DOC: time of events and first_samp

[pdeman]

hi,

how can i plot the events on top of the eeg versus time plot ?
i found the events/versus sample using find events. but when i divide versus the sampling rate i don't have at all the same values in time than the eeg. there must be an offset.
I'ld like to plot some eeg data with the "events" on top of it.

regards

[agramfort]

raw.plot(events=events) but don't resample raw after finding the evens otherwise they won't match.

[jona-sassenhagen]

but don't resample raw after finding the evens otherwise they won't match.

To be precise, we recommend

raw, events = raw.resample(new_sfreq, events=events)

[pdeman]

Thanks but in this particular case I don't plot using mne.
I would need to find the time for raws recording and time for events in the same unit.

[agramfort]

if you have events you can time in seconds with events_times = (events[:, 0] - raw.first_samp) / raw.info['sfreq'] time of 0s will be beginning of the file. HTH

[pdeman]

ok thanks

[larsoner]

At some point we should add a tutorial that discusses:

1. find_events, with event masking
2. events column values
3. raw.first_samp, what it is, how it changes with .crop, how to get event times relative to the start of the instance
4. Annotations and orig_time

I'll re-open and re-title this issue so we do this at some point.

[larsoner]

@massich you have been working on these things quite a bit recently, could you make a plot_timing.py tutorial or something similar to cover these topics in the next two weeks to make it into 0.17?

[massich]

yes

[larsoner]

Moving to 0.18, not a blocker