Mirrors/modelscope
1
0
Fork 0
mirror of https://github.com/modelscope/modelscope.git synced 2025-12-23 19:49:24 +01:00
Code Issues Packages Projects Releases Wiki Activity
2,825 Commits 263 Branches 106 Tags
fix/patch_dynamic_module
Commit Graph

801 Commits

Author SHA1 Message Date
Xingjun.Wang
055496c597 Fix CI 2025-08-07 19:26:32 +08:00
tastelikefeet
d8bb003681 fix bug (#1441) 2025-08-04 22:05:49 +08:00
Koko-ry
d6f11c1c22 Skip UT TextToSpeechSambertHifigan16kPipelineTest temporarily (#1420) 2025-07-17 18:57:44 +08:00
co63oc
8323fc5185 Fix typos in multiple files (#1357) 2025-06-05 14:04:29 +08:00
Z-yq
f9b8d4b9d2 vc ssr (#1208) 
* vc ssr

* Add more patches for hf (#1160)

* clone and lint #1205 (#1209)

* 更新格式

* fix cache path (#1211)

Co-authored-by: Yingda Chen <yingda.chen@alibaba-inc.com>

* “update"

* fix create_commit login (#1210)

* support multiple include/exclude filter patterns in command line (#1214)

Co-authored-by: Yingda Chen <yingda.chen@alibaba-inc.com>

* Use legacy cache (#1215)

* fix name (#1216)

Co-authored-by: Yingda Chen <yingda.chen@alibaba-inc.com>

* fix path name for log accuracy (#1217)


* change log msg

---------

Co-authored-by: Yingda Chen <yingda.chen@alibaba-inc.com>

* fix visibility (#1222)

Co-authored-by: Yingda Chen <yingda.chen@alibaba-inc.com>

* Merge 1.23 hotfix to master (#1227)

* 修复格式问题

* fix 路径问题

* Update test_speech_super_resolution.py

* Update test_voice_conversion.py

---------

Co-authored-by: tastelikefeet <58414341+tastelikefeet@users.noreply.github.com>
Co-authored-by: Yingda Chen <yingdachen@apache.org>
Co-authored-by: Yingda Chen <yingda.chen@alibaba-inc.com>
Co-authored-by: zhongyuqi <zhongyuqi@microbt.com>
 2025-06-04 16:08:20 +08:00
tastelikefeet
75d54927e1 Support trust_remote_code for pipeline and model (#1333) 2025-05-13 22:52:57 +08:00
suluyana
1d1463434d fix: pipeline builder (#1280) 
* fix pipeline builder
---------

Co-authored-by: suluyan <suluyan.sly@aliabab-inc.com>
 2025-03-21 12:29:06 +08:00
chenzhaofei01
51d27d7d04 face recognition pipline adds use_det parameter to control whether to use face detection model (#1212) 
* face recognition pipline adds use_det parameter to control whether to use face detection model

---------

Co-authored-by: zhaoshao <chenzhaofei.czf@alibaba-inc.com>
 2025-03-03 14:15:36 +08:00
suluyana
1fe211ffe5 fix pipeline builder when model is not supported (#1125) 
* fix pipeline builder when model is not supported

* fix ci & skip
---------

Co-authored-by: suluyan.sly@alibaba-inc.com <suluyan.sly@alibaba-inc.com>
 2024-12-12 19:24:38 +08:00
tastelikefeet
d2ef1003ea Skip obsolete sd pipeline (#1131) 2024-12-05 20:07:32 +08:00
Yingda Chen
2b1c839918 Format llm pipeline (#1094) 
* format llm pipeline

Co-authored-by: Yingda Chen <yingda.chen@alibaba-inc.com>
 2024-11-22 20:04:59 +08:00
Mark
250b72fce7 fix: text error correction batch run bug (#1052) 
Co-authored-by: Mark <smartmark-pro@qq.com>
 2024-11-01 09:35:19 +08:00
Mashiro
e71057cacf feat(audio/ans): Add ZipEnhancer and related layers for acoustic nois… (#1019) 
* feat(audio): Add acoustic noise suppression pipeline and tests for zipenhancer model

* Introduce `ZipEnhancer` module and associated layers (`ZipEnhancerLayer`, `Generator`, `ZipFormer`, ...).
* Add `speech_zipenhancer_ans_multiloss_16k_base` pipeline for `ZipEnhancer` module.
* Add new test cases and update metainfo.


