-
-
Notifications
You must be signed in to change notification settings - Fork 25.3k
/
decision_boundary.py
375 lines (308 loc) · 13.1 KB
/
decision_boundary.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
from functools import reduce
import numpy as np
from ...preprocessing import LabelEncoder
from ...utils import check_matplotlib_support
from ...utils import _safe_indexing
from ...base import is_regressor
from ...utils.validation import (
check_is_fitted,
_is_arraylike_not_scalar,
_num_features,
)
def _check_boundary_response_method(estimator, response_method):
"""Return prediction method from the `response_method` for decision boundary.
Parameters
----------
estimator : object
Fitted estimator to check.
response_method : {'auto', 'predict_proba', 'decision_function', 'predict'}
Specifies whether to use :term:`predict_proba`,
:term:`decision_function`, :term:`predict` as the target response.
If set to 'auto', the response method is tried in the following order:
:term:`decision_function`, :term:`predict_proba`, :term:`predict`.
Returns
-------
prediction_method: callable
Prediction method of estimator.
"""
has_classes = hasattr(estimator, "classes_")
if has_classes and _is_arraylike_not_scalar(estimator.classes_[0]):
msg = "Multi-label and multi-output multi-class classifiers are not supported"
raise ValueError(msg)
if has_classes and len(estimator.classes_) > 2:
if response_method not in {"auto", "predict"}:
msg = (
"Multiclass classifiers are only supported when response_method is"
" 'predict' or 'auto'"
)
raise ValueError(msg)
methods_list = ["predict"]
elif response_method == "auto":
methods_list = ["decision_function", "predict_proba", "predict"]
else:
methods_list = [response_method]
prediction_method = [getattr(estimator, method, None) for method in methods_list]
prediction_method = reduce(lambda x, y: x or y, prediction_method)
if prediction_method is None:
raise ValueError(
f"{estimator.__class__.__name__} has none of the following attributes: "
f"{', '.join(methods_list)}."
)
return prediction_method
class DecisionBoundaryDisplay:
"""Decisions boundary visualization.
It is recommended to use
:func:`~sklearn.inspection.DecisionBoundaryDisplay.from_estimator`
to create a :class:`DecisionBoundaryDisplay`. All parameters are stored as
attributes.
Read more in the :ref:`User Guide <visualizations>`.
.. versionadded:: 1.1
Parameters
----------
xx0 : ndarray of shape (grid_resolution, grid_resolution)
First output of :func:`meshgrid <numpy.meshgrid>`.
xx1 : ndarray of shape (grid_resolution, grid_resolution)
Second output of :func:`meshgrid <numpy.meshgrid>`.
response : ndarray of shape (grid_resolution, grid_resolution)
Values of the response function.
xlabel : str, default=None
Default label to place on x axis.
ylabel : str, default=None
Default label to place on y axis.
Attributes
----------
surface_ : matplotlib `QuadContourSet` or `QuadMesh`
If `plot_method` is 'contour' or 'contourf', `surface_` is a
:class:`QuadContourSet <matplotlib.contour.QuadContourSet>`. If
`plot_method` is 'pcolormesh', `surface_` is a
:class:`QuadMesh <matplotlib.collections.QuadMesh>`.
ax_ : matplotlib Axes
Axes with confusion matrix.
figure_ : matplotlib Figure
Figure containing the confusion matrix.
See Also
--------
DecisionBoundaryDisplay.from_estimator : Plot decision boundary given an estimator.
