Calibrators

class ollin.calibration.velocity.VelocityCalibrator(movement_model, config=None)

Class to calibrate velocity parameters.

This class also holds all data generated in calibration simulations.

config

dict – Dictionary holding all configuration settings. See config to see all relevant settings.

movement_model

MovementModel – Reference to Movement model instance being calibrated.

velocity_info

array –

Numpy array of size:

[num_velocities, num_niches, num_worlds, trials_per_world]

containing velocity information. Here:

Num_velocities:Refers to the number of simulated mean velocities held in the configuration array config['velocities']. Num_niches:Refers to the number of simulated niche sizes held in the configuration array config['niche_sizes']. Num_worlds:Refers to the number of sites created per selection of (velocity, niche_size). Trials_per_world:  Refers to the number of simulations made per site.

Hence if:

vel = self.velocity_info[i, j, k, l]

this means that vel was the mean velocity in the l-th simulation at the k-th world generated with niche size config['niche_sizes'][j] and “model”-mean velocity config['velocities'][i].

calculate_velocity_info()

Simulate multiple scenarios in parallel and record mean velocity.

fit()

Fit correction parameter to simulated velocity data.

Returns:fit – Dictionary holding the fitted parameters. Return type:dict

plot(cmap='Set2', figsize=(10, 10), ax=None, plotfit=True)

Plot graph of generated velocity data and fit.

Plots a graph of internal Model’s “mean-velocity” parameter versus resulting simulated mean velocity. Adds a fitted line to the plot if desired to visually check calibration.

Parameters:

• cmap (str, optional) – Name of colormap to use for diferent niche sizes.
• ax (matplotlib.axes.Axes, optional) – Ax in which to draw the plot. If not given a new one will be created.
• fisize (tuple, optional) – Size of figure to create. Used only if no ax is given.
• plotfit (bool, optional) – If True will plot a line fitted to velocity data. Defaults to True.

Returns:

ax – Ax object for further plotting.

Return type:

matplotlib.axes.Axes

class ollin.calibration.home_range.HomeRangeCalibrator(movement_model, config=None)

Class to calibrate Home Range parameters.

This class also holds all data generated in calibration simulations.

config

dict – Dictionary holding all configuration settings. See config to see all relevant settings.

movement_model

MovementModel – Reference to Movement model instance being calibrated.

home_range_info

array –

Numpy array of size:

[num_velocities, num_niches, num_worlds, trials]

containing occupancy information. Here:

Num_velocities:Refers to the number of simulated mean velocities held in the configuration array config['velocities']. Num_niches:Refers to the number of simulated niche sizes held in the configuration array config['niche_sizes']. Num_worlds:Refers to the number of sites created per selection of (velocity, niche_size). Trials:Refers to the number of simulations made per site.

Hence if:

hr = home_range_info[i, j, k, l]

this means that hr was the mean home range in the l-th simulation at the k-th world generated with niche size config['niche_sizes'][j] and mean velocity config['velocities'][i].

calculate_hr_info()

Simulate multiple scenarios in parallel and record mean home range.

fit()

Fit correction parameter to simulated home range data.

Returns:fit – Dictionary holding the fitted parameters. Return type:dict

plot(cmap='Set2', figsize=(10, 10), ax=None, plotfit=True)

Plot graph of generated home range data and fit.

Plots a graph of mean velocity versus resulting simulated home ranges. Adds a fitted line to the plot if desired to visually check calibration.

Parameters:

• cmap (str, optional) – Name of colormap to use for diferent niche sizes.
• ax (matplotlib.axes.Axes, optional) – Ax in which to draw the plot. If not given a new one will be created.
• fisize (tuple, optional) – Size of figure to create. Used only if no ax is given.
• plotfit (bool, optional) – If True will plot a line fitted to velocity data. Defaults to True.

Returns:

ax – Ax object for further plotting.

Return type:

matplotlib.axes.Axes

class ollin.calibration.occupancy.OccupancyCalibrator(movement_model, config=None)

Class to calibrate Occupancy parameters.

This class also holds all data generated in calibration simulations.

config

dict – Dictionary holding all configuration settings. See config to see all relevant settings.

movement_model

MovementModel – Reference to Movement model instance being calibrated.

occupancy_info

array –

Numpy array of size:

[num_home_ranges, num_niches, num_nums, num_worlds, trials]

containing occupancy information. Here:

Num_home_ranges:  Refers to the number of simulated home ranges held in the configuration array config['home_ranges']. Num_niches:Refers to the number of simulated niche sizes held in the configuration array config['niche_sizes']. Num_nums:Refers to the number of different number of individuals simulated which are held in the configuration array config['nums`]. Num_worlds:Refers to the number of sites created per selection of (home_range, niche_size). Trials:Refers to the number of simulations made per site.

Hence if:

oc = occupancy_info[i, j, k, l, m]

this means that oc was the occupancy generated by config['nums'][k] individuals, randomly selected from the the m-th simulation at the l-th world generated with niche size config['niche_sizes'][j] and home range config['home_ranges'][k].

calculate_oc_info()

Simulate multiple scenarios in parallel and record occupancy.

fit()

Fit model parameters to simulated occupancy data.

Returns:fit – Dictionary holding the fitted parameters. Return type:dict

plot(figsize=(10, 10), ax=None, x_var='density', w_target=True, xscale=None, yscale=None, lwidth=0.1, wtext=False)

Plot graph of generated occupancy data and fit.

Plots a grid of graphs of the relation between population density and resulting simulated occupancy. Adds a fitted line to the plot if desired to visually check calibration.

Parameters:

• ax (matplotlib.axes.Axes, optional) – Ax in which to draw the plot. If not given a new one will be created.
• fisize (tuple, optional) – Size of figure to create. Used only if no ax is given.
• x_var (str, optional) – Variable to use in the x-axis. Options are: ‘density’, ‘home_range’, ‘niche_sizes’. Defaults to ‘density’.
• w_target (bool, optional) – If True will plot a line fitted to velocity data. Defaults to True.
• xscale (str, optional) – Scale to use in the x-axis. Options are: ‘linear’, ‘log’. Defaults to ‘linear’.
• yscale (str, optional) – Scale to use in the y-axis. Options are: ‘linear’, ‘log’, ‘logit’. Defaults to ‘linear’.
• lwidth (str, optional) – Width of fitted line. Defaults to 0.1.
• wtext (bool, optional) – If True will add a text description to each plot, specifying the niche size and home_range.

Returns:

ax – Ax object for further plotting.

Return type:

matplotlib.axes.Axes