Co-authored-by: Haoxu Wang <wanghaoxu.whx@alibaba-inc.com>
 2024-10-24 15:38:09 +08:00
Ranqing
543a2b32d7 upload high quality human normal estimation model (#903) 
* upload human normal estimation model

* upload human normal estimation unittest

* update human normal estimation

* update test data

---------

Co-authored-by: 葭润 <ranqing.rq@alibaba-inc.com>
Co-authored-by: ranqing <ranqing@Sundays-Mac.local>
Co-authored-by: suluyana <suluyan_sly@163.com>
 2024-07-31 10:19:41 +08:00
Xingjun.Wang
210ab40c54 Upgrade datasets (#921) 
* del _datasets_server import in hf_dataset_util

* fix streaming for youku-mplug and adopt latest datasets

* fix download config copy

* update ut

* add youku in test_general_datasets

* update UT for general dataset

* adapt to datasets version: 2.19.0 or later

* add assert for youku data UT

* fix disable_tqdm in some functions for 2.19.0 or later

* update get_module_with_script

* set trust_remote_code is True in load_dataset_with_ctx

* update print info

* update requirements for datasets version restriction

* fix _dataset_info

* add pillow

* update comments

* update comment

* reuse _download function in DataDownloadManager

* remove unused code

* update test_run_modelhub in Human3DAnimationTest

* set datasets>=2.18.0
 2024-07-23 22:26:12 +08:00
Firmament-cyou
8cdbeec3f6 Support swift model inference for llm pipeline (#880) 
* Support swift model inference for llm pipeline

* fix bug

* Add 'swfit' llm_framework and fix stream bug

* For pass the unittest run action
 2024-06-30 19:43:13 +08:00
yfchenmodelscope
f65f45959d Add Eres2netv2ep4 (#892) 
* add eres2netv2ep4
 2024-06-27 21:09:51 +08:00
yfchenmodelscope
f9451bfe38 add sdpn and tdnn (#865) 
* add sdpn and tdnn
 2024-05-24 14:39:16 +08:00
liuyhwangyh
82ee20f447 fix issue #845 (#861) 
* fix #845

Co-authored-by: mulin.lyh <mulin.lyh@taobao.com>
 2024-05-23 20:34:52 +08:00
yfchenmodelscope
3a0072bef8 add eres2netv2 (#830) 2024-04-15 16:53:53 +08:00
pipikk
802f90d2e1 feat: add raft model (#702) 
* first version

* fix minor bugs

* add test images

* TorchModel & docstr

* modify pipeline implementation

* minor

* add datasets

* add extractor

* add update

* add augmentor

* add flow_viz

* add frame_utils

* add utils

* add raft_model

* add dense_optical_flow_estimation_pipeline

* add image_utils

* add test_dense_optical_flow_estimation

* test

* update cv/__init__

* [3]update cv/__init__

* update submodule data/test

* correct yapf

* fix bugs

* move test data

* update submodule

* minor

* update submodule

---------

Co-authored-by: kejie <kejie.qkj@alibaba-inc.com>
 2024-04-08 15:27:20 +08:00
Ranqing
1ade52df17 upload marigold monocular depth estimation core files (#703) 
* upload marigold monocular depth estimation core files

* fix lint

* remove unused files.

* update marigold model files

* update marigold core files to fix review comments

* fix lint

* fix lint

* fix lint

* format code

* format code

---------

Co-authored-by: 葭润 <ranqing.rq@alibaba-inc.com>
Co-authored-by: wenmeng zhou <wenmeng.zwm@alibaba-inc.com>
 2024-03-05 11:21:29 +08:00
Firmament-cyou
a0fe5947b7 Support stream_generate for LLMPipeline (#768) 
* support streaming output for llm_pipeline
* add qwen2 format_messages
 2024-03-05 10:05:18 +08:00
Zhe Sheng
f79a5b5924 Features/cv_geomvsnet_multi_view_depth_estimation_general (#790) 
* add mvs_depth_estimation model: GeoMVSNet

* file headers

* modify unitest class name

---------

Co-authored-by: shengzhe.sz <shengzhe.sz@alibaba-inc.com>
 2024-03-04 19:36:00 +08:00
yfchenmodelscope
2ecb97ac4f add res2net resnet models (#772) 
* add res2net resnet models

* add paper link and model introduction
 2024-02-23 16:19:13 +08:00
heyyxd
158d72bfd2 add self supervised depth completion. (#711) 
* add self supervised depth completion.

* update.

* fix the problem of key inconsistency.

* delete args parser.

* rename metrics to test_metrics.
 2024-02-22 22:30:48 +08:00
mulin.lyh
588e41c787 fix lint issue 2024-01-22 15:52:30 +08:00
liuyhwangyh
672c32e7bd fix ci compatible issues，fix llmpipeline lazy import issue (#725) 
* fix ci issue

* fix case issue

* modify lint to python3.10

* fix case issue

---------

Co-authored-by: mulin.lyh <mulin.lyh@taobao.com>
 2024-01-17 22:19:05 +08:00
Yisheng (Ethan) He
94ce1ebd7a Feature/LoFTR_image_local_feature_matching (#687) 
* Add loftr image local feature matching.

* add pipeline doc str and remove example data as examples exists in data/test

* update pipeline doc str.

