Mirrors/modelscope
1
0
Fork 0
mirror of https://github.com/modelscope/modelscope.git synced 2026-09-02 03:59:31 +02:00
Code Issues Packages Projects Releases Wiki Activity
2,547 Commits 280 Branches 124 Tags
fix/cache_dir
Commit Graph

2547 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Wang Qiang
de33f4dc87 Support sdxl finetune by lora method (#468) 
* support sdxl finetune by lora

* remove useless imports

* support sdxl finetune

* upgrade diffusers to 0.19.0

* sdxl finetune

* fix bugs

* pre commit

* diffusers>=0.19.0
 2023-08-23 11:23:34 +08:00
mulin.lyh
cba4e40bc1 fix numpy pandas compatible issue 
明确受影响的模型(damo)：  
ONE-PEACE-4B	ModuleNotFoundError: MyCustomPipeline: MyCustomModel: No module named 'one_peace'，缺少依赖。
cv_resnet50_face-reconstruction	 不兼容tf2  
nlp_automatic_post_editing_for_translation_en2de	tf2.0兼容性问题，tf1.x需要  
cv_resnet18_ocr-detection-word-level_damo	tf2.x兼容性问题  
cv_resnet18_ocr-detection-line-level_damo	tf兼容性问题  
cv_resnet101_detection_fewshot-defrcn	模型限制必须detection0.3+torch1.11.0"  
speech_dfsmn_ans_psm_48k_causal	"librosa， numpy兼容性问题  
cv_mdm_motion-generation	"依赖numpy版本兼容性问题：   File ""/opt/conda/lib/python3.8/site-packages/smplx/body_models.py"",  
cv_resnet50_ocr-detection-vlpt	numpy兼容性问题  
cv_clip-it_video-summarization_language-guided_en	tf兼容性问题

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/13744636
* numpy and pandas no version

* modify compatible issue

* fix numpy compatible issue

* modify ci

* fix lint issue

* replace Image.ANTIALIAS to Image.Resampling.LANCZOS pillow compatible

* skip uncompatible cases

* fix numpy compatible issue, skip cases that can not compatbile numpy or tensorflow2.x

* skip compatible cases

* fix clip model issue

* fix body 3d keypoints compatible issue
 2023-08-22 23:04:31 +08:00
tastelikefeet
df53a6a89f Add bloom text generation model (#492) 
* add bloom text-generation

* pre-commit passed
 2023-08-22 16:56:15 +08:00
Kang
cccd502fa2 fix video output of image2video (#488) 
* fix video output

* fix logger.error

* fix log error
 2023-08-22 10:57:23 +08:00
Xingjun.Wang
82902b8758 add slicing method for NativeIterableDataset (#490) 2023-08-22 09:29:51 +08:00
Kang
040698e201 add video2video (#486) 
* add video2video

* fix bugs of pre-commit

* update some files

* fix video write module

* fix max_frames
 2023-08-21 18:44:14 +08:00
yanyi.ys
be26a62c48 人像生成代码评审 
该CR提交了关于人像生成的代码，能够通过给定人像图片以及相应的target姿势数据生成相应姿势的图片。

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/13715612
 2023-08-21 18:42:24 +08:00
ryan
5d0f85a9ba update image-face-fusion module for facechain (#487) 
* update image-face-fusion module for facechain

* add model not found warning

* revise lint error

---------

Co-authored-by: 嘉弈 <ryan.yy@alibaba-inc.com>
 2023-08-21 18:29:45 +08:00
kangxiaoyang.kxy
6c3973b9a3 fix image colorization model load issue 
之前的评审提交（https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/13669766?tab=changes）可能导致 modelscope hub 的模型无法读取，对于不同 modelscope 版本使用者产生兼容性问题，因此在上次评审提交的基础上做一定修改确保兼容性。
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/13716448
* fix image colorization model load issue
 2023-08-21 09:47:32 +08:00
wenmeng.zwm
7f7573f15d fix ut error 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/13716099
* fix ut error
 2023-08-19 21:16:07 +08:00
kangxiaoyang.kxy
1c76007da7 update image colorization model: remove redundant layers for less memory use 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/13669766
* update image colorization model: remove redundant layers for less memory use
 2023-08-19 16:27:09 +08:00
Wang Qiang
78f5e6a8bf Switching model from huggingface to modelscope hub of efficient tuning (#479) 2023-08-18 20:30:04 +08:00
Wang Qiang
f321804ab2 Merge pull request #472 from kangzhao2/baishao_test 
Add image2video
 2023-08-18 20:29:36 +08:00
kangzhao2
2643d985dc update test_image2video.py 2023-08-18 20:13:20 +08:00
kangzhao2
8ff95a5dc8 add xformer in unet_i2v.py 2023-08-18 19:41:35 +08:00
liangym
c8826ab04b modify judgment in_audios type (#474) 
* Delete judgment in_audios type

* use input

---------

Co-authored-by: wantt <wantt2@qq.com>
 2023-08-18 18:22:21 +08:00
kangzhao2
5f0f63eb37 fix init.py 2023-08-18 17:55:40 +08:00
kangzhao2
626f345eff add __init__.py file 2023-08-18 17:13:01 +08:00
kangzhao2
2605935797 fix pre-commit 2023-08-18 11:47:45 +08:00
kangzhao2
4ca76b0a85 fix comments again 2023-08-17 20:34:43 +08:00
kangzhao2
b8c76a426f fix comments 2023-08-17 20:03:05 +08:00
Wang Qiang
4ed1111d70 Fix bugs of configs file path and duration (#476) 
* fix bugs of configs file path and duration

