Kimirock/PaddleNLP

Files

History

sneaxiy 206403950e Cherry-pick PRs from incubate/paddlenlp_fleety (#11049 )

* add offload opt util (#10607)

* fix offload optimizer (#10608)

* fix sharding reshard bug (#10613)

* reset ema (#10790)

* Move fused_quanted_ops and token_dispatcher_utils to FleetY (#10803)

* move setup_fp8.py (#10820)

* Add tokens_zip_unique_add_subbatch and merge_subbatch_cast ops (#10822)

* add tokens_zip_unique_add_subbatch and merge_subbatch_cast ops

* enhance ut

* enhance ut again

* remove duplicate codes

* enhance ut again

* Sharding reshard supports mismatch parameter name (#10479)

* Sharding Reshard Supports N -> 1 (#10532)

* cherry-pick https://github.com/PaddlePaddle/PaddleNLP/pull/10531

* support tpdp-ep sharding reshard (#10568)

* fix typo

* revert sharding_first

2025-09-03 15:44:15 +08:00

..

Merge prepare.benchmark.sh and prepare.sh (#1938 )

2022-04-13 12:15:56 +08:00

[BugFixes] update converter command & unittest (#4367 )

2023-01-10 12:02:04 +08:00

[CI] skip unit case for hang (#10642 )

2025-06-11 14:18:37 +08:00

fix dataaug test (#5532 )

2023-04-04 19:13:53 +08:00

Fix zero_padding (#9003 )

2024-08-26 21:04:22 +08:00

experimental/autonlp

[PIR] Fix test cases (#10456 )

2025-04-23 10:05:10 +08:00

LoRA-Pro Implemention (#10146 )

2025-05-28 19:25:24 +08:00

[Typos] Fix (#10494 )

2025-05-22 17:53:07 +08:00

Consolidate LoRA and Prefix into Peft package (#6051 )

2023-05-31 11:45:16 +08:00

Fix hidden states and quant kv cache (#10854 )

2025-08-21 15:35:25 +08:00

Longlora (#9970 )

2025-03-04 10:34:39 +08:00

[Mergekit]update & add LoRA merge (#9811 )

2025-02-11 17:46:33 +08:00

Download重构 (#8020 )

2024-03-08 16:41:40 +08:00

[Typos] Fix (#10494 )

2025-05-22 17:53:07 +08:00

Cherry-pick PRs from incubate/paddlenlp_fleety (#11049 )

2025-09-03 15:44:15 +08:00

LoRA-Pro Implemention (#10146 )

2025-05-28 19:25:24 +08:00

Revised PoSE (#8822 )

2024-11-21 16:06:41 +08:00

Fix failed ci. part1 (#10341 )

2025-04-15 10:10:20 +08:00

[TaskFlow] Fix pir for taskflow (#9822 )

2025-02-20 10:21:53 +08:00

clear flag (#10870 )

2025-07-21 19:47:12 +08:00

[Integration] add swanlab logger (#10594 )

2025-05-27 11:25:16 +08:00

[Typos] Fix (#10494 )

2025-05-22 17:53:07 +08:00

[tests] migrate to unittest.assertEqual (#10670 )

2025-07-04 11:30:44 +08:00

[CI]update build.sh (#10941 )

2025-08-20 16:51:05 +08:00

__init__.py

Add unittest (#179 )

2021-03-31 10:49:13 +08:00

common_test.py

[Refined Recompute] Remove refined recompute deep copy (#9617 )

2025-01-02 14:50:39 +08:00

parallel_launch.py

Fix unified_checkpoint bug (#7770 )

2024-01-04 19:42:05 +08:00

prepare.sh

update bos url (#1437 )

2021-12-10 18:37:04 +08:00

README.md

first commit without legacy files

2021-02-05 22:16:18 +08:00

requirements.txt

[CI] fix workflow (#10861 )

2025-07-18 11:20:45 +08:00

run_all_test.sh

Add unittest (#179 )

2021-03-31 10:49:13 +08:00

run_coverage.sh

Add unittest (#179 )

2021-03-31 10:49:13 +08:00

sample_text.txt

Add unittest for bert modeling and tokenizer (#2624 )

2022-07-21 10:35:55 +08:00

test_model.sh

Update transformer ci/ce (#812 )

2021-07-29 15:04:31 +08:00

testing_utils.py

[Typos] Fix (#10494 )

2025-05-22 17:53:07 +08:00

transformer_base_dygraph_params.txt

Use double precisionto train in full procedure (#1017 )

2021-09-14 11:57:56 +08:00

transformer_base_static_params.txt

Add custom vocab file for ce (#963 )

2021-09-07 21:47:57 +08:00

transformer_big_dygraph_params.txt

Use double precisionto train in full procedure (#1017 )

2021-09-14 11:57:56 +08:00

transformer_big_static_params.txt

Add custom vocab file for ce (#963 )

2021-09-07 21:47:57 +08:00

README.md

UnitTest for PaddleNLP

TBD