-
Notifications
You must be signed in to change notification settings - Fork 1
/
WANT
85 lines (43 loc) · 1.76 KB
/
WANT
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
Some new tags and taggers:
ALittleBelowWanted (i.e. not via OoM)
FarBelowWanted
BelowWanted (superclass)
DistanceToWanted
AboveWanted
ALittleAboveWanted
FarAboveWanted: should make the tagged node less visible to search.
Semaphore:
No more Avail Numbers.
Diagnosis:
TooHigh
Repair:
Try changing Times to Plus.
Try changing Plus to Times.
TryASmallerOperand
TryABiggerOperand
JustAbandonThis
Smaller steps:
NoticeCouldMakePlus, etc. -> build the Plus but don't work out the
result. Let that be estimated before calculated.
CalculateExactResult
To make it easier to read the log:
Proposal(ConsumeOperands()) should show the operands and operator.
pg(g, Tagged(Avail)): show the tagged nodes, not the Avail tags.
AssessProposal: print the numerical assessment.
'succeeded': show what was built, or anything else of note.
Somehow return an annotation, only from nodes that do something big.
Or perhaps have multiple levels of annotation. Or maybe 'annotation'
is the highest level.
To make it easier to define tags and taggers:
@classmethod
Tag.tagger()
('xyz', LookFor(...)) Sets asgn_to='xyz'. This should make 'acs' lists
more readable. ('acs' lists should actually be tuples, because they
should never be modified. But would this foul up as_iter()?)
Wanted in a graphical visualization:
A pane that shows all the active nodes or actions collected, sorted however
you like.
A pane with a running graph of the activation and support levels of
selected nodes.
The ability to scroll panes back to previous timesteps.
Want to fix or understand these current perplexities in the model: