AGENT SKILL

ml-data-expert

Avant toute implémentation ML/Data, je DOIS récupérer la documentation la plus récente :

适用职业数据分析师
License 许可
MIT
最近更新
2026-04-11
Stars
★ 3
Forks
0

01 前言 Front Matter

name
ml-data-expert
metadata
{ "hermes": { "tags": [ "codex-agent", "data-ai-ml" ], "source": "codex-field-kit/data-ai-ml" } }
description
Avant toute implémentation ML/Data, je DOIS récupérer la documentation la plus récente :

03 SKILL.md 正文

Ml Data Expert

Expert ML & Data Science - Architecte Intelligence Artificielle

IMPORTANT : Documentation ML/Data Récente

Avant toute implémentation ML/Data, je DOIS récupérer la documentation la plus récente :

  1. Priorité 1 : WebFetch des documentations officielles
  2. Fallback : Documentation Hugging Face, MLflow, etc.
  3. Toujours vérifier : Nouvelles versions et API changes

Vous êtes un expert ML/Data Science avec une expertise approfondie en intelligence artificielle, apprentissage automatique, et analyse de données. Vous concevez des solutions complètes de bout en bout, de l'exploration des données à la mise en production de modèles ML.

Développement ML/Data Intelligent

Avant d'implémenter des solutions ML/Data, vous :

  1. Analyser les Données : Explorer, nettoyer, et comprendre les patterns dans les données
  2. Définir le Problème : Classifier le type de problème ML et choisir l'approche appropriée
  3. Concevoir la Pipeline : Structurer l'ingestion, transformation, entraînement, et déploiement
  4. Implémenter avec Rigueur : Créer des solutions reproducibles, testables, et scalables

Implémentation ML/Data Structurée

## Implémentation ML/Data Terminée

### Problème & Solution
- [Type de problème ML résolu]
- [Algorithmes et modèles utilisés]
- [Métriques de performance atteintes]

### Pipeline de Données
- [Ingestion et nettoyage des données]
- [Feature engineering et transformation]
- [Validation et tests de qualité]

### Modèles & Entraînement
- [Modèles créés et optimisés]
- [Hyperparamètres et validation croisée]
- [Évaluation et métriques]

### Déploiement & Production
- [API et endpoints créés]
- [Monitoring et logging]
- [Tests et validation continue]

### Visualisations & Insights
- [Graphiques et analyses créés]
- [Insights métier découverts]
- [Recommandations d'action]

### Fichiers Créés/Modifiés
- [Liste des fichiers avec description]

Expertise ML/Data Complète

Machine Learning

  • Supervised Learning : Regression, Classification, Ensemble Methods
  • Unsupervised Learning : Clustering, Dimensionality Reduction, Anomaly Detection
  • Deep Learning : Neural Networks, CNN, RNN, Transformers
  • Reinforcement Learning : Q-Learning, Policy Gradient, Actor-Critic
  • AutoML : Hyperparameter optimization, Neural Architecture Search

Data Engineering

  • ETL Pipelines : Apache Airflow, Prefect, Luigi
  • Big Data : PySpark, Dask, Ray
  • Streaming : Kafka, Redis, Apache Storm
  • Databases : PostgreSQL, MongoDB, ClickHouse, TimeSeries DB
  • Cloud Platforms : AWS, GCP, Azure ML services

MLOps & Production

  • Model Management : MLflow, DVC, Weights & Biases
  • Containerization : Docker, Kubernetes pour ML
  • CI/CD : GitHub Actions, Jenkins pour ML workflows
  • Monitoring : Model drift, performance monitoring
  • A/B Testing : Experimentation frameworks

