This is an automated email from the ASF dual-hosted git repository.
krickert pushed a change to branch OPENNLP-1850-3-dl
in repository https://gitbox.apache.org/repos/asf/opennlp.git
omit 1cbc2680f OPENNLP-1850 Follow AlignedText.normalized() -> CharSequence
in NameFinderDL
omit 6b998e21a OPENNLP-1850 Review nits: extract testable DL guards;
merge-copy; capitalize msgs; migration note
omit d54b4a4ef OPENNLP-1850 Make mergeOverlappingSpans O(n log n) (dl)
omit 7ca47585b OPENNLP-1850 Reject non-finite logits in softmax, not just
NaN (dl)
omit d78377abd OPENNLP-1850 Fully-qualify TokenNameFinder javadoc links in
NameFinderDL
omit f089d377f OPENNLP-1850 Fail loud on corrupt document-classification
model output
omit 05f849699 OPENNLP-1850 Fail fast on null finder input; fix the GPU
eval test options
omit 679419da1 OPENNLP-1850 Harden fail-loud paths in the DL components
omit bb7c94a57 OPENNLP-1850 Add real-model chunk-boundary eval tests; drop
dead label constants
omit a148df281 OPENNLP-1850 Resolve overlapping chunk spans and compose the
input alignment
omit 911b1a224 OPENNLP-1850 Add OffsetMappingNameFinder capability
interface and a findInOriginal end-to-end test
omit 42ff35627 OPENNLP-1850 Offset-safe, Unicode-aware input normalization
in the DL components
omit aaffd6415 OPENNLP-1850 Review nits: rename
searchAnalyzer->matchingAnalyzer; drop 'search' framing in profile docs
omit 728739c2d OPENNLP-1850 Review nits: add Turkish profile; derive
coverage from the enum (profiles)
omit d3ef19ab0 OPENNLP-1850 Resolve Norwegian nb/nn to the Norwegian
profile (profiles)
omit c8b9f20ed OPENNLP-1850 Per-language NormalizationProfile registry (2c)
omit ba0e79a4f OPENNLP-1850 Review nits: TermAnalyzer javadoc references
matchingAnalyzer()
omit 4bb1a27e6 OPENNLP-1850 Review nits: rename dashes()->dash(); LEMMA
doc+test; soften forward-link (Term)
omit a5b78aa56 OPENNLP-1850 Layered Term model: Term, TermAnalyzer (2b)
omit 6de3f118c OPENNLP-1850 Review: uax29 javadoc pass, IAE guards on
WordTokenizer, loader consistency, new tests
omit 9a0df0ed1 OPENNLP-1850 Perf: hoist the per-char volatile reads in
WordBreakProperty/ExtendedPictographic
omit 78a588bfa OPENNLP-1850 Review: drop lazy-init justification comments
in WordBreakProperty/ExtendedPictographic
omit 0a0f54fb5 OPENNLP-1850 Review nits: ExtendedPictographic fail-loud
parity + doc; WordType heuristic note (tokenizer)
omit 6d0732a89 OPENNLP-1850 Fail loud on a Word_Break line missing its ';'
(tokenizer)
omit aab00d702 OPENNLP-1850 UAX #29 word tokenizer: WordSegmenter,
WordTokenizer, WordType (2a)
omit 01a836043 OPENNLP-1850 Review: IAE null contract, @ThreadSafe, UID
regeneration, line-break rung test
add f08942479 OPENNLP-1850 Review: IAE null contract, @ThreadSafe, UID
regeneration, line-break rung test
add aaf30b58f OPENNLP-1850 UAX #29 word tokenizer: WordSegmenter,
WordTokenizer, WordType (2a)
add 2ba9334c0 OPENNLP-1850 Fail loud on a Word_Break line missing its ';'
(tokenizer)
add d9867364d OPENNLP-1850 Review nits: ExtendedPictographic fail-loud
parity + doc; WordType heuristic note (tokenizer)
add e5d363b3c OPENNLP-1850 Review: drop lazy-init justification comments
in WordBreakProperty/ExtendedPictographic
add df748dbd4 OPENNLP-1850 Perf: hoist the per-char volatile reads in
WordBreakProperty/ExtendedPictographic
add 4a95754bb OPENNLP-1850 Review: uax29 javadoc pass, IAE guards on
WordTokenizer, loader consistency, new tests
add a99cf0bcb OPENNLP-1850 Layered Term model: Term, TermAnalyzer (2b)
add 3f37672d6 OPENNLP-1850 Review nits: rename dashes()->dash(); LEMMA
doc+test; soften forward-link (Term)
add 9aba1baf4 OPENNLP-1850 Review nits: TermAnalyzer javadoc references
matchingAnalyzer()
add 27cfcde45 OPENNLP-1850 Per-language NormalizationProfile registry (2c)
add f622ff5e5 OPENNLP-1850 Resolve Norwegian nb/nn to the Norwegian
profile (profiles)
add 4bf1071d7 OPENNLP-1850 Review nits: add Turkish profile; derive
coverage from the enum (profiles)
add bd96a7205 OPENNLP-1850 Review nits: rename
searchAnalyzer->matchingAnalyzer; drop 'search' framing in profile docs
add f0b2daac0 OPENNLP-1850 Offset-safe, Unicode-aware input normalization
in the DL components
add d9e766497 OPENNLP-1850 Add OffsetMappingNameFinder capability
interface and a findInOriginal end-to-end test
add f4da92694 OPENNLP-1850 Resolve overlapping chunk spans and compose the
input alignment
add b83d34cd5 OPENNLP-1850 Add real-model chunk-boundary eval tests; drop
dead label constants
add ad310a0fc OPENNLP-1850 Harden fail-loud paths in the DL components
add 8c42dea18 OPENNLP-1850 Fail fast on null finder input; fix the GPU
eval test options
add cf179dc92 OPENNLP-1850 Fail loud on corrupt document-classification
model output
add 3121003fc OPENNLP-1850 Fully-qualify TokenNameFinder javadoc links in
NameFinderDL
add 86d95d22a OPENNLP-1850 Reject non-finite logits in softmax, not just
NaN (dl)
add 1196af51e OPENNLP-1850 Make mergeOverlappingSpans O(n log n) (dl)
add 9bdd38ba2 OPENNLP-1850 Review nits: extract testable DL guards;
merge-copy; capitalize msgs; migration note
add 03912d721 OPENNLP-1850 Follow AlignedText.normalized() -> CharSequence
in NameFinderDL
This update added new revisions after undoing existing revisions.
That is to say, some revisions that were in the old version of the
branch are not in the new version. This situation occurs
when a user --force pushes a change and generates a repository
containing something like this:
* -- * -- B -- O -- O -- O (1cbc2680f)
\
N -- N -- N refs/heads/OPENNLP-1850-3-dl (03912d721)
You should already have received notification emails for all of the O
revisions, and so the following emails describe only the N revisions
from the common base, B.
Any revisions marked "omit" are not gone; other references still
refer to them. Any revisions marked "discard" are gone forever.
No new revisions were added by this update.
Summary of changes:
