-
Notifications
You must be signed in to change notification settings - Fork 1.6k
/
Copy pathtest_target_space.py
311 lines (235 loc) · 10.3 KB
/
test_target_space.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
from __future__ import annotations
import numpy as np
import pytest
from bayes_opt.constraint import ConstraintModel
from bayes_opt.exception import NotUniqueError
from bayes_opt.target_space import TargetSpace
def target_func(**kwargs):
# arbitrary target func
return sum(kwargs.values())
PBOUNDS = {"p1": (0, 1), "p2": (1, 100)}
def test_keys_and_bounds_in_same_order():
pbounds = {"p1": (0, 1), "p3": (0, 3), "p2": (0, 2), "p4": (0, 4)}
space = TargetSpace(target_func, pbounds)
assert space.dim == len(pbounds)
assert space.empty
assert space.keys == ["p1", "p3", "p2", "p4"]
assert all(space.bounds[:, 0] == np.array([0, 0, 0, 0]))
assert all(space.bounds[:, 1] == np.array([1, 3, 2, 4]))
def test_params_to_array():
space = TargetSpace(target_func, PBOUNDS)
assert all(space.params_to_array({"p1": 2, "p2": 3}) == np.array([2, 3]))
assert all(space.params_to_array({"p2": 2, "p1": 9}) == np.array([9, 2]))
with pytest.raises(ValueError):
space.params_to_array({"p2": 1})
with pytest.raises(ValueError):
space.params_to_array({"p2": 1, "p1": 7, "other": 4})
with pytest.raises(ValueError):
space.params_to_array({"other": 1})
def test_array_to_params():
space = TargetSpace(target_func, PBOUNDS)
assert space.array_to_params(np.array([2, 3])) == {"p1": 2, "p2": 3}
with pytest.raises(ValueError):
space.array_to_params(np.array([2]))
with pytest.raises(ValueError):
space.array_to_params(np.array([2, 3, 5]))
def test_to_float():
space = TargetSpace(target_func, PBOUNDS)
x = space._to_float({"p2": 0, "p1": 1})
assert x.shape == (2,)
assert all(x == np.array([1, 0]))
with pytest.raises(ValueError):
x = space._to_float([0, 1])
with pytest.raises(ValueError):
x = space._to_float([2, 1, 7])
with pytest.raises(ValueError):
x = space._to_float({"p2": 1, "p1": 2, "other": 7})
with pytest.raises(ValueError):
x = space._to_float({"p2": 1})
with pytest.raises(ValueError):
x = space._to_float({"other": 7})
def test_register():
PBOUNDS = {"p1": (0, 10), "p2": (1, 100)}
space = TargetSpace(target_func, PBOUNDS)
assert len(space) == 0
# registering with dict
space.register(params={"p1": 1, "p2": 2}, target=3)
assert len(space) == 1
assert all(space.params[0] == np.array([1, 2]))
assert all(space.target == np.array([3]))
# registering with dict out of order
space.register(params={"p2": 4, "p1": 5}, target=9)
assert len(space) == 2
assert all(space.params[1] == np.array([5, 4]))
assert all(space.target == np.array([3, 9]))
# registering with array
space.register(params=np.array([0, 1]), target=1)
assert len(space) == 3
assert all(space.params[2] == np.array([0, 1]))
assert all(space.target == np.array([3, 9, 1]))
with pytest.raises(NotUniqueError):
space.register(params={"p1": 1, "p2": 2}, target=3)
with pytest.raises(NotUniqueError):
space.register(params={"p1": 5, "p2": 4}, target=9)
def test_register_with_constraint():
constraint = ConstraintModel(lambda x: x, -2, 2, transform=lambda x: x)
space = TargetSpace(target_func, PBOUNDS, constraint=constraint)
assert len(space) == 0