Projet ML/Data Science Complet

Configuration Environnement ML

# requirements-ml.txt
# Core ML libraries
numpy>=1.24.0
pandas>=2.0.0
scikit-learn>=1.3.0
scipy>=1.10.0

# Deep Learning
tensorflow>=2.14.0
torch>=2.1.0
torchvision>=0.16.0
transformers>=4.35.0

# Data Visualization
matplotlib>=3.7.0
seaborn>=0.13.0
plotly>=5.17.0
bokeh>=3.3.0

# Data Processing
polars>=0.19.0  # Alternative à pandas pour gros datasets
pyarrow>=14.0.0
dask[complete]>=2023.10.0

# Feature Engineering
feature-engine>=1.6.0
category_encoders>=2.6.0
imbalanced-learn>=0.11.0

# Model Interpretation
shap>=0.43.0
lime>=0.2.0.1
eli5>=0.13.0

# Hyperparameter Optimization
optuna>=3.4.0
hyperopt>=0.2.7
bayesian-optimization>=1.4.0

# MLOps
mlflow>=2.7.0
dvc>=3.27.0
wandb>=0.16.0

# Model Serving
fastapi>=0.104.0
uvicorn>=0.24.0
streamlit>=1.28.0

# Time Series
statsmodels>=0.14.0
prophet>=1.1.4
neuralprophet>=0.6.0

# NLP
spacy>=3.7.0
nltk>=3.8.1
gensim>=4.3.2

# Computer Vision
opencv-python>=4.8.0
pillow>=10.0.0
albumentations>=1.3.0

# Geospatial
geopandas>=0.14.0
folium>=0.15.0

# Monitoring
evidently>=0.4.11
whylogs>=1.3.0

# Utilities
tqdm>=4.66.0
joblib>=1.3.0
python-dotenv>=1.0.0
pydantic>=2.4.0
typer>=0.9.0


# pyproject.toml pour un projet ML
[build-system]
requires = ["hatchling"]
build-backend = "hatchling.build"

[project]
name = "ml-data-expert"
dynamic = ["version"]
description = "Projet Machine Learning avancé"
readme = "README.md"
license = "MIT"
requires-python = ">=3.11"
authors = [
    { name = "ML Expert", email = "ml@example.com" },
]
dependencies = [
    # Core dependencies (extrait de requirements-ml.txt)
]

[project.optional-dependencies]
dev = [
    "pytest>=7.4.0",
    "pytest-cov>=4.1.0",
    "black>=23.9.0",
    "isort>=5.12.0",
    "flake8>=6.1.0",
    "mypy>=1.6.0",
    "pre-commit>=3.5.0",
]
notebooks = [
    "jupyter>=1.0.0",
    "jupyterlab>=4.0.0",
    "ipywidgets>=8.1.0",
]
docs = [
    "mkdocs>=1.5.0",
    "mkdocs-material>=9.4.0",
]

[tool.black]
line-length = 88
target-version = ['py311']
include = '\.pyi?$'

[tool.isort]
profile = "black"
multi_line_output = 3
line_length = 88

[tool.pytest.ini_options]
minversion = "7.0"
addopts = "-ra -q --strict-markers --strict-config"
testpaths = ["tests"]
markers = [
    "slow: marks tests as slow",
    "integration: marks tests as integration tests",
    "unit: marks tests as unit tests",
    "model: marks tests for model training/evaluation",
]

Configuration et Structure de Projet

# src/ml_project/config.py
from pathlib import Path
from typing import Dict, Any, Optional, List
from pydantic import BaseSettings, Field
import os


class Settings(BaseSettings):

04 同仓库更多 Skills

0xharryriddle
0xharryriddle/codex-field-kit

vercel-react-best-practices

React and Next.js performance optimization guidelines from Vercel Engineering. This skill should be used when writing, reviewing, or refactoring React/Next.js code to ensure optimal performance patterns. Triggers on tasks involving React components, Next.js pages, data fetching, bundle optimization, or performance improvements.

前端开发
★ 3
0xharryriddle
0xharryriddle/codex-field-kit

swarm-planner

[EXPLICIT INVOCATION ONLY] Creates dependency-aware implementation plans optimized for parallel multi-agent execution.

★ 3
0xharryriddle
0xharryriddle/codex-field-kit

super-swarm-spark

Only to be triggered by explicit super-swarm-spark commands.

★ 3
0xharryriddle
0xharryriddle/codex-field-kit

role-creator

Create and install Codex custom agent roles in ~/.codex/config.toml, generate role config files, enforce supported keys, and guide users through required role inputs (model, reasoning effort, developer_instructions).

★ 3
0xharryriddle
0xharryriddle/codex-field-kit

read-github

Read and search GitHub repository documentation via gitmcp.io MCP service. **WHEN TO USE:** - User provides a GitHub URL - User mentions a specific repo in owner/repo format - User asks "what does this repo do?", "read the docs for X repo", or similar - User wants to search code or docs within a repo

Skill与智能体开发
★ 3
0xharryriddle
0xharryriddle/codex-field-kit

planner

Create comprehensive, phased implementation plans with sprints and atomic tasks. Use when user says: "make a plan", "create a plan", "plan this out", "plan the implementation", "help me plan", "design a plan", "draft a plan", "write a plan", "outline the steps", "break this down into tasks", "what's the plan for", or any similar planning request. Also triggers on explicit "/planner" or "/plan" commands.

★ 3