skills.rest
Ayoub-ouederni
Ayoub-ouederniCOMMUNITY
Explore all skills by this author →
GitHub

train-model

Train an XGBoost AML model on the Elliptic dataset with cross-validation and evaluation.

Updated Apr 3, 2026
Search Tags:#pytest#machine learning#model training#cross-validation#xgboost#aml#elliptic dataset
One-click install
npx skills add https://github.com/Ayoub-ouederni/SENTINEL --skill train-model-ayoub-ouederni
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: train-model
Source: https://github.com/Ayoub-ouederni/SENTINEL/tree/main/.claude/skills/train-model
Command: npx skills add https://github.com/Ayoub-ouederni/SENTINEL --skill train-model-ayoub-ouederni

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Building an anti-money-laundering risk scoring model requires a full ML pipeline: dataset acquisition, feature engineering, model training, and evaluation. This Skill automates that entire workflow for the SENTINEL decentralized AML protocol. ## Core Features & Use Cases - Dataset Preparation: Downloads the Elliptic dataset if not present and performs preprocessing with feature engineering across 5 risk dimensions. - Model Training & Evaluation: Trains an XGBoost classifier with 5-fold cross-validation and computes precision, recall, F1, and AUC-ROC metrics. - Export & Testing: Exports the trained model as a pickle file and runs the pytest test suite. - Use Case: A developer working on the SENTINEL protocol needs a fresh AML scoring model after updating features; invoke this Skill to retrain, evaluate, and export the model in one step. ## Quick Start Train the AML model on the Elliptic dataset and export the evaluated XGBoost model as a pickle file.

Frequently Asked Questions about train-model

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I train an XGBoost model for AML detection?▼

Invoke this Skill to download the Elliptic dataset, engineer features across 5 risk dimensions, and train XGBoost with 5-fold cross-validation. The trained model is exported as a pickle file after evaluation.

What dataset is used for training the AML model?▼

The Skill uses the Elliptic dataset, a labeled dataset of Bitcoin transactions commonly used for illicit activity detection. It is downloaded automatically if not already present locally.

Which metrics evaluate the AML model performance?▼

The model is evaluated with precision, recall, F1 score, and AUC-ROC. These metrics are computed after training with 5-fold cross-validation to measure classification quality on imbalanced fraud data.

Does the training workflow run tests after exporting the model?▼

Yes, after exporting the model as a pickle file, the Skill runs the pytest test suite. This validates the training pipeline and model artifacts before downstream use.

What are the limitations of XGBoost for AML scoring?▼

XGBoost depends on the quality and coverage of the labeled Elliptic dataset, so performance may degrade on unseen laundering patterns. Retraining with fresh labeled data is needed as attack behaviors evolve.

Related Skills

Explore More Data & Analytics Skills→
kridaydavekridaydave
1

trove

Builds datasets through sourcing, sampling, labeling, deduplication, splitting, and datasheet documentation.

Community
Advanced
Sarnav07Sarnav07

okx-agentic-wallet

Automate wallet operations across EVM and Solana networks with the Onchain OS Wallet.

Community
Advanced
archyl-comarchyl-com
1

archyl-predict

Forecast architecture risk from Archyl drift, DORA, conformance, and C4 data.

Official
Advanced
ZK-TheoryZK-Theory
1

tda-graph-network-analysis

Constructs and analyzes Mapper graphs, transition graphs, and network metrics for topological data analysis.

Official
Intermediate