Examples
--------
>>> import matplotlib.pyplot as plt
>>> import numpy as np
>>> from sklearn.datasets import load_iris
>>> from sklearn.inspection import DecisionBoundaryDisplay
>>> from sklearn.tree import DecisionTreeClassifier
>>> iris = load_iris()
>>> feature_1, feature_2 = np.meshgrid(
... np.linspace(iris.data[:, 0].min(), iris.data[:, 0].max()),
... np.linspace(iris.data[:, 1].min(), iris.data[:, 1].max())
... )
>>> grid = np.vstack([feature_1.ravel(), feature_2.ravel()]).T
>>> tree = DecisionTreeClassifier().fit(iris.data[:, :2], iris.target)
>>> y_pred = np.reshape(tree.predict(grid), feature_1.shape)
>>> display = DecisionBoundaryDisplay(
... xx0=feature_1, xx1=feature_2, response=y_pred
... )
>>> display.plot()
<...>
>>> display.ax_.scatter(
... iris.data[:, 0], iris.data[:, 1], c=iris.target, edgecolor="black"
... )
<...>
>>> plt.show()
"""
def __init__(self, *, xx0, xx1, response, xlabel=None, ylabel=None):
self.xx0 = xx0
self.xx1 = xx1
self.response = response
self.xlabel = xlabel
self.ylabel = ylabel
def plot(self, plot_method="contourf", ax=None, xlabel=None, ylabel=None, **kwargs):
"""Plot visualization.
Parameters
----------
plot_method : {'contourf', 'contour', 'pcolormesh'}, default='contourf'
Plotting method to call when plotting the response. Please refer
to the following matplotlib documentation for details:
:func:`contourf <matplotlib.pyplot.contourf>`,
:func:`contour <matplotlib.pyplot.contour>`,
:func:`pcolormesh <matplotlib.pyplot.pcolormesh>`.
ax : Matplotlib axes, default=None
Axes object to plot on. If `None`, a new figure and axes is
created.
xlabel : str, default=None
Overwrite the x-axis label.
ylabel : str, default=None
Overwrite the y-axis label.
**kwargs : dict
Additional keyword arguments to be passed to the `plot_method`.
Returns
-------
display: :class:`~sklearn.inspection.DecisionBoundaryDisplay`
Object that stores computed values.
"""
check_matplotlib_support("DecisionBoundaryDisplay.plot")
import matplotlib.pyplot as plt # noqa
if plot_method not in ("contourf", "contour", "pcolormesh"):
raise ValueError(
"plot_method must be 'contourf', 'contour', or 'pcolormesh'"
)
if ax is None:
_, ax = plt.subplots()
plot_func = getattr(ax, plot_method)
self.surface_ = plot_func(self.xx0, self.xx1, self.response, **kwargs)
if xlabel is not None or not ax.get_xlabel():
xlabel = self.xlabel if xlabel is None else xlabel
ax.set_xlabel(xlabel)
if ylabel is not None or not ax.get_ylabel():
ylabel = self.ylabel if ylabel is None else ylabel
ax.set_ylabel(ylabel)
self.ax_ = ax
self.figure_ = ax.figure
return self
@classmethod
def from_estimator(
cls,
estimator,
X,
*,
grid_resolution=100,
eps=1.0,
plot_method="contourf",
response_method="auto",
xlabel=None,
ylabel=None,
ax=None,
**kwargs,
):
"""Plot decision boundary given an estimator.
Read more in the :ref:`User Guide <visualizations>`.
Parameters
----------
estimator : object
Trained estimator used to plot the decision boundary.
X : {array-like, sparse matrix, dataframe} of shape (n_samples, 2)
Input data that should be only 2-dimensional.
grid_resolution : int, default=100
Number of grid points to use for plotting decision boundary.
Higher values will make the plot look nicer but be slower to
render.
eps : float, default=1.0
Extends the minimum and maximum values of X for evaluating the
response function.
plot_method : {'contourf', 'contour', 'pcolormesh'}, default='contourf'
Plotting method to call when plotting the response. Please refer
to the following matplotlib documentation for details:
:func:`contourf <matplotlib.pyplot.contourf>`,
:func:`contour <matplotlib.pyplot.contour>`,
:func:`pcolormesh <matplotlib.pyplot.pcolormesh>`.
response_method : {'auto', 'predict_proba', 'decision_function', \
'predict'}, default='auto'
Specifies whether to use :term:`predict_proba`,
:term:`decision_function`, :term:`predict` as the target response.
If set to 'auto', the response method is tried in the following order:
:term:`decision_function`, :term:`predict_proba`, :term:`predict`.
