gwlearn.linear_model.GWLogisticRegression¶

class gwlearn.linear_model.GWLogisticRegression(bandwidth=None, fixed=False, kernel='bisquare', include_focal=True, graph=None, n_jobs=-1, fit_global_model=True, strict=False, keep_models=False, temp_folder=None, batch_size=None, min_proportion=0.2, undersample=False, leave_out=None, random_state=None, verbose=False, **kwargs)[source]¶

Geographically weighted logistic regression

Fits one sklearn.linear_model.LogisticRegression per focal observation using spatially varying sample weights.

Notes

y must be binary ({0, 1} or boolean).
To enable prediction on new data via predict()/predict_proba(), you must set keep_models=True (store in memory) or keep_models=Path(...) (serialize to disk).
Only point geometries are supported.

Parameters:¶

bandwidth : float | int | None¶

Bandwidth for defining neighborhoods.

If fixed=True, this is a distance threshold.
If fixed=False, this is the number of nearest neighbors used to form the local neighborhood.

If graph is provided, bandwidth is ignored.

fixed : bool, optional¶

True for distance based bandwidth and False for adaptive (nearest neighbor) bandwidth, by default False

kernel : str | Callable, optional¶

Type of kernel function used to weight observations, by default “bisquare”

include_focal : bool, optional¶

Include focal in the local model training. Excluding it allows assessment of geographically weighted metrics on unseen data without a need for train/test split, hence providing value for all samples. This is needed for further spatial analysis of the model performance (and generalises to models that do not support OOB scoring). However, it leaves out the most representative sample. By default True

graph : Graph, optional¶

Custom libpysal.graph.Graph object encoding the spatial interaction between observations in the sample. If given, it is used directly and bandwidth, fixed, kernel, and include_focal keywords are ignored. Either geometry or graph need to be specified. To allow prediction, it is required to specify geometry. Potentially, both can be specified where graph encodes spatial interaction between observations in geometry.

n_jobs : int, optional¶

The number of jobs to run in parallel. -1 means using all processors by default -1

fit_global_model : bool, optional¶

Determines if the global baseline model shall be fitted alongside the geographically weighted, by default True

strict : bool | None, optional¶

Do not fit any models if at least one neighborhood has invariant y, by default False. None is treated as False but provides a warning if there are invariant models.

keep_models : bool | str | Path, optional¶

Keep all local models (required for prediction), by default False. Note that for some models, like random forests, the objects can be large. If string or Path is provided, the local models are not held in memory but serialized to the disk from which they are loaded in prediction.

temp_folder : str | None, optional¶

Folder to be used by the pool for memmapping large arrays for sharing memory with worker processes, e.g., /tmp. Passed to joblib.Parallel, by default None

batch_size : int | None, optional¶

Number of models to process in each batch. Specify batch_size if your models do not fit into memory. By default None

min_proportion : float, optional¶

Minimum proportion of minority class for a model to be fitted, by default 0.2

undersample : bool | float, optional¶

Whether to apply random undersampling to balance classes.

If True, undersample the majority class to match the minority class (i.e., minority/majority ratio = 1.0).

If a float alpha > 0, target a minority/majority ratio of alpha after resampling, i.e. alpha = N_min / N_resampled_majority. By default False

leave_out : float | int, optional¶

Leave out a fraction (when float) or a set number (when int) of random observations from each local model to be used to measure out-of-sample log loss based on pooled samples from all the models. This is useful for bandwidth selection for cases where some local models are not fitted due to local invariance and resulting information criteria are not comparable.

random_state : int | None, optional¶

Random seed for reproducibility, by default None

verbose : bool, optional¶

Whether to print progress information, by default False

**kwargs¶

Additional keyword arguments passed to model initialisation

proba_[source]¶

Probability predictions for focal locations based on a local model trained around the point itself.

Type:¶: pd.DataFrame

pred_[source]¶

Binary predictions for focal locations based on a local model trained around the location itself.

Type:¶: pd.Series

hat_values_[source]¶

Hat values for each location (diagonal elements of hat matrix)

Type:¶: pd.Series

effective_df_[source]¶

Effective degrees of freedom (sum of hat values)

Type:¶: float

log_likelihood_[source]¶

Global log likelihood of the model

Type:¶: float

aic_[source]¶

Akaike information criterion of the model

Type:¶: float

aicc_[source]¶

Corrected Akaike information criterion to account for model complexity (smaller bandwidths)

Type:¶: float

bic_[source]¶

Bayesian information criterion

Type:¶: float

local_coef_[source]¶

Local coefficient of the features in the decision function for each feature at each location

Type:¶: pd.DataFrame

local_intercept_[source]¶

Local intercept values at each location

Type:¶: pd.Series

prediction_rate_[source]¶

Proportion of models that are fitted, where the rest are skipped due to not fulfilling min_proportion.

Type:¶: float

left_out_y_[source]¶

Array of y values left out when leave_out is set.

