coco_pipe.io.utils

Helper functions for IO operations.

Attributes

logger
mne
BIDSPath
read_raw_bids

Functions

_get_mne()	Lazily import MNE so base IO structures stay lightweight.
_get_bids_path()	Lazily import mne-bids' BIDSPath.
_get_read_raw_bids()	Lazily import the MNE-BIDS raw reader.
row_quality_score(→ pandas.Series)	Calculate a 'badness' score for each row (NaNs + Infs + Zeros).
make_strata(→ pandas.Series)	Create a single stratification label from multiple covariates.
sample_indices(→ pandas.Index)	Sample indices for each class based on size_map.
split_column(→ Tuple[str, str])	Split a column into (unit, feature) using sep and reverse.
read_bids_entry(→ Tuple[numpy.ndarray, numpy.ndarray, ...)
load_participants_tsv(→ Dict[str, Dict[str, Any]])	Reads participants.tsv and returns dict: {sub_id: {col: val, ...}}.
detect_subjects(→ List[str])
detect_sessions(→ List[str])
detect_runs(→ List[str])	Detect available runs for a given subject/session/task.
smart_reader(→ Any)
default_id_extractor(→ str)

Module Contents

coco_pipe.io.utils.logger

coco_pipe.io.utils.mne = None

coco_pipe.io.utils.BIDSPath = None

coco_pipe.io.utils.read_raw_bids = None

coco_pipe.io.utils._get_mne()

Lazily import MNE so base IO structures stay lightweight.

coco_pipe.io.utils._get_bids_path()

Lazily import mne-bids’ BIDSPath.

coco_pipe.io.utils._get_read_raw_bids()

Lazily import the MNE-BIDS raw reader.

coco_pipe.io.utils.row_quality_score(df: pandas.DataFrame, exclude_cols: List[str] | None = None, count_zero: bool = True) → pandas.Series

Calculate a ‘badness’ score for each row (NaNs + Infs + Zeros). Lower is better.

coco_pipe.io.utils.make_strata(df: pandas.DataFrame, covariates: List[str], n_bins: int = 5, binning: str = 'quantile') → pandas.Series

Create a single stratification label from multiple covariates. Numeric covariates are binned.

coco_pipe.io.utils.sample_indices(df: pandas.DataFrame, target: str, size_map: Dict[Any, int], rng, replace: bool, prefer_clean: bool, exclude: List[str]) → pandas.Index

Sample indices for each class based on size_map.

coco_pipe.io.utils.split_column(name: str, sep: str, reverse: bool) → Tuple[str, str]

Split a column into (unit, feature) using sep and reverse.

coco_pipe.io.utils.read_bids_entry(bids_path: Any, is_pre_epoched: bool, is_evoked: bool, mode: str, window_length: float | None, stride: float | None, event_id: Dict[str, int] | str | List[str] | None = None, tmin: float = -0.2, tmax: float = 0.5, baseline: Tuple[float | None, float | None] | None = None) → Tuple[numpy.ndarray, numpy.ndarray, List[str], float, numpy.ndarray | None]

coco_pipe.io.utils.load_participants_tsv(root: pathlib.Path) → Dict[str, Dict[str, Any]]

Reads participants.tsv and returns dict: {sub_id: {col: val, …}}.

coco_pipe.io.utils.detect_subjects(root: pathlib.Path) → List[str]

coco_pipe.io.utils.detect_sessions(root: pathlib.Path, subject: str) → List[str]

coco_pipe.io.utils.detect_runs(root: pathlib.Path, subject: str, session: str | None = None, task: str | None = None, datatype: str = 'eeg') → List[str]

Detect available runs for a given subject/session/task.

coco_pipe.io.utils.smart_reader(path: pathlib.Path) → Any

coco_pipe.io.utils.default_id_extractor(path: pathlib.Path) → str