Skip to content

Commit 187fd08

Browse files
till-mtill-m
committed
Save files while moving device
1 parent b765b5d commit 187fd08

File tree

7 files changed

+150
-176
lines changed

7 files changed

+150
-176
lines changed

‎bayes_opt/__init__.py‎

Lines changed: 0 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -9,7 +9,6 @@
99
from bayes_opt.constraint import ConstraintModel
1010
from bayes_opt.domain_reduction import SequentialDomainReductionTransformer
1111
from bayes_opt.logger import JSONLogger, ScreenLogger
12-
from bayes_opt.parameter import BayesParameter
1312
from bayes_opt.target_space import TargetSpace
1413

1514
__version__ = importlib.metadata.version("bayesian-optimization")
@@ -24,5 +23,4 @@
2423
"ScreenLogger",
2524
"JSONLogger",
2625
"SequentialDomainReductionTransformer",
27-
"BayesParameter",
2826
]

‎bayes_opt/parameter.py‎

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -477,7 +477,7 @@ def wrap_kernel(kernel: kernels.Kernel, transform: Callable[[Any], Any]) -> kern
477477
kernel_type = type(kernel)
478478

479479
class WrappedKernel(kernel_type):
480-
@copy_signature(getattr(kernel_type.__init__, "deprecated_original", kernel_type.__init__))
480+
@_copy_signature(getattr(kernel_type.__init__, "deprecated_original", kernel_type.__init__))
481481
def __init__(self, **kwargs: Any) -> None:
482482
super().__init__(**kwargs)
483483

@@ -492,8 +492,8 @@ def __reduce__(self) -> str | tuple[Any, ...]:
492492
return WrappedKernel(**kernel.get_params())
493493

494494

495-
def copy_signature(source_fct: Callable[..., Any]) -> Callable[[Callable[..., Any]], Callable[..., Any]]:
496-
"""Clones a signature from a source function to a target function.
495+
def _copy_signature(source_fct: Callable[..., Any]) -> Callable[[Callable[..., Any]], Callable[..., Any]]:
496+
"""Clone a signature from a source function to a target function.
497497
498498
via
499499
https://stackoverflow.com/a/58989918/

‎bayes_opt/target_space.py‎

Lines changed: 11 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -242,22 +242,33 @@ def make_params(self, pbounds: BoundsMapping) -> dict[str, BayesParameter]:
242242
A dictionary with the parameter names as keys and the corresponding
243243
parameter objects as values.
244244
"""
245+
any_is_not_float = False # TODO: remove in an upcoming release
245246
params: dict[str, BayesParameter] = {}
246247
for key in pbounds:
247248
pbound = pbounds[key]
248249

249250
if isinstance(pbound, BayesParameter):
250251
res = pbound
252+
if not isinstance(pbound, FloatParameter):
253+
any_is_not_float = True
251254
elif (len(pbound) == 2 and is_numeric(pbound[0]) and is_numeric(pbound[1])) or (
252255
len(pbound) == 3 and pbound[-1] is float
253256
):
254257
res = FloatParameter(name=key, bounds=(float(pbound[0]), float(pbound[1])))
255258
elif len(pbound) == 3 and pbound[-1] is int:
256259
res = IntParameter(name=key, bounds=(int(pbound[0]), int(pbound[1])))
260+
any_is_not_float = True
257261
else:
258262
# assume categorical variable with pbound as list of possible values
259263
res = CategoricalParameter(name=key, categories=pbound)
264+
any_is_not_float = True
260265
params[key] = res
266+
if any_is_not_float:
267+
msg = (
268+
"Non-float parameters are experimental and may not work as expected."
269+
" Exercise caution when using them and please report any issues you encounter."
270+
)
271+
warn(msg, stacklevel=4)
261272
return params
262273

263274
def make_masks(self) -> dict[str, NDArray[np.bool_]]:

‎docsrc/index.rst‎

Lines changed: 6 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -11,6 +11,7 @@
1111
Basic Tour </basic-tour>
1212
Advanced Tour </advanced-tour>
1313
Constrained Bayesian Optimization </constraints>
14+
Parameter Types </parameter_types>
1415
Sequential Domain Reduction </domain_reduction>
1516
Acquisition Functions </acquisition_functions>
1617
Exploration vs. Exploitation </exploitation_vs_exploration>
@@ -26,6 +27,7 @@
2627
reference/constraint
2728
reference/domain_reduction
2829
reference/target_space
30+
reference/parameter
2931
reference/exception
3032
reference/other
3133