Type:¶: np.ndarray

left_out_proba_[source]¶

Array of probabilites on left out observations in local models when leave_out is set.

Type:¶: np.ndarray

left_out_w_[source]¶

Array of weights on left out observations in local models when leave_out is set.

Type:¶: np.ndarray

Examples

>>> import geopandas as gpd
>>> from geodatasets import get_path
>>> from gwlearn.linear_model import GWLogisticRegression

>>> gdf = gpd.read_file(get_path('geoda.guerry'))
>>> X = gdf[['Crm_prp', 'Litercy', 'Donatns', 'Lottery']]
>>> y = gdf["Region"] == 'E'

>>> gw = GWLogisticRegression(
...     bandwidth=30,
...     fixed=False,
...     keep_models=True,
...     max_iter=200,
... ).fit(X, y, geometry=gdf.representative_point())
>>> gw.pred_.head()
0     True
1    False
2    False
3     True
4     True
dtype: boolean

Methods

`__init__`([bandwidth, fixed, kernel, ...])
`fit`(X, y[, geometry])	Fit geographically weighted local classification models.
`get_metadata_routing`()	Get metadata routing of this object.
`get_params`([deep])	Get parameters for this estimator.
`local_metric`(func, args, *kwargs)	Compute a metric per fitted local model.
`predict`(X, geometry[, bandwidth, ...])	Predict classes for new observations.
`predict_proba`(X, geometry[, bandwidth, ...])	Predict class probabilities for new observations.
`score`(X, y, geometry[, bandwidth, ...])	Return the mean accuracy on the given test data and labels.
`set_fit_request`(*[, geometry])	Configure whether metadata should be requested to be passed to the `fit` method.
`set_params`(**params)	Set the parameters of this estimator.
`set_predict_proba_request`(*[, bandwidth, ...])	Configure whether metadata should be requested to be passed to the `predict_proba` method.
`set_predict_request`(*[, bandwidth, ...])	Configure whether metadata should be requested to be passed to the `predict` method.
`set_score_request`(*[, bandwidth, geometry, ...])	Configure whether metadata should be requested to be passed to the `score` method.

Attributes

fit(X, y, geometry=None)[source]¶

Fit geographically weighted local classification models.

This fits one local model per focal observation (subject to local invariance checks and min_proportion) and stores focal predictions.

Parameters:¶

X : pandas.DataFrame¶: Feature matrix.
y : pandas.Series¶: Binary target encoded as boolean or {0, 1}.
geometry : geopandas.GeoSeries | None¶: Geographic location of the observations in the sample. Used to determine the spatial interaction weight based on specification by bandwidth, fixed, kernel, and include_focal keywords. If None, a precomputed graph needs to be specified. To allow prediction, it is required to specify geometry. If both graph and geometry are specified, graph is used at the fit time, while geometry is used for prediction.

Returns:¶

Fitted estimator.

Return type:¶

self

Notes

The neighborhood definition comes from either self.graph or from geometry + (bandwidth, fixed, kernel, include_focal).

get_metadata_routing()[source]¶

Get metadata routing of this object.

Please check User Guide on how the routing mechanism works.

Returns:¶: routing – A MetadataRequest encapsulating routing information.
Return type:¶: MetadataRequest

get_params(deep=True)[source]¶

Get parameters for this estimator.

Parameters:¶

deep : bool, default=True¶: If True, will return the parameters for this estimator and contained subobjects that are estimators.

Returns:¶

params – Parameter names mapped to their values.

Return type:¶

dict

local_metric(func, *args, **kwargs)[source]¶

Compute a metric per fitted local model.

Parameters:¶

func : callable¶: Callable with a signature func(y_true, y_pred, *args, **kwargs).

Returns:¶

One value per focal location (NaN for skipped / unfitted local models).

Return type:¶

numpy.ndarray

predict(X, geometry, bandwidth='nearest', global_model_weight=0)[source]¶

Predict classes for new observations.

This is equivalent to predict_proba(...).idxmax(axis=1).

Prediction can be retrieved either from the nearest local model or based on the ensemble of local models. In the latter case, the prediction process works as follows:

For a new location on which you want a prediction, identify local models within the bandwidth used to train the model.
Apply the kernel function used to train the model to derive weights of each of the local models.
Make prediction using each of the local models in the bandwidth.
Make weighted average of predictions based on the kernel weights.
Normalize the result to ensure sum of probabilities is 1.

The results from the nearest and ensemble predictions are typically similar, with the ensemble being significantly slower due to the required number of inference calls.

Further the prediction can be a result of a fusion of local and global models when global_model_weight is set to a non-zero value, following Georganos et al. [2021].

