cma.optimization_tools.Sections

class documentation

class Sections(object):

plot sections through an objective function.

A first rational thing to do, when facing an (expensive) application. By default 6 points in each coordinate are evaluated. This class is still experimental.

Examples

import cma, numpy as np
s = cma.Sections(cma.ff.rosen, np.zeros(3)).do(plot=False)
s.do(plot=False)  # evaluate the same points again, i.e. check for noise
try:
    s.plot()
except:
    print('plotting failed: matplotlib.pyplot package missing?')

Details

Data are saved after each function call during do. The filename is attribute name and by default str(func), see __init__.

A random (orthogonal) basis can be generated with cma.Rotation()(np.eye(3)).

CAVEAT: The default name is unique in the function name, but it should be unique in all parameters of __init__ but plot_cmd and load. If, for example, a different basis is chosen, either the name must be changed or the .pkl file containing the previous data must first be renamed or deleted.

s.res is a dictionary with an entry for each "coordinate" i and with an entry 'x', the middle point. Each entry i is again a dictionary with keys being different dx values and the value being a sequence of f-values. For example s.res[2][0.1] == [0.01, 0.01], which is generated using the difference vector s .basis[2] like

s.res[2][dx] += func(s.res['x'] + dx * s.basis[2]).

See Also
`__init__`

Method	`__init__`	objective function
Method	`do`	generates, plots and saves function values `func(y)`, where `y` is 'close' to `x` (see `__init__()`). The data are stored in the `res` attribute and the class instance is saved in a file with (the weired) name `str(func)`...
Method	`flattened`	return flattened data `(x, f)` such that for the sweep through coordinate `i` we have for data point `j` that `f[i][j] == func(x[i][j])`
Method	`load`	load from file
Method	`plot`	plot the data we have, return `self`
Method	`save`	save to file
Instance Variable	`args`	Undocumented
Instance Variable	`basis`	Undocumented
Instance Variable	`func`	Undocumented
Instance Variable	`name`	Undocumented
Instance Variable	`plot_cmd`	Undocumented
Instance Variable	`res`	Undocumented
Instance Variable	`x`	Undocumented

def __init__(self, func, x, args=(), basis=None, name=None, plot_cmd=None, load=True): ¶

func: objective function
x: point in search space, middle point of the sections
args: arguments passed to func
basis: evaluated points are func(x + locations[j] * basis[i]) for i in len(basis) for j in len(locations), see do()
name: filename where to save the result
plot_cmd: command used to plot the data, typically matplotlib pyplots plot or semilogy
load: load previous data from file str(func) + '.pkl'

def do(self, repetitions=1, locations=np.arange(-0.5, 0.6, 0.2), plot=True): ¶

generates, plots and saves function values func(y), where y is 'close' to x (see __init__()). The data are stored in the res attribute and the class instance is saved in a file with (the weired) name str(func).

Parameters

repetitions: for each point, only for noisy functions is >1 useful. For repetitions==0 only already generated data are plotted.
locations: coordinated wise deviations from the middle point given in __init__

def flattened(self): ¶

return flattened data (x, f) such that for the sweep through coordinate i we have for data point j that f[i][j] == func(x[i][j])

def load(self, name=None): ¶

load from file

def plot(self, plot_cmd=None, tf=(lambda y: y)): ¶

plot the data we have, return self

def save(self, name=None): ¶

save to file

args = ¶

Undocumented

basis = ¶

Undocumented

func = ¶

Undocumented

name = ¶

Undocumented

plot_cmd = ¶

Undocumented

res = ¶

Undocumented

x = ¶

Undocumented