Skip to content

Latest commit

 

History

History
143 lines (79 loc) · 4.6 KB

list_of_plotting_fuctions.md

File metadata and controls

143 lines (79 loc) · 4.6 KB
Raw

List of RatInABox plotting functions:

In this markdown we describe teh plotting functions available with RatInABox. The following definitions hold:

  • Env: a 2D Environment() class with a wall and solid boundary conditions
  • Env1D: a 1D Environment() class with periodic boundary conditions

Environment

Environment.plot_environment()

Displays the environment. Works for both 1 or 2D environments. Examples:

  • Env.plot_environment()

Agent

Agent.plot_trajectory()

Plots the agent trajectory. Works for 1 or 2D.

  • Ag.plot_trajectory(t_end=120)

  • Ag1D.plot_trajectory(t_end=120)

Agent.animate_trajectory()

Makes an animation of the agents trajectory.

Agent.plot_position_heatmap()

Plots a heatmap of the Agents past locations (2D and 1D example shown)

Agent.plot_histogram_of_speeds()

Agent.plot_histogram_of_rotational_velocities()

Neurons

Neurons.plot_rate_timeseries()

Plots a timeseries of the firing rates

Neurons.plot_rate_timeseries(imshow=True)

Plots a timeseries of the firing rates as an image

Plots a timeseries of the firing rates

Neurons.animate_rate_timeseries()

Makes an animation of the firing rates timeseries

Neurons.plot_ratemap()

Depending on the parameters passed this function will either

  1. Analytically calculate the rate map (method = 'analytic'),
  2. Infer the rate map from past activity (method = 'history') or
  3. Plot the observed spikes (spikes=True). As an example here we show this function for a set of 3 (two dimensional) grid cells and 10 (one-dimensional) place cells.
  • Neurons.plot_ratemap(method=analytic`)

  • Neurons.plot_ratemap(method=history`)

  • Neurons.plot_ratemap(method=neither`, spikes=True)

PlaceCells.plot_place_cell_centres()

Scatters where the place cells are centres

BoundaryVectorCells.plot_BVC_receptive_field()

Other details:

  • All plotting functions return a tuple (fig, ax) of matplotlib figure objects.
  • Most plotting functions support being passed fig and ax and will plot whatever it is they plot ontop of that. This means we can make:
  1. Layered figures (e.g. plot a trajectory on top of a rate map):
fig, ax = Neurons.plot_rate_map(chosen_neuron="1")
fig, ax = Ag.plot_trajectory(fig=fig, ax=ax)

  1. Multipanel figures:
fig, axes = plt.subplots(1,5,figsize=(20,4))
Ag.plot_trajectory(fig=fig,ax=axes[0])
Neurons.plot_rate_map(fig=fig,ax=[axes[1],axes[2],axes[3]],chosen_neurons='3') #<-- to plot more than 1 neuron pass an array of axes
Neurons.plot_rate_timeseries(fig=fig,ax=axes[4])

  • For rate maps and timeseries' by default all the cells will be plotted. This may take a long time if the number of cells is large. Control this with the chosen_neurons argument

  • We use the matplotlib wrapper tomplotlib to format and save figures. This is not necesary for the workings of ratinabox but note your figures might look slightly different without it.