Ecosyste.ms: Timeline

Browse the timeline of events for every public repo on GitHub. Data updated hourly from GH Archive.

zRzRzRzRzRzRzR

zRzRzRzRzRzRzR closed an issue on THUDM/CogVideo
Improved convenience of prompt input
### Feature request / 功能建议 This is not a feature of CogVideo, but it is a Tips to improve the convenience of prompt input. By including the example below in the examples for the AI, it will tran...
zRzRzRzRzRzRzR closed an issue on THUDM/CogVideo
卡数增加，loss逐渐变大
超级棒的工作！！！在finetune的时候遇到一些问题，想请教下，在训练的时候使用1e-5学习率，40卡训练还正常，loss比较稳定；但是换成100卡就开始loss增大，逐渐发散，想问下这种情况可以怎么改善吗？
zRzRzRzRzRzRzR created a comment on an issue on THUDM/CogVideo
@zhipuch

View on GitHub

zRzRzRzRzRzRzR created a comment on an issue on THUDM/CogVideo
emm，我想知道你的diffusers版本是不是源代码安装，且你的模型仓库是否是最新的HF权重，sample height 和width是300吗。

View on GitHub

zRzRzRzRzRzRzR pushed 4 commits to CogVideoX_dev THUDM/CogVideo
  • Merge pull request #507 from THUDM/CogVideoX_dev diffusers version 2fdc59c
  • [Fix] fix rope temporal patch size 2fb763d
  • Merge pull request #538 from spacegoing/fix_rope_finetune_shape [Fix] fix rope temporal patch size d82922c
  • Merge pull request #556 from THUDM/main new announced f1a2b48

View on GitHub

zRzRzRzRzRzRzR closed a pull request on THUDM/CogVideo
new announced
zRzRzRzRzRzRzR opened a pull request on THUDM/CogVideo
new announced
zRzRzRzRzRzRzR created a comment on an issue on THUDM/CogVideo
仓库更新了吗，模型仓库，检查 transformers中config是不是300 300还是 96 170

View on GitHub

zRzRzRzRzRzRzR deleted a branch zRzRzRzRzRzRzR/transformers

glm-4-1108

zRzRzRzRzRzRzR pushed 2 commits to glm-4-1108 zRzRzRzRzRzRzR/transformers
  • use q_rot dd47bb0
  • Merge branch 'glm-4-1108' of github.com:zRzRzRzRzRzRzR/transformers into glm-4-1108 1ae053c

View on GitHub

zRzRzRzRzRzRzR pushed 4 commits to glm-4-1108 zRzRzRzRzRzRzR/transformers
  • VideoLLaVA: add default values (#34916) add default values 73b4ab1
  • Skipping aqlm non working inference tests till fix merged (#34865) 0e805e6
  • [Whisper] Fix whisper integration tests (#34111) * fix test_tiny_timestamp_generation * fix test_large_timestamp_... 4d1d0f2
  • Merge branch 'huggingface:main' into glm-4-1108 73614df

View on GitHub

zRzRzRzRzRzRzR pushed 1 commit to glm-4-1108 zRzRzRzRzRzRzR/transformers

View on GitHub

zRzRzRzRzRzRzR pushed 1 commit to glm-4-1108 zRzRzRzRzRzRzR/transformers

View on GitHub

zRzRzRzRzRzRzR created a comment on an issue on THUDM/CogVideo
The gradio demo has not been upgraded to 1.5 because the 1.5 model itself is very slow in inference, and it takes 15 minutes even without using any memory-saving work (please check our readme).

View on GitHub

zRzRzRzRzRzRzR created a comment on an issue on THUDM/CogVideo
Yes, but this was not done using the native capabilities of our CogVideoX model; it used our T2V capabilities. This should be an engineering operation, not a function of the model's original struct...

View on GitHub

zRzRzRzRzRzRzR created a comment on a pull request on huggingface/transformers
This modification should meet the requirements, and I have tried to remove all unnecessary code. The remaining code is all the code that will be used.

