xgb-modeling
OfficialTrain XGBoost models with robust financial-grade evaluation
Data & Analytics#model evaluation#xgboost#feature selection#auc#binary classification#ks#psi stability
Authoraliyun
Version1.0.0
Installs0
System Documentation
What problem does it solve?
When you need to build a binary classification model quickly and convincingly, this Skill turns your chosen feature set into an XGBoost training workflow with standardized performance, stability, and model card reporting.
Core Features & Use Cases
- Multi-feature-scheme modeling: run one feature list, run multiple feature sets, or generate feature schemes via IV/PSI/correlation/statistics selectors, then compare results side-by-side.
- Decision-grade evaluation: produce AUC/KS/Gini plus business-oriented Lift and Bad Capture Rate metrics, with PSI and stability analysis across time windows.
- Portable, no-platform-coupling workflow: outputs a structured result.json manifest and a human-readable modeling report in one run for downstream skills.
Quick Start
Train an XGBoost baseline on your dataset and get a full report by telling the AI: "Use xgb-modeling to train a binary model with --data_path ./examples/toy.parquet, --target y_label, and --output_dir ./outputs/run1."
Dependency Matrix
Required Modules
xgboostscikit-learnoptbinningpandasnumpyscipy
Components
scripts
💻 Claude Code Installation
Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.
Please help me install this Skill: Name: xgb-modeling Download link: https://github.com/aliyun/qwen-dianjin/archive/main.zip#xgb-modeling Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.