binmakeswell
binmakeswell
### Describe the feature Fewer users are able to train large models directly from scratch. We need to provide an example of fine-tuning. For example, how to load pre-training parameters...
### 🐛 Describe the bug Delete redundant code, eg. test code to 2D selfattention, which is [misleading](https://github.com/hpcaitech/ColossalAI/issues/602). ### Environment _No response_
### 🐛 Describe the bug Add clear preventive error message for users, eg. [num heads is not divisible](https://github.com/hpcaitech/ColossalAI/issues/602) and [batch size](https://github.com/hpcaitech/ColossalAI/issues/524), etc. ### Environment _No response_
[Colossal-AI](https://github.com/hpcaitech/ColossalAI) is a unified deep learning system for the big model era, which integrates many efficient techniques like multi-dimensional tensor parallelism, sequence parallelism, heterogeneous memory management, etc. By using Colossal-AI,...
### Describe the feature An interesting application can attract a lot of attention. https://github.com/openai/CLIP https://github.com/openai/DALL-E DALL·E 2: https://cdn.openai.com/papers/dall-e-2.pdf https://mp.weixin.qq.com/s/FN9KjuHot547fa68usS98Q
# Colossal-AI 贡献指南 欢迎来到开源摘星计划-Colossal-AI 贡献专栏 。[Colossal-AI](https://github.com/hpcaitech/ColossalAI) 已加入开源摘星计划，该计划提供奖励以鼓励你加入我们的社区：所有参与本项目贡献的开发者都可获得相应贡献积分、奖励以及被评选为年度摘星100 人的机会，活动及奖励详情请见：[开源摘星计划](https://github.com/weopenprojects/WeOpen-Star)。 📅活动时间：2022年5月-12月 ## 🌟关于 Colossal-AI > AI模型在几年内已增大万倍，远超硬件能力数倍的增长，如何高效利用分布式技术实现AI大模型的并行训练加速已成为所有与计算相关行业的关键痛点。 [Colossal-AI](https://github.com/hpcaitech/ColossalAI)通过高效多维并行、异构内存管理、大规模优化库、自适应任务调度等方式，仅需几行代码，便可与已有项目结合，高效快速部署AI大模型训练，兼容低端计算设备，可显著缩短AI大模型训练时间、降低训练成本，减少学习和部署的人力成本，性能全面超越已有解决方案，如微软 DeepSpeed和英伟达Megatron-LM。 Colossal-AI上线一周便在全球最大开源社区GitHub登顶Python（人工智能）方向世界第一，随着版本更新，数月内多次再登GitHub热榜世界第一，目前已有GitHub Star3千多颗，开源社区发展速度超过同期竞品。项目核心团队均来自国际一流高校和科技巨头，已与多家世界知名企业建立合作。 ### **以下是项目可参与的方向：** 1. 我们列出了适合首次加入 Colossal-AI社区的贡献者的问题、适合对项目有足够了解的贡献者的特殊 issue 挑战（参见下面的列表）以及一些社区日常任务。如果你愿意解决，请点击该 issue 并在评论区留言“WeOpen Star”（日常及其他任务无需此操作）。提交 PR 时，请同时将...
- add optional affiliation information tips
Thank you for your rapid and outstanding contribution to Stable Diffusion 2.0! AIGC has recently risen to be one of the hottest topics in AI. Unfortunately, large hardware requirements and...
**Is your feature request related to a problem? Please describe.** AIGC has recently risen to be one of the hottest topics in AI. Unfortunately, large hardware requirements and training costs...