Fix/pd variance plot test coverage (#820)

* Fixed minor bug for conditional plots.

* Included importance explanation fixture.

* Included interaction explanation fixture.

* Included test for hbar ncols. Solved flake8 issues.

* Included value error ax for hbar plot.

* Included interaction plot tests.

* Included plot_pd_variance tests.

* Updated the explanation objects with random values. Fixed test that failed before because of simple explanation values.

* More random values.

* Removed lazy fixture.

Author: RobertSamoilescu

alibi/explainers/pd_variance.py

@@ -11,8 +11,7 @@
 import numpy as np
 from sklearn.base import BaseEstimator
 
-from alibi.api.defaults import (DEFAULT_DATA_PD, DEFAULT_DATA_PDVARIANCE,
-                                DEFAULT_META_PD, DEFAULT_META_PDVARIANCE)
+from alibi.api.defaults import (DEFAULT_DATA_PD, DEFAULT_DATA_PDVARIANCE, DEFAULT_META_PDVARIANCE)
 from alibi.api.interfaces import Explainer, Explanation
 from alibi.explainers import plot_pd
 from alibi.explainers.partial_dependence import (Kind, PartialDependence,
@@ -559,10 +558,9 @@ def _plot_feature_importance(exp: Explanation,
         feature_importance = feature_importance[sorted_indices][:top_k]
 
     # construct pd explanation object to reuse `plot_pd` function
-    meta = copy.deepcopy(DEFAULT_META_PD)
-    data = copy.deepcopy(DEFAULT_DATA_PD)
-    meta.update(exp.meta)
+    meta = copy.deepcopy(exp.meta)
     meta['params']['kind'] = 'average'
+    data = copy.deepcopy(DEFAULT_DATA_PD)
     data.update(feature_names=feature_names,
                 feature_values=feature_values,
                 pd_values=pd_values,
@@ -673,10 +671,9 @@ def _plot_feature_interaction(exp: Explanation,
     merged_features = list(itertools.chain.from_iterable(merged_features))  # type: ignore[arg-type]
 
     # construct pd explanation object to reuse `plot_pd` function
-    meta = copy.deepcopy(DEFAULT_META_PD)
-    data = copy.deepcopy(DEFAULT_DATA_PD)
-    meta.update(exp.meta)
+    meta = copy.deepcopy(exp.meta)
     meta['params']['kind'] = 'average'
+    data = copy.deepcopy(DEFAULT_DATA_PD)
     data.update(feature_names=merged_feature_names,
                 feature_values=merged_feature_values,
                 pd_values=merged_pd_values,
@@ -704,11 +701,11 @@ def _plot_feature_interaction(exp: Explanation,
 
         # set title for the first conditional importance plot
         ax = axes.flatten()[step * i + 1]
-        ax.set_title('inter({}|{}) = {:.3f}'.format(ft_name2, ft_name1, conditional_importance[i][0].item()))
+        ax.set_title('inter({}|{}) = {:.3f}'.format(ft_name2, ft_name1, conditional_importance[i][0][target_idx]))
 
         # set title for the second conditional importance plot
         ax = axes.flatten()[step * i + 2]
-        ax.set_title('inter({}|{}) = {:.3f}'.format(ft_name1, ft_name2, conditional_importance[i][1].item()))
+        ax.set_title('inter({}|{}) = {:.3f}'.format(ft_name1, ft_name2, conditional_importance[i][1][target_idx]))
 
     return axes