Ecosyste.ms: Timeline
Browse the timeline of events for every public repo on GitHub. Data updated hourly from GH Archive.
zRzRzRzRzRzRzR closed an issue on THUDM/CogVideo
CogVideoX1.5-5B-SAT全量微调失败，位置编码部分对不上
### System Info / 系統信息 非硬件问题，代码逻辑有些问题 ### Information / 问题信息 - [X] The official example scripts / 官方的示例脚本 - [ ] My own modified scripts / 我自己修改的脚本和任务 ### Reproduction / 复现过程 启动脚本：cd sat; bash ...
zRzRzRzRzRzRzR closed an issue on THUDM/CogVideo
train_text_to_video_sft 中的设置
你好，请问下# Single GPU 训练时 train_text_to_video_sft.sh ACCELERATE_CONFIG_FILE="accelerate_configs/uncompiled_1.yaml" 训练没有问题。我将uncompiled_1.yaml-->deepspeed.yaml换为deepspeed报错， 企业微信截图_17325422283473 ...
zRzRzRzRzRzRzR created a comment on an issue on THUDM/CogVideo
1. sat部分现在代码没有进行维护，在之后的新模型以及框架，sat方案会完全弃用。 2.diffusers方案会逐渐实现多卡训练，支持目前的所有模型，我们会有人员维护。
zRzRzRzRzRzRzR pushed 3 commits to main MetaGLM/FinGLM2
zRzRzRzRzRzRzR closed an issue on THUDM/CogVideo
prepare_rotary_positional_embeddings
### Feature request / 功能建议 train video with different resolution ### Motivation / 动机 train video with different resolution ### Your contribution / 您的贡献 Hi, Thank you for your nice work aga...
zRzRzRzRzRzRzR closed an issue on THUDM/CogVideo
sampling is different between diffusers and sat ?
Hello, I am currently encountering an issue where the effect becomes noticeably blurry after using conversion code to transform a SAT training model into diffusers. I have pinpointed that the mode...
zRzRzRzRzRzRzR closed an issue on THUDM/CogVideo
经验分享：Windows下安装报错的解决方法
### System Info / 系統信息 Windows 11， Python 3.11, Cuda 12.1 ### Information / 问题信息 - [X] The official example scripts / 官方的示例脚本 - [ ] My own modified scripts / 我自己修改的脚本和任务 ### Reproduction / 复现过程...
zRzRzRzRzRzRzR closed an issue on THUDM/CogVideo
Why is get_velocity applied to model_output during the training of cogvideo?
### Feature request / 功能建议 The common way of writing is: ： ``` if noise_scheduler.config.prediction_type == "epsilon": target = noise elif noise_scheduler.config.prediction_type == "v_pr...
zRzRzRzRzRzRzR created a comment on an issue on THUDM/CogVideo
No, only CogVideoX1.5-I2V is supported; other models must strictly adhere to the resolution mentioned in the readme and cannot use portrait orientation.
zRzRzRzRzRzRzR closed an issue on THUDM/CogVideo
Preparing dataset for training CogVideo1.5 I2V
### System Info / 系統信息 Linux, otherwise N/A ### Information / 问题信息 - [X] The official example scripts / 官方的示例脚本 - [ ] My own modified scripts / 我自己修改的脚本和任务 ### Reproduction / 复现过程 [Here](https...
zRzRzRzRzRzRzR created a comment on an issue on THUDM/CogVideo
The relevant conversion scripts have been updated, you can view them in the diffusers library, and our repository has also updated this part of the code.
zRzRzRzRzRzRzR closed an issue on THUDM/CogVideo
VRAM requirement + Generation Speed for CogVideoX 1.5?
### Feature request / 功能建议 I think it would be helpful to have some info on what to expect when running inference with 1.5. Especially with I2V, since it's supposed to generate at any resolution,...
zRzRzRzRzRzRzR closed an issue on THUDM/CogVideo
Using cogvideox-5b to make text2video inference video with size 9:16 appears blurry
### System Info / 系統信息 diffusers latest version ### Information / 问题信息 - [ ] The official example scripts / 官方的示例脚本 - [ ] My own modified scripts / 我自己修改的脚本和任务 ### Reproduction / 复现过程 ''' imp...
