physbo.gp.core.learning module

class physbo.gp.core.learning.adam(gp, config)

ベースクラス: physbo.gp.core.learning.online

default

get_one_update(params, X, t)

パラメータ:

• params (numpy.ndarray) -- Parameters for optimization. Array of real elements of size (n,), where ‘n’ is the number of independent variables.
• X (numpy.ndarray) -- N x d dimensional matrix. Each row of X denotes the d-dimensional feature vector of search candidate.
• t (numpy.ndarray) -- N-dimensional vector that represents the corresponding negative energy of search candidates.

reset()

class physbo.gp.core.learning.batch(gp, config)

ベースクラス: object

basis class for batch learning

init_params_search(X, t)

パラメータ:

• X (numpy.ndarray) -- N x d dimensional matrix. Each row of X denotes the d-dimensional feature vector of search candidate.
• t (numpy.ndarray) -- N-dimensional vector that represents the corresponding negative energy of search candidates.

戻り値:

The parameters which give the minimum marginal likelihood.

戻り値の型:

numpy.ndarray

one_run(params, X, t, max_iter=None)

パラメータ:

• params (numpy.ndarray) -- Initial guess for optimization. Array of real elements of size (n,), where ‘n’ is the number of independent variables.
• X (numpy.ndarray) -- N x d dimensional matrix. Each row of X denotes the d-dimensional feature vector of search candidate.
• t (numpy.ndarray) -- N-dimensional vector that represents the corresponding negative energy of search candidates.
• max_iter (int) -- Maximum number of iterations to perform.

戻り値:

The solution of the optimization.

戻り値の型:

numpy.ndarray

run(X, t)

Performing optimization using the L-BFGS-B algorithm

パラメータ:

• X (numpy.ndarray) -- N x d dimensional matrix. Each row of X denotes the d-dimensional feature vector of search candidate.
• t (numpy.ndarray) -- N-dimensional vector that represents the corresponding negative energy of search candidates.

戻り値:

The solution of the optimization.

戻り値の型:

numpy.ndarray

class physbo.gp.core.learning.online(gp, config)

ベースクラス: object

base class for online learning

disp_marlik(params, eval_X, eval_t, num_epoch=None)

Displaying marginal likelihood

パラメータ:

• params (numpy.ndarray) -- Parameters for optimization. Array of real elements of size (n,), where ‘n’ is the number of independent variables.
• eval_X (numpy.ndarray) -- N x d dimensional matrix. Each row of X denotes the d-dimensional feature vector of search candidate.
• eval_t (numpy.ndarray) -- N-dimensional vector that represents the corresponding negative energy of search candidates.
• num_epoch (int) -- Number of epochs

get_one_update(params, X, t)

init_params_search(X, t)

Initial parameter searchs

パラメータ:

• X (numpy.ndarray) -- N x d dimensional matrix. Each row of X denotes the d-dimensional feature vector of search candidate.
• t (numpy.ndarray) -- N-dimensional vector that represents the corresponding negative energy of search candidates.

戻り値:

The parameter which gives the minimum likelihood.

戻り値の型:

numpy.ndarray

one_run(params, X, t, max_epoch=None)

パラメータ:

• params (numpy.ndarray) -- Parameters for optimization. Array of real elements of size (n,), where ‘n’ is the number of independent variables.
• X (numpy.ndarray) -- N x d dimensional matrix. Each row of X denotes the d-dimensional feature vector of search candidate.
• t (numpy.ndarray) -- N-dimensional vector that represents the corresponding negative energy of search candidates.
• max_epoch (int) -- Maximum candidate epochs

戻り値:

The solution of the optimization.

戻り値の型:

numpy.ndarray

run(X, t)

Run initial search and hyper parameter running.

パラメータ:

• X (numpy.ndarray) -- N x d dimensional matrix. Each row of X denotes the d-dimensional feature vector of search candidate.
• t (numpy.ndarray) -- N-dimensional vector that represents the corresponding negative energy of search candidates.

戻り値:

The solution of the optimization.

戻り値の型:

numpy.ndarray