Skip to content

Commit 930804e

Browse files
rkishoreclaude
andcommitted
Link the RemoveNonSpacingMarks report, dotnet/machinelearning#7728
Co-Authored-By: Claude Opus 5 (1M context) <[email protected]>
1 parent 2f85799 commit 930804e

2 files changed

Lines changed: 6 additions & 1 deletion

File tree

‎README.md‎

Lines changed: 3 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -154,6 +154,9 @@ that `PREDICTIONS.md` was written for.
154154
- [dotnet/machinelearning#7728](https://github.com/dotnet/machinelearning/issues/7728):
155155
`RemoveNonSpacingMarks = true` fails to strip accents from precomposed characters under
156156
invariant globalization.
157+
- [dotnet/machinelearning#7728](https://github.com/dotnet/machinelearning/issues/7728):
158+
`RemoveNonSpacingMarks = true` fails to strip accents from precomposed characters under
159+
invariant globalization.
157160
- [elbruno/elbruno.localembeddings#56](https://github.com/elbruno/elbruno.localembeddings/issues/56):
158161
accented words become `[UNK]` with the default model.
159162
- [iyulab/lm-supply#12](https://github.com/iyulab/lm-supply/issues/12): embeddings diverge

‎RESULTS.md‎

Lines changed: 3 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -123,7 +123,9 @@ precomposed probe: the option silently does nothing, and nothing throws. It stil
123123
text that arrives already decomposed, because the mark test (`CharUnicodeInfo`) does not
124124
depend on ICU; only the decomposition does. The `accents` value, 0.333027, is the value
125125
measured for ElBruno and for SK under invariant mode. For SK this is consistent with the
126-
same mechanism, but SK's tokenizer was not read or instrumented here.
126+
same mechanism, but SK's tokenizer was not read or instrumented here. Reported as
127+
[dotnet/machinelearning#7728](https://github.com/dotnet/machinelearning/issues/7728) on
128+
2026-09-22.
127129

128130
| id | prediction | outcome |
129131
|---|---|---|

0 commit comments

Comments
 (0)