added ssvep to cvep method #232
Conversation
AidaFakhry
force-pushed
the
aida_test_branch
branch
from
8829ae6
to
01f5fac
Compare
There was a problem hiding this comment.
I don't understand what the returned list represents in the ssvep_to_cvep function and how the calling function would use the results. I would expect that for a given refresh_rate it would return the same number of items, but
ssvep_to_cvep(refresh_rate=50, flicker_rate=10)
=> [0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0]
ssvep_to_cvep(refresh_rate=50, flicker_rate=25)
=> [0, 0, 1, 1]
If the intent is to communicate a flash pattern, we need to know how much time each entry in the list represents, so the units are important to consider. For instance, if we are returning a pattern for 1 second display, the items in the list could each be 1hz worth of data (50 items for a 50hz refresh rate). Alternatively, this could be a tuple with the on/off duration in seconds . For the second example, this would be (0.5, 0.5) for half second off, followed by half second on.
@sci-tab, how would it be easiest to consume this data when writing displays?
bcipy/helpers/stimuli.py
Outdated
| @@ -679,3 679,25 @@ def get_fixation(is_txt: bool) -> str: | |||
| return ' ' | |||
| else: | |||
| return DEFAULT_FIXATION_PATH | |||
|
|
|||
|
|
|||
| def ssvep_to_cvep(refresh_rate: int = 60, flicker_rate: float = 10.0): | |||
There was a problem hiding this comment.
We need some documentation for this function to give a summary of what it's for (particularly if someone doesn't know what SSVEP and CVEP are abbreviations for), it's parameters (units?), etc.
Also, there should be a return type (List[float]), and the documentation should explain what this represents.
bcipy/helpers/stimuli.py
Outdated
|
|
||
| length_flicker = refresh_rate / flicker_rate | ||
|
|
||
| # determine if lenth_flickerr is an integer using .is_integer() |
There was a problem hiding this comment.
There's a typo in the comment: lenth_flickerr -> length_flicker. But you can probably just remove this comment since it's redundant with the code.
bcipy/helpers/stimuli.py
Outdated
|
|
||
| # determine if lenth_flickerr is an integer using .is_integer() | ||
| if not length_flicker.is_integer(): | ||
| raise Exception("Provided flicker rate must divide into refresh rate.") |
There was a problem hiding this comment.
length_flicker is only used as a float to check the whether it's evenly divisible so we might want to immediately cast it as an int:
if length_flicker.is_integer():
length_flicker = int(length_flicker)
else:
raise Exception("Provided flicker rate must divide into refresh rate.")
bcipy/helpers/stimuli.py
Outdated
| t = 0 | ||
| else: | ||
| t = 1 | ||
| return test_array |
There was a problem hiding this comment.
Python has a couple of list methods that may be helpful here. The operator can be used to concatenate lists ([1,2] [3,4] => [1,2,3,4]), and the * operator is a shortcut for constructing a list of consecutive elements with the same value ([0] * 3 => [0,0,0]).
Another observation is that we're basically just toggling between 0 (for even indices) and 1 (for odd indices). We can determine if a number is even or odd by using the modulus operator %.
So with all that, one way to shorten the above code is something like:
test_array = []
for i in range(length_flicker):
test_array = [i % 2] * length_flicker
That is the intended output and how the VEP display would be able to use it. I'll pick it up from here and see if I can spot the issue! |
tab-cmd
force-pushed
the
aida_test_branch
branch
2 times, most recently
from
3246299
to
71f91f7
Compare
linting and test fix error fixing error fixing
tab-cmd
force-pushed
the
aida_test_branch
branch
from
71f91f7
to
e4daec2
Compare
Co-authored-by: Aida Fakhry <[email protected]> Co-authored-by: Tab Memmott <[email protected]>
* add timer to prevent double clicking (#218) * #181963356 ; added better alerting to inform users when fake data mode is on * Added changelog entry * Refactoring to reduce duplication * #182040775 ; option to summarize session at the end of a copy phrase task * Update README.md High level GPT2 documentation added, word level prediction described, contact information added * Matrix Display Fixes (#221) * fixes bug with last stimuli lagging in matrix calibration, added entire grid flash before each inquiry * Address PR comment cleanup Co-authored-by: Tab Memmott <[email protected]> * Refactored session helper to generate spreadsheets without first creating a sqlite database * ERP plotting improvements and MNE data support (#220) * #182005453 ; added parameter for signal model path; updated initialization code to use this for typing tasks * #182406904 ; Fix bug when generating a spreadsheet from active session data * [Bug fix] Two presses needed to start copy phrase (#225) * Documentation and formatting * #182200897 ; check for internet connection and confirm before proceeding with the session * Added logging for GPU information, if available * Whitespace fix * Added changes * Move session validation to validate module; update internet connectivity test to work with VPN * #182106570 ; added functionality for configuration channel specifications for a given device. * Added a test mock to prevent a GUI interaction during testing * Fix test issue by switching to a different acquisition client for mocking * use a better default timeout (#228) * #182776686 ; prevent widgets scrolling on parameters form * Jitter (#233) Jitter for calibration parameter. copy phrase ip cleanup stimuli tests * User ID reset bug and cleanup (#235) User id reset fix * Add support for Tobii Nano (#234) * Add support for Tobii Nano * address PR feedback * update changelog * added ssvep to cvep method (#232) Co-authored-by: Aida Fakhry <[email protected]> Co-authored-by: Tab Memmott <[email protected]> * #183037159 ; alert when using battery power * Better documentation on return values and potential impacts * VEP Display (#237) * upgrade psychopy (#239) * ERP viz demo fix (#240) * #175193924 infrastructure for changed parameters ; #183134229 Fixed issue with params_form rounding some float values ; #183133776 allow static offset to be input with more precision * Remove unneeded parameters (#241) * signal model loaded log (#244) better signal model logging * Devices config update && BciPy running outside of root (#242) * Refactor param changes GUI to use a scroll area for displaying a large number of changes * Refinements; parameters changes start out collapsed if empty; minor refactorings * #182408060 ; Ensure that spelled text is provided to the language model when using fake data * Export the device used as a useable config file (#247) * Convert: add BDF support and fix static offset bug (#246) * #182031676 ; remove TCP-based acquisition code * Added PR number to changelog * Added function type information; better string representations of channel specs and device specs; removed unused code in tests * Updated max_buffer_len parameter name for lsl_client * fix rebase issue in parameters module * Bug fixes: jitter, offline analysis, task complete alerting (#250) * Matrix Bug: fix for scp flashing (#252) * Matrix documentation (#253) Co-authored-by: lawhead <[email protected]> Co-authored-by: Shijia Liu <37078018 [email protected]> Co-authored-by: jcgangemi1 <87331189 [email protected]> Co-authored-by: AidaFakhry <92689243 [email protected]> Co-authored-by: Aida Fakhry <[email protected]> Co-authored-by: Basak Celik <[email protected]>
Overview
A method that uses the refresh rate on the monitor and returns a code for SSVEP. every frame of / on for x seconds.
Ticket
Ticket 2
Contributions
Test
Documentation
Are documentation updates required? In-line, README, or documentation? Verify the updates you did here.
Changelog
Is the CHANGELOG.md updated with your detailed changes? N/a