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 f3f545a2d OPENNLP-1850 DL review nits: document arg-validation
contract, javadoc locate(), hoist span-ordering constant
omit 5e00d4d8d OPENNLP-1850 DL hardening: IllegalArgumentException null
contract and null-element guards
omit 20520ec9c OPENNLP-1850 Follow AlignedText.normalized() -> CharSequence
in NameFinderDL
omit 1e3ef2c06 OPENNLP-1850 Review nits: extract testable DL guards;
merge-copy; capitalize msgs; migration note
omit 29a83c891 OPENNLP-1850 Make mergeOverlappingSpans O(n log n) (dl)
omit aee7377e4 OPENNLP-1850 Reject non-finite logits in softmax, not just
NaN (dl)
omit 73d3b2a54 OPENNLP-1850 Fully-qualify TokenNameFinder javadoc links in
NameFinderDL
omit 013254420 OPENNLP-1850 Fail loud on corrupt document-classification
model output
omit 39615df87 OPENNLP-1850 Fail fast on null finder input; fix the GPU
eval test options
omit 094c1d4f9 OPENNLP-1850 Harden fail-loud paths in the DL components
omit 0165d1f27 OPENNLP-1850 Add real-model chunk-boundary eval tests; drop
dead label constants
omit 9d8f948a0 OPENNLP-1850 Resolve overlapping chunk spans and compose the
input alignment
omit 9542aa3d1 OPENNLP-1850 Add OffsetMappingNameFinder capability
interface and a findInOriginal end-to-end test
omit 9512a32d7 OPENNLP-1850 Offset-safe, Unicode-aware input normalization
in the DL components
add d5d37dca2 OPENNLP-205: Refactor the end-of-sentence position to span
mapping in SentenceDetectorME (#1141)
add b42de7b68 OPENNLP-1850 Offset-safe, Unicode-aware input normalization
in the DL components
add c96cb1cca OPENNLP-1850 Add OffsetMappingNameFinder capability
interface and a findInOriginal end-to-end test
add 7b326c45e OPENNLP-1850 Resolve overlapping chunk spans and compose the
input alignment
add 44f54e9b8 OPENNLP-1850 Add real-model chunk-boundary eval tests; drop
dead label constants
add b03e39224 OPENNLP-1850 Harden fail-loud paths in the DL components
add 22c3e1825 OPENNLP-1850 Fail fast on null finder input; fix the GPU
eval test options
add b9abbcc03 OPENNLP-1850 Fail loud on corrupt document-classification
model output
add 45175b917 OPENNLP-1850 Fully-qualify TokenNameFinder javadoc links in
NameFinderDL
add 1aca3cb6e OPENNLP-1850 Reject non-finite logits in softmax, not just
NaN (dl)
add 81816d194 OPENNLP-1850 Make mergeOverlappingSpans O(n log n) (dl)
add 8d810786c OPENNLP-1850 Review nits: extract testable DL guards;
merge-copy; capitalize msgs; migration note
add 5ed944991 OPENNLP-1850 Follow AlignedText.normalized() -> CharSequence
in NameFinderDL
add f260fe712 OPENNLP-1850 DL hardening: IllegalArgumentException null
contract and null-element guards
add e23658141 OPENNLP-1850 DL review nits: document arg-validation
contract, javadoc locate(), hoist span-ordering constant
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 (f3f545a2d)
\
N -- N -- N refs/heads/OPENNLP-1850-3-dl (e23658141)
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:
.../tools/sentdetect/SentenceDetectorME.java | 161 ++++++-----
.../SentenceDetectorMESpanMappingTest.java | 308 +++++++++++++++++++++
2 files changed, 404 insertions(+), 65 deletions(-)
create mode 100644
opennlp-core/opennlp-runtime/src/test/java/opennlp/tools/sentdetect/SentenceDetectorMESpanMappingTest.java
