Agent Auto Sci AI ML
Use this subskill when the project needs modeling rather than only descriptive analysis.
Fast Workflow
- Define prediction or explanation target.
- Build a data dictionary and leakage checklist.
- Establish simple baselines before complex models.
- Choose model family: classical ML, deep learning, time series, spatial ML, or NLP/transformer.
- Split data correctly: spatial, temporal, group, or external validation as needed.
- Evaluate with task-appropriate metrics and uncertainty.
- Interpret with SHAP, permutation importance, partial dependence, or ablation.
- Translate model results into manuscript claims only after robustness checks.
Read references/ml_ai_workflows.md.
For full paper projects, this skill owns baseline design, model/method selection, leakage checks, train/test split logic, hyperparameter search, ablation, robustness, error-case analysis, XAI, and methods/results reporting boundaries. It must work with agent-auto-sci-methodology before turning feature importance into mechanism language.
For deeper K-Dense-style encapsulation:
references/k_dense_ml_ai_mapping.md: how scikit-learn, PyTorch Lightning, transformers, SHAP, time-series, and XAI skills are adapted.references/sport_geography_ml_playbook.md: sport geography modeling patterns, leakage checks, baselines, and manuscript reporting.
Related Helper Skills
Use installed helper skills when useful:
scikit-learnpytorch-lightningtransformersshapaeontimesfm-forecastingstatistical-analysisagent-auto-sci-data-vizagent-auto-sci-geospatial
Must Not Do
- Do not use ML if the research question only needs transparent spatial/statistical analysis.
- Do not report accuracy without baselines and uncertainty.
- Do not mix train/test spatial units when spatial leakage is possible.
- Do not use SHAP as causal evidence.
1---2name: agent-auto-sci-ai-ml3description: "机器学习与人工智能科研子 skill。用于地理学+体育学项目中的 classical ML、深度学习、时间序列、模型评估、数据泄露检查、基线设计、超参数搜索、SHAP/可解释 AI、公平性诊断、实验复现和论文方法写作。触发于 machine learning、AI、scikit-learn、PyTorch Lightning、transformers、SHAP、model evaluation、XAI、spatial ML、time series 等任务。"4metadata:5 source: https://skillmd.com/skills/lzy599775/agent-auto-sci-ai-ml6---78# Agent Auto Sci AI ML910Use this subskill when the project needs modeling rather than only descriptive analysis.1112## Fast Workflow13141. Define prediction or explanation target.152. Build a data dictionary and leakage checklist.163. Establish simple baselines before complex models.174. Choose model family: classical ML, deep learning, time series, spatial ML, or NLP/transformer.185. Split data correctly: spatial, temporal, group, or external validation as needed.196. Evaluate with task-appropriate metrics and uncertainty.207. Interpret with SHAP, permutation importance, partial dependence, or ablation.218. Translate model results into manuscript claims only after robustness checks.2223Read `references/ml_ai_workflows.md`.2425For full paper projects, this skill owns baseline design, model/method selection, leakage checks, train/test split logic, hyperparameter search, ablation, robustness, error-case analysis, XAI, and methods/results reporting boundaries. It must work with `agent-auto-sci-methodology` before turning feature importance into mechanism language.2627For deeper K-Dense-style encapsulation:2829- `references/k_dense_ml_ai_mapping.md`: how scikit-learn, PyTorch Lightning, transformers, SHAP, time-series, and XAI skills are adapted.30- `references/sport_geography_ml_playbook.md`: sport geography modeling patterns, leakage checks, baselines, and manuscript reporting.3132## Related Helper Skills3334Use installed helper skills when useful:3536- `scikit-learn`37- `pytorch-lightning`38- `transformers`39- `shap`40- `aeon`41- `timesfm-forecasting`42- `statistical-analysis`43- `agent-auto-sci-data-viz`44- `agent-auto-sci-geospatial`4546## Must Not Do4748- Do not use ML if the research question only needs transparent spatial/statistical analysis.49- Do not report accuracy without baselines and uncertainty.50- Do not mix train/test spatial units when spatial leakage is possible.51- Do not use SHAP as causal evidence.52
