菜单
Run Echo-Memory memory-baseline and context training recipes on Wan 2.1 1.3B. Use when training spatial/SSM/compression/context rows, editing train/*.sh launchers, or configuring DATASET_BASE_PATH for static or dynamic pools.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/Echo-Team-Joy-Future-Academy-JD/Echo-Memory --skill echo-memory-train命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Publish Echo-Memory GitHub Pages, update bilingual docs/i18n, and align doc/checkpoints.md with Hugging Face Echo-Team/Echo-Memory. Use for gh-pages releases, project page edits, or maintainer checkpoint documentation.
Run Echo-Memory replay, in-domain loop/revisit, and open-domain eval v2 scripts. Use when checking HF checkpoints, editing eval/v2, or tracing CKPT → memory profile via env/memory_baseline_runtime.py.
Scope Cursor Agent prompts for Echo-Memory development — memory families, entry scripts, project skills, and public-repo constraints. Use when vibe coding, writing .cursor/rules, or planning multi-file agent tasks in this repo.
基于 SOC 职业分类
正在显示 SKILL.md
| name | echo-memory-train |
| description | Run Echo-Memory memory-baseline and context training recipes on Wan 2.1 1.3B. Use when training spatial/SSM/compression/context rows, editing train/*.sh launchers, or configuring DATASET_BASE_PATH for static or dynamic pools. |
export WAN_BASE_MODEL=/path/to/Wan2.1-T2V-1.3B
export DATASET_BASE_PATH=data/Context-as-Memory-Dataset # static in-domain pool
export PYTHONPATH=$PWD:${PYTHONPATH:-}
export OUTPUT_BASE_ROOT=$PWD/outputs
data/Context-as-Memory-Dataset; see doc/dataset_preprocessing.mddata/dynamic-memory-dataset; see doc/dynamic_dataset_preprocessing.md| Family | Directory | Example |
|---|---|---|
| Spatial / SSM / compression | train/memory_baselines_basic/ | run_spatial_memory_baseline.sh, run_ablation_block_wise_ssm_two_chunk.sh |
| Context K=1/5/20 | train/context_learning/ | run_pre_qkv_ctx1.sh, run_pre_qkv_ctx20.sh |
Run from repository root: bash train/memory_baselines_basic/run_spatial_memory_baseline.sh
Shared env: train/_shared/common_env_memory.sh — no private paths baked in.
DATASET_BASE_PATH to the correct training pool name in docs (Echo terms only in public markdown).run_*.sh patterns over new one-off Python entrypoints.outputs/ unless OUTPUT_BASE_ROOT is set.data/, outputs/, checkpoints, or machine-local paths.train/README.md — launcher indextrain/memory_baselines_basic/README.md — ablation set