shamo.core.surrogate.abc.SurrABC¶

class shamo.core.surrogate.abc.SurrABC(name, parent_path, **kwargs)[source]¶

Bases: shamo.core.objects.dir.ObjDir

Generate a Gaussian process from a set of training data.

Parameters

namestr: The name of the generated surrogate model.
parent_pathstr, byte or os.PathLike: The path to the parent directory of the surrogate model.

Other Parameters

paramslist [tuple [str, shamo.DistABC]]: A list of tuples containing the names of the parameters and the corresponding distributions as values.
sol_json_pathstr: The path to the parametric solution the surrogate is built of.

Methods

`clear`
`copy`
`fit`	Generate a Gaussian process from a set of training data.
`fromkeys`	Create a new dictionary with keys from iterable and values set to value.
`get`	Return the value for key if key is in the dictionary, else default.
`get_gp`	Load the Gaussian process.
`get_relative_path`	Return the relative path from the object to a file or directory.
`items`
`keys`
`load`	Load an object from a JSON file.
`pop`	If key is not found, d is returned if given, otherwise KeyError is raised
`popitem`	Remove and return a (key, value) pair as a 2-tuple.
`predict`	Evaluate the Gaussian process on new points.
`save`	Save the object to a JSON file.
`setdefault`	Insert key with a value of default if key is not in the dictionary.
`update`	If E is present and has a .keys() method, then does: for k in E: D[k] = E[k] If E is present and lacks a .keys() method, then does: for k, v in E: D[k] = v In either case, this is followed by: for k in F: D[k] = F[k]
`values`

Attributes

`gp_path`	Return the path of the Gaussian process file.
`json_path`	Return the path to the object JSON file.
`name`	Return the name of the object.
`params`	Return the parameters of the surrogate model.
`parent_path`	Return the path to the parent directory of the object.
`path`	Return the path to the object directory.
`sol_json_path`

clear() → None. Remove all items from D.¶

copy() → a shallow copy of D¶

classmethod fit(name, parent_path, sol, **kwargs)[source]¶

Generate a Gaussian process from a set of training data.

Parameters

namestr: The name of the generated surrogate model.
parent_pathstr, byte or os.PathLike: The path to the parent directory of the surrogate model.
solshamo.core.solutions.parametric.SolParamABC: The parametric solution to generate a surrogate model for.

Returns

shamo.Surrogate: The generated surrogate model.

Other Parameters

kernelsklearn.gaussian_process.kernels.Kernel: The kernel used to generate the Gaussian process.
n_restarts_optimizerint: The number of restarts for the optimisation step.
random_stateint: The seed for the random state.
alphafloat: The added diagonal to account for noise on training points.
n_procint: The number of jobs to run in parallel. None means 1 job runs at a time and -1 means all cores are used.

See also

sklearn.gaussian_process.GaussianProcessRegressor

fromkeys(iterable, value=None, /)¶: Create a new dictionary with keys from iterable and values set to value.

get(key, default=None, /)¶: Return the value for key if key is in the dictionary, else default.

get_gp()[source]¶

Load the Gaussian process.

Returns

sklearn.gaussian_process.GaussianProcessRegressor: The Gaussian process.

get_relative_path(path)¶

Return the relative path from the object to a file or directory.

Parameters

pathstr, byte or os.PathLike: The path to the file or directory to compute the relative path for.

Returns

pathlib.Path: The relative path to the file or directory.

property gp_path¶

Return the path of the Gaussian process file.

Returns

pathlib.Path: The path of the Gaussian process file.

items() → a set-like object providing a view on D’s items¶

property json_path¶

Return the path to the object JSON file.

Returns

pathlib.Path: The path to the object JSON file.

keys() → a set-like object providing a view on D’s keys¶

classmethod load(json_path)¶

Load an object from a JSON file.

Parameters

json_pathstr, byte or os.PathLike: The path to the JSON file containing the object data.

Raises

TypeError: If argument json_path is not a str, byte or os.PathLike.

property name¶

Return the name of the object.

Returns

str: The name of the object.

property params¶

Return the parameters of the surrogate model.

Returns

list [tuple [str, shamo.DistABC]]: A list of tuples containing the names of the parameters and the corresponding distributions as values.

property parent_path¶

Return the path to the parent directory of the object.

Returns

pathlib.Path: The path to the parent directory of the object.

property path¶

Return the path to the object directory.

Returns

pathlib.Path: The path to the object directory.

pop(k[, d]) → v, remove specified key and return the corresponding value.¶: If key is not found, d is returned if given, otherwise KeyError is raised

popitem(/)¶

Remove and return a (key, value) pair as a 2-tuple.

Pairs are returned in LIFO (last-in, first-out) order. Raises KeyError if the dict is empty.

predict(x, **kwargs)[source]¶

Evaluate the Gaussian process on new points.

Parameters

xnumpy.ndarray: The coordinates of the evaluation points in the parameter space. Each row represents an evaluation.

Notes

To change the behaviour of this method, one should subclass Surrogate and change the _post_pro method.

save(exist_ok=True)¶

Save the object to a JSON file.

Parameters

exist_okbool, optional: If set to True, any already existing object is overriden. Otherwise, if the object already exists, a FileExistsError. The default is True.

Raises

FileExistsError: If exist_ok is set to False and the object already exists.
TypeError: If any of the keys/values to be stored is not a str, int, float, bool or None.

setdefault(key, default=None, /)¶

Insert key with a value of default if key is not in the dictionary.

Return the value for key if key is in the dictionary, else default.

update([E, ]**F) → None. Update D from dict/iterable E and F.¶: If E is present and has a .keys() method, then does: for k in E: D[k] = E[k] If E is present and lacks a .keys() method, then does: for k, v in E: D[k] = v In either case, this is followed by: for k in F: D[k] = F[k]

values() → an object providing a view on D’s values¶