@@ -121,11 +123,13 @@ section. We suggest that you:
121123
to learn how to use the package's most important features.
122124
- Take a look at the `advanced tour
123125
notebook <advanced-tour.html>`__
124-
to learn how to make the package more flexible, how to deal with
125-
categorical parameters, how to use observers, and more.
126+
to learn how to make the package more flexible or how to use observers.
126127
- To learn more about acquisition functions, a central building block
127128
of bayesian optimization, see the `acquisition functions
128129
notebook <acquisition_functions.html>`__
130+
- If you want to optimize over integer-valued or categorical
131+
parameters, see the `parameter types
132+
notebook <parameter_types.html>`__.
129133
- Check out this
130134
`notebook <visualization.html>`__
131135
with a step by step visualization of how this method works.

‎docsrc/reference/parameter.rst‎

Lines changed: 5 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,5 @@
1+
:py:mod:`bayes_opt.parameter`
2+
--------------------------------
3+
4+
.. automodule:: bayes_opt.parameter
5+
:members:

‎examples/advanced-tour.ipynb‎

Lines changed: 28 additions & 133 deletions
Original file line numberDiff line numberDiff line change
@@ -96,7 +96,7 @@
9696
"name": "stdout",
9797
"output_type": "stream",
9898
"text": [
99-
"Next point to probe is: {'x': -0.331911981189704, 'y': 1.3219469606529486}\n"
99+
"Next point to probe is: {'x': np.float64(-0.331911981189704), 'y': np.float64(1.3219469606529486)}\n"
100100
]
101101
}
102102
],
@@ -167,12 +167,12 @@
167167
"name": "stdout",
168168
"output_type": "stream",
169169
"text": [
170-
"-18.503835804889988 {'x': 1.953072105336, 'y': -2.9609778030491904}\n",
171-
"-1.0819533157901717 {'x': 0.22703572807626315, 'y': 2.4249238905875123}\n",
172-
"-6.50219704520679 {'x': -1.9991881984624875, 'y': 2.872282989383577}\n",
173-
"-5.747604713731052 {'x': -1.994467585936897, 'y': -0.664242699361514}\n",
174-
"-2.9682431497650823 {'x': 1.9737252084307952, 'y': 1.269540259274744}\n",
175-
"{'target': 0.7861845912690544, 'params': {'x': -0.331911981189704, 'y': 1.3219469606529486}}\n"
170+
"-18.707136686093495 {'x': np.float64(1.9261486197444082), 'y': np.float64(-2.9996360060323246)}\n",
171+
"0.750594563473972 {'x': np.float64(-0.3763326769822668), 'y': np.float64(1.328297354179696)}\n",
172+
"-6.559031075654336 {'x': np.float64(1.979183535803597), 'y': np.float64(2.9083667381450318)}\n",
173+
"-6.915481333972961 {'x': np.float64(-1.9686133847781613), 'y': np.float64(-1.009985740060171)}\n",
174+
"-6.8600832617014085 {'x': np.float64(-1.9763198875239296), 'y': np.float64(2.9885278383464513)}\n",
175+
"{'target': np.float64(0.7861845912690544), 'params': {'x': np.float64(-0.331911981189704), 'y': np.float64(1.3219469606529486)}}\n"
176176
]
177177
}
178178
],
@@ -190,112 +190,7 @@
190190
"cell_type": "markdown",
191191
"metadata": {},
192192
"source": [
193-
"## 2. Dealing with discrete parameters\n",
194-
"\n",
195-
"**There is no principled way of dealing with discrete parameters using this package.**\n",
196-
"\n",
197-
"Ok, now that we got that out of the way, how do you do it? You're bound to be in a situation where some of your function's parameters may only take on discrete values. Unfortunately, the nature of bayesian optimization with gaussian processes doesn't allow for an easy/intuitive way of dealing with discrete parameters - but that doesn't mean it is impossible. The example below showcases a simple, yet reasonably adequate, way to dealing with discrete parameters."
