skills.rest
gizix
gizixCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

ml-pipeline

Build end-to-end scikit-learn pipelines for tabular data preprocessing and model training.

2|2|Updated Nov 4, 2025
Search Tags:#production#pipeline#preprocessing#ml-pipeline#feature-engineering#sklearn
One-click install
npx skills add https://github.com/gizix/cc_projects --skill ml-pipeline
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ml-pipeline
Source: https://github.com/gizix/cc_projects/tree/main/jupyter-ml-template/.claude/skills/ml-pipeline
Command: npx skills add https://github.com/gizix/cc_projects --skill ml-pipeline

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you design, implement, and reproduce machine learning pipelines combining preprocessing, modeling, and evaluation in a coherent workflow.

Core Features & Use Cases

  • Pipeline Design: Build scikit-learn Pipelines with preprocessing and models.
  • Column Transformers: Handle numeric and categorical features cleanly.
  • Custom Transformers & Feature Engineering: Extend pipelines with domain-specific logic.
  • Complete Pipeline: End-to-end steps from raw data to model predictions.

Quick Start

Create a simple Pipeline with a StandardScaler and RandomForestClassifier, then fit on training data and predict on test data.

Frequently Asked Questions about ml-pipeline

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

FAQPage Schema
How do I build an end-to-end machine learning pipeline with scikit-learn?▼

Build an end-to-end ML pipeline by chaining preprocessing, feature engineering, and model training steps using scikit-learn's Pipeline and ColumnTransformer. This ensures reproducible, modular workflows from raw tabular data to predictions with automatic data transformation consistency.

Can I handle both numeric and categorical features in a single sklearn pipeline?▼

Yes, use ColumnTransformer to apply different preprocessing steps to numeric and categorical columns separately, then combine them before model training. This keeps your pipeline clean and ensures proper feature handling without data leakage.

How do I add custom feature engineering logic to a scikit-learn pipeline?▼

Create custom transformers by subclassing BaseEstimator and TransformerMixin, then insert them into your Pipeline or ColumnTransformer. This lets you encapsulate domain-specific transformations while maintaining reproducibility and serialization.

What's the best way to perform hyperparameter tuning on a complete ML pipeline?▼

Use GridSearchCV or RandomizedSearchCV with your fitted Pipeline to tune parameters across preprocessing and model steps simultaneously. This ensures hyperparameters are optimized on the actual pipeline transformation sequence, avoiding bias.

How do I save and reload a trained ML pipeline for production?▼

Serialize your sklearn Pipeline using joblib or pickle to persist the entire fitted preprocessing and model chain. Reloading preserves all transformations and model coefficients, enabling consistent predictions on new data in production.

Can I inspect and select important features from my pipeline?▼

Extract feature importances from tree-based models or coefficients from linear models within your Pipeline, then use feature selection techniques to identify which transformed features drive predictions. This clarifies model behavior and reduces dimensionality.

Related Skills

Explore More Data & Analytics Skills→
KaranKathur06KaranKathur06

Data Cleaning Pipeline

Clean tabular datasets by imputing missing values, removing duplicates, detecting outliers, and standardizing data types.

Community
Intermediate
snoodleboot-iosnoodleboot-io
2

model-monitoring

Monitor machine learning model performance and detect prediction or data drift.

Official
Intermediate
h4vzzh4vzz
33

hyperparameter-tuning

Explore hyperparameter configurations using grid, random, Bayesian, or Hyperband strategies.

Community
Advanced
scanadyscanady

data-ai-ml-pipeline

Design end-to-end ML pipelines with data ingestion, feature engineering, training, and deployment.

Community
Advanced