parrt
clfviz throws KeyError, points to color_map index.
UPDATE:
found problem and fix. to fix change line 319 in classifiers.py:
color_map = {v: class_colors[i] for i, v in enumerate(class_values)} ## changed from color_map = {i: class_colors[i] for i, v in enumerate(class_values)}
I tried to do a pull request, but in the newest clone it is already fixed! this seems to be a problem only with pip3 install ... version
This error happens with may data with two features (X.shape = (39,2)) and one feature (X.shape = (39, 1)). I printed and gave you X,y at the end if you want to recreate.
Here's the code:
dt = DecisionTreeClassifier(max_depth=2)
dt.fit(X, y)
from dtreeviz import clfviz
clfviz(dt, X, y, feature_names=["a","b"], markers=['o','X','s','D'], target_name="y")
And error message:
/usr/local/lib/python3.9/site-packages/sklearn/base.py:450: UserWarning: X does not have valid feature names, but DecisionTreeClassifier was fitted with feature names
warnings.warn(
/usr/local/lib/python3.9/site-packages/sklearn/base.py:450: UserWarning: X does not have valid feature names, but DecisionTreeClassifier was fitted with feature names
warnings.warn(
---------------------------------------------------------------------------
KeyError Traceback (most recent call last)
Input In [374], in <module>
----> 1 clfviz(dt, X, y, feature_names=features, target_name=target)
File /usr/local/lib/python3.9/site-packages/dtreeviz/classifiers.py:103, in clfviz(model, X, y, ntiles, tile_fraction, binary_threshold, show, feature_names, target_name, class_names, markers, boundary_marker, boundary_markersize, fontsize, fontname, dot_w, yshift, sigma, colors, ranges, ax)
88 clfviz_univar(model=model, x=X, y=y,
89 ntiles=ntiles,
90 binary_threshold=binary_threshold,
(...)
100 colors=colors,
101 ax=ax)
102 elif len(X.shape) == 2 and X.shape[1] == 2:
--> 103 clfviz_bivar(model=model, X=X, y=y,
104 ntiles=ntiles, tile_fraction=tile_fraction,
105 binary_threshold=binary_threshold,
106 show=show,
107 feature_names=feature_names, target_name=target_name,
108 class_names=class_names,
109 markers=markers,
110 boundary_marker=boundary_marker,
111 boundary_markersize=boundary_markersize,
112 fontsize=fontsize, fontname=fontname,
113 dot_w=dot_w, colors=colors,
114 ranges=ranges,
115 ax=ax)
116 else:
117 raise ValueError(f"Expecting 2D data not {X.shape}")
File /usr/local/lib/python3.9/site-packages/dtreeviz/classifiers.py:188, in clfviz_bivar(model, X, y, ntiles, tile_fraction, binary_threshold, show, feature_names, target_name, class_names, markers, boundary_marker, boundary_markersize, fontsize, fontname, dot_w, colors, ranges, ax)
184 if 'misclassified' in show:
185 # Show correctly classified markers
186 good_x = x_[class_X_pred[i] == class_values[i],:]
187 ax.scatter(good_x[:, 0], good_x[:, 1],
--> 188 s=dot_w, c=color_map[i],
189 marker=markers[i],
190 alpha=colors['scatter_marker_alpha'],
191 edgecolors=colors['scatter_edge'],
192 lw=.5)
193 # Show misclassified markers (can't have alpha per marker so do in 2 calls)
194 bad_x = x_[class_X_pred[i] != class_values[i],:]
KeyError: 0
data:
X = np.array([[2.], [1.], [3.], [3.], [1.], [1.], [2.], [3.], [1.], [2.], [3.], [1.], [1.], [1.], [1.], [1.], [3.], [2.], [3.], [2.], [1.], [1.], [2.], [1.], [1.], [1.], [1.], [1.], [3.], [3.], [1.], [1.], [3.], [2.], [1.], [1.], [2.], [1.], [3.], [1.], [1.], [1.], [2.], [1.], [1.], [3.], [2.], [2.], [3.], [3.], [3.], [3.], [1.], [1.], [1.], [1.], [3.], [1.], [3.]]) y = np.array([3, 1, 4, 1, 4, 2, 1, 1, 4, 3, 4, 4, 1, 2, 3, 2, 2, 3, 3, 2, 3, 3, 2, 2, 3, 2, 4, 2, 2, 1, 4, 2, 3, 4, 1, 4, 4, 1, 2, 3, 3, 3, 4, 1, 1, 1, 4, 4, 3, 2, 4, 1, 1, 4, 2, 3, 1, 1, 2])