198-
]
199-
},
200-
{
201-
"cell_type": "code",
202-
"execution_count": 9,
203-
"metadata": {},
204-
"outputs": [],
205-
"source": [
206-
"def func_with_discrete_params(x, y, d):\n",
207-
" # Simulate necessity of having d being discrete.\n",
208-
" assert type(d) == int\n",
209-
" \n",
210-
" return ((x + y + d) // (1 + d)) / (1 + (x + y) ** 2)"
211-
]
212-
},
213-
{
214-
"cell_type": "code",
215-
"execution_count": 10,
216-
"metadata": {},
217-
"outputs": [],
218-
"source": [
219-
"def function_to_be_optimized(x, y, w):\n",
220-
" d = int(w)\n",
221-
" return func_with_discrete_params(x, y, d)"
222-
]
223-
},
224-
{
225-
"cell_type": "code",
226-
"execution_count": 11,
227-
"metadata": {},
228-
"outputs": [],
229-
"source": [
230-
"optimizer = BayesianOptimization(\n",
231-
" f=function_to_be_optimized,\n",
232-
" pbounds={'x': (-10, 10), 'y': (-10, 10), 'w': (0, 5)},\n",
233-
" verbose=2,\n",
234-
" random_state=1,\n",
235-
")"
236-
]
237-
},
238-
{
239-
"cell_type": "code",
240-
"execution_count": 12,
241-
"metadata": {},
242-
"outputs": [
243-
{
244-
"name": "stdout",
245-
"output_type": "stream",
246-
"text": [
247-
"| iter | target | w | x | y |\n",
248-
"-------------------------------------------------------------\n",
249-
"| \u001b[30m1 | \u001b[30m-0.06199 | \u001b[30m2.085 | \u001b[30m4.406 | \u001b[30m-9.998 |\n",
250-
"| \u001b[35m2 | \u001b[35m-0.0344 | \u001b[35m1.512 | \u001b[35m-7.065 | \u001b[35m-8.153 |\n",
251-
"| \u001b[30m3 | \u001b[30m-0.2177 | \u001b[30m0.9313 | \u001b[30m-3.089 | \u001b[30m-2.065 |\n",
252-
"| \u001b[35m4 | \u001b[35m0.1865 | \u001b[35m2.694 | \u001b[35m-1.616 | \u001b[35m3.704 |\n",
253-
"| \u001b[30m5 | \u001b[30m-0.2187 | \u001b[30m1.022 | \u001b[30m7.562 | \u001b[30m-9.452 |\n",
254-
"| \u001b[35m6 | \u001b[35m0.2488 | \u001b[35m2.684 | \u001b[35m-2.188 | \u001b[35m3.925 |\n"
255-
]
256-
},
257-
{
258-
"name": "stdout",
259-
"output_type": "stream",
260-
"text": [
261-
"| \u001b[35m7 | \u001b[35m0.2948 | \u001b[35m2.683 | \u001b[35m-2.534 | \u001b[35m4.08 |\n",
262-
"| \u001b[35m8 | \u001b[35m0.3202 | \u001b[35m2.514 | \u001b[35m-3.83 | \u001b[35m5.287 |\n",
263-
"| \u001b[30m9 | \u001b[30m0.0 | \u001b[30m4.057 | \u001b[30m-4.458 | \u001b[30m3.928 |\n",
264-
"| \u001b[35m10 | \u001b[35m0.4802 | \u001b[35m2.296 | \u001b[35m-3.518 | \u001b[35m4.558 |\n",
265-
"| \u001b[30m11 | \u001b[30m0.0 | \u001b[30m1.084 | \u001b[30m-3.737 | \u001b[30m4.472 |\n",
266-
"| \u001b[30m12 | \u001b[30m0.0 | \u001b[30m2.649 | \u001b[30m-3.861 | \u001b[30m4.353 |\n",
267-
"| \u001b[30m13 | \u001b[30m0.0 | \u001b[30m2.442 | \u001b[30m-3.658 | \u001b[30m4.599 |\n",
268-
"| \u001b[30m14 | \u001b[30m-0.05801 | \u001b[30m1.935 | \u001b[30m-0.4758 | \u001b[30m-8.755 |\n",
269-
"| \u001b[30m15 | \u001b[30m0.0 | \u001b[30m2.337 | \u001b[30m7.973 | \u001b[30m-8.96 |\n",
270-
"| \u001b[30m16 | \u001b[30m0.07699 | \u001b[30m0.6926 | \u001b[30m5.59 | \u001b[30m6.854 |\n",
271-
"| \u001b[30m17 | \u001b[30m-0.02025 | \u001b[30m3.534 | \u001b[30m-8.943 | \u001b[30m1.987 |\n",
272-
"| \u001b[30m18 | \u001b[30m0.0 | \u001b[30m2.59 | \u001b[30m-7.339 | \u001b[30m5.941 |\n",
273-
"| \u001b[30m19 | \u001b[30m0.0929 | \u001b[30m2.237 | \u001b[30m-4.535 | \u001b[30m9.065 |\n",
274-
"| \u001b[30m20 | \u001b[30m0.1538 | \u001b[30m0.477 | \u001b[30m2.931 | \u001b[30m2.683 |\n",
