ml-training-recipes
仓库创建 2026年3月17日最近提交 2 个月前SkillHot 收录 20 天前
▸ 精选理由
集成生产级训练模式与调试技巧，便于快速落地训练
▸ 风险提示
部分配置需大量算力或特定 GPU 环境
这个 Skill 做什么
提供面向多领域的 PyTorch 训练实战模板与调参建议。
提供面向各类任务的 PyTorch 训练模板和调参实战建议，涵盖 LLMs、vision、diffusion、医疗影像等领域。适合训练或微调模型、排查 loss 突变或 OOM、选优化器（AdamW、Muon）、调学习率与混合精度时参考。亮点是基于生产级代码库和系统化实验流程，能把研究思路迅速转成可复现的训练脚本与调试步骤。
▸ 展开 SKILL.md 英文原文
Battle-tested PyTorch training recipes for all domains — LLMs, vision, diffusion, medical imaging, protein/drug discovery, spatial omics, genomics. Covers training loops, optimizer selection (AdamW, Muon), LR scheduling, mixed precision, debugging, and systematic experimentation. Use when training or fine-tuning neural networks, debugging loss spikes or OOM, choosing architectures, or optimizing GPU throughput.
1.5k
Stars
104
Forks
16
仓库内 Skill
+0
7 日增星
安装 / 使用
给你的 Agent 一句话（通用）
帮我安装这个 skill：https://raw.githubusercontent.com/OpenRaiser/NanoResearch/main/skills/vendor-ai-research/ml-training-recipes/SKILL.md或 curl 直取 SKILL.md
curl -fsSL "https://raw.githubusercontent.com/OpenRaiser/NanoResearch/main/skills/vendor-ai-research/ml-training-recipes/SKILL.md"SKILL.MD 节选查看完整文件 ↗
# ML Training Recipes Battle-tested patterns for PyTorch training across domains. Drawn from production codebases (Karpathy's autoresearch/nanochat, torchvision, HuggingFace) and modern training practice. ## Reference files (read when needed) - `references/architecture.md` — Transformer/LLM architecture code patterns, weight init - `references/optimizers.md` — Muon, AdamW hybrid, per-group LR, compiled optimizer steps - `references/domain-specific.md` — Vision, diffusion, contrastive, distributed, checkpointing, data loading - `references/scaling-and-selection.md` — Scaling laws, compute budget tables, decision trees, DGX Spark - `references/biomedical.md` — Drug discovery, pro
via SKILL·HOT · 数据来自 GitHub 公开信息 · 原文版权归作者所有