Skip to content

Conversation

@jkim0731
Copy link
Collaborator

Getting ophys timestamps from utilities assumed 4 FOVs (8 planes).
For pilot data, this may not be true (can have any number of FOVs).

Added a function to get the number of FOVs,
and applied this number for getting ophys frames timestamps.

@jkim0731
Copy link
Collaborator Author

Seems like there is a lot of conflicts to resolve with the main branch - should we just use this branch for learning mFISH data?

@matchings
Copy link
Collaborator

@jkim0731 yes we need to stay on this branch for learning mFISH. The main branch is still configured for Visual Behavior project dataset, and needs to stay like that since we are still working on the platform paper. it is unfortunate to have two divergent branches, but it is what it is. We can clean it up and make it work for either project once the platform paper is done, but i dont want to change anything on main until then.

thanks for making this function more flexible though!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants