| name | xdf |
| description | XDF (.xdf) — Extensible Data Format from LSL recordings |
| layer | L0 |
| group | consumer_eeg |
| metadata | {"tags":["io","xdf","lsl","recording","multi_stream"],"formats":[".xdf"],"modalities":["eeg","meg","fnirs","spike"]} |
XDF — Extensible Data Format
Format Overview
XDF is the offline storage format for LSL-recorded streams. Stores
multiple parallel streams (EEG + Markers + Audio + ...) with per-stream
metadata + per-sample timestamps with LSL clock-correction.
Loader
Primary: pyxdf.load_xdf(path). EEG-only path: mne.io.read_raw_xdf
(if installed).
Common Pitfalls
- Stream selection. A single XDF can contain EEG + Markers + Audio +
...; pick the right stream by name.
- Clock correction.
pyxdf returns timestamps already clock-corrected
to a common LSL clock; check that the marker stream's timestamps align.
- Format zoo. XDF supports many channel formats; ensure dtype is
float-castable.
EasyBCI Path
- load:xdf,stream=EEG
References
- Kothe, C. A. (2014). Lab Streaming Layer (LSL). https://github.com/sccn/labstreaminglayer.
- pyxdf documentation. https://github.com/xdf-modules/pyxdf.
Boundary
lsl_stream: live-stream version; XDF is the recorded archive.