Parameters:¶

X : pandas.DataFrame¶: Feature matrix for new observations.
geometry : geopandas.GeoSeries¶: Point geometries for new observations.
bandwidth : "nearest", float or None¶: Prediction method. Nearest uses the nearest location available at the fit time and does prediction using its single model. When set to a numeric value, uses an ensemble of local models available within the bandwidth, with predictions from individual models being weighted based on the distance and a set kernel. When None, uses the bandwidth set at the fit time.
global_model_weight : float¶: Weight of the prediction from the global model. When non-zero, the resulting prediction is a weighted average of the values from local model(s) and from global model, where local prediction has a weight of 1 and global model has a weight equal to global_model_weight.

Returns:¶

Predicted class.

Return type:¶

pandas.Series

Notes

Requires the estimator to have been fit with keep_models=True (or a Path) so local models can be used at prediction time.

predict_proba(X, geometry, bandwidth='nearest', global_model_weight=0)[source]¶

Predict class probabilities for new observations.

Prediction can be retrieved either from the nearest local model or based on the ensemble of local models. In the latter case, the prediction process works as follows:

For a new location on which you want a prediction, identify local models within the bandwidth used to train the model.
Apply the kernel function used to train the model to derive weights of each of the local models.
Make prediction using each of the local models in the bandwidth.
Make weighted average of predictions based on the kernel weights.
Normalize the result to ensure sum of probabilities is 1.

The results from the nearest and ensemble predictions are typically similar, with the ensemble being significantly slower due to the required number of inference calls.

Further the prediction can be a result of a fusion of local and global models when global_model_weight is set to a non-zero value, following Georganos et al. [2021].

Parameters:¶

X : pandas.DataFrame¶: Feature matrix for new observations.
geometry : geopandas.GeoSeries¶: Point geometries for new observations.
bandwidth : "nearest", float or None¶: Prediction method. Nearest uses the nearest location available at the fit time and does prediction using its single model. When set to a numeric value, uses an ensemble of local models available within the bandwidth, with predictions from individual models being weighted based on the distance and a set kernel. When None, uses the bandwidth set at the fit time.
global_model_weight : float¶: Weight of the prediction from the global model. When non-zero, the resulting prediction is a weighted average of the values from local model(s) and from global model, where local prediction has a weight of 1 and global model has a weight equal to global_model_weight.

Returns:¶

Predicted probabilities with columns equal to the global classes observed during fit.

Return type:¶

pandas.DataFrame

Notes

Requires the estimator to have been fit with keep_models=True (or a Path) so local models can be used at prediction time.

score(X, y, geometry, bandwidth='nearest', global_model_weight=0)[source]¶

Return the mean accuracy on the given test data and labels.

Parameters:¶

X : pandas.DataFrame¶: Feature matrix for new observations.
y : pandas.Series¶: True labels for X.
geometry : geopandas.GeoSeries¶: Point geometries for new observations.
bandwidth : "nearest", float or None¶: Prediction method. See predict().
global_model_weight : float¶: Weight of the prediction from the global model.

Returns:¶

Mean accuracy of self.predict(X, geometry).

Return type:¶

float

set_fit_request(*, geometry='$UNCHANGED$')[source]¶

Configure whether metadata should be requested to be passed to the fit 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 fit 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 fit.
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:¶

geometry : str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED¶: Metadata routing for geometry parameter in fit.

Returns:¶

self – The updated object.

Return type:¶

object

set_params(**params)[source]¶

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as Pipeline). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Parameters:¶

**params : dict¶: Estimator parameters.

Returns:¶

self – Estimator instance.

Return type:¶

estimator instance

set_predict_proba_request(*, bandwidth='$UNCHANGED$', geometry='$UNCHANGED$', global_model_weight='$UNCHANGED$')[source]¶

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

The options for each parameter are:

True: metadata is requested, and passed to predict_proba 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 predict_proba.
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:¶

bandwidth : str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED¶: Metadata routing for bandwidth parameter in predict_proba.
geometry : str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED¶: Metadata routing for geometry parameter in predict_proba.
global_model_weight : str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED¶: Metadata routing for global_model_weight parameter in predict_proba.

Returns:¶

self – The updated object.

Return type:¶

object

set_predict_request(*, bandwidth='$UNCHANGED$', geometry='$UNCHANGED$', global_model_weight='$UNCHANGED$')[source]¶

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

The options for each parameter are:

True: metadata is requested, and passed to predict 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 predict.
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:¶

bandwidth : str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED¶: Metadata routing for bandwidth parameter in predict.
geometry : str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED¶: Metadata routing for geometry parameter in predict.
global_model_weight : str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED¶: Metadata routing for global_model_weight parameter in predict.

Returns:¶

self – The updated object.

Return type:¶

object

set_score_request(*, bandwidth='$UNCHANGED$', geometry='$UNCHANGED$', global_model_weight='$UNCHANGED$')[source]¶

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

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:¶

bandwidth : str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED¶: Metadata routing for bandwidth parameter in score.
geometry : str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED¶: Metadata routing for geometry parameter in score.
global_model_weight : str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED¶: Metadata routing for global_model_weight parameter in score.

Returns:¶

self – The updated object.

Return type:¶

object