zRzRzRzRzRzRzR closed an issue on THUDM/CogVideo
Improving quality of Image to Video generation CogVideoX1.5-5B
I was wondering if there is any special way to improve the quality of video generated form static images. I followed the suggestion provided on the [example](https://github.com/THUDM/CogVideo/blob/...
zRzRzRzRzRzRzR created a comment on an issue on THUDM/CogVideo
We have not maintained the SAT code for fine-tuning CogVideoX1.5, as we are no longer using this framework; we are maintaining the fine-tuning of diffusers. Can you use the model with diffusers ve...
zRzRzRzRzRzRzR created a comment on an issue on THUDM/CogVideo
我们SAT的微调代码没有适配CogVideoX1.5，你需要使用diffusers的模型和实现来微调哦
zRzRzRzRzRzRzR closed an issue on THUDM/CogVideo
cogvideox-1.5-5b微调报错
### System Info / 系統信息 diffuser: 0.32.0.dev0 python: 3.11 cuda: 12.0 ### Information / 问题信息 - [ ] The official example scripts / 官方的示例脚本 - [ ] My own modified scripts / 我自己修改的脚本和任务 ### Reprod...
zRzRzRzRzRzRzR pushed 3 commits to main THUDM/CogVideo
- [bugfix] fix specific resolution setting Different models use different resolutions, for example, for the CogVideoX1... 92a5892
- [docs] improve help messages in argument parser Fix and clarify help documentation in parser.add_argument() to bette... 1811c50
- Merge pull request #615 from THUDM/CogVideoX_dev Cog video x dev 7935bd5
zRzRzRzRzRzRzR pushed 101 commits to main zRzRzRzRzRzRzR/diffusers
- [Fix] Syntax error (#10068) fix syntax error 784b351
- [CI] Add quantization (#9832) * add quantization to nightly CI. * prep. * fix lib name. * remove deps that ... 827b6c2
- Add `sigmas` to Flux pipelines (#10081) 8d386f7
- Fixed Nits in Evaluation Docs (#10063) Minor fixes and script improvement in evaluation docs. 922c5f5
- fix link in the docs (#10058) * fix link in the docs * fix same issue for ko c44fba8
- fix offloading for sd3.5 controlnets (#10072) * add cd34439
- [Single File] Fix SD3.5 single file loading (#10077) update a9d3f6c
- Fix `num_images_per_prompt>1` with Skip Guidance Layers in `StableDiffusion3Pipeline` (#10086) beb8566
- [Single File] Pass token when fetching interpreted config (#10082) update 6db3333
- Interpolate fix on cuda for large output tensors (#10067) * Workaround for upscale with large output tensors. Fix... 2312b27
- Convert `sigmas` to `np.array` in FlowMatch set_timesteps (#10088) 30f2e9b
- fix: missing AutoencoderKL lora adapter (#9807) * fix: missing AutoencoderKL lora adapter * fix --------- C... 963ffca
- Let server decide default repo visibility (#10047) 0763a7e
- Fix some documentation in ./src/diffusers/models/embeddings.py for demo (#9579) * Fix some documentation in ./src/di... fc72e0f
- Don't stale close-to-merge (#10096) Re: https://github.com/huggingface/diffusers/discussions/10046#discussioncomment... acf79b3
- Add StableDiffusion3PAGImg2Img Pipeline + Fix SD3 Unconditional PAG (#9932) * fix progress bar updates in SD 1.5 PAG... 63b631f
- Notebooks for Community Scripts-4 (#10094) * Add Diffuser Notebooks for Community Scripts. * Add missing link. ... cf25894
- Fix Broken Link in Optimization Docs (#10105) Update broken link. 2be66e6
- DPM++ third order fixes (#9104) * Fix wrong output on 3n-1 steps count * Add sde handling to 3 order * make ... 8ac6de9
- update (#7067) * add data_dir parameter to load_dataset --------- Co-authored-by: Sayak Paul <spsayakpaul@gmai... b58f67f
- and 81 more ...
zRzRzRzRzRzRzR closed an issue on THUDM/GLM-4-Voice
GLM-4-Voice报错
[出错.pdf](https://github.com/user-attachments/files/18090689/default.pdf) 我们使用的是Windows操作系统 1、将main、glm-4-voice-9b、glm-4-voice-decoder、glm-4-voice-tokenizer下载到了本地 2、安装了conda，创建了虚拟环境voice，安装了pynin...