@camstack/develop 0.1.6 → 0.1.7
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/package.json +1 -1
- package/skills/track-review.md +16 -0
package/package.json
CHANGED
package/skills/track-review.md
CHANGED
|
@@ -122,6 +122,22 @@ fails does not stop the others, and an authorisation failure stops the run and
|
|
|
122
122
|
shows the hub's message. It returns a per-track log and the corpus KPI
|
|
123
123
|
(total / annotated / complete frames) before and after; quote both.
|
|
124
124
|
|
|
125
|
+
A box taken from the pipeline is saved as the pipeline's (`source: 'pipeline'`,
|
|
126
|
+
with its model, score and the `origin` it was offered as), so the hub can tell
|
|
127
|
+
"kept as drawn" from "relabelled". The save names the frame revision apply read;
|
|
128
|
+
if the operator edited the same frame in the admin page meanwhile, the hub
|
|
129
|
+
refuses it (`retrain-revision-conflict`) and that track is reported `failed` —
|
|
130
|
+
re-plan it rather than forcing it.
|
|
131
|
+
|
|
132
|
+
### Annotations saved before origins existed
|
|
133
|
+
|
|
134
|
+
`track_review_backfill_origin { collectionName | frameIds }` is a **dry-run**: per
|
|
135
|
+
frame it lists the rows that would get an origin (an EXACT rectangle match against
|
|
136
|
+
the pipeline box the hub re-derives), the rows that stay unmatched, and the rows
|
|
137
|
+
that already have one. Show that to the operator; only then repeat it with
|
|
138
|
+
`confirm: true`. It writes the origin column only and never overwrites an existing
|
|
139
|
+
origin. Rows left unmatched keep an approximate review verdict.
|
|
140
|
+
|
|
125
141
|
## Done when
|
|
126
142
|
|
|
127
143
|
- every suspect and every control track has a verdict;
|