Smoothing trajectories: Smoothing uses a 1D Savitzky-Golay filter. Specify smoothing as a parameter when extracting trajectories in a session or directly smooth a single trial from the pd.Dataframe of session extracted trajectories Smoothing functions can be found in utils.smooth_trajectory

Extracting head angles: Extract head angle from sleap pose data as the angle of the vector from neck to nose relative to the horizontal (going CCW). Functions can be found in head_angle.py, with functions to extract head angle at specified frames in each trial (e.g. stim start frames or trial end frames),or just a single specified frame index.
Functions take track number as an input, so must run separately on each track in social sessions.

logistic_regression.py: makes use of head_angle.py function head_angle_to_wall to find the head angles to both of the walls, record them in om, and compare them. This ONLY works for social sessions, as it assumes track 0 in the sleap trajectory data# SLEAP_project