added working 'get_foraging_bouts' function #410
Conversation
Allow forwarding of load function kwargs to reader
Additional pre-commit and ruff fixes
* Persist venv across job steps * Update codecov-action version * Remove `env_config`
Clear all CI checks on all platforms
| ) | ||
| foraging_mask = max_windowed_wheel_vals > (patch_spun_df["cum_wheel_dist"] + min_wheel_movement) | ||
| # Discretize into foraging bouts | ||
| bout_start_indxs = np.where(np.diff(foraging_mask.astype(int), prepend=0) == 1)[0] |
There was a problem hiding this comment.
these should be shifted up by win_len
There was a problem hiding this comment.
bout_start_indxs = np.where(np.diff(foraging_mask.astype(int), prepend=0) == 1)[0] + win_len
There was a problem hiding this comment.
if the wheel starts moving after a long period of inactivity but for fewer than min_wheel_movement BUT a pellet is delivered as below, should this (first) pellet be counted in the foraging bout, i.e. should n=3 or n=2? If we shift bout_start_idxs by win_len-1, n=2.
There was a problem hiding this comment.
I think the foraging bout should be started when the wheel actually starts moving.
Let me look at this particular case in more detail later.
The bouts that the previous rolling max approach detects will differ from just shifting the cum_wheel_dist, and I think this may be one of them?
There was a problem hiding this comment.
Iirc, using rolling max and direct shift produce the same results in terms of the number of pellets. But feel free to verify this.
| foraging_mask = max_windowed_wheel_vals > (patch_spun_df["cum_wheel_dist"] + min_wheel_movement) | ||
| # Discretize into foraging bouts | ||
| bout_start_indxs = np.where(np.diff(foraging_mask.astype(int), prepend=0) == 1)[0] | ||
| bout_end_indxs = np.where(np.diff(foraging_mask.astype(int), prepend=0) == -1)[0] |
There was a problem hiding this comment.
Need to account for final pellet delivery in bout: time for bonsai to deliver pellet / beambreak to occur after threshold crossing
There was a problem hiding this comment.
bout_start_indxs shift is done in a96231c and left a TODO note for changing the constant for bout_end_indxs
…eshold update for the block
jkbhagatio
force-pushed
the
block_foraging_bout_detection
branch
from
7f5b51a
to
baa5336
Compare
No description provided.