Index of values

Alias for cart.

Alias for interval.

A kind of reverse polish notation.

Affects the distribution to a cluster and returns the id of the latter.

Called only once.

To restore configurations after each perturbation.

See source code

bernouilli prng length density

2 independent bernouilli in parallel

binary len n is base 2 representation of n, truncated to length len.

let random_bool = bool prng gives a pseudo-random boolean generator of type unit -> bool faster than the one provided by stdlib.

create an array of neigbourhoods according to the finite-size, empty boundary, topology.

build_lc_index neighbourhoods depth cell builds the light cone of the given cell in terms of indexes.

Since this function is going to be heavily used, we optimize it by returning an int array instead of an int option array.

let cs = chi_square_table significance_level initialize a small table.

chi_squared distrib1 distrib2 A small value of the returned probability value indicates that the two distributions are likely to be different.

call let test = close significance_level to compute the table (which will be extended as needed).

Free energy for each cell, see the paper.

LLE for each cell.

Stores the $f$ configurations needed to compute the LLE.

Create an empty cluster collection.

Removes the duplicates from a list.

dichotomy f inf sup

dichotomy_extended middle succ pred f inf sup

Same as Experiment.Stat_cmplx.Experiment.discard_cell.

All future light cones must be isomorph, so we discard the cells to close to the border.

This function isn't used in the Periodic topology.

distance_to_boundary width height index

Does the measurement defined by the experiment and returns two output functions (printing and drawing).

Computes the next configuration in pof_configurations, shifts all configurations and thus loses the oldest configuration of past_configurations.

Called once per time step.

Same as Array.init, but begins the initialisation at the end of the array.

draw_configuration () color_of_state configuration time () needed to keep polymorphism (?!).

extract_bit nth i is true iff the nth bit of i is 1

extremum (>) list gives the max of the list.

See the paper.

get width height array x y

Probability for a cell to get a grain, at each step.

gray_gradient min max value is simply (max - value) / (max - min).

round height so that it's even

histogram nb_columns list

histogram_full interval_min interval_max nb_columns Assumes sorted input

Returns the unique identifier of lc_state.

Quantity of information, according to Shanon's definition.

init width height f

init_drawings simulation_time is guaranted to be called in the experiment before any call to draw_configuration.

initialize nb_cells, whatever the dimension.

int_of_bool false is 0 and int_of_bool true is 1.

iteri height f

lcState_of_lcIndex depth present_configuration pf_configurations lc converts the index representation of cells into a state representation.

Affects a unique integer identifier to each lc, for easier manipulation.

Computes the LLE for one cell.

mapi height f

Like Measure.Statistic.mean_float but discard the values (-1.) if User_interface.Text.Experiment_parameters.topology_boundary is Bounded.

Mean of the floats in the array, ignoring the nan ones.

merges distribution into cluster, assuming that Array.fold_left (+) 0 distribution = nb_samples_distribution

Like List.merge, but takes more than only two lists as input.

See source code

mix a b 0. is a.

mod_squared_distance modulus x y is (x - y)², considering x and y on a ring of length modulus.

nb_samples clusters.(i) is the nb of samples of the cluster with id i.

neighbourhoods is the neighbourhood of each cell, each neighbourhood being encoded as an array of indexes.

Returns the sphere (not ball) of radius 2 centered on each cell

next_step transition_rule configuration_in configuration_out doesn't modify configuration_in and writes the updated configuration in configuration_out.

next_step prng nb_updates transition_rule configuration_in configuration_out updates nb_updates cells one by one.

next_step prng update_rate transition_rule configuration_in configuration_out chooses the cells to update according to a bernouilli law of parameter update_rate and updates all at the same time.

next_step transition_rule configuration_in configuration_out updates all cells at the same time.

Computes the max possible complexity.

number_of states configuration gives the number of cells in configuration which are in one of the states states.

of_1d width height index

perturbed_next_step current_depth cells Computes the next step for the cells cells in the configuration configurations.(current_depth).

pof stands for Present Or Future

Parse the read parameters and pack them into modules

order parameter

remove_elements prng proba list remove some elements list, each with probability proba

restore_configurations light_cone copies the states of backup_configurations into configurations, for the cells given after perturbation_range +1 in light_cone.

The inverse function of Graphics.rgb.

create an array of neigbourhoods according to the ring (periodic) topology.

set width height array x y elt

specified_ints prng probabilities returns an array filled with random numbers between 0 (inclusive) and Array.length probabilities (exclusive), with probabilities specified by probabilities.

Removes the duplicates from a list.

Removes the duplicates from a list.

All the possible states

swap a i j modify the array a in place, swapping elements at indexes i and j.

to_1d width height x y

Used to compute the probability of each causal state.

Transition rule.

Transition rule.

transition_rule neighbourhoods Do some computation on the neigbourhoods if needed, then returns a function tr such that tr configuration index returns the new state of the cell number index in configuration.

Can be used only on a 2D periodic topology with no holes nor missing links, need Sphere2 Meta topology

Counts the area of the difference plume for one perturbation (see the paper).

uniform_ints prng length min max returns an array filled with random numbers between min and max (inclusive), with uniform probabilities.