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
741 Commits 263 Branches 106 Tags
acba1786b0ac263140e2743b51fc7fc7c36d0980
Commit Graph

363 Commits

Author SHA1 Message Date
shichen.fsc
1483c64638 [to #42322933] Fix ASR error when resample failed, and add all asr models UT, add apply-cmvn for pytorch models 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10465241
 2022-10-20 12:54:37 +08:00
xianzhe.xxz
865397763e [to #42322933]add damoyolo model in tinynas-object-detection 
接入damyolo系列检测模型
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10377688
 2022-10-18 16:53:29 +08:00
shuying.shu
cb570d586c add referring video object segmentation pipeline 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10400324
 2022-10-18 16:10:10 +08:00
hemu.zp
3b1f1a0252 [to #42322933] Add GPT3 tensor parallel inference 
添加基于 Megatron-v3 的 GPT3 tensor 并行的推理代码
复用 DistributedPipeline 与 megatron-util
适用模型：1.3B/2.7B/13B 参数的 GPT-3 预训练生成模型
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10416721
 2022-10-18 15:58:33 +08:00
leyuan.hjy
172522d196 [to #42322933]video-object-detection init 
新增video-object-detection 算法
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10247489
 2022-10-17 20:58:23 +08:00
hemu.zp
271e2a2a99 [to #42322933] Add gpt_neo model 
1. 添加 gpt_neo 模型，因 checkpoint 归属于 Langboat 还未上传到模型库，已线下完成测试
2. 添加 text-generation task models 与 head，后续会将 gpt3，palm 等已上线文本生成模型统一为 backbone + head 结构的 task models
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10404249
 2022-10-17 20:54:29 +08:00
jiangnana.jnn
7720ae50e2 return dict values when input single sample for easycv pipeline 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10423383
 2022-10-17 20:30:42 +08:00
shichen.fsc
542c4ce1b3 [to #42322933] Fix bug in KWS when setting customized keyword 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10412829
 2022-10-17 14:07:05 +08:00
wendi.hwd
674e1a7878 [to #42322933]cv/cvdet_fix_outputs->master fix outputs 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10421413

    * fix outputs
 2022-10-17 14:06:07 +08:00
wenqi.oywq
88a7599efb [to #42322933]change output channels from RGB to BGR, to consistent with demo-service 
默认输出为array的，通道格式统一为BGR格式，本次修改是为了与这个格式一致
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10422508
 2022-10-17 14:05:12 +08:00
wenmeng.zwm
f6e542cdcb refine pipeline input to support demo service 
* image_captioninig support single image and dict input
* image_style_transfer use dict input

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10417330
 2022-10-17 10:40:08 +08:00
caorongyu.cry
202fcdf298 [to #42322933] change tableqa output 
修改output的结构，直接返回可转化成json format的结构
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10415403
 2022-10-14 23:11:19 +08:00
zhangzhicheng.zzc
0eb823b764 [to #42322933] support t5_with_translation 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10383770

    * T5 support translate
 2022-10-13 10:52:40 +08:00
jiaqi.sjq
144ffee2cf [to #42322933] Add explict model id in tts UT 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10371244
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10371244
 2022-10-13 10:16:07 +08:00
yuxiang.tyx
2989492bc0 [to #42322933]更新face_detection_scrfd模型并支持finetune, 新增card_detection模型 
1. 调整face_detection的文件层级(scrfd与其余新增face_detection方法平级);
2. 增加极大脸/旋转脸的检测方法，更新了新模型；
3. 支持读入数据集并finetune和eval；
4. 新增card_detection模型，支持读入datasethub数据集并finetune
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10244540
 2022-10-12 19:58:50 +08:00
shouzhou.bx
4cb5f8a2cd [to #42322933] add human whole body model and image object detection auto model 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10319306
 2022-10-12 19:53:14 +08:00
caorongyu.cry
7145990054 [to #42322933] reivse model problem and remove history sql for demo 
相比于master上的tableqa，做出了如下修复：
1. 修复了schema linking中的问题。
2. 同时设置了有history sql和没有history sql的两种输入
3. 增加了sqlite执行逻辑，可以返回sql执行结果
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10365114
 2022-10-12 15:18:35 +08:00
huizheng.hz
922f4c589b [to #42322933]图像去噪using msdataset to load dataset 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10338265
 2022-10-11 22:46:30 +08:00
bin.xue
0d97f8959d [to #42322933] test: unify kws pipeline input type to AUDIO 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10362437
 2022-10-11 22:24:19 +08:00
ashui.cbh
69da8f91ac [to #42322933]suport image inpainting 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10111615
 2022-10-11 20:49:13 +08:00
james.wjg
4c99363804 [to #42322933]video summarization 添加 license & header; 修改 output for demo service 
video summarization:
1. 添加 license & header; 
2. 修改 output for demo service
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10260946
 2022-10-11 17:24:46 +08:00
jiaqi.sjq
65be443e98 [to #41669377] Add more models to test in tts UT 
Link https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10360754#tab=detail
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10360754
 2022-10-11 17:22:58 +08:00
zhangzhicheng.zzc
e240edea7e [to #42322933]t5 bug fixex 2022-10-11 17:20:11 +08:00
hanyuan.chy
67d6fa001d [to #42322933] unify keys forbody_3d_keypoints 
统一关键点检测输出key的名字
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10359335
 2022-10-11 17:17:51 +08:00
ryan.yy
ff69439c4f [to #42322933]add image_body_reshaping code 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10217723

