@@ -35,7 +35,20 @@ the constraints are currently silently ignored.
35
35
36
36
(CLC)=
37
37
### ContinuousLinearConstraint
38
+
38
39
The [ ` ContinuousLinearConstraint ` ] ( baybe.constraints.continuous.ContinuousLinearConstraint )
40
+ is used to model linear relationship between between continuous parameters.
41
+ BayBE support two different kinds of continuous linear constraints:
42
+ 1 . ** Intrapoint constraints:** These constraints model the relationship between the parameters
43
+ within (* intra-* ) individual experiments (* -points* ) and are what people typically think of when
44
+ simply speaking about "constraints". For that reason, a ` ContinuousLinearConstraint ` is
45
+ interpreted as an intrapoint constraint if not explicitly declared otherwise.
46
+ 2 . ** Interpoint constraints:** These constraints model the relationship of different experiments
47
+ across the whole batch (* inter-* ) of exeriments (* -points* ).
48
+
49
+ #### Intrapoint constraints
50
+ When not explicitly declaring it an interpoint constraint, the
51
+ [ ` ContinuousLinearConstraint ` ] ( baybe.constraints.continuous.ContinuousLinearConstraint )
39
52
asserts that the following kind of equations are true (up to numerical rounding errors):
40
53
41
54
$$
@@ -82,20 +95,23 @@ ContinuousLinearConstraint(
82
95
A more detailed example can be found
83
96
[ here] ( ../../examples/Constraints_Continuous/linear_constraints ) .
84
97
85
- ### Continuous Interpoint Constraints
98
+ #### Interpoint constraints
86
99
87
- The constraints discussed in the previous paragraph belong to the class of so-called
88
- "intrapoint constraints". That is, they impose conditions on each individual point of a batch.
89
- In contrast to this, interpoint constraints do so ** across** the points of the batch.
90
- That is, an interpoint constraint of the form `` x + y <= 1 `` technically encodes
100
+ In contrast to the intrapoint constrains discussed in the previous paragraph,
101
+ interpoint constraints models constraints ** across** the points of the batch.
102
+ That is, an interpoint constraint of the form $x + y <= 1$ encodes
91
103
$$
92
104
\sum_{i\text{ in batch}}x_i + \sum_{i\text{ in batch}}y_i <= 1
93
105
$$
106
+
107
+ They can be defined by using the ` interpoint ` argument of the
108
+ [ ` ContinuousLinearConstraint ` ] ( baybe.constraints.continuous.ContinuousLinearConstraint )
109
+ class.
110
+
94
111
A possible relevant constraint might be that only 100ml of a given solvent are available for
95
112
a full batch.
96
113
97
- They can be defined by using the ` interpoint ` argument of the [ ` ContinuousLinearConstraint ` ] ( baybe.constraints.continuous.ContinuousLinearConstraint )
98
- class.
114
+
99
115
``` python
100
116
from baybe.constraints import ContinuousLinearConstraint
101
117
0 commit comments