xgbse._base.XGBSEBaseEstimator¶
Base class for all estimators in xgbse. Implements explainability through prototypes.
get_neighbors(self, query_data, index_data=None, query_id=None, index_id=None, n_neighbors=30)
¶
Search for portotypes (size: n_neighbors) for each unit in a dataframe X. If units array is specified, comparables will be returned using its identifiers. If not, a dataframe of comparables indexes for each sample in X is returned.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
query_data |
pd.DataFrame |
Dataframe of features to be used as input |
required |
query_id |
[pd.Series, np.array] |
Series or array of identification for each sample of query_data. Will be used in set_index if specified. |
None |
index_id |
[pd.Series, np.array] |
Series or array of identification for each sample of index_id. If specified, comparables will be returned using this identifier. |
None |
n_neighbors |
int |
Number of neighbors/comparables to be considered. |
30 |
Returns:
| Type | Description |
|---|---|
comps_df (pd.DataFrame) |
A dataframe of comparables/neighbors for each evaluated sample. If units identifier is specified, the output dataframe is converted to use units the proper identifier for each sample. The reference sample is considered to be the index of the dataframe and its comparables are its specific row values. |