modelscope
/
ModelScope
Not watched
1
0
Fork 0
Code
Releases 0 Wiki evaluate Activity
Issues 0 Pull Requests 0 Datasets Model Cloudbrain HPC
You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.
1128 Commits
1 Branch
10 MB
0 Download
Branch: master
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from 'master'
${ noResults }
ModelScope/requirements/multi-modal.txt

multi-modal.txt 393 B
Raw Permalink Normal View History

[to #42322933] add image_caption_pipeline with OFA 1. add OFA whl for image caption pipeline 2. fix a bug in pipelines/builder.py Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/8930942 * [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 * [to #41669377] add pipeline tutorial and fix bugs 1. add pipleine tutorial 2. fix bugs when using pipeline with certain model and preprocessor Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/8814301 * refine doc * refine doc * upload ofa for caption(with source code but not whl) * remove data in gitignore * append uncommitted data dir in ofa * remove ofa_dir , use ofa.whl instead. * update BPE * rollback changes used in debugging. * Merge branch 'master' into ofa/image_caption # Conflicts: # docs/README.md # docs/source/conf.py # docs/source/index.rst # docs/source/tutorials/pipeline.md # maas_lib/models/nlp/sequence_classification_model.py # maas_lib/pipelines/builder.py # maas_lib/version.py # setup.py # tests/pipelines/test_text_classification.py * 1. fix a bug in pipelines/builder.py. 2. modify model_path to model in image_captioning.py. * 1. rename test_image_captioning.py. * format all files using pre-commit. * add fairseq in requirements.txt * add fairseq in requirements.txt * change fairseq path to git repo to a whl on oss in ofa.txt. * change module_name to 'ofa' * Merge remote-tracking branch 'origin/master' into ofa/image_caption # Conflicts: # maas_lib/pipelines/builder.py * optim requirements for ofa / refine image_captioning.py * uncommited change. * feat: Fix confilct, auto commit by WebIDE
3 years ago
adapt to different wav input Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10886461
2 years ago
[to #42322933] add image_caption_pipeline with OFA 1. add OFA whl for image caption pipeline 2. fix a bug in pipelines/builder.py Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/8930942 * [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 * [to #41669377] add pipeline tutorial and fix bugs 1. add pipleine tutorial 2. fix bugs when using pipeline with certain model and preprocessor Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/8814301 * refine doc * refine doc * upload ofa for caption(with source code but not whl) * remove data in gitignore * append uncommitted data dir in ofa * remove ofa_dir , use ofa.whl instead. * update BPE * rollback changes used in debugging. * Merge branch 'master' into ofa/image_caption # Conflicts: # docs/README.md # docs/source/conf.py # docs/source/index.rst # docs/source/tutorials/pipeline.md # maas_lib/models/nlp/sequence_classification_model.py # maas_lib/pipelines/builder.py # maas_lib/version.py # setup.py # tests/pipelines/test_text_classification.py * 1. fix a bug in pipelines/builder.py. 2. modify model_path to model in image_captioning.py. * 1. rename test_image_captioning.py. * format all files using pre-commit. * add fairseq in requirements.txt * add fairseq in requirements.txt * change fairseq path to git repo to a whl on oss in ofa.txt. * change module_name to 'ofa' * Merge remote-tracking branch 'origin/master' into ofa/image_caption # Conflicts: # maas_lib/pipelines/builder.py * optim requirements for ofa / refine image_captioning.py * uncommited change. * feat: Fix confilct, auto commit by WebIDE
3 years ago
​
[to #45892407]fix: fix pytorch_lighting incompatible with taming-transformers-rom1504 Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10604329 * [to #45892407]fix: fix pytorch_lighting incompatible with taming-transformers-rom1504
2 years ago
​
[to #42322933] set rouge_score version 2022/7/22 更新的 rouge_socre 0.0.7 版本存在 import 报错的问题，暂时将 rouge_score 库版本设置为 <= 0.0.4
3 years ago
​
​
[to #42322933]caption finetune done, add belu metric Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10318299
2 years ago
ofa增加asr任务 ofa增加asr任务infer Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10761019
2 years ago
[to #42322933] Add ofa-text-to-image-synthesis to maas lib Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/9590940
3 years ago
[to #42322933] add image_caption_pipeline with OFA 1. add OFA whl for image caption pipeline 2. fix a bug in pipelines/builder.py Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/8930942 * [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 * [to #41669377] add pipeline tutorial and fix bugs 1. add pipleine tutorial 2. fix bugs when using pipeline with certain model and preprocessor Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/8814301 * refine doc * refine doc * upload ofa for caption(with source code but not whl) * remove data in gitignore * append uncommitted data dir in ofa * remove ofa_dir , use ofa.whl instead. * update BPE * rollback changes used in debugging. * Merge branch 'master' into ofa/image_caption # Conflicts: # docs/README.md # docs/source/conf.py # docs/source/index.rst # docs/source/tutorials/pipeline.md # maas_lib/models/nlp/sequence_classification_model.py # maas_lib/pipelines/builder.py # maas_lib/version.py # setup.py # tests/pipelines/test_text_classification.py * 1. fix a bug in pipelines/builder.py. 2. modify model_path to model in image_captioning.py. * 1. rename test_image_captioning.py. * format all files using pre-commit. * add fairseq in requirements.txt * add fairseq in requirements.txt * change fairseq path to git repo to a whl on oss in ofa.txt. * change module_name to 'ofa' * Merge remote-tracking branch 'origin/master' into ofa/image_caption # Conflicts: # maas_lib/pipelines/builder.py * optim requirements for ofa / refine image_captioning.py * uncommited change. * feat: Fix confilct, auto commit by WebIDE
3 years ago
[to #42322933] lazy load on trainer
3 years ago
[to #42362425] fix audio_requirement and refine quickstart, changelog doc * make audio requirements optional * add changelog for version v0.2 * add numpy constraint for compatibility with tensorflow1.15 * update faq * fix nlp requiring tensorflow * add torchvision to multimodal dependency * bump version from 0.2.1 to 0.2.2 * add warning msg when tensorflow is not installed Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/9268278
3 years ago
[to #42322933] lazy load on trainer
3 years ago
[to #42322933]fix ocr prepreocess & conflict 修复ocr预处理逻辑不一致问题 Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10581697
2 years ago
​
 123456789101112131415161718 
  1. ftfy>=6.0.3

  2. librosa

  3. pycocoevalcap>=1.2

  4. pycocotools>=2.0.4

  5. # compatible with taming-transformers-rom1504

  6. pytorch_lightning<=1.7.7

  7. # rough-score was just recently updated from 0.0.4 to 0.0.7

  8. # which introduced compatability issues that are being investigated

  9. rouge_score<=0.0.4

  10. sacrebleu

  11. soundfile

  12. taming-transformers-rom1504

  13. timm

  14. tokenizers

  15. torchvision

  16. transformers>=4.12.0

  17. unicodedata2

  18. zhconv

致力于通过开放的社区合作，开源AI模型以及相关创新技术，推动基于模型即服务的生态繁荣发展