Cellular automata¶

cromosim.ca.compute_exit_times(dt, people, results)[source]¶

Compute exit times for all the individuals from the results array

Parameters:	dt: float time step people: numpy masked arrays equal to 1 if the cell (i,j) is occupied, 0 elsewhere results: numpy array positions at each time
Returns:	exit_times : all the exit times

cromosim.ca.exit(domain, people, people_ij)[source]¶

To update people and people_ij arrays in removing individuals who have leaved the domain

Parameters:	domain: Domain contains everything for managing the domain people: numpy masked arrays equal to 1 if the cell (i,j) is occupied, 0 elsewhere people_ij: numpy array (i,j) for each individual
Returns:	people: numpy masked arrays new positions equal to 1 if the cell (i,j) is occupied, 0 elsewhere people_id: numpy masked arrays new people index (i,j)

cromosim.ca.parallel_update(people, people_ij, weight, friction=0, randomstate=None)[source]¶

To move all individuals in parallel according to the following rule:  first, desired moves are precomputed and then the conflicts (two individuals at the same position) are resolved

Parameters:

people: numpy masked arrays equal to 1 if the cell (i,j) is occupied, 0 elsewhere people_ij: numpy array (i,j) for each individual weight: numpy array weight for the probabilities in order to move in such a way like to reach the door friction: float to designate the effect induced by a modified handling of conflicts, friction is the probability that a conflict remains unresolved (no one moves) randomstate: numpy randomstate create a new one or reuse the given random state

Returns:

people: numpy masked arrays new positions equal to 1 if the cell (i,j) is occupied, 0 elsewhere people_id: numpy masked arrays new people index (i,j)

cromosim.ca.plot_people_according_to_current_door_distance(ifig, people, domain, axis=None, savefig=False, filename='fig.png')[source]¶

To draw occupied cells with colors depend to current door distances

Parameters:	ifig: int figure number people: numpy masked arrays equal to 1 if the cell (i,j) is occupied, 0 elsewhere domain: Domain contains everything for managing the domain axis: list matplotlib axis savefig: boolean writes the figure as a png file if true filename: string png filename used to write the figure

cromosim.ca.plot_people_according_to_exit_times(ifig, dt, people, domain, results, axis=None, savefig=False, filename='fig.png')[source]¶

To draw occupied cells with colors depend to the exit times

Parameters:	ifig: int figure number dt: float time step people: numpy masked arrays equal to 1 if the cell (i,j) is occupied, 0 elsewhere domain: Domain contains everything for managing the domain results: numpy array positions at each time axis: list matplotlib axis savefig: boolean writes the figure as a png file if true filename: string png filename used to write the figure

cromosim.ca.plot_people_according_to_initial_door_distance(ifig, people, domain, results, axis=None, savefig=False, filename='fig.png')[source]¶

To draw occupied cells with colors depend to initial (time=0) door distances

Parameters:	ifig: int figure number people: numpy masked arrays equal to 1 if the cell (i,j) is occupied, 0 elsewhere domain: Domain contains everything for managing the domain axis: list matplotlib axis savefig: boolean writes the figure as a png file if true filename: string png filename used to write the figure

cromosim.ca.plot_people_paths(ifig, dt, pixel_size, people, domain, results, axis=None, savefig=False, filename='fig.png')[source]¶

To draw all the individual paths from intial time to final time

Parameters:

ifig: int figure number dt: float time step pixel_size: float size of one pixel in meters people: numpy masked arrays equal to 1 if the cell (i,j) is occupied, 0 elsewhere domain: Domain contains everything for managing the domain results: numpy array positions at each time axis: list matplotlib axis savefig: boolean writes the figure as a png file if true filename: string png filename used to write the figure

cromosim.ca.sequential_update(people, people_ij, weight, shuffle=None, randomstate=None)[source]¶

To move all individuals sequentially according to the following rule:  the update of individual i is determined in a stochastic way by computing transition probabilities on neighboring cells (including the current position of i).

Parameters:

people: numpy masked arrays equal to 1 if the cell (i,j) is occupied, 0 elsewhere people_ij: numpy array (i,j) for each individual weight: numpy array weight for the probabilities in order to move in such a way like to reach the door shuffle: string shuffle kind (‘random’ or ‘random_frozen’): if the sequential order change or not at each time randomstate: numpy randomstate create a new one or reuse the given random state

Returns:

people: numpy masked arrays new positions equal to 1 if the cell (i,j) is occupied, 0 elsewhere people_id: numpy masked arrays new people index (i,j)