Sync every head count to the 190-head roster - #260
Merged
Merged
Conversation
The retrain in #258 took the aroma roster from 164 to 167, which moved three numbers that were quoted in nineteen places across the README, five docs, the workbench UI, the MCP server, the e2e conftest and the pipeline comments in predict.py and app.py. Several were already stale before this — the profile vector had been described as 170-dimensional since before mouthfeel shipped, and one workbench card claimed 167 heads in its title while still saying 164 survive in its body. The three canonical numbers, so future drift is easy to spot: 190 trained heads in all — 6 taste + 167 aroma + 5 mouthfeel + 12 safety 178 dimensions in the flavor-profile vector — taste + aroma + mouthfeel, tox deliberately excluded because safety is not a flavor-match dimension 167 aroma descriptor heads clearing CV-AUROC >= 0.70 (The loader reports 191 artifacts: the 190 heads plus the sweetness-intensity regressor, which is a regressor rather than a head and so is not counted as one.) Also corrects the aroma training corpus size in AROMA.md, which still read 1994 molecules against an actual 2403 after the supplement work. Signed-off-by: Austin L. <86896075+rvnminers-A-and-N@users.noreply.github.com>
rvnminers-A-and-N
added
area:data
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The retrain in #258 took the aroma roster 164 → 167, which moved three numbers quoted in nineteen places — README, five docs, the workbench UI, the MCP server, the e2e conftest, and the pipeline comments in
predict.py/app.py.Several were already stale before this: the profile vector had been described as 170-dimensional since before mouthfeel shipped, and one workbench card claimed 167 heads in its title while still saying "164 survive" in its body.
The three canonical numbers, stated once so future drift is easy to spot:
The loader reports 191 artifacts: the 190 heads plus the sweetness-intensity regressor, which is a regressor rather than a head and so isn't counted as one.
Also corrects the aroma corpus size in
AROMA.md— still read 1,994 molecules against an actual 2,403 after the supplement work.Verified live against the running app after restart:
/api/headsreturns 6 + 167 + 5 + 12 = 190, andblackberry/fig/magnoliaare serving.