View on GitHub

zRzRzRzRzRzRzR pushed 1 commit to glm-4-1108 zRzRzRzRzRzRzR/transformers
  • fix wrong change of examples/research_projects 8703374

View on GitHub

zRzRzRzRzRzRzR pushed 25 commits to main zRzRzRzRzRzRzR/transformers
  • Fix: Enable prefill phase key value caching of nemotron/minitron models (#34742) * modeling nemotron kv caching bugf... 318fe25
  • Add safe_globals to resume training on PyTorch 2.6 (#34632) Starting from version 2.4 PyTorch introduces a stricter ... 1339a14
  • Cache: init empty cache when `use_cache` (#34274) * fix * fix tests * fix copies * add docs * Revert "ad... c1a8520
  • BLIP: fix generation after hub update (#34876) * fix blip generation * dont remove it yet * Update src/transfo... 098962d
  • [`Deberta/Deberta-v2`] Refactor code base to support compile, export, and fix LLM (#22105) * some modification for r... 857d46c
  • 🔴 Mllama: fix base prefix (#34874) fix base prefix 1e492af
  • Sum gathered input tokens (#34554) * sum gathered input tokens * ruff line-length is 119, format the code ----... 4dc1a69
  • allow unused input parameters passthrough when chunking in asr pipelines (#33889) * allow unused parameter passthrou... a0f4f31
  • prepare_fa2_from_position_ids function bugfix (#33269) contiguous() is called before view() for key and value within... c50b567
  • Bump tornado from 6.4.1 to 6.4.2 in /examples/research_projects/visual_bert (#34887) Bump tornado in /examples/resea... 62ab94d
  • chore: fix some typos (#34891) Signed-off-by: wanxiangchwng <[email protected]> 97514a8
  • Fix convert_tokens_to_string when decoder is None (#34569) * Fix convert_tokens_to_string when decoder is None * ... 74db22f
  • [`peft`] Given that `self.active_adapter` is deprecated, avoid using it (#34804) * Given that self.active_adapter is... 11cc229
  • Fix Qwen2 failing tests (#34819) * fix: qwen2 model ids * fix: line * fix: more format * update: reformat f4c04ba
  • Bump tornado from 6.4.1 to 6.4.2 in /examples/research_projects/lxmert (#34917) Bumps [tornado](https://github.com/t... 1de3598
  • Rename OLMo November to OLMo2 (#34864) * Rename/move OLMo Nov files to OLMo2 * Rename Olmo1124 and its variants t... 9121ab8
  • Fix : BitNet tests (#34895) * fix_tests_bitnet * fix format 4e6b19c
  • [AWQ, CI] Bump AWQ version used in docker image (#34922) The old AWQ version is failing with the latest (unreleased)... b13916c
  • fix static cache data type miss-match (#34799) * fix gptj data type missmatch Signed-off-by: jiqing-feng <jiqing.... a464afb
  • Fix `test_auto_backbone_timm_model_from_pretrained` (#34877) fix Co-authored-by: ydshieh <[email protected]... a830df2
  • and 5 more ...

View on GitHub

zRzRzRzRzRzRzR pushed 1 commit to glm-4-1108 zRzRzRzRzRzRzR/transformers
  • remove partial_rotary_factor of function partial_rotary_factor 0c44372

View on GitHub

zRzRzRzRzRzRzR pushed 26 commits to glm-4-1108 zRzRzRzRzRzRzR/transformers
  • Fix: Enable prefill phase key value caching of nemotron/minitron models (#34742) * modeling nemotron kv caching bugf... 318fe25
  • Add safe_globals to resume training on PyTorch 2.6 (#34632) Starting from version 2.4 PyTorch introduces a stricter ... 1339a14
  • Cache: init empty cache when `use_cache` (#34274) * fix * fix tests * fix copies * add docs * Revert "ad... c1a8520
  • BLIP: fix generation after hub update (#34876) * fix blip generation * dont remove it yet * Update src/transfo... 098962d
  • [`Deberta/Deberta-v2`] Refactor code base to support compile, export, and fix LLM (#22105) * some modification for r... 857d46c
  • 🔴 Mllama: fix base prefix (#34874) fix base prefix 1e492af
  • Sum gathered input tokens (#34554) * sum gathered input tokens * ruff line-length is 119, format the code ----... 4dc1a69
  • allow unused input parameters passthrough when chunking in asr pipelines (#33889) * allow unused parameter passthrou... a0f4f31
  • prepare_fa2_from_position_ids function bugfix (#33269) contiguous() is called before view() for key and value within... c50b567
  • Bump tornado from 6.4.1 to 6.4.2 in /examples/research_projects/visual_bert (#34887) Bump tornado in /examples/resea... 62ab94d
  • chore: fix some typos (#34891) Signed-off-by: wanxiangchwng <[email protected]> 97514a8
  • Fix convert_tokens_to_string when decoder is None (#34569) * Fix convert_tokens_to_string when decoder is None * ... 74db22f
  • [`peft`] Given that `self.active_adapter` is deprecated, avoid using it (#34804) * Given that self.active_adapter is... 11cc229
  • Fix Qwen2 failing tests (#34819) * fix: qwen2 model ids * fix: line * fix: more format * update: reformat f4c04ba
  • Bump tornado from 6.4.1 to 6.4.2 in /examples/research_projects/lxmert (#34917) Bumps [tornado](https://github.com/t... 1de3598
  • Rename OLMo November to OLMo2 (#34864) * Rename/move OLMo Nov files to OLMo2 * Rename Olmo1124 and its variants t... 9121ab8
  • Fix : BitNet tests (#34895) * fix_tests_bitnet * fix format 4e6b19c
  • [AWQ, CI] Bump AWQ version used in docker image (#34922) The old AWQ version is failing with the latest (unreleased)... b13916c
  • fix static cache data type miss-match (#34799) * fix gptj data type missmatch Signed-off-by: jiqing-feng <jiqing.... a464afb
  • Fix `test_auto_backbone_timm_model_from_pretrained` (#34877) fix Co-authored-by: ydshieh <[email protected]... a830df2
  • and 6 more ...