    * add image_body_reshaping code
 2022-10-10 17:42:41 +08:00
moyu.wk
42896de709 [to #42322933]add pipelines for APE\QE\Domain Classifications 
Add pipelines for the following new models:
- [Multilingual Quality Estimation](https://modelscope.cn/models/damo/nlp_translation_quality_estimation_multilingual/summary)
- [Automatic Post-Editing (En-De)](https://modelscope.cn/models/damo/nlp_automatic_post_editing_for_translation_en2de/summary)
- [Domain classification (Zh)](https://modelscope.cn/models/damo/nlp_domain_classification_chinese/summary)
- [Style classification (Zh)](https://modelscope.cn/models/damo/nlp_style_classification_chinese/summary)
- [Style classification (En)](https://modelscope.cn/models/damo/nlp_style_classification_english/summary)
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10315370
 2022-10-10 11:36:39 +08:00
wenmeng.zwm
65bea053af [to #44902165] feat: Add input signature for pipeline 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10233235

* add input definition for pipeline


* support multiple input format for single pipeline


*  change param for body_3d_keypoints
 2022-10-09 18:59:07 +08:00
tingwei.gtw
a079ab922f [to #42322933] add product-segmentation pipeline 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10252583
 2022-10-01 21:46:40 +08:00
tingwei.gtw
5343c899fb [to #42322933] add face-emotion pipeline 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10202117
 2022-10-01 18:35:42 +08:00
hanyuan.chy
7f468acca3 [to #42322933]style(license): add license + render result poses with video 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10263904
 2022-10-01 18:34:23 +08:00
tingwei.gtw
4199af337e [to #42322933] add face-human-hand detection model 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10260332
 2022-10-01 15:57:12 +08:00
lllcho.lc
9fa761d7a6 [to #42322933] add PST action recognition model 
Add patch shift transformer model for action recognition task.
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10282964
 2022-10-01 11:10:36 +08:00
chaojie.mcj
0e52a20d28 [to #42322933]update license 
以下算法进行了header变更：
modelscope.models.cv.cmdssl_video_embedding
modelscope.models.cv.action_recognition 
modelscope.models.cv.animal_recognition
modelscope.models.multi_modal.multi_stage_diffusion
modelscope.models.multi_modal.gemm

modelscope.pipelines.cv.live_category_pipeline
modelscope.pipelines.cv.video_category_pipeline
modelscope.models.cv.image_to_image_translation
modelscope.models.cv.image_to_image_generation

modelscope.models.cv.video_inpainting
modelscope.models.multi_modal.diffusion
modelscope.models.multi_modal.team
modelscope.models.cv.shop_segmentation
modelscope.models.cv.text_driven_segmentation
modelscope.models.cv.action_recognition 