For multiclass problems, :term:`predict` is selected when
`response_method="auto"`.
xlabel : str, default=None
The label used for the x-axis. If `None`, an attempt is made to
extract a label from `X` if it is a dataframe, otherwise an empty
string is used.
ylabel : str, default=None
The label used for the y-axis. If `None`, an attempt is made to
extract a label from `X` if it is a dataframe, otherwise an empty
string is used.
ax : Matplotlib axes, default=None
Axes object to plot on. If `None`, a new figure and axes is
created.
**kwargs : dict
Additional keyword arguments to be passed to the
`plot_method`.
Returns
-------
display : :class:`~sklearn.inspection.DecisionBoundaryDisplay`
Object that stores the result.
See Also
--------
DecisionBoundaryDisplay : Decision boundary visualization.
ConfusionMatrixDisplay.from_estimator : Plot the confusion matrix
given an estimator, the data, and the label.
ConfusionMatrixDisplay.from_predictions : Plot the confusion matrix
given the true and predicted labels.
Examples
--------
>>> import matplotlib.pyplot as plt
>>> from sklearn.datasets import load_iris
>>> from sklearn.linear_model import LogisticRegression
>>> from sklearn.inspection import DecisionBoundaryDisplay
>>> iris = load_iris()
>>> X = iris.data[:, :2]
>>> classifier = LogisticRegression().fit(X, iris.target)
>>> disp = DecisionBoundaryDisplay.from_estimator(
... classifier, X, response_method="predict",
... xlabel=iris.feature_names[0], ylabel=iris.feature_names[1],
... alpha=0.5,
... )
>>> disp.ax_.scatter(X[:, 0], X[:, 1], c=iris.target, edgecolor="k")
<...>
>>> plt.show()
"""
check_matplotlib_support(f"{cls.__name__}.from_estimator")
check_is_fitted(estimator)
if not grid_resolution > 1:
raise ValueError(
"grid_resolution must be greater than 1. Got"
f" {grid_resolution} instead."
)
if not eps >= 0:
raise ValueError(
f"eps must be greater than or equal to 0. Got {eps} instead."
)
possible_plot_methods = ("contourf", "contour", "pcolormesh")
if plot_method not in possible_plot_methods:
available_methods = ", ".join(possible_plot_methods)
raise ValueError(
f"plot_method must be one of {available_methods}. "
f"Got {plot_method} instead."
)
num_features = _num_features(X)
if num_features != 2:
raise ValueError(
f"n_features must be equal to 2. Got {num_features} instead."
)
x0, x1 = _safe_indexing(X, 0, axis=1), _safe_indexing(X, 1, axis=1)
x0_min, x0_max = x0.min() - eps, x0.max() + eps
x1_min, x1_max = x1.min() - eps, x1.max() + eps
xx0, xx1 = np.meshgrid(
np.linspace(x0_min, x0_max, grid_resolution),
np.linspace(x1_min, x1_max, grid_resolution),
)
if hasattr(X, "iloc"):
# we need to preserve the feature names and therefore get an empty dataframe
X_grid = X.iloc[[], :].copy()
X_grid.iloc[:, 0] = xx0.ravel()
X_grid.iloc[:, 1] = xx1.ravel()
else:
X_grid = np.c_[xx0.ravel(), xx1.ravel()]
pred_func = _check_boundary_response_method(estimator, response_method)
response = pred_func(X_grid)
# convert classes predictions into integers
if pred_func.__name__ == "predict" and hasattr(estimator, "classes_"):
encoder = LabelEncoder()
encoder.classes_ = estimator.classes_
response = encoder.transform(response)
if response.ndim != 1:
if is_regressor(estimator):
raise ValueError("Multi-output regressors are not supported")
# TODO: Support pos_label
response = response[:, 1]
if xlabel is None:
xlabel = X.columns[0] if hasattr(X, "columns") else ""
if ylabel is None:
ylabel = X.columns[1] if hasattr(X, "columns") else ""
display = DecisionBoundaryDisplay(
xx0=xx0,
xx1=xx1,
response=response.reshape(xx0.shape),
xlabel=xlabel,
ylabel=ylabel,
)
return display.plot(ax=ax, plot_method=plot_method, **kwargs)