View on GitHub

zRzRzRzRzRzRzR created a comment on an issue on THUDM/CogVideo
是，22帧算 和21帧算都可以，用22帧算吧

View on GitHub

zRzRzRzRzRzRzR pushed 2 commits to main THUDM/GLM-4
  • correct compute_metrics function 7a14e5d
  • Merge pull request #655 from zhipuch/main correct compute_metrics function c23abb0

View on GitHub

zRzRzRzRzRzRzR created a comment on an issue on THUDM/GLM-4
微调吗，默认是截断

View on GitHub

zRzRzRzRzRzRzR created a comment on an issue on huggingface/transformers
We have observed that many open-source frameworks, such as [xinference](https://github.com/xorbitsai/inference) and [vLLM](https://github.com/vllm-project/vllm), currently do not support the `Glmmo...

View on GitHub

zRzRzRzRzRzRzR pushed 1 commit to main MetaGLM/LawGLM

View on GitHub

zRzRzRzRzRzRzR created a comment on an issue on THUDM/CogVideo
Please set the default height and width, otherwise it will be 2400 * 2400 (in order to achieve different I2V, for normal operation, if you are t2v, please set weight = 1360, height = 768 to work pr...

View on GitHub

zRzRzRzRzRzRzR created a comment on an issue on THUDM/CogVideo
The computational load of CogVideoX-5B is about one-fourth of CogVideoX1.5-5B. Therefore, you should multiply the time of CogVideoX-5B by 4. If your generation time is 10 seconds, then you also nee...

View on GitHub

zRzRzRzRzRzRzR created a comment on an issue on THUDM/CogVideo
Oh, this issue is likely because you haven't installed the latest main branch of the diffusers library, please pull the latest diffusers and install it.

View on GitHub

zRzRzRzRzRzRzR closed an issue on THUDM/CogVideo
Why does the result of t2v model so terrible？
### System Info / 系統信息 Package Version Editable project location ------------------------ ------------- ----------------------------------------------- accelerate ...
zRzRzRzRzRzRzR pushed 10 commits to glm-4-1108 zRzRzRzRzRzRzR/transformers
  • Watermarking: fix order (#34849) fix watermarking order 6a912ff
  • Update checks for torch.distributed.tensor to require torch >= 2.5 (#34816) * Update checks for torch.distributed.te... 1867be6
  • Remove quantization related config from dequantized model (#34856) * Remove quantization related config from dequant... d9e6f30
  • Auto compile when static cache (#34247) * generate with compile * nits * simple * generate with compile ... 597efd2
  • Speculative decoding: Test the target distribution (to prevent issues like #32867) (#34553) * Update test_utils.py ... 42b36d7
  • smol improvements to support more flexible usage (#34857) * smol improvements to support more flexible usage * ruff 861758e
  • [CI] Skip EETQ tests while package is broken with latest transformers (#34854) * CI Skip EETQ tests while package is... 286ffaa
  • Bitnet test fix to avoid using gated model (#34863) small test fix 54be2d7
  • Fix support for image processors modifications in modular (#34866) * add fix and examples * fix camel case naming 3a8eb74
  • Merge branch 'huggingface:main' into glm-4-1108 c57cd93

View on GitHub

Load more