milia.recommend.action#

Case-based recommendation: suggest the action that fixed the most similar past cases.

Classes

CaseRecommender([n_neighbors, metric])

Majority action among the k nearest past cases, ties going to the nearer case.

class milia.recommend.action.CaseRecommender(n_neighbors=3, metric='minkowski')#

Bases: ClassifierMixin, BaseEstimator

Majority action among the k nearest past cases, ties going to the nearer case.

The notebook version averaged integer-coded actions and rounded, which depends on how actions happen to be numbered; this is a true vote.

Parameters:
  • n_neighbors (int) – Past cases consulted.

  • metric (str) – Distance metric.

Examples

>>> X = np.array([[0.0], [0.1], [5.0], [5.1]])
>>> CaseRecommender(n_neighbors=2).fit(X, ["voltvar", "voltvar", "storage", "storage"]).predict(
...     [[4.9]]
... ).tolist()
['storage']
fit(X, y)#

Index past cases and the action that fixed each.

Parameters:
  • X (ArrayLike) – Features of past cases.

  • y (ArrayLike) – The action that fixed each case.

Return type:

CaseRecommender

Returns:

The fitted recommender.

votes(X)#

Vote share of each action among the nearest past cases.

Equal votes, nudged by a negligible amount towards nearer cases so that ties go to the nearest.

Parameters:

X (ArrayLike) – Features of new cases.

Return type:

ndarray

Returns:

Array of shape (n, actions); rows sum to one.

predict(X)#

The majority action among each case’s nearest past cases.

Parameters:

X (ArrayLike) – Features of new cases.

Return type:

ndarray

Returns:

One action per case.

Examples

>>> import numpy as np
>>> X = np.array([[0.0], [0.1], [0.2], [5.0], [5.1]])
>>> actions = ["voltvar", "storage", "storage", "voltwatt", "voltwatt"]
>>> CaseRecommender(n_neighbors=3).fit(X, actions).predict([[0.05], [5.05]]).tolist()
['storage', 'voltwatt']
predict_proba(X)#

Vote share of each action (see votes()).

Parameters:

X (ArrayLike) – Features of new cases.

Return type:

ndarray

Returns:

Array of shape (n, actions) in the order of classes_.

set_score_request(*, sample_weight='$UNCHANGED$')#

Configure whether metadata should be requested to be passed to the score method.

Note that this method is only relevant when this estimator is used as a sub-estimator within a meta-estimator and metadata routing is enabled with enable_metadata_routing=True (see sklearn.set_config()). Please check the User Guide on how the routing mechanism works.

The options for each parameter are:

  • True: metadata is requested, and passed to score if provided. The request is ignored if metadata is not provided.

  • False: metadata is not requested and the meta-estimator will not pass it to score.

  • None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

  • str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Parameters#

sample_weightstr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED

Metadata routing for sample_weight parameter in score.

Returns#

selfobject

The updated object.

Return type:

CaseRecommender