Gradient model simulations

For each simulation type there is one .fishexp pickle file which contains per-frame position data across the simulations and one _bouts.pkl file which contains swim bout information for the simulation. Each file contains a pickled (via Python pickle module) dictionary. Keys within this dictionary identify the gradient type. For each gradient type there is a sub-dictionary with keys for each individual simulation that was run (200 each).The different names are for the following content:Experiment_fish_model_gradient: Simulatins using swim kineamtics binned according to tempeature and previous bout temperautre change from gradient experimentsFish_model_0_order: Constant transition model Navigation model simulationsFish_model_transition: Stimulus driven transition model Navigation model simulationFish_model_fullsim_transition: Same as Stimulus model however contains a full range gradient simulationFor each individual simulation the .fishexp file contains a matrix with n-simulation frames rows (at simulated 100 Hz same as in actual experiments) and 2 columns, the x-coordinate in pixels and the y-coordinate in pixels (resolution is 7 px/mm as in the experiments).For each simulation the _bouts.pkl file contains a dataframe with the following structure:StateOriginal indexStartStopDisplacementAngle changeIBIThe annotated swim modeThe original index of the bout (since bouts close to the arena edge were filtered out)The start simulation frame of the boutThe end simulationframe of the boutThe displacement of the bout in pixelsThe angle change of the bout in radiansThe waiting time since the previous bout (interbout interval) in framesTemperaturePrev Delta TDelta XDelta YX PositionY PositionHeadingThe temperature at the start of the boutThe temperature change across the previous boutThe movement in the X direction in pixelsThe movement in the Y direction in pixelsThe x position at the start of the boutThe y position at the start of the boutThe fish heading at the start of the bout