You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Commit bfe3f2f
Browse filesBrowse the repository at this point in the historyBrowse files
Copy file name to clipboardExpand all lines: .zdev/open-search/TASKS.md
+3-3Lines changed: 3 additions & 3 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -3,9 +3,9 @@
3
3
# Tasks: open-search
4
4
5
5
- Total: 45
6
-
- Ready: 1
6
+
- Ready: 0
7
7
- Blocked: 0
8
-
- Done: 44
8
+
- Done: 45
9
9
10
10
| ID | Task | State | Blocked by |
11
11
| --- | --- | --- | --- |
@@ -53,4 +53,4 @@
53
53
|[open-search-042](tasks/042-simplify-goal-handling-in-the-parent-15-puzzle-example.md)| Simplify goal handling in the parent 15-puzzle example | done | — |
54
54
|[open-search-043](tasks/043-use-a-fixed-buffer-for-puzzle-neighbors.md)| Use IntArgs for puzzle neighbors | done | — |
55
55
|[open-search-044](tasks/044-split-puzzle-goal-implications-with-a-shared-boolvar.md)| Split puzzle goal implications with a shared BoolVar | done | — |
56
-
|[open-search-045](tasks/045-offer-manhattan-value-selection-for-the-puzzle.md)| Offer Manhattan value selection for the puzzle |ready| — |
56
+
|[open-search-045](tasks/045-offer-manhattan-value-selection-for-the-puzzle.md)| Offer Manhattan value selection for the puzzle |done| — |
Copy file name to clipboardExpand all lines: .zdev/open-search/tasks/045-offer-manhattan-value-selection-for-the-puzzle.md
+14-4Lines changed: 14 additions & 4 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -3,7 +3,7 @@ schema_version = 1
3
3
id = "open-search-045"
4
4
key = "puzzle-manhattan-values"
5
5
area = "open-search"
6
-
status = "open"
6
+
status = "done"
7
7
complexity = "standard"
8
8
afk = true
9
9
blocked_by = []
@@ -25,12 +25,22 @@ The user asks what the fixed-model branch filter does and whether Manhattan valu
25
25
26
26
## Done when
27
27
28
-
-[] Both open and fixed models expose -branching min and -branching manhattan; the latter chooses a legal next blank position minimizing resulting board Manhattan distance, with deterministic ties.
29
-
-[] Explain the fixed-model filter and compare node/failure/propagation statistics for both orders on identical bounded cases; all reported incumbents replay to the goal and final optima agree.
30
-
-[] Normal parent and child builds and focused checks pass; parent source commit is incorporated into the child and user edits are preserved.
28
+
-[x] Both open and fixed models expose -branching min and -branching manhattan; the latter chooses a legal next blank position minimizing resulting board Manhattan distance, with deterministic ties.
29
+
-[x] Explain the fixed-model filter and compare node/failure/propagation statistics for both orders on identical bounded cases; all reported incumbents replay to the goal and final optima agree.
30
+
-[x] Normal parent and child builds and focused checks pass; parent source commit is incorporated into the child and user edits are preserved.
31
31
32
32
## Validation
33
33
34
34
- Build the parent and child fifteen-puzzle targets normally. Use bounded BFS/path-replay checks for both branching options and open/fixed models, including BAB and child A*/CABS.
35
35
- Measure deterministic statistics on a small fixed set with single-thread execution; bound total measurement solver time to 120 seconds, do not infer runtime gains from node counts.
Added optional Manhattan value selection to both puzzle models through ordinary INT_VAL callbacks, with numerical minimum unchanged as default. Parent commit 0d80c51e0b is incorporated into the search branch.
41
+
42
+
Validation:
43
+
44
+
- Both normal builds and 96 BFS/path replay checks passed; independent verification passed at W5519716bf265e4d8. Parent user edits are preserved; whitespace and selected clang-tidy checks passed.
45
+
- On scramble 12, seed 1, limit 12, one thread: nodes min to Manhattan are BAB 5426 to 11, A* 33 to 29, CABS 33 to 19 in both models. Open BAB failures 2712 to 2 and propagations 606170 to 3127.
46
+
- The 24-run comparison used 18.891 seconds of solver time. On scramble 18, seed 3, limit 18, CABS nodes increased 43176 to 50669 and A* changed 7985 to 7981; four BAB runs stopped without solutions at 3500ms. All 20 completed runs agreed on final optima, and every incumbent replayed to the goal. No permanent tests or local sanitizer work.
0 commit comments