Things to change for next iterations

[niksirbi]

Summary of feedback from 2023-11-29

Participants found the course useful and liked the format of mixing theory bits with practical exercises.
They asked for more time allocated to the theory behind extracting behaviours.

Ideas for the next course iteration in 2024

• Expand the theory section on different approaches for classifying behaviours/action segmentation
• Add one more notebook with practical exercises on the above topic, with keypoint-moseq or some similar tool.
• Update the existing notebook with the latest movement features (such as they are by the time of the course)

@lochhh feel free to comment with any takeaways or ideas you might have about this. I'll leave this issue open as a reminder to our future selves.

[lochhh]

Two small things:

• Fix typo in image
• Add practical exercise solutions to the notebook (e.g., as collapsed cells for future courses, to help people who are less familiar with Python stay up to speed during the course) - I think @niksirbi you have these on a local branch during the course, would be nice to have this as a separate EPM_analysis_solutions.ipynb for now.

[niksirbi]

Actually, I had already deleted that local branch before I saw your comments, but re-implementing the solutions should be fast.

[lochhh]

• Update schedule 10am-1pm; 2pm-5pm
• Reset menti responses
• Simplify neural activity and behaviour slide, update what how why explanation (Marr’s three levels of analysis)
• Reduce Krakauer quotes
• Multi-animal identity tracking (apperance typo)
• Update current notebook with newer movement features
• Add keypoint-moseq notebook

[niksirbi]

Also:

• possibly update the final slides on behavioural classification.
• Review SLEAP installation and screenshots to see if they are valid for newest SLEAP version