275-
"| \u001b[30m21 | \u001b[30m0.0 | \u001b[30m0.9999 | \u001b[30m4.397 | \u001b[30m-3.971 |\n",
276-
"| \u001b[30m22 | \u001b[30m-0.01894 | \u001b[30m3.764 | \u001b[30m-7.043 | \u001b[30m-3.184 |\n",
277-
"| \u001b[30m23 | \u001b[30m0.03683 | \u001b[30m1.851 | \u001b[30m5.783 | \u001b[30m7.966 |\n",
278-
"| \u001b[30m24 | \u001b[30m-0.04359 | \u001b[30m1.615 | \u001b[30m-5.133 | \u001b[30m-6.556 |\n",
279-
"| \u001b[30m25 | \u001b[30m0.02617 | \u001b[30m3.863 | \u001b[30m0.1052 | \u001b[30m8.579 |\n",
280-
"| \u001b[30m26 | \u001b[30m-0.1071 | \u001b[30m0.8131 | \u001b[30m-0.7949 | \u001b[30m-9.292 |\n",
281-
"| \u001b[30m27 | \u001b[30m0.0 | \u001b[30m4.969 | \u001b[30m8.778 | \u001b[30m-8.467 |\n",
282-
"| \u001b[30m28 | \u001b[30m-0.1372 | \u001b[30m0.9475 | \u001b[30m-1.019 | \u001b[30m-7.018 |\n",
283-
"| \u001b[30m29 | \u001b[30m0.08078 | \u001b[30m1.917 | \u001b[30m-0.2606 | \u001b[30m6.272 |\n",
284-
"| \u001b[30m30 | \u001b[30m0.02003 | \u001b[30m4.278 | \u001b[30m3.8 | \u001b[30m8.398 |\n",
285-
"=============================================================\n"
286-
]
287-
}
288-
],
289-
"source": [
290-
"optimizer.set_gp_params(alpha=1e-3)\n",
291-
"optimizer.maximize()"
292-
]
293-
},
294-
{
295-
"cell_type": "markdown",
296-
"metadata": {},
297-
"source": [
298-
"## 3. Tuning the underlying Gaussian Process\n",
193+
"## 2. Tuning the underlying Gaussian Process\n",
299194
"\n",
300195
"The bayesian optimization algorithm works by performing a gaussian process regression of the observed combination of parameters and their associated target values. The predicted parameter $\\rightarrow$ target hyper-surface (and its uncertainty) is then used to guide the next best point to probe."
301196
]
@@ -304,14 +199,14 @@
304199
"cell_type": "markdown",
305200
"metadata": {},
306201
"source": [
307-
"### 3.1 Passing parameter to the GP\n",
202+
"### 2.1 Passing parameter to the GP\n",
308203
"\n",
309204
"Depending on the problem it could be beneficial to change the default parameters of the underlying GP. You can use the `optimizer.set_gp_params` method to do this:"
310205
]
311206
},
312207
{
313208
"cell_type": "code",
314-
"execution_count": 13,
209+
"execution_count": 9,
315210
"metadata": {},
316211
"outputs": [
317212
{
@@ -320,12 +215,12 @@
320215
"text": [
321216
"| iter | target | x | y |\n",
322217
"-------------------------------------------------\n",
323-
"| \u001b[30m1 | \u001b[30m0.7862 | \u001b[30m-0.3319 | \u001b[30m1.322 |\n",
324-
"| \u001b[30m2 | \u001b[30m-18.19 | \u001b[30m1.957 | \u001b[30m-2.919 |\n",
325-
"| \u001b[30m3 | \u001b[30m-12.05 | \u001b[30m-1.969 | \u001b[30m-2.029 |\n",
326-
"| \u001b[30m4 | \u001b[30m-7.463 | \u001b[30m0.6032 | \u001b[30m-1.846 |\n",
327-
"| \u001b[30m5 | \u001b[30m-1.093 | \u001b[30m1.444 | \u001b[30m1.096 |\n",
328-
"| \u001b[35m6 | \u001b[35m0.8586 | \u001b[35m-0.2165 | \u001b[35m1.307 |\n",
218+
"| \u001b[39m1 \u001b[39m | \u001b[39m0.7862 \u001b[39m | \u001b[39m-0.331911\u001b[39m | \u001b[39m1.3219469\u001b[39m |\n",
219+
"| \u001b[39m2 \u001b[39m | \u001b[39m-18.34 \u001b[39m | \u001b[39m1.9021640\u001b[39m | \u001b[39m-2.965222\u001b[39m |\n",
220+
"| \u001b[35m3 \u001b[39m | \u001b[35m0.8731 \u001b[39m | \u001b[35m-0.298167\u001b[39m | \u001b[35m1.1948749\u001b[39m |\n",
221+
"| \u001b[39m4 \u001b[39m | \u001b[39m-6.497 \u001b[39m | \u001b[39m1.9876938\u001b[39m | \u001b[39m2.8830942\u001b[39m |\n",
222+