# registering with dict
space.register(params={"p1": 1, "p2": 2}, target=3, constraint_value=0.0)
assert len(space) == 1
assert all(space.params[0] == np.array([1, 2]))
assert all(space.target == np.array([3]))
assert all(space.constraint_values == np.array([0]))
# registering with array
space.register(params={"p1": 0.5, "p2": 4}, target=4.5, constraint_value=2)
assert len(space) == 2
assert all(space.params[1] == np.array([0.5, 4]))
assert all(space.target == np.array([3, 4.5]))
assert all(space.constraint_values == np.array([0, 2]))
with pytest.raises(ValueError):
space.register(params={"p1": 0.2, "p2": 2}, target=2.2)
def test_register_point_beyond_bounds():
PBOUNDS = {"p1": (0, 1), "p2": (1, 10)}
space = TargetSpace(target_func, PBOUNDS)
with pytest.warns(UserWarning):
space.register(params={"p1": 0.5, "p2": 20}, target=2.5)
def test_probe():
PBOUNDS = {"p1": (0, 10), "p2": (1, 100)}
space = TargetSpace(target_func, PBOUNDS, allow_duplicate_points=True)
assert len(space) == 0
# probing with dict
space.probe(params={"p1": 1, "p2": 2})
assert len(space) == 1
assert all(space.params[-1] == np.array([1, 2]))
assert all(space.target == np.array([3]))
# probing with array
space.probe(np.array([5, 4]))
assert len(space) == 2
assert all(space.params[1] == np.array([5, 4]))
assert all(space.target == np.array([3, 9]))
# probing same point with dict
space.probe(params={"p1": 1, "p2": 2})
assert len(space) == 3
assert all(space.params[2] == np.array([1, 2]))
assert all(space.target == np.array([3, 9, 3]))
# probing same point with array
space.probe(np.array([5, 4]))
assert len(space) == 4
assert all(space.params[1] == np.array([5, 4]))
assert all(space.target == np.array([3, 9, 3, 9]))
space = TargetSpace(target_func, PBOUNDS, allow_duplicate_points=False)
# register wrong target to check probe doesn't recompute a duplicate point
space.register(params={"p1": 1, "p2": 2}, target=5)
# probing same point with dict
target_ = space.probe(params={"p1": 1, "p2": 2})
assert target_ == 5
# probing same point with array
target_ = space.probe(np.array([1, 2]))
assert target_ == 5
def test_random_sample():
pbounds = {"p1": (0, 1), "p3": (0, 3), "p2": (0, 2), "p4": (0, 4)}
space = TargetSpace(target_func, pbounds, random_state=8)
for _ in range(50):
random_sample = space.random_sample()
assert len(random_sample) == space.dim
assert all(random_sample >= space.bounds[:, 0])
assert all(random_sample <= space.bounds[:, 1])
def test_y_max():
space = TargetSpace(target_func, PBOUNDS)
assert space._target_max() is None
space.probe(params={"p1": 1, "p2": 7})
space.probe(params={"p1": 0.5, "p2": 1})
space.probe(params={"p1": 0, "p2": 1})
assert space._target_max() == 8
def test_y_max_with_constraint():
PBOUNDS = {"p1": (0, 10), "p2": (1, 100)}
constraint = ConstraintModel(lambda p1, p2: p1 - p2, -2, 2)
space = TargetSpace(target_func, PBOUNDS, constraint)
assert space._target_max() is None
space.probe(params={"p1": 1, "p2": 2}) # Feasible
space.probe(params={"p1": 5, "p2": 1}) # Unfeasible
space.probe(params={"p1": 0, "p2": 1}) # Feasible
assert space._target_max() == 3
def test_y_max_within_pbounds():
PBOUNDS = {"p1": (0, 2), "p2": (1, 100)}
space = TargetSpace(target_func, PBOUNDS)
assert space._target_max() is None
space.probe(params={"p1": 1, "p2": 2})