* pre commit

* delete configs

* test videocomposer model version
 2023-08-16 21:03:11 +08:00
kangzhao2
90f7a5c6c0 update files 2023-08-16 11:35:15 +08:00
kangzhao2
037e73fe6e baishao 2023-08-15 21:32:30 +08:00
kangzhao2
e6404f29a2 baishao 2023-08-15 21:17:00 +08:00
Wang Qiang
a67d339e3b support float16 training and pipeline for stable diffusion (#447) 
* support float16 traing and pipeline for stable diffusion

* pre commit

* fix bugs

* add torch type example

* fix bugs of torch type

* support type float16

* fix bugs of load pipeline

* change type to fp16

* lora rank

---------

Co-authored-by: 翊靖 <yijing.wq@alibaba-inc.com>
 2023-08-15 20:04:32 +08:00
Zhicheng Zhang
d212ced3f3 Update qwen QA example (#470) 2023-08-15 17:25:25 +08:00
wenmeng.zwm
8db088d2cb fix virtural tryon ut error 2023-08-15 16:49:15 +08:00
Wang Qiang
ee8afd2d62 VideoComposer: Compositional Video Synthesis with Motion Controllability (#431) 
* VideoComposer: Compositional Video Synthesis with Motion Controllability

* videocomposer pipeline

* pre commit

* delete xformers
 2023-08-15 12:01:03 +08:00
xiangpeng.wxp
344d2a2e5b [to #42322933] add truct remote code for polylm 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/13628684
* add truct remote code for polylm

* [to #42322933] add truct remote code for polylm

* [to #42322933] add truct remote code for polylm
 2023-08-14 15:37:42 +08:00
Jintao
18d33a4825 fix copytree python37 bug (#464) 
* fix copytree python37 bug

* add copytree_py37 function
 2023-08-14 11:45:33 +08:00
Yingda Chen
b18288cd6b update example 2023-08-13 11:06:41 +08:00
wenmeng zhou
74d8317bb0 fix pipeline check error (#455) 
* fix pipeline check error

* update
 2023-08-11 15:52:53 +08:00
Jintao
c06f47f384 add readme and warning (#462) 
* add readme and warning

* fix bug

* update

* update readme
 2023-08-11 14:55:24 +08:00
Ran Zhou
026a9ef227 Add machine reading comprehension model, preprocessor and pipeline (#451) 
* Add machine reading comprehension model, preprocessor and pipeline

* fix precommit errors

* Optimize mrc preprocessor, add mrc input output definition, add mrc pipeline docstr

---------

Co-authored-by: seadamo <ran.zhou@alibaba-inc.com>
 2023-08-11 13:47:26 +08:00
Zhicheng Zhang
0e8be51fef add qwen qa example with langchain 2023-08-11 11:53:17 +08:00
xiangpeng.wxp
5e83523c9a [to #42322933] Add polylm lora trainer to modelscope 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/13509769
* [to #42322933] Add polylm lora trainer to modelscope
 2023-08-10 20:57:47 +08:00
Zhicheng Zhang
ad5b134bfe update qwen qa example 2023-08-10 19:34:39 +08:00
Zhicheng Zhang
2baef2110d update QA example 2023-08-10 19:03:20 +08:00
chenyafeng.cyf
33605de759 eres2net_lre_v2 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/13602081

* add eres2net_base_large_lre

* eres2net_language_identification

* eres2net_lre_v2
 2023-08-10 17:44:43 +08:00
liuyhwangyh
75a14a36ba get github diff files (#446) 
* get github diff files

* add github environment

---------

Co-authored-by: mulin.lyh <mulin.lyh@taobao.com>
 2023-08-10 16:22:37 +08:00
wenmeng zhou
e3abfe0564 Update framework.txt (#453) 
remove version restrict for numpy and pandas
 2023-08-09 18:00:44 +08:00
wenmeng zhou
6f371a2d29 bumpy version to 1.8.2 (#449) 2023-08-09 17:50:26 +08:00
mushenL
75dd355ad7 llama2 max_length change (#452) 
default max_length set to 2048
 2023-08-09 11:07:09 +08:00
wenmeng.zwm
725521a2af skip test_text_to_360panorama_image test 2023-08-08 16:41:19 +08:00
Zhicheng Zhang
a36c25dbd3 Merge remote-tracking branch 'github/master' into master-github 2023-08-08 15:11:02 +08:00
Zhicheng Zhang
9854759a55 add doc search QA with dashscope 2023-08-08 15:05:50 +08:00
wenmeng zhou
cba1156f0d fix chatglm2b rope_ratio config is missing (#440) 2023-08-08 14:53:29 +08:00
zsl01670416
43a57fe110 support load dataset for llama 
support loading dataset for llama:
1.load dataset by MsDataset when parameters train dataset name and val dataset name were set. but there is no suitable dataset in hub.
2.load dataset by MsDataset when only parameter train dataset name was set, and then split into train dataset and validation dataset . 
3.load dataset by MsDataset when user set parameter src_txt, which is a file path such as 'alpaca_data.json', and then split into training dataset and validation dataset.
4.load dataset by build dataset from file in flex training.

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/13505335
 2023-08-07 19:48:36 +08:00
lingcai.wl
7757f3c64a [to #51336898] support eas deploy, fix minor problems 
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/13541904
* [to #51336898] support eas deploy

* [to #51336898] support eas deploy, fix minor problems
 2023-08-07 19:48:07 +08:00