"| \u001b[39m5 \u001b[39m | \u001b[39m-4.286 \u001b[39m | \u001b[39m-1.995643\u001b[39m | \u001b[39m-0.141769\u001b[39m |\n",
223+
"| \u001b[39m6 \u001b[39m | \u001b[39m-6.781 \u001b[39m | \u001b[39m-1.953302\u001b[39m | \u001b[39m2.9913127\u001b[39m |\n",
329224
"=================================================\n"
330225
]
331226
}
@@ -348,7 +243,7 @@
348243
"cell_type": "markdown",
349244
"metadata": {},
350245
"source": [
351-
"### 3.2 Tuning the `alpha` parameter\n",
246+
"### 2.2 Tuning the `alpha` parameter\n",
352247
"\n",
353248
"When dealing with functions with discrete parameters,or particularly erratic target space it might be beneficial to increase the value of the `alpha` parameter. This parameters controls how much noise the GP can handle, so increase it whenever you think that extra flexibility is needed."
354249
]
@@ -358,7 +253,7 @@
358253
"cell_type": "markdown",
359254
"metadata": {},
360255
"source": [
361-
"### 3.3 Changing kernels\n",
256+
"### 2.3 Changing kernels\n",
362257
"\n",
363258
"By default this package uses the Matern 2.5 kernel. Depending on your use case you may find that tuning the GP kernel could be beneficial. You're on your own here since these are very specific solutions to very specific problems. You should start with the [scikit learn docs](https://scikit-learn.org/stable/modules/gaussian_process.html#kernels-for-gaussian-processes)."
364259
]
@@ -376,7 +271,7 @@
376271
},
377272
{
378273
"cell_type": "code",
379-
"execution_count": 14,
274+
"execution_count": 10,
380275
"metadata": {},
381276
"outputs": [],
382277
"source": [
@@ -385,7 +280,7 @@
385280
},
386281
{
387282
"cell_type": "code",
388-
"execution_count": 15,
283+
"execution_count": 11,
389284
"metadata": {},
390285
"outputs": [],
391286
"source": [
@@ -399,7 +294,7 @@
399294
},
400295
{
401296
"cell_type": "code",
402-
"execution_count": 16,
297+
"execution_count": 12,
403298
"metadata": {},
404299
"outputs": [],
405300
"source": [
@@ -411,7 +306,7 @@
411306
},
412307
{
413308
"cell_type": "code",
414-
"execution_count": 17,
309+
"execution_count": 13,
415310
"metadata": {},
416311
"outputs": [],
417312
"source": [
@@ -433,7 +328,7 @@
433328
},
434329
{
435330
"cell_type": "code",
436-
"execution_count": 18,
331+
"execution_count": 14,
437332
"metadata": {},
438333
"outputs": [],
439334
"source": [
@@ -449,7 +344,7 @@
449344
},
450345
{
451346
"cell_type": "code",
452-
"execution_count": 19,
347+
"execution_count": 15,
453348
"metadata": {},
454349
"outputs": [
455350
{
@@ -476,7 +371,7 @@
476371
},
477372
{
478373
"cell_type": "code",
479-
"execution_count": 20,
374+
"execution_count": 16,
480375
"metadata": {},
481376
"outputs": [
482377
{
@@ -485,7 +380,7 @@
485380
"['optimization:start', 'optimization:step', 'optimization:end']"
486381
]
487382
},
488-
"execution_count": 20,
383+
"execution_count": 16,
489384
"metadata": {},
490385
"output_type": "execute_result"
491386
}
@@ -497,7 +392,7 @@
497392
],
498393
"metadata": {
499394
"kernelspec": {
500-
"display_name": "Python 3 (ipykernel)",
395+
"display_name": "bayesian-optimization-t6LLJ9me-py3.10",
501396
"language": "python",
502397
"name": "python3"
503398
},
@@ -511,7 +406,7 @@
511406
"name": "python",
512407
"nbconvert_exporter": "python",
513408
"pygments_lexer": "ipython3",
514-
"version": "3.1.undefined"
409+
"version": "3.10.13"
515410
},
516411
"nbdime-conflicts": {
517412
"local_diff": [

0 commit comments

Comments
 (0)