* add pipeline doc str

add pipeline doc str

---------

Co-authored-by: 翼生 <heyisheng.hys@alibaba-inc.com>
Co-authored-by: wenmeng zhou <wenmeng.zwm@alibaba-inc.com>
 2024-01-17 21:51:29 +08:00
zhengyi-z
a9d5b88407 add image matching fast model based on lightglue (#694) 
* add image matching fast model based on lightglue

* add model-paper doc str and pipeline doc str

---------

Co-authored-by: bushe.zzy <bushe.zzy@alibaba-inc.com>
 2024-01-17 21:46:10 +08:00
Weihao Yuan
105247140c Feature/image normal estimation (#683) 
* image_normal_estimation

* image_normal_estimation

* update according to pr review

* update submodule data test

---------

Co-authored-by: Weihao Yuan <qianmu.ywh@alibaba-inc.com>
 2024-01-09 11:53:02 +08:00
slin000111
2d528ed482 fix anydoor pre-commit flake8 and isort errors (#707) 
* fix anydoor pre-commit flake8 and isort errors

* Add noqa to resolve conflicts

* fix image_to_3d_pipeline test error

---------

Co-authored-by: slin000111 <zhaoshilin2015@sina.con>
 2024-01-09 11:52:39 +08:00
zhangwlgq
0b1a974800 add rife-video-frame-interpolation and model (#685) 
* add rife-video-frame-interpolation pipeline and model

* add doc, change test level

---------

Co-authored-by: miracle.zjf <miracle.zjf@alibaba-inc.com>
Co-authored-by: wenmeng zhou <wenmeng.zwm@alibaba-inc.com>
 2024-01-02 21:50:11 +08:00
wenmeng.zwm
39562dc555 merge master 2023-12-27 00:08:20 +08:00
Firmament-cyou
b473decba0 update ckpt to general_v0.1 (#696) 2023-12-26 16:59:47 +08:00
Firmament-cyou
b3c6eebdf3 Add AnyDoor support (#688) 
* support anydoor

* add dinov2

* fix bug

* convert rgb

* update anydoor_pipeline and add docstr
 2023-12-25 19:01:50 +08:00
hitsz-zuoqi
2ef12d1e88 add syncdreamer as a image-to-3d pipeline (#679) 
* add syncdreamer

* delete demo_output

* remove useless files and hide clip install

* update pipeline

* update clip package

* update unitest
 2023-12-21 22:12:32 +08:00
mulin.lyh
86fcbb8fbf Merge branch 'master-gitlab' into master-merge-internal20231208 2023-12-08 21:24:27 +08:00
Firmament-cyou
8f0f9d4a33 Register llm format map (#659) 
* add LLMAdapterRegistry

* fix bug

* replace traceback with cache
 2023-12-06 16:22:50 +08:00
myf272609
00eb4219a0 [to #42322933] fix issues for 3dhuman models 
- 角色驱动：添加自定义blender路径支持；移除模型位置标准化
- 角色渲染：添加自定义渲染分辨率支持；添加模型位置标准化
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/14459360
* fix some issues

* fix
 2023-11-08 21:11:21 +08:00
Firmament-cyou
fafb0fe013 llm pipeline support chatglm3 (#618) 2023-11-04 20:44:55 +08:00
chenyafeng.cyf
e8b34a3bcd add_eres2net_speaker_diarization 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/14412468
* add_eres2net_speaker_diarization

* fix lint issue

* update code

* update pipeline
 2023-10-25 20:36:36 +08:00
wenmeng.zwm
881c98d751 feat: Resolve conflict, auto committed by Code 2023-10-24 19:31:23 +08:00
zhangyanzhao.zyz
ebd6ddb530 更新sentence embedding model，支持gte，bloom sentence embedding 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/14375781

* fix linter

* bloom embedding
 2023-10-24 15:23:50 +08:00
huizheng.hz
0911283dde add freeU model 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/14307648


* support sd21, sdxl
 2023-10-24 15:23:50 +08:00
mulin.lyh
d30ef8b202 fix huggingface position_ids compatible issue 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/14406558
* fix compatible issues

* fix transformer compatible issue

* skip case for huggingface link issue

* fix hf autotokenlizer case

* Merge branch 'fix_ci_issue' of http://gitlab.alibaba-inc.com/Ali-MaaS/MaaS-lib into fix_ci_issue
 2023-10-24 15:18:55 +08:00
Firmament-cyou
a3ae55577f Support swift in llm_pipeline. (#595) 
* support swift in llm_pipeline

* adjust function name
 2023-10-24 14:31:24 +08:00
zhangyanzhao.zyz
fb7328f4ec 更新sentence embedding model，支持gte，bloom sentence embedding 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/14375781

* fix linter

* bloom embedding
 2023-10-20 19:56:01 +08:00
huizheng.hz
690473ce85 add freeU model 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/14307648


* support sd21, sdxl
 2023-10-20 16:10:54 +08:00
hemu.zp
4cad376298 Add llm_first parameter for pipeline 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/14264249

* support llm_first parameter

* register_module(Tasks.text_generation)

* fix bug

* update format & fix out_base64 for int4

* pre-commit
 2023-10-13 14:04:04 +08:00