modelscope.models.cv.face_emotion
modelscope.models.cv.hand_static
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10268474
 2022-09-28 14:30:37 +08:00
Yingda Chen
11b33164c3 [to #42322933] disable t5 test temporarily 2022-09-28 09:26:44 +08:00
zhangzhicheng.zzc
d721fabb34 [to #42322933]bert with sequence classification / token classification/ fill mask refactor 
1.新增支持原始bert模型（非easynlp的 backbone prefix版本）
2.支持bert的在sequence classification/fill mask /token classification上的backbone head形式
3.统一了sequence classification几个任务的pipeline到一个类
4.fill mask 支持backbone head形式
5.token classification的几个子任务（ner，word seg， part of speech）的preprocessor 统一到了一起TokenClassificationPreprocessor
6. sequence classification的几个子任务（single classification， pair classification）的preprocessor 统一到了一起SequenceClassificationPreprocessor
7. 改动register中 cls的group_key 赋值位置，之前的group_key在多个decorators的情况下，会被覆盖，obj_cls的group_key信息不正确
8. 基于backbone head形式将 原本group_key和 module同名的情况尝试做调整，如下在modelscope/pipelines/nlp/sequence_classification_pipeline.py 中 
原本
 @PIPELINES.register_module(
    Tasks.sentiment_classification, module_name=Pipelines.sentiment_classification)
改成
@PIPELINES.register_module(
    Tasks.text_classification, module_name=Pipelines.sentiment_classification)
相应的configuration.json也有改动，这样的改动更符合任务和pipline（子任务）的关系。
8. 其他相应改动为支持上述功能
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10041463
 2022-09-27 23:08:33 +08:00
tingwei.gtw
372adb3936 [to #42322933] support hand-static model 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10244616
 2022-09-27 23:04:38 +08:00
jiaqi.sjq
e90ff9e479 [to #42322933] tts sambert am changs from tensorfow to PyTorch and add licenses 
* [to #41669377] docs and tools refinement and release 

1. add build_doc linter script
2. add sphinx-docs support
3. add development doc and api doc
4. change version to 0.1.0 for the first internal release version

Link: https://code.aone.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/8775307
 2022-09-27 22:09:30 +08:00
wendi.hwd
26df8f1988 [to #42322933]add semantic-segmentation task output is numpy mask for demo-service 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10265856
 2022-09-27 15:01:05 +08:00
tianchu.gtc
bd4127bc27 [to #42322933]segformer 接入demo接口更改 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10253628
 2022-09-26 15:49:35 +08:00
zhangzhicheng.zzc
5e4894870b [to #42322933]add t5 model / text2text generation task 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10191736

    * add T5 for generation
 2022-09-26 12:23:28 +08:00
pengyu.lpy
3ed0c9c8d8 [to #42322933] relax un-determinsitic test validation constraints 
放松了tests/pipelines/easycv_pipelines/test_segmentation_pipeline.py里面对于识别非determinsitic的校验条件，一方便后续模型更新
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10245940
 2022-09-24 17:58:42 +08:00
hemu.zp
69f8928dd2 [to #42322933] Replace mplug input 'question' with 'text' 
mplug 相关任务 pipeline 输入字段统一为 'image' + 'text'
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10236282
 2022-09-23 15:54:17 +08:00
bin.xue
470a1989bc [to #42322933] feat: far field KWS accept mono audio for online demo 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10211100
 2022-09-22 23:01:14 +08:00
jiangnana.jnn
376ba9fef9 [to #42322933]update easycv pipelines 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10208603

    * update easycv pipelines
 2022-09-22 15:15:29 +08:00
yichang.zyc
b537bb8c27 fix vg return value 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10207239
 2022-09-21 18:57:34 +08:00
wenmeng.zwm
6808e9a301 [to #44902099] add license for framework files 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10189613
 2022-09-20 17:49:31 +08:00
myf272609
0041ab0ab8 [to #42322933] add multi-style cartoon models to ut 
1. 卡通化接入多风格模型（原始日漫风、3D、手绘风、素描风、艺术特效风格），添加ut接入测试
2. 修改pipeline中模型文件名称至通用名
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10153717
 2022-09-19 11:28:01 +08:00
lingcai.wl
0548d92de8 [to #44657982] fix some demo problems 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10083156
 2022-09-19 10:27:26 +08:00
wenmeng.zwm
7fb25d7bbb [to #42322933]fix UT error for 830 version 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10142442
 2022-09-16 22:42:39 +08:00
caorongyu.cry
77cfcf0a9a [to #42322933] commit nlp_convai_text2sql_pretrain_cn inference process to modelscope 
commit nlp_convai_text2sql_pretrain_cn inference process to modelscope
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10025155
 2022-09-14 19:04:56 +08:00