space.probe(params={"p1": 0, "p2": 1})
with pytest.warns(UserWarning):
space.probe(params={"p1": 5, "p2": 1})
assert space._target_max() == 3
def test_max():
PBOUNDS = {"p1": (0, 10), "p2": (1, 100)}
space = TargetSpace(target_func, PBOUNDS)
assert space.max() is None
space.probe(params={"p1": 1, "p2": 2})
space.probe(params={"p1": 5, "p2": 4})
space.probe(params={"p1": 2, "p2": 3})
space.probe(params={"p1": 1, "p2": 6})
assert space.max() == {"params": {"p1": 5, "p2": 4}, "target": 9}
def test_max_with_constraint():
PBOUNDS = {"p1": (0, 10), "p2": (1, 100)}
constraint = ConstraintModel(lambda p1, p2: p1 - p2, -2, 2)
space = TargetSpace(target_func, PBOUNDS, constraint=constraint)
assert space.max() is None
space.probe(params={"p1": 1, "p2": 2}) # Feasible
space.probe(params={"p1": 5, "p2": 8}) # Unfeasible
space.probe(params={"p1": 2, "p2": 3}) # Feasible
space.probe(params={"p1": 1, "p2": 6}) # Unfeasible
assert space.max() == {"params": {"p1": 2, "p2": 3}, "target": 5, "constraint": -1}
def test_max_with_constraint_identical_target_value():
PBOUNDS = {"p1": (0, 10), "p2": (1, 100)}
constraint = ConstraintModel(lambda p1, p2: p1 - p2, -2, 2)
space = TargetSpace(target_func, PBOUNDS, constraint=constraint)
assert space.max() is None
space.probe(params={"p1": 1, "p2": 2}) # Feasible
space.probe(params={"p1": 0, "p2": 5}) # Unfeasible, target value is 5, should not be selected
space.probe(params={"p1": 5, "p2": 8}) # Unfeasible
space.probe(params={"p1": 2, "p2": 3}) # Feasible, target value is also 5
space.probe(params={"p1": 1, "p2": 6}) # Unfeasible
assert space.max() == {"params": {"p1": 2, "p2": 3}, "target": 5, "constraint": -1}
def test_res():
PBOUNDS = {"p1": (0, 10), "p2": (1, 100)}
space = TargetSpace(target_func, PBOUNDS)
assert space.res() == []
space.probe(params={"p1": 1, "p2": 2})
space.probe(params={"p1": 5, "p2": 4})
space.probe(params={"p1": 2, "p2": 3})
space.probe(params={"p1": 1, "p2": 6})
expected_res = [
{"params": {"p1": 1, "p2": 2}, "target": 3},
{"params": {"p1": 5, "p2": 4}, "target": 9},
{"params": {"p1": 2, "p2": 3}, "target": 5},
{"params": {"p1": 1, "p2": 6}, "target": 7},
]
assert len(space.res()) == 4
assert space.res() == expected_res
def test_set_bounds():
pbounds = {"p1": (0, 1), "p3": (0, 3), "p2": (0, 2), "p4": (0, 4)}
space = TargetSpace(target_func, pbounds)
# Ignore unknown keys
space.set_bounds({"other": (7, 8)})
assert all(space.bounds[:, 0] == np.array([0, 0, 0, 0]))
assert all(space.bounds[:, 1] == np.array([1, 3, 2, 4]))
# Update bounds accordingly
space.set_bounds({"p2": (1, 8)})
assert all(space.bounds[:, 0] == np.array([0, 0, 1, 0]))
assert all(space.bounds[:, 1] == np.array([1, 3, 8, 4]))
def test_no_target_func():
target_space = TargetSpace(None, PBOUNDS)
with pytest.raises(ValueError, match="No target function has been provided."):
target_space.probe({"p1": 1, "p2": 2})
def test_change_typed_bounds():
pbounds = {
"p1": (0, 1),
"p2": (1, 2),
"p3": (-1, 3, int),
"fruit": ("apple", "banana", "mango", "honeydew melon", "strawberry"),
}
space = TargetSpace(None, pbounds)
with pytest.raises(ValueError):
space.set_bounds({"fruit": ("apple", "banana", "mango", "honeydew melon")})
with pytest.raises(ValueError):
space.set_bounds({"p3": (-1, 2, float)})