Stats
Actions
Tags
Help us improve
Share bugs, ideas, or general feedback.
From external-gitcode-ascend-skills
Routes AI for Science requests on Ascend NPU to profiling, model migration, or TensorFlow/PyTorch workflows based on user input.
npx claudepluginhub ascend-ai-coding/awesome-ascend-skills --plugin mindspeed-llm-train-profilerHow this skill is triggered — by the user, by Claude, or both
Slash command
/external-gitcode-ascend-skills:ai4s-mainThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
本 Skill 只负责路线判断和子 skill 分流，不展开具体迁移或调优细节。
Migrates CUDA PyTorch/TensorFlow/vLLM projects to Huawei Ascend NPU: environment checks, code analysis, auto migration, manual adaptation, distributed changes, validation, and skill creation.
Guides Next.js Cache Components and Partial Prerendering (PPR): 'use cache' directives, cacheLife(), cacheTag(), revalidateTag() for caching, invalidation, static/dynamic optimization. Auto-activates on cacheComponents: true.
Migrates code, prompts, and API calls from Claude Sonnet 4.0/4.5 or Opus 4.1 to Opus 4.5, updating model strings on Anthropic, AWS, GCP, Azure platforms.
Share bugs, ideas, or general feedback.
本 Skill 只负责路线判断和子 skill 分流，不展开具体迁移或调优细节。 当用户只给出一个宽泛的 AI for Science 需求、模型名、TensorFlow/Keras 项目，或只说“帮我迁到昇腾/采集 profiling”时，先从这里判断进入哪个子 skill。
| 方向 | 进入条件 | 推荐子 Skill |
|---|---|---|
| Profiling 采集 | 代码已经能训练或推理，只需要采集 trace、分析热点算子、调用栈、内存或瓶颈 | ai4s-profiling |
| 模型迁移 | 已知模型名，或要把 AI4S 模型从 GPU/CUDA 迁移到昇腾 NPU | ai4s-basic 或模型专属 skill |
| TF 框架 | 原项目是 TensorFlow/Keras，需要决定保留 TensorFlow 还是改写到 PyTorch | ascend-tf-community / tf-to-pytorch |
| 模型或任务 | 进入的 Skill | 说明 |
|---|---|---|
| Boltz2 | boltz2 | 蛋白结构预测与端到端推理复现 |
| BoltzGen | boltzgen | 生成式蛋白设计与逆折叠 |
| DeepFRI，保留 TensorFlow | deepfri-tf-npu | 保留 TF 运行时和原始实现 |
| DeepFRI，迁移到 PyTorch | deepfri | 做 TF 到 PyTorch 改写与权重转换 |
| DiffSBDD | diffsbdd | 结构化药物设计与扩散推理 |
| GENERator | generator | DNA 序列生成模型迁移 |
| OligoFormer | oligoformer | siRNA 效能预测与 RNA-FM 依赖适配 |
| ProteinBERT | proteinbert | 蛋白语言模型权重转换、embedding 与微调 |
| 未沉淀的新模型 | ai4s-basic | 先走通用迁移流程，再沉淀模型专属 skill |
torch_npu 生态、统一训练推理流程，进入 tf-to-pytorch。