeglab_data.set")

EEG, filter_com = pop_eegfiltnew(
    EEG,
    locutoff=1,
    hicutoff=40,
    plotfreqz=False,
    return_com=True,
)
EEG, resample_com = pop_resample(EEG, 64, return_com=True)
EEG, clean_com = pop_clean_rawdata(
    EEG,
    FlatlineCriterion=5,
    ChannelCriterion=0.8,
    LineNoiseCriterion=4,
    Highpass=(0.25, 0.75),
    BurstCriterion=20,
    WindowCriterion=0.25,
    return_com=True,
)

history = [filter_com, resample_com, clean_com]

The same rule applies to ICA, ICLabel, and BIDS preprocessing:

from eegprep import bids_preproc, pop_iclabel, pop_runica

EEG, ica_com = pop_runica(EEG, icatype="picard", gui=False, return_com=True)
EEG, label_com = pop_iclabel(EEG, "default", return_com=True)

bids_preproc(
    "/path/to/bids-root",
    OutputDir="/path/to/bids-root/derivatives/eegprep",
    Subjects=["01"],
    Tasks=["rest"],
    WithInterp=True,
    WithICA=False,
    NumJobs=1,
)

Reusable Presets#

Represent lab presets as plain dictionaries and pass them explicitly:

CLEANING_PRESETS = {
    "resting": {
        "FlatlineCriterion": 5,
        "ChannelCriterion": 0.8,
        "LineNoiseCriterion": 4,
        "Highpass": (0.25, 0.75),
        "BurstCriterion": 20,
        "WindowCriterion": 0.25,
    },
    "erp": {
        "FlatlineCriterion": 5,
        "ChannelCriterion": 0.85,
        "LineNoiseCriterion": 4,
        "Highpass": (0.1, 0.5),
        "BurstCriterion": 20,
        "WindowCriterion": 0.25,
    },
}

EEG, com = pop_clean_rawdata(EEG, **CLEANING_PRESETS["resting"], return_com=True)

Keep these preset dictionaries in your lab package, notebook, or analysis repo so parameter changes are version-controlled with the study.

When to Use Which Surface#

Need	Use
Change menu visibility or storage behavior	`EEG_OPTIONS` or `pop_editoptions`.
Change filtering, cleaning, resampling, ICA, ICLabel, or BIDS behavior	Explicit function arguments.
Record replayable commands	`return_com=True` and keep the returned command strings.
Share settings across scripts	A normal Python dictionary or module constant in your analysis code.
Use the same state model as the GUI	`EEGPrepSession`; see GUI and Console Together.

Troubleshooting#

Option Did Not Affect a Processing Step#

Check whether the function actually reads EEG_OPTIONS. Most processing functions are configured through explicit keyword arguments.

Memory-Mapped Loading Did Not Happen#

option_memmapdata applies to two-file .set/.fdt datasets. Save with savemode="twofiles" or set option_savetwofiles before saving, then load again